Coronary heart disease (CHD) affects millions of patients around the world. In recent years 1.6 million hospital admissions were for acute coronary syndromes (ACS) in the United States alone.¹ As more nations “Westernize” and their populations are exposed to increasing risk factors for coronary artery disease (CAD), this will be a substantial and perhaps the greatest international health concern in the next several decades.

The prevalence of CHD makes the variety of risk factors that lead to this disease process an important focus in prevention and understanding of important intervention strategies. Well-known major risk factors for CHD include cigarette smoking, hypertension (blood pressure ≥140/90 mmHg or on an antihypertensive medication), diabetes mellitus, high low-density lipoprotein (LDL), low high-density lipoprotein (HDL), family history of premature CHD (CHD in male first degree relative <55 years or in female first degree relative <65 years), and age (men ≥45 years, women ≥55 years).²

The Adult Treatment Panel III (ATP III) has classified diabetes mellitus as a CHD risk equivalent and previous studies have noted that diabetics have a two- to eightfold increased rate of future cardiovascular events than those of age and ethnically matched nondiabetics.³ Data has also revealed that nearly 75% of the deaths in diabetic patients result from CHD.⁴ There is ever increasing evidence that obesity, physical inactivity, and dyslipidemia (elevated triglyceride, small LDL, low HDL, increased blood pressure, and insulin resistance) coalesce to create a metabolic syndrome conferring a multitude of risk factors for CHD. Diagnosis and management of the metabolic syndrome is a major challenge for CHD prevention and treatment. In this review we describe the pathophysiology of dyslipidemias. We discussed the clinical assessment and management of dyslipidemia in a previously published review in the Journal of Clinical Metabolism and Diabetes (April 2011).

REVIEW

The relationship between cholesterol and CHD has been well established.^5,6 The Framingham Heart Study, initiated in 1948, revealed over the course of follow-up of the study population that cholesterol was one of the major culprits in the development of CHD,⁵ and studies published over the past several decades, like REVERSAL, PROVE-IT, CARE, and the Lipid Research Clinic study, have confirmed that pharmacological intervention to lower cholesterol indeed lowered CHD morbidity and mortality.^7–10

Fats are essential to life as we need these substances for energy storage, energy utilization, steroid hormone synthesis, bile acid synthesis, and cell membrane production. Dietary fats are necessary components that the body will utilize to synthesize essential fats it cannot obtain from the diet. Hyperlipidemia (elevated concentrations of any or all lipid components in the plasma) and dyslipidemia (average or increased total cholesterol, increased LDL, with low HDL and/or elevated triglycerides) are terms used in clinical practice to describe clinically the disorders of lipid metabolism. Dyslipoproteinemia more accurately describes the pathophysiology surrounding abnormalities of lipid metabolism and lipoprotein transport pathways that lead to CHD. There are several major types of lipids that circulate systemically and constitute the makeup of molecules involved in dyslipoproteinemia including cholesterol and cholesterol esters, triglycerides, and phospholipids. Cholesterol is a vital component of cell membranes and also is involved in the production of steroid hormones and bile acids. Cholesterol circulates in the plasma in the form of cholesterol esters as a part of lipoproteins. Triglycerides (TG) transport fatty acids that once hydrolyzed can be used for immediate metabolic needs or stored for future energy consumption. Phospholipids also are a component of all cell membranes and participate in signal transduction pathways. These lipid types and their components are circulating in the plasma in the form of lipoproteins, which are made up of lipids and apolipoproteins all arranged in a spherical structure. The apolipoproteins provide structure to the lipoprotein, allow binding to proteins for cellular uptake, are responsible for assembly and secretion of lipoproteins, and serve as coactivators or inhibitors of enzymes (Table 1).¹¹

An understanding of lipoprotein assembly and transport is a key to understanding the pathophysiology and clinical disease state of dyslipoproteinemia. Fat as previously mentioned is essential for human life. Fat, once ingested through the diet, is exposed to secreted pancreatic lipases. Pancreatic lipases hydrolyze triglycerides to free fatty acids (FFA). Bile acids, released from the gall bladder upon dietary intake, emulsify the TG and FFA, which facilitates the formation of micelles. Micelles are taken up by the intestinal brush border and FFA are re-esterfied to form TG. Once re-esterification is complete the TG are packaged into chylomicrons and enter the portal circulation. This process occurs rapidly and once in the plasma the chylomicrons may be exposed to lipoprotein lipase (LPL). The LPL cleaves the FFA from the TG and muscles rapidly take up the FFA to use for energy. Adipose tissue stores the TG for future energy utilization, a process that is dependent on and requires insulin. The FFA released from the cleavage of TG by LPL may also bind to fatty acid binding proteins that are transported back to the liver and repackaged as very low-density lipoproteins (VLDL). The VLDL is the body's way of ensuring readily available TG during times of need. Very low-density lipoproteins are secreted by the liver and are TG rich lipoproteins. With insulin resistance, the availability of insulin to store TG in adipose tissue is diminished or ineffective, and subsequently there is an increase in FFA delivered to the liver. This increase in FFA delivery leads to an increase in VLDL. Another potential pathway for the fat from our diet is its uptake by macrophages.¹² Macrophages express receptors that bind oxidized or remnant lipoproteins, which mediate the uptake of oxidized LDL into the macrophage. There is no negative feedback mechanism for suppressing this pathway thus allowing intimal macrophages to accumulate cholesterol. The process where remnant particles are incorporated into macrophages potentiates foam cell development and deposition of fatty streaks in the intima.¹³ Endothelial cells may also uptake remnant lipoproteins through the LOX-1 receptor.

Two important lipoproteins in the transport of cholesterol between the liver and the peripheral tissues are LDL and HDL. They also have important clinical indications for the development of CHD as HDL has an inverse relationship with CHD and LDL has been linked to cardiovascular mortality.^5,6 Low-density lipoprotein is responsible for cholesterol transport from the portal circulation to the periphery and in the peripheral cells cholesterol uptake is regulated by the LDL receptor. The LDL is compromised of predominantly cholesterol esters. When there are elevated levels of circulating TG in the plasma these cholesterol ester rich LDL lipoproteins change composition. They become deplete in their core cholesterol esters and saturated in TG. This process leads to smaller and denser LDL particles, which are more pathogenic in CHD. The mechanism is thought to be related to increased susceptibility of the smaller LDL particles to oxidation, increased circulating time, and a higher affinity to the extracellular matrix. The LDL once taken up in the periphery by the cell is degraded in the lysosome. The cholesterol produced from LDL acts to suppress the transcription of the HMG CoA reductase gene through the sterol regulatory element-binding protein (SREBP).¹⁴ High-density lipoprotein is responsible for reverse cholesterol transport from the periphery to the liver and is an important component in controlling excess storage of cholesterol in the periphery. The main components of HDL are apolipoprotein A-I and II. The HDL lipoproteins must acquire phospholipids and cholesterol to form HDL particles and LPL and ATP binding cassette A1 gene (ABCA1) are the primary mechanism. Mature HDL is formed once the cholesterol core is esterified by lecithin-cholesterol acyltransferase (LCAT). High-density lipoprotein is primarily catabolized by the liver via scavenger receptor class BI (SR-BI). Dysfunction of HDL mediated transport is an independent risk factor for CHD as there is decreased transport back to the liver for excretion in the bile and feces.

Lipoprotein disorders constitute a disruption or alteration to the above transport or processing process. These alterations lead to the clinical disease states that one witnesses clinically in patients. Lipoprotein disorders can be classified into primary causes of dyslipoproteinemia and secondary causes. Primary causes are genetically based and usually inherited (Table 2). Secondary causes result from metabolic derangements, chronic disease states, or medications. Familial Hypercholesterolemia (FH) is an autosomal co-dominant disorder caused by a defect in the LDL receptor. This defect leads to accumulation of LDL particles in the plasma. The prevalence is 1 in 500 and is associated with a LDL that is greater than the 95th percentile for age and gender matched controls. Patients with this disorder have an increased risk of CHD by their 30–40s for men and 40–50s for women. Patients often have xanthalasmas, xanthomas on the extensor tendons, and corneal arcus. Familial Hypertriglyceridemia (FHTG) is characterized by elevated TG, elevated VLDL, decreased LDL, low HDL, and total cholesterol that is normal or increased. The etiology is overproduction of VLDL by the liver. There is not as strong as a relationship between premature CAD as with FH. The FHTG is genetically linked to first degree relatives but there is varying phenotypic expression. FHTG is potentiated by the diet and alcohol consumption. Fasting TG levels in FHTG are typically 200–500 mg/dL. Familial Hyperchylomicronemia (FHC) is a rare disorder of elevated circulating TG secondary to decreased or absent LPL activity. Deficient LPL activity leads to insufficient hydrolysis of the chylomicrons and subsequent accumulation in the plasma. Fasting TG levels are in excess of 1000 mg/dL and if serum is left to settle, a thin milky white band can be noted on the top of the collected serum. Patients afflicted with this also tend to have decreased HDL levels. Patients with FHTG often have recurrent pancreatitis, xanthomas, xerophthalmia, xerostomia, and even psychological disorders. Familial Combined Hyperlipidemia (FCH) occurs secondary to overproduction of apolipoproteins B containing lipoproteins. This is one of the most common forms of primary lipoprotein disorders affecting 1 in 50 individuals. Analysis of lipoproteins reveals elevated total cholesterol and TG. The FCH is present in 10–20% of individuals presenting with premature CAD.¹⁵ Tangier Disease and Familial High-Density Lipoprotein Deficiency are disorders characterized by deficient or nearly absent HDL. In Tangier's Disease and familial HDL deficiency, patients have severely reduced HDL secondary to reduced cholesterol exit from the cell membrane. The cause is from a defect in ATP binding cassette A1 gene, which is responsible for transporting cellular cholesterol to the cell membrane for export. The genetic absence of this gene leads to decreased lipidification of the HDL apoliproteins and thus reduced functioning HDL. Familial Dysbetalipoproteinemia (FDB) is caused by a defect in apolipoprotein E mediated clearance of remnant lipoproteins. This results in elevated serum LDL and TG that leads to enriched VLDL.¹⁶ Less than 1% of those with CHD have this disorder. The above disorders are important clinically as patients presenting with premature CAD or who have family members with premature CAD should be screened for the above disorders. The goal of identification is for early lipid screening and aggressive medical treatment.

Secondary causes of the lipoprotein disorders often are reversible or, at the least, controllable. Hormonal causes include hypothyroidism, which is often characterized by elevated LDL, TG, or both, and can be corrected with treatment of the hypothyroidism. Estrogens tend to increase TG and HDL and they may decrease LDL, but are not used clinically for treatment due to the increased risk of cardiovascular events. Renal disorders including glomerulonephritis and chronic kidney disease are associated with elevated LDL and elevated TG with decreased HDL respectively. Liver disease can lead to the abnormal formation of lipoproteins leading to dyslipoproteinemia. Sedentary lifestyle with disproportionate fat and refined sugar intake contributes a large part to elevated lipoprotein and lipid levels. Medications including thiazide diuretics, beta blockers, and corticosteroids are a few examples of medications that all increase TG levels. One of the increasingly present secondary causes are metabolic related disorders. Glycogen storage diseases and lipodystrophies fit into this category, but by far the most common are those with diabetes and metabolic syndrome. Metabolic syndrome is now the most common secondary cause of dyslipoproteinemia. This syndrome is characterized by abdominal obesity, elevated blood pressure, dyslipidemia, and peripheral insulin resistance. As mentioned previously, insulin is a vital component in the storage of circulating TG in adipose tissue and when absent leads to increased FFA uptake by the liver and increased production of VLDL. Patients with metabolic syndrome often have elevated TG and low HDL. The role of adipose tissue has been recognized as an endocrine organ releasing substances like adiponectin, leptin, and TNF-α in response to differing stimuli.¹⁷ Obesity and insulin resistance seem to play the main role in the pathogenesis of metabolic syndrome and the subsequent development of dyslipidemia and increased cardiovascular risk.¹⁸

CONCLUSION

The assembly, storage, and transportation of lipoproteins is a complex system with a multitude of components. Understanding the basic role of lipoproteins in the body and how disorders of lipoprotein assembly, storage, and transport manifest clinically are important for the prevention and treatment of dyslipidemia.¹⁹ By understanding each lipoprotein and its role in the body one can gain a better understanding into the mechanism and targets for treating and identifying those with dyslipidemia. Dyslipidemia is an important, but treatable and controllable risk factor for CHD and with better knowledge of lipid pathophysiology and lipoprotein transport clinicians can better serve their patients. Our group has recently published a review describing the clinical management of dyslipidemia in the Journal of Clinical Metabolism and Diabetes.¹⁹

REFERENCES

1. Roger, Veronique, et al. Heart Disease and Stroke Statistics. 2011 Update. Circulation. 2011;123:e18–e209.
2. National Institutes of Health. Third report of the national cholesterol education program expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (ATP III), 2007; Bethesda, MD: National Institutes of Health.
3. Howard BV, Rodriguez BL, Bennett PH, et al. Prevention conference VI: Diabetes and cardiovascular disease: Writing group I: Epidemiology. Circulation. 2002;105:e132.
4. Gu K, Cowie CC, Harris MI. Mortality in adults with and without diabetes in a national cohort of the U.S. population, 1971–1993. Diabetes Care. 2002;25:1129.
5. Dawber T, Meadors G, Moore F.; National Heart Institute, National Institutes of Health, Public Health Service, Federal Security Agency, Washington, DC. Epidemiological Approaches to Heart Disease: The Framingham Study. Presented at a Joint Session of the Epidemiology, Health Officers, Medical Care, and Statistics Sections of the American Public Health Association, at the 78th Annual Meeting, St. Louis, MO, November 3, 1950.
6. Stamler J, Wentworth D, Neaton JD. Is relationship between serum cholesterol and risk of premature death from coronary heart disease continuous and graded? Findings in 356,222 primary screenees of the Multiple Risk Factor Modification Trial (MRFIT). JAMA. 1986;256:2823– 2928.
7. Nissen SE, Tuzcu EM, Schoenhagen P, et al. Effect of intensive compared with moderate lipid-lowering therapy on progression of coronary atherosclerosis. JAMA. 2004;291:1071.
8. Cannon CP, Braunwald E, McCabe CH, et al. Intensive versus moderate lipid lowering with statins after acute coronary syndromes. NEJM. 2004; 350:1495.
9. Sacks FM, Pfeffer MA, Moye LA, et al. The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. Cholesterol and Recurrent Events Trial investigators. NEJM. 1996;335(14):1001–1009.
10. Williams O, Stinnett S, Chambless L, et al. Populations and methods for assessing dyslipoproteinemia and its correlates. The Lipid Research Clinics Program Prevalence Study. Circulation. 1986;73(suppl 1):4–11.
11. Libby P, Genest J, Gotto A, Jr. Braunwald’s heart disease: Lipoprotein disorders and cardiovascular disease. in D P Zippes, P Libby, R O Bonow, E Braunwald (Eds.), A text book of cardiovascular medicine, (pp. 1013– 1033). 2005; Philadelphia: Elsevier Saunders.
12. Davies MJ. The composition of coronary artery plaques. NEJM. 1997; 336:1312.
13. Gianturco SH, Bradley WA. Pathophysiology of triglyceride-rich lipoproteins in atherothrombosis: cellular aspects. Clin Cardiol. 1999; 22(suppl II):7–14.
14. Brown MS, Goldstein JL. A proteolytic pathway that controls the cholesterol content of membranes, cells, and blood. Proc Natl Acad Sci USA. 1999;96:11041–11048.
15. Genest JJ, Martin-Munsky SS, McNamara JR, et al. Familial lipoprotein disorders in patients with premature coronary artery disease. Circulation. 1992;85:2025–2033.
16. Miller M. Current perspectives on the management of hypertriglyceridemia. Am Heart J. 2000;140:232–240.
17. Ronti T. The endocrine factor of adipose tissue: an update. Clin Endocrinol. 2006;64:355–365.
18. Schindler C. The metabolic syndrome as an endocrine disease: is there an effective pharmacotherapeutic strategy optimally targeting the pathogenesis? Therapeutic Adv Cardiovasc Dis. 2007;1:7–26.
19. Sloan K, Vacek J. Treatment of dyslipidemia. J Clin Metab Diabetes. 2011; 2:1–4.

Proteinuria is a classic sign of kidney disease, which was considered as a marker of progressive kidney disease, whereas microalbuminuria was established first as an early sign that predicts overt proteinuria in diabetic nephropathy. Later, several evidence proved that pathological “albumin-leakage” appears not only in diabetes but also means an increased risk for further vascular complications in all types of cardiovascular diseases reflecting widespread endothelial dysfunction.^1–4 Albuminuria is probably due to damage of the glomerular endothelial cell glycocalyx and/or the glomerular podocytes caused by local or systemic factors such as inflammatory mediators, glycemic products, free radicals, and others.⁵ Nowadays, microalbuminuria is mainly recognized as a cardiovascular risk factor, with no lower limit.⁶ According to the latest clinical studies, detection of albuminuria and proteinuria is suggested not only for monitoring of diabetic nephropathy but these laboratory test are necessary in all cases when early cardiovascular complications may develop.^7,8 Even though the above normal urinary albumin level predicts the development of micro- and macroalbuminuria, a higher baseline means higher mortality later.⁹ Above this, there is no doubt that albuminuria is a good marker of progression of kidney disease, whereas proteinuria reflects increased cardiovascular morbidity and mortality also. Decreased albuminuria or proteinuria resulted in decreased risk for both cardiovascular events and end-stage kidney disease.

The “epidemic” of chronic kidney disease (CKD) affects about 10%–15% of the adult population nowadays. Most of such patients are asymptomatic until severe renal failure develops (stage 4–5 CKD). Due to the introduction of automated estimated glomerular filtration rate (eGFR) reporting by laboratories, facilitated by Kidney Disease Outcomes Quality Initiative study of National Kidney Foundation (KDOQI) in 2002, large number of asymptomatic stage 3 CKD patients (6%–8% of population) can be identified within primary and secondary care. This practice was started in our institute in 2006 and was followed nation wide 2 years later.¹⁰ This resulted in significantly increased workload of nephrological providers, but many of these patients had a low risk for progressive kidney disease and did not require any special treatment. However, this practice is not able to identify the large number of patients with stage 1–2 CKD (5%–6% of population) at increased risk for renal disease progression and cardiovascular disease.

Testing urine samples for the presence of protein or albumin is a valuable tool for identifying CKD patients with eGFR above 60 ml/min/1.73 m², and a subpopulation of CKD patients with greatly increased risk for progressive nephropathy and cardiovascular disease or death. Therefore, national and international guidelines uniformly recommend testing for proteinuria or albuminuria for patients at high risk of these diseases.^11–16

However, there are several problems and disagreements that need to be resolved:

• Should asymptomatic urinary track infection be excluded, when urine is screened for proteinuria?

It is obvious that in certain circumstances (e.g. acute illnesses), screening should be postponed because of transient proteinuria or albuminuria, but guidelines are not consistent as to whether asymptomatic urinary tract infection (UTI) has to be excluded or not before testing.

• Can urinary strip test be used for screening?

Some guidelines continuously recommend its use as first-line tests, but most of them are neither sensitive enough for urinary albumin and nor cost-effective.

• Which urine sample should be tested in laboratory?

Traditionally, these tests are performed from the 24-hour collected urine. However, in an ambulatory setting, collection of urine for 24 hours is not convenient. Therefore, other samples (overnight, random, or first morning urine) are also widespread. Most guidelines prefer usage of random or first morning urine sample with simultaneous creatinine determination.

• Which methods are preferred in central laboratory for measuring urinary albumin and total protein?

Interpretation of urinary albumin results depends on laboratory methods and quality of the reagent and calibrator.

• Is testing for albuminuria or for proteinuria preferred?

There is no consensus among guidelines and laboratories in respect of which is the preferable test—total protein or albumin measurement. Some prefer albumin for diabetic nephropathy and total protein for nondiabetic nephropathy, whereas others prefer albumin for all patient with CKD.

• How can urine samples be stored for albumin or protein determination?

• What are normal reference ranges and critical levels of albumin and protein excretion with respect to nephrological referral?

Although different biochemical methods result in great variations, upper limits of reference ranges in laboratory diagnostic method descriptions hardly change. As urinary albumin is an independent marker of vascular damage, recently lower and method-dependent clinical decision limits were suggested.^17–19

For these reasons, we have studied the comprehensive guidelines and recent publications, compared analytical sensitivity and reproducibility of up-to-date clinical biochemical methods, and surveyed the relevant methods in Hungarian central laboratories. On the basis of these data and our diagnostic experiences, in 2011 a clinical recommendation was made in agreement with the Hungarian Association of Nephrology and Nephrology Workgroup of the Hungarian Association of Laboratory Medicine.

SHOULD ASYMPTOMATIC UTI BE TESTED WHEN URINE IS SCREENED FOR PROTEINURIA?

Symptomatic UTI is frequently associated with proteinuria and albuminuria. Febrile pyelonephritis is usually associated with tubular proteinuria, sometimes also includes albuminuria that may be present rarely in cystitis also. Therefore, it was usually recommended to exclude UTI, when urine is screened for proteinuria or albuminuria. However, there is no evidence to confirm such a causal relationship between asymptomatic UTI and proteinuria or even microalbuminuria.²⁰ It is possible that individuals with asymptomatic UTI and proteinuria/albuminuria may have an underlying kidney disease (eg, reflux or diabetic nephropathy) that predisposes to both proteinuria/albuminuria and UTI. Overemphasis of the importance of exclusion of UTI increases costs and may delay the establishment of proteinuria. We do not recommend screening for asymptomatic UTI.

CAN URINARY STRIP TEST BE USED FOR SCREENING?

For decades, urinary dipsticks have been used for the detection of proteinuria in daily routine. However, if we apply dipsticks, preanalytical circumstances that can produce misleading results should be taken into consideration. For example:

• In concentrated urine, the results are often false positive, whereas in dilute samples, we can obtain false-negative results.

• Some drugs and metabolites can interact with analytes and thus modify the results.

• Developing color on the strip depends on the urinary pH.

• Urinary protein strips are sensitive to albumin, but most of them do not react in the microalbuminuria range and may result in false-negative reactions at Bence-Jones proteinuria.

There is consensus among guidelines that the major problem with this test is the poor negative predictive value, high false-negative result rate when the urine is diluted. In spite of this, KDOQI allowed its usage in most cases, and the UK CKD guideline also continuously incorporated it (except in diabetic patients). Other guidelines and, in their recent excellent review, Lamb et al.²¹ did not recommend the usage of conventional dipsticks when screening individuals at high risk of CKD. Albumin-specific reagent strips could be used but they are not widespread. If dipsticks are accepted, the positive results usually are followed by laboratory test and that approach is not cost-effective.²² We are against the usage of urinary dipsticks for proteinuria in screening of asymptomatic individuals at high risk for CKD.

WHICH URINE SAMPLE SHOULD BE TESTED IN LABORATORY?

Intraindividual albumin excretion changes in a wide range, and diurnal variance of albumin excretion is relatively high. Therefore, the gold standard method for proteinuria was the measurement of protein in a 24-hour collected urine. However, in an ambulatory setting, a 24-hour collection of urine is not convenient. The imprecision of collection is responsible mostly for the high individual coefficient of variation that may reach up to 150%. Therefore, these results seem to be unreliable and not comparable.

To avoid this bias, the use of random, preferably first morning urine sample was suggested. From first morning urine, a single determination of albumin is not enough because urine can be more or less concentrated with considerable individual variance. Therefore, critical differences (clinically significant change in a parameter compared to the previous one) in urinary albumin concentration may reach 140%. To improve this, correction to the urine creatinine content was suggested. A large number of clinical studies suggested that protein and albumin determination in 24-hour collected urine should be replaced by protein and creatinine ratio (PCR) or albumin and creatinine ratio (ACR) measured from the same random urine sample.^23–29 The ACR determined from a random first morning urine is the most reliable result, critical difference in this case is 40%. Today, all guidelines suggest the use of ACR or PCR. We recommend that laboratories automatically calculate and report these parameters also, when a proteinuria or albuminuria measurement is ordered.

WHICH CENTRAL LABORATORY METHODS ARE PREFERRED FOR MEASURING URINARY ALBUMIN AND TOTAL PROTEIN?

Today, the majority of routine laboratory methods for urinary albumin are immunoassays. Degradated products of albumin have epitopes with different immunological activity. For that reason, “non-immunogenic” forms of albumin may be detected by high performance liquid chromatography (HPLC) in the urine of diabetic patients. Recent studies predict that these nonalbumin proteins measured by HPLC coelute with albumin.^30–39 Other evidence indicates that the HPLC test has no advantages over immunoassay as a predictive marker for cardiovascular events or mortality.^32,33 Therefore, currently nonimmunological methods of urinary albumin determination can be considered as research methods.

Despite general use of immunoassays, the interlaboratory variation in urinary albumin determination is relatively high, due to usage of different calibrators, although albumin calibrators traceable to the international serum standard are available. On the other hand, medium is different in urine and serum, accordingly urinary albumin fractions differ from the serum albumin due to posttranslational modification and degradation. The International Federation of Clinical Chemistry considered this situation and suggested the use of calibrators traceable to a serum albumin reference material (CRM470, Institute for Reference Materials and Measurements, Belgium).¹⁸ Today, the modern urine albumin calibrators are traceable to CRM 470 that is diluted to the concentrations measured in urine, although there are no standard diluents or standard procedures for dilution. Recently, a new reference material for urinary albumin has been developed by a Japanese group and measured by isotope dilution mass spectrometry. It is supposed that evaluation of these reference systems by the Joint Committee for Traceability in Laboratory Medicine will further improve the accuracy of urinary albumin measurements.¹⁹

Measurement of total protein seems more difficult in urine by the immunoassay approach. Because a variable mixture of protein is measured, it is difficult to define a standardized reference material. There is no reference measurement, and turbidimetric and colorimetric methods are widely used currently. Therefore, there is significant inter- and intralaboratory variation that decreases at higher concentrations of urinary total protein. At lower concentrations (at range of microalbuminuria), urinary album in assays are clearly superior to total protein assays.²¹

During 2007–2008, in the Hungarian external quality control examination, the interlaboratory variation of the immunoturbidimetric urinary albumin measurement was 9%–41% at microalbumin level and 3%–28% at macroalbumin level. The albumin results might be scattered because 37 participants purchased the reagent from 13 different manufacturers, and some of them used calibrators not traceable to the international albumin standard (CRM470). Although majority of our laboratories used immunoturbidimetry (?91%) and only 3% applied nephelometry, these results are not comparable because nephelometry produces 2–3 times higher albumin values. For determination of urinary protein, majority of our laboratories applied the more sensitive pyrogallol reagent (74%), with a 10–20 mg/l lower limit of detection, and ?12% of the laboratories used turbidimetry with 40–60 mg/l lower limit of detection (Figure 1 and 2). Hungarian interlaboratory variation of urinary protein was 6% for the pirogallol method and 11% for the turbidimetry at normal levels of protein, and 5%–7% for both methods at protein >450 mg/l. Fortunately, these sensitive protein methods are reliable, economical, and then we found the results more comparable than in the case of urinary albumin.

Testing for Albuminuria or for Proteinuria is Preferred?

There is no consensus among guidelines and laboratories in respect of the most preferable test—total protein or albumin measurement. Some of them prefer albumin for diabetic nephropathy and total protein in nondiabetic nephropathy Scottish Intercollegiate Guidlines Network (SIGN), Caring for Australasians with Renal Impairment (CARI), (UK CKD), whereas others prefer albumin for all patient with CKD (KDOQI, KDIGO, NICE).

It is accepted that total protein measurement is insufficient to detect the onset of diabetic nephropathy; therefore, microalbuminuria has to be tested. Large epidemiological data show that microalbuminuria is also common in the nondiabetic population and was associated with cardiovascular morbidity and mortality. Although there is evidence of beneficial effects of Angiotensin Converting Enzyme Inhibitor (ACEI)/Angiotensin II Receptor Blockers (ARB) treatment for lower than 0.5 g/day, proteinuria is still lacking in the nondiabetic population, this is likely not different from that seen in diabetic patients. Consequently, detection of microalbuminuria seems preferable in nondiabetic population also; therefore, NICE and KDOQI suggested usage of ACR for all patients with CKD.

However, there are arguments against the universal use of ACR. Theoretically, the nonalbuminuric (tubular and overflow) proteinuria could be missed, but relative increases of low molecular weight protein in tubulointerstitial disease are usually accompanied by significant albuminuria due to the decreased reabsorption of filtered albumin. Testing for PCR may also miss tubular proteinuria; therefore, this problem seemed overestimated.²¹ Recently, Methven et al. found that effectivities of PCR and ACR are equal as predictors of renal outcomes and mortality in patients with CKD in a 3.5 years follow-up.^34,35 Unexpectedly, PCR performed well at low levels (0.15–0.45 g/day) of proteinuria. Nonalbumin proteinuria has some additional prognostic significance that is not captured by albumin measurement (such as the worse prognosis of nonselective proteinuria in glomerular disease).

ACR is 2–10 times more expensive than PCR depending on local financial agreements between hospitals and suppliers (in Hungary at present laboratories receive 36 + 36 points for PCR and 473 + 36 points for ACR). Economic aspects are very important nowadays especially in countries with lower economical background. Use of ACR at high proteinuric levels additionally increases the costs because the upper limit of urinary albumin measurement range generally is not greater than 500 mg/l and that results in dilutions of samples. Additionally, samples with very high albumin concentration may be falsely reported as low, due to the antigen excess (“prozone phenomenon”).

When all these aspects are considered, we prefer usage of ACR for screening of CKD, but we suggest the use of PCR when proteinuria is present (Table 1).

HOW CAN URINE SAMPLES BE STORED FOR ALBUMIN OR PROTEIN DETERMINATION?

During storage, the albumin content of the sample is decreased because of degradation in spite of cooling; therefore, albumin and total protein examinations are best performed on the day of sampling. If it is not possible, the sample can be stored at +4°C for approximately 7 days. For a longer period, the samples have to be stored at –70°C. At a temperature of –20°C, degradation of albumin may cause a ?40% decrease in concentration compared to the initial value. Adsorbing character of the sample tube also modifies the albumin content, and after prolonged storage a considerable part of albumin is irreversibly adsorbed to the walls of the tube. The loss of albumin is less in hydrophilic glass vials or containers.

WHAT ARE NORMAL REFERENCE RANGES AND CRITICAL LEVELS OF ALBUMIN AND PROTEIN EXCRETION WITH RESPECT TO NEPHROLOGICAL REFERRAL?

Traditionally, microalbuminuria is defined as a daily albumin excretion of between 30 and 300 mg. Guidelines are quite consistent in accepting the equivalent lower limit of ACR ( ?2.5 mg/mmol for male, ?3.5 mg/mmol for female) and 30 mg/mmol as upper limit, but there are some differences in the definition of proteinuria in terms of PCR. We suggest the use of 45 mg/mmol value (Table 2), as in the UK CKD guide, that is slightly lower than 50 mg/mmol reported in NICE. There are different suggestions as to the use of the terms microalbuminuria and macroalbuminuria (instead of proteinuria) or to universal use of albuminuria. We are agreeing with Miller³⁶ that this term may cause confusion and do not recommend the use of macroalbuminuria, but the term microalbuminuria is already widely used in general practice. The risk stratifications also need some classification.

Most of the guidelines suggest nephrological referral when proteinuria is above 1 g/day, but in diabetic patients it could be even higher, above 3.5 g/day, independently of eGFR.

Recent analysis in the general US population (more than 13 000 individuals from National Health and Nutrition Examination Survey (NHANES) III, followed for 13 years) showed that combination of eGFR and albuminuria resulted in precise estimation for cardiovascular risk and all-cause mortality. ³⁷ Even larger population-based cohorts of more than 65 000 subjects followed up for 10 years indicated that all levels of eGFR should be accompanied by urinary ACR measurement.³⁸ The predictive power of combined ACR and eGFR measurements was very high for future kidney failure. Adjustment for age, sex, diabetes, hypertension, and other potential risk factors did not improve predictions further. They suggested a 12-category matrix (eGFR > 60, 45–60, 30–45, <30, subdivided by ACR into normo-, micro-, and macroalbuminuria) to the risk stratification. This table was more useful than the current KDOQI CKD classification. For example, the combined ACR–eGFR strategy could detect the same number of future ESRD; 66% versus 69%, with much fewer nephrological referrals: 14 versus 47 per 1000 subjects screened, if compared with a screening strategy of referring all CKD stage 3–4 patients. This table also reflects cardiovascular risk very well. Subjects with low, medium, or high risk for future ESRD had low (RR < 2), medium (RR 2-3), and high (RR > 3) risk, respectively for future cardiovascular deaths also.³⁹

Nephrologists are well familiar with nephrotic range proteinuria being the most severe with regard to both renal disease progression and accelerated atherosclerosis also. Therefore, we suggest adding nephrotic range proteinuria to the eGFR–ACR table for the estimation of relative risk of kidney disease progression or cardiovascular events (Table 3).

OUR SUGGESTIONS FOR SCREENING AND EVALUATION OF PROTEINURIA/MICROALBUMINURIA ON A DAILY ROUTINE BASIS:

• For assessment of progressive CKD and cardiovascular disease risk, screening of proteinuria and microalbuminuria is suggested in high-risk populations beside the eGFR determination.

• Acute illnesses,¯ fever, exercise, uncontrolled hypertension or hyperglycaemia, and symptomatic UTI cause transient albuminuria, proteinuria, and/or hematuria. Therefore, these should be treated first while screening is suggested after recovery. Asymptomatic UTI should not be tested.

• To establish correct urinary albumin and protein excretion, when these tests are ordered laboratories should automatically measure the creatinine concentrations also in the same urinary samples. The calculated albumin and creatinine ratio (ACR) and protein creatinine ratio (PCR) should be reported in mg/mmol units.

• For screening of proteinuria, PCR measurement is suggested from a random sample. If proteinuria is verified (PCR > 45 mg/mmol), retest from the first morning urine is needed to confirm it, except when PCR > 100 mg/mmol. If the possibility of microalbuminuria arises (PCR is between 15 and 45 mg/mmol), the first morning sample has to be sent for ACR determination.

• For screening of microalbuminuria, measurement of ACR is suggested from a first morning urine sample. If it is positive (ACR > 2.5 mg/mmol in male and >3.5 mg/mmol in female), a retest always from a first morning urine sample is needed to confirm it, except when ACR >70 mg/mmol.

• For follow-up of albuminuria or proteinuria, ACR or PCR measurement from a first morning urine sample is suggested. In general practice and outpatient clinics, 24-hour urine collection is not necessary. In the range of microalbuminuria (ACR < 30 mg/mmol, PCR < 45 mg/mmol), ACR is preferable. When proteinuria is detected (ACR > 30 mg/mmol, PCR > 45 mg/mmol), PCR is preferable.

• In the case of newly diagnosed proteinuria/albuminuria or CKD (eGFR < 60 ml/min/1.73 m²), screening of hematuria by urinary strips or urinary sediment examination is also suggested.

• Nephrological referral is suggested (because possibility of kidney biopsy arises), if proteinuria is significant (PCR > 100 mg/mmol) in a nondiabetic patient or if it is in nephritic range (PCR > 350 mg/mmol) in a diabetic patient, and in all cases when hematuria and proteinuria (ACR >30 mg/mmol, PCR > 45 mg/mmol) are present simultaneously. Asymptomatic microalbuminuria (ACR < 30 mg/mmol/l) warrants nephrological referral only when eGFR is less than 45 ml/min/1.73 m² or when microscopic hematuria is present, but urological investigation did not identify the reasons. Macroscopic hematuria should be checked first also by an urologist, but in the case of concomitant proteinuria and increasing serum creatinine, urgent nephrological referral is suggested (because of the possibility of rapid progressive glomerulonephritis).

CONCLUSIONS

The protein content of urine depends on the diurnal changes in excretion. For this reason, measurement of urinary protein and albumin concentrations is not enough for classification of proteinuria and albuminuria. Application of ACR and PCR means an adjustment to the diurnal changes, and 24-hour collection of urine can be ignored in general practice. These calculated parameters are reliable markers of protein and albumin excretion and generally show good correlation. Even so, they are not convertible because the ratio of albumin and other proteins depends on the clinical state. Generally, the sensitivity of albumin strips is not accepted for the detection of initial stages of albumin excretion, and they are not applicable for diagnosis of microalbuminuria. For the laboratories, we recommend that the albumin/creatinine values be determined preferably from the first morning urine and report them together with gender-dependent reference ranges. At the stage of proteinuria, the determination of protein/creatinine value from the first morning urine is recommended. These are early, independent markers of cardiovascular diseases; therefore, the precise sampling and application of up-to-date laboratory methods combined with eGFR determinations have a great impact in the early diagnosis and treatment of hypertension, diabetes mellitus, kidney, and vascular diseases.

Acknowledgements: Data on Hungarian External Quality Control examination were supplied with the kind help of Dr Erika Sárkány, QualiCont Public Company, Szeged, Hungary.

Disclosure: This work was supported by the TAMOP 4.2.1./B-091/1/KONV-2010-0007 project to A. V. Oláh

REFERENCES

1. Jarrett RJ, Viberti GC, Argyropoulos A, et al. Microalbuminuria predicts mortality in non-insulin dependent diabetics. Diabet Med. 1984;1:17–19.
2. Gerstein HC, Mann JFE, Pogue J, et al. Prevalence and determinants of microalbuminuria in high-risk diabetic and non-diabetic patients in the Heart Outcomes Prevention Evaluation Study. Diabet Care. 2000; 23,S2:SB35–SB39.
3. Gerstein HC, Mann JFE, Qilong Y, et al. Albuminuria and risk of cardiovascular events, death and hearth failure in diabetic and nondiabetic individuals. JAMA. 2001;286:421–426.
4. Hillege HL, Fidler V, Diercks GF, et al. Urinary albumin excretion predicts cardiovascular and non cardiovascular mortality in general population. Circulation. 2002;106:1777–1782.
5. Satchell SC, Tooke JE. What is the mechanism of microalbuminuria in diabetes: a role for the glomerular endothelium? Diabetologia. 2008;51: 714–725.
6. Klausen K, Borch-Johnsen K, Feldt-Rasmussen B, et al. Very low levels of microalbuminuria are associated with increased risk of coronary heart disease and death independently of renal function, hypertension, and diabetes. Circulation. 2004;110:32–35.
7. Mann JFE, Gerstein HC, Qilong Y, et al. Development of renal disease in people at high cardiovascular risk. J Am Soc Nephrol. 2003;14:641–647.
8. De Zeeuw D. Albuminuria, not only a cardiovascular/renal risk marker, but also a target for treatment? Kidney Int. 2004; S92:S2–S6.
9. Murussi M, Campagnolo N, Beck MO, et al. High-normal levels of albuminuria predict the development of micro- and macro-albuminuria and increased mortality in Brasilian Type 2 diabetic patients: an 8 year follow-up study. Diabet Med. 2007;24:1136–1142.
10. Mátyus J, Oláh AV, Ujhelyi L, et al. The epidemic of chronic kidney disease requires the estimation of glomerular filtration rate. Hungarian Med J. 2008;149:77–82.
11. National Kidney Foundation. K/DOQI Clinical Practice Guidelines for Chronic Kidney Disease: evaluation, classification and stratification. Am J Kidney Dis. 2002;39(S1):S1–S266.
12. The caring for Australasians with renal impairment guidelines: urine protein as a diagnostic test. Nephrology. 2004;9(S3):S3–S7.
13. Levey AS, Eckardt KU, Tsukamoto Y, et al. Definition and classification of chronic kidney disease: a position statement from Kidney Disease: improving Global Outcomes (KDIGO). Kidney Int. 2005;67:2089–2100.
14. Joint Specialty Committee on Renal Medicine of the Royal College of Physicians and the Renal Association and the Royal College of General Practitioners: Chronic Kidney Disease in Adults: UK Guidelines for identification, management and referral (2008) www.renal.org/CKD guide/full/UKCKDfull.pdf 15. National Institute for Health and Clinical Excellence: Chronic Kidney Disease. National clinical guideline for early identification and management of adults in primary and secondary care. CG73. London: National Institute for Health and Clinical Excellence; 2008. http://guidance.nice.org.uk/ CG73 16. Scottish Intercollegiate Guidelines Network. Diagnosis and management of chronic kidney disease: a national guideline. Edinburgh: Scottish Intercollegiate Guidelines Network; 2008. http://www.sign.ac.uk/guidelines/ fulltext/103/index.html 17. Gansevoort RT, Verhave JC, Hillege HL, et al. The validity of screening based on spot morning urine samples to detect subjects with microalbuminuria in the general population. Kidney Int. 2005;67:S28–S35.
18. Whicher JT, Baudner S, Bienvenu J, et al. New initiatives in the standardization of protein measurements. Pure & Appl Chem. 1996;68(10): 1851–1856.
19. Database of higher-order reference materials, measurement methods/ procedures, and services. http://www.bipm.org/jctlm/ Accessed January 27, 2009.
20. Carter LJ, Thomson CRV, Stevens PE, et al. Does urinary tract infection cause proteinuria or microalbuminura? A systematic review. Nephrol Dial Transplant. 2006;21:3031–3037.
21. Lamb E, Mackenzie F, Stevens P. How should proteinuria be detected and measured? Ann Clin Biochem. 2009;46:205–217.
22. Chronic kidney disease, costing template of implementating NICE guidance, NICE clinical guideline 73 (2008), http://guidance.nice. org.uk/CG73/CostingTemplate/xls/English 23. Ginsberg JM, Chang BS, Matarese RA, Garella S. Use of single voided urine samples to estimate quantitative proteinuria. N Engl J Med. 1983; 309:1543–1546.
24. Marshall SM, Alberti KG. Screening for early diabetic nephropathy. Ann Clin Biochem. 1986;23:195–197.
25. Lemann J Jr, Doumas BT. Proteinuria in health and disease assessed by measuring the urinary protein/creatinine ratio. Clin Chem. 1987; 33:297–299.
26. Hutchison AS, O’Reilly DS, MacCuish AC. Albumin excretion rate, albumin concentration, and albumin/creatinine ratio compared for screening diabetics for slight albuminuria. Clin Chem. 1988;34:2019–2021.
27. Rodby RA, Rohde RD, Sharon Z, et al. The urine protein to creatinine ratio as a predictor of 24-hour urine protein excretion in type 1 diabetic patients with nephropathy. Am J Kidney Dis. 1995;26(6):904–909.
28. Price CP, Newall RG, Boyd JC. Use of protein:creatinine ratio measurements on random urine samples for prediction of significant proteinuria. Clin Chem. 2005;51(9):1577–1586.
29. Ruggenenti P, Gaspari F, Perna A, et al. Cross sectional longitudinal study of spot morning urine protein. British Med J. 1998;316:504–509.
30. Sviridov D, Meilinger B, Drake SK, et al. Co-elution of other proteins with albumin during size-exclusion HPLC: implications for urine albumin analysis. Clin Chem. 2006;52:389–397.
31. Shaikh A, Seegmiller J, Borland T, et al. Comparison between immunoturbidimetry, size-exclusion chromatography, and LC-MS to quantify urinary albumin. Clin Chem. 2008;54:1504–1510.
32. McQeen MJ, Gerstein HC, Pogue J, et al. Reevaluation by high performance liquid chromatography: clinical significance of microalbuminuria in individuals at high risk of cardiovascular disease in the Heart Outcomes Prevention Evaluation (HOPE) study. Am J Kidney Dis. 2006;48:889–896.
33. Wang Z, Hoy WE, Nicol JL, et al. Predictive value of nephelometric and high performance liquid chromatography assays of urine albumin for mortality in a high-risk aboriginal population. Am J Kidney Dis. 2008;52:672–682.
34. Methven S, Macgregor MS, Traynor JP, O’Reilly DS, Deighan CJ. Assessing proteinuria in chronic kidney disease: protein-creatinine ratio versus albumin-creatinine ratio. Nephrol Dial Transplant. 2010; 25:2991–2996.
35. Methven S, MacGregor MS, Traynor JP, et al. Comparison of urinary albumin and urinary total protein as predictors of patients outcomes in CKD. Am J Kidney Dis. 2010;57:21–28, DOI: 10.1053/j.ajkd.2010.08.009.
36. Miller WG, Bruns DE. Laboratory issues in measuring and reporting urine albumin. Nephrol Dial Transplant. 2009;24:717–718.
37. Astor BC, Hallan SI, Miller ER, et al. Glomerular filtration rate, albuminuria and risk of cardiovascular and all-cause mortality in the US population. Am J Epidemiol. 2008;167:1226–1234.
38. Hallan SI, Ritz E, Lydersen S, et al. Combining GFR and albuminuria to classify CKD improves prediction of ESRD. J Am Soc Nephrol. 2009;20:1069–1077.
39. Hallan SI, Stevens P. Screening for chronic kidney disease: which strategy? J Nephrol. 2010;23:147–155.

Insulin-induced oedema is a recognised but uncommon complication of insulin therapy especially in the paediatric age group (16 years and below).¹ At present, the pathogenesis is poorly understood and its epidemiology is unknown. It affects mainly patients with newly diagnosed type 1 diabetes mellitus or patients with poorly controlled diabetes mellitus after starting intensive insulin therapy or grossly underweight patients on large doses of insulin.^2,3 It has also been reported in patients with diabetic ketoacidosis.^4,5 Its clinical presentation is quiet variable ranging from mild peripheral oedema to cardiac failure and massive serosal effusions.⁶ DeFronzo et al. ^7–9 have attempted to clarify its underlying pathophysiological mechanism by demonstrating insulin-induced anti-natriuresis in man. Insulin-induced oedema is usually self-limiting but may occasionally progress to overt cardiac failure and development of pleural effusion.² Sometimes, too, the oedema may be gradually replaced by fat tissue with persistent weight gain.¹⁰

A survey of available standard textbooks of Paediatrics^1,1 as well as Paediatric Endocrinology¹² revealed that there is no reference to insulin in a discussion of differential diagnosis of oedema. In addition, oedema is not mentioned in the list of complications of insulin therapy. The lack of awareness among clinicians concerning insulin therapy in the differential diagnosis of oedema was further confirmed during an Update Course in Paediatrics for Resident Doctors from various training centres in Nigeria who were shown a case of insulin-induced oedema in an adolescent girl being treated for diabeticketoacidotic coma for discussion. None of the 30 candidates mentioned insulin-induced oedema in their discussion of the differential diagnosis of the oedema in this patient.

The purpose of this report is to review the scanty existing medical literature on the subject of insulin-induced oedema and raise awareness of clinicians about its continued occurrence.

CASE REPORT

The patient is a 15-year-old adolescent Nigerian girl who was diagnosed of type 1 diabetes mellitus 4 years ago. Compliance with therapy has been poor because of lack of support from the father and some misconception on the part of the patient and her mother, resulting in a poor glycaemic control. Before she defaulted from follow-up, she was on premixed combination of human insulin 70% intermediate-acting insulin plus 30% short-acting (regular) insulin given twice daily (20 units in the morning and 10 units in the evening).

The patient presented with abdominal pain and fever of 1 day duration, difficulty in breathing of 8 hours duration and loss of consciousness of 5 hours duration. No history of jaundice, chest pain, palpitation, cyanosis or urinary symptoms was found. She has not achieved menarche.

Physical examination revealed an acutely ill-looking adolescent girl who was pyrexial (temperature 39.6°C) and dehydrated. No facial puffiness or oedema of the limb was found. She was underweight (weighed 29 kg) with a sexual maturity rating of Tanner stage 1. She was unconscious (Glassgow coma scale of 3/15) with global hypotonia and hyporeflexia. Abdominal reflexes were absent. No evidence of diabetic retinopathy. Her pulse rate was 148/minute, regular and good volume. Blood pressure was 100/60 mmHg. Heart sounds were normal. No cardiac murmur was found. She was tachypnoeic (RR 34 cycles/minute) and dyspnoeic with acidotic breathing. Lung fields were clear on auscultation. Laboratory investigations showed blood glucose 18.2 mmol/L (327 mg/dl), ketonuria 3+, and acidosis (bicarbonate 10 mmol/L), confirming diabetic ketoacidosis. Serum electrolytes values were normal except for the bicarbonate. Serum creatinine level was 0.7 mg/dl. Urinalysis did not reveal any significant proteinuria or bilirubinuria. Urine glucose was 2+ and pH was 5. Full blood count, peripheral blood film and chest radiograph were essentially normal. Blood culture yielded no growth. Dehydration was corrected with normal saline (0.9% sodium chloride). She was given 0.9% sodium chloride 20 ml/kg (bolus) over the first 1 hour, then (85 ml/kg + 100% maintenance) minus fluid bolus over the remaining 23 hours. She was treated with insulin infusion at a rate of 0.1 unit/kg/h, and the standard protocol for management of diabetic ketoacidosis was followed.^11,12 Insulin infusion was discontinued on the second day of admission and replaced with subcutaneous insulin(soluble) at 1 unit/kg/day 6 hourly. Antibiotics I.V. Ceftriaxone and I.V. Gentamycin were administered. She regained consciousness between the fifth and sixth hour of treatment and the fever subsided on the fourth day of admission. On the fifth day on admission, she developed facial swelling that was followed 2 days later by pitting ankle and pretibial oedema. The weight had increased from 29 kg on admission to 34 kg. A repeat serum creatinine level was normal and a repeat urinalysis did not reveal proteinuria or bilirubinuria. Repeat urea and electrolytes revealed normal values, including bicarbonate level. Urine output remained adequate. Reassurance was given to the patient and her mother. The oedema resolved over the next 7 days without any specific medical intervention such as administration of diuretics. She discharged from the ward after spending 14 days on admission and is currently being followed up in Paediatric Endocrine metabolic clinic of the hospital.

DISCUSSION

The diagnosis of insulin-induced oedema in this patient was based on: (1) the exclusion of all other major causes of oedema; (2) its temporal relationship to improved glycaemic control and resolution of diabetes ketoacidosis; and (3) its benign nature as evidenced by its spontaneous resolution. For instance, in this patient, there was no clinical or laboratory evidence of renal, hepatic or cardiovascular disease. Oedema appeared following improved glycaemic control with insulin infusion, and the odema resolved spontaneously without any specific medical intervention, indicating its transient/benign nature.

Some of the epidemiological observations in the present case were in tandem with those of previous reports. For instance, she is an underweight adolescent girl with type 1 diabetes mellitus with ketoacidosis.^1,4,5 The natural outcome of insulin-induced oedema was observed in our patient because no specific medical intervention was instituted before the resolution of the oedema. This observation is in keeping with several other previous reports.^6,10,13,14 In contrast, there are also reports of resolution of oedema following specific therapies such as administration of loop diuretics¹ or ephedrine.⁴ However, the clinicians did not observe spontaneous resolution. The clinical implication is that in cases where the oedema fails to resolve spontaneously, loop diuretics or ephedrine may be administered. Similarly, Suzuki et al.¹⁵ reported four cases with diabetes mellitus due to the 3243 mitochondrial tRNA mutation who developed oedema of the lower extremities following glycaemic control, and their oedema responded to the administration of Co-enzyme Q 10. In the present case, facial puffiness was noticed on the fifth day of admission. Two days later, pitting ankle and pretibial oedema developed. The oedema resolved spontaneously 7 days later. A review of the literature revealed that duration of the insulin therapy before onset of oedema as well as duration of oedema before spontaneous resolution varied widely. For instance, the duration of insulin therapy before onset of oedema ranged from 1 to 14 days,^1,6,14 whereas the duration of oedema before resolution ranged from 4 to 10 days.^1,14 In the present case, both the duration (5 days) of insulin therapy before onset of oedema and duration (7 days) of oedema before resolution were keeping with previous reports.^1,6,14

Although insulin-induced oedema has been recognised for a long time as an uncommon complication of insulin therapy, its pathogenesis remains poorly understood, even today. Some authors have suggested that it may involve vasomotor changes induced by the rapid glycaemic control or that a rapid improvement of glycaemic control might have induced hepatic re-oxygenation, resulting in the production of reactive oxygen species in the liver that ultimately contribute to cell damage.¹⁵ Other authors think that insulin induces oedema through its effect on vascular permeability and renal tubules.¹⁴

In conclusion, there is a need to raise the awareness of clinicians with regard to the existence of insulin-induced oedema, especially with the current trend towards intensive insulin therapy. Clinicians should document its occurrence and include this entity in the differential diagnosis of oedema in children and adolescents with type 1 diabetes mellitus.

Disclosure: The authors declare no conflict of interest.

REFERENCES

1. Mamoulakis D, Bitsori M, Galanakis E, Raissaki M, Kalmanti M. Insuininduced oedema in children and adolescents. J Paediatr Child Health. 2006;42(10):655–657.
2. Cheliah A, Burge MR. Insulin edema in the twenty-first century: review of existing literature. J Investig Med. 2004;52(2):104–108..
3. Juliusson PB, Bjerknes R, Sovik O, Kvistad PH. Generalized edema following insulin treatment of newly diagnosed diabetes mellitus. Tidsskr Nor Laegeforen. 2001;121(8):919–920. Abstract. [Article in Norwegian].
4. Hopkins DF, Cotton SJ, Williams G. Effective treatment of insulininduced edema using ephedrine. Diabetes Care. 1993;16(7):1026–1028.
5. Khalangot ND, Koka MA, Latypova GA, Bakhtiiarova AA. Insulin edema in patients with diabetes mellitus and recent diabetic ketoacidosis (epidemiology and case reports). Lik Sprava. 2004;8:39–43.
6. Lee P, Kinsella J, Borkman M, Carter J. Bilateral pleural effusion, ascites, and facial and peripheral oedema in a 19-year-old woman two weeks following commencement of insulin lispro and determir—an unusual presentation of insulin oedema. Diabetic Med. 2007;24(11):1282–1285.
7. DeFronzo RA, Cooke RC, Andres R, Faloona GR, Davis PJ. The effect of insulin on renal handling of sodium, potassium, calcium, and phosphate in man. J Clin Investig. 1975;55:845–846.
8. DeFronzo RA. The effect of insulin on renal sodium metabolism: a review with clinical implications. Diabetologia. 1981;21:165–171.
9. DeFronzo RA, Goldberg M, Agus ZS. The effect of glucose and insulin on renal electrolyte transport. J Clin Investig. 1976;58:83–90.
10. Hirshberg B, Muszkat M, Marom T, Shalit M. Natural course of insulin oedema. J Endocrinol Investig. 2000;23(3):187–188.
11. Kliegman RM, Behrman RE, Jenson HB, Stanton BF, eds. Nelson Textbook of Pediatrics. 18th ed. Philadelphia, PA: Saunder Elsevier; 2007:1–3147.
12. Sperling MA. Diabetes mellitus. In: Kaplan SA, ed. Clinical Pediatric Endocrinology. 2nd ed. Philadelphia, PA: WB Saunders Company; 1990: 127–164.
13. Cosmi F, Bianco C, Mollaili M, Aimi M, Corbacelli C, Ricca M. Edematogenic syndrome in diabetics treated with insulin: a little-known complication. Minerva Med. 1986;77(5–6):171–174.
14. Ran TS, Nelli P, Rai P, Kundaje GN. Insulin oedema: an uncommon complication of insulin therapy. J Assoc Physicians India. 1991;39(5): 411–412.
15. Suzuki Y, Kadowaki H, Taniyama M, et al. Insulin oedema in diabetes mellitus associated with 3243 mitochondrial tRNALeu(UUR) mutation: case reports. Diabetes Res Clin Pract. 1995;29(2):137–142.

Moderate caloric restriction elicits a myriad of metabolic alterations that positively influence the prognosis of obesity and type 2 diabetes. A common factor in these aforementioned conditions is a progressive reduction in insulin sensitivity, defined in this review as the extent to which a specified quantity of insulin is able to affect metabolic and signaling events. Although often considered to be a negative outcome, insulin resistance may alternatively be viewed as an adaptation to an environmental stressor—an overabundance of nutrients in this case. Relief from said stress via a modest reduction in energy intake (while providing sufficient macro- and micronutrients for normal cellular function) increases insulin sensitivity in a rapid and sustainable manner. ^1–4 Deciphering the underlying mechanisms that drive this phenomenon has powerful implications for the development of therapies to counter obesity and type 2 diabetes. This review focuses on changes in the molecular mediators of insulin signaling in the context of obesity and caloric restriction, with an emphasis on the kinases that translate stress and nutrient signals into cellular, and ultimately, organism-wide events. A brief overview of insulin action, signaling, and the role of serine phosphorylation of insulin receptor substrate (IRS) proteins in insulin resistance will be followed by a detailed review of current research linking increased nutrient- and stress-related serine kinase activity with insulin resistance, and reduced serine kinase activity with moderate caloric restriction and improved insulin sensitivity.

Insulin action

In higher organisms, insulin is normally secreted in response to an influx of nutrients. In skeletal muscle, insulin stimulates glucose uptake and glycogen synthesis in addition to inducing a metabolic shift from oxidation of fatty acids to glucose (reviewed in⁵). In adipose tissue, insulin promotes lipid synthesis and uptake and inhibits lipolysis.⁶ Finally, in liver, insulin suppresses glucose and lipoprotein production and output^{7, 8} while stimulating glycogen synthesis.⁹ Therefore, the principal role of the hormone is to properly partition exogenous nutrients among the three primary insulin-responsive peripheral tissues by inducing nutrient uptake/storage while inhibiting endogenous nutrient release/production. These effects are acutely mediated through cellular signaling pathways and, in the long term, are subject to genomic regulation.

Insulin signaling

The key mediators of insulin signaling—the IRS proteins— reside upstream in the pathway, proximal to the insulin receptor. When insulin binds to its receptor, the tyrosine kinase activity of the receptor phosphorylates (activates) the IRS proteins on tyrosine residues, which in turn amplifies and transduces the insulin signal downstream. Conversely, serine phosphorylation of IRS proteins by a number of serine kinases generally antagonizes this action, attenuating propagation of the insulin signal.¹⁰ The critical role of the IRS proteins with respect to insulin function has been established in a number of in vitro and in vivo models utilizing knockout ^11–14 as well as siRNA-mediated knockdown;^{15, 16} a detailed review discussing the significance of the IRS proteins, as well as their structural and regulatory aspects, can be found in.¹⁰ Although there are several IRS isoforms, in this review, we focus on the ramifications of serine phosphorylation of IRS-1, as its regulation, phosphorylation sites, and role in insulin sensitivity are most well characterized. Given the central position of the IRS proteins in insulin action, it is not surprising that their regulation is governed by upstream, downstream, and cross-pathway feedback. Two signaling nodes of particular interest are the stress (MAPK) and nutrient-sensitive (mTOR/p70S6K) kinases, due to their potent inhibitory effects on insulin signaling,^3,17–20 their strong positive association with chronic energy excess, ^21–24 and an abundance of data showing that their inhibition or ablation can reverse or prevent insulin resistance.^{20, 23, 25, 26}

SERINE PHOSPHORYLATION OF IRS PROTEINS AND INSULIN RESISTANCE

Transgenic animal models, siRNA and adenovirus-mediated knockdown studies, and the use of specific kinase inhibitors provide substantial evidence to support a mechanism in which activation of serine kinases results in insulin resistance through IRS serine phosphorylation. ^{12–16 27} The exact mechanism(s) by which serine phosphorylation of IRS proteins leads to insulin resistance is/are not completely understood. A wealth of data indicates that IRS becomes a poor substrate for the insulin receptor upon serine phosphorylation; other potential mechanisms include serine phosphorylation-induced degradation of IRS protein via the ubiquitin–proteasome pathway and translocation of IRS proteins from the plasma membrane to the cytosol (reviewed in¹⁰).

At the molecular level, hyperinsulinemia,^{28, 29}, cytokines,^{30, 31}, growth factors,^{32, 33} oxidative stress,^{34, 35} free fatty acids,^{36, 37} hyperaminoacidemia,^{38, 39} and osmotic stress⁴⁰ have all been found to induce insulin resistance and enhance serine phosphorylation of IRS proteins. Physiologically, high-fat diets/overfeeding,^{23, 41} hypertension^{32, 42} and abnormal stressors (e.g., infection,⁴³ trauma,⁴⁴ and burn injuries⁴⁵) result in insulin resistance and IRS serine phosphorylation as well. These effectors signal through diverse upstream pathways, but share in common the activation of stress- and/or nutrient-sensitive kinases, which in turn directly and/or indirectly increase serine phosphorylation of the IRS proteins.

In the context of energy flux and insulin action, the serine kinases most potently induced by chronic nutrient excess— the MAPKs and mTOR/p70S6K—act as key determinants of insulin sensitivity. The ability of these kinases to serine phosphorylate IRS-1 is well documented; within the MAPKs, Erk-1/2 phosphorylate mouse and rat IRS-1 at serine 612/632 (S616/636 in humans), whereas JNK acts at serine 302/307 in mice and rats (serine 307/312 in humans). mTOR/p70S6K have been shown to target IRS-1 at serine 302 and S632/635 in mice and rats (serine 636/639 in humans) (reviewed in¹⁰). The regulation of these kinases and the significance of their effects on IRS-1 and insulin signaling will be discussed in detail in the following sections.

STRESS SIGNALING AND INSULIN RESISTANCE

The mitogen-activated protein kinases (MAPKs) are a family of serine/threonine kinases that mediate cellular response to extracellular stress (i.e., oxidative stress, UV radiation, inflammatory hormones, etc.), growth factors (i.e., insulin, insulin-like growth factors (IGFs), platelet-derived growth factor (PDGF), epidermal growth factor (EGF), etc.), and longchain saturated fatty acids (i.e., palmitate and stearate). There are three main classes of MAPKs, all of which are phosphorylated by upstream MAPK kinases: the extracellular-regulated kinases-1 and -2, the p38 MAPKs (a, b, d, and c), and the c-Jun amino-terminal kinases-1, -2, and -3 (also known as stressactivated protein kinases, SAPK). Erk-1/2, activated upstream by MEK1, seem to be the primary mediators of growth factor signaling. JNK 1–3, whose upstream kinase is MKK7, appear to mediate general stress response. The p38 MAPK family, directly downstream of MKK3/6, generally governs differentiation and development (for an excellent review, see⁴⁶). MAPK signaling may be initiated by a variety of cytoplasmic receptors, including receptor tyrosine kinases and G protein-coupled receptors; specificity is mediated in part by different scaffolding proteins which couple upstream receptor activation to MAPKKKs and MAPKKs. Furthermore, treatments that upregulate PKC, including palmitate and the diacylglycerol analog PMA or TPA, induce Erk-1/2 (and JNK) activity and lead to serine phosphorylation of IRS-1 at the known Erk-1/2 site (S612). ^47–49

We focus on Erk-1/2 in this review because of our observation that caloric restriction results in drastic changes in the activity of Erk-1/2, but not JNK or p38 MAPK.³ The role of JNK in insulin resistance, particularly that associated with high-fat feeding, is well established, and the reader is directed to a recent review.⁵⁰ Furthermore, in the majority of studies, Erk-1/2 and JNK activation occurs in parallel, with only a few studies (in which both kinases were investigated) showing activation of one kinase but not the other.^{3, 51} The p38 MAPK appears to have a role in insulin resistance as well,^{28, 34, 52, 53} but the lack of a well-defined phenotype in the knockout models of p38 precludes its discussion in this review.

Erk-1/2 and insulin resistance

Erk-1/2 has been found to be upregulated in several in vivo models of insulin resistance related to nutrient excess. The hyperglycemic, hyperinsulinemic ob/ob mouse has severalfold higher basal hepatic Erk-1/2 phosphorylation.⁵¹ Aging, which has been shown to decrease the efficiency of nutrient partitioning,⁵⁴ is also associated with enhanced Erk-1/2 phosphorylation, which is increased in older, insulin-resistant mice compared with younger, more sensitive controls.⁵⁵ Myotubes isolated from type 2 diabetic humans have reduced insulin-stimulated PI3K activity and significantly higher basal Erk-1/2 phosphorylation,⁵⁶ and myotubes from subjects with polycystic ovary syndrome (PCOS; an endocrine disorder featuring hyperandrogenism and insulin resistance, among other abnormalities) have increased MEK1 (the upstream kinase for Erk-1/2) and Erk-1/2 phosphorylation;⁵⁷ the same pattern can be observed in adipocytes isolated from type 2 diabetics.²⁸ The fact that Erk-1/2 activity is strongly enhanced in these diverse insulin-resistant models suggests that its role may be (at least partially) causative, rather than simply correlative.

In vitro studies of insulin resistance support the in vivo observations with respect to Erk-1/2. Treatment of many different cell types with growth factors, cytokines, and free fatty acids known to antagonize insulin resistance, including angiotensin II (AT II),^{32, 58} tumor necrosis factor (TNF)a,²⁶ C-reactive protein,³¹ prostaglandin E2,⁵⁹ palmitate,^{49, 60} and (chronic) insulin enhances Erk-1/2 phosphorylation. In the majority of these studies, the serine phosphorylation of IRS-1 at the known Erk target site (S616/636 in humans and S612/632 in mice and rats) is enhanced as well. The mechanism underlying AT II-, cytokine-, and prostaglandin-induced resistance appears to be activation of cognate receptors with the resulting signaling cascade leading to Erk-1/2 (or JNK) activation; receptor transactivation may also play a role.³² Palmitate, however, appears to induce stress signaling via PKC activation secondary to MLK and MKK upregulation.⁴⁸ Additional studies support the ability of fatty acids to activate Erk-1/2; Erk-2 immunoprecipitated from PMA (a diacylglycerol analog commonly used to activate PKC)-treated cells is able to serine phosphorylate an IRS-1 peptide containing the S612 site.⁴⁷ A subsequent study using an IRS-1 S612A mutant (unphosphorylatable) confirms that activation of PKC is sufficient to induce phosphorylation of IRS-1 S612.⁴⁹ These findings show that nutrient and stress signals are potent inducers of insulin resistance in a controlled in vitro environment, enhance Erk-1/2 activity in parallel with their ability to enhance serine phosphorylation of IRS-1, and antagonize insulin signaling; furthermore, the ability of Erk to directly phosphorylate an IRS-1 fragment at a key serine site argues in favor of a mechanistic role for this kinase.

Erk-1/2 and insulin sensitivity

Just as Erk-1/2 activity positively associates with obesity, diabetes, and stress factor-induced insulin resistance, a decrease in Erk-1/2 phosphorylation strongly correlates with insulin sensitivity. Recently, it was discovered that Zucker fatty rats placed on a calorie-restricted diet have dramatically reduced hepatic Erk-1/2 activity compared with ad libitum-fed controls, with a concomitant reduction in serine phosphorylation of IRS-1 at S612 and S632, known Erk-1/2 target sites in rats.³ Likewise, caloric restriction is able to bring Erk-1/2 activity in aged mice down to the levels seen in young controls.⁵⁵ When myotubes isolated from type 2 diabetics are treated with PD98059, an inhibitor of MEK1, Erk-1/2 phosphorylation is almost completely suppressed, and IRS-1 S636 phosphorylation significantly reduced;⁵⁶ a MEK1 inhibitor was shown to have similar effects on myotubes from PCOS subjects, restoring the association of PI3K with IRS-1.⁵⁷ Finally, the Erk-1 knockout mouse, which is resistant to diet-induced obesity and insulin resistance, lends support for a mechanistic and not a merely associative role for Erk in mediating insulin sensitivity in vivo.²⁰

In vitro, insulin-stimulated Akt phosphorylation in endothelial cells made resistant with angiotensin II is restored by PD98059, and the ATII-induced IRS-1 S616 phosphorylation is prevented.^{32, 58} Twenty-four-hour TNFa treatment causes a marked decrease in insulin receptor and IRS protein in 3T3-L1 adipocytes, which can be reversed by coincubation with PD98059.²⁶ In L6 myotubes, the C-reactive protein-incited reduction in insulin-stimulated Akt and GKS3b phosphorylation, as well as glycogen synthesis, is reversed by a MEK1 inhibitor; in line with this, IRS-1 S612 phosphorylation is reduced by more than 80%.³¹ In the HepG2 cell line, prostaglandin E2 suppression of insulin-stimulated Akt phosphorylation and serine phosphorylation of S636 can be abrogated by inhibition of MEK1.⁵⁹ Palmitate activates both MEK1 and JNK, and this can be reversed by PD98059.⁶⁰ PMA treatment of 293 cells induces Erk-2 activity and IRS-1 S612 phosphorylation, which is also abrogated by a MEK1 inhibitor.⁴⁷ Hyperinsulinemia in 3T3-L1 cells augments IRS-1 S612 phosphorylation, which again can be prevented almost entirely by PD98059.²⁹ Finally, when serine 612 is mutated to alanine in an IRS-1 fragment, the resulting peptide can no longer be phosphorylated (at residue 612); however, the mutant IRS-1 is capable of being tyrosine phosphorylated by the insulin receptor and signals normally to PI3K, even in the presence of PMA.⁴⁹ Thus, experiments in cultured cells are consistent with in vivo findings: that is, suppression of Erk-1/2, whether by pharmacological or molecular means, reverses or prevents IRS-1 serine phosphorylation and insulin resistance.

In summary, obesity and type 2 diabetes, two conditions characterized by chronic nutrient excess, lead to elevated serum levels of cytokines, growth factors, free fatty acids, glucose, insulin, and hypertension-associated hormones. These factors have been shown to induce insulin resistance, with concomitant elevation in Erk-1/2 activity and, when examined, serine phosphorylation of IRS-1. Ablation of Erk-1/2 through wholebody knockout, siRNA, or inhibitors is able to prevent and/or reverse these deleterious effects on insulin action and signaling. Furthermore, treatments that prevent or reverse insulin resistance have the same abrogating effect on the known Erk-1/2 serine phosphorylation sites in IRS-1. Finally, site-directed mutation of a primary Erk-1/2 IRS-1 phosphorylation site permits normal downstream insulin signaling in the presence of a resistance-inducing diacylglycerol analog. The sum of these findings suggests that factors that activate Erk-1/2 induce insulin resistance through site-specific serine phosphorylation of IRS proteins, and imply that inhibition of said phosphorylation events might at least partially abrogate insulin resistance secondary to Erk-1/2 activation.

NUTRIENT SIGNALING AND INSULIN RESISTANCE

There are two principal intracellular ‘‘sensors’’ of energy status: the mammalian target of rapamycin, or mTOR, which exists in two complexes, mTORC-1 and mTORC-2, and the AMP-activated protein kinase (AMPK). The role of AMPK in energy sensing and insulin resistance is well established, but a comprehensive evaluation of its structure and regulation is beyond the scope of this review. Thus, AMPK will be discussed only within the context of its role in mTOR regulation; interested readers are directed to recent reviews.^{61, 62}

mTOR governs protein synthesis, cell growth, cytoskeleton organization, and cell survival and proliferation (see^{63, 64} for recent reviews) and has been proposed to be a nutrient sensor.⁶⁴ mTOR is activated by a high ratio of ATP:AMP, amino acids (in particular the branched-chain amino acids, leucine, valine, and isoleucine), and growth factors, including insulin.⁶⁴ A decrease in the cellular energy state inhibits mTOR, an effect that is almost certainly mediated by an increase in AMP leading to activation of AMPK, rather than a decrease in ATP leading to inactivation of mTOR. This results from the intracellular ratio of ATP:AMP, which is on the order of 100:1, in conjunction with the adenylate kinase reaction, which interconverts equimolar quantities of AMP and ATP to ADP and vice versa; this implies that changes in AMP are of greater relative significance than changes in ATP.⁶⁵ Thus, when cellular AMP levels rise, AMPK becomes activated and phosphorylates the TSC-1/2 complex as well as the mTORC-1 scaffolding protein Raptor, the net result of which is inhibition of the mTORC-1 complex.⁶⁶

mTOR is the catalytic component of two multiprotein complexes, mTORC-1 and mTORC-2. The mTORC-1 complex is composed of four main components: mTOR, the catalytic subunit, Raptor, a scaffolding protein, mLST8 (also known as GbL), a binding protein, and PRAS40, a regulatory protein.⁶⁶ The mTORC-2 complex also contains mTOR and mLST8, with Rictor as a scaffolding protein and mSIN1 as a regulatory protein.^{66, 67} There are several key differences between the two complexes, in terms of both regulation and downstream targets. mTORC-1 is rapamycin sensitive and is linked to S6K1 and 4EBP-1 activation, whereas mTORC-2 is rapamycin insensitive and modulates cytoskeleton organization as well as acting as the putative PDK-2 activity leading to phosphorylation of Akt on serine 473.⁶⁸ The focus of this review will be on the mTORC-1 complex, on account of its ability to phosphorylate and activate p70S6K, a downstream mediator of protein synthesis that plays a crucial role in the attenuation of insulin signaling through serine phosphorylation of IRS-1. ^17–19

mTOR/p70S6K and insulin resistance

Hypercaloric, high-fat diets are commonly used to induce insulin resistance in mice, and also lead to several-fold higher p70S6K activation and IRS-1 phosphorylation at S632/635, an mTOR target site.^{21,, 23} High-fat diets would be expected to augment not only plasma glucose (and insulin) and fatty acid concentrations, but also amino acid levels, which when elevated, have been shown to induce insulin resistance in vitro and in vivo. High-fat diet-induced hyperaminoacidemia is mediated either by increased net food intake or by the sparing effect fatty acids and/or carbohydrates have on amino acid catabolism.^{69, 70} Recently, it was reported that high-fat/ high-branched-chain amino acid feeding induces a greater degree of resistance than high-fat feeding alone.⁷¹ This regimen elevates mTOR, p70S6K, and IRS-1 S302 phosphorylation (another known mTOR/p70S6K site in mice¹⁷). In humans, amino acid infusion elevates p70S6K and IRS-1 S612 and S636 phosphorylation.⁷² In further support of diet-induced activation of the mTOR/p70S6K signaling pathway and insulin resistance, p70S6K and IRS-1 serine 632/635 (the mouse equivalent of human S636/639) phosphorylation is significantly augmented in hyperphagic mouse models of obesity²³ and type 2 diabetes,^{23, 73} even when the animals are placed on a standard chow diet. A direct role for hyperlipidemia with respect to mTOR/p70S6K-mediated resistance is less well defined, with most studies focusing rather on the role of mTOR on lipid regulatory genes. These studies emphasize the close association between overfeeding and mTOR/p70S6K activation leading to serine phosphorylation of IRS-1 at the mTOR/p70S6K target sites and insulin resistance.

In vitro, treatments that induce insulin resistance, including angiotensin II³² and hyperinsulinemia,^{32, 74} particularly hyperinsulinemia in the presence of elevated leucine,⁷¹ have effects that parallel those seen in vivo. Glucose also plays a role in mTOR/p70S6K-induced insulin resistance, possibly through suppression of AMPK, as treatment of L6 myotubes with high glucose augments IRS-1 S636/639 phosphorylation and tempers insulin-stimulated Akt S308 phosphorylation in culture.¹⁷ The effects of elevated ATP:AMP ratios and mTOR/p70S6K activation leading to insulin resistance are difficult to evaluate independently of the numerous effects that ATP-depleting treatments have on intracellular metabolism; however, studies in which TSC-1/2 (negative regulators of mTORC and targets of AMPK) have been knocked out or knocked down^{18, 75} show a significant increase in mTOR/p70S6K activity, a decrease in insulin signaling efficiency, and an elevation in IRS-1 S302 and S636/639 phosphorylation. Likewise, when Rheb (a positive effector of mTOR) is constitutively overexpressed, IRS-1 S636/639 is enhanced and insulin signaling to Akt dampened.¹⁷ The in vitro findings point towards a primary role for mTOR/p70S6K in inducing insulin resistance; whether mTOR/p70S6K activity is augmented by treatment with individual nutrients or through suppression of inhibitory mTORC regulators, the resulting effect is insulin resistance coincident with site-specific IRS-1 serine phosphorylation.

mTOR/p70S6K and insulin sensitivity

In the same vein as Erk-1/2, the inhibition of p70S6K not only correlates with improved insulin sensitivity, but can act as the primary mediator of this effect. The p70S6K knockout mouse provides compelling support for this claim; the lack of p70S6K prevents weight and fat gain in the context of a highfat diet, preserves insulin action, maintains insulin signaling, and ablates serine 632/635 phosphorylation of IRS-1.²³ In mice fed a high-fat/high-branched-chain amino acid diet, rapamycin restores glucose clearance to the level of chow diet controls.⁷¹ Oral rapamycin, an immunosuppressant and chemotherapeutic commonly used to inhibit mTOR, successfully prevents IRS-1 serine phosphorylation in the skeletal muscle of humans infused with amino acids, and modestly increases the glucose infusion rate needed to clamp plasma glucose concentrations.⁷² Finally, in insulin-resistant Zucker fatty rats, hepatic IRS-1 S632/635 phosphorylation is among one of only two sites to show significant reduction following caloric restriction (the other being the Erk-1/2 site, S612).³ These data show that inhibition or ablation of mTOR/p70S6K signaling is sufficient to prevent or reverse insulin resistance in vivo.

The in vivo findings concerning mTOR/p70S6K and insulin sensitivity are largely paralleled by in vitro studies. When p70S6K is knocked down in HeLa cells, insulin-stimulated Akt phosphorylation increases concomitant with a significant reduction in IRS-1 S636/639 (the human equivalent of murine 632/635) phosphorylation,²³ providing direct evidence for the ability of mTOR/p70S6K signaling to influence insulin signaling even in a basal state. Additional support may be found in studies utilizing rapamycin: L6 myotubes made resistant by incubation with elevated concentrations of amino acids demonstrate normal insulin-stimulated glucose uptake and IRS-1-associated PI3K activity when coincubated with rapamycin.³⁹ The hyperinsulinemia-induced increase in total IRS-1 serine phosphorylation is abrogated by rapamycin in L6 myotubes¹⁷ and rat liver;¹⁹ in both instances, insulin-stimulated glucose disposal is restored as well. In myotubes coincubated with high glucose or high glucose plus leucine, rapamycin reverses IRS-1 S632/635 phosphorylation and restores insulin stimulation of Akt;¹⁷ similar effects may be observed in white adipose tissue²⁹ and 3T3-L1 adipocytes⁷⁴ exposed to hyperinsulinemia. Rapamycin significantly reduces or completely prevents IRS-1 serine 632/635 phosphorylation and restores insulin-stimulated Akt when TSC-1 and/or TSC-2 are deleted,^{18, 75} or when the mTOR agonist, Rheb, is constitutively expressed.¹⁷ Interestingly, rapamycin is as effective as a MEK1 inhibitor in preventing angiotensin II-induced insulin resistance in epithelial cells,³² and can also prevent serine phosphorylation of IRS-1 at residue 307, a classic JNK target, in TNFatreated hepatoma cells⁷⁶ and insulin-treated 3T3-L1 adipocytes.²⁹

Although the majority of studies use rapamycin to abrogate the impact of mTOR/p70S6K on insulin signaling, a recent study provides direct evidence that increasing AMPK activity may have similarly beneficial effects. Overexpression of a constitutively active AMPK abrogates glucose-induced IRS-1 S636/639 phosphorylation and inhibition of insulin-stimulated Akt S308 in HEK 293 cells.¹⁷ Similarly, overexpression of an active upstream activator of AMPK, LKB-1, reverses the effect of Rheb overexpression on IRS-1 S636/639. Finally, metformin, a pharmacological AMPK agonist and a front-line medication in the treatment of insulin resistance, is able to prevent insulinincited IRS-1 serine phosphorylation and restore insulin signaling towards Akt. An additional finding of interest is that treating 3T3-L1 adipocytes with 2-deoxyglucose, a non-metabolizable glucose analog commonly used to mimic energy deprivation (in vitro and in vivo), activates AMPK and ablates IRS-1 phosphorylation on S632/635.

In total, these data provide compelling evidence to support a mechanism by which activation of mTOR/p70S6K signaling (by general overfeeding or treatment with excessive quantities of individual nutrients), most likely via inhibition of AMPK secondary to changes in the ATP:AMP ratio, results in IRS-1 serine phosphorylation, inhibition of downstream insulin signaling, and insulin resistance. On the other hand, suppression of mTOR/p70S6K activity, whether by treatment with rapamycin or through activation of AMPK, reduces IRS-1 serine phosphorylation on the mTOR/p70S6K target sites, normalizes insulin signaling, and ameliorates resistance.

CONCLUSION

In vivo models support the central role of Erk-1/2 and mTOR/p70S6K in mediating insulin resistance: high-fat feeding strongly upregulates the activity of these kinases, whereas caloric restriction has the opposite effect; insulin sensitivity, in turn, inversely tracks the activity of Erk-1/2 and mTOR/p70S6K. Knockout animal models provide mechanistic evidence for the ability of these kinases to control insulin sensitivity, as the absence of either Erk-1/2 or p70S6K protects against high-fat diet-induced IRS-1 serine phosphorylation and insulin resistance. These data are supported by a wealth of in vitro studies demonstrating the close inverse association between Erk-1/2 and mTOR/p70S6K activity and insulin action, with targeted knockdown and site-directed mutagenesis experiments confirming the primary role of the kinases and site-specific IRS-1 serine phosphorylation in the determination of insulin signaling.

Stress- and nutrient-related signals induce insulin resistance through augmentation of the Erk-1/2 and p70S6K/mTOR signaling pathways (Fig. 1). An increase in the activity of the serine kinases, which govern these aforementioned pathways, results in serine phosphorylation of the IRS proteins at specific residues. This attenuates propagation of the insulin signal through mechanisms that are not completely defined. Evidence suggests that serine phosphorylation of IRS proteins precludes their association with the tyrosine kinase activity of the insulin receptor; as tyrosine phosphorylation of IRS is necessary for the recruitment of immediate downstream signaling molecules (i.e., PI3K), the entire signaling chain would be compromised. Serine phosphorylation additionally appears to antagonize the binding between IRS and downstream signaling partners, induces degradation of IRS, and may alter its cellular localization. IRS signaling to PI3KRAktRFoxo1/GLUT4/ GSK3B is largely responsible for insulin inhibition of hepatic glucose output, stimulation of glucose uptake/clearance, and storage of glucose as glycogen. Thus, any impairment in insulin signaling may result in systemic insulin resistance; this effect, in turn, becomes a causative agent in the worsening of resistance, due to the ability of various nutrients to antagonize insulin signaling—an effect largely secondary to upregulation of Erk-1/2 and mTOR/p70S6K activity.

Insulin signaling is a complex network of interconnected signaling nodes and, as such, it would be presumptuous to exclude the roles of other mediating kinases, phosphatases, secondary messengers, etc. in the regulation of insulin sensitivity. However, serine phosphorylation of IRS molecules emerges as a common theme in cellular insulin resistance: further understanding of the factors that determine the extent to which IRS proteins are serine phosphorylated may provide valuable insight into targeted approaches towards the treatment and prevention of insulin resistance.

Disclosure:<break>We declare no conflicts of interest.</break>

Acknowledgments: We would like to thank Dr. Matthew Brady for his constructive suggestions and insight.

REFERENCES

1. Lee CK, Klopp RG, Weindruch R, Prolla TA. Gene expression profile of aging and its retardation by caloric restriction. Science. 1999;285(5432):1390–1393.
2. Wang ZQ, Floyd ZE, Qin J, et al. Modulation of skeletal muscle insulin signaling with chronic caloric restriction in cynomolgus monkeys. Diabetes. 2009;58(7):1488–1498.
3. Zheng Y, Zhang W, Pendleton E, et al. Improved insulin sensitivity by calorie restriction is associated with reduction of ERK and p70S6K activities in the liver of obese Zucker rats. J Endocrinol. 2009;203(3): 337–347.
4. Bodkin NL, Ortmeyer HK, Hansen BC. Long-term dietary restriction in older-aged rhesus monkeys: effects on insulin resistance. J Gerontol A Biol Sci Med Sci. 1995;50(3):B142–B147.
5. Corpeleijn E, Saris WH, Blaak EE. Metabolic flexibility in the development of insulin resistance and type 2 diabetes: effects of lifestyle. Obes Rev. 2009;10(2):178–193.
6. Campbell PJ, Carlson MG, Hill JO, Nurjhan N. Regulation of free fatty acid metabolism by insulin in humans: role of lipolysis and reesterification. Am J Physiol. 1992;263(6 Pt 1):E1063–E1069.
7. Basu R, Chandramouli V, Dicke B, Landau B, Rizza R. Obesity and type 2 diabetes impair insulin-induced suppression of glycogenolysis as well as gluconeogenesis. Diabetes. 2005;54(7):1942–1948.
8. Pavlic M, Xiao C, Szeto L, Patterson BW, Lewis GF. Insulin acutely inhibits intestinal lipoprotein secretion in humans in part by suppressing plasma free fatty acids. Diabetes 2010 Mar;59(3):580–7.
9. Aiston S, Coghlan MP, Agius L. Inactivation of phosphorylase is a major component of the mechanism by which insulin stimulates hepatic glycogen synthesis. Eur J Biochem. 2003;270(13):2773–2781.
10. Sun XJ, Liu F. Phosphorylation of IRS proteins Yin-Yang regulation of insulin signaling. Vitam Horm. 2009;80:351–387.
11. Yamauchi T, Tobe K, Tamemoto H, et al. Insulin signalling and insulin actions in the muscles and livers of insulin-resistant, insulin receptor substrate 1-deficient mice. Mol Cell Biol. 1996;16(6):3074–3084.
12. Araki E, Lipes MA, Patti ME, et al. Alternative pathway of insulin signalling in mice with targeted disruption of the IRS-1 gene. Nature. 1994;372(6502):186–190.
13. Sadagurski M, Weingarten G, Rhodes CJ, White MF, Wertheimer E. Insulin receptor substrate 2 plays diverse cell-specific roles in the regulation of glucose transport. J Biol Chem. 2005;280(15):14536–14544.
14. Tamemoto H, Kadowaki T, Tobe K, et al. Insulin resistance and growth retardation in mice lacking insulin receptor substrate-1. Nature. 1994;372(6502):182–186.
15. Bouzakri K, Zachrisson A, Al-Khalili L, et al. siRNA-based gene silencing reveals specialized roles of IRS-1/Akt2 and IRS-2/Akt1 in glucose and lipid metabolism in human skeletal muscle. Cell Metab. 2006;4(1):89–96.
16. Taniguchi CM, Ueki K, Kahn R. Complementary roles of IRS-1 and IRS-2 in the hepatic regulation of metabolism. J Clin Invest. 2005;115(3): 718–727.
17. Tzatsos A, Kandror KV. Nutrients suppress phosphatidylinositol 3-kinase/Akt signaling via raptor-dependent mTOR-mediated insulin receptor substrate 1 phosphorylation. Mol Cell Biol. 2006;26(1):63–76.
18. Shah OJ, Wang Z, Hunter T. Inappropriate activation of the TSC/Rheb/ mTOR/S6K cassette induces IRS1/2 depletion, insulin resistance, and cell survival deficiencies. Curr Biol. 2004;14(18):1650–1656.
19. Ueno M, Carvalheira JB, Tambascia RC, et al. Regulation of insulin signalling by hyperinsulinaemia: role of IRS-1/2 serine phosphorylation and the mTOR/p70 S6K pathway. Diabetologia. 2005;48(3):506–518.
20. Bost F, Aouadi M, Caron L, et al. The extracellular signal-regulated kinase isoform ERK1 is specifically required for in vitro and in vivo adipogenesis. Diabetes. 2005;54(2):402–411.
21. Khamzina L, Veilleux A, Bergeron S, Marette A. Increased activation of the mammalian target of rapamycin pathway in liver and skeletal muscle of obese rats: possible involvement in obesity-linked insulin resistance. Endocrinology. 2005;146(3):1473–1481.
22. Ono H, Pocai A, Wang Y, et al. Activation of hypothalamic S6 kinase mediates diet-induced hepatic insulin resistance in rats. J Clin Invest. 2008;118(8):2959–2968.
23. Um SH, Frigerio F, Watanabe M, et al. Absence of S6K1 protects against age- and diet-induced obesity while enhancing insulin sensitivity. Nature. 2004;431(7005):200–205.
24. Wang J, Ma H, Tong C, Zhang H, Lawlis GB, Li Y, et al. Overnutrition and maternal obesity in sheep pregnancy alter the JNK-IRS-1 signaling cascades and cardiac function in the fetal heart. FASEB J 2010 Jun;24(6):2066–76.
25. Adochio R, Leitner JW, Hedlund R, Draznin B. Rescuing 3T3-L1 adipocytes from insulin resistance induced by stimulation of Aktmammalian target of rapamycin/p70 S6 kinase (S6K1) pathway and serine phosphorylation of insulin receptor substrate-1: effect of reduced expression of p85alpha subunit of phosphatidylinositol 3-kinase and S6K1 kinase. Endocrinology. 2009;150(3):1165–1173.
26. Fujishiro M, Gotoh Y, Katagiri H, et al. Three mitogen-activated protein kinases inhibit insulin signaling by different mechanisms in 3T3-L1 adipocytes. Mol Endocrinol. 2003;17(3):487–497.
27. Kubota N, Kubota T, Itoh S, et al. Dynamic functional relay between insulin receptor substrate 1 and 2 in hepatic insulin signaling during fasting and feeding. Cell Metab. 2008;8(1):49–64.
28. Carlson CJ, Koterski S, Sciotti RJ, Poccard GB, Rondinone CM. Enhanced basal activation of mitogen-activated protein kinases in adipocytes from type 2 diabetes: potential role of p38 in the downregulation of GLUT4 expression. Diabetes. 2003;52(3):634–641.
29. Gual P, Gremeaux T, Gonzalez T, Le Marchand-Brustel Y, Tanti JF. MAP kinases and mTOR mediate insulin-induced phosphorylation of insulin receptor substrate-1 on serine residues 307, 612 and 632. Diabetologia. 2003;46(11):1532–1542.
30. Plomgaard P, Bouzakri K, Krogh-Madsen R, Mittendorfer B, Zierath JR, Pedersen BK. Tumor necrosis factor-alpha induces skeletal muscle insulin resistance in healthy human subjects via inhibition of Akt substrate 160 phosphorylation. Diabetes. 2005;54(10):2939–2945.
31. D’Alessandris C, Lauro R, Presta I, Sesti G. C-reactive protein induces phosphorylation of insulin receptor substrate-1 on Ser307 and Ser 612 in L6 myocytes, thereby impairing the insulin signalling pathway that promotes glucose transport. Diabetologia. 2007;50(4):840–849.
32. Arellano-Plancarte A, Hernandez-Aranda J, Catt KJ, Olivares-Reyes JA. Angiotensin-induced EGF receptor transactivation inhibits insulin signaling in C9 hepatic cells. Biochem Pharmacol. 2010;79(5):733–745.
33. Rui L, Aguirre V, Kim JK, et al. Insulin/IGF-1 and TNF-alpha stimulate phosphorylation of IRS-1 at inhibitory Ser307 via distinct pathways. J Clin Invest. 2001;107(2):181–189.
34. Archuleta TL, Lemieux AM, Saengsirisuwan V, et al. Oxidant stressinduced loss of IRS-1 and IRS-2 proteins in rat skeletal muscle: role of p38 MAPK. Free Radic Biol Med. 2009;47(10):1486–1493.
35. Chen B, Wei J, Wang W, et al. Identification of signaling pathways involved in aberrant production of adipokines in adipocytes undergoing oxidative stress. Arch Med Res. 2009;40(4):241–248.
36. Jove M, Planavila A, Sanchez RM, Merlos M, Laguna JC, Vazquez-Carrera M. Palmitate induces tumor necrosis factor-alpha expression in C2C12 skeletal muscle cells by a mechanism involving protein kinase C and nuclear factor-kappaB activation. Endocrinology. 2006;147(1):552–561.
37. Yu C, Chen Y, Cline GW, et al. Mechanism by which fatty acids inhibit insulin activation of insulin receptor substrate-1 (IRS-1)-associated phosphatidylinositol 3-kinase activity in muscle. J Biol Chem. 2002;277(52):50230–50236.
38. Patti ME, Brambilla E, Luzi L, Landaker EJ, Kahn CR. Bidirectional modulation of insulin action by amino acids. J Clin Invest. 1998;101(7):1519–1529.
39. Tremblay F, Marette A. Amino acid and insulin signaling via the mTOR/ p70 S6 kinase pathway. A negative feedback mechanism leading to insulin resistance in skeletal muscle cells. J Biol Chem. 2001;276(41):38052–38060.
40. Gual P, Gonzalez T, Gremeaux T, Barres R, Le Marchand-Brustel Y, Tanti JF. Hyperosmotic stress inhibits insulin receptor substrate-1 function by distinct mechanisms in 3T3-L1 adipocytes. J Biol Chem. 2003;278(29):26550–26557.
41. Danielsson A, Fagerholm S, Ost A, et al. Short-term overeating induces insulin resistance in fat cells in lean human subjects. Mol Med. 2009;15 (7– 8):228–234.
42. Sartori M, Ceolotto G, Papparella I, et al. Effects of angiotensin II and insulin on ERK1/2 activation in fibroblasts from hypertensive patients. Am J Hypertens. 2004;17(7):604–610.
43. Agwunobi AO, Reid C, Maycock P, Little RA, Carlson GL. Insulin resistance and substrate utilization in human endotoxemia. J Clin Endocrinol Metab. 2000;85(10):3770–3778.
44. Xu J, Kim HT, Ma Y, et al. Trauma and hemorrhage-induced acute hepatic insulin resistance: dominant role of tumor necrosis factor-alpha. Endocrinology. 2008;149(5):2369–2382.
45. Li L, Messina JL. Acute insulin resistance following injury. Trends Endocrinol Metab. 2009;20(9):429–435.
46. Aouadi M, Binetruy B, Caron L, Le Marchand-Brustel Y, Bost F. Role of MAPKs in development and differentiation: lessons from knockout mice. Biochimie. 2006;88(9):1091–1098.
47. De FK, Roth RA. Modulation of insulin receptor substrate-1 tyrosine phosphorylation and function by mitogen-activated protein kinase. J Biol Chem. 1997;272(50):31400–31406.
48. Jaeschke A, Davis RJ. Metabolic stress signaling mediated by mixedlineage kinases. Mol Cell. 2007;27(3):498–508.
49. De FK, Roth RA. Protein kinase C modulation of insulin receptor substrate-1 tyrosine phosphorylation requires serine 612. Biochemistry. 1997;36(42):12939–12947.
50. Yang R, Trevillyan JM. c-Jun N-terminal kinase pathways in diabetes. Int J Biochem Cell Biol. 2008;40(12):2702–2706.
51. Gum RJ, Gaede LL, Heindel MA, et al. Antisense protein tyrosine phosphatase 1B reverses activation of p38 mitogen-activated protein kinase in liver of ob/ob mice. Mol Endocrinol. 2003;17(6):1131–1143.
52. Fujishiro M, Gotoh Y, Katagiri H, et al. MKK6/3 and p38 MAPK pathway activation is not necessary for insulin-induced glucose uptake but regulates glucose transporter expression. J Biol Chem. 2001;276(23): 19800–19806.
53. Liu Z, Cao W. p38 mitogen-activated protein kinase: a critical node linking insulin resistance and cardiovascular diseases in type 2 diabetes mellitus. Endocr Metab Immune Disord Drug Targets. 2009;9(1):38–46.
54. Park S, Komatsu T, Hayashi H, et al. Calorie restriction initiated at middle age improved glucose tolerance without affecting age-related impairments of insulin signaling in rat skeletal muscle. Exp Gerontol. 2006;41(9):837–845.
55. Kim HJ, Jung KJ, Yu BP, Cho CG, Chung HY. Influence of aging and calorie restriction on MAPKs activity in rat kidney. Exp Gerontol. 2002;37(8–9):1041–1053.
56. Bouzakri K, Roques M, Gual P, et al. Reduced activation of phosphatidylinositol- 3 kinase and increased serine 636 phosphorylation of insulin receptor substrate-1 in primary culture of skeletal muscle cells from patients with type 2 diabetes. Diabetes. 2003;52(6):1319–1325.
57. Corbould A, Zhao H, Mirzoeva S, Aird F, Dunaif A. Enhanced mitogenic signaling in skeletal muscle of women with polycystic ovary syndrome. Diabetes. 2006;55(3):751–759.
58. Andreozzi F, Laratta E, Sciacqua A, Perticone F, Sesti G. Angiotensin II impairs the insulin signaling pathway promoting production of nitric oxide by inducing phosphorylation of insulin receptor substrate-1 on Ser312 and Ser616 in human umbilical vein endothelial cells. Circ Res. 2004;94(9):1211–1218.
59. Henkel J, Neuschafer-Rube F, Pathe-Neuschafer-Rube A, Puschel GP. Aggravation by prostaglandin E2 of interleukin-6-dependent insulin resistance in hepatocytes. Hepatology. 2009;50(3):781–790.
60. Coll T, Jove M, Rodriguez-Calvo R, Eyre E, Palomer X, Sanchez RM, et al. Palmitate-mediated downregulation of peroxisome proliferator-activated receptor-gamma coactivator 1alpha in skeletal muscle cells involves MEK1/2 and nuclear factor-kappaB activation. Diabetes. 2006;55(10):2779– 2787.
61. Zhang BB, Zhou G, Li C. AMPK: an emerging drug target for diabetes and the metabolic syndrome. Cell Metab. 2009;9(5):407–416.
62. Viollet B, Athea Y, Mounier R, et al. AMPK: Lessons from transgenic and knockout animals. Front Biosci. 2009;14:19–44.
63. Rivas DA, Lessard SJ, Coffey VG. mTOR function in skeletal muscle: a focal point for overnutrition and exercise. Appl Physiol Nutr Metab. 2009;34(5):807–816.
64. Sarbassov DD, Ali SM, Sabatini DM. Growing roles for the mTOR pathway. Curr Opin Cell Biol. 2005;17(6):596–603.
65. Hardie DG. Minireview: the AMP-activated protein kinase cascade: the key sensor of cellular energy status. Endocrinology. 2003;144(12): 5179–5183.
66. Gwinn DM, Shackelford DB, Egan DF, et al. AMPK phosphorylation of raptor mediates a metabolic checkpoint. Mol Cell. 2008;30(2):214–226.
67. Sabatini DM. mTOR and cancer: insights into a complex relationship. Nature Rev Cancer. 2006;6(9):729–734.
68. Yan L, Mieulet V, Lamb RF. mTORC2 is the hydrophobic motif kinase for SGK1. Biochem J. 2008;416(3):e19–e21.
69. Krotkiewski M. Value of VLCD supplementation with medium chain triglycerides. Int J Obes Relat Metab Disord. 2001;25(9):1393–1400.
70. Hart DW, Wolf SE, Zhang XJ, et al. Efficacy of a high-carbohydrate diet in catabolic illness. Crit Care Med. 2001;29(7):1318–1324.
71. Newgard CB, An J, Bain JR, et al. A branched-chain amino acid-related metabolic signature that differentiates obese and lean humans and contributes to insulin resistance. Cell Metab. 2009;9(4):311–326.
72. Krebs M, Brunmair B, Brehm A, et al. The Mammalian target of rapamycin pathway regulates nutrient-sensitive glucose uptake in man. Diabetes. 2007;56(6):1600–1607.
73. Koketsu Y, Sakoda H, Fujishiro M, et al. Hepatic overexpression of a dominant negative form of raptor enhances Akt phosphorylation and restores insulin sensitivity in K/KAy mice. Am J Physiol Endocrinol Metab. 2008;294(4):E719–E725.
74. Berg CE, Lavan BE, Rondinone CM. Rapamycin partially prevents insulin resistance induced by chronic insulin treatment. Biochem Biophys Res Commun. 2002;293(3):1021–1027.
75. Harrington LS, Findlay GM, Gray A, et al. The TSC1–2 tumor suppressor controls insulin-PI3K signaling via regulation of IRS proteins. J Cell Biol. 2004;166(2):213–223.
76. Bae EJ, Yang YM, Kim JW, Kim SG. Identification of a novel class of dithiolethiones that prevent hepatic insulin resistance via the adenosine monophosphate-activated protein kinase-p70 ribosomal S6 kinase-1 pathway. Hepatology. 2007;46(3):730–739.

Coronary heart disease (CHD) affects millions of patients around the world. In recent years 1.6 million hospital admissions were for acute coronary syndromes (ACS) in the United States alone.¹ As more nations “Westernize” and their populations are exposed to increasing risk factors for coronary artery disease (CAD), this will be a substantial and perhaps the greatest international health concern in the next several decades.

The prevalence of CHD makes the variety of risk factors that lead to this disease process an important focus in prevention and understanding of important intervention strategies. Well-known major risk factors for CHD include cigarette smoking, hypertension (blood pressure ≥140/90 mmHg or on an antihypertensive medication), diabetes mellitus, high low-density lipoprotein (LDL), low high-density lipoprotein (HDL), family history of premature CHD (CHD in male first degree relative <55 years or in female first degree relative <65 years), and age (men ≥45 years, women ≥55 years).²

The Adult Treatment Panel III (ATP III) has classified diabetes mellitus as a CHD risk equivalent and previous studies have noted that diabetics have a two- to eightfold increased rate of future cardiovascular events than those of age and ethnically matched nondiabetics.³ Data has also revealed that nearly 75% of the deaths in diabetic patients result from CHD.⁴ There is ever increasing evidence that obesity, physical inactivity, and dyslipidemia (elevated triglyceride, small LDL, low HDL, increased blood pressure, and insulin resistance) coalesce to create a metabolic syndrome conferring a multitude of risk factors for CHD. Diagnosis and management of the metabolic syndrome is a major challenge for CHD prevention and treatment. In this review we describe the pathophysiology of dyslipidemias. We discussed the clinical assessment and management of dyslipidemia in a previously published review in the Journal of Clinical Metabolism and Diabetes (April 2011).

REVIEW

The relationship between cholesterol and CHD has been well established.^{5, 6} The Framingham Heart Study, initiated in 1948, revealed over the course of follow-up of the study population that cholesterol was one of the major culprits in the development of CHD,⁵ and studies published over the past several decades, like REVERSAL, PROVE-IT, CARE, and the Lipid Research Clinic study, have confirmed that pharmacological intervention to lower cholesterol indeed lowered CHD morbidity and mortality. ^7–10 

Fats are essential to life as we need these substances for energy storage, energy utilization, steroid hormone synthesis, bile acid synthesis, and cell membrane production. Dietary fats are necessary components that the body will utilize to synthesize essential fats it cannot obtain from the diet. Hyperlipidemia (elevated concentrations of any or all lipid components in the plasma) and dyslipidemia (average or increased total cholesterol, increased LDL, with low HDL and/or elevated triglycerides) are terms used in clinical practice to describe clinically the disorders of lipid metabolism. Dyslipoproteinemia more accurately describes the pathophysiology surrounding abnormalities of lipid metabolism and lipoprotein transport pathways that lead to CHD. There are several major types of lipids that circulate systemically and constitute the makeup of molecules involved in dyslipoproteinemia including cholesterol and cholesterol esters, triglycerides, and phospholipids. Cholesterol is a vital component of cell membranes and also is involved in the production of steroid hormones and bile acids. Cholesterol circulates in the plasma in the form of cholesterol esters as a part of lipoproteins. Triglycerides (TG) transport fatty acids that once hydrolyzed can be used for immediate metabolic needs or stored for future energy consumption. Phospholipids also are a component of all cell membranes and participate in signal transduction pathways. These lipid types and their components are circulating in the plasma in the form of lipoproteins, which are made up of lipids and apolipoproteins all arranged in a spherical structure. The apolipoproteins provide structure to the lipoprotein, allow binding to proteins for cellular uptake, are responsible for assembly and secretion of lipoproteins, and serve as coactivators or inhibitors of enzymes (Table 1).¹¹

An understanding of lipoprotein assembly and transport is a key to understanding the pathophysiology and clinical disease state of dyslipoproteinemia. Fat as previously mentioned is essential for human life. Fat, once ingested through the diet, is exposed to secreted pancreatic lipases. Pancreatic lipases hydrolyze triglycerides to free fatty acids (FFA). Bile acids, released from the gall bladder upon dietary intake, emulsify the TG and FFA, which facilitates the formation of micelles. Micelles are taken up by the intestinal brush border and FFA are re-esterfied to form TG. Once re-esterification is complete the TG are packaged into chylomicrons and enter the portal circulation. This process occurs rapidly and once in the plasma the chylomicrons may be exposed to lipoprotein lipase (LPL). The LPL cleaves the FFA from the TG and muscles rapidly take up the FFA to use for energy. Adipose tissue stores the TG for future energy utilization, a process that is dependent on and requires insulin. The FFA released from the cleavage of TG by LPL may also bind to fatty acid binding proteins that are transported back to the liver and repackaged as very low-density lipoproteins (VLDL). The VLDL is the body's way of ensuring readily available TG during times of need. Very low-density lipoproteins are secreted by the liver and are TG rich lipoproteins. With insulin resistance, the availability of insulin to store TG in adipose tissue is diminished or ineffective, and subsequently there is an increase in FFA delivered to the liver. This increase in FFA delivery leads to an increase in VLDL. Another potential pathway for the fat from our diet is its uptake by macrophages.¹² Macrophages express receptors that bind oxidized or remnant lipoproteins, which mediate the uptake of oxidized LDL into the macrophage. There is no negative feedback mechanism for suppressing this pathway thus allowing intimal macrophages to accumulate cholesterol. The process where remnant particles are incorporated into macrophages potentiates foam cell development and deposition of fatty streaks in the intima.¹³ Endothelial cells may also uptake remnant lipoproteins through the LOX-1 receptor.

Two important lipoproteins in the transport of cholesterol between the liver and the peripheral tissues are LDL and HDL. They also have important clinical indications for the development of CHD as HDL has an inverse relationship with CHD and LDL has been linked to cardiovascular mortality.^{5, 6} Low-density lipoprotein is responsible for cholesterol transport from the portal circulation to the periphery and in the peripheral cells cholesterol uptake is regulated by the LDL receptor. The LDL is compromised of predominantly cholesterol esters. When there are elevated levels of circulating TG in the plasma these cholesterol ester rich LDL lipoproteins change composition. They become deplete in their core cholesterol esters and saturated in TG. This process leads to smaller and denser LDL particles, which are more pathogenic in CHD. The mechanism is thought to be related to increased susceptibility of the smaller LDL particles to oxidation, increased circulating time, and a higher affinity to the extracellular matrix. The LDL once taken up in the periphery by the cell is degraded in the lysosome. The cholesterol produced from LDL acts to suppress the transcription of the HMG CoA reductase gene through the sterol regulatory element-binding protein (SREBP).¹⁴ High-density lipoprotein is responsible for reverse cholesterol transport from the periphery to the liver and is an important component in controlling excess storage of cholesterol in the periphery. The main components of HDL are apolipoprotein A-I and II. The HDL lipoproteins must acquire phospholipids and cholesterol to form HDL particles and LPL and ATP binding cassette A1 gene (ABCA1) are the primary mechanism. Mature HDL is formed once the cholesterol core is esterified by lecithin-cholesterol acyltransferase (LCAT). High-density lipoprotein is primarily catabolized by the liver via scavenger receptor class BI (SR-BI). Dysfunction of HDL mediated transport is an independent risk factor for CHD as there is decreased transport back to the liver for excretion in the bile and feces.

Lipoprotein disorders constitute a disruption or alteration to the above transport or processing process. These alterations lead to the clinical disease states that one witnesses clinically in patients. Lipoprotein disorders can be classified into primary causes of dyslipoproteinemia and secondary causes. Primary causes are genetically based and usually inherited (Table 2). Secondary causes result from metabolic derangements, chronic disease states, or medications. Familial Hypercholesterolemia (FH) is an autosomal co-dominant disorder caused by a defect in the LDL receptor. This defect leads to accumulation of LDL particles in the plasma. The prevalence is 1 in 500 and is associated with a LDL that is greater than the 95th percentile for age and gender matched controls. Patients with this disorder have an increased risk of CHD by their 30–40s for men and 40–50s for women. Patients often have xanthalasmas, xanthomas on the extensor tendons, and corneal arcus. Familial Hypertriglyceridemia (FHTG) is characterized by elevated TG, elevated VLDL, decreased LDL, low HDL, and total cholesterol that is normal or increased. The etiology is overproduction of VLDL by the liver. There is not as strong as a relationship between premature CAD as with FH. The FHTG is genetically linked to first degree relatives but there is varying phenotypic expression. FHTG is potentiated by the diet and alcohol consumption. Fasting TG levels in FHTG are typically 200–500 mg/dL. Familial Hyperchylomicronemia (FHC) is a rare disorder of elevated circulating TG secondary to decreased or absent LPL activity. Deficient LPL activity leads to insufficient hydrolysis of the chylomicrons and subsequent accumulation in the plasma. Fasting TG levels are in excess of 1000 mg/dL and if serum is left to settle, a thin milky white band can be noted on the top of the collected serum. Patients afflicted with this also tend to have decreased HDL levels. Patients with FHTG often have recurrent pancreatitis, xanthomas, xerophthalmia, xerostomia, and even psychological disorders. Familial Combined Hyperlipidemia (FCH) occurs secondary to overproduction of apolipoproteins B containing lipoproteins. This is one of the most common forms of primary lipoprotein disorders affecting 1 in 50 individuals. Analysis of lipoproteins reveals elevated total cholesterol and TG. The FCH is present in 10–20% of individuals presenting with premature CAD.¹⁵ Tangier Disease and Familial High-Density Lipoprotein Deficiency are disorders characterized by deficient or nearly absent HDL. In Tangier's Disease and familial HDL deficiency, patients have severely reduced HDL secondary to reduced cholesterol exit from the cell membrane. The cause is from a defect in ATP binding cassette A1 gene, which is responsible for transporting cellular cholesterol to the cell membrane for export. The genetic absence of this gene leads to decreased lipidification of the HDL apoliproteins and thus reduced functioning HDL. Familial Dysbetalipoproteinemia (FDB) is caused by a defect in apolipoprotein E mediated clearance of remnant lipoproteins. This results in elevated serum LDL and TG that leads to enriched VLDL.¹⁶ Less than 1% of those with CHD have this disorder. The above disorders are important clinically as patients presenting with premature CAD or who have family members with premature CAD should be screened for the above disorders. The goal of identification is for early lipid screening and aggressive medical treatment.

Secondary causes of the lipoprotein disorders often are reversible or, at the least, controllable. Hormonal causes include hypothyroidism, which is often characterized by elevated LDL, TG, or both, and can be corrected with treatment of the hypothyroidism. Estrogens tend to increase TG and HDL and they may decrease LDL, but are not used clinically for treatment due to the increased risk of cardiovascular events. Renal disorders including glomerulonephritis and chronic kidney disease are associated with elevated LDL and elevated TG with decreased HDL respectively. Liver disease can lead to the abnormal formation of lipoproteins leading to dyslipoproteinemia. Sedentary lifestyle with disproportionate fat and refined sugar intake contributes a large part to elevated lipoprotein and lipid levels. Medications including thiazide diuretics, beta blockers, and corticosteroids are a few examples of medications that all increase TG levels. One of the increasingly present secondary causes are metabolic related disorders. Glycogen storage diseases and lipodystrophies fit into this category, but by far the most common are those with diabetes and metabolic syndrome. Metabolic syndrome is now the most common secondary cause of dyslipoproteinemia. This syndrome is characterized by abdominal obesity, elevated blood pressure, dyslipidemia, and peripheral insulin resistance. As mentioned previously, insulin is a vital component in the storage of circulating TG in adipose tissue and when absent leads to increased FFA uptake by the liver and increased production of VLDL. Patients with metabolic syndrome often have elevated TG and low HDL. The role of adipose tissue has been recognized as an endocrine organ releasing substances like adiponectin, leptin, and TNF-α in response to differing stimuli.¹⁷ Obesity and insulin resistance seem to play the main role in the pathogenesis of metabolic syndrome and the subsequent development of dyslipidemia and increased cardiovascular risk.¹⁸

CONCLUSION

The assembly, storage, and transportation of lipoproteins is a complex system with a multitude of components. Understanding the basic role of lipoproteins in the body and how disorders of lipoprotein assembly, storage, and transport manifest clinically are important for the prevention and treatment of dyslipidemia.¹⁹ By understanding each lipoprotein and its role in the body one can gain a better understanding into the mechanism and targets for treating and identifying those with dyslipidemia. Dyslipidemia is an important, but treatable and controllable risk factor for CHD and with better knowledge of lipid pathophysiology and lipoprotein transport clinicians can better serve their patients. Our group has recently published a review describing the clinical management of dyslipidemia in the Journal of Clinical Metabolism and Diabetes.¹⁹

REFERENCES

1. Roger, Veronique, et al. Heart Disease and Stroke Statistics. 2011 Update. Circulation. 2011;123:e18–e209.
2. National Institutes of Health. Third report of the national cholesterol education program expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (ATP III), 2007; Bethesda, MD: National Institutes of Health.
3. Howard BV, Rodriguez BL, Bennett PH, et al. Prevention conference VI: Diabetes and cardiovascular disease: Writing group I: Epidemiology. Circulation. 2002;105:e132.
4. Gu K, Cowie CC, Harris MI. Mortality in adults with and without diabetes in a national cohort of the U.S. population, 1971–1993. Diabetes Care. 2002;25:1129.
5. Dawber T, Meadors G, Moore F.; National Heart Institute, National Institutes of Health, Public Health Service, Federal Security Agency, Washington, DC. Epidemiological Approaches to Heart Disease: The Framingham Study. Presented at a Joint Session of the Epidemiology, Health Officers, Medical Care, and Statistics Sections of the American Public Health Association, at the 78th Annual Meeting, St. Louis, MO, November 3, 1950.
6. Stamler J, Wentworth D, Neaton JD. Is relationship between serum cholesterol and risk of premature death from coronary heart disease continuous and graded? Findings in 356,222 primary screenees of the Multiple Risk Factor Modification Trial (MRFIT). JAMA. 1986;256:2823– 2928.
7. Nissen SE, Tuzcu EM, Schoenhagen P, et al. Effect of intensive compared with moderate lipid-lowering therapy on progression of coronary atherosclerosis. JAMA. 2004;291:1071.
8. Cannon CP, Braunwald E, McCabe CH, et al. Intensive versus moderate lipid lowering with statins after acute coronary syndromes. NEJM. 2004; 350:1495.
9. Sacks FM, Pfeffer MA, Moye LA, et al. The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. Cholesterol and Recurrent Events Trial investigators. NEJM. 1996;335(14):1001–1009.
10. Williams O, Stinnett S, Chambless L, et al. Populations and methods for assessing dyslipoproteinemia and its correlates. The Lipid Research Clinics Program Prevalence Study. Circulation. 1986;73(suppl 1):4–11.
11. Libby P, Genest J, Gotto A, Jr. Braunwald’s heart disease: Lipoprotein disorders and cardiovascular disease. in D P Zippes, P Libby, R O Bonow, E Braunwald (Eds.), A text book of cardiovascular medicine, (pp. 1013– 1033). 2005; Philadelphia: Elsevier Saunders.
12. Davies MJ. The composition of coronary artery plaques. NEJM. 1997; 336:1312.
13. Gianturco SH, Bradley WA. Pathophysiology of triglyceride-rich lipoproteins in atherothrombosis: cellular aspects. Clin Cardiol. 1999; 22(suppl II):7–14.
14. Brown MS, Goldstein JL. A proteolytic pathway that controls the cholesterol content of membranes, cells, and blood. Proc Natl Acad Sci USA. 1999;96:11041–11048.
15. Genest JJ, Martin-Munsky SS, McNamara JR, et al. Familial lipoprotein disorders in patients with premature coronary artery disease. Circulation. 1992;85:2025–2033.
16. Miller M. Current perspectives on the management of hypertriglyceridemia. Am Heart J. 2000;140:232–240.
17. Ronti T. The endocrine factor of adipose tissue: an update. Clin Endocrinol. 2006;64:355–365.
18. Schindler C. The metabolic syndrome as an endocrine disease: is there an effective pharmacotherapeutic strategy optimally targeting the pathogenesis? Therapeutic Adv Cardiovasc Dis. 2007;1:7–26.
19. Sloan K, Vacek J. Treatment of dyslipidemia. J Clin Metab Diabetes. 2011; 2:1–4.

Thyroid autoimmunity is the most prevalent immunological process affecting subjects with type 1 diabetes mellitus. The association between diabetes and thyroid autoimmunity has long been recognized in many populations: in several studies of children and adults with type 1 diabetes, a high prevalence of thyroid autoantibodies (8–44%) has been found as an indicator of autoimmune thyroid disease.^1–10 Heterogeneity has been described in the natural history of thyroid autoimmunity in type 1 diabetes: it may be disclosed either at the onset of diabetes or during the follow-up. Most of the antibody-positive diabetic patients are clinically and biochemically euthyroid. It is uncertain how many of them will later develop thyroid dysfunction but long-term follow-up suggests that the majority of type 1 diabetic patients with positive thyroid antibodies will develop hypothyroidism;¹⁰ hyperthyroidism is much less commonly reported.^{11, 12}

The prevalence of autoimmune thyroid disease in subjects with type 1 diabetes mellitus has been scarcely studied in the Italian population so far. Some studies conducted in the pediatric age group^{7, 13–18} showed a prevalence of about 4–18%.

The objective of this study was to investigate the prevalence of thyroid peroxidase antibodies (TPO-Ab) in a population of children and adults with type 1 diabetes all coming from western Sicily (Italy), as well as their clinical significance for the development of autoimmune thyroid dysfunctions.

PATIENTS AND METHODS

The study design was cross-sectional and descriptive, analyzing children and adults with a previous type 1 diabetes diagnosis. This study was carried out at the Division of Diabetology of the Borsellino Hospital, Marsala (Western Sicily, Italy) during the period from January 2008 to June 2010. All patients with type 1 diabetes, as diagnosed by the American Diabetes Association criteria,¹⁹ attended the outpatient clinic of our division every 3–4 months as part of their routine follow-up. Only the patients with available complete thyroid screening (FT3, FT4, TSH, and TPO-Ab) were considered eligible for the study. A detailed history was obtained for each patient. Information was collected concerning age, gender, time of onset of diabetes, time of disclosure of thyroid autoimmunity, and presence of other autoimmune diseases.

An anti-TPO titer exceeding 100 units/mL was considered significantly elevated. Hypothyroidism was defined as clinically overt if TSH was >10 mIU/L (or in presence of previous therapy with l-thyroxine) and subclinical when TSH was between 4.0 and 10 mIU/L (in absence of therapy with l-thyroxine). Hyperthyroidism was considered if TSH <0.1 mIU/L.

Data were expressed as mean±standard deviation (SD) and/or as percentage. Student's t-test and the χ² test were used to evaluate differences between groups, according to the characteristics of the variables analyzed. Statistical significance was posted at level P<.05.

RESULTS

During the period of the survey around 500 type 1 diabetic patients presented to consultations at our metabolic service. Thyroid screening was available for 374 of them. These patients were enrolled in the study. Among these type 1 diabetic patients 186 were males (49.7%) and 188 were females (51.3%), the mean age was 31.1±15.3 years (range 1–68), and the duration of diabetes was 12.1±10.7 years (range 0–54).

Prevalence of thyroperoxidase antibodies positivity (TPO-Ab+) in our diabetic population was 28.8% (108 patients) and the female-to-male ratio was > 2:1 (75 females, 33 males). The remaining 266 diabetic patients who were autoantibody negative represented the control group. Demographic and clinical characteristics of these subjects are shown in Table 1. Of 108 patients, five had an autoimmune thyroid disease diagnosed and treated before the diagnosis of type 1 diabetes (three Graves’ disease, two hypothyroidism); thyroid autoantibodies positivity was disclosed at the onset of diabetes only in five patients, whereas in the remaining 98 it was revealed after a mean duration of diabetes of 9.7±7.8 years (range 1–28 years).

The TPO-Ab+ patients showed an age at diabetes onset significantly higher versus controls. The TPO-Ab+ patients were not more likely to have celiac disease or other autoimmune diseases compared with those with negative thyroid autoantibodies (Table 1). The thyroid status of most of the patients with positive antibodies was euthyroidism (54 subjects) but overt hypothyroidism (38 subjects) and subclinical hypothyroidism (13 subjects) were also present. Characteristics of these subgroups of patients are shown in Table 2. Subjects with overt hypothyroidism were significantly older than others and showed an age at diabetes diagnosis higher than euthyroid subjects. No significant difference was found in diabetes duration among the groups. In patients with overt or subclinical hypothyroidism a higher prevalence of another autoimmune disease was not found (Table 2). Hyperthyroidism was found in only three adult subjects; in all these cases, thyroid dysfunction was pre-existing to diabetes. There were no subjects who showed thyroid dysfunction in absence of thyroid antibodies.

The TPO-Ab+ subjects were also grouped by gender to analyze clinical features. The results showed no differences between males and females regardless of age, illness duration, and age at diabetes onset (data not shown).

Finally, all the patients were divided into three age groups according to developmental stage: age group 1 (0–12 years; n=47), age group 2 (12–18 years; n=44), age group 3 (>18 years; n=283). Prevalence of thyroperoxidase antibody positivity was 23.4% in group 1, it was 22.7% in group 2, and 30.7% in group 3. Characteristics of these groups of patients are shown in Table 3.

DISCUSSION

The prevalence of thyroid autoimmunity has been reported to be increased in subjects with type 1 diabetes mellitus compared with the general population; however, frequencies differ among studies.^1–10 Age, sex, time of evolution of diabetes, and genetic and geographical background can explain the differences across studies. Those published have emanated largely from pediatric referral departments rather than general diabetes centers. Moreover, prevalence variation among Italian studies was reported.^{7, 13–18}

In our study, performed in the western part of Sicily (Italy), the presence of thyroid antibodies was found in 108 out of the 374 subjects (children and adult patients) with type 1 diabetes, while thyroid dysfunction occurred in 50% of those with positive TPO-Ab. Therefore, we found serological markers of thyroid autoimmunity in 28.8% of our patients, a higher percentage than previously reported by others in Italy, but our results were hardly comparable to these studies conducted mostly in the pediatric age group and in different geographical areas with different iodine disposability.

By dividing diabetic subjects in relation to age, there was a higher frequency of thyroperoxidase antibody positivity, although not significant, in the group of adults than in children and adolescents groups. Therefore we found that younger patients were more likely to be TPO-Ab negative than older patients; this is in agreement with the majority of previous reports.

In accordance with the prevalent literature, our data confirm that autoimmune thyroiditis is significantly more prevalent in female subjects with type 1 diabetes than in males;^{1, 8, 10, 14–18} however, other studies did not find such difference.^{6, 20} It is known that type 1 diabetes is equally prevalent among males and females in most populations. This unequal gender distribution of thyroid autoimmunity prevalence in patients with diabetes suggests that etiological risk factors for developing antibodies against thyroid could not be the same as for type 1 diabetes.

Among the novelties of our study is the association of thyroid autoimmunity with the age at diabetes diagnosis. To our knowledge there are very limited data in the literature on the above topic. Goodwin reported that younger age at onset of type 1 diabetes was associated with increased risk for autoimmune thyroid disease, but this work was conducted in a selected population of concordant sibling pairs²¹. On the contrary, Verge et al²² found that TPO-Ab prevalence rose with age of diabetes diagnosis and in a recent paper, Machnica⁹ noticed that newly diagnosed diabetic patients with thyroid autoimmunity were significantly older at the time of presentation of diabetes. In a large multicenter survey, Kordonouri¹⁰ showed that patients with thyroid antibodies were significant older, had a longer duration of diabetes, and developed diabetes later in life than those without antibodies. Our TPO-Ab+ patients had a higher age at clinic visit and at diabetes onset of diabetes than TPO-Ab negative subjects; duration of type 1 diabetes seemed to have no influence on the presence of thyroid autoimmunity in our case report. Therefore, one can assume that a subgroup of type 1 diabetic patients associated with thyroid autoimmunity might present in the female patients of older age at diagnosis of diabetes; this result could be explained by the longer period of exposure to risk factors that, coexisting with a genetic predisposition, may initiate autoimmunity processes but the factors responsible for this interesting phenomenon need to be investigated further.

Different results were published with respect to the prevalence of hypothyroidism in subjects with type 1 diabetes and positive TPO-Ab. We have found a very high prevalence of thyroid dysfunction in the group of patients with type 1 diabetes and thyroid autoimmunity. Among them, overt hypothyroidism was predominant (35%), while subclinical hypothyroidism was present in 12% of them. Hyperthyroidism was far less common. Patients who had overt hypothyroidism were older at diabetes diagnosis than those who did not. Similar results were previously reported by Gonzalez et al.²³

Interestingly, in patients with type 1 diabetes and another autoimmune disease, diabetes onset mostly precedes the diagnosis of the other disorder. Similarly, in the present study, except for five subjects, all patients developed autoimmune thyroiditis after the manifestation of diabetes.

The natural history of thyroid disorders in our diabetic patients has not yet been established, but the progression thyroid autoimmunity seems very slow. In our patients, autoimmune thyroid disease typically follows the diagnosis of diabetes by an average of 10 years (range 1–28 years); these data are in agreement with those reported by Umpierrez et al²⁴ in its longitudinal study; thus, screening for thyroid disease should be considered for many years after diabetes diagnosis. Therefore, we recommend that if the initial thyroid screening is negative, future tests should be performed in type 1 diabetic patients, even in the absence of clinical signs.

Also, thyroid autoimmunity has been frequently described in patients with other autoimmune diseases such as celiac disease. However, our diabetic patients with thyroid autoimmunity did not show a higher prevalence of celiac disease than those without significant thyroid antibodies; our data are in agreement with previous reports²⁵ but in contrast with others.¹⁷

In conclusion, the presence of thyroid antibody positivity and the subsequent development of subclinical or clinical autoimmune thyroiditis were quite prevalent among the type 1 diabetic patients of our study, while the possible risk factors for its development were older age at diabetes diagnosis and female gender. Although a considerable number of patients (28.8%) have markers of autoimmune thyroid disease, few of them have positive thyroid autoimmunity at diagnosis. These data support the recommendation for regular (ie, yearly) examinations of thyroid antibodies, particularly of TPO-Ab, in all patients with type 1 diabetes mellitus; screening should start at diagnosis of diabetes but should also be repeated for decades after clinical onset.

Keywords

type 1 diabetes mellitus, autoimmune thyroid disease, thyroid autoimmunity

Disclosure: The authors declare no conflict of interest.

REFERENCES

1. Riley WJ, Maclaren NK, Lezotte DC, Spillar RP, Rosenbloom AN. Thyroid autoimmunity in insulin-dependent diabetes mellitus: the case for routine screening. J Pediatr. 1981;99:350–354.
2. McKenna MJ, Herskowitz R, Wolfsdorf JI. Screening for thyroid disease in children with IDDM. Diabetes Care. 1990;13:801–803.
3. Kontiainen S, Schlenzka A, Koskimies S, Rilva A, Maenpaa J. Autoantibodies and autoimmune diseases in young diabetics. Diabetes Res. 1990;13:151–156.
4. Darendeliler FF, Kadioglu A, Bas F, et al. Thyroid ultrasound in IDDM. J Pediatr Endocrinol. 1994;7:33–37.
5. Hansen D, Bennedbaek FN, Hansen LK, Hoier-Madsen M, Jacobsen BB, Hegedus L. Thyroid function, morphology and autoimmunity in young patients with insulin-dependent diabetes mellitus. Eur J Endocrinol. 1999; 140:512–518.
6. Lindberg B, Ericsonn UB, Ljung R, Ivarsson SA. High prevalence of thyroid autoantibodies at diagnosis of insulin-dependent diabetes mellitus in Swedish children. J Lab Clin Med. 1997;130:585–589.
7. Lorini R, D’Annunzio G, Vitali L, Scaramuzza A. IDDM and autoimmune thyroid disease in the pediatric age group. J Pediatr Endocrinol Metab. 1996;9(suppl 1):89–94.
8. Chang CC, Huang CN, Chuang LM. Autoantibodies to thyroid peroxidase in patients with type 1 diabetes in Taiwan. Eur J Endocrinol. 1998;139: 44–48.
9. Machnica L, Osior A, Jarosz-Chobot P, Deja G, Polanska J, Otto-Buczkowska E. An analysis of the prevalence of thyroid autoantibodies: thyroid peroxidase antibodies (ATA) and thyroglobulin antibodies (ATG) in children with newly diagnosed diabetes mellitus type 1 during 2000–2004 in the upper Silesia region, Poland. Acta Diabetol. 2008;45:37–40.
10. Kordonouri O, Klinghammer A, Lang EB, Gruters-Kieslich A, Grabert M, Holl RW. Thyroid autoimmunity in children and adolescents with type 1 diabetes. A multicenter survey. Diabetes Care. 2002;25:1346–1350.
11. Perros P, McCrimmon RJ, Shaw G, Frier BM. Frequency of thyroid dysfunction in diabetic patients: value of annual screening. Diabet Med. 1995;12:622–627.
12. McCanlies E, O’Leary LA, Foley TP, et al. Hashimoto’s thyroiditis and insulin-dependent diabetes mellitus: differences among individuals with and without abnormal thyroid function. J Clin Endocrinol Metab. 1998;83: 1548–1551.
13. Betterle C, Zanette F, Pedini B, et al. Clinical and subclinical organspecific autoimmune manifestations in type 1 (insulin-dependent) diabetic patients and their first-degree relatives. Diabetologia. 1984;26: 431–436.
14. Trimarchi F, De Luca F, Vanelli M, et al. Circulating thyroid antibodies and thyroid function studies in children and adolescents with insulindependent diabetes mellitus. Eur J Pediatr. 1984;142:253–256.
15. Lorini R, Larizza D, Livieri C, et al. Auto-immunity in children with diabetes mellitus and in their relatives. Eur J Pediatr. 1986;145:182–184.
16. Radetti G, Paganini C, Gentili L, et al. Frequency of Hashimoto’s thyroiditis in children with type 1 diabetes mellitus. Acta Diabetol. 1995;32: 121–124.
17. Franzese A, Buono P, Mascolo M, Leo AL, Valerio G. Thyroid autoimmunity starting during the course of type 1 diabetes denotes a subgroup of children with more severe diabetes. Diabetes Care. 2000;23: 1201–1202.
18. Lenzi L, Mirri S, Generoso M, et al. Thyroid autoimmunity and type 1 diabetes in children and adolescents: screening data from Juvenile Diabetes Tuscany Regional Centre. Acta Biomedica. 2009;80:203–206.
19. American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care. 2010;33(suppl 1):S62–S69.
20. Glastras SJ, Craig ME, Verge CF, Chan AK, Cusumano JM, Donaghue KC. The role of autoimmunity at diagnosis of type 1 diabetes in the development of thyroid and celiac disease and microvascular complications. Diabetes Care. 2005;28:2170–2175.
21. Goodwin G, Volkening LK, Laffel LMB. Younger age at onset of type 1 diabetes in concordant sibling pairs is associated with increased risk for autoimmune thyroid disease. Diabetes Care. 2006;29:1397–1398.
22. Verge CF, Howard NJ, Rowley MJ, et al. Anti-glutamate decarboxylase and other antibodies at the onset of childhood IDDM: a population-based study. Diabetologia. 1994;37:1113–1120.
23. Gonzalez GC, Capel I, Rodriguez-Espinosa J, Mauricio D, De Leiva A, Perez A. Thyroid autoimmunity at onset of type 1 diabetes as a predictor of thyroid dysfunction. Diabetes Care. 2007;30:1611–1612.
24. Umpierrez GE, Latif KA, Murphy MB, et al. Thyroid dysfunction in patients with type 1 diabetes. A longitudinal study. Diabetes Care. 2003;26:1181–1185.
25. Li Voon Chong JS, Leong KS, Wallymahmed M, Sturgess R, MacFarlane IA. Is coeliac disease more prevalent in young adults with coexisting type 1 diabetes mellitus and autoimmune thyroid disease compared with those with type 1 diabetes mellitus alone? Diabet Med. 2002;19:334–337.

Diabetes mellitus has become a leading public health problem in the second half of the last century and it will continue to be so well into the second millennium ¹. This diabetes epidemic is mainly attributable to type-2 diabetes that has accompanied the upsurge of obesity, sedentary life style, and global changes in dietary habits ². A totally new phenomenon is that a significant element of the diabetes syndrome has started to appear among children and teenage populations ^{3, 4}. This, in its own right, will represent a significant future public health problem ⁵ as diabetes is a chronic disease with significant chronic complications known to be related to diabetes duration ⁶ plus other factors including glycemic control ^{7, 8}. Vascular disease, as a consequence, will be a major problem to tackle in people with diabetes in years to come, an area which has not been a priority for health management in pediatric populations ⁵. The enigma of the plausible rise of vascular disease stemming from such a rise in the rate of type-2 diabetes in young people will be further compounded by a lack of insight into the interaction of various risk factors of vascular disease in this new population with metabolic disease coming to the forefront.

This review will focus on the latest state of the art in the field of endothelial dysfunction, the forerunner of vascular disease in young people with diabetes, with special emphasis on those who are free from any diabetes complications. We shall also discuss prospective strategies to deal with this dysfunction.

THE FUNCTIONS OF THE ENDOTHELIUM

In the wake of advances in vascular biology and molecular cell research, the vascular endothelium is no longer a passive barrier between the lumen and vascular wall and it is now well established that it is a proactive organ with dynamic and protean spectrum of functions ⁹. The various functions of the vascular endothelial cells (ECs) have been unraveled in the last three decades, and this has resulted in a revolution of the way physicians currently regard the vascular endothelium in clinical perspective. Furthermore, the insight gained from both clinical and basic science research has added a great deal to the way various vascular disease and vascular risk are perceived and managed. The functions of the vascular endothelium can be summarized as follows.

Regulation of vascular tone

The endothelium plays the most pivotal role in regulating vascular tone, which is an important component of blood flow, blood pressure, and hemostasis. The endothelium synthesizes both vasodilator and vasoconstrictor agents and it regulates the balance between the two. The major vasodilator agent produced by endothelium is nitric oxide (NO; endothelium-derived relaxing factor). NO is synthesized from the amino acid l-arginine by the enzyme nitric oxide synthase (NOS). There are several isoforms of NOS of which the most relevant to this theme are NOS type III and NOS type I. The former is isolated from endothelial cells (Ecs) and the later was isolated from the brain. The NOS type III is the enzyme responsible for synthesis and release of NO from the EC. Following its release, NO exerts its effect on the vascular smooth muscle cells (VSMCs) in a paracrine fashion. This will result in relaxation of the vascular smooth muscle bed resulting in dilatation of the vessel wall. Such a process occurs in response to various stimuli, mainly mechanical (shear stress), and in response to other factors that are receptor-dependent (acetylcholine, metacholine) or receptor-independent stimuli (calcium iontophore) ¹⁰. These effects are generated through the constitutive NOS type III, and these mechanisms represent an important regulatory factor for blood pressure. Further, the endothelium also produces vasoconstrictor agents that include endothelin ¹¹, prostaglandin ¹², and angiotensin-II ¹³. The balance between the vasodilator and vasoconstrictor agents regulates vascular tone and is crucial for progression of vascular disease ¹⁴. In conditions where there is endothelial dysfunction the balance will tip toward vasoconstriction.

Role in hemostasis regulation

The endothelium plays a crucial role in the regulation of blood rheology and hemostasis. Endothelium synthesizes important factors that are involved in the process of coagulation and fibrinolysis and the balance between them ^15–17. NO is the most crucial substance that is involved in the regulation of hemostasis. The bradykinin mediated NO release inhibits platelet adhesion to the vascular EC ¹⁸, and this is executed by NO even under flow conditions ¹⁹. The NO is well documented to inhibit platelet aggregation ²⁰. Elevation of cyclic 3′5′-guanosine monophosphate (cGMP) mediated by NO plays a crucial role in preventing platelet activation ²¹ reduce the expression of P-selectin expression and glycoprotein IIb-IIIa ²². The thrombin generated NO release acts further in inhibiting adhesion of platelets to the endothelium ²³.

The coagulation cascade with its two pathways that culminates in the formation of thrombin and plasmin represents an important cofactor. Tissue type plasminogen activator is an important agent in this cascade and its determining factor is plasminogen activator inhibitor-1 (PAI-1). Both are synthesized by the endothelium.

Role in inflammation

Another important function of the endothelium is the regulation of inflammation ²⁴. The endothelium synthesizes the cell adhesion molecules family, the intercellular cell adhesion molecule-1 (ICAM-1), vascular cell adhesion molecules (VCAM), and the E-selectin ²⁵. These adhesion molecules play an important role in cell-cell interaction and in regulation of the inflammatory process. Further, NO and prostacyclin are also important players in the inflammatory process ^{16, 26}. The net result of these functions is a complex array of regulation of vascular tone, increase vascular permeability, vascular smooth muscle growth and differentiation, and tissue inflammation.

ASSESSMENT OF ENDOTHELIAL FUNCTION

Biophysical markers of endothelial cell function

This broadly shows the capacity of the ECs to respond to various noxious stimuli that may result in shedding of these surrogates, either directly via agents that stimulate the endothelium (endothelium-dependent vascular responses) or indirectly through other pathways that do not directly involve the endothelium (endothelium-independent vascular responses. Such assessments can be carried at the level of the microcirculation utilizing sophisticated techniques or at the macrovascular level. Assessment of conduit EF can be carried out using the noninvasive technique of flow-mediated dilatation (FMD), where changes in the brachial artery diameter in response to occlusion by applying a pressure cuff and the subsequent hyperemic response following the release of pressure are measured by high resolution ultrasonography ²⁷. This will assess the endothelium-dependent response as the release of NO will follow the net increase in shear stress and this will then stimulate relaxation of the vascular smooth muscle layer of the artery ^{27, 28}. The FMD is combined with the assessment of endothelium-independent responses achieved by a sublingual intake of glyceryl trinitrite (GTN). More sophisticated measures at the microvascular level utilize the laser Doppler technique to assess both endothelium-dependent and endothelium-independent vascular responses ²⁹. This could be achieved either via the hyperemic response following heating stimulus or via delivering a vasoactive reagent utilizing the technique of iontophoresis through the skin. The technique of laser Doppler flowmetry involved utilizing low current iontophoresis of various stimuli and has been well validated by our group ^{30, 31}. In this study acetylcholine iontophoresis as an endothelium-dependent stimulus and sodium nitroprusside as an endothelium-independent stimulus were used. Recently another method has been developed via assessment of hyperemic response on the finger tip by using tonometry ³².

Biochemical markers of endothelium activation/perturbation

These show the ability of ECs to respond to various noxious stimuli and in its aftermath markers that are indicative of ECs dysfunction and injury can be assayed. Classically these include molecules that are synthesized by ECs. Of these, the most important are cell adhesion molecules that include ICAM-1, VCAM, and E-selectin ^{33, 34}, von Willebrand factor (vWF) ³⁵, and thrombomodulin (TM) ³⁶. Other surrogates include tPA and PAI-1 ³⁷. Several of these agents are either present in the ECs or other cell lines in concentrations compatible with their physiological roles. The adhesion molecules, for instance, are needed for cell-cell interaction at basal physiological milieu. Increasing levels of the above molecules are generally taken to be a manifestation of ongoing ECs activation/perturbation with eventual cellular damage ²⁵.

Intima media thickness

This method utilizes high resolution ultrasonography to measure the carotid intima-media thickness as a crude method of early atherosclerosis. The IMT in young people may probably represent a more advanced surrogate of endothelial dysfunction ³⁸, with early biophysical changes of endothelial dysfunction and increased arterial pulse wave velocity (PWV) occurring in advance of IMT ³⁹. It is a well-known marker for early atherosclerosis in adults ⁴⁰. In children with type-1 diabetes it has been shown to predict coronary artery disease ^{41, 42}, and also to be closely associated with vascular risk markers in children and adolescents with overweight and obesity ⁴³. Further, in patients with type-1 diabetes it showed increment with the presence of microvascular complications, suggesting that it might be a link between micro- and macrovascular disease ⁴⁴. Worth mentioning is that there is no ideal method of assessing the endothelial dysfunction either at the macro- or at the microvascular level. Each method has its disadvantages and limitations. The discussion of these is beyond the scope of this paper. However, it would be ideal to combine the assessment of the biophysical and biochemical markers in order to get more insight when studying any given population.

ENDOTHELIAL DYSFUNCTION IN YOUNG PEOPLE WITH DIABETES

In recent years and in contrast to previously held concepts it was observed that youngsters with diabetes are not immune from early evidence of vascular dysfunction ⁴⁵. Endothelial dysfunction has been shown to be present in young patients with diabetes including children and adolescents. This dysfunction has been shown to be present very early in the course of diabetes and well in advance of any clinical microvascular dysfunction. In an earlier report, we have shown that children with type-1 diabetes have an impaired skin blood flow response to heat ⁴⁶. In a subsequent study involving 56 youngsters with type-1 diabetes (age range 9-22 years), 26 matched healthy normal controls. We further elaborated on this earlier finding of assessing the biophysical markers that involve both the functional and structural markers by utilizing the sophisticated technique of laser Doppler flowmetry in addition to the heating stimulus ⁴⁷. The study subjects were normotensive, nonobese, had an average diabetes duration of 7 years, and were of average mean age of 14 years (range 9-22). None of them had any evidence of clinical microvascular disease (retinopathy, clinical neuropathy with no evidence of microalbuminuria). The index and control group in this study had similar skin perfusion at baseline. However, the diabetes youngsters exhibited evidence of impaired vascular responses to both endothelium-dependent and endothelium-independent stimuli. The reduced vascular responses to acetylcholine correlated with diabetes duration and with glycemic control (measured by HbA1c), but it did not show any correlation with age or other markers of excess vascular risk namely total cholesterol, systolic or diastolic blood pressure, heart rate, or insulin dose. There was no correlation between those aforementioned variables and the reduced response to the endothelium-independent stimuli obtained after inotopheresis of sodium nitroprusside. Further, the subjects with diabetes also showed reduced response to local heating (at 44°C) confirming the earlier observation by our group ⁴⁶. We postulated that the reduction in all three components of the biophysical markers may point to a combined functional and structural changes in the microcirculation in these young patients with diabetes. In a study by Donaghue et al ⁴⁸, brachial artery FMD assessing the endothelium-dependent responses and endothelium-independent responses following sublingual GTN were both reduced in 12 out of 20 young patients with type-1 diabetes compared with 20 nondiabetic healthy controls. However, in contrast to our study, Donaghue et al assessed the macrocirculation rather than the microcirculation and, indeed, their findings were in line with our own despite studying a different vascular bed. Further, their finding did not correlate with HbA1c. A recent study by Odermarsky et al ⁴⁹ involving 59 youngsters with type-1 diabetes in Sweden, reduced cutaneous microvascular responses to Ach but not to sodium nitroprusside was observed in their cohort. Compared with our study, their group were relatively older and had had a relatively longer duration of diabetes than the group in our study (average 14 years vs 18 years, 1-18 years vs 3-22 years, respectively). Their cohort also had delayed myocardial repolarization assessed by the duration of QT interval corrected for heart rate and also showed increased carotid intima media thickness. Heilman et al ⁵⁰ studied a group of 30 youngsters with type-1 diabetes (age range 4.7-18.6 years) and a group of 30 healthy children and assessed the carotid intima-media thickness and carotid PWV. The diabetes group had higher IMT and PWV. There was close correlation between IMT, HbA1c, and PWV. Ladeia et al ⁵¹, however, were unable to show impaired endothelium-dependent responses denoted by FMD of the brachial artery in a group of young type-1 diabetes patients from Italy who had diabetes duration less than 5 years. But in contrast to the above studies, they have a relatively small number of subjects in their study, which may explain their negative finding. The authors were able to confirm a correlation between the degree of microalbuminuria and the obtained hyperemic response. Furthermore, they were also able to show positive correlation with HbA1c, suggesting a modulatory role of chronic hyperglycemia. Studies assessing biophysical markers of endothelial dysfunction are summarized in Table 1.

Markers of EC activation have also been studied in youngsters with diabetes in the absence of clinical vascular disease. In a study involving 51 children, adolescents, and young adults with type-1 diabetes and 29 healthy nondiabetic control (matched for age, sex, blood pressure, and total cholesterol) we studied markers of EC activation that included ICAM-1, E-selectin, and von Willebrand factor ⁵². The group with diabetes had levels of ICAM-1 and E-selectin, which were significantly greater than those in the healthy normal control group. The vWF levels were not different between the two study groups. Further, there was no correlation between E-selectin, ICAM-1, and diabetes duration or HbA1c in the diabetic group. Stemming from these finding we were able to postulate that there is evidence of significant EC activation in youngsters with diabetes that may set the scene for future diabetic complications. Romana et al ⁵³ confirmed the above finding of endothelial perturbation in youngsters with diabetes early in the course of the disease in a study that involved 40 young patients with type-1 diabetes and age- and sex-matched healthy controls. The diabetic group had either <1 year or >1 year of diabetes duration ⁵³. In this study, markers of EC activation, including vWF and tPA as well as markers of inflammation including TNF-α and CRP, were measured. Intriguingly, the group with <1 year of diabetes duration showed higher values of CRP, TNF-α, vWF, tPA, and prothrombin fragments 1 and 2 (antigen fragments of prothrombin that are indicative of EC perturbation) when compared with those with diabetes duration >1 year and the healthy control. Those markers were normal in 45% of the diabetic group after >1 year (Group B). The authors argued that whether the observed phenomenon is a reflection of the risk factor for future vascular dysfunction or whether it is a transient phenomenon. They went further to hypothesize that early perturbation of EC function coincides with the rise in immunological and inflammatory markers that may set the scene for future angiopathy. In a study by Carrizo Tdel et al ⁵⁴ carried in diabetic youngsters between the ages of 6 and 15 years, they found that levels of E-selectin were 66% higher in the group with diabetes compared with age- and sex-matched healthy controls. Further, levels of E-selectin in the diabetic group were even higher in diabetic youngsters with poor control compared with those who had a lower level of HbA1c, and levels of E-selectin correlated with HbA1c in the overall group with diabetes. Further, the diabetic subjects have significant abnormalities of the lipoprotein subfractions including high total cholesterol, low HDL and high LDL cholesterol. Earlier, our group has shown that plasma levels of both TM and VEGF were significantly increased in juvenile diabetic patients with no clinical evidence of vascular disease compared to normal age- and sex-matched control subjects ⁵⁵. Wiltshire et al ⁵⁶ were not able to show any difference between the levels of vWF and TM in a group of 35 young people with type-1 diabetes when compared to matched healthy normal control. However, they were able to show that those who had a higher folate level will have lower TM, but no effect on VWF was found. Dimeglio et al ⁵⁷ conducted assessment of the endothelial integrity and perturbation by an indirect novel way of circulating endothelial progenitor biomarkers; the circulating colony forming unit-endothelial cells (CFU-ECs) including CD33, CD133, CD31, and CD34+CD45–. The study group had lower CD34+CD133+CD31+ cells. These later are considered as the harbinger of future macrovascular disease risk and the higher CD34+CD45– reflect an ongoing EC damage. The studies assessing biochemical markers of EC activation are summarized in Table 2.

Taking the structural changes of the vascular endothelium in the macrocirculation from a different perspective via assessing the intima-media thickness that is a sensitive marker of early atherosclerosis rather than a marker of endothelial dysfunction per se, several studies have involved youngsters with diabetes. Järvisalo et al ⁴² combined the assessment of brachial artery FMD with assessing carotid IMT in 45 subjects with diabetes compared to 30 healthy nondiabetic youngsters with a mean age of the combined group of 11±2 years. The authors were able to document impaired endothelial function and increased IMT in 16 of those with type-1 diabetes. In this study, changes in IMT correlated with reduction in FMD and high LDL cholesterol.

An earlier study by Singh et al ⁵⁸ reported similar findings to that of Järvisalo and coworkers. They studied a group of 31 teenagers with diabetes and 35 nondiabetic matched controls and, using the same technique, they found that flow mediated endothelium-dependent dilatation of the brachial artery was significantly impaired in those with diabetes. No difference was found in the endothelium-independent dilatory response with reference to the control group. Further, such impaired vasodilatory responses correlated with diabetes duration, blood glucose levels, total cholesterol, and LDL cholesterol. However, their diabetic group showed no increase in IMT compared to the nondiabetic group. A study among young Hispanic population with type-1 diabetes from Mexico also showed an increased intima-media thickness and reduced carotid flow velocities (end diastolic velocity and peak systolic velocity) in the patients group who was composed of 52 patients (mean age 11±3 years) compared to an age, sex, and body weight matched healthy control ⁵⁹. In the study by Gül et al ⁴⁴ that assessed IMT in patients with type-1 diabetes there was exponential increase in carotid IMT with the appearance of microvascular complications.

Type-2 diabetes, a disease well known to be associated with insulin resistance is appearing more frequently in youngsters in accompaniment of the global diabetes epidemic ^{3, 4}. Endothelial dysfunction is a regular proximate of insulin resistance, so it would be expected that youngsters with type-2 diabetes may exhibit evidence of endothelial dysfunction. However, there is paucity of data in this area. Shah et al ⁶⁰, in a study involving 129 young people with type-2 diabetes of mixed ethnicity that assessed markers of early atherosclerosis using IMT as a surrogate, were able to show that IMT is significantly increased and this correlated with diabetes duration and glycemic control. Such correlation was independent of the traditional risk factors for vascular disease that included male sex, LDL cholesterol, and blood pressure. Further, the abnormality was more prominent in male subjects.

The studies relating to IMT and both type-1 and type-2 diabetes in the young people are summarized in Table 3.

PUTATIVE MECHANISMS OF ED IN YOUNGSTERS WITH DIABETES

As diabetes mellitus is a state of metabolic disturbance and this would carry significant correlates, one could argue that these could be possible contributors to endothelial dysfunction, with the logical argument of chronic hyperglycemia as the initiating factor and others are interplayers/perpetuators. Here we shall elaborate on such possible mechanisms.

Hyperglycemia and its cellular consequences

The mechanisms underlying the role of hyperglycemia in triggering endothelial dysfunction are both complex and multifactorial ⁶¹. In experimental animals it has been shown that hyperglycemia is the major determinant of the development of ED ⁶². Hyperglycemia results in an excess generation of vasoconstrictor prostanoids and abnormalities of the cellular pathways involving diacyl glycerol (DAC) and protein kinase-C (PKC) ^{63, 64}. Correction of the hyperglycemia and attainment of normoglycemia by insulin therapy has been shown to abrogate such an increase in PKC and DAG ⁶⁵.

In normal children, high postfeeding glucose is associated with impaired microvascular responses ⁶⁶. In this study, which is a subset study from an earlier study set up to examine the effect of infant feeding practices on children's health, plasma glucose levels 2 h after the feeding challenge varied considerably amongst the subjects (No=159), ranging from 3.9 to 11.4 mmol l/1. Over this range, there were no significant correlations between glucose levels and microvascular responses. However, stratifying subjects into quintiles according to the 2-h glucose level did show significant differences in microvascular responses to ACh and SNP amongst the groups. In a subgroup analysis, comparison of microvascular responses between subjects in the lower quintile of 2-h glucose (3.9-4.9 mmol l/1, n=29) with subjects from the upper quintile (7.4-11.4 mmol l/1, n=29) showed a similar baseline skin perfusion in the two groups but significantly reduced microvascular responses to ACh and SNP. There was a highly significant correlation between ACh and SNP responses in all subjects and also within subjects in the upper quintile of 2-h glucose. Furthermore, subjects in the upper glucose quintile had significantly elevated 2-h insulin levels, higher fasting insulin resistance index, a greater waist-to-hip ratio, and a trend toward a greater low density lipoprotein (LDL) cholesterol level, than subjects in the lower glucose quintile. In subjects in the upper glucose quintile, fasting triglyceride levels correlated with fasting insulin levels, with the fasting insulin resistance index, and total cholesterol correlated with 2-h glucose. So, the message from this clinical study could provide significant insight into the impact postprandial glucose rise even within the normal reference range on the microvascular function. It is plausible from the result of this study that there appears some interaction of the multiple risk factor of vascular disease in children and young people well in advance of any development of abnormalities of glucose tolerance.

On the other hand, in healthy adults without diabetes, oral glucose loading was found to attenuate endothelium-dependent vasodilatation, an effect that was prevented by antioxidants ⁶⁷. Children with type-1 diabetes are reported to have lack of compensation of increase in vascular resistance secondary to the tendency to have excess blood viscosity and mean arterial blood pressure ⁶⁸.

Free radical generation and oxidative stress

Diabetes induced free radical generation and oxidative stress is documented to contribute to vascular injury and complications in diabetes ⁶⁹. We reported earlier that youngsters with type-1 diabetes have excess levels of markers of free radical generation and oxidative stress well in advance of any clinical vasculopthy ⁵². In that study, levels of superoxide dismutase (SOD), plasma thiol (PSH), and glutathione (GSH) were significantly perturbed in a group of young patients with diabetes, with levels of SOD correlating negatively with age and diabetes duration. Suys et al ⁷⁰ has shown that free radical generation in young people with type-1 diabetes show a correlation with impaired FMD of the brachial artery. A subgroup of their study cohort had higher circulating Cu/Zn SOD, implying that higher SOD status has a protective role for ED seen with low levels of SOD. Young people with type-1 diabetes even had higher levels of free radical markers than age- and sex-matched obese children ⁷¹.

Mitochondrial dysfunction

The mitochondrial organelles in healthy cells are in a state of dynamic activities, with a continuous cascade of mitochondrial fusion to form larger cellular organelles, mitochondrial fission to form daughter mitochondria. Those later are removed by the process of autophagy. This process of “mitochondrial biosynthesis” occurs in a balanced and controlled state termed “mitochondrial dynamics.” ⁷² The formation of new mitochondria (mitochondrial biosynthesis) is regulated by peroxisome proliferator activator gamma coactivator 1-alpha (PPARγ coactivator 1-α), and nuclear respiratory factor-1 (NRF-1) ⁷³. The mitochondria represent the hub of generation of the major fuel element, the adenosine triphosphate (ATP), which is essential for cell growth, differentiation, and programmed cell death ⁷⁴.

Diabetes has been shown to be associated with disruption of the “mitochondrial dynamics.” Cells will have reduced mitochondrial mass, impaired autophagy, disrupted mitochondrial fusion, and predominance of mitochondrial fragmentations ⁷⁵. This will result in subsequent disturbances in mitochondrial energy generation that is termed “mitochondrial dysfunction.” Hyperglycemia per se will lead to increased mitochondrial membrane potential, excess free radical generation in situ of the mitochondrial organelle, mitochondrial DNA damage, and eventual suppression of mitochondrial function ⁷⁶. Studies of skeletal muscles in patients with diabetes have shown small mitochondria and decreased mitochondrial mass ⁷⁷. Furthermore, decreased expression of genes related to mitochondrial biosynthesis and those involved in oxidative phosphorylation ⁷⁸. Life style measures including exercise and caloric restriction were reported to improve mitochondrial biogenesis in patients with diabetes ^{79, 80}.

Dyslipidemia

Abnormal lipid metabolism is a well-established cause for endothelial dysfunction. In recent years, evidence has accumulated on the prevalence of dyslipidemia in young people with diabetes. Maahs et al ⁸¹ have reported that 18.6% of children with type-1 diabetes had levels of total cholesterol (TC) >200 mg/dl or levels of HDL <35mg/dl. A cross-sectional, population-based study “The SEARCH for Diabetes in Youth Study,” ⁸² which was conducted in 2448 youth with diabetes who had fasting measures of TC, LDLc, and TG, showed that 5% had TC concentration >240 mg/dl, 3% had LDLc, and 2% had TG >400 mg/dl. About half of the participants (48%) had an LDL-C concentration above the optimal level of 100 mg/dl. Further, among the youth ages 10+, the prevalence of abnormal lipids was higher in type-2 than in type-1 diabetes: 33 vs 19% had TC concentration >200 mg/dl; 24 vs 15% had LDL-C concentration >130 mg/dL; 29 vs 10% had triglyceride concentration >150 mg/dl; and 44 vs 12% had HDL-C concentration <40 mg/dl. More interestingly is that in this study, only 1% of youth were receiving pharmacologic therapy for dyslipidemia. A study by Mahmud et al ⁸³ showed that microvascular responses assessed by peripheral arterial tonometry were abnormal in a group of 23 youngsters with type-1 diabetes (aged 12-18 years) in the fasting state and these worsened further following a high fat meal (comprising of total fat, saturated fat, and cholesterol). In the review by Pinhas-Hamil and Zeitler ⁸⁴, children and young people with type-1 diabetes who had various diabetic complications were found to have lipid abnormalities ranging between 15%-62%.

Various studies have documented correlation of the conventional risk factors in children with the development of atherosclerosis in adulthood, including intima media thickness, arterial compliance, and pathological lesions of atherosclerosis ^85–87. Despite some recommendations to address the documented abnormalities of lipid metabolism in youth with diabetes, there are no data to date to confirm the safety of current lipid lowering agents in this group of patients.

Sex hormones

There are reports of gender differences in EDs suggesting that sex hormones may have a modulatory role on endothelial function. There appears to be differences between males and females with some studies suggesting less favorable responses of EDs in boys ^{66, 70}. On the other hand, a pubertal surge of sex hormones may have detrimental effects on EC function. We have earlier reported that puberty has a modulatory role on markers of EC activation, vascular responses, and oxidative stress in young people with type-1 diabetes ⁸⁸. In this study that has three groups of diabetes youngsters assessed by Tanner Classification—a prepubertal group, an adolescent group, and a young adults group—the adolescent group had the worst responses of microvascular reactivity to Ach, higher ICAM-1 and E-selectin, and lower SOD and GSH. These findings were independent of the conventional risk factors including blood pressure and total cholesterol. Puberty has been well documented before to have a modulatory effect on insulin resistance ⁸⁹, as well as lipid metabolism with increased glycerol and lipid oxidation ⁹⁰. Clinical studies have long documented the detrimental effects of development and progression of diabetic complications in young people ^{91, 92}.

High homocystiene levels

Hyperhomocysteinemia is known to be an independent risk factor for vascular disease and this may contribute to the ED in patients with type-1 diabetes ⁹³. Folate status has been associated with endothelial dysfunction in children and adolescents with type-1 diabetes. Wiltshire et al ⁵⁶ found that the reduced FMD and GTN-induced responses of brachial artery in a group of children and adolescents with type-1 diabetes correlates independently with the folate status, and also the vessel diameter correlated inversely with total homocysteine level. The authors concluded that folate supplements may confer a protective effect on ED in young people with diabetes.

STRATEGIES FOR TREATMENT AND PREVENTION

Strict glycemic control

It is well established from the results of DCCT that strict glycemic control can retard progression or prevent the development of microvascular complication in patients with type-1 diabetes ⁷. In a recent study conducted in young people with type-1 diabetes, we have documented that intensive insulin therapy can result in improvement of markers of microvascular dysfunction including levels of E-selectin and microvascular reactivity assessed by laser Doppler flowmetry ⁹⁴. Three groups of children and adolescents were allocated either to conventional insulin therapy (CIT) or to CIT, but with “sweet talk” and a third group allocated to intensive insulin therapy (IIT; either by basal bolus regime or insulin pump) together with “sweet talk” the three groups had their microvascular responses, markers of EC activation, and conventional risk factors of vascular disease measured before and after the study that lasted for 1 year. Microvascular responses were assessed by laser Doppler flowmetry. The group with IIT and sweet talk achieved a significantly better control of diabetes than the other two groups, and that was accompanied by significant improvement of the biophysical and biochemical markers of endothelial dysfunction. However, there was no improvement in the conventional risk factors of vascular disease (including total cholesterol, LDL, TG, and HDL). We were able to conclude from this study that intensive insulin therapy is capable of achieving improvement in the perturbed endothelial function in young people with type-1 diabetes by additional means. Earlier, Troseid et al ⁹⁵ reported improved levels of E-selectin with improved levels of glycemic control and reduction of weight in subjects with metabolic syndrome.

Life style measures

Exercise and life style measures are well known to be beneficial in improving vascular disease. There is strong evidence to suggest that exercise in children and young people can typically mimic that of the adults. We have shown before that obesity in normal children has a modulatory role on the microvascular responses ⁶⁶. Physical activity has been shown to have a beneficial effect on FMD in normal children ⁹⁶. Young people with obesity with a mean age of 14 years were found to have a reduced FMD values of 50% less than their normal weight counterparts ⁹⁷. In the same obese group, aerobic exercise for 8 weeks brought their FMD to values nearer to their normal weight control group despite no significant change in weight. Interestingly, the observed improvement was lost following abstinence from exercise suggesting a direct causal effect. In the study by Roche et al ⁹⁸ in youth with type-1 diabetes, they documented significant improvement in microvascular reactivity following aerobic exercises. On the other hand, Burns and Arslanian ⁹⁹ were able to show correlation between ICAM-1 and E-selectin with waist circumference in normal young people of multiethnic origin. In the same study, children on the highest percentile for waist circumference also had an atherogenic lipoprotein profile.

Lipid lowering agents

There is some preliminary evidence to suggest that the lipid lowering agents, the statins, may have beneficial effects on endothelial dysfunction. Dogra et al ¹⁰⁰, studied 16 type-1 diabetic patients who had microalbuminuria and they were able to show that FMD and endothelium-dependent responses of the brachial artery were better following the use of atorvastatin for 6 weeks. We have also shown this in a group of subjects with peripheral arterial disease ¹⁰¹. However, it would not be possible to extrapolate these findings to young patients with diabetes as the use of lipid lowering agents is very limited.

Antioxidants and multivitamins

Earlier, Cerellio et al ⁶⁷. Further, in youngsters with type-1 diabetes there is some preliminary evidence to suggest that folate and vitamin B6 supplements alone or in combination can result in improvement of impaired FMD of the brachial artery ¹⁰³. In this paper, such effect has been shown to occur within 2 hours and be sustained for up to 8 weeks. The effect of folate supplementation appears to be more beneficial to children with diabetes than to obese children without diabetes. Wiltshire et al ⁵⁶ found that the abnormalities of FMD of the brachial artery in young people with type 1 diabetes is related to their folate status, those with a better folate status had better FMD responses—lower TM. They were able to hypothesize that folate may therefore protect against endothelial dysfunction in young people with diabetes. However, on the other hand, Pena et al ¹⁰⁴ failed to show any beneficial effect following folate supplements in obese nondiabetic children. It is plausible that folate supplements may confer better protection in diabetic rather than nondiabetic subjects. However, no clinical trial has shown a sustained improvement with such treatments and so the evidence in support is lacking. In animal models with hypercholesterolemia, antioxidants were shown to improve NO induced relaxation ¹⁰⁵.

Other agents

Stemming from their preliminary effect on improvement of endothelial dysfunction in patients with CAD ¹⁰⁶, ACE inhibitors were postulated to be useful in improving ED in patients with diabetes. The effect of enalapril on FMD was examined in a group of young adults with type-1 (age range 18-44 years), who were free from any evidence of clinical vascular disease ¹⁰⁷. The investigators failed to show any effect on either endothelium-dependent or endothelium-independent vascular responses. Breast feeding has been recently shown by our group to confer beneficial effects on microvascular function in children ¹⁰⁸. In this study, 11-14 year children who were breast fed showed better skin microvascular responses by iontophoresis of acetylcholine versus those who were fed by infant milk formula. The children who were breast fed had a lower systolic blood pressure than those who were not breast fed, although this difference in blood pressure was not associated with an effect on vascular responses. The role of infant feeding on cardiovascular function has been investigated previously, looking mainly at the effect on later blood pressure. Studies show that breastfeeding is associated with a lowering of blood pressure at the age of 13 to 16 years in children born prematurely ¹⁰⁹ and at 6-9 years in children born at term ^{110, 111}, while the consumption of formula milk is associated with an increase in blood pressure at the age of 23-27 years ¹¹². The exact mechanisms by which breast feeding augments endothelial function are not known. A possible effect may be related to long-chain polyunsaturated fatty acids (LCPUFAs) that are present in breast milk. The endothelial system is membrane-rich and uses LCPUFAs for its structural and functional integrity (to synthesize prostaglandins, for example), and a lack of LCPUFAs at critical periods of development might therefore lead to imbalances in endothelial function. We have shown previously that LCPUFA-supplemented formula has a beneficial effect on blood pressure in 6-year-old children ¹¹³ and we have also demonstrated that dietary supplementation with eicosapentaenoic acid and docosahexaenoic acid augments ACh-mediated vasodilatation in normal adults ¹¹⁴. Thus, it is possible that some of the beneficial effects of breast feeding on endothelial function are mediated by LCPUFAs.

CONCLUSION

There is an accumulating body of evidence to suggest that the vascular endothelium in young people with diabetes is perturbed well in advance of any clinical vascular disease. The emergence of type-2 diabetes and insulin resistance stemming from the accompanying obesity epidemic among young people will have an added impact to the one already posed by type-1 diabetes in young people. The huge implications on the risk of vascular disease in the growing young populations have to be addressed by policy makers. The current dilemma of how best to tackle this new menace amongst young people remains a challenging task. On the other hand, the current available evidence shows there are some promising indications that endothelial dysfunction in young people with diabetes or insulin resistance is amenable to intervention either via tightening the metabolic control or addressing the various conventional risk factors of vascular disease. In respect of the latter, efforts will be hampered by the prescribing limitations in young patients.

Keywords

endothelial dysfunction, diabetes mellitus, adolescent diabetes, atherosclerosis, oxidative stress, nitric oxide, free radical markers, intima media thickness

Disclosure: The authors declare no conflict of interest.

Funding: None

Acknowledgements: Authors would like to acknowledge the help of administrative and nursing staff of Pediatric Clinic, Ninewells Hospital and the medical and scientific staff of Vascular & Inflammatory Diseases Research Unit, Ninewells Hospital & Medical School Dundee.

REFERENCES

1. Wild S, Roglic G, Green A, Sicree R, King H. Diabetes pandemic. Global prevalence of diabetes: estimates for the year 2000 and projections for 2030. Diabetes Care. 2004;5:1047–1053.
2. Roglic G, Unwin N, Bennett PH, et al. The burden of mortality attributable to diabetes: realistic estimates for the year 2000. Diabetes Care. 2005;9:2320–2321.
3. Pinhas-Hamiel O, Zeitler P. The global spread of type-2 diabetes mellitus in children and adolescents. J Pediatr. 2005;146:693–700.
4. EURODIAB ACE Study Group. Variations and trends in incidence of childhood diabetes in Europe. Lancet. 2000;355:873–876.
5. Vaughan TB, Ovalle F, Moreland E. Vascular disease in pediatric type-2 diabetes: the state of the art. Diab Vasc Dis Res. 2007;4(4):297–304.
6. Pirart J. Diabetes mellitus and its degenerative complications: a prospective study of 4,400 patients observed between 1947 and 1973 [author’s translation]. Diabete Metab. 1977;2:97–107.
7. The Diabetes Control and Complication Trial Research Group. The effect of intensive insulin treatment of diabetes in the development and progression of long-term complications of insulin dependent diabetes mellitus. N Eng J Med. 1993;29:977–986.
8. UK Prospective Diabetes Study (UKPDS). Intensive blood glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type-2 diabetes (UKPDS 33). Lancet. 1998;352(9131):837–853.
9. Haller H. Endothelial function: general consideration. Drugs. 1997;53(suppl 1):1–10.
10. Furchgott RF. Introduction to EDRF research. J Cardiovascul Pharmacol. 1993;22(suppl 7):S1–S2.
11. Cacoub P, Carayon A, Dorent R, et al. Endothelin: the vasoconstrictor of the 1990’s? Rev Med Interne. 1990;14:229–232.
12. Goldin E, Casadevall M, Mourelle M, et al. Role of prostaglandins and nitric oxide in gastrointestinal hyperemia of diabetic rats. Am J Physiol. 1996;270:G684–G690.
13. McFarlane R, McCredie RJ, Bonney MA, et al. Angiotensin converting enzyme inhibition and arterial endothelial function in adults with type-1 diabetes mellitus. Diabet Med. 1999;16:62–66.
14. Cohen RA. The role of nitric oxide and other endothelium-derived vasoactive substances in vascular disease. Prog Cardiovasc Dis. 1995;38(2):105–128.
15. Emeis JJ. The vascular wall and fibrinolysis. Haemostasis. 1979;8(3–5): 332–339.
16. Chan TK, Chan V. Antithrombin III the major modulator of intravascular coagulation is synthesized by human endothelial cells. Sem Thromb Haemost. 1987;13(4):425–433.
17. Chesterman CN. Vascular endothelium, haemostasis and thrombosis. Blood Review. 1988;2(2):80–94.
18. Sneddon JM, Vane JR. Endothelium-derived relaxing factor reduces platelet adhesion to bovine endothelial cells. Proc Natl Acad Sci USA. 1988;85(8):2800–2804.
19. de Graaf JC, Banga JD, Moncada S, Palmer RM, de Groot PG, Sixma JJ. Nitric oxide functions as inhibitor of platelet adhesion under flow conditions. Circulation. 1992;85(6):2284–2290.
20. Stamler J, Mendelsohn ME, Amarnte P, et al. N-acetylcysteine potentiates platelet inhibition by endothelium derived relaxing factor. Cir Res. 1989;65:789–795.
21. Ignarro LJ, Byrns RE, Buga GM, Wood KS. Endothelium derived relaxing factor from pulmonary artery and vein possesses pharmacologic and chemical properties identical to those of nitric oxide radical. Cir Res. 1987;61:866–879.
22. Freedman JE, Loscalzo J, Barnard MR, Alpert C, Keaney JF, Michelson AD. Nitric oxide released from activated platelets inhibits platelet recruitment. J Clin Invest. 1997;100(2):350–356.
23. Venturini CM, Del Vecchio PJ, Kaplan JE. Thrombin induced platelet adhesion to endothelium is modified by endothelial derived relaxing factor (ERDF). Biochem Biophys Res Commun. 1989;159(1):349–354.
24. Bevilacqua MP, Gimbrone MA Jr. Inducible endothelial function in inflammation and coagulation. Sem Thromb Hemost. 1987;13(4): 425–433.
25. Biegelsen ES, Loscalzo J. Endothelial functions and atherosclerosis. Coron Artery Dis. 1999;10:241–256.
26. Kawashima S. The two faces of endothelial nitric oxide synthase in the pathophysiology of atherosclerosis. Endothelium. 2004;11(2):99–107.
27. Celemajer DS. Testing endothelial function using ultrasound. J Cardiovas Pharmacol. 1998;32(suppl 3):S29–S32.
28. Corretti MC, Anderson TJ, Benjamin EJ, et al. Guidelines for the ultrasound assessment of endothelial dependent flow mediated vasodilation of the brachial artery: a report of the International Brachial ARTERY Reactivity Task Force. J Am Coll Cardiol. 2002;39:257–265.
29. Newton DJ, Khan F, Belch JJF. Assessment of microvascular endothelial function in human skin. Clin Sci. 2001;101:567–572.
30. Khan F, Davidson NC, Littleford RC, Litchfield SJ, Struthers AD, Belch JJF. Cutaneous vascular responses to acetylcholine are mediated by prostacyclin-dependent mechanism in man. Vasc Med. 1997;2:82–86.
31. Khan F, Litchfield SJ, McLaren M, Veale DJ, Littleford RC, Belch JJF. Oral l-arginine supplementation and cutaneous vascular responses in patients with primary Raynaud’s phenomenon. Arthritis Rheum. 1997;40:352–357.
32. Haller MJ, Stein J, Shuster J, et al. Peripheral artery tonometry demonstrates altered endothelial function in children with type-1 diabetes. Pediatr Diab. 2007;8:193–198.
33. Gearing AT, Hemingway I, Piggot R, Hughes J, Rees AJ, Cashman SJ. Soluble forms of vascular adhesion molecules, E-selectin, ICAM-1, & VCAM-1: pathological significance. Ann NY Acad Sci. 1992;667:324–331.
34. Bevilacqua MP, Stenglen S, Gimbrone MA, et al. Endothelial leukocyte adhesion molecule 1: an inducible receptor for neutrophils related to compliment regulatory proteins and lectins. Science. 1989;243:1160–1165.
35. Jones TR, Kao KJ, Pizzo SV, Bigner DD. Endothelial cell surface expression and binding of factor VIII/von Willebrand factor. Am J Pathol. 1981;103(2):304–308.
36. Takano S, Kimora S, Odama S, Aoki N. Plasma thrombomodulin in health and disease. Blood. 1990;76(10):2024–2029.
37. Kaplan KL, Bini A, Fenoglio J Jr, Kudryk B. Fibrin and the vessel wall. Adv Exp Med Biol. 1990;281:313–318.
38. Litwin M, Trelweicz J, Wawer Z, et al. Intima media thickness and arterial elasticity in hypertensive children: controlled study. Pediatr Nephrol. 2004;19:767–774.
39. Riggio S, Mandaraffino G, Sardo MA, et al. Pulse wave velocity and augmentation index, but not intima-media thickness, are early indicators of vascular damage in hypercholesterolemic children. Eur J Clin Invest. 2010;40(3):250–257.
40. Bots ML, Hoes AW, Koudst PJ, Hofman A, Grobbee DE. Common carotid intima media thickness and risk of stroke and myocardial infarction: The Rotterdam Study. Circulation. 1997;96:1432–1437.
41. Abdelgaffar S, Amir M, Hadidi A, El Mougi F. Carotid intima-media thickness: an index for subclinical atherosclerosis in type-1 diabetes. J Trop Pediatr. 2005;52:39–45.
42. Järvisalo MJ, Raitakari M, Toikka JO, et al. Endothelial dysfunction and increased arterial intima-media thickness in children with type-1 diabetes. Circulation. 2004;109(14):17505.
43. Schiel R, Beltschikow W, Radon S, Kramer G, Perenthaler T, Stein G. Increased intima media thickness and association with cardiovascular risk factors in obese and overweight children and adolescents. Eur J Med Res. 2007;12(10):503–508.
44. Gül K, Ustün I, Aydin Y, et al. Carotid intima media thickness and its relations with the complications in patients with type-1 diabetes mellitus. Anadolu Kardiyol Derg. 2010;10(1):52–58.
45. McNally PG, Raymond NT, Swift PG, Hearnshaw JR, Burden AC. Does the prepubertal duration of diabetes influence the onset of microvascular complications? Diabet Med. 1993;10:906–908.
46. Belch JJF, Greene SA, Littleford R, Jennings PE, Khan F. Impaired skin blood flow response to heat in children with insulin dependent diabetes. Int Angiol. 1996;5:189–191.
47. Khan F, Elhadd TA, Greene SA, Belch JJF. Impaired skin microvasular function in children, adolescents and young adults with type-1 diabetes insulin dependent diabetes mellitus. Diabetes Care. 2000;23:215–219.
48. Donaghue KC, Robinson J, McCreadie R, Fung A, Siolinik M, Celermajer DS. Macroangiopathy—does it play a role in young people? Horm Res. 1998;50(suppl 1):38–40.
49. Odermarsky M, Lykkesfeldt J, Liuba P. Poor vitamin C status is associated with increased carotid intima-media thickness, decreased microvascular function, and delayed myocardial repolarization in young patients with type-1 diabetes. Am J Clin Nutr. 2009;90(2):447–452.
50. Heilman K, Zilmer M, Zilmer K, et al. Arterial stiffness, carotid artery intima-media thickness and plasma myeloperoxidase level in children with type-1 diabetes. Diabetes Res Clin Pract. 2009;84(2):168–173.
51. Ladeia AM, Ladeia-Frota C, Pinho L, Stefanelli E, Adan L. Endothelial dysfunction is correlated with microalbuminuria in children with short duration type-1 diabetes. Diabetes Care. 2005;28(8):2048–2050.
52. Elhadd TA, Kennedy G, Hill A, et al. Abnormal markers of endothelial cell activation and oxidative stress in children, adolescents and young adults with type-1 diabetes with no clinical vascular disease. Diabetes Metabol Res Rev. 1999;15:405–411.
53. Romano M, Pomilio M, Vigneri S, et al. Endothelial perturbation in children and adolescents with type-1 diabetes. Diabetes Care. 2001;24(9):1674–1678.
54. Carrizo Tdel R, Prado MM, Velarde MS, Diaz EI, Bazán MC, Abergu˜ AV. Soluble E-selectin in children and adolescents with type-1 diabetes. Medicina (B Aires). 2008;68(3):193–197.
55. McLaren M, Elhadd TA, Greene SA, Belch JJF. Elevated thrombomodulin and vascular endothelial growth factor in juvenile type-1 diabetic patients. Clin Appl Thromb Haemost. 1999;5(1):21–24.
56. Wiltshire EJ, Gent R, Hirte C, Pena A, Thomas DW, Couper JJ. Endothelial dysfunction relates to folate status in children and adolescents with type-1 diabetes. Diabetes. 2002;51:2282–2286.
57. Dimeglio LA, Tosh A, Saha C, et al. Endothelial abnormalities in adolescents with type-1 diabetes: a biomarker for vascular sequelae? J Pediatr. 2010;157(4):540–6.
58. Singh TP, Groehn H, Kazmers A. Vascular function and carotid intimamedia thickness in children with insulin dependent diabetes mellitus. J Am Coll Cardiol. 2003;41(4):661–665.
59. Rodriguez RR, Gomez-Diaz RA, Haj JT, et al. Carotid intima media thickness in pediatric type-1 diabetic patients. Diabetes Care. 2007;30(10):2599–2602.
60. Shah AS, Dolan LM, Kimball TR, et al. Influence of duration of diabetes, glycaemic control, and traditional cardiovascular risk factors on early atherosclerotic vascular changes in adolescents and young adults with type-2 diabetes mellitus. J Clin Endo Metabol. 2009;94(10):3740–3745.
61. Hink U, Li H, Mollnau H, et al. Mechanisms underlying endothelial dysfunction in diabetes mellitus. Circ Res. 2001;88(2):E14–E22.
62. Tesfamariam B, Brown ML, Deykin D, Cohen RA. Elevated glucose promotes generation of endothelium-derived vasoconstrictor prostanoids in rabbit aorta. J Clin Invest. 1990;85:929–932.
63. Tesfamariam B, Brown ML, Cohen RA. Free radicals mediate endothelial cell dysfunction caused by elevated glucose. AM J Physiol. 1992;263 (2 pt 2):H321–H326.
64. Tesfamariam B, Brown ML, Cohen RA. Elevated glucose impairs endothelium-dependent relaxation by activating protein kinase C. J Clin Invest. 1991;87(5):1643–1648.
65. Inoguchi T, Xia P, Kunisaki M, Higashi S, Feener EP, King GL. Insulin effect on protein kinase C and diacylglycerol induced by diabetes and glucose in vascular tissue. Am J Physiol. 1994;267:E369–E379.
66. Khan F, Green FC, Forsyth S, Greene SA, Morris AD, Belch JJF. Impaired microvascular function in normal children: effects of adiposity and poor glucose handling. J Physiol. 2003;705–711.
67. Title LM, Cummings PM, Giddens K, Nassar BA. Oral glucose loading acutely attenuates endothelium-dependent vasodilation in healthy adults without diabetes: an effect prevented by vitamins C and E. J Am Coll Cardiol. 2000;36(7):2185–2191.
68. Vasquez BY, Vazquez MA, Jaquez MG, Huemoeller AH, Intagliatta M, Cabrales P. Blood pressure correlates with blood viscosity in diabetes type-1 children but not in normals. Clin Hemorheol Microcirc. 2010;44(1):55–61.
69. Wolff SP. The potential role of oxidative stress in diabetes and its complications: novel implications for theory and therapy. In: Crabbe MJC, ed. Diabetic Complications. 1st ed. Edinburgh, Churchill Livingstone; 1987:167–220.
70. Suys B, de Beech LO, Rooman R, et al. Impact of oxidative stress on the endothelial dysfunction of children and adolescents with type-1 diabetes: protection by superoxide dismutase. Pediatr Res. 2007;62(4):456–461.
71. Codon¡er-Franch P, Pons-Morales S, Boix-Garcia L, Valls-Bellés V. Oxidant/antioxidant status in obese children compared to pediatric patients with type-1 diabetes mellitus. Pediatr Diabetes. 2010;11(4): 251–257.
72. Twig G, Hyde B, Shirihai OS. Mitochondrial fusion, fission and autophagy as a quality control axis: the bioenergetic view. Biochim Biophys Acta. 2008;1777:1092-1097. [PubMed: 18519024]
73. Nisoli E, Clementi E, Paolucci C, et al. Mitochondrial biogenesis in mammals: the role of endogenous nitric oxide. Science. 2003;299: 896–899. [PMID:12574632]
74. Duchen MR. Roles of mitochondria in health and disease. Diabetes. 2004;53:S96–S102.
75. Goldman S, Zhang Y, Jin S. Autophagy and adipogenesis: implications in obesity and type II diabetes. Autophagy. 2010;6:179–181.
76. Makino A, Scott BT, Dillmann WH. Mitochondrial fragmentation and superoxide anion production in coronary endothelial cells from a mouse model of type-1 diabetes. Diabetologia. 2010;53:1783–1794.
77. Kelley DE, He J, Menshikova EV, Ritov VB. Dysfunction of mitochondria in human skeletal muscles in type-2 diabetes. Diabetes. 2002;51: 2944–2950.
78. Mootha VK, Lindergren CM, Eriksson KF, et al. PGC-1 alpha responsive genes involved in oxidative phosphorylation are co-ordinately down regulated in human diabetes. Nature Genetics. 2003;34:267–273.
79. Kim JA, Wei Y, Sowers JR. Role of mitochondrial dysfunction in insulin resistance. Cir Res. 2008;102:401–414.
80. Pilegaard H, Saltin B, Neufer PD. Exercise induces transient transcriptional activation of the PGC-1alpha gene in human skeletal muscle. J Physiol. 2003;546:851–858.
81. Maahs DM, Wadwa RP, Bishop F, Daniels SR, Rewers M, Klingensmith G. Dyslipidaemia in youth with diabetes: to treat or not to treat. J Pediatr. 2008;153(4):458–465.
82. Kershnar AK, Daniels SR, Imperatore G, et al. Lipid abnormalities are prevalent in youth with type 1 and type 2 diabetes: the SEARCH for Diabetes in Youth Study. J Pediatr. 2006;149(3):314–319.
83. Mahmud FH, Van Uum S, Kanji N, Thiessen-Philbrook H, Clarson CL. Impaired endothelial function in adolescents with type-1 diabetes mellitus. J Pediatr. 2008;152(4):557–562.
84. Pinhas-Hamil O, Zeitler P. Acute and chronic complications of type-2 diabetes mellitus in children and adolescents. Lancet. 2007;369: 1823–1831.
85. McGill HC Jr, McMahan CA, Malcolm GT, Oalman MC, Strong JP. Effects of serum lipoproteins and smoking on atherosclerosis in young men and women. The PDAY Research Group, pathological determinants of atherosclerosis in youth. Arterioscler Thromb Vasc Biol. 1997;17:95–106.
86. Raitakari OT, Juonala M, Kahonen M, et al. Cardiovascular risk factors in childhood and carotid artery intima-media thickness in adulthood: the Cardiovascular Risk in Young Finns Study. JAMA. 2003;290: 2277–2283.
87. Juonala M, Järvisalo MJ, Maki-Torkko N, Kahonen M, Viikari JS, Raitakari OT. Risk factors identified in childhood and decreased carotid artery elasticity in adulthood. The Cardiovascular Risk in Young Finns Study. Circulation. 2005;112:1486–1493.
88. Elhadd TA, Khan F, Kirk G, et al. Influence of puberty on endothelial dysfunction and oxidative stress in young patients with type-1 diabetes mellitus. Diabetes Care. 1998;21:1990–1996.
89. De Fronzo RA, Hendler R, Simonson D. Insulin resistance is a prominent feature of insulin dependent diabetes. Diabetes. 1982;31: 795–801.
90. Arslanian SA, Kalhan SC. Correlation between fatty acids and glucose metabolism: potential explanation of insulin resistance of puberty. Diabetes. 1994;43:908–914.
91. Rogers DG, White NH, Shawitz RA, Palmberg P, Smith ME, Santiago JV. The effect of puberty on development of early diabetic microvascular disease in insulin dependent diabetes. Diabetes Res Clin Pract. 1987;3: 39–44.
92. Dahlquist G, Rudberg S. The prevalence of microalbuminuria in diabetic children and adolescents and its relation to puberty. Acta Paediatr Scand. 1987;76:795–800.
93. Hoffmann MA, Kohl B, Zumbach MS, et al. Hyperhomocysteinaemia and endothelial dysfunction in IDDM. Diabetes Care. 1998;21:841–848.
94. Franklin VL, Khan F, Kennedy G, Belch JJF, Greene SA. Intensive insulin therapy improves endothelial dysfunction and microvascular reactivity in young people with type-1 diabetes. Diabetologia. 2008;51:353–360.
95. Troseid M, Lappegard KT, Mollnes TE, Arnesen H, Seljeflot I. Changes in serum levels of E-selectin correlate to improved glycemic control and reduced obesity in subjects with metabolic syndrome. Scand J Clin Lab Invest. 2005;65(4):283–290.
96. Hopkins N, Stratton G, Tinken T, et al. Seasonal reduction in physical activity and flow mediated dilation in children [published online ahead of print]. Med Sci Sports Exerc. 2010.
97. Watts K, Beye P, Siafirakas A, et al. Exercise training normalizes vascular dysfunction and improves central adiposity in obese adolescents. J Am Coll Cardiol. 2004;43:1823–1827.
98. Roche DM, Edmunds S, Cable T, Didi M, Stratton G. Skin microvascular reactivity in children and adolescents with type-1 diabetes in relation to levels of physical activity and aerobic fitness. Pediatr Exerc Sci. 2008;20(4):426–438.
99. Burns SF, Arslanian SA. Waist circumference, atherogenic lipoproteins, and vascular smooth muscle biomarkers in children. J Clin Endo Metabol. 2009;94(12):4914–4922.
100. Dogra GK, Watts GF, Chan DC, Stanton K. Statin therapy improves brachial artery vasodilator function in patients with type-1 diabetes and microalbuminuria. Diabet Med. 2005;22(3):239–242.
101. Kirk G, McLaren M, Muir A, Stonebridge PA, Belch JJF. P-selectin levels in patients with hypercholesterolemia and peripheral arterial occlusive disease after lipid lowering treatment. Vasc Med. 1999;4(1):23–26.
102. Cerellio A, Curcio F, Colombatti A, Dello Russo P, Pegoraro I, Guigliano D. Hyperglycemia induced endothelial cell alterations are reversed by antioxidants [abstract]. Diabetologia. 1992;35(suppl 1):103A.
103. Mackenzie KE, Wiltshire EJ, Gent R, Hirte C, Piotto L, Couper JJ. Folate and vitamin B6 rapidly normalizes endothelial dysfunction in children with type 1 diabetes. Pediatrics. 2006;118(1):242–253.
104. Pen˜a AS, Wiltshire E, Gent R, Piotto L, Couper JJ. Folic acid does not improve endothelial function in obese children and adolescents. Diabetes Care. 2007;30(8):2122–2127.
105. Adachi T, Matsui R, Xu S, et al. Antioxidants improves smooth muscle sacro/endoplasmic reticulum Ca(2)-ATPase function and lowers tyrosine nitration in hypercholesterolemia and improves nitric oxide induced relaxation. Cir Res. 2002;90(10):1114–1121.
106. Liang W, Tan CY, Ang L, et al. Ramipril improves oxidative stress-related endothelial dysfunction in db/db mice. J Physiol Sci. 2008;58(6):405–411.
107. Mullen MJ, Clarkson P, Donald AE, et al. Effect of enalapril on endothelial function in young insulin dependent diabetic patients: a randomized, double blind study. J Am Coll Cardiol. 1998;31(6):1330–1335.
108. Khan F, Green FC, Forsyth JS, Greene SA, Newton DJ, Belch JJF. The beneficial effects of breastfeeding on microvascular function in 11-14 year old children. Vasc Med. 2009;14(2):137–142.
109. Singhal A, Cole TJ, Lucas A. Early nutrition in preterm infants and later blood pressure: two cohorts after randomised trials. Lancet. 2001;357:413–419.
110. Wilson AC, Forsyth JS, Greene SA, Irvine L, Hau C, Howie PW. Relation of infant diet to childhood health: seven year follow up of cohort of children in Dundee infant feeding study. Br Med J. 1998;316:21–25.
111. Martin RM, Ness AR, Gunnell D, Emmett P, Davey Smith G. for the ALSPAC Study Team. Does breast feeding in infancy lower blood pressure in childhood? The Avon Longitudinal Study of Parents and Children (ALSPAC). Circulation. 2004;109:1259–1266.
112. Martin RM, McCarthy A, Smith GD, Davies DP, Ben-Shlomo Y. Infant nutrition and blood pressure in early adulthood: the Barry Caerphilly Growth Study. Am J Clin Nutr. 2003;77:1489–1497.
113. Forsyth JS, Willatts P, Agostoni C, Bissenden J, Casaer P, Boehm G. Long chain polyunsaturated fatty acid supplementation in infant formula and blood pressure in later childhood: follow up of a randomised controlled trial. Br Med J. 2003;326:953–957.
114. Khan F, Elherik K, Bolton-Smith C, et al. The effects of dietary fatty acid supplementation on endothelial function and vascular tone in healthy subjects. Cardiovasc Res. 2003;59:955–962.

Severe obstetric hemorrhage is the most feared obstetric emergency that can occur to any woman at childbirth. If unattended, the hemorrhage can kill even a healthy woman within two hours ^{1, 2}. The Hemorrhage accounts for nearly one-quarter of all maternal deaths, and for almost half of all postpartum deaths in low-income countries ^2–4. Recent studies in high resource settings ⁵ including Canada ⁶, the USA ⁷, Australia ⁸, and the UK ⁹ indicate an unexpected and unexplained increase in obstetric hemorrhage over the last 10 years. Although maternal deaths are extremely rare in high resource settings, the morbidity associated with a severe hemorrhage is still a major problem ^9–13. Many more women survive but suffer serious illness as a result, not only from the effects of acute hypoperfusion and anemia, but also from the interventions that a severe hemorrhage may necessitate ^14–16. Severe obstetric hemorrhage is, however, the most preventable complication. Nonetheless, suboptimal obstetric care was identified in more than half the deaths due to hemorrhage ^{17, 18}. It is therefore suggested as a complementary indicator for the assessment of the quality of obstetric care ^{19, 20}. The aim of this article is to review the incidence, causes, and risk factors of severe obstetric hemorrhage.

DEFINING SEVERE OBSTETRIC HEMORRHAGE

Obstetric hemorrhage refers to excessive blood loss from the genital tract, occurring antepartum, intrapartum, or in the postpartum period. The most common type of obstetric hemorrhage is postpartum hemorrhage (PPH), mainly primary PPH, occurring within 24 h postpartum. Primary PPH is the focus of this article. Secondary PPH is less common, occurring between 24 h and 6 weeks postpartum, most likely due to infection secondary to retained placental products ²¹.

Any review of obstetric hemorrhage is complicated by the lack of agreement on what constitutes excessive blood loss. Primary PPH is defined according to WHO (World Health Organization) as blood loss >500 ml in the first 24 h postpartum ²². This is debatable, because nearly half of all women who deliver vaginally shed that amount of blood, or more, when measured objectively. Blood loss of more than 500 ml is not, therefore, necessarily unusual for vaginal delivery ²³. Furthermore, women with a low body mass index usually have low blood volume, and women who are anemic or having severe preeclampsia, might have fewer physiological reserves with which to withstand blood loss as low as 500 ml or less ^{24, 25}. Other proposed definitions of hemorrhage include a 10% decrease in hemoglobin or hematocrit level or the need for a blood transfusion ²⁶. Given the delay in obtaining laboratory values, this information would not reflect the patient's current hemodynamic status. The change in hematocrit depends on the timing of the test and the amount of fluid previously administered ²⁷. It could also be affected by extraneous factors such as prepartum hemoconcentration ²⁵. Any definition based on the need for blood transfusion may reflect differences in provider practice patterns rather than patient clinical status ²⁸.

Obstetric hemorrhage may best be defined as excessive bleeding that makes the patient symptomatic (lightheadedness, syncope) and/or results in signs of hypovolemia (hypotension, tachycardia, or oliguria) ²⁹ ( Table 1 ). Classifying hemorrhage according to severity is another problem. Some researchers have used a strict definition including only women admitted to an intensive care unit (ICU) or having a hysterectomy ^30–32. This, however, underestimates the real proportion of cases due to differences in management protocols. It was shown that only one-third of the cases of severe morbidity are transferred to the ICU ³³. Even those who used clinical definitions based on the amount of blood loss have used different limits, varying from ≥1000 ml ³⁴ to ≥2500 ml ^{9, 33}. Blood loss >1500 ml represents 25% of blood volume, and this amount would lead to hemodynamic decompensation ³⁵. It has been suggested that the definition should take into account any blood loss that causes a major physiological change that threatens a mother's life ²⁹. This is especially important in cases of concealed intra-abdominal bleeding.

ESTIMATION OF BLOOD LOSS

Visual estimation is the most universal method used to assess blood loss at delivery. It is relatively straightforward and requires no expenditure ³⁶. The major advantage of this method is that it is a real-time assessment. It enables the birth attendant to correlate findings, on an individualized basis, with the clinical presentation. However, visual estimation of blood loss is known to underestimate the actual loss by 30%-50% ^{23, 37}. Standardized visual estimation is an attempt to rectify this error, based on training of providers and standardization of the size and quality of the pads used during delivery. Instruction in this method has significantly reduced the error in blood loss estimation for inexperienced as well as experienced clinicians ³⁸.

The acid hematin method and measurement of tagged erythrocytes were referred to in earlier studies ²³. However, they were not used in practice as they require larger resources and consist of several impractical procedures. The BRASS-V Drape is a special drape that has a calibrated and funneled collecting pouch, incorporated within a plastic sheet that is placed under the woman's buttocks, immediately after delivery of the baby. This simple, practical tool has the potential for a more accurate detection of blood loss and would lead to earlier interventions contributing to the reduction of both mortality and severe morbidity ³⁹. However, the use of this method would take longer time to apply in practice universally, as compared with a visual estimation.

THE PROPORTION OF SEVERE OBSTETRIC HEMORRHAGE

The proportion of severe obstetric hemorrhage is influenced by the study design and population characteristics as well as obstetric management. A systematic review of international studies covering the period 1997-2006 found that the percentage of PPH (blood loss >500 ml) and of severe PPH (blood loss >1000 ml) were 6% and 1.86%, respectively ⁴⁰. The incidence of severe PPH was 1.67% and 2.95%, in population-based and institution-based studies, respectively. Severe PPH was 2.94% for vaginal deliveries and 6.38% for cesarean section. The incidence across global regions was 2.21% in Africa, 1.78% in Asia, 1.75% in Europe, 5.33% in Latin America, and 4.33% in Oceania. The incidence of severe hemorrhage varies considerably, even among countries with high resources ( Table 2 ).This can be partly due to the use of different definitions as well as differences in registration methods and management aspects. However, other factors such as different population characteristics may also contribute to such variations. The MOthers Mortality and Severe Morbidity (MOMS) Survey ¹¹ was conducted during the 1990s by an international team spanning specific regions in 11 European countries during 1-year. Using unified clinical definitions of severe obstetric hemorrhage (blood loss ≥1500 ml, blood/plasma expanders transfusion, or death from 24 weeks gestations), the survey found a total incidence of severe hemorrhage of 4.6/1000 deliveries. However the incidences varied widely from 0.7/1000 in Austria to 8.8/1000 in Finland. Demographic and genetic profiles vary as certain populations have increased immigrants, increased hereditary coagulation disorders, or increased severe preeclampsia predisposing to increased severe obstetric hemorrhage.

According to a population-based study in Norway ⁴¹, severe obstetric hemorrhage occurred in 11.38/1000 of 307 415 mothers ( Table 2 ). This was relatively higher than earlier studies, partly due to a larger denominator of gestations after 16 weeks, in addition to including cases of placenta previa and abruption in the Norwegian study. However, there are several possible factors underlying this relatively higher proportion. The study from Norway might have had a better case ascertainment as it used data of the total population and not only selected regions over a limited time period. Norway participated in the MOMS study, with data from two hospitals in Oslo. Mothers referred to these hospitals from other areas were excluded. This might have contributed to the low reported number of severe hemorrhage (2.7/1000) ¹¹. Eggebø and Gjessing ^{42, 43} performed a hospital-based study in Stavanger University Hospital during a 3-year period and reported a higher proportion of severe hemorrhage (8.5/1000). This might be due to a larger representative sample or the restricted use of prophylactic oxytocin in the third stage of labor in a limited period covered by this study.

The second possibility is that Norway has, in fact, a higher proportion of severe hemorrhage cases than other countries. However, the CS rate of 17.1% in Norway ⁴⁴ is lower than in other countries with similar resources, such as 30.8% in the USA ⁴⁵, and 24% in the UK ⁴⁶. The global increase in mean maternal age, maternal obesity, induction rates, and multiple pregnancies is seen in Norway as in other countries ^{44, 47, 48}. We cannot preclude, however, the possibility of a genetic predisposition to hemorrhage in the Norwegian mothers.

Inadequate management of the third stage might contribute to the high proportion of severe hemorrhage. Active management of the third stage of labor as recommended by International Federation of Gynaecology and Obstetrics (FIGO) involves the use of prophylactic uterotonics such as syntocinon or syntometrine with the delivery of the fetal anterior shoulder, clamping of the umbilical cord once pulsations stopped, and controlled cord traction using the Brandt-Andrews technique once uterine contraction is achieved ⁴⁹. In contrast, expectant management involves waiting for spontaneous separation of the placenta from the uterine wall and avoidance of synthetic uterotonics. A meta-analysis indicated that active management of the third stage resulted in reduction in maternal blood loss and a reduction in the risks of PPH ⁵⁰. According to a European survey ⁵¹, only 11% of responding units in Norway used active management of the third stage as FIGO recommended ⁴⁹. However, the survey showed that only 3%-20% of units in Austria, Denmark, Finland, France, Hungary, Italy, and Portugal actively managed the third stage. The third possibility lies in an overestimation due to misclassification in registration. This is highly unlikely as such random misclassification would result in underestimation rather than overestimation ⁵².

CAUSES OF SEVERE OBSTETRIC HEMORRHAGE

The most common causes of obstetric hemorrhage are those related to primary PPH (the Four Ts: Tone, Trauma, Tissue, and Thrombin) ⁵³. Uterine atony (Tone) accounts for more than 70% of cases; retained placental products (Tissue) accounts for approximately 10%; genital-tract injuries as uterine rupture and inversion, cervical, and perineal injuries (Trauma) accounts for 20%; and preexistent or acquired coagulation disorders and platelets dysfunction (Thrombin) account for 1% of cases ⁵³. Ante/intrapartum hemorrhage was reported to occur in about 3%-4% of pregnant population ⁴⁴, of which 30% was due to placental abruption, and 20% was due to placenta previa. Both are associated with increased risk for postpartum hemorrhage ⁵⁴. Placenta abruption is associated with increased risk for disseminated intravascular coagulation (DIC), resulting in the high case fatality rate. Placenta previa may increase atonic PPH from the placental implantation site due to a weak contractile lower segment ⁵⁴.

Placenta previa may, moreover, be associated with abnormal adherent placenta (placenta accrete/increta or percreta), especially in the presence of a uterine scar. Uterine atony may result in retained placentas and the latter may result in uterine atony through deficient contractility of focal areas of myometrium. Uterine atony is poor contractility of the uterus, resulting from uterine overdistension as in macrosomia or multiple pregnancies or from uterine exhaustion as in prolonged labor.

Identifying exact causes of severe hemorrhage may be challenging in the presence of multiple causes or clinically unrecognized and undocumented causes. Although uterine atony is the most common cause of postpartum hemorrhage, it was only reported in 30% and 48% of severe PPH cases in the Norwegian study ⁴¹ and a Scottish audit, respectively ⁹. Uterine atony accounted for 79% of PPH in the United States ⁵⁵.Moreover, a large percentage of severe PPH had no identified causes, either due to lack of clinical recognition or documentation. Most of the cases with no identified causes were deliveries by CS in Norway ⁵⁶, indicating inaccurate documentation of exact causes of severe PPH at CS. The cesarean section is associated with a higher risk for uterine atony and surgical bleeding at hysteretomy sites. Surgical bleeding is usually underreported due to the absence of specific diagnostic causes. In addition, placenta accreta, highly associated with placenta previa, has no specific international diagnostic code.

RISK FACTORS OF SEVERE OBSTETRIC HEMORRHAGE

Risk factors should be studied even though up to two-thirds of cases had no identifiable risk factors ⁵⁵. These included demographic, medical, pregnancy, and labor and delivery variables predisposing to, or exaggerating, the main causes and mechanisms of severe obstetric hemorrhage ( Figure 1 ). The complex interrelation between different risk factors, as, for example, the association between increasing maternal age and delivery by CS or increased medical diseases, is important to remember when determining their independent contribution to severe hemorrhage risk ( Figure 2 ).

Demographic factors

As maternal age increases, the risk for severe PPH is reported to increase in several studies ^{10, 41, 57}. The increase is highest when mothers are >40 years old ⁴¹. Increased dysfunctional labor with older age is one of the suggested underlying mechanisms. However older mothers have also increased risk for medical diseases, placenta previa, and abruption and uterine rupture ^{58, 59}. According to recent studies, primiparas have a higher risk for severe PPH due to increased risk for prolonged labor (uterine exhaustion), operative vaginal deliveries, and perineal traumas ^{41, 56, 57, 60}. Immigrant Asian women, and especially those from South East Asia, had the highest risk for severe PPH ^{34, 41}.

Previous obstetric history

Previous obstetric hemorrhage is a significant risk factor for severe PPH. An Australian population-based study (125 295 mothers), found that women with PPH in their first pregnancy have a threefold increased risk of recurring PPH in the second pregnancy ⁶¹. Having a previous CS is shown in several studies to increase severe hemorrhage and peripartum hysterectomy risk ^{41, 56, 62}.

Medical and pregnancy factors

Cardiac disease was shown to be associated with increased risk for severe PPH ^{41, 63}. Clinical studies are warranted to find whether the use of anticoagulation or inadequate prophylaxis with oxytocin in the third stage of labor contributes to this risk of hemorrhage.

Von Willebrand's disease, the most common hereditary blood disorder with a prevalence of 0.6% to 1.3% ⁶⁴, needs special attention. These mothers had 50% higher risk for PPH and five times the increased risk for blood transfusion than women without a bleeding disorder in the United States ⁶⁵ and fourfold increased risk for severe obstetric hemorrhage in Norway ⁴¹. Von Willebrand's disease is an autosomally inherited congenital bleeding disorder involving a qualitative or quantitative deficiency of von Willebrand factor (vWF). Dominant and recessive patterns for transmission exist. The most common presenting symptom is menorrhagia (84%). Of women with menorrhagia, 5%-20% have been found to have previously undiagnosed vWB disease ⁶⁴. It is important for obstetricians and gynecologists to be aware of this problem, as these women have a higher risk of primary PPH, perineal hematomas, and especially secondary PPH.

The significant increase of severe hemorrhage by HELLP syndrome suggests coagulopathy as the underlying cause of hemorrhage ^{41, 54, 66}. Anemia during pregnancy is a significant risk factor, with the greatest impact in low resource settings ^{24, 41, 66}. Multiple pregnancy, polyhydramnios and macrosomia were shown to increase severe atonic PPH through uterine overdistension and exhaustion ^{10, 34, 41–43, 54, 55, 66–68}. The rate of multiple pregnancy is rising due to increased assisted reproduction in growing numbers of older women seeking pregnancy. Macrosomia, also associated with increased genital injuries, is increasing partly as a result of increasing obesity and sedentary lifestyle.

Labor and delivery factors

Induced and prolonged labor are significant risk factors for severe hemorrhage, even when controlled for relevant confounders ^{34, 41, 56, 66, 68}. Chorioamnionitis has been repeatedly shown to result in a poor contractile uterus, likely in part due to inflammation ^{55, 66, 68}.

Delivery by emergency CS carries the highest risk for severe obstetric hemorrhage ^{7, 10, 14},^{41, 43, 55, 56, 66, 68}. Uterine atony and surgical bleeding are expected to be the highest at an emergency CS, especially in late labor. A previous Norwegian population-based cohort study found that an unexpectedly high number of cesareans were performed in the late stages of labor ⁶⁹. Moreover, a UK study showed that 10% of emergency peripartum hysterectomies were performed after a cesarean section at full dilation due to failed progress in labor or failed delivery using instruments ¹⁴. This emphasizes the importance of performing emergency cesareans at the correct time and for the correct indications. Recent studies have shown that CS, even if planned, was associated with severe postpartum hemorrhage versus vaginal deliveries ^{41, 55, 56, 66, 68, 70}. Moreover, prelabor CS was found to have a higher risk for severe PPH versus spontaneous labor in mothers with or without previous CS ⁵⁶.

Screening for underlying medical causes of severe obstetric hemorrhage

Women who have had an unexplained massive obstetric hemorrhage should be investigated thoroughly for preexistent or acquired bleeding disorders. Inherited bleeding disorders should be considered especially in those with menorrhagia. Von Willebrand's disease is the most frequent, followed by a mild coagulation factor deficiency, as prothrombin, fibrinogen, and factors V, VII, X, and XI, mild platelets disorder, and carriers of hemophilia A and B ^{64, 65, 71}. In cooperation with the hematologist, the following tests should be performed: hemoglobin/hematocrit, platelet count, ferritin, PT (INR), and APTT, vWD workup (factor VIII, vWF antigen, and vWF functional assay), and eventually factor VII, factor IX, factor XI, factor XIII, factor α2-antiplasmin, bleeding time, and specific platelet functional assay ⁷¹. Once diagnosed, all efforts should be made to reduce menorrhagia in order to prevent anemia during pregnancy. These women should have genetic counseling prior to pregnancy and should have conjoined management of the hematologist, obstetrician, and an anesthesiologist during pregnancy and labor as well as in the postpartum period.

Immune thrombocytopenia is diagnosed by positive antibodies against platelets surface antigens provided that other causes of thrombocytopenia have been excluded ⁷². Severe PPH lasting for several weeks may be due to an acquired autoimmune coagulopathy during pregnancy such as the rare acquired hemophilia (spontaneous development of autoantibodies against factor VIII coagulant protein). This is diagnosed by the positive factor VIII inhibitors ⁷³. Other autoimmune coagulations include very rare lupus anticoagulant-hypoprothrombinemia, resulting in acquired prothrombin deficiency and life-threatening PPH, or other disorders with inhibitors for factors V, VII, IX, XI, or XII ⁷². Women with unexplained severe PPH should be screened for infection, drugs, liver and renal disease, or connective tissue diseases such as systemic lupus erythemetosus (SLE) or antiphospholipid syndrome that may result in autoimmune coagulation disorder, thrombocytopenia, or DIC ⁷².

CONCLUSION

Severe obstetric hemorrhage is a relatively frequent complication associated with serious maternal outcome. This necessitates performing larger population-based studies in different resource settings across the world and regular audits in order to review labor ward management and protocols. Several challenges are found in epidemiological studies of severe obstetric hemorrhage due to the lack of unified definition, different population characteristics, and management aspects as well as the inaccuracy in both estimation of blood loss and documentation of data. The frequency and impact of severe hemorrhage can be effectively reduced by reducing avoidable risk factors, especially those related to obstetric interventions as increased CS rate and induction of labor. Other risk factors not amenable to change such as age, ethnic origin, and preexisting medical diseases or bleeding disorders can be minimized by extra vigilance and planned conjoined management. Increasing our knowledge of severe obstetric hemorrhage would contribute toward decreasing severe maternal morbidity and mortality.

Keywords

severe obstetric hemorrhage, risk factors, causes, coagulopathy, uterine atony

Disclosure: The authors declare no conflict of interest.

Acknowledgment: Special thanks to Pernille Frese, from the National Resource Centre for Women's Health in Rikshospitalet for her help in drawing the figure provided. We would like to thank the Norwegian Foundation for Health and Rehabilitation and the Norwegian Women's Public Health Association for funding part of the research in this article.

REFERENCES

1. Chong YS, Su LL, Arulkumaran S. Current strategies for the prevention of postpartum haemorrhage in the third stage of labour. Curr Opin Obstet Gynecol. 2004;16:143–150.
2. Abou Zahr C. Global burden of maternal death and disability. Br Med Bull. 2003;67:1–11.
3. Khan KS, Wojdyla D, Say L, Gülmezoglu AM, Van Look PF. WHO analysis of causes of maternal death: a systematic review. Lancet. 2006;367:1066–1074.
4. Tsu VD, Langer A, Aldrich T. Postpartum hemorrhage in developing countries: is the public health community using the right tools? Int J Gynaecol Obstet. 2004;85:S42–S51.
5. Knight M, Callaghan WM, Berg C, et al. Trends in postpartum haemorrhage in high resource countries: a review and recommendations from the International Postpartum Haemorrhage Collaborative Group. BMC Pregnancy and Childbirth. 2009;9:1–10.
6. Joseph KS, Rouleau J, Kramer MS, et al. Investigation of an increase in postpartum haemorrhage in Canada. BJOG. 2007;114:751–759.
7. Berg CJ, Mackay AP, Qin C, Callaghan WM. Overview of maternal morbidity during hospitalisation for labor and delivery in the United States: 1993–1997 and 2001–2005. Obstet Gynecol. 2009;113:1075–1081.
8. Ford JB, Roberts CL, Simpson JM, Vaughan J, Cameron CA. Increased postpartum haemorrhage rates in Australia. Int J Gynaecol Obstet. 2007;98:237–243.
9. Brace V, Kernaghan D, Penney G. Learning from adverse clinical outcomes: major obstetric haemorrhage in Scotland, 2003–2005. BJOG. 2007;114:1388–1396.
10. Waterstone M, Bewley S, Wolfe C. Incidence and predictors of severe obstetric morbidity: case-control study. BMJ. 2001;322:1089–1093.
11. Zhang WH, Alexander S, Bouvier-Colle MH, Macfarlane A; MOMS-B Group. Incidence of severe pre-eclampsia, postpartum haemorrhage and sepsis as a surrogate marker for severe maternal morbidity in a European population-based study: the MOMS-B Survey. BJOG. 2005;112:89–96.
12. Baskett TF, O’Connell CM. Severe obstetric maternal morbidity: a 15-year population-based study. J Obstet Gynaecol. 2005;25:7–9.
13. Wen SW, Huang L, Liston R, et al. Severe maternal morbidity in Canada, 1991–2001. CMAJ. 2005;173:759–764.
14. Knight M; UKOSS. Peripartum hysterectomy in the UK: management and outcomes of the associated haemorrhage. BJOG. 2007;114:1380–1387.
15. Smith J, Mousa HA. Peripartum hysterectomy for primary postpartum haemorrhage: incidence and maternal morbidity. J Obstet Gynaecol. 2007;27:44–47.
16. Walvekar V, Virkud A. Familial consequences. In: B-Lynch C, Keith LG, Lalonde AB and Karoshi M, eds. A Text Book of Postpartum Haemorrhage. Duncow, UK: Sapiens Publishing; 2006:372–375.
17. Liston W. Haemorrhage. In: Lewis G, ed. Saving Mother’s Lives: Reviewing Maternal Deaths to Make Motherhood Safer 2003-2005. The Seventh Report of the Confidential Enquiries into Maternal Deaths in the UK. London: CEMACH; 2007:78–85.
18. Berg CJ, Harper MA, Atkinson SM, et al. Preventability of pregnancyrelated death. Results of a State-wide review. Obstet Gynecol. 2005;106:1228–1234.
19. Pattinson RC, Hall M. Near misses: a useful adjunct to maternal death enquiries. Br Med Bull. 2003;67:231–243.
20. Stones W, Lim W, Al-Azzawi F, Kelly M. An investigation of maternal morbidity with the identification of life-threatening ‘near miss’ episodes. Health Trends. 1991;23:13–15.
21. Collins S, Arulkumaran S, Hayes K, Jackson S, Impy L, eds. Oxford Handbook of Obstetrics and Gynaecology. Oxford Handbooks Series. 2nd ed. Oxford: Oxford University Press; 2008.
22. World Health Organisation. The prevention and management of postpartum haemorrhage. Report of a technical working group, Geneva, July 3–6, 1989. Unpublished document. WHO/MCH/90.7. Geneva: World Health Organisation, 1990.
23. Pritchard JA, Baldwin RM, Dickey JC, Wiggins KM. Blood volume changes in pregnancy and the puerperium. II Red blood cell loss and changes in apparent blood volume during and following vaginal delivery, cesarean section, and cesarean plus total hysterectomy. Am J Obstet Gynecol. 1962;84:1271–1282.
24. Lawson JB. Obstetric haemorrhage. In: Lawson JB, Stewart DB, eds. Obstetrics and Gynaecology in the Tropics. London: Edward Arnold; 1967: 155–159.
25. Zeeman GG, Cunningham FG, Pritchard JA. The magnitude of hemoconcentration with eclampsia. Hypertens Pregnancy. 2009;28:127–137.
26. American College of Obstetricians and Gynecologists. Quality Assurance in Obstetrics and Gynecology. Washington, DC: American College of Obstetricians and Gynecologists; 1989.
27. Cunningham FG, Gant NF, Leveno KJ, et al., eds. Conduct of normal labour and delivery. In: Williams Obstetrics. 21st ed. New York, NY: McGraw- Hill; 2001:320–325.
28. Schuurmans N, Mackinnon C, Lane C, Etches D. Prevention and management of postpartum haemorrhage. J Soc Obstet Gynaecol Canada. 2000;22:271–281.
29. Bonnar J. Massive obstetric haemorrhage. Baillieres Best Pract Res Clin Obstet Gynaecol. 2000;14:1–18.
30. Baskett TF, Sternadel J. Maternal intensive care and near-miss mortality in obstetrics. Br J Obstet Gynaecol. 1998;105:981–984.
31. Lapinsky SE, Kruczynski K, Seaward GR, Farine D, Grossman RF. Critical care management of the obstetric patient. Can J Anaesth. 1997;44:325–329.
32. Hazelgrove JF, Price C, Pappachan VJ, et al. Multicenter study of obstetric admissions to 14 intensive care units in Southern England. Crit Care Med. 2001;29:770–775.
33. Brace V, Hall M, Penney G. Quantifying severe maternal morbidities: a Scottish population study. BJOG. 2004;111:481–484.
34. Magann EF, Evans S, Hutchinson M, et al. Postpartum haemorrhage after vaginal birth: an analysis of risk factors. South Med J. 2005;98:419–422.
35. Benedetti T. Obstetric haemorrhage. In: Gabbe SG, Niebyl JR, Simpson JL, eds. A Pocket Companion to Obstetrics. 4th ed. New York, NY: Churchill Livingstone; 2002:217–248 Chap.
17. 36. Maine D. Safe Motherhood Programs: Options and Issues. Columbia University, NY: Center for Population & Family Health; 1993:42:1–60.
37. Glover P. Blood loss at delivery: how accurate is your estimation? Aust J Midwifery. 2003;16:21–24.
38. Luegenbiehl DL. Improving visual estimation of blood volume on peripads. MCN Am J Matern Child Nurs. 1997;22:294–298.
39. Patel A, Goudar SS, Geller SE, et al. Drape estimation versus visual assessment for estimating postpartum haemorrhage. Int J Gynaecol Obstet. 2006;93:220–224.
40. Carroli G, Cuesta C, Abalos E, Gulmezoglu AM. Epidemiology of postpartum haemorrhage: a systematic review. Best Bract Res Clin Obstet Gynaecol. 2008;22:999–1012.
41. Al-Zirqi I, Vangen S, Forsén L, Stray-Pedersen B. Prevalence and risk factors of severe obstetric haemorrhage. BJOG. 2008;115:1265–1272.
42. Eggebø TM, Gjessing LK. Hemorrhage after vaginal delivery [In Norwegian]. Tidsskr Nor Laegeforen. 2000;120(24):2860–2863.
43. Eggebø TM, Gjessing LK. Hemorrhage after caesarean delivery [In Norwegian]. Tidsskr Nor Laegeforen. 2000;120(24):2864–2866.
44. Medical Birth Registry of Norway. Statistic bank. Available at: http://mfrnesstar. uib.no/mfr/. Accessed March 10, 2010.
45. Hamilton BE, Martin JA, Ventura SJ. Births: preliminary data for 2007. National Vital Statistics Report 2009. 57 (12). Available at: http://www.cdc.gov/nchs/data/nvsr/nvsr57/nvsr57_12.pdf. Accessed March 10, 2010.
46. The Health and Social Care Information Centre. NHS maternity statistics, 2008-09. Available at: http://www.ic.nhs.uk/statistics-anddata-collections. Accessed March 10, 2010.
47. Eurostat NewCronos. UNICEF Innocenti Research Center (IRS). The statistical year book of the economic commission for Europe. Available at: http://www.unece.org/stats/trends2005/Sources. Accessed March 10, 2010.
48. International Association of Obesity, London. Global prevalence of adult obesity. Available at: http://www.iaso.org/. Accessed March 10, 2010.
49. International Confederation of Midwives, International Federation of Gynecology and Obstetrics, Society of Obstetricians and Gynaecologists of Canada. Management of the third stage of labour to prevent postpartum haemorrhage. J Obstet Gynaecol Can. 2003;25:952–955.
50. Prendivikke WJ, Elbourne D, McDonald S. Active versus expectant management in the third stage of labour. Cochrane Database Syst Rev. 2000;(2):CD000007.
51. Winter C, Macfarlane A, Deneux-Tharaux C, et al. Variations in policies for management of the third stage of labour and the immediate management of postpartum haemorrhage in Europe. BJOG. 2007;114:845–854.
52. Rothman KJ, Greenland S, eds. . Precision and validity in epidemiological studies. In: Modern Epidemiology. 2nd ed. Philadelphia, PA: Lippincott-Raven Publishers; 1998:115–134.
53. Anderson J, Etches D, Smith D. Postpartum haemorrhage. In: Damos JR, Eisinger SH, eds. Advanced Life Support in Obstetrics (ALSO) Provider Course Manual. Clay Center, KS: American Academy of Family Physicians; 2000:1–15.
54. Stones RW, Paterson CM, Saunders NJ. Risk factors for major obstetric haemorrhage. Eur J Obstet Gynecol Reprod Biol. 1993;48:15–18.
55. Bateman BT, Berman MF, Riley LE, Leffert LR. The epidemiology of postpartum haemorrhage in a large, nationwide sample of deliveries. Anesth Analg. 2010;110(5):1368–1373.
56. Al-Zirqi I, Vangen S, Forsén L, Stray-Pedersen B. Effects of onset of labour and mode of delivery on severe postpartum haemorrhage. Am J Obstet Gynecol. 2009;201:273.e.l–273.e.9.
57. Ohkuchi A, Onagawa T, Usui R, et al. Effect of maternal age on blood loss during parturition: a retrospective multivariate analysis of 10,053 cases. J Perinat Med. 2003;31:209–215.
58. Federiksen MC, Glassenberg R, Stika CS. Placenta previa: a 22 year analysis. Am J Obstet Gynecol. 1999;180:1432–1437.
59. Shipp TD, Zelop C, Repke JT, et al. The association of maternal age and symptomatic uterine rupture during a trial of labor after prior cesarean delivery. Obstet Gynecol. 2002;99:585–588.
60. 60. Malkiel A, Pnina M, Aloni H, Gdansky E, Grisaru-Granovsky S. Primiparity: a traditional intrapartum obstetric risk reconfirmed. Isr Med Assoc J. 2008;10:508–511.
61. Ford JB, Roberts CL, Bell JC, et al. Postpartum haemorrhage occurrence and recurrence: a population-based study. Med J Aust. 2007;187:391–393.
62. Kwee A, Bots ML, Visser GH, Bruinse HW. Emergency peripartum hysterectomy: a prospective study in The Netherlands. Eur J Obstet Gynecol Reprod Biol. 2006;124:187–192.
63. Ouyang DW, Khairy P, Fernandes SM, Landzberg MJ, Economy KE. Obstetric outcomes in pregnant women with congenital heart disease. Int J Cardiol. 2009. Doi: 10.1016/j.ijcard.2009.04.006.
64. The American College of Obstetricians and Gynecologists. ACOG Committee opinion: Von Willebrand disease in women. Obstet Gynecol. 2009;114:1439–1443.
65. James AH, Jamison MG. Bleeding events and other complications during pregnancy and childbirth in women with von Willebrand disease. J Thromb Haemost. 2007;5:1165–1169.
66. Cunningham FG, Leveno KJ, Bloom SL, et al., eds. Obstetrical hemorrhage. In: Williams Obstetrics. 22nd ed. New York, NY: McGraw- Hill; 2005:809–854.
67. Jolly MC, Sebire NJ, Harris JP, Regan L, Robinson S. Risk factors for macrosomia and its clinical consequences: a study of 350,311 pregnancies. Eur J Obstet Gynecol Reprod Biol. 2003;111:9–14.
68. Oyelese Y, Ananth CV. Postpartum haemorrhage: epidemiology, risk factors, and causes. Clin Obstet Gynecol. 2010;53:147–156.
69. Häger RM, Daltveit AK, Hofoss D, et al. Complications of cesarean deliveries: rates and risk factors. Am J Obstet Gynecol. 2004;190:428–434.
70. Pallasmaa N, Ekblad U, Gissler M. Severe maternal morbidity and mode of delivery. Acta Obstet Gynecol Scand. 2008;87:662–668.
71. The Society of Obstetricians and Gynaecologists of Canada. Gynaecological and obstetric management of women with inherited bleeding disorders. Int J Gynaecol Obstet. 2005;95:75–87.
72. Scott-Timperley LJ, Haire WD. Autoimmune coagulation disorders. Rheum Dis Clin N Am. 1997;23:411–423.
73. Kashyap R, Choudhry VP, Mahapatra M, et al. Postpartum acquired haemophilia: clinical recognition and management. Haemophilia. 2001;7:327–330.

1.1 Diabetes—forms, complications, and management

Diabetes mellitus (DM), aglobal public health problem, is now emerging as an epidemic world over. According to a widely accepted estimation, the number of diabetic patients would reach 366 million by the year 2030 ¹. The situation is particularly grim in developing countries like India where unprecedented economic growth has been accompanied with an unfortunate byproduct of that prosperity in the form of diabetes. India now has the world's largest diabetic population, encompassing an estimated 35 million people out of an overall population of 1 billion. Another 79 million people have impaired glucose tolerance (IGT). In just over 20 years (ie, 2025) the country will have almost 200 million people (approximately 15% of the population) affected by diabetes or its precursor ^{2, 3}.

There are mainly two types of diabetes—type 1 and type 2. In type 1 diabetes, in the absence of pancreatic beta cells the hormone insulin is not produced, while type 2 diabetes mellitus (T2DM) is characterized by a progressive impairment of insulin secretion by pancreatic beta cells and by a relative decreased sensitivity of target tissues to the action of this hormone ⁴.

DM is a major worldwide health problem predisposing to markedly increased cardiovascular mortality. Other serious morbidities and mortalities are related to development of nephropathy (kidney damage), neuropathy (nerve damage), and retinopathy (blindness) ^5–7 due to diabetes. Increased oxidative stress has been implicated in the pathogenesis of DM. Hyperglycemia-induced protein glycation generates superoxide-free radicals ^{5, 6, 8, 9}. The generation of active oxygen species may lead to lipid per-oxidation and formation of reactive products that may be involved in severe damage of cell molecules and structures.

As the prevalence of T2DM continues to increase worldwide, there is an enhanced need for effective disease management. T2DM is managed through a stepwise program of intensive therapy that consists of lifestyle modification including appropriate diet and exercise programs and sequential addition of oral antihyperglycemic agents (OHAs) and insulin as required. Improvement in blood glucose control through a combination of lifestyle and oral modifications may decrease the rate of this progression and enhance the quality of life for people with T2DM ¹⁰. About one-third of type 2 diabetic patients are treated with oral hypoglycemic agents to stimulate insulin secretion. These drugs however risk inducing hypoglycemia and, over time, lose their efficacy ⁴. Although, oral hypoglycemic agents/insulin are the mainstay of treatment of diabetes and are effective in controlling hyperglycemia, they have prominent side effects and fail to significantly alter the course of diabetic complications. The common side effects associated with the main classes of drugs used for the treatment of T2DM are hypoglycemia, weight gain, gastrointestinal disorders, peripheral edema, and liver disease ¹¹.

While the pharmacological therapies are in use for management, the diabetes prevention trials in China ¹², Finland ¹³ and USA ¹⁴ remind us that nutrition and lifestyle approaches can be more effective in delaying the onset of this disease.

1.2 Dietary therapy for managing type 2 diabetes—recommendations

Dietary therapy, especially, is showing a bright future in the management of T2DM. With this background, this paper reviews the commonly consumed cereals, which form the important component of a balanced diet and that have been shown to possess antidiabetic properties.

Currently, the ADA recommends the use of diabetes food pyramid for the T2DM patients. The food pyramid divides food into six groups, which vary in size. The largest group—grains, beans, and starchy vegetables is at the bottom. This implies that consuming adequate quantity of these as compared to other foods is beneficial for diabetic patients. The next group is that of the vegetables and fruits followed by milk and meat products. The smallest group—fats, sweets, and alcohol—is at the top of the pyramid. This implies that it is advisable to eat very few servings from these food groups ¹⁵. The use of low-glycemic index (GI) diets (comprising of whole grain cereals and legumes) in the management of diabetes have been recommended around the world ¹⁶.

It is beneficial to incorporate more grains in the diet for effectively managing T2DM. Several studies from the literature corroborate this fact. A study from Peru evaluated the health-relevant functionality of 10 thermally processed Peruvian Andean grains (five cereals, three pseudocereals, and two legumes) for potential T2DM-relevant antihyperglycemia and antihypertension activity using in vitro enzyme assays. Inhibition of enzymes relevant for managing early stages of T2DM such as hyperglycemia-relevant alpha-glucosidase and alpha-amylase and hypertension-relevant angiotensin I-converting enzyme (ACE) were assayed along with the total phenolic content, phenolic profiles, and antioxidant activity based on the 1,1-diphenyl-2-picrylhydrazyl radical assay. Purple corn (Zea mays L.) (cereal) exhibited high free radical scavenging-linked antioxidant activity (77%) and had the highest total phenolic content (8±1mg of gallic acid equivalents/g of sample weight) and alpha-glucosidase inhibitory activity (51% at 5 mg of sample weight). The major phenolic compound in this cereal was protocatechuic acid (287±15 microg/g of sample weight). Pseudocereals such as Quinoa (Chenopodium quinoa Willd) and Kañiwa (Chenopodium pallidicaule Aellen) were rich in quercetin derivatives (1131±56 and 943±35 microg [expressed as quercetin aglycone]/g of sample weight, respectively) and had the highest antioxidant activity (86% and 75%, respectively). Andean legumes (Lupinus mutabilis cultivars SLP-1 and H-6) inhibited significantly the hypertension-relevant ACE (52% at 5 mg of sample weight). No alpha-amylase inhibitory activity was found in any of the evaluated grains. This in vitro study indicates the potential of combination of whole grain cereals, pseudocereals, and legumes to develop effective dietary strategies for managing T2DM and associated hypertension, and provides the rationale for animal and clinical studies ¹⁷.

In addition to the above-mentioned hypertensive action, Lupin is also believed to possess significant antidiabetic activity. Conglutin-gamma, a lupin seed glycoprotein has been credited with insulin-mimetic action. In order to assess its insulin mimetic biological activity, insulin-activated kinases, myoblastic C2C12 cells were incubated with 100 nM insulin or 10 muM conglutin-gamma. It was observed that insulin or conglutin-gamma cell stimulation resulted in the persistent activation of protein synthetic pathway kinases and increased glucose transport, glut4 translocation, and muscle-specific gene transcription regulation. The study proposed the potential therapeutic use of this natural legume protein in the treatment of diabetes and other insulin-resistant conditions ¹⁸.

1.3 Whole grains—types, structure, nutritive value, and role in managing diabetes

The “fibre hypothesis,” published in the early 1970s, suggested that whole foods, such as whole grains, fruits, and vegetables provide dietary fiber along with other components that have health benefits ^{19, 20}. The major cereal grains include wheat, rice, and maize, with oats, rye, barley, triticale, sorghum, and millet as minor grains. Buckwheat, wild rice, and amaranth are not true grains (botanically) but are typically associated with the grain family due to their similar composition.

All grains have a bark-like, protective hull, beneath which lie the endosperm, bran, and the germ. The germ contains the plant embryo. The endosperm supplies food for the growing seedling. Bran is the outer covering that surrounds the germ and the endosperm. It protects the grain from its environment, including the weather, insects, moulds, and bacteria ²⁰.

The American Association of Cereal Chemists has defined a whole-grain ingredient as “…the intact, ground, cracked or flaked caryopsis, whose principal anatomical components, the starchy endosperm, germ and bran, are present in substantially the same relative proportions as they exist in the intact caryopsis.” Thus, for whole-grain ingredients such as flour, the three major components (bran, germ, and endosperm) must be present in the same amounts that occur in the grain's native state. A whole-grain health claim has also been approved in the USA. For a whole grain food to meet the whole-grain health-claim standards, the food must include 51% whole grain flour by weight of final product and must contain 1.7 g dietary fiber ²⁰.

Grain products comprise the base of the US Department of Agriculture's Food Guide Pyramid ²¹, which suggests that several of the recommended six to 11 servings of grain products should be from whole grains. The 2000 Dietary Guidelines for Americans ^{20, 22} established a separate guideline for grains with a particular emphasis on eating more whole grain foods. It is recommended that at least three servings, or one-half of grain foods consumed daily, be whole grains.

Many consumers are unaware of the health benefits of whole grains or of the recommendations regarding increased intake. Also, there is much confusion about which products are truly whole grain. The bran portion of a whole grain may be highly colored and contain stringent, intensely flavored compounds that are not always appealing in taste. Other barriers to whole-grain consumption include price, softness, texture, and moisture content ²⁰.

The bran and germ fractions derived from conventional milling provide a majority of the biologically active compounds found in a grain. Specific nutrients include high concentrations of B vitamins (thiamin, niacin, riboflavin, and pantothenic acid) and minerals (Ca, Mg, K, P, Na, and Fe), elevated levels of basic amino acids (for example, arginine and lysine), and elevated tocol levels in the lipids ²⁰.

Numerous phytochemicals, some common in many plant foods (phytates and phenolic compounds) and some unique to grain products (avenanthramides, avenalumic acid) are responsible for the high antioxidant activity of whole grain foods ²³.

Components in whole grains associated with improved health status include lignans, tocotrienols, phenolic compounds, and antinutrients including phytic acid, tannins, and enzyme inhibitors. In the grain-refining process the bran is removed, resulting in the loss of dietary fiber, vitamins, minerals, lignans, phyto-oestrogens, phenolic compounds, and phytic acid. Thus, refined grains are more concentrated in starch since most of the bran and some of the germ is removed in the refining process ²⁰.

It is well accepted that glucose and insulin are linked to chronic diseases, especially DM. Whole-grain consumption is part of a healthy diet described as the “prudent” diet. Epidemiological studies consistently show that the risk for T2DM decreases with the consumption of whole grains ^{24, 25}. Whole grains are now recommended by the American Diabetes Association for T2DM prevention ²⁶.

Pereira et al ²⁷ tested the hypothesis that whole grain consumption improves insulin sensitivity in overweight and obese adults. Eleven overweight or obese hyperinsulinaemic adults aged 25–56 years consumed two diets, each for 6 weeks. The diets were identical, except that refined-grain products were replaced by whole products.

At the end of each treatment, subjects consumed 355 ml of a liquid mixed meal, and blood samples were taken over 2 h. Fasting insulin was 10% lower during the consumption of the whole-grain diet. The authors concluded that insulin sensitivity may be an important mechanism whereby whole grain foods reduce the risk of T2DM and heart disease.

2 ANTIDIABETIC POTENTIAL OF CEREAL GRAINS

With several studies suggesting the beneficial role of whole grains on human health, it is important to examine their antidiabetic potential. The following section discusses the clinical and dietary intervention studies conducted on the commonly consumed grains.

2.1 Avena sativa—oat

Oat bran lowers serum lipid concentrations in healthy and hyperlipidemic subjects. The addition of oatmeal to subjects’ usual diets lowered serum total cholesterol significantly by 5–8% ^{28, 29}.

To determine the effects of a ready-to-eat oat-bran cereal on lipid concentrations, control (corn flakes) and oat-bran cereal diets for 2 weeks were fed to 12 men with undesirably high serum total-cholesterol concentrations. After completing the first diet, subjects completed 2 weeks on the alternate diet. Intakes of carbohydrate, protein, fat, and cholesterol were virtually identical on the two diets. The oat-bran cereal provided 25g oat bran/day. The oat-bran cereal diet compared with the corn flakes diet lowered serum total-cholesterol and serum LDL-cholesterol concentrations significantly by 5.4% (P<.05) and 8.5% (P<.025), respectively. Hence, it was concluded that ready-to-eat oat-bran cereal provides a practical means to incorporate soluble fiber into the diet to lower serum cholesterol ³⁰.

Fasting control subjects and subjects with T2DM were fed porridge meals containing either wheat farina, wheat farina plus oat gum or oat bran. Blood samples were collected for 3 h after the test meals and plasma glucose and insulin were measured. Oat bran and wheat farina plus oat gum meals reduced the postprandial plasma glucose excursions and insulin levels when compared with the control wheat farina meal in both control and T2DM subjects. This study showed that both the native cell wall fiber of oat bran and isolated oat gum, when incorporated into a meal, act similarly by lowering postprandial plasma glucose and insulin levels. A diet rich in β-glucan may therefore be of benefit in the regulation of postprandial plasma glucose levels in subjects with type 2 diabetes ³¹.

Earlier studies incorporating oat gum in the diet of subjects also revealed a lowering in glucose and insulin levels. Jenkins et al ³² reported that oat gum reduced glucose and insulin responses of healthy adults when added to a glucose solution. Even Braaten et al ³¹ tested responses to glucose and glucose with oat gum and found reductions in glucose and insulin when the nine healthy subjects consumed solutions to which oat gum had been added. The high viscosity of the solution containing oat gum was concluded to be the property that delays gastric emptying and/or intestinal absorption resulting in these lower responses ³³.

Granfeldt et al ³⁴ tested responses of nine older men to raw rolled oats, boiled rolled oats, boiled intact oat kernels, and white bread. Only consumption of boiled intact oat kernels resulted in glucose and insulin response reductions below the responses to white bread. Consumption of rolled oats or oat flour showed reductions in glucose response by 15–30% in moderately overweight women.

2.2 Hordeum vulgare—barley

Barley exhibited suitable properties as a cereal for NIDDM subjects with a low glycemic index (68.7 in 18 healthy volunteers and 53.4 in 14 NIDDM subjects) and a high insulinaemic index (105.2) in NIDDM subjects. It seemed to mobilize insulin in NIDDM subjects at 0.5 h after ingestion ³⁵.

Comparison of barley and oat foods to white bread in nine healthy subjects found no effect of porridges, but consumption of high-fiber barley breads resulted in glycemic indices of 57–72% of white bread and insulin indices of 42–72% of index of white bread ³⁶.

Substantial reductions in insulin and glucose responses were found in seven young subjects after consumption of varying amounts of boiled barley compared to white bread ³⁷.

Yokoyama et al ³⁸ compared responses of five subjects to pastas containing wheat or wheat and 12g β-glucans from barley. Consumption of barley-containing pasta resulted in lower glycemic and insulin indices.

Bourdon et al ³⁹ compared glucose and insulin responses of 11 healthy men (28–42 years of age). Subjects consumed traditional wheat pasta or a pasta in which 40% of the wheat flour had been replaced with either a highly viscous barley cultivar (Prowashonupana), which naturally has about 15% soluble β-glucan or Waxbar barley enriched in soluble fiber by repeated milling. Insulin responses were lower after consuming the barley pastas than after taking the traditional wheat pasta during the first hour. Though the glucose areas did not differ, the decline in glucose after taking barley pastas was more gradual than after the wheat pasta.

In a study, the effects of a barley diet containing high dietary fiber on the onset and development of DM in spontaneously diabetic rats was investigated by comparing with a rice diet containing low dietary fiber and an alpha-corn starch diet containing very low dietary fiber. Thirty male Goto-Kakizaki (GK) strain rats (8 weeks of age) were randomly assigned to three groups; high barley (HB) group on a barley diet (dietary fiber intake, 1.79 g/day/rat), rice (R) group on a rice diet (dietary fiber intake, 0.46 g/day/rat), and alpha-corn starch (CS) group on an alpha-corn starch diet (dietary fiber intake, 0.24 g/day/rat). Feeding for 3 months showed that fasting plasma glucose level in the HB group was significantly lower than in the R and CS groups; the glucose tolerance in the HB group was markedly improved. Moreover, the plasma cholesterol and triglyceride levels in the HB group were significantly lower than those of the R and CS groups. It was concluded that barley enabled glycemic control and improved glucose tolerance compared with rice or alpha-corn starch ⁴⁰.

2.3 Paspalum scrobiculatum—kodo millet

Whole grain flour of kodo millet incorporated at 55% by wt. in the basal diet fed to alloxan-induced diabetic rats over a period of 28 days showed a considerable reduction in blood glucose (42%) and cholesterol (27%). The levels of enzymatic and non-enzymatic antioxidants and lipid peroxides were significantly reduced in diabetic rats and restored to normal levels in the millet-fed groups ⁴¹.

2.4 Oryza sativa—rice

Rasmussen et al ⁴² compared responses of seven diabetics to 25 and 50 g carbohydrate from white rice and white bread and found significantly lower glucose and insulin responses after the 50 g rice compared to 50 g white bread. Similar reductions were found in responses to 100 g of rice compared to white bread in men and women ⁴².

In a recent study the consumption of pre-germinated brown rice has been associated with antidiabetic potential, improving hyperglycemia and preventing neuropathy of diabetes in experimental animals ⁴³.

2.5 Secale cereal—rye

Rye is relatively high in soluble fiber, but there are few human studies reporting responses to rye consumption ⁴⁴.

Whole kernel rye has been tested in a few groups of diabetics ^{45, 46}. As compared to white bread, GI of whole kernel rye bread was 42–56 in diabetics (both type 1 and 2).

Juntunen et al ⁴⁷ fed high-fiber rye bread and white-wheat bread to postmenopausal women and measured glucose and insulin metabolism. The acute insulin response increased significantly more during the rye-bread periods than during the wheat-bread period. They suggested that high-fiber rye bread appears to enhance insulin secretion, possibly indicating an improvement of β-cell function.

2.6 Triticum aestivum—wheat

McIntosh et al ⁴⁸ compared responses to consumption of wheat and barley foods for 4 weeks in each of 21 mildly hypercholesterolemic men. Although cholesterol levels were significantly lowered by barley, there was no difference in glucose responses to either wheat or barley. Total dietary fiber in this study was significantly increased from 21 to 38 g/day during both periods, a circumstance which could account for the lack of difference.

The comparison of four types of wheat—whole grain, cracked grain, coarse, and fine wholemeal flour—in 10 healthy subjects resulted in glucose responses to whole grain of approximately one-third the response of the fine flour. Insulin responses were found to be similar ⁴⁹.

Thirty NIDDM females after a month of control period were fed 125 g of instant wheat meal (45 g in breakfast, 40 g in mid morning, and 40 g in evening) on weekly basis for 2 months. The body weight, BMI, and waist hip ratio reduced significantly but were still higher than the recommended standard after supplementation. Significant decrease in blood pressure of the subject was also recorded after supplementation of instant wheat meal. It was inferred from the results that supplementation of instant wheat meal helped in the reduction of anthropometric parameters and blood pressure of the diabetic subjects and thus helped in the retardation of secondary complications associated with T2DM ⁵⁰.

2.7 Zea mays—maize

Fermented preparation of Zea mays styles showed hypoglycemic effect in rabbits ⁵¹. Macerated decoction extract of styles produced consistent hypoglycemic effect in fasting rabbits; the hypoglycemic effect compared well with that of crystalline insulin ⁵². Some Native American corn products such as tortillas (54%) and hominy (57%) have much lower glycemic indices than white bread ⁵³.

The effect of soluble corn bran hemicellulose (CBH, 10 g/day) on glucose control and serum insulin in three groups: patients with IGT with (20 subjects) or without (8 subjects) obesity and with healthy non-obese controls (10 subjects), was studied. Long-term supplementation (6 months) with CBH decreased the post-OGTT curve for patients with impaired mild type 2 diabetes, but not that for the controls. Hemoglobin A1c decreased significantly during CBH supplementation in the obese patients, while the fasting glucose level decreased in all three groups, although not significantly. A decreased serum insulin response by OGTT was found in those patients with IGT. The improved OGTT result was associated with improved insulin release and perhaps with peripheral insulin sensitivity. These findings suggest that CBH at a low dose might contribute to glycemic control and would play a useful role in treating type 2 diabetes patients ⁵⁴.

To assess the effect of novel maize-based dietary fibers on postprandial glycemia in a clinical study, 12 healthy volunteers were fed seven test beverages containing maize-based fiber ingredients (25 g total carbohydrate) along with two control meals on separate occasions in random order. It was observed that all test fibers resulted in significantly lower glycemic and insulinemic responses for the incremental area under the curve (iAUC) and at all time points compared with the control (P<.05). The in vitro digestibility curves were comparable to the cumulative in vivo iAUCs. In vitro data expressed as percent digestion correlated significantly with the in vivo iAUC for the first 30 min of the test meal (P<.05). It was concluded that these novel maize-based dietary fibers produced lower postprandial glycemic and insulinemic responses than the control. While further assessment is necessary in beverage and foods containing these fibers, they may be effective in applications for dietary strategies to control diabetes and other chronic diseases ⁵⁵.

The effect of Zea mays L. saponin (ZMLS) on ultrastructure of kidney and pancreas and blood glucose in the diabetes rats induced by streptozocin was studied.

It was found that as compared with the model group, the large, middle-dose ZMLS group could remarkably decrease the blood glucose (P<.01), and could remarkably prevent the pancreatic islet beta-cell from the injury induced by Streptozotocin. ZMLS showed beneficial effect on decreasing blood glucose and protective action on the kidney and pancreas injury induced by STZ ⁵⁶.

A recent study by Robertson et al ⁵⁷ found that the consumption of Hi-maize-resistant starch from National Starch Food Innovation, Bridgewater, NJ, significantly increased insulin sensitivity in subjects with insulin resistance and metabolic syndrome. The 8-week randomized, crossover study showed that 10 overweight subjects with insulin resistance and metabolic syndrome who consumed 40 g of dietary fiber from Hi-maize-resistant starch/day increased their hepatic insulin sensitivity by 54%, their peripheral (muscle) insulin sensitivity by 24%, and their glucose uptake into forearm muscle by 68%. They also had reduced fasting insulin levels, reduced postprandial insulin responses to a standardized meal, and significantly lower levels of fasting non-esterified fatty acids.

From the above-mentioned details it is clear that cereal grains exhibit their antidiabetic effect by glucose lowering, insulin-stimulating, and delayed gastric emptying and intestinal absorption on account of high dietary fiber content. These are also effective in controlling oxidative damage and other secondary complications related to diabetes. Hence, cereal grains are effective in applications for dietary strategies to control diabetes and other chronic diseases.

3 CONCLUSION

The incidences of modern lifestyle diseases like type 2 diabetes widely prevalent in industrialized countries are on the rise in developing countries. The burden of T2DM is enormous when the costs of diagnosis and treatment are considered. Among the various lifestyle approaches for the management of this disease, dietary intervention with a vegetarian diet seems to be an economical, physiological, and safe approach for the prevention and possible management of T2DM.

Consumption of a number of grains and grain extracts has been reported to control or improve glucose tolerance and reduce insulin resistance. Although dietary goals recommend the consumption of three servings of whole grains per day, the average consumption is less than one serving per day. There are a number of mechanisms by which grains may improve glucose metabolism and delay or prevent the progression of IGT to insulin resistance and diabetes. These mechanisms are related to the physical properties and structure of grains. The composition of the grain, including particle size, amount and type of fiber, viscosity, amylose and amylopectin content all affect the metabolism of carbohydrates from grains. Increasing whole-grain intake in the population can result in improved glucose metabolism and delay or reduce the risk of developing T2DM. Whole grains can provide a substantial contribution to the improvement of the diets of people. A number of whole grain foods and grain fiber sources are beneficial in reduction of insulin resistance and improvement in glucose tolerance.

Keywords

glucose, insulin, grains, glycemic index, diabetes

Disclosure: The authors declare no conflict of interest.

REFERENCES

1. Wild S, Roglic G, Green A, Sicree R, King H. Global prevalence of diabetes: estimates for the year 2000 and projections for 2030. Diab Care. 2004;27:1047–1053.
2. ADA 2004. http://www.diabetes.org/uedocuments/PSQF04.pdf [Accessed November 9, 2006].
3. Kaushik G, Satya S, Khandelwal RK, Naik SN. Commonly consumed Indian plant food materials in the management of diabetes mellitus. Diab Metab Synd: Clin Res Rev. 2010;4(1):21–40.
4. Burcelin R, Rolland E, Dolci W, Germain S, Carrel V, Thorens B. Encapsulated, genetically engineered cells, secreting glucagon-like peptide-1 for the treatment of non-insulin dependent diabetes mellitus. Ann N Y Acad Sci. 1999;875(18):277–285.
5. Atalay M, Laaksonen DE. Diabetes, oxidative stress and physical exercise. J Sports Sci Med. 2002;1:1–14.
6. Memisogullari R, Taysi S, Bakan E, Capoglu I. Antioxidant status and lipid peroxidation in Type II diabetes mellitus. Cell Biochem Funct. 2003;21(3):291–296.
7. Raskin P, Jovanovic L, Berger S, Schwartz S, Woo V, Ratner R. Repaglinide/troglitazone combination therapy: improved glycemic control in Type 2 diabetes. Diab Care. 2000;23(7):979–983.
8. Cunningham J, Leffell M, Mearkle P, Harmatz P. Elevated plasma ceruloplasmin in insulin-dependent diabetes mellitus: evidence for increased oxidative stress as a variable complication. Metabolism. 1995;44(8):996–999.
9. Lipinski B. Pathophysiology of oxidative stress in diabetes mellitus. J Diab Compl. 2001;15:203–210.
10. Warren RE. The stepwise approach to the management of Type 2 diabetes. Diabetes Res Clin Pract. 2004;65(1):S3–S8.
11. Mallare JT, Karabell AH, Velasquez-Mieyer P, Stender SRS, Christensen ML. Current and future treatment of metabolic syndrome and Type 2 diabetes in children and adolescents. Diabetes Spectr. 2005;18(4):221–225.
12. Pan XR, Li GW, Hu YH, et al. Effects of diet and exercise in preventing NIDDM in people with impaired glucose tolerance: the Da Qing IGT and Diabetes Study. Diab Care. 1997;20:537–544.
13. Tuomilehto J, Lindström J, Eriksson JG, Valle TT, Hamalainen H, Ilanne- Parikka P. Prevention of Type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med. 2001;344: 1343–1350.
14. Diabetes Prevention Program Research Group. Reduction in the incidence of Type 2 diabetes with lifestyle intervention or metformin. N Engl J Med. 2002;346:393–403.
15. ADA 2005. http://www.diabetes.org/nutrition-and-recipes/nutrition/foodpyramid.jsp [Accessed November 9, 2006].
16. The Diabetes, Nutrition Study: Group of the European Association for the Study of Diabetes (EASD). Recommendations for the nutritional management of patients with diabetes mellitus. Eur J Clin Nutr. 2000;54:353–355.
17. Ranilla LG, Apostolidis E, Genovese MI, Lajolo FM, Shetty K. Evaluation of indigenous grains from the Peruvian Andean region for antidiabetes and antihypertension potential using in vitro methods. J Med Food. 2009;12(4):704–713.
18. Terruzzi I, Senesi P, Magni C, et al. Insulin-mimetic action of conglutingamma, a lupin seed protein, in mouse myoblasts. Nutr Metab Cardiovasc Dis. 2010 (in press).
19. Trowell H. Ischemic heart disease and dietary fiber. Am J Clin Nutr. 1972;25:926–932.
20. Slavin J. Whole grains and human health. Nutr Res Rev. 2004;17:99–110.
21. United States Department of Agriculture. Pyramid Servings Data – Results from USDA’s 1994–1996 Continuing Survey of Food Intake by Individuals. Riverdale, CA: Food Surveys Research Group; 1997.
22. United States Department of Agriculture (2000) Dietary Guidelines for Americans. Dietary Guidelines Advisory Committee 2000. http://www.ars.usda.gov/dgac/2kdiet.pdf [Accessed December 10, 2006].
23. Miller G, Prakash A, Decker E. Whole-grain micronutrients. In: Marquart L, Slavin JL, Fulcher RG, eds. Whole-Grain Foods in Health and Disease. St Paul, MN: Eagan Press; 2002:243–258.
24. Van Dam RM, Rimm EB, Willett WC, Stampfer MJ, Hu FB. Dietary patterns and risk for Type 2 Diabetes mellitus in US men. Ann Intern Med. 2002;136:201–209.
25. Murtaugh MA, Jacobs DR, Jacob B, Steffen LM, Marquart L. Epidemiological support for the protection of whole grains against diabetes. Proc Nutr Soc. 2003;62:143–149.
26. Franz MJ, Bantle JP, Beebe CA, et al. Evidence-based nutrition principles and recommendations for the treatment and prevention of diabetes and related complications. Diab Care. 2002;25:148–198.
27. Pereira MA, Jacobs DR, Pins JJ, et al. Effect of whole grains on insulin sensitivity in overweight hyperinsulinemic adults. Am J Clin Nutr. 2002;75:848–855.
28. De Groot AP, Luyken R, Pikaar NA. Cholesterol-lowering effect of robled oats. Lancet. l963;2:303–304.
29. Luyken R, de Wijn JF, Pikaar NA, von der Meer R. Effect of oatmeal on the serum cholesterol content of the blood. Voeding. 1965;26:229–244.
30. James WA, Donna BS, Carlene CH, et al. Oat-bran cereal lowers serum total and LDL cholesterol in hypercholesterolemic men. Am J Clin Nutr. 1990;52:495–499.
31. Braaten JT, Scott FW, Wood PJ, et al. High b-glucan oat bran and oat gum reduce postprandial blood glucose and insulin in subjects with and without type 2 diabetes. Diab Med. 1994;11(3):312–318.
32. Jenkins DJA, Wolever TMS, Leeds AR, et al. Dietary fibers, fiber analogues, and glucose tolerance: importance of viscosity. Brit Med J. 1978;1:1392–1394.
33. Wood PJ, Anderson JW, Braaten JT, Cave NA, Scott FW, Vachon C. Physical effects of b-D-glucan rich fractions from oats. Cer Foods World. 1989;34:878–882.
34. Granfeldt Y, Hagander B, Bjorck I. Metabolic responses to starch in oat and wheat products. On the importance of food structure, incomplete gelatinization or presence of viscous dietary fibre. Eur J Clin Nutr. 1995;49:189–199.
35. Shukla K, Narain JP, Puri P, et al. Glycaemic response to maize, bajra and barley. Indian J Physiol Pharmacol. 1991;35(4):249–254.
36. Liljeberg HG, Granfeldt YE, Bjorck IM. Products based on a high fiber barley genotype, but not on common barley or oats, lower postprandial glucose and insulin responses in healthy humans. J Nutr. 1996;126: 458–466.
37. Wolever TMS, Bolognesi C. Source and amount of carbohydrate affect postprandial glucose and insulin in normal subjects. J Nutr. 1996;126:2798–2806.
38. Yokoyama WH, Hudson CA, Knuckles B, et al. Effect of barley b-glucan in durum wheat pasta on human glycemic response. Cer Chem. 1997;74: 293–296.
39. Bourdon I, Yokoyama W, Davis P, et al. Postprandial lipid, glucose, insulin, and cholecystokinin responses in men fed barley pasta enriched with (-glucan. Am J Clin Nutr. 1999;69:55–63.
40. Li J, Kaneko T, Wang Y, Qin LQ, Sato A. Effects of dietary fiber on the glucose tolerance in spontaneously diabetic rats—comparison among barley, rice, and corn starch. Nippon Eiseigaku Zasshi. 2003;58(2):281–286.
41. Prashant SH, Namakkal SR, Chandra TS. Effects of the antioxidant properties of millet species on oxidative stress and glycemic status in alloxan-induced rats. Nutr Res. 2005;25(12):1109–1120.
42. Rasmussen O, Gregersen S, Hermansen K. Influence of the amount of starch on the glycaemic index to rice in non-insulin-dependent diabetic subjects. Brit J Nutr. 1992;67:371–377.
43. Usuki S, Ariga T, Dasgupta S, et al. Structural analysis of novel bioactive acylated steryl glucosides in pre-germinated brown rice bran. J Lipid Res. 2008;49(10):2188–2196.
44. Hallfrisch J, Behall KM. Mechanisms of the effects of grains on insulin and glucose responses. J Am Coll Nutr. 2000;19(3):320S–325S.
45. Jenkins DJA, Wolever TMS, Jenkins AL, et al. Low glycemic response to traditionally processed wheat and rye products: bulgur and pumpernickel bread. Am J Clin Nutr. 1986;43:516–520.
46. Brand JC, Foster KA, Crossman S, Truswell AS. The glycaemic and insulin indices of realistic meals and rye breads tested in healthy subjects. Diab Nutr Metab. 1990;3:137–142.
47. Juntunen KS, Laaksonen DE, Poutanen KS, Niskanen LK, Mykkanen HM. High fiber rye bread and insulin secretion and sensitivity in healthy postmenopausal women. Am J Clin Nutr. 2003;77(2):385–391.
48. McIntosh GH, Whyte J, McArthur R, Nestel PJ. Barley and wheat foods: influence on plasma cholesterol concentrations in hypercholesterolemic men. Am J Clin Nutr. 1991;53:1205–1209.
49. Behall KM, Scholfield DJ, Hallfrisch J. The effect of particle size of whole grain flour on plasma glucose, insulin, and TSH in human subjects. J Am Coll Nutr. 2000;18:591–597.
50. Anita K, Kanchandeep G, Rajbir S. Impact of instant wheat meal supplementation and nutrition counselling on anthropometry and blood pressure of non insulin dependent diabetics. Ethno-Med. 2009;3(1):69–74.
51. Menczel E. Hypoglycemic substances. Bull. Res. Coun. Israel Sect. E. 1963; 102(1963):235–236.
52. Menczel E, Sulman FG. Hypoglycaemic effect of Zeu styles. Proc So. Exp Bio & Med. 1962;1110(1):178–181.
53. Miller JB. International tables of glycemic index. Am J Clin Nutr. 1995;62:871S–893S.
54. Hanai H, Ikuma M, Sato Y, et al. Long-term effects of water-soluble corn bran hemicellulose on glucose tolerance in obese and non-obese patients: improved insulin sensitivity and glucose metabolism in obese subjects. Biosci Biotechnol Biochem. 1997;61(8):1358–1361.
55. Kendall CW, Esfahani A, Hoffman AJ, et al. Effect of novel maize-based dietary fibers on postprandial glycemia and insulinemia. J Am Coll Nutr. 2008;27(6):711–718.
56. Miao MS, Zhang GL, Miao YY, Shi JJ, Liu HL. Influence of Zea mays L. saponin (ZMLS) on ultrastructure of kidney and pancreas in diabetes rats induced by streptozocin. Zhongguo Zhong Yao Za Zhi. 2008;33(10):1179– 1183.
57. Robertson MD, Wright JW, Batt J, Russell-Jones D, Umpleby AM. Dietary resistant starch is an insulin sensitizer A37(P37). Diab Med. 2009;26(suppl l):14.

Hypogonadism in patients with type 2 diabetes mellitus (T2DM) is an underestimated medical problem that is associated with decreased sexual function, bone mineral density, and sense of well-being. It is also associated with increased incidence of T2DM and the cardiac metabolic syndrome (CMS).¹³ Usually, hypogonadism in patients with T2DM remains undiagnosed until patients complain of erectile dysfunction (ED), which affects one-third of all patients.^{4, 5} The etiology of ED in diabetic patients is often secondary to vascular disease or autonomic neuropathy. However, hypogonadism is also playing a role in the etiology of ED in patients with DM as well as in the metabolic derangements that these patients are suffering from. The current Endocrine Society guidelines for testosterone therapy in adult men with androgen deficiency syndromes recommend testosterone treatment to maintain secondary sex characteristics, improve sexual function, sense of well-being, and bone mineral density.⁶ However, the actual benefit of testosterone replacement therapy in patients with T2DM may go beyond these effects.

ASSOCIATION BETWEEN LOW TESTOSTERONE AND T2DM AND COMPONENTS OF METABOLIC SYNDROME

Pathophysiology

The relationship between hypogonadism with obesity and insulin resistance is bidirectional (Figure 1).⁷

Adipose Tissue Leading to Hypogonadism and Insulin Resistance

In adipose tissue, testosterone is aromatized to 17β estradiol, which feeds back and decreases gonadotropin-releasing hormone (GnRH) release leading to hypogonadotropic hypogonadism.⁷ In addition, the adipocytokines, interleukin (IL)-6, tumor necrosis factor (TNF)-α, as well as leptin, which are increased in obesity, blunt the GnRH-gonadotropin response to hypogonadism.⁸ There are also leptin receptors in Leydig cells that inhibit luteinizing hormone (LH) and human chorionic gonadotropin (hCG) stimulated testosterone release.^{9, 10}

The adipocytokines, TNF-α and IL-6 as well as leptin, upregulate suppressor of cytokine signaling-3 (SOCS-3) that results in impaired insulin receptor (IRS) tyrosine kinase phosphorylation and proteosomal degradation of IRS1 and IRS2 leading to impaired insulin transduction at multiple levels, which most likely also include the hypothalamus.¹¹ The consequent insulin resistance then induces hypogonadism.

Insulin Resistance Leading to Hypogonadism

The reduced gonadotropin response to low testosterone levels seems to be at the hypothalamic level. Further support of a blunted GnRH response due to insulin resistance comes from studies in neuron-specific insulin receptor knockout (NIRKO) rats in which LH levels are reduced but there is restoration upon direct GnRH stimulation.¹² The reason for a reduced GnRH response could be due to insulin resistance, as a consequence of elevated levels of TNF-α and IL-6 that are increased in obesity and T2DM.¹¹ Gonadal androgens at physiological levels act by inhibiting macrophage IL-1, IL-6, and TNF-α.¹³

There are insulin and leptin receptors in Leydig cells and insulin resistance additionally reduces testosterone secretion at this level both directly and indirectly through hyperinsulinemia-mediated leptin release.¹⁰

Hypogonadism Leading to Obesity and Insulin Resistance

Hypogonadism induces lipoprotein lipase at the adipose tissue level promoting TG storage and inhibits hormone sensitive lipase, which limits β oxidation that ultimately increases fat mass.^{14, 15}

Studies with androgen receptor knockout (ARKO) rats have also demonstrated that obesity develops secondary to decreased activity and decreased oxygen consumption¹⁶ and treatment of hypogonadism has been shown to prevent visceral adipose tissue accumulation.¹⁷ Hypogonadism promotes the commitment of mesenchymal pluripotent cells into myogenic lineage and inhibits their differentiation into adipogenic lineage.¹⁸

It has also been shown that testosterone inhibits IL-6 production and hypogonadism is associated with release of inflammatory mediators that could contribute to HH through the above described mechanisms.¹³

Studies of Hypogonadism in Diabetes Mellitus

Testosterone levels are lower in men with T2DM, CMS, and in obese males even in the absence of T2DM or CMS.^{19, 20} This relationship is bidirectional as hypogonadism leads to T2DM as well as CMS and the converse is also true as demonstrated by Laaksonen et al.²⁰ He showed that men with metabolic syndrome at baseline and that still met criteria at an 11-year follow-up had a 5.7–7.4 times higher risk of developing hypogonadism defined as total testosterone (TT) levels less than 11 nmol/L. Each of the parameters of metabolic syndrome individually increased the risk of hypogonadism. Laaksonen also showed that those who no longer had the metabolic syndrome at follow-up didn't remain at increased risk.

In another study involving 3156 men aged 45–84 years who participated in the Multi-Ethnic Study of Atherosclerosis showed that after adjusting for age, ethnicity, BMI, and waist circumference, impaired fasting glucose (IFG) and diabetes were associated inversely with TT. The odds ratio (OR) for the highest quartile compared to the lowest ranged from 0.26 to 0.77 for diabetes and from 0.50 to 0.85 for IFG. Further evidence of the direct relationship of hypogonadism and insulin resistance, independent of changes in body composition is supported by a trial in which testosterone withdrawal for a 2-week period in patients with idiopathic hypogonadotropic hypogonadism resulted in a homeostatic model assessment of insulin resistance (HOMA-IR) increase from 1.07±0.2 to 1.4±0.1 (P<.005).²¹

Several studies including the Massachusetts Male Aging Study (MMAS), Multiple Risk Factor Intervention Trial (MRFIT), and Rancho Bernardo Study have revealed a direct relationship between hypogonadism with hyperinsulinemia, glucose intolerance, and incident T2DM and CMS.^1–3 It has also been shown that this association remains even after adjustment for BMI.² In a study conducted by Haffner, in 87 men in Finland, he demonstrated that total whole-body glucose disposal was associated negatively with waist to hip ratio (WHR) and positively associated with TT and sex hormone binding globuline (SHBG).²²

The Hypogonadism in Males (HIM) study estimated the prevalence of hypogonadism, defined as a TT of less than 300 ng/dl (<10.4nmol/L) in men aged 45 years visiting primary care practices in the United States to be 38.7%.²³ This biochemical definition of hypogonadism, although imprecise, has been adopted as the standard by the Food and Drug Administration (FDA) and the Endocrine Society when it is accompanied by signs and symptoms of hypogonadism.⁶ In this study, the prevalence rate for hypogonadism in diabetes was of 50% (odds ratio of 2.09) and the prevalence rate for hypogonadism in obesity was 52.4%.²³ In a consecutive series of 1134 male patients with sexual dysfunction with a mean age of 52 years, the prevalence of hypogonadism in patients with diabetes and metabolic syndrome was 41% and in patients with metabolic syndrome alone was 29% based on a TT<10.4nmol/L.²⁴

Some of the studies assessed TT and SHBG levels. Since TT levels are decreased with insulin resistance and obesity, firm conclusions require assessment of free testosterone (FT), which represents 0.5%–3% of circulating testosterone. Measurement of FT through radioimmunoassay (RIA) is inaccurate and the gold standard, which is equilibrium dialysis, is not readily available. Dhindsa et al,²⁵ confirmed an increased prevalence of hypogonadism of 33% among 103 males with T2DM, by determining FT through equilibrium dialysis. If only TT had been used to define hypogonadism, there would have been 36% false positives and 12% false negatives. In addition, gonadotropin levels were lower confirming the hypogonadotropic nature of this disorder.²⁵

Dhindsa et al demonstrated that the prevalence of low TT and FT was 0% and 6%, respectively, in type 1 diabetic patients, highlighting that hypogonadotropic hypogonadism is not due to hyperglycemia per se. This could be related to lack of hyperinsulinemic and obesity decreased SHBG.²⁶

THE EFFECTS OF TESTOSTERONE REPLACEMENT THERAPY ON DIABETES AND COMPONENT OF THE METABOLIC SYNDROME

Effects on Diabetes and Glucose Homeostasis

There is a growing evidence of the favorable effect of testosterone replacement therapy (TRT) on glucose homeostasis in general and glycemic control in patients with T2DM. Many studies have demonstrated the favorable effect of TRT on blood glucose and insulin sensitivity. Marin P study group studied the effect of TRT (vs placebo) in 23 middle-aged men with abdominal obesity.²⁷ The insulin sensitivity, measured by the euglycemic/hyperinsulinemic glucose clamp method, improved in the treatment group. More improvement was noted in men with relatively low serum testosterone levels at the outset. Naharci et al²⁸ demonstrated that untreated patients (n=24) with idiopathic hypogonadotrophic hypogonadism (IHH) had higher fasting plasma glucose levels, higher fasting insulin levels, a higher HOMA-IR score, and a lower quantitative insulin sensitivity check index (QUICKI) when compared with the control group (n=20, age, weight-matched eugonadal men). The HOMA-IR score decreased dramatically to the level of the control group after 6 months treatment with testosterone. Another study examined the effect of acute withdrawal of TRT in healthy men with IHH, on insulin sensitivity, and glucose homeostasis.²¹ Fasting glucose levels, fasting insulin levels, HOMA-IR score increased, and insulin sensitivity index decreased from baseline 2 weeks after discontinuation of testosterone treatment.

Hypogonadal men with type 2 diabetes who were treated with TRT, showed improvement in hemoglobin A1c (HbA1c) and fasting blood glucose in several studies. Boyanov et al,²⁹ an open-label study, included (n=24) middle-aged, type 2 diabetic patients with mild androgen deficiency. Participants were treated with 120mg oral testosterone undecanoate for 3 months. The control group (n=24) included similar patients but they did not receive any treatment. The treatment group showed significant improvement in blood glucose levels and mean HbA1c (10.4% to 8.6% P<.05). Heufelder et al,³⁰ a single blind, randomized clinical trial, examined the effect of the addition of 5 grams testosterone gel to diet and exercise (D&E) in 32 mildly hypogonadal men with newly diagnosed T2DM. All patients underwent a supervised diet and exercise program and did not receive any glucose lowering agent. Half of the participants (n=16) received testosterone gel treatment. At the end of the 52 weeks study period, all of the patients treated with combined D&E plus testosterone reached the HbA1c target value of less than 7.0% and 87.5% reached less than 6.5%, whereas only 40.4% of the D&E alone participants reached less than 7.0% and none reached less than 6.5% (P<.001 for both comparisons). Kapoor et al³¹ is a double-blind, placebo-controlled crossover study. It included 24 hypogonadal men with T2DM (10 were on treatment with insulin) who were treated with intramuscular (IM) 200mg testosterone injections every 2 weeks for 3 months. It showed reduction in HbA1c (−0.37±0.17%, P=.03), fasting blood glucose (−1.58±0.68mmol/l, P=.03), the HOMA index (−1.73±0.67, P=.02, n=14 not on insulin), and an average of 7 units/day reduction in insulin dose in those on insulin (n=10). More recently, Corona et al,³² a meta-analysis of the randomized controlled trials, included the three mentioned trials,^29–31 confirmed the above results by showing a significant reduction in fasting plasma glucose and HbA1c associated with TRT.

On the contrary, Corrales et al³³ showed neutral effects on overall glycemic control in patients with T2DM and partial androgen deficiency when treated with 150 mg of IM testosterone injections every 2 weeks for 6 months. Similar results were concluded from a small study of 11 patients with T2DM and relative hypogonadism before and after treatment with 100 mg IM testosterone injection every 3 weeks for 12 weeks.³⁴ The results showed that before and after treatment there was insignificant difference in (HOMA) of insulin sensitivity and beta-cell function as well as insignificant change in insulin sensitivity, glucose effectiveness, and acute insulin response (AIR) after glucose load. The author concluded that the insignificant results might be secondary to the duration of treatment and the dose used. As a matter of fact, in a pilot study to explore the relationship between androgens and glucose tolerance in obese men,³⁵ a single IM injection of testosterone 500mg resulted in decreased glucose tolerance, while the response to a single dose of 250mg was suggesting improved insulin sensitivity. This suggests a possible role for the dose of testosterone used and that higher levels of testosterone may result in unfavorable effects.

Effects on Adipose Tissue and Body Composition

Adipose tissue is an endocrine organ that secret several hormones that play a role in glucose homeostasis. Visceral fat has a higher metabolic activity with respect to insulin-induced glucose uptake and is strongly linked to insulin resistant and the development of coronary artery disease. Sex steroids exert their effects on the adipose tissue by increasing lipolysis through activating hormone sensitive lipase. In a study of the effect of testosterone on the adipocytes in a male rat module, testosterone was shown to increase lipolysis and the number of adrenoreceptors.³⁶ A study in mouse pluripotent stem cells indicates that testosterone regulates body composition by promoting the commitment of these mesenchymal cells into the myogenic lineage and inhibiting their differentiation into the adipogenic lineage.³⁷ A significant decrease in body fat mass along with significant increases in body mass index and body lean mass was observed after TRT in 24 men with IHH (28). A meta-analysis, Isidori et al,³⁸ of randomized controlled trials evaluating the effects of TRT on body composition in middle-aged and aging men showed a reduction of −1.6 kg (CI: 2.5–0.6) of total body fat, corresponding to −6.2% (CI: 9.2–3.3) variation of initial body fat, an increase in fat-free mass of 1.6 kg (CI: 0.6–2.6), corresponding to +2.7% (CI: 1.1–4.4) increase over baseline and no change in body weight.

The randomized trials that were conducted in patients with T2DM and hypogonadism to examine the effect of TRT have shown similar results. In Heufelder et al,³⁰ waist circumferences declined in both groups but more so in the diet/exercise plus testosterone group. A significant reduction in visceral adiposity as assessed by waist circumference (−1.63±0.71cm, P=.03) and waist/hip ratio (−0.03±0.01, P=.01) was reported by Kapoor et al.³¹ In Boyanov et al,²⁹ while negligible change was found in the control group, a statistically significant (P<.05) reduction in body weight (−2.66%), visceral obesity (−3.96%; measured by waist–hip ratio), and body fat (−5.65%; assessed in the fasting state by bioelectrical impedance on a tetrapolar body composition analyzer) was detected in the treatment group. A meta-analysis, Corona et al,³² confirmed the above results.

Effects on Lipids

The relationship between the change in testosterone levels and the lipid profile was demonstrated in epidemiological studies as well as in studies in patients with prostate cancer receiving androgen ablation. Haidar et al³⁹ is a retrospective study of 29 patients with prostate cancer, with a mean age of 75 years. All patients received standard LHRH agonist therapy. The lipids profile from before and after the ablation therapy were compared. The results showed an increase in total cholesterol from 252.0+41.14 to 322.3+41.09mg/dL, low density lipoprotein (LDL) cholesterol from 184.5+16.31 to 229.1+21.00mg/dL, triglycerides (TG) from 207.4+39.3 to 83.9+49.68mg/dL, and a decrease in high density lipoprotein (HDL) cholesterol from 31.4+6.35 to 20.9+2.22mg/dL.

A meta-analysis of trials conducted to examine the effect of IM testosterone injections on lipids, Whitsel et al⁴⁰ showed a small, dosage-dependent decrease in HDL cholesterol and concomitant declines in total cholesterol and LDL cholesterol. The decrease in HDL cholesterol was not explained by dosing frequency or duration, concomitant elevation of plasma TT, or aromatization of testosterone to estradiol. In the meta-analysis by Isidori et al,³⁸ total cholesterol decreased by 0·23 mmol/l (CI:−0.37 to −0.10), especially in men with lower baseline testosterone concentrations with no change in LDL cholesterol. A significant reduction of HDL cholesterol was found only in studies with higher mean testosterone levels at baseline (−0.085mmol/l, CI:−0.017 to −0.003). Sensitivity and meta-regression analysis revealed that the dose/type of testosterone used, in particular the possibility of aromatization, explained the heterogeneity in findings observed on HDL cholesterol among studies.

The randomized trials that studied the effects of TRT on lipid profile in patients with T2DM and hypogonadism has shown variable results. Boyanov et al²⁹ showed no significant change in the lipid profile. In Heufelder et al³⁰ there was a significant reduction in serum TG and a significant increase in HDL cholesterol in the D&E with TRT group. In Kapoor et al³¹ there was no significant change in LDL cholesterol, TG, or HDL cholesterol between the placebo and treatment group. However, there was a statistically significant decrease in total cholesterol (−0.4±0.17mmol/l, P=0.03) between the two groups. The meta-analysis, Corona et al³² that included the above three trials, showed a significant reduction in TG, while no significant change was noted in total of HDL cholesterol.

Effect on Blood Pressure

Exogenous sex hormone use, including oral contraceptives, postmenopausal hormonal therapy, and anabolic steroids, has been associated with blood pressure (BP) changes in both sexes. However, levels of endogenous testosterone were shown to have inverse relationship with BP.⁴¹ The Tromsø study⁴² showed that lower levels of testosterone in men are associated with higher blood pressure, higher left ventricular mass, and left ventricular hypertrophy.

The Marin P study group had published several studies that showed that administration of TRT in abdominally obese, middle-aged hypogonadal men resulted in improvement in diastolic BP²⁷ and significant decrease in diastolic BP.⁴³

A clinical trial in patients with hypogonadism and T2DM treated with testosterone showed variable results. While Boyanov et al²⁹ and Kapoor et al³¹ showed no effect on BP, Heufelder et al³⁰ showed a significant decrease in diastolic BP and no significant change in systolic BP in the treatment group (supervised D&E with TRT) vs the control group (supervised D&E alone). The meta-analysis, Corona et al included the three trials,^29–31 showed no significant change in BP.

Effect on Inflammation

Although it does not constitute a formal diagnostic criterion for metabolic syndrome, inflammation has been recognized as a player in the development of insulin resistance in metabolic syndrome. Leptin, a product of the adipose tissue, is a potent proinflammatory hormone with circulating levels strongly correlated to the amount of body fat. Other atherogenic proinflammatory cytokines include TNFα, IL-1, and IL-6. Although adiponectin is also secreted from the adipose tissue, obesity is associated with lower levels of it. It sensitizes the body to insulin and is an important anti-inflammatory adipokine.

Testosterone has immunomodulating properties. A randomized, single-blind, placebo-controlled, crossover study of testosterone replacement (vs placebo) in 27 men with symptomatic androgen deficiency, showed that testosterone induced reductions in TNFα and IL-1β and an increase in IL-10 (anti-inflammatory).⁴⁴ In another study, the observed poststent elevation in IL-6 and hs-CRP levels in the control group was significantly attenuated in the treatment group who received a short-term (3 weeks) testosterone treatment.⁴⁵

In a study of testosterone deficient men with T2DM, although no significant effects on TNF-a, IL-6, or CRP levels were detected after TRT, the baseline testosterone levels significantly inversely correlated with IL-6 and CRP levels.⁴⁶ In the same study, both leptin and adiponectin levels decreased. A more recent study, Heufelder et al,³⁰ showed improvement in the levels of adiponectin and hs-CRP in D&E with TRT group. In Corrales et al,⁴⁷ TRT therapy depresses the ex vivo production of IL-1β, IL-6, and TNFα by circulating antigen-presenting cells in aging type 2 diabetic men with partial androgen deficiency.

Effects on Coagulation

The hypogonadal state carries a higher risk of hypercoagulability.⁴⁸ Low levels of testosterone are associated with a hypercoagulable state. In men with low levels of testosterone, levels of plasminogen activator inhibitor type 1 (PAI-1), Factor VII, and fibrinogen all are negatively correlated with testosterone levels, whereas tissue plasminogen activator (tPA) is positively correlated.

TESTOSTERONE THERAPY IN OLDER MEN

A recent study conducted by Basaria et al⁴⁹ randomized community-dwelling men, 65 years of age or older, with limited mobility, and a testosterone level of 3.5 to 12.1 nmol per liter, to transdermal testosterone gel vs placebo. Among the participants, 24% vs 26% in the testosterone and placebo group, respectively, had diabetes. Higher rates of cardiovascular adverse events in the testosterone group led to early termination of the study. There was, however, a higher baseline hyperlipidemia and statin use in the testosterone group and there was lack of assessment of estrogen levels, which could have been increased via aromatization of administered testosterone. This is important, since hyperestrogenemia is a documented risk factor for thrombosis and cardiovascular mortality in elderly men.⁵⁰ In addition, the target testosterone level in the treatment group was 17.4 to 34.7nmol/l, which is substantially higher than the usual recommendation of achieving testosterone levels in the mid-normal range for age (18.15±6.83nmol/L for ages 65–74), and these higher levels could have accounted for the observed adverse effects.⁵¹ Importantly, the testosterone group had greater leg and arm strength and the higher cardiovascular event rate could be explained by a protection factor accounted for the limited physical vigor in hypogonadal men when compared to those who received testosterone treatment.⁵² The results of this trial were surprising given that numerous studies with a higher number of patients and that have achieved the same or higher testosterone levels have not reproduced these findings. Additional larger studies will be needed to clarify the important dilemma of testosterone replacement in elderly men.⁵³ Another concern when using TRT is the effect of testosterone on the prostate. Holmang et al⁵⁴ showed that the prostate volume increased by 12% (P<.012) in eugonadal middle-aged men treated with testosterone undecanoate (160mg/day). However, there were no changes in the urine flow or in the serum concentration of prostate-specific antigen (PSA). Meikle et al⁵⁵ also noted increased prostate volume when hypogonadal men were treated with transdermal testosterone. There was no change in the urine flow, and the prostate volume on treatment was comparable to that reported for normal men. Although there are several case reports for prostate cancer diagnosed in patients on TRT, the evidence of causality is still not well established. A recently published meta-analysis to evaluate the adverse effects of TRT showed no significant difference in the risk of prostatic and urologic outcomes between patients who received TRT and the control group.⁵³

SUMMARY

The relationship between hypogonadism with obesity and insulin resistance is bidirectional.⁷ Many studies report increased incident of T2DM and CMS in patients with hypogonadism^1–3 and CMS as well as T2DM predispose to the development of hypogonadism.²⁰

Testosterone replacement therapy seems to have several favorable effects on glucose homeostasis and the other components of the CMS in patients with T2DM. The TRT improves fasting blood glucose levels, HbA1c, insulin sensitivity, and reduces the amount of insulin used. These effects seem to be dose dependent and that may explain why these effects were not observed in Lee et al.³⁴ Also, supraphysiologic doses may result in decreased glucose tolerance.³⁵ A reduction in visceral adipose tissue with TRT was reported across all studies in patients with T2DM and hypogonadism. Also, it resulted in a decrease in total cholesterol and TG in several studies and meta-analysis. There is good evidence to suggest a favorable effect in decreasing BP, especially the diastolic BP.³⁰ However, other studies did not show significant change.^{29, 31, 32} Testosterone might be exerting some of its effects through its immune modulating effects as shown in some studies that TRT decreased CRP, IL-6, IL-1β, TNFα, and increases IL-10 that are thought to play a major rule in the pathophysiology of the CMS.

The results from Basaria et al⁴⁹ should caution in careful testosterone administration in older men with CV disease and immobility but it certainly should not deter practitioners from prescribing testosterone replacement for well-established late-onset hypogonadism, particularly in patients with T2DM where there has been proven benefit as exemplified above. Given that many men are willing to accept risks in light of the benefits of testosterone replacement, additional larger studies of testosterone administration in well-characterized groups of older men, particularly with T2DM, need to be conducted to more clearly outline benefits (prevention of bone fractures, improvement in muscle strength, well-being, and sexual function as well as avoidance of falls, prevention of psychiatric disease, and possibly improvement in insulin sensitivity), and balance them against the risks of prostatic and CV disease as well as other adverse outcomes.

Keywords

cardiac metabolic syndrome, type 2 diabetes mellitus, obesity, hypogonadotropic hypogonadism, insulin resistance, testosterone

Disclosure: The Authors declare no conflict of interest.

Acknowledgement: This research was supported by NIH (ROI HL73101-01A1) and Veterans Affairs Merit System 0018 (JRS)

REFERENCES

1. Oh J-Y, Barrett-Connor E, Wedick N, et al. Endogenous sex hormones and the development of type 2 diabetes in older men and women: the Rancho Bernardo Study. Diabetes Care. 2002;25(1):55–60.
2. Stellato RK, Feldman HA, Hamdy O, et al. Testosterone, sex hormonebinding globulin, and the development of type 2 diabetes in middle-aged men: prospective results from the Massachusetts male aging study. Diabetes Care. 2000;23(4):490–494.
3. Laaksonen D, Niskanen L, Punnonen K, et al. Testosterone and sex hormone-binding globulin predict the metabolic syndrome and diabetes in middle-aged men. Diabetes Care. 2004;27(5):1036–1041.
4. McCulloch DK, Campbell IW, Wu FC, et al. The prevalence of diabetic impotence. Diabetologia. 1980;18(4):279–283.
5. Fedele D, Coscelli C, Santeusanio F, et al. Erectile dysfunction in diabetic subjects in Italy. Diabetes Care. 1998;21:1973–1977.
6. The Endocrine Society Clinical Practice Guidelines. Testosterone therapy in adult men with androgen deficiency syndromes. J Clin Endocrinol Metab. 2010;95(6):2536–2559.
7. Cohen PG. The hypogonadal-obesity cycle. Med Hypotheses. 1999;52: 49–51.
8. Jones TH. Testosterone associations with erectile dysfunction, diabetes and metabolic syndrome. Eur Urol Suppl. 2007;6:847–857.
9. Isidori AM, Caprio M, Strollo F, et al. Leptin and androgens in male obesity: evidence for leptin contribution to reduced androgen levels. J Clin Endocrinol Metab. 1999;84:3673–3680.
10. Caprio M, Isidori AM, Carta AR, et al. Expression of functional leptin receptors in rodent Leydig cells. Endocrinology. 1999;140:4939–4947.
11. Emanuelli B, Peraldi P, Filloux C, et al. Suppressor of cytokine signaling 3 is a physiological regulator of adipocyte insulin signaling. J Biol Chem. 2001;276:47944–47949.
12. Bruning JC, Gautam D, Burks DJ, et al. Role of brain insulin receptor in control of body weight and reproduction. Science. 2000;289:2122–2125.
13. Li ZG, Danis VA, Brooks PM. Effect of gonadal steroids on the production of IL-1 and IL-6 by blood mononuclear cells in vitro. Clin Exp Rheumatol. 1993;11:157–162.
14. Blouin K, Boivin A, Tchernof A. Androgens and body fat distribution. J Steroid Biochem Mol Biol. 2008;108:272–280.
15. Pitteloud N, Hardin M, Dwyer AA, et al. Increasing insulin resistance is associated with a decrease in Leydig cell testosterone secretion in men. J Clin Endocrinol Metab. 2005;90(5):2636–2641.
16. Fan W, Yanase T, Nomur M, et al. Androgen receptor null male mice develop late-onset obesity caused by decreased energy expenditure and lipolytic activity but show normal insulin sensitivity with high adiponectin secretion. Diabetes. 2005;54:1000–1008.
17. Allan C, Strauss B, Burger H. Testosterone therapy prevents gain in visceral adipose tissue and loss of skeletal muscle in nonobese aging men. J Clin Endocrinol Metab. 2008;93:139–146.
18. Sing R, Artaza JN, Taylor WE, et al. Androgens stimulate myogenic differentiation and inhibit adipoenesis in C3H 10T1/2 pluripotent cells through an androgen receptor-mediated pathway. Endocrinology. 2003; 144:5081–5088.
19. Kaplan SA, Meehan AG, Shah A. The age related decrease in testosterone is significantly exacerbated in obese men with the metabolic syndrome. What are the implications for the relatively high incidence of erectile dysfunction observed in these men? J Urol. 2006;176(4, pt 1): 1524–1527.
20. Laaksonen D, Niskanen L, Punnonen K, et al. The metabolic syndrome and smoking in relation to hypogonadism in middle-aged men: a prospective cohort study. Eur J Endocrinol. 2003;149:601–609.
21. Yialamas M, Dwyer A, Hanley E, et al. Acute sex steroid withdrawal reduces insulin sensitivity in healthy men with idiopathic hypogonadotropic hypogonadism. J Clin Endocrinol Metab. 2007;92(11):4254– 4259.
22. Haffner S, Karhapaa P, Mykkanen L. Insulin resistance, body fat distribution, and sex hormones in men. Diabetes. 1994;43:212–219.
23. Mulligan T, Frick M, Zuraw Q, et al. Prevalence of hypogonadism in males aged at least 45 years: the HIM study. Int J Clin Pract. 2006;60 (7):762–769.
24. Corona G, Mannucci E, Petrone L, et al. NCEP-ATPIII-defined metabolic syndrome, type 2 diabetes mellitus, and prevalence of hypogonadism in male patients with sexual dysfunction. J Sex Med. 2007;4:1038–1045.
25. Dhindsa S, Prabhakar S, Sethi M, et al. Frequent occurrence of hypogonadotropic hypogonadism in type 2 diabetes. J Clin Endocrinol Metab. 2004;89:5462–5468.
26. Tomar R, Dhindsa S, Chaudhuri A, et al. Contrasting testosterone concentrations in type 1 and type 2 diabetes. Diabetes Care. 2006;29(5): 1120–1122.
27. Marin P, Holmang S, Jonsson L, et al. The effects of testosterone treatment on body composition and metabolism in middle-aged obese men. Int J Obes Relat Metab Disord. 1992;16(12):991–997.
28. Naharci MI, Pinar M, Bolu E, et al. Effect of testosterone on insulin sensitivity in men with idiopathic hypogonadotropic hypogonadism. Endocr Pract. 2007;13(6):629–635.
29. Boyanov M, Boneva Z, Christov V. Testosterone supplementation in men with type 2 diabetes, visceral obesity and partial androgen deficiency. Aging Male. 2003;6:1–7.
30. Heufelder AE, Saad F, Bunck MC, et al. 52-week treatment with diet and exercise plus transdermal testosterone reverses the metabolic syndrome and improves glycaemic control in men with newly diagnosed type 2 diabetes and subnormal plasma testosterone. J Androl. 2009; 30:726–733.
31. Kapoor D, Goodwin E, Channer KS, et al. Testosterone replacement therapy improves insulin resistance, glycaemic control, visceral adiposity and hypercholesterolaemia in hypogonadal men with type 2 diabetes. Eur J Endocrinol. 2006;154:899906.
32. Corona G, Monami M, Rastrelli G, et al. Type 2 diabetes mellitus and testosterone: a meta-analysis study [published online ahead of print October 24, 2010]. Int J Androl. 2010.
33. Corrales JJ, Burgo RM, Garcý´a-Berrocal B, et al. Partial androgen deficiency in aging type 2 diabetic men and its relationship to glycemic control. Metabolism. 2004;53(5):666–672.
34. Lee C, Kuo S, Hung Y, et al. The effects of testosterone supplement on insulin sensitivity, glucose effectiveness, and acute insulin response after glucose load in type 2 diabetic patients. Endocr Res. 2005;31(2):139– 148.
35. Mårin P, Krotkiewski M, Björntorp P. Androgen treatment of middleaged, obese men: effects on metabolism, muscle and adipose tissues. Eur J Med. 1992;1(6):329–336.
36. Xu XF, De Pergola G, Björntorp P. Testosterone increases lipolysis and the number of beta-adrenoreceptors in male rat adipocytes. Endocrinology. 1991;128(1):379–382.
37. Singh R, Artaza JN, Taylor W, et al. Androgens stimulate myogenic differentiation and inhibit adipogenesis in C3H 10T1/2 pluripotent cells through an androgen receptor-mediated pathway. Endocrinology. 2003; 144(11):5081–5088.
38. Isidori AM, Giannetta E, Greco EA, et al. Effects of testosterone on body composition, bone metabolism and serum lipid profile in middle-aged men: a meta-analysis. Clin Endocrinol (Oxf). 2005;63:280–293.
39. Haidar A, Yassin A, Saad F, et al. Effects of androgen deprivation on glycaemic control and on cardiovascular biochemical risk factors in men with advanced prostate cancer with diabetes. Aging Male. 2007; 10(4):189–196.
40. Whitsel E, Boyko E, Matsumoto A, et al. Intramuscular testosterone esters and plasma lipids in hypogonadal men: a meta-analysis. Am J Med. 2001;111:261–269.
41. Khaw KT, Barrett-Connor E. Blood pressure and endogenous testosterone in men: an inverse relationship. J Hypertens. 1988;6(4):329–332.
42. Svartberg J, von Muehlen D, Schirmer H, et al. Association of endogenous testosterone with blood pressure and left ventricular mass in men. The Tromsø study. Eur J Endocrinol. 2004;150:65–71.
43. Marin P, Holmang S, Gustafsson C, et al. Androgen treatment of abdominally obese men. Obesity Res. 1993;1(4):245–251.
44. Malkin CJ, Pugh PJ, Jones RD, et al. The effect of testosterone replacement on endogenous inflammatory cytokines and lipid profiles in hypogonadal men. J Clin Endocrinol Metab. 2004;89(7):3313– 3318.
45. Niyazi G, Talantbek B, Haluk D, et al. The effects of short term (3 weeks) testosterone treatment on serum inflammatory markers in men undergoing coronary artery stenting. Int J Cardiol. 2006;109:339–343.
46. Kapoor D, Clarke S, Stanworth R, et al. The effect of testosterone replacement therapy on adipocytokines and C-reactive protein in hypogonadal men with type 2 diabetes. Eur J Endocrinol. 2007;156(5): 595–602.
47. Corrales JJ, Almeida M, Burgo R, et al. Androgen-replacement therapy depresses the ex vivo production of inflammatory cytokines by circulating antigen-presenting cells in aging type-2 diabetic men with partial androgen deficiency. J Endocrinol. 2006;189(3):595–604.
48. Saad F. The role of testosterone in type 2 diabetes and metabolic syndrome in men. Arq Bras Endocrinol Metab. 2009:53–58.
49. Basaria S, Coviello AD, Travison TG, et al. Adverse events associated with testosterone administration. N Engl J Med. 2010;363:109–122.
50. Phillips G. Correspondence. Adverse events associated with testosterone administration. N Engl J Med. 2010;363:1865–1867.
51. Khoo T. Correspondence. Adverse events associated with testosterone administration. N Engl J Med. 2010;363:1865–1867.
52. Weikert C, Pischon T. Correspondence. Adverse events associated with testosterone administration. N Engl J Med. 2010;363:1865–1867.
53. Fernandez-Balsells MM, Murad MH, Lane M, et al. Clinical review 1: adverse effects of testosterone therapy in adult men: a systematic review and meta-analysis. J Clin Endocrinol Metab. 2010;95:2650–2675.
54. Holmang S, Marin P, Lindstedt G, et al. Effect of long-term oral testosterone undeconoate treatment on prostate volume and serum prostate-specific antigen concentration in eugonadal middle aged men. Prostate. 1993;23(2):99–106.
55. Meikle AW, Arver S, Dobs AS, et al. Prostate size in hypogonadal men treated with a nonscrotal permeation-enhanced testosterone transdermal system. Urology. 1997;49(2):191–196.