Associations of IR and Insulin With MAU and CKD. Many early small case-control studies in subjects with various components of the CMS investigated the association among the underlying disorder of the CMS, IR, and MAU. These studies were primarily designed to examine whether MAU was another component of the syndrome. With few exceptions, IR was significantly higher in microalbuminuric than in normoalbuminuric patients with various concomitant conditions of various origins.45,46 In cross-sectional studies, a similar trend was observed. In 982 nondiabetic subjects from the cohort of the Insulin Resistance Atherosclerosis Study,47 an investigative group showed that subjects with MAU had lower insulin sensitivity and higher fasting insulin concentrations compared with subjects without MAU. In regression analysis, an increasing degree of insulin sensitivity was related to a decreasing prevalence of MAU (odds ratio [OR], 0.86).47 In a recent study in 712 patients with DM, ACR was independently related to the homeostasis model assessment of IR (HOMA-IR) index.48
Some studies have also examined cross-sectional and prospective associations between MAU and hyperinsulinemia. In a small prospective study in patients with DM, serum insulin concentration at baseline independently and significantly predicted the UAE rate after 6 years of follow-up.49 In 497 clinically healthy, nondiabetic Koreans, those with MAU had higher levels of fasting insulin than those with normoalbuminuria and, in multiple logistic regression analyses, fasting insulin and systolic BP were the only variables independently associated with MAU.50 In a subsequent study from the same group in 1006 Koreans, subjects with MAU had higher fasting plasma insulin and proinsulin levels, while in multiple regression analyses, fasting insulin was independently related to UAE rate and hyperinsulinemia was independently associated with the presence of MAU.51
Other studies however, did not confirm these associations. In two populations with high risk of DM, Wanigelas from Papua New Guinea and Nauruans of the South Pacific, the relationships among UAE and markers of IR were inconsistent in most groups and were no longer significant after adjusting for fasting glucose and body mass index (BMI).52 In 271 never-treated, nondiabetic, uncomplicated hypertensive Caucasian men, the HOMA-IR index and insulin levels did not differ across ascending urine albumin quartiles, whereas BMI and BP were significantly greater in the upper compared with the lower quartiles.53
As far as the relationship between IR and CKD is concerned, in a case-control study in Japanese subjects, there was no significant difference in insulin sensitivity (measured with the clamp technique) among patients with DM with normoalbuminuria, microalbuminuria or macroalbuminuria. However, a fourth group of patients who had DM with CKD (serum creatinine levels >2.0 mg/dL) demonstrated more than two times lower insulin sensitivity than the normoalbuminuric group.54 In another case-control study, insulin sensitivity measured with the clamp technique was lower in nondiabetic patients with CKD than in healthy controls. Moreover, in CKD patients there is a positive correlation between insulin sensitivity and creatinine clearance.55 A recent study including 227 nondiabetic patients with CKD and different degrees of renal function and 76 matched healthy controls showed that the HOMA-IR index was also markedly higher in CKD patients than in healthy subjects. However, HOMA-IR in CKD patients was significantly correlated with BMI, triglycerides, and age, but not with the level of GFR.56
Investigators examined the relationship of fasting serum insulin, glycated hemoglobin (HbA1c), and HOMA-IR index to risk of CKD in a subgroup of the Third National Health and Nutrition Examination Survey (NHANES III) population, including 6453 adult individuals without DM.57 The prevalence of CKD was significantly and progressively higher with increasing levels of serum insulin, HbA1c, and HOMA-IR. After adjustment for potential confounders, the ORs of CKD for the highest compared with the lowest quartile were 4.03, 2.67, and 2.65 for serum insulin, HbA1c levels, and HOMA-IR, respectively. Furthermore, for one SD higher level of serum insulin, HbA1c, and HOMA-IR, the increases in the risk of CKD were 35%, 69%, and 30%, respectively. These findings support the notion that IR and concomitant hyperinsulinemia are present in CKD patients without DM. However, this analysis cannot separate the effect on the development of CKD of the two above factors from that of blood glucose levels.
Associations of the CMS With MAU and CKD. Numerous observational studies have examined the relationship between MAU, or CKD, and either the major components of CMS or the syndrome as a single entity, defined with the use of different diagnostic tools (Table). In early case-control studies, UAE was found higher in patients with many features of the CMS, and vice versa,58 but since no regression analyses were applied to adjust for possible confounders, the conclusions that can be drawn from those studies are limited.
Table Table. Cardiometabolic Syndrome Manifestations Associated With Cardiovascular Disease
|Clinical components of cardiometabolic syndrome|
| Insulin resistance/hyperinsulinemia|
| Visceral obesity|
| Chronic kidney disease|
| Increased serum apolipoprotein B levels|
| Small, dense low-density lipoprotein cholesterol particles|
| Increased plasminogen activator inhibitor/plasminogen activator ratio|
| Increased serum fibrinogen levels|
| Increased production of interleukin-6|
| Increased systolic and pulse pressure|
| Premature atherosclerosis|
| Enhanced tissue renin-angiotensin-aldosterone system|
| Salt sensitivity|
| Endothelial dysfunction|
Data from cross-sectional studies performed in subjects with various concomitant conditions are more interesting. One such study examined the effects of the factors included in the “deadly quartet” described by Kaplan in 198959 (upper-body obesity, glucose intolerance, hypertriglyceridemia, and hypertension) on overnight UAE ratios in 317 patients with DM. In logistic regression analyses controlling for possible confounders, elevated HbA1c, or the addition of one component of the deadly quartet, to pure DM doubled or tripled the odds of an elevated UAE ratio. The addition of two components to DM further increased the ORs, while the presence of DM and all three factors were associated with an OR of 9.34 for elevated UAE.60 Another cross-sectional study examined associations of MAU and macroalbuminuria with the CMS among Australian Aboriginal people and showed highly significant linear associations of both microalbuminuria and macroalbuminuria with an increasing number of components of the syndrome.61 Among people with none, one, two, and three to five of these components, the prevalence of MAU was 16%, 20%, 36%, and 32%, and the prevalence of macroalbuminuria was 2%, 6%, 12%, and 32%, respectively. However, MAU was independently associated only with hypertension and DM, and macroalbuminuria was independently associated only with hypertension, DM, abdominal obesity, and less strongly with IR—findings indicating that hypertension and DM were the major contributors to high rates of albuminuria in this population.
In the sixth examination (1995–1998) of the Framingham Offspring Study,62 urine ACR was measured in 1592 subjects and categorized as having none, one, two, or all three of the phenotypes of IR syndrome they used: impaired glucose tolerance, hypertension, and/or a central metabolic syndrome (two or more traits of obesity, dyslipidemia, or hyperinsulinemia). The groups of participants with all three phenotypes of the syndrome had higher age- and sex-adjusted ACRs and a greater proportion of ACR >30 mg/g than those with no phenotypes. Moreover, groups with two or three phenotypes of the syndrome had higher ACRs and a greater proportion of subjects with ACR >30 mg/g than groups with none or one of the phenotypes—and groups with hyperinsulinemia tended to have a greater proportion of subjects with ACR >30 mg/g compared with those without hyperinsulinemia.62 Another investigative group explored the association between MAU and the IR syndrome in 934 nondiabetic Native Americans. After controlling for possible confounders, the OR for MAU was 1.8 for one IR syndrome component, 1.8 for two, and 2.3 for three or more, compared with the presence of no syndrome components.63
As with IR, however, not all the studies examining the relationship between the CMS and MAU documented a close association. A Dutch study used an age-, sex- and glucose tolerance-stratified random sample of the 50–75-year-old general population (n=622) to examine the relationship between MAU and several variables related to the syndrome.64 In multiple logistic regression analyses, MAU showed independent associations with hypertension, DM, and waist-to-hip ratio, but no associations with impaired glucose tolerance, hyperinsulinemia, IR, or dyslipidemia. The authors concluded that in this population, MAU was likely a complication of hypertension and DM, and not an integral part of the IR syndrome. Another study examined this relationship in 1031 young adults (61% Caucasian, 39% African American) from the population of the Bogalusa Heart Study.65 After controlling for age and gender, African Americans with MAU had higher systolic and diastolic B P, prevalence of hypertension, and HDL cholesterol than those without MAU, whereas Caucasians showed no such associations. In both races, none of the other variables of the CMS measured (BMI, waist circumference, triglycerides, glucose, insulin, HOMA-IR index, and uric acid) displayed any relation to MAU, suggesting that MAU is not necessarily an intrinsic component of the syndrome.
In recent years, studies examining the relationship between the CMS and CKD have also been released. In further analyses in subsamples of the NHANES III population, including more than 6000 adult subjects aged 20 years and older, investigators observed that the syndrome, defined according to the NCEP/ATP III criteria, was associated with higher risk of both MAU and CKD.66 In particular, the multivariate-adjusted OR of MAU in participants with the CMS compared with participants without it was 1.89 and compared with participants with no or one component of the syndrome, those with three, four, and five components had ORs for MAU of 1.62, 2.45, and 3.19, respectively. Moreover, in participants with the syndrome, the OR of CKD was 2.60 compared with those without it, whereas subjects with two, three, four, and five components had ORs of 2.21, 3.38, 4.23, and 5.85, respectively, compared with participants with no or one component. In multivariate models, however, elevated BP and plasma glucose levels are associated with an increased OR of MAU, whereas elevated B P, low HDL cholesterol, high triglycerides, and abdominal obesity are each associated with an increased OR of CKD,66 thus making it difficult for this analysis to separate the effects of the CMS from those of the individual components on the development of MAU and CKD.
Yet another cross-sectional analysis, a nested cohort of the NHANES III database including more than 7300 participants, examined the prevalence of non-traditional CVD risk factors, including CKD, across DM status and for persons with and without the CMS. After adjustment for multiple confounders, the presence of CMS was again significantly associated with CKD (OR, 2.27).67 A more recent study examined the association of the syndrome defined according to the NCEP/ATP III criteria with diabetic nephropathy in 2415 Finnish patients with type 1 DM. Participants were classified as having a normal UAE rate, MAU, macroalbuminuria, or end-stage renal disease, and the prevalence of the syndrome rose significantly from 28% to 44%, 62%, and 68%, respectively. Moreover, patients with the CMS had an almost four times higher risk of diabetic nephropathy, and each of the separate components of the CMS was independently associated with diabetic nephropathy.68
Finally, the association between the CMS and CKD was also examined in two recent prospective studies.69,70 In the first, including 10,096 nondiabetic subjects with normal baseline kidney function from the original cohort of the Atherosclerosis Risk in Communities (ARIC) study, the presence of the CMS was again associated with a higher risk of CKD after 9 years of follow-up.69 Participants who fulfilled the NCEP/ ATP III criteria for the CMS at baseline had a 43% greater risk of developing CKD than those without it. Compared with participants with no components of the syndrome, those with one, two, three, four, or five components of it had an OR for CKD of 1.13, 1.53, 1.75, 1.84, and 2.45, respectively. Moreover, after adjustment for the subsequent development of DM and hypertension during the 9 years of follow-up, the participants with the CMS at baseline still had a 24% greater risk of developing CKD compared with those without it.69 The second study included subjects without CKD from the Framingham Heart Study offspring cohort and followed them for an average of 7 years for development of CKD. After adjusting for multiple confounders among participants without DM at baseline, the presence of the syndrome was almost significantly associated with the development of CKD during the follow-up period (OR, 1.46; p=0.06).70