 |
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
|
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
*Department of Epidemiology, Rollins School of Public Health, Emory University, Atlanta, Georgia; Cardiovascular Health Branch, Division of Adult and Community Health, National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention, Atlanta, Georgia; Bemidji Area Indian Health Service, Bemidji, Minnesota.
Correspondence to Dr. William M. McClellan, 57 Executive Park South, NE, Suite 200, Atlanta, GA 30329. Phone: 404-982-7573; Fax: 404-982-7591; E-mail: bmcclell{at}gmcf.org
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although microalbuminuria has been associated with an increased rate of decline in creatinine clearance in nondiabetic hypertensive patients (13–14), there is generally less information about its association with the IRS in persons without diabetes. If the metabolic and hemodynamic abnormalities characteristic of the IRS are causally related to microalbuminuria, we would expect to find an association between IRS and microalbuminuria in nondiabetic persons similar to that observed among patients with diabetes. We studied this possibility by examining the association between the IRS and prevalent microalbuminuria among nondiabetic members of a population of Native Americans.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Subjects
An age-stratified random sample (ages 25 to 44 yr and 
45 yr) was drawn from the Indian Health Service clinic lists for each community. These lists included all persons who had used the local Indian Health Service/Tribal healthcare facility at least once during the three preceding years. We excluded persons who reported histories of kidney failure, pregnant women, persons with diabetes mellitus (by history or with a fasting blood glucose >126 mg/dl), and those with macroalbuminuria (albumin:creatinine ratio [ACR] 300 mg/g).
Data
Examinations took place in the tribal clinics after informed consent was obtained. Interviewers at each site ascertained medical, psychosocial, behavioral, and sociodemographic information, and trained technicians obtained physical measures and blood and urine samples from each participant.
Participants were instructed to fast 12 h before the examination. Blood and urine samples were collected and shipped to the Medical Research Laboratory (Highland Heights, KY) for analysis. Urine albumin concentration was measured on a spot urine sample using nephelometry. The alkaline picrate method was used to measure urine creatinine concentration.
Measures of glucose, serum total cholesterol, HDL cholesterol, triglycerides, and serum creatinine were determined by enzymatic methods using a Hitachi 737 chemistry analyzer and reagents from Boehringer Mannheim Diagnostics (Indianapolis, IN). LDL cholesterol was calculated using the formula of Friedewald et al. (15). Serum insulin was determined by RIA (Diagnostic Products, Los Angeles, CA) and was reported in mIU/ml.
BP was measured by a mercury sphygmomanometer three times at 30-s intervals after an initial 5-min rest. The second and third readings were averaged to estimate systolic (first Korotkoff sound) and diastolic (fifth Korotkoff sound) BP. Mean arterial pressure was defined as the sum of the diastolic pressure and one third of the difference between the systolic and diastolic values. Body height (cm) and weight (kg) were measured with the participant dressed in light clothing and without shoes. Abdominal girth was determined at the level of the umbilicus with the participant supine. The anthropometric measurements yielded two adiposity measurements: body mass index (BMI), equal to weight (kg) divided by height squared (m2), and waist-to-hip ratio (WHR).
Educational attainment was categorized as less than, equal to, or more than a high school education. Participants were classified as current smokers if they smoked presently and had smoked at least 100 cigarettes in their lifetime, former smokers if they did not currently smoke but had smoked at least 100 cigarettes in their lifetime, or non-smokers. Self-reported family histories of diabetes and kidney disease among biologic parents or siblings were recorded.
For this study, we defined the IRS as a phenotype comprised of five discrete conditions—hypertension, impaired fasting glucose (IFG), high fasting insulin, low HDL cholesterol, and high triglycerides (16). We defined hypertension as a systolic pressure 140 mmHg, a diastolic pressure 90 mmHg, or current use of antihypertensive medication. For fasting blood glucose (FBG), <110 mg/dl (6.1 mmol/L) was considered normal, and 110 mg/dl (6.1 mmol/L) but <126 mg/dl (7.0 mmol/L) was considered IFG (17). All persons with FBG concentration 126 mg/dl (7.0 mmol/L) or a diagnosis of diabetes were considered diabetic (17). High insulin was defined as a fasting insulin concentration >14 mIU/ml, equivalent to the 75th percentile (18). Low HDL was defined as <35 mg/dl (0.9 mmol/L), and high triglycerides as 250 mg/dl (2.8 mmol/L) (19).
Participants with a BMI 30 kg/m2 were considered obese (20). Obesity, however, was not included in the IRS definition because of its high prevalence in this population (41.3% with BMI 30 kg/m2; 76.5% with BMI 30 kg/m2 and/or WHR > 0.9 in women or WHR > 1.0 in men). In addition, we wanted to determine if the association between the IRS and microalbuminuria was the same or different for obese and non-obese participants.
We defined microalbuminuria as an ACR of 30 to 299 mg/g with normal values <30 mg/g (21). Persons whose albumin concentrations were undetectable were set to the minimum value of 0.89 mg/g.
Statistical Analyses
Statistical analyses were performed using the SAS statistical software (version 8.0; SAS Institute, Cary, NC). A descriptive analysis included either proportions or means and SD for selected characteristics. Bivariate analyses used two-tailed t tests and ANOVA for continuous variables and 
2 tests for categorical variables to assess differences in characteristics between groups defined by either microalbuminuria or IRS traits (22).
Logistic regression was performed to measure the association between the IRS and microalbuminuria by (1) those with particular numbers of IRS traits compared with those having none, and (2) the individual IRS traits and the presence of microalbuminuria (23). An unadjusted model and fully adjusted model were examined. Covariates in the full model included age, gender, BMI, education, smoking history, and family histories of diabetes and kidney disease. To comply with confidentiality agreements, we did not adjust for reservation site in the models. Log likelihood ratios were used to determine the significance of the interaction terms, namely the interactions of the IRS with gender and the IRS with BMI. The final model was derived using a hierarchical backwards elimination procedure in which higher order (interaction) terms were dropped before lower order terms.
Finally, because information regarding the distribution of ACR was lost when we created the dichotomous variable for microalbuminuria (yes, no), we examined the association between the log-transformed ACR and individual IRS traits using Pearson correlation coefficients and the general linear model. The distributions of all continuous variables were examined, and those that displayed significant degree of skewness, including the urinary ACR, were log-transformed. In the full model, the independent associations between the log-transformed ACR and each of the individual IRS traits, namely systolic and diastolic BP, fasting blood glucose, HDL cholesterol concentrations, and the log-transformed values for triglyceride and fasting insulin, were assessed (24). This model adjusted for BMI, WHR, education, smoking history, and family histories of kidney disease and diabetes.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
25 yr were eligible to participate in the ITHP; 399 (19.3%) declined and 294 (14.2%) agreed to participate but did not appear for examination, leaving 1376 (527 men; 849 women). We excluded persons with a history of kidney failure (n = 29) and pregnancy (n = 14), those with incomplete data on microalbuminuria (n = 21), hypertensive status (n = 12), diabetes status (n = 1), or serum insulin concentrations (n = 21), and those with diabetes (n = 338) or macroalbuminuria (n = 6). Thus, the final study population numbered 934. Microalbuminuria was observed in 15% of the 934 nondiabetic participants (Table 1). Persons with microalbuminuria were older (P = 0.001) and had higher systolic BP (P = 0.004) than those without microalbuminuria. Participants without microalbuminuria were more likely to have at least a high school education compared with their counterparts (P = 0.007).
|
|
The mean (± SD) ACR for men and women combined was 20.3 (± 29.0) mg/g (Table 3). The mean ACR increased from 16.8 mg/g among those with no traits to 25.5 mg/g among those with two and 25.1 mg/g among the 69 persons with three or more traits. Similar increases in mean ACR by number of IRS traits were observed by gender. Among men and women combined, mean ACR was significantly higher among those with hypertension and IFG. Mean ACR was significantly higher among men with high fasting insulin yet was similar among women with and without high fasting insulin. In addition, mean ACR was significantly lower among women with low HDL cholesterol but was comparable among men with and without low HDL cholesterol. No differences in mean ACR were observed among participants with and without high triglycerides.
|
|
30 kg/m2 versus <30 kg/m2) and microalbuminuria (OR, 0.80; 95% CI, 0.52 to 1.2) (data not shown). Because the pattern of association was different between men and women, the results are presented separately by gender. Although the interaction did not attain statistical significance (P between 0.05 and 0.10), the association between microalbuminuria and the IRS, by number of traits, appeared stronger among men and resembled more of a dose response relationship compared with that among women. Figure 1 further illustrates the differences between men and women in the prevalence of microalbuminuria by number of IRS traits. No interaction between the IRS and obesity was observed.
|
After controlling for age, gender, smoking history, educational status, and family histories of diabetes and kidney disease in the general linear model, only systolic BP (P = 0.03) was independently associated with the urinary ACR. No association was found for fasting blood glucose (P = 0.15), diastolic BP (P = 0.71), HDL (P = 0.20), triglycerides (P = 0.40), insulin (P = 0.36), BMI (P = 0.30), or WHR (P = 0.44).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Commonly associated with diabetes, the IRS, as represented by three or more traits, was present in a small proportion (7.4%) of our nondiabetic population. In general, BMI increased with increasing number of IRS traits. As expected, insulin concentrations and the metabolic and physiologic abnormalities associated with the IRS increased as participants went from zero to three or more IRS traits.
The link between elevated insulin concentrations and the phenotypic traits of the IRS is still only partially understood. Hyperinsulinemia may contribute to the development of hypertension through multiple mechanisms, including Na+ retention (25), activation of the sympathetic nervous system (26), decreased Na+-K+-ATPase activity (27), and increased Na+-H+ pump activity (28). Similarly, hyperinsulinemia may contribute to dyslipidemia by increasing the synthesis of VLDL by the liver (29), resulting in increased concentrations of triglycerides. Moreover, the low concentrations of HDL in the IRS may reflect an increased rate of catabolism of apoA1 seen in persons with elevated insulin (30–31). Finally, the progressive impairment of glucose tolerance is thought to reflect the loss of beta-cell reserves due to persistent hyperinsulinemia (32).
It has been suggested that microalbuminuria could be another component of the IRS (33–38). Mechanisms that might link hyperinsulinemia to greater urinary albumin excretion include increased glomerular hemodynamic pressure, greater permeability of the filtration barrier due to advanced glycosylation end products, and endothelial dysfunction that results in increased transcapillary escape of albumin.
Several studies that were not population-based have found associations between hyperinsulinemia and microalbuminuria in persons without diabetes, but population-based studies have sometimes found different results. In the former group, for example, Bianchi et al. (39) measured insulin responsiveness in response to a glucose load in persons without diabetes and found a direct correlation between increased insulin secretion and urinary albumin excretion rates. In addition, nondiabetic persons with microalbuminuria have been reported to have higher insulin values in response to a glucose load (40–41), and their albumin urinary excretion rates have been found to be directly proportional to serum glucose concentrations (42). An association between microalbuminuria and hyperinsulinemia has also been reported for young black subjects (43).
Population-based studies, on the other hand, have found less consistent evidence for an association between insulin and microalbuminuria in persons without diabetes. In one study, Haffner et al. (44) examined the association between microalbuminuria and IRS traits in nondiabetic persons living in low-income communities in Mexico City. Microalbuminuria was highly prevalent (55.1%), but no association was found between this condition and mean systolic BP or fasting insulin. In contrast, those with microalbuminuria had higher triglyceride values, and higher fasting and 2-h blood glucose concentrations. After controlling for age (P = 0.289) and parental history of diabetes (P = 0.036), neither the presence of glucose intolerance (P = 0.524) nor a 10 mmHg difference in systolic BP was associated with the presence of microalbuminuria.
In another population-based study, this one of European whites living in the Netherlands, a much lower prevalence of microalbuminuria was found among nondiabetic persons with normal (6.8%) or impaired glucose tolerance (8.7%) than in persons with diabetes (21.3%) (45). In both nondiabetic groups, median triglycerides, HDL cholesterol, fasting blood glucose, and fasting insulin were comparable in those with and without microalbuminuria. After controlling for age, gender, impaired glucose tolerance, a history of hypertension, fasting hyperinsulinemia, the joint distribution of HDL cholesterol and triglyceride values, and WHR, only a history of hypertension and WHR were significantly associated with the presence of microalbuminuria.
Our study’s main finding demonstrating an increased prevalence OR for microalbuminuria and a higher mean ACR for those with three or more IRS traits (versus none) supports a positive association between the IRS and microalbuminuria. However, the gradient of the effect from one to three or more traits was minimal. At the same time, evidence consistent with previous reports supporting no association with individual IRS traits was found. For one, insulin, triglycerides, and HDL cholesterol concentrations were similar between those with and without microalbuminuria, and no association between urinary ACR and these traits was detected.
The differences in the association of the IRS and microalbuminuria between men and women were unexpected yet intriguing. The relationship appeared to be restricted to men only, with an apparently weak relationship in women. If this phenomenon were found in other studies, it would be reasonable that it might reflect a different biologic response between men and women, in that the IRS imparts a stronger, adverse pathophysiologic reaction on the body’s vessels in men as compared with women. Yet, because the interaction was not statistically significant in our models, we can only recommend further investigation on this topic.
Aside from high fasting insulin, the definition of which was based on the population distribution, hypertension was the most prevalent IRS trait both in the total population and among participants with one (38.9%) trait. Moreover, hypertension was the only IRS trait that was significantly (P < 0.05) associated with microalbuminuria after adjusting for the other risk factors in the total population. In this particular population of nondiabetic Native Americans, the association between the IRS and microalbuminuria may be due primarily to one IRS trait, hypertension. However, in men, the joint effect of the IRS traits on microalbuminuria exceeded that of any single metabolic or hemodynamic abnormality, including hypertension. Among women, on the other hand, it appeared that hypertension predominately contributed to the presence of microalbuminuria because it was the only trait significantly associated with this condition.
The prevalence of microalbuminuria in our population is comparable to that previously reported for other non-European, nondiabetic populations (46). For example, the highest reported prevalence of microalbuminuria among nondiabetic persons has been found among Nauru Islanders (21%), New Zealand Maori (11%), and New Zealand Pacific Islanders (10%) (46). The prevalence among nondiabetic Native Americans aged 
45 yr who participated in the Strong Heart Study ranged from 3.9% among those in Oklahoma with normal glucose tolerance (ages, 55 to 64 yr) to 23.1% among those in Arizona with impaired glucose tolerance (ages, 65 to 74 yr) (47). Comparable rates among nondiabetic European populations range between 2% and 10% (46).
Four limitations to our study should be noted. First, clinic lists were used to determine the study population. All persons who had used the clinics within the previous 3 yr were eligible to be selected for participation in the study, and persons who had not used the clinics for health care were not eligible. The lists, however, were considered most representative of the Native Americans living on the reservations at the time of the study due to several factors including geographic isolation of the reservation communities, limited access to non-Indian Health Service facilities, the comprehensive range of services provided by the clinics, and the prerequisite that individuals register with the clinics for reimbursement of services provided outside of the clinics.
A second limitation was the substantial nonparticipation rate. For this to have biased our findings, however, persons with the IRS and microalbuminuria would have to have declined participation at rates greater than other persons. Although possible, this event seems unlikely due to the asymptomatic nature of microalbuminuria.
A third limitation was that we measured microalbuminuria only once and, thus, some persons may have been misclassified. Current American Diabetes Association recommendations suggest 2 of 3 positive tests for diagnosis of microalbuminuria (21). In addition, false positive urine albumin specimens (e.g., bacteria, blood, or white blood cells) may be present. For this to have biased our findings, however, the rate of misclassification would have to differ between men and women with and without the IRS. For example, the rate of false positives would either have to be disproportionately higher among men with the IRS or, conversely, among women without the IRS.
The possibility for insufficient power to detect associations between microalbuminuria and the individual IRS traits, specifically high insulin, high triglycerides, and low HDL cholesterol, served as a fourth limitation. However, the point estimates for these traits all wavered around the null value, which suggested no association.
In summary, we found that in this particular nondiabetic Native American population with a high prevalence of microalbuminuria, the IRS was associated with increased prevalence of microalbuminuria. Our findings suggest that the risk of early proteinuria is increased twofold by the presence of three or more IRS traits. Only one of the syndrome’s traits, elevated BP, was independently associated with increased risk of prevalent microalbuminuria. In general, health promotion efforts should focus on lowering the prevalence of hypertension, as well as glucose intolerance and obesity, in this particular Native American population, which is at high risk for renal and cardiovascular disease.
|
Acknowledgments |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  K. A. Griffin, H. Kramer, and A. K. Bidani Adverse renal consequences of obesity Am J Physiol Renal Physiol, April 1, 2008; 294(4): F685 - F696. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Nguyen, C. McCulloch, P. Brakeman, A. Portale, and C.-y. Hsu Being Overweight Modifies the Association Between Cardiovascular Risk Factors and Microalbuminuria in Adolescents Pediatrics, January 1, 2008; 121(1): 37 - 45. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Whaley-Connell, B. S. Pavey, K. Chaudhary, G. Saab, and J. R. Sowers Review: Renin-angiotensin-aldosterone system intervention in the cardiometabolic syndrome and cardio-renal protection Therapeutic Advances in Cardiovascular Disease, October 1, 2007; 1(1): 27 - 35. [Abstract] [PDF]
|
|
|
|
 |
|
 A. Rashidi, A. Ghanbarian, and F. Azizi Are Patients Who Have Metabolic Syndrome without Diabetes at Risk for Developing Chronic Kidney Disease? Evidence Based on Data from a Large Cohort Screening Population Clin. J. Am. Soc. Nephrol., September 1, 2007; 2(5): 976 - 983. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Nogueira and M. Weir The Unique Character of Cardiovascular Disease in Chronic Kidney Disease and Its Implications for Treatment with Lipid-Lowering Drugs Clin. J. Am. Soc. Nephrol., July 1, 2007; 2(4): 766 - 785. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Scavini, C. A. Stidley, S. S. Paine, V. O. Shah, F. Tentori, A. Bobelu, T. K. Welty, J. W. MacCluer, and P. G. Zager The Burden of Chronic Kidney Disease among the Zuni Indians: The Zuni Kidney Project Clin. J. Am. Soc. Nephrol., May 1, 2007; 2(3): 509 - 516. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Cirillo, W. Sato, S. Reungjui, M. Heinig, M. Gersch, Y. Sautin, T. Nakagawa, and R. J. Johnson Uric Acid, the Metabolic Syndrome, and Renal Disease J. Am. Soc. Nephrol., December 1, 2006; 17(12_suppl_3): S165 - S168. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Locatelli, P. Pozzoni, and L. Del Vecchio Renal Manifestations in the Metabolic Syndrome. J. Am. Soc. Nephrol., April 1, 2006; 17(4_suppl_2): S81 - S85. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. H. Jee, L. E. Boulware, E. Guallar, I. Suh, L. J. Appel, and E. R. Miller III Direct, Progressive Association of Cardiovascular Risk Factors With Incident Proteinuria: Results From the Korea Medical Insurance Corporation (KMIC) Study Arch Intern Med, October 24, 2005; 165(19): 2299 - 2304. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Kurella, J. C. Lo, and G. M. Chertow Metabolic Syndrome and the Risk for Chronic Kidney Disease among Nondiabetic Adults J. Am. Soc. Nephrol., July 1, 2005; 16(7): 2134 - 2140. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. R. Tuttle Renal manifestations of the metabolic syndrome Nephrol. Dial. Transplant., May 1, 2005; 20(5): 861 - 864. [Full Text] [PDF]
|
|
|
|
|
|
G. C. Shearer, J. W. Newman, B. D. Hammock, and G. A. Kaysen Graded Effects of Proteinuria on HDL Structure in Nephrotic Rats J. Am. Soc. Nephrol., May 1, 2005; 16(5): 1309 - 1319. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. R. Schelling and J. R. Sedor The Metabolic Syndrome as a Risk Factor for Chronic Kidney Disease: More than a Fat Chance? J. Am. Soc. Nephrol., November 1, 2004; 15(11): 2773 - 2774. [Full Text] [PDF]
|
|
|
|
|
|
F. A. El-Atat, S. N. Stas, S. I. McFarlane, and J. R. Sowers The Relationship between Hyperinsulinemia, Hypertension and Progressive Renal Disease J. Am. Soc. Nephrol., November 1, 2004; 15(11): 2816 - 2827. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Sayeed, H. Mahtab, P. A. Khanam, K. A. Ahsan, A. Banu, A.N.M. B. Rashid, and A.K. A. Khan Diabetes and Impaired Fasting Glycemia in the Tribes of Khagrachari Hill Tracts of Bangladesh Diabetes Care, May 1, 2004; 27(5): 1054 - 1059. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. T. Lane Microalbuminuria as a marker of cardiovascular and renal risk in type 2 diabetes mellitus: a temporal perspective Am J Physiol Renal Physiol, March 1, 2004; 286(3): F442 - F450. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Chen, P. Muntner, L. L. Hamm, D. W. Jones, V. Batuman, V. Fonseca, P. K. Whelton, and J. He The Metabolic Syndrome and Chronic Kidney Disease in U.S. Adults Ann Intern Med, February 3, 2004; 140(3): 167 - 174. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. O. Shah, M. Scavini, C. A. Stidley, F. Tentori, T. K. Welty, J. W. MacCluer, A. S. Narva, A. Bobelu, C. P. Albert, D. S. Kessler, et al. Epidemic of Diabetic and Nondiabetic Renal Disease among the Zuni Indians: The Zuni Kidney Project J. Am. Soc. Nephrol., May 1, 2003; 14(5): 1320 - 1329. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. R. Loon Diabetic Kidney Disease: Preventing Dialysis and Transplantation Clin. Diabetes, April 1, 2003; 21(2): 55 - 62. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
HOME
CURRENT ISSUE
ARCHIVES
JASN Express
ONLINE SUBMISSION
AUTHOR INFO
EDITORIAL BOARD SUBSCRIB |
