The clinical implications of microalbuminuria (MA) in nondiabetic persons with the metabolic syndrome (MS) are largely unknown. The present post hoc analysis of the Multiethnic Study of Atherosclerosis (MESA) included 5,809 nondiabetic persons with no history of cardiovascular disease aged 45 to 84 years. The study population was divided according to the presence or absence of MS and MA into 4 study groups: no MS and no MA, MA only, MS only, and MS plus MA. The measurements included markers of systemic inflammation, subclinical atherosclerosis, left ventricular mass index, composite and individual cardiovascular end points, and all-cause mortality. Prospective and cross-sectional analyses were performed to ascertain the association of study groups with these covariates. The MS plus MA group showed a consistently stronger association with the markers of systemic inflammation, subclinical atherosclerosis, and most clinical end points compared to the other study groups. In conclusion, stratification by MA can help identify a high-risk subset of nondiabetic patients with the MS.
We hypothesized that microalbuminuria (MA) in patients with the metabolic syndrome (MS) in the absence of diabetes mellitus would result in a pro-inflammatory milieu that would result in a greater prevalence of subclinical atherosclerosis and incident cardiovascular events. This subset of the MS population might therefore represent a high-risk category that would benefit from more aggressive medical therapy and lifestyle modifications.
Methods
The Multiethnic Study of Atherosclerosis (MESA) was a multicenter, longitudinal cohort study designed to investigate the risk factors and progression of subclinical atherosclerosis in a population free of cardiovascular disease at baseline. A total of 6,814 asymptomatic, self-identified men and women (52.85%) aged 45 to 84 years, including whites, Chinese, blacks, and Hispanics, were enrolled from July 2000 to August 2002 from 6 United States communities. Details of the study design and execution have been previously published. The institutional review board of Wayne State University approved the present study.
Subjects with a urine albumin/creatinine ratio (UAC) measured at the baseline visit were screened for the present analysis. The exclusion criteria included the unavailability of the urine albumin/creatinine ratio (n = 39); urinary infection within 2 weeks (n = 123); treated (n = 680) or untreated diabetes mellitus (n = 179), defined by a fasting glucose level of ≥126 or the use of oral hypoglycemic agents at the baseline visit ; and macroalbuminuria (n = 100); numbers were not mutually exclusive. A total of 5,809 nondiabetic subjects were included in the final analysis.
A random urine sample was collected at the first visit. Nephelometry and the Jaffe method were used to measure urine albumin and creatinine, respectively. MA was defined as a UAC of 17 to 250 mg/g in men or 25 to 355 mg/g in women according to the National Kidney Foundation Disease Outcomes Quality Initiative guidelines. The International Diabetes Federation guidelines were used to define the MS according to central adiposity defined by race-specific cutoffs, abnormal lipid profile, impaired glucose tolerance, and hypertension.
The study population was divided into 4 groups: (1) no MS and no MA, defined as the absence of both MS and MA, and served as the reference group for statistical comparisons; (2) MA- only, the presence of MA but the absence of the MS; (3) MS-only, the presence of the MS but the absence of MA; and (4) MS plus MA, the presence of both the MS and MA.
Coronary artery calcification was measured in each participant at the baseline visit using electron beam computed tomography or multidetector row helical computed tomography, as previously reported. The scans were quantified using the Agatston scoring system. The risk of having age-, gender-, and race-specific coronary artery calcification greater than the seventy-fifth percentile, as described by McClelland et al in the MESA population, were calculated using logistic regression analysis. The maximum internal carotid artery intimal-medial wall thickness (IMT) and common carotid arterial distensibility were measured by ultrasonography in all participants using a previously described protocol. Gender- and race-specific percentiles for both IMT and common carotid arterial distensibility were calculated. Significant atherosclerosis was defined as IMT greater than the seventy-fifth percentile or common carotid arterial distensibility less than the twenty-fifth percentile.
Cardiac magnetic resonance imaging was performed for the measurement of the left ventricular mass and volumes. The left ventricular mass was adjusted for height, weight, and gender using a method previously described by Bluemke et al to obtain the left ventricular mass index. The association of the study groups with a left ventricular mass index greater than the seventy-fifth percentile was tested using logistic regression analysis.
The inflammatory markers included interleukin-6, C-reactive protein, and fibrinogen. Gender- and race-specific seventy-fifth percentiles were calculated and used for analysis.
The participants were interviewed by a telephone interviewer every 9 to 12 months, in addition to the 3 MESA follow-up visits, to obtain information about hospitalizations, cardiovascular events, or death. The self-reported end points were verified by obtaining the medical records. The hospital records were obtained for an estimated 98% of the hospitalized cardiovascular events and 96% of the outpatient diagnoses. The clinical end points used for analysis included (1) all-cause cardiovascular disease (CVDA), a composite of myocardial infarction, resuscitated cardiac arrest, definite angina, probable angina (if followed by revascularization), stroke, stroke death, coronary heart disease death, and other atherosclerotic and cardiovascular death; (2) all-cause coronary heart disease, including myocardial infarction, resuscitated cardiac arrest, definite angina, probable angina, and coronary heart disease death; (3) myocardial infarction (fatal or nonfatal); (4) congestive heart failure; (5) stroke; and (6) all-cause mortality. CVDA was the primary end point in the present analysis, and all other outcomes were considered secondary.
Demographic information was obtained using questionnaires. Smoking was defined as current, previous, or never, and alcohol by the number of drinks per week. Hypertension was presented as a binary variable and defined by the current use of antihypertensive medications or blood pressure >140/90 mm Hg. Family history was defined as a history of heart attack at any age in a subject’s parents, siblings, or children. Total intentional exercise was presented as the number of metabolic equivalent tasks-hours/week. The estimated glomerular filtration rate was estimated using the abbreviated Modification of Diet in Renal Disease Study equation. Insulin resistance was calculated using the homeostasis model assessment formula.
All covariates were tested for normality by visual inspection using frequency distribution curves. The variables are presented as the mean ± SD and the median (interquartile range) based on normality. The baseline characteristics were compared using analysis of variance with planned contrasts, and the chi-square test for linear trend for categorical variables.
The 4 comparison groups were treated as a categorical variable. The no-MS plus no-MA group was the reference group in all comparisons unless otherwise specified. Odds ratios for the association of the comparison groups with (1) markers of subclinical atherosclerosis (coronary artery calcification greater than the seventy-fifth percentile, IMT greater than the seventy-fifth percentile, and common carotid arterial distensibility less than the twenty-fifth percentile); (2) left ventricular mass index greater than the seventy-fifth percentile; and (3) biomarkers (C-reactive protein, interleukin-6, and fibrinogen greater than the seventy-fifth percentile) were obtained using univariate and multivariate logistic regression analyses. Log transformation of skewed dependent variables was performed for multivariate linear regression analyses to ensure that significant associations were not missed (data not shown).
Gender- and race-stratified analyses were performed to test for effect modification. Confounding variables were identified through a literature search, and differential distribution in the comparison groups was ensured before introduction into the multivariate model. When comparing the MS plus MA group with the MS-only group, we also adjusted for insulin resistance (homeostasis model assessment). This was done because insulin resistance was noted to be greater in the MS plus MA group than in the MS group and could potentially confound the association between study groups and outcomes.
For the clinical end points, Kaplan-Meier estimates of the time-to-event end points were obtained. Cox proportional hazards regression analyses (stratified according to study groups) were used to estimate the hazard ratios. Two-tailed p values <0.05 were considered statistically significant. All statistical procedures were performed using Statistical Analysis Systems, version 9.1 (SAS Institute, Cary, North Carolina).
Results
The prevalence of the MS as defined by the International Diabetes Federation criteria was 30.24% in this nondiabetic population, with a range of 23.11% in the Chinese to 40.19% in the Hispanic subgroup ( Table 1 ). The prevalence, as defined by the National Cholesterol Education Program (National Cholesterol Education Program criteria), ranged from 25.86% in the entire population to 19.17% and 32.06% in the Chinese and Hispanic subgroups, respectively. MA was significantly less prevalent in the population without MS (7.7%) than in those with the MS (14.23%). As expected, a significant correlation between systolic blood pressure and UAC was noted (r = 0.21, p <0.0001) in the entire study cohort.
| Variable | All Participants (n = 5,809) | No MS + No MA (n = 3,740) | MA-Only (n = 312) | MS-Only (n = 1,507) | MS + MA (n = 250) | p Value for Linear Trend |
|---|---|---|---|---|---|---|
| Race | ||||||
| White | 2,421 (41.68%) | 1,607 (66.38%) | 98 (4.05%) | 621 (25.65%) | 95 (3.92%) | <0.0001 |
| Chinese | 688 (11.84%) | 465 (67.59%) | 64 (9.30%) | 132 (19.19%) | 27 (3.92%) | <0.0001 |
| Black | 1,513 (26.05%) | 1,010 (66.75%) | 98 (6.48%) | 342 (22.60%) | 63 (4.16%) | <0.0001 |
| Hispanic | 1,187 (20.43%) | 658 (55.43%) | 52 (4.38%) | 412 (34.71%) | 65 (5.48%) | <0.0001 |
| Age (years) | 61.76 ± 10.26 | 60.78 ± 10.16 | 66.34 ± 10.62 | 62.46 ± 9.95 | 66.94 ± 9.53 | <0.0001 |
| Women (%) | 53.41% | 56.47% | 43.25% | 49.51% | 42.37% | <0.0001 |
| Body mass index (kg/m 2 ) | 27.98 ± 5.29 | 27.02 ± 5.09 | 26.7 ± 5.02 | 30.36 ± 4.85 | 30.79 ± 5.23 | <0.0001 |
| Current smokers | 13.07% | 13.01% | 13.58% | 13.25% | 11.91% | 0.589 |
| Current alcohol use | 58.06% | 59.51% | 57.45% | 55.34% | 51.71% | 0.0628 |
| Intentional exercise >5 hours/week | 71.17% | 73.8% | 66.26% | 66.67% | 62.29% | <0.0001 |
| Low-density lipoprotein (mg/dl) | 118.97 ± 31.01 | 118.31 ± 30.27 | 115.89 ± 32.98 | 118.04 ± 32.38 | 115.74 ± 31.71 | 0.377 |
| Estimated glomerular filtration rate (ml/min/1.73 m 2 ) | 81.41 ± 17.38 | 79.48 ± 19.15 | 79.63 ± 16.28 | 79.63 ± 16.28 | 77.14 ± 18.10 | <0.0001 |
| Family history | 42.80% | 42.06% | 39.4% | 45.35% | 43.12% | <0.0001 |
| At least high school (%) | 83.75% | 86.43% | 79.14% | 79.32% | 72.88% | <0.0001 |
| Household income >30,000/year (%) | 62.31% | 65.12% | 58.9% | 57.59% | 48.31% | <0.0001 |
| Antihypertensive medications | 33.07% | 24.5% | 46.93% | 48.77% | 57.63% | <0.0001 |
| Statin use | 13.12% | 11.51% | 13.5% | 16.96% | 16.1% | <0.0001 |
| Currently using aspirin regularly | 22.57% | 20.87% | 23.77% | 26.04% | 28.63% | <0.0001 |
| Interleukin-6 (pg/ml) | 1.162 (0.75–1.81) | 1.03 (0.67–1.65) | 1.29 (0.87–2.04) | 1.39 (0.93–2.05) | 1.58 (1.10–2.48) | <0.0001 |
| C-reactive protein (mg/L) | 1.83 (0.80–4.12) | 1.49 (0.69–3.64) | 1.65 (0.83–3.56) | 2.59 (1.23–5.18) | 3.44 (1.63–5.5) | <0.0001 |
| Fibrinogen (mg/dl) | 343.03 ± 71.95 | 337.99 ± 70.78 | 357.36 ± 73.38 | 349.25 ± 72.94 | 369.56 ± 73.23 | <0.0001 |
| Blood pressure ≥130/85 mm Hg ⁎ | 40.81% | 23.88% | 71.17% | 77.71% | 88.98% | <0.0001 |
| Waist circumference (cm) | 97.08 ± 14.06 | 93.84 ± 13.46 | 94.36 ± 13.47 | 104.71 ± 11.83 | 106.57 ± 13.14 | <0.0001 |
| High-density lipoprotein cholesterol (mg/dl) | 51.71 ± 15 | 55.35 ± 15.04 | 53.04 ± 12.92 | 43.2 ± 11.32 | 42.72 ± 11.68 | <0.0001 |
| Triglycerides (mg/dl) | 126.62 ± 76.78 | 104.81 ± 57.65 | 107.13 ± 55.85 | 188.05 ± 89.72 | 186.17 ± 93.54 | <0.0001 |
| Homeostasis model assessment | 1.45 ± 1.10 | 1.15 ± 0.81 | 1.31 ± 0.87 | 2.13 ± 1.33 | 2.4 ± 1.69 | <0.0001 |
The study populations differed significantly for most risk factors. Compared to the MS-only and reference group, those in the MS plus MA group were more likely to be older, hypertensive, less physically active, less educated, and poor and to have lower renal function and greater insulin resistance, and to use medications more frequently.
All 3 groups (MA-only, MS-only, and MS plus MA) had significantly greater odds of having greater than seventy-fifth percentile levels of inflammatory biomarkers than the reference group. The levels of all 3 inflammatory markers were significantly greater in the MS plus MA groups compared to the MS-only group ( Table 2 ).
| Model | Interleukin-6 >75th Percentile ⁎ (n = 5,659) | C-Reactive Protein >75th Percentile ⁎ (n = 5,809) | Fibrinogen >75th Percentile ⁎ (n = 5,809) | |||
|---|---|---|---|---|---|---|
| OR | 95% CI | OR | 95% CI | OR | 95% CI | |
| A (unadjusted HR) | ||||||
| MA only † | 1.71 ‡ | 1.33–2.2 | 1.21 | 0.93–1.59 | 1.54 ‡ | 1.2–1.97 |
| MS-only † | 1.88 ‡ | 1.64–2.16 | 2.02 ‡ | 1.76–2.31 | 1.35 ‡ | 1.18–1.55 |
| MA + MS † | 3.17 ‡ | 2.42–4.16 | 3.37 ‡ | 2.58–4.4 | 2.31 ‡ | 1.76–3.02 |
| MA + MS vs MS-only | 1.69 ‡ | 1.27–2.23 | 1.67 ‡ | 1.26–2.2 | 1.70 ‡ | 1.28–2.26 |
| B (adjusted for age, gender, and ethnicity) | ||||||
| MA only † | 1.55 ‡ | 1.2–2.0 | 1.25 ‡ | 0.95–1.64 | 1.34 ‡ | 1.04–1.73 |
| MS only † | 1.83 ‡ | 1.59–2.10 | 2.04 ‡ | 1.78–2.34 | 1.31 ‡ | 1.14–1.50 |
| MA + MS † | 2.89 ‡ | 2.19–3.82 | 3.48 ‡ | 2.65–4.56 | 2.1 ‡ | 1.52–2.64 |
| MA + MS vs MS-only | 1.58 ‡ | 1.18–2.11 | 1.75 ‡ | 1.32–2.33 | 1.56 ‡ | 1.17–2.09 |
| C (adjusted for variables in model B and known confounders) § | ||||||
| MA only † | 1.46 ‡ | 1.09–1.96 | 1.05 | 0.79–1.41 | 1.32 ‡ | 1.01–1.73 |
| MS-only † | 1.67 ‡ | 1.42–1.96 | 1.95 ‡ | 1.68–2.26 | 1.21 ‡ | 1.04–1.41 |
| MA + MS † | 2.76 ‡ | 2.02–3.77 | 3.15 ‡ | 2.35–4.22 | 1.74 ‡ | 1.30–2.35 |
| MA + MS vs MS-only ¶ | 1.62 ‡ | 1.18–2.24 | 1.65 ‡ | 1.22–2.24 | 1.43 ‡ | 1.05–1.95 |
⁎ Gender- and race-specific seventy-fifth percentiles for IL-6, C-reactive protein, and fibrinogen.
† Reference group: no MS and no MA.
‡ Statistically significant (p <0.05).
§ Adjusted in addition for income, education, intentional exercise, alcohol use, antihypertensive medications, aspirin, statins, family history of coronary artery disease, and estimated glomerular filtration rate.
The prevalence of coronary artery calcification greater than the seventy-fifth percentile, IMT greater than the seventy-fifth percentile, and common carotid arterial distensibility less than the twenty-fifth percentile were greatest in the MS plus MA group. The MS plus MA group was most strongly associated with all markers of subclinical atherosclerosis before and after adjustments ( Table 3 ). A weaker association was found between the MS-only group and subclinical atherosclerosis. The MA-only group, in contrast, was associated only with common carotid arterial distensibility, with no association noted with IMT and coronary artery calcification. Compared to the MS-only group, the MS plus MA group had significantly greater coronary artery calcification, IMT, and lower common carotid arterial distensibility. The prevalence and odds of a left ventricular mass index greater than the seventy-fifth percentile were greater in the groups with MA (MA-only and MS plus MA; Table 3 ). Compared to the MS-only group, the MS plus MA group was noted to have a significantly greater left ventricular mass index.
| Model | CAC >75th Percentile | LVMI >75th Percentile | CAD c <25th Percentile | IMT >75th Percentile |
|---|---|---|---|---|
| A (unadjusted HR) | ||||
| MA only ⁎ | 1.34 (1.03–1.75) † | 1.75 (1.35–2.27) † | 2.75 (2.18–3.46) † | 1.68 (1.32–2.16) † |
| MS-only ⁎ | 1.58 (1.37–1.83) † | 1.09 (0.93–1.27) | 1.61 (1.41–1.85) † | 1.54 (1.34–1.77) † |
| MA + MS ⁎ | 2.58 (1.96–3.4) † | 2.03 (1.52–2.72) † | 4.66 (2.8–4.77) † | 2.86 (2.19–3.74) † |
| MA + MS vs MS only | 1.63 (1.22–2.17) † | 1.87 (1.37–2.55) † | 2.26 (1.71–2.99) † | 1.86 (1.40–2.46) † |
| B (adjusted for age gender and ethnicity) | ||||
| MA only ⁎ | 1.25 (0.95–1.64) | 1.82 (1.39–2.37) † | 1.96 (1.52–2.53) † | 1.70 (0.90–1.52) |
| MS-only ⁎ | 1.58 (1.37–1.83) † | 1.087 (0.92–1.28) | 1.52 (1.32–1.76) † | 1.43 (1.24–1.66) † |
| MA + MS ⁎ | 2.44 (1.84–3.23) † | 2.11 (1.57–2.85) † | 2.66 (1.99–3.55) † | 2.06 (1.55–2.74) † |
| MA + MS vs MS-only | 1.64 (1.22–2.2) † | 1.98 (1.44–2.71) † | 1.78 (1.32–2.39) † | 1.48 (1.10–1.99 † |
| C (adjusted for above variables and known confounders) ‡ | ||||
| MA only ⁎ | 1.29 (0.97–1.73) | 1.77 (1.34–2.36) † | 1.99 (1.52–2.60) † | 1.13 (0.86–1.5) |
| MS-only ⁎ | 1.53 (1.29–1.76) † | 0.99 (0.84–1.19) | 1.44 (1.23–1.68) † | 1.25 (1.07–1.47) † |
| MA + MS ⁎ | 2.3 (1.7–3.12) † | 2.08 (1.52–2.86) † | 2.73 (2–3.71) † | 1.88 (1.39–2.55) † |
| MA + MS vs MS-only § | 1.59 (1.16–2.18) † | 2.15 (1.53–3.02) † | 1.87 (1.36–2.57) † | 1.51 (1.10–2.07) † |
⁎ Reference group: no MS + no MA.
† Statistically significant (p <0.05).
‡ Also adjusted for income, education, intentional exercise, alcohol use, antihypertensive medication use, aspirin use, statin use, family history of coronary artery disease, estimated glomerular filtration rate.
A total of 226 CVDA events, 166 all-cause coronary heart disease events, 168 deaths (all cause), 79 myocardial infarctions, 69 strokes, and 71 new-onset congestive heart failure events occurred during a mean follow-up period of 4.74 years. The crude incidence rates of CVDA, all-cause coronary heart disease events, and myocardial infarction were greatest in the MS plus MA group than in the remaining groups. The incidence of all-cause mortality was greatest in the MA group alone, followed by the MS plus MA group ( Table 4 ). However, cardiovascular death constituted only 26 (15.47%) of the 168 deaths. The hazard of noncardiovascular death was again greatest in the MA-only group (data not shown).
| Group | Events | |||||||
|---|---|---|---|---|---|---|---|---|
| Death | CVDA | CHDA | MI | |||||
| Crude No. | Incidence | Crude No. | Incidence | Crude No. | Incidence | Crude No. | Incidence | |
| No MS + no MA (n = 3,740) | 82 | 4.51 | 105 | 5.94 | 77 | 4.34 | 38 | 2.13 |
| MA-only (n = 312) | 29 | 19.33 | 20 | 13.81 | 12 | 8.22 | 4 | 2.72 |
| MS-only (n = 1,507) | 41 | 6.16 | 76 | 11.96 | 60 | 9.38 | 27 | 4.18 |
| MA + MS (n = 250) | 16 | 14.80 | 25 | 24.69 | 17 | 16.47 | 10 | 9.60 |
| Total | 168 | 6.13 | 226 | 8.53 | 166 | 6.151 | 79 | 2.87 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
