The hypertriglyceridemic waist phenotype among women

Atherosclerosis 171 (2003) 123–130

The hypertriglyceridemic waist phenotype among women Michael J. LaMonte a,∗ , Barbara E. Ainsworth b , Katrina D. DuBose b , Peter W. Grandjean c , Paul G. Davis d , Frank G. Yanowitz a , J. Larry Durstine b a

Division of Cardiology, LDS Hospital, 8th Ave. and C Street, University of Utah School of Medicine, Salt Lake City, UT 84143, USA b Norman J. Arnold School of Public Health, University of South Carolina, Columbia SC, USA c Department of Health and Human Performance, Auburn University, Auburn, AL, USA d Department of Exercise and Sport Science, University of North Carolina at Greensboro, Greensboro, NC, USA Received 25 August 2002; received in revised form 30 June 2003; accepted 24 July 2003

Abstract Background: Elevated plasma triglycerides (TG) and waist girth (hypertriglyceridemic waist (HTGW)) has been associated with elevated insulin, small dense low-density lipoprotein (sLDL) particles, and Apo B in men. The HTGW has not been reported for women and the effect of cardiorespiratory fitness (“fitness”) on associations between HTGW and coronary risk factors is unknown. Purpose: To determine the prevalence of HTGW and the influence of fitness on the relationship between HTGW and coronary risk among 137 healthy women (54 ± 9 year; body mass index (BMI) = 28 ± 6 kg/m2 ). Methods: HTGW was defined as waist girth >88 cm and TG >150 mg/dl. The metabolic triad was defined as insulin >31 pmol/l, Apo B >69 mg/dl and LDL-C >84 mg/dl. Fitness was assessed with a maximal treadmill exercise test. Results: The sample prevalence of HTGW (n = 15) was 11% (95% CI = 5.7–16.0%). Apo B (P = 0.04) and insulin (P = 0.0001) increased across quintiles of waist girth, and LDL-C (P = 0.004) increased across quintiles of TG. Metabolic triad prevalence was highest (67%, n = 10) among HTGW women and lowest (22%, n = 26) among non-HTGW women. A trend for higher coronary heart disease CHD risk factors was observed among HTGW compared with non-HTGW women. Among the HTGW group, a trend for lower CHD risk factors was observed among fit (≥6.5 METs, n = 7) versus unfit women (<6.5 METs, n = 8). Sample size limitations prohibited meaningful tests of significant differences in CHD risk factors when stratified simultaneously on HTGW and fitness status. Conclusions: HTGW is associated with increased coronary risk factors similarly among women as reported for men. Higher fitness may improve the CHD risk profile among women with HTGW. © 2003 Elsevier Ireland Ltd. All rights reserved. Keywords: Cardiovascular disease; Women; Exercise; Lipids; Obesity

1. Introduction Despite progress in its prevention, detection and treatment, coronary heart disease (CHD) continues to exact an enormous economic and public health toll as the leading cause of death among US adults [1]. The latest report of the National Cholesterol Education Program Adult Treatment Panel III (NCEP ATP-III [2]) indicates that all CHD risk factors contribute to coronary disease similarly in women and men, and that most premature CHD in women occurs among those with co-existing risk factors [3]. Additional concern exists over increases in obesity and type 2 diabetes rates among women and minorities [1,4]. Clustering of CHD ∗ Corresponding author. Tel.: +1-801-408-1617; fax: +1-801-408-1229. E-mail address: [email protected] (M.J. LaMonte).

risk factors accompanies both conditions [5,6] and multiplicatively increases the risk of coronary events [7,8]. For many individuals, sudden death or nonfatal myocardial infarction is the first symptom of CHD [9]. Therefore, leading health organizations have called for the development of clinical assessment strategies to identify high-risk individuals without established CHD in order to initiate aggressive primary preventive therapies [1,9]. Lemiuex et al. [10] described a hypertriglyceridemic waist phenotype (HTGW) among men in the Quebec Cardiovascular Study, and suggested that this measurement might be an inexpensive clinical method of identifying elevated CHD risk among asymptomatic individuals. Lemiuex et al. [10] characterized the HTGW as a waist girth ≥90 cm and a plasma triglyceride (TG) concentration ≥177 mg/dl. Among 185 healthy Canadian men, the atherogenic metabolic triad of elevated plasma insulin, apolipoprotein B (Apo B), and

0021-9150/$ – see front matter © 2003 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.atherosclerosis.2003.07.008

124

M.J. LaMonte et al. / Atherosclerosis 171 (2003) 123–130

small dense low-density lipoprotein particles (sLDL) was highly prevalent among men with (84%) versus men without (10%) the HTGW. The presence of angiographically defined CHD was nearly four times more likely among men with the HTGW versus men without HTGW. While elevated TG and waist circumference have previously been associated with increased risk for CHD and diabetes among women [11–13], it is unknown if the HTGW predicts increased risk for CHD and diabetes similarly in women as men. The objective of this study was to examine cross-sectional data for the presence of the HTGW phenotype and its relationship with CHD risk factors among a tri-ethnic sample of healthy middle-aged women. We also investigated the influence of cardiorespiratory fitness on the relationship between the HTGW and selected CHD risk factors.

2. Methods

CAPS was to develop culturally sensitive physical activity surveys for use in epidemiological studies of health outcomes among diverse populations of women [15]. Blood samples were obtained from study participants so that parameters such as HDL cholesterol or fasting insulin could serve as potential sources of indirect validity of physical activity levels. The CAPS study objectives and design have been described in greater detail elsewhere [15–18]. Briefly, CAPS inclusion criteria were self-reported AA, NA, or CA ethnicity, absence of conditions that would preclude daily physical activities such as walking or house and family care, and the ability to read and write English well enough to complete a physical activity diary. Data collection was standardized according to an operations manual used at clinic sites in Albuquerque, New Mexico and Columbia, South Carolina. The Institutional Review Board at the Universities New Mexico and of South Carolina approved the study and all participants provided written informed consent. The study population is described in Table 1.

2.1. Study participants 2.2. Physical measures This study consisted of 44 African–American (AA), 45 native American (NA), and 46 Caucasian (CA) women enrolled in the Cross-Cultural Activity Participation Study (CAPS). CAPS was a 5-year study funded by the National Institutes of Health and Centers for Disease Control and Prevention as part of the Community Trials arm of the Women’s Health Initiative [14]. The primary objective of

Interviewer-administered surveys were used to obtain demographic information, personnel and family health histories. Body height and weight were measured with a calibrated clinical scale and stadiometer and body mass index (BMI) was computed as weight (kg) divided by height squared (m2 ). Waist circumference was taken midway

Table 1 Characteristics of study participants (mean ± S.D.)

n Age (year) BMI (kg/m2 ) Waist girth (cm) Systolic BP (mmHg) Diastolic BP (mmHg) Insulin (pmol/l) Glucose (mg/dl) Triglycerides (mg/dl) geometric mean Total cholesterol (mg/dl) LDL-C (mg/dl) HDL-C (mg/dl) Total/HDL-C ApoB (g/l) CRP (mg/dl) geometric mean Maximal METs Framingham 10 year (%) Current smoker (%) Estrogenic drug user (%) CVD (%) Diabetes (%)

African–American

Native American

Caucasian

46 56.6 30.9 89.0 129 79 79.9 93.6 80.6 165.3 82.5 65.3 2.5 0.8 0.44 7.2 4.6 8.7 47.8 8.7 10.8

45 50.7 ± 28.7 ± 88.9 ± 118 ± 76 ± 74.8 ± 93.6 ± 115.3 ± 169.2 ± 92.6 ± 50.9 ± 3.5 ± 0.8 ± 0.25 ± 9.1 ± 4.3 ± 4.4 11.1∗ 4.5∗ 15.5

46 54.3 ± 25.2 ± 78.4 ± 116 ± 76 ± 43.3 ± 84.6 ± 98.4 ± 170.6 ± 90.7 ± 57.6 ± 3.1 ± 0.7 ± 0.23 ± 10.0 ± 4.2 ± 6.5 47.8† 10.8† 0∗,†

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

10.1 6.1 13.1 18.6 8.7 57.7 30.4 35.4 34.5 33.7 18.6 0.8 0.4 0.34 1.4 0.5

8.9∗ 5.7 11.5 12.7∗ 9.2 60.2 36.1 52.8∗ 29.1 25.9 13.5∗ 0.9 0.2 0.18∗ 1.8∗ 0.5

10.1 4.8∗,† 11.2 18.7∗ 9.2 29.8∗,† 10.8 52.8 27.8 25.5 14.4∗,† 0.9 0.2 0.19∗ 1.9∗,† 0.4∗

HDL-C, high density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; ApoB, apolipoprotein B; CRP, C-reactive protein; CVD, prevalent cardiovascular disease; diabetes, prevalent diabetes. ∗ P < 0.05 with African–American. † P < 0.05 with native American.

M.J. LaMonte et al. / Atherosclerosis 171 (2003) 123–130

between the last rib and iliac crest [17] and the average of two measures was recorded to the nearest 0.1 cm. Resting blood pressure was auscultated as the first and fifth Korotkof sounds following 5 min of quiet sitting [20]. The average of two measures within 4 mmHg was recorded. Cardiorespiratory fitness (“fitness”) was quantified as the maximal METs (1 MET = 3.5 ml O2 kg−1 min−1 ) determined from the final speed and grade of a physician supervised maximal treadmill exercise test [18]. The average percentage of age-predicted maximal heart rate and average maximal perceived exertion (20-point Borg scale) achieved during the exercise tests were 96 ± 10 and 18 ± 2%, respectively [18]. Fitness levels were age adjusted with linear regression prior to further analysis. The Framingham 10-year probability of a coronary event was computed as previously described [18]. 2.3. Blood chemistry measures Following a 12 h fast and 24 h abstinence from exercise and smoking, 40 ml of antecubital venous blood was collected in Vacutainers containing K3-EDTA, centrifuged at 1500 × g at 4 ◦ C for 20 min, and frozen at −80 ◦ C until analysis [16]. Blood samples taken in New Mexico were appropriately packaged and shipped to the University of South Carolina’s exercise biochemistry laboratories where all blood chemistry analyses were performed according to CDC Lipid Research Clinics standards. Plasma total cholesterol (TC) was determined spectrophotometrically using a stable Liebermann–Burchard reagent [21]. Following precipitation with manganese chloride [22], high-density lipoprotein cholesterol (HDL-C) was determined with the same method described for TC. Low-density lipoprotein cholesterol (LDL-C) was estimated with the Friedewald equation [23] as plasma triglyceride concentrations were <400 mg/dl. Plasma triglyceride and glucose concentrations were determined with commercially available enzymatic assay kits (Sigma Diagnostics, St. Louis, MO, kit No. 339-10 and 315-100, respectively). Plasma insulin was measured by radioimmunoassay using the Coat-A-Count Insulin procedure (Diagnostic Products, Los Angeles, CA) [24]. Apo B concentrations were measured in the fasting plasma with an immunoelectrophoretic procedure [25]. C-reactive protein (CRP) was determined with the Dade-Behring high sensitivity immunoassay [26]. Intra-assay coefficients of variation were as follows: total cholesterol (1.8%), HDL-C (2.3%), TG (3.6%), glucose (4.9%), insulin (2.6%), Apo B (2.5%), and CRP (3.7% for a 1.78 mg/dl standard and 3% for a 0.74 mg/dl standard). 2.4. Variable definitions 2.4.1. HTGW Lemieux et al. [10] used data from their cohort study to define a waist girth of ≥90 cm and TG ≥177 mg/dl as being

125

strongly associated with increased coronary risk. CAPS was a cross-sectional study of healthy women and we were not able to identify specific values for waist and TG that correlated with increased prospective CHD risk. Therefore, we used waist girth ≥88 cm because this value was similar to Lemiuex et al. and conforms to NIH recommendations for a high-risk waist girth among women [19]. TG ≥150 mg/dl was used in accord with NCEP ATP-III recommendations [2]. 2.4.2. Metabolic triad Lemieux et al. [10] arbitrarily defined the metabolic triad as higher than median values for fasting insulin (≥48.5 pM/l), ApoB (≥0.96 g/l) and LDL peak particle size (<255.5 Å) among 38 nonobese (BMI <25 kg/m2 ) men. Similarly, we defined the metabolic triad based on higher than median values of fasting insulin (≥31 pM/l), Apo B (≥69 mg/dl), and LDL-C (≥81 mg/dl) among 52 CAPS women whose BMI was <25 kg/m2 . We did not have sLDL measures for CAPS women at the time of these analyses, therefore total LDL-C was used in this report. The correlation between sLDL and LDL-C is positive, but small (e.g., r = 0.13 [27]). Therefore, the prevalence of the metabolic triad may be underestimated in our study, which in turn, could result in weaker associations between the HTGW and the metabolic triad among CAPS women than reported by Lemieux et al.

2.5. Statistical analysis Summary statistics (mean, S.D., frequency) were computed for variables in accord with the assumptions of a normal distribution. TG and CRP values were highly skewed, therefore geometric means are reported descriptively. Non-transformed TG values are included in analyses of association with the metabolic triad because extreme elevations in triglyceride are a focus of this study. Spearman correlations were computed to determine bivariate associations between continuous variables. The prevalence of women with the metabolic triad was examined among subgroups stratified on quintiles of waist circumference and triglyceride concentration defined from the sample distributions for each variable. After assigning participants the median value of their respective waist and triglyceride category, tests for linear trends were performed by regressing each of the continuous metabolic risk factors across categories of waist and triglyceride treated as ordinal variables. To examine the influence of fitness on the relationship between the HTGW and CHD risk factors, we created a categorical variable that defined fit women as those in the upper 75th percentile of age-adjusted maximal METs (≥6.5 METs). All analyses were performed in SAS statistical software (version 8, SAS Institute, Carey, NC). P-values are two-sided with an α rate of 0.05.

126

M.J. LaMonte et al. / Atherosclerosis 171 (2003) 123–130 Table 4 CHD risk factors according to HTGW status

3. Results Participants were middle-aged, overweight, and had relatively low CHD risk factors (Table 1). AA women were older, had larger BMIs, higher systolic blood pressure, insulin, HDL-C and CRP concentrations, and lower fitness compared with NA and CA women. Estrogenic medication use was highest among AA and CA women. Among all women, Spearman correlations revealed waist girth was associated with insulin (r = 0.62, P < 0.01), TG (r = 0.36, P < 0.01), HDL-C (r = −0.44, P < 0.01), and fitness (r = −0.60, P < 0.01). TG was associated with insulin (r = 0.37, P < 0.01), Apo B (r = 0.64, P < 0.01), and HDL-C (r = −0.55, P < 0.01). Insulin was associated with HDL-C (r = −0.28, P < 0.01), and fitness (r = −0.53, P < 0.01). LDL-C was associated with Apo B (r = 0.57, P < 0.01) and both LDL-C and Apo B were associated with HDL-C (r = −0.23 and r = −0.31, respectively, P < 0.01). The prevalence of the HTGW was 11, 4.4, 17.8 and 10.9% for the overall sample, AA, NA, and CA, respectively (Table 2). The difference in HTGW prevalence between AA and NA 2 was marginally significant (χd.f.=1 = 4.2, P = 0.05). However the 95% confidence intervals were wide and overlapped

Table 2 Hypertriglyceridemic waist prevalence for the entire group and by race

Group Race African–American Native–American Caucasian

n

Prevalence (%)

95% CI

137

11.0

5.7–16.0

44 45 46

4.4 17.8 10.9

1.6–10.5 6.6–28.9 1.9–19.9

√ CI, confidence interval. 95% CI computed as P ± 1.96 [P(1 − P)/n]. Comparisons: African–American vs. native American, χ2 (d.f. = 1) = 4.2, P = 0.05. African–American vs. Caucasian, χ2 (d.f. = 1) = 0.89, P = 0.35. Native American vs. Caucasian, χ2 (d.f. = 1) = 1.4, P = 0.24.

n Age (year) BMI (kg/m2 ) Waist (cm) Maximal METs Systolic BP (mmHg) Glucose (mg/dl) Insulin (pM/l) Triglyceride (mg/dl) Total cholesterol (mg/dl) LDL-C (mg/dl) HDL-C (mg/dl) Total/HDL-C Apo B (mg/dl) CRP (mg/dl)a Framingham 10 year (%)

Non-HTGW

HTGW

122 54.3 (52.5–56.1) 27.6 (26.5–28.6) 83.7 (81.5–85.9) 8.8 (8.4–9.1) 121.5 (118.3–124.8) 87.8 (83.8–91.7) 61.5 (52.2–70.8) 93.9 (87.2–100.6) 168.3 (161.9–172.7) 88.5 (83.3–93.7) 60.0 (57.1–62.9) 2.9 (2.8–3.1) 72.0 (68.7–75.2) 0.28 (0.23–0.33) 4.1 (3.6–4.7)

15 50.8 (45.8–55.7) 33.7 (31.3–36.1) 99.3 (94.5–104.1) 8.0 (7.1–8.9) 118.9 (112.0–125.8) 113.5 (81.7–145.3) 102.5 (73.2–131.7) 232.3 (188.1–276.6) 177.7 (158.8–194.5) 89.0 (73.5–104.5) 41.0 (38.1–44.7) 4.5 (3.8–4.9) 93.6 (81.1–106) 0.38 (0.21–0.58) 6.6 (4.9–8.8)

Data presented as mean (95% confidence interval). HTGW (waist ≥ 88 cm, triglyceride ≥ 150 mg/dl). Non-HTGW (waist < 88 cm, triglyceride < 150 mg/dl). a Geometric mean.

making us less confident in the statistical significance of the point estimate probability. We, therefore, collapsed race for the remaining analyses. Table 3 presents mean concentrations of Apo B, insulin and LDL-C stratified across quintiles of waist girth and triglyceride concentration. Insulin levels were greater with increasing waist girth (P = 0.0001) and TG (P = 0.0002). Apo B elevations were more strongly associated with TG (P = 0.0001) than waist girth (P = 0.04). A steep rise in LDL-C was observed across increasing categories of TG (P = 0.002), however no relationship was observed between LDL-C and waist girth. Fig. 1 presents the percentage of women with the metabolic triad according to subgroups of waist girth and triglyceride concentration. Metabolic triad prevalence was higher (28.6–66.7%) among women who

Table 3 Mean levels of Apo B, insulin, and LDL-C according to categories of waist girth and triglyceride concentration (n = 137)

Waist quintiles 1 (median = 69.9 cm) 2 (median = 77.4 cm) 3 (median = 83.6 cm) 4 (median = 91.5 cm) 5 (median = 104.9 cm) Triglyceride quintiles 1 (median = 54.1 mg/dl) 2 (median = 73.2 mg/dl) 3 (median = 89.2 mg/dl) 4 (median = 122.5 mg/dl) 5 (median = 201.2 mg/dl)

Apo B (mean = 74 mg/dl)

Insulin (mean = 66 pM/l)

LDL-C (mean = 88.6 mg/dl)

62.8 78.3 76.8 77.7 75.5 P for trend = 0.04

32.9 36.9 61.7 83.5 110.6 P for trend = 0.0001

89.4 92.4 91.8 88.3 91.6 P for trend = 0.98

55.6 66.8 72.5 85.8 89.6 P for trend = 0.0001

48.2 50.5 55.9 81.3 91.9 P for trend = 0.0002

76.4 85.2 95.7 97.8 91.8 P for trend = 0.002

M.J. LaMonte et al. / Atherosclerosis 171 (2003) 123–130

127

Fig. 1. Prevalence of the metabolic triad (elevated fasting insulin, Apo B, and LDL-C) among subgroups of women according to waist girth and fasting triglyceride concentration. Number above each bar is the prevalence of the metabolic triad and the number on each bar is the 95% confidence interval √ computed as P ± 1.96 [P(1 − P)/n].

had either elevated waist girth (>88 cm) or TG (>150 mg/dl) compared to women with a waist girth <88 cm and TG <150 mg/dl (22.4%). CHD risk factors are shown according to HTGW status in Table 4. Sample size limitations precluded meaningful statistical tests for mean differences, therefore, data are shown de-

scriptively as the mean and 95% confidence interval. A trend for elevated CHD risk factors and increased 10-year Framingham probability was observed among HTGW women. The small number of women with the HTGW should, however, be taken into account when interpreting the meaningfulness of this trend.

Table 5 CHD risk factors among HTGW women stratified by level of fitness Non-HTGW

n Age (year) BMI (kg/m2 ) Waist (cm) Maximal METs Systolic BP (mmHg) Glucose (mg/dl) Insulin (pM/l) Triglyceride (mg/dl) Total cholesterol (mg/dl) LDL-C (mg/dl) HDL-C (mg/dl) Total/HDL-C Apo B (mg/dl) CRP (mg/dl)a Framingham 10-year (%)

HTGW

Unfit

Fit

Unfit

Fit

16 54.5 (49.8–59.1) 34.7 (31.3–38.2) 97.1 (90.4–103.7) 5.9 (5.3–6.4) 133.6 (123.7–143.6) 106.1 (79.5–132.6) 96.3 (63.1–129.5) 92.6 (77.6–107.6) 163.8 (146.7–181.0) 82.7 (65.1–100.3) 59.6 (56.7–62.5) 2.7 (2.2–3.8) 73.6 (60.5–86.8) 0.42 (0.15–0.69) 5.8 (3.5–7.7)

106 54.2 (52.2–56.2) 26.5 (25.6–27.5) 81.6 (79.5–83.7) 10.5 (8.8–12.5) 119.7 (116.3–123.1) 85.0 (82.6–87.4) 56.2 (46.9–62.1) 94.1 (86.6–101.6) 167.8 (162.1–173.6) 90.5 (85.3–95.7) 62.6 (49.8–75.4) 2.6 (2.3–3.1) 71.7 (68.5–75.0) 0.32 (0.24–0.40) 3.9 (3.3–4.4)

8 51.8 (42.6–61.1) 35.1 (30.8–39.4) 102.2 (93.9–110.5) 5.9 (5.7–6.5) 129.8 (122.1–137.4) 135.8 (74.7–196.8) 125.1 (74.9–175.3) 230.2 (143.8–316.5) 190.6 (174.1–207.2) 90.7 (60.5–120.9) 41.8 (36.4–47.2) 4.6 (3.8–5.2) 94.5 (74.3–114.6) 0.46 (0.20–0.39) 6.9 (4.2–9.4)

7 49.6 (43.7–55.4) 32.0 (29.3–34.7) 96.0 (90.3–101.7) 9.5 (8.4–10.7) 119.5 (116.3–122.8) 88.1 (81.6–94.6) 76.5 (50.1–103) 234.7 (189.3–280.1) 164.4 (132.8–196-1) 87.1 (71.2–102.9) 41.1 (35.6–46.6) 4.0 (3.3–4.8) 92.7 (72.3–113.3) 0.29 (0.23–0.79) 5.3 (2.2–10.5)

Data presented as mean (95% confidence interval). HTGW (waist ≥ 88 cm, triglyceride ≥ 150 mg/dl). Non-HTGW (waist < 88 cm, triglyceride < 150 mg/dl). Fit ≥ 6 METs; unfit < 6 METs. a Geometric mean.

128

M.J. LaMonte et al. / Atherosclerosis 171 (2003) 123–130

We next examined the influence of fitness on CHD risk factor profile among HTGW women. Fitness categories were defined according to the sample distribution of age-adjusted maximal METs, with women in the upper 75th percentile defined as fit (≥6.5 METs) and those in the lower 25th percentile defined as unfit (Table 5). Again, sample size limitations restricted our data analysis to only a descriptive presentation. Fit HTGW women tended to have better CHD risk factors than their unfit HTGW counterparts.

4. Discussion The American Heart Association has called for development of clinical methods to identify asymptomatic individuals at high-risk for new onset CHD [9]. Lemieux et al. [10] recently reported that two simple measures, waist girth >90 cm and fasting TG >177 mg/dl (the “hypertriglyceridemic waist phenotype”), might be an inexpensive clinical method for identifying men with elevated insulin, Apo B and sLDL concentrations, and hence, increased CHD and diabetes risk. The current study was undertaken to examine cross-sectional data for the presence of the HTGW phenotype and its relationship with CHD risk factors among a sample of healthy middle-aged women. We also investigated the influence of cardiorespiratory fitness on CHD risk factors among women with HTGW. The prevalence of the HTGW among our sample of healthy women was 11% and varied somewhat by race. Although the differences in HTGW prevalence across race were not statistically significant, our observation of race-related phenotypic variation in body fat distribution and lipid concentrations is consistent with other reports among racially diverse populations of women [28]. Lemieux et al. [10] did not directly report HTGW prevalence, however, based on their graphical data (Fig. 3, p. 181) we estimated the HTGW prevalence to be approximately 49% among their cohort. Differences in HTGW prevalence between the present study and Lemieux et al. is likely due to differences in case definition and the small sample size. Data from large ethnically diverse populations of men and women are needed to establish expected population prevalence estimates of the HTGW. Because waist girth and fasting TG measures are relatively inexpensive and readily available measures in a clinical setting, the possibility of using these indices to identify asymptomatic individuals at high-risk for CHD and diabetes has important public health implications for primary prevention. Lemieux et al. [10] showed CHD risk factor profiles worsened as waist girth and TG concentrations increased among Canadian men. Men whose waist was >90 cm and TG concentration was >177 mg/dl were nearly four times more likely to have angiographically defined CHD compared to men with a waist <90 cm and TG <177 mg/dl. A large proportion (>80%) of Canadian men with the HTGW also displayed the atherogenic metabolic triad of increased

fasting insulin, Apo B, and sLDL. This pattern of coronary risk factors has been associated with a five-fold increase in the 5-year incidence of CHD events among men [29]. In the current study, (Fig. 1), more than 66% of women with waist girths >88 cm and TG concentrations >150 mg/dl demonstrated the atherogenic metabolic triad. Our observation is consistent with national data showing a high prevalence of CHD and diabetes risk factor clustering, including elevated waist girth and TG, among women and minorities [5]. Differences in metabolic triad prevalence between our study and Lemieux et al.’s [10] might be related to the use of LDL-C rather than sLDL in our analysis. The correlation between sLDL and LDL-C is positive, but weak (r = 0.13 [27]), therefore, metabolic triad prevalence may be underestimated among CAPS women. Among CAPS women, TG was strongly associated with higher concentrations of all three metabolic triad components (Table 3), whereas TG was most strongly associated with sLDL among Canadian men [10]. In both studies elevated TG seemed to be better than waist girth in discriminating women with the metabolic triad (Fig. 1). These observations agree with recent findings of Austin et al. [11] who showed TG is an important, and perhaps overlooked CHD risk factor, with a slightly stronger effect on CHD risk among women than men. The CHD risk factor profile tended to be worse among HTGW versus non-HTGW women in CAPS (Table 4). This finding was also observed in Lemieux’s study of Canadian men [10]. The small number of women with HTGW precluded meaningful tests of statistical significance, however, some clinical relevance may exist. In our study, HDL-C concentrations among non-HTGW women met NCEP-ATP III [2] recommendations for cardioprotection (≥60 mg/dl), whereas HDL-C levels among HTGW women bordered on being clinically low (<40 mg/dl). Fasting glucose concentrations were in the clinically normal range (<110 mg/dl) among non-HTGW women, but were in the prediabetic range (110–126 mg/dl) among HTGW women. The Framingham 10-year probability was 6.6% (95% CI = 4.9–8.8%) among HTGW women compared to 4.1% (95% CI = 3.6–4.7%) among non-HTGW women. A point worth noting is the combination of high triglyceride/low HDL-C seen with the HTGW appears to confer increased CHD risk even among women whose TC and LDL-C are clinically acceptable (Table 4). Although the current study is hampered by a small sample size, taken together, data from CAPS and the Quebec Cardiovascular study support the HTGW as a reasonable method for identifying increased CHD risk among asymptomatic men and women. Lemieux et al. [10] did not include a measure of physical activity or fitness in their study, and therefore did not examine the influence of activity or fitness on the association between the HTGW and CHD risk factors. This aspect of our analysis was highly underpowered and statistical inferences could not be made. However, our findings are discussed to illustrate potentially meaningful clinical observations. When

M.J. LaMonte et al. / Atherosclerosis 171 (2003) 123–130

we stratified CAPS women with the HTGW into fitness categories based on maximal treadmill exercise tests, we observed a trend for lower CHD risk factors among fit versus unfit women (Table 5). Fasting insulin and glucose among fit HTGW women were clinically acceptable whereas unfit HTGW women had elevated insulin levels and fasting glucose concentrations that met diagnostic criteria for overt diabetes (≥126 mg/dl). Fit HTGW women also had lower total to HDL cholesterol ratios. Of interest was the lower CRP levels observed among fit compared with unfit HTGW women. This observation is particularly intriguing given the direct association between CRP and abdominal fat [30], CHD risk [31] and diabetes risk [32] previously reported among apparently healthy women. Higher fitness may attenuate large increases in CHD risk factors even among individuals with a high-risk phenotype such as the HTGW. This relationship may translate into lower risk for future coronary events. Because sample limitations prohibited meaningful tests as to the statistical significance of the effect fitness had on the relationship between HTGW and CHD risk, caution must be taken when interpreting our observations. Notwithstanding, routine fitness assessment may provide additional information that is clinically useful as a prognostic measure of individual CHD risk beyond traditional risk factor assessment [9,18]. Furthermore, fitness testing may motivate individuals to modify their physical activity levels to achieve goals currently ascribed for the primary and secondary prevention of CHD and diabetes [33,34]. The current study was a post hoc analysis of data collected during a larger investigation aimed at developing a culturally sensitive physical activity survey rather than examining associations between measures of activity and fitness with CHD risk factors. Consequently, the relatively small sample and cross-sectional design restricted the power of statistical analyses and limited the conclusions that could be drawn. Lack of objective data on coronary status, such as angiography or prospective CHD events, limited our ability to examine the precision of the HTGW in identifying increased CHD risk among asymptomatic individuals. However, we believe our analysis is important because it is the first study to consider the HTGW in a racially diverse sample of asymptomatic women, and to examine the potential modifying effects of fitness on the HTGW-CHD risk factor relationship. We conclude that HTGW is definable among women and appears to be an inexpensive and sensitive clinical tool for identifying asymptomatic individuals who have elevated CHD and diabetes risk factors. Higher cardiorespiratory fitness may attenuate risk factor elevations among individuals with the HTGW.

Acknowledgements The authors express gratitude to Drs. Vivian Heyward, Lisa Stolarczyk, Cheryl Addy, Jennifer Hootman, Melinda

129

Irwin, and Melicia Whitt, and Ms. Angela Morgan, for their work in collecting, compiling, and managing the CAPS data set. This work was supported by NIH WHI-SIP # 22W-U48/CCU409664 awarded to Dr. Ainsworth.

References [1] Cooper R, Cutler J, Desvigne-Nickens P, et al. Trends and disparities in coronary heart disease, stroke, and other cardiovascular diseases in the United States. Circulation 2000;102:3137–47. [2] Expert Panel on Detection, Evaluation, and Treatment of High Blood Pressure in Adults. Executive summary of the third report of the National Cholesterol Education Program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults. JAMA. 2001;285:2486–97. [3] Mosca L, Manson JE, Sutherland SE, et al. Cardiovascular disease in women. Circulation 1997;96:2468–82. [4] Mokdad AH, Bowman BA, Ford ES, et al. The continuing epidemics of obesity and diabetes in the United States. JAMA 2001;286:1195– 200. [5] Ford ES, Giles WH, Dietz WH. Prevalence of the metabolic syndrome among US adults. JAMA 2002;287:356–9. [6] Meigs JB, D’Agostino RB, Wilson PWF, et al. Risk variable clustering in the insulin resistance syndrome. Diabetes 1997;46:1594–600. [7] Trevisan M, Liu J, Bahsas FB, et al. Syndrome X and mortality: a population-based study. Am J Epidemiol 1998;148:958–66. [8] Wilson PFW, Kannel WB, Sibershatz H, D’Agostino RB. Clustering of metabolic factotrs and coronary heart disease. Arch Intern Med 1999;159:1104–9. [9] Greenland P, Smith SC, Grundy SM. Improving coronary heart disease risk assessment in asymptomatic people. Circulation 2001;104:1863–7. [10] Lemieux I, Pascot A, Couillard C, et al. Hypertriglyceridemic waist. A marker of the atherogenic metabolic triad (hyperinsulinemia; hyperapolipoprotein B; small, dense LDL) in men? Circulation 2000;102:179–84. [11] Austin MA, Hokanson JE, Edwards KL. Hypertriglyceridemia as a cardiovascular risk factor. Am J Cardiol 1998;81(4A):7B–12B. [12] Lissner L, Bjorkelund C, Heitmann BL, et al. Larger hip circumference independently predicts health and longevity in a Swedish female cohort. Obes Res 2001;9:644–6. [13] Boyko EJ, De Courten M, Zimmet PZ, et al. Features of the metabolic syndrome predict higher risk of diabetes and impaired glucose tolerance. Diabetes Care 2000;23:1242–8. [14] Riley PL, Finnegan LP. The prevention research centers program: collaboration in women’s health. J Women’s Health 1997;6:281–3. [15] Ainsworth BE, Irwin ML, Addy CL, et al. Moderate physical activity patterns of minority women: the Cross-Cultural Activity Participation Study. J Women’s Health 1999;8:805–13. [16] Drowatzky KL, Durstine JL, Irwin ML, et al. The association between physical activity, cardiorespiratory fitness, and lipoprotein(a) concentrations in a tri-ethnic sample of women: The Cross-Cultural Activity Participation Study. Vascular Med 2001;6:15–21. [17] Irwin ML, Mayer-Davis EJ, Addy CL, et al. Moderate-intensity physical activity and fasting insulin levels in women. Diabetes Care 2000;23:449–54. [18] LaMonte MJ, Durstine JL, Addy CL, et al. Physical activity, physical fitness, and Framingham 10-year risk score: The Cross-Cultural Activity Participation Study. J Cardiopulm Rehab 2001;21:63–70. [19] National Heart, Lung, and Blood Institute. Clinical guidelines in the identification, evaluation, and treatment of overweight and obesity in adults: the evidence report. Rockville, MD: National Institutes of Health, National Heart, Lung, and Blood Institute; 1998.

130

M.J. LaMonte et al. / Atherosclerosis 171 (2003) 123–130

[20] Joint National Committee on Prevention, Detection, Evaluation and Treatment of High Blood Pressure. The sixth report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Arch Intern Med. 1997;157:2413–45. [21] Kim E, Goldberg M. Serum cholesterol assay using a stable Leibermann-Buchard reagent. Clin Chem 1969;15:1171–9. [22] Warnick GR, Albers JJ. A comprehensive evaluation of the heparin-manganese precipitation procedure for estimation of high-density lipoprotein. J Lipid Res 1978;19:65–76. [23] Freidewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein in plasma without the use of the preparative ultracentrifuge. Clin Chem 1972;18:499–502. [24] Tietz N (Ed.), Fundamentals of Clinical Chemistry. Philadelphia, PA: WB Saunders; 1986. [25] Laurell CB. Quantitative estimation of proetines by electrophoresis in agarose gel containing antibodies. Ann Biochem 1966;15:45–52. [26] Rifai N, Tracy RP, Ridker PM. Clinical efficacy of an automated high-sensitivity C-reactive protein assay. Clin Chem 1999;45:2136– 41. [27] Campos H, Moye LA, Glasser SP, Stampfer MJ, Sacks FM. Low-density lipoprotein size, pravastatin, and coronary events. JAMA 2001;286:1468–74.

[28] Conway JM, Yanovski SZ, Avila NA, Hubbard VS. Visceral adipose tissue differences in black and white women. Am J Clin Nutr 1995;61:765–71. [29] LaMarche B, Tchernof A, Mauriege P, et al. Fasting insulin, and apolipoprotein B levels and low-density lipoprotein particle size as risk factors for ischemic heart disease. JAMA 1998;279:1955–61. [30] LaMonte MJ, Durstine JL, Yanowitz FG, et al. Cardiorespiratory fitness and C-reactive protein among a tri-ethnic sample of women. Circulation 2002;106:403–6. [31] Ridker PM, Hennekens CH, Buring JE, et al. C-reactive protein and other markers of inflammation in prediction of cardiovascular disease in women. N Engl J Med 2000;342:836–43. [32] Pradhan AD, Manson JE, Rifai N, et al. C-reactive protein, interleukin 6, and risk of developing type 2 diabetes mellitus. JAMA 2001;286:327–34. [33] Pearson TA, Blair SN, Daniels SR, et al. AHA guidelines for primary prevention fo cardiovascular disease and stroke: 2002 update. Circulation 2002;106:388–91. [34] Smith SC, Blair SN, Bonow RO, et al. AHA/ACC guidelines for preventing heart attack and death in patients with atherosclerotic cardiovascular disease: 2001 update. Circulation 2001;104: 1577–9.