Abstract
Background
Metabolic syndrome is a combination of risk factors for cardiovascular disease and diabetes, It has been reported to be increased in human immunodeficiency virus (HIV)–infected individuals.
Methods
In a cohort of HIV-infected adults we examined parameters that contribute to defining the metabolic syndrome and to estimating the 10-year risk of coronary heart disease (CHD). The study group consisted of 296 participants (217 men and 79 women) of mixed ethnicity with a mean age of 45.3 years.
Results
There was an appreciable prevalence of metabolic syndrome (30.0%), with the frequency increasing to 42.5% in those over 50 years of age. Those with the metabolic syndrome had a lower viral load. More women had abdominal obesity (59.5%) than men (20.7%, P < 0.001). The frequency of elevated plasma glucose was higher in females (37.2%) compared to males (16.9%, P = 0.004). High frequencies of decreased high-density lipoprotein cholesterol (HDL-C) and elevated blood pressure were seen in both sexes. Hypertriglyceridemia was less prevalent in African Americans. In those under 50 years of age, the 10-year CHD risk score for men was double that for women (6.2% vs 2.7%, P < 0.001). In older participants, the risk was similar between the sexes, with a third having scores over 10%.
Conclusions
The prevalence of metabolic syndrome was higher than in most other HIV cohorts. Those with the syndrome had significantly lower viral loads. Mean 10-year Framingham Cardiovascular Risk (FCR) scores were nearly doubled for those with metabolic syndrome. Both researchers and clinicians should consider age as well as sex when assessing patients with HIV infection for risks associated with metabolic syndrome.
Introduction
It is well established that dyslipidemia is a common metabolic hallmark of human immunodeficiency virus (HIV) infection and antiretroviral therapy (ART).1 Fat maldistribution (previously known as lipodystrophy) is a common comorbidity in HIV infection,2 and it is also associated with insulin resistance, dyslipidemia, and cardiovascular disease.3 Numerous reports have confirmed that ART is associated with changes in body fat. Studies have indicated some drug combinations to be more strongly associated with these changes.4 Body image is an important part of one's self-esteem, quality of life, and adherence to treatment.5 Some people may normally prefer a larger body size to avoid any possible hint of HIV infection and stigma, whereas others may strive for an underweight body size and even refuse therapy to maintain their weight. Either scenario, being overweight or underweight, has negative health effects.
Metabolic syndrome is a constellation of risk factors for cardiovascular disease. In addition to dyslipidemia and central obesity, the other risk factors are hypertension and insulin resistance. Assessment of metabolic syndrome risk factors and risk of coronary artery disease (CAD) in the context of HIV is important for optimum clinical management and quality of life. Metabolic perturbations of ART impacting glucose homeostasis and atherogenic profiles include: Elevated plasma glucose, decreased insulin sensitivity, elevated total cholesterol (TC), very-low-density lipoprotein cholesterol (VLDL-C), and low-density lipoprotein cholesterol (LDL-C), lower high-density lipoprotein cholesterol (HDL-C), and elevated plasma triglycerides.6 Furthermore, sex differences have been reported.7 Problems associated with these metabolic abnormalities are numerous and pose life-compromising and -threatening conditions, including CAD, stroke, diabetes mellitus, and renal failure.
The presence of metabolic syndrome is a risk factor for the development of nonalcoholic fatty liver disease (NAFLD).8 NAFLD can progress to the more severe nonalcoholic steato-hepatitis, and is itself an independent risk factor for cardiovascular disease (CVD).8 In the context of HIV, individuals have been shown to be at greater risk for NAFLD, probably due to use of ART and coinfection with hepatitis C.9,10
There is no consistent definition of metabolic syndrome. The World Health Organization (WHO) guidelines were originally proposed in 1999 as a working definition intended as a starting point for further refinement.11 One commonly used definition is that of the Adult Treatment Panel III (ATP III) of the National Cholesterol Education Program (NCEP),12 and this is the definition we use. The WHO and National Heart, Lung, and Blood Institute have recently issued an interim statement13 with the ultimate aim of a unified criteria. In this statement, the waist circumference cut off has country-specific definitions and the fasting glucose cut point has been lowered. In addition to waist circumference, we have used the ratio of waist to thigh, which is strongly associated with type 2 diabetes.14
Adeyemi and co-workers15 recently reported on the prevalence of metabolic syndrome in a cohort of HIV-infected individuals aged over 50 years. They also calculated the 10-year Framingham Cardiac Risk (FCR) score and found that women had much lower FCR scores than men.
Previously, we have described our cohort of HIV-infected adults participating in the Symptoms and Genetics Study of HIV-infected adults.16 In that report, we described personal characteristics and symptom prevalence in our sample of individuals living with HIV. In this report, we determine how our cohort compares with other samples of HIV infection in regard to parameters that define the presence of metabolic syndrome. We have assessed whether these parameters associated with the metabolic syndrome are associated with age, sex, viral load, CD4 count, or HIV medication use. We stratified our cohort to look at the prevalence of metabolic syndrome and FCR score in both younger and older patients, enabling a comparison with the previous study.15
Methods
Study design
This is a cross-sectional analysis of factors related to metabolic syndrome in the previously described Symptoms and Genetics Study cohort of 350 HIV-infected adults.16 Eligible participants were English-speaking adults at least 18 years of age who had been diagnosed with HIV at least 3 months prior to enrollment. They were recruited using flyers posted at approved local HIV clinics and approved community sites. Data were collected at the baseline study visit between April, 2005, and December, 2007. Study visits were conducted at the University of California San Francisco (UCSF) General Clinical Research Center. In addition to completing the demographic and symptom questionnaires, the baseline assessment included a fasting blood sample for genetic and metabolic analysis, and obtaining anthropometric measures of height, weight, and circumference of neck, waist, hips, and thighs. Participants also provided urine samples for toxicology screening using RediCup® (Redwood Toxicology Laboratory, Inc, Santa Rosa CA). Written informed consent was obtained for all participants, and the UCSF Committee of Human Research approved the study protocol.
The study sample was not limited to self-identified men and women, but also included transgender adults. The criteria for metabolic syndrome are different for men and women; no guidelines are available for transgender individuals. Given that most of the transgender participants were taking hormone therapy that could influence lipid metabolism, categorizing them by sex would be inaccurate. Therefore, the 23 self-identified transgender adults were excluded from this analysis. Participants who tested positive for cocaine or methamphetamine by urine toxicology screening (RediCup,® Redwood Toxicology Laboratory, Inc, Santa Rosa CA) during data collection were also excluded. Inclusion of these individuals might have led to an over-diagnosis of metabolic syndrome because these drugs are known to elevate blood pressure. We failed to get a blood sample from 1 participant.
Measures
Participants reported demographic information, health and HIV history, and current HIV medication regimen. Medication regimens were subsequently coded into one of five categories: (1) Nucleoside/nucleotide reverse transcriptase inhibitors (NRTI) only; (2) non NRTI-based regimen with an NRTI backbone; (3) protease inhibitor (PI) based with an NRTI backbone; (4) other regimens; or (5) no HIV medications. Most recent CD4, viral load, and fasting glucose values were obtained from the subject's medical record. Trained research staff obtained anthropometric and blood pressure measurements. Body mass index (BMI) was calculated as the weight in kilograms divided by squared height in meters.
A fasting blood sample was also collected at the study visit. The cholesterol and triglyceride (TG) content of plasma and lipoproteins was measured by automated chemical analysis using standards provided by the Centers for Disease Control.17 HDL-C was determined after precipitation of apoliprotein B (apoB)-containing lipoproteins with dextran sulfate and magnesium.18 LDL-C was calculated as total cholesterol (TC) minus HDL-C plus VLDL-C or by using the Friedewald equation when the TG was <400 mg/dL.19
Metabolic syndrome was defined according to the NCEP ATP III guidelines.12 Three metabolic abnormalities out of a possible total of five were required: (1) Waist circumference >102 cm for men or >88 cm for women; (2) plasma triglycerides ≥150 mg/dL; (3) HDL-C <40 mg/dL for men or <50 mg/dL for women; (4) blood pressure ≥130/85 mmHg; (5) fasting glucose ≥110 mg/dL. The FRC score was calculated for each participant as described in the Framingham Heart Study.20
Statistical analysis
Anthropometric, demographic, and clinical characteristics were summarized by descriptive statistics. Data are presented throughout as means ± standard deviation (SD). Group differences were assessed using chi-squared tests, Mann–Whitney tests, t-tests, or analysis of variance (ANOVA), as appropriate. CD4 counts were square root–transformed prior to analysis. Logistic regression analysis was conducted to identify predictors of metabolic syndrome while controlling for other factors. For all analyses, P < 0.05 was considered statistically significant. Statistical tests were performed using SPSS version 16.0 (SPSS Inc., Chicago, IL) and STATA (StataCorp LP, College Station, TX).
Results
After exclusion of 23 transgender individuals and 31 of those who tested positive for cocaine or methamphetamines, the sample for this analysis included a total of 296 HIV-infected participants. The demographic and clinical characteristics are described in Table 1. A total of 267 individuals had sufficient data to enable an evaluation of presence or absence of metabolic syndrome. Compared to the 187 participants without metabolic syndrome, those categorized as having metabolic syndrome (n = 80) were older, had a lower viral load, and fewer were taking NRTI ART. There were no differences in CD4 count. A third of participants had been diagnosed with hepatitis C, but there was no difference in the prevalence of metabolic syndrome. The percentage of those taking hormone or steroid therapy was also similar. Although almost twice as many participants with metabolic syndrome were taking lipid-controlling medication, this was not a statistically significant finding. The overall frequency of reported alcohol use was similar. Those with metabolic syndrome were at almost twice the risk of developing coronary heart disease (CHD), as assessed using the Framingham 10-year percentage risk scores compared to those without (see Table 1), with a mean difference in the scores of 4.43 [95% confidence interval (CI) 2.99–5.88; P < 0.001]. If we had used the new interim guidelines for defining metabolic syndrome,13 with the cut point of 100 mg/dL for fasting glucose, 5 additional participants would have met the criteria.
Table 1.
Demographic Characteristics of Study Participants, HIV Status, and Medication History
| |
|
|
Metabolic syndrome |
||
|---|---|---|---|---|---|
| n | Total 296 | +80 | −187 | P value | |
| Female (%) | 26.7 | 30.0 | 23.5 | 0.266 | |
| Age (years) | 45.3 ± 8.3 | 47.0 ± 9.3 | 44.5 ± 7.6 | 0.032 | |
| Range | 22–77 | 26–77 | 22–64 | ||
| Race | 0.457 | ||||
| Caucasian (%) | 42.6 | 37.5 | 43.3 | ||
| African American (%) | 37.2 | 36.2 | 38.5 | ||
| Mixed/Asian/Other (%) | 10.5 | 15.0 | 9.1 | ||
| Hispanic/Latino (%) | 9.8 | 11.2 | 9.1 | ||
| Time since HIV diagnosis (years) | 12.1 ± 6.9 | 12.0 ± 7.1 | 12.2 ± 6.8 | 0.805 | |
| Ever diagnosed with AIDS (%) | 53.7 | 55.0 | 54.5 | 0.946 | |
| Time since AIDS diagnosis (years) | 7.8 ± 5.8 (157) | 6.9 ± 5.8 (45) | 8.4 ± 5.8 (100) | 0.135 | |
| Hepatitis C diagnosis (%) | 33.8 (281) | 36.8 (76) | 33.5 (176) | 0.611 | |
| CD4 count (cells/mL) | 455 ± 267 (281) | 478 ± 260 (76) | 447 ± 266 (179) | 0.324 | |
| Range | 4-1740 | 27-1088 | 4-1740 | ||
| CD4 <200 (%) | 16.7 | 14.5 | 17.3 | 0.575 | |
| Viral load (VL, copies/mL) | |||||
| Detectable VL (≥50; %) | 47.6 (275) | 43.2 (74) | 47.4 (175) | 0.545 | |
| VL ≥10,000 (%) | 18.9 (275) | 10.8 (74) | 21.1 (175) | 0.053 | |
| Median detectable VL | 4529 (131) | 2990 (32) | 5533 (83) | 0.045 | |
| HIV medications | 0.331 | ||||
| NRTI based (%) | 5.1 | 1.2 | 7.5 | ||
| NNRTI based (%) | 19.3 | 18.8 | 18.7 | ||
| Protease inhibitor based (%) | 42.2 | 46.2 | 42.8 | ||
| Other regimen (%) | 3.7 | 2.5 | 3.7 | ||
| No current HIV medication (%) | 29.1 | 31.2 | 27.3 | ||
| Taking hormones or steroids (%) | 16.6 | 15.0 | 16.6 | 0.748 | |
| Taking lipid-controlling medication (%) | 11.8 | 15.2 | 8.6 | 0.115 | |
| Framingham 10-year % CHD riska | 6.7 ± 5.5 (278) | 10.0 ± 7.6 (74) | 5.6 ± 4.1 (183) | <0.001 | |
| Relative risk | 0.92 ± 0.58 | 1.25 ± 0.71 | 0.79 ± 0.46 | <0.001 | |
Values are percentages or means ± standard deviation (SD) with the numbers of subjects in parentheses (except median is provided for viral load). Differences between those with or without metabolic syndrome tested by chi-squared test or t-test, with the exception of viral load where the MannWhitney test was used. CD4 count was square root–transformed prior to testing.
As previously defined 20.
Abbreviations: HIV, Human immunodeficiency virus; AIDS, acquired immunodeficiency syndrome; NRTI, nucleoside/nucleotide reverse trancriptase inhibitor; NNRTI, non-NRTI; CHD, coronary heart disease.
There were significant differences in the racial/ethnic composition between male and female participants (Table 2). Women had a higher mean BMI than men, which was reflected in a higher percentage of men with normal weight (48.4%) compared to women (29.1%), and a higher frequency of obesity among women (39.2%) compared to men (15.7%). The percentages of those who were overweight (BMI values, 25–29.9 kg/m2) were similar between men and women (Table 2). Both waist-to-hip ratio, and waist-to-thigh ratio were higher in men than women. We found, in men, that the waist-to-thigh ratio was strongly associated with the presence of diabetes. The ratios in diabetics and nondiabetics were 2.14 ± 0.21 (n = 13) and 1.87 ± 0.20 (n = 203), respectively (P < 0.001). There was a less pronounced effect in women with the ratio in diabetics being 1.91 ± 0.25 (n = 9) and nondiabetics 1.75 ± 0.22 (n = 70) (P = 0.041).
Table 2.
Characteristics of Study Participants by Sex: All Participants
| n | Males 217 | Females 79 | P value | |
|---|---|---|---|---|
| Age (years) | 45.1 ± 8.3 | 46.0 ± 8.2 | 0.390 | |
| Race | <0.001 | |||
| Caucasian (%) | 49.8 | 22.8 | ||
| African American (%) | 28.1 | 62.0 | ||
| Mixed/Asian/Other (%) | 11.5 | 7.6 | ||
| Hispanic/Latino (%) | 10.6 | 7.6 | ||
| BMI (kg/m2) | 26.0 ± 4.8 | 29.1 ± 6.3 | <0.001 | |
| BMI <25 | 48.4 | 29.1 | 0.003 | |
| BMI 25–29.9 | 35.9 | 31.6 | 0.492 | |
| BMI ≥30 | 15.7 | 39.2 | <0.001 | |
| Neck circumference (cm) | 38.4 ± 2.8 | 34.5 ± 3.0 | <0.001 | |
| Waist circumference (cm) | 93.9 ± 12.2 | 93.8 ± 14.4 | 0.946 | |
| Waist-to-hip ratio | 0.92 ± 0.07 | 0.86 ± 0.08 | <0.001 | |
| Waist-to-thigh ratio | 1.89 ± 0.21 | 1.77 ± 0.23 | <0.001 | |
| Central obesity (%)a | 61.3 | 64.6 | 0.608 | |
| Systolic blood pressure (mmHg) | 126 ± 16 | 124 ± 19 | 0.248 | |
| Diastolic blood pressure (mmHg) | 75.6 ± 10.3 | 71.7 ± 11.2 | 0.005 | |
| Diabetes (%) | 6.0 (216) | 11.4 | 0.12 | |
| Current smoker (%) | 54.9 (213) | 54.5 (77) | 0.954 | |
| Metabolic syndrome (%)b | 28.1 (199) | 35.3 (68) | 0.266 | |
| Glucose > 110 mg/dL (%) | 16.9 | 37.2 | 0.004 | |
| Waist circumference >102 cm men; >88 cm women (%) | 20.7 | 59.5 | <0.001 | |
| TG ≥150 mg/dL (%) | 49.1 | 32.1 | 0.01 | |
| HDL-C <40 mg/dl men; <50 mg/dl women (%) | 40.7 | 42.3 | 0.805 | |
| Blood pressure ≥130/85 mmHg (%) | 53.5 | 45.6 | 0.23 | |
| Framingham 10-year CHD risk (%)c | 7.6 ± 5.4 (201) | 4.6 ± 5.4 (77) | <0.0001 | |
| Relative riskd | 0.87 ± 0.52 | 1.03 ± 0.71 | 0.077 |
Values are percentages or means ± standard deviation (SD) with the numbers of subjects in parentheses.
Post hoc analysis (Bonferonni-adjusted) showed a significant difference between Caucasians and African Americans (P < 0.01) and African Americans versus Mixed/Asian/Other (P = 0.024).
As defined by the WHO.11
As defined by the ATP III guidelines.12
As defined by ref.20
The relative 10-year CHD risk was calculated by dividing an individual's risk score by the average value for a person of the same sex and age in the Framingham population.
Abbreviations: BMI, Body mass index; TG, triglycerides; HDL-C, high-density lipoprotein cholesterol; CHD, coronary heart disease; WHO, World Health Organization.
Although there was no difference in systolic blood pressure, diastolic blood pressure was higher in men (Table 2). There was no sex difference in the frequency of diabetes or smoking. Although the prevalence of metabolic syndrome was similar for men (28.1%) and women (35.3%), there were significant differences in the underlying parameters that define this disorder. Women were more likely to have metabolic syndrome than men because of relatively high waist circumference or raised levels of glucose, whereas men were more likely to have elevated TGs. The prevalence of low HDL-C was similar between the sexes. Overall, 63.3% of participants had a total of two or more metabolic abnormalities; 59.9% of men and 72.5% of women (P = 0.064). Similarly, 11.0% had a total of four or more metabolic abnormalities—9.5% of men and 15.3% of women (P = 0.180). Men reported a higher alcohol use than women, but this difference was not significant (15.2% among men and 7.6% among women, P = 0.087; data not shown).
Both male and female participants had similar levels of TC [191 ± 52 vs. 190 ± 45 mg/dL, respectively; not significant]. These values are lower than those of comparable age and sex in the U.S. population at large21; TC was 10% lower in our sample of men and 8% lower in our sample of women. Male and female participants also have similar levels of LDL-C (110 ± 37 vs. 114 ± 35 mg/dL, respectively; N.S.). These values are lower than the U.S. population at large21; LDL-C was 23% lower in our sample of men and 12% lower in our sample of women.
We calculated the 10-year Framingham CHD risk scores20 (Table 2)and found a difference, as expected, between men (7.6 ± 5.4%) and women (4.6 ± 5.4%; P < 0.001). Occurrence of smoking, one of the major factors influencing the score, was similar between men and women. The relative risk was calculated by dividing an individual's risk score by the average value for a person of the same sex and age in the Framingham population.20 We found that men had somewhat lower relative risk (0.87 ± 0.52) than women (1.03 ± 0.71), but this was not statistically different (Table 2).
The occurrence of metabolic syndrome was not significantly different among the four ethnic/racial groups: Caucasians = 27.0%; African Americans = 28.7%; Hispanic/Latinos = 34.6%; others = 41.4% (P = 0.457). The overall 10-year CHD risk was also similar: Caucasians = 7.4 ± 5.8%; African Americans = 6.0 ± 5.4%; Hispanic/Latino = 5.8 ± 3.5%; others = 7.6 ±6.1% (P = 0.162).
Logistic regression analysis showed that with men age was the only significant and strong predictor of odds for having metabolic syndrome. The model only explained 3.5% of the variance for having metabolic syndrome. For women, age was not a predictor. Regression analysis allowed an examination of the extent to which ethnicity confounded the differences seen between men and women. With the exception of frequency of elevated TGs, the differences in parameters between the sexes seen in Table 2 were still highly significant after accounting for ethnicity. The waist-to-thigh ratio was lower among African Americans, in both men and women, compared to all other ethnicities. The waist-to- hip ratio was also lower, but only in African-American men when compared to men of other ethnicities. Diastolic blood pressure was higher among African-American women compared to women of other ethnicities. The frequency of elevated TGs was lower among African Americans compared to other ethnicities, and within each ethnic group there were no sex differences.
To compare the prevalence of metabolic syndrome and CHD risk scores in younger and older participants, the sample was dichotomized to less than 50 years of age and 50 years of age or older. The differences in ethnic/racial origin were more pronounced in the younger group (Table 3)compared to the older group (Table 4). The relative sex differences in BMI, neck circumference, waist-to-hip ratio, and waist-to-thigh ratio observed in the total population were seen in both the younger and older groups. The higher diastolic blood pressure among men was more pronounced among older participants. There was a trend toward a higher occurrence of diabetes in younger women (9.3%) compared to younger men (2.6%; P = 0.057). The prevalence of diabetes in men and women was similar in the older group of participants. The proportion of those who smoked was higher in younger men (60.4%) compared to older men (42.2%; P = 0.014), and there was a similar trend among women (61.5% vs 40.0%; P = 0.076). Although there was no significant difference in the prevalence of metabolic syndrome between males and females in either age group, the occurrence in older men was nearly doubled (41.0%) compared to younger men (22.5%, P = 0.007). In older women, although the prevalence was higher than in younger women (47.4% vs 30.6%), it was not significantly so (P = 0.194).
Table 3.
Characteristics of Study Participants by Sex: Participants under 50 Years of Age
| n | Males 151 | Females 54 | P value | |
|---|---|---|---|---|
| Age (years) | 40.8 ± 5.6 | 42.0 ± 6.0 | 0.200 | |
| Race | <0.001 | |||
| Caucasian (%) | 46.4 | 13.0 | ||
| African American (%) | 27.2 | 66.7 | ||
| Mixed/Asian/Other (%) | 11.9 | 9.3 | ||
| Hispanic/Latino (%) | 14.6 | 11.1 | ||
| BMI (kg/m2) | 26.0 ± 4.9 | 29.1 ± 6.2 | 0.001 | |
| Neck circumference (cm) | 38.3 ± 2.6 | 34.6 ± 3.2 | <0.001 | |
| Waist circumference (cm) | 92.4 ± 12.2 | 93.5 ± 14.3 | 0.627 | |
| Waist-to-hip ratio | 0.90 ± 0.07 | 0.85 ± 0.07 | <0.001 | |
| Waist-to-thigh ratio | 1.83 ± 0.19 | 1.74 ± 0.22 | 0.008 | |
| Central obesity (%)a | 55.0 | 59.3 | 0.585 | |
| Systolic blood pressure (mmHg) | 126 ± 15 | 122 ± 18 | 0.141 | |
| Diastolic blood pressure (mmHg) | 75.4 ± 9.9 | 72.2 ± 11.4 | 0.052 | |
| Diabetes (%) | 2.6 | 9.3 | 0.057 | |
| Current smoker (%) | 60.4 (149) | 61.5 (52) | 0.885 | |
| Metabolic syndrome (%)b | 22.5 (138) | 30.6 (49) | 0.255 | |
| Glucose >110 mg/dL (%) | 12.5 | 36.4 | 0.002 | |
| Waist circumference >102cm men; >88 cm women (%) | 17.2 | 55.6 | <0.001 | |
| TG ≥150 mg/dL (%) | 46.7 | 26.4 | 0.01 | |
| HDL-C <40 mg/dL men; <50 mg/dL women (%) | 42.1 | 37.7 | 0.583 | |
| Blood pressure ≥130/85 mmHg (%) | 50.3 | 42.6 | 0.329 | |
| Framingham 10-year CHD risk (%)c | 6.2 ± 4.1 (143) | 2.7 ± 1.7 (52) | <0.001 | |
| Relative riskd | 0.94 ± 0.55 | 1.10 ± 0.66 | 0.099 |
Values are percentages or means ± standard deviation (SD). If the number of subjects differs from that at the top of the column, the correct number is in parentheses.
Post hoc analysis (Bonferonni-adjusted) showed a significant difference between Caucasians and African Americans (P < 0.01).
As defined by the WHO.
As defined by the ATP III guidelines.12
As previously defined.20
The relative 10-year CHD risk was calculated by dividing an individual's risk score by the average value for a person of the same sex and age in the Framingham population.
Abbreviations: BMI, Body mass index; TG, triglycerides; HDL-C, high-density lipoprotein cholesterol; CHD, coronary heart disease; WHO, World Health Organization; ATP III, Adult Treatment Panel III.
Table 4.
Characteristics of Study Participants by Sex: Participants 50 Years of Age or Over
| n | Males 66 | Females 25 | P value | |
|---|---|---|---|---|
| Age (years) | 54.8 ± 4.5 | 54.7 ± 4.3 | 0.917 | |
| Race | 0.235 | |||
| Caucasian (%) | 57.6 | 44.0 | ||
| African American (%) | 30.3 | 52.0 | ||
| Mixed/Asian/Other (%) | 10.6 | 4.0 | ||
| Hispanic/Latino (%) | 1.5 | 0.0 | ||
| BMI (kg/m2) | 26.2 ± 4.6 | 29.3 ± 6.7 | 0.044 | |
| Neck circumference (cm) | 38.5 ± 3.2 | 34.2 ± 2.7 | <0.001 | |
| Waist circumference (cm) | 97.4 ± 11.5 | 94.5 ± 14.9 | 0.326 | |
| Waist-to-hip ratio | 0.95 ± 0.07 | 0.87 ± 0.09 | <0.001 | |
| Waist-to-thigh ratio | 2.03 ± 0.21 | 1.82 ± 0.23 | <0.001 | |
| Central obesity (%)a | 75.8 | 76.0 | 0.981 | |
| Systolic blood pressure (mmHg) | 127 ± 17 | 127 ± 20 | 0.98 | |
| Diastolic blood pressure (mmHg) | 76.0 ± 11.0 | 70.5 ± 10.9 | 0.035 | |
| Diabetes (%) | 13.6 | 16.0 | 0.774 | |
| Current smoker (%) | 42.2 (64) | 40.0 | 0.851 | |
| Metabolic syndrome(%)b | 41.0 (61) | 47.4 (19) | 0.623 | |
| Glucose >110 mg/dL (%) | 26.0 | 40.0 | 0.37 | |
| Waist Circumference >102cm men; >88cm women (%) | 28.8 | 68.0 | 0.001 | |
| TG ≥150 mg/dL (%) | 54.8 | 44.0 | 0.36 | |
| HDL-C <40 mg/dL men; <50 mg/dL women (%) | 37.3 | 52.0 | 0.211 | |
| Blood pressure ≥130/85 mm Hg (%) | 60.6 | 52.0 | 0.457 | |
| Framingham 10-year CHD risk (%)c | 10.9 ± 6.7 (57) | 8.5 ± 7.8 | 0.159 | |
| Relative riskd | 0.70 ± 0.41 | 0.89 ± 0.80 | 0.266 |
Values are percentages or means ± standard deviation (SD) with the numbers of subjects in parentheses.
As defined by the WHO.11
As defined by the ATP III guidelines.12
As previously defined.20
The relative 10-year CHD risk was calculated by dividing an individual's risk score by the average value for a person of the same sex and age in the Framingham population.
Abbreviations: BMI, Body mass index; TG, triglycerides; HDL-C, high-density lipoprotein cholesterol; CHD, coronary heart disease; WHO, World Health Organization; ATP III, Adult Treatment Panel III.
There were some differences between the two age groups in the parameters that define the metabolic syndrome. As in the total population, more women had elevated levels of glucose than men of a similar age. In the younger group, there was a pronounced effect, with three times more women affected than men (36.4% vs. 12.5%, P = 0.002) (Table 3). However, there was no significant sex difference in the older group. Whereas the percentage of participants, both male and female, with a high waist circumference increased with age, the large sex disparity was maintained, with three times as many women in the younger group and over twice as many in the older group affected (Tables 3 and 4). There were more men in the under 50-year-old group with raised TGs (46.7%) than women (26.4%; P = 0.010). There was no significant sex difference in the older group. As with the overall cohort, in the two age groups, there were no significant sex differences in the prevalence of decreased levels of HDL-C or of elevated blood pressure, nor were there changes in these parameters with age.
The FCR CHD score in younger men was double that seen in younger women (Table 3). This sex difference was not seen in the older age group, but the score increased in older versus younger men (10.9 ± 6.7 vs. 6.2 ± 4.1; p < 0.001) and older versus younger women (8.5 ± 7.8 vs. 2.7 ± 1.7; P < 0.001). The occurrence of high CHD risk scores (over 10%) in younger men compared to younger women was particularly striking; whereas 8.4% of men were in this category, there were no women in this category (P = 0.031). In older participants, both males and females, the proportion with risk scores over 10% had risen sharply and there was no longer any difference by sex (32.8% vs. 32.0%). Virtually absent in the younger age group, the frequency of individual risk scores over 20% was similar in older men (10.3%) and older women (12.0%).
Discussion
In this sample of HIV-infected adults, there was an appreciable occurrence of metabolic syndrome, which confers an increased risk of developing type II diabetes and symptoms of CVD. Our population was ethnically diverse, with fewer Caucasians and more African Americans among the women. There was no significant difference in the occurrence of metabolic syndrome among the four ethnic groups.
Women were significantly more overweight than men, as indicated by their higher mean BMI values. Interestingly, as others have shown,14 waist-to-thigh ratio was associated with diabetes especially in men. Women, particularly those under 50 years of age, had a higher prevalence of diabetes and elevated levels of glucose than men, and had (in both age groups) a higher frequency of increased waist circumference. With men, notably those under 50 years of age, the prevalence of elevated TGs was higher than for women. Higher TGs might, in part, be explained by the trend toward a higher prevalence of reported alcohol use among men, The pattern of metabolic syndrome differed between the sexes, a reflection of differences in the occurrence of metabolic abnormalities.
Those with metabolic syndrome were slightly older than those without, had a lower viral load, but had a similar CD4 count, similar frequency of acquired immunodeficiency syndrome (AIDS) diagnosis, and similar time since HIV or AIDS diagnosis. The two groups showed no difference in the percentage receiving ART, similar to a previous study,22 although fewer of those with metabolic syndrome were taking NRTI-based medication, probably reflecting targeted clinical management of their metabolic abnormalities.
In total, 30.0% of participants had metabolic syndrome, with more women (35.3%) affected than men (28.1%), This high prevalence in women is slightly higher than the 33% in the Women's Interagency Study.23 In one report, a higher frequency in HIV-positive women was observed compared to men,24 whereas in other studies there was no difference.25–28 The prevalence we found is higher than reported in other studies of HIV patients. Pao et al.29 cataloged of 12 studies of metabolic syndrome in HIV cohorts. Three reports, from Pavia, Italy, cited frequencies above 30.0%. The prevalence in the other nine studies ranged from 11.4% to 25.5%, In comparison, the frequency in the general U.S. population reported in 2002 was 21.8%.30 However, in a more recent study, 34% of U.S. adults were reported to meet the criteria.31 There is considerable disparity in the literature concerning whether the prevalence of metabolic syndrome is indeed higher among HIV-infected individuals. Two reports, both using matched control groups, demonstrate this conundrum: One found a similar frequency in patients (25.5%) compared to controls (26.5%),27 whereas the other found a higher occurrence in patients (15.8%) than controls (3.2%).32
We also examined the frequency of two or more, and four or more metabolic abnormalities 63.3% had at least two abnormalities, This is somewhat more than the 49% in a multicenter study of HIV-infected patients.33 For those with four or more metabolic abnormalities, men had a slightly lower frequency (9.5%) than reported for the male population at large (11.1%),30 but for women it was higher (15.3% vs. 9.6%).
Older men (>50 years) had nearly double the prevalence of metabolic syndrome compared to younger men, and there was a trend toward a higher frequency in women. The percentages of men (41.0%) and women (47.4%) affected is somewhat higher than reported for a similar population of where the respective values were 34% and 36%.15 We found a much higher percentage of older women with elevated waist circumference compared to men. However, Adeyemi and colleagues observed a much higher frequency of low HDL-C in men (61%) compared to women (21%), whereas our study showed a trend toward a higher frequency in women. Despite a similar high frequency of raised glucose levels in older women, in both studies we saw a nearly three-fold higher prevalence of diabetes (16% vs. 6%). Adeyemi et al.15 reported a low mean 10-year cardiac risk scores (FCR) for the older female participants (3.1%) compared to 11.8% for the men. We found much higher mean scores in older women (8.5%), closer to what might be expected for this age group in the general population.20
In conclusion, we examined the parameters that define metabolic syndrome in a sample of HIV-infected men and women. The prevalence was higher than in most other studies, with the frequency increasing to 42.5% in those over 50 years of age. Those with the syndrome, and detectable HIV virus, had significantly lower median viral loads. Mean 10-year FCR scores were nearly doubled for those with metabolic syndrome. Both researchers and clinicians should consider age as well as sex when assessing patients with HIV-infection for risks associated with metabolic syndrome.
Author Disclosure Statement
No competing financial interests exist for any of the authors.
Acknowledgments
This research was supported by a grant from the National Institute of Mental Health (NIMH, 5 R01 MH074358), from the Dhanem Foundation (J.P.K.), and by a gift from the Mildred V. Strouss Charitable Trust (J.P.K.). Data collection was supported by the General Clinical Research Center in the UCSF CTSA (1 UL RR024131). Dr. Aouizerat is also supported by an National Institutes of Health (NIH) Roadmap K12 (KL2 RR024130); Dr. Davis is supported by an NIH Research Infrastructure in Minority Institutions (RIMI) award (5P20 MD0005444). Authors wish to acknowledge the contributions to the study from Ryan Kelly, Yeonsu Song, Kristen Nelson, and Matthew Shullick.
References
- 1.Klein D. Hurley LB. Sidney S. Cardiovascular Disease and HIV Infection. N Engl J Med. 2003;349:1869–1870. doi: 10.1056/NEJM200311063491918. [DOI] [PubMed] [Google Scholar]
- 2.Balt CA. Nixon H. Complications of HIV: Lipodystrophy, anemia, renal, cardiovascular, and bone diseases. Nurse Pract Forum. 2001;12:199–213. [PubMed] [Google Scholar]
- 3.Tanwani LK. Mokshagundam SL. Lipodystrophy, insulin resistance, diabetes mellitus, dyslipidemia, and cardiovascular disease in human immunodeficiency virus infection. South Med J. 2003;96:180–188. doi: 10.1097/01.SMJ.0000051731.69719.2E. [DOI] [PubMed] [Google Scholar]
- 4.Blanco F. García-Benayas T. José de la Cruz J. González-Lahoz J. Soriano V. First-line therapy and mitochondrial damage: different nucleosides, different findings. HIV Clinical Trials. 2003;4:11–19. doi: 10.1310/HF1J-3P6K-1K9H-AGPY. [DOI] [PubMed] [Google Scholar]
- 5.Calista D. Boschini A. Cutaneous side effects induced by indinavir. Eur J Dermatol. 2000;10:292–296. [PubMed] [Google Scholar]
- 6.Koppel K. Bratt G. Eriksson M. Sandstrom E. Serum lipid levels associated with increased risk for cardiovascular disease is associated with highly active antiretroviral therapy (HAART) in HIV-1 infection. Int J STD AIDS. 2000;11:451–455. doi: 10.1258/0956462001916236. [DOI] [PubMed] [Google Scholar]
- 7.Lainka E. Oezbek S. Falck M. Ndagijimana J. Niehues T. Marked dyslipidemia in human immunodeficiency virus-infected children on protease inhibitor-containing antiretroviral therapy. Pediatrics. 2002;110:e56. doi: 10.1542/peds.110.5.e56. [DOI] [PubMed] [Google Scholar]
- 8.Erickson SK. Nonalcoholic fatty liver disease. J Lipid Res. 2009;50:S415–S416. doi: 10.1194/jlr.R800089-JLR200. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Merriman RB. Nonalcoholic fatty liver disease and HIV infection. Curr HIV/AIDS Rep. 2006;3:113–117. doi: 10.1007/BF02696654. [DOI] [PubMed] [Google Scholar]
- 10.Guaraldi G. Squillace N. Stentarelli C. Orlando G. D'Amico R. Ligabue G. Fiocchi F. Zona S. Loria P. Esposito R. Palella F. Nonalcoholic fatty liver disease in HIV-infected patients referred to a metabolic clinic: prevalence, characteristics, and predictors. Clin Infect Dis. 2008;47:250–257. doi: 10.1086/589294. [DOI] [PubMed] [Google Scholar]
- 11.World Health Organization. Definition, Diagnosis and Classification of Diabetes Mellitus and its Complications. Report of a WHO Consultation. Geneva: World Health Organization; 1999. [Google Scholar]
- 12.Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) final report. Circulation. 2002;106:3143–3421. [PubMed] [Google Scholar]
- 13.Alberti KGMM. Eckel RH. Grundy SM. Zimmet PZ. Cleeman JI. Donato KA. Fruchart J-C. James WPT. Loria CM. Smith SC. Harmonizing the metabolic syndrome: A joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention, National Heart, Lung, and Blood Institute, American Heart Association, World Heart Federation, International Atherosclerosis Society, and International Association for the Study of Obesity. Circulation. 2009;120:1640–1645. doi: 10.1161/CIRCULATIONAHA.109.192644. [DOI] [PubMed] [Google Scholar]
- 14.Chuang YC. Hsu KH. Hwang CJ. Hu PM. Lin TM. Chiou WK. Waist-to-thigh ratio can also be a better indicator associated with type 2 diabetes than traditional anthropometrical measurements in Taiwan population. Ann Epidemiol. 2006;16:321–331. doi: 10.1016/j.annepidem.2005.04.014. [DOI] [PubMed] [Google Scholar]
- 15.Adeyemi O. Rezai K. Bahk M. Badri S. Thomas-Gossain N. Metabolic syndrome in older HIV-infected patients: data from the CORE50 cohort. AIDS Patient Care STDS. 2008;22:941–945. doi: 10.1089/apc.2008.0119. [DOI] [PubMed] [Google Scholar]
- 16.Lee KA. Gay C. Portillo CJ, et al. Symptom experience in HIV-infected adults: A function of demographic and clinical characteristics. J Pain Symptom Manag. 2009;38:882–893. doi: 10.1016/j.jpainsymman.2009.05.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Kane JP. Malloy MJ. Ports TA. Phillips NR. Diehl JC. Havel RJ. Regression of coronary atherosclerosis during treatment of familial hypercholesterolemia with combined drug regimens. JAMA. 1990;264:3007–3012. [PubMed] [Google Scholar]
- 18.Warnick GR. Benderson J. Albers JJ. Dextran sulfate-Mg2+ precipitation procedure for quantitation of high-density-lipoprotein cholesterol. Clin Chem. 1982;28:1379–1388. [PubMed] [Google Scholar]
- 19.Friedewald WT. Levy RI. Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem. 1972;18:499–502. [PubMed] [Google Scholar]
- 20.Wilson PW. D'Agostino RB. Levy D. Belanger AM. Silbershatz H. Kannel WB. Prediction of coronary heart disease using risk factor categories. Circulation. 1998;97:1837–1847. doi: 10.1161/01.cir.97.18.1837. [DOI] [PubMed] [Google Scholar]
- 21.U.S. Department of Health and Human Services, Public Health Service, National Institutes of Health (NIH publication number 80-1527); Washington, DC: 1980. Population Studies Data Book: Vol. I, The Prevalence Study. [Google Scholar]
- 22.Jones CY. Metabolic syndrome in HIV-infected patients: No different than the general population? Clin Infect Dis. 2007;44:735–738. doi: 10.1086/511691. [DOI] [PubMed] [Google Scholar]
- 23.Sobieszczyk ME. Hoover DR. Anastos K. Mulligan K. Tan T. Shi Q. Gao W. Hyman C. Cohen MH. Cole SR. Plankey MW. Levine AM. Justman J the Women's Interagency HIV Study. Prevalence and predictors of metabolic syndrome among HIV-infected and HIV-uninfected women in the Women's Interagency HIV Study. J Acquir Immune Defic Syndr. 2008;48:272–280. doi: 10.1097/QAI.0b013e31817af461. [DOI] [PubMed] [Google Scholar]
- 24.Baum MK. Rafie C. Lai S. Xue L. Sales S. Page JB. Berkman R. Karas L. Campa A. Coronary heart disease (CHD) risk factors and metabolic syndrome in HIV-positive drug users in Miami. Am J Infect Dis. 2006;2:173–179. [PMC free article] [PubMed] [Google Scholar]
- 25.Bonfanti P. Ricci E. de Socio G. Zeme D. Carradori S. Penco G. Parruti G. Grosso C. Madeddu G. Vichi F. Bini T. Martinelli C. Melzi S. Quirino T for the CISAI Study Group. Metabolic syndrome: A real threat for HIV-positive patients?: Results from the SIMONE study. J Acq Immune Defic Syndr. 2006;42:128–131. doi: 10.1097/01.qai.0000219775.20174.2d. [DOI] [PubMed] [Google Scholar]
- 26.Jericó C. Knobel H. Montero M. Ordoñez-Llanos J. Guelar A. Gimeno JL. Saballs P. López-Colomés JL. Pedro-Botet J. Metabolic syndrome among HIV-infected patients: Prevalence, characteristics, and related factors. Diabetes Care. 2005;28:132–137. doi: 10.2337/diacare.28.1.132. [DOI] [PubMed] [Google Scholar]
- 27.Mondy K. Overton ET. Grubb J. Tong S. Seyfried W. Powderly W. Yarasheski K. Metabolic syndrome in HIV-infected patients from an urban, midwestern US outpatient population. Clin Infect Dis. 2007;44:726–734. doi: 10.1086/511679. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Palacios R. Santos J. González M. Ruiz J. Márquez M. Incidence and prevalence of the metabolic syndrome in a cohort of naive HIV-infected patients: Prospective analysis at 48 weeks of highly active antiretroviral therapy. Int J STD & AIDS. 2007;18:184–187. doi: 10.1258/095646207780132415. [DOI] [PubMed] [Google Scholar]
- 29.Pao V. Lee GA. Grunfeld C. HIV therapy, metabolic syndrome, and cardiovascular risk. Curr Atheroscler Rep. 2008;10:61–70. doi: 10.1007/s11883-008-0010-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Ford ES. Giles WH. Dietz WH. Prevalence of the metabolic syndrome among US adults: Findings from the third National Health and Nutrition Examination Survey. JAMA. 2002;287:356–359. doi: 10.1001/jama.287.3.356. [DOI] [PubMed] [Google Scholar]
- 31.Ervin RB. Prevalence of metabolic syndrome among adults 20 years of age and over, by sex, age, race and ethnicity, and body mass index: United States, 2003–2006. Natl Health Stat Rep. 2009;13:1–7. [PubMed] [Google Scholar]
- 32.Estrada V. Martínez-Larrad MT. González-Sánchez JL. de Villar N. Zabena C. Fernández C. Serrano-Ríos M. Lipodystrophy and metabolic syndrome in HIV-infected patients treated with antiretroviral therapy. Metab Clin Exp. 2006;55:940–945. doi: 10.1016/j.metabol.2006.02.024. [DOI] [PubMed] [Google Scholar]
- 33.Samaras K. Wand H. Law M. Emery S. Cooper D. Carr A. Prevalence of metabolic syndrome in HIV-infected patients receiving highly active antiretroviral therapy using International Diabetes Foundation and Adult Treatment Panel III criteria: Associations with insulin resistance, disturbed body fat compartmentalization, elevated C-reactive protein, and [corrected] hypoadiponectinemia. Diabetes Care. 2007;30:113–119. doi: 10.2337/dc06-1075. [DOI] [PubMed] [Google Scholar]