Metabolic syndrome increases the risk of developing cardiovascular outcomes and type II diabetes. It was estimated to be present in 23.7%1 of adults in the United States between 1988 and 1994, and the rate is increasing. The syndrome is defined as the occurrence of 3 or more of the following abnormalities: hypertriglyceridemia, low high-density lipoprotein (HDL) cholesterol, hypertension, abdominal obesity, and high serum glucose. Metabolic syndrome in the United States is more prevalent with increasing age and is highest among Mexican Americans. Furthermore, African-American and Mexican-American women have higher rates than men in each of these ethnic groups.1
The components of metabolic syndrome have been recognized in patients infected with HIV. Low HDL and low-density lipoprotein (LDL) cholesterol and elevations in triglycerides were observed in untreated HIV-infected patients early in the HIV epidemic, before the advent of highly active antiretroviral therapy (HAART).2-4 Although these abnormalities were seen most often in those with more advanced disease, they also occurred in patients with asymptomatic HIV disease.4 Since the mid 1980s, several additional abnormalities have been noted in HIV, including hypertension,5 visceral adiposity6,7 and insulin resistance.8,9 Although the relative frequency of these abnormalities compared with the general population is debated,8 and the etiology is unclear, specific antiretroviral treatments9 have been implicated as causative agents in some studies. Jerico et al10 showed a positive association between metabolic syndrome and stavudine (d4T) and lopinavir/ritonavir in HIV-infected adults in Spain.
The cumulative effects of HIV may play a role in the occurrence of metabolic abnormalities because HIV-infected patients are living longer as a result of successful treatment, and they are then also at risk for type II diabetes and cardiovascular disease associated with aging in the general population. Certainly, risk factors for cardiovascular disease and diabetes, such as poor dietary habits,11 low level of physical activity,8 and smoking,12 are common in HIV-infected adults.
The prevalence and incidence of metabolic syndrome in HIV-infected adults in the United States is unknown. Two Italian studies found a higher prevalence of metabolic syndrome among Italian HIV-infected patients treated with HAART13,14 compared with HIV-negative blood donors14 and the general United States population.13
The purpose of this analysis was (1) to compare the prevalence of metabolic syndrome and each of its 5 components in HIV-infected adults with that in the US population (National Health and Nutrition Survey [NHANES] 1999-2002)15 (2) to determine the incidence and predictors of newly developing metabolic syndrome in HIV-infected adults.
Nutrition for Healthy Living Study
The Nutrition for Healthy Living (NFHL) study is a cohort of HIV-infected adult volunteers who were followed semiannually to evaluate the relation between HIV infection and nutrition. Eligible participants included HIV-positive adults (aged 18 years or older) living in the greater Boston area or Rhode Island. Participants were recruited through advertisements on the radio as well as in local newspapers, health clinics, and physician networks as described previously.16-18 Individuals were excluded if they had any of the following conditions at enrollment: pregnancy, thyroid disease, or any malignancies other than Kaposi sarcoma. Participants were also excluded if they were not fluent in English. Further details of this study have been reported elsewhere.19-21 Beginning in July 2000, metabolic measurements were collected, including fasting serum glucose and lipids, waist circumference, and blood pressure. The Institutional Review Boards at Tufts University School of Medicine and Miriam Hospital, Rhode Island, approved this study, and written informed consent was obtained from each participant. NFHL data for these analyses began at the first visit for which participants had all the parameters necessary to diagnose the metabolic syndrome (n = 477). In this study, we included participants who were between 25 and 65 years old (actual range: 25-63 years) and were seen between September 2000 and November 2003. The cross-sectional analysis was performed on the first visit in this data set. The longitudinal analysis was performed on those without metabolic syndrome at the first visit and included all subsequent visits or until a participant met the definition of metabolic syndrome or died.
Trained interviewers collected information from patients on demographics, dietary intake by 3-day food records, HIV-related clinical events, and therapy.7 HAART use was defined as current use at the first study visit in this data set. From reported household income, we determined which participants were living below the US federal poverty level (yes/no).22 Daily caloric intake was determined from the 3-day food records using the Minnesota Nutrition Data System version 4.06_34. Physical activity was assessed by the physical activity recall instrument developed by Sallis et al,23 which assesses usual activity over the past 7 days. We defined “regular exercise” as any moderate, hard, or very hard physical activity in the past week and “strength training” as any strength training in the past week.
Plasma triglycerides, serum HDL, and plasma glucose levels were obtained after a 5-hour fast.18,20,21,24 Using the Beckman LX-20 (Beckman Coulter, Inc., Brea, CA), plasma glucose was measured by the glucose oxidase method, plasma triglycerides by the 2-step enzymatic method, and HDL by an enzymatic method (separated by dextran sulfate, 500,000 molecular weight). HIV RNA was measured by the Roche Amplicor Monitor reverse transcriptase polymerase chain reaction (PCR) assay (Roche Molecular Systems, Somerville, NJ), with a lower detection limit of 400 copies/mL.
Blood pressure was measured with the Omron Automatic Inflation Blood Pressure Monitor (HEM-722CR, Omron Healthcare, Inc., Bannockburn, IL) after the participant had been seated for 5 minutes with his or her feet on the floor and his or her arms supported at heart level. Subjects were weighed (kilograms) on a digital scale fully dressed but without shoes, heavy clothing, or objects before eating or drinking (minimum 5-hour fast). Height (centimeters) was measured without shoes by a stadiometer. Body mass index (BMI) was calculated as weight divided by height squared (kg/m2) and then categorized into <25 (normal or less), 25 to 29.9 (overweight), or ≥30 (obese).
Waist circumference was taken without outer clothing, using a tape measure in light contact with but not compressing, the skin.25 It was measured at the natural waist indentation at the end of a normal expiration while subjects were standing. Research personnel were trained and recertified semiannually in anthropometry. The reliability of these measures has been shown to be high.7
Dual X-Ray Absorptiometry
Yearly dual X-ray absorptiometry (DXA) scans were performed on the Hologic QDR-2000 machine (Hologic, Waltham, MA). A series of cutoff lines positioned at anatomic markers was used to define trunk fat and limb fat (arms and legs). The ratio of trunk fat to limb fat was calculated, and the distribution was then divided into quartiles for analysis. The quartiles were determined separately for men and women.
National Health and Nutrition Examination Survey (1999-2002)
National Health and Nutrition Examination Survey Participants
The NHANES (1999-2002) is a representative sample of nonincarcerated adults in the United States selected on the basis of a complex multistage design so as to include an adequate number of minorities.15 We included NHANES (1999-2002) participants who were between 25 and 65 years old, had an examination in the mobile van rather than at home, and who had blood drawn after fasting for at least 8 hours (n = 1876). To be consistent with the NFHL study, we included all Hispanic/Latinos except Mexican Americans, because only (0%) of NFHL participants were Mexican. In the study by Ford et al, 1 Mexican Americans had the highest risk of metabolic syndrome, and therefore differ from other Hispanics.
Data were collected by means of personal interviews. Poverty (yes/no) was determined by US Census Bureau calculations (available at: http://www.census.gov/hhes/www/poverty/povdef.html). Participants were asked if they had engaged in moderate or vigorous physical activity and muscle-strengthening activities over the past 30 days. We defined regular exercise as any moderate or vigorous activity over the past month and strength training as any strength training over the past month. Average daily caloric intake was determined from 24-hour dietary recall. Daily caloric intake per body weight was calculated (kilocalories per kilogram per day) for the NFHL study and NHANES.
Definition of Metabolic Syndrome
The definition of metabolic syndrome is based on the “Report of the National Heart, Lung, and Blood Institute/American Heart Association Conference on Scientific Issues Related to Definition.”26 It is defined as the presence of 3 or more of the following 5 abnormalities: (1) abdominal obesity (waist circumference >102 cm for men and >88 cm for women, (2) hypertriglyceridemia (>150 mg/dL), (3) low HDL cholesterol (<40 mg/dL for men and <50 mg/dL for women), and (4) high BP (systolic >130 mm Hg and/or diastolic ≥85 mm Hg), (5) high fasting glucose (≥100 mg/dL). In the NFHL study and NHANES, persons with a diagnosis of diabetes or being treated for diabetes were classified as having “high fasting glucose” (n = 9 in the NFHL study, n = 108 in the NHANES) and persons taking medication for high blood pressure were defined as having high blood pressure (n = 27 in the NFHL study, n = 412 in the NHANES).
We evaluated metabolic syndrome and each of its 5 components (metabolic abnormalities) as separate outcomes. Thus, the outcomes include metabolic syndrome, low HDL, hypertriglyceridemia, high blood pressure, abdominal obesity, and high blood glucose.
Nutrition for Healthy Living Study: Prevalence
Unadjusted Prevalence of Each Outcome in Nutrition for Healthy Living Study
The unadjusted prevalence of each outcome within the NFHL cohort was the number of cases of the outcome divided by the total number of participants in the NFHL study (n = 477). To obtain the 95% confidence intervals (CIs) around each estimate of prevalence, we used generalized estimating equations (GEEs) for the binomial with the Poisson distribution, and we used the log link to obtain the robust variance in PROC GENMOD.27 All analyses were conducted in SAS 9.1 (SAS Institute, Cary, NC).
Nutrition for Healthy Living Study Versus National Health and Nutrition Examination Survey: Characteristics and Prevalence
To determine if there were significant differences in characteristics of HAART users and non-HAART users in the NFHL study relative to characteristics of NHANES participants, we used PROC SURVEYREG for continuous variables and PROC SURVEYFREQ for categoric variables. Because participants in the NHANES are selected by a complex multistage design, analytic models using NHANES data must include sample weights to produce correct national estimates for the NHANES and the corresponding variances. We specified the strata, cluster, and weight of the sampling scheme for the NHANES using the NHANES variables “sdmvstra,” “sdmvpsu,” and “wtmec4yr,” respectively. For the NFHL cohort, we set weight equal to 1 and “sdmvpsu” to the participant's identification number.
Unadjusted and Adjusted Odds Ratios for Each Outcome: Nutrition for Healthy Living Study Versus National Health and Nutrition Examination Survey
For each outcome, we determined the odds ratio (OR) comparing HAART users in the NHFL study with participants in the NHANES and non-HAART users in NFHL study with participants in the NHANES (PROC SURVEYLOGISTIC, SAS 9.1; SAS Institute)28 using the same weighting scheme as explained previously. We first obtained an unadjusted OR and then an OR adjusted for age in years, race, gender, poverty, regular exercise, strength training, and caloric intake. These variables are associated with metabolic syndrome, and most differ between the NFHL study and NHANES. We performed separate models adjusted for BMI in addition to the other variables mentioned previously. When there was significant evidence for nonlinearity between age in years and each outcome, we included the significant restricted cubic spline terms in the model29 in addition to age in years. When the prevalence of a metabolic abnormality is common, the OR may overestimate the true prevalence ratio.30 In this study, all the metabolic abnormalities are common, with the exception of high glucose.
Combinations of Components of Metabolic Syndrome
We determined the combination of metabolic abnormalities for each person with metabolic syndrome. The frequency of each combination was determined separately in the NFHL study and NHANES.
Nutrition for Healthy Living Study: Incidence of Metabolic Syndrome
The incidence rate of metabolic syndrome was determined among those without the syndrome at baseline who were followed over time. The incidence rate was the number of new cases over the total person-months of follow-up. There were 7026 person-months of follow-up. The median (25th and 75th percentiles) time of follow-up was 19 months (25th 12, 75th 31). Because the NHANES is a cross-sectional study, there are no incidence data available for comparison on HIV-negative individuals.
We evaluated the relative risk of developing metabolic syndrome across levels of demographic, clinical, and nutritional predictors using Cox proportional hazards regression with time-varying covariates (PROC PHREG, SAS 9.1; SAS Institute). Clinical and nutritional predictors were assessed at the beginning of the interval before an event (BMI and trunk-to-limb fat ratio) or over the interval before an event (change in viral load, weight, exercise, diet, or medication use). We categorized a change in viral load >0.5 log10 copies/mL as reflecting clinical treatment failure. We also tested for duration of continuous medication use up until the end of the interval. We tested for types of HAART use (eg, protease inhibitor [PI] based, nonnucleoside reverse transcriptase inhibitor [NNRTI] based, mixed) and specific medications. Predictors that were associated with developing metabolic syndrome at P < 0.2 on univariate analysis were tested in the multivariate model. Gender, age, and race were included in the final multivariate model even though they were not significant independent predictors so as to adjust for confounding.
The characteristics of NFHL study participants, by HAART use, and those of NHANES participants are shown in Table 1. Significant differences between the cohorts are described in the footnote for Table 1.
Unadjusted Prevalence of Each Outcome in Nutrition for Healthy Living Study
The unadjusted prevalence of metabolic syndrome in the NFHL study was 24% (95% CI: 21% to 28%). The most common metabolic abnormalities in the NFHL study were low HDL (prevalence = 54%, 95% CI: 49% to 58%) and high triglycerides (prevalence = 47%, 95% CI: 42% to 51%), and the least common was high blood glucose (prevalence = 4%, 95% CI: 3% to 7%). The prevalence of high blood pressure and abdominal obesity were 33% (95% CI: 29% to 37%) and 25% (95% CI: 22% to 30%), respectively.
Number and Frequency of Different Combinations of Abnormalities: Nutrition for Healthy Living Study Versus National Health and Nutrition Examination Survey
The percentages of participants in each cohort (NFHL study vs. NHANES) with the following number of abnormalities were as follows: ≥1 abnormality (81.6% vs. 81.3%); ≥2 abnormalities (51.8% vs. 56.7%); ≥3 abnormalities (24.3% vs. 34.1%); ≥4 abnormalities (5.2% vs. 15.8%), and all 5 abnormalities (0% and 4.5%). Among those with metabolic syndrome in each cohort, 76.7% of NFHL participants and 47.0% of NHANES participants had hypertriglyceridemia and low HDL plus at least 1 additional criterion. The distribution of the various combinations of abnormalities in those with metabolic syndrome is shown in Table 2.
Odds Ratio of Each Outcome in HAART and Non-HAART Users in Nutrition for Healthy Living Study Versus National Health and Nutrition Examination Survey
HAART users (adjusted OR = 0.67, 95% CI: 0.47 to 0.96) and non-HAART users (adjusted OR = 0.55, 95% CI: 0.33 to 0.93) in the NFHL study were less likely to have metabolic syndrome compared with NHANES participants (Table 3). Those using Kaletra (lopinavir/ritonavir) were equally likely to have metabolic syndrome as those in the NHANES (OR = 1.2, 95% CI: 0.46 to 2.8; P = 0.76), whereas those not using lopinavir/ritonavir were less likely to have metabolic syndrome compared with those in the NHANES (OR = 0.55, 95% CI: 0.33 to 0.92; P = 0.02). After adjusting for BMI, the difference in metabolic syndrome between NFHL and NHANES participants was no longer statistically significant, but the same trend still exists (adjusted OR = 0.77, 95% CI: 0.5 to 1.2).
The odds of having low HDL was 1.6 times higher in HAART users (adjusted OR = 1.6, 95% CI: 1.1 to 2.1) in the NFHL study compared with NHANES participants, and the odds of having high triglycerides was 2.5 times greater in HAART users in the NFHL study than in NHANES participants (adjusted OR = 2.5, 95% CI: 1.9 to 3.4). Non-HAART users in the NFHL study were also more likely to have low HDL (adjusted OR = 2.7, 95% CI: 1.7 to 4.3) than NHANES participants, but they were equally likely to have high triglycerides (adjusted OR = 0.99, 95% CI: 0.63 to 1.6). High blood pressure was equally common in HAART users (adjusted OR = 0.92, 95% CI: 0.69 to 1.2) and non-HAART users (adjusted OR = 0.83, 95% CI: 0.53 to 1.3) in the NFHL study compared with NHANES participants. Abdominal obesity was significantly less likely in HAART users (adjusted OR = 0.44, 95% CI: 0.32 to 0.60) and non-HAART users (adjusted OR = 0.42, 95% CI: 0.26 to 0.67) in the NFHL study compared with NHANES participants. High blood glucose was also much less common in NFHL participants compared with NHANES participants regardless of HAART use (adjusted OR = 0.06, 95% CI: 0.04 to 0.11) or non-HAART use (adjusted OR = 0.07, 95% CI: 0.03 to 0.17). Adjustment for BMI when evaluating the differences in each component of the metabolic syndrome between participants of the NFHL study and NHANES did not change the results.
Incidence of Metabolic Syndrome in Nutrition for Healthy Living Study
The baseline characteristics of NFHL participants without metabolic syndrome at baseline (n = 338) are shown in Table 4 (baseline frequency).
There were 88 new cases of metabolic syndrome over a total of 7026 person-months of follow-up for an incidence rate of 1.2 per 100 person-months. Among the 88 new cases of metabolic syndrome, 84% participants had hypertriglyceridemia, 89% had low HDL, 72% had high blood pressure, 37% had abdominal obesity, and 32% had high glucose.
In the final multivariate model, the risk of developing metabolic syndrome over follow-up (Table 4) was significantly higher with an increasing viral burden, lopinavir/ritonavir or didanosine (ddI) use, increasing weight, greater BMI, and greater trunk-to-limb fat ratio, adjusted for race, gender, and age. The risk was significantly lower among those with a college education. The risk was 80% higher among those with at least a 0.5-log increase in viral load in the previous 6 months. The risk was 2 times higher in those using lopinavir/ritonavir or ddI. Even after adjusting for BMI, the risk was 2-fold higher among those who gained more than 2 kg over the previous 6 months, and the risk increased from 2.3-fold to 4.9-fold with increasing quartile of trunk fat relative to limb fat.
Duration of individual medications or types of HAART was not an independent predictor of metabolic syndrome.
To our knowledge, this is one of the first large studies of the incidence and prevalence of metabolic syndrome in HIV. Our results suggest that HIV itself, treatment, and host factors affect the risk of metabolic syndrome in HIV disease. Among our cohort of HIV-infected adults, the risk of developing metabolic syndrome was higher in those with a clinically relevant increase in viral load (≥0.5 log) in the previous 6 months. This association is most likely related to the effect of HIV on HDL levels,31 because increasing viral load was associated with decreasing HDL in this cohort. In contrast, the relation between increasing viral load and metabolic syndrome does not seem to be mediated by general inflammation caused by chronic infection for 2 reasons. First, high viral load did not predict development of metabolic syndrome. Second, in a separate analysis in our cohort, neither absolute viral load nor change in viral load was associated with C-reactive protein (CRP), a marker of generalized inflammation that is associated with cardiovascular outcomes in the general population.
Specific antiretroviral therapies are known to affect individual components of the metabolic syndrome adversely, such as increasing triglycerides (lopinavir/ritonavir,32 nevirapine,33 and nelfinavir33) and increasing fasting glucose (indinavir34 and lopinavir/ritonavir35). Other therapies have been associated with improvements in metabolic profiles, such as increases in HDL cholesterol (nevirapine and nelfinavir).33 In our cohort, lopinavir/ritonavir users had a higher risk of developing metabolic syndrome. This is supported by results from a cross-sectional study by Jerico et al.10 In contrast to that study, we found no association between d4T and metabolic syndrome. There was no association between metabolic syndrome and indinavir, nelfinavir, or nevirapine in our study. Although ddI use has not been shown to cause metabolic changes, the increased risk of metabolic syndrome among our ddI users may reflect the effect of previous treatments on individuals or may be related to greater disease severity. HIV patients are generally prescribed ddI after they have failed other antiretroviral medications because of metabolic side effects, other toxicities, or uncontrolled viral load.
Obesity is a leading cause of metabolic syndrome in the general population, and we found that it is an important risk factor for metabolic syndrome in HIV-infected persons as well. Many HIV-infected patients maintain or gain weight as they survive longer with improved treatments. In our cohort, more than 35% were overweight (BMI: 25-29) and 18% were obese (BMI ≥ 30). Less than 2% had a BMI < 18. Body shape was an equally important risk factor for metabolic syndrome. A greater amount of trunk fat relative to limb fat increased the risk of metabolic syndrome, even after adjusting for BMI and weight gain. Peripheral fat atrophy, as measured by low triceps skinfold, was not a predictor of metabolic syndrome, however. This suggests that greater trunk fat is an important risk factor for metabolic syndrome whether or not there is a significant loss of peripheral fat. Body shape abnormalities have been widely noted in HIV since the mid-1990s, including peripheral fat atrophy and central fat deposition. Our previous data7 and those of others36,37 suggest that these 2 abnormalities are the result of separate processes that do not commonly occur together. The cardiovascular outcomes of lipodystrophy need further study.
Despite the adverse effects of HIV and antiretroviral treatments on some components of metabolic syndrome, the prevalence of metabolic syndrome was lower in our HIV-infected cohort compared with the US population, even after adjusting for differences in demographics, physical activity, and diet between the 2 populations. There are several reasons why the rate might be lower in HIV. One possible explanation is that only 2 of the 5 components of metabolic syndrome were more likely in HIV-infected adults compared with NHANES participants: metabolic syndrome was mostly (77%) driven by hypertriglyceridemia and low HDL in HIV. Both abnormalities have been associated with HIV disease before and after the introduction of HAART.2-4,38 The present analysis suggests a strong effect of HIV on low HDL, because HAART and non-HAART had a higher prevalence of low HDL compared with that in the NHANES. It also suggests a strong effect of therapy (but not HIV) on triglyceride levels, because only HAART users differed from NHANES participants on hypertriglyceridemia. Another possible explanation for differences in metabolic syndrome between the NFHL study and the NHANES is the higher rate of obesity in the NHANES. After adjusting for BMI, the difference in metabolic syndrome between NFHL and NHANES participants was no longer statistically significant, but a trend still exists. Adjustment for BMI when evaluating the differences in each component of the metabolic syndrome between NFHL and NHANES participants did not change the results.
Our findings differ from those of 2 small Italian studies on metabolic syndrome in HIV. These studies compared HAART patients with HIV-negative blood donors14 and NHANES participants13 and found an increased prevalence of metabolic syndrome associated with HIV infection. In contrast to these studies, our population was not limited to HAART users and it included a more diverse demographic group of men and women. These studies may also differ from ours because HIV-negative blood donors are generally healthier than the general population, and this could amplify differences in prevalence. In addition, in contrast to our analyses, in comparing HAART users with NHANES participants, the authors of these small Italian studies did not account for the NHANES sampling scheme in their analysis, which would affect the standard errors and diminish differences, nor did they adjust for confounding factors.
When comparing the prevalence of the other components of metabolic syndrome with that of the NHANES, we found a similar rate of hypertension. Bergersen et al39 found that patients on HAART had a similar rate of hypertension as HAART-naive patients and HIV-negative controls. In contrast, Gazzaruso et al5 found a higher rate of hypertension in HAART patients compared with age- and sex-matched HIV-negative blood donors. In another study of HIV-infected patients on HAART, patients with HIV-associated lipodystrophy were more likely to have hypertension than those without lipodystrophy.40 Abdominal obesity is a commonly reported component of HIV-associated lipodystrophy,7 and specific HAART therapies may be implicated in the etiology of this abnormality.41 Yet, in our cohort, the rates of abdominal obesity were lower than in the NHANES, even after adjusting for demographics, physical activity, and dietary factors. Although lower BMI in the NFHL study may partially explain the lower prevalence of metabolic syndrome in the NFHL study compared with the NHANES, the mechanism is unclear because the prevalence of abdominal obesity remained lower in the NFHL study even after adjustment for BMI. This is supported by the findings of Bachetti et al36 and Gripsolver et al,37 who found a similar prevalence of abdominal obesity in HIV compared with population controls in the Cardia Study. High blood glucose was rarely observed in our cohort (4%) and was significantly less frequent than in the NHANES. Indinavir has been associated with insulin resistance,34 but this agent was only used by 11.5% of persons in our cohort. Indinavir users did not have a higher prevalence of high glucose compared with NHANES participants. Thus, loss of control of glucose seems to be infrequent and rarely contributes to the diagnosis of metabolic syndrome in our HIV cohort.
There are several limitations to our analyses. We were not able to compare our population with NHANES participants living in the northeastern United States directly, because this level of detail is not publically available. Our results may not be generalizable to the entire population of HIV-infected adults in the United States. Our study protocol called for a minimum fast of 5 hours before collecting fasting lipids and glucose, whereas an 8-hour fast was required of participants in the NHANES. Neither used the 12-hour standard fast. Chylomicrons may have still been present in the sera, exaggerating triglyceride levels. In that case, the true rate of metabolic syndrome might actually be lower in our study, showing a greater difference between the NFHL study and NHANES, because participants in the NHANES had a longer fast. Finally, some of our participants may have had preexisting risk factors for metabolic syndrome before becoming HIV-positive. The exact dates of HIV infection and development of diabetes or hypertension cannot be known for sure. Nevertheless, at the time of enrollment in the NFHL study, all 4 people who reported a past history of diabetes said they were diagnosed with diabetes before they knew they were HIV-positive. Among those with a previous diagnosis of hypertension at study entry (n = 37), 17 said they were diagnosed with hypertension before they knew they were HIV-positive. If we exclude those people with a diagnosis of diabetes or hypertension before they knew they were infected with HIV, we again observe a lower rate of metabolic syndrome in HIV-infected individuals compared with NHANES participants. We have no data on past waist circumference or lipid levels.
In summary, HIV-infected patients are at risk of metabolic syndrome, even though the rate is lower than that in the general population. HIV providers should monitor patients for factors that indicate a risk of cardiovascular disease outcomes.
1. 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
2. Grunfeld C, Feingold KR. The role of the cytokines, interferon alpha and tumor necrosis factor in the hypertriglyceridemia and wasting of AIDS. J Nutr
3. Grunfeld C, Kotler DP, Shigenaga JK, et al. Circulating interferon-alpha levels and hypertriglyceridemia in the acquired immunodeficiency syndrome. Am J Med
4. Shor-Posner G, Basit A, Lu Y, et al. Hypocholererolemia is associated with immune dysfunction in early human immunodeficiency virus-1 infection. Am J Med
5. Gazzaruso C, Bruno R, Garzaniti A, et al. Hypertension among HIV
and relationships to insulin resistance and metabolic syndrome
. J Hypertens
6. Saint-Marc T, Partisani M, Poizot-Martin I, et al. A syndrome of peripheral fat wasting (lipodystrophy) in patients receiving long-term nucleoside analogue therapy. AIDS
7. Jacobson DL, Knox TA, Spiegelman D, et al. Prevalence
, evolution and risk factors
for fat atrophy and fat deposition in a cohort of HIV
-infected men and women. Clin Infect Dis
8. Jones CY, Wilson I, Greenberg AS, et al. Insulin resistance in HIV
-infected men and women in the Nutrition for Healthy Living Cohort. J Acquir Immune Defic Syndr
9. Carr A, Samaras K, Burton S, et al. A syndrome of peripheral lipodystrophy, hyperlipidaemia and insulin resistance in patients receiving HIV
protease inhibitors. AIDS
10. Jerico C, Knobel H, Montero M, et al. Metabolic syndrome
-infected patients. Diabetes Care
11. Kim JH, Spiegelman DL, Rimm E, et al. The correlates of dietary intake in HIV
-infected adults. Am J Clin Nutr
12. Galai N, Park LP, Wesch J, et al. Effect of smoking on the clinical progression of HIV
-1 infection. J Acquir Immune Defic Syndr Hum Retrovirol
13. Gazzaruso C, Sacchi P, Garzaniti A, et al. Prevalence
of metabolic syndrome
patients. Diabetes Care
14. Bruno R, Gazzaruso C, Sacchi P, et al. High prevalence
of metabolic syndrome
-infected patients: link with the cardiovascular risk. J Acquir Immune Defic Syndr
15. US Department of Health and Human Services, Centers for Disease Control and Prevention National Center for Health Statistics. National Health and Examination Survey 1999-2000, 2001-2002
. Hyattsville, MD; 2005.
16. Forrester JE, Spiegelman D, Woods M, et al. Weight and body composition in a cohort of HIV
-positive men and women. Public Health Nutr
17. Forrester JE, Spiegelman D, Tchetgen E, et al. Weight loss and body-composition changes in men and women infected with HIV
. Am J Clin Nutr
18. Knox TA, Spiegelman D, Skinner SC, et al. Diarrhea and abnormalities of gastrointestinal function in a cohort of men and women with HIV
infection. Am J Gastroenterol
19. Shevitz AH, Knox TA, Spiegelman D, et al. Elevated resting energy expenditure among HIV
-seropositive persons receiving highly active antiretroviral therapy. AIDS
20. Silva M, Skolnik PR, Gorbach SL, et al. The effect of protease inhibitors on weight and body composition in HIV
-infected patients. AIDS
21. Wilson IB, Roubenoff R, Knox TA, et al. Relation of lean body mass to health-related quality of life in persons with HIV
. J Acquir Immune Defic Syndr
22. The 1998 HHS poverty guidelines. Fed Regist
23. Sallis JF, Haskell WL, Wood PD, et al. Physical activity assessment methodology in the five-city project. Am J Epidemiol
24. Wanke CA, Silva M, Knox TA, et al. Weight loss and wasting remain common complications in individuals infected with human immunodeficiency virus in the era of highly active antiretroviral therapy. Clin Infect Dis
25. Lohman TG, Roche AF, Marttorell R. Anthropometric Standardization Manual
. Champaign, IL: Human Kinetics Publishers; 1988.
26. Grundy SM, Brewer HB Jr, Cleeman JI, et al. Definition of metabolic syndrome
: report of the National Heart, Lung, and Blood Institute/American Heart Association Conference on Scientific Issues Related to Definition. Circulation
27. Diggle PJ, Liang K-J, Zegar SL. Analysis of Longitudinal Data
. Oxford, UK: Clarendon Press; 1994.
28. Binder DA. On the variances of asymptotically normal estimators from complex surveys. Surv Methodol
29. Durrleman S, Simon R. Flexible regression models with cubic splines. Stat Med
30. Szklo M, Nieto FJ. Measure of disease occurrence and association. In: Epidemiology: Beyond the Basics
. Gaithersburg, MD: Aspen Publishers; 2000:87-89.
31. El-Sadr WM, Mullin CM, Carr A, et al. Effects of HIV
disease on lipid, glucose and insulin levels: results from a large antiretroviral-naive cohort. HIV Med
32. Lee GA, Seneviratne T, Noor MA, et al. The metabolic effects of lopinavir/ritonavir in HIV
-negative men. AIDS
33. Fisac C, Virgili N, Ferrer E, et al. A comparison of the effects of nevirapine and nelfinavir on metabolism and body habitus in antiretroviral-naive human immunodeficiency virus-infected patients: a randomized controlled study. J Clin Endocrinol Metab
34. Noor MA, Lo JC, Mulligan K, et al. Metabolic effects of indinavir in healthy HIV
-seronegative men. AIDS
35. Noor MA, Parker RA, O'Mara E, et al. The effects of HIV
protease inhibitors atazanavir and lopinavir/ritonavir on insulin sensitivity in HIV
-seronegative healthy adults. AIDS
36. Bacchetti P, Gripshover B, Grunfeld C, et al. Fat distribution in men with HIV
infection. J Acquir Immune Defic Syndr
37. Gripshover B, Tien PC, Saag M, et al, for the Investigators of the Fat Redistribution and Metabolic Change in HIV
Infection (FRAM) Study. Lipoatrophy is the dominant feature of the lipodystrophy syndrome in HIV
-infected men. Presented at: 10th Conference on Retroviruses and Opportunistic Infections, Boston, MA; 2003.
38. Mildvan D, Machado SG, Wilets I, et al. Endogenous interferon and triglyceride concentrations to assess response to zidovudine in AIDS and advanced AIDS-related complex. Lancet
39. Bergersen BM, Sandvik L, Dunlop O, et al. Prevalence
of hypertension in HIV
-positive patients on highly active anti-retroviral therapy (HAART) compared with HAART-naive and HIV
-negative controls: results from a Norwegian study of 721 patients. Eur J Clin Microbiol Infect Dis
40. Sattler FR, Qian D, Louie S, et al. Elevated blood pressure in subjects with lipodystrophy. AIDS
41. Heath KV, Hogg RS, Singer J, et al. Antiretroviral treatment patterns and incident HIV
-associated morphologic and lipid abnormalities in a population-based cohort. J Acquir Immune Defic Syndr