Transcript HIV-positive

Prevalence and Predictors of Metabolic Syndrome among HIV-Positive and
Negative Women in the Women’s Interagency HIV Study
XVI International AIDS Conference
Toronto, 2006
ME. Sobieszczyk1, D. Hoover2, K. Anastos3, K. Mulligan4 , T. Tan5, Q. Shi6, C. Hyman7, M. Cohen8, SR. Cole9, J. Justman10, and
Women's Interagency HIV Study
Contact Information:
Magdalena Sobieszczyk, MD, MPH
630 W. 168th St., PH 8W-876
New York, NY 10032
Email: [email protected]
From the 1Columbia University College of Physicians & Surgeons, New York, NY; 2Rutgers University, Piscataway, NJ; 3Montefiore Medical Center and Lincoln Medical and Mental Health
Center, Bronx, NY;4University Of California, San Francisco, CA; 5Data Solutions, LLC, Bronx, NY; 6New York Medical College, Valhalla, NY; 7 Kings County Hospital Center, Brooklyn, NY;
8Cook County Hospital, Chicago, IL; 9Johns Hopkins Bloomberg School of Public Health, Baltimore, MD; 10Mailman School of Public Health, Columbia University, New York, NY.
UPDATED ABSTRACT
METHODS
Background
The prevalence of the Metabolic Syndrome (MetSynd) has been well described
in the general population and has been rising, particularly among women. This
constellation of metabolic disturbances is frequently associated with increased
risk of cardiovascular morbidity and mortality in the general population.
There are limited data, however, about the prevalence of MetSynd among HIVinfected persons, but it appears to be higher than in the general population. Few
prior studies have been performed to evaluate the risks and prevalence of
MetSynd in the growing population of HIV-infected women.
Methods
Prevalence for MetSynd was assessed among HIV-positive and negative women
in the Women’s Interagency HIV Study (WIHS) seen from October 2000 to
October 2004.
Results
Of the 668 HIV-infected and 1726 HIV-uninfected women who provided fasting
samples at one visit, MetSynd was more prevalent among the infected (33%)
than uninfected (22%) women (p<0.0001). In the HIV-infected compared to the
uninfected women, mean triglyceride level was higher (154 mg/dl vs. 102 mg/dl,
p<0.0001) and mean HDL cholesterol was lower (47 mg/dl vs. 55 mg/dl,
p<0.0001). No difference was observed in mean fasting glucose or blood
pressure. Increased triglycerides and low HDL were the most common
components present among those with MetSynd for both HIV-positive and
negative women. 32% and 29% of HIV-infected women were on PI-based and
non-PI based HAART, respectively.
In multivariate analyses restricted to HIV-infected women, factors significantly
associated with MetSynd were: older age (OR=1.42 per 5 years, p<0.0001),
White (OR=1.90, p<0.001) or Hispanic (OR=1.50, p<0.001) compared to Black
race/ethnicity; higher BMI (OR=3.27 and 9.15, for BMI 26-30 and >30 kg/m2
respectively; p<0.001 for both), current smoking (OR=1.29, p<0.05) and use of
stavudine (p=1.29, p<0.05). There was a trend to higher prevalence of MetSynd
in women with a CD4 count >350 cells/μL (OR=1.20, p=0.05), or with higher HIV1 RNA (OR=1.20, p=0.05, and OR=1.36, p<0.05, for HIV-1 RNA 10,001-50,000
and >50,000, respectively). Among the HIV-uninfected women, BMI, age, parity
and smoking status were associated with metabolic syndrome.
Conclusions
Metabolic syndrome is more prevalent in the HIV-infected than HIV-uninfected
women. Increased MetSynd risk is associated with older age and other traditional
risk factors in both groups. For HIV infected women higher HIV RNA levels and
use of Stavudine were associated with an increased prevalence of MetSynd.
Study Population
The WIHS is a prospective, multicenter cohort study of HIV-positive and high-risk HIV-negative in six innercity sites across the United. 668 HIV-uninfected and 1726 HIV-infected women provided fasting samples at
>1visit between October 2000 and October 2004. Although a woman may have met MetSynd criteria at more
than one visit, she was counted only once in the analysis. An index visit was defined as the participant’s first
visit with complete waist, blood pressure and fasting blood values during the study period. Women reporting
pregnancy at any study visit were excluded
Primary Outcome Variable and Definitions
This was a cross sectional study and the main outcome was prevalence of metabolic syndrome at a visit.
The recently updated National Cholesterol Education Program (NCEP) Adult Treatment Panel III (ATP III)
definition was used to identify participants with the metabolic syndrome. According to this definition,
participants who had three or more of the following criteria were defined as having the metabolic syndrome:
(1) waist measurement >88 cm (> 35 inches) for women; (2) hypertriglyceridemia of 150 mg/dL or higher or
drug treatment for elevated triglycedrides; (3) reduced HDL cholesterol of less than 50 mg/dL or drug
treatment for reduced HDL-C; (4) elevated blood pressure (systolic > 130 mmHg or diastolic > 85 mmHg
using the average of 2 seated measurements) or antihypertensive drug treatment in a patient with a history of
hypertension; (5) elevated fasting glucose > 100 mg/dL or currently using antidiabetic medication.
Statistical Methods
Robust covariance logistic regression models with repeated visits nested within study subjects were used to
evaluate association between metabolic syndrome and covariates both overall and separately within HIV
serostatus groups. Univariate modeling was initially performed to assess the contribution of HIV status, age,
BMI, smoking, alcohol use and other variables. Variables found significant in the univariate analysis were
then fit simultaneously in multivariate models. Multivariate analyses were performed separately for each
serogroup. Analyses were performed using SAS software (Version 9.0).
CONCLUSIONS
Metabolic syndrome is more prevalent in the HIV-infected than HIV-uninfected women. This
increased prevalence was due primarily to lipids rather than other factors
Among HIV-uninfected women the traditional risk factors of BMI, smoking, age and parity
were associated with the metabolic syndrome, while race/ethnicity was not
Among HIV-infected women, white race and Hispanic ethnicity, higher CD4 count, higher HIV1 RNA levels and stavudine use were significantly associated with the metabolic syndrome in
a multivariate model
Nevirapine and light to moderate alcohol use were significantly associated with lower
prevalence of MetSynd
In the multivariate models, ritonavir boosted PI use was not significantly associated with the
metabolic syndrome
RESULTS
Table 1: Patient Characteristics
Mean Age (years)
HIVnegative
N=668
36 (10)*
HIVpositive
N=1726
40 (9)*
28
59
10
4
28
57
12
3.0
Mean BMI (kg/m2 )
29 (8)*
28 (7)*
6
25
26
41
7
29
29
31
Characteristic
HIV-negative HIV-positive
% (N)
% (N)
P
value
HIV-negative
60
HIV-positive
56%
55
<.00
01
0.3
Ethnicity: (%)
Hispanic
African American
White
Other
BMI Category (%)
<20
21-25
26-30
>30
P
valu
e
Metabolic
Syndrome
Waist
circumference >88
cm
0.00
3
0.00
02
22
33
<.0001
50
44%
43%
45
42 (281)
42 (724)
39%
ns
40
34%
35 (596)
% of Women
Variable
Figure 1. Proportion of Women with
Components of Metabolic Syndrome
Table 2. Prevalence of Metabolic Syndrome
and its components by HIV status
<.0001
Triglycerides >150
mg/dl
16 (106)
HDL-C < 50 mg/dl
40 (269)
63 (1085)
<.0001
Elevated BP
26 (172)
32 (552)
<.0001
35
30%
27% 28%
30
25
20
15
10
Elevated fasting
glucose
19 (124)
18 (305)
ns
5
0
Hepatitis C Ab positive
(%)
Smoking status (%)
Never
Past
Current
Alcohol use (%)
Light (<3 drinks/wk)
Moderate (3-13 drinks/wk)
Heavy (>14 drinks/wk)
None
Postmenopausal Status
(%)
Parity Status (% live
births)
None
1-2
3-4
>5
Mean CD4 cell count
(cells/µL)
18
28
18
52
33
20
46
31
17
7
45
27
12
3
57
0.3
1
29
39
22
11
NA
NA
ART use (%)
No ART
ART naive
HAART/PI
HAART/no PI
Efavirenz
Stavudine
Nevirapine
Ritonavir boosted PI
NA
1 Index
0
<.00
01
1
2
>=3
Number of Components of the Metabolic Syndrome
0.02
Mean HIV-1 RNA level
(log10 copies/ml))
Illicit Drug use (%) 2
Crack
Cocaine
Heroin 3
IVDU
29
19
42
27
12
Table 3: Components of Metabolic Syndrome
Mean values by HIV-1 status
<.00
01
0.04
<.00
01
HIVpositive
Mean Waist
circumference
(cm)
91 (17)1
90 (15)
<.0001
Mean fasting
Triglycerides
(mg/dl)
102 (64)
154
(134)
<.0001
Mean fasting HDL
cholesterol
(mg/dl)
428
Mean Sys/Diastolic BP
(mm HG)
3.1
Mean fasting blood
glucose
(mg/dl)
23
17
13
10
P
value
HIVnegative
Characteristic
34
18
32
29
15
24
14
8
26
20
16
9
Table 4. Multivariate Analysis of Factors
Associated with Metabolic Syndrome
1 std,
55 (16)
47 (18)
117/75
91(33)
117/75
91(33)
visit defined as the first visit at which fasting blood samples were available;
Current or past use; 3 Smoking or intranasal use; *standard deviation
2
HIV-negative
[95% CI]
Race/Ethnicity
White
Hispanic
Other
Black
---------
<.0001
0.75/0.
63
0.64
standard deviation
Data in this manuscript were collected by the Women's Interagency HIV
Study (WIHS) Collaborative Study Group with centers (Principal
Investigators) at New York City/Bronx Consortium (Kathryn Anastos);
Brooklyn, NY (Howard Minkoff); Washington DC Metropolitan Consortium
(Mary Young); The Connie Wofsy Study Consortium of Northern California
(Ruth Greenblatt); Los Angeles County/Southern California Consortium
(Alexandra Levine); Chicago Consortium (Mardge Cohen); Data
Coordinating Center (Stephen Gange). The WIHS is funded by the National
Institute of Allergy and Infectious Diseases with contributions from the
National Cancer Institute, the National Institute on Drug Abuse, the National
Institute of Child Health and Human Development. Magdalena Sobieszczyk
was funded by the CIDER ID-Epidemiology Training Grant (NIH T32,
AI49821 ).
Smoking Status
Past
Current
1.40 [1.26-1.54]1
1.50 [0.94,2.41] 4
2.10 [1.36,9.11] 2
Alcohol use
Light
Moderate
Heavy
CD4 count (cells/µL)
200-350
>350
NA
Plasma HIV RNA
(cells/ml)
81-10,000
10,001- 50,000
>50,000
NA
2
p<0.05;
0.95 [0.75,1.22] 4
1.29 [1.04,1.60] 2
reference
1.80 [0.36,5.16] 4
8.10 [2.65,24.70] 1
28.30 [9.30,86.04] 1
NA
p<0.0001;
1
0.81 [0.69,0.94] 3
0.62 [0.49,0.78] 2
0.70 [0.43,1,.05] 4
HIV Treatment
No ART*
Stavudine
Nevirapine
PI/ritonavir
1
1.42 [1.34,1.50]
----
BMI (kg/m2)
<20
21-25
26-30
>30
Parity Status
1-2 live births
3-4 live births
> 5 live births
HIV-positive
[95% CI]
1.90 [1.44,2.55] 1
1.50 [1.20-1.84] 1
1.27 [0.75-2.18] 4
reference
Age (per 5 years)
ACKNOWLEDGEMENTS
0.07
0.04
0.23
0.60
Characteristic
reference
1.61 [1.13,2.28] 2
3.27 [2.27,4.71] 1
9.15 [6.31,13.28]1
reference
1.29 [1.07,1.55] 2
0.75 [0.60,0.95] 2
1.15 [0.93,1.43] 4
1.11 [0.86,1.35] 4
1.20 [1.05,1.61] 3
1.02 [0.88-1.18] 4
1.20 [1.00,1.48] 3
1.36 [1.06,1.76] 2
---4
1.40 [0.84,2.29]
1.40 [0.78,2.42] 4
2.40 [1.32,4.32] 2
3p=0.05, 4
p=ns * No ART at the index visit