759 
A Prospective Study of Abnormal Glucose Tolerance among Older Adults with or at-risk for HIV
Andrea Howard*, R Klein, M Floris-Moore, J Arnsten, and E Schoenbaum
Montefiore Med Ctr, Albert Einstein Coll of Med, Bronx, NY, US
Background: The
risk of impaired glucose tolerance (IGT) and diabetes in HIV+ patients
receiving HAART has not been well defined.
Methods: We
performed 2 oral glucose tolerance tests (OGTT) a median of 18.5 months (IQR
18.1 to 19.5) apart in 198 HIV+ and 125 at-risk HIV
adults without a history of diabetes at baseline. Standardized interviews were
administered to assess sociodemographic and medical
data including HAART use. Height and weight were measured. IGT was defined as
fasting glucose <126 mg/dL and 2-hour glucose 140
to 199 mg/dL; diabetes was defined as fasting glucose
³126
mg/dL or 2-hour glucose ³200 mg/dL,
or self-reported use of antidiabetic medication. c2 tests and
logistic regression analysis were performed to assess the associations of HIV
and HAART (none vs protease inhibitor [PI] vs non-PI) with incident abnormal glucose tolerance (IGT or
diabetes).
Results:
Participants were 42% male, 55% black, and 32% Hispanic. At baseline,
median age was 49 years (range 35 to 73); 34% were overweight (body mass index
25.0 to 29.9 kg/m2) and 31% obese (body mass index ³30.0
kg/m2). With no difference by HIV status, 46% reported a history of
injection drug use (p = 0.98). Among
HIV+ participants, 52% were on PI-HAART, 26% non-PI HAART, and 22%
were HAART-naïve; median CD4+ count was 461
cells/mm3 (range 36 to 1369). On the baseline OGTT, of 323
participants, 23 (7%) had diabetes and an additional 51 (16%) had IGT. On
follow-up, 4% (12 of 300) had incident diabetes; 10 of 12 were diagnosed by
OGTT and 2 self-reported new use of antidiabetic medication. Of 249
participants without IGT or diabetes at baseline, 23 (9%) had incident abnormal
glucose tolerance. There was no difference in the incidence of abnormal glucose
tolerance by HIV status (8% for HIV+, 12% for HIV, p = 0.34), or among HIV+
participants, by HAART use (7% for no HAART, 9% for PI-HAART, 8% for non-PI
HAART, p = NS). In a model including
HIV status and HAART use, factors associated with incident abnormal glucose
tolerance were age ³50
(ORadj 3.9, 95%CI 1.4, 10.9) and obesity (ORadj 3.2, 95%CI 1.3, 8.3).
Conclusions: In
this prospective study of mostly black or Hispanic older adults, classic
diabetes risk factors, rather than HIV or HAART, predicted the development of
abnormal glucose tolerance. In the HIV primary care setting, lifestyle
interventions aimed at obtaining and maintaining a normal body weight should be
implemented to reduce the risks of abnormal glucose tolerance and associated
cardiovascular complications.
|