icon-    folder.gif   Conference Reports for NATAP  
  18th CROI
Conference on Retroviruses
and Opportunistic Infections
Boston, MA
February 27 - March 2, 2011
Back grey_arrow_rt.gif
Diabetes Incidence in People With HIV Falling Since 1999-2000 in France
  18th Conference on Retroviruses and Opportunistic Infections, February 27-March 2, 2011, Boston
Mark Mascolini
Diabetes incidence (the new diagnosis rate) in a large French cohort of antiretroviral-treated people peaked in 1999-2000 and fell steadily through 2007-2009 [1]. At the turn of the millennium, diabetes incidence stood above 20 cases per 1000 person-years, much higher than rates reported in the Swiss HIV Cohort Study (SHCS) and in D:A:D, a multinational longitudinal cohort study, which had respective rates of 4.4 and 5.7 per 1000 person-years. Now incidence in the French cohort is similar to rates in SHCS and D:A:D, perhaps partly because of a move away from more toxic antiretrovirals.
The French APROCO-COPILOTE cohort enrolled protease inhibitor (PI)-naive people from 1997 through 1999. Everyone started a PI regimen. This analysis focused on 1046 cohort members who did not have diabetes when they enrolled and who had at least two glycemia measures. The investigators defined diabetes as at least two fasting glucose levels at or above 7.0 mmol/L (126 mg/dL) or at least two nonfasting levels above 11.1 mmol/L (200 mg/dL). They separated cohort members into four glycemia groups:
Normal glycemia: normal glycemia on all measures (fasting and nonfasting glucose always below 5.6 or 7.8 mmol/L)
Impaired glycemia: At least one fasting glucose between 5.6 and 7 mmol/L or one nonfasting glucose between 7.8 and 11.1 mmol/L
Diabetic glycemia once: one fasting or nonfasting glucose in the diabetic range
Diabetes: as defined above
Of the 1046 cohort members studied, 111 (11%) had diabetes, 111 (11%) had diabetic glycemia once, 450 (43%) had impaired glycemia, and 374 (36%) had normal glycemia. Overall diabetes incidence for the entire study period was 14.1 cases per 1000 person-years. Diabetes incidence rose from about 15 cases per 1000 person-years in 1997-1998 to above 20 cases per 1000 person-years in 1999-2000. After that, incidence fell steadily to about 5 cases per 1000 person-years in 2005-2006 and 2007-2009.
Multivariate analysis identified seven independent predictors of diabetes incidence at the following hazard ratios (HR) (and 95% confidence intervals):
Older age at cohort inclusion
-- 40-50 versus under 40: HR 2.13 (1.36 to 3.33), P < 10(-4)
-- 50 or older versus under 40: HR 3.15 (2.16 to 5.72, P < 10(-4)
Body mass index (kg/m2) at cohort inclusion
-- 25 to 30 versus under 25: HR 1.98 (1.27 to 3.08), P = 0.003
--30 or over versus under 25: HR 2.27 (1.08 versus 4.75), P = 0.003
Waist-to-hip ratio
-- At or above 0.97 for men and 0.92 for women versus lower: HR 3.95 (2.36 to 6.58), P < 10(-4)
Lipoatrophy at least once
-- HR 1.92 (1.20 to 3.09), P = 0.007
Didanosine use
-- HR 1.51 (1.01 to 2.26), P = 0.05
Stavudine use
-- HR 2.77 (1.68 to 4.57), P < 10(-4)
Indinavir use
-- HR 1.92 (1.30 to 2.83), P = 0.001
Factors that did not independently affect risk of diabetes diagnosis included lipohypertrophy, hypertension, HCV coinfection, family history of diabetes, CD4 count, CD4/CD8 ratio, viral load, an AIDS diagnosis, or being naive to combination antiretroviral therapy when entering the cohort.
Although the APROCO team did not say so, the correlations with three older antiretrovirals and the declining incidence since 2000 suggest that newer combinations pose a lessened threat to glucose control. Also, growing awareness of diabetes as a threat to people with HIV probably improved prevention and treatment of high glucose, factors not measured in this analysis.
Because of the heightened risk of heart disease with diabetes, the investigators stressed the importance of "screening, close surveillance, and targeted intervention on glucose parameters in long-term treated HIV-infected patients."

1. Capeau J, Bouteloup V, Raffi F, et al. Diabetes mellitus in treated HIV-infected patients: incidence over ten years in 1,046 patients from the ANRS CO8 APROCO-COPILOTE cohort. 18th Conference on Retroviruses and Opportunistic Infections. February 27-March 2, 2011. Boston. Abstract 850.