Diabetes in a hospital cohort of persons living with HIV: a descriptive and comparative study in French Guiana.

ABSTRACT

BACKGROUND: In French Guiana (population 294,000) the prevalence of type 2 diabetes (10%) and of HIV(1.1%) are very high. Our objective was to determine the prevalence of diabetes and its complications in a HIV cohort. MATERIALS AND METHODS: We enrolled HIV-infected persons followed in Cayenne, Kourou, and Saint Laurent du Maroni hospitals between January 1, 1992 and December 31, 2021 in the French Hospital Database for HIV (FHDH) a national database compiling data from all French regions. RESULTS: There was no difference of diabetes prevalence between men (8.2%) and women (8.8%), P = 0.4. Patients with diabetes were older (56 years ± 13.4) than those without diabetes (44.7 years ± 13.6) and prevalence increased with age. The proportion of persons with diabetes was greater among virologically suppressed persons (10%) than those with a detectable viral load under antiretroviral treatment (5.8%). Persons with diabetes had substantially greater CD4 counts at diagnosis than persons without diabetes. The majority of macro and microvascular complications were observed in people with diabetes. Persons with diabetes and HIV were significantly less likely to have had AIDS (1.6 versus 2.2 per 100 person-years, respectively). Overall, 374 persons living with HIV of 4167 had died (9%) the proportion of persons with diabetes among the dead was greater than those who did not die 11.7% versus 8.1%, respectively, p = 0.017. However, persons with diabetes were older and hence died older, 62.3 years (SD = 1.9) for deceased persons with diabetes versus 50.4 years (SD = 0.8), P < 0.0001. However, using Cox regression to adjust for age, initial CD4 count, country of birth there was no significant difference in the Hazard for death between persons with diabetes and persons without diabetes (aHR = 0.99, 95%CI = 0.65-1.5), P = 0.9. CONCLUSIONS: The prevalence of diabetes in our HIV cohort was high. Persons with diabetes had greater CD4 counts, earlier care, and greater virological suppression than persons without diabetes. There were no significant differences between persons with diabetes and without diabetes in terms of survival.