The objectives of our study were to estimate the prevalence of RI in a large and unselected cohort of HIV-infected patients in care and to identify associated factors that could lead to specific preventive or control measures. We performed a cross-sectional survey within the French Agency buy PD98059 of AIDS and Hepatitis Research (ANRS) CO3 Aquitaine Cohort of HIV-infected patients living and followed in South-western France. Patients were enrolled

prospectively in this cohort through a hospital-based surveillance system, if they were aged 13 years or more and provided informed consent. Standardized epidemiological, clinical, biological and therapeutic data collection were completed by attending physicians at time of enrolment and at each hospital follow-up visit, generally every 3 or 6 months (in agreement with French recommendations for standards of care) or more frequently in case of an intercurrent event, then verified and coded by research nurses with an annual audit for quality control. In our study, the main outcome of interest was the renal filtration rate assessed by a single measurement of the clearance of creatinine (CC) using the Cockcroft–Gault (CG) formula [11] owing to the fact that creatininemia was routinely registered in our database from January 2004.

CC was measured using Jaffé methodology in the three laboratories where measurements have been carried out and calibrations have been performed to assure comparability. We did not standardize CG measurement for body surface area as there is no general consensus of whether or not this has to be performed [9]. The lack of data related to ethnicity in our systematic survey KPT-330 nmr did not allow the use of the Modification of Diet in Renal Disease (MDRD) formula to assess the renal function; nevertheless crude prevalence of HAS1 RI was calculated using the modified MDRD formula [12], which does not need to know ethnicity, to allow comparisons with other studies: CC mL/min=175 × (serum creatinine μM/L × 0.0113)−1.154× age−0.203× 0.742 (if female). In the analysis,

we included data of the cohort participants at the time of first follow-up where a simultaneous measurement of variables allowing the calculation of their CC was collected between January 2004 and September 2006. We then excluded patients with incomplete data on body weight, height and creatininemia. We also excluded patients with a body mass index (BMI) <18 or >30 kg/m2, ascites and pregnant women in order to ensure the validity of the CG formula. According to the recommendations of the HIV Medicine Association of the Infectious Diseases Society of America [12], we assigned normal renal function to patients with a CC value >90 mL/min and RI to those with a CC <90 mL/min. Four stages of RI were defined: mild RI for a CC between 60 and 90 mL/min; moderate RI for a CC between 30 and 60 mL/min; severe RI for a CC between 15 and 30 mL/min; and end-stage RI for a CC <15 mL/min.