Among 1976 pre-dialyzed HIV subjects, 661 were prospectively followed-up for 4 years to determine incidence of composite outcomes, including all-cause mortality, cardiovascular disease and

a decline over 25% from baseline in eGFR. Four risk categories (0 to 3) were constructed using the combination of 5 stages of eGFR and 3 grades of albuminuria. The cumulative incidence of the outcomes was analyzed with Kaplan-Meier method, and hazard risk (HR) of risk categories for the outcome incidence was calculated using multivariable proportional hazards regression analysis, adjusted for some known risk factors. Results: The frequency of each CKD category was shown ABT 888 in Figure 1. The prevalence of HIV infection was 0.024% in the chronic HD patients. The Kaplan-Meier estimates were significantly increased over time in the risk categories 2 and 3, compared with the risk categories 0 and 1 (Figure 2). The HR of risk categories 2 and 3 was 2-fold greater (HR = 2.00; its 95% confidence interval, 1.08–3.57; P = 0.0277), as compared to risk categories 0 and 1. Conclusion: The new CKD classification may facilitate targeting of high-risk CKD in the HIV-infected population as well as in the general population. WU SUNNY1, MASSON PHILIP1,

DUTHIE FIONA2, PALMER SUETONIA1,3, STRIPPOLI GIOVANNI1, WHITELEY WILL2, WEBSTER ANGELA1 1University of Sydney; 2University of Edinburgh; 3University of Otago Introduction: Cognition affects quality of life, medication management and survival. We aimed to summarise how CKD affects cognitive function. Methods: We searched databases Z-IETD-FMK molecular weight (Jan 2014) for studies measuring cognitive function using validated neuropsychological tools in participants with CKD. We extracted measures of cognition and synthesised results for cognitive domains stratified by glomerular filtration rate (GFR) using random effects, expressed as

standardized mean differences (SMD) with 95% confidence intervals (CI). Depending on the study design, we assessed quality using either the Newcastle-Ottawa scale or the Cochrane risk of bias tool. Results: In interim Tenoxicam analyses, we included 28 studies (112,714 participants): 17 cross-sectional studies (37,889 participants), 10 cohort studies (46,441 participants) and 1 randomised control trial (28,384 participants). Studies measured cognition using 43 different tools. Cognitive domains were measured with varying frequencies: global cognition (23 studies), executive function (14 studies), attention (14 studies), processing speed (14 studies), memory (13 studies), language (9 studies), visuo-spatial perception (5 studies) and intelligence (2 studies). No study measured psychomotor function. Overall, methodological quality of cohort studies was better than cross-sectional studies where analyses were unadjusted for potential confounders, making meaningful comparison challenging. Compared to people with GFR >60 ml/min/1.