BACKGROUND: People with chronic kidney disease (CKD) have an increased risk of stroke but the magnitude of increased risk and the independent effects of glomerular filtration rate (GFR) and albuminuria are unclear. We aimed to quantify the association between the independent and combined effects of GFR and albuminuria on stroke risk. METHODS: We searched MEDLINE and EMBASE (February 2014) for cohort studies or randomized controlled trials (RCTs) which reported stroke incidence in adults with a baseline measurement of GFR and/or albuminuria. We extracted study and participant characteristics, risk of bias and relative risks (RR, with confidence interval; CI) of stroke associated with GFR and/or quantity of albuminuria, synthesized data using random effects meta-analysis and explored heterogeneity using meta-regression. RESULTS: We identified 83 studies; 63 cohort studies (2 085 225 participants) and 20 RCTs (168 516 participants) reporting 30 392 strokes. There was an inverse linear relationship between GFR and risk of stroke, with risk of stroke increasing 7% (RR: 1.07, CI: 1.04-1.09) for every 10 mL/min/1.73 m2 decrease in GFR. A 25 mg/mmol increase in albumin-creatinine ratio was associated with a 10% increased risk of stroke (RR: 1.10, 95% CI: 1.01-1.20). The effect of albuminuria was independent of GFR. Results were not different across subtypes of stroke, sex and varying prevalence of cardiovascular risk factors. CONCLUSIONS: Stroke risk increases linearly and additively with declining GFR and increasing albuminuria. CKD staging may also be a useful clinical tool for identifying people who may benefit most from interventions to reduce cardiovascular risk.
a0931-0509 (Print)