Renal dysfunction in stroke patients: A hospital‐based cohort study and systematic review

Background and purpose Renal dysfunction (i.e. a reduced estimated glomerular filtration rate, e GFR ) is commonly found in hospitalized stroke patients but its associations with patients' characteristics and outcome require further investigation.Methods We linked clinical data from stroke patients enrolled between 2005 and 2008 into two prospective hospital registers with routine laboratory e GFR data. The e GFR was calculated using the M odification of D iet in R enal D isease method and renal dysfunction was defined as <60 ml/min/1·73 m 2 . In addition we systematically reviewed studies investigating the association between e GFR and outcome after stroke. Results Of 2520 patients who had an e GFR measured on admission hospital, 805 (32%) had renal dysfunction. On multivariate analysis, renal dysfunction was significantly less likely in those with a predicted good outcome ( OR 0·27, 95% CI 0·21, 0·36) based on the well‐validated six simple variable model. After adjustment for other predictive factors, stroke patients with renal dysfunction were more likely to die in hospital compared with those without (odds ratio 1·59, 95% confidence intervals 1·26, 2·00). Of the 31 studies involving 41 896 participants included in the systematic review, 18 studies found that low e GFR was an independent predictor of death and 6 reported a significant association with death and disability.Conclusion Our findings suggest that renal dysfunction on admission is common and associated with poor outcomes over the first year. Further work is required to establish to what extent these associations are causal and whether treating impaired renal function improves outcomes.