Dehydration is an independent predictor of discharge outcome and admission cost in acute ischaemic stroke

Background and purpose: Our aim was to investigate the influence of admission dehydration on the discharge outcome in acute ischaemic and hemorrhagic stroke. Methods: Between January 2009 and December 2011, 4311 ischaemic and 1371 hemorrhagic stroke patients from the stroke registry of Chang Gung healthcare system were analyzed. The eligible patients were identified according to inclusion/exclusion criteria. In total, 2570 acute ischaemic and 573 acute hemorrhagic stroke patients were finally recruited. According to the blood urea nitrogen (BUN) to creatinine (Cr) ratio (BUN/Cr), these patients were divided into dehydrated (BUN/Cr ≥ 15) and non-dehydrated (BUN/Cr < 15) groups. Demographics, admission costs and discharge outcomes including modified Rankin scale (mRS) and Barthel index (BI) were examined. Data were analyzed using multivariate analysis of two-stage least squares including logistic and linear regression. Results: Acute ischaemic stroke with admission dehydration had higher infection rates (P = 0.006), worse discharge BI (62.8 ± 37.4 vs. 73.4 ± 32.4, P < 0.001, adjusted P < 0.001), worse mRS (2.7 ± 1.6 vs. 2.3 ± 1.5, P < 0.001, adjusted P = 0.009) and higher admission costs (2470.8 ± 3160.8 vs. 1901.2 ± 2046.8 US dollars, P < 0.001, adjusted P = 0.013) than those without dehydration. However, acute hemorrhagic stroke with or without admission dehydration showd no difference in admission costs (P = 0.618) and discharge outcomes (BI, P = 0.058; mRS, P = 0.058). Conclusion: Admission dehydration is associated with worse discharge outcomes and higher admission costs in acute ischaemic stroke but not in hemorrhagic stroke.