The objective of this study was to develop a simplified scoring system to p
redict 30-day mortality in patients with acute ischemic stroke. A retrospec
tive cohort study was performed in a tertiary referral hospital in the Hunt
er Region of Australia. A prognostic index was created by assigning points
to the variables in a Cox model. The index included impaired consciousness
(5 points), dysphagia (3 points), urinary incontinence (4 points), admissio
n body temperature higher than 36.5 degreesC (2 points). and hyperglycemia
without a clinical history of diabetes (2 points). A score of II or more de
fined a high-risk group. The index achieved a sensitivity, specificity, and
positive predictive value of 68%, 98% and 75 %, respectively, in the deriv
ation sample and 57%. 97% and 68%, respectively. in the validation sample.
The results provide a simple risk stratification instrument for clinical re
search and practice. Further evaluation of the model in a prospective cohor
t is warranted (C) 2001 Elsevier Science Inc. All rights reserved.