Predictors of clinical deterioration during hospitalization following acute ischemic stroke

Eur Neurol. 2012;67(3):186-92. doi: 10.1159/000334723. Epub 2012 Feb 23.

Abstract

Background/aims: A number of risk factors for early worsening of neurological symptoms have been identified. We aimed to evaluate the influence of hemorheologic, biochemical, and metabolic factors on neurological deterioration during hospitalization following acute ischemic stroke and develop a model of neurological deterioration.

Methods: Worsening of stroke was defined as a deleterious increase in NIH Stroke Scale (NIHSS) score of ≥4 points during hospitalization. We performed multivariate logistic regression analysis and constructed a prediction model based on chart data of 2,398 patients admitted at five medical centers; 203 of the patients had worsening of stroke and 2,186 had not.

Results: The results of multivariate logistic regression analysis showed that hemoglobin (odds ratio: 0.529) and albumin (odds ratio: 0.024) were significantly associated with stroke deterioration, as were the modified Rankin Scale on emergency department admission (odds ratio: 4.956) and length of hospitalization (odds ratio: 1.201). After adjusting for age, gender and NIHSS on emergency department admission, only hemoglobin (odds ratio: 0.894, 95% confidence interval: 0.814-0.981, p = 0.018) was associated with worsening.

Conclusion: Hemoglobin and albumin were found to be risk factors for persistent neurological deterioration during hospitalization following acute ischemic stroke, suggesting that blood viscosity may be related to neurological deterioration.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adolescent
  • Adult
  • Aged
  • Aged, 80 and over
  • Brain Ischemia / diagnosis*
  • Disease Progression*
  • Female
  • Hospitalization*
  • Humans
  • Male
  • Middle Aged
  • Predictive Value of Tests
  • Prognosis
  • Recovery of Function
  • Retrospective Studies
  • Risk Factors
  • Severity of Illness Index
  • Stroke / diagnosis*