Background: Stroke is the third cause of death in older individuals living in Western Countries. The identification of predictors for mortality after stroke has a major importance for clinicians in order to allow the implementation of therapeutic and preventive strategies.
Objective: To evaluate the association between clinical and laboratory parameters and 30-days total mortality in a large sample of older patients with stroke.
Methods: 469 older patients (median age: 80.0 years) consecutively hospitalized for acute ischemic stroke were enrolled. The data recorded included: (1) clinical features of stroke; (2) routine clinical chemistry analyses; (3) medical history, and (4) 12-lead ECG. All patients underwent computed tomography scan of the brain. Stroke type was classified by the Oxfordshire Community Stroke Project system.
Results: 130 subjects died within 30 days after stroke, with an overall mortality of 27.7%. At univariate analysis, altered levels of consciousness (ALC), congestive heart failure, atrial fibrillation, previous stroke, high blood glucose, fibrinogen and blood sedimentation rate levels, higher white blood cell count, lower serum albumin and iron levels were associated with mortality. Multivariate logistic regression analysis indicated that short-term mortality was associated with ALC (OR: 11.80; CI 95%: 5.50-24.00), congestive heart failure (OR: 3.06; CI 95%: 1.04-8.80), and age (OR: 1.04; CI 95%: 1.002-1.09) independent of gender, previous stroke, AF, fasting blood glucose, serum albumin, serum iron, and white blood cell count. In patients with ALC (high-mortality rate: 63.6%), only hyperglycemia (III vs. I tertile, OR: 9.60; CI 95%: 1.65-52.50) was associated with mortality after multivariate adjustment.
Conclusion: Our study highlights the role of ALC and congestive heart failure in the short-term prognostic stratification of older patients with acute ischemic stroke. Furthermore, our results support the value of post-stroke hyperglycemia as a marker for short-term mortality also in advanced age, and particularly in the presence of ALC and in nondiabetic individuals.