C-reactive protein on admission as a predictor of in-hospital death in the elderly with acute myocardial infarction.

Abstract

Background: The mean age of patients with acute myocardial infarction is increasing and the associated in-hospital mortality is exponentially age-related. Inflammation markers have been related to cardiovascular short and long-term prognosis. The aim of this study was to evaluate the short-term prognostic value of C-reactive protein (CRP) levels on admission in the oldest segment of the patients with acute myocardial infarction.

Methods: CRP was prospectively measured on admission by immunonephelometry in 205 consecutive old women (mean age 82 +/- 5 years) with definite acute myocardial infarction; values were then related to in-hospital mortality and the causes of death.

Results: CRP levels ranged from 0.1 to 31.9 mg/dl and were raised in 71% of the patients. It showed no significant correlation with baseline clinical variables such as age, history of diabetes or hypertension or prior myocardial infarction, infarct location, and time from symptom onset to admission. The overall in-hospital mortality rate was 25% and rose from 15% among patients in the lower three quartiles of CRP levels (cut point 6.4 mg/dl) to 55% among those in the upper quartile (p < 0.001). By univariate logistic-regression, the odds ratio for early death was 0.84 (95% confidence interval 0.78-0.89) for every increase by 1 mg/dl of CRP, and 5.7 (95% confidence interval 2.7-11.9) for a CRP level in the upper quartile. Multivariate analysis demonstrated the independence of the relation between CRP levels and in-hospital mortality (p = 0.0001). No significant differences in CRP level were found among deceased patients classified by cause of death (heart rupture 44%, pump failure 41%, comorbidity 5%).

Conclusions: CRP concentration is raised in many old patients with acute myocardial infarction and seems to independently stratify patients for in-hospital mortality risk. This prognostic information may assist in providing the appropriate level and duration of close monitoring and be an additional support to evaluate the risk-benefit ratio of thrombolytic therapy in some cases.