An assessment of individual risk factors may identify a subgroup of postinfarction patients at low risk, i.e. patients appropriate for early discharge. Using a large unselected population of the national registry, 'The 60-Minutes Myocardial Infarction Project', we (1) attempted to provide a retrospective analysis of clinical factors and in-hospital mortality in a population living on the 6th hospital day following admission to define a low-risk patient group with a residual in-hospital mortality of less than 1% eligible for early discharge, and (2) to analyze the current impact of risk stratification based on these clinical factors on the length of hospitalization. The study group consisted of 12,045 survivors on the 6th day after admission out of 14,980 patients of the registry with proven Q-wave myocardial infarction. Risk modeling was performed with multiple logistic regression.
Results: A total of 873 patients (7.3%) died after day 6 in hospital. The most important prognostic factors were cardiopulmonary resuscitation prior to admission (odds ratio, OR: 7.2, confidence interval, CI: 5.11-10.22), thrombolysis complicated by severe bleedings (OR: 6.2, CI: 1.2-31. 2) and age >70 years (OR 4.7, CI 3.51-6.39). The other more significant independent predictors of increased mortality were end-stage renal disease, age between 56 and 70 years, systolic blood pressure <95 mm Hg on admission, history of trauma </=2 months, cancer and left-bundle-branch block. Summarizing these nine groups of patients with the strongest association to in-hospital mortality, we defined a high-risk group comprising 79% of the AMI patients with a residual in-hospital mortality of 8.8%. On the other hand, by excluding these nine high-risk patient groups, a low-risk group of 21% of all AMI patients seems to be appropriate for early discharge (residual in-hospital mortality = 1.07%). However, in the current practice, there was no difference regarding the median length of hospital stay between the two risk groups. The low-risk patients were hospitalized 20 days compared to 22 days in the high-risk patients.
Conclusion: Using a simple logistic regression model, which considers clinical factors of the early hospital phase, one fifth of the infarction patients can be stratified to be at low risk, and might be eligible for early hospital discharge. Currently, an individual risk stratification has no impact on the length of hospital stay in Germany.