[Community-acquired pneumonias. Admission criteria and complicated course indicators]

To evaluate the admission criteria and to select indicators that identify patients for whom hospitalization is not necessary, we studied 75 patients with community acquired pneumonia (CAP) who were admitted to a clinical service. According to Appropriateness Evaluation Protocol (AEP) only 60% of our patients justified their hospitalization (Group A) while 40% did not (Group B). The most frequent hospitalization criteria found in Group A were tachypnea (> 30x min.) (40%), respiratory failure (38%) and encefalopathy (18%). The average age in Group A was 62 versus 47 in Group B (p < 0.001). Comorbid conditions were present in 100% of Group A and 71% had two or more while only 33% of patients in Group B had two or more (p < 0.01). During the evolution, Group A had more organ failure than B (53 vs. 17%) (p < 0.001) and a longer period of hospitalization (14 vs. 9 days) (p < 0.01). The differences between groups A and B is best visualized in the incidence of sepsis (4 vs. 0%), and mortality rates (15% vs. 0%) (p < 0.05). Using the Fine risk criteria for a complicated course, we selected 14 patients from Group B, with one or more criteria (Group C) that were compared with 16 patients without them (Group D). The presence of a poor clinical status at admission was the only difference between Group D and C (79 vs. 0%) (p < 0.001). When three or more risk factors were present the differences were significant (79 vs. 6%) (p < 0.001). We conclude that the utilization of hospitalization criteria together with the risk factors for a complicated course, specifically when two or more factors per patient are present, permit the identification of a population with CAP that needs hospitalization with 71.4% sensitivity and 100% specificity. The presence of two or less risk factors in patients without admission criteria has a highly predictable negative value (100%) and anticipates an uneventful evolution without complications.