Evaluation of Acute-phase Reactants, Immunologic Markers and other Clinical and Laboratory Parameters in Patients with Pneumonia and Non-pneumonic Lower Respiratory Tract Infections

More efficacious tools to distinguish pneumonia from other acute lower respiratory tract infections (OALRTI) in facilities where radiologic studies are not easily or rapidly available are desirable to select the patients who should undergo chest radiographs, to avoid unnecessary visits to the emergency wards of hospitals and to optimize health resources. To this end we analyzed the relevance of many clinical and laboratory parameters, including acute-phase reactants and immune activation markers, in 98 patients with pneumonia and 149 with OALRTI seen at the emergency ward of our hospital. Many clinical and laboratory parameters were associated with the diagnosis of pneumonia in the univariate analysis. Among them, C-reactive protein proved to be the most discriminant for the differentiation between the two conditions (area under the ROC curve 0.83, 95%CI 0.78-0.89, P<0.0001). A multivariate logistic regression analysis revealed that C-reactive protein, presence of suggestive auscultatory findings, lower age, presence of pleuritic pain and lower percent of eosinophils were independently associated with the diagnosis of pneumonia. A formula was derived from this analysis, which, for the most discriminant cut-off level, correctly classified pneumonic and non-pneumonic patients with a sensitivity of 88% and specificity 90%. The area under the ROC curve of this predictive model was 0.93 (95% CI 0.89-0.96, P<0.0001). C-reactive protein, especially if combined with other easily obtained parameters, constitutes a useful adjunct for the differentiation of pneumonia from OALRTI. Routine measurement of these parameters could result in a more adequate utilization of resources.