Utility of procalcitonin in predicting mortality among cases of hospital-acquired pneumonia: a North Indian study

Abstract

Bckground Data regarding the role of biomarkers like procalcitonin (PCT) in predicting treatment outcomes in hospital-acquired pneumonia (HAP) are limited in an Indian setting. We set out to determine the usefulness of PCT in predicting mortality among the cases of nosocomial pneumonia, at an 800-bed, apex tertiary care centre, in Kashmir (North India). Patients and methods Of the 318 confirmed cases of HAP, 60 consenting cases were selected randomly. Quantitative determination of PCT was done using immunofluorescence assays. Levels greater than 0.5 ng/ml were taken as positive. Data were collected on clinical parameters, and Acute Physiology and Chronic Health Evaluation II (APACHE-II) scores pneumonia severity index were calculated. Appropriate blood and respiratory cultures were performed. Results Of the 60 cases included in the study, 19 (32%) died during the hospital stay, of which 14 (74%) deaths occurred in PCT-positive cases (P=0.001). The median PCT level was higher in the in-hospital mortality group (1.07 vs. 0.25), with a mean value of 1.2±2.8 vs. 1.2±2.5 in the group with no mortality (P=0.000). Using multivariate analysis, positive PCT level was strongly associated with in-hospital mortality (odds ratio: 6.767, 95%CI: 1.992–22.984, P=0.002) and APACHE-II score greater than 20 (n=14, odds ratio=4.5, 95%CI=1.448–13.984, P=0.009). Using receiver operating characteristic analysis, PCT had apropos discrimination power for in-hospital mortality (0.713 of area under the curve) and higher APACHE-II scores (0.753 of area under the curve). Using Cox regression model for mortality in PCT-positive group, the calculated hazard ratio was 3.273 (95%CI: 1.076–9.951, P=0.037). Conclusion PCT might have a vital role in the management of HAP, as a predictor of the poor treatment outcome.