From the authors:
We thank S. Krüger and co-workers for their thoughtful comments and suggestions.
With regard to the Pneumonia Severity Index (PSI), most subsequent risk scores and prognostic markers in community-acquired pneumonia (CAP) research focused on 30-day mortality despite the well-known limitations of this end-point. Mortality may not be a directly related to infection, but rather due to comorbidities, advanced age or secondary complications within the follow-up period. In fact, within the Procalcitonin-Guided Antibiotic Therapy and Hospitalisation in Patients with Lower Respiratory Tract Infections (ProHOSP) study, deaths were evenly distributed within the 30-day follow-up, as demonstrated in the Kaplan–Meier plot. When restricting our analysis to short-term mortality within the first 3 days after admission, procalcitonin (PCT) showed a superior prognostic performance (area under the curve (AUC) 0.68). Also, considering not only initial but also follow-up PCT levels within our study significantly improved the prognostic performance of PCT. This was also true in a previousLegionellasp. CAP study [1] and intensive care unit (ICU) studies [2,3]. Therefore, we advocate the use of PCT for prognostication in an in-patient CAP setting primarily in combination with a clinical risk score that also incorporates static risk factors, such as age and comorbidities, and consideration of the kinetic of the marker …