Introduction of Procalcitonin Testing Did Not Reduce Antibiotic Utilization for Acute Exacerbations of Chronic Obstructive Pulmonary Disease
Acute exacerbations of chronic obstructive pulmonary disease (AECOPDs) are a frequent cause of inpatient admission and antibiotics are often used for treatment, even without a clear indication. Procalcitonin (PCT), a biomarker for bacterial infections, has been used to safely discontinue antibiotic therapy in AECOPDs. We sought to characterize the impact of PCT’s introduction to our medical center on the management of AECOPDs.
Patients admitted from October 1, 2013 (the date PCT became available) to March 31, 2014 were included in the study if an AECOPD was present on admission and they were not transferred from another facility. Data were extracted through structured query and chart review. The primary outcome was antibiotic days on therapy (DOT) and secondary outcomes included 30-day readmission and mortality. Linear and logistic regression were used to model the outcomes.
A total of 238 patients met inclusion criteria. Mean age was 66.7, standard deviation ±12.3, 124 (52%) patients were female, and systemic inflammatory response syndrome (SIRS) was present in 115 (48%). PCT was measured on 73 patients. Presence of SIRS and length of stay (LOS) associated with PCT measurement, while age and gender did not. Overall, PCT measurement associated with more antibiotic DOT (+56%, standard error [SE] ±9%, P <.001) compared to patients on whom no PCT was measured, even when adjusted for SIRS and LOS. Among those on whom PCT was measured, low PCT levels (<0.25 ng/mL) associated with fewer DOT for intravenous (IV) antibiotics (-108%, SE ±22%, P<.001). When adjusted for LOS, PCT measurement was not associated with readmission or mortality.
Measurement of PCT during episodes of AECOPD did not reduce overall antibiotic utilization, though lower levels associated with fewer days on IV therapy. The impact of PCT on antibiotic utilization demonstrated in controlled studies may not be immediately realized in local clinical practice. Whether increased provider familiarity with PCT over time and/or targeted educational efforts can affect these findings deserves further study.
B. Wang, None
S. Winters, None
S. Flanders, None