Medical News Stories
Compliance with Procalcitonin Algorithm Antibiotic Recommendations for Patients in Medical Intensive Care Unit
Ammar AA, Lam SW, Duggal A, et al. Pharmacotherapy. 2017;37:177-86.
Study Question: What is the rate of provider compliance with a procalcitonin (PCT) algorithm that describes antibiotic recommendations, and is there a difference in compliance if PCT is measured when initiating versus continuing antibiotics?
Study Description: This was a single-center, retrospective cohort study involving adult patients who had a PCT value measured at least once while in the medical ICU from November 2011 to October 2013. The initiation cohort involved PCT values resulting before or within 24 hours of starting antibiotics while the continuation cohort involved PCT values resulting ≥24 hours after antibiotics started. Compliance was determined using an algorithm similar to the PRORATA trial (which evaluated PCT to reduce antibiotic exposure in the ICU) 24 hours before and after the PCT value resulted. Adherence to the PCT algorithm and choice of antibiotics was at the discretion of the provider.
Results: A total of 957 PCT values in 527 patients were included in the study. Approximately 56% and 44% were ordered in the initiation and continuation cohorts, respectively. The median day to obtaining a PCT were 2.1 days (IQR 1.7-5.9) and 6.4 days (IQR 3.3-12.1) in the initiation and continuation cohorts, respectively. There was no difference in compliance between the initiation and continuation cohorts (49% vs. 48%; p=0.68), nor was either of these cohorts impacted by actual PCT values (positive or negative). While algorithm compliance was lower with the first PCT value compared to subsequent PCT values on a patient (45% vs. 55%, p=0.002), the actual PCT value had no impact on compliance. Studied patient or PCT-related factors also did not independently affect algorithm compliance in a logistic regression analysis.
Conclusion(s): Overall PCT algorithm compliance amongst the antibiotic initiation and continuation cohorts was low.
Perspective: Algorithm compliance in this study was lower than the >75% rates reported in many previously published randomized controlled trials. Several reasons could have led to the low compliance rate including the retrospective application of an algorithm to patients cared for by prescribers not compelled to adherence, the inclusion of higher acuity patients compared to those in published literature, and the lack of an antibiotic stewardship program at the time of the study. Institutions implementing PCT for antibiotic guidance should evaluated compliance with established algorithms.