Serum Procalcitonin Measurement and Viral Testing to Guide Antibiotic Use for Respiratory Infections in Hospitalized Adults: A Randomized Controlled Trial

Abstract

Background: Viral lower respiratory tract illness (LRTI) frequently causes adult hospitalization and is linked to antibiotic overuse. European studies suggest that the serum procalcitonin (PCT) level may be used to guide antibiotic therapy. We conducted a trial assessing the feasibility of using PCT algorithms with viral testing to guide antibiotic use in a US hospital.

Methods: Three hundred patients hospitalized with nonpneumonic LRTI during October 2013-April 2014 were randomly assigned at a ratio of 1:1 to receive standard care or PCT-guided care and viral PCR testing. The primary outcome was antibiotic exposure, and safety was assessed at 1 and 3 months.

Results: Among the 151 patients in the intervention group, viruses were identified in 42% (63), and 83% (126) had PCT values of <0.25 µg/mL. There were no significant differences in antibiotic use or adverse events between intervention patients and those in the nonintervention group. Subgroup analyses revealed fewer subjects with positive results of viral testing and low PCT values who were discharged receiving antibiotics (20% vs 45%; P = .002) and shorter antibiotic durations among algorithm-adherent intervention patients versus nonintervention patients (2.0 vs 4.0 days; P = .004). Compared with historical controls (from 2008-2011), antibiotic duration in nonintervention patients decreased by 2 days (6.0 vs 4.0 days; P < .001), suggesting a study effect.

Conclusions: Although antibiotic use was similar in the 2 arms, subgroup analyses of intervention patients suggest that physicians responded to viral and biomarker data. These data can inform the design of future US studies.

Clinical trials registration: NCT01907659.

Keywords: antibiotic use; procalcitonin; respiratory infections; viral testing.

Figure 1.

Flow of patients through the study. Reasons for exclusion from the study included intensive care unit (ICU) stay, antibiotic use for >24 hours prior to enrollment, active chemotherapy, conditions known to increase procalcitonin level (eg, renal failure, pancreatitis, and trauma), definite infiltrate on a chest radiograph (according to radiology report), >15% bands on a peripheral blood smear, and a systolic blood pressure (SBP) of <90 mm Hg at enrollment. Abbreviation: PCT, procalcitonin.

Figure 2.

Time series plot of total antibiotic days in intervention and nonintervention patients and a subgroup analysis of virus-positive intervention subjects with low procalcitonin (PCT) levels, compared with the nonintervention arm. Abbreviation: IQR, interquartile range.

Figure 3.

Provider response to the procalcitonin (PCT)–guided treatment algorithm. A circle represents an individual PCT value for each intervention study subject. The horizontal bar represents the threshold for PCT values (0.24 ng/mL), which defines levels as either low or high. The algorithm discourages antibiotic use below this threshold and recommends antibiotics for values above this level. Results are segregated by provider response to the algorithm and are designated as “algorithm followed” and “algorithm rejected.”