Journal Club: Should we be Procalcitonin or Anti?

An overview of Huang, DT et al. (2018). Procalcitonin-Guided Use of Antibiotics for Lower Respiratory Tract Infection. N Engl J Med, 379(3), pp.236-249

https://www.ncbi.nlm.nih.gov/pubmed/29781385

Background: Procalcitonin is an acute phase reactant used to differentiate bacterial from viral infections. Previous studies have shown that decreasing procalcitonin level correlates with resolution of infection. In 2017, the FDA approved a procalcitonin assay to guide clinicians in deciding to start and/or discontinue antibiotics for patients with suspected lower respiratory tract infections. This was based on several European trials that demonstrated that procalcitonin-based antibiotic prescription guidelines reduced the use of antibiotics without further harm. The ProACT trial investigated whether the use of procalcitonin-based antibiotic guidelines would reduce the total antibiotic exposure days for patients with suspected lower respiratory tract infections in the US.

Study design: Randomized, controlled noninferiority trial

Study population: 1,656 adult patients with suspected acute lower respiratory tract infection for whom clinicians had not yet decided to use antibiotics were chosen from 14 major urban academic medical centers. Patients were excluded based on conditions where physicians were unlikely to withhold antibiotics (recent antibiotics, vasopressor use, mechanically ventilated, severe immunosuppression, coinfections, known lung abscess or empyema), conditions where procalcitonin can be elevated without infection (chronic dialysis, metastatic cancer, recent surgery), and conditions rendering follow-up or consent challenging (incarcerated or homeless patients). Patients were also excluded if the treating physician was unwilling to consider procalcitonin in antibiotic decision-making. 2,122 patients who did not meet exclusion criteria were still excluded from the study, mostly for logistical issues. Patients were randomized to two groups – the procalcitonin group and the usual-care group.

Intervention: Clinicians (residents, hospitalists, PCPs, PAs, and NPs) were provided with procalcitonin tier-based treatment guidelines and national antibiotic prescribing guidelines (see figure). Procalcitonin levels for patients in the procalcitonin arm were reported to clinicians in the ED, and at 6-24 hours, 3 days, 5 days, and 7 days if patient was hospitalized. Procalcitonin levels were drawn but not provided to clinicians in the usual-care group. Clinicians retained autonomy for prescribing antibiotics if deemed appropriate, regardless of procalcitonin level.

Final diagnoses: 39.3% asthma exacerbation, 31.9% COPD exacerbation, 24.2% acute bronchitis, and 19.9% community-acquired pneumonia

Results: There was no statistically significant difference in the antibiotic exposure days between the two groups during the 30-day trial period. There was also no difference in adverse events that could be attributable to withholding antibiotics. For acute bronchitis, the proportion of patients who received an antibiotics in the ED was lower in the procalcitonin group than in the usual-care group (17.3% vs 32.1%). The rate of clinician adherence to the procalcitonin guideline recommendation in the procalcitonin arm was 72.9% in the emergency department and 64.8% for the remainder of the time points.

Discussion Points:

This trial was registered in ClinicalTrials.gov. Remember to look for this when evaluating the quality of RCTs as registered trials must adhere to strict protocols and document study methods in advance of data collection. Beware of trials that do not pre-publish their methods as there is no way to be sure the methods are not modified mid-trial. After a midway analysis demonstrating a larger than expected percent of patients who were lost to follow-up, the ProACT protocol was transparently adjusted to allow for additional participants to achieve the pre-planned power for detection of noninferiority.
Patients were included on the basis of provider suspicion for lower respiratory tract infection, but this was a judgment call by providers, and recruitment was not standardized.
Only a small percentage of the study population (20%) had pneumonia for which procalcitonin-guided treatment may have had the greatest potential effect. There was not enough power for a subgroup analysis, so this is an area for potential further research.
The lack of difference in antibiotic exposure days between the two groups may be because physicians do not have enough confidence in the procalcitonin assay to base their clinical decisions on the test outcomes.
Evidence behind practice guidelines for duration of antibiotic use is extremely limited. Further research in this field may help reduce unnecessary antibiotic exposure days.

Bottom line: The ProACT study may not change your practice, and it raises concerns regarding the utility of procalcitonin among US providers. Further studies are needed to investigate the utility of procalcitonin testing in patients with pneumonia, as well as optimal antibiotic course durations for other common infections.