Pulmonary infection is one of the most common reasons for hospitalization in the US. Overprescription of antibiotics, due either to inappropriate initiation or excessive treatment duration, is an important contributor to antibiotic resistance and complications of therapy.^1.2

Procalcitonin is released from endothelial cells in the setting of bacterial infection, and was approved by the FDA as a blood-based biomarker of lower respiratory tract infection in 2017.³ Procalcitonin levels rise within 6-12 hours of infection and fall by around 50% per day with resolution of infection.^4.5 Furthermore, release of procalcitonin is decreased by interferon gamma, meaning that levels are typically lower in viral infections. ^4,5 These features of procalcitonin release make it an appealing biomarker to guide antibiotic therapy decisions in bacterial infection.

It has been most thoroughly studied in pulmonary infection and a number of clinical algorithms have been proposed to guide interpretation of procalcitonin levels. These recommend various cutoffs to estimate the likelihood of bacterial infection and provide recommendations as to whether antibiotics are indicated or not, in the appropriate clinical setting. However, trials are still ongoing to evaluate the utility and safety of these cutoffs in real world clinical practice with diverse groups of patients. As an example, one such proposed algorithm⁶ is outlined below*:

(Adapted from Sager et al, BMC. 2017.)

*Please review the original publications prior to clinical application. Important caveats: there are no robust data on the use of procalcitonin in patients on immunosuppression. Also, those with renal impairment may show altered kinetics due to delayed excretion of procalcitonin, and diagnostic cutoffs have not been defined.⁷

The Study

However, while several trials have shown decreased antibiotic exposure using a procalcitonin-based approach, there were no data examining changes in mortality and concerns remained about the safety of this approach. Given this landscape, Schuetz et al. carried out a patient-level meta-analysis of 6,708 patients enrolled in 26 separate trials, each examining antibiotic administration based on procalcitonin levels versus control.⁸ Enrolled patients had upper or lower respiratory tract infection, including community-acquired pneumonia, hospital-acquired pneumonia, ventilator-associated pneumonia, exacerbation of COPD, or bronchitis. The primary aim of the meta-analysis was to look for differences in 30-day mortality and treatment failures; secondary outcomes included antibiotic use, duration, and hospitalization data. The trials included patients across a range of clinical settings including primary care, ED, and ICU admissions.

Results

In terms of the primary results, mortality was significantly lower in the procalcitonin group, (9% versus 10% in control group, OR 0.83, 95% CI 0.7-0.99), and with the greatest relative mortality benefit seen in ICU patients (14% decrease in mortality) and in those with community-acquired pneumonia. Mortality could not be assessed in primary care due to the very small number of events, where only one patient died.

Indications

While the incidence of treatment failure was similar between the two groups, the authors postulated that those in the procalcitonin group may have benefited from an earlier recognition of failure to respond, or treatment failure, than those in the control group, allowing their therapy to be tailored more quickly. Furthermore, low procalcitonin levels may have prompted a search for alternative diagnoses that might not have been otherwise entertained in complex, critically ill patients. Prior studies would support this hypothesis, where in patients admitted with heart failure, those who had procalcitonin checked had improved outcomes, felt to be due to more appropriate clinical decision making at the time of admission.⁶

In their meta-analysis, Schuetz et al. also showed that in the procalcitonin group, there was a significant decrease in antibiotic exposure, where fewer patients were prescribed antibiotics, and the duration of therapy was shorter.⁸ Lower rates of initial antibiotic prescriptions were seen in patients treated at primary care (63% versus 23%) and ED (83% versus 69%); and shorter duration of antibiotic courses were seen in ED and ICU patients.⁸ Finally, there was a significant reduction in antibiotic-related side effects in procalcitonin-treated patients (16% versus 22% in control group).

In summary, this meta-analysis showed that procalcitonin-guided antibiotic therapy showed significant benefits, particularly in high-risk patients with respiratory tract infection, without evidence of harm despite lower rates of antibiotic use.⁸ No cost benefit analysis was carried out, and this may be important, particularly in the primary care setting. Also, no immunocompromised patients were included in the study, and steroids for example, can suppress procalcitonin levels.⁷ Therefore, it is unclear if these data would be applicable to such patients. However, despite these considerations, this study suggests that unnecessary antibiotic prescribing could be safely decreased in a broad section of the population using this approach, a prospect of considerable public health benefit.