Procalcitonin guidance improves antibiotic stewardship

The case

A 72-year-old male with COPD presents to the emergency department with increased dyspnea and cough. He is afebrile, and oxygen saturation is 87% on room air. WBC count is 9.5 with a normal differential, and chest x-ray is read by the radiologist as atelectasis versus early consolidation in the left lower lobe. Should antibiotics be initiated?

Background

The problem: Antibiotic overuse

With the increasing prevalence of antibiotic resistance in our nation’s hospitals, the need for robust antibiotic stewardship programs has continued to rise in importance. In 2016, the CDC reported a fatal case of septic shock due to a carbapenem-resistant strain of Klebsiella resistant to all tested antibiotics.¹ This case received much media coverage; moreover, this patient represented only one of the approximately 23,000 patients infected with antibiotic-resistant bacteria in the United States who die each year. Although various approaches to curbing antibiotic resistance are being pursued, judicious antibiotic use is central to success. Current evidence suggests that up to 30% of antibiotics are not optimally prescribed,² leaving a significant opportunity for improvement.

Lower respiratory infections account for a substantial proportion of antibiotic utilization in the United States. In a recent study, acute respiratory conditions generated 221 antibiotic prescriptions per 1,000 population, but only half of these were deemed appropriate.² The inability to reliably discern viral from bacterial etiology is a driver of excess antibiotic use.

^{Dr Graham Beards/en.wikipedia/CC BY-SA 4.0}

Antibiotic resistance tests; the bacteria in the culture on the left are sensitive to the antibiotics contained in the white paper discs. The bacteria on the right are resistant to most of the antibiotics.

The procalcitonin assay has been touted as a possible solution to this problem. Multiple studies have evaluated its utility as a tool to help discriminate between bacterial infection and viral or noninfectious etiologies.

What is procalcitonin?

Thyroidal c-cells convert the prohormone procalcitonin to calcitonin, which is stored in secretory granules for release in response to fluctuations in calcium levels via a classical neuroendocrine feedback loop. Alternatively, procalcitonin can be synthesized in nonthyroidal parenchymal cells, and high levels of proinflammatory mediators secreted in response to bacterial endotoxin drive increased procalcitonin production. Interestingly, interferon gamma, up-regulated in viral infections, reduces procalcitonin production. Nonthyroidal parenchymal cells lack mechanisms for efficient conversion of procalcitonin to calcitonin and do not contain secretory granules to facilitate its regulated release. Hence bacterial infections correlate with higher serum procalcitonin levels.³

Evidence

Can procalcitonin guide antibiotic therapy in patients with acute respiratory illness while reducing antibiotic utilization?

The ability of procalcitonin to selectively identify bacterial infection makes it a potentially promising tool to advance the antibiotic stewardship agenda. Multiple randomized controlled trials have explored the use of procalcitonin-guided antibiotic therapy for treatment of lower respiratory tract infections such as acute bronchitis, exacerbations of COPD, and pneumonia. Each study discussed below was done in Switzerland, involved the same key investigator (Mirjam Christ-Crain, MD, PhD), and shared a similar design in which a threshold for low procalcitonin values (less than 0.1 mcg/L) and high procalcitonin values (greater than 0.25 mcg/L) was prespecified. Antibiotic therapy was strongly discouraged for patients with low procalcitonin and encouraged for those with high procalcitonin; antibiotics were not recommended for patients with intermediate values, but the treating physician was allowed ultimate discretion (Figure 1). All studies compared a procalcitonin-guided treatment group to a standard care group, in which antibiotics were prescribed by the treating physician based on established clinical guidelines.

Figure 1. Procalcitonin treatment algorithm

Figure 1. Procalcitonin treatment algorithm

In a study focusing on outpatients presenting to their primary care physicians with acute respiratory tract infection, 53 primary care physicians in Switzerland recruited 458 patients. There was no significant difference in time to symptom resolution, as determined by patient report during an interview 14 days after initial presentation; however, 97% of patients in the standard-care group received antibiotics, compared with 25% in the procalcitonin-guided group. Equal numbers of patients (30% in each group) reported persistent symptoms at 28-day follow-up. Among the cohort of patients with upper respiratory infections or acute bronchitis, procalcitonin guidance reduced antibiotic prescriptions by 80%.⁴

Dr. Bryan Huang

In a blinded, single-center, randomized, controlled trial of 226 patients presenting to a university hospital with a COPD exacerbation severe enough to require a change in the baseline medication regimen, procalcitonin-guided therapy allowed for an absolute reduction of antibiotic use by 32% without an impact on outcomes. Rates of clinical improvement, ICU utilization, recurrent exacerbations, hospital length of stay, and mortality did not differ between the groups.⁵

Dr. Greg Seymann

Another study by Dr. Christ-Crain looked at whether procalcitonin could be used to determine duration of antibiotic therapy in hospitalized pneumonia patients. In a similarly designed randomized, nonblinded trial with two arms, a procalcitonin group (n = 151) and a standard care group (n = 151), procalcitonin levels were checked at 4, 6, and 8 days, with similar cutoffs regarding levels for which antibiotics were encouraged or discouraged. Antibiotic treatment duration was reduced from a median of 12 days in the standard-care group to 5 days in the procalcitonin group (P less than .001). There was no difference in the success rate of treatment, readmission rate, or death rate between the two groups.⁶These initial studies were limited by their relatively small size and narrow scope. The ProHOSP study was the first large, multicenter study to address the utility of procalcitonin-guided antibiotic therapy. Design was similar, although the decision to treat with antibiotics was more rigorously controlled by the centralized study personnel. The study enrolled 1,359 patients in the emergency departments of six Swiss tertiary-care hospitals. Most patients had pneumonia, but 17% had COPD exacerbation and 11% had acute bronchitis. No difference in death, ICU admission, readmission, or disease-specific complications was noted. Antibiotic exposure was reduced by 34.8% in the procalcitonin group, with a mean of 8.7 days versus 5.7 days on antibiotics.⁷In combination, the studies above support the use of procalcitonin to guide decisions about antibiotic use in patients with lower respiratory tract infections; antibiotic use can be significantly reduced without adverse outcomes. Further, sequential monitoring of procalcitonin levels may help guide duration of antibiotic therapy. These studies all had fairly high rates of follow-up and the ProHOSP study, in particular, had a large, representative sample and a rigorous methodology to standardize antibiotic prescription in the control group.

Limitations include the possible impact of the Hawthorne effect, as physicians knew their antibiotic usage patterns were being monitored, which may impact generalizability of the findings to a real-world setting. Similarly, it is difficult to control for a spillover effect as providers exposed to the procalcitonin-guided algorithm became more comfortable with a restrictive prescribing approach. The costs of the additional procalcitonin assay must be weighed against the benefits. Incidence and cost of other adverse effects of antibiotic use (rates of Clostridium difficile, renal insufficiency, urticarial drug eruptions, etc.) were not addressed. The rapid assay currently has limited availability in the United States, though that is changing. Finally, recent additional studies (unrelated to procalcitonin) have suggested shorter antibiotic treatment durations for patients with pneumonia.⁸