Let’s face it, we have an unhealthy relationship with antibiotics. We are all aware that we are currently in an antibiotic resistance crisis that is driven by overuse. And despite our enlightenment we cannot seem to stop prescribing antibiotics for complaints that have been clearly shown not to require antibiotics. And so, given our appalling unstructured attempts at antibiotic stewardship, it is not hard to imagine that some guidance, practically any form of guidance, would reduce the rates of antibiotic use in patients who clearly would not benefit from antibiotics.
Enter everyone’s new favorite biomarker procalcitonin, the white blood cell count for the millennials. Procalcitonin is a peptide precursor of the hormone calcitonin, that is involved in calcium homeostasis. The level of procalcitonin in the serum is typically undetectable and only rises to detectable levels during inflammatory states. What differentiates procalcitonin from other inflammatory markers like, WBC, CRP, ESR, D-dimer, is that its rise is more precipitous with bacterial infections. Because of this unique property, it has been offered up as an effective method of identifying patients presenting to the Emergency Department with a bacterial infection.
Huang et al conducted a 14 center RCT enrolling patients 18 years or older with the ED diagnosis of lower-respiratory tract infection, in whom the treating physician had not yet decided to give antibiotics (1). The authors excluded patients in whom the treating physicians were unlikely to withhold antibiotics (prior antibiotics, vasopressor use, mechanical ventilation via endotracheal tube, known severe immunosuppression, accompanying non-respiratory infection, known lung abscess/empyema), or in cases where procalcitonin could be elevated without bacterial infection (chronic dialysis, metastatic cancer, surgery in the past seven days).
Physicians of patients whom were randomized to the procalcitonin group were given the results of the procalcitonin assay as well as a standardized protocol to aid in antibiotic decisions. And while physicians were given the clinical autonomy to decide whether to prescribe antibiotics, research assistants recorded the reasons they chose to contradict the procalcitonin pathway. 1656 patients were included in the final analysis. The final diagnosis in descending order included asthma exacerbation (39.3%), COPD exacerbation (31.9%), acute bronchitis (24.2%), and community-acquired pneumonia (19.9%). 47.2% of the patients were hospitalized, 36.5% received antibiotics in the Emergency Department, and 59.4% received antibiotics within 30 days of their ED visit.
The authors found no difference in their primary outcome, the difference in antibiotic exposure within 30-days of ED presentation (4.2 and 4.3 days in the procalcitonin and control groups, respectively). There was also no difference in any of the secondary endpoints, the percentage of patients receiving any antibiotics within 30 days (57.0% vs 61.8%), the percentage of patients receiving an antibiotic in ED (34.1% vs 38.7), or the mean hospital antibiotic-days among hospitalized patients (2.6 vs 2.7 days). The authors did note that in the subset of patients with the final diagnosis of acute bronchitis, the rate of antibiotic use was significantly lower in the procalcitonin group than in the controls (17.3% vs. 32.1). No difference was observed in the rates of adverse events between the procalcitonin and control groups.
And so, the question becomes is procalcitonin a test with mediocre test characteristics that adds nothing to clinical value as all the patients it identifies as needing or not needing antibiotics have clinically obvious presentations. Or is it the fact that as physicians we have a general dislike for being told what to do, and in this case the clinicians disregarded the procalcitonin treatment guideline and administered antibiotics to whomever they pleased? Overall physicians adhered to the procalcitonin guidelines in 72.9% of the cases. Adherence varied dependent on the final diagnosis of the patient in question, asthma exacerbation, 64.2%; acute exacerbation of COPD, 49.2%; acute bronchitis, 82.4%; and community acquired pneumonia, 39.4%. The most common reasons for prescribing antibiotics contrary to the procalcitonin guidelines recommendations were that the clinician believed that a bacterial infection was present despite a low procalcitonin level, or the clinician believed that the patient had an acute COPD exacerbation that would benefit from antibiotics.
That being said if you examine the subset of patients in whom the clinicians were compliant with the procalcitonin guided therapy recommendations, the results were consistent with those of the entire cohort. The authors found no difference in the number of patients who received antibiotics (antibiotic days 4.2 vs 4.3).
Given the noncompliance observed in this cohort, a definitive interpretation of procalcitonin’s role in the Emergency Department is unclear. Despite this, I find it somewhat telling that procalcitonin failed to outperform the straw man that is our current culture of extreme over-prescribing. Especially given the fact that even in the 72.9% of the patients in which the clinicians followed the guidelines, the authors were unable to find a benefit with the use of procalcitonin. This speaks to a bigger question, why do we require a mediocre biomarker that asks the wrong clinical question to do the right thing? After all, the question we should be asking is not, does this patients have a bacterial infection? But rather, will this patient benefit from antibiotics? Countless studies have demonstrated that most forms of lower respiratory infections presenting to the ED do not require antibiotics independent of the infectious agent. A far more logical approach would be to simply not prescribe antibiotics to patients with lower respiratory tract infections without a strong clinical indication. Essentially only for patients with clinical or radiological signs of pneumonia or a clinically ill appearing patient with a COPD exacerbation, should antibiotics be given. We have known this microbiological fact for quite some time, and what is required is not a new means of detecting an infection but rather a shift in the mentality of both physicians and patients. Not every cough and cold requires a course of antibiotics. In fact, most do not. But this is far too complex a conversation, so instead we offer a biomarker and its meaningless results as a inferior surrogate.
- Huang DT, Yealy DM, Filbin MR, et al. Procalcitonin-Guided Use of Antibiotics for Lower Respiratory Tract Infection. N Engl J Med. 2018;