The Role of Procalcitonin in the ED for Antibiotic Management

Approximately 30-50% of inpatient antimicrobial use is considered inappropriate.¹ This is further compounded by the fact that the length of treatment for most infections is not clearly defined. Some consequences of inappropriate antimicrobial use are: drug toxicity, increased drug resistance, super infections, and delayed appropriate treatment. Misuse and overuse of antibiotics is a pressing public health concern affecting not just the US, but the entire world. As a result, multiple key organizations – including the CDC, AHRQ, IDSA, AHA – have promoted the need for antimicrobial stewardship programs. Like many other leading healthcare institutions across the country, SUNY Downstate now has an Antimicrobial Stewardship Program (ASP) as of September 2015 to optimize the use of antibiotics and decrease resistance.

Patients often present with an unclear picture of bacterial infection vs viral infection vs autoinflammatory process vs malignancy. These factors discourage a clear and accurate diagnosis, which can lead to a delay in therapy and worsening of a patient’s progress. Pitfalls of bacterial infection diagnosis include inappropriate timing of cultures, low yield of positive cultures, and unclear source of infection. In a perfect world, the ideal diagnostic agent for bacterial infections would be easy to use, rapidly available, highly sensitive, highly specific, independent of other comorbidities, inexpensive, and reproducible. Enter: procalcitonin.

Below is a table comparing procalcitonin with other commonly used biomarkers/diagnostic tools. ²

Procalcitonin (PCT) is a 116-amino-acid precursor of calcitonin produced by parafollicular cells (C cells) of the thyroid and by neuroendocrine cells (K cells) in the lung and GI tract. Once enzymatically cleaved, PCT is converted to calcitonin then stored in endocrine granules awaiting release. Bacterial toxins from both gram-positive and gram-negative organisms and the ensuing inflammatory host response (i.e. TNF-alpha, interleukins, interferons) stimulate procalcitonin’s release. Dandona illustrates the rapid rise of PCT in normal subjects after injection with bacterial endotoxin over a 24 hour period.⁵ PCT levels start to rise approximately 3-6 hours after infectious challenge and peak after 6-12 hours. Depending on the severity of the infection and initiation of appropriate antibiotics, peak levels may stay elevated past 24 hours.⁶ Furthermore, several studies have shown a strong correlation between mortality and peak PCT levels.^6-7 Similar to troponin levels, this predictable time curve allows for clinicians to use PCT objectively for both diagnosis and prognosis.

Major advantages of procalcitonin are: specificity for bacterial infections, rapid increase in presence of infection and rapid decrease with control of infection, not impaired by neutropenia or immunosuppression, reliability, and predictability. However, every hero has his/her weakness – likewise, procalcitonin has some confounders hindering its use. In settings of massive physiologic stress (e.g. trauma, shock, burns), PCT levels can be variable. For newborns, PCT levels are not reliable until 72 hours after birth. Non-bacterial cytokine activation such as vasculitis, acute graft vs host disease, malaria, and chronic kidney disease may increase PCT at baseline.¹

The most evidence for PCT utilization has been published in patients with lower respiratory tract infections (LRTI’s) and sepsis. Especially in patients with COPD exacerbation and bronchitis, the clinical picture may not be clear whether it is truly an exacerbation or pneumonia. There are multiple studies and meta-analyses proving the benefit of PCT-guided algorithm in reducing antibiotic exposure and costs while having no effect on patient outcomes.

The ProResp trial published by Christ-Crain, et al. was a single-center, randomized, blinded study on patients presenting with lower respiratory tract infections (i.e. CAP, bronchitis, asthma, COPD). Patients were randomized to either standard antibiotic treatment (n=119) or a procalcitonin-guided treatment algorithm (n=124). If PCT < 0.25 mcg/L, antibiotics were discouraged. If PCT ≥ 0.25 mcg/L, antibiotics were encouraged. Re-evaluation of antibiotic use by trending PCT levels was possible after 6-24 hours in both groups. The results of the study showed that there was no difference in mortality or length of stay between both groups. Additionally, 81% of infections were serologically determined to be of viral origin. Compared to the standard treatment group, the PCT-guided group had significantly decreased use of antibiotics (83% vs 44%, p<0.0001). Thus, the authors concluded that withholding antimicrobial treatment minimized patient antimicrobial exposure and did not compromise patient outcomes.⁹

The ProHOSP trial was a multicenter, non-inferiority study that looked at adults with LRTI presenting to the ED and treated according to a PCT algorithm. PCT levels were obtained upon admission and if levels ≥ 0.25 mcg/L, antibiotics were started empirically. If PCT < 0.25 mcg/L, antibiotics were not initiated unless the patient was clinically unstable, needed ICU admission, or had a localized infection. For patients started on antibiotics, PCT levels were redrawn on days 3, 5, 7. Using the same cutoffs, antibiotics were recommended to be discontinued if levels decreased appropriately below the 0.25 mcg/L margin. Compliance to the algorithm was 90.8%. There was an overall reduction in mean antibiotic exposure, reduction in number of antibiotic prescriptions, and no difference in adverse outcomes or mortality. The PCT-algorithm was most impactful in patients presenting with COPD exacerbations and bronchitis.¹⁰

Below is a sample PCT algorithm to guide antibiotic use for lower respiratory tract infections:

PCT can also assist in the diagnosis of sepsis. However, it is important to note that decision to initiate antimicrobial therapy should NOT be delayed based solely on PCT serum levels. PCT should be placed into the clinical context of each patient’s scenario – the site of infection, the likelihood of bacterial infection, severity of illness, and any other pertinent clinical data.

Finally, the PRORATA trial was a multicenter, prospective, randomized, open-label trial looking specifically at septic patients admitted to the ICU not yet started on antibiotics. Patients were randomized to a PCT-guided algorithm (n=307) similar to ProResp and ProHOSP trials; antibiotics were started or stopped based on the predefined cutoff ranges of 0.5mcg/L. (**Notice the range is different from LRTI algorithm.**) The control group (n=314) received antibiotics as usual and duration was at the discretion of the physician. Although adherence to the protocol was lower than the LRTI trials, there was no difference in mortality and ICU length of stay between the two groups. However, once again, there was a significant decrease in antimicrobial use. Any deaths that did occur were attributed to the patient’s underlying disease and not infection.¹¹ Below is the sepsis algorithm that was used:

Procalcitonin is an in-house lab test! Cost is about $20. Blood should be sent in a red top tube and sent to the hematology lab. Turnaround time is approximately 24 hours, which is why it is perfect to order in the ED before the patient is admitted. By the time PCT levels have returned, the primary team can assess the patient’s status and use PCT level to continue or discontinue antibiotics.