Procalcitonin in the management of lower respiratory tract infection and sepsis

Introduction

In the USA, over 50% of hospitalized patients receive antimicrobial therapy at some stage of their hospitalization, most of which represents inappropriate use due to an absence of clear symptoms or microbiologic evidence of infection, treatment directed at colonizing or contaminating microorganisms or durations beyond recommended courses. More concerning is that this statistic has not improved, over the past 20 years.^1,2

Acute respiratory infections are the syndrome for which antibiotics are most commonly prescribed, yet the majority of these are caused by viruses.^3,4 During the COVID-19 pandemic, for example, less than 10% of patients presenting with COVID-19 pneumonia had documented bacterial co-infections, yet over 80% received antibiotic treatment.⁵ Suspected sepsis is another common reason for broad-spectrum antibiotic use, but approximately 30%–40% of patients treated empirically for sepsis have non-bacterial causes for their illness.⁶ Given the challenges of inaccurately diagnosing sepsis and the risks of undertreatment, however, most clinicians opt to prescribe empiric antibiotics for patients with possible sepsis at the risk of antimicrobial resistance, Clostridioides difficile infection and other antibiotic-associated toxicities. Subsequent antibiotic de-escalation is often challenging since many patients never have a pathogen identified, and the optimal duration of therapy is often unclear.

Biomarkers of bacterial infection, such as procalcitonin (PCT), provide objective laboratory data to improve clinicians’ decision-making around starting, withholding, or discontinuing antibiotics. PCT has been available in the USA for 14 years and longer in other parts of the world. It has a remarkable feature in that it generally rises in the setting of bacterial infection but not viral infection. Other biomarkers, such as C-reactive protein (CRP), tumour necrosis factor-alpha, interleukin 6 (IL-6) and cluster of differentiation 64 (CD64), all share a common downside, namely, the inability to differentiate viral from bacterial pathogens. The ability for PCT's specificity is the production and release in response to bacterial components through the activation of immune cells by bacterial antigens through pattern-recognition receptors such as Toll-like receptors. On the other hand, there is a negative feedback mechanism in pure viral infections that downregulates the production of PCT through interferon gamma.⁷

PCT levels typically rise quickly, within 3–6 h after the onset of bacterial infection. The time to peak is 24 h, and the period for the elimination half-life is the same. The degree of rise of PCT is also proportional to the intensity of the bacterial stimulus, which explains its value as a prognostic marker.⁷ In addition, the change in PCT levels over time allows insight into the response to antibiotics: a lack of PCT decay suggests a poor response to treatment, whereas reductions of PCT of 80% from its maximal value or an absolute value of less than 0.25–0.5 ng/mL (coupled with clinical improvement) generally indicate that it is safe to discontinue therapy. In line with the FDA-approved use of PCT, we discussed PCT-guided antibiotic management strategies for non-immunocompromised adult patients.

Moderator

Michael Mansour, MD, PhD

Associate Professor of Medicine, Division of Infectious Diseases, Massachusetts General Hospital, Boston, MA, USA

Department of Medicine, Harvard Medical School, Boston, MA, USA

Speakers (in order of appearance)

Mike Broyles, PharmD

Joshua P. Metlay, MD, PhD

Chanu Rhee, MD, MPH

Evangelos J. Giamarellos-Bourboulis, MD

Ramy H. Elshaboury, PharmD

Themes discussed

Challenges and guidelines related to antibiotic stewardship, specifically for lower respiratory tract infections

Acute respiratory infections are a heterogeneous group of conditions and one of the most common reasons people seek urgent care in the outpatient or emergency department. Given the lack of advanced respiratory diagnostics, we face a two-sided problem. On the under-diagnosis side, we know that pneumonia, representing less than 10% of all patients presenting with an acute cough, remains a life-threatening illness requiring prompt antibiotic treatment.⁸ However, the low prevalence of pneumonia and fear of missed treatment leads to overuse of antibiotics for the majority of patients with acute respiratory infections.

The challenge is to parse out those individuals who would benefit from an antibiotic and those who would not. There has been a need for some time to identify a biomarker(s), including white blood cell count, erythrocyte sedimentation rate, CRP and PCT, to assist with understanding how we can distinguish the people who need an antibiotic from those who do not.

The most recent ATS-IDSA guideline for managing adults with community-acquired pneumonia (CAP) included questions related to the role of biomarkers in CAP management. Of note is that the guideline was developed using a formal structure for evaluating available studies to influence recommendations. In addition, the Population, Intervention, Comparator and Outcomes (PICO) framework emphasizes patient outcomes, including mortality, symptom resolution and length of hospitalization, in assessing the evidence in support of different interventions.^9,10 Two questions were asked when using the PICO framework to evaluate the role of biomarkers in pneumonia management. The first involves the role of the biomarker at the time of diagnosis and whether to start antibiotics. The second is if PCT could shorten the duration of antibiotics. The guideline committee determined insufficient data to recommend using PCT to guide decisions regarding antimicrobial drug initiation in patients with pneumonia. However, an important question is whether the combination of PCT plus rapid molecular tests for viral and bacterial pneumonia pathogens could safely guide the initial decision regarding antibiotic treatment. Additional trials are needed in this space for future guideline development.

The guideline recommendation for treating uncomplicated pneumonia is 5 days for antibiotic duration. There is subsequent data that suggests even 3 days may be sufficient.¹¹ If clinical guidelines are followed for the duration of antibiotic therapy, it is less clear that PCT guidance will help shorten the duration of antibiotic use. In settings where prescribers are still routinely giving greater than 7 days of antibiotic therapy for uncomplicated pneumonia, there may be a role for PCT to help guide antibiotic duration.⁹

Table 1 provides a list of previous studies where PCT was used to guide antibiotic therapy for patients with suspected or confirmed lower respiratory tract infection (LRTI), and Table 2 provides a list of previous studies where PCT was used to guide antibiotic discontinuation for patients with suspected or confirmed sepsis.

Table 1.

Previous studies where PCT was used to guide antibiotic therapy for patients with suspected or confirmed LRTI

Publication	Sample size	Country	Endpoints
Huang, 201812	1556	USA	Antibiotic use was about 4 days in both groups (PCT: 4.2 ± 5.8 days; UC: 4.3 ± 5.6 days)
DeSear, 202113	7272	USA	Antibiotic reductions of 1.47 days in PCT group versus usual care group
			Meta-analysis results per FDA public docket FDA-2016-N-2880 from 2015
Long, 200914	172	China	19.2% reduction in relative antibiotic initiation for all patients 38% reduction in overall antibiotic exposure (i.e. total days of antibiotic therapy) for inpatients 51% reduction in overall antibiotic exposure (i.e. total days of antibiotic therapy) for patients who presented to the emergency department and other associated clinics, but were not admitted 2.9-day reduction in antibiotic duration (1.25-day reduction in study-level) 3.6-day reduction in total antibiotic exposure (2.79-day reduction in study-level) No negative effects in regard to mortality, complications or length of stay
Bouadma, 201015	630	France
Burkhardt, 201016	550	Germany
Hochreiter, 200917	110	Germany
Kristoffersen, 200918	223	Denmark
Long, 201119	127	China
Schroeder, 200920	27	Germany
Schuetz, 200921	1381	Switzerland
Briel, 200822	300	Switzerland
Nobre, 200823	79	Switzerland
Stolz, 200724	226	Switzerland
Christ-Crain, 200625	302	Switzerland
Christ-Crain, 200426	243	Switzerland

Table 2.

Previous studies where PCT was used to guide antibiotic discontinuation for patients with suspected or confirmed sepsis

Publication	Sample size	Country	Endpoints
Kyriazopoulou, 202027	266	Greece	PCT-guided early discontinuation of antimicrobials led to reduction in 28-day mortality, early discharge from hospital and lower hospitalization cost
de Jong, 201628	1575	Netherlands	Antibiotic reductions (5 days versus 7 days) and 28-day mortality reductions (19.6% versus 25%) in PCT-guided group versus control group
Dark, 202429	2760	UK	Mean reduction of 0.88 days in the PCT group (95% CI, 0.19–1.58, P = 0.01). For all-cause mortality up to 28 days, the daily PCT-guided protocol was non-inferior to standard care
			Meta-analysis results per FDA public docket FDA-2016-N-2880 from 2015
Bouadma, 201015	630	France	1.5-day reduction in antibiotic duration 3.2-day reduction in total antibiotic exposure 23% reduction in overall antibiotic exposure (i.e. total days of antibiotic therapy) No negative effects in regard to mortality, hospital length of stay or ICU length of stay
Hochreiter, 200917	110	Germany
Stolz, 200930	101	USA, Switzerland
Schroeder, 200920	27	Germany
Nobre, 200823	79	Switzerland

Evidence supporting the use of PCT in Europe

In some European countries, PCT has been in mainstream use in managing pneumonia and sepsis for many years. Several randomized controlled trials (RCTs) show high-quality data suggesting the positive impact of PCT-based antibiotic decision-making. The most robust evidence comes from the ProHOSP trial conducted in Switzerland, which showed that an algorithm for withholding or stopping antibiotics based on a PCT level falling to less than 0.25 ng/mL (or >80% decline from peak) was safe for stopping antibiotic therapy and led to a shorter duration of antibiotics and antibiotic-associated adverse events versus standard care.²¹ Three RCTs conducted by Stolz 2007 and Christ-Crain 2004, 2006 listed in Table 1, are congruent, validating a cut-off of improved, safe outcomes with a PCT < 0.25 ng/mL. If an antibiotic-stopping rule using PCT 0.25 ng/mL is used, there is a significant decrease in the overconsumption of antibiotics.

Similar level evidence exists for patients in the intensive care unit (ICU). Six small-scale RCTs showed that the use of one PCT rule affected a substantial decrease in antibiotic cessation. Two large RCTs, one from the Netherlands (SAPS trial), demonstrate a significant drop in the consumption of antibiotics using a PCT cut-off of 0.5 ng/mL or an 80% decrease from the peak for stopping antibiotics. However, a striking difference is that the SAPS trial also demonstrated a mortality benefit in the PCT arm compared to usual care.²⁸ It is unclear whether this mortality benefit was due to decreased antibiotic exposure, improved diagnostic clarity with PCT use or other reasons. In Greece, another study, PROGRESS, showed a decrease in 28-day mortality despite an environment with high antimicrobial resistance.²⁷ Multidrug-resistant bacterial colonization decreased by shortening treatment with antibiotics using PCT. In turn, fewer antibiotic-associated diarrhoea, acute kidney injury, C. difficile infection and electrolyte disturbances likely result in the 28-day mortality improvement, stressing the crucial nature of early antibiotics cessation. Both the SAPS and PROGRESS trials studied the same rule of PCT, i.e. antibiotics stop when baseline PCT drops by at least 80% or when PCT drops to less than 0.5 ng/mL. Another study from the UK, the ADAPT-sepsis trial, was published at the end of 2024 after our roundtable convened, confirming the effectiveness of a daily PCT measurement to reduce antibiotics in the ICU.²⁹

Evidence supporting the use of PCT in the USA

There is less data available in US-based studies, although several well-designed studies have yielded mixed results. In 2009, the ProVAP study enrolled patients from ICUs of three hospitals (including one US hospital) with ventilator-associated pneumonia. It showed that PCT-guided antibiotic therapy was safe and reduced antibiotic exposure by about 27%.³⁰ A second single-centre study in 2015 examined the utility of PCT algorithms with viral testing in hospitalized patients with non-pneumonia LRTI.³¹ Although the trial was negative overall regarding antibiotic use between the intervention arm and standard care, an interesting finding is that patients with low PCT and positive respiratory viral assay received substantially shorter antibiotic durations, suggesting that clinicians are more likely to respond to the combination of these findings. A third RCT, ProACT, published in 2019, compared patients with LRTI across 14 hospitals in the USA using a similar study structure to ProHOSP.¹² This showed no difference in antibiotic utilization between the PCT-guided care versus the usual care arm. That said, both arms received an average of 4 days of antibiotics, but with multiple caveats.

Given the positive clinical outcomes without safety or mortality concerns in several trials using PCT-guided pneumonia management, it is unclear why the USA-based ProACT, with a similar study as ProHOSP, did not show benefit. One potential reason is that the average duration of antibiotic therapy for suspected LRTIs has decreased over time, particularly in hospitals with good antibiotic stewardship programmes (as in ProACT). As evidence of this, the median duration of antibiotic therapy was just over 4 days in both the PCT and intervention arms in this trial.

It is also possible that, as many new biomarkers are introduced, adoption bias in the healthcare setting may explain US-Europe practice differences and that US providers need more guidance on interpreting and reacting to PCT levels. To answer this question, a US-based, multicentre RCT, ProSAVE, was launched in the USA, and it compares the standard of care to an antimicrobial stewardship PCT-guided arm.¹² The antimicrobial stewardship team at each site interprets the PCT and guides the hospitalist team. The ProSAVE preliminary results show promise, and we hope to conclude the study shortly.³²

Practical implementation challenges and solutions regarding the use of PCT

While data may suggest the safe and impactful use of a biomarker such as PCT, practical implementation remains challenging. Multiple phases of education and guidance creation are necessary to accomplish proper implementation. The first phase is to educate, including efforts to delineate PCT optimal utility, appropriate patient populations and interpretation. Education efforts to clinicians must be ongoing to ensure sustained compliance. The second phase is to provide guidance and electronic decision support systems when ordering the test. This step also requires a large volume of human input. Electronic decision support tools are available to guide clinical care in real time and, in some cases, discourage ordering the test in an inappropriate patient population. An adjunctive approach would be to use flags to identify patients where the test may be helpful and prompt ordering when appropriate. Finally, the last phase is guiding at the time of result interpretation to support antibiotic duration and antibiotic cessation decisions.³³ Additionally, it is important to identify possible discrepancies between clinical decision-making and the PCT result or factors where the false elevation of PCT is unrelated to infections but related to a kinetic difference in unique populations. The panel agreed that any successful stewardship programme will require these three phases.

The commitment to stewardship can be intense, especially if the antimicrobial stewardship clinician alone is to screen, communicate and intervene on critical results. A response to this burdensome process is adopting new tools that integrate AI technology, which will require IT support. When functioning as intended, the broad application of PCT, or other biomarkers, diagnostics, or point-of-care tests, can be successful without requiring the cumbersome human interaction and the human staffing components to care for many patients. This is especially impactful in a large healthcare setting, a fast-paced emergency department or an ICU setting.

Conclusion

The panel identified persistent unmet needs in the area of antibiotic overuse when managing patients with acute respiratory infections. During this roundtable, we discussed the use of biomarkers, such as PCT, to potentially provide some insight in addition to clinical acumen and other diagnostics to understand how to curb antibiotic overuse. RCTs will help provide the guidance, safety and clinical impact the medical community seeks in biomarkers. Another challenge identified is implementation, in which the panel stressed that sustained education is a key requirement to achieve successful adoption and appropriate biomarker use.

Funding

This Virtual Roundtable was supported by an educational grant from BRAHMS GmbH, part of Thermo Fisher Scientific. No funding was provided for the video roundtable.

Transparency declarations

M.K.M. is the recipient of an unrestricted research grant for ProSAVE (Thermo Fisher Scientific). C.R. and M.K.M. report royalties from UpToDate, Inc., for authoring chapters related to procalcitonin use in lower respiratory tract infections. No other authors have declarations to report.