Procalcitonin use in lower respiratory tract infections

INTRODUCTION — Procalcitonin is a serum biomarker that helps distinguish bacterial infection from other causes of infection or inflammation. In patients with lower respiratory tract infections, procalcitonin can serve as a helpful adjunct to clinical judgment for guiding antibiotic therapy and resolving diagnostic uncertainty. The study of procalcitonin is evolving, and the approach to procalcitonin use varies among institutions and experts.

This topic reviews the role of procalcitonin in the evaluation and management of adults with lower respiratory tract infections, including pneumonia, acute bronchitis, and acute exacerbations of chronic obstructive pulmonary disease.

The general evaluation and management of lower respiratory tract infections are reviewed separately.

●(See "Clinical evaluation and diagnostic testing for community-acquired pneumonia in adults".)

●(See "Treatment of community-acquired pneumonia in adults who require hospitalization".)

●(See "Acute bronchitis in adults".)

●(See "Evaluation for infection in exacerbations of chronic obstructive pulmonary disease".)

●(See "COPD exacerbations: Management".)

●(See "Treatment of community-acquired pneumonia in adults in the outpatient setting", section on 'General approach'.)

RATIONALE FOR USE — Reducing antibiotic use for the treatment of respiratory tract infections is a global health care priority [1-6]. Lower respiratory tract infections (LRTIs) are among the most common reasons for antibiotic prescription [7]. An estimated 30 to 85 percent of these prescriptions are unnecessary or inappropriate [8-16]. Even when indicated, antibiotic treatment courses often exceed recommended durations [16].

Antibiotic overuse for LRTIs is in part due to the difficulty in distinguishing between viral and bacterial infections. A substantial fraction of LRTIs are viral [17-23] and do not require treatment with antibiotics. However, clinical signs and symptoms of bacterial and viral LRTIs are similar and often cannot be distinguished based on clinical features alone. Microbiologic testing can be helpful, but results from culture or other assays often take days to obtain and, in many cases, a pathogen is not identified. (See "Clinical evaluation and diagnostic testing for community-acquired pneumonia in adults", section on 'Microbiologic testing'.)

Procalcitonin has good discriminatory value for distinguishing between viral and bacterial infections, and results can be obtained in hours or less. In patients with community-acquired pneumonia, procalcitonin is about 65 to 70 percent accurate in distinguishing bacterial from viral pathogens [24]. When used as part of an algorithm in combination with clinical judgment in patients with LRTIs, procalcitonin has been shown in some studies to reduce unnecessary antibiotic use by about 25 to 50 percent without increasing morbidity or mortality [25-29].

PROCALCITONIN BIOLOGY

Synthesis — Procalcitonin synthesis pathways vary in different inflammatory states. In the absence of systemic inflammation, procalcitonin synthesis is restricted to thyroid neuroendocrine cells, and the protein is not released into the blood until it is cleaved into its mature form, calcitonin [30-33]. Thus, serum procalcitonin is typically undetectable in healthy persons when standard assays are used [32].

When systemic inflammation is caused by bacterial infection, procalcitonin synthesis is induced in nearly all tissues and released into the blood. Known triggers for synthesis include bacterial toxins, such as endotoxin [34], and cytokines including tumor necrosis factor (TNF)-alpha, interleukin 1-beta, and interleukin 6 [30,35,36]. In contrast, procalcitonin synthesis is not induced in most viral infections [30-33]. The lack of induction is likely due to cytokines released in viral infections that inhibit TNF-alpha production, such as interferon-gamma [31,33,35-39]. Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) may be an exception, though the degree of procalcitonin rise with SARS-CoV-2 infection is variable and likely correlates with severity [40]. Less commonly, other respiratory viruses may be associated with a rise in procalcitonin; this appears to correlate with severity [24,41].

Not all bacterial infections cause procalcitonin to rise, or rise to the same degree. Typical bacteria, such as Streptococcus pneumoniae or Haemophilus influenzae, tend to cause greater rises in procalcitonin than atypical bacteria [24,42,43]. Certain fungi, such as Pneumocystis jirovecii [44,45] and Candida species [46-48], and parasites, such as malaria [49,50], have also been reported to cause elevations in procalcitonin.

Noninfectious causes of systemic inflammation, such as shock, trauma, surgery, burn injury, and chronic kidney disease can also induce procalcitonin production but are less closely correlated with procalcitonin induction than bacterial infection [39,51]. A number of other causes of elevated procalcitonin levels have been reported.

Kinetics — Serum procalcitonin levels rise within two to four hours of an inflammatory stimulus, typically peaking within 24 to 48 hours [39,52]. Peak levels roughly correlate with the severity of infection, with higher levels observed in patients with septic shock and sepsis than with uncomplicated pneumonia or other localized infections [52-54].

With resolution of inflammation, procalcitonin levels quickly decline at a predictable rate. After reaching peak, levels decline by about 50 percent every 1 to 1.5 days [52]. When the inflammatory stimulus is ongoing, procalcitonin production continues and levels plateau [52]. These kinetics are altered in patients with renal dysfunction. (See 'Chronic kidney disease' below.)

CLINICAL USE — Procalcitonin can serve as a helpful adjunct to clinical judgment for guiding antibiotic therapy and resolving diagnostic uncertainty in patients with known or suspected lower respiratory tract infections (LRTIs). However, the study of procalcitonin is evolving, the approach to procalcitonin use varies among experts, and the assay is not available at all institutions.

Procalcitonin's greatest utility may be for guiding early antibiotic discontinuation in patients with community-acquired pneumonia (CAP). Discontinuation of antibiotics based on defined procalcitonin thresholds has been shown in some studies to reduce antibiotic use without adverse outcomes (algorithm 1) [27,28]. In most other circumstances, however, we interpret procalcitonin levels qualitatively (eg, as high, low, rising, or falling) (table 1), weighting its value similarly to other clinical findings.

In all circumstances, clinicians should also be aware of the limitations of the assay, including noninfectious factors that cause procalcitonin to rise or fall, how procalcitonin levels vary among pathogens, and patient populations in which procalcitonin has not been well studied (see 'Limitations' below). All decisions to stop antibiotics should be made in combination with clinical judgment, noting that certain infections require prolonged duration of antibiotics regardless of procalcitonin levels (eg, CAP complicated by Staphylococcus aureus bacteremia).

Guiding antibiotic therapy

Community-acquired pneumonia in hospitalized patients — We measure procalcitonin to help determine when to discontinue antibiotic therapy in immunocompetent patients who are hospitalized with known or suspected CAP. We generally obtain a procalcitonin level at the time of diagnosis and repeat the level every one to two days, depending on severity of illness. We determine the need for continued antibiotic therapy based on the patient's clinical course, suspected etiology, serial procalcitonin levels, and microbiologic tests results.

We do not typically use procalcitonin to inform antibiotic initiation in patients with known or suspected CAP.

●In critically ill patients, empiric antibiotic therapy should never be delayed.

●Although most clinically stable patients with suspected CAP should also receive empiric antibiotic therapy, some experts withhold antibiotics in selected low-risk patients with very low procalcitonin levels when a viral cause is suspected and close follow-up is arranged.

Our approach is based on a large body of evidence that supports the use of procalcitonin for guiding antibiotic decision making [27-29,55-59]. In a patient-level meta-analysis comparing a procalcitonin algorithm with standard care in over 6000 patients with LRTIs of any kind, there was a substantial reduction in antibiotic use without increasing adverse outcomes in the procalcitonin group [29,59]. In an analysis of a subset of 2910 patients with CAP, the procalcitonin algorithm reduced antibiotic exposure by 2.9 days when compared with standard care (7.5 versus 10.4 days, 95% CI -2.02 to -2.87). There were no significant differences in mortality, treatment failure, length of stay, or antibiotic-associated side effects. Consistent results were found across treatment settings (eg, primary care, emergency department, and intensive care unit [ICU]) and in trials that had high (>70 percent) or low adherence to the procalcitonin algorithm. When procalcitonin algorithms are used, the greatest reduction in antibiotic use comes from earlier antibiotic discontinuation rather than curtailing antibiotic initiation [29,59].

Our approach is similar to the American Thoracic Society (ATS) and Infectious Diseases Society of America (IDSA) guidelines, which do not recommend procalcitonin to guide initiation or withholding of antibiotic therapy for patients with CAP. The guidelines suggest that serial procalcitonin measurement may be most useful in settings where the average length of stay for patients with CAP exceeds five to seven days [60]. The guidelines also suggest that a low procalcitonin level can help guide early antibiotic discontinuation in a patient who has achieved early clinical stability, has a positive test for influenza, and no evidence of bacterial infection. In our experience, this also applies to other respiratory viruses.

Clinically stable patients

●Antibiotic discontinuation – We use the following procalcitonin thresholds, along with clinical judgment, to help guide antibiotic discontinuation in clinically stable patients:

•For patients with procalcitonin levels that are persistently <0.25 ng/mL in whom the initial diagnosis of CAP was uncertain and an alternative diagnosis has been made (eg, congestive heart failure), we generally discontinue antibiotics.

•For patients with procalcitonin levels <0.25 ng/mL with probable viral CAP or a noninfectious syndrome (based on history or test results), we generally discontinue antibiotics. Low procalcitonin levels suggest that concurrent bacterial infection is improbable.

•For patients with known or suspected bacterial CAP who have received at least five days of appropriate antibiotic therapy, we consider discontinuing antibiotics for patients who have improved clinically and who have a procalcitonin level <0.25 ng/mL (or a ≥80 percent drop from peak level). However, reaching a level of <0.25 ng/mL is not a requirement for antibiotic discontinuation; clinical judgment alone is adequate in patients with clinically resolved pneumonia.

•For patients with known or suspected bacterial CAP with procalcitonin levels that are declining but still ≥0.25 ng/mL, we generally continue antibiotic therapy. Declining procalcitonin levels suggest response to antibiotic therapy. However, as noted above, reaching a level of <0.25 ng/mL (or a ≥80 percent drop from peak level) is not a requirement for antibiotic discontinuation; clinical judgment alone is adequate in patients with clinically resolved pneumonia.

•For patients with rising levels, or levels that fail to decline with antibiotic therapy, we consider other causes of elevated procalcitonin levels and assess whether to continue or change antibiotic regimens based on the individual patient characteristics (see 'Limitations' below). In patients who are not clinically improving and have known or suspected bacterial CAP, persistently high or rising procalcitonin levels suggest poorer prognosis or uncontrolled infection.

Optimal thresholds for discontinuing antibiotics have not been precisely determined. Some experts use a lower threshold, typically 0.1 ng/mL, when deciding to discontinue antibiotics. Others take into account the rate of decline and discontinue antibiotics when the procalcitonin level has decreased by ≥80 percent from its peak. The latter approach is most often used when initial values are >5 ng/mL and the time period it would take to decline to <0.25 ng/mL would exceed a reasonable duration of antibiotics [61].

The procalcitonin thresholds evaluated in most trials (and outlined above) are derived from the procalcitonin-guided antibiotic therapy and hospitalization in patients with lower respiratory tract infections (ProHOSP) study [28]. In this trial, 1359 adults with an LRTI of any kind were randomized to procalcitonin-guided antibiotic management or standard care. Clinicians were allowed to overrule the procalcitonin algorithm and start antibiotics in patients with respiratory or hemodynamic instability, with pneumonia due to Legionella pneumophila, or with other risk factors for poor outcomes regardless of the procalcitonin level. Among the 925 patients with CAP, mean antibiotic exposure was reduced by 30 percent (95% CI -37.6 to -26.9) with no difference in adverse events. Clinician adherence to the algorithm was about 90 percent.

In contrast, the more recent United States-based ProACT Trial, which enrolled patients presenting to the emergency department with any suspected lower respiratory tract infection, including CAP, and used the same algorithm as ProHOSP, did not show reduction in antibiotic utilization [62]. This may be a result of increased provider knowledge about appropriate antibiotic prescribing and shorter baseline treatment courses in hospitals with antibiotic stewardship programs and high adherence to pneumonia quality measures. (See 'Practical considerations' below.)

●Antibiotic initiation – Although most procalcitonin algorithms provide an option for withholding antibiotics at the time of diagnosis, using procalcitonin to determine whether to start antibiotics is more controversial.

•For most patients with known or suspected CAP, we treat with empiric antibiotics regardless of the initial procalcitonin level because of the high morbidity associated with CAP and the imperfect diagnostic accuracy of procalcitonin.

•For selected clinically stable patients without comorbidities, some experts consider withholding antibiotics when the clinical presentation and/or laboratory results strongly suggest a viral infection (or if suspicion of bacterial infection is otherwise low) and the procalcitonin level is <0.25. The decision to withhold antibiotics should also take into account the patient's severity of illness and the degree of suspicion for bacterial CAP [61].

If antibiotics are withheld and the patient is hospitalized, a repeat procalcitonin can be checked within 6 to 24 hours to ensure that the initial procalcitonin was not drawn too early in the course of the patient's illness.

In patients with CAP, procalcitonin is about 65 to 70 percent accurate in distinguishing bacterial from viral pathogens [24,63]. In addition, the optimal threshold for diagnosing bacterial pneumonia has not been determined [64-66]. Because the assay's predictive value is augmented when used in combination with clinical judgment and other test results, such as viral respiratory panels [67], we withhold antibiotics only if other factors strongly suggest a nonbacterial etiology.

●Prognosis – Procalcitonin also has prognostic value in patients with CAP [53,68-72]. In a patient-level meta-analysis of 14 trials and 4211 patients with respiratory tract infection, initially elevated procalcitonin levels were associated with an increased risk of treatment failure (odds ratio [OR] 1.66, 95% CI 1.44-1.90) and mortality (OR 1.69, 95% CI 1.41-2.04) in patients with CAP. However, the optimal threshold for predicting adverse events has not been determined. Serial procalcitonin levels may be more predictive than single values. Several cohort studies suggest that rising levels predict mortality and other adverse outcomes, either independently [69] or in combination with disease severity scores [70,71]. Thus, when procalcitonin levels fail to decline with antibiotic therapy, we generally reevaluate our diagnostic and therapeutic approach [53,63-65,68-74].

Additional detail on the evaluation and management of CAP is provided separately. (See "Clinical evaluation and diagnostic testing for community-acquired pneumonia in adults" and "Treatment of community-acquired pneumonia in adults in the outpatient setting" and "Treatment of community-acquired pneumonia in adults who require hospitalization".)

Critically ill patients — We also use procalcitonin when considering early antibiotic discontinuation in patients with CAP who were septic or critically ill at the time of diagnosis. The thresholds that we use to guide antibiotic discontinuation differ in critically ill than stable patients. We consider stopping antibiotics when procalcitonin levels fall below 0.5 ng/mL (or decrease by ≥80 percent from peak if initial value >5 ng/mL). The procalcitonin threshold used to stop antibiotics in most trials evaluating the critically ill is higher (0.5 ng/mL) than that used for stable patients with respiratory tract infections (0.25 ng/mL), presumably because baseline levels for critically ill patients are expected to be abnormal. Other experts, including some UpToDate authors, use a lower threshold, typically 0.25 ng/mL, for antibiotic discontinuation [75].

We do not use procalcitonin to inform antibiotic initiation in this population. In any patient with known or suspected CAP who is septic or critically ill, antibiotics should not be delayed regardless of the initial procalcitonin level.

The recommendations are based on randomized trials evaluating procalcitonin in the critically ill [57,76-81] and on the body of evidence that supports the use of procalcitonin for guiding antibiotic use in patients with LRTIs in general [28,29]. As an example, in a patient-level meta-analysis evaluating over 2400 patients admitted to the intensive care unit with LRTI of any kind, procalcitonin-guided antibiotic decision making was associated with 1.23 day reduction in antibiotic duration (8.8 versus 9.5 days, 95% CI 0.82 to -0.65) when compared with placebo [28,29]. There were no significant differences in mortality, treatment failure, length of stay, or antibiotic-associated side effects. Validation of these findings in the broader clinical settings outside of randomized trials is limited.

The algorithm we use to manage patients who are critically ill is derived from the Stop Antibiotics on Procalcitonin guidance Study (SAPS) trial, which evaluated 1575 patients admitted to the ICU with suspected or known infection (65 percent had a respiratory tract infection) [76]. Procalcitonin levels were checked daily, and clinicians were advised to stop antibiotics when levels were ≤0.5 ng/mL or if the level decreased by ≥80 percent from peak. Compared with controls, the procalcitonin group had significantly lower median antibiotic exposure (7.5 versus 9.3 defined daily doses) and lower 28-day mortality (19.6 versus 25 percent). Clinician adherence to the algorithm within 48 hours was 53 percent; the main reason for nonadherence was concern about stopping antibiotics in patients who were not yet clinically stable. Identical thresholds of ≤0.5 ng/mL or ≥80 percent reduction from peak on day 5 were used in the Procalcitonin-Guided Antimicrobial Therapy to Reduce Long-Term Sequelae of Infections (PROGRESS) trial, which evaluated 266 patients with sepsis (over 60 percent from pneumonia). Compared with usual care, procalcitonin-guided antibiotic discontinuation had lower antibiotic exposure (median 5 versus 10 days) and lower rates of infection-related adverse events at 180 days and 28-day mortality [82].  

Our recommendations are largely consistent with the 2021 Surviving Sepsis Campaign and the 2016 IDSA antimicrobial stewardship guidelines, which both give a weak recommendation for using serial procalcitonin levels to guide antibiotic discontinuation in patients with suspected infections in the ICU [83,84].

Additional detail on the evaluation and management of CAP in patients who are critically ill is provided separately. (See "Treatment of community-acquired pneumonia in adults who require hospitalization" and "Evaluation and management of suspected sepsis and septic shock in adults".)

Other lower respiratory tract infections

Community-acquired pneumonia in outpatients — The approach to using procalcitonin levels to guide antibiotic-decision making for outpatients with CAP varies among experts. When the test is available and turnaround time is quick, some experts use procalcitonin in conjunction with a positive test for a viral pathogen (eg, influenza) to withhold antibiotics or stop them early, provided that there are no other signs of bacterial infection (eg, dense consolidation on chest imaging, markedly high white blood cell count, or positive microbiologic test) and that patient has close follow-up.

Other experts prefer to treat empirically with antibiotics because bacterial CAP is hard to exclude definitively, the morbidity association with CAP is high, and outpatient treatment courses for CAP are short (eg, five days).

While data directly supporting procalcitonin-guided antibiotic decision-making in outpatients with CAP are limited, one randomized trial comparing >150 outpatients with CAP found that clinical outcomes were similar and antibiotic exposure was decreased when procalcitonin was used [56]. This approach is further supported by a larger trial evaluating >450 patients with any LRTI (including CAP, though the proportion of patient with CAP is not known); clinical outcomes were similar and antibiotic exposure was reduced when procalcitonin values were used to guide antibiotic decision-making [85].

Ventilator-associated pneumonia — The utility of procalcitonin for guiding antibiotic therapy in patients with ventilator-associated pneumonia (VAP) is less certain than for CAP. Because patients with VAP are often critically ill, we typically start antibiotics when the diagnosis is suspected, regardless of the procalcitonin level. This recommendation is derived from observational data that suggest that procalcitonin has poor predictive value for the diagnosis of VAP [86,87].

For patients with known or suspected VAP, we use procalcitonin to help determine when to discontinue antibiotic therapy. The optimal threshold for guiding antibiotic discontinuation in patients with VAP has not been determined. When interpreting results, we generally consider procalcitonin levels qualitatively (eg, as high, low, rising, or falling) and in conjunction with other clinical parameters (table 1) [86-89].

A single randomized trial (ProVAP) directly evaluated use of procalcitonin algorithms versus standard care in 101 patients with known or suspected VAP. In the procalcitonin group, stopping antibiotics when the procalcitonin level was <0.5 ng/mL or had decreased by ≥80 percent from peak resulted in a significant 27 percent reduction in antibiotic use (median 10 versus 15 days) without increasing adverse outcomes. While similar findings have been observed in subgroup analyses of patients with VAP in larger randomized trials evaluating procalcitonin use in critically ill patients, less than 400 patients have been studied overall [29,57,59,77,78,90].

Our approach is consistent with the 2016 IDSA and ATS guidelines, which strongly recommend using clinical criteria alone, rather than procalcitonin plus clinical criteria, to decide whether to start antibiotics [91]. These guidelines also make a weak recommendation for using procalcitonin in addition to clinical criteria when discontinuing antibiotics. The 2017 combined European and Latin American (European Respiratory Society, European Society of Intensive Care Medicine, European Society of Clinical Microbiology and Infectious Diseases, and Asociación Latinoamericana del Tórax) guidelines do not recommend procalcitonin use in patients with VAP when the anticipated antibiotic course is 7 to 8 days but do recommend use in selected cases when care needs to be individualized (eg, patients on toxic or second-line agents such as colistin or those infected with highly drug resistant pathogens) [92].

Additional detail on the evaluation and management of hospital-acquired and ventilator-associated pneumonia is provided separately. (See "Clinical presentation and diagnostic evaluation of ventilator-associated pneumonia" and "Treatment of hospital-acquired and ventilator-associated pneumonia in adults".)

Acute exacerbations of chronic obstructive pulmonary disease — Use of procalcitonin to help guide antibiotic therapy in patients with acute exacerbation of chronic obstructive pulmonary disease (AECOPD) is controversial. The clinical utility and safety of using procalcitonin for this patient population is not firmly established. However, a substantial fraction of acute COPD exacerbations are caused by viruses, and some experts use procalcitonin to help guide antibiotic discontinuation in patients with nonsevere AECOPD, particularly at institutions where comprehensive molecular respiratory viral panels are available.

While procalcitonin use in patients hospitalized with AECOPD has been shown to reduce antibiotic exposure without increasing adverse events in several trials [29,93-98], this benefit has not been consistently demonstrated in all trials and treatment settings [99]. In addition, the studied population is small overall and most trials have methodologic limitations. In one trial comparing procalcitonin algorithms with guideline-concordant care in 208 patients with AECOPD in the emergency department, algorithm use reduced antibiotic exposure by 30 percent without increasing adverse events, including the need for subsequent antibiotic use over the next six months (relative risk [RR] 0.56, 95% CI 0.43-0.73) [94]. In this trial, procalcitonin levels were low at admission in the majority of patients (median 0.096 ng/mL; interquartile range 0.070 to 0.200), which is consistent with epidemiologic data showing that most AECOPD are caused by viruses [22]. Median procalcitonin levels did not differ with the AECOPD severity, suggesting that bacterial infections may not be more prevalent in severe exacerbations. In a noninferiority trial, 302 patients admitted to the ICU with severe AECOPDs and suspected LRTI were randomized to procalcitonin-guided antibiotic initiation and discontinuation versus guideline-concordant care [99]. There were no significant differences in baseline characteristics, procalcitonin levels, or microbiologic findings between groups. However, short-term mortality was higher in patients who received procalcitonin-guided care (20 versus 14 percent; adjusted difference [AD] 6.6 percent, 90% CI -0.3 to 13.5). The mortality difference was highest in the subgroup of 119 patients who did not receive antibiotics at initial presentation (31 versus 12 percent; AD 19.2 percent, 90% CI 7.2-31.1) but not detectable among the 182 patients who received antibiotics at initial presentation. This study highlights the importance of early antibiotic administration in severely ill patients with suspected infection regardless of procalcitonin levels and that the promise of procalcitonin likely lies in guiding antibiotic discontinuation rather than initiation in patients with COPD.

A confounder in the use of procalcitonin in AECOPD is that infections in AECOPD are generally less invasive than in CAP and the spectrum of bacterial pathogens differs. The degree to which procalcitonin levels correlate with the disease process is uncertain [100] because the procalcitonin levels appear to correlate with invasiveness [51] and levels may be lower in AECOPD when compared with CAP [101]. In one trial, no benefit for antibiotic use was found in patients with AECOPD and procalcitonin levels <0.1 ng/mL [102]. In contrast, in a second trial, antibiotics appeared to be beneficial even when procalcitonin levels were low [100], possibly owing to anti-inflammatory effects of specific antimicrobials. Additional data in clinical settings are needed before broad use can be recommended.

The evaluation and management of infection in COPD exacerbations are provided separately. (See "Evaluation for infection in exacerbations of chronic obstructive pulmonary disease" and "Management of infection in exacerbations of chronic obstructive pulmonary disease".)

Acute bronchitis — We do not routinely use procalcitonin for the evaluation and management of patients with acute bronchitis. The majority of cases of acute bronchitis are caused by respiratory viruses, and antibiotics are not routinely recommended. We reserve procalcitonin testing for cases in which the diagnosis of acute bronchitis is uncertain and the need for antibiotics is unclear.

Procalcitonin has not been directly evaluated as an adjunct to clinical diagnosis in patients with acute bronchitis. In four within-study subgroup analyses of trials comparing procalcitonin-guided therapy with standard care in >500 patients with acute bronchitis, there were reductions in unnecessary antibiotic use (ranging from 37 to 80 percent) without increasing adverse outcomes in the procalcitonin group [27,28,59,103]. These findings provide indirect evidence that antibiotics can be safely withheld in patients with low procalcitonin levels (eg, <0.25 ng/mL) when the diagnosis of acute bronchitis is uncertain based on clinical features alone.

The evaluation and management of acute bronchitis are discussed separately. (See "Acute bronchitis in adults".)

Evaluation of dyspnea or other respiratory illnesses — Rarely, we use procalcitonin to help narrow the differential diagnosis in patients presenting with nonspecific symptoms, such as dyspnea. Determining the etiology of dyspnea can be challenging, particularly in patients with comorbidities, such as heart failure, in whom pulmonary edema and LRTI can co-occur. When using procalcitonin diagnostically, we interpret results qualitatively (table 1), weighting its value similarly to other clinical parameters.

In patients with dyspnea, procalcitonin has been studied as a diagnostic aid to distinguish pneumonia from heart failure. Observational data suggest adding procalcitonin to clinical judgment, and other laboratory parameters can enhance diagnostic certainty [73,104-106]. In an analysis of 453 patients presenting to the emergency department with dyspnea, median procalcitonin values were significantly higher in patients with than without clinically diagnosed pneumonia (0.38 [0.12 to 1.40] versus 0.06 [0.04 to 0.09]). A single procalcitonin level was found to be 84 percent accurate for distinguishing heart failure from pneumonia when a threshold of 0.1 ng/mL was used. Consistent results have been shown in other observational studies [104,105], though diagnostic accuracy may wane with the severity of heart failure.

There are few data evaluating clinical outcomes. In a post-hoc analysis of a cohort study evaluating 1641 patients with dyspnea, patients with confirmed acute heart failure who had procalcitonin levels >0.21 ng/mL had worse outcomes when they did not receive antibiotics [105].

The accuracy of procalcitonin for differentiating other respiratory illnesses is not well established. As an example, procalcitonin does not appear to reliably differentiate aspiration pneumonitis from aspiration pneumonia [107,108]. In a prospective cohort, 65 patients admitted with pulmonary aspiration underwent bronchoalveolar lavage at time of admission; no difference in procalcitonin levels were found on day 1 or day 3 between those with positive and negative cultures [107]. Overall, further study on the diagnostic value of procalcitonin is needed.

Coronavirus disease 2019 (COVID-19) — The utility of procalcitonin in patients with COVID-19 is uncertain. Procalcitonin values are low in most patients with COVID-19 [109-112], and values appear to rise with disease severity [40,113-116]. This rise may be due to generic systemic inflammation and is not specific for secondary bacterial infection.

In general, we do not treat patients with COVID-19 with empiric antibiotics unless bacterial superinfection is suspected based on microbiologic testing, chest imaging (eg, dense consolidations on imaging), clinical instability, or other factors. When clinical suspicion for bacterial infection is low, a low procalcitonin value is reassuring and supports the decision to withhold or discontinue antibiotics. By contrast, an elevated procalcitonin value has poor predictive value for bacterial infection and we make the decision to start or stop antibiotics based on other factors. (See "COVID-19: Management in hospitalized adults", section on 'Empiric treatment for bacterial pneumonia in selected patients'.)

Limitations

False positive and false negatives — Procalcitonin is more specific for bacterial infections than other inflammatory markers, such as white blood cell count, erythrocyte sedimentation rate, and C-reactive protein [117]. However, false positives can still occur. Major stressors that cause systemic inflammation, such as severe trauma [118], cardiac arrest or circulatory shock [119], surgery [120], burns [121], pancreatitis [122], and intracranial hemorrhage [123] can also elevate procalcitonin levels, possibly due to gut translocation of lipopolysaccharide or other bacterial products [39]. Procalcitonin can also be elevated in the immediate postnatal period [124], after receipt of immunomodulatory agents (such as T cell antibodies, alemtuzumab, interleukin 2, and granulocyte transfusions) [125-127],with severe liver disease [128], and with certain neoplasms including medullary thyroid cancer [129] and other neuroendocrine tumors [130,131]. Certain autoimmune diseases, such as Kawasaki disease, can raise procalcitonin levels [132], though this does not appear to be the case with most immune disorders (eg, rheumatoid arthritis or systemic lupus erythematosus) [133-135].

Other infectious nonbacterial etiologies that increase procalcitonin include malaria [136,137] and invasive Candida infections [138]. However, procalcitonin levels appear to be lower in these infections when compared with bacterial infections [139,140]. Other pulmonary mold infections, including aspergillosis, mucormycosis, and coccidioidomycosis can cause low-level elevations [141,142].

Procalcitonin may not rise in contained localized infections such as tonsillitis, sinusitis, cystitis, uncomplicated skin/soft tissue infections, abscesses, or empyemas [143,144]. False negatives can also occur if procalcitonin is drawn too early in the course of infection [31]. Notably, procalcitonin production is not impaired in immunocompromised states such as neutropenia, corticosteroids, bone marrow or solid organ transplantation, and human immunodeficiency virus (HIV) [145-149].

Variability among pathogens — The degree to which procalcitonin rises varies among pathogens, with higher levels observed in patients with LRTIs caused by typical bacteria when compared with atypical bacteria or other pathogens [24].

In one multicenter study evaluating 1735 hospitalized patients with CAP, median procalcitonin levels were higher in patients with pneumonia caused by typical bacteria (2.5 ng/mL) compared with atypical bacteria (0.20 ng/mL) and viruses (0.09 ng/mL) [24]. Among atypical bacteria, Legionella species cause modest elevations in procalcitonin [42,150,151], while Mycoplasma and Chlamydia species may not result in detectable elevations [24,152]. The degree of rise provoked by infection with other pathogens is less well studied, although early observations suggest that invasive infection caused by typical bacteria tend to be higher than infections caused by less common pathogens including Mycobacteria tuberculosis, Candida species, P. jirovecii, and Plasmodium species [44,45]. Less commonly, viral infections lead to a rise in procalcitonin [24,41]. This is most pronounced with SARS-CoV-2 infection and appears to correlate with the severity of infection [40].

Use in special populations — It is unclear how to interpret procalcitonin levels in patients with renal impairment; in immunocompromised, pregnant, or surgical patients; or in other patients under severe physiologic stress. Procalcitonin use is not precluded in these populations, but clinicians should be aware of the assay's limitations when interpreting results.

Chronic kidney disease — Persons with chronic kidney disease (CKD) have higher baseline levels of procalcitonin, believed to be due to higher levels of circulating inflammatory cytokines [153,154]. In one study, the average procalcitonin level in healthy persons with CKD prior to starting renal replacement was 1.82 ± 0.39 ng/mL [155]. Levels drop with renal replacement therapy. Average levels in patients with CKD receiving hemodialysis range from 0.26 ng/ mL to 1.0 ng/mL prior to dialysis sessions [153]. After dialysis, levels decline by 20 to 80 percent, varying with the mode of dialysis used. Despite higher baseline levels, procalcitonin does rise in the setting of infection in patients with CKD [156]. However, the rate of rise may be slower than in healthy patients [57,153]. Elimination is marginally prolonged in patients with CKD, with a mean half-life of 28.9 hours in healthy patients compared with 33 hours in patients with creatine clearance <30 mL/minute.

Because of these dynamics, some experts suggest that higher procalcitonin thresholds be used for patients with renal dysfunction [153]. Others suggest that single procalcitonin levels are unreliable in patients with CKD and propose that trends in procalcitonin levels have greater predictive value [155]. Additional studies are needed to determine how to best use procalcitonin in this population and whether use is associated with reduction in unnecessary antibiotic prescription.

Immunocompromise — Immunocompromised patients, apart from patients receiving corticosteroids, have generally been excluded from trials evaluating procalcitonin efficacy. However, observational studies suggest that procalcitonin levels correlate with the presence of bacterial infection and the severity of infection in this population [146,157-160]. Baseline procalcitonin levels tend to be higher in this population (possibly due to other concurrent causes of systemic inflammation), and optimal thresholds for guiding antibiotic therapy have not yet been determined [161]. Because morbidity from bacterial infections can be high in this population, and because immunocompromised patients may require longer durations of antibiotics in general, we typically do not use procalcitonin to guide antibiotic discontinuation in immunocompromised individuals.

Surgery and trauma — Severe physiologic stressors such as surgery, trauma, or burns can lead to procalcitonin elevations [51,162]. While patients with these conditions were not excluded from trials evaluating procalcitonin, they represent a minority of cases in large trials [28,29]. Aggregated data from small studies suggest that procalcitonin has moderate diagnostic accuracy for sepsis in surgical and trauma patients [163] but, as with other special populations, further study is needed to inform practice [66].

PRACTICAL CONSIDERATIONS — Procalcitonin assays are not available at all institutions. Because the usefulness of the assay is dependent on rapid turnaround time, testing should only be obtained at centers where results can be obtained in a timely manner. Several different commercial procalcitonin assays are available in the United States. Most demonstrate good concordance at clinically relevant thresholds [164]. One large randomized controlled trial allowed the use of different analyzers according to site availability, with no difference in recommended thresholds among assays [76].

Overall, knowledge of procalcitonin is evolving and additional studies performed in clinical settings are needed to refine its use. As an example, in the ProACT randomized trial evaluating 1656 patients presenting to the emergency departments with an acute lower respiratory tract infection of any kind (ie, asthma exacerbation, acute chronic obstructive pulmonary disease exacerbation, acute bronchitis, and community-acquired pneumonia), procalcitonin-guided antibiotic initiation or discontinuation did not result in decreased antibiotic use when compared with usual care [62]. Importantly, the hospitals included in this trial all had high adherence to quality measures for the treatment of pneumonia. Similarly, in a pragmatic randomized trial evaluating 285 patients with presumed CAP in the emergency departments of 12 French hospitals, antibiotic duration was similar when comparing procalcitonin-guided antibiotic use versus serial guideline-based clinical assessment (10 versus 9 days) [165]. Clinical success and adverse event rates did not differ between groups. The lack of benefit for procalcitonin use in these trials may reflect increased provider knowledge of appropriate antibiotic prescribing practices.

As with these trials, many existing studies on procalcitonin are limited by a necessary lack of blinding, lack of a reliable reference standard for the diagnosis of bacterial lower respiratory tract infections, variable baseline antibiotic prescription rates, and incomplete algorithm adherence. It remains to be seen if procalcitonin algorithms will continue to be effective as rigorous antimicrobial stewardship programs are implemented and knowledge of appropriate antibiotic use grows. Preliminary data suggest that procalcitonin use can add value to antimicrobial stewardship programs [166].

Use of procalcitonin algorithms should never override clinical judgment. In most trials, algorithms were often overruled by clinical judgment, underscoring the fact that the assay should be used as an adjunct to clinical judgment and not a replacement.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Community-acquired pneumonia in adults" and "Society guideline links: Hospital-acquired pneumonia and ventilator-associated pneumonia in adults" and "Society guideline links: Antimicrobial stewardship" and "Society guideline links: Chronic obstructive pulmonary disease".)

SUMMARY AND RECOMMENDATIONS

●Background − Procalcitonin is a serum biomarker that helps distinguish bacterial infection from other causes of infection or inflammation. In patients with lower respiratory tract infections (LRTIs), procalcitonin can serve as a helpful adjunct to clinical judgment for guiding antibiotic therapy and resolving diagnostic uncertainty. (See 'Introduction' above and 'Rationale for use' above.)

●Guiding early antibiotic discontinuation in patients hospitalized with CAP − We primarily use procalcitonin to guide early antibiotic discontinuation in hospitalized patients with community-acquired pneumonia (CAP) (algorithm 1).

•Clinically stable patients − We generally obtain a level at the time of diagnosis and repeat the level every two days in patients who are clinically stable. We determine the need for continued antibiotic therapy based on clinical improvement and serial procalcitonin levels. (See 'Clinically stable patients' above.)

•Critically ill patients − For patients with CAP who are critically ill or septic at the time of hospitalization, we generally obtain a procalcitonin level daily and consider discontinuing antibiotics when levels are <0.5 ng/mL (or decreased by ≥80 percent from peak when then initial level was >5 ng/mL) and the patient has stabilized. (See 'Critically ill patients' above.)

●Uncertain value for guiding antibiotic initiation − Using procalcitonin levels to help determine whether to start antibiotics in a patient with CAP is controversial. For most patients with known or suspected CAP, we treat with empiric antibiotics regardless of the initial procalcitonin level. For selected patients who are clinically stable and lack major comorbidities, we consider withholding antibiotics when the clinical presentation and/or laboratory results strongly suggest a viral infection and the procalcitonin level is <0.25.

●Use in outpatients with CAP – The approach to using procalcitonin levels to guide antibiotic-decision making for outpatients with CAP varies among experts. When the test is available and turnaround time is quick, some experts use procalcitonin in conjunction with a positive test for a viral pathogen (eg, influenza) to withhold antibiotics or stop them early (provided the patient has close follow-up). Other experts prefer to treat empirically with antibiotics because bacterial CAP is hard to exclude definitively, the morbidity association with CAP is high, and outpatient treatment courses for CAP are short (eg, five days). (See 'Community-acquired pneumonia in outpatients' above.)

●Other respiratory tract infections − For patients with other lower respiratory tract infections such as ventilator-associated pneumonia, acute bronchitis, and chronic obstructive pulmonary disease exacerbations, we do not routinely use procalcitonin in evaluation and management. We suggest that procalcitonin testing be reserved for selected cases. (See 'Other lower respiratory tract infections' above.)

●Helping narrow a differential diagnosis − Occasionally, we use procalcitonin to help narrow the differential diagnosis in patients presenting with nonspecific symptoms such as dyspnea (eg, to help distinguish pneumonia from heart failure). When using procalcitonin diagnostically, we interpret results qualitatively (table 1) and in conjunction with other clinical findings. (See 'Evaluation of dyspnea or other respiratory illnesses' above.)

●Key caveats – The study of procalcitonin is evolving and use of procalcitonin algorithms should never override clinical judgment. In most trials, algorithms were commonly overruled by clinical judgment, underscoring the fact that the assay should be used as an adjunct to clinical judgment and not a replacement. (See 'Practical considerations' above.)

●False positives and false negatives − When interpreting procalcitonin levels, clinicians should be aware of factors that cause procalcitonin levels to rise apart from bacterial LRTI and how the magnitude of procalcitonin elevation varies among pathogens (table 2). (See 'False positive and false negatives' above and 'Variability among pathogens' above.)

●Use in special populations − Procalcitonin has not been well studied in certain populations, such as immunocompromised patients, surgical patients, and patients with chronic kidney disease. While procalcitonin use is not precluded in these populations, clinicians should be aware of the assay's limitations when interpreting results. (See 'Use in special populations' above.)

ACKNOWLEDGMENT — UpToDate gratefully acknowledges John G Bartlett, MD, who contributed on earlier versions of this topic and was a founding Editor-in-Chief for UpToDate in Infectious Diseases.