How to best use procalcitonin to diagnose infections and manage antibiotic treatment

Introduction

Acute respiratory tract illnesses and suspected sepsis often prompt initiation of empiric antibiotic treatment. In many cases, however, a bacterial pathogen cannot be detected, as viruses account for a large proportion of respiratory illnesses [1, 2]. The same is true in patients presenting with systemic inflammatory response syndrome (SIRS) and possible sepsis – where other inflammatory or viral illnesses may be the cause of a significant proportion of cases. Despite technological advances and the wide availability of rapid molecular diagnostics [3], antibiotics are often over-prescribed due to concerns about bacterial coinfections and the risk of withholding therapy. Once started, physicians often prescribe prolonged courses of antibiotics due to lack of proof that the infection has been resolved. Unnecessarily long treatment also often results from the application of antibiotic regimens advocated by standard practice guidelines. Individualizing antibiotic treatment may potentially improve antibiotic stewardship efforts to encourage judicious and correct usage of these agents. This would further mitigate the emergence of multi-drug resistant pathogens – a situation directly linked to antibiotic overuse and one of the most urgent health threats worldwide [4].

Regarding clinical care of patients with acute respiratory illnesses and sepsis, the integration of host response markers which correlate with the likelihood of bacterial infection has strong potential to improve individual antibiotic therapy decision-making [5]. Among such host-response markers, Procalcitonin (PCT) has generated much interest due to its higher specificity towards bacterial infection compared to C-reactive protein (CRP) or white blood cell count (WBC) [6, 7]. PCT also has a superior prognostic [8] and kinetic profile to track infections – making it useful to assess response to treatment [7, 9], [10], [11], [12].

The goal of this narrative review is to discuss different studies and clinical trials that have investigated clinical effects of using PCT for the purpose of antibiotic stewardship, and to provide practical advice on how to best integrate PCT into clinical care. Thereby, several milestone studies about PCT that were published in Clinical Chemistry and Laboratory Medicine (CCLM) over the last years are particularly highlighted and critically discussed.

Clinical trials investigating PCT-guided antibiotic stewardship

In addition to a large body of observational data [13], several interventional trials have compared PCT protocols with control groups regarding antibiotic use and clinical outcomes [14, 15]. Most trials focused on patients with respiratory infection and/or sepsis. Although the protocols followed slightly different PCT guidelines, there were some similarities: PCT cut-off values in low risk patients were used to identify individuals for whom bacterial infection was unlikely and antibiotic treatment could be stopped, or withheld [16]. If the probability for bacterial infection was high, the protocol focused on PCT kinetics and cessation of antibiotics when PCT levels dropped to normal ranges, or by at least 80–90%. PCT protocols had varying cut-offs for low or higher acuity/severity: for emergency department and medical ward patients, a PCT<0.25 μg/L recommended no use of antibiotics, while in intensive care patients a PCT<0.5 μg/L advocated discontinuation of antibiotics.

There are several meta-analyses which have investigated the effectiveness and safety of utilizing PCT to guide antibiotic treatment. One individual patient data meta-analysis of 26 trials focusing on respiratory infections reported that PCT guidance was associated with a 2.4-day reduction in antibiotic exposure (5.7 vs. 8.1 days). This was due to decreased rates of antibiotic initiation in low-risk patient (i.e., bronchitis or COPD patients) and shorter treatment courses in high risk patients (i.e., pneumonia patients) [7, 15]. Importantly, use of PCT resulted in reductions of antibiotic-related side-effects (16 vs. 22%), and in significantly lowered mortality (286 [9%] deaths in 3,336 procalcitonin-guided patients vs. 336 [10%] in 3,372 controls [p=0.037]). Similar effects were also seen for other respiratory infections and clinical settings.

Another meta-analysis of 11 trials and 4,482 patients investigated the effects of applying PCT in patients with sepsis [14]. Those randomized to the PCT protocol experienced an earlier discontinuation of antibiotics and significant reductions in mean treatment duration from 10.4 to 9.3 days. Here again, mortality in PCT-guided patients was significantly lower compared to the control group (21.1 vs. 23.7%, p=0.03). In a further meta-analysis of individuals with positive blood cultures, PCT reduced both antibiotic exposure (by 2.86 days) and mortality (16.6 vs. 20.0%) [17]. A subgroup analysis demonstrated the strongest effects for pneumonia caused by Streptococcus pneumoniae (−4.52 days) and urogenital infections caused by Escherichia coli (−4.21 days).

How to implement PCT in clinical practice

Based on these trials, important considerations for PCT use in clinical practice must be made. The interpretation of PCT results varies according to clinical setting, specific patient situation, and type of illness.

In low risk patients in primary care, a marker such as PCT can rule out bacterial disease and reduce antibiotic initiation [18]. Two large primary care PCT trials examined a total of 1,008 patients with lower and upper respiratory tract infections – whereby a low PCT of ≤0.25 μg/L prohibited the use of antibiotics [18, 19]. Antibiotic initiation was reduced from 63% in control group patients to 23% in PCT-guided patients; with a significant reduction in antibiotic exposure from 4.6 to 1.6 days and no difference in clinical outcomes. Currently, the most significant limitation to the use of PCT in primary care is the lack of highly sensitive point-of-care tests. Furthermore, other markers such as CRP – where tests are more widely available and less expensive – have shown benefit in primary care patients despite their lower specificity regarding bacterial infections [20]. As corresponding large scale head-to-head trials are lacking, further study is needed to compare PCT-guided care with other biomarkers such as CRP. Cost-effectiveness and country-specific reimbursement of PCT must also be addressed.

Several trials of mostly respiratory infection patients have investigated PCT in the emergency department setting. A study of over 500 patients with bronchitis found significant reduction in initiation of antibiotics (26 vs. 66%) when PCT was used as guideline [15]. Similarly for COPD exacerbation, an analysis of more than 1,200 patients found significant reduction in the application of antibiotics with PCT-guided care (43 vs. 72%). PCT may also help emergency department personnel to discriminate between chronic heart failure patients with decompensation vs. respiratory infection [21, 22]. For cases with pneumonia, baseline PCT levels obtained in the emergency department help assess kinetics over time and guide duration of antibiotic therapy.

Several trials on pneumonia patients admitted to medical wards have found that serial PCT measurements and the termination of antibiotics when PCT dropped by ≥80% (or to <0.25 μg/L) resulted in shorter treatment durations, lower risk of adverse drug events, and improved survival rates [23]. One meta-analysis of more than 3,000 pneumonia patients reported reductions in antibiotic treatment from 10.4 day to 7.5 days, and improvements in treatment failure (22 vs. 26%) and mortality (12 vs. 14%) [15]. The positive impact of PCT-guided antibiotic management on mortality may be explained by various factors; including a reduction of direct toxic effects of antibiotics, a lowering of risk for antibiotic related Clostridium difficile infections, and changes in diagnostic and therapeutic monitoring and management based on prognostic information. Less trial data is available regarding the use of PCT in surgical patients [24] – where it is important to understand that inflammatory stress-induced increases in PCT concentrations correlate with the extent of surgery [10]. One study indicated that PCT was highest on the second day postop and typically declined thereafter in patients with uncomplicated recoveries. Persistently elevated PCT levels have been conversely associated with the development of postsurgical infection [25, 26]. The acquisition of more data is necessary before PCT is broadly implemented to assist management of post-surgical complications.

The intensive care unit is another important setting where clinical trial data has demonstrated the usefulness of PCT for antibiotic decision-making. Here, PCT-guided care does not focus on the initiation of antibiotics, but rather on determining clinical response to treatment and possible discontinuation of therapy [27]. Several large multicenter trials have found PCT helpful in reducing antibiotic exposure [28, 29]. Most trials stopped antibiotic treatment when PCT decreased by ≥80% from its peak value; or to a level of ≤0.5 μg/L. A recent meta-analysis of PCT-guided antibiotic treatment found reductions in both duration of treatment and mortality [14, 30].

Practical considerations

Decisions regarding antibiotic use in individual patients is complex and should be based on several considerations including: the pretest probability for bacterial infection (based on clinical examination and microbiological findings), the severity of presentation, and PCT test results. Figure 1 provides a practical guide for the rational use of PCT in both high and low risk settings; in conjunction with clinical assessment including interpretation of PCT, and recommendations for antibiotic use as discussed at a consensus conference of international experts [16, 31].

Figure 1:

Clinical algorithm for PCT use in different clinical settings. (A) Suggested PCT protocol in low and moderate severity patients [31]. (B) Suggested PCT protocol in critical care patients [31].

In low-risk situations (e.g., patient in primary care or on the medical ward) and a minimal pretest probability for bacterial infections (e.g., individuals presenting with bronchitis), low PCT levels <0.25 μg/L help rule out bacterial infection and empiric antibiotic therapy should be avoided. PCT should be retested if the patient does not improve clinically. Antibiotics should be considered if PCT increases, or initial clinical assessment indicates high probability of bacterial infection. PCT testing can then be undertaken every 24–48 h, with discontinuation of antibiotics if levels drop to ≤0.25 μg/L or decrease by 80% or more from peak values.

Concerning high-risk patient with sepsis, initial antibiotics should be used irrespective of PCT results – but low PCT values may prompt additional diagnostic measures to rule out other non-bacterial causes of illness. In these situations, monitoring of PCT over time helps track resolution of infection and decision-making regarding early termination of antibiotic treatment.

Finally, while most studies are performed in Europe and the US, it is important to adapt existing algorithms for differences in type of infections in regions with tropical diseases [32].

Limitations of procalcitonin

Most PCT studies investigate individuals with respiratory infections or sepsis, and data is limited for immunosuppressed patients [11]. In addition, various non-infectious conditions such as C-cell carcinoma or trauma can cause systemic inflammation and increases in PCT [33]. Also, PCT-guided stewardship should not be applied to patients with chronic infections such as osteomyelitis or endocarditis, as observational studies have not shown positive results and interventional research is lacking [31]. For emerging infections such as COVID-19, the role of PCT as a stewardship tool requires further study [34]. PCT is also relatively low in some atypical infections – including mycoplasma [35]. Some clinical conditions such as kidney failure and dialysis may influence PCT kinetics [36], although one large individual patient meta-analysis suggested that use of PCT in patients with impaired kidney function works well and is associated with shorter antibiotic courses and lower mortality rates [37]. Finally, a large number of new PCT immunoassays are currently being developed (including point-of-care tests [38]), which will require careful analysis and clinical evaluation [39], [40], [41], [42], [43], [44], [45]. The expansion of PCT assays also necessitates evaluation of result comparability among tests and standardized calibration to support safe usage of PCT in clinical routine [46], [47], [48]. Finally, cost-effectiveness analyses are needed to evaluate how widespread PCT testing would affect health care budgets – in comparison with the resulting benefits for clinical outcomes and antibiotic resistance patterns [49, 50].

Concluding remarks

The medical community is highly interested in reducing unnecessary antibiotic exposure in patients with acute respiratory illness and low risk for bacterial infection, as well as shortening therapy in cases of known bacterial infections. Essentially, the aim is to shift from fixed antibiotic doses to individualized treatment. As a marker of host defense response, PCT has shown promising results for respiratory infections and sepsis. PCT should not be used as a substitute for good clinical practice, but rather be part of the overall assessment of a patient. Decisions pertaining to initiation and cessation of antibiotic treatment remains strongly dependent on evaluating all available clinical and diagnostic parameters, including a thorough assessment of the patient and severity of illness. Furthermore, PCT usage should not delay or impede initiation of empirical treatment in high-risk situations. Host response markers such as PCT, however, remain the best line of defense against diagnostic uncertainty and antibiotic overuse. Further research is needed to explore the optimal application of biomarkers in combination with pathogen-directed tests. Importantly, medical Journals such as Clinical Chemistry and Laboratory Medicine (CCLM) have played a critical role in reviewing and dissemination the high-quality evidence about assays for PCT measurement, observational research regarding association with outcomes among different populations, and interventional research proofing its effectiveness for patient care.