Early diagnosis and initiation of resuscitation measures, antibiotic treatment and/or source control remains the cornerstone of sepsis care (1). Once treatment is initiated, close monitoring of sepsis patients is of utmost importance to identify patients with adverse disease course who may require alternative therapeutic approaches. Secondarily, monitoring may identify patients with a favorable course trajectory who are at low risk for complications warranting discharge from the ICU and possibly deescalation of antibiotic therapy. Appropriate deescalation of the intensity of care with transfer to the floor setting is an important decision because ICU readmissions due to treatment failure on the floor are associated with adverse prognosis and prolonged in-hospital stay (2). Daily assessment of patient risk using objective prognostic data may aid in these intensification and deescalation decisions.
In addition to clinical signs and symptoms, blood markers may assist in patient monitoring (3–6). While serum lactic acid is a biomarker commonly used to help guide response to resuscitation measures (7), procalcitonin (PCT) has been previously demonstrated to be helpful in antibiotic stewardship decisions (8–10). PCT is a host-response marker that is up-regulated by microbial toxins and certain proinflammatory mediators (e.g., interleukin-1b, tumor necrosis factor-α, interleukin-6) and is down-regulated during recovery (11). The expression of PCT is attenuated by the cytokines typically released in response to a viral infection (e.g., interferon-γ); thus, an elevated PCT is typically indicative of a bacterial infection (12). In addition to its diagnostic value, the kinetics of PCT have also been shown to predict mortality and treatment failure in sepsis (13–19). In a recent retrospective analysis involving 256 sepsis patients from two ICUs in the United States, a lack of PCT decrease by more than 80% over the first 72 hours was associated with an increase in ICU and in-hospital mortality, independent of prognostic ICU risk scores (5).
Prior to this study, Food and Drug Administration (FDA) approved use of PCT was limited to single point measurement in the ICU. Therefore, we sought to expand the indication to apply to serial measurements of severe sepsis and septic shock patients, across all hospital settings. Herein, we performed a large, prospective U.S.-based multicenter study to validate the prognostic accuracy of an inability to decrease PCT by more than 80% from baseline to day 4 to predict 28-day mortality in patients with severe sepsis or septic shock who were being admitted to the ICU from the emergency department (ED) or other wards.
MATERIALS AND METHODS
This report adheres to the Reporting of Observational trials (STrengthening the Reporting of OBservational studies in Epidemiology) (20). The study protocol was approved by the U.S. FDA and registered at ClinicalTrials.gov (https://clinicaltrials.gov/ct2/show/NCT01523717, Registration date: January 19, 2012).
Inclusion criteria for this study were adult patients (age, ≥ 18) who were diagnosed with severe sepsis or septic shock according to adapted American College of Chest Physicians/Society of Critical Care Medicine criteria (21–23) (Supplemental trial definitions, Supplemental Digital Content 1, http://links.lww.com/CCM/C411); patients treated in the ICU or a decision to admit the patient from the ED or other wards; blood sampling performed within 12 hours after diagnosis of severe sepsis or septic shock; and willingness to provide written informed consent. We excluded patients without initial blood draw and patients missing data regarding the primary outcome (28-d mortality). For the intention-to-diagnose (ITD) analysis (analysis mandated by the FDA), patients who died or were discharged from the hospital prior to the day 4 blood draw were also excluded (Supplemental Fig. S1, Supplemental Digital Content 1, http://links.lww.com/CCM/C411). We conducted a sensitivity analysis using the per-protocol population (PP) where we excluded patients who did not meet our sepsis definition or were not transferred to the ICU (Supplemental Fig. S1, Supplemental Digital Content 1, http://links.lww.com/CCM/C411).
In addition to blood specimens, we collected clinical data obtained at admission and during the hospital stay. All participating study centers had evidence-based sepsis protocols in place supported by the latest Surviving Sepsis Campaign guidelines for management of patients with severe sepsis and septic shock (24–26).
The primary endpoint of this study was 28-day all-cause mortality. To verify vital status, patients were followed during the hospital stay and contacted by telephone at day 28.
Sample Collection and PCT Measurement
Blood samples were collected within 12 hours after diagnosis of severe sepsis or septic shock and then daily for a total of 5 days. PCT was measured with the B·R·A·H·M·S PCT sensitive KRYPTOR (B·R·A·H·M·S GmbH, Hennigsdorf, Germany) (27). We tested the prognostic accuracy of a lack in PCT decrease of more than 80% (ΔPCT in %) from baseline to day 4. Because in some patients PCT may still increase within the first 24 hours before reaching the maximum level, we also investigated the kinetics from the maximum level on baseline or days 1–4 (Supplemental trial definitions, Supplemental Digital Content 1, http://links.lww.com/CCM/C411). In addition, we investigated the prognostic ability of baseline PCT levels as well as short-term kinetics of PCT from baseline to day 1 for mortality prediction.
The primary study population was the ITD population. The analysis was repeated in the per-PP excluding patients with protocol violations (Supplemental Fig. S1, Supplemental Digital Content 1, http://links.lww.com/CCM/C411). Based on a previous study that identified a decrease in PCT by 80% or lower as the best threshold for predicting mortality, we designed this study to validate this cutoff (5). Therefore, we hypothesized a priori that an inability to decrease PCT by more than 80% from baseline to day 4 would predict 28-day all-cause mortality. Please see online supplement for the details of the statistical approach and sample size calculations.
From December 2011 to March 2014, 858 patients were enrolled in the study, with 38 patients excluded because of missing or withdrawn written informed consent, missing initial blood draw, or missing information regarding the primary outcome (Supplemental Fig. S1, Supplemental Digital Content 1, http://links.lww.com/CCM/C411). There were another 174 patients who were excluded due to a missing day 4 blood draw (median PCT value at baseline was 7.5 μg/L [interquartile range (IQR), 1.7–21.3 μg/L] for the 73 patients excluded due to death and 1.7 μg/L [IQR, 0.3–11.0 μg/L] for the 101 patients excluded due to hospital discharge prior to day 4 blood draw). The main ITD analysis included 646 patients.
Patients had a mean age of 64 years and 57% were male gender. Infections were mainly community acquired (92%) and slightly more than half of the patients had severe sepsis (54%) when compared with septic shock (46%) (Table 1). Among the 820 patients enrolled were 184 deaths for a 22% all-cause 28-day mortality rate. Among the 646 patients in the ITD population, there were a total of 107 deaths for a 17% overall 28-day mortality rate. The median stay in the ICU was 4 days and patients stayed for a median of 11 days in the hospital. Table 1 shows detailed patient characteristics in the overall ITD population stratified by 28-day survival status. Supplemental Table S1 (Supplemental Digital Content 1, http://links.lww.com/CCM/C411) also shows baseline results in the per-PP and the overall population.
PCT Kinetics and 28-Day Mortality
In regard to the primary hypothesis, the mortality rate was nearly double for those who did not decrease their PCT by more than 80% from baseline to day 4 when compared with those who did decrease by more than 80% (20% vs 10%; p = 0.001) (Table 2). The prognostic measures at this cutoff showed a sensitivity of 77% (95% CI, 69–85%) with a specificity of 39% (35–43%), a negative predictive value of 90% (86–94%), and a positive predictive value of 20% (16–24%). In our sensitivity analysis, comparable results were obtained in the per-protocol patient population (Supplemental Fig. S2, a–c, Supplemental Digital Content 1, http://links.lww.com/CCM/C411).
Kaplan-Meier and Cox Regression Analysis
Results were further analyzed in a time-to-event analysis. Figure 1 shows Kaplan-Meier plots for the overall population and stratified by patient location at day 4. Again, the 80% PCT decrease cutoff from baseline to day 4 significantly separated survivors from nonsurvivors.
To assess whether the PCT decrease provides prognostic information beyond that of other clinical outcome predictors, we calculated multivariate Cox regression models including Acute Physiology and Chronic Health Evaluation (APACHE) II or maximum Sequential Organ Failure Assessment (SOFA) score, as well as nine clinical variables associated with severity and/or adverse outcome (appropriateness of antibiotic therapy, sepsis syndrome, microbiology, nosocomial vs community infection, blood culture positivity, initial PCT level, age, gender, and patient location at day 4). In the unadjusted analysis, the hazard ratio (HR) for 28-day mortality of patients without a PCT decrease of more than 80% until day 4 was 2.05 (95% CI, 1.30–3.24) (Table 3). This result remained robust in the model including adjustment for APACHE II and nine other variables with a HR of 1.97 (95% CI, 1.18–3.30; p < 0.01). Results were similar when using maximum SOFA score instead of APACHE II in the fully adjusted model (Supplemental Table S2, Supplemental Digital Content 1, http://links.lww.com/CCM/C411). Furthermore, PCT remains a significant predictor in models incorporating change in WBC count between baseline and day 4 (Supplemental Table S3, Supplemental Digital Content 1, http://links.lww.com/CCM/C411), as well as a model incorporating change in SOFA score between baseline and day 4 (Supplemental Table S4, Supplemental Digital Content 1, http://links.lww.com/CCM/C411). We also note that the maximum SOFA score over the first 4 days (stratified by maximum SOFA, > 8) is also associated with a approximately two-fold mortality risk (Supplemental Table S2, Supplemental Digital Content 1, http://links.lww.com/CCM/C411). We repeated the same analyses in the per-protocol patient population and achieved comparable results (Supplemental Table S5, Supplemental Digital Content 1, http://links.lww.com/CCM/C411).
Shorter Term Change in PCT (Baseline to Day 1) and Baseline PCT. In addition to the primary analysis where we analyzed the prognostic value of a PCT decrease of more than 80% between baseline and day 4, we assessed the prognostic value of a shorter term change of PCT between baseline and day 1. Among the 752 patients available for analysis, patients who died have an average mean increase of 30% (95% CI, 15–47%) when compared with 0% (95% CI, –7% to +6%) for those who survived (p < 0.001). The area under the curve (AUC) for the short-term increase was 0.64 (95% CI, 0.59–0.69). When simply stratifying by patients with an initial increase in PCT from baseline to day 1 (n = 323) when compared with a decrease (n = 429), patients with a PCT increase had an almost three-fold higher mortality (mortality, 29% vs 12%; p < 0.0001).
In addition to the concept of serial PCT measurements, we investigated the ability of a single PCT value to predict 28-day all-cause mortality at baseline (n = 820). Although nonsurvivors had higher mean baseline PCT levels compared with survivors (5.2 μg/L [95% CI, 3.9–7.0 μg/L] vs 3.4 μg/L [95% CI, 2.8–4.0 μg/L]; p < 0.02), a single PCT value at baseline was less predictive for 28-day mortality than serial measurements (Supplemental Table S6, Supplemental Digital Content 1, http://links.lww.com/CCM/C411). The AUC for baseline PCT as a predictor of mortality was 0.56 (95% CI, 0.51–0.60). Furthermore, when comparing the PCT values over the first 5 days among survivors versus nonsurvivors, the PCT values for the nonsurvivors were higher and stayed higher on all days 1–5 (Fig. 2).
Combined Initial PCT, PCT Change, and ICU Status. As an additional exploratory analysis requested by the FDA, we assessed whether the prognostic performance of our main covariate of interest, PCT reduction of more than 80% at day 4, was different when stratified by patient location at day 4 (ICU vs non-ICU) (Table 2). Mortality was higher in patients still hospitalized in the ICU at day 4 (26%) than in those patients already discharged to the floor at day 4 (9%). Among patients discharged from the ICU by day 4 who had a high baseline PCT value of greater than 2 μg/L, mortality was more than three-fold increased if PCT did not drop by more than 80% (19% vs 5%) (Supplemental Table S7, Supplemental Digital Content 1, http://links.lww.com/CCM/C411). Similarly, for patients still residing in the ICU at day 4 and low baseline PCT of less than or equal to 2 μg/L, mortality was about three-fold higher if PCT did not drop by more than 80% compared with PCT that decreased by more than 80% (26% vs 10%).
Within this large, multicenter study including 13 study sites across the United States, we found that kinetics of PCT over the first 4 days were predictive for survival of patients diagnosed with severe sepsis or septic shock. These results remained significant after multivariate adjustment for other known prognostic variables and risk factors. These findings validate previous retrospective research from the United States (5), as well as other smaller studies (14, 16–19). In addition to daily routine clinical assessment, monitoring of PCT in this patient population was demonstrated to aid in risk assessment, which may translate into better informed clinical decisions regarding intensification of care or ICU discharge.
Previous Related PCT Studies
Several studies have investigated the prognostic utility of PCT in systemic infection and sepsis (reviewed in ). In one retrospective analysis from the United States, PCT change over the first 72 hours was also a predictor for ICU and in-hospital mortality in sepsis patients, independent of ICU risk scores (APACHE IV and Simplified Acute Physiology Score II) (5). In addition to general sepsis, studies also found PCT to improve risk stratification in respiratory infections, mainly community-acquired pneumonia and chronic obstructive pulmonary disease exacerbation (15, 16). An individual patient data meta-analysis including more than 4,000 patients from 14 trials found an association between admission PCT levels and treatment failure across different types of respiratory infections and treatment settings (16). In a large U.S. pneumonia cohort (28), the greatest benefit of PCT was found in patients classified as high risk by the pneumonia severity index. A PCT level less than 0.1 μg/L virtually excluded mortality in these high-risk patients. Our current analysis is in line with these previous investigations and supports the prognostic utility of PCT when measured serially.
Implications of Findings
Early recognition and the start of appropriate antibiotic treatment, fluid resuscitation, source control, and close patient monitoring remain the cornerstone of care to lower sepsis related morbidity and mortality (1). In real-life practice, this remains challenging because clinical variables lack specificity for sepsis etiology and prognosis (29). Additionally, microbiologic diagnostics have low sensitivity and modalities such as cultures do not provide needed information in a timely fashion. Therefore, sepsis biomarkers are promising tools to help monitor patient response to therapy and may help guide therapeutic decisions in individual patients. It is particularly challenging during initial patient evaluation to determine which patients with sepsis will not respond well to therapeutic measures, and thus will have worsening of their clinical status. Specifically, the development of new or worsening organ dysfunction portends poor outcome and is a common pathway to death in these patients (30, 31). Novel strategies that improve a clinician’s ability to accurately risk stratify patients with suspected sepsis facilitate early and appropriate therapeutic intervention, improve important triage decisions (e.g., admission to the hospital vs discharge home, or admission to ICU vs non-ICU bed, or continuation of care in the ICU vs discharge to the hospital ward), and provide a means to follow response to therapy, including antibiotic stewardship (32). PCT has generated much interest as a sepsis biomarker that is associated with increased risk and severity of bacterial infection. PCT has also been found to correlate with risk of culture-proven bacteremia (33). Although the biological function of PCT in host defense is incompletely understood, this peptide influences the immune system in various ways including a decrease in phagocytic and candidacidal activity of neutrophils and also leads to an increase in the concentration of intracellular calcium ions (34, 35) which facilitate the host response.
Importantly, previous research demonstrates that PCT can be used to inform antibiotic stewardship decisions, mainly by reducing antibiotic initiation in low-risk patients (i.e., bronchitis, upper respiratory infection) and by early stoppage of antibiotics in patients with pneumonia and sepsis (10, 32, 36). A recent 1,500 patient multicenter, ICU trial from the Netherlands which included more than 1,500 patients with suspected or confirmed infection found a significant reduction in antibiotic use and a mortality in the group that followed a PCT-based antibiotic stewardship algorithm (37, 38). Our observation that a drop in PCT of at least 80% was associated with survival outcomes further supports the approach of monitoring PCT levels over time. Interestingly, our secondary analysis identified that simply assessing whether PCT decreases or increases from baseline to day 1 revealed a three-fold higher mortality in patients with a short-term increase in PCT levels. This simple finding could prove particularly useful during early critical care management. However, while PCT may contribute to prognosis, it is not able to identify nonsurvivors such that one would, for example, withdraw care if the PCT level does not decrease by more than 80%. Still, beyond antibiotic deescalation in patients with clinical improvement and decreasing PCT levels, it remains unclear which therapeutic and/or diagnostic measures should be used in patients with nondecreasing PCT levels to improve outcomes (39, 40). One large critical care trial investigating whether PCT-guided initiation and escalation of antimicrobial therapy would improve survival did not find any outcomes benefit for this strategy (13). However, further interventional studies are needed to investigate whether monitoring of PCT kinetics over longer time periods will improve outcomes in patients with sepsis. Such a study should also investigate costs of PCT measurement and possible costs savings. Based on a literature review, a previous study has estimated of the total cost of performing a PCT assay, including assay material, reagents, technician time, purchase, maintenance of a bench top analyzer, and overhead, to be approximately Can$49.42 (approximately $50 USD) per test (41).
Strengths and Limitations
The strengths of this study are the rigorous study design with a large sample size, prospective inclusion of patients being treated for sepsis in different experienced centers across the United States, and the relatively low number of patients who were lost to follow-up. The study hypothesis was defined a priori, based on retrospective data (5), which strengthens our conclusions and provides external validation. As a limitation, we were not able to use patients who died or were discharged prior to the day 4 blood draw in our main analysis and the prognostic accuracy in these patients is not clearly defined. Also, as an observational study, we cannot make conclusions about clinical effects of using a PCT algorithm for patient care. An interventional study is needed to test the hypothesis that serial monitoring of PCT will improve clinical decisions and outcomes (10, 40).
The results of this large, U.S.-based prospective multicenter study in several U.S. hospitals indicate that inability to decrease PCT by at least 80% is a significant independent predictor of mortality and may aid in sepsis care.
We are grateful to all the physicians, nursing staff, and patients who participated in this study.
1. Dellinger RP, Levy MM, Rhodes A, et al; Surviving Sepsis
Campaign Guidelines Committee including The Pediatric Subgroup: Surviving Sepsis
Campaign: International guidelines for management of severe sepsis
and septic shock, 2012. Intensive Care Med 2013; 39:165–228
2. Ouanes I, Schwebel C, Français A, et al; Outcomerea Study Group: A model to predict short-term death or readmission after intensive care unit discharge. J Crit Care 2012; 27:422.e1–422.e9
3. Jun KR, Lee JN, Song SA, et al. Serial changes in serum procalcitonin
, interleukin 6, and C-reactive protein levels according to non-specific surgical stimulation. Clin Chem Lab Med 2015; 53:549–558
4. Henriquez-Camacho C, Losa J. Biomarkers for sepsis
. Biomed Res Int 2014; 2014:547818
5. Schuetz P, Maurer P, Punjabi V, et al. Procalcitonin
decrease over 72 hours in US critical care units predicts fatal outcome in sepsis
patients. Crit Care 2013; 17:R115
6. Schuetz P, Litke A, Albrich WC, et al. Blood biomarkers for personalized treatment and patient management decisions in community-acquired pneumonia. Curr Opin Infect Dis 2013; 26:159–167
7. Jones AE, Shapiro NI, Trzeciak S, et al; Emergency Medicine Shock Research Network (EMShockNet) Investigators: Lactate clearance vs central venous oxygen saturation as goals of early sepsis
therapy: A randomized clinical trial. JAMA 2010; 303:739–746
8. Schuetz P, Mueller B. Procalcitonin
: An effective screening tool and safe therapeutic decisionmaking aid for emergency department patients with suspected sepsis
. Ann Emerg Med 2015; 66:318–319
9. Schuetz P, Mueller B. Biomarker
-guided de-escalation of empirical therapy is associated with lower risk for adverse outcomes. Intensive Care Med 2014; 40:141
10. Schuetz P, Briel M, Christ-Crain M, et al. Procalcitonin
to guide initiation and duration of antibiotic treatment in acute respiratory infections: An individual patient data meta-analysis. Clin Infect Dis 2012; 55:651–662
11. Schuetz P, Albrich W, Mueller B. Procalcitonin
for diagnosis of infection and guide to antibiotic decisions: Past, present and future. BMC Med 2011; 9:107
12. Linscheid P, Seboek D, Schaer DJ, et al. Expression and secretion of procalcitonin
and calcitonin gene-related peptide by adherent monocytes and by macrophage-activated adipocytes. Crit Care Med 2004; 32:1715–1721
13. Jensen JU, Hein L, Lundgren B, et al; Procalcitonin
And Survival Study (PASS) Group: Procalcitonin
-guided interventions against infections to increase early appropriate antibiotics and improve survival in the intensive care unit: A randomized trial. Crit Care Med 2011; 39:2048–2058
14. Karlsson S, Heikkinen M, Pettilä V, et al; Finnsepsis Study Group: Predictive value of procalcitonin
decrease in patients with severe sepsis
: A prospective observational study. Crit Care 2010; 14:R205
15. Schuetz P, Amin DN, Greenwald JL. Role of procalcitonin
in managing adult patients with respiratory tract infections. Chest 2012; 141:1063–1073
16. Kutz A, Briel M, Christ-Crain M, et al. Prognostic value of procalcitonin
in respiratory tract infections across clinical settings. Crit Care 2015; 19:74
17. Ruiz-Rodríguez JC, Caballero J, Ruiz-Sanmartin A, et al. Usefulness of procalcitonin
clearance as a prognostic biomarker
in septic shock. A prospective pilot study. Med Intensiva 2012; 36:475–480
18. Suberviola B, Castellanos-Ortega A, González-Castro A, et al. [Prognostic value of procalcitonin
, C-reactive protein and leukocytes in septic shock]. Med Intensiva 2012; 36:177–184
19. Pieralli F, Vannucchi V, Mancini A, et al. Procalcitonin
kinetics in the first 72 hours predicts 30-day mortality in severely ill septic patients admitted to an intermediate care unit. J Clin Med Res 2015; 7:706–713
20. von Elm E, Altman DG, Egger M, et al; Iniciativa STROBE: [The Strengthening the Reporting of Observational Studies in Epidemiology [STROBE] statement: Guidelines for reporting observational studies]. Gac Sanit 2008; 22:144–150
21. Levy MM, Fink MP, Marshall JC, et al. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis
Definitions Conference. Crit Care Med 2003; 31:1250–1256
22. Bone RC, Balk RA, Cerra FB, et al. Definitions for sepsis
and organ failure and guidelines for the use of innovative therapies in sepsis
. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care Medicine. Chest 1992; 101:1644–1655
23. Vincent JL, Moreno R, Takala J, et al. The SOFA (Sepsis
-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis
-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med 1996; 22:707–710
24. Shapiro NI, Howell M, Talmor D. A blueprint for a sepsis
protocol. Acad Emerg Med 2005; 12:352–359
25. Levy MM, Dellinger RP, Townsend SR, et al. The Surviving Sepsis
Campaign: Results of an international guideline-based performance improvement program targeting severe sepsis
. Intensive Care Med 2010; 36:222–231
26. Dellinger RP, Levy MM, Carlet JM, et al; International Surviving Sepsis
Campaign Guidelines Committee; American Association of Critical-Care Nurses; American College of Chest Physicians; American College of Emergency Physicians; Canadian Critical Care Society; European Society of Clinical Microbiology and Infectious Diseases; European Society of Intensive Care Medicine; European Respiratory Society; International Sepsis
Forum; Japanese Association for Acute Medicine; Japanese Society of Intensive Care Medicine; Society of Critical Care Medicine; Society of Hospital Medicine; Surgical Infection Society; World Federation of Societies of Intensive and Critical Care Medicine: Surviving Sepsis
Campaign: International guidelines for management of severe sepsis
and septic shock: 2008. Crit Care Med 2008; 36:296–327
27. Christ-Crain M, Jaccard-Stolz D, Bingisser R, et al. Effect of procalcitonin
-guided treatment on antibiotic use and outcome in lower respiratory tract infections: Cluster-randomised, single-blinded intervention trial. Lancet 2004; 363:600–607
28. Huang DT, Weissfeld LA, Kellum JA, et al; GenIMS Investigators: Risk prediction with procalcitonin
and clinical rules in community-acquired pneumonia. Ann Emerg Med 2008; 52:48–58.e2
29. Vincent JL. Dear SIRS, I’m sorry to say that I don’t like you. Crit Care Med 1997; 25:372–374
30. Abraham E, Laterre PF, Garg R, et al; Administration of Drotrecogin Alfa (Activated) in Early Stage Severe Sepsis
(ADDRESS) Study Group: Drotrecogin alfa (activated) for adults with severe sepsis
and a low risk of death. N Engl J Med 2005; 353:1332–1341
31. Shapiro N, Howell MD, Bates DW, et al. The association of sepsis
syndrome and organ dysfunction with mortality in emergency department patients with suspected infection. Ann Emerg Med 2006; 48:583–90, 590.e1
32. Schuetz P, Chiappa V, Briel M, et al. Procalcitonin
algorithms for antibiotic therapy decisions: A systematic review of randomized controlled trials and recommendations for clinical algorithms. Arch Intern Med 2011; 171:1322–1331
33. Laukemann S, Kasper N, Kulkarni P, et al. Can we reduce negative blood cultures with clinical scores and blood markers? results from an observational cohort study. Medicine (Baltimore) 2015; 94:e2264
34. Christ-Crain M, Müller B. Calcitonin peptides–the mediators in sepsis
or just another fairy tale? Crit Care Med 2008; 36:1684–1687
35. Becker KL, Snider R, Nylen ES. Procalcitonin
and systemic inflammation: A harmful biomarker
and a therapeutic target. Br J Pharmacol 2010; 159:253–264
36. Bouadma L, Luyt CE, Tubach F, et al; PRORATA trial group: Use of procalcitonin
to reduce patients’ exposure to antibiotics in intensive care units
(PRORATA trial): A multicentre randomised controlled trial. Lancet 2010; 375:463–474
37. de Jong E, van Oers JA, Beishuizen A, et al. Efficacy and safety of procalcitonin
guidance in reducing the duration of antibiotic treatment in critically ill patients: A randomised, controlled, open-label trial. Lancet Infect Dis 2016; 16:819–827
38. Schuetz P, Mueller B. Procalcitonin
in critically ill patients: Time to change guidelines and antibiotic use in practice. Lancet Infect Dis 2016; 16:758–760
39. Schuetz P, Mueller B. To escalate or to de-escalte–that is the question. Crit Care Med 2011; 39:2590; author reply 2591
40. Jensen JU, Heslet L, Jensen TH, et al. Procalcitonin
increase in early identification of critically ill patients at high risk of mortality. Crit Care Med 2006; 34:2596–2602
41. Heyland DK, Johnson AP, Reynolds SC, et al. Procalcitonin
for reduced antibiotic exposure in the critical care setting: A systematic review and an economic evaluation. Crit Care Med 2011; 39:1792–1799