Adherence to a procalcitonin-guided antibiotic treatment protocol in patients with severe sepsis and septic shock

Adherence to a procalcitonin-guided antibiotic treatment protocol in patients with severe sepsis... Background: In randomised controlled trials, procalcitonin (PCT )-guided antibiotic treatment has been proven to significantly reduce length of antibiotic therapy in intensive care unit (ICU) patients. However, concern was raised on low protocol adherence and high rates of overruling, and thus the value of PCT-guided treatment in real clinical life outside study conditions remains unclear. In this study, adherence to a PCT protocol to guide antibiotic treatment in patients with severe sepsis and septic shock was analysed. Methods: From 2012 to 2014, surgical ICU patients with severe sepsis or septic shock were retrospectively screened for PCT measurement series appropriate to make treatment decisions on antibiotic therapy. We compared (1) patients with appropriate PCT measurement series to patients without appropriate series; (2) patients who reached the antibi- otic stopping advice threshold (PCT < 0.5 ng/mL and/or decrease to 10% of peak level) to patients who did not reach a stopping advice threshold; and (3) patients who were treated adherently to the PCT protocol to non-adherently treated patients. The groups were compared in terms of antibiotic treatment duration, PCT kinetics, and other clinical outcomes. Results: Of 81 patients with severe sepsis or septic shock, 14 were excluded due to treatment restriction or short course in the ICU. The final analysis was performed on 67 patients. Forty-two patients (62.7%) had appropriate PCT measurement series. In patients with appropriate PCT series, median initial PCT (p = 0.001) and peak PCT levels (p < 0.001) were significantly higher compared to those with non-appropriate series. In 26 patients with appropriate series, PCT levels reached an antibiotic stopping advice. In 8 of 26 patients with stopping advice, antibiotics were discontinued adherently to the PCT protocol (30.8%). Patients with adherently discontinued antibiotics had a shorter antibiotic treatment (7d [IQR 6–9] vs. 12d [IQR 9–16]; p = 0.002). No differences were seen in terms of other clinical outcomes. Conclusion: In patients with severe sepsis and septic shock, procalcitonin testing was irregular and adherence to a local PCT protocol was low in real clinical life. However, adherently treated patients had a shorter duration of antibi- otic treatment without negative clinical outcomes. Procalcitonin peak values and kinetics had a clear impact on the regularity of PCT testing. Keywords: Sepsis, Procalcitonin, Antibiotic consumption, Protocol adherence *Correspondence: andreas.hohn@uk-koeln.de Department of Anaesthesiology and Intensive Care Medicine, University Hospital of Cologne, Kerpener Str. 62, 50937 Cologne, Germany Full list of author information is available at the end of the article © The Author(s) 2018. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creat iveco mmons .org/licen ses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Hohn et al. Ann. Intensive Care (2018) 8:68 Page 2 of 10 If antibiotics were discontinued when a PCT stopping Background advice was reached, patients were classified as being Length of antibiotic therapy in critically ill patients treated adherently to the PCT protocol. Finally, adher- can be safely guided by PCT-guided protocols. A ently treated patients were compared to non-adherently recent large study and several reviews and meta-anal- treated patients again in terms of duration of antibiotic yses demonstrated that implementation of PCT-guided treatment, PCT kinetics, and other clinical outcomes. protocols leads to a significant reduction in antibiotic exposure of 2–3.5 days [1–8]. Despite this reduction, a recent Cochrane Database review found no beneficial Study design and patients effects on mortality, duration of mechanical ventilation, In this retrospective cohort study, patients admitted to or reinfection [9]. However, the use of PCT to guide the surgical intensive care unit (ICU) between 2012 and treatment decisions is recommended in current guide- 2014 with the diagnosis of severe sepsis or septic shock lines [10, 11]. were included. To define the first episode of sepsis, the Concern was raised on low protocol adherence in pro- time of sepsis diagnosis and the source of infection, data spective trials. The low rate of compliance of PCT-guided were taken from a review of patient records by four of the algorithms and the high rate of exclusion weaken the real authors. Furthermore, data on antibiotic treatment were impact of such protocols in the clinical decision-making analysed in the same way. Information on the following process [12]. However, own data revealed that in clini- variables was available from the hospital information sys- cal routine, duration of antibiotic therapy in septic ICU tem and the patients’ chart: patients decreased after implementation of a PCT proto- col [13], and combination with an antibiotic stewardship • Age, gender and Simplified Acute Physiology Score programme had positive impact on antibiotic use density (SAPS) II at admission to ICU. and the spectrum of antibiotic classes used in the ICU • Maximum sequential organ failure score (SOFA). [14]. Nonetheless, data are scarce regarding adherence • Type of infection. to PCT protocols for guidance of antibiotic treatment. A • Length of stay in ICU and length of stay in hospital. retrospective analysis from seven German intensive care • PCT measurements and results during ICU stay. units (ICU) showed that the use of PCT measurement • Duration of mechanical ventilation. to monitor sepsis treatment is not well established in • Doses of antibiotics and antibiotic use density during clinical routine [15]. The aim of this retrospective cohort ICU stay. study from a surgical ICU was to analyse adherence to • ICU survival status. a PCT-guided antibiotic treatment protocol in patients • Final diagnosis at discharge. with severe sepsis and septic shock. Furthermore, we sought to assess factors affecting protocol adherence, to get a better understanding on the use of PCT in real clin- Description of the local PCT protocol and PCT stopping ical life. advice Procalcitonin-guided antibiotic treatment was intro- duced in our ICU in 2011 [14]. The PCT protocol used Methods from 2012 contains the following rules: In brief, patients with severe sepsis and septic shock were retrospectively identified by a database query from the • Initiation of antibiotic therapy based on clinical deci- hospital information system by their primary diagnosis sion. (i.e. severe sepsis/septic shock). In septic patients, pro- • Daily PCT (day 1–3) measurement in patients under- calcitonin measurement series were analysed to assess going antibiotic treatment started on admission to whether measurement series were appropriate to guide ICU or clinical diagnosis of sepsis, severe sepsis or antibiotic treatment or not. Patients with appropriate septic shock. PCT measurement series were compared to patients • From day 4 on, PCT samples every other day in without appropriate series in terms of antibiotic treat- patients under antibiotic treatment. ment duration, PCT kinetics, and other clinical out- comes. Afterwards, among patients with appropriate Antibiotic stopping advice: If PCT is < 0.50  ng/mL or PCT measurement series, those with PCT stopping PCT decreases to ≤ 10% from peak level, discontinu- advice (PCT < 0.5 ng/mL and/or decrease to 10% of peak ation of antibiotic treatment is recommended. If PCT level) to discontinue antibiotics according to the local is ≥ 0.5  ng/mL and does not decrease to ≤ 10% of peak PCT protocol were compared to patients which did not level, or even increases, discontinuation of antibiotics reach a stop threshold. is only suggested assuming other conditions (unrelated Hohn et al. Ann. Intensive Care (2018) 8:68 Page 3 of 10 to bacterial infections) leading to increased PCT levels Antibiotic therapy and PCT measurements were evalu- (Additional file 1: Figure S1). ated only for the first antibiotic episode of infection. Definition of appropriate PCT measurement series The local PCT protocol recommended daily PCT meas - Definition of adherence to the PCT protocol urements in patients under antibiotic treatment on day According to the implemented PCT algorithm, patients 1–3, and PCT measurements every other day from day 4 with appropriate PCT measurement series were consid- on. However, since it is not mandatory to have daily PCT ered to have been treated adherently if antibiotics were measurements, to guide antibiotic treatment sufficiently, discontinued within 24 h after a PCT decrease to ≤ 10% we decided that it was appropriate if PCT results were from peak level or a PCT value declining to < 0.5  ng/ available at regular intervals (e.g. on day 2, 3, or 4, and mL. The PCT decrease was calculated by the following then on 5, 6, 7, or 8 etc.). formula: A selection of patients with appropriate PCT meas- PCT decrease at day i urement series was then done accordingly to a recently published analysis [15] (Fig. 1). Patients with restriction max PCT before day i − PCT at day i of critical care treatment, or short stay (≤ 1–2  days) in max PCT before day i the ICU, were removed from the analysis beforehand. Fig. 1 Selection algorithm to identify patients with appropriate procalcitonin (PCT ) measurement series according to [15] Hohn et al. Ann. Intensive Care (2018) 8:68 Page 4 of 10 The PCT value at day i was compared to all available (n = 42) compared to those with non-appropriate PCT PCT values at the days before. measurement series (n = 25). Patients with appropri- ate PCT measurement series had statistically significant Group comparisons and statistical analysis longer ventilation times and a longer duration of anti- Comparisons regarding antibiotic treatment duration, biotic treatment in the first episode of sepsis. ICU mor - PCT kinetics, and other clinical outcomes were per- tality was lower in the group with non-appropriate PCT formed for three different groups of patients: series, and the rate of pneumonias was higher in the non- appropriate group. 1. Patients with appropriate PCT measurement series Median initial and peak PCT values of the first septic versus patients without appropriate PCT measure- episode were statistically significant higher in the group ment series. of patients with appropriate PCT measurement series. 2. Patients who reached antibiotic stopping advice ver- sus patients who did not reach antibiotic stopping Factors affecting achievement of PCT stopping advice advice according to the local PCT protocol. threshold 3. For patients who reached antibiotic stopping advice: Table  2 gives the patient characteristics and outcomes Patients treated adherently to the PCT protocol ver- of patients who reached an antibiotic stopping advice sus patients treated non-adherently (i.e. antibiotics threshold according to the PCT protocol (n = 26) com- stopped vs. not stopped when PCT stopping advice pared to those without reaching the stopping advice occurred). threshold (n = 16). Patients who reached the antibiotic stopping advice Variables were analysed by had a statistically significant longer stay in the hospital and a relevant lower ICU mortality. SAPS II and maxi- • Counts and proportions for categorical variables, mum SOFA scores were higher in patients who did not • means and standard deviations and medians and reach the stopping advice. Median initial and peak PCT quartiles for numerical variables. values were higher in patients with stopping advice, but these differences were not statistically significant. Length 95% confidence intervals for proportions were deter - of antibiotic therapy was comparable in both groups. In mined with the exact method of Clopper and Pearson. 24 of 26 cases, patients reached the stopping advice by a Pearson’s χ test was applied for group comparisons of 90% decrease in PCT. A 90% decrease plus PCT < 0.5 ng/ binary variables. The Wilcoxon rank-sum test was applied mL occurred only in 2 of 26 patients. A drop of PCT for comparing two groups, and p values are reported. below < 0.5  ng/mL alone was in none of the cases a rea- Differences with p < 0.05 were considered statistically son for a stopping advice. In patients who did not reach significant. All statistical analyses were performed with a stopping advice in the PCT protocol, a secondary PCT R version 3.1.2 (R Foundation for Statistical Computing, increase occurred in 81.3% and only in 42.3% of patients Vienna, Austria). who reached the stopping advice threshold. Results Between 2012 and 2014, 81 patients with the final diag - Factors affecting adherence to the PCT protocol nosis of severe sepsis and septic shock were identified Table  3 shows the patient characteristics and outcomes from the hospital information system. Of these, 14 were of patients treated adherently to the antibiotic stopping excluded due to treatment restriction or short stay in advice (n = 8) compared to those with non-adherent the ICU (Fig.  2). The final analysis was performed for 67 treatment (n = 18). patients. According to our definition (Fig.  1), 25 patients Both groups were comparable with respect to SAPS (37.3%) did not have appropriate PCT measurement II, maximum SOFA scores, length of hospital and ICU series. Of the remaining 42 patients, 26 patients reached stay, and ventilation hours. The rate of male patients was an antibiotic stopping advice threshold according to the lower in the adherent group, and median length of anti- local PCT protocol. In 8 cases (30.8%) antibiotics were biotic treatment was statistically significant shorter (7 vs. stopped adherently to the PCT protocol. 12 days). In the adherently treated group, 88% per cent had Factors affecting appropriate versus non‑appropriate PCT abdominal sepsis. A trend was seen towards a lower measurement series ICU mortality and higher median initial PCT values in Table  1 shows the patient characteristics and outcomes adherently treated patients. In 75% of adherently treated of patients with appropriate PCT measurement series patients, the initial PCT value and peak value were Hohn et al. Ann. Intensive Care (2018) 8:68 Page 5 of 10 Fig. 2 Flow diagram of septic ICU patients and distribution to different subgroups of procalcitonin testing and protocol adherence. pt(s) patient(s) concordant, whereas in non-adherent patients only in In only 8 out of 81 patients with severe sepsis and sep- 44% the initial and peak values of PCT were concordant. tic shock during the entire study period, or in 67 patients who entered the final statistical analysis, the algorithm Discussion was fully applied. However, in patients with a low PCT Factors affecting adherence to the PCT protocol (< 0.5 ng/mL) or a marked decrease of PCT levels (≤ 10% Currently, there are only limited data on PCT protocol of peak level), a shorter antibiotic treatment was not adherence in critically ill patients outside of study con- associated with worse clinical outcomes. Thus, our study ditions. One study in medical ICU patients revealed confirms the results of a recent large study [1] and several an adherence below 50% to a local PCT protocol [16]. reviews and meta-analyses [2–6]. Procalcitonin is able to However, this study did not compare adherently treated support clinical decision-making to discontinue antibiot- patients to non-adherently treated patients. Thus, our ics when PCT decrease goes along with clinical improve- study reveals important insights on whether and how ment. However, in our study, antibiotics were continued the promising strategy of PCT-guidance is transferred to in nearly 70% of patients although antibiotic discontinu- clinical routine practice. ation was recommended by the PCT protocol. Algo- We detected a low adherence to discontinue antibiotics rithm overruling was also of concern in prospective trials adherently to a PCT stopping advice in clinical real life. where overruling rates reached from 16 to 53% [17–19]. Hohn et al. Ann. Intensive Care (2018) 8:68 Page 6 of 10 Table 1 Demographics and  clinical data of  patients with  appropriate and  non-appropriate procalcitonin (PCT) measurement series Patients with appropriate PCT Patients with non‑appropriate PCT p value series (n = 42) series (n = 25) Age, years (median [IQR]) 68 [59–76] 76 [68–79] 0.172 SAPS II score (median [IQR]) 40 [29–51] 38 [32–47] 0.807 Male gender (% [95% CI]) 59.5 [43.3–74.4] 60.0 [38.7–78.9] 0.999 ICU mortality (% [95% CI]) 40.5 [26–57] 32.0 [14.9–53.5] 0.604 Ventilation hours (median [IQR]) 423 [68–840] 102 [8–335] 0.012 ICU LOS, days (median [IQR]) 17 [5–41] 11 [4–21] 0.264 Hospital LOS, days (median [IQR]) 42 [25–73] 39 [25–54] 0.351 SOFA score max (median [IQR]) 11 [9–13] 9 [8–11] 0.112 Length of antibiotic treatment, days (median [IQR]) 10 [7–13] 5 [4–6] < 0.001 Antibiotic use density, DDD (median [IQR]) 11 [9–23] 6 [4–9] < 0.001 Initial PCT value, ng/mL (median [IQR]) 9.15 [3.1–32.4] 1.55 [0.9–4.6] 0.001 Peak PCT value, ng/mL (median [IQR]) 30.2 [7.1–60.7] 3.8 [1.6–10.3] < 0.001 % [95% CI] (n) % [95% CI] (n) Cause of infection Abdominal sepsis 52.4 [36.4–67.9] (22) 44.0 [24.4–65.1] (11) Bloodstream infection 0 4.0 [0.1–20.3] (1) Pleural empyema 9.5 [2.6–22.6] (4) 0 Pneumonia 19.0 [8.6–34.1] (8) 32.0 [14.9–53.5] (8) Prosthetic infection 0 8.0 [1.0–26.0] (2) Urogenital sepsis 9.5 [2.6–22.6] (4) 8.0 [1.0–26.0] (2) Soft tissue infection 2.4 [0.6–12.6] (1) 4.0 [0.1–20.3] (1) Unknown 7.1 [1–19] (3) 0 Expectably, overruling rates were higher in real life since in adherently treated patients. In 75% of the adherently the PCT protocol was not applied under controlled treated patients, the initial PCT corresponded to the study conditions and populations are not comparable peak value. Contrasting to that, in 56% of the not adher- due to strict in- and exclusion criteria in randomised ently treated patients the PCT value rises after the initial trials. Exclusion rates in randomised PCT studies range value. For clinicians, high initial PCT levels seem to be from 0.3% [20] to 84% [21], thus not necessarily reflect - no barrier to stop antibiotics according to the PCT pro- ing clinical real life. For example, patients with common tocol, but secondarily increasing PCT seems to prompt ICU infections, e.g. caused by Pseudomonas aeruginosa clinicians to continue antibiotics. Similarly, to the study and Acinetobacter baumannii, which were excluded in by Jensen [20], in clinical real life, secondary increases several studies [17–19] were not excluded in our study. seem to be interpreted as “alert PCT”. For most patients, Furthermore, probably we cannot exclude that inappro- a decline of PCT to ≤ 10% of peak value was a trigger for priate septic source control contributed to low adherence an antibiotic stopping advice and not the absolute limit of to antibiotic stopping advices in our study, and we did PCT < 0.5 ng/mL. Thus, in patients with severe sepsis and not analyse the impact of multi drug resistant pathogens septic shock, relative variations of PCT seem to be more on antibiotic treatment duration. important to the clinician than absolute values. Further- Nonetheless, the low rate of less than 10% of sep- more, in patients with secondary PCT increases, the pro- tic patients in which discontinuation of antibiotics was calcitonin protocol was more frequently overruled and accompanied by a PCT stop signal challenges the clini- antibiotic treatment continued despite reaching a stop- cal relevance of PCT-guided antibiotic stewardship and ping advice. efforts should be made to improve adherence to PCT Our data suggest that certain types of infection like protocols in clinical practice. pneumonia or abdominal sepsis might affect adherence Kinetics of procalcitonin also affect adherence to dis - to PCT protocols as well. In patients who reached a stop- continue antibiotics when a stopping advice thresh- ping advice, those with abdominal sepsis were treated old is reached. Median initial PCT levels were higher adherently in 43% cases (7/15), whereas patients with Hohn et al. Ann. Intensive Care (2018) 8:68 Page 7 of 10 Table 2 Demographics and clinical data of patients with and without procalcitonin (PCT) stopping advice Patients with PCT stopping Patients without PCT stopping p value advice (n = 26) advice (n = 16) Age, years (median [IQR]) 68 [59–75] 70 [62–79] 0.475 SAPS II score (median [IQR]) 40 [29–51] 44 [33–52] 0.597 Male gender (% [95% CI]) 69.2 [48.2–85.7] 43.8 [19.8–70.1] 0.121 ICU mortality (% [95% CI]) 26.9 [11.6–47.8] 62.5 [35.4–84.8] 0.029 Ventilation hours (median [IQR]) 597 [96–1260] 238 [47–461] 0.068 ICU LOS, days (median [IQR]) 23 [5–56] 11 [5–20] 0.212 Hospital LOS, days (median [IQR]) 56 [35–100] 23 [17–44] 0.002 SOFA score max (median [IQR]) 11 [9–12] 13 [9–15] 0.320 Length of antibiotic treatment, days (median [IQR]) 10 [8–13] 10 [7–12] 0.398 Antibiotic use density, DDD (median [IQR]) 13 [10–24] 10 [8–15] 0.136 Initial PCT value, ng/mL (median [IQR]) 20.0 [3.3–42.7] 5.1 [2.8–10.8] 0.136 Peak PCT value, ng/mL (median [IQR]) 40.5 [9.2–69.7] 16.2 [6.0–32.5] 0.095 % [95% CI] (n) % [95% CI] (n) Cause of infection Abdominal sepsis 57.7 [36.9,76.6] (15) 43.8 [19.8–70.1] (7) Bloodstream infection 0 0 Pleural empyema 11.5 [2.4–30.2] (3) 6.3 [0.2–30.2] (1) Pneumonia 26.9 [11.5–47.8] (7) 6.3 [0.2–30.2] (1) Prosthetic infection 0 0 Urogenital sepsis 3.8 [0.1–19.6] (1) 18.8 [4.0–45.6] (3) Soft tissue infection 0 6.3 [0.2–30.2] (1) Unknown 0 18.8 [4.0–45.6] (3) pneumonia were treated adherently only in 14% (1/7). that patients without appropriate PCT measurement However, due to the small number of patients with other series were in a better clinical condition, and probably infection sites than abdomen or lung, this could not be regularly PCT testing was not considered necessary by proven statistically in this study. Nonetheless, these the clinician. In patients with inappropriate PCT series, results are interesting since mortality in critically ill sep- median duration of antibiotic treatment was 5 days, and tic patients with pneumonia or abdominal sepsis are thus more regular PCT testing likely would not have comparable [22, 23]. had beneficial impact on further reduction in antibiotic exposure. Factors affecting appropriate versus non‑appropriate PCT measurement series Factors affecting achievement of PCT stopping advice A prerequisite to PCT-guided antibiotic stewardship is threshold regular measurements. In our study, 62.7% had appropri- Beside its value for biomarker-guided antibiotic steward- ate PCT measurement series. Median initial PCT values ship, PCT can be used as a prognostic value. Patients who and peak values were statistically significantly higher in did not reach the PCT stopping advice threshold had the group with appropriate measurement series. These higher mortality rates (63%) compared to those where results suggest that frequency of PCT testing is affected PCT declined indicating that the infection is under con- by PCT levels, and high PCT values might prompt cli- trol (27%). Median initial PCT levels and peak levels were nicians to request PCT testing more frequently. In the higher in patients with stopping advice. However, these group of patients with non-appropriate PCT series, there results were statistically not significant. Interestingly, in was a higher number of patients with pneumonia. It is 81.3% (13/16) of patients without reaching the PCT stop- well known that peak values of PCT are lower in res- ping advice, PCT levels increased after the initial value. piratory tract infections compared to other kind of infec- In patients with stopping advice, this occurred only in tions [24–26]. Lower duration of mechanical ventilation, 42.3% (11/26). These results are in line with previous shorter stay in the ICU, and lower mortality rates reflect studies which showed that high PCT levels are associated Hohn et al. Ann. Intensive Care (2018) 8:68 Page 8 of 10 Table 3 Demographics and  clinical data of  patients with  procalcitonin (PCT) stopping advice treated adherently and non-adherently to the PCT protocol Patients adherent to PCT Patients non‑adherent to PCT p value stopping advice (n = 8) stopping advice (n = 18) Age, years (median [IQR]) 67 [59–75] 68 [59–75] 0.560 SAPS II score (median [IQR]) 35 [31–43] 44 [28–51] 0.655 Male gender (% [95% CI]) 50.0 [15.7–84.3] 77.8 [30.6–69.4] 0.197 ICU mortality (% [95% CI]) 12.5 [0.3–52.7] 33.3 [13.3–59.0] 0.375 Ventilation hours (median [IQR]) 353 [68–621] 735 [209–1508] 0.317 ICU LOS, days (median [IQR]) 24 [3–38] 23 [12–59] 0.780 Hospital LOS, days (median [IQR]) 57 [47–79] 54 [34–100] 0.824 SOFA score max (median [IQR]) 11 [9–12] 12 [9–13] 0.557 Length of antibiotic treatment, days (median [IQR]) 7 [6–9] 12 [9–16] 0.002 Antibiotic use density, DDD (median [IQR]) 10 [7–12] 18 [12–26] 0.001 Initial PCT value, ng/mL (median [IQR]) 20.0 [9.4–38.2] 16.8 [2.0–42.7] 0.560 Peak PCT value, ng/mL (median [IQR]) 32.3 [9.4–64.2] 44.9 [13.9–69.7] 0.718 % [95% CI] (n) % [95% CI] (n) Cause of infection Abdominal sepsis 87.5 [47.3–99.7] (7) 44.4 [21.5–69.2] (8) Bloodstream infection 0 0 Pleural empyema 0 16.7 [3.5–41.4] (3) Pneumonia 12.5 [0.3–52.7] (1) 33.3 [13.3–59.0] (6) Prosthetic infection 0 0 Urogenital sepsis 0 5.6 [0.1–27] (1) Soft tissue infection 0 0 Unknown 0 0 with an unfavourable outcome [27, 28]. Furthermore, a to guide antibiotic treatment [15]. Furthermore, we can not adequately declining PCT is an independent predic- only assess correlation and cannot comment on causality, tor of mortality in septic patients [29, 30]. In our study, and thus, probably additional factors may have contrib- the shorter ICU and hospital lengths of stay in patients uted to the results. without stopping advice might be founded in the fact that Furthermore, we can only comment on the first episode those patients had a higher mortality. of sepsis and did not analyse the impact of PCT guidance on further septic episodes during the stay in ICU. As our data were collected retrospectively, diagnosis of sepsis Limitations was not yet according to the new sepsis definition [31], There are limitations of the present study which should and thus we may have missed patients because one in be mentioned. Taken the overall low sample size together eight patients admitted to intensive care units with infec- with the retrospective study design, all results and their tion and new organ failure does not meet the condition interpretation should be taken with caution. As possibly of at least two systemic inflammatory response syndrome not all septic cases might have been registered correctly criteria. in the hospital’s database, we cannot exclude a selec- tion bias in our analysis. The selection of patients was Conclusion also possibly affected by the definition of an appropriate Our study showed that a PCT protocol works outside measurement series. However, today, there is no universal clinical studies, but these benefits along with interna - definition how often and in what intervals PCT should be tional guideline recommendations have not yet entered measured to be sufficient to guide antibiotic treatment. the daily routine in intensive care medicine. In patients In addition, there are little data for the frequency of PCT with severe sepsis and septic shock, PCT testing was testing and adherence to PCT protocols in clinical real irregular and adherence to a PCT protocol for discon- life. In our study, we used a recently published definition tinuation of antibiotic treatment was low, but when to classify a series of PCT measurements as appropriate applied it was associated with a considerably shorter Hohn et al. Ann. Intensive Care (2018) 8:68 Page 9 of 10 Funding duration of antibiotic treatment and a lower antibiotic Data were generated and analysed as part of the routine work, and no other consumption for the first septic episode without nega - kind of funding has been received. tive clinical outcomes. Severity of illness, PCT peak levels, and PCT kinetics, respectively, might affect Publisher’s Note frequency of PCT testing and protocol adherence. Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. Patients at high risk without sufficiently decreasing or even increasing PCT levels seem to prompt clinicians Received: 6 March 2018 Accepted: 28 May 2018 to continue antibiotics and to overrule PCT protocols despite reaching antibiotic stopping advice thresholds. Additional file References 1. de Jong E, van Oers JA, Beishuizen A, Vos P, Vermeijden WJ, Haas LE et al. Efficacy and safety of procalcitonin guidance in reducing the duration Additional file 1: Figure S1. Local PCT protocol. Start of antibiotic treat - of antibiotic treatment in critically ill patients: a randomised, controlled, ment is based on a clinical decision. According to this algorithm, antibiot- open-label trial. Lancet Infect Dis. 2016;16(7):819–27. ics should be discontinued when clinical improvement goes along with 2. Matthaiou DK, Ntani G, Kontogiorgi M, Poulakou G, Armaganidis A, Dimo- decreasing PCT levels. The protocol can be overruled by the means of the poulos G. An ESICM systematic review and meta-analysis of procalci- attending physician due to clinical reasons or when conditions are pre- tonin-guided antibiotic therapy algorithms in adult critically ill patients. sent which require a prolonged antibiotic treatment. Daily PCT samples Intensive Care Med. 2012;38(6):940–9. were recommended on day 1–3 in patients on antibiotic treatment since 3. Prkno A, Wacker C, Brunkhorst FM, Schlattmann P. Procalcitonin-guided admission to ICU or since clinical suspicion of systemic bacterial infection. therapy in intensive care unit patients with severe sepsis and septic From day 4 on, PCT samples were recommended every other day in shock—a systematic review and meta-analysis. Crit Care. 2013;17(6):R291. patients under antibiotic treatment. 4. Kopterides P, Siempos II, Tsangaris I, Tsantes A, Armaganidis A. Procalci- tonin-guided algorithms of antibiotic therapy in the intensive care unit: a systematic review and meta-analysis of randomized controlled trials. Crit Abbreviations Care Med. 2010;38(11):2229–41. CI: Confidence interval; ICU: Intensive care unit; pt(s): Patient(s); PCT: Procalci- 5. Tang H, Huang T, Jing J, Shen H, Cui W. Eec ff t of procalcitonin-guided tonin; SAPS: Simplified acute physiology score; SOFA: Sequential organ failure treatment in patients with infections: a systematic review and meta- score. analysis. Infection. 2009;37(6):497–507. 6. Iankova I, Thompson-Leduc P, Kirson NY, Rice B, Hey J, Krause A et al. Effi- Authors’ contributions cacy and safety of procalcitonin guidance in patients with suspected or AH, NB, BH and SS were responsible for the conception, study design, study confirmed sepsis: a systematic review and meta-analysis. Crit Care Med. execution, data management, data analysis, data interpretation, and manu- 2018;46(5):691–8. script writing. SH and JCW were responsible for conception, study design, 7. Huang HB, Peng JM, Weng L, Wang CY, Jiang W, Du B. Procalcitonin- data management, data analysis, data interpretation, and critically revising guided antibiotic therapy in intensive care unit patients: a systematic the manuscript. MH was responsible for data analysis, data interpretation and review and meta-analysis. Ann Intensive Care. 2017;7(1):114. manuscript writing. All authors read and approved the final manuscript. 8. Brechot N, Hekimian G, Chastre J, Luyt CE. Procalcitonin to guide antibi- otic therapy in the ICU. Int J Antimicrob Agents. 2015;46(Suppl 1):S19–24. Author details 9. Andriolo BN, Andriolo RB, Salomao R, Atallah AN. Eec ff tiveness and Department of Anaesthesiology and Intensive Care Medicine, University safety of procalcitonin evaluation for reducing mortality in adults with Hospital of Cologne, Kerpener Str. 62, 50937 Cologne, Germany. Department sepsis, severe sepsis or septic shock. Cochrane Database Syst Rev. of Infectiology and Hospital Hygiene, Hospital Düren gem. GmbH, Roonstraße 2017;1:Cd010959. 30, 52351 Düren, Germany. Thermo Fisher Scientific, Thermo Scientific 10. Rhodes A, Evans LE, Alhazzani W, Levy MM, Antonelli M, Ferrer R, et al. Biomarkers, Neuendorfstr. 25, 16761 Hennigsdorf, Germany. Department Surviving sepsis campaign: international guidelines for management of of Anaesthesiology, University Hospital Heidelberg, Im Neuenheimer Feld sepsis and septic shock: 2016. Intensive Care Med. 2017;43(3):304–77. 110, 69120 Heidelberg, Germany. Department of Anaesthesiology, Surgical 11. de With K, Allerberger F, Amann S, Apfalter P, Brodt HR, Eckmanns T, Intensive Care, Emergency Medicine and Pain Management, Hospital Düren et al. Strategies to enhance rational use of antibiotics in hospital: a gem. GmbH, Roonstraße 30, 52351 Düren, Germany. guideline by the German society for infectious diseases. Infection. 2016;44(3):395–439. Acknowledgements 12. Povoa P, Salluh JI. Biomarker-guided antibiotic therapy in adult critically ill None. patients: a critical review. Ann Intensive Care. 2012;2(1):32. 13. Hohn A, Schroeder S, Gehrt A, Bernhardt K, Bein B, Wegscheider K, et al. Competing interests Procalcitonin-guided algorithm to reduce length of antibiotic therapy in SH and JW as employees of Thermo Fisher Scientific performed the statistical patients with severe sepsis and septic shock. BMC Infect Dis. 2013;13:158. analysis. SS served as consultant and has received payments from BRAHMS AG 14. Hohn A, Heising B, Hertel S, Baumgarten G, Hochreiter M, Schroeder S. for speaking engagements. All other authors declare no conflict of interest. Antibiotic consumption after implementation of a procalcitonin-guided antimicrobial stewardship programme in surgical patients admitted to Availability of data and materials an intensive care unit: a retrospective before-and-after analysis. Infection. The datasets used and/or analysed during the current study are available from 2015;43(4):405–12. the corresponding author on reasonable request. 15. Bodmann KF, Schenker M, Heinlein W, Wilke MH. Procalcitonin as a tool for the assessment of successful therapy of severe sepsis: an analysis Consent for publication using clinical routine data. Med Klin Intensivmed Notfmed. 2016. https :// Not applicable. doi.org/10.1007/s0006 3-016-0183-7. 16. Ammar AA, Lam SW, Duggal A, Neuner EA, Bass SN, Guzman JA, et al. Ethics approval and consent to participate Compliance with procalcitonin algorithm antibiotic recommenda- Ethics approval was waived by the Ethics Committee of the State Chamber of tions for patients in medical intensive care unit. Pharmacotherapy. Physicians of North Rhine (Germany) due to the retrospective nature of the 2017;37(2):177–86. study (Ethics Committee No 136-2015). Hohn et al. Ann. Intensive Care (2018) 8:68 Page 10 of 10 17. Bouadma L, Luyt CE, Tubach F, Cracco C, Alvarez A, Schwebel C, et al. Use 25. Hedlund J, Hansson LO. Procalcitonin and C-reactive protein levels in of procalcitonin to reduce patients’ exposure to antibiotics in intensive community-acquired pneumonia: correlation with etiology and progno- care units (PRORATA trial): a multicentre randomised controlled trial. sis. Infection. 2000;28(2):68–73. Lancet. 2010;375(9713):463–74. 26. Luyt CE, Guerin V, Combes A, Trouillet JL, Ayed SB, Bernard M, et al. 18. Nobre V, Harbarth S, Graf JD, Rohner P, Pugin J. Use of procalcitonin to Procalcitonin kinetics as a prognostic marker of ventilator-associated shorten antibiotic treatment duration in septic patients: a randomized pneumonia. Am J Respir Crit Care Med. 2005;171(1):48–53. trial. Am J Respir Crit Care Med. 2008;177(5):498–505. 27. Klingele M, Bomberg H, Schuster S, Schafers HJ, Groesdonk HV. Prog- 19. Stolz D, Smyrnios N, Eggimann P, Pargger H, Thakkar N, Siegemund M, nostic value of procalcitonin in patients after elective cardiac surgery: a et al. Procalcitonin for reduced antibiotic exposure in ventilator-associ- prospective cohort study. Ann Intensive Care. 2016;6(1):116. ated pneumonia: a randomised study. Eur Respir J. 2009;34(6):1364–75. 28. Peschanski N, Chenevier-Gobeaux C, Mzabi L, Lucas R, Ouahabi S, Aqui- 20. Jensen JU, Hein L, Lundgren B, Bestle MH, Mohr TT, Andersen MH, et al. lina V, et al. Prognostic value of PCT in septic emergency patients. Ann Procalcitonin-guided interventions against infections to increase early Intensive Care. 2016;6(1):47. appropriate antibiotics and improve survival in the intensive care unit: a 29. Harbarth S, Holeckova K, Froidevaux C, Pittet D, Ricou B, Grau GE, et al. randomized trial. Crit Care Med. 2011;39(9):2048–58. Diagnostic value of procalcitonin, interleukin-6, and interleukin-8 in 21. Svoboda P, Kantorova I, Scheer P, Radvanova J, Radvan M. Can procalci- critically ill patients admitted with suspected sepsis. Am J Respir Crit Care tonin help us in timing of re-intervention in septic patients after multiple Med. 2001;164(3):396–402. trauma or major surgery? Hepatogastroenterology. 2007;54(74):359–63. 30. Schuetz P, Birkhahn R, Sherwin R, Jones AE, Singer A, Kline JA, et al. Serial 22. Zahar JR, Timsit JF, Garrouste-Orgeas M, Francais A, Vesin A, Descorps- procalcitonin predicts mortality in severe sepsis patients: results from Declere A, et al. Outcomes in severe sepsis and patients with septic the multicenter procalcitonin monitoring sepsis (MOSES) study. Crit Care shock: pathogen species and infection sites are not associated with Med. 2017;45(5):781–9. mortality. Crit Care Med. 2011;39(8):1886–95. 31. Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, 23. Leligdowicz A, Dodek PM, Norena M, Wong H, Kumar A, Kumar A. Asso- Bauer M, et al. The third international consensus definitions for sepsis and ciation between source of infection and hospital mortality in patients septic shock (sepsis-3). JAMA. 2016;315(8):801–10. who have septic shock. Am J Respir Crit Care Med. 2014;189(10):1204–13. 24. Brunkhorst FM, Al-Nawas B, Krummenauer F, Forycki ZF, Shah PM. Procalcitonin, C-reactive protein and APACHE II score for risk evaluation in patients with severe pneumonia. Clin Microbiol Infect. 2002;8(2):93–100. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Annals of Intensive Care Springer Journals

Adherence to a procalcitonin-guided antibiotic treatment protocol in patients with severe sepsis and septic shock

Free
10 pages

Loading next page...
 
/lp/springer_journal/adherence-to-a-procalcitonin-guided-antibiotic-treatment-protocol-in-BCSoIEcHCS
Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s)
Subject
Medicine & Public Health; Intensive / Critical Care Medicine; Emergency Medicine; Anesthesiology
eISSN
2110-5820
D.O.I.
10.1186/s13613-018-0415-5
Publisher site
See Article on Publisher Site

Abstract

Background: In randomised controlled trials, procalcitonin (PCT )-guided antibiotic treatment has been proven to significantly reduce length of antibiotic therapy in intensive care unit (ICU) patients. However, concern was raised on low protocol adherence and high rates of overruling, and thus the value of PCT-guided treatment in real clinical life outside study conditions remains unclear. In this study, adherence to a PCT protocol to guide antibiotic treatment in patients with severe sepsis and septic shock was analysed. Methods: From 2012 to 2014, surgical ICU patients with severe sepsis or septic shock were retrospectively screened for PCT measurement series appropriate to make treatment decisions on antibiotic therapy. We compared (1) patients with appropriate PCT measurement series to patients without appropriate series; (2) patients who reached the antibi- otic stopping advice threshold (PCT < 0.5 ng/mL and/or decrease to 10% of peak level) to patients who did not reach a stopping advice threshold; and (3) patients who were treated adherently to the PCT protocol to non-adherently treated patients. The groups were compared in terms of antibiotic treatment duration, PCT kinetics, and other clinical outcomes. Results: Of 81 patients with severe sepsis or septic shock, 14 were excluded due to treatment restriction or short course in the ICU. The final analysis was performed on 67 patients. Forty-two patients (62.7%) had appropriate PCT measurement series. In patients with appropriate PCT series, median initial PCT (p = 0.001) and peak PCT levels (p < 0.001) were significantly higher compared to those with non-appropriate series. In 26 patients with appropriate series, PCT levels reached an antibiotic stopping advice. In 8 of 26 patients with stopping advice, antibiotics were discontinued adherently to the PCT protocol (30.8%). Patients with adherently discontinued antibiotics had a shorter antibiotic treatment (7d [IQR 6–9] vs. 12d [IQR 9–16]; p = 0.002). No differences were seen in terms of other clinical outcomes. Conclusion: In patients with severe sepsis and septic shock, procalcitonin testing was irregular and adherence to a local PCT protocol was low in real clinical life. However, adherently treated patients had a shorter duration of antibi- otic treatment without negative clinical outcomes. Procalcitonin peak values and kinetics had a clear impact on the regularity of PCT testing. Keywords: Sepsis, Procalcitonin, Antibiotic consumption, Protocol adherence *Correspondence: andreas.hohn@uk-koeln.de Department of Anaesthesiology and Intensive Care Medicine, University Hospital of Cologne, Kerpener Str. 62, 50937 Cologne, Germany Full list of author information is available at the end of the article © The Author(s) 2018. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creat iveco mmons .org/licen ses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Hohn et al. Ann. Intensive Care (2018) 8:68 Page 2 of 10 If antibiotics were discontinued when a PCT stopping Background advice was reached, patients were classified as being Length of antibiotic therapy in critically ill patients treated adherently to the PCT protocol. Finally, adher- can be safely guided by PCT-guided protocols. A ently treated patients were compared to non-adherently recent large study and several reviews and meta-anal- treated patients again in terms of duration of antibiotic yses demonstrated that implementation of PCT-guided treatment, PCT kinetics, and other clinical outcomes. protocols leads to a significant reduction in antibiotic exposure of 2–3.5 days [1–8]. Despite this reduction, a recent Cochrane Database review found no beneficial Study design and patients effects on mortality, duration of mechanical ventilation, In this retrospective cohort study, patients admitted to or reinfection [9]. However, the use of PCT to guide the surgical intensive care unit (ICU) between 2012 and treatment decisions is recommended in current guide- 2014 with the diagnosis of severe sepsis or septic shock lines [10, 11]. were included. To define the first episode of sepsis, the Concern was raised on low protocol adherence in pro- time of sepsis diagnosis and the source of infection, data spective trials. The low rate of compliance of PCT-guided were taken from a review of patient records by four of the algorithms and the high rate of exclusion weaken the real authors. Furthermore, data on antibiotic treatment were impact of such protocols in the clinical decision-making analysed in the same way. Information on the following process [12]. However, own data revealed that in clini- variables was available from the hospital information sys- cal routine, duration of antibiotic therapy in septic ICU tem and the patients’ chart: patients decreased after implementation of a PCT proto- col [13], and combination with an antibiotic stewardship • Age, gender and Simplified Acute Physiology Score programme had positive impact on antibiotic use density (SAPS) II at admission to ICU. and the spectrum of antibiotic classes used in the ICU • Maximum sequential organ failure score (SOFA). [14]. Nonetheless, data are scarce regarding adherence • Type of infection. to PCT protocols for guidance of antibiotic treatment. A • Length of stay in ICU and length of stay in hospital. retrospective analysis from seven German intensive care • PCT measurements and results during ICU stay. units (ICU) showed that the use of PCT measurement • Duration of mechanical ventilation. to monitor sepsis treatment is not well established in • Doses of antibiotics and antibiotic use density during clinical routine [15]. The aim of this retrospective cohort ICU stay. study from a surgical ICU was to analyse adherence to • ICU survival status. a PCT-guided antibiotic treatment protocol in patients • Final diagnosis at discharge. with severe sepsis and septic shock. Furthermore, we sought to assess factors affecting protocol adherence, to get a better understanding on the use of PCT in real clin- Description of the local PCT protocol and PCT stopping ical life. advice Procalcitonin-guided antibiotic treatment was intro- duced in our ICU in 2011 [14]. The PCT protocol used Methods from 2012 contains the following rules: In brief, patients with severe sepsis and septic shock were retrospectively identified by a database query from the • Initiation of antibiotic therapy based on clinical deci- hospital information system by their primary diagnosis sion. (i.e. severe sepsis/septic shock). In septic patients, pro- • Daily PCT (day 1–3) measurement in patients under- calcitonin measurement series were analysed to assess going antibiotic treatment started on admission to whether measurement series were appropriate to guide ICU or clinical diagnosis of sepsis, severe sepsis or antibiotic treatment or not. Patients with appropriate septic shock. PCT measurement series were compared to patients • From day 4 on, PCT samples every other day in without appropriate series in terms of antibiotic treat- patients under antibiotic treatment. ment duration, PCT kinetics, and other clinical out- comes. Afterwards, among patients with appropriate Antibiotic stopping advice: If PCT is < 0.50  ng/mL or PCT measurement series, those with PCT stopping PCT decreases to ≤ 10% from peak level, discontinu- advice (PCT < 0.5 ng/mL and/or decrease to 10% of peak ation of antibiotic treatment is recommended. If PCT level) to discontinue antibiotics according to the local is ≥ 0.5  ng/mL and does not decrease to ≤ 10% of peak PCT protocol were compared to patients which did not level, or even increases, discontinuation of antibiotics reach a stop threshold. is only suggested assuming other conditions (unrelated Hohn et al. Ann. Intensive Care (2018) 8:68 Page 3 of 10 to bacterial infections) leading to increased PCT levels Antibiotic therapy and PCT measurements were evalu- (Additional file 1: Figure S1). ated only for the first antibiotic episode of infection. Definition of appropriate PCT measurement series The local PCT protocol recommended daily PCT meas - Definition of adherence to the PCT protocol urements in patients under antibiotic treatment on day According to the implemented PCT algorithm, patients 1–3, and PCT measurements every other day from day 4 with appropriate PCT measurement series were consid- on. However, since it is not mandatory to have daily PCT ered to have been treated adherently if antibiotics were measurements, to guide antibiotic treatment sufficiently, discontinued within 24 h after a PCT decrease to ≤ 10% we decided that it was appropriate if PCT results were from peak level or a PCT value declining to < 0.5  ng/ available at regular intervals (e.g. on day 2, 3, or 4, and mL. The PCT decrease was calculated by the following then on 5, 6, 7, or 8 etc.). formula: A selection of patients with appropriate PCT meas- PCT decrease at day i urement series was then done accordingly to a recently published analysis [15] (Fig. 1). Patients with restriction max PCT before day i − PCT at day i of critical care treatment, or short stay (≤ 1–2  days) in max PCT before day i the ICU, were removed from the analysis beforehand. Fig. 1 Selection algorithm to identify patients with appropriate procalcitonin (PCT ) measurement series according to [15] Hohn et al. Ann. Intensive Care (2018) 8:68 Page 4 of 10 The PCT value at day i was compared to all available (n = 42) compared to those with non-appropriate PCT PCT values at the days before. measurement series (n = 25). Patients with appropri- ate PCT measurement series had statistically significant Group comparisons and statistical analysis longer ventilation times and a longer duration of anti- Comparisons regarding antibiotic treatment duration, biotic treatment in the first episode of sepsis. ICU mor - PCT kinetics, and other clinical outcomes were per- tality was lower in the group with non-appropriate PCT formed for three different groups of patients: series, and the rate of pneumonias was higher in the non- appropriate group. 1. Patients with appropriate PCT measurement series Median initial and peak PCT values of the first septic versus patients without appropriate PCT measure- episode were statistically significant higher in the group ment series. of patients with appropriate PCT measurement series. 2. Patients who reached antibiotic stopping advice ver- sus patients who did not reach antibiotic stopping Factors affecting achievement of PCT stopping advice advice according to the local PCT protocol. threshold 3. For patients who reached antibiotic stopping advice: Table  2 gives the patient characteristics and outcomes Patients treated adherently to the PCT protocol ver- of patients who reached an antibiotic stopping advice sus patients treated non-adherently (i.e. antibiotics threshold according to the PCT protocol (n = 26) com- stopped vs. not stopped when PCT stopping advice pared to those without reaching the stopping advice occurred). threshold (n = 16). Patients who reached the antibiotic stopping advice Variables were analysed by had a statistically significant longer stay in the hospital and a relevant lower ICU mortality. SAPS II and maxi- • Counts and proportions for categorical variables, mum SOFA scores were higher in patients who did not • means and standard deviations and medians and reach the stopping advice. Median initial and peak PCT quartiles for numerical variables. values were higher in patients with stopping advice, but these differences were not statistically significant. Length 95% confidence intervals for proportions were deter - of antibiotic therapy was comparable in both groups. In mined with the exact method of Clopper and Pearson. 24 of 26 cases, patients reached the stopping advice by a Pearson’s χ test was applied for group comparisons of 90% decrease in PCT. A 90% decrease plus PCT < 0.5 ng/ binary variables. The Wilcoxon rank-sum test was applied mL occurred only in 2 of 26 patients. A drop of PCT for comparing two groups, and p values are reported. below < 0.5  ng/mL alone was in none of the cases a rea- Differences with p < 0.05 were considered statistically son for a stopping advice. In patients who did not reach significant. All statistical analyses were performed with a stopping advice in the PCT protocol, a secondary PCT R version 3.1.2 (R Foundation for Statistical Computing, increase occurred in 81.3% and only in 42.3% of patients Vienna, Austria). who reached the stopping advice threshold. Results Between 2012 and 2014, 81 patients with the final diag - Factors affecting adherence to the PCT protocol nosis of severe sepsis and septic shock were identified Table  3 shows the patient characteristics and outcomes from the hospital information system. Of these, 14 were of patients treated adherently to the antibiotic stopping excluded due to treatment restriction or short stay in advice (n = 8) compared to those with non-adherent the ICU (Fig.  2). The final analysis was performed for 67 treatment (n = 18). patients. According to our definition (Fig.  1), 25 patients Both groups were comparable with respect to SAPS (37.3%) did not have appropriate PCT measurement II, maximum SOFA scores, length of hospital and ICU series. Of the remaining 42 patients, 26 patients reached stay, and ventilation hours. The rate of male patients was an antibiotic stopping advice threshold according to the lower in the adherent group, and median length of anti- local PCT protocol. In 8 cases (30.8%) antibiotics were biotic treatment was statistically significant shorter (7 vs. stopped adherently to the PCT protocol. 12 days). In the adherently treated group, 88% per cent had Factors affecting appropriate versus non‑appropriate PCT abdominal sepsis. A trend was seen towards a lower measurement series ICU mortality and higher median initial PCT values in Table  1 shows the patient characteristics and outcomes adherently treated patients. In 75% of adherently treated of patients with appropriate PCT measurement series patients, the initial PCT value and peak value were Hohn et al. Ann. Intensive Care (2018) 8:68 Page 5 of 10 Fig. 2 Flow diagram of septic ICU patients and distribution to different subgroups of procalcitonin testing and protocol adherence. pt(s) patient(s) concordant, whereas in non-adherent patients only in In only 8 out of 81 patients with severe sepsis and sep- 44% the initial and peak values of PCT were concordant. tic shock during the entire study period, or in 67 patients who entered the final statistical analysis, the algorithm Discussion was fully applied. However, in patients with a low PCT Factors affecting adherence to the PCT protocol (< 0.5 ng/mL) or a marked decrease of PCT levels (≤ 10% Currently, there are only limited data on PCT protocol of peak level), a shorter antibiotic treatment was not adherence in critically ill patients outside of study con- associated with worse clinical outcomes. Thus, our study ditions. One study in medical ICU patients revealed confirms the results of a recent large study [1] and several an adherence below 50% to a local PCT protocol [16]. reviews and meta-analyses [2–6]. Procalcitonin is able to However, this study did not compare adherently treated support clinical decision-making to discontinue antibiot- patients to non-adherently treated patients. Thus, our ics when PCT decrease goes along with clinical improve- study reveals important insights on whether and how ment. However, in our study, antibiotics were continued the promising strategy of PCT-guidance is transferred to in nearly 70% of patients although antibiotic discontinu- clinical routine practice. ation was recommended by the PCT protocol. Algo- We detected a low adherence to discontinue antibiotics rithm overruling was also of concern in prospective trials adherently to a PCT stopping advice in clinical real life. where overruling rates reached from 16 to 53% [17–19]. Hohn et al. Ann. Intensive Care (2018) 8:68 Page 6 of 10 Table 1 Demographics and  clinical data of  patients with  appropriate and  non-appropriate procalcitonin (PCT) measurement series Patients with appropriate PCT Patients with non‑appropriate PCT p value series (n = 42) series (n = 25) Age, years (median [IQR]) 68 [59–76] 76 [68–79] 0.172 SAPS II score (median [IQR]) 40 [29–51] 38 [32–47] 0.807 Male gender (% [95% CI]) 59.5 [43.3–74.4] 60.0 [38.7–78.9] 0.999 ICU mortality (% [95% CI]) 40.5 [26–57] 32.0 [14.9–53.5] 0.604 Ventilation hours (median [IQR]) 423 [68–840] 102 [8–335] 0.012 ICU LOS, days (median [IQR]) 17 [5–41] 11 [4–21] 0.264 Hospital LOS, days (median [IQR]) 42 [25–73] 39 [25–54] 0.351 SOFA score max (median [IQR]) 11 [9–13] 9 [8–11] 0.112 Length of antibiotic treatment, days (median [IQR]) 10 [7–13] 5 [4–6] < 0.001 Antibiotic use density, DDD (median [IQR]) 11 [9–23] 6 [4–9] < 0.001 Initial PCT value, ng/mL (median [IQR]) 9.15 [3.1–32.4] 1.55 [0.9–4.6] 0.001 Peak PCT value, ng/mL (median [IQR]) 30.2 [7.1–60.7] 3.8 [1.6–10.3] < 0.001 % [95% CI] (n) % [95% CI] (n) Cause of infection Abdominal sepsis 52.4 [36.4–67.9] (22) 44.0 [24.4–65.1] (11) Bloodstream infection 0 4.0 [0.1–20.3] (1) Pleural empyema 9.5 [2.6–22.6] (4) 0 Pneumonia 19.0 [8.6–34.1] (8) 32.0 [14.9–53.5] (8) Prosthetic infection 0 8.0 [1.0–26.0] (2) Urogenital sepsis 9.5 [2.6–22.6] (4) 8.0 [1.0–26.0] (2) Soft tissue infection 2.4 [0.6–12.6] (1) 4.0 [0.1–20.3] (1) Unknown 7.1 [1–19] (3) 0 Expectably, overruling rates were higher in real life since in adherently treated patients. In 75% of the adherently the PCT protocol was not applied under controlled treated patients, the initial PCT corresponded to the study conditions and populations are not comparable peak value. Contrasting to that, in 56% of the not adher- due to strict in- and exclusion criteria in randomised ently treated patients the PCT value rises after the initial trials. Exclusion rates in randomised PCT studies range value. For clinicians, high initial PCT levels seem to be from 0.3% [20] to 84% [21], thus not necessarily reflect - no barrier to stop antibiotics according to the PCT pro- ing clinical real life. For example, patients with common tocol, but secondarily increasing PCT seems to prompt ICU infections, e.g. caused by Pseudomonas aeruginosa clinicians to continue antibiotics. Similarly, to the study and Acinetobacter baumannii, which were excluded in by Jensen [20], in clinical real life, secondary increases several studies [17–19] were not excluded in our study. seem to be interpreted as “alert PCT”. For most patients, Furthermore, probably we cannot exclude that inappro- a decline of PCT to ≤ 10% of peak value was a trigger for priate septic source control contributed to low adherence an antibiotic stopping advice and not the absolute limit of to antibiotic stopping advices in our study, and we did PCT < 0.5 ng/mL. Thus, in patients with severe sepsis and not analyse the impact of multi drug resistant pathogens septic shock, relative variations of PCT seem to be more on antibiotic treatment duration. important to the clinician than absolute values. Further- Nonetheless, the low rate of less than 10% of sep- more, in patients with secondary PCT increases, the pro- tic patients in which discontinuation of antibiotics was calcitonin protocol was more frequently overruled and accompanied by a PCT stop signal challenges the clini- antibiotic treatment continued despite reaching a stop- cal relevance of PCT-guided antibiotic stewardship and ping advice. efforts should be made to improve adherence to PCT Our data suggest that certain types of infection like protocols in clinical practice. pneumonia or abdominal sepsis might affect adherence Kinetics of procalcitonin also affect adherence to dis - to PCT protocols as well. In patients who reached a stop- continue antibiotics when a stopping advice thresh- ping advice, those with abdominal sepsis were treated old is reached. Median initial PCT levels were higher adherently in 43% cases (7/15), whereas patients with Hohn et al. Ann. Intensive Care (2018) 8:68 Page 7 of 10 Table 2 Demographics and clinical data of patients with and without procalcitonin (PCT) stopping advice Patients with PCT stopping Patients without PCT stopping p value advice (n = 26) advice (n = 16) Age, years (median [IQR]) 68 [59–75] 70 [62–79] 0.475 SAPS II score (median [IQR]) 40 [29–51] 44 [33–52] 0.597 Male gender (% [95% CI]) 69.2 [48.2–85.7] 43.8 [19.8–70.1] 0.121 ICU mortality (% [95% CI]) 26.9 [11.6–47.8] 62.5 [35.4–84.8] 0.029 Ventilation hours (median [IQR]) 597 [96–1260] 238 [47–461] 0.068 ICU LOS, days (median [IQR]) 23 [5–56] 11 [5–20] 0.212 Hospital LOS, days (median [IQR]) 56 [35–100] 23 [17–44] 0.002 SOFA score max (median [IQR]) 11 [9–12] 13 [9–15] 0.320 Length of antibiotic treatment, days (median [IQR]) 10 [8–13] 10 [7–12] 0.398 Antibiotic use density, DDD (median [IQR]) 13 [10–24] 10 [8–15] 0.136 Initial PCT value, ng/mL (median [IQR]) 20.0 [3.3–42.7] 5.1 [2.8–10.8] 0.136 Peak PCT value, ng/mL (median [IQR]) 40.5 [9.2–69.7] 16.2 [6.0–32.5] 0.095 % [95% CI] (n) % [95% CI] (n) Cause of infection Abdominal sepsis 57.7 [36.9,76.6] (15) 43.8 [19.8–70.1] (7) Bloodstream infection 0 0 Pleural empyema 11.5 [2.4–30.2] (3) 6.3 [0.2–30.2] (1) Pneumonia 26.9 [11.5–47.8] (7) 6.3 [0.2–30.2] (1) Prosthetic infection 0 0 Urogenital sepsis 3.8 [0.1–19.6] (1) 18.8 [4.0–45.6] (3) Soft tissue infection 0 6.3 [0.2–30.2] (1) Unknown 0 18.8 [4.0–45.6] (3) pneumonia were treated adherently only in 14% (1/7). that patients without appropriate PCT measurement However, due to the small number of patients with other series were in a better clinical condition, and probably infection sites than abdomen or lung, this could not be regularly PCT testing was not considered necessary by proven statistically in this study. Nonetheless, these the clinician. In patients with inappropriate PCT series, results are interesting since mortality in critically ill sep- median duration of antibiotic treatment was 5 days, and tic patients with pneumonia or abdominal sepsis are thus more regular PCT testing likely would not have comparable [22, 23]. had beneficial impact on further reduction in antibiotic exposure. Factors affecting appropriate versus non‑appropriate PCT measurement series Factors affecting achievement of PCT stopping advice A prerequisite to PCT-guided antibiotic stewardship is threshold regular measurements. In our study, 62.7% had appropri- Beside its value for biomarker-guided antibiotic steward- ate PCT measurement series. Median initial PCT values ship, PCT can be used as a prognostic value. Patients who and peak values were statistically significantly higher in did not reach the PCT stopping advice threshold had the group with appropriate measurement series. These higher mortality rates (63%) compared to those where results suggest that frequency of PCT testing is affected PCT declined indicating that the infection is under con- by PCT levels, and high PCT values might prompt cli- trol (27%). Median initial PCT levels and peak levels were nicians to request PCT testing more frequently. In the higher in patients with stopping advice. However, these group of patients with non-appropriate PCT series, there results were statistically not significant. Interestingly, in was a higher number of patients with pneumonia. It is 81.3% (13/16) of patients without reaching the PCT stop- well known that peak values of PCT are lower in res- ping advice, PCT levels increased after the initial value. piratory tract infections compared to other kind of infec- In patients with stopping advice, this occurred only in tions [24–26]. Lower duration of mechanical ventilation, 42.3% (11/26). These results are in line with previous shorter stay in the ICU, and lower mortality rates reflect studies which showed that high PCT levels are associated Hohn et al. Ann. Intensive Care (2018) 8:68 Page 8 of 10 Table 3 Demographics and  clinical data of  patients with  procalcitonin (PCT) stopping advice treated adherently and non-adherently to the PCT protocol Patients adherent to PCT Patients non‑adherent to PCT p value stopping advice (n = 8) stopping advice (n = 18) Age, years (median [IQR]) 67 [59–75] 68 [59–75] 0.560 SAPS II score (median [IQR]) 35 [31–43] 44 [28–51] 0.655 Male gender (% [95% CI]) 50.0 [15.7–84.3] 77.8 [30.6–69.4] 0.197 ICU mortality (% [95% CI]) 12.5 [0.3–52.7] 33.3 [13.3–59.0] 0.375 Ventilation hours (median [IQR]) 353 [68–621] 735 [209–1508] 0.317 ICU LOS, days (median [IQR]) 24 [3–38] 23 [12–59] 0.780 Hospital LOS, days (median [IQR]) 57 [47–79] 54 [34–100] 0.824 SOFA score max (median [IQR]) 11 [9–12] 12 [9–13] 0.557 Length of antibiotic treatment, days (median [IQR]) 7 [6–9] 12 [9–16] 0.002 Antibiotic use density, DDD (median [IQR]) 10 [7–12] 18 [12–26] 0.001 Initial PCT value, ng/mL (median [IQR]) 20.0 [9.4–38.2] 16.8 [2.0–42.7] 0.560 Peak PCT value, ng/mL (median [IQR]) 32.3 [9.4–64.2] 44.9 [13.9–69.7] 0.718 % [95% CI] (n) % [95% CI] (n) Cause of infection Abdominal sepsis 87.5 [47.3–99.7] (7) 44.4 [21.5–69.2] (8) Bloodstream infection 0 0 Pleural empyema 0 16.7 [3.5–41.4] (3) Pneumonia 12.5 [0.3–52.7] (1) 33.3 [13.3–59.0] (6) Prosthetic infection 0 0 Urogenital sepsis 0 5.6 [0.1–27] (1) Soft tissue infection 0 0 Unknown 0 0 with an unfavourable outcome [27, 28]. Furthermore, a to guide antibiotic treatment [15]. Furthermore, we can not adequately declining PCT is an independent predic- only assess correlation and cannot comment on causality, tor of mortality in septic patients [29, 30]. In our study, and thus, probably additional factors may have contrib- the shorter ICU and hospital lengths of stay in patients uted to the results. without stopping advice might be founded in the fact that Furthermore, we can only comment on the first episode those patients had a higher mortality. of sepsis and did not analyse the impact of PCT guidance on further septic episodes during the stay in ICU. As our data were collected retrospectively, diagnosis of sepsis Limitations was not yet according to the new sepsis definition [31], There are limitations of the present study which should and thus we may have missed patients because one in be mentioned. Taken the overall low sample size together eight patients admitted to intensive care units with infec- with the retrospective study design, all results and their tion and new organ failure does not meet the condition interpretation should be taken with caution. As possibly of at least two systemic inflammatory response syndrome not all septic cases might have been registered correctly criteria. in the hospital’s database, we cannot exclude a selec- tion bias in our analysis. The selection of patients was Conclusion also possibly affected by the definition of an appropriate Our study showed that a PCT protocol works outside measurement series. However, today, there is no universal clinical studies, but these benefits along with interna - definition how often and in what intervals PCT should be tional guideline recommendations have not yet entered measured to be sufficient to guide antibiotic treatment. the daily routine in intensive care medicine. In patients In addition, there are little data for the frequency of PCT with severe sepsis and septic shock, PCT testing was testing and adherence to PCT protocols in clinical real irregular and adherence to a PCT protocol for discon- life. In our study, we used a recently published definition tinuation of antibiotic treatment was low, but when to classify a series of PCT measurements as appropriate applied it was associated with a considerably shorter Hohn et al. Ann. Intensive Care (2018) 8:68 Page 9 of 10 Funding duration of antibiotic treatment and a lower antibiotic Data were generated and analysed as part of the routine work, and no other consumption for the first septic episode without nega - kind of funding has been received. tive clinical outcomes. Severity of illness, PCT peak levels, and PCT kinetics, respectively, might affect Publisher’s Note frequency of PCT testing and protocol adherence. Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. Patients at high risk without sufficiently decreasing or even increasing PCT levels seem to prompt clinicians Received: 6 March 2018 Accepted: 28 May 2018 to continue antibiotics and to overrule PCT protocols despite reaching antibiotic stopping advice thresholds. Additional file References 1. de Jong E, van Oers JA, Beishuizen A, Vos P, Vermeijden WJ, Haas LE et al. Efficacy and safety of procalcitonin guidance in reducing the duration Additional file 1: Figure S1. Local PCT protocol. Start of antibiotic treat - of antibiotic treatment in critically ill patients: a randomised, controlled, ment is based on a clinical decision. According to this algorithm, antibiot- open-label trial. Lancet Infect Dis. 2016;16(7):819–27. ics should be discontinued when clinical improvement goes along with 2. Matthaiou DK, Ntani G, Kontogiorgi M, Poulakou G, Armaganidis A, Dimo- decreasing PCT levels. The protocol can be overruled by the means of the poulos G. An ESICM systematic review and meta-analysis of procalci- attending physician due to clinical reasons or when conditions are pre- tonin-guided antibiotic therapy algorithms in adult critically ill patients. sent which require a prolonged antibiotic treatment. Daily PCT samples Intensive Care Med. 2012;38(6):940–9. were recommended on day 1–3 in patients on antibiotic treatment since 3. Prkno A, Wacker C, Brunkhorst FM, Schlattmann P. Procalcitonin-guided admission to ICU or since clinical suspicion of systemic bacterial infection. therapy in intensive care unit patients with severe sepsis and septic From day 4 on, PCT samples were recommended every other day in shock—a systematic review and meta-analysis. Crit Care. 2013;17(6):R291. patients under antibiotic treatment. 4. Kopterides P, Siempos II, Tsangaris I, Tsantes A, Armaganidis A. Procalci- tonin-guided algorithms of antibiotic therapy in the intensive care unit: a systematic review and meta-analysis of randomized controlled trials. Crit Abbreviations Care Med. 2010;38(11):2229–41. CI: Confidence interval; ICU: Intensive care unit; pt(s): Patient(s); PCT: Procalci- 5. Tang H, Huang T, Jing J, Shen H, Cui W. Eec ff t of procalcitonin-guided tonin; SAPS: Simplified acute physiology score; SOFA: Sequential organ failure treatment in patients with infections: a systematic review and meta- score. analysis. Infection. 2009;37(6):497–507. 6. Iankova I, Thompson-Leduc P, Kirson NY, Rice B, Hey J, Krause A et al. Effi- Authors’ contributions cacy and safety of procalcitonin guidance in patients with suspected or AH, NB, BH and SS were responsible for the conception, study design, study confirmed sepsis: a systematic review and meta-analysis. Crit Care Med. execution, data management, data analysis, data interpretation, and manu- 2018;46(5):691–8. script writing. SH and JCW were responsible for conception, study design, 7. Huang HB, Peng JM, Weng L, Wang CY, Jiang W, Du B. Procalcitonin- data management, data analysis, data interpretation, and critically revising guided antibiotic therapy in intensive care unit patients: a systematic the manuscript. MH was responsible for data analysis, data interpretation and review and meta-analysis. Ann Intensive Care. 2017;7(1):114. manuscript writing. All authors read and approved the final manuscript. 8. Brechot N, Hekimian G, Chastre J, Luyt CE. Procalcitonin to guide antibi- otic therapy in the ICU. Int J Antimicrob Agents. 2015;46(Suppl 1):S19–24. Author details 9. Andriolo BN, Andriolo RB, Salomao R, Atallah AN. Eec ff tiveness and Department of Anaesthesiology and Intensive Care Medicine, University safety of procalcitonin evaluation for reducing mortality in adults with Hospital of Cologne, Kerpener Str. 62, 50937 Cologne, Germany. Department sepsis, severe sepsis or septic shock. Cochrane Database Syst Rev. of Infectiology and Hospital Hygiene, Hospital Düren gem. GmbH, Roonstraße 2017;1:Cd010959. 30, 52351 Düren, Germany. Thermo Fisher Scientific, Thermo Scientific 10. Rhodes A, Evans LE, Alhazzani W, Levy MM, Antonelli M, Ferrer R, et al. Biomarkers, Neuendorfstr. 25, 16761 Hennigsdorf, Germany. Department Surviving sepsis campaign: international guidelines for management of of Anaesthesiology, University Hospital Heidelberg, Im Neuenheimer Feld sepsis and septic shock: 2016. Intensive Care Med. 2017;43(3):304–77. 110, 69120 Heidelberg, Germany. Department of Anaesthesiology, Surgical 11. de With K, Allerberger F, Amann S, Apfalter P, Brodt HR, Eckmanns T, Intensive Care, Emergency Medicine and Pain Management, Hospital Düren et al. Strategies to enhance rational use of antibiotics in hospital: a gem. GmbH, Roonstraße 30, 52351 Düren, Germany. guideline by the German society for infectious diseases. Infection. 2016;44(3):395–439. Acknowledgements 12. Povoa P, Salluh JI. Biomarker-guided antibiotic therapy in adult critically ill None. patients: a critical review. Ann Intensive Care. 2012;2(1):32. 13. Hohn A, Schroeder S, Gehrt A, Bernhardt K, Bein B, Wegscheider K, et al. Competing interests Procalcitonin-guided algorithm to reduce length of antibiotic therapy in SH and JW as employees of Thermo Fisher Scientific performed the statistical patients with severe sepsis and septic shock. BMC Infect Dis. 2013;13:158. analysis. SS served as consultant and has received payments from BRAHMS AG 14. Hohn A, Heising B, Hertel S, Baumgarten G, Hochreiter M, Schroeder S. for speaking engagements. All other authors declare no conflict of interest. Antibiotic consumption after implementation of a procalcitonin-guided antimicrobial stewardship programme in surgical patients admitted to Availability of data and materials an intensive care unit: a retrospective before-and-after analysis. Infection. The datasets used and/or analysed during the current study are available from 2015;43(4):405–12. the corresponding author on reasonable request. 15. Bodmann KF, Schenker M, Heinlein W, Wilke MH. Procalcitonin as a tool for the assessment of successful therapy of severe sepsis: an analysis Consent for publication using clinical routine data. Med Klin Intensivmed Notfmed. 2016. https :// Not applicable. doi.org/10.1007/s0006 3-016-0183-7. 16. Ammar AA, Lam SW, Duggal A, Neuner EA, Bass SN, Guzman JA, et al. Ethics approval and consent to participate Compliance with procalcitonin algorithm antibiotic recommenda- Ethics approval was waived by the Ethics Committee of the State Chamber of tions for patients in medical intensive care unit. Pharmacotherapy. Physicians of North Rhine (Germany) due to the retrospective nature of the 2017;37(2):177–86. study (Ethics Committee No 136-2015). Hohn et al. Ann. Intensive Care (2018) 8:68 Page 10 of 10 17. Bouadma L, Luyt CE, Tubach F, Cracco C, Alvarez A, Schwebel C, et al. Use 25. Hedlund J, Hansson LO. Procalcitonin and C-reactive protein levels in of procalcitonin to reduce patients’ exposure to antibiotics in intensive community-acquired pneumonia: correlation with etiology and progno- care units (PRORATA trial): a multicentre randomised controlled trial. sis. Infection. 2000;28(2):68–73. Lancet. 2010;375(9713):463–74. 26. Luyt CE, Guerin V, Combes A, Trouillet JL, Ayed SB, Bernard M, et al. 18. Nobre V, Harbarth S, Graf JD, Rohner P, Pugin J. Use of procalcitonin to Procalcitonin kinetics as a prognostic marker of ventilator-associated shorten antibiotic treatment duration in septic patients: a randomized pneumonia. Am J Respir Crit Care Med. 2005;171(1):48–53. trial. Am J Respir Crit Care Med. 2008;177(5):498–505. 27. Klingele M, Bomberg H, Schuster S, Schafers HJ, Groesdonk HV. Prog- 19. Stolz D, Smyrnios N, Eggimann P, Pargger H, Thakkar N, Siegemund M, nostic value of procalcitonin in patients after elective cardiac surgery: a et al. Procalcitonin for reduced antibiotic exposure in ventilator-associ- prospective cohort study. Ann Intensive Care. 2016;6(1):116. ated pneumonia: a randomised study. Eur Respir J. 2009;34(6):1364–75. 28. Peschanski N, Chenevier-Gobeaux C, Mzabi L, Lucas R, Ouahabi S, Aqui- 20. Jensen JU, Hein L, Lundgren B, Bestle MH, Mohr TT, Andersen MH, et al. lina V, et al. Prognostic value of PCT in septic emergency patients. Ann Procalcitonin-guided interventions against infections to increase early Intensive Care. 2016;6(1):47. appropriate antibiotics and improve survival in the intensive care unit: a 29. Harbarth S, Holeckova K, Froidevaux C, Pittet D, Ricou B, Grau GE, et al. randomized trial. Crit Care Med. 2011;39(9):2048–58. Diagnostic value of procalcitonin, interleukin-6, and interleukin-8 in 21. Svoboda P, Kantorova I, Scheer P, Radvanova J, Radvan M. Can procalci- critically ill patients admitted with suspected sepsis. Am J Respir Crit Care tonin help us in timing of re-intervention in septic patients after multiple Med. 2001;164(3):396–402. trauma or major surgery? Hepatogastroenterology. 2007;54(74):359–63. 30. Schuetz P, Birkhahn R, Sherwin R, Jones AE, Singer A, Kline JA, et al. Serial 22. Zahar JR, Timsit JF, Garrouste-Orgeas M, Francais A, Vesin A, Descorps- procalcitonin predicts mortality in severe sepsis patients: results from Declere A, et al. Outcomes in severe sepsis and patients with septic the multicenter procalcitonin monitoring sepsis (MOSES) study. Crit Care shock: pathogen species and infection sites are not associated with Med. 2017;45(5):781–9. mortality. Crit Care Med. 2011;39(8):1886–95. 31. Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, 23. Leligdowicz A, Dodek PM, Norena M, Wong H, Kumar A, Kumar A. Asso- Bauer M, et al. The third international consensus definitions for sepsis and ciation between source of infection and hospital mortality in patients septic shock (sepsis-3). JAMA. 2016;315(8):801–10. who have septic shock. Am J Respir Crit Care Med. 2014;189(10):1204–13. 24. Brunkhorst FM, Al-Nawas B, Krummenauer F, Forycki ZF, Shah PM. Procalcitonin, C-reactive protein and APACHE II score for risk evaluation in patients with severe pneumonia. Clin Microbiol Infect. 2002;8(2):93–100.

Journal

Annals of Intensive CareSpringer Journals

Published: Jun 4, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off