You are on page 1of 11


Procalcitonin-Guided Antibiotic Therapy: A Systematic Review and
Nilam J. Soni, MD1*, David J. Samson, MS2, Jodi L. Galaydick, MD3, Vikrant Vats, PhD2, Elbert S. Huang, MD, MPH4,
Naomi Aronson, PhD2, David L. Pitrak, MD5

Division of Hospital Medicine, Department of Medicine, University of Texas Health Science Center San Antonio, San Antonio, Texas; 2Blue Cross
and Blue Shield Association Technology Evaluation Center Evidence-Based Practice Center, Chicago, Illinois; 3Division of Critical Care Medicine,
Department of Surgery, Icahn School of Medicine at Mt. Sinai, New York, New York; 4Section of General Internal Medicine, Department of Medicine,
University of Chicago, Chicago, Illinois; 5Section of Infectious Diseases & Global Health, Department of Medicine, University of Chicago, Chicago,

BACKGROUND: The utility of procalcitonin to manage [CI]: 22.59 to 21.52) without increasing morbidity or mortal-
patients with infections is unclear. A systematic review of ity. In contrast, procalcitonin-guided intensification of antibi-
comparative studies using procalcitonin-guided antibiotic otics in adult ICU patients increased antibiotic usage and
therapy in patients with infections was performed. morbidity. In adult patients with respiratory tract infections,
procalcitonin guidance significantly reduced antibiotic dura-
METHODS: Randomized, controlled trials comparing tion by 2.35 days (95% CI: 24.38 to 20.33), antibiotic pre-
procalcitonin-guided initiation, intensification, or discontinu- scription rate by 22% (95% CI: 241% to 24%), and total
ation of antibiotic therapy to clinically guided therapy were antibiotic exposure without affecting morbidity or mortality.
included. Outcomes were antibiotic usage, morbidity, and A single, good quality study of neonates with suspected
mortality. MEDLINE, EMBASE, the Cochrane Database, sepsis demonstrated reduced antibiotic duration by 22.4
National Institute for Clinical Excellence, the National hours (P 5 0.012) and reduced the proportion of neonates
Guideline Clearinghouse, and the Health Technology on antibiotics for 72 hours by 27% (P 5 0.002) with procal-
Assessment Programme were searched from January 1, citonin guidance.
1990 to December 16, 2011.
CONCLUSION: Procalcitonin guidance can safely reduce
RESULTS: Eighteen randomized, controlled trials were antibiotic usage when used to discontinue antibiotic therapy
included. Data were pooled into clinically similar patient in adult ICU patients and when used to initiate or discon-
populations. In adult intensive care unit (ICU) patients, tinue antibiotics in adult patients with respiratory tract infec-
procalcitonin-guided discontinuation of antibiotics reduced tions. Journal of Hospital Medicine 2013;000:000–000.
antibiotic duration by 2.05 days (95% confidence interval C 2013 Society of Hospital Medicine

Many serum biomarkers have been identified in recent roid gland. Conversion of procalcitonin to calcitonin
years with a wide range of potential applications, is inhibited by various cytokines and bacterial endo-
including diagnosis of local and systemic infections, dif- toxins. Procalcitonin’s primary diagnostic utility is
ferentiation of bacterial and fungal infections from viral thought to be in establishing the presence of bacterial
syndromes or noninfectious conditions, prognostic infections, because serum procalcitonin levels rise and
stratification of patients, and enhanced management fall rapidly in bacterial infections.3–5 In healthy indi-
of antibiotic therapy. Currently, there are at least viduals, procalcitonin levels are very low. In systemic
178 serum biomarkers that have potential roles to infections, including sepsis, procalcitonin levels are
guide antibiotic therapy, and among these, pro- generally greater than 0.5 to 2 ng/mL, but often reach
calcitonin has been the most extensively studied levels 10 ng/mL, which correlates with severity of
biomarker.1,2 illness and a poor prognosis. In patients with respira-
Procalcitonin is the prohormone precursor of calci- tory tract infections, procalcitonin levels are less ele-
tonin that is expressed primarily in C cells of the thy- vated, and a cutoff of 0.25 ng/mL seems to be most
predictive of a bacterial respiratory tract infection
requiring antibiotic therapy.6–8 Procalcitonin levels
*Address for correspondence and reprint requests: Nilam J. Soni, MD, decrease to <0.25 ng/mL as infection resolves, and a
Division of Hospital Medicine, University of Texas Health Science Center decline in procalcitonin level may guide decisions
San Antonio, 7703 Floyd Curl Drive, MC 7982, San Antonio, TX 78229-
3900; Telephone: 210–567-4815; Fax: 210–358-0647; E-mail: about discontinuation of antibiotic therapy.5 The purpose of this systematic review was to syn-
Additional Supporting Information may be found in the online version of thesize comparative studies examining the use of pro-
this article. calcitonin to guide antibiotic therapy in patients with
Received: March 3, 2013; Revised: May 23, 2013; Accepted: May 30, suspected local or systemic infections in different
2013 Society of Hospital Medicine DOI 10.1002/jhm.2067 patient populations. We are aware of 6 previously
Published online in Wiley Online Library ( published systematic reviews evaluating the utility of

An Official Publication of the Society of Hospital Medicine Journal of Hospital Medicine Vol 00 | No 00 | Month 2013 1 Data elements were abstracted into the following categories: quality Search Strategy assessment. subgroup and sensitivity analyses were Systematic Reviews. graded as high. studies. and (5) secondary outcomes included morbidity Research and Quality’s Effective Health Care (antibiotic adverse events.15 A standard protocol consistent with the care unit length of stay). controlled trial or nonrandomized com- review board reviewed and exempted this study. and outcome assessment. (3) interventions included initiation. systemic erature search. Gray literature. Studies were included if they ful- filled all of the following criteria: (1) randomized. duration of This review is based on a comparative effectiveness antibiotic therapy. febrile neutropenia.Soni et al | Procalcitonin-Guided Antibiotic Therapy procalcitonin guidance in the management of infec. and quality of life. Three quality categories were used: good. applicability and clinical diversity assess- MEDLINE and EMBASE were searched from January 1. this review was made after completing the formal lit- tion(s). con. or insufficient. and postoperative patients at risk of tions. directness. nin AND chronic obstructive pulmonary disease. tive studies using the following search terms: procalcito. 2011. Methods Guide for Effectiveness and Comparative Studies with any of the following criteria were Effectiveness Reviews16 was followed. hospital and/or intensive Program. not a www. intensive care unit. and/or discontinuation of antibiotic therapy groups compared to other systematic reviews. (4) pri- mary outcomes included antibiotic usage (antibiotic METHODS prescription rate. ICU. 2006 to June 28. RESULTS trolled trial or nonrandomized comparative study. grants and federally funded Methods Guide. a search for systematic reviews questions and outcomes. An investigational randomized. (2) Of the 2000 studies identified through the literature adult and/or pediatric patients with known or sus. unknown source. In addition. and manufacturing information was searched Recommendations Assessment. Search of gray literature yielded 4 published ill patients with sepsis syndrome or ventilator. Study Selection and precision. 2011.16 a system based on the Grading of research. neonates with sepsis. abstracts The strength of evidence was graded using the and conference papers. low. fair. and the decision to pool studies was inflammatory response syndrome OR postoperative based on the specific number of studies with similar infection. regulatory information. the Cochrane Database of performed. sepsis. postoperative compli. guided by procalcitonin plus clinical criteria. and the Health Technology Assessment ing method involved inverse variance weighting and a Programme. 1986 were excluded and 14 studies19–32 were pected local or systemic infection. including critically included. not reporting relevant outcomes.effectivehealthcare. including databases with random effects model. Development and from January 1. Study Question Data Extraction and Quality Assessment In selected populations of patients with suspected local or systemic infection. final strength of evidence grading was made by con- Second and third reviewers were consulted to screen sensus adjudication among the authors.33–36 A total of 18 randomized. critical illness. ment. search. based on clinical similarity of available studies and tute for Clinical Excellence. clinical trial registries. A detailed excluded: published in non-English language. If a meta-analysis could be was conducted in MEDLINE. the National Guideline reporting of mean and standard deviation. not description of the methods is available online (http:// reporting primary data from original research. intensifi- and pooled patients into the most clinically similar cation. surgical wound infection. parative study. reviewers. septic. controlled tri- associated pneumonia.18 The following domains were addressed: risk of bias. Disagreements between for infection to guide initiation. Quality of included 1990 through December 16. intensification. and poor. and/or reviewers were resolved by group discussion among discontinuation of antibiotic therapy when compared the research team and final quality rating was assigned to clinical criteria for infection alone? by consensus adjudication. children with fever of criteria alone to manage antibiotic therapy in patients 2 An Official Publication of the Society of Hospital Medicine Journal of Hospital Medicine Vol 00 | No 00 | Month 2013 . The decision to incorporate formal data synthesis in cation(s). adults with respiratory tract als comparing procalcitonin guidance to use of clinical infections. The ies looking at procalcitonin-guided antibiotic therapy. and the Cochrane studies was assessed using the US Preventive Services Controlled Trials register was searched with no date Task Force framework17 by at least 2 independent restriction for randomized and nonrandomized compara. articles when needed. postsurgical infec. COPD. and days without antibiotic ther- review prepared for the Agency for Healthcare apy). moderate. The overall strength of evidence was A single reviewer screened abstracts and selected stud. and Web sites of the National Insti. consistency. Data Synthesis and Analysis intensive care. what are the effects of A single reviewer abstracted data and a second using procalcitonin measurement plus clinical criteria reviewer confirmed accuracy.9–14 Our systematic review included more studies infection. mortality. The pool- Clearinghouse. critically ill. total antibiotic exposure. postoperative infection(s). Evaluation Working Group.

and (5) postoperative patients at duration of antibiotic therapy. morbidity. Four studies conducted superiority anal. 2. and outcomes that were not Procalcitonin-guided antibiotic discontinuation did adequately studied in randomized. 1. including ICU length of stay Data were pooled into clinically similar groups that (LOS). The Jensen et al. controlled trials. associated pneumonia. (3) neonates with suspected sepsis.ahrq. respectively. We sought. (2) adult patients with respira- tory tract infections.43% favoring able online (http://www. are displayed in Tables 1. study found a 2-day increase. Meta-analysis of ICU LOS is displayed in were reviewed separately: (1) adult intensive care unit Figure 2B. and 3. and the relative reduction ranged intensification. whereas 1 study conducted a noninferiority analysis. were reported (Table 2). and mortality outcomes interval (CI) of 26% to 5% (Figure 2C). with infections were included. in the duration of antibiotic therapy The absolute reduction in duration of antibiotic usage and a 7. An Official Publication of the Society of Hospital Medicine Journal of Hospital Medicine Vol 00 | No 00 | Month 2013 3 . PRISMA diagram of the literature search. Morbidity parisons. showed a mortality difference of 20. ICU patients. to their standard practice. Mortality (4) children between 1 to 36 months of age with fever There was no increase in mortality as a result of shorter of unknown source. com. including patients with ventilator. The PRISMA diagram from 21% to 38%. Tables summarizing study quality and term mortality (28-day or in-hospital mortality) outcome measures with strength of evidence are avail.33 (N 5 1272) addressed procalcitonin- Discontinuation guided intensification of antibiotic therapy in adult Five studies19–23 (N 5 938) addressed procalcitonin. non- randomized comparative studies of populations. Adult ICU Patients: Procalcitonin-Guided Antibiotic Intensification Adult ICU Patients: Procalcitonin-Guided Antibiotic Two studies24. Meta-analysis of antibiotic duration (Figure 1) depicts the flow of search screening and in adult ICU patients was performed (Figure 2A).7 to 5 days. study selection. and a 95% confidence Antibiotic usage. Limited data on adverse antibiotic events (ICU) patients. algorithm for broadening antibiotic therapy in yses for mortality with procalcitonin-guided therapy. Meta-analysis of short- risk of infection. not increase morbidity. Procalcitonin-Guided Antibiotic Therapy | Soni et al FIG. interventions.002) in the num- with procalcitonin guidance in these studies ranged ber of days on 3 antibiotics with procalcitonin-guided from 1. but did not find. The Jensen et al. Physicians study24 was too small (N 5 72) and lacked sufficient in control groups administered antibiotics according details to be informative. or 50% Antibiotic Usage relative increase. study also educated physicians about empiric therapy Absolute procalcitonin values for discontinuation of and intensification of antibiotic therapy.9% absolute increase (P 5 0. high-quality study that used a detailed ICU patients. Antibiotic Usage The Jensen et al. patients with elevated procalcitonin-guided therapy.effectivehealthcare. A second antibiotics ranged from 0.25 to 1 ng/mL. study33 was a large guided discontinuation of antibiotic therapy in adult (N 5 1200).

‡ Per protocol analysis.0) 0.001 § Total ABT exposure Nobre.9 to 27.0 (22. 200728 208 41/102 (40.001 Stolz.1 to 22.5 67.002 Adult patients with respiratory tract infections ABT duration.3) <0.9 to 227.7) 0. not statistically significant. Year N PCT-Guided Therapy* Control* Difference PCT-CTRL (95% CI) P Value Critically ill adult patients: procalcitonin-guided antibiotic discontinuation ABT Duration. 200926 210 5. 201133 1200 3570/5447 (65.4 220.4 0. 201023 621 14.1) <0.001 Critically ill adult patients: procalcitonin-guided antibiotic intensification ABT duration.6%) 212. 201031 All neonates (N 5 121) 79.1% 0.9) 0.0001 Christ-Crain.3 to 218. § Per 1000 inpatient days.8 Christ-Crain.5 to 20.9 to 1. 0. 200629 302 5.9 to 217.1 (212.0001 Days without ABTs.5% (244.6 8.5 9 23. intention to treat. 13. NSS.001 Burkhardt. ITT. 200921 101 1077 1341 Bouadma.0 <0. per protocol. control. 200728 208 NR NR 231.5 (1. serious blood infection.049 Bouadma. 200430 243 332# 661# — — Neonates with sepsis ABTs 72 hours. 201036 All children (N 5 384) 48/192 (25%) 54/192 (28. Specifically.15. % Stocker.0 to 5. SBI.49 No SBI or neutropenia (N 5 312) 14/158 (9%) 16/154 (10%) 21.0 to 9.4 13. PP.8) <0.05 Long.3jj (1.004 Stolz.05 Long.8 27 NSS Infection possible (N 5 40) 83.3 11. % Manzano. 201023 621 10. 200819 79 15.4 111.004) and a significant days on mechanical ventilation (P < 0.8 7.3) 0.4%) 99/119 (83.1) 0. jj Rate ratios.7 to 20.9%) 272% (278 to 266) <0.8 to 211.3. confidence interval.28.7) 0.3 23.5 0.3).05 Christ-Crain.0001) and 5% 4 An Official Publication of the Society of Hospital Medicine Journal of Hospital Medicine Vol 00 | No 00 | Month 2013 .7 (22.001 Infection unlikely (N 5 60) 46.7) <0.4%) NR (97.5) <0. 200629 302 136# 323# — — Christ-Crain.2 (21.1 to 2 5.9 27.8 12. patients The Jensen et al. 200920 27 6.1 101.5–17)† 3.65 Adult postoperative patients at risk of infection ABT duration.4) 0.0% (24.9%) 19/19 (100%) 238. d Hochreiter.6 (251.2 7. 3. h Stocker.9 12.80) <0. 200827 458 6.8 (0. *Values are mean unless specified. 200925 1359 5.9 (23.1 to 25.27 NOTE: Abbreviations: ABT.3) 0.012 Infection proven/probably (N 5 21) 177.5)‡ 0.5 228.9 7.0 (21.002 Infection proven/probably (N 5 21) 9/9 (100%) 12/12 (100%) 0% (0 to 0) NA Infection possible (N 5 40) 13/21 (61. 200632 All patients (N 5 20) 5.5%) 213.04 Stolz.8) <0.3). CTRL.1 to 7. NR.0) <0.4 to 25.7 23.1jj (0.1 6.20 to 21.7 8.001 Children ages 1–36 months with fever of unknown source Antibiotic prescription rate.1 <0.2%) 76/106 (71. 201133 1200 6 (3–11)† 4 (3–10)† NR NR Days spent in ICU on 3 ABTs Jensen. 200925 1359 506/671 (75. % Schuetz.6 (5.07.0001 Antibiotic prescription rate.049 Bouadma.7%) 19/30 (63.5 (28. † Median (interquartile range).5%) 2721/4717 (57. ICU. 0.7 0.3 (25. ¶ Adjusted for potential confounding and possible cluster effects. 201034 550 84/275 (30. # Mean per 1000 days of follow-up.5 222.4 to 4. 200921 101 13 (2–21)† 9. CI.7%) 231.8).0001 Long. Morbidity increase in organ dysfunction.1 21. 200430 243 55/124 (44. day 28 Nobre.3 to 16.1) <0.7%) 7. 201034 550 7.3 13. study showed a significant 1-day had a highly statistically significant 5% increase in increase in ICU LOS (P 5 0.3 to 28. 201031 All neonates (N 5 121) 33/60 (55%) 50/61 (82%) 227. antibiotic.6 2.8 21. 1.1 (20. d* Schuetz. 17.003 Schroeder.1 to 1.0 — Christ-Crain.6 2.7) 0. 200819 79 66 9. 201135 162 NR (84.003 Infection unlikely (N 5 60) 11/30 (36.701 Total ABT exposure Stolz.3%) 226.7 to 44.0 (24.5%) 89/275 (32.0002 504 655 Stolz.0001 Burkhardt.4%) 603/688 (87.5 to 21.4) <0.0 (242.0%) 219/226 (96.7 (1. 200629 302 128/151 (84. 200922 110 5. 0. 23.2) 0.79%) 213. 201135 162 NR NR NR — Christ-Crain.3 21.1 to 20.7 — Briel.4%) 6.4%) 21.5% (18.1(28.9% (219.8 21.0001 Kristoffersen.8% (29.8 170. 201023§ 621 653 812 2159 (2185 to 2131) <0. procalcitonin.1 (258.002 Kristoffersen. 200926 210 88/103 (85. 200430 243 10.038 ABT duration. 10 (PP) 22. 200921 101 10 (6–16)† 15 (10–23)† 25 0. 201135 162 5 (3–6)¶ 7 (5–9)¶ 22.9 to 1.9) 0.4 to 21.25 Briel.7 to 0.4%) 85/107 (79.8% (249.9) <0. d Chromik. intensive are unit.2%) 238.5 (ITT).8%) 149/151 (98.5 0. 200819 79 541 644 1.9% (6.001 Nobre. not reported.2% (216.5)‡ 0.0%) 23.6 to 5. 200827 458 58/232 (25. PCT. Summary of Antibiotic Usage Outcomes Outcome Author.Soni et al | Procalcitonin-Guided Antibiotic Therapy TABLE 1. days Jensen.9 0.9 22.

7 20.9 23. diarrhea. 200724 72 16.7 20. 200632 20 2/10 8/10 260.7%) 22.8 0. 201023 621 1.22 Stolz.5% (210.6 6 6.2%) 21/151 (13.6 (28. % Chromik. PCT-CTRL (95% CI) P Value Critically ill adult patients: procalcitonin-guided antibiotic discontinuation ICU LOS.9% (3 to 6. 200925 1359 9. not applicable. ¶ Days during the first 14 days of illness that work and leisure activities were restricted. Mortality Adult Patients With Respiratory Tract Infections The Jensen et al.004 SOFA score* Svoboda. 200926 210 7/103 (6.03) >0. study was a superiority trial powered Eight studies25–30. mechanical ventilation.4) 0.02 Schroeder.08 Jensen.6 to 0. 200926 210 5. § Abdominal pain.1 13. 201036 All children (N 5 384) 50/192 (26%) 48/192 (25%) 1 (28 to 10) 0.2 0.5)† 2.2) 0.71 Kristoffersen. 200819 79 4 7 24.6) >0. GFR.4 16.5 0. length of stay.7 to 23.4 to 1.1) <0.2 to 0. 200632 20 1/10 2/10 210 (241. 200921 101 21 (2–24)† 19 (8. 201023 621 16.94%) 20.3 (27.8%) 5/107 (4.0 % (3. Sepsis-Related Organ Failure Assessment. and rash. 201031 121 32% 39% 27 0.65 ¶ Restricted activity. MV. d* Svoboda.53 Christ-Crain.9 9.9 6. and/or discontinuation of antibiotics in adult patients An Official Publication of the Society of Hospital Medicine Journal of Hospital Medicine Vol 00 | No 00 | Month 2013 5 .51 Stolz.0 6 9.5 20.3 to 5.34 ICU admission.45 Children ages 1–36 months with fever of unknown source Hospitalization rate Manzano.05 Days without MV Stolz. and rash. CTRL.2 to 8.5 22.2 to 4.2 (20. 201034jj 550 11 /59 (18.4 23. 200925‡ 1359 133/671 (19.1 8.7) — Briel. SOFA.1 days 21.1 to 1.7%) 22. but no sig.15 Bouadma.53 SOFA day 28 Bouadma.1 to 20.9 0.1) 0.0%) 6/119 (5.5–22. ICU.8–22.09 Jensen. ‡ Nausea.8% (29.4%) 60/688 (8.9 0.4) 0.12 Christ-Crain.6 (0. Procalcitonin-Guided Antibiotic Therapy | Soni et al TABLE 2. days Hochreiter.3 21.5 (23. 200629 302 20/151 (13. 200921 101 26 (7–21)† 26 (16. 200920 27 16.4 20.4%) 22.87 Antibiotic adverse events Schuetz. days Nobre.0001 Percent days in ICU with GFR <60 Jensen.1 to 224.0. glomerular filtration rate.5% vs 32.7 6 8.52 to 1.3 (25.1 19.0) 0.0 0.3) 0. 200827 458 8.1 40.3% (25.4 9. diarrhea.1%) 28.3 13.4 to 5.5%) 2861 (60.0 (295.6%) 16/101 (15.0) 0.7) NSS SAPS II score Hochreiter. NA.69 Kristoffersen.5% decrease in 28-day mortality.83).007 NOTE: Abbreviations: CI.06 Days on MV* Svoboda. pro- calcitonin. 200728 208 9 (1–15)† 10 (1–15)† 21 0. *All values are mean unless specified.0 (280.8%) 2. systemic inflammatory response syndrome.9) <0. 201023 621 26.9 11.7% (28. PCT.05 Neonates with sepsis Recurrence of infection Stocker. vomiting.35 (N 5 3492) addressed initiation to test a 7.5 to 1.9 to 3.2% (212.7 (22.53 Systemic infection. LOS.2 to 0.5 17.85 Stolz.9 8.6 6 3.4 (22.0 to 6.2 — Christ-Crain. P 5 0. 201133 1200 2796 (51.4 (26.2 6 10. 200632 20 3/10 7/10 240.1 days (22.2 to 21.2 16. 200827§ 458 2. 200430 224 10. 200922 110 40. control. days Chromik. 201133 1200 3569 (65. 200728 208 8/102 (7.0001). 200632 20 18 30 212 0. 200925 1359 43/671 (6. 200921 101 10 (0–18)† 8.1) 0.8 to 0.3 8. NSS. 1.9 14. SAPS.0001 Adult patients with respiratory tract infections Hospital LOS.0 21 (23. 200430 224 5/124 (4.7) 0.8 0. d Briel. 200920 27 7.6 to 5. Year N PCT* Control* Difference. 200819 79 17 23.5 (26.4 to 15. 201023 621 15.7) <0.5 (0–18)‡ 1.7 22.9) 0. 200922 110 15.0) 0.3)† 0 0.2 to 2. Summary of Morbidity Outcomes Outcome Author.1 (24.3%) 2187 (46.1) 0.0) 0.6 days 3.037 SOFA score max Schroeder. procalcitonin-guided intensification (31.4%) 5.6 (27. d* Schuetz. 201133 1200 6 (3–12)† 5 (3–11)† 1 0.0) 0.81 No SBI or neutropenia (N 5 312) 16/158 (10%) 11/154 (7%) 3 (23 to 10) 0.1) 0.0) 0. 201034 550 9.3 (23.7 6 3.9) >0. 200724 72 7.0) NSS Bouadma.1 26.8%) 11/106 (10.2 0.46 Bouadma. SIRS.4 1.0.4 to 0.07 Sepsis/SIRS. confidence interval.23 Hospital LOS. jj Antibiotic adverse events not defined.4) 0. 200629 302 12.0% (26. % Chromik.47 Critically ill adult patients: procalcitonin-guided antibiotic intensification ICU LOS. † Median (interquartile range). 200724 72 10.34 Adult postoperative patients at risk of infection Hospital LOS.046 Nobre.0 6 9.8%) 193/688 (28. % Schuetz.0 to 1.1% (24.2) 0. % Chromik.3 6 4.05 Burkhardt. intensive care unit.6 20.0 to 21.0 to 0.057 Local wound infection. not statistically significant.5 (20.7%) 4.3 28.0%) 21.0 to 2.34.25 (20.87 ICU-free days alive.9 20.4) 0. increase in days with abnormal renal function nificant difference in mortality was observed with (P < 0. Simplified Acute Physiology Score. 1–28 Stolz.5 0.4 to 7. nausea.96 Christ-Crain.5) 0.05 Burkhardt.

0% (22.29 In-hospital mortality Nobre. Meta-analysis of prescription rates from 8 and clinical signs and symptoms.Soni et al | Procalcitonin-Guided Antibiotic Therapy TABLE 3.07 NOTE: Abbreviations: CI. including community-acquired pneumonia. Neo- 2% to 7% and relative reductions ranging from 1.1% (24.1 (217. Settings included pri.4.1.83 Adult patients with respiratory tract infections 6-month mortality Stolz.1%) 3. 200728 208 5/102 (4.95 6/31 (19.5)* 0.9) 0.2%) 211.99 Stolz. 201133 1200 190/604 (31. 201023 621 92/307 (30.28 28-day mortality Jensen. procalcitonin. 2009 (30-day)25 1359 34/671(5.1) 0.1%) 21.1%) 33/688(4. emergency departments. with acute upper and lower respiratory tract infec.6 (24.8 to 6.4 to 9.7.2) 0. studies (Figure 3B) yielded a statistically significant tions. Year N PCT-Guided Therapy Control PCT-CTRL (95% CI) P Value Critically ill adult patients: procalcitonin-guided antibiotic discontinuation 28-day mortality Nobre.6 to 16. not available. NA.0%) 0.2) 0.81 60-day mortality Bouadma.0%) 28. 200827 458 0/231(0%) 1/224 (0. Physicians in control groups administered Morbidity antibiotics according to their own standard practices Procalcitonin guidance did not increase hospital LOS and/or evidence-based guidelines.5) 0. Absolute reduction in antibiotic duration ranged Procalcitonin guidance did not increase mortality.2 to 18.2) 0.2%) 64/314 (20.4) 0. (216.95 5/31 (16.6% (210.0%) 82/314 (26. CTRL.8% natal sepsis was suspected on the basis of risk factors to 72%. control. Mortality sure.6%) 14/50 (28.2%) 20. Meta-analysis of ICU admis- aged initiation of antibiotics with procalcitonin levels sion rates from 5 studies (Figure 3C) produced a risk >0. antibiotic prescription rate. Summary of Mortality Outcomes Mortality Mortality Difference Outcome Author. 201135 162 0/81 (0%) 0/81 (0%) 0 — Neonates with sepsis Mortality (in-hospital) Stocker.1) 0.1%) 9/40 (22. SBI. 200632 20 1/10 (10%) 3/10 (30%) 220 (254.29 Bouadma.4%) 0.25 ng/mL.3% (28.9%) 0.5%) 23.9 to 19.2%) 4/119 (3. difference of 21%.7 to 17.2 to 10.31 Burkhardt. (233. (224. 200922 110 15/57 (26. mary care clinics.9 (23.0) 0. and total antibiotic expo.35 days favoring procalcitonin (95% CI: 24.8 to 7.2) 0.5%) 8/40 (20.6 to 4.4%) 3/13 (23. and hospi- tal wards.9%) 9/106 (8.95 Schuetz. 1%).5 ng/mL. 200926 210 2/103(1.4) 0. 200920 27 3/14 (21.2 to 4.4) 0.9 (233.9%) 1.4%) 7/37 (18. *Per protocol analysis. PCT.96 Stolz. and relative reductions ranged from meta-analysis of 4 studies (Figure 3D) produced a risk 13% to 55%.8 (25.9 to 8.41 (218. 201036 384 All children 0% 0% 0 (0 to 0) Adult postoperative patients at risk of infection Mortality Chromik.1%) 6/37 (16. with a statistically significant pooled mean difference of 22.73 28-day mortality Christ-Crain.6 (217. and acute bronchitis.30 6-week mortality Christ-Crain.2) 0.3 to 3.6) 0.99 Schroeder. 200724 72 10/38 (26.3 to 0. 200921 101 10/51 (19. or ICU admission rates.32 Hochreiter.3%) 14/53 (26.0 to 14. 200430 243 4/124(3.2). Four of the 8 studies reported sufficient difference of 0.1 to 29. 201034 550 0/275(0%) 0/275 (0%) 0 — Kristoffersen. 201031 121 0% 0% 0 (0 to 0) NA Children ages 1–36 months with fever of unknown source Mortality Manzano.8) 0. and 4 studies strongly encouraged anti. 201023 621 65/307 (21. 0.3%) 13/34 (38. serious blood infection.38 to Neonates With Sepsis 20.92 Critically ill adult patients: procalcitonin-guided antibiotic intensification 28-day mortality Svoboda. reduced in the 4 studies reporting this outcome. 200819 79 8/39 (20.3 to 19. Total antibiotic exposure was consistently disease.4%) 20.3% with a narrow 95% CI (21% to details to be pooled into a meta-analysis (Figure 3A) 2%). 200629 302 18/151 (11.9%) 20/151 (13.7%) 12/50 (24.4%) 20.4%) 20.2%) 0. There was insufficient evidence to judge the effect on days of restricted activity or antibiotic Antibiotic Usage adverse events.3 (223. with no statistical heterogeneity (I2 5 0%).6 to 7.5%) 191/596 (32.3% (22.9%) 1/107 (0. Procalcitonin guidance reduced antibiotic duration. confidence interval. with a narrow 95% CI (24% to biotics with procalcitonin levels >0.7 to 5.1 to 2.0%) 20. pooled risk difference of 222% (95% CI: 241% to acute exacerbation of chronic obstructive pulmonary 24%). All studies encour.5%) 0.82 Briel. 200819 79 9/39 (23.54 Long. Procalcitonin guidance also reduced antibiotic One study31 (N 5 121) evaluated procalcitonin-guided prescription rate with absolute reductions ranging from antibiotic therapy for suspected neonatal sepsis.4% (21.33). 200921 101 8/51 (15. Antibiotic initiation 6 An Official Publication of the Society of Hospital Medicine Journal of Hospital Medicine Vol 00 | No 00 | Month 2013 .9) 0.2%) 21.0%) 28.5 (217.8%) 0. and from 1 to 7 days.

but the overall strength of hours (P 5 0. Unknown Source One study36 (N 5 384) evaluated procalcitonin-guided Antibiotic Usage antibiotic therapy for fever of unknown source in chil- Duration of antibiotic therapy was decreased by 22. Abbreviations: CI.012). standard deviation. and no sig- recurrence of infection was seen with procalcitonin. and the evidence was judged insufficient to draw conclusions. proportion of neonates on antibiotics 72 hours was reduced by 27% (P 5 0. or discontinuation was based on a procalcitonin nom. Procalcitonin-Guided Antibiotic Therapy | Soni et al FIG. guided antibiotic therapy (P 5 0.4 dren 1 to 36 months of age. this study was rated good. Morbidity Morbidity A statistically insignificant 7% reduction in rate of Rate of hospitalization was relatively low. nificant difference was seen between procalcitonin and guided antibiotic therapy (P 5 0. Meta-analyses of procalcitonin-guided antibiotic discontinuation in adult intensive care unit (ICU) patients. inverse variance weighted. An Official Publication of the Society of Hospital Medicine Journal of Hospital Medicine Vol 00 | No 00 | Month 2013 7 .49).002).1% in anti- neonates who were categorized as having possible or biotic prescription rate was seen with procalcitonin- infection or unlikely to have infection. IV. control groups. The largest reduction in Antibiotic Usage antibiotic duration was seen in the 80% to 85% of A statistically insignificant reduction of 3. and strength of evidence was rated moderate for antibiotic usage and insuffi. Children Ages 1 to 36 Months With Fever of cient for morbidity and mortality outcomes. confidence interval. SD. 2. a 24% relative reduction. Mortality ogram. Antibiotic therapy in the control group was No in-hospital deaths occurred in either the procalci- based on the physician’s assessment. The quality of tonin or control group.45).

standard deviation.5 ng/ One study32 (N 5250) monitored procalcitonin in mL) were randomized to receive prophylactic antibi- consecutive patients after colorectal surgery to identify otic therapy with ceftriaxone or no antibiotics. inverse variance weighted. confidence interval. IV. Meta-analyses of adults with respiratory tract infections.Soni et al | Procalcitonin-Guided Antibiotic Therapy FIG. Twenty Adult Postoperative Patients at Risk of Infection patients with elevated procalcitonin levels (>1. The 8 An Official Publication of the Society of Hospital Medicine Journal of Hospital Medicine Vol 00 | No 00 | Month 2013 . 3. prophylactic antibiotic therapy. Abbreviations: CI. Two hundred thirty patients had normal procalcitonin levels. SD. Mortality patients at risk of infection who might benefit from In-hospital mortality was reported as 0% in both arms.

we days with abnormal renal function. sions: procalcitonin-guided antibiotic decision making Clinical and microbiological evaluations are neither compared to clinical criteria-guided antibiotic decision sensitive nor specific for differentiating bacterial from making reduces antibiotic usage without increasing viral respiratory tract infections. randomized. Although these findings are systemic inflammatory response syndrome by 60% promising. Procalcitonin can morbidity or mortality. Despite these differences. the use of procalcitonin as an indicator included and pooled the Jensen et al. duration of antibiotic therapy. Comparison to Other Systematic Reviews ically ill patients is challenging because clinical criteria Six systematic reviews of procalcitonin guidance in for diagnosis overlap with noninfectious causes of the the management of patients with infections were pub- systemic inflammatory response syndrome. Previous systematic reviews therapy. needed. The authors con. and mortality in adult ICU patients. Our Limitations review showed that procalcitonin guidance signifi. discontinuation.07.5% who received prophylactic antibiotic therapy had a but was not statistically insignificant (P 5 0. guide initiation of antibiotic therapy in adults with suspected bacterial respiratory tract infection. respectively). study33 with of need for intensification of antibiotic therapy in other studies of adult ICU patients.9–14 Our systematic review of antibiotic therapy for presumed sepsis is necessary differs from past reviews in the number of studies while diagnostic evaluation is ongoing.007). included and the pooling of studies according to ing antibiotic therapy is associated with increased patient population. controlled trials. whereas ours included 18 otics. either guidance significantly reduced antibiotic usage in adult for initiation. an increase in rately from adult patients. The incidences of local and systemic infec. We addressed pediatric populations sepa- increased morbidity: a longer ICU LOS. whereas patients with normal pro- Morbidity calcitonin levels did not require any additional surgi- Procalcitonin guidance reduced the incidence of sepsis/ cal or medical therapy. Summary of the Main Findings Diagnosis of sepsis or other serious infections in crit. tainty about the noninferiority margin for morbidity otic prescription rate. P 5 0. defined a margin for noninferiority for 28. there was insufficient evidence that procalci- tonin guidance does not increase morbidity or DISCUSSION mortality. Meta-analysis of all 5 ICU studies showed populations is less clear. not statistically significant (P 5 0. We evaluated the adult ICU patients should be discouraged because this Jensen et al. or intensification of ICU patients by reducing the duration of antibiotic antibiotic therapy. and recognizing that there days on mechanical ventilation.43% in mortality and a An Official Publication of the Society of Hospital Medicine Journal of Hospital Medicine Vol 00 | No 00 | Month 2013 9 . and an increase in are distinct groups within the pediatric population. Initiation lished prior to our review. Procalcitonin-Guided Antibiotic Therapy | Soni et al strength of evidence was judged insufficient to draw colorectal surgery patients reported that elevated pro- conclusions from this study.53. One study31 40%.07). looked at procalcitonin-guided antibiotic intensifica- The large.37–39 Our review found that procalcitonin different uses of procalcitonin measurements. calcitonin levels may identify patients at risk for infec- tion who benefit from prophylactic antibiotic Antibiotic Usage therapy. more data in postoperative patients are (p50. Only the and total exposure to antibiotic therapy in adult Bouadma et al.32 Patients with elevated procalcitonin levels Duration of antibiotic therapy was reduced by 3. However. study separately because it uniquely approach was associated with increased morbidity. P 5 0. in neonates with suspected sepsis showed a significant decrease in the proportion of neonates started on Mortality empiric antibiotic therapy and a decrease in the dura- Mortality was 20% higher in the control arm but was tion of antibiotic therapy with procalcitonin guidance. in contrast to other studies that antibiotic intensification in response to elevated that looked at procalcitonin-guided antibiotic discon- procalcitonin measurement was associated with tinuation. One study in postoperative a pooled point estimate of 20.and 60-day The role of procalcitonin-guided therapy in other mortality. type and severity of infection. without increasing morbidity or mortality. rather than decreasing the initiation of antibi. One previous review13 In contrast. and mortality. our review and explained by clinical harms of increased exposure to other systematic reviews. An important limitation of this review was the uncer- cantly reduced antibiotic usage with respect to antibi. we came to similar conclu- broad-spectrum antibiotics. significant decrease in the incidence and severity of systemic infections. and significant gap in the present literature. tion were reduced with procalcitonin guidance but The utility of procalcitonin in pediatric settings is a were not statistically significant (10%. study23 did a power analysis and pre- patients with respiratory tract infections.27). because delay. separately grouped neonates and children ages 1 to 36 cluded that the increased morbidity could only be months. included 7 to 14 studies. well-designed study by Jensen33 showed tion in adult ICU patients.

Total antibiotic exposure is conventionally risk for infection. differences. prognostic markers and adverse events and emergence of antibiotic resistance mediators. et al. Eur Respir J. 2 large trials are in pro. procalcitonin guidance can sig- may be too high. but algorithm adherence was sig.40. Samson DJ. 78. nostic marker of ventilator-associated pneumonia. Rockville. a systematic review and meta-analysis. evidence to recommend procalcitonin-guided antibi- Differences in reporting of total antibiotic exposure otic therapy in neonates with sepsis. nificantly less in the United States. children with and morbidity outcomes limited our ability to pool fever of unknown source. Effect of procalcitonin- events. and Chronic Health Evaluation II. our systematic review found that Reviews. 2011. Arch Intern Med.44. 12. 2009. Biomarkers of sepsis.(9):CD007498. 11. Gauvin F. Newer observational studies Prepared by the Blue Cross and Blue Shield Association Technology Evaluation Center Evidence-based Practice Center under contract have demonstrated reduced antibiotic usage with no. Galaydick JL. Intensive Care Med. Available at: www. Christ-Crain M. Dandona P. studies are needed. 10(11)-EHC063-EF.effectivehealthcare.38(11):2229–2241. AHRQ publication no. Pierrakos C. such as cystic fibrosis. Infection. Agency for Healthcare Research and Quality. Crit Care Med. Kontogiorgi M. A 10% noninferiority margin for mortality has nin should not be used to guide intensification of anti- been recommended by the Infectious Diseases Society biotic therapy in adult ICU patients because this of America and American College of Chest Physicians. Simplified Acute official position of AHRQ or of the US Department of Health and Physiology (SAP) II. Shen H. Heinz U. 9. Likewise. Kopterides P. Greenwald JL. 2007.37.45 final. Forycki ZF. 1998. 2011. Soni NJ. Finally. Marshall JC. Clin Infect allergic reactions or adverse events of antibiotic ther.42 One randomized controlled trials and recommendations for clinical algo- rithms. opportunities for future research. Guerin V. 2011. Pitrak DL. study-based algorithms.38:940–949. generalizing results from 14.171(1):48–53. Dis. Sepsis biomarkers: a review. Agarwal R. 2. Schuetz P. There is insufficient ity in the future. Luyt CE. Chiesa C. Biomarkers in respiratory tract infections: cally. Huang T. 206–217. from current randomized controlled studies. Muller B. Research Gaps 14(1):R15. implementation of procalcitonin algorithms in real-life Comparative effectiveness review No. 4. Methods Guide for Effectiveness and Comparative Effectiveness In summary. 8. Armaganidis A. Cochrane Database Syst Rev 2012. Chiappa V. Nix D. Reinhart K. nificantly reduce antibiotic usage without adversely gress that may yield more precise estimates of mortal. 2009. single-arm study describes how procalcitonin can be 13. Clin been associated with reductions in antibiotic adverse Infect Dis. such as Clostridium difficile colitis and super. Tsangaris I. Matthaiou DK. SAP III. meta-analysis.26(3):664–672. Crit Care. Serum procal- citonin and C-reactive protein levels as markers of bacterial infection: a significant percentage of community-acquired respi. 2005. et al.43 but additional procalcitonin-guided antibiotic therapy algorithms in adult critically ill patients. Vats V. cle are responsible for its content. In adults with but there is concern that a 10% margin for mortality respiratory infections. guided treatment in patients with infections: a systematic review and infection with multi-drug resistant Gram-negative bac. The authors of this arti- different severity of illness scales. Aronson N. Pro- calcitonin-guided algorithms of antibiotic therapy in the intensive care ance. 12(13)-EHC124-EF. Jing J. 290–2007-10058-I. teria. Tsantes A. Kinetics of procalcitonin in iat- efficacy of procalcitonin-guided antibiotic therapy rogenic sepsis. 10 An Official Publication of the Society of Hospital Medicine Journal of Hospital Medicine Vol 00 | No 00 | Month 2013 . Procalcitonin to guide duration of antimi- crobial therapy in intensive care units: a systematic review. affecting morbidity or mortality. Procalcito- care. chronic conditions. Briel M. 2004. Wilson MF.39(2): ratory tract infections and are often treated empiri. An ESCIM systematic review and meta-analysis of the duration of antibiotic therapy. Cui W. No statement in this article should be construed as an Related Organ Failure Assessment. Intensive Care Med. II. Dimopoulos G. et al. Panero A. Saint-Louis P. We identified gaps in the available literature and 37(7):2290–2298.Soni et al | Procalcitonin-Guided Antibiotic Therapy 95% CI of 26% to 5% for difference in mortality cantly reduce antibiotic usage in adult ICU patients between procalcitonin-guided therapy versus standard without affecting morbidity or mortality. Christ-Crain M.34 and only 1 reported antibiotic resist. Lacroix J. standardized reporting of antibiotic diagnostic guides to antibiotic prescription. J Clin Endocrinol Metab. Procalcitonin increase after needs to be studied in patient populations excluded endotoxin injection in normal subjects. Strategies to reduce antibiotic usage have centrations for the diagnosis of sepsis in critically ill neonates. Procalcitonin kinetics as a prog- immunocompromised patients and pregnant women.30 Second. and Quality. Amre DK. 5.19 Third. Simon L. including any clinical treatment rec- ommendations. Poulakou G. Procalcitonin-guided antibiotic therapy. morbidity was reported with US Department of Health and Human Services. Rossi N. 16. or postoperative patients at data. including Sepsis. AHRQ publication no.41 Few studies in our review reported 10. such as structured implementation of practice rithms for antibiotic therapy decisions: a systematic review of guidelines and antibiotic stewardship programs. Brunkhorst FM. Siempos. Reliability of procalcitonin con- is needed. Schwartz DN. 2010.53(4):379–387. There are no conflicts of interest reported by any of the authors. Am J Respir Crit Patients who are immunocompromised or have Care Med. approach may increase morbidity.24(8):888–889.171(15):1322–1331. MD: Agency for Healthcare Research settings in Europe.27. procalcitonin guidance should be unit: a systematic review and meta-analysis of randomized controlled compared to other strategies to reduce antibiotic trials. Rockville. Muller B. such as 1994. et al. Presently. would depend on similar use of and adherence to 15. 2010.cfm. Clin Infect Dis.79(6):1605–1608.ahrq. which limited com- parisons across studies. First. and Acute Physiology Human Services. 7. Published Accessed October 2012. apy. reported as mean days per 1000 days of follow-up. Procalcitonin algo- usage. 25. account for 6. 1998. Tang H. Vincent JL. MD: procalcitonin-guided antibiotic therapy can signifi. Schuetz P.30(3):556–573.37(6):497–507. References 1. Armaganidis used in antibiotic stewardship programs to decrease A. Procalcitonin to initiate or those studies that were conducted primarily in Europe discontinue antibiotics in acute respiratory tract infections.29. Ntani G. but some studies only reported relative or absolute Disclosures: This project was funded under contract HHSA 290–2007- 10058 from the Agency for Healthcare Research and Quality (AHRQ). Combes A. the safety and 3. Crit Care Med.

Cousineau J. Mittelkotter U. et al. Sogaard OS.168(18):2000–2007. Nobre V. Procalcitonin guidance of sepsis is an antimicrobial stewardship program (ASP). Graf JD. Procalcitonin to guide 819–824. Tubach F. Am J Prev vention trial. based guidelines vs. et al. et al. procalcitonin-guidance with standard therapy. Eur J Clin antibiotic therapy in community-acquired pneumonia: a randomized Microbiol Infect Dis. Chromik AM. 44.29(6):1109–1115. et al. trial): a multicentre randomised controlled trial. lower respiratory tract infections: cluster-randomised. admitted with suspected sepsis. Fraser VJ. Effectiveness and safety of 30. Procalcitonin-guided interven- nin to shorten antibiotic treatment duration in septic patients: a tions against infections to increase early appropriate antibiotics and randomized trial. Crit Care. Current methods of the US in suspected neonatal early-onset sepsis: prospective randomized inter- Preventive Services Task Force: a review of the process. Lancet. Kantorova I. nin (PCT) to guide discontinuation of antibiotic use in an unspecified 29. et al. Procalcitonin-guided antibiotic antibiotic therapy for ventilator-associated pneumonia in adults: a use vs a standard approach for acute respiratory tract infections in pri. Langenbecks Arch Surg. Wolff M. Sherman G. Lohr KN. Schuetz P. Reith HB. 2009. Chlebicki MP. AHRQ series paper 5: grading Pre-emptive antibiotic treatment vs “standard” treatment in patients the strength of a body of evidence when comparing medical interven. Langenbecks Arch Surg. Smyrnios N. Christ-Crain M. 18. Ward S. Lundgren B. Bingisser R. Long W. Barza M. Burkhardt O. Holeckova K. 2010. Batschwaroff M. Infect Control Hosp Epidemiol. 2008. 2004. Eur J Clin of procalcitonin-guided decision-making to shorten antibiotic therapy Microbiol Infect Dis. 2006. Schweiger AM.131(1): 43. et al. et al. Schroeder S. Eggimann P. 26. improve survival in the intensive care unit: a randomized trial. health-care program. 2012.20(3 suppl):21–35. Wegscheider K. Schuetz P. Use of procalcito- 9–19.174(1):84–93. et al. Use of procalcitonin to reduce aged 1 to 36 months presenting with fever without source: a random- patients’ exposure to antibiotics in intensive care units (PRORATA ized controlled trial. ized trial. with community-acquired pneumonia. 164(3):396–402. Chastre J. Impact of procalcitonin on the management of children 23. Ewig S. Infectious Diseases Soci- 2007–2008. Christ-Crain M. Zhang Y.34(6):1364–1375. Eur randomized study. Manzano S. guided algorithm reduces length of antibiotic treatment in surgical 34. Radvan M. et al. Clin Infect Dis. standard guidelines on antibiotic use in lower 39. calcitonin algorithm to guide antibiotic therapy in respiratory tract 31. et al.13(3):R83. discussion 42. Clin Microbiol Infect. 32. 2009. Eur Respir J. Albrich WC. Inadequate antimicrobial respiratory tract infections: the ProHOSP randomized controlled trial. 7 years. et al. Thomann R. Stolz D. Harbarth S. 21. 2012. Harris RP. Procalcitonin for reduced 35. Use infections outside of study conditions: a post-study survey. controlled trial comparing antimicrobial stewardship. Crit 20. Christ-Crain M. Care Med. 2003. Haagen U. Procalcitonin (PCT). Can pro. 2007. 33. 1999. Lee W. Neonatology. Hochreiter M. et al. 2001. ically ill patients. Jaccard-Stolz D. Favorable impact of interruption of suspected lower respiratory tract infections based on a a multidisciplinary antibiotic management program conducted during single procalcitonin measurement at hospital admission—a random. Bouadma L.16(5): 22. 2009. Briel M.44(2):159–177. Mueller B. Endter F. trial. Am J Respir Crit Care Med. Rohner P. Atkins D. Dusemund F. An Official Publication of the Society of Hospital Medicine Journal of Hospital Medicine Vol 00 | No 00 | Month 2013 11 . Effect of procalcitonin. Med. 2001. mary care. Xie J. Kollef MH. Diagnostic value of multiple trauma or major surgery? Hepatogastroenterology. Svoboda P. Procalcitonin antibiotic exposure in ventilator-associated pneumonia: a randomised guidance for reduction of antibiotic use in low-risk outpatients study.” Arch Intern Med. 37. Use of procalcito. Procalcitonin guidance and intensive care patients with severe sepsis: results of a prospective reduction of antibiotic use in acute respiratory tract infection. procalcitonin. 19. Am J Respir Crit Care Med. Lancet. Stolz D. Dellit TH. Thiede A. Tang J. ety of America and the Society for Healthcare Epidemiology of Amer- 28. 2010. Hein L. duration of antibiotic therapy in intensive care patients: a randomized 36.290(19):2588–2598. 2007.394(2):221–226. with elevated serum procalcitonin levels after elective colorectal sur- tions—agency for healthcare research and quality and the effective gery: a prospective randomised pilot study. Lu G. Koehler T.302(10):1059–1066. 2010. Bucher B. Chest. randomized trial. 2011. 41. Scheer P. Arch Intern Med. Am J Emerg Med. Jensen JU.177(5):498–505. Respirology. et al. Bingisser R. Experience with a clinical guideline for the treatment of ventilator- 24. Terrin N.24(9):699–706. Schuetz P. 375(9713):463–474. 2009. Ward S. Uhl W. Stocker M. Kwa AL. Owens RC. Hsu LY. Wejse C. Bailey B. Carling P.97(2):165–174.363(9409):600–607. Comparison of 8 vs 15 days of 27. calcitonin help us in timing of re-intervention in septic patients after 38. Stolz D. 2009. 2011. Fagon JY. 2010. et al. Effect of procalcitonin-guided antibiotic therapy in lower respiratory tract procalcitonin-guided treatment on antibiotic use and outcome in infections in “real life. Killion A. Antibiotic treatment 40. Kohler T. 2007. Am J Respir Crit Care Med. Fung T. Hochreiter M. 2001. Respir J. et al. Woolf SH.15(5):481–487. Procalcitonin-Guided Antibiotic Therapy | Soni et al 17. Antibiotic treatment of ica guidelines for developing an institutional program to enhance exacerbations of COPD: a randomized. 25. Fraser VJ.63(5):513–523. Delvin E. Dusemund F. Kollef MH. Chest. and interleukin-8 in critically ill patients 54(74):359–363.391(3):187–194. Pugin J. Effectiveness of a pro- intervention trial. et al. Bingisser R. Sherman G. Liew YX. Girodias JB. Radvanova J.29(3):269–277. Christ-Crain M.115(2):462–474. 2011. Galetto-Lacour A. et al. 2006. McGowan JE Jr.172(9):715–722.39(9):2048–2058. et al. Harbarth S. JAMA. Crit Care Med. Ibrahim EH. Schaiff R. et al. prospective controlled trial. Froidevaux C.30(7):853–855. El Helou S. interleukin-6. single-blinded 45. Kristoffersen KB.28(6):647–653. Deng X. Luyt CE. Berger TM. Fontana M. Helfand M. treatment of infections: a risk factor for hospital mortality among crit- JAMA. Owens DK. 2008. 2010. associated pneumonia.36(3):601–607. 2003. J Clin Epidemiol.