AACN Advanced Critical Care Volume 26, Number 2, pp. 99-106 © 2015, AACN
Procalcitonin-Guided Antibiotic Therapy Duration in Critically Ill Adults Caroline Walker, RN, MSN, BSN, AGACNP-BC
ABSTRACT Procalcitonin is a promising biomarker for antibiotic therapy because its levels rise and fall quickly with bacterial infections. A multidatabase literature search was reviewed with 3 primary prospective randomized control trials used in further analysis. The results indicated that a procalcitonin-guided antibiotic protocol reduces the number of days a patient has to take antibiotics while having no effect on mortality when compared with control groups. Short-term studies did not show a difference in the intensive care unit length of stay, infection relapse rate, super-
infection rate, or multidrug-resistant bacteria rate between the procalcitonin-protocol and control group. Because procalcitonin-guided antibiotic therapy has been shown to reduce the duration of treatment with antibiotics in critically ill patients without worsening the mortality rate or other outcomes, the implementation of a procalcitonin-guided antibiotic therapy should be considered for patients with proven or highly suspected bacterial infections in the intensive care unit. Keywords: antibiotic duration, critically ill, procalcitonin, sepsis
A
reduction of antibiotic-resistant bacteria.3 The purpose of this review is to determine whether the use of procalcitonin-guided antibiotic therapy for critically ill patients in the ICU reduces patients’ exposure to antibiotics without worsening outcomes.
ntibiotic stewardship has been gaining ground in health care decision making. Antibiotic stewardship involves judicial use and careful monitoring of antibiotic regimens to ensure that every patient receives optimal antibiotic therapy.1 One way to implement antibiotic stewardship would be to reduce the length of time patients receive antibiotics. Research on using procalcitonin-guided antibiotic therapy in patients who are not critically ill to reduce antibiotic use has been promising.2 However, less research has been done on the use of procalcitonin-guided antibiotic therapy for critically ill patients to reduce their exposure to antibiotics. Reducing antibiotic use for critically ill patients can be a difficult decision for providers.3 Because antibiotics are some of the most widely prescribed medications in the intensive care unit (ICU), providers can play a significant role in antibiotic stewardship and the
What Is Procalcitonin? Procalcitonin has emerged as a promising biomarker for bacterial infections. In healthy individuals, procalcitonin is produced in the thyroid parafollicular cells (C cells) as a precursor to
Caroline Walker is Adult Gerontology Acute Care Nurse Practitioner, University of Pennsylvania School of Nursing, 418 Curie Boulevard, Philadelphia, PA 19104 (caroline.walker@ alumni.upenn.edu). The author declares no conflicts of interest. DOI: 10.1097/NCI.0000000000000079
99 Copyright © 2015 American Association of Critical-Care Nurses. Unauthorized reproduction of this article is prohibited.
NCI-D-14-00040_LR.indd 99
07/04/15 7:15 PM
W W W.A A CNA D VA NCE D CRIT ICA LCA RE .COM
WA LK E R
calcitonin, which regulates calcium and phosphate in the body. The 116-amino-acid procalcitonin is cleaved into the 32-amino-acid calcitonin before the hormone leaves the thyroid cell. Only a very small amount of procalcitonin escapes the thyroid cells into the bloodstream; therefore, healthy individuals should have only 0.02 ng/mL of procalcitonin in their blood.4 During a bacterial infection, procalcitonin levels can increase. Bacterial endotoxins stimulate macrophages, which cause the release of interleukin-1β, tumor necrosis factor-α, and interleukin-6.5 Animal studies show that these mediators stimulate nonthyroid tissues, including the liver, kidneys, pancreas, adipocytes, and muscle tissue, to produce procalcitonin.4,6 Because these cells lack the ability to cleave procalcitonin into calcitonin, a large amount of procalcitonin leaves the cells and circulates in the plasma.4 In as little as 3 to 6 hours, procalcitonin levels can range from 0.05 ng/mL to as high as 10 ng/mL or greater, depending on the severity of the infection.7 In addition, procalcitonin is not produced in response to viral infections because of interferon gamma.5 Research also has shown that procalcitonin levels are not affected by corticosteroids, nonsteroidal anti-inflammatory drugs, or the number of circulating white blood cells.3,5 However, conditions with a large stress response, such as trauma, surgery, cardiac shock, and systemic inflammatory response syndrome, can increase procalcitonin levels without a bacterial infection. Antibiotic pretreatment and early or localized bacterial infections might show misleadingly low levels of procalcitonin. Thankfully, most of these conditions cause only a moderate rise or depression in procalcitonin levels, and a rapid change is usually spotted with repeat testing in as little as 6 hours.5 Nevertheless, procalcitonin remains an exciting biomarker for bacterial infections. Procalcitonin levels can rise in as quickly as 3 hours, and the half-life of procalcitonin can range from 22 to 26 hours, which makes it an ideal biomarker for testing.6 As the infection resolves, the procalcitonin level falls back to normal. Numerous quantitative and semiquantitative testing options can produce procalcitonin level results in as little as 30 minutes.8 A recent assay has a functional sensitivity of 0.06 μg/L.3,9 Procalcitonin has an 89% sensitivity (true positives identified) and 94% specificity (true negatives identified) when used to diagnose
sepsis.3 Procalcitonin’s ability to predict sepsis (94% positive predictive value) and its probability to rule it out (90% negative predictive value) are remarkably high.3 Procalcitonin not only can be used to identify patients with a bacterial infection but also can reflect the severity of the infection and the volume of bacterial load.5 Most patients with sepsis will have a procalcitonin level between 0.5 and 2 ng/L, although clinically significant lower respiratory tract infections can demonstrate levels between 0.25 and 0.5 ng/L.9 Literature Review
A literature review was conducted using 3 databases: PubMed, the Cumulative Index of Nursing and Allied Health (CINAHL), and Scopus. PubMed was searched using the phrase procalcitonin antibiotic duration sepsis, resulting in 34 articles, 9 of which were included in this review. On CINAHL, the term procalcitonin was searched with the filters abstract available, peer reviewed, and published between 2009-2014, which resulted in 150 articles, of which 3 articles were used in this article, with 1 as a duplicate from the PubMed results. On Scopus, the term procalcitonin AND sepsis AND antibiotics was searched with the filter 2009-2014, limited to medical and nursing articles and reviews written in English. Of the returned 650 articles, 11 were included in this review. Additional articles were found using back tracing of citations from obtained articles. Articles that were excluded from the review were not specific for the critically ill population, focused on procalcitonin as solely a diagnostic marker, evaluated the role of procalcitonin in noninfectious conditions, or were classified as a commentary or letters to editors. Non-English articles and articles focused on pediatric or neonatal populations were also excluded. Three studies were used for the primary analysis—Deliberato and colleagues,10 Bouadma and colleagues,11 and Hochreiter and colleagues.12 These 3 studies were chosen for their focus on critically ill adult patients with suspected or proven bacterial infections, including but not limited to sepsis, with comparable protocols and endpoints of durations of treatment with antibiotics, mortality, and ICU length of stay. Other studies were excluded from the primary analysis but were included as references, because of their focus on only patients with sepsis.
100 Copyright © 2015 American Association of Critical-Care Nurses. Unauthorized reproduction of this article is prohibited.
NCI-D-14-00040_LR.indd 100
07/04/15 7:15 PM
V OL UME 2 6 • N U MBER 2 • APRIL- JUNE 2015
P ROCA LCIT ONIN- GUID E D D URAT ION OF A NT IBIOT IC T H E RA P Y
Primary Analysis
Patients included in the studies for primary analysis10-12 were male and female adults from medical and surgical ICUs with suspected or proven serious bacterial infections. The patients were critically ill and had a wide range of comorbidities. Admission diagnoses for the 3 studies mainly included sepsis, severe sepsis, septic shock, pulmonary infections, urinary system infections, and peritonitis.10-12 The most common exclusion criteria in the 3 studies were patients who started antibiotic therapy more than 24 hours before the study, more than 48 hours before cultures were taken, or before admission to the ICU. Also, the studies excluded patients who needed long-term treatment with antibiotics for infections (such as infective endocarditis), patients with nonbacterial infections, patients with minor or localized infections, patients younger than 18 years, patients with known pregnancy, or patients who did not give consent.10-12 Patients were randomized into a procalcitonin-protocol group or a control group, with standard antibiotic care at the discretion of the provider. All 3 articles used comparable criteria for their procalcitonin protocols. See Table 1 for individual study information. Findings
The findings of the 3 studies are not entirely consistent, but they are encouraging for the use of procalcitonin-guided antibiotic therapy.10-12 Mounting evidence has shown that the procalcitonin protocol reduces the duration of antibiotic therapy for adults in the ICU without affecting mortality rate. However, studies have not yet shown a decrease in ICU length of stay and rates of infection relapse, superinfection, or multidrug-resistant bacteria during a procalcitonin-guided antibiotic therapy protocol. Duration of Antibiotic Therapy
All 3 studies found a reduction in the duration of days of treatment with antibiotics between the procalcitonin-protocol and control groups. The largest study, by Bouadma and colleagues,11 found that use of the procalcitonin protocol reduced the duration of antibiotic therapy by 3 days when compared with the duration of antibiotic therapy in the control group (10.3 days vs 13.3 days; P < .0001), even with a limited adherence rate to both procalcitonin and control protocols. In addition, the procalcitonin-protocol group in the study by Hochreiter and col-
leagues12 had a reduction of 2 antibiotic-treatment days (5.9 days vs 7.9 days; P < .0001). However, the intent-to-treat analysis in the study by Deliberato and colleagues10 did not result in a statistically significant difference in antibiotic-therapy duration, which may be due to the limited protocol-adherence rates found in the study, with only 47% of procalcitonin group patients and 79% of control group patients following protocol. When a separate per-protocol analysis was performed, Deliberato and colleagues10 found a statistically significant 4-day difference between the median days of antibiotic-therapy duration in the procalcitonin and control groups (9 days for the procalcitonin-protocol group vs 13 days for the control group; P = .008). The fewer antibiotic-therapy days in the study by Hochreiter and colleagues,12 5.9 days compared with 9 to 10 days in the other studies, may be explained by the discontinuation of antibiotic therapy when procalcitonin levels were less than 1 ng/mL in that study12 versus less than 0.5 ng/mL in the other 2 studies.10,11 Despite using different procalcitonin levels for the discontinuation of antibiotic therapy, these researchers found a consistent decrease in antibiotic usage when compared with the control groups. These results provide encouraging evidence that procalcitoninguided antibiotic therapy reduces patients’ antibiotic exposure (see Figure 1). Mortality
No statistically significant difference was found in mortality rate when comparing the procalcitonin-protocol and control groups in any of the 3 studies.10-12 However, Bouadma and colleagues11 reported both a 28-day and 60-day mortality rate, whereas Deliberato and colleagues10 and Hochreiter and colleagues12 reported only in-hospital mortality rate. More evidence supports no change to shortterm mortality rate than evidence supporting no change to long-term mortality rate. Bouadma and colleagues’11 28-day mortality rate (21.2% procalcitonin vs 20.4% control, which is within the confidence interval), Deliberato and colleagues’10 in-hospital mortality rate (4.8% procalcitonin vs 10.3% control, P = 0.42), and Hochreiter and colleagues’12 inhospital mortality rate (26.3% procalcitonin and 26.4% control) all support the conclusion that there is no change in the short-term mortality rate. Interestingly, no in-hospital deaths were reported for either the procalcitonin or
101 Copyright © 2015 American Association of Critical-Care Nurses. Unauthorized reproduction of this article is prohibited.
NCI-D-14-00040_LR.indd 101
07/04/15 7:15 PM
W W W.A A CNA D VA NCE D CRIT ICA LCA RE .COM
WA LK E R
Table 1: Study Details Study
Deliberato et al10
Bouadma et al11
Hochreiter et al12
Year
2013
2010
2009
Design
Randomized
Randomized
Randomized
Size
81 patients: 42 PCT vs 39 control; per protocol: 20 PCT vs 31 control
621 patients: 307 PCT vs 314 control
110 patients: 57 PCT vs 53 control
Site
Mixed medical-surgical ICU in Brazil
5 medical ICUs, 2 surgical ICUs, 1 mixed medical-surgical ICU in 5 French hospitals
Surgical ICU in Germany
Procalcitonin protocol
Discontinue antibiotic therapy when procalcitonin level
Procalcitonin-Guided Antibiotic Therapy Duration in Critically Ill Adults Caroline Walker, RN, MSN, BSN, AGACNP-BC
ABSTRACT Procalcitonin is a promising biomarker for antibiotic therapy because its levels rise and fall quickly with bacterial infections. A multidatabase literature search was reviewed with 3 primary prospective randomized control trials used in further analysis. The results indicated that a procalcitonin-guided antibiotic protocol reduces the number of days a patient has to take antibiotics while having no effect on mortality when compared with control groups. Short-term studies did not show a difference in the intensive care unit length of stay, infection relapse rate, super-
infection rate, or multidrug-resistant bacteria rate between the procalcitonin-protocol and control group. Because procalcitonin-guided antibiotic therapy has been shown to reduce the duration of treatment with antibiotics in critically ill patients without worsening the mortality rate or other outcomes, the implementation of a procalcitonin-guided antibiotic therapy should be considered for patients with proven or highly suspected bacterial infections in the intensive care unit. Keywords: antibiotic duration, critically ill, procalcitonin, sepsis
A
reduction of antibiotic-resistant bacteria.3 The purpose of this review is to determine whether the use of procalcitonin-guided antibiotic therapy for critically ill patients in the ICU reduces patients’ exposure to antibiotics without worsening outcomes.
ntibiotic stewardship has been gaining ground in health care decision making. Antibiotic stewardship involves judicial use and careful monitoring of antibiotic regimens to ensure that every patient receives optimal antibiotic therapy.1 One way to implement antibiotic stewardship would be to reduce the length of time patients receive antibiotics. Research on using procalcitonin-guided antibiotic therapy in patients who are not critically ill to reduce antibiotic use has been promising.2 However, less research has been done on the use of procalcitonin-guided antibiotic therapy for critically ill patients to reduce their exposure to antibiotics. Reducing antibiotic use for critically ill patients can be a difficult decision for providers.3 Because antibiotics are some of the most widely prescribed medications in the intensive care unit (ICU), providers can play a significant role in antibiotic stewardship and the
What Is Procalcitonin? Procalcitonin has emerged as a promising biomarker for bacterial infections. In healthy individuals, procalcitonin is produced in the thyroid parafollicular cells (C cells) as a precursor to
Caroline Walker is Adult Gerontology Acute Care Nurse Practitioner, University of Pennsylvania School of Nursing, 418 Curie Boulevard, Philadelphia, PA 19104 (caroline.walker@ alumni.upenn.edu). The author declares no conflicts of interest. DOI: 10.1097/NCI.0000000000000079
99 Copyright © 2015 American Association of Critical-Care Nurses. Unauthorized reproduction of this article is prohibited.
NCI-D-14-00040_LR.indd 99
07/04/15 7:15 PM
W W W.A A CNA D VA NCE D CRIT ICA LCA RE .COM
WA LK E R
calcitonin, which regulates calcium and phosphate in the body. The 116-amino-acid procalcitonin is cleaved into the 32-amino-acid calcitonin before the hormone leaves the thyroid cell. Only a very small amount of procalcitonin escapes the thyroid cells into the bloodstream; therefore, healthy individuals should have only 0.02 ng/mL of procalcitonin in their blood.4 During a bacterial infection, procalcitonin levels can increase. Bacterial endotoxins stimulate macrophages, which cause the release of interleukin-1β, tumor necrosis factor-α, and interleukin-6.5 Animal studies show that these mediators stimulate nonthyroid tissues, including the liver, kidneys, pancreas, adipocytes, and muscle tissue, to produce procalcitonin.4,6 Because these cells lack the ability to cleave procalcitonin into calcitonin, a large amount of procalcitonin leaves the cells and circulates in the plasma.4 In as little as 3 to 6 hours, procalcitonin levels can range from 0.05 ng/mL to as high as 10 ng/mL or greater, depending on the severity of the infection.7 In addition, procalcitonin is not produced in response to viral infections because of interferon gamma.5 Research also has shown that procalcitonin levels are not affected by corticosteroids, nonsteroidal anti-inflammatory drugs, or the number of circulating white blood cells.3,5 However, conditions with a large stress response, such as trauma, surgery, cardiac shock, and systemic inflammatory response syndrome, can increase procalcitonin levels without a bacterial infection. Antibiotic pretreatment and early or localized bacterial infections might show misleadingly low levels of procalcitonin. Thankfully, most of these conditions cause only a moderate rise or depression in procalcitonin levels, and a rapid change is usually spotted with repeat testing in as little as 6 hours.5 Nevertheless, procalcitonin remains an exciting biomarker for bacterial infections. Procalcitonin levels can rise in as quickly as 3 hours, and the half-life of procalcitonin can range from 22 to 26 hours, which makes it an ideal biomarker for testing.6 As the infection resolves, the procalcitonin level falls back to normal. Numerous quantitative and semiquantitative testing options can produce procalcitonin level results in as little as 30 minutes.8 A recent assay has a functional sensitivity of 0.06 μg/L.3,9 Procalcitonin has an 89% sensitivity (true positives identified) and 94% specificity (true negatives identified) when used to diagnose
sepsis.3 Procalcitonin’s ability to predict sepsis (94% positive predictive value) and its probability to rule it out (90% negative predictive value) are remarkably high.3 Procalcitonin not only can be used to identify patients with a bacterial infection but also can reflect the severity of the infection and the volume of bacterial load.5 Most patients with sepsis will have a procalcitonin level between 0.5 and 2 ng/L, although clinically significant lower respiratory tract infections can demonstrate levels between 0.25 and 0.5 ng/L.9 Literature Review
A literature review was conducted using 3 databases: PubMed, the Cumulative Index of Nursing and Allied Health (CINAHL), and Scopus. PubMed was searched using the phrase procalcitonin antibiotic duration sepsis, resulting in 34 articles, 9 of which were included in this review. On CINAHL, the term procalcitonin was searched with the filters abstract available, peer reviewed, and published between 2009-2014, which resulted in 150 articles, of which 3 articles were used in this article, with 1 as a duplicate from the PubMed results. On Scopus, the term procalcitonin AND sepsis AND antibiotics was searched with the filter 2009-2014, limited to medical and nursing articles and reviews written in English. Of the returned 650 articles, 11 were included in this review. Additional articles were found using back tracing of citations from obtained articles. Articles that were excluded from the review were not specific for the critically ill population, focused on procalcitonin as solely a diagnostic marker, evaluated the role of procalcitonin in noninfectious conditions, or were classified as a commentary or letters to editors. Non-English articles and articles focused on pediatric or neonatal populations were also excluded. Three studies were used for the primary analysis—Deliberato and colleagues,10 Bouadma and colleagues,11 and Hochreiter and colleagues.12 These 3 studies were chosen for their focus on critically ill adult patients with suspected or proven bacterial infections, including but not limited to sepsis, with comparable protocols and endpoints of durations of treatment with antibiotics, mortality, and ICU length of stay. Other studies were excluded from the primary analysis but were included as references, because of their focus on only patients with sepsis.
100 Copyright © 2015 American Association of Critical-Care Nurses. Unauthorized reproduction of this article is prohibited.
NCI-D-14-00040_LR.indd 100
07/04/15 7:15 PM
V OL UME 2 6 • N U MBER 2 • APRIL- JUNE 2015
P ROCA LCIT ONIN- GUID E D D URAT ION OF A NT IBIOT IC T H E RA P Y
Primary Analysis
Patients included in the studies for primary analysis10-12 were male and female adults from medical and surgical ICUs with suspected or proven serious bacterial infections. The patients were critically ill and had a wide range of comorbidities. Admission diagnoses for the 3 studies mainly included sepsis, severe sepsis, septic shock, pulmonary infections, urinary system infections, and peritonitis.10-12 The most common exclusion criteria in the 3 studies were patients who started antibiotic therapy more than 24 hours before the study, more than 48 hours before cultures were taken, or before admission to the ICU. Also, the studies excluded patients who needed long-term treatment with antibiotics for infections (such as infective endocarditis), patients with nonbacterial infections, patients with minor or localized infections, patients younger than 18 years, patients with known pregnancy, or patients who did not give consent.10-12 Patients were randomized into a procalcitonin-protocol group or a control group, with standard antibiotic care at the discretion of the provider. All 3 articles used comparable criteria for their procalcitonin protocols. See Table 1 for individual study information. Findings
The findings of the 3 studies are not entirely consistent, but they are encouraging for the use of procalcitonin-guided antibiotic therapy.10-12 Mounting evidence has shown that the procalcitonin protocol reduces the duration of antibiotic therapy for adults in the ICU without affecting mortality rate. However, studies have not yet shown a decrease in ICU length of stay and rates of infection relapse, superinfection, or multidrug-resistant bacteria during a procalcitonin-guided antibiotic therapy protocol. Duration of Antibiotic Therapy
All 3 studies found a reduction in the duration of days of treatment with antibiotics between the procalcitonin-protocol and control groups. The largest study, by Bouadma and colleagues,11 found that use of the procalcitonin protocol reduced the duration of antibiotic therapy by 3 days when compared with the duration of antibiotic therapy in the control group (10.3 days vs 13.3 days; P < .0001), even with a limited adherence rate to both procalcitonin and control protocols. In addition, the procalcitonin-protocol group in the study by Hochreiter and col-
leagues12 had a reduction of 2 antibiotic-treatment days (5.9 days vs 7.9 days; P < .0001). However, the intent-to-treat analysis in the study by Deliberato and colleagues10 did not result in a statistically significant difference in antibiotic-therapy duration, which may be due to the limited protocol-adherence rates found in the study, with only 47% of procalcitonin group patients and 79% of control group patients following protocol. When a separate per-protocol analysis was performed, Deliberato and colleagues10 found a statistically significant 4-day difference between the median days of antibiotic-therapy duration in the procalcitonin and control groups (9 days for the procalcitonin-protocol group vs 13 days for the control group; P = .008). The fewer antibiotic-therapy days in the study by Hochreiter and colleagues,12 5.9 days compared with 9 to 10 days in the other studies, may be explained by the discontinuation of antibiotic therapy when procalcitonin levels were less than 1 ng/mL in that study12 versus less than 0.5 ng/mL in the other 2 studies.10,11 Despite using different procalcitonin levels for the discontinuation of antibiotic therapy, these researchers found a consistent decrease in antibiotic usage when compared with the control groups. These results provide encouraging evidence that procalcitoninguided antibiotic therapy reduces patients’ antibiotic exposure (see Figure 1). Mortality
No statistically significant difference was found in mortality rate when comparing the procalcitonin-protocol and control groups in any of the 3 studies.10-12 However, Bouadma and colleagues11 reported both a 28-day and 60-day mortality rate, whereas Deliberato and colleagues10 and Hochreiter and colleagues12 reported only in-hospital mortality rate. More evidence supports no change to shortterm mortality rate than evidence supporting no change to long-term mortality rate. Bouadma and colleagues’11 28-day mortality rate (21.2% procalcitonin vs 20.4% control, which is within the confidence interval), Deliberato and colleagues’10 in-hospital mortality rate (4.8% procalcitonin vs 10.3% control, P = 0.42), and Hochreiter and colleagues’12 inhospital mortality rate (26.3% procalcitonin and 26.4% control) all support the conclusion that there is no change in the short-term mortality rate. Interestingly, no in-hospital deaths were reported for either the procalcitonin or
101 Copyright © 2015 American Association of Critical-Care Nurses. Unauthorized reproduction of this article is prohibited.
NCI-D-14-00040_LR.indd 101
07/04/15 7:15 PM
W W W.A A CNA D VA NCE D CRIT ICA LCA RE .COM
WA LK E R
Table 1: Study Details Study
Deliberato et al10
Bouadma et al11
Hochreiter et al12
Year
2013
2010
2009
Design
Randomized
Randomized
Randomized
Size
81 patients: 42 PCT vs 39 control; per protocol: 20 PCT vs 31 control
621 patients: 307 PCT vs 314 control
110 patients: 57 PCT vs 53 control
Site
Mixed medical-surgical ICU in Brazil
5 medical ICUs, 2 surgical ICUs, 1 mixed medical-surgical ICU in 5 French hospitals
Surgical ICU in Germany
Procalcitonin protocol
Discontinue antibiotic therapy when procalcitonin level
