The Clinical Use of Procalcitonin: AACC Guidance Document On

AACC Guidance Document on
the Clinical Use of Procalcitonin

AUTHORS
Allison Chambliss, PhD, DABCC, FAACC Joshua Hayden, PhD, DABCC, FAACC Alison Woodworth, PhD, DABCC,
Department of Pathology and Department of Laboratories FAACC
Laboratory Medicine Norton Healthcare Global Laboratory Services
University of California, Los Angeles Louisville, KY CTI Clinical Trials and Consulting
Los Angeles, CA Cincinnati, OH
Sophie E, Katz, MD, MPH
Khushbu Patel, PhD, DABCC, FAACC Division of Infectious Diseases,
Department of Pathology and Department of Pediatrics
Laboratory Medicine Vanderbilt University Medical Center
Children’s Hospital of Philadelphia Nashville, TN
Philadelphia, PA
Emi Minejima, PharmD
Jessica M Colón-Franco, PhD, DABCC, Department of Clinical Pharmacy
FAACC University of Southern California
Department of Laboratory Medicine School of Pharmacy
Cleveland Clinic Los Angeles, CA
Cleveland, OH
CITATION Chambliss AB, Patel K, Colón-Franco JM, Hayden J, Katz SE, Minejima E, Woodworth A. AACC guidance document on
the clinical use of procalcitonin. J Appl Lab Med 2023;8:598-634.
TABLE OF CONTENTS
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Can PCT Results Be Utilized to Inform Treatment Decisions in Both Initiation and Cessation of
Antimicrobial Therapy in Adult Patients with Sepsis or Respiratory Infections?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Is PCT an Accurate Predictor of Outcomes (e.g., Mortality, Respiratory Failure, Shock) in Adult Populations?. . . . . . . . . . 8
Can PCT Results Be Utilized to Inform Treatment Decisions in Both Initiation and Cessation of
Antimicrobial Therapy in Neonatal and Pediatric Patients with Sepsis or Respiratory Infections?. . . . . . . . . . . . . . . . . . . 12
Is PCT an Accurate Predictor of Outcomes (Mortality, Respiratory Failure, Shock) in Pediatric Populations? . . . . . . . . . . 14
When and How Often Should PCT Be Measured? Which Cutoff(s) Should Be Used? . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
How Should PCT Be Incorporated into Antimicrobial Stewardship Efforts?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
What Preanalytical Factors Affect PCT Results and/or Interpretation?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
What FDA-Cleared Methods Are Available to Measure PCT and How Do They Compare? . . . . . . . . . . . . . . . . . . . . . . . 19
Are Clinical Decision Points (Cutoffs) Comparable across PCT Assays?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
What Are Possible Confounding Factors for the Interpretation of PCT Results? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Supplemental Material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
INTRODUCTION
Procalcitonin (PCT), the precursor of the hormone calcitonin, is a to be eliminated primarily by the kidneys. The aforementioned
116 amino acid protein encoded by the CALC-1 gene. The product attributes led to PCT's emergence as a biomarker of infection
of this gene is prePCT, which undergoes sequential proteolytic specifically able to distinguish bacterial from viral infections.
cleavage to produce PCT and calcitonin (Fig. 1). Under normal PCT has been regarded as potentially useful across some clinical
physiologic conditions, transcription of the CALC-1 gene occurs settings to aid in the diagnosis of sepsis; predict disease severity
in neuroendocrine cells, primarily in the thyroid C-cells, and and outcomes, including mortality; and guide antibiotic therapy,
circulating concentrations of PCT are undetectable. In response as demonstrated by a number of studies and clinical trials that are
to stimuli such as bacterial infection and systemic inflammation, referenced in this document. However, for several clinical settings
production of PCT is activated in numerous nonthyroidal cells and patient populations, the utility of PCT remains undefined.
(i.e., adipocytes and fibroblasts) unable to process PCT into Although the need for biomarkers to aid in managing septic
calcitonin, leading to its accumulation (2). Production of patients is well accepted, PCT adoption as routine standard of
PCT in response to bacterial infection is cytokine mediated. care has not been straightforward. PCT use in the United States
Interleukin-1β, tumor necrosis factor-α, and interleukin-6 was not widespread until recently, when additional assays
activate PCT production. Interferon-γ, secreted during viral and clinical indications were cleared by the Food and Drug
infection, counter-regulates its expression (Fig. 1). Expression Administration (FDA). However, PCT was widely used across
of PCT increases within hours of the inflammatory insult, peaks European countries, where most of the original published clinical
at approximately 12 hours, and has a half-life of about 24 hours trials took place. At the time of manuscript preparation, the FDA
(Fig. 2) (4). The extent of PCT increase correlates with the had cleared 42 PCT assays. Approved indications were initially
severity of disease, while decreasing concentrations indicate limited to assessment for risk of disease progression to severe
disease resolution. As a small protein (14.5 kDa), PCT is believed sepsis and/or septic shock among critically ill patients. More
AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

Published in The Journal of Applied Laboratory Medicine, May 4, 2023 ∙ academic.oup.com/jalm 2
recently, some of the PCT assay manufacturers have added the • Can PCT results be utilized to inform treatment decisions in
cumulative 28-day risk of all-cause mortality and antimicrobial both initiation and cessation of antimicrobial therapy in adult
stewardship in patients with respiratory illnesses and/or sepsis patients with sepsis or respiratory infections?
as intended uses as well. Factors inhibiting widespread use of PCT • Is PCT an accurate predictor of outcomes (mortality,
may include lack of specificity for sepsis/infection, overall lack respiratory failure, shock) in adult populations?
of consensus on the utility of PCT across various clinical settings • Can PCT results be utilized to inform treatment decisions
and patient populations, and variability in its interpretation (i.e., in both initiation and cessation of antimicrobial therapy in
cutoffs, when to measure). neonatal and pediatric patients with sepsis or respiratory
This document aims to provide evidence-based guidance in infections?
how to best use PCT across various clinical settings and patient • Is PCT an accurate predictor of outcomes (e.g., mortality,
populations to improve patient outcomes. The document is respiratory failure, shock) in pediatric populations?
intended for clinical and laboratory stakeholders, particularly • When and how often should PCT be measured? Which
those caring for adult and pediatric patients with suspected cutoff(s) should be used?
sepsis and respiratory infections (e.g., physicians/assistants, • How should PCT be incorporated into antimicrobial
nurses, pharmacists, laboratorians). A detailed review of the stewardship efforts?
English literature was conducted for PCT studies in adult (≥ 100 • What preanalytical factors affect PCT results and/ or
patients), pediatric, and neonatal populations in order to answer interpretation?
the following questions addressing key aspects for PCT-guided • What FDA-approved methods are available to measure PCT
management of patients with suspected sepsis, respiratory, or and how do they compare?
defined site-specific infections. In preparing this document, we • Are clinical decision points (cutoffs) comparable across
focused on applying the FDA-cleared indications for PCT in both PCT assays?
adults and pediatrics. • What are possible confounding factors for the interpretation
of PCT results?
FIGURE 1. Schematic of cytokine-mediated inflammatory host response pathway leading to adipocyte secretion of
procalcitonin (top) compared to normal physiological secretion of calcitonin from thyroidal C-cells (bottom). Reprinted
from Linscheid P, et al. In Vitro and In Vivo Calcitonin I Gene Expression in Parenchymal Cells: A Novel Product of
Human Adipose Tissue. Endocrinology. 2003;144:5582 (1), by permission of Oxford University Press and the
Endocrine Society.

CAN PCT RESULTS BE UTILIZED TO INFORM studies demonstrating little benefit with the addition of PCT
TREATMENT DECISIONS IN BOTH INITIATION (17-20). This section focuses on studies that included evaluation
AND CESSATION OF ANTIMICROBIAL THERAPY IN of outcomes based on using PCT to inform decisions regarding
ADULT PATIENTS WITH SEPSIS OR RESPIRATORY antibiotic initiation and cessation.
INFECTIONS?
Multiple meta-analyses have evaluated the utility of PCT to Antibiotic Initiation in the Critically Ill
guide decisions on antibiotic initiation for sepsis and respiratory There were a limited number of studies that addressed the
tract infections (RTIs) (5-9). Data about the use of PCT to safely impact of PCT utilization on rates of antibiotic initiation in
reduce antibiotic treatment for patients with sepsis and lower critically ill patients in the ICU. Layios et al. implemented a PCT
respiratory tract infection (LRTI) is encouraging. In patients protocol to guide antibiotic initiation and specifically reported
with sepsis admitted to an intensive care unit (ICU), there is the rate of initiation of antibiotics as an outcome (21). This
relatively strong evidence to support the use of PCT to reduce was a randomized controlled trial (RCT) of 5 ICUs and 509
antibiotic duration (10-16). The data regarding use of PCT in total patients and evaluated the rate of initiation of antibiotics,
patients with LRTI is less straightforward, with more recent which was not significantly decreased by the availability of PCT
FIGURE 2. Relative kinetic expression pattern of procalcitonin upon inflammatory insult as compared to other
inflammatory markers. Procalcitonin increases in the plasma within 2 to 6 hours, peaks at approximately 12
hours, and has a half-life of about 24 hours. Adapted with permission from Meisner (3).

compared to standard of care (PCT 62.6% vs control 57.7%, had a significantly higher 3-month mortality compared with the
P = 0.11). Possible explanations for the lack of benefit included control group (PCT 19/61, 31% vs control 7/58, 12%, P = 0.015).
a low rate (25%) of PCT results that were <0.25 ng/mL, critical In summary, the data on the utility of PCT to guide initiation of
illness of patients that prompted clinicians to overrule the antibiotics in critically ill patients is currently limited and does
PCT treatment algorithm, and a low rate (57%) of initiation of not show benefit in decreasing antibiotic prescriptions.
antibiotics in the control arm compared to prior studies. Jensen
et al. also implemented an antibiotic initiation PCT protocol in Antibiotic Cessation in the Critically Ill
a multicenter RCT study of critically ill patients in the ICU in We reviewed 5 RCTs comparing PCT-guided antibiotic duration
Denmark (n = 1200) (22). In the PCT arm, an "alert procalcitonin" with standard antibiotic duration (Supplemental Table 1) with
notification was provided when the initial PCT was ≥ 1 ng/mL sample sizes that ranged from 110 to 1546 patients (10-16, 23).
to initiate antibiotics or when subsequent PCT levels were Overall, 3757 patients were enrolled in these trials. All studies
not decreasing by 10% from the previous day to intensify the demonstrated a reduction in antibiotic use with PCT guidance
antibiotic course. In this study, they found the PCT group had a compared to standard antibiotic duration, with reductions in
longer antibiotic course by a median of 2 days [PCT median 6 days on antibiotics of about 2 days.
days (interquartile range [IQR] 3-11)] vs standard of care of 4 Major limitations to study interpretation included the
days (IQR 310). Antibiotic initiation rates were only reported variation in timing of PCT measurements among studies and
for guideline concordant cases with PCT ≥ 1 ng/ mL, which was variation in the PCT concentration thresholds utilized to drive
82.1% in the PCT group and 82.4% in the standard of care group. discontinuation of antibiotic therapy among studies. One study
As 28-day mortality rates were comparable between the 2 groups by Hochreiter et al. did not report frequency of PCT measurement
(PCT vs standard of care: hazard ratio [HR] 0.98; 95% CI, 0.83- or turnaround time for PCT result notification (15). The other 7
1.16), the authors concluded that PCT was not effective to guide studies each obtained PCT measurements on enrollment and
initiation or escalation of antibiotic therapy over standard of care then at variable time intervals ranging from daily to every 5 to
practices and led to increased use of broad-spectrum antibiotics. 7 days until ICU or hospital discharge (10-14, 16, 23). Absolute
Three studies that had a mixed initiation and cessation PCT cut points for antibiotic discontinuation were 0.25 ng/mL (16),
protocol reported the rate of antibiotic initiation compared 0.5 ng/mL (11-14), or 1 ng/ mL (10, 15, 23). Most studies
to a control group. In the PRORATA trial, a large randomized also allowed for antibiotic discontinuation based on a relative
clinical trial (n = 621) by Bouadma et al. that included critical decrease in PCT concentrations over time. The majority of studies
care patients in France, 28/307 (9%) in the PCT group did not used an 80% to 90% decrease from peak value to direct antibiotic
receive antibiotics at study inclusion in accordance with the cessation, although one study used a 50% drop compared to the
prespecified algorithm (<0.5 ng/mL) (14). Eight of the PCT previous value (10), and another used a 25% decrease from peak
group then proceeded to be given antibiotics within 5 days, and concentration (15). Only 2 studies reported turnaround time for
7/8 survived. In comparison, 15/314 (5%) of the control group PCT result notifications from lab personnel to study providers (14,
did not receive antibiotics at inclusion, and 8/15 proceeded to 16). In both studies, results were available within 2 to 3 hours of
be given antibiotics within 5 days; 1/5 survived. Otherwise, the specimen collection (15). The other 4 studies each obtained PCT
rate of initiation of antibiotics in patients who had PCT levels that measurements on enrollment and then at variable time intervals
fell outside of the prespecified range included 65/307 (21%) ranging from daily to every 3 days until ICU or hospital discharge
patients who were initiated on antibiotics when PCT was <0.5 ng/ (10-14, 16, 23). Absolute cut points for antibiotic discontinuation
mL and 4/307 (1%) who were not given antibiotics although the were 0.5 ng/mL (11-14) or 1 ng/mL (10, 15, 23). Most studies
PCT was ≥ 0.5 ng/mL. Additional details on antibiotic exposure also allowed for antibiotic discontinuation based on a relative
outcomes related specifically to antibiotic initiation were not decrease in PCT concentration over time. Only one study reported
teased out from the total use combining the initiation and turnaround time for PCT result notifications from lab personnel
cessation protocol. Overall algorithm adherence was 53% in the to study providers, and results were available within 2 to 3 hours
PCT group. In a second RCT with mixed initiation and cessation of specimen collection (14).
PCT protocol of ICU patients with severe acute exacerbation of Compliance rates with PCT-driven antibiotic cessation
chronic obstructive pulmonary disease (AECOPD) (n = 302), algorithms also varied greatly among studies, ranging from 28.7%
although there was no significant difference in the proportion to 97%. Most often, providers chose to continue antibiotics despite
of those who received antibiotics at baseline (PCT group 58% a low PCT level. Noncompliance with algorithm guidance may
vs control 62%, P = ns), by day 1 there were significantly fewer have skewed study results. However, as noted earlier, all studies
patients in the PCT arm who remained on antibiotics (P < 0.001) demonstrated significant decrease in antibiotic use when using PCT
(18). In a subgroup analysis, those who were not on antibiotic guidance, and noncompliance with algorithms would have more
therapy initially and used the PCT to guide antibiotic initiation often led to similar results in the PCT and standard of care groups.

In a 2018 meta-analysis, Meier et al. describe 523 patients Antibiotic Initiation in Respiratory Tract Infections
with positive blood cultures from 13 clinical trials that randomly A 2018 meta-analysis by Schuetz et al. included 26 randomized
assigned patients to either PCT-guided treatment or a standard clinical trials of PCT in RTIs, of which 13 were conducted in
of care group (24). The mean duration of antibiotic therapy the ICU setting, 11 in emergency departments (EDs), and 2 in
was 2.86 days shorter in the PCT-guided group compared to primary care (26). Infection types included community-acquired
the control group (95% CI, -4.88 to -0.84 days, P = -0.006), and pneumonia (CAP), AECOPD, acute bronchitis, hospital-acquired
mortality was again similar between groups (16.6% PCT vs 20% pneumonia, and ventilator-associated pneumonia (VAP).
control, P = 0.263). This study highlights that PCT guidance for Initiation of antibiotics was significantly reduced overall in the
antibiotic duration may be a safe and effective way to decrease PCT group compared to the control (70% vs 85%, adjusted
antibiotic use among patients with bacteremia. odds ratio [OR] 0.27 [0.24-0.32], P < 0.0001). Lower antibiotic
In summary, all studies evaluated demonstrated a reduction in prescription rates were seen in primary care (PCT 23% vs
antibiotic use with PCT guidance compared to standard antibiotic control 63%, adjusted OR 0.13 [0.09-0.18], P < 0.0001), ED (PCT
duration. Limitations found across studies included differences 69% vs control 83%, adjusted OR 0.49, P < 0.0001), and ICU
in compliance rates for the antibiotic cessation algorithms that settings (PCT 92% vs control 99%, 0.02 [0.01-0.05], P < 0.0001).
may have limited the benefit in certain populations. Antibiotic prescription rates were similarly lower when analyzed
by infection type for all included infection types except for
Other Biomarkers Evaluated to Guide Antibiotic VAP, where antibiotic initiation was 100% in both the PCT and
Initiation or Cessation in the Critically Ill control arms.
Like PCT, C-reactive protein (CRP) is an acute phase reactant, After the publication of the meta-analysis by Schuetz et al. (26),
upregulated in response to inflammation (Fig. 2). Due to its Huang et al. published the ProACT study, a 14-center, randomized
longer half-life, its utility in antibiotic monitoring is limited. clinical trial (17) evaluating PCT use in acute LRTI of 1656 adult
Several studies have investigated its potential to guide antibiotic patients in the United States. This study excluded severely ill
therapy. For example, in a RCT in 2 ICUs in Brazil, investigators patients requiring endotracheal intubation or intravenous
compared PCT vs CRP for the primary outcome of antibiotic vasopressors. Of the patients evaluated in the ED, there was 72.9%
duration in critically ill patients with severe sepsis or septic shock PCT algorithm adherence, with the highest algorithm adherence
(25). Due to its longer half-life, its utility in antibiotic monitoring for acute bronchitis (82.4%) and the lowest in CAP (39.4%).
is limited. In the PCT arm, discontinuation of antibiotic therapy There was no difference in the primary outcome of antibiotic
was recommended if PCT fell to <0.1 ng/mL (if the initial PCT was exposure during the first 30 days between the PCT group vs
<1 ng/mL) or if there was a decrease in PCT by 90% from peak control group (mean antibiotic days 4.2 vs 4.3 days, difference
(if initial PCT was >1 ng/mL). In the CRP arm, discontinuation of -0.05 days; 95% CI, -0.6-0.5, P = 0.87). The secondary outcome
antibiotic therapy was recommended if CRP fell to <25 mg/L (if of percentage of patients receiving an antibiotic prescription
the initial CRP was <100 mg/L) or if CRP decreased by 50% from in acute bronchitis in the ED was significantly lower in the PCT
peak (if the initial CRP was >100 mg/L). Of the 94 total patients group (17.3% vs 32.1%; risk difference, -14.8%; 99.86% CI, -28.5
included, the authors found the median duration of antibiotics to to -1.1). The authors noted that a study limitation was that PCT
be similar between the groups (PCT median 7 days vs CRP 6 days, information was not available to all prescribers prior to when
P = 0.06), even after adjusting for severity of illness. Mortality decisions about antibiotic initiation could be made, thus the
was similar in both groups (21 patients died in each group, true effect of PCT information to guide decisions on whether to
P = 0.86). Thus, the authors concluded that CRP, which costs initiate antibiotics may be lacking. Also of note is the lower rate of
significantly less than PCT per test, was as useful as PCT in guiding antibiotic prescription even in the control arm in the ED (38.7%)
antibiotic cessation. compared to control arms in the studies in the meta-analysis by
Currently there is limited data available on the utility of CRP Schuetz et al. (85%) (26).
compared to PCT in guiding antibiotic initiation or cessation in Patients presenting with acute heart failure represent
a critically ill cohort across different ICU settings. In addition, a challenging population to evaluate the need for antibiotic
as CRP is typically a less expensive inflammatory marker than therapy for RTIs due to diagnostic uncertainty. Mockel et al.
PCT and may be more readily available in clinical laboratories, evaluated whether PCT-guided initiation of antibiotics in a
additional cost-effectiveness studies are needed to address study population of patients with suspected or confirmed heart
whether PCT is a more clinically advantageous inflammatory failure could provide benefit by decreasing unnecessary starts
marker to guide antibiotic initiation or cessation in the of antibiotics (27). In heart failure patients presenting to the
critically ill. ED with primary symptom of dyspnea, antibiotic initiation was
recommended if PCT was >0.2 ng/mL. In this RCT of 742 total
patients, the initiation rate of antibiotic was similar between

the study groups (PCT guided 18% vs standard of care 14%, therapy for treatment of patients with LRTI, including CAP, COPD,
P = 0.145), and the primary outcome of all- cause 3-month nonpneumonia LRTI, and VAP (Supplemental Table 2) (17, 20,
mortality was similar (PCT guided 14% vs standard of care 6.6%; 30-34). Sample sizes ranged from 101 to 1656 patients. Overall,
90% CI, -0.313.5%). PCT algorithm adherence was 83% in the 3905 patients were enrolled in these trials. There were an
PCT group. The overall rate of pneumonia in this study was only additional 3 retrospective studies evaluating PCT for antibiotic
7.5%, which the authors predict was likely due to study clinicians discontinuation (35, 36) and one prospective cohort study (37).
avoiding randomization of patients with a high suspicion of The impact of PCT measurement on antibiotic use varied among
pneumonia, where immediate initiation of antibiotics was studies, with 6/11 demonstrating reduction in antibiotic use
warranted without having to wait for the PCT result. Thus, the with PCT guidance. None of the 4 RCTs published since 2016
authors concluded that PCT-guided initiation of antibiotics have demonstrated benefit in using PCT to decrease antibiotic
was not more effective than standard of care in the rates of use in patients with RTIs (17-20). Possible reasons for the
antibiotic utilization. lack of benefit seen may be due to a different population being
AECOPD represents another population that poses a investigated (severe AECOPD vs mild to moderate AECOPD) (18),
diagnostic challenge in LRTI. A meta-analysis by Ni et al. included lower algorithm adherence rates (19) compared to prior studies
23 studies of PCT usage in patients presenting with severe (32), or more contemporary studies including antimicrobial
AECOPD with the primary outcome of length of stay and treatment stewardship in the standard of care arm that reduced the
failure (28). Six RCTs involving a total of 942 patients found no antibiotic duration.
difference in treatment failure rates (relative risk 0.85; 95% CI, Timing of PCT measurements and frequency with which PCT
0.66-1.09) or length of stay (weighted mean difference = -0.1; was measured varied among studies. The first measurement was
95% CI. -0.980.79) between the PCT-guided group and the control obtained at or within 24 hours of enrollment in all studies. The
group. PCT-guided treatment significantly reduced the antibiotic timing of the next measurement varied in all studies, ranging
prescription rate by 34% (relative risk 0.66; 95% CI, 0.62-0.71). from 12 hours to 5 days after the first. No studies required PCT
Bremmer et al. evaluated patients with low PCT (<0.25 ng/ measurements after day 10, and most stopped testing on day
mL) in noncritically ill AECOPD in a retrospective study on the 7. Of the studies that mentioned timing from sample collection
effects on 30-day all-cause hospital readmissions (29). This study to provider notification of results, most were available within 1
excluded ICU patients, patients with immunocompromising hour. Cut points used to discontinue antibiotics also varied across
conditions, patients on mechanical ventilation, and patients with studies. Ten of the 11 studies used an absolute value of 0.25 ng/
pneumonia. Comparing patients who received <24 h vs >24 h of mL to recommend antibiotic discontinuation (17-20, 31, 32, 38-
antimicrobial therapy, there was no difference in all-cause 30-day 40). The remaining study used an absolute value of 0.5 ng/mL
re-admission (15.5% vs 17.4%, P = 0.63) and chronic obstructive (41). Five of the studies utilizing LRTI patients also allowed for
pulmonary disease (COPD)-related 30-day readmission rates antibiotic discontinuation based on a relative decrease in PCT
(11.2% vs 12.3%, P = 0.743), concluding that in a noncritically ill concentration of ≥ 80% (20, 32, 41, 42) or ≥ 90% (18) when
cohort of AECOPD admissions, antibiotics may be withheld safely compared to the peak PCT concentration or the concentration
utilizing PCT guidance. at randomization. Compared to studies using PCT for antibiotic
In summary, PCT-guided therapy can significantly decrease discontinuation in sepsis, compliance rates with PCT algorithm
antibiotic initiation without compromising safety in LRTI. The guidance were higher in pneumonia studies, ranging from 61%
clinical benefit was more consistently seen in lower acuity to 85%. Compliance rates were not reported in 2 of the studies
patients such as those in the primary care setting with CAP, (18, 20). A more detailed discussion of PCT collection timing and
AECOPD, or acute bronchitis. More data is needed to support its frequency is included in a later section of this document.
use in antibiotic initiation for critically ill patients or patients All included RCTs published prior to 2011 demonstrated
with immunocompromising conditions. Data is currently not reduction in antibiotic use with PCT guidance and did not
available to support the use of PCT in patients with the suspicion demonstrate increased risk of mortality (if included as an
of VAP to guide antibiotic initiation decisions. outcome) (32, 39-41). However, RCTs published since 2016 have
not demonstrated reduced antibiotic use with PCT guidance
Antibiotic Cessation in Respiratory Tract Infections compared to control (17-19, 38, 42). Proposed reasons for this
There were 11 RCTs comparing PCT-guided therapy with standard decrease in effect size include shorter baseline treatment durations
antibiotic therapy for treatment of patients with LRTI, including for most patients with LRTI and low algorithm compliance
CAP, COPD, non-pneumonia LRTI, and VAP (Supplemental Table rates. Two recently published retrospective studies with more
2) (17, 19, 20, 30-34). Sample sizes ranged from 45 to 1656 than 300 patients in each study demonstrated significantly
patients; only one study had fewer than 100 patients. There were shorter antibiotic duration in patients with LRTI in whom a PCT-
10 RCTs comparing PCT-guided therapy with standard antibiotic guided antibiotic cessation algorithm was followed (35, 36).

Schuetz et al. published in 2017 a Cochrane review of 26 KEY SUMMARY POINTS
randomized clinical trials on RTIs with a patient level meta- • In patients admitted to the ICU, PCT should be used to
analysis of 6708 participants (7). All-cause 30-day mortality was reduce antibiotic duration. Data does not support using
significantly lower with the PCT-guided therapy (adjusted OR PCT to guide initiation of antibiotics in these patients.
0.83; 95% CI, 0.70-0.99, P = 0.037). Reduction in total antibiotic • In patients with LRTI, PCT may be used to safely reduce
exposure (mean 8.1 days compared to 5.7 days, regression antibiotic exposure and duration, but there is less evidence
coefficient -2.43 days [95% CI, -2.71 to -2.15, P < 0.001]) was to support this recommendation, particularly in VAP.
• Studies to date have shown significant variation in PCT
observed, although the rate of initiation of antibiotics was
testing algorithms in terms of cut points and timing of PCT
not reported.
measurements.
In summary, conclusions regarding the impact of PCT • Other inflammatory markers, such as CRP, have shown
measurement on antibiotic cessation in LRTI are mixed among benefit in decreasing antibiotic initiation in primary care
studies. Earlier studies and meta-analysis show overall decrease settings and AECOPD, although conclusions are mixed.
in antibiotic use with PCT, while the more recent RCTs have Direct comparison of CRP and PCT is still lacking in LRTI.
shown no difference relative to standard of care without PCT. As CRP is typically less expensive and may be a more
widely available biomarker than PCT, additional cost-
Other Biomarkers Evaluated to Guide Antibiotic effectiveness studies are also needed.
Initiation or Cessation in Respiratory Tract Infections
There are limited studies available that evaluated the IS PCT AN ACCURATE PREDICTOR OF OUTCOMES
performance of other biomarkers (e.g., CRP) compared to PCT on (E.G., MORTALITY, RESPIRATORY FAILURE, SHOCK)
antibiotic prescribing patterns for LRTI. However, CRP has been IN ADULT POPULATIONS?
evaluated in RTIs for its utility to guide antibiotic prescribing PCT is upregulated in response to proinflammatory signals,
decisions, most notably in the primary care settings, where it has and its concentrations increase with increased disease severity
demonstrated decreased antibiotic prescriptions (43-45). In an (3). Concentration decreases over time are associated with
outpatient setting, point-of-care CRP testing led to significantly disease recovery, while consistently elevated or increasing PCT
fewer antibiotic prescriptions for acute LRTI and rhinosinusitis concentrations are a signal of persistent or more severe illnesses.
as compared to standard of care (43.4% vs 56.6%, relative risk Thus, researchers have studied the ability of PCT to predict
0.77; 95% CI, 0.56-0.98) in the United Kingdom (46). outcomes in a variety of disease states. Among the most commonly
Butler et al. evaluated CRP to guide antibiotic prescription studied outcomes are mortality and disease progression.
decisions in AECOPD in a multicenter RCT (n = 653) (47). The literature investigating the potential utility of PCT in
A CRP point-of-care test was performed at presentation in predicting outcomes is significant. Mortality is an important
the intervention arm, with guidance that for CRP <20 mg/L, outcome among critically ill patients given that approximately
antibiotics were unlikely to be beneficial, for CRP 20-40 mg/L, 270 000 septic patients die each year in the United States alone
antibiotics may be beneficial in the presence of purulent sputum, (49). With recent updates to FDA-approved intended uses of
and for CRP >40 mg/L, antibiotics were likely to be beneficial. these assays, we chose to look more closely at publications since
The availability of rapid CRP results significantly decreased 2010 with at least 100 patients enrolled that investigated the
the number of patients who received an antibiotic prescription correlation between PCT concentrations and mortality. Fourteen
(47.7% vs 69.7%, adjusted OR 0.31; 95% CI, 0.21-0.45). Similarly, of these looked at 28/30-day all-cause mortality (Supplemental
Prins et al. found that using CRP significantly decreased antibiotic Table 3). Among the 28-day mortality studies, the patient
prescriptions compared to using Global Initiative for Chronic populations ranged from ED, ICU, and/or hospital in-patients
Obstructive Lung Disease guidelines (based on symptoms of with sepsis, RTIs, and other infections. Two of these studies were
purulent sputum) in their RCT of 220 adults in the Netherlands separate large meta-analyses of >3000 patients each looking at
(CRP 31.7% vs Global Initiative for Chronic Obstructive Lung PCT in septic and RTI patients (50, 51). One of these studies was
Disease 46.2%, P = 0.028) (48). However, neither study directly an RCT; however, the investigation of 28-day mortality was a
compared CRP vs PCT performance. secondary analysis (52).
Additional data is needed to compare CRP vs PCT for antibiotic
initiation or cessation in other LRTI, such as pneumonia, and Outcomes from Single PCT Measurements
in the inpatient setting. As CRP is typically a less expensive All studies demonstrated an association between initial PCT
inflammatory marker test than PCT, cost-effectiveness studies concentrations measured at presentation and/or evolution over
comparing the 2 markers would be helpful to guide laboratories time (discussed later) and 28/30-day mortality (Supplemental
considering implementation of PCT. Table 3). Most of the studies show that, independent of patient
population, PCT is significantly higher in patients that will go on

to die within 28/30 days compared to those who survived (10, AUC of 0.79 (0.75-0.83), and pooled relative risk of mortality
53-57). The prognostic utility of initial PCT measurements to was 3.05 (2.35-3.95) (51). The authors concluded that while
predict 28-day mortality is often assessed by generating receiver the prognostic utility of the initial PCT value is limited, PCT
operator characteristic curves and calculating the Area under nonclearance better predicts mortality and its performance
the Curve (AUC). These AUCs range from 0.56 to 0.82 among is superior to following clearance of other biomarkers like
studies where the time of initial PCT measurement was clearly lactate. Subsequently, one essential study of 13 US medical
delineated (10, 50-53, 56, 58-60). In their meta-analysis, Liu et centers including 858 patients admitted to the ICU with severe
al. demonstrated in a diverse patient population that a single sepsis or septic shock from the ED or other hospital locations
PCT measurement had a moderate ability to predict mortality. demonstrated that a lack of PCT decrease over time was a good
The area under the summary receiver operating characteristic predictor of 28/30-day mortality. The study found that although
curve was 0.73 (0.69-0.77) (n = 13 studies) (51). Follow-up PCT PCT was higher in nonsurvivors (mean 5.2 [95% CI, 3.9-7.0] vs
measurements in patients with signs of infection at 72 hours after 3.4 ng/mL [95% CI, 2.8-4.0, P< 0.02]), this baseline value was
the initial test were also significantly higher in patients who died a poor predictor of 28-day mortality (AUC 0.56; 95% CI, 0.51-
(60). However, in patients with autoimmune disease and sepsis, 0.60) (63). In patients in which PCT did not decrease by >80%
PCT concentrations at 72 hours were not significantly different in between baseline and day 4, 28-day mortality was 20.0% (hazard
survivors compared to nonsurvivors (61). This group also looked ratio 1.97, multivariate adjustment, P = 0.009), twice as high as
at the correlation between PCT peak concentration and mortality the group with this decline (P = 0.001). At this cutoff, sensitivity
and found variable results depending on patient population (61). was 77% (95% CI, 65-81) and specificity was 39% (95% CI, 35-
Similarly, in their large multicenter prospective observational 43) with similar performance regardless of whether the patients
study with more than 1700 septic shock patients, Ryoo et al. were in the ICU at day 4. Notably, ICU residency by day 4 was
were unable to demonstrate that elevated PCT, measured at a strong independent predictor of mortality (hazard ratio 2.69,
presentation to the ED, was significantly predictive of 28-day multivariate adjustment, P < 0.0001), with much higher mortality
mortality (62). The authors attribute this apparent discrepancy, than among those discharged to the hospital floors (26 vs 9%). In
compared to similar studies, to PCT being measured too soon a secondary analysis comparing PCT at baseline and day 1, PCT
after presentation and their patient population (ED vs ICU). In increased by 30% (95% CI, 15-47) and by 0% (95% CI, -7 to 6)
summary, a single PCT measurement at presentation is associated for those who died and survived, respectively (P < 0.0001), and
with higher mortality rates; however, patient population, time of mortality increased 3-fold in patients with an increase in PCT
PCT measurement, and other factors confound study results and compared to a decrease in PCT (29 vs 12%, P < 0.0001). Mortality
limit the utility of a single PCT measurement to predict mortality. was approximately 3 times higher if PCT did not decrease by
80%, regardless of whether the initial PCT concentration was
Outcomes from Sequential PCT Measurements and above or below 2 ng/mL. This study demonstrated that both
PCT Clearance short and longer serial approach/PCT clearance are stronger
PCT has a distinct expression pattern following an infectious insult prognosticators than an initial single PCT measurement. Finally,
(Fig. 2). Further, PCT expression persists throughout infection in their large RCT of 1089 patients with severe sepsis or septic
and decreases as the infection is cleared (3). Therefore, many shock, Elke et al. demonstrated that a PCT decrease < 20% from
have advocated monitoring PCT kinetics/concentration change baseline to day 1 was associated with a significantly higher 28-
over time as a prognostic marker in patients with infections (3). day mortality rate compared to those in whom PCT declined by
Sometimes referred to as the PCT delta, this change in PCT over > 20%. In this same patient population, there was a significantly
time has been associated with clinical outcomes. Specifically, a lower risk of mortality for those in whom PCT declined by >
PCT decrease over time, referred to as PCT clearance, suggests 50% between days 0 and 4 compared to those with a < 50% PCT
that a patient is responding to antimicrobial or other therapy. By decline. Although the parameters of the PCT delta calculation
contrast, persistent PCT expression over time, also called PCT are not standardized, the lack of PCT clearance over time does
nonclearance, is associated with poor outcomes in critically ill predict 28-day mortality in diverse patient populations.
and septic patients (51).
Among recent studies, 5 demonstrated that a lack of PCT Outcomes for Patients Presenting to the ED
clearance over time was a good predictor of 28/30-day mortality Some differences are noted in the utility of PCT among different
(51, 52, 61, 63, 64). Of these, one was a large meta-analysis patient populations. In general, there are far fewer studies in ED
where a subset analysis of 9 studies with 868 septic patients patients looking at the ability of initial PCT concentrations to
demonstrated that lack of PCT clearance, defined by a decrease predict mortality. The meta-analysis by Liu et al. included only
in PCT of at least 25% (range 25%-70%) in 48 hours to 7 days, 3 studies in the ED, which limited their ability to evaluate PCT's
could predict mortality (28 day and hospital) with a summary ability to predict mortality in ED patients (51). An individual

patient meta-analysis including 2605 ED patients concluded analysis, Elke and colleagues demonstrated that mortality rates
that PCT measurement at presentation predicted mortality (AUC differed among sepsis populations depending on the definition
0.67, P < 0.001, OR 1.82) and correlated with treatment failure utilized. However, regardless of the sepsis definition (Sepsis-1
(i.e., death, ICU admission, re-hospitalization, and complications vs Sepsis-3), PCT concentrations measured within 24 hours of
or recurrent or worsening infection within 28 days) (AUC a severe sepsis or septic shock diagnosis were a poor predictor
0.64, P < 0.01, OR 1.85) (50). In reality, this performance is of 28-day mortality (AUC was 0.56 for both populations) (52).
modest for predicting treatment failure and mortality and Interestingly, in their meta-analysis of septic patients, while
awaits confirmation by other studies. In their meta-analysis of PCT concentrations measured on days 1 and 3 were both
septic patients (ED and hospitalized), while pooled mean PCT significantly different between survivors and nonsurvivors in the
concentrations measured on days 1 and 3 were both significantly total population, in a subgroup analysis of patients with severe
different between survivors and non-survivors and able to predict sepsis or septic shock, PCT concentrations were not significantly
mortality, the correlation was stronger once ED patients were different in patients who died (82).
excluded (82). In their multicenter trial, Saeed et al. investigated Among patients with RTI, initial PCT concentrations
several biomarkers' abilities to predict 28-day mortality when correlated with 30-day mortality in patients with COPD and
measured at initial assessment in ED patients with suspected CAP but not in patients with acute bronchitis or VAP (50).
infections. PCT measurement at initial ED evaluation showed Elke et al. demonstrated that PCT, measured at baseline, had
moderate ability to predict 28-day mortality (AUC 0.72-0.75), marginally improved mortality prediction among patients with
which is similar to other studies in diverse patient populations pneumological compared to intraabdominal infection (AUC 0.58
(53). Of note, the mortality rate in this patient population was vs 0.52). Further, initial PCT concentrations were significantly
low at approximately 7%. In their multicenter, multinational ED correlated with 28-day mortality among patients with gram-
cohort study, Sager et al. demonstrated that PCT concentrations, positive and -negative infections but not among patients with
when measured during the ED stay, predicted 30-day mortality fungal infections (52). Most studies noted differences in the PCT's
(56). In contrast, in a third study looking at ED patients with ability to predict 28/ 30-day mortality across different diagnosis,
septic shock (20.7% mortality), Ryoo et al. showed that initial likely due to differences in disease severity, treatments, and
PCT measurement was not an independent predictor of 28-day mortality rates in the individual populations.
mortality. The authors suggest that their findings may differ One other source of confusion regarding the prognostic value
from other studies because all PCT measurements were collected of PCT is that significant heterogeneity exists among data sets in
prior to initiation of antimicrobial therapy (62). Unlike Ryoo individual studies, making it difficult to draw conclusions. In their
and colleagues' population, most other studies utilized patients meta-analysis, Liu et al. demonstrated significant heterogeneity
having received prior antibiotics (51, 52, 63). In their large across data sets in individual studies, especially their mortality
meta-analysis of patients with RTI, initial PCT concentrations rates, which ranged from 17% to 66.7% (51). In a large meta-
predicted mortality at 30 days when measured at ED admission analysis of patients with RTI the mortality rate was only 6%,
but not at ICU admission, with AUCs of 0.67 and 0.5, respectively leading to high negative predictive values for PCT (50). In another
(50). In their study, Yu et al. demonstrated that in patients large meta-analysis conducted by Arora and colleagues looking
with a suspected infection in the ED or hospital floor, addition at the utility of PCT to predict mortality among septic patients,
of an initial PCT concentration to qSOFA score ≥ 2 significantly statistical heterogeneity of the patient populations across studies
improved prediction of 30-day mortality (58). was high, and mortality rates ranged from 13% to 69% (82).
In general, most studies comparing the prognostic utility of More recently, in their study of patients with suspected infection,
initial PCT concentrations and PCT clearance showed differences Yu et al. reported a mortality rate of 9% and were also able to
in its ability to predict 28/30-day mortality depending on patient demonstrate high negative predictive values for prediction of 30-
location (i.e., ED vs ICU). This is likely due to differences in disease day mortality (58). In contrast, in their study of septic patients,
severity and treatments noted in several studies. Elke et al. demonstrated that initial PCT concentrations were a
poor predictor of 28-day mortality, and their mortality rate was
Outcomes Differ by Patient Diagnoses, Sepsis 27% (52). As expected, heterogeneity in patient populations
Definitions, and Study Populations among different studies is significantly impacted by mortality
Groups have also identified differences in PCT's prognostic utility rates and thus the performance characteristics of PCT to predict
across patients' diagnoses. In septic patients, discrepancies mortality (51).
among studies could be due to the sepsis definition utilized.
Studies published prior to 2016 likely utilized either the Sepsis-1 Outcomes Other Than 28/30-Day Mortality
or -2 definition to classify patients, whereas the Sepsis-3 definition While most of the previous discussion has focused on 28/30-day
may have been used in more recent studies (83). In their meta- mortality, studies have investigated other outcomes as well. For

example, several studies addressed PCT and its correlation with mid-regional proadrenomedullin (MR-proADM), presepsin, and
mortality during the patient's hospitalization (51, 52, 60, 63) multibiomarker models. Next we summarize the evidence for the
and/or at other time points, including 7, 14, or 90 days; 1-year biomarkers currently used clinically and provide an outlook on
post-presentation; or undefined mortality endpoints (46, 48, 51, promising early findings for the newer biomarkers..
58, 60, 61). Lactate is a key biomarker routinely used for outcome
There has also been significant interest in utilizing PCT to prediction in patients with sepsis despite the low quality of
predict response to therapy. In general, PCT concentrations and/or evidence (72). Although it can be elevated in other contexts,
evolution of PCT concentrations over time are good predictors of increases in blood lactate indicate tissue hypoxia, and its
successful or failed treatments in a variety of patient populations measurement is a surrogate marker of hypoperfusion. Patients
(50, 84). PCT concentrations at presentation were also correlated with elevated lactate concentrations have poor outcomes while
with general disease progression, admission to the ICU, and/ sufficient decreased lactate over time (also referred to as lactate
or length of hospital stay (60). Zaccone et al. showed that PCT clearance) is associated with decreased mortality. In adults
measured within 12 hours of admission among 1063 critically ill patients admitted to the ICU, the likelihood of mortality decreased
patients was an accurate predictor of ICU transfer (65). Similarly, by 11% for every 10% increase in lactate clearance (73). In
in a large population of patients with lower acute respiratory children with septic shock, failure to achieve a lactate clearance
infection (ARI), initial PCT concentrations correlated with of >10% increased the risk of mortality (likelihood ratio 2.83;
treatment failure at 30 days. Further initial PCT concentrations 95% CI, 1.824.41) (74). Surviving Sepsis Campaign guidelines
even correlated with treatment failure among patients with recommend measuring lactate promptly after sepsis is suspected
certain upper ARIs like the common cold or rhinosinusitis (50). or identified and remeasuring it if elevated >2 mmol/L (72). A
In contrast, in a population of patients with febrile urinary tract lactate value of ≥ 4 mmol/L warrants fluid resuscitation and
infection, PCT concentrations at presentation, day 3, or PCT normalization is targeted (72). RCTs evaluating lactate clearance
clearance over time were able to predict treatment failure with for therapy guidance and outcomes vary in their conclusions.
AUCs of 0.52, 0.55, and 0.58, respectively (66). For in-hospital mortality, one RTC found a reduction only when
In summary, PCT concentrations increase with disease adjusted for risk factors (75), while another did not find an effect
severity in patients with sepsis and RTIs as well as in other select (76). Tian et al. reported that 10% and 30% lactate clearance was
patient populations. Elevated PCT concentrations measured not associated with a reduction in 7-day mortality rate, but the
at ED or hospital admission in patients with sepsis or LRTI 28-day mortality was significantly lower in patients with ample
are associated with a greater risk for 28- to 30-day mortality. lactate clearance (77). A meta-analysis including these studies
Similarly, a lack of PCT clearance over time is also associated (547 patients) concluded the use of lactate clearance to guide
with a great risk of mortality. Similar trends were observed with therapy reduces the risk of mortality (risk ratio of 0.65; 95% CI,
other mortality outcomes as well as treatment response and/or 0.49-0.85) (78). It is important to consider that lactate clearance
disease progression. However, significant heterogeneity in study is only appropriate for use in patients with severe sepsis and/or
populations across studies, especially related to mortality rates, septic shock as it is not elevated in early and/or mild sepsis.
limits our ability to formally recommend the use of PCT as a CRP has been heavily studied as a sepsis bio-marker and
predictor of prognosis. is frequently utilized to monitor numerous inflammatory
disorders. The evidence is controversial regarding its role as an
Other Biomarkers Evaluated to Predict Outcomes in outcomes predictor in patients with sepsis and/or RTIs. Several
Patients with Sepsis and/or Respiratory Tract Infections studies report that, in comparison to PCT, CRP measurements at
Algorithms that use biomarker results to stratify patients by study enrollment and/or admission are not significantly higher
mortality risk and provide actionable information for patient in survivors compared to nonsurvivors (52, 58, 60). In 2 studies
management are a promising tool for patient care. PCT is likely comparing prognostic utility of several biomarkers, including
the best-studied biomarker for this purpose in the context of PCT and CRP, in patients presenting to the ED, PCT and CRP
sepsis and/or RTIs; however, its performance varies across were significantly associated with 28-day mortality, but neither
studies (Supplemental Table 3). Although PCT is FDA cleared was considered an independent predictor of mortality (53, 62).
for predicting outcomes and disease progression in patients In general, CRP lacks significant clinical utility as a prognostic
with sepsis, factors such as optimal cutoffs and recommended marker.
frequency of testing vary by clinical context and thus complicate Other new biomarkers may have promise but have not
its operationalization (67-71). These variables are further been adopted into clinical practice. MR-proADM is a product
discussed in later sections. Other biomarkers with prognostic of proADM, generated in a 1:1 ratio with adrenomedulin, a
roles include standard of care tests such as lactate and CRP, as well calcitonin peptide family mostly known for its vasodilatory
as a growing list of candidate biomarkers including interluekin-6, activity. Like PCT, MR-proADM elevations are not specific to

infections. Several studies have established a relationship studies and approximately 1500 patients in ED and ICU settings
between MR-proADM and outcomes such as disease progression concluded that presepsin is not superior to PCT for mortality
and mortality and its superiority compared to other bio-markers prediction (81). The AUC of PCT was 0.81 (95% CI, 0.78-0.84) with
including PCT, CRP, copeptin, and presepsin (67-71). One such a pooled sensitivity of 0.76 (95% CI, 0.55-0.89) and specificity
study, a prospective multicenter study, reported that MR-proADM of 0.74 (95% CI, 0.33-0.94), and the AUC of presepsin was 0.77
outperformed PCT and CRP, and clinical scores such as SOFA/ (95% CI, 0.73-0.81) with pooled sensitivity and specificity of 0.83
qSOFA and National Early Warning Score for ICU admission and (95% CI, 0.72-0.90) and 0.69 (95% CI, 0.63-0.74), respectively.
28-day mortality in ED patients (n = 684) at presentation and These studies utilized a unified point-of-care assay (PATHFAST,
3 days after (60). MR-proADM had the strongest association LSI Medience Corp.), yet cutoffs and clearance strategy are not
by univariate analysis with requirement for ICU admission (OR standardized.
4.1 [2.3-7.1] vs PCT OR 2.2 [1.5-3.4] vs CRP 2.1 [1.2-3.6]) and As discussed with only a few examples, new biomarkers,
28-day mortality (MR-proADM OR 4.1 [2.6-6.5] vs PCT OR 1.9 alone or in combination, show early promising results for roles
[1.32.7] vs CRP OR 1.0 [0.7-1.5]). Adding MR-proADM and PCT in predicting outcome. However, the evidence to support the
increases the correlation with mortality (hazard radio 5.7 [2.8- utility of these biomarkers for outcome prediction strengthened
11.6]), a combination that could be explored further. The study is an active area of research. Of note, these biomarkers have
utilizes a noncommercial^ available point-of-care analyzer for almost exclusively been studied in developed counties and
PCT and MR-proADM. Although the availability of rapid testing in relatively small sample cohorts. Although some have been
could increase access to these biomarkers for decision-making, published as mentioned previously, future studies on these
more data will be needed to demonstrate the required analytical candidate biomarkers in RCTs, across different medical centers,
performance characteristics to support such applications. and with larger sample size will be needed to demonstrate if
Studies using a laboratory assay available for MR-proADM (79) these promising early results can be validated. Moreover, these
demonstrated consistent MR-proADM performance. Mearelli and studies should focus on establishing evidence-based cutoffs,
colleagues showed that MR-proADM improves qSOFA's outcome and interpretative criteria is necessary to draw meaningful
prediction ability (80). Nine biomarkers were evaluated in total conclusions for real-live applications. Until then, PCT is the only
(CRP, lactate, PCT, soluble interleukin 2 receptor alpha, soluble assay in the United States cleared by the FDA for 28-day mortality
triggering receptor expressed on myeloid cell-1, secretory prediction in critically ill patients. Despite the FDA approval, the
phospholipase A2 group II, presepsin, MR-proADM, and soluble lack of uniformity in the studies make recommendation of a
tumor necrosis factor receptor-1) as a secondary analysis of a specific clearance cut-off challenging.
prospective study conducted in 5 EDs in Italy. The analysis found
the highest AUROC of 0.83 (95% CI 0.80.87) when combining KEY SUMMARY POINTS
clinical parameters with CRP, lactate, and MR-proADM. The • PCT concentrations increase with disease severity in
study by Elke and colleagues also reported MR-proADM was the patients with sepsis and RTIs.
strongest mortality predictor at baseline relative to PCT and CRP • Elevated PCT concentrations measured at ED presentation
in patients with sepsis (AUC 0.73 vs 0.56 for PCT and 0.55 for or hospital admission in patients with sepsis or LRTI are
CRP) and septic shock (AUC 0.72 vs 0.50 for PCT and 0.53 for associated with a greater risk of 28-to 30-day mortality.
• In patients with sepsis and/or RTI, the lack of PCT
CRP); MR-proADM remained the strongest predictor at days 1, 4,
clearance overtime is associated with a greater risk of 28 to
7, and 10. In summary, there is some data to suggest a potential
30-day mortality.
role for MR-proADM in outcomes prediction in septic patients,
• The heterogeneity of patient populations in the studies
however, currently, MR-proADM assays are not widely available makes is difficult to form uniform recommendations for the
or cleared for this use. use of PCT as an outcome predictor.
Presepsin, a subtype of soluble cluster-of-differentiation
marker protein 14, is released into the blood after
lipopolysaccharides in microorganisms bind cluster-of- CAN PCT RESULTS BE UTILIZED TO INFORM
differentiation marker protein 14 in monocytes and macrophages. TREATMENT DECISIONS IN BOTH INITIATION
Its prognostic role has been extensively described but mainly AND CESSATION OF ANTIMICROBIAL THERAPY IN
in observational studies with small sample sizes. Presepsin is NEONATAL AND PEDIATRIC PATIENTS WITH SEPSIS
significantly higher in nonsurvivors in ED and ICU settings and OR RESPIRATORY INFECTIONS?
with a weighted pooled standardized mean difference of 1.09 There are numerous studies evaluating the role of PCT in
(95% CI, 0.78-1.41) for 30-day mortality. When compared to critically children in the pediatric ICU (PICU). Many of these
PCT, there is no conclusive evidence pointing to the superiority studies have investigated the use of PCT to help differentiate
of presepsin for mortality prediction. A meta-analysis of 9 serious bacterial infections and distinguish bacterial from viral

meningitis (85-92). Since the use of PCT in this manner informs 89.8% and 55.5% for serious bacterial infections, with an AUC
decisions on treatment and initiation of antibiotics, there has of 0.726 when tested in a cohort of 1247 infants presenting to
been much interest in using PCT to identify or rule out sepsis and a pediatric ED in Spain (93). The authors caution the use of this
serious bacterial infections. Unlike the adult studies, the pediatric prediction rule in young infants with a short history of fever.
studies did not directly evaluate the rate of antibiotic initiation Similar observations regarding the ability of PCT to rule out
and instead focused on evaluating the sensitivity, specificity, and bacterial infections have been made in children <18 years old
negative predictive value of PCT in detecting bacterial infections. with LRTI. Using a PCT cutoff of <0.25 ng/mL, a 96% negative
predictive value, 85% sensitivity, and 45% specificity was
PCT for Detecting Bacterial Infections and Initiating observed for ruling out typical bacterial CAP in a study of 532
Antibiotics in Pediatric and Neonatal Patients hospitalized children with radiographically confirmed CAP
A meta-analysis of 12 studies involving more than 7000 children (94). Another group had evaluated the utility of PCT in guiding
(<18 years) reported a sensitivity and specificity of 55% and 85%, antibiotic treatment in 319 children with pneumonia. A PCT-
respectively, for detecting serious bacterial infections (86). For guided algorithm using a threshold <0.25 ng/mL for withholding
invasive bacterial infections, sensitivity and specificity was 82% antibiotics in the study group (n = 155) resulted in 85.8%
and 86%, respectively, using a PCT cutoff of 0.5 ng/ mL. In this children receiving fewer antibiotic prescriptions and a shorter
analysis, serious bacterial infections included a broad spectrum exposure time (3.9% vs 25.2%) compared to the control group
of infections ranging from bacterial meningitis to urinary tract without a significant difference in recurrence of respiratory
infections as well as a subgroup of severe invasive bacterial symptoms or new antibiotic prescriptions in the following month
infections including bacterial meningitis, sepsis, and bacteremia. (95). However, another study of children (1 month-18 years old)
The negative predictive value of PCT was approximately 99% for presenting with LRTI found that using PCT-guided algorithms did
invasive bacterial infections and ranged from 79.5% to 96.7% for not alter antibiotic prescription rates but did reduce duration of
serious bacterial infections. antibiotic exposure by approximately 3 days. In this study, a PCT
A meta-analysis of 28 studies that included more than 2600 >0.5 ng/mL was used to initiate antibiotics and a PCT 0.25 to 5
neonates with suspected sepsis PCT showed a sensitivity of 85% ng/mL was left to clinical discretion (96).
for detecting sepsis, and, when combined with CRP, the sensitivity Collectively these studies suggest that PCT has insufficient
improved to 91% (87). In a prospective trial of 80 children, sensitivity in detecting sepsis, serious bacterial infections, or
PCT was better able to detect severe infections among PICU LRTI and should not be used as a stand-alone marker to make
patients when compared to CRP or white blood cell count (88). decisions on initiation of antibiotic therapy. Measuring PCT alone
Another prospective study of 64 PICU patients showed that PCT may miss patients with localized but serious infections (91). It
outperforms CRP alone in detecting bacterial infections; however, is important to note that PCT normal ranges can vary in healthy
PCT was insufficiently sensitive with an AUC of 0.71 (89). A study neonates within the first days of life (peaking at day 1 of life and
of 85 PICU patients with suspected sepsis demonstrated a negative decreasing to normal by day 2 and 3) (Table 1). Additionally, PCT
predictive value of 90% when a combination of CRP (<4 mg/dL) is not well studied in immunocompromised critically ill children,
and PCT (<1.75 ng/mL) was used to identify critically ill children and it is unknown whether these findings can be extrapolated to
with a low risk of bacterial infection (90). A retrospective study of this population.
more than 600 PICU patients reported a negative likelihood ratio
of 0.3 for PCT in ruling out infection (91). There is an increasing Antibiotic Cessation in Neonatal and Pediatric Populations
amount of evidence that suggests PCT in combination with other While there is no strong evidence to support using PCT
biomarkers may be useful in ruling out infection or identifying concentrations to guide antibiotic initiation in pediatric and
infants with low risk for serious infections. Kuppermann and neonatal populations, there is evidence that suggests PCT is useful
colleagues developed a clinical prediction rule (Pediatric to safely discontinue antibiotics in this population. Neonatal and
Emergency Care Applied Research Network) for identifying <60 pediatric PCT trials are summarized in Supplemental Table 4.
day-old infants across 26 EDs that are at low risk for serious The largest multicenter RCT in neonates that assessed the impact
bacterial infections to avoid unnecessary lumbar punctures, of PCT-guided decision-making on duration of antibiotic therapy
antibiotic exposure, and hospitalizations. This algorithm was the NeoPIns trial (97), which enrolled 1710 neonates of >34
incorporated urinalysis, absolute neutrophil count, and PCT, and weeks gestational age suspected of early-onset sepsis in the first
demonstrated a sensitivity of 97.7% and specificity of 60.0%, 72 hours of life. Eight hundred sixty-six neonates were assigned
with a negative predictive value of 99.6% (92). However, further to the experimental arm and 844 to the control arm. In the
validation of such multibiomarker algorithms is necessary. An experimental arm, PCT was measured at 12, 24, 36 to 72, and every
external validation of the Pediatric Emergency Care Applied 24 to 48 hours thereafter until discontinuation of antibiotics.
Research Network rule yielded a sensitivity and specificity of If 2 consecutive PCT values fell within the normal range for the

TABLE 1. Normal hourly values of post-birth PCT. for PCT in guiding discontinuation of antibiotics in both neonatal
and pediatric populations. Though the studies did not report any
TIME AFTER BIRTH (HOURS) PCT (ng/mL)a
adverse outcomes, the rate of reinfection or death in the control
0-6 0.5 arms was too low for the impact to be assessed.
6–12 2
KEY SUMMARY POINTS
12–18 5 • PCT should not be used as a stand-alone test for the
18–36 10 diagnosis of sepsis or to guide antibiotic initiation in
pediatric patients.
36–48 5 • PCT can guide safe cessation of antibiotics in neonates
and pediatric patients with suspected sepsis who show
48–60 2
clinical improvement.
60–72 1 • In neonates, PCT concentrations rise and fall rapidly, thus
cutoffs need to be stratified by age (hours after birth).
>72 0.5 • There is no consensus for PCT cutoffs or clearance rates to
PCT cutoffs used in NeoPIns trial (97).
a
guide duration of antibiotic therapy in pediatric patients.
age, antibiotics were discontinued. PCT-guided decision-making IS PCT AN ACCURATE PREDICTOR OF OUTCOMES
led to a 10-hour reduction in antibiotics exposure. However, (MORTALITY, RESPIRATORY FAILURE, SHOCK) IN
the impact on reinfection and death during the first month of PEDIATRIC POPULATIONS?
life could not be entirely assessed due to low occurrence rates A few studies have examined the role of PCT as a prognostic
of these adverse events. The cutoffs for PCT in this study ranged predictor in pediatric populations. As with adult populations,
from 0.5 to 10 ng/ mL, stratified based on hours after birth (Table higher PCT concentrations are associated with sepsis severity
1). It is worth noting that the PCT discontinuation or continuation and increased risk of death (100-103). Conversely, low or normal
recommendation per protocol was overruled in 25% of neonates PCT values have excellent negative predictive values for adverse
in the experimental arm. A 22.4-hour reduction in antibiotics outcomes. In an observational prospective study of 65 children
exposure was observed in an earlier single-center study with 121 with meningococcal infections, patients with PCT concentrations
neonates (61 in control arm and 60 in experimental arm) (98). <10 ng/mL survived, whereas all patients with PCT ≥ 10 ng/mL
The study criteria and design were the same as the NeoPIns trial; developed multiple organ dysfunction syndrome or died (104).
however, a single <10 ng/mL PCT cutoff was used in this study. Another study looking at PCT kinetics in pediatric patients with
Additionally, the sample size was not sufficient to assess safety systemic inflammatory response syndrome and organ failure
measures. One of the limitations of the NeoPIns trial is that this after open heart surgery showed similar outcomes, where PCT
study included only neonates >34 weeks of gestational age, and concentrations <10 ng/mL post-surgery survived (105).
the utility of PCT in preterm infants is unknown.
A similar reduction of antibiotic exposure has been reported Outcomes from Single PCT Measurements
in children <18 years old. A single-center study from Spain Single PCT measurements at admission cannot predict the
examined the impact of implementation of a PCT-guided protocol likelihood of a patient developing severe sepsis or septic shock.
in antibiotic decision-making (99). One hundred fourteen patients In a retrospective single-center study evaluating 109 critically
were examined prior to implementation of the protocol and 112 ill children who had a PCT measurement within 48 hours of
after implementation. Antibiotics were discontinued if there was admission, 61 patients with septic shock had a median PCT
a 50% decrease in PCT value or PCT values dropped below <0.5 concentration of 7.16 ng/mL with an IQR of 2.21 to 42.28 ng/mL,
ng/mL. Implementation of this PCT-guided protocol resulted in whereas another 48 patients without septic shock had a median
a reduction of 1.1 days of antibiotic exposure without adverse PCT concentration of 0.91 ng/mL and an IQR of 0.10 to 10.80 ng/
outcomes. The compliance of antibiotic de-escalation in the PCT mL (91). Though PCT concentrations were higher in patients who
protocol was only 54.8%; however, prior to PCT implementation, progressed to severe sepsis or septic shock, there was significant
de-escalation only occurred in 26% of the patients. Using a PCT overlap in PCT concentrations between the 2 groups. Therefore,
cutoff of <0.25 ng/mL in the pediatric LRTI population has also PCT measurements alone are unable to predict probability of
demonstrated a reduction in antibiotic exposure (95, 96). developing severe sepsis or septic shock. Similarly, in another
The studies evaluating the role of PCT for antibiotic cessation study of 64 patients with meningococcal sepsis and septic shock,
are limited in pediatric populations, and further investigation is median PCT levels on admission were higher in children with
warranted. Overall, the studies to date suggest that there is a role septic shock compared to children with sepsis (270 ng/mL and

64.4 ng/mL, respectively). However, in this study there was no WHEN AND HOW OFTEN SHOULD PCT BE
significant difference in PCT ranges for survivors (n = 42, 5.7- MEASURED? WHICH CUTOFF(S) SHOULD BE USED?
672.3 ng/mL) and nonsurvivors (n = 13, 55646.4) with septic PCT-guided algorithms have been investigated to optimize
shock (106). Similarly, in a study of 75 children with septic shock, antimicrobial therapy and predict outcomes including mortality,
though nonsurvivors had higher PCT concentrations (median 273 disease progression, and length of stay. This strategy has safely
ng/mL for nonsurvivors vs 82 ng/mL for survivors), the range reduced antibiotic treatment in septic patients in different
of PCT concentrations among survivors and nonsurvivors was clinical settings (i.e., ED, ICU) and various etiologies, particularly
similar (3.3-759.8 ng/mL and 5.1-736.4 ng/mL, respectively). respiratory infections (Supplemental Table 1 and 2). However,
The maximum length of ICU stay in this study was 32 days (100). recent trials did not confirm these findings (17-20, 38).
Algorithms evaluated across studies differ in positive cutoff(s),
Outcomes from Sequential PCT Measurements timing of serial testing, and PCT assays used. Commonly, PCT
Hatherill and colleagues also evaluated changes in PCT algorithms consist of recommendations to initiate or discontinue
concentrations after treatment. They found that of 39 children antibiotics, typically using different cutoffs based on acuity,
with sequential PCT measurements, 16 (41%) showed no fall clinical setting (i.e., ED, ICU), and patient population.
in PCT after 24 hours of treatment, and the observed mortality
in this group was 44% compared to 9% in 23 patients who Timing and Frequency for Antibiotic Initiation and
showed a decline (100). The RESOLVE phase III trial examined Cessation in Adults
the biomarker response in children with severe sepsis after The recommended PCT cutoff tiers for patients in the ED and
treatment with either placebo or Drotrecogin alfa. In this hospital wards are <0.1, 0.1 to 0.25, 0.26 to 0.5, and >0.5 ng/
study, 251 survivors showed a decline in PCT concentrations mL (Supplemental Tables 1 and 2). In patients with uncertain
compared to the 37 nonsurvivors who showed an increase clinical suspicion of infection, PCT concentrations <0.1 and 0.1
in PCT concentrations 24 hours post-treatment regardless of to 0.25 ng/mL indicate that initiation of antibiotics is strongly
whether patients were in the trial or control arm (107). Overall, discouraged and discouraged, respectively. However, if there is
these studies demonstrated that a decline in PCT values after no clinical improvement, the studies referenced in Supplemental
treatment is associated with better survival rates; however, the Tables 1 and 2 support that PCT should be rechecked after 6 to 24
percent change in PCT concentrations were not provided in these hours. PCT concentrations 0.26 to 0.5 and >0.5 ng/mL indicate
studies. Poddar et al. evaluated whether reduction in PCT can that antibiotics are encouraged and strongly encouraged,
predict 28-day mortality in 20 children admitted to the PICU with respectively, and remeasuring PCT every 2 to 3 days to assess for
severe sepsis or septic shock. Of the 14 children who survived the opportunity of discontinuation of antibiotics is recommended.
to 28 days, the percent reduction in PCT was 75.5% compared Upon re-evaluation, PCT concentrations <0.1 and 0.1 to 0.25
to a 200% increase in PCT concentrations in the 6 nonsurvivors ng/mL strongly encourages continuation and discontinuation
between day of admission and 72 to 96 hours later(108). of antibiotics, respectively. When PCT drops by >80% from its
Given the limited number of patients evaluated in these peak value, antibiotic discontinuation is also recommended.
studies and the low rate of mortality, larger studies are needed PCT concentrations 0.26 to 0.5 and >0.5 ng/mL discourage and
to determine extent of PCT reduction required to accurately strongly discourage discontinuing antibiotics, respectively. A
predict mortality in pediatric patients. Additionally, outcomes simplified approach for patients with moderate illness outside of
assessment was only a secondary measure assessed in most of the ICU consists of empiric antibiotic initiation based on clinical
these studies, and therefore details of PCT kinetics and percent practice guidelines and initial assessment or, if PCT is 0.25 ng/
reduction were largely missing. mL, measuring PCT at least daily and stopping antimicrobial
therapy if PCT decreases to <0.25 ng/mL or by at least 80%
KEY SUMMARY POINTS from peak concentration combined with improvement of clinical
• An elevated PCT is generally suggestive of a worse symptoms (109). Not surprisingly, different RCTs attempting to
outcome in pediatric patients with severe sepsis or septic assess the feasibility of using PCT for initiating or discontinuing
shock. antibiotics across different patient populations, and settings have
• A single PCT measurement has limited prognostic value utilized different algorithms and cutoffs, as discussed earlier. Less
since many studies have shown significant overlap in PCT is known about the broader applicability of these algorithms for
concentrations among survivors and nonsurvivors.
predicting other outcomes such as mortality.
• Serial PCT measurements may be predictive of mortality
during ICU stay; however, additional studies are needed to
define interpretive criteria in pediatric and neonatal patients.
Timing and Frequency for Outcomes Prediction in Adults
In the United States, most PCT assays were initially FDA cleared
to predict disease progression. More recently, some intended

uses have expanded to mortality risk assessment and antibiotic Change in PCT Concentrations over Time to Predict
management decision-making. A higher 28-day risk of all-cause Outcomes in Adults
mortality is predicted in patients with PCT concentrations that The MOSES study found that while baseline PCT was a poor
increase or decline < 80% from the day severe sepsis or septic predictor of 28-day mortality (AUC 0.56;95% CI, 0.51-0.60), failed
shock was first diagnosed (day 0) or day 1 to day 4 (63). In this PCT clearance (≤80%) between baseline and day 4 doubled the
context, the main overlap with the cutoffs from the antibiotic mortality (63). On day 1, mortality was 3-fold lower in patients
initiation/cessation algorithms discussed previously is the with decreased PCT (29% vs 12%), regardless of the initial
assessment of PCT clearance in patients with high acuity disease. PCT concentration. In a study in patients with intra-abdominal
Not all studies that demonstrated an association between sepsis, 5-day 70% PCT clearance predicted mortality while PCT
PCT concentrations at presentation and 28/30-day mortality clearance at days 3 and 4 did not differentiate survivors and
reported a concentration cutoff, and those that did used a variety nonsurvivors (64). Lower PCT clearance cutoffs have also been
of cutoffs, complicating our ability to recommend an evidence- reported to effectively predict mortality. Elke et al. demonstrated
based cutoff for mortality prediction. Moreover, reported cutoffs that patients with PCT clearance by 20% at day 1 or by 50% at
differ across patient populations and diseases. day 4 had lower mortalities in the ICU and in-hospital (16.8%
Some studies used a cutoff of 0.25 ng/mL (60, 109), and 24.1% vs 28.9% and 30.4%, respectively) (52). In patients
while others used cutoffs such as those described earlier with autoimmune disease, PCT peak concentrations did not
for antimicrobial stewardship (i.e., 0.1, 0.25, 0.5, 2.0 ng/mL) differ between survivors and nonsurvivors while PCT clearance
(50, 56), and others reported a variety of cutoffs (51, 54, 55) on days 5 and 7 was significant (P = 0.06 and 0.005, respectively)
(Supplemental Table 3). The study by Kutz and colleagues, a meta- (61). In this study, clearance was calculated from the PCT peak
analysis of 14 trials, found an association between increasing concentration, an approach difficult to operationalize as it is
PCT concentrations at presentation and adverse outcomes challenging to predict when the peak PCT will occur prospectively.
such as treatment failure and mortality but only in ED patients The value of both PCT concentrations and clearance for
and in patients with ARI and CAP but not in ICU or primary 28/30-day, ICU, or in-hospital mortality prediction was evaluated
care settings. A cutoff of 0.1 ng/mL resulted in sensitivities of in a meta-analysis, which included 23 studies (up to 2014) with
86.1 (95% CI, 82.4-89.3) and 92.5% (95% CI, 86.2-96.5) in ED 3994 patients (51). The studies associating absolute values
patients for treatment failure and mortality, respectively, and of PCT with mortality included different clinical settings (ED
>90% for mortality in patients in the ICU or with ARI or CAP. and various ICUs) and PCT cutoffs (0.1214.27 ng/mL). For the
For these same parameters, specificity approaches 80% at a evaluation of PCT clearance, the studies were mostly ICU (one
PCT cutoff of 2.0 ng/mL. Similarly, a multinational prospective surgical ICU) and the clearance cutoffs ranged from 25% to 70%.
study evaluating several PCT cutoffs (0.05, 0.1, 0.25, 0.5 ng/mL) In both cases, sample size and mortality varied significantly. The
in samples collected at ED admission from nearly 7000 patients AUCs were 0.77 (95% CI, 0.73-0.80) and 0.79 (95% CI, 0.75-0.83)
showed association between 30-day mortality and increasing for a single PCT concentration vs PCT clearance, respectively.
PCT concentrations (56). The study reported OR of 7.31 (95% Nonclearance is associated with a relative risk for mortality of
CI, 3.3214.75) for patients with an admission PCT >0.5 ng/mL. 3.05 (95% CI, 2.35-3.95). The meta-analysis provides helpful
In comparison, samples with PCT < 0.1 ng/mL or <0.05 ng/mL insight into the value of PCT for mortality prediction but does
had OR of 1.71 (95% CI, 0.87-3.34) and 1.0, respectively. A large not help elucidate the optimal cutoff to use or the definition of
multi-center prospective study of more than 1700 patients with clearance and time points used.
septic shock and PCT measured in the ED at diagnosis derived Unfortunately, the degree of heterogeneity in these studies
an optimal PCT cutoff of 0.17 ng/mL but found that PCT was not is significant (82). With such heterogeneity and inconsistency
an independent predictor of 28-day mortality (62). In their study across studies, it is difficult to make a uniform recommendation
of severely ill trauma patients, those with PCT of 5 ng/mL or regarding the utility of PCT in predicting prognosis. Further,
greater were at higher risk of dying (OR 3.65; 95% CI, 1.03-12.9) even if a consistent cutoff for the PCT delta calculation were
(55). A study in patients with tuberculosis reported that baseline identified to predict 28-day mortality, it is unclear as to whether
PCT concentrations >0.13 ng/mL predicted mortality with OR of these predictions have any impact on patient care. Additional
7.9 (95% CI, 3.2-19.7); however, reported cutoffs for mortality studies are needed to establish standardized cutoffs and/ or PCT
ranged from 0.05 to 0.12 on days 7, 14, and 28, highlighting the clearance parameters in the prediction of 28-day mortality and to
need for a simplified approach to integrating PCT into mortality determine the clinical utility of a 28-day mortality prediction in
prediction, such as an integrated risk score. patients with sepsis and LRTI patients.

Timing and Frequency for Antibiotics Initiation and The NeoPIns trial provides the strongest evidence and
Cessation in Pediatric Patients guidance for cessation of antibiotics in neonates >34 weeks
The majority of evidence described earlier in the pediatric of gestational age and with onset of sepsis within the first 72
populations demonstrates that PCT has some utility in identifying hours of life. In the trial when PCT concentrations fell within the
patients with sepsis and severe bacterial infections; however, normal range for the age (Table 1), antibiotics were discontinued.
PCT does not have adequate sensitivity to serve as a stand-alone PCT was measured at 12, 24, 36 to 72 hours, and every 24 to 48
test to guide decisions on antibiotic therapy initiation. A meta- hours until discontinuation of antibiotics (97). The NeoPIns trial
analysis consisting of more than 7000 children showed that a demonstrated a 22.4-hour reduction in antibiotic exposure. In
PCT cutoff of 0.5 ng/mL had a sensitivity of only 55% (specificity children <18 years old, one single-center study evaluated a PCT
85%) for detecting serious bacterial infections, and the negative reduction criteria of 50% or PCT decrease below 0.5 ng/mL.
predictive value ranged from 79.5% to 96.7% (86). A prospective PCT was measured at baseline, 24, 48, and 72 hours of antibiotic
study of 64 PICU patients showed that a PCT cutoff of 2.5 ng/ mL at treatment. Using this criteria for antibiotic cessation, a 1.1-day
admission had optimal sensitivity and specificity (68% and 74%, reduction in antibiotic exposure was observed without adverse
respectively), with a negative predictive value of 78%, whereas a outcomes (99). While more studies are needed in pediatric and
CRP cutoff of 40 mg/L had a sensitivity of 95% and specificity of neonatal populations, the evidence for PCT use in antibiotic
42% with a negative predictive value of 94% (89). Another study cessation is encouraging.
of 85 PICU patients demonstrated a negative predictive value of Guidelines from the American Academy of Pediatrics on
90% for ruling out bacterial infection in patients with systemic assessment of febrile well-appearing infants 8 to 60 days old
inflammatory response syndrome when PCT and CRP were recommend using PCT >0.5 ng/mL for initiating antibiotic
used together, with a CRP cutoff of <4 mg/ DL and PCT cutoff of treatment, citing evidence that PCT is an independent predictor of
<1.75 ng/mL measured <4 hours after initiation or expansion of bacterial infections with better performance characteristics than
antibiotics (90). A retrospective analysis of 600 PICU patients CRP, absolute neutrophil count, and white blood cell count (110).
reported a negative likelihood ratio of 0.3 for PCT using a cutoff of
<0.1 ng/mL. When used in combination with a CRP < 0.8 mg/dL, Timing and Frequency for Outcomes Prediction in
the negative likelihood ratio was 0.1 for bacterial infection (91). Pediatric Patients
A meta-analysis of 2600 neonates showed that using PCT The prognostic ability of single increased PCT measurement is
in combination had a sensitivity of 91% for detecting neonatal poor in predicting mortality, as the studies mentioned previously
sepsis compared to using CRP and PCT alone, with a sensitivity of showed that even though the median PCT concentrations were
71% and 85%, respectively. The cutoff intervals for PCT proposed higher in nonsurvivors compared to survivors, the ranges of PCT
by subanalysis was 0.5 to 2 ng/ mL, and a cutoff value of >10 concentrations in both groups overlapped significantly (111).
mg/L for CRP yielded the highest sensitivity and specificity (87). Thus, a PCT cutoff concentration for mortality predication cannot
Due to the rapid changes in PCT concentrations in neonates, be recommended. As evidenced by the MOSES trial in the adult
caution should be used for assigning an absolute PCT cutoff in population, serial PCT measurements are likely more useful
neonates <72 hours of age. As in the pediatric population, the than single PCT measurements in predicting the risk of death in
cutoff for PCT used to detect bacterial infections varies by study pediatric populations. However, large prospective multicenter
in neonates and will be dependent on the algorithmic approach. studies examining the prognostic accuracy of PCT reduction are
For example, in the prediction rule described by Kuppermann lacking for the pediatric population. Additionally, small sample
and colleagues to rule out serious bacterial infections in infants, sizes, low death rates, and differences in study design hinder
a PCT concentration of <1.71 ng/mL was used in conjunction the ability to make any conclusions on timing, frequency, cutoffs,
with a negative urinalysis result and absolute neutrophil count and reduction for PCT measurements needed to make accurate
of <4090/^L (92). Collectively, these studies suggest that PCT in predictions of mortality (100, 107). Only one single-center study
combination with CRP has superior performance in ruling out evaluated the ability of serial PCT measurements 4 days apart to
bacterial infection compared to PCT measurements alone in both predict 28-day morality. The authors found that a 75.5% reduction
pediatric and neonatal populations. However, the timing of PCT in PCT concentrations was seen in survivors;however, the study
measurement and proposed cutoffs vary greatly study to study. size was small, consisting of only 20 children (14 survivors and
Overall, the major limitations for comparing studies examining 6 nonsurvivors) (108). It is also important to note that PCT is
the value of PCT in detecting sepsis or bacterial infections are usually not the only factor in predicting mortality in septic ICU
a lack of a uniform definition for sepsis, varied inclusion and patients as other clinical information (cause of illness, other
exclusion criteria, and lack of harmonized criteria for time of underlying conditions, and other test results) are considered.
sampling and interpretation.

KEY SUMMARY POINTS published, tailoring and developing an institution-specific
• Decision thresholds for antimicrobial use/discontinuation algorithm is important based on the patient population, type
vary by diagnosis and acuity and should considered in the of infections, or level of acuity seen at the institution, as well
context of other clinical signs and symptoms. as the analytical assay used. In the real-world clinical setting,
• No consistent PCT cutoff concentration has been overruling of an algorithm was commonly seen in critically ill
established to predict mortality. patients, thus the development of different algorithms based
• Insufficient PCT clearance over time is associated with a
on PCT interpretation in mild, moderate, or severe presentation
significantly greater risk of 28- to 30-day mortality in septic
with differing antibiotic recommendations based on severity may
and LRTI patients.
be warranted (109). As with any other antimicrobial stewardship
• No consistent PCT clearance calculation parameters have
been established to predict 28- to 30-day mortality. intervention implemented at an institution, continued
• Routine measurement of PCT to predict mortality is not prospective audit with provider feedback on appropriate use and
recommended due to lack of consistent PCT cutoffs and/ interpretation of PCT will likely be needed (116).
or PCT clearance parameters and insufficient evidence Moradi et al. describes the implementation of a clinical
demonstrating a benefit to estimating 28-day mortality risk decision support tool to alert a prescriber in the electronic
in septic and LRTI patients. medical record in a pre- vs post- intervention quasi-experimental
study (117). If the patient met 3 criteria: (1) had a positive
HOW SHOULD PCT BE INCORPORATED INTO viral PCR result by FilmArray® Respiratory Panel, (2) had a
ANTIMICROBIAL STEWARDSHIP EFFORTS? PCT result of <0.25 ng/mL, and (3) had one or more antibiotics
As adherence to a predefined PCT algorithm has been reported to ordered, an alert would populate when a prescriber opened the
vary in real-world PCT studies, active antimicrobial stewardship patient's chart indicating "the results suggest a viral infection,
intervention paired with PCT results may be needed. Other rapid please reassess necessity of antibiotics as indicated." Without
diagnostic technology, primarily in microbial identification from real-time intervention made by an infectious diseases physician
cultures, has shown mortality benefit only when an antimicrobial or pharmacist, this intervention provided a significant decrease
stewardship intervention was paired with the result (112). in inpatient antibiotic days of therapy (post- intervention mean
Effective antimicrobial stewardship requires a multidisciplinary 5.8 days vs 8 days, P < 0.001) and rate of discharge antibiotic
team effort involving infectious disease physicians, nurses, prescription (post- intervention 20% vs 47.8%, P < 0.001) and
pharmacists, and laboratorians (113). The Center for Disease duration of outpatient antibiotic therapy (0.9 days vs 2.4 days,
Control and Prevention's 2019 Core Elements of Hospital P< 0.001). Antimicrobial stewardship programs should similarly
Antibiotic Stewardship Programs guidelines indicate that the use implement strategic use of clinical decision support tools in
of procalcitonin might help identify patients in whom antibiotics conjunction with rapid diagnostic tools to optimize the use of
can be stopped because bacterial pneumonia is unlikely (114). PCT in antibiotic decision-making.
These guidelines indicate that laboratory and stewardship
personnel can work collaboratively to present lab data in a way KEY SUMMARY POINTS
that supports optimal antibiotic use and is consistent with the • Antimicrobial stewardship programs should implement
institution's expected practices. Further, the laboratory can strategic use of clinical decision support tools in
collaborate with stewardship program personnel to develop conjunction with rapid diagnostic tools to optimize the use
guidance for clinicians around clinical decision-making based on of PCT in antibiotic decision-making.
laboratory results.
We reviewed the literature for studies that assessed the WHAT PREANALYTICAL FACTORS AFFECT PCT
effectiveness of incorporating PCT into antimicrobial stewardship RESULTS AND/OR INTERPRETATION?
efforts. Many of the reviewed studies describe providing extensive Acceptable Sample Types
education and feedback to the prescribers and the development The predominant sample used for the analysis of PCT is serum
of a clinical decision pathway or algorithm to provide guidance or plasma obtained from the collection of venous whole blood.
prior to implementation of PCT at their institution (17, 29, 35, Most data from regulatory filings of PCT assays suggest no
115). Examples of how education was provided include didactic difference between serum or plasma (EDTA, lithium heparin,
educational sessions, case reviews, and creation of pocket cards sodium heparin) when using the acceptance criteria: slope =
(115). Notification to the prescriber of the PCT result through 1.0 ± 0.1, r2 ≥ 0.95 (118-122). Arterial samples have also been
text page or email with guidance on interpretation of the result reported to yield comparable PCT values to venous samples
or prospective audit and direct feedback to the prescriber may (123). The validity of capillary blood samples is an important
provide benefit (38). Although the specific cutoff of PCT values consideration for potential point-of-care testing applications of
to recommend antibiotic therapy varied among the studies PCT, and one study showed that capillary and venous samples

were comparable when analyzed on the same platform (slope = WHAT FDA-CLEARED METHODS ARE AVAILABLE TO
1.01 [95% CI, 1.00-1.05], intercept = 0 [95% CI, 0-0], r2 = 0.98) MEASURE PCT AND HOW DO THEY COMPARE?
(124). However, direct comparison between capillary blood on Numerous FDA-cleared immunoassays are available to measure
the investigated device and venous blood on reference methods PCT (Table 2). A detailed analytical overview of many of these
showed greater bias, as discussed further later. assays has been provided by Schuetz and colleagues as well
as the International Federation of Clinical Chemistry and
Stability and Storage Laboratory Medicine Working Group on Standardization of
The only known enzymes to process PCT are found intracellularly, Procalcitonin Assays (127, 128). While many PCT assays were
thus there is little known contribution of extracellular proteases initially approved only for indications of risk assessment (e.g., for
to PCT degradation in vitro. Published data have shown no progression to severe sepsis and septic shock) and/or risk of 28-
significant change in PCT concentration during room temperature day all- cause mortality, most assays are now also approved for
storage for up to 24 hours (123, 125). PCT stability can be further antibiotic therapy decision-making in patients with suspected or
prolonged by storage at lower temperatures, such as up to 5 days confirmed sepsis or LRTI (CAP, acute bronchitis, and AECOPD).
at 4°C and more than 2 weeks at < -20°C (123, 125, 126). Limited We reviewed the literature to compile reported statistics from
studies have demonstrated that up to 3 freeze-thaw cycles do not method correlation experiments performed between the various
significantly impact PCT levels (123, 125). available PCT assays (Supplemental Table 5). For the purposes of
this review, we considered the BRAHMS PCT-sensitive Kryptor to
KEY SUMMARY POINTS be the reference method, as it was an early-available method that
• PCT shows excellent agreement between arterial and was used in most of the initial clinical trials that first established
venous samples in common vacutainer tubes. PCT clinical decision points (39, 40). Our subsequent review
• Limited data suggest that capillary blood could be an of method correlation studies is divided into assays that use
acceptable PCT sample type, though more studies BRAHMS-licensed PCT reagents, Diazyme-licensed PCT reagents,
are required. and other PCT reagents.
TABLE 2. Features of POCT and laboratory testing hCG.a

MANUFACTURER (PLATFORM) BRAHMS PCTa DIAZYME PCTb OTHERb
Abbott (Alinity) X
Abbott (Architect) X X
Beckman Coulter (AU) X
Beckman Coulter (Unicel, Access) X (Access PCT)
bioMerieux (VIDAS) X
BRAHMS (Kryptor) X
DiaSorin (LIAISON) X
Diazyme (DZ-Lite) X
Fujirebio (Lumipulse) X
Ortho-Clinical (VITROS) X
Roche (cobas) X X
Siemens (Atellica) X
FDA-approved indications: to aid in the risk assessment of critically ill patients on their first day of ICU admission for progression to severe sepsis and
a
septic shock; to aid in assessing the cumulative 28-day risk of all-cause mortality for patients diagnosed with severe sepsis or septic shock in the ICU or
when obtained in the emergency department or other medical wards prior to ICU admission, using a change in PCT level over time; to aid in decision-
making on antibiotic therapy for patients with suspected or confirmed lower respiratory tract infections defined as community-acquired pneumonia,
acute bronchitis, and acute exacerbation of chronic obstructive pulmonary disease in an inpatient setting or an emergency department; to aid in
decision-making on antibiotic discontinuation for patients with suspected or confirmed sepsis.
b
to aid in the risk assessment of critically ill patients on their first day of ICU admission for progression to severe sepsis and septic shock.

BRAHMS PCT Assays (Supplemental Table 5). Interestingly, the 2 most extreme slopes
Ten published studies that compared various PCT assays to the were generated from studies on the same analyzer (Roche cobas c
BRAHMS PCT-sensitive Kryptor were included in our review (124, 702), though the negative bias was generated from serum studies
125, 129-136). Nine of these studies included a test method(s) and the positive bias was generated from lithium-heparin plasma
that utilized BRAHMS-licensed reagents, and these studies studies, each from separate investigators. Overall, correlation
were performed across several immunoassay analyzers. For the coefficients were lower with the Diazyme reagent methods
BRAHMS assays, as compared to the Kryptor reference method, than with the BRAHMS reagent methods. Analysis of categorial
reported slopes ranged from 0.795 to 1.40, and correlation characterization often showed lower agreement at clinical
coefficients (r) ranged from 0.8864 to 0.997 (Supplemental Table decision points for the Diazyme assays than for the BRAHMS
5). The Abbott ARCHITECT method repeatedly showed slight to assays. For example, Dipalo et al. found agreement ranging
moderate negative proportional biases across multiple studies from 83% to 86% at 0.5 ng/mL for 4 Diazyme methods, while
(slopes ranged from 0.8060.97), while the bioMerieux VIDAS and agreement increased to 90% to 92% at 2.0 ng/mL and to 98%
MINI VIDAS consistently showed positive proportional biases at 10.0 ng/mL (136). Similarly, Lippi et al. reported agreement
(slopes ranged from 1.188-1.40). Correlation of the Roche Elecsys between 83% (at 0.25 ng/mL) and 96% (10 ng/mL) (130).
cobas methods to the Kryptor varied by the analyzer model; the
lower-throughput e411 showed a negative proportional bias Other PCT Assays
(slope = 0.795) while the higher-throughput e600 series and e801 Assays not classified as using BRAHMS or Diazyme-licensed
exhibited slopes closer to 1. Intercepts that may be indicative of PCT reagents include the Beckman Access assays and Snibe
clinically significant onstant biases included the ADVIA Centaur MAGLUMI. One published study was included that compared each
(intercept = 0.40) and the cobas e 601 (intercept = -0.47). Other of these methods to the Kryptor method (130). While the Access
cobas analyzers did not demonstrate significant intercepts. methods showed reasonable agreement to the Kryptor method,
Several of the studies further characterized method the MAGLUMI showed a large positive proportional bias (slope =
correlation by quantifying the percent agreement of categorical 1.51) that yielded overestimates from the Kryptor (Supplemental
interpretation of PCT results relative to the Kryptor at commonly Table 5). Agreement with the Kryptor at clinical decision points
used clinical decision points. Dipalo et al. found optimal was at least 96% across all concentrations (0.10, 0.25, 0.5. 2.0,
agreement ranging from 94% to 98% for the Centaur, cobas and 10.0 ng/mL) for the Access assays, while agreement for the
e601, LIASION, and VIDAS at the following concentrations: 0.5, MAGLUMI ranged from 91% to 96%.
2.0, and 10.0 ng/mL (136). Similarly, Lippi et al. found agreement
of 96% to 99% for the BRAHMS methods tested at 0.10, 0.25, 0.5, KEY SUMMARY POINTS
2.0, and 10.0 ng/mL (130). Conversely, Gruzdys et al. found that • BRAHMS-licensed PCT immunoassays have demonstrated
an overall negative bias of the ARCHITECT method translated to good overall correlation with the Kryptor method and have
predicted medical decision concentrations for the ARCHITECT shown high categorical agreement around common clinical
that were significantly lower than their Kryptor counterparts, decision points. Non-BRAHMS assays have generally
particularly at 0.50 and 2.00 ng/mL (125). However, not all shown reduced correlation to the Kryptor.
• Biases across methods may impact reference intervals and
ARCHITECT studies found reduced agreement at clinical decision
the interpretation of PCT results across clinical decision
points. The varied results of agreement studies are likely due, in
ranges.
part, to the small number of data points included within each
concentration range.
A point-of-care testing method, BRAHMS PCT direct, showed ARE CLINICAL DECISION POINTS (CUTOFFS)
reasonable correlation to the Kryptor lab-based method, more COMPARABLE ACROSS PCT ASSAYS?
so in venous whole blood (slope = 0.98) than in capillary (slope While there is currently no true reference method for PCT, an
= 0.90) samples (124). It should be noted that this study used assay's agreement with the BRAHMS PCT-sensitive Kryptor
a combination of both the BRAHMS PCT-sensitive Kryptor and assay provides a basis to determine the feasibility of use of
Elecsys BRAHMS PCT assays as the reference method and did not common PCT clinical decision points. Several published studies,
distinguish between them for this analysis. described previously, demonstrate that while the various PCT
methods generally compare to the BRAHMS PCT-sensitive
Diazyme PCT Assays Kryptor with reasonable agreement, there are minor to moderate
Four published studies were included that compared Diazyme biases in patient sample materials that suggest that method
PCT reagents on various immunoassay analyzers to the Kryptor harmonization has not been achieved. These biases preclude
method (129, 130, 133, 136). Reported slopes ranged from 0.6543 PCT from being trended across multiple methods and require
to 1.19, and correlation coefficients (r) ranged from 0.85 to 0.960 the use of assay-dependent reference intervals. Importantly,

these biases also impact the interpretation of PCT results across • Labs implementing PCT or changing PCT methods should
clinical decision ranges, especially at low PCT concentrations, verify or establish the reference interval of their method
which have the highest implications for diagnostic purposes and and perform robust correlation studies to the BRAHMS
antibiotic decision-making. Our review indicates that assays that PCT-sensitive Kryptor or another BRAHMS method using
use BRAHMS-licensed reagents have demonstrated stronger patient samples across clinically relevant concentration
ranges.
correlation to the BRAHMS PCT-sensitive Kryptor method than
other assays that have been introduced more recently. The
analytical differences observed in these correlation studies could WHAT ARE POSSIBLE CONFOUNDING FACTORS
be attributed to many factors, including the various detection FOR THE INTERPRETATION OF PCT RESULTS?
methods used by the available immunoassays as well as the Proper interpretation of PCT results requires careful
PCT antibodies used in the assay reagents. Assay manufacturers consideration of the patient's clinical condition and the myriad
typically provide limited to no information regarding the of factors that can influence PCT. While PCT is often used as a
antibody design or specific PCT epitopes targeted in their surrogate marker for bacterial infections, there are conditions
assays, though BRAHMS assays have been reported to employ 2 where elevated PCT can be observed in the absence of bacterial
monoclonal antibodies on the calcitonin and katacalcin segments, infection. These clinical situations include inflammatory events
respectively (127). such as severe trauma, various major surgeries, and cardiogenic
Notably, the International Federation of Clinical Chemistry shock. Patients under these circumstances typically require
and Laboratory Medicine has an active Working Group on monitoring for systemic infection. However, PCT should be
Standardization of Procalcitonin Assays seeking to develop interpreted with caution in the context of these nonbacterial
and validate standard reference materials and a reference elevations, particularly if acute inflammatory conditions occur
measurement procedure for PCT by stable isotope dilution in between trended PCT measurements. While PCT may show
mass spectrometry (137). The International Federation of general elevations in burns, some studies have suggested that
Clinical Chemistry and Laboratory Medicine Working Group PCT may still be able to distinguish septic from nonseptic burn
on Standardization of Procalcitonin Assays has indicated that patients (111, 138). Some nonbacterial infections, such as malaria
PCT assays belonging to the BRAHMS family participate in and some fungal infections, have also shown to non-consistently
a harmonization program using the BRAHMS PCTsensitive elevate PCT (139, 140). PCT typically remains low in viral
Kryptor as a reference method, which supports our findings infections, as it is inhibited by interferon-y, and this is the basis
that BRAHMS assays correlate better with the Kryptor than for the proposed use of PCT to distinguish bacterial from viral
non-BRAHMS assays (127). However, there remains a need for infections. However, PCT elevations have been observed in some
higher order reference material such that all PCT assays could viral infections, including in patients with SARS-CoV-2 infection
be harmonized. In the meantime, laboratories seeking to newly (141). The etiology of PCT elevation in some patients with severe
implement PCT methods should perform robust accuracy COVID-19 is not well understood, and further studies are needed
studies to a reference method, preferably the BRAHMS PCT- to determine the impact of other confounding factors, such as
sensitive Kryptor or a BRAHMS method, and should carefully bacterial co-infection. Additionally, patients with significantly
consider the implications for clinical interpretation. Further, compromised renal function may not be able to clear PCT at a
each laboratory should establish or verify a reference interval normal rate, and this can potentially cause PCT elevations (142).
specific to the chosen platform. Laboratories should work closely Finally, there is a natural elevation of PCT in healthy neonates
with clinical colleagues, such as those in infectious diseases and just after birth (143). While there is growing evidence that PCT
on antimicrobial stewardship teams, to align PCT interpretive may have a role for guidance of antibiotic therapy in neonates,
algorithms with the chosen analytical method. neonatal-specific reference ranges should be used. Generally,
in cases of PCT elevations due to nonbacterial inflammatory
KEY SUMMARY POINTS processes, traditional PCT clinical decision points for outcome
• There is not yet a reference method for PCT. In the prediction and antimicrobial therapy decisions will not be valid.
meantime, the BRAHMS PCT-sensitive Kryptor assay Alternatively, the kinetics (increases or decreases over time) of
should be considered the gold standard since it was used PCT may be monitored and considered in the context of potential
in most of the initial clinical trials that first established PCT background increases.
clinical decision points.
• The same PCT method should be used to trend PCT
KEY SUMMARY POINTS
values for the same patient, as there is currently an
• Various clinical scenarios outside of bacterial infection may
absence of method harmonization and biases exist
cause elevated PCT results, including trauma, surgery,
between current methods.
shock, and renal dysfunction.

• PCT should be interpreted with caution in clinical settings
Author Contributions
of nonbacterial inflammatory processes and should not be
The corresponding author takes full responsibility that all
trended around acute inflammatory events.
• Traditional PCT clinical decision points should not be used authors on this publication have met the following required
in the setting of nonbacterial PCT elevations. criteria of eligibility for authorship: (a) significant contributions
• Age-specific reference ranges should be used to interpret to the conception and design, acquisition of data, or analysis
PCT in neonates. and interpretation of data; (b) drafting or revising the article for
intellectual content; (c) final approval of the published article;
and (d) agreement to be accountable for all aspects of the article
CONCLUSIONS thus ensuring that questions related to the accuracy or integrity
While the literature has accumulated with RCTs investigating the of any part of the article are appropriately investigated and
use of PCT, increasingly in the United States with the more recent resolved. Nobody who qualifies for authorship has been omitted
availability of FDA-approved assays, there exists considerable from the list.
variabilities in study designs and study populations that make Allison Chambliss (Conceptualization-Equal, Investigation-
it difficult to provide specific evidence-based recommendations Equal, Methodology-Equal, Writing—original draft-Equal,
for PCT protocols. In general, evidence to support the use of Writing—review and editing-Equal); Khushbu Patel
PCT to guide antibiotic cessation is compelling, particularly (Conceptualization-Equal, Data curation-Equal, Formal analysis-
in the critically ill and in some LRTIs, but is lacking in other Equal, Validation-Equal, Writing —original draft-Equal,
clinical scenarios. Current data on the utility of PCT to guide the Writing—review and editing-Equal); Jessica Colon-Franco
initiation of antibiotics is limited and does not demonstrate a (Formal analysis-Equal, Investigation-Equal, Methodology-Equal,
benefit. In the pediatric and neonatal populations, some studies Writing—original draft-Equal, Writing—review and editing-
have established a role for PCT-guided protocols in reducing Equal); Joshua Hayden (Conceptualization-Equal, Data curation-
antibiotic exposure; however, the utility of PCT has not yet been Equal, Formal analysis-Equal, Investigation-Equal, Methodology-
well-studied in preterm infants. While elevations in PCT are Equal, Writing—original draft-Equal, Writing— review and
generally correlative with poor outcomes, no consistent PCT editing-Equal); Sophie Katz (Conceptualization-Equal, Data
concentration(s) have been established to predict mortality. PCT curation-Equal, Investigation-Equal, Writing—original draft-
interpretation guidance should consider the analytical method Equal, Writing—review and editing-Equal); Emi Minejima
used, ideally its comparison to the BRAHMS Kryptor method, (Conceptualization-Equal, Data curation-Equal, Formal analysis-
and patients should be trended using the same analytical Equal, Investigation-Equal, Writing—original draft-Equal,
assay. Improved outcomes from PCT implementation are more Writing—review and editing-Equal); and Alison Woodworth
likely to be realized when the test is used in conjunction with (Conceptualization-Equal, Data curation-Equal, Formal analysis-
antimicrobial stewardship programs, institutional interpretive Equal, Methodology-Equal, Project administration-Equal,
algorithms, and clinical decision support tools. Successful Visualization-Equal, Writing—original draft-Equal, Writing—
implementation of clinical PCT requires a multidisciplinary review and editing-Equal).
effort among laboratorians, pharmacists, and infectious disease
providers. Authors’ Disclosures or Potential Conflicts of Interest
Upon manuscript submission, all authors completed the author
SUPPLEMENTAL MATERIAL disclosure form. Disclosures and/or potential conflicts of interest:
Supplemental material is available at The Journal of Applied Employment or Leadership: A.B. Chambliss, The Journal of
Laboratory Medicine online. Applied Laboratory Medicine, AACC; K. Patel, Clinical Chemistry,
AACC; A. Woodworth, AACC Board of Directors; J.M. Colon-
Nonstandard Abbreviations Franco, CLSI —Sepsis guidelines. Consultant or Advisory Role:
PCT, procalcitonin; FDA, Food and Drug Administration; LRTI, A.B. Chambliss, Roche Diagnostics; S.E. Katz, Optum; J.M. Colon-
lower respiratory tract infection; RTI, respiratory tract infection; Franco, Cytovale. Stock Ownership: None declared. Honoraria:
ICU, intensive care unit; RCT, randomized controlled trial; CRP, A.B. Chambliss, AACC; S.E. Katz, Washington University St.
C-reactive protein; ED, emergency department; CAP, community- Louis, East Tennessee Children's Hospital; A. Woodworth, AACC,
acquired pneumonia; AECOPD, acute exacerbation of chronic Diasorin, Patient Centered Laboratory Utilization Guidance.
obstructive pulmonary disease; VAP, ventilator-associated Research Funding: S.E. Katz, Pfizer. Expert Testimony: None
pneumonia; OR, odds ratio; COPD, chronic obstructive pulmonary declared. Patents: None declared. Other Remuneration: A.B.
disease; AUC, area under curve; ARI, acute respiratory infection; Chambliss, support for attending meetings and/or travel from
MR-proADM, mid-regional proadrenomedullin; PICU, pediatric AACC; A. Woodworth, support for attending meetings and/or
intensive care unit. travel from Vizient.

Role of Sponsor procalcitonin to shorten antimicrobial therapy in septic
The funding organizations played no role in the design of study, patients with proven bacterial infection in an intensive care
choice of enrolled patients, review and interpretation of data, setting. Diagn Microbiol Infect Dis 2013; 76:266-71.
preparation of manuscript, or final approval of manuscript. 14. Bouadma L, Luyt C-E, Tubach F, Cracco C, Alvarez A, Schwebel
C, et al. Use of procalcitonin to reduce patients' exposure
to antibiotics in intensive care units (PRORATA trial): a
REFERENCES multicentre randomised controlled trial. Lancet Lond Engl
1. Linscheid P, Seboek D, Nylen ES, Langer I, Schlatter M, Becker 2010;375:463-74.
KL, et al. In vitro and in vivo calcitonin I gene expression in
parenchymal cells: a novel product of human adipose tissue. 15. Hochreiter M, Kohler T, Schweiger AM, Keck FS, Bein B, von
Endocrinology 2003;144: 5578-84. Spiegel T, et al. Procalcitonin to guide duration of antibiotic
therapy in intensive care patients: a randomized prospective
2. Becker KL, Snider R, Nylen ES. Procalcitonin in sepsis and controlled trial. Crit Care Lond Engl 2009;13:R83.
systemic inflammation: a harmful biomarker and a therapeutic
target. Br J Pharmacol 2010;159:253-64. 16. Nobre V, Harbarth S, Graf J-D, Rohner P, Pugin J. Use of
procalcitonin to shorten antibiotic treatment duration in
3. Meisner M. Pathobiochemistry and clinical use of septic patients: a randomized trial. Am J Respir Crit Care Med
procalcitonin. Clin Chim Acta 2002;323:17-29. 2008;177:498-505.
4. Meisner M, Lohs T, Huettemann E, Schmidt J, Hueller M, 17. Huang DT, Yealy DM, Filbin MR, Brown AM, Chang C-CH, Doi
Reinhart K. The plasma elimination rate and urinary secretion Y, et al. Procalcitonin-guided use of antibiotics for lower
of procalcitonin in patients with normal and impaired renal respiratory tract infection. N Engl J Med 2018;379: 236-49.
function. Eur J Anaesthesiol 2001;18: 79-87.
18. Daubin C, Valette X, Thiolliere F, Mira J-P, Hazera P, Annane
5. Huang H-B, Peng J-M, Weng L, Wang C-Y, Jiang W, Du B. D, et al. Procalcitonin algorithm to guide initial antibiotic
Procalcitonin-guided antibiotic therapy in intensive care unit therapy in acute exacerbations of COPD admitted to the
patients: a systematic review and meta-analysis. Ann Intensive ICU: a randomized multicenter study. Intensive Care Med
Care 2017;7:114. 2018;44:428-37.
6. Peng F, Chang W, Xie J-F, Sun Q, Qiu H-B, Yang Y. Ineffectiveness 19. Montassier E, Javaudin F, Moustafa F, Nandjou D, Maignan M,
of procalcitonin-guided antibiotic therapy in severely critically Hardouin J-B, et al. Guideline-based clinical assessment versus
ill patients: a meta-analysis. Int J Infect Dis 2019;85:158-66. procalcitonin-guided antibiotic use in pneumonia: a pragmatic
7. Schuetz P, Wirz Y, Sager R, Christ-Crain M, Stolz D, Tamm M, randomized trial. Ann Emerg Med 2019;74:580-91.
et al. Procalcitonin to initiate or discontinue antibiotics in 20. Wussler D, Kozhuharov N, Oliveira MT, Bossa A, Sabti Z, Nowak
acute respiratory tract infections. Cochrane Database Syst Rev A, et al. Clinical utility of procalcitonin in the diagnosis of
2017;10:CD007498. pneumonia. Clin Chem 2019;65:1532-42.
8. Hey J, Thompson-Leduc P, Kirson NY, Zimmer L, Wilkins D, 21. Layios N, Lambermont B, Canivet J-L, Morimont P, Preiser J-C,
Rice B, et al. Procalcitonin guidance in patients with lower Garweg C, et al. Procalcitonin usefulness for the initiation of
respiratory tract infections: a systematic review and meta- antibiotic treatment in intensive care unit patients. Crit Care
analysis. Clin Chem Lab Med 2018;56:1200-9. Med 2012;40:2304-9.
9. Odermatt J, Friedli N, Kutz A, Briel M, Bucher HC, Christ-Crain 22. Jensen JU, Hein L, Lundgren B, Bestle MH, Mohr TT, Andersen
M, et al. Effects of procalcitonin testing on antibiotic use and MH, et al. Procalcitonin-guided interventions against
clinical outcomes in patients with upper respiratory tract infections to increase early appropriate antibiotics and
infections. An individual patient data meta-analysis. Clin Chem improve survival in the intensive care unit: a randomized trial.
Lab Med 2017;56: 170-7. Crit Care Med 2011;39: 2048-58.
10. Bloos F, Trips E, Nierhaus A, Briegel J, Heyland DK, Jaschinski 23. Shehabi Y, Sterba M, Garrett PM, Rachakonda KS, Stephens
U, et al. Effect of sodium selenite administration and D, Harrigan P, et al. Procalcitonin algorithm in critically ill
procalcitonin-guided therapy on mortality in patients with adults with undifferentiated infection or suspected sepsis.
severe sepsis or septic shock: A randomized clinical trial. A randomized controlled trial. Am J Respir Crit Care Med
JAMA Intern Med. 2016;176:1266-76. 2014;190:1102-10.
11. de Jong E, van Oers JA, Beishuizen A, Vos P, Vermeijden WJ, 24. Meier MA, Branche A, Neeser OL, Wirz Y, Haubitz S, Bouadma
Haas LE, et al. Efficacy and safety of procalcitonin guidance L, et al. Procalcitonin-guided antibiotic treatment in patients
in reducing the duration of antibiotic treatment in critically with positive blood cultures: a patient-level meta-analysis of
ill patients: a randomised, controlled, open-label trial. Lancet randomized trials. Clin Infect Dis 2019;69:388-96.
Infect Dis 2016;16: 819-27.
25. Oliveira CF, Botoni FA, Oliveira CRA, Silva CB, Pereira HA,
12. Annane D, Maxime V, Faller JP, Mezher C, Clec'h C, Martel P, et Serufo JC, et al. Procalcitonin versus C-reactive protein for
al. Procalcitonin levels to guide antibiotic therapy in adults guiding antibiotic therapy in sepsis: a randomized trial. Crit
with non-microbiologically proven apparent severe sepsis: a Care Med 2013;41:2336-43.
randomised controlled trial. BMJ Open 2013;3:e002186.
26. Schuetz P, Wirz Y, Sager R, Christ-Crain M, Stolz D, Tamm M,
13. Deliberato RO, Marra AR, Sanches PR, Martino MDV, Ferreira et al. Effect of procalcitonin-guided antibiotic treatment on
CEDS, Pasternak J, et al. Clinical and economic impact of mortality in acute respiratory infections: a patient level meta-

analysis. Lancet Infect Dis 2018;18: 95-107. infections: cluster-randomised, single-blinded intervention
27. Mockel M, de Boer RA, Slagman AC, von Haehling S, Schou M, trial. Lancet Lond Engl 2004; 363:600-7.
Vollert JO, et al. Improve management of acute heart failure 40. Briel M, Schuetz P, Mueller B, Young J, Schild U, Nusbaumer
with ProcAlCiTonin in Europe: results of the randomized C, et al. Procalcitonin-guided antibiotic use vs a standard
clinical trial IMPACT EU biomarkers in cardiology (BIC) 18. approach for acute respiratory tract infections in primary care.
Eur J Heart Fail 2020; 22:267-75. Arch Intern Med 2008;168:2000-7.
28. Ni W, Bao J, Yang D, Xi W, Wang K, Xu Y, et al. Potential of serum 41. Stolz D, Christ-Crain M, Bingisser R, Leuppi J, Miedinger D,
procalcitonin in predicting bacterial exacerbation and guiding Muller C, et al. Antibiotic treatment of exacerbations of COPD:
antibiotic administration in severe COPD exacerbations: a a randomized, controlled trial comparing procalcitonin-
systematic review and meta-analysis. Infect Dis 2019;51:639- guidance with standard therapy. Chest 2007;131:9-19.
50. 42. Corti C, Fally M, Fabricius-Bjerre A, Mortensen K, Jensen
29. Bremmer DN, Moffa MA, Ma K, Bean HR, Snatchko J, Trienski BN, Andreassen HF, et al. Point-of-care procalcitonin test
TL, et al. Acute exacerbations of chronic obstructive to reduce antibiotic exposure in patients hospitalized with
pulmonary disease with a low procalcitonin concentration: acute exacerbation of COPD. Int J Chron Obstruct Pulmon Dis
impact of antibiotic therapy. Clin Infect Dis 2019;68:725-30. 2016;11:1381-9.
30. Branche AR, Walsh EE, Vargas R, Hulbert B, Formica MA, 43. Aabenhus R, Jensen J-US, J0rgensen KJ, Hrobjartsson A,
Baran A, et al. Serum procalcitonin measurement and viral Bjerrum L. Biomarkers as point-of-care tests to guide
testing to guide antibiotic use for respiratory infections in prescription of antibiotics in patients with acute respiratory
hospitalized adults: a randomized controlled trial. J Infect Dis infections in primary care. Cochrane Database Syst Rev
2015;212:1692-700. 2014;2014:CD010130.
31. Long W, Deng X, Zhang Y, Lu G, Xie J, Tang J. Procalcitonin 44. Tonkin-Crine SK, Tan PS, van Hecke O, Wang K, Roberts
guidance for reduction of antibiotic use in low-risk outpatients NW, McCullough A, et al. Clinician-targeted interventions
with community-acquired pneumonia. Respirol Carlton Vic to influence antibiotic prescribing behaviour for acute
2011;16:819-24. respiratory infections in primary care: an overview of
32. Schuetz P, Christ-Crain M, Thomann R, Falconnier C, Wolbers systematic reviews. Cochrane Database Syst Rev 2017;
M, Widmer I, et al. Effect of procalcitonin-based guidelines 2017:CD012252.
vs standard guidelines on antibiotic use in lower respiratory 45. Engel MF, Paling FP, Hoepelman AIM, van der Meer V,
tract infections: the ProHOSP randomized controlled trial. Oosterheert JJ. Evaluating the evidence for the implementation
JAMA 2009;302:1059-66. of C-reactive protein measurement in adult patients with
33. Stolz D, Smyrnios N, Eggimann P, Pargger H, Thakkar N, suspected lower respiratory tract infection in primary care: a
Siegemund M, et al. Procalcitonin for reduced antibiotic systematic review. Fam Pract 2012;29:383-93.
exposure in ventilator-associated pneumonia: a randomised 46. Cals JWL, Butler CC, Hopstaken RM, Hood K, Dinant G-J.
study. Eur Respir J 2009;34:1364-75. Effect of point of care testing for C reactive protein and
34. Christ-Crain M, Stolz D, Bingisser R, Muller C, Miedinger D, training in communication skills on antibiotic use in lower
Huber PR, et al. Procalcitonin guidance of antibiotic therapy in respiratory tract infections: cluster randomised trial. BMJ
community-acquired pneumonia. Am J Respir Crit Care Med 2009;338:b1374.
2006;174:84-93. 47. Butler CC, Gillespie D, White P, Bates J, Lowe R, Thomas-
35. Newton JA, Robinson S, Ling CLL, Zimmer L, Kuper K, Trivedi Jones E, et al. C-reactive protein testing to guide antibiotic
KK. Impact of procalcitonin levels combined with active prescribing for COPD exacerbations. N Engl J Med
intervention on antimicrobial stewardship in a community 2019;381:111-20.
hospital. Open Forum Infect Dis 2019;6:ofz355. 48. Prins HJ, Duijkers R, van der Valk P, Schoorl M, Daniels JMA,
36. Akagi T, Nagata N, Wakamatsu K, Harada T, Miyazaki H, Takeda van der Werf TS, et al. CRP-guided antibiotic treatment in
S, et al. Procalcitonin-guided antibiotic discontinuation acute exacerbations of COPD in hospital admissions. Eur
might shorten the duration of antibiotic treatment without Respir J 2019;53:1802014.
increasing pneumonia recurrence. Am J Med Sci 2019;358:33- 49. Rhee C, Dantes R, Epstein L, Murphy DJ, Seymour CW,
44. Iwashyna TJ, et al. Incidence and trends of sepsis in US
37. Beye F, Vigneron C, Dargent A, Prin S, Andreu P, Large A, et hospitals using clinical vs claims data, 2009-2014. JAMA
al. Adhering to the procalcitonin algorithm allows antibiotic 2017;318:1241.
therapy to be shortened in patients with ventilator-associated 50. Kutz A, Briel M, Christ-Crain M, Stolz D, Bouadma L, Wolff
pneumonia. J Crit Care 2019;53: 125-31. M, et al. Prognostic value of procalcitonin in respiratory
38. Branche A, Neeser O, Mueller B, Schuetz P. Procalcitonin tract infections across clinical settings. Crit Care Lond Engl
to guide antibiotic decision making. Curr Opin Infect Dis 2015;19:74.
2019;32:130-5. 51. Liu D, Su L, Han G, Yan P, Xie L. Prognostic value of
39. Christ-Crain M, Jaccard-Stolz D, Bingisser R, Gencay MM, Huber procalcitonin in adult patients with sepsis: a systematic
PR, Tamm M, et al. Effect of procalcitonin-guided treatment review and meta-analysis. PloS One 2015;10:e0129450.
on antibiotic use and outcome in lower respiratory tract 52. Elke G, Bloos F, Wilson DC, Brunkhorst FM, Briegel J, Reinhart

K, et al. The use of mid-regional proadrenomedullin to sepsis. Eur J Inflamm 2020;18: 2058739220942631.
identify disease severity and treatment response to sepsis—a 65. Zaccone V, Falsetti L, Nitti C, Gentili T, Marchetti A, Piersantelli
secondary analysis of a large randomised controlled trial. Crit MN, et al. The prognostic role of procalcitonin in critically ill
Care 2018; 22:79. patients admitted in a medical stepdown unit: a retrospective
53. Saeed K, Wilson DC, Bloos F, Schuetz P, van der Does cohort study. Sci Rep 2020;10:4531.
Y, Melander O, et al. The early identification of disease 66. Stalenhoef JE, van Nieuwkoop C, Wilson DC, van der Starre
progression in patients with suspected infection presenting WE, van der Reijden TJK, Delfos NM, et al.Procalcitonin,
to the emergency department: a multi-centre derivation and mid-regional proadrenomedullin and C-reactive protein in
validation study. Crit Care 2019;23:40. predicting treatment outcome in community-acquired febrile
54. Osawa T, Watanabe M, Morimoto K, Okumura M, Yoshiyama urinary tract infection. BMC Infect Dis 2019;19:161.
T, Ogata H, et al. Serum procalcitonin levels predict mortality 67. Schuetz P, Hausfater P, Amin D, Amin A, Haubitz S, Faessler L,
risk in patients with pulmonary tuberculosis: a single-center et al. Biomarkers from distinct biological pathways improve
prospective observational study. J Infect Dis 2020;222:1651-4. early risk stratification in medical emergency patients: the
55. Sakran JV, Michetti CP, Sheridan MJ, Richmond R, Waked T, multinational, prospective, observational TRIAGE study. Crit
Aldaghlas T, et al. The utility of procalcitonin in critically ill Care Lond Engl 2015;19:377.
trauma patients. J Trauma Acute Care Surg 2012;73: 413-8. 68. Enguix-Armada A, Escobar-Conesa R, La Torre AG-D, De
56. Sager R, Wirz Y, Amin D, Amin A, Hausfater P, Huber A, et La Torre-Prados MV. Usefulness of several biomarkers
al. Are admission procalcitonin levels universal mortality in the management of septic patients: C-reactive
predictors across different medical emergency patient protein, procalcitonin, presepsin and mid-regional pro-
populations? Results from the multi-national, prospective, adrenomedullin. Clin Chem Lab Med 2016;54:163-8.
observational TRIAGE study. Clin Chem Lab Med 69. Andaluz-Ojeda D, Nguyen HB, Meunier-Beillard N, Cicuendez
2017;55:1873-80. R, Quenot J-P, Calvo D, et al. Superior accuracy of mid-regional
57. Shim B-S, Yoon Y-H, Kim J-Y, Cho Y-D, Park S-J, Lee E-S, et al. proadrenomedullin for mortality prediction in sepsis with
Clinical value of whole blood procalcitonin using point of varying levels of illness severity. Ann Intensive Care 2017;7:15.
care testing, quick sequential organ failure assessment score, 70. Charles P-E, Peju E, Dantec A, Bruyere R, Meunier-Beillard N,
C-reactive protein and lactate in emergency department Dargent A, et al. MR-proADM elevation upon ICU admission
patients with suspected infection. J Clin Med 2019;8:833. predicts the outcome of septic patients and is correlated with
58. Yu H, Nie L, Liu A, Wu K, Hsein Y-C, Yen DW, et al. Combining upcoming fluid overload. Shock Augusta Ga 2017;48:418-26.
procalcitonin with the qSOFA and sepsis mortality prediction. 71. Onal U, Valenzuela-Sanchez F, Vandana KE, Rello J. Mid-
Medicine (Baltimore) 2019;98:e15981. regional pro-adrenomedullin (MR-proADM) as a biomarker
59. Zhou H, Guo S, Lan T, Ma S, Zhang F, Zhao Z. Risk stratification for sepsis and septic shock: narrative review. Healthcare
and prediction value of procalcitonin and clinical severity (Basel) 2018;6:110.
scores for community-acquired pneumonia in ED. Am J Emerg 72. Rhodes A, Evans LE, Alhazzani W, Levy MM, Antonelli M,
Med 2018;36:2155-60. Ferrer R, et al. Surviving sepsis campaign: international
60. Castillo JGD, Wilson DC, Clemente-Callejo C, Roman F, Bardes- guidelines for management of sepsis and septic shock. Crit
Robles I, Jimenez I, et al. Biomarkers and clinical scores to Care Med 2017;45:486-552.
identify patient populations at risk of delayed antibiotic 73. Nguyen HB, Rivers EP, Knoblich BP, Jacobsen G, Muzzin A,
administration or intensive care admission. Crit Care Lond Ressler JA, et al. Early lactate clearance is associated with
Engl 2019;23:335. improved outcome in severe sepsis and septic shock. Crit Care
61. Shi Y, Peng J, Hu X, Wang Y. The utility of initial procalcitonin Med 2004;32:1637-42.
and procalcitonin clearance for prediction of bacterial 74. Choudhary R, Sitaraman S, Choudhary A. Lactate clearance as
infection and outcome in critically ill patients with the predictor of outcome in pediatric septic shock. J Emerg
autoimmune diseases: a prospective observational study. BMC Trauma Shock 2017;10:55.
Anesthesiol 2015;15:137.
75. Jansen TC, van Bommel J, Schoonderbeek FJ, Sleeswijk Visser
62. Ryoo SM, Han KS, Ahn S, Shin TG, Hwang SY, Chung SP, et SJ, van der Klooster JM, Lima AP, et al. Early lactate-guided
al. The usefulness of C-reactive protein and procalcitonin therapy in intensive care unit patients: a multicenter, open-
to predict prognosis in septic shock patients: a multicenter label, randomized controlled trial. Am J Respir Crit Care Med
prospective registry-based observational study. Sci Rep 2010;182:752-61.
2019;9:6579.
76. Jones AE. Lactate clearance vs central venous oxygen
63. Schuetz P, Birkhahn R, Sherwin R, Jones AE, Singer A, Kline saturation as goals of early sepsis therapy: a randomized
JA, et al. Serial procalcitonin predicts mortality in severe clinical trial. JAMA 2010;303:739.
sepsis patients: results from the multicenter procalcitonin
77. Tian H, Han S, Lv C, Wang T, Li Z, Hao D, et al. [The effect of
MOnitoring SEpsis (MOSES) study. Crit Care Med
early goal lactate clearance rate on the outcome of septic
2017;45:781-9.
shock patients with severe pneumonia]. Zhongguo Wei Zhong
64. Peng J-C, Xu Q-Y, Ding J, Li Z, Wang J-M, Gao Y, et al. Usefulness Bing Ji Jiu Yi Xue 2012;24:42-5.
of procalcitonin clearance to predict mortality in abdominal
78. Gu W-J, Zhang Z, Bakker J. Early lactate clearance-guided

therapy in patients with sepsis: a meta-analysis with trial intensive care unit. J Pediatr 2016;179: 74-81.e2.
sequential analysis of randomized controlled trials. Intensive 92. Kuppermann N, Dayan PS, Levine DA, Vitale M, Tzimenatos
Care Med 2015;41:1862-3. L, Tunik MG, et al. A clinical prediction rule to identify febrile
79. ThermoFisher Scientific. MR-proADM. https://www. infants 60 days and younger at low risk for serious bacterial
mrproadm.com/about-mrproadm/brahms-mr-proadm. infections. JAMA Pediatr 2019;173:342.
html(Accessed August 2021). 93. Velasco R, Gomez B, Benito J, Mintegi S. Accuracy of PECARN
80. Mearelli F, Barbati G, Casarsa C, Giansante C, Breglia A, Spica rule for predicting serious bacterial infection in infants with
A, et al. The integration of qSOFA with clinical variables and fever without a source. Arch Dis Child 2021; 106:143-8.
serum biomarkers improves the prognostic value of qSOFA 94. Stockmann C, Ampofo K, Killpack J, Williams DJ, Edwards
alone in patients with suspected or confirmed sepsis at ED KM, Grijalva CG, et al. Procalcitonin accurately identifies
admission. J Clin Med 2020;9:1205. hospitalized children with low risk of bacterial community-
81. Zhu Y, Li X, Guo P, Chen Y, Li J, Tao T. The accuracy assessment acquired pneumonia. J Pediatr Infect Dis Soc 2018;7:46-53.
of presepsin (sCD14-ST) for mortality prediction in adult 95. Esposito S, Tagliabue C, Picciolli I, Semino M, Sabatini C,
patients with sepsis and a head-to-head comparison to PCT: a Consolo S, et al. Procalcitonin measurements for guiding
meta-analysis. Ther Clin Risk Manag 2019;15:741-53. antibiotic treatment in pediatric pneumonia. Respir Med
82. Arora S, Singh P, Singh PM, Trikha A. Procalcitonin levels in 2011;105:1939-45.
survivors and nonsurvivors of sepsis: systematic review and 96. Baer G, Baumann P, Buettcher M, Heininger U, Berthet G,
meta-analysis. Shock Augusta Ga 2015;43: 212-21. Schafer J, et al. Procalcitonin guidance to reduce antibiotic
83. Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, treatment of lower respiratory tract infection in children and
Annane D, Bauer M, et al. The third international consensus adolescents (ProPAED): a randomized controlled trial. PloS
definitions for sepsis and septic shock (sepsis-3). JAMA One 2013;8: e68419.
2016;315:801. 97. Stocker M, van Herk W, El Helou S, Dutta S, Fontana MS,
84. Asirvatham AR, Menon U, Pavithran PV, Vasukutty JR, Kumar Schuerman FABA, et al. Procalcitonin-guided decision making
H, Bhavani N, et al. Role of procalcitonin as a predictor of for duration of antibiotic therapy in neonates with suspected
clinical outcome in acute diabetic foot infections: a prospective early-onset sepsis: a multicentre, randomised controlled trial
study. Indian J Endocrinol Metab 2019;23:122-7. (NeoPIns). Lancet Lond Engl 2017;390:871-81.
85. Yu H, Qi Z, Hang C, Fang Y, Shao R, Li C. Evaluating the value 98. Stocker M, Fontana M, El Helou S, Wegscheider K, Berger
of dynamic procalcitonin and presepsin measurements for TM. Use of procalcitonin-guided decision-making to shorten
patients with severe sepsis. Am J Emerg Med 2017;35:835-41. antibiotic therapy in suspected neonatal early-onset sepsis:
86. Trippella G, Galli L, De Martino M, Lisi C, Chiappini E. prospective randomized intervention trial. Neonatology
Procalcitonin performance in detecting serious and invasive 2010;97:165-74.
bacterial infections in children with fever without apparent 99. Bobillo-Perez S, Sole-Ribalta A, Balaguer M, Esteban E, Girona-
source: a systematic review and meta-analysis. Expert Rev Alarcon M, Hernandez-Platero L, et al. Procalcitonin to stop
Anti Infect Ther 2017;15:1041-57. antibiotics after cardiovascular surgery in a pediatric intensive
87. Ruan L, Chen G-Y, Liu Z, Zhao Y, Xu G-Y, Li S-F, et al. The care unit—the PROSACAB study. PLoS One 2019;14:e0220686.
combination of procalcitonin and C-reactive protein or 100. Hatherill M, Tibby SM, Turner C, Ratnavel N, Murdoch IA.
presepsin alone improves the accuracy of diagnosis of Procalcitonin and cytokine levels: relationship to organ
neonatal sepsis: a meta-analysis and systematic review. Crit failure and mortality in pediatric septic shock. Crit Care Med
Care 2018;22:316. 2000;28:2591-4.
88. Casado-Flores J, Blanco-Quiros A, Asensio J, Arranz E, Garrote 101. Rey C, Los Arcos M, Concha A, Medina A, Prieto S, Martinez
JA, Nieto M. Serum procalcitonin in children with suspected P, et al. Procalcitonin and C-reactive protein as markers
sepsis: a comparison with C-reactive protein and neutrophil of systemic inflammatory response syndrome severity in
count. Pediatr Crit Care Med 2003;4: 190-5. critically ill children. Intensive Care Med 2007; 33:477-84.
89. Simon L, Saint-Louis P, Amre DK, Lacroix J, Gauvin F. 102. Han YY, Doughty LA, Kofos D, Sasser H, Carcillo JA.
Procalcitonin and C-reactive protein as markers of bacterial Procalcitonin is persistently increased among children with
infection in critically ill children at onset of systemic poor outcome from bacterial sepsis. Pediatr Crit Care Med
inflammatory response syndrome. Pediatr Crit Care Med 2003;4:21-5.
2008;9:407-13. 103. Milcent K, Faesch S, Gras-Le Guen C, Dubos F, Poulalhon C,
90. Downes KJ, Weiss SL, Gerber JS, Klieger SB, Fitzgerald JC, Badier I, et al. Use of procalcitonin assays to predict serious
Balamuth F, et al. A pragmatic biomarker-driven algorithm bacterial infection in young febrile infants. JAMA Pediatr
to guide antibiotic use in the pediatric intensive care unit: 2016;170:62.
the optimizing antibiotic strategies in sepsis (OASIS) study. J 104. Casado-Flores J, Blanco-Quiros A, Nieto M, Asensio J,
Pediatr Infect Dis Soc 2017;6: 134-41. Fernandez C. Prognostic utility of the semi-quantitative
91. Lautz AJ, Dziorny AC, Denson AR, O'Connor KA, Chilutti procalcitonin test, neutrophil count and C-reactive protein
MR, Ross RK, et al. Value of procalcitonin measurement for in meningococcal infection in children. Eur J Pediatr
early evidence of severe bacterial infections in the pediatric 2006;165:26-9.

105. Celebi S, Koner O, Menda F, Balci H, Hatemi A, Korkut K, et 2021).
al. Procalcitonin kinetics in pediatric patients with systemic 119. US Food & Drug Administration. 510(k) Premarket
inflammatory response after open heart surgery. Intensive Notification. https://www.accessdata.fda.gov/scripts/ cdrh/
Care Med 2006;32:881-7. cfdocs/cfPMN/pmn.cfm?ID=55055 (Accessed February 2021).
106. Van der Kaay D, De Kleijn E, De Rijke Y, Hop W, De Groot 120. US Food & Drug Administration. 510(k) Premarket
R, Hazelzet J. Procalcitonin as a prognostic marker in Notification. https://www.accessdata.fda.gov/scripts/ cdrh/
meningococcal disease. Intensive Care Med 2002;28: 1606-12. cfdocs/cfpmn/pmn.cfm?id=K162297 (Accessed February
107. Dalton HJ, Carcillo JA, Woodward DB, Short MA, Williams MD. 2021).
Biomarker response to drotrecogin alfa (activated) in children 121. US Food & Drug Administration. 510(k) Premarket
with severe sepsis: results from the RESOLVE clinical trial. Notification. https://www.accessdata.fda.gov/scripts/ cdrh/
Pediatr Crit Care Med 2012;13:639-45. cfdocs/cfpmn/pmn.cfm?ID=K181002 (Accessed February
108. Poddar B, Gurjar M, Singh S, Aggarwal A, Baronia A. Reduction 2021).
in procalcitonin level and outcome in critically ill children 122. US Food & Drug Administration. 510(k) Premarket
with severe sepsis/septic shock: a pilot study. J Crit Care Notification. https://www.accessdata.fda.gov/scripts/ cdrh/
2016;36:230-3. cfdocs/cfpmn/pmn.cfm?ID=K200236 (Accessed February
109. Schuetz P, Beishuizen A, Broyles M, Ferrer R, Gavazzi G, Gluck 2021).
EH, et al. Procalcitonin (PCT)-guided antibiotic stewardship: 123. Meisner M, Tschaikowsky K, Schnabel S, Schmidt J, Katalinic A,
an international experts consensus on optimized clinical use. Schuttler J. Procalcitonin—influence of temperature, storage,
Clin Chem Lab Med 2019;57: 1308-18. anticoagulation and arterial or venous asservation of blood
110. Pantell RH, Roberts KB, Adams WG, Dreyer BP, Kuppermann samples on procalcitonin concentrations. Eur J Clin Chem Clin
N, O'Leary ST, et al. Evaluation and management of well- Biochem 1997;35: 597-601.
appearing febrile infants 8 to 60 days old. Pediatrics 124. Kutz A, Hausfater P, Oppert M, Alan M, Grolimund E, Gast
2021;148:e2021052228. C, et al. Comparison between BRAHMS PCT direct, a new
111. Cabral L, Afreixo V, Meireles R, Vaz M, Chaves C, Caetano M, sensitive point-of-care testing device for rapid quantification
et al. Checking procalcitonin suitability for prognosis and of procalcitonin in emergency department patients and
antimicrobial therapy monitoring in burn patients. Burns established reference methods—a prospective multinational
Trauma 2018;6:10. trial. Clin Chem Lab Med 2016;54:577-84.
112. Timbrook TT, Morton JB, McConeghy KW, Caffrey AR, 125. Gruzdys V, Cahoon K, Pearson L, Lehman CM. Method
Mylonakis E, LaPlante KL. The effect of molecular rapid verification shows a negative bias between 2 procalcitonin
diagnostic testing on clinical outcomes in bloodstream methods at medical decision concentrations. J Appl Lab Med
infections: a systematic review and meta-analysis. Clin Infect 2019;4:69-77.
Dis 2017;64:15-23. 126. Schuetz P, Christ-Crain M, Huber AR, Muller B. Long-term
113. American Nurses Association and Centers for Disease Control stability of procalcitonin in frozen samples and comparison
and Prevention. Redefining the antibiotic stewardship team: of Kryptor® and VIDAS® automated immunoassays. Clin
recommendations from the American Nurses Association/ Biochem 2010;43:341-4.
Centers for Disease Control and Prevention Workgroup on the 127. Huynh H-H, Breuf A, Vinh J, Delatour V, Delatour V, Breuf A,
Role of Registered Nurses in Hospital Antibiotic Stewardship et al. Evaluation of the necessity and the feasibility of the
Practices. JAC-Antimicrob Resist. 2019;1:dlz037. standardization of procalcitonin measurements: activities of
114. Centers for Disease Control and Prevention. Core Elements of IFCC WG-PCT with involvement of all stakeholders. Clin Chim
Hospital Antibiotic Stewardship Programs. https://www.cdc. Acta 2021;515:111-21.
gov/antibiotic-use/core-elements/ hospital.html (Accessed 128. Schuetz P, Bretscher C, Bernasconi L, Mueller B. Overview of
March 2022). procalcitonin assays and procalcitonin-guided protocols for
115. Broyles MR. Impact of procalcitonin-guided antibiotic the management of patients with infections and sepsis. Expert
management on antibiotic exposure and outcomes: real-world Rev Mol Diagn 2017;17:593-601.
evidence. Open Forum Infect Dis 2017;4:ofx213. 129. Dipalo M, Buonocore R, Gnocchi C, Picanza A, Aloe R, Lippi G.
116. Lindsay PJ, Rohailla S, Taggart LR, Lightfoot D, Havey T, Analytical evaluation of diazyme procalcitonin (PCT) latex-
Daneman N, et al. Antimicrobial stewardship and intensive enhanced immunoturbidimetric assay on Beckman Coulter
care unit mortality: a systematic review. Clin Infect Dis AU5800. Clin Chem Lab Med 2015;53:593-7.
2019;68:748-56. 130. Lippi G, Salvagno GL, Gelati M, Pucci M, Lo Cascio C, Demonte
117. Moradi T, Bennett N, Shemanski S, Kennedy K, Schlachter A, D, et al. Two-center comparison of 10 fully-automated
Boyd S. Use of procalcitonin and a respiratory polymerase commercial procalcitonin (PCT) immunoassays. Clin Chem Lab
chain reaction panel to reduce antibiotic use via an EMR alert. Med 2019;58:77-84.
Clin Infect Dis 2019;71: 1684-9. 131. Soh A, Binder L, Clough M, Hernandez MH, Lefevre G, Mostert
118. US Food & Drug Administration. 510(k) Premarket K, et al. Comparison of the novel ARCHITECT procalcitonin
Notification. https://www.accessdata.fda.gov/scripts/ cdrh/ assay with established procalcitonin assay systems. Pr Lab
cfdocs/cfpmn/pmn.cfm?ID=K070310 (Accessed February Med 2018;12:e00110.

132. Chambliss AB, Hayden J, Colby JM. Evaluation of procalcitonin
immunoassay concordance near clinical decision points. Clin
Chem Lab Med 2019;27:1414-21.
133. Ceriotti F, Marino I, Motta A, Carobene A. Analytical evaluation
of the performances of diazyme and BRAHMS procalcitonin
applied to Roche cobas in comparison with BRAHMS PCT-
sensitive Kryptor. Clin Chem Lab Med 2017;56:162-9.
134. Dipalo M, Gnocchi C, Avanzini P, Musa R, Di Pietro M, Aloe
R. Comparison of procalcitonin assays on KRYPTOR and
LIAISON® XL analyzers. Diagnostics 2019;9:94.
135. Dupuy AM, Ne M, Bargnoux AS, Badiou S, Cristol JP. Analytical
evaluation of lumipulse® BRAHMS PCT CLEIA assay
and clinical performances in an unselected population
as compared with central lab PCT assay. Clin Biochem
2017;50:248-50.
136. Dipalo M, Guido L, Micca G, Pittalis S, Locatelli M, Motta A,
et al. Multicenter comparison of automated procalcitonin
immunoassays. Pr Lab Med 2015;17: 22-8.
137. IFCC. Standardization of Procalcitonin assays (WG-PCT).
https://www.ifcc.org/ifcc-scientific-division/sd-working-
groups/wg-pct (Accessed February 2021).
138. Cabral L, Afreixo V, Almeida L, Paiva JA. The use of
procalcitonin (PCT) for diagnosis of sepsis in burn patients: a
meta-analysis. PloS One 2016;11:e0168475.
139. Al-Nawas B, Shah P. Procalcitonin in acute malaria. Eur J Med
Res 1997;2:206-8.
140. Dornbusch HJ, Strenger V, Kerbl R, Lackner H, Schwinger W,
Sovinz P, et al. Procalcitonin—a marker of invasive fungal
infection? Support Care 2005;13:343-6.
141. Lippi G, Plebani M. Procalcitonin in patients with severe
coronavirus disease 2019 (COVID-19): a meta-analysis. Clin
Chim Acta 2020;505:190-1.
142. Meisner M, Tschaikowsky K, Hutzler A, Schick C, Schuttler J.
Postoperative plasma concentrations of procalcitonin after
different types of surgery. Intensive Care Med 1998;24:680-4.
143. Chiesa C, Pellegrini G, Panero A, Osborn JF, Signore F, Assumma
M, et al. C-reactive protein, interleukin-6, and procalcitonin in
the immediate postnatal period: influence of illness severity,
risk status, antenatal and perinatal complications, and
infection. Clin Chem 2003; 49:60-8.


The Clinical Use of Procalcitonin: AACC Guidance Document On

Uploaded by

Copyright:

Available Formats

The Clinical Use of Procalcitonin: AACC Guidance Document On

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Clinical Use of Procalcitonin: AACC Guidance Document On

Uploaded by

Copyright:

Available Formats

AACC Guidance Document on

the Clinical Use of Procalcitonin

AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

TABLE 2. Features of POCT and laboratory testing hCG.a

Beckman Coulter (AU) X

Beckman Coulter (Unicel, Access) X (Access PCT)

AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

AACC GUIDANCE DOCUMENT ON THE CLINICAL USE OF PROCALCITONIN

You might also like