RESEARCH ARTICLE

Pneumonia Severity in Children: Utility of

Procalcitonin in Risk Stratification
Laura F. Sartori, MD, MPH,a,b Yuwei Zhu, MD, MS,a Carlos G. Grijalva, MD, MPH,a Krow Ampofo, MBCHB,c Per Gesteland, MD, MSc,c Jakobi Johnson, BS,a
Rendie McHenry, MS,a Donald H. Arnold, MD, MPH,a Andrew T. Pavia, MD,c Kathryn M. Edwards, MD,a Derek J. Williams, MD, MPHa

ABSTRACT OBJECTIVES: To determine if serum procalcitonin, an indicator of bacterial etiology in pneumonia

in all ages and a predictor of severe pneumonia in adults, is associated with disease severity in
children with community-acquired pneumonia.
METHODS: We prospectively enrolled children 2 months to ,18 years with clinical and
radiographic pneumonia at 2 children’s hospitals (2014–2019). Procalcitonin samples were obtained
at presentation. An ordinal outcome scale of pneumonia severity was defined: very severe
(intubation, shock, or death), severe (intensive care admission without very severe features and/or
high-flow nasal cannula), moderate (hospitalization without severe or very severe features), and mild
(discharge). Hospital length of stay (LOS) was also examined. Ordinal logistic regression was used to
model associations between procalcitonin and outcomes. We estimated adjusted odds ratios (aORs)
for a variety of cut points of procalcitonin ranging from 0.25 to 3.5 ng/mL.
RESULTS: The study included 488 children with pneumonia; 30 (6%) were classified as very severe,
106 (22%) as severe, 327 (67%) as moderate, and 25 (5%) as mild. Median procalcitonin in the very
severe group was 5.06 (interquartile range [IQR] 0.90–16.83), 0.38 (IQR 0.11–2.11) in the severe
group, 0.29 (IQR 0.09–1.90) in the moderate group, and 0.21 (IQR 0.12–1.2) in the mild group.
Increasing procalcitonin was associated with increasing severity (range of aORs: 1.03–1.25) and
increased LOS (range of aORs: 1.04–1.36). All comparisons were statistically significant.
CONCLUSIONS: Higher procalcitonin was associated with increased severity and LOS.
Procalcitonin may be useful in helping clinicians evaluate pneumonia severity.

www.hospitalpediatrics.org
DOI:https://doi.org/10.1542/hpeds.2020-001842
Copyright © 2021 by the American Academy of Pediatrics
Address correspondence to Laura F. Sartori, MD, MPH, Division of Pediatric Emergency Medicine, Children’s Hospital of Philadelphia,
3401 Civic Center Blvd, Philadelphia, PA 19104. E-mail: sartoril@email.chop.edu
HOSPITAL PEDIATRICS (ISSN Numbers: Print, 2154-1663; Online, 2154-1671).
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: Supported in part by funding from the National Institutes of Health under grant awards R01AI125642 and T32HD060554. The
National Institutes of Health did not participate in the design and concept of the study. Procalcitonin assays were provided by
a
Vanderbilt University bioMérieux. The authors have no other financial relationships relevant to this article to disclose. Funded by the National Institutes of
Medical Center, Nashville, Health (NIH).
Tennessee; bChildren’s
Hospital of Philadelphia, POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.
Philadelphia,
Dr Sartori conceptualized and designed the study, drafted the initial manuscript, and reviewed and revised the manuscript; Dr Zhu
Pennsylvania; and
c
Division of Infectious
designed data collection tools, performed statistical calculations and modeling, and reviewed and revised the manuscript; Drs Grijalva,
Diseases and Emergency Ampofo, Gesteland, Arnold, Pavia, Edwards, and Williams conceptualized and designed the study and reviewed and revised the
Medicine, Department of manuscript; Mr Johnson designed, coordinated, and supervised data collection and critically reviewed and revised the manuscript; Ms
Pediatrics University of McHenry designed and coordinated laboratory collection, collected data, conducted the initial analyses, and reviewed and revised the
Utah, Salt Lake City, Utah manuscript; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

HOSPITAL PEDIATRICS Volume 11, Issue 3, March 2021 215

Downloaded from http://publications.aap.org/hospitalpediatrics/article-pdf/11/3/215/814552/hpeds_2020-001842.pdf

by guest
 Pneumonia is a major cause of morbidity in STUDY DESIGN ED triage) and frozen at 280°C. All other
children in the United States, with 1% to 4% Overview testing was at the discretion of the provider
of all pediatric emergency department (ED) Data were obtained from 2 prospective and not uniform among subjects. Additional
visits and .100 000 hospitalizations cohorts of children with pneumonia data collected from the electronic health
attributed to pneumonia each year.1–5 presenting for emergency care. The first record included physical examination,
Despite the high prevalence of pediatric was a pilot study performed at a single radiology and laboratory findings, vital
pneumonia, hospitalization rates vary tertiary children’s hospital in the signs, ED disposition, LOS, need for intensive
dramatically, with adjusted analyses in southeastern United States from December care, and development of respiratory failure
previous studies revealing hospitalization 2014 to January 2017. These data were and/or shock.
rates between 19% and 69%.6 Studies in combined with data from the same
adults with pneumonia have similarly Exposure
southeastern children’s hospital and a
revealed wide variation in hospitalization Procalcitonin was the primary exposure of
second tertiary children’s hospital in the
rates (38%–79%), with providers often interest, measured in nanograms per
western United States from September
overestimating the risk of adverse milliliter. Frozen samples were batched and
2017 to May 2019. This study represents a
outcomes.7,8 This suggests that risk analyzed by using the bioMérieux VIDAS
subset of both of these cohorts because
perceptions are often imprecise and may 3 platform, according to standard protocols
subjects were still eligible for each cohort if
lead to unnecessary or prolonged by an institutional research laboratory.16
they declined to provide blood samples.
hospitalizations and intensive therapies. Procalcitonin values below the detection
Study protocols were approved by the
National pediatric pneumonia management limit of 0.05 ng/mL were set at 0.025 ng/mL.
institutional review board of each
guidelines highlight this decision-making Procalcitonin assays were not conducted in
institution.
process as a key knowledge gap, real time and were not available to
emphasizing the need for objective Study Population clinicians. Seven children had procalcitonin
risk stratification tools, including the obtained during clinical care. Because
Eligibility criteria were identical for both
evaluation of biomarkers, to accurately clinical and study procalcitonin values were
cohorts. Children 2 months to ,18 years
assess disease severity and the risk for not statistically different, only the study
presenting to the ED with signs and
adverse outcomes.9 procalcitonin values were used during
symptoms of acute lower respiratory
analysis.
Procalcitonin, a precursor to calcitonin, is a infection, radiographic evidence of
leading potential biomarker for risk pneumonia, and a provider diagnosis of Outcome
stratification in pneumonia. Although pneumonia were considered for inclusion. The primary outcome of interest was an
studies in adults with pneumonia indicate Radiographic evidence of pneumonia was ordinal pneumonia severity outcome
that procalcitonin provides useful determined by the treating clinical provider measure, defined as very severe (intubation,
prognostic information independently and at the time of enrollment, in concert with shock requiring vasoactive medications, or
in conjunction with existing prognostic the clinician’s interpretation of available death), severe (intensive care admission
tools, similar data in children are clinical radiology reports. A secondary without very severe features and/or high-
limited.10–12 In previous studies in children, review of all radiology reports was flow nasal cannula), moderate
researchers have yielded conflicting results, performed by the 2 site primary (hospitalization without severe or very
with 2 studies revealing associations investigators to further classify pneumonia severe features), and mild (discharge from
between procalcitonin and hospitalization, by infiltrate pattern. Children ,2 months of the hospital).17 Use of continuous or bilevel
need for intensive care, and longer length of age were excluded, as were those who were positive pressure outside of normal home
stay (LOS).13,14 In a recent study, researchers immunocompromised, had cystic fibrosis, settings required intensive care at both
failed to demonstrate an association or had a tracheostomy. In an effort to institutions. The outcome was assigned
between procalcitonin and less severe exclude hospital-acquired pneumonias, according to the most severe outcome
outcomes but did note higher median children transferred from other hospitals experienced at any point during the
procalcitonin concentration in children with and those recently hospitalized or encounter. LOS in hours was a secondary
septic shock and effusion requiring previously enrolled were also excluded outcome and was measured in hours from
drainage, with significance of these (Supplemental Table 3). the time of ED triage to the time of
individual outcomes limited by a small discharge from either the ED or inpatient
sample size.15 We sought to explore those Study Procedures
hospital unit.
findings in a multicenter study of children Enrolled children and caregivers were
with pneumonia presenting for emergency interviewed to collect baseline Covariates
care across the spectrum of illness. We sociodemographic, clinical, and historical Covariates chosen for analysis in the model
hypothesized that increasing procalcitonin data. Blood samples for procalcitonin were selected on the basis of a previous
levels would be associated with increasing measurement were collected at the time of study evaluating the use of 20 candidate
pneumonia severity. enrollment (up to 72 hours from the time of predictors, including demographics,

216 SARTORI et al

Downloaded from http://publications.aap.org/hospitalpediatrics/article-pdf/11/3/215/814552/hpeds_2020-001842.pdf

by guest
 comorbidities, vital signs and other clinical very severe pneumonia to explore how procalcitonin in the very severe group was
factors, radiographic characteristics, and procalcitonin performs in those with less 5.06 ng/mL (IQR 0.90–16.83), 0.38 ng/mL (IQR
laboratory results in the prediction of severe disease. We also explored 0.11–2.11) in the severe group, 0.29 ng/mL
severe pediatric pneumonia.17 We selected procalcitonin results in those with (IQR 0.09–1.90) in the moderate group, and
covariates from the reduced 10-predictor bacteremia and those with parapneumonic 0.21 ng/mL (IQR 0.12–1.20) in the mild group
model for inclusion in our analyses, effusion requiring a drainage (Fig 1). In unadjusted comparison,
excluding altered mental status, which was procedure. Lastly, we examined the timing procalcitonin was weakly correlated with
rarely observed. The covariates included of antibiotic administration and LOS (r 5 0.21; P , .001).
were the ratio of PaO2 to fraction of procalcitonin collection. Associations Between Procalcitonin
inspired oxygen (PF ratio) (estimated and Clinical Outcomes
from the pulse oxygen saturation to RESULTS
For the ordinal severity outcome, increasing
fraction of inspired oxygen ratio), Study Population
procalcitonin was associated with
age, triage heart rate (HR), triage A total of 1116 children were enrolled in the increasing severity with aORs comparing
respiratory rate (RR), triage systolic blood combined cohorts, with 488 children (44%) multiple procalcitonin cut points ranging
pressure (SBP), comorbidities, chest with procalcitonin samples of sufficient from 1.03 (95% confidence interval [CI]
indrawing, infiltrate pattern on chest quantity for analysis: 167 from the pilot 1.01–1.06) for 0.25 ng/mL (reference) versus
radiograph, and presence of pleural cohort and 321 from the main study 0.5 ng/mL to 1.25 (95% CI 1.04–1.51) for
effusion.17,18 (Supplemental Fig 4). Compared with 0.10 vs 2.0 ng/mL; all comparisons were
children without blood samples, those with statistically significant (Table 2, Fig 2).
Statistical Analysis procalcitonin samples were older (64.5 vs Increasing procalcitonin was also
Baseline characteristics were evaluated by 33.5 months; P , .001), more likely to be associated with longer LOS, with aOR
severity classification by using Wilcoxon admitted to the hospital (95% vs 78%; P , ranging from 1.04 (95% CI 1.02–1.07) for
rank and Kruskal-Wallis tests and x2 tests .001), more likely to require ICU care (26% 0.25 vs 0.50 ng/mL to 1.36 (95% CI 1.17–1.59)
for continuous and categorical variables, vs 16%; P , .001), and had a longer average for 0.1 vs 2.0 ng/mL; all comparisons were
respectively. Associations between LOS (median 50.5 vs 40.5 hours; P , .001). statistically significant (Table 2, Fig 3). When
procalcitonin and severity outcomes were Among children included in the study, the limiting our analysis to procalcitonin
evaluated by using ordinal logistic distribution of outcomes was as follows: samples obtained within 24 hours of ED
regression, adjusting for the covariates 30 children (6%) were classified as very triage, results for both the ordinal severity
listed above. Restricted cubic spline severe, 106 as severe (22%), 327 (67%) as outcome and LOS outcome were similar and
functions were applied on age, PF ratio, and moderate, and 25 (5%) as mild. Median LOS remained significant (results not shown). In
procalcitonin with 3 knots to relax linearity was 50.5 hours (interquartile range [IQR] the secondary analysis excluding the very
assumptions. To account for known age- 32.0–96.0). Among the 30 subjects in the severe group, aORs remained .1.0 for all
based differences in normal HR, RR, and very severe group, 28 (6% of total cohort) comparisons, although point estimates
blood pressure, interaction terms between required intubation, 13 (3%) developed were reduced in magnitude and
each of these variables and age were shock, and 2 (0.05%) died. Baseline comparisons were no longer statistically
included. The proportional odds assumption demographic characteristics were largely significant for the ordinal severity
was confirmed by visual inspection of similar among the 4 severity groups, outcome, whereas associations remained
partial residual plots. Because a cubic although those in the very severe group significant for the LOS outcome
spline outcome was used for the primary were younger than those in other groups. (Supplemental Table 4).
exposure, we estimated adjusted odds Baseline clinical characteristics varied
ratios (aORs) for a variety of cut points of Associations Between Procalcitonin
between groups, with the most severely ill
procalcitonin. We also generated predicted and Suspected Bacterial Etiology
children demonstrating higher HRs and RRs,
probability plots to demonstrate how lower PF ratios, increased incidence of Bacteremia was identified in only 5 (1.7%) of
varying procalcitonin changes the estimated chest indrawing and pleural effusion, and 290 patients for whom a blood culture
probabilities for severe or very severe more comorbidities (Table 1). The median result was obtained, all with very high
outcomes and LOS .3 days, holding all LOS for all children was 50.5 hours (IQR procalcitonin levels (142.3, 21.25, 18.18,
other covariates constant. Several 32.0–96.0). In total, 164 children (34%) had a 13.32, and 11.13 ng/mL). Bacterial
additional analyses were also conducted. LOS .3 days. pathogens detected were Streptococcus
Because a small number of children had pneumoniae (3), Staphylococcus aureus (1),
procalcitonin collected .24 hours after Procalcitonin and S pyogenes (1). Seventy-four children
triage, we repeated our primary analyses in Overall, procalcitonin concentrations among had parapneumonic effusion noted on
those whose procalcitonin was collected subjects were low, with a procalcitonin of imaging, with 17 (23%) requiring drainage.
within 24 hours of triage. Next, we repeated ,0.25 ng/mL in 44% of subjects and Median procalcitonin of those requiring
our main analysis after excluding those with ,1.0 ng/mL in 65% of subjects. Median drainage was 10.45 ng/mL (IQR 3.05–20.01).

HOSPITAL PEDIATRICS Volume 11, Issue 3, March 2021 217

Downloaded from http://publications.aap.org/hospitalpediatrics/article-pdf/11/3/215/814552/hpeds_2020-001842.pdf

by guest
by guest
TABLE 1 Characteristics of Study Population by Ordinal Outcome

218
Characteristic Mild (n 5 25) Moderate (n 5 327) Severe (n 5 106) Very Severe (n 5 30) P
Median age (IQR), mo 65.0 (40.0–114) 67.0 (28.5–118.5) 64.0 (28.2–132.2) 25.0 (14.2–134.0) .28
Male sex, % (n) 48 (12) 56 (184) 53 (56) 50 (15) .75
Race, % (n) .37
White 80 (20) 71 (233) 83 (88) 63 (19) —
Black or African American 8 (2) 15 (50) 10 (11) 27 (8) —
Other 12 (3) 14 (44) 7 (7) 10 (3) —
Ethnicity, % (n) .58
Hispanic 12 (3) 14 (46) 21 (22) 10 (3) —
Not Hispanic 88 (22) 84 (276) 78 (83) 90 (27) —
Unknown 0 (0) 2 (5) 1 (1) 0 (0) —
Household smoking exposure, % (n) 28 (7) 27 (89) 27 (29) 20 (6) .86
No. comorbidities, % (n) .14
0 56 (14) 54 (176) 47 (50) 43 (13) —
1 32 (8) 28 (90) 26 (28) 17 (5) —
2 4 (1) 12 (40) 13 (14) 23 (7) —
$3 8 (2) 6 (21) 13 (14) 17 (5) —
Median PF ratio (IQR)a 474 (462–479) 457 (428–474) 394 (251–445) 410 (244–451) ,.001
Median HR (IQR) 137.0 (118.0–156.0) 136.0 (120.0–155.0) 145.0 (127.0–160.8) 147.0 (140.0–168.0) .006

Downloaded from http://publications.aap.org/hospitalpediatrics/article-pdf/11/3/215/814552/hpeds_2020-001842.pdf

Median RR (IQR) 26.0 (22.0–34.0) 32.0 (24.0–40.0) 39.5 (32.0–52.0) 41.5 (32.0–54.8) ,.001
Median SBP (IQR) 109.0 (102.0–119.5) 108.0 (99.2–117.0) 110.0 (103.2–119.0) 101.0 (96.2–113.0) .034
Chest indrawing, % (n) 16 (4) 38 (123) 65 (69) 60 (18) ,.001
Infiltrate pattern, % (n) .003
Single lobar 60 (15) 53 (174) 30 (32) 47 (14) —
Multilobar 12 (3) 20 (67) 32 (34) 13 (4) —
Interstitial 20 (5) 9 (30) 13 (14) 13 (4) —
Mixed 0 (0) 5 (16) 8 (8) 0 (0) —
Unknown 8 (2) 12 (40) 17 (18) 28 (8) —
Presence of effusion 8 (2) 14 (46) 13 (14) 40 (12) ,.001
Antibiotics administration before procalcitonin 60 (15) 84 (276) 77 (83) 90 (27) .006
collectionb
—, not applicable.
a
Ratio of PaO2 to FiO2, estimated from the SpO2 to FiO2 ratio
b
Oral or intravenous antibiotic.

SARTORI et al
 before procalcitonin collection in the ED or
while admitted, the mean time from
antibiotic administration to procalcitonin
collection was 15.8 hours (95% CI 13.7–17.9).
Procalcitonin concentration was not
different (P 5 .65) between the no antibiotic
pretreatment group (median 0.55 ng/mL;
IQR 0.15–2.75) and the antibiotic
pretreatment group (median 0.33 ng/mL;
IQR 0.10–2.29).
DISCUSSION
In our study of 488 children with pneumonia
presenting for emergency care, increased
procalcitonin was associated with both our
primary outcome of increased pneumonia
severity and our secondary outcome of
increased hospital LOS. In our secondary
analysis excluding those with very severe
pneumonia, procalcitonin was no longer
significant for the ordinal severity outcome
but remained significant for the LOS
outcome.
In our adjusted models, procalcitonin was a
significant predictor of severe pneumonia
FIGURE 1 Procalcitonin concentration by ordinal severity outcome. The log10 procalcitonin outcomes. This is in line with adult studies
concentration among children with pneumonia was stratified by ordinal severity associating elevated procalcitonin levels
outcome. Dark lines inside the boxes denote the median and box borders denote the
with increased risk of mechanical
IQR. Bars represent upper and lower adjacent values. Circles represent individual
values. The y-axis is presented in log10 scale. ventilation or vasopressor support within
72 hours of presentation and smaller
pediatric studies associating elevated
Of the 17 undergoing drainage procedures, Procalcitonin and Antibiotic Use procalcitonin with hospital admission, ICU
4 (24%) had positive bacterial growth Eighty-eight children (18%) had admission, and LOS.13,14,19 These findings are
results on pleural culture, all with high procalcitonin levels obtained before the similar to those observed by Florin et al,15
procalcitonin levels (36.09, 31.21, 17.49, and administration of any antibiotic (including who also demonstrated associations
8.58 ng/mL). Bacterial pathogens detected outpatient oral antibiotics). Of those who between procalcitonin and an ordinal
in pleural fluid were S pyogenes (2), S had not received an antibiotic before severity outcome (most severe defined as
pneumoniae (1), and S intermedius with presentation to the enrolling ED, but for ICU care, vasoactive infusions, effusion
whom an antibiotic was administered drainage, or severe sepsis) adjusted for
Parvimonas micra coinfection (1).
age, antibiotic receipt before arrival, and
TABLE 2 aORs of Ordinal Severity Outcomes and LOS Outcome by Varying Procalcitonin Cut length of fever.
Points Very severe outcomes were rare in our
Low or Reference to High, ng/mL Ordinal Severity Outcome LOS Outcome cohort, with only 6% requiring mechanical
Odds Ratio (95% CI) Odds Ratio (95% CI)
ventilation and 3% requiring use of
0.1–0.5 1.05 (1.01–1.10) 1.07 (1.04–1.11) vasoactive medications. This group
0.1–1.0 1.12 (1.02–1.23) 1.17 (1.08–1.26) demonstrated significantly higher median
0.1–2.0 1.25 (1.04–1.51) 1.36 (1.17–1.59) procalcitonin concentrations (median
0.25–0.5 1.03 (1.01–1.06) 1.04 (1.02–1.07) 5.06 ng/mL) compared with other groups,
0.25–1.0 1.10 (1.01–1.18) 1.14 (1.07–1.21) all of which had median procalcitonin
0.25–2.0 1.23 (1.03–1.46) 1.33 (1.16–1.53) concentrations ,0.40 ng/mL. Although our
secondary analysis excluding the very
0.5–1.0 1.06 (1.01–1.12) 1.09 (1.04–1.13)
severe group was limited by a smaller
0.5–2.0 1.19 (1.03–1.38) 1.27 (1.13–1.43)
sample size, it provides further context to
2.0–3.5 1.16 (1.03–1.32) 1.23 (1.11–1.36)
our results and suggests that very severe

HOSPITAL PEDIATRICS Volume 11, Issue 3, March 2021 219

Downloaded from http://publications.aap.org/hospitalpediatrics/article-pdf/11/3/215/814552/hpeds_2020-001842.pdf

by guest
 our subjects with bacteremia supports
bacterial etiology association with
procalcitonin elevation.25 This is further
supported by the high incidence of pleural
effusion (40%) in the very severe group. In
those with parapneumonic effusion severe
enough to require drainage, procalcitonin
values were very high (median 10.45 ng/mL
[IQR 3.05–20.01]). Florin et al15 similarly
demonstrated associations between
procalcitonin and those requiring chest
drainage. Although only 4 subjects had
bacterial growth on pleural culture (all with
procalcitonin .8 ng/mL), we suspect that
nearly all of these children had bacterial
pneumonia and that bacterial growth was
limited by antibiotic pretreatment. This
latter hypothesis is supported by the
Etiology of Pneumonia in the Community
study, in which researchers reported a
dramatic decrease in pleural culture yield
when antibiotics were administered before
drainage (71% vs 37%).2
In contrast, low procalcitonin values
(,0.25 ng/mL) are associated with low risk
for bacterial etiology.26 In our study,
approximately half of the children had a
procalcitonin concentration ,0.25 ng/mL,
FIGURE 2 Predicted probability of severe or very severe pneumonia by initial procalcitonin suggesting bacterial disease was unlikely.
concentration. Predicted probabilities were estimated from the model by using an This assumption aligns with findings from
age of 5 years, HR of 100, SBP 100, and PF ratio 450. All other covariates are included
the Etiology of Pneumonia in the Community
at reference values. Procalcitonin was modeled by using a restricted cubic spline with
3 knots. The blue line represents predicted probability; the gray shaded area study, with the majority of pediatric
represents 95% CIs. For example, for a 5-year-old patient with an HR of 100, SBP of pneumonias in the pneumococcal conjugate
100, and a normal PF ratio of 450, an increase in procalcitonin from 0.1 to 1.0 ng/mL vaccine era caused by viral pathogens.2
results in an increase in predicted probability of severe or very severe pneumonia Although viral pneumonias may also
from 2.2% to 2.8%. contribute to severe outcomes, especially in
younger children,27 whether procalcitonin
outcomes are a key driver of the tumor necrosis factor and interleukin-1 and predicts these more severe outcomes in the
associations between procalcitonin and interleukin-6 and inhibited in viral infections absence of bacterial disease is unclear.
pneumonia severity. In the study by Florin through the action of interferon.10,20,21 These Recent investigations suggest an
et al,15 procalcitonin also did not unique qualities make procalcitonin a association between procalcitonin and
discriminate between those with less severe leading candidate for the differentiation of severe coronavirus disease 2019 illness, but
outcomes. Thus, although procalcitonin is bacterial and viral etiologies, with these studies are small and limited to
associated with increasing disease severity, procalcitonin used to guide antibiotic adults.28
its utility for differentiating among those administration in sepsis and other bacterial Difficulties identifying causative pathogens
with less severe pneumonia outcomes in the processes and revealing superior ability to in pediatric pneumonia highlight the
clinical setting may be more limited. differentiate these etiologies from other importance of using procalcitonin in
We hypothesize that the utility of biomarkers such as C-reactive protein.22–24 concert with other clinical decision-making
procalcitonin as a predictor of severe Procalcitonin values .2.0 ng/mL have been tools. The wide variation in procalcitonin
pneumonia is attributable in part to associated with bacteremic pneumococcal concentrations observed in our study, with
procalcitonin’s association with bacterial pneumonia, and although our study was not some mildly ill children demonstrating high
pathogens. Procalcitonin is upregulated in designed to assess pneumonia etiology, the procalcitonin values and some severely ill
bacterial infections through the release of very high procalcitonin concentrations in children demonstrating low procalcitonin

220 SARTORI et al

Downloaded from http://publications.aap.org/hospitalpediatrics/article-pdf/11/3/215/814552/hpeds_2020-001842.pdf

by guest
 useful in terms of how clinicians assess
patient needs but may not fully reflect the
physiology of illness severity. However,
our findings support previous work
revealing association between increased
procalcitonin and need for ICU admission in
pediatric community-acquired pneumonia.13
As previously discussed, our study was not
designed to systematically assess etiology
of community-acquired pneumonia through
comprehensive microbiologic testing. As a
result, we were not able to determine if
pneumonia etiology modifies associations
between procalcitonin and severe
pneumonia outcomes. Finally, 82% of
children received some form of antibiotic
before procalcitonin collection. Although
procalcitonin values did not differ by receipt
of antibiotics, it is possible that antibiotic
pretreatment could have falsely lowered
procalcitonin. Antibiotic treatment would
not be expected to influence procalcitonin
values in those with viral pneumonia, and
given that the majority of pneumonias in US
children are caused by viral pathogens, the
influence of antibiotics on procalcitonin in
our study may be minimal.2
FIGURE 3 Predicted probability of LOS .3 days by initial procalcitonin concentration. Predicted
probabilities were estimated from the model by using an age of 5 years, HR 100, SBP CONCLUSIONS
100, and PF ratio 450. All other covariates are included at reference values.
Procalcitonin was modeled by using restricted cubic spline with 3 knots. The blue line Our findings indicate that increased
represents predicted probability; the gray shaded area represents 95% CIs. For procalcitonin concentration is associated
example, for a 5-year-old patient with an HR of 100, SBP of 100, and a normal PF ratio with more severe outcomes and longer LOS
of 450, an increase in procalcitonin from 0.1 to 1.0 ng/mL results in an increase in among children with pneumonia.
predicted probability of LOS .3 days from 10% to 12%. Procalcitonin may be useful in helping
clinicians evaluate pneumonia severity,
values, reveals that procalcitonin alone is discharged from the ED provided blood although further study is needed,
imperfect in predicting disease outcomes. In samples, suggesting our population likely particularly as it relates to procalcitonin in
addition, the independent association represents a more severely ill cohort than those with noncritical illness. In future
between procalcitonin and pneumonia the average child presenting to the ED with studies, researchers should explore the
disease severity in children appears pneumonia. This is not unexpected and utility of procalcitonin integration into risk
modest. Nonetheless, in adult studies, aligns with current clinical practice, with stratification models and other decision
researchers have found that adding well-appearing children quickly discharged support applications to determine if these
procalcitonin to established adult from the hospital without laboratory tools improve clinical decision-making.
pneumonia risk classification systems evaluations. Nonetheless, we acknowledge
REFERENCES
improves prognostic accuracy, with that our study may not fully characterize the
procalcitonin levels in the ED more useful range of procalcitonin values in mild illness. 1. Self WH, Grijalva CG, Zhu Y, et al. Rates of
than measurements in the ICU or primary Enrollment occurred at 2 tertiary care emergency department visits due to
care locations.19,29 It will be important to test children’s hospitals, with the majority (86%) pneumonia in the United States, July
the utility of procalcitonin in similar coming from a single institution. This may 2006-June 2009. Acad Emerg Med. 2013;
pediatric risk stratification tools. limit the generalizability of the study’s 20(9):957–960
Our study has limitations to consider. We findings. Our outcome groups also 2. Jain S, Williams DJ, Arnold SR, et al; CDC
were not able to enroll every eligible child incorporated hospital location (floor versus EPIC Study Team. Community-acquired
presenting with pneumonia. Few children ICU) into severity classification, which is pneumonia requiring hospitalization

HOSPITAL PEDIATRICS Volume 11, Issue 3, March 2021 221

Downloaded from http://publications.aap.org/hospitalpediatrics/article-pdf/11/3/215/814552/hpeds_2020-001842.pdf

by guest
 among U.S. children. N Engl J Med. 2015; prediction of hospitalized patients with 20. Self WH, Balk RA, Grijalva CG, et al.
372(9):835–845 community acquired pneumonia. Ann Procalcitonin as a marker of etiology in
Thorac Med. 2014;9(3):162–167 adults hospitalized with community-
3. Lee GE, Lorch SA, Sheffler-Collins S,
11. Huang DT, Weissfeld LA, Kellum JA, et al; acquired pneumonia. Clin Infect Dis.
Kronman MP, Shah SS. National
GenIMS Investigators. Risk prediction 2017;65(2):183–190
hospitalization trends for pediatric
pneumonia and associated with procalcitonin and clinical rules in 21. Becker KL, Nylén ES, White JC, Müller B,
complications. Pediatrics. 2010;126(2): community-acquired pneumonia. Ann Snider RH Jr. Clinical review 167:
204–213 Emerg Med. 2008;52(1):48–58.e2 procalcitonin and the calcitonin gene
12. Müller B, Harbarth S, Stolz D, et al. family of peptides in inflammation,
4. Agency for Healthcare Research and
Diagnostic and prognostic accuracy of infection, and sepsis: a journey
Quality. National estimates on use of
clinical and laboratory parameters in from calcitonin back to its precursors.
hospitals by children from the HCUP
community-acquired pneumonia. BMC J Clin Endocrinol Metab. 2004;89(4):
Kids’ Inpatient Database (KID). Available
Infect Dis. 2007;7:10 1512–1525
at: http://hcupnet.ahrq.gov/. Accessed
January 12, 2014 13. Stockmann C, Ampofo K, Killpack J, et al. 22. Vijayan AL, Vanimaya , Ravindran S, et al.
Procalcitonin accurately identifies Procalcitonin: a promising diagnostic
5. Keren R, Luan X, Localio R, et al; Pediatric
hospitalized children with low risk of marker for sepsis and antibiotic therapy.
Research in Inpatient Settings (PRIS)
bacterial community-acquired J Intensive Care. 2017;5:51
Network. Prioritization of comparative
effectiveness research topics in hospital pneumonia. J Pediatric Infect Dis Soc. 23. Simon L, Gauvin F, Amre DK, Saint-Louis P,
pediatrics. Arch Pediatr Adolesc Med. 2018;7(1):46–53 Lacroix J. Serum procalcitonin and
2012;166(12):1155–1164 14. Don M, Valent F, Korppi M, et al. Efficacy C-reactive protein levels as markers of
of serum procalcitonin in evaluating bacterial infection: a systematic review
6. Bourgeois FT, Monuteaux MC, Stack AM,
severity of community-acquired and meta-analysis. Clin Infect Dis. 2004;
Neuman MI. Variation in emergency
pneumonia in childhood. Scand J Infect 39(2):206–217
department admission rates in US
children’s hospitals. Pediatrics. 2014; Dis. 2007;39(2):129–137 24. Jin M, Khan A. Procalcitonin: uses in the
134(3):539–545 15. Florin TA, Ambroggio L, Brokamp C, et al. clinical laboratory for the diagnosis of
Biomarkers and disease severity in sepsis. Labmedicine. 2010;41:173–177
7. Dean NC, Jones JP, Aronsky D, et al.
Hospital admission decision for patients children with community-acquired 25. Baumann P, Baer G, Bonhoeffer J, et al.
with community-acquired pneumonia: pneumonia. Pediatrics. 2020;145(6): Procalcitonin for diagnostics and
variability among physicians in an e20193728 treatment decisions in pediatric lower
emergency department. Ann Emerg Med. 16. Katz SE, Sartori LF, Szeles A, et al. respiratory tract infections. Front
2012;59(1):35–41 Agreement between two procalcitonin Pediatr. 2017;5:183

8. Fine MJ, Hough LJ, Medsger AR, et al. The assays in hospitalized children. Infect 26. Katz SE, Sartori LF, Williams DJ. Clinical
hospital admission decision for patients Dis Ther. 2019;8(3):463–468 progress note: procalcitonin in the
with community-acquired pneumonia. 17. Williams DJ, Zhu Y, Grijalva CG, et al. management of pediatric lower
Results from the pneumonia Patient Predicting severe pneumonia outcomes respiratory tract infection. J Hosp Med.
Outcomes Research Team cohort study. in children. Pediatrics. 2016;138(4): 2019;14(11):688–690
Arch Intern Med. 1997;157(1):36–44 e20161019 27. Shan W, Shi T, Chen K, et al. Risk factors
9. Bradley JS, Byington CL, Shah SS, et al; 18. Rice TW, Wheeler AP, Bernard GR, Hayden for severe community-aquired
Pediatric Infectious Diseases Society and DL, Schoenfeld DA, Ware LB; National pneumonia among children hospitalized
the Infectious Diseases Society of Institutes of Health, National Heart, Lung, with CAP younger than 5 years of
America. The management of and Blood Institute ARDS Network. age. Pediatr Infect Dis J. 2019;38(3):
community-acquired pneumonia in Comparison of the SpO2/FIO2 ratio and 224–229
infants and children older than the PaO2/FIO2 ratio in patients with 28. Lippi G, Plebani M. Procalcitonin in
3 months of age: clinical practice acute lung injury or ARDS. Chest. 2007; patients with severe coronavirus
guidelines by the Pediatric Infectious 132(2):410–417 disease 2019 (COVID-19): a meta-
Diseases Society and the Infectious analysis. Clin Chim Acta. 2020;505:
19. Self WH, Grijalva CG, Williams DJ, et al.
Diseases Society of America. Clin Infect 190–191
Procalcitonin as an early marker of the
Dis. 2011;53(7):e25–e76
need for invasive respiratory or 29. Kutz A, Briel M, Christ-Crain M, et al.
10. Andrijevic I, Matijasevic J, Andrijevic L, vasopressor support in adults with Prognostic value of procalcitonin in
Kovacevic T, Zaric B. Interleukin-6 and community-acquired pneumonia. Chest. respiratory tract infections across
procalcitonin as biomarkers in mortality 2016;150(4):819–828 clinical settings. Crit Care. 2015;19:74

222 SARTORI et al

Downloaded from http://publications.aap.org/hospitalpediatrics/article-pdf/11/3/215/814552/hpeds_2020-001842.pdf

by guest