Content uploaded by Daniela Carla de Souza
Author content
All content in this area was uploaded by Daniela Carla de Souza on Dec 17, 2020
Content may be subject to copyright.
Copyright © 2016 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies.
Unauthorized reproduction of this article is prohibited
Pediatric Critical Care Medicine www.pccmjournal.org 727
Objectives: To report the prevalence of sepsis within the first 24
hours at admission and the PICU sepsis-related mortality among
critically ill children admitted to PICU in South America.
Design: A prospective multicenter cohort study.
Setting: Twenty-one PICU, located in five South America coun-
tries.
Patients: All children from 29 days to 17 years old admitted to
the participating PICU between June 2011 and September 2011.
Clinical, demographic, and laboratory data were registered within
the first 24 hours at admission. Outcomes were registered upon
PICU discharge or death.
Interventions: None.
Measurements and Main Results: Of the 1,090 patients included
in this study, 464 had sepsis. The prevalence of sepsis, severe
sepsis, and septic shock were 42.6%, 25.9%, and 19.8%,
respectively. The median age of sepsis patients was 11.6 months
(interquartile range, 3.2–48.7) and 43% had one or more prior
chronic condition. The prevalence of sepsis was higher in infants
(50.4%) and lower in adolescents (1.9%). Sepsis-related mortal-
ity was 14.2% and was consistently higher with increased disease
severity: 4.4% for sepsis, 12.3% for severe sepsis, and 23.1% for
septic shock. Twenty-five percent of deaths occurred within the
first 24 hours at PICU admission. Multivariate analysis showed
that higher Pediatric Risk of Mortality and Pediatric Logistic Organ
Dysfunction scores, the presence of two or more chronic con-
ditions, and admission from pediatric wards were independently
associated with death.
Conclusions: We observed high prevalence of sepsis and sepsis-
related mortality among this sample of children admitted to PICU
in South America. Mortality was associated with greater severity
of illness at admission and potentially associated with late PICU
referral. (Pediatr Crit Care Med 2016; 17:727–734)
Key Words: mortality; pediatrics; prevalence; septic shock; severe
sepsis
Sepsis is a complex and dynamic process that may present
with a variety of nonspecific signs and symptoms, which
may differ in severity among individuals or in the same
individual throughout the course of the disease. The different
stages of sepsis represent a continuum of disease severity that,
unless early recognized and treated, may result in multiple
organ dysfunctions and, eventually, death (1, 2).
Sepsis consumes a substantial portion of ICUs’ resources,
is a common cause of death in children, and is considered a
public health problem (3–5). More than 50% of the deaths in
children under 5 years old result from severe infectious dis-
eases, such as pneumonia, diarrhea, and malaria, all of which
may lead to sepsis (6, 7). Sepsis, therefore, may be considered a
major cause of death in children worldwide.
Over the last few decades, there has been well documented
rise in pediatric sepsis prevalence (8–10). An 81% increase in
the number of pediatric severe sepsis cases was reported in the
United States between 1995 and 2005. In 2005, the reported
incidence of sepsis in patients under 20 years in that country
was 0.89: 1,000 children, resulting in 75,000 sepsis admissions
(8). The rising number of sepsis admissions may be related to
the increased survival of high-risk patients, such as very low
birthweight infants and children with complex chronic condi-
tions (3, 8).
Despite the relevance and severity of sepsis, the available
data neither allow a global perspective on the differences in
the epidemiology, etiology, and outcomes of pediatric sep-
sis among different geographic areas, nor of the influence of
Epidemiology of Sepsis in Children Admitted to
PICUs in South America*
Daniela Carla de Souza, MD1; Huei Hsin Shieh, MD1; Eliane Roseli Barreira, MD1;
Andrea Maria Cordeiro Ventura, MD1; Albert Bousso, MD, PhD1;
Eduardo Juan Troster, MD, PhD2,3; on behalf of the LAPSES Group
*See also p. 794.
11Pediatric Intensive Care Unit, Department of Pediatrics, Hospital Univer-
sitário da Universidade de São Paulo, Sao Paulo, SP, Brazil.
2Department of Pediatrics, Medical School, Universidade de São Paulo,
Sao Paulo, SP, Brazil.
3Pediatric Intensive Care Unit, Department of Pediatrics, Hospital Israelita
Albert Einstein, Sao Paulo, SP, Brazil.
This work was performed at the Department of Pediatrics, Hospital Univer-
sitário da Universidade de São Paulo, Sao Paulo, SP, Brazil.
The authors have disclosed that they do not have any potential conflicts
of interest.
For information regarding this article, E-mail: daniela@hu.usp.br
Pediatric Critical Care Medicine
Pediatr Crit Care Med
1529-7535
10.1097/PCC.0000000000000847
17
8
727
734
2016
Copyright © 2016 by the Society of Critical Care Medicine and the World
Federation of Pediatric Intensive and Critical Care Societies
DOI: 10.1097/PCC.0000000000000847
Saritha
xxx
August
2016
de Souza et al
Copyright © 2016 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies.
Unauthorized reproduction of this article is prohibited
de Souza et al
728 www.pccmjournal.org August2016•Volume17•Number8
regional socioeconomic features on such aspects. These issues
assume greater importance in developing countries, where
infectious diseases represent the leading cause of life-threaten-
ing conditions among children, and the rational distribution
of limited economic resources is essential for planning health-
care policies aiming to reduce childhood mortality (6, 7, 11).
Considering the importance and the scarce knowledge on
several aspects of pediatric sepsis in developing countries, we
conducted the present study to evaluate the prevalence of sep-
sis at PICU admission, the risk factors for sepsis and the PICU
sepsis-related mortality in children admitted to PICU in South
America.
METHODS
This was a prospective multicenter observational study con-
ducted between June 1, 2011, and September 30, 2011. Sixty-
five PICU affiliated with Sociedade Latino-Americana de
Cuidados Intensivos Pediátricos—the Latin American Society
of Pediatric Intensive Care—were invited to join this study.
Twenty-one PICU located in five countries—10 in Brazil, seven
in Argentina, two in Equator, one in Chile and one in Para-
guay—agreed to participate, comprising 257 PICU beds. For
those centers that did not agree to participate, the main rea-
son for nonparticipation was the lack of local researchers. All
the centers were located in state capitals or metropolitan areas
and 81% of them had 100–500 total hospital beds. Among the
participating centers, 13 were public, eight were private, and
seven were university hospitals. Seven hospitals had either local
or international accreditation. Seventy-six percent of the PICU
beds were located in general hospitals, and one PICU provided
care for oncologic patients only. The study was approved by
the Committee of Ethics in Research from each participating
center, and written informed consent was obtained from all the
included patients. Due to different timings for approval of the
study protocol by each local Committee of Ethics on Research,
the duration of data collection differed among the participat-
ing PICU: 4 months in nine, 3 months in eight, and 2 months
in four PICU.
In this study, all children from 29 days to 17 years old con-
secutively admitted to the participating PICU were included.
Children who were admitted to the PICU only for procedures,
who were readmitted within 72 hours after initial PICU dis-
charge, in palliative care or from whom informed consent was
not obtained were excluded.
Clinical, demographic, and laboratory data were registered
within the first 24 hours at PICU admission and included the
following: presence of sepsis, age, gender, prior chronic con-
ditions, nutritional and immunization status, type of admis-
sion (for medical or surgical conditions), origin of the patient
(emergency department [ED], pediatric wards, or operating
room), maternal schooling (used as a surrogate for socioeco-
nomic status), Pediatric Index of Mortality (PIM) 2, Pediatric
Risk of Mortality (PRISM), and Pediatric Logistic Organ
Dysfunction (PELOD) severity scores. PICU length of stay and
PICU mortality were registered upon PICU discharge or death.
The data were registered in an electronic database available at
the study website (https://www.lapsesgroup.com).
Sepsis was defined according to the 2005 International
Pediatric Sepsis Consensus Conference (12). Patients could
be classified into more than one category depending on sep-
sis severity: patients with septic shock were classified into the
three sepsis categories (septic shock, severe sepsis, and sepsis),
patients with severe sepsis into two categories (severe sepsis
and sepsis), and patients with sepsis with no signs of hypoper-
fusion were classified as having sepsis only. Chronic condition
was defined as any medical condition that could be reason-
ably expected to persist for at least 12 months (unless death
occurred) and to involve either several different organ systems
or one organ system severely enough to require specialty pedi-
atric care, and likely to involve a period of hospitalization in
a tertiary care center (13). Multiple organ dysfunction syn-
drome was defined as the presence of two or more organ dys-
functions according to the PELOD score. Nutritional status
was classified according to the Bulletin of the World Health
Organization, using the weight-for-height for children 1–60
months and the Body Mass Index for children greater than 60
months old (14, 15).
Continuous variables were described as mean ± sd or
median and interquartile ranges according to their distribu-
tion, and categorical variables were reported as frequencies,
percentages, and their respective 95% CIs. Comparisons were
performed using the Mann-Whitney U test for continuous data
and the chi-square or Fisher exact tests for categorical data.
Sepsis-related mortality was analyzed using a multiple logistic
regression model that included any variable showing associa-
tion with mortality with a p value less than 0.2 as determined
by univariate analysis. Patients for whom the information
on PICU mortality was missing were excluded from mortal-
ity analysis. Twenty-eight day survival was estimated using
Kaplan-Meier curves and compared with log rank tests. Any
p value less than 0.05 was considered statistically significant.
Statistical analysis was performed using the SPSS 20.0 statisti-
cal software (SPSS Statistics for Windows, V20; Chicago, IL).
RESULTS
During the study period, 1,583 patients were admitted to the
participating PICU, 1,203 patients were eligible, and 1,090
patients were included. Among the study patients, 464 PICU
admissions (42.6%) fulfilled the sepsis criteria, 282 (25.9%)
had severe sepsis, and 216 (19.8%) had septic shock (Fig. 1).
Detailed data are reported for 462 patients; two patients had
missing data on demographics.
Comparisons Between Sepsis and Nonsepsis
Patients
The clinical, demographic, and therapeutic data for sepsis and
nonsepsis patients are shown in Table 1. Compared with non-
sepsis children, sepsis patients were younger (11.6 vs 34.1 mo;
p < 0.001), had higher PRISM (10 vs 5; p < 0.001), PIM 2 (5.6
vs 1.6; p < 0.001), and PELOD (10.1 vs 4.7; p < 0.001) scores
Copyright © 2016 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies.
Unauthorized reproduction of this article is prohibited
Feature Articles
Pediatric Critical Care Medicine www.pccmjournal.org 729
and had a greater number of organ dysfunctions upon admis-
sion (1.9 vs 1.1; p < 0.001). Maternal illiteracy (32 vs 20.2%;
p < 0.001), admission for medical conditions (91.6 vs 66.8%;
p < 0.001), and admission from pediatric wards (32.3 vs 18.8%;
p < 0.001) were also more common in sepsis patients. Multiple
logistic regression analysis showed that female gender (odds
ratio [OR], 2.25; 95% CI, 1.12–4.55; p = 0.023), maternal illit-
eracy (OR, 2.54; 95% CI, 1.37–4.72; p = 0.003), and acquired
immunodeficiency (OR, 2.39; 95% CI, 1.32–4.34; p = 0.004)
were independent risk factors associated with the occurrence
of sepsis in our population. Mechanical ventilation (p < 0.001),
vasoactive drugs (p < 0.001), and blood components transfu-
sion (p < 0.001) were more frequently used for sepsis than for
nonsepsis patients.
Characteristics and Outcomes of Sepsis Patients
Among the study patients, the prevalence of sepsis was higher
in infants (50.4%) and markedly lower (1.9%) in adoles-
cents. Most sepsis patients (91.6%) were admitted for medi-
cal conditions, and the source of admission for 209 patients
(45.2%) was the ED. Two or more organ dysfunctions upon
PICU admission were observed in 255 sepsis patients (55%).
Respiratory and cardiovascular dysfunctions were the most
frequently observed in 310 (67.1%) and 222 (48.1%) patients,
respectively. Prior complex chronic conditions were reported
in 196 children (43.1%) with sepsis, with congenital or genetic
defects (9.9%), respiratory (9%), neuromuscular (9%), and
cardiovascular (7.5%) conditions representing the most
frequent comorbidities. A
history of prematurity was
reported in 44 sepsis patients
(9.7%). The source and etiol-
ogy of the infection that led to
the sepsis episode are reported
in Table 2.
Mortality data were avail-
able for 458 patients. Six
patients (1.3%) had missing
information on mortality and
were excluded from the analy-
sis. Overall sepsis-related mor-
tality was 14.2% (65 deaths).
We observed a stepwise
increase in mortality according
to sepsis severity: 4.4% for sep-
sis with no signs of hypoperfu-
sion, 12.3% for severe sepsis,
and 23.1% for septic shock.
Importantly, one out of four
sepsis-related deaths occurred
within the first 24 hours at
PICU admission.
Comparisons between survi-
vors and nonsurvivors showed
that higher severity and organ
dysfunction scores (p < 0.001),
septic shock (p < 0.001), female gender (p = 0.03), admission
from wards (p = 0.009), immunodeficiency (p = 0.004), the pres-
ence of two or more comorbidities (p < 0.001), and admission to
public hospitals (p = 0.005) were associated with mortality in sep-
sis patients (Table 3). The presence of any organ dysfunction and
the number of organ dysfunctions were related to increased mor-
tality in our patients. After multivariate analysis, higher PRISM
(OR, 1.06; 95% CI, 1.02–1.11; p = 0.005) and PELOD scores (OR,
1.06; 95% CI, 1.02–1.11; p = 0.001), the presence of two or more
chronic conditions (OR, 2.74; 95% CI, 1.4–5.36; p = 0.003) and
admission from wards (OR, 2.44; 95% CI, 1.19–5.01; p = 0.015)
remained significantly associated with mortality. In our sample
of children with sepsis, the presence of septic shock (OR, 1.70;
95% CI, 0.68–4.27; p = 0.256) was not associated with mortal-
ity after multivariate analysis. As shown in Figure 2, 28-day sur-
vival was significantly lower for sepsis than for nonsepsis patients
(hazard ratio, 1.98; 95% CI, 1.20–2.35; long rank p = 0.006).
DISCUSSION
This is the first prospective multicenter study to evaluate the
prevalence of sepsis at PICU admission and PICU sepsis-
related mortality among critically ill children in South America.
We observed high prevalence of both sepsis and sepsis-related
mortality, as well as a high frequency of early deaths among
sepsis patients.
Prior studies have shown wide variability in the prevalence
of pediatric sepsis. Different methodologies, settings, and
Figure 1. Flowchart of patients in this study.
Copyright © 2016 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies.
Unauthorized reproduction of this article is prohibited
de Souza et al
730 www.pccmjournal.org August2016•Volume17•Number8
heterogeneous populations may explain the wide ranges in the
reported sepsis prevalence. Importantly, different numbers of
patients with varying degrees of severity may be identified with
sepsis in a given population, depending on the sepsis definition
adopted. This was shown in two recent studies conducted in
the United States with similar populations of children over the
same time period, which observed differences of up to seven-
fold in sepsis prevalence rates, depending on the definition
adopted for sepsis classification (16, 17). According to Gaieski
et al (18), more inclusive definitions can identify more cases
with lower severity, which more accurately reflect the clinical
experience with sepsis patients, while strict definitions capture
fewer but more severe cases.
In 2005, the publication of the IPSSC pediatric sepsis defi-
nitions aimed to address this heterogeneity and allow a uni-
form evaluation on the epidemiology of pediatric sepsis (12).
Since 2005, seven epidemiologic studies on pediatric sepsis
have been published, two of them conducted in developed
countries (19, 20), three in developing countries (21–23), and
two in various countries worldwide (24, 25). Despite adopting
the same definitions, the reported prevalence of sepsis ranged
from 1% to 27% (Table 4). Compared with those studies,
TABLE 1. Baseline Characteristics and Therapeutics of Sepsis and Nonsepsis Patients
Variable Sepsis Nonsepsis p
Age, mo, median (IQR) 11.6 (3.2–48.7) 34.1 (6.3–101.1) < 0.001
Female gender, n (%) 220/462 (47.6) 285/622 (45.8) 0.557
Pediatric Risk of Mortality, median (IQR) 10 (6–17) 5 (2–10) < 0.001
Pediatric Index of Mortality 2, median (IQR) 5.6 (1.6–13.7) 1.6 (0.8–5.8) < 0.001
Pediatric Logistic Organ Dysfunction, mean ± sd 10.1 ± 9.6 4.7 ± 7.3 < 0.001
No. of organ dysfunctions, mean ± sd 1.9 ± 1.4 1.1 ± 1.2 < 0.001
PICU length of stay in days, median (IQR) 7 (3–15) 3 (2–7) < 0.001
Complete vaccination, n (%) 247/324 (76.2) 332/441 (75.3) 0.762
Nutritional status 0,317
Normal 238/396 (60.1) 327/507 (64.5)
Malnutrition 115/396 (29.0) 125/507 (24.7)
Obesity 43/396 (10.9) 55/507 (10.8)
Maternal schooling < 0.001
Illiteracy 119/372 (32.0) 89/441 (20.2)
Elementary/middle school 134/372 (36.0) 160/441 (36.3)
High school 98/372 (26.3) 150/441 (34.0)
Higher education/post graduation 21/372 (5.6) 42/441 (9.5)
Admission for medical conditions, n (%) 423/462 (91.6) 413/618 (66.8) < 0.001
Source of admission, n (%) < 0.001
Emergency department 209/462 (45.2) 251/618 (40.6)
Wards 149/462 (32.3) 116/618 (18.8)
Operating room 23/462 (5.0) 154/618 (24.9)
Other hospital 81/462 (17.5) 97/618 (15.7)
Chronic conditions, n (%) 196/455 (43.1) 313/618 (50.6) 0,014
Therapies, n (%)
Mechanical ventilation 313/462 (67.7) 241/617 (39.1) < 0,001
Vasoactive drugs 202/462 (43.7) 91/626 (14.5) < 0.001
Blood products 183/462 (39.6) 146/620 (23.5) < 0.001
Renal replacement therapy 8/462 (1.7) 17/624 (2.7) 0.050
IQR = interquartile range.
Copyright © 2016 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies.
Unauthorized reproduction of this article is prohibited
Feature Articles
Pediatric Critical Care Medicine www.pccmjournal.org 731
sepsis prevalence among the patients in our population was
markedly higher (42.6%).
Socioeconomic features and healthcare policies are known
to influence the prevalence of sepsis in different countries.
Additionally, the role of immunization programs in the inci-
dence of infectious diseases in childhood can never be overes-
timated. In our study, we observed low immunization coverage
rates (76%) in our study population. Streptococcus pneumoniae
and Bordetella pertussis, both preventable pathogens that may
cause invasive diseases, were frequent etiologic agents of infec-
tion among our sepsis patients. Low vaccine coverage may,
therefore, be related to the high prevalence of sepsis observed
in our study. Additional markers of socioeconomic develop-
ment, such as sanitary conditions and maternal schooling,
have also been shown to correlate with sepsis incidence in
children (26–28). Although our study did not evaluate sani-
tary aspects, we observed that maternal illiteracy was associ-
ated with the occurrence of sepsis. Finally, the role of seasonal
variations on the incidence of infectious diseases may also be a
contributing factor to the high prevalence of sepsis observed in
our study, given that data collection were conducted during the
winter season, when there is a marked increase in the incidence
of bacterial and viral respiratory diseases (29). Interestingly,
respiratory syncytial virus was the most frequent isolate found
in our sepsis patients, confirming the importance of viruses
in the etiology of pediatric sepsis, as has been shown in prior
studies (17, 23–25).
Despite the high prevalence of sepsis in our study, the
observed sepsis-related mortality rate (14.2%) did not differ
from the 4.5% to 25% mortality rates reported in previous
studies (16, 17, 19–24). According to the World Federation
Pediatric Intensive Critical Care Society Global Pediatric
Sepsis Initiative study, different from the frequency, socio-
economic features do not play a pivotal role in sepsis-related
mortality. Instead, differences in the adherence to the sepsis
treatment guidelines (23.8% vs 51.9%) were associated with
the differences in the sepsis-related mortality (30% vs 11%)
observed between developing and developed countries, respec-
tively (30). Similarly, in the Sepsis Prevalence, Outcomes, and
Therapies study, no differences in sepsis-related mortality
were observed between developed and developing countries
(23% vs 29%) (25).
Several studies have associated sepsis mortality with late
sepsis diagnosis, late hospital admissions, and low adherence
to the published sepsis treatment guidelines (21, 23, 31–36).
We hypothesize that these factors may have influenced the
mortality in our patients, due to the high incidence of early
deaths and the increased mortality in patients admitted to
the PICU coming from pediatric wards instead of the ED. An
intriguing finding in our study is that septic shock was no lon-
ger associated with mortality after multivariate analysis (OR,
1.70; 95% CI, 0.68–4.27; p = 0.256). It is likely that this finding
is related to the small absolute number of deaths in each sepsis
group, resulting in large CIs. Another possible explanation is
that, according to the International Pediatric Sepsis Consensus
Conference definition, both severe sepsis and septic shock can
be defined as sepsis associated with cardiovascular dysfunc-
tion, allowing both groups to overlap.
This is the first prospective study to evaluate the prevalence
of sepsis and sepsis-related mortality in several countries in
TABLE 2. Site of Infection and Microbiologic
Etiology of Sepsis (n = 444)
Characteristic n (%)
Primary site of infection
Respiratory 312 (70.3)
Gastrointestinal 48 (10.8)
Bloodstream 41 (9.2)
CNS 28 (6.3)
Genitourinary 23 (5.2)
Skin 14 (3.2)
Catheter-related infection 4 (0.9)
Cardiac 4 (0.9)
Microbiologic
Gram negative bacteria 93 (20.9)
Neisseria meningitidis 6 (1.4)
Haemophilus influenzae 5 (1.1)
Escherichia coli 15 (3.4)
Pseudomonas aeruginosa 19 (4.3)
Bordetella pertussis 15 (3.4)
Klebsiella pneumoniae 9 (2.0)
Others 24 (5.4)
Gram positive bacteria 74 (16.7)
Streptococcus pneumoniae 23 (5.2)
Group A Streptococcus 2 (0.5)
Methicillin-susceptible Staphylococcus
aureus
19 (4.3)
Methicillin-resistant S. aureus 12 (2.7)
Enterococcus 6 (1.4)
Others 12 (2.7)
Virus 76 (17.1)
Respiratory syncytial virus 60 (13.5)
Influenza A 5 (1.1)
Adenovirus 2 (0.5)
Other virus 9 (2.0)
Fungus 16 (3.6)
Candida albicans 7 (1.6)
Candida tropicalis 2 (0.5)
Aspergillus species 2 (0.5)
Others 5 (1.1)
Copyright © 2016 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies.
Unauthorized reproduction of this article is prohibited
de Souza et al
732 www.pccmjournal.org August2016•Volume17•Number8
South America. The uniformity of the diagnostic criteria
adopted and the large number of patients included support
the accuracy of our results. Importantly, we have demon-
strated the need for a decrease in the incidence of sepsis and
sepsis-related mortality in children admitted to PICU in
South America. Some limitations of this study, nevertheless,
must be discussed. First, our study included hospitals from
five countries in South America, all of them located in met-
ropolitan areas, and only for a few months of the year that
included the winter season—when there is a marked increase
in the incidence of respiratory viral infections—which may
limit the generalizability of the results. Second, nearly one
third of the patients admitted to the PICU were not included
in the study, most of them for admission on weekends, when
the researchers were not present. Given that most of the
patients were lost by chance (admission on weekends), it is
likely that the main characteristics of those patients did not
differ from the studied population although we cannot posi-
tively affirm it. Third, sepsis prevalence was evaluated within
the first 24 hours at PICU admission, and mortality was reg-
istered at PICU discharge. Sepsis patients who were treated
in other settings than the PICU, who developed sepsis after
the first 24 hours at PICU admission or who died after PICU
discharge were not evaluated, which may have underestimated
sepsis prevalence and mortality. In the other hand, we believe
that PICU mortality represents a more accurate estimate of
sepsis-related mortality, since either hospital or late mortality
may be influenced by other factors besides the sepsis episode
that led to the PICU admission. Finally, we neither evaluate
the effect of therapeutic interventions, nor of the time elapsed
from the onset of symptoms or from the initiation of treat-
ment to PICU admission, and the influence of these factors on
the outcomes of sepsis patients.
Epidemiologic studies are necessary to attain a comprehen-
sive understanding of the frequency and impact of a specific
disease in different geographic areas, in order to improve its
Figure 2. Kaplan-Meier curve for survival from sepsis until 28 d after
PICU admission.
TABLE 3. Comparisons Between Sepsis Survivors and Nonsurvivors
Variable Nonsurvivors Survivors p
Age, mo, median 16.6 11.5 0.398
Female gender, n (%) 39/65 (60.0) 178/391 (45.5) 0.030
Pediatric Risk of Mortality, median 17.0 9.0 < 0.001
Pediatric Index of Mortality 2, median 13.8 4.8 < 0.001
Pediatric Logistic Organ Dysfunction, median 21 10 < 0.001
No. of organ dysfunctions, mean ± sd 3.0 ± 1.4 1.7 ± 1.3 < 0.001
PICU length of stay in days, mean ± sd 8.3 ± 11.9 11.9 ± 15.4 < 0.001
Admission for medical conditions, n (%) 63/65 (96.9) 355/390 (91.0) 0.311
Source of admission, n (%) 0.009
Emergency department 23/65 (35.4) 184/391 (47.1)
Wards 33/65 (50.8) 114/391 (29.2)
Operating room 2/65 (3.1) 21/391 (5.4)
Other hospitals 7/65 (10.8) 72/391 (18.4)
No. of chronic conditions, n (%) < 0.001
< 2 40/65 (61.5) 321/392 (81.9)
≥ 2 25/65 (38.5) 71/392 (18.1)
Copyright © 2016 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies.
Unauthorized reproduction of this article is prohibited
Feature Articles
Pediatric Critical Care Medicine www.pccmjournal.org 733
outcomes and to optimize the use of financial resources. Our
study presents preliminary data on the epidemiology and out-
comes of pediatric sepsis in South America, and our results
suggest that efforts to improve the early diagnosis and treat-
ment of pediatric sepsis should be considered when planning
strategies to reduce sepsis-related mortality in this popula-
tion. Further studies are necessary to confirm our findings.
CONCLUSION
Our study showed high prevalence of sepsis within the first
24 hours at admission and PICU sepsis-related mortality in
this sample of critically ill children admitted to PICUs in South
America. Mortality was associated with greater severity of illness
at admission and potentially associated with late PICU referral.
ACKNOWLEDGMENTS
The authors acknowledge the invaluable contribution of all
the researchers who participated in this study and formed the
LAPSES group: Adriana de Oliveira Mukai (Hospital Univer-
sitário de Taubaté, Taubaté, SP, Brasil), Adriana Claudia Bor-
dogna (HIAEP “Sor María Ludovica”, La Plata, Buenos Aires,
Argentina), Ana Paula de Carvalho Panzeri Carlotti (Hospital
Universitário e Unidade de Emergência do Hospital das Clíni-
cas, USP, Ribeirão Preto, SP, Brasil), Arthur F. Delgado (Instituto
da Criança da Faculdade de Medicina da Universidade de São
Paulo, São Paulo, SP, Brasil), Cintia Vranjac (Hospital Cruz Azul,
Sao Paulo, SP, Brasil), Edward Sugo (Unidade de Emergência do
Hospital das Clínicas, USP, Ribeirão Preto, SP, Brasil), Gisella P.
Manjarrés (Hospital Español, Buenos Aires, Argentina; Clinica
Privada Independencia e Sanatorio Privado Figueroa Paredes,
Gran Buenos Aires, Argentina), Gonzalo Soto Germani (Hos-
pital Guillermo Grant Benavente, Concepción, Chile), Gustavo
A. González (Complejo Médico de la Policía Federal Argentina
“Churruca-Visca”, Buenos Aires, Argentina), Hassel Jimmy
Jimenez Rolòn (Hospital de Clinicas San Lorenzo, Faculdade
de Ciências Médicas, Universidade de Assunção, San Lorenzo,
Paraguai), Karina Andrea Cinquegrani (Hospital Nestor Kirch-
ner, Florencio Varela, Gran Buenos Aires, Argentina), Leila
Costa Volpon (Hospital Universitário, USP, Ribeirão Preto, SP,
Brasil), Lilian Elizabeth Hickmann Opazo (Hospital Guillermo
Grant Benavente, Concepción, Chile), Lygia Maria Coimbra
(Hospital Vita, Curitiba, PR, Brasil), Mariana Luquez (Hospi-
tal Lucio Melendez, Argentina), Mónica G. Garea (Complejo
Médico de la Policía Federal Argentina “Churruca-Visca”, Bue-
nos Aires, Argentina), Nilzete Liberato Bresolin (Hospital Infan-
til Joana de Gusmão, Florianópolis, SC, Brasil), Priscila Helena
Félix (Hospital Infantil Sabará, São Paulo, SP, Brasil), Rocío
Yerovi Santos (Hospital de los Valles, Quito, Ecuador), Salma
Brito Saraty (Santa Casa de Belém, Belém, PA, Brasil), Santiago
Campos Miño (Hospital de los Valles, Quito, Ecuador), San-
tiago Hermogenes Esquivel (Hospital Español, Buenos Aires,
Argentina; Clinica Privada Independencia e Sanatorio Privado
Figueroa Paredes, Gran Buenos Aires, Argentina), Sônia Ferraz
de Andrade (Hospital Infantil Joana de Gusmão, Florianópolis,
SC, Brasil), Sonia Noemì Alonso Martìnez (Hospital de Clini-
cas San Lorenzo, Faculdade de Ciências Médicas, Universidade
de Assunção, San Lorenzo, Paraguai), Thiago Schioba (Instituto
da Criança da Faculdade de Medicina da Universidade de São
Paulo, São Paulo, SP, Brasil), and Dr. Werther B. de Carvalho
(Instituto da Criança da Faculdade de Medicina da Universi-
dade de São Paulo, São Paulo, SP, Brasil).
REFERENCES
1. Carcillo JA: Pediatric septic shock and multiple organ failure. Crit
Care Clin 2003; 19:413–440, viii
TABLE 4. Epidemiologic Studies of Pediatric Sepsis in Various Countries—International
Pediatric Sepsis Consensus Conference 2005 Definition
References Country No. of Centers Population
Frequency Mortality
Sepsis
Severe
Sepsis
(%)
Septic
Shock
(%) Overall
Septic
Shock
Jaramillo-
Bustamante
et al (21)
Colombia 19 PICU > 1 mo to < 18 yr old 27.3% 24.8 4 7. 9 18.3 30.4
Shime et al (20) Japan Nine PICU Children < 15 yr old 1.4 18.9
Khan et al (22) Pakistan One PICU 1 mo to 14 yr old 17.3% 24.0 32.6
Van de Voorde
et al (24)
12 countries 16 emergency
department
Children 0.06 4.5
Pérez et al (19) Spain Seven PICU 7 d to 18 yr old 5.6/100,000
inhab/yr
64.7 12.5
Wang et al (23) China 11 Hospital 28 d to 15 yr old 181/100,000
children
(5.5%)
7. 9 2.1 3.5
34.6
34.6
Weiss et al (25) 26 countries 128 PICU Children < 18 yr old 8.2 25.0
Copyright © 2016 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies.
Unauthorized reproduction of this article is prohibited
de Souza et al
734 www.pccmjournal.org August2016•Volume17•Number8
2.VincentJL,MartinezEO,SilvaE:Evolvingconceptsinsepsisdeni-
tions. Crit Care Clin 2009; 25:665–675, vii
3. Watson RS, Carcillo JA, Linde-Zwirble WT, et al: The epidemiology of
severe sepsis in children in the United States. Am J Respir Crit Care
Med 2003; 167:695–701
4. Angus DC, Linde-Zwirble WT, Lidicker J, et al: Epidemiology of severe
sepsis in the United States: Analysis of incidence, outcome, and
associated costs of care. Crit Care Med 2001; 29:1303–1310
5. Thompson GC, Kissoon N: Sepsis in Canadian children: A national
analysis using administrative data. Clin Epidemiol 2014; 6:461–469
6. Liu L, Johnson HL, Cousens S, et al; Child Health Epidemiology
Reference Group of WHO and UNICEF: Global, regional, and
national causes of child mortality: An updated systematic analysis for
2010 with time trends since 2000. Lancet 2012; 379:2151–2161
7. Liu L, Oza S, Hogan D, et al: Global, regional, and national causes of
child mortality in 2000-13, with projections to inform post-2015 pri-
orities: An updated systematic analysis. Lancet 2015; 385:430–440
8. Hartman ME, Linde-Zwirble WT, Angus DC, et al: Trends in the epi-
demiology of pediatric severe sepsis. Pediatr Crit Care Med 2013;
14:686–693
9. Martin GS: Sepsis, severe sepsis and septic shock: Changes in inci-
dence, pathogens and outcomes. Expert Rev Anti Infect Ther 2012;
10:701–706
10. Martin GS, Mannino DM, Eaton S, et al: The epidemiology of sepsis
in the United States from 1979 through 2000. N Engl J Med 2003;
348:1546–1554
11. Kissoon N, Argent A, Devictor D, et al: World Federation of Pediatric
Intensive and Critical Care Societies-its global agenda. Pediatr Crit
Care Med 2009; 10:597–600
12. Goldstein B, Giroir B, Randolph A, et al: International pediatric sepsis
consensus conference: Definitions for sepsis and organ dysfunction
in pediatrics. Pediatr Crit Care Med 2005; 6:2–8
13. Feudtner C, Christakis DA, Connell FA: Pediatric deaths attribut-
able to complex chronic conditions: A population-based study of
Washington State, 1980-1997. Pediatrics 2000; 106:205–209
14. World Health Organization: The WHO Child Growth Standards.
2007. Available at: http://www.who.int/childgrowth/en/. Accessed
March 18, 2013
15. World Health Organization: Growth Reference Data for 5–19 Years.
Available at: http://www.who.int/growthref/en/. Accessed March 18,
2013
16. Balamuth F, Weiss SL, Neuman MI, et al: Pediatric severe sepsis in
U.S. children’s hospitals. Pediatr Crit Care Med 2014; 15:798–805
17. Ruth A, McCracken CE, Fortenberry JD, et al: Pediatric severe sepsis:
Current trends and outcomes from the Pediatric Health Information
Systems database. Pediatr Crit Care Med 2014; 15:828–838
18. Gaieski DF, Edwards JM, Kallan MJ, et al: Benchmarking the inci-
dence and mortality of severe sepsis in the United States. Crit Care
Med 2013; 41:1167–1174
19. Pérez DV,Garcia IJ, TorneEE,etal:Prognosticfactorsinpediatric
sepsis study, from the Spanish Society of Pediatric Intensive Care.
Pediatr Infect Dis J 2014; 33:152–157
20. Shime N, Kawasaki T, Saito O, et al: Incidence and risk factors for
mortality in paediatric severe sepsis: Results from the national pae-
diatric intensive care registry in Japan. Intensive Care Med 2012;
38:1191–1197
21. Jaramillo-Bustamante JC, Marín-Agudelo A, Fernández-Laverde
M, et al: Epidemiology of sepsis in pediatric intensive care units:
First Colombian multicenter study. Pediatr Crit Care Med 2012;
13:501–508
22. Khan MR, Maheshwari PK, Masood K, et al: Epidemiology and out-
come of sepsis in a tertiary care PICU of Pakistan. Indian J Pediatr
2012; 79:1454–1458
23. Wang Y, Sun B, Yue H, et al: An epidemiologic survey of pediatric
sepsis in regional hospitals in China. Pediatr Crit Care Med 2014;
15:814–820
24. VandeVoordeP,EmersonB,GomezB,etal:Paediatriccommunity-
acquired septic shock: Results from the REPEM network study. Eur J
Pediatr 2013; 172:667–674
25. Weiss SL, Fitzgerald JC, Pappachan J, et al; Sepsis Prevalence,
Outcomes, and Therapies (SPROUT) Study Investigators and
Pediatric Acute Lung Injury and Sepsis Investigators (PALISI)
Network: Global epidemiology of pediatric severe sepsis: The sepsis
prevalence, outcomes, and therapies study. Am J Respir Crit Care
Med 2015; 191:1147–1157
26. Mangia CM, Kissoon N, Branchini OA, et al: Bacterial sepsis in
Brazilian children: A trend analysis from 1992 to 2006. PLoS One
2011; 6:e14817
27. GavidiaR,FuentesSL,VasquezR,etal:Lowsocioeconomicstatusis
associated with prolonged times to assessment and treatment, sep-
sis and infectious death in pediatric fever in El Salvador. PLoS One
2012; 7:e43639
28. Schell CO, Reilly M, Rosling H, et al: Socioeconomic determinants
of infant mortality: A worldwide study of 152 low-, middle-, and high-
income countries. Scand J Public Health 2007; 35:288–297
29. Danai PA, Sinha S, Moss M, et al: Seasonal variation in the epidemiol-
ogy of sepsis. Crit Care Med 2007; 35:410–415
30. Kissoon N, Carcillo JA, Espinosa V, et al; Global Sepsis Initiative
Vanguard Center Contributors: World Federation of Pediatric
Intensive Care and Critical Care Societies: Global sepsis initiative.
Pediatr Crit Care Med 2011; 12:494–503
31. Wolfler A, Silvani P, Musicco M, et al; Italian Pediatric Sepsis
Study (SISPe) group: Incidence of and mortality due to sepsis,
severe sepsis and septic shock in Italian Pediatric Intensive Care
Units: A prospective national survey. Intensive Care Med 2008;
34:1690–1697
32. Han YY, Carcillo JA, Dragotta MA, et al: Early reversal of pediatric-
neonatal septic shock by community physicians is associated with
improved outcome. Pediatrics 2003; 112:793–799
33. Oliveira CF, Nogueira de Sá FR, Oliveira DS, et al: Time- and fluid-
sensitive resuscitation for hemodynamic support of children in sep-
tic shock: Barriers to the implementation of the American College of
Critical Care Medicine/Pediatric Advanced Life Support Guidelines
in a pediatric intensive care unit in a developing world. Pediatr Emerg
Care 2008; 24:810–815
34. Santhanam I, Kissoon N, Kamath SR, et al: GAP between knowledge
and skills for the implementation of the ACCM/ PALS septic shock
guidelines in India: Is the bridge too far? Indian J Crit Care Med
2009; 13:54–58
35. Ninis N, Phillips C, Bailey L, et al: The role of healthcare delivery in the
outcome of meningococcal disease in children: Case-control study of
fatal and non-fatal cases. BMJ 2005; 330:1475
36. Inwald DP, Tasker RC, Peters MJ, et al; Paediatric Intensive Care
Society Study Group (PICS-SG): Emergency management of chil-
dren with severe sepsis in the United Kingdom: The results of the
Paediatric Intensive Care Society sepsis audit. Arch Dis Child 2009;
94:348–353