Epidemiology and Outcome of Sepsis in a Tertiary Care PICU of Pakistan

Muhammad Rehan Khan & Prem Kumar Maheshwari & Komal Masood & Farah
Naz Qamar & Anwar-ul Haque Received: 23 April 2011 /Accepted: 7 February
2012 / Published online: 6 March 2012 # Dr. K C Chaudhuri Foundation 2012
Abstract Objective To determine the epidemiology and outcome of sepsis in
children admitted in pediatric intensive care unit (PICU) of a tertiary care hospital.
Methods Retrospective review of children 1 mo to 14 y old, admitted to the PICU
with severe sepsis or septic shock from January 2007 through December 2008 was
done. Demographic, clinical and laboratory features of subjects were reviewed.
The primary outcome was mortality at the time of discharge from PICU. The
independent predictors of mortality were modeled using multiple logistic
regression. Results In 2 years, 17.3% (133/767) children admitted to the PICU had
sepsis. Median age was 18 mo (IQR 693 mo), with male: female ratio of 1.6:1.
Mean PRISM III score was 9 (7.8). One third had culture proven infection,
majority (20%) having bloodstream infection. The frequency of multi-organ
dysfunction syndrome (MODS) was 81% (108/133). The case specific mortality
rate of sepsis was 24% (32/133). Multi-organ dysfunction (Adjusted OR 18.0,
95% CI 2.2144), prism score of >10 (Adjusted OR 1.5, 95% CI 0.64.0) and the
need for>2 inotropes (Adjusted OR 3.5, 95% CI 1.39.2) were independently
associated with mortality due to sepsis. Conclusions The presence of septic shock
and MODS is associated with high mortality in the PICU of developing countries.
Keywords Sepsis. Septic shock . Tertiary care . Pediatric intensive care . Pakistan
Introduction Sepsis in children remains to be a significant cause of morbidity and
mortality worldwide [1]. Watson et al. reported frequency of pediatric severe
sepsis to be more than 42000 cases annually and associated mortality rate to be
10% in the United States [2]. Recently published report from the United Kingdom
showed that 17% of children die from with severe sepsis and septic shock in
pediatric intensive care unit (PICU) [3]. The mortality rate of sepsis in children
from PICU of developing countries is higher than 50% [4, 5]. World Health
Organization statistics have shown that 80% of death in children less than 4 y can
be classified as sepsis related deaths [6]. To improve the outcome of pediatric

sepsis, World Federation of Pediatric Intensive Critical Care (WFPICC) sepsis

initiative in developing countries emphasized the importance of simple
interventions like early rapid fluid administration, early antibiotics therapy,
oxygen supplementation, and early use of inotropes through peripheral
intravenous access [7]. Several recent studies from PICUs have shown that there
was a significant reduction in the mortality rate of children with severe sepsis and
septic shock [8, 9]. There is a need to document the outcome of children with
severe sepsis and septic shock from an organized PICU of Pakistan. Material and
Methods The study was carried out at PICU of Aga Khan University Hospital
(AKUH), Karachi. AKUH is a 500-beded tertiarycare referral center with
combined adult-pediatric intensive care unit. This PICU is a closed-unit, where
patient care is delivered by pediatric residents under supervision of an attending
physician board certified in pediatric critical care medicine. There are on an
average yearly more than 450 admissions in the PICU, out of which the major
bulk (about 17%) is due to infections. The authors retrospectively reviewed the
medical records of all children aged one-month to 14-y admitted to the PICU
either from emergency room or pediatric wards with diagnosis of sepsis, severe
sepsis and septic shock from January 2007 through December 2008. As per policy
of PICU, patients are categorized into severe sepsis and septic shock according to
the criteria laid down by the International Pediatric Sepsis Consensus Conference
2005[10] and are managed according to the clinical practice guidelines of the
American College of Critical Care Medicine-Pediatric Advanced Life Support
(ACCM/PALS) published in 2002 [11]. The therapeutic interventions are started
from the time diagnosis is suspected and includes early aggressive fluid
resuscitation by pull-push method, early administration of empiric antibiotics,
oxygen supplement, and early infusion of inotropic agent through peripheral
intravenous access. For each patient, demographic data like age, gender,
diagnosis, type of admission (medical or surgical) length of stay, primary
diagnosis and associated co-morbidities was extracted. A severity score (pediatric
risk of mortality, PRISM III) and the organ dysfunction criteria are routinely
calculated for all patients admitted to the PICU [12]. Organ dysfunction criteria

used, are defined according to the International Pediatric Sepsis Consensus

Conference 2005. Blood and urine cultures of all the cases were sent at the time of
admission to the PICU. CSF culture was sent for all sick infants 7 d). The
demographic and clinical characteristics of the patients are described in table 1.
Almost 1/3 of children (40/133) had culture proven infection. Blood stream
infection was common, seen in 20.3% (27/133) while 10.5% (14/33) grew
organisms in tracheal culture. The results of blood, tracheal, urine and CSF culture
of the study subjects is shown in table 2. Multi-organ dysfunction was seen in
81% of the patients (108/133). The increasing number of organ dysfunction was
associated with higher mortality rate as shown in fig. 1. The case specific
mortality rate of sepsis was 24% (32/ 133). In patients with septic shock, the
mortality rate was about 32.6% (31/95) while in severe sepsis group it was 2.7%
(1/37). Out of the patients having MODS, 30% expired (32/108). The mean
PRISM score of non-survivors was higher than that of survived patients (p value
10 (Adjusted Odds Ratio 1.5, 95% CI 0.64.0) and the need for>2 inotropes
(Adjusted Odds Ratio 3.5, 95% CI 1.39.2) The results of univariate and
multivariate analysis are shown in tables 3 and 4. Discussions To the best of
authors knowledge, this is the first report on epidemiology and outcome of sepsis
in children from a tertiary-care PICU of Pakistan where diagnosis is established
on the criteria as defined by International Pediatric Sepsis Consensus Conference
2005 and the patients are managed according to clinical practice guidelines of
pediatric septic shock of ACCM/PALS 2002. The mortality related to sepsis in
pediatric age group has decreased significantly over the past few decades [15].
This is largely due to implementation of standard protocols, guidelines, clinical
practice parameters, goal directed therapies and educational programs
implemented throughout the world to combat sepsis associated mortality [16, 17].
Despite all the improvement in developed world, sepsis is still a significant health
care problem and is associated with high morbidity and mortality in PICU settings
of developing countries [4, 5]. The case specific mortality of sepsis in the present
study was 24%. Due to the lack of uniform definition of sepsis in the past, there
was a wide range of variability in reporting the true incidence of sepsis. The

authors used the diagnostic criteria of the first consensus on pediatric sepsis, so it
can be helpful in comparing the epidemiological study on pediatric sepsis
internationally. The hospitalization rate for severe sepsis has almost doubled in the
last decade [18]. The frequency of severe sepsis and septic shock was slightly
high in the present study as compared to previous data which have reported it to
be from 1.6% to 4% and 2% to 3% respectively [19]. There was no significant
difference with respect to age and sex of children in the present study as compared
to available literature. Infants had the highest frequency of sepsis in the present
study. Similar results have been shown by Watson et al. in their study in which
48% of the patients with sepsis were less than 1 year old [2]. There is a male
preponderance reported in previous studies (55%59%) in patients with sepsis
which is comparable to the present study [20]. Only one fifth of the patients had a
positive blood culture as compared to a high culture yield of about 50%60% in
previous studies[20, 21]. This could be due to use of antibiotics prior to
presentation as these drugs are readily available over the counter in this part of the
world. However, the possibility of contamination during the blood collection from
peripheral vessel for the high incidence of coagulasenegative staphylococcus
bacteremia cannot be ruled out. The frequency of sepsis-related MODS in the
present study was significantly higher than previously published reports in late
1990s in which the frequency of MODS was around 18% [22]. However, the
recent studies have shown that there is a significant increase in the incidence of
MODS [20]. This could be either due to increasing awareness, better
documentation of organ dysfunction or delay in presentation to intensive care
setting. The overall mortality in the present study is comparable to figures from
United Kingdom and Italy [3, 20]. Much higher mortality rates have been
observed from PICUs of developing countries like Brazil and India [4, 5]. The
present low mortality rate could be due to strict adherence to the international
guidelines and standardized case management. There were few limitations and
strengths in this study. Firstly, this was a retrospective study and secondly, it was
limited to a single private sector hospital which caters for only a limited
population and thus, there is a selection bias and the results may not be the true

reflection of actual epidemiology and outcome of sepsis in the country. Moreover,

patients with sepsis from pediatric wards were not included because of variability
in clinical practices. Prospective and multi-centered studies are needed for precise
identification of epidemiological trends and estimation of burden of sepsis in this
part of the world. The strength of this study is that all patients admitted to the
pediatric intensive care unit were assessed and managed according to standard
guidelines by the same investigator. This is the first study based on diagnosis of
severe sepsis and septic shock according to criteria laid down by the International
Pediatric Sepsis Consensus 2005. Moreover, these patients were managed
according to clinical practice guidelines published by ACCM in 2002. Use of
standard definitions and treatment will help in comparing the results with
international data. Conclusions Sepsis in children is associated with high mortality
despite aggressive treatment strategies. The multiorgan dysfunction syndrome,
high PRISM score and need for multiple inotropes are important risk factors for
mortality. Early recognition and prompt treatment is the key to improve outcome
of sepsis. More prospective studies should be done on the basis of new sepsis
definitions and clinical practice guidelines to evaluate the true burden and
outcome of sepsis in the developing countries. Conflict of Interest and Role of
Funding Source Dr Farah Naz Qamar, received research training support from the
National Institute of Healths Fogarty International Center (1 D43 TW007585-01).
The sponsors did not have any role in study design, data analysis or report writing