The Epidemiology of Sepsis in Childhood

Abstract

Background: 

Sepsis, or systemic inflammatory response to infection, is a major childhood disease and a common cause of death in children. Despite its importance, a global perspective on the epidemiology and mortality of pediatric sepsis across the world is still lacking.

Methods: 

A non-systematic review of the medical articles published in Medline from 2005 to 2015.

Results: 

Studies suggest that there has been a rise in the number of pediatric sepsis cases along the last two decades, which may relate to the increased survival of preterm and low birth-weight infants and children with severe chronic conditions. Children living in low-income countries represent a vulnerable population for sepsis. Despite several initiatives to improve the diagnosis and early treatment of pediatric sepsis, the mortality resulting from pediatric sepsis remains high, ranging from 5% to 40%. Poor outcomes, however, do not seem to be related to the limitation of resources but to the delay in the recognition and early treatment of sepsis.

Conclusions: 

Educational efforts aiming to increase the awareness on sepsis by the general public and the adherence to the treatment guidelines by healthcare providers may result in significant improvements in sepsis survival. The global attention to pediatric sepsis, however, can only be achieved with the standardization of the definitions and the use of simple and sensitive diagnostic criteria that incorporate the differences in the necessities among different settings and the availability of local resources.

INTRODUCTION

Sepsis, the systemic inflammatory response of the organism to infection, has long been recognized as a life-threatening condition. The correlation between bacterial invasive infection and systemic severe disease has been well established for over a century ^{(1, 2)}. Since then, advances in the understanding of the disease have shown that sepsis is a complex and dynamic process that presents with a myriad of non-specific signs and symptoms, which vary among individuals and in a same individual along the course of the disease, and is closely related to intrinsic aspects of the host immunologic response. Unless early recognized and treated, sepsis may progress to the further stages of severe sepsis and septic shock, leading to multiple organs dysfunction and, eventually, death ^{(1, 3)}.

Epidemiologic studies suggest that sepsis is a major cause of death in childhood worldwide and is often an overlooked public health problem ^{(4, 5)}. Despite the medical, social, and economic burden related to pediatric sepsis, epidemiologic data are still scarce and incomplete, since several studies that investigated the epidemiology of pediatric sepsis were conducted with specific populations, had small sample sizes, or used different diagnostic criteria, making difficult the comparison of results.

Understanding the epidemiology of pediatric sepsis is essential to identify vulnerable populations, to develop diagnosis and treatment guidelines, to draw prevention strategies, and to optimize the distribution of healthcare resources. One major hindrance to this understanding lies on the use of different diagnostic criteria among the studies. Several definitions, such as the sepsis definition proposed by Hayden ⁽⁶⁾, the Bone criteria modified for children ⁽⁷⁾, and various other diagnostic criteria, were used to define sepsis in pediatric studies conducted up to 2004. In 2005, the publication of pediatric definitions for sepsis and septic shock by the International Pediatric Sepsis Consensus Conference (IPSCC) ⁽⁸⁾ resulted in the standardization of the diagnostic criteria, allowing comparisons across the studies.

This article aims to provide an overview on the epidemiology and outcomes of pediatric sepsis in the light of studies published since 2005 and to discuss the current challenges to the understanding on the epidemiology of pediatric sepsis and to the reduction of sepsis-related mortality in childhood.

METHODS

We conducted a non-systematic review of all the epidemiologic studies on pediatric sepsis conducted among hospitalized children that were published in PubMed (Medline) from 2005 to 2015 using the MeSh terms: “sepsis,” “epidemiology,” “prevalence,” “incidence,” “child,” “mortality,” and “outcomes.” When considered relevant, the references cited in these studies were also included.

The frequency of sepsis in childhood

Hartman et al. ⁽⁹⁾ reported that in 2005, 17,542 children were hospitalized for sepsis in seven US states, which represented an 81% increase in the number of severe sepsis cases in children compared with 1995. Although the overall pediatric sepsis incidence increased from 0.56 to 0.89 cases per 1,000 children in that period, the incidence of sepsis in newborns more than doubled, from 4.5 to 9.7 cases per 1,000 live births. In 2005, it was estimated that over 75,000 hospitalizations for pediatric sepsis occurred in the United States. Similarly, Thompson and Kissoon ⁽¹⁰⁾ reported that, from 2004 to 2009, sepsis accounted for 22% of hospital admissions in Canadian children, and that higher prevalence rates were observed for newborns (56.3%) and infants (18.8%).

Since 2005, eight multicenter studies that investigated the epidemiology of pediatric sepsis using the IPSCC definition criteria have been published. Three of these studies were conducted in developed countries ^(11–13), three in developing countries ^(14–16), and two studies in countries with varying degrees of socio-economic development ^{(17, 18)}. Despite adopting the same diagnostic criteria, these studies used different methodologies, recruited patients in different settings, some included newborns and others not, while a few were performed in inpatient units and others in pediatric intensive care units (PICU). These differences may explain the wide variation in the prevalence of sepsis—from 1% to 27%—observed among the reports.

Significant disparities in the incidence and prevalence of sepsis have been observed among populations with similar socio-economic conditions, suggesting that, although socio-economic features may play a pivotal role in the incidence of infectious diseases, other factors may also influence the frequency of sepsis in a given population. In a retrospective study that analyzed the incidence of sepsis in children under 16 years in Australia and New Zealand, Schlapbach et al. ⁽¹¹⁾ reported that among the 97,127 children admitted to the PICU from 2002 to 2013, 11.9% had severe infections, 2.9% had sepsis, and 2.1% had septic shock. The authors also observed an increase of 0.09 and 0.08 cases per 100,000 children per year of sepsis and septic shock, respectively, along that period. Lower prevalence rates were observed in a prospective study conducted in Japan, where Shime et al. ⁽¹³⁾ reported that severe sepsis represented 1.4% of total PICU admissions in patients under 16 years, including newborns. In a prospective study that excluded patients with early neonatal sepsis conducted in Spain, Vila Pérez et al. ⁽¹²⁾ estimated an incidence of 5.6 cases per 100,000 inhabitants per year.

Epidemiologic studies conducted in developing countries have shown higher prevalence of pediatric sepsis than reported for developed countries. In a multicenter prospective study conducted in Colombia, Jaramillo-Bustamante et al. ⁽¹⁴⁾ observed that among the 1,051 children from 1 month to 17 years who had sepsis diagnosed within the first 24 h of PICU admission, 25% had severe sepsis and 48% had septic shock. In that study, the patients were classified according to the worst sepsis severity stage. Lower prevalence rates were observed in a single center retrospective study conducted in Paquistan by Khan et al. ⁽¹⁵⁾, where 17.3% children from 1 month to 14 years had sepsis of any severity stage. In a large prospective study conducted among hospitalized children in China, Wang et al. ⁽¹⁶⁾ found a 5.5% sepsis prevalence among 27,836 children admitted to the 11 participating hospitals. From the 1,530 sepsis patients, the authors reported that 7.9% had severe sepsis and 2.1% had septic shock and estimated an incidence of 181 cases per 100,000 children per year.

Recently, the SPROUT study ⁽¹⁸⁾, a cross-sectional study conducted in 26 countries in four continents, investigated the prevalence, therapies, and outcomes of severe sepsis in 6,925 children admitted to 128 PICU. The SPROUT study showed 8.2% prevalence of severe sepsis among children, similar to the rates reported for adults. There was a wide variation in sepsis prevalence among the different continents, ranging from 6.2% in Europe to 23.1% in Africa (P <0.001). The results of this study suggest that, in a typical 16-bed PICU, at least one critically ill child is likely being treated for severe sepsis at any given time.

Several aspects influence the prevalence of sepsis, which may partially explain the disparities in the results of the epidemiologic studies. Such aspects include the variations in clinical practice, the subject selection, the diagnostic criteria, the study design, the season and source of data collection, and the clinic and demographic characteristics of the population under study. The difficulties in comparing epidemiologic studies on pediatric sepsis were demonstrated by Weiss et al. ⁽¹⁹⁾ in a study that evaluated the extent of agreement among three distinct diagnostic criteria for pediatric sepsis in a single PICU. In that study, 9.2% children fulfilled the diagnostic criteria for severe sepsis according to at least one definition: 5.2% according to the IPSCC definition, 5.6% according to clinical criteria, and 6% when the ICD-9-MC criteria modified by Angus et al. ⁽²⁰⁾ (combined ICD-9-CM code: code for infection plus codes for organ dysfunction) were used. They also observed that nearly one-third of the children who had sepsis according to the clinical criteria would not be identified using the IPSCC definition or the ICD-9-MC code only, and that only 2.5% of the patients were identified according to all the three diagnostic criteria. Similar results were found in two other studies that used the ICD-9-CM code specific for severe sepsis (995.92) and septic shock (785.52) and the combined ICD-9-CM code to report the prevalence and outcomes of pediatric sepsis in a same population. Balamuth et al. ⁽²¹⁾ observed a 7-fold difference in the prevalence of sepsis depending on the diagnostic criteria used, with 3.1% prevalence for all hospitalized children using the combined ICD-9-CM code and 0.45% according to the ICD-9-CM code for sepsis. In the study by Ruth et al. ⁽²²⁾, sepsis prevalence in the PICU was 6.2% and 3.1% using the combined ICD-9-CM and the ICD-9-CM specific for severe sepsis/septic shock, respectively.

Despite the disparities in the published results, the available data evidence the high frequency of sepsis in the pediatric population and suggest that sepsis represents a public health problem, not only in developing but also in developed countries. Ten years after the publication of the IPSCC criteria, however, reliable estimates on the frequency of pediatric sepsis are still lacking. Studies suggest an increase in the number of sepsis cases along the last two decades—which may relate, among other factors, to the increased survival of preterm and low birth-weight infants and children with severe chronic conditions—and that children living in low-income countries represent a vulnerable population for sepsis ^{(9, 11, 18, 21, 22)}.

Sepsis-related mortality in children

Despite several initiatives to improve the diagnosis and early treatment of pediatric sepsis, such as the ACCM/PALS Pediatric Sepsis Treatment Guidelines ^{(23, 24)}, the Global Pediatric Sepsis Initiative ⁽²⁵⁾, and the Surviving Sepsis Campaign ⁽²⁶⁾, the mortality resulting from pediatric sepsis remains high. The impact of sepsis on childhood mortality is even more unsettling in low-income countries, where low immunization rates and poor sanitary conditions are associated with higher prevalence of infectious diseases.

According to the World Health Organization, in 2013 nearly 6.3 million deaths occurred in children under 5 years old, most of them taking place in developing countries. Over 50% of the deaths were related to severe infectious diseases such as pneumonia, diarrhea, and malaria, where the term “severe” is used to identify conditions accompanied by signs of hypoperfusion, such as acidosis and hypotension, which are the hallmarks of sepsis ⁽²⁷⁾. Sepsis, therefore, may be considered an important cause of death in children, mainly in low-income areas.

Aiming to reduce the mortality of pediatric sepsis, in 2007 the World Federation of Pediatric Intensive and Critical Care Societies (WFPICCS) launched the Global Pediatric Sepsis Initiative ⁽²⁵⁾. The rationale of this program was based on studies that demonstrated that simple interventions, such as immunization programs, early antibiotics administration, fluid resuscitation, and, in selected cases, the use of inotropic drugs, may significantly reduce sepsis-related mortality. The initiative also acknowledges the disparities in the availability of healthcare resources among different regions, and suggests specific recommendations for the therapeutic approach of pediatric sepsis based on the socio-economic characteristics and the accessibility to healthcare resources. According to the WFPICCS, differently from what is observed for prevalence, mortality resulting from sepsis is not related to socio-economic features, but to the adherence to the guidelines for treatment of severe sepsis and septic shock in children, which is still worryingly low in both developed and developing countries ⁽²⁸⁾.

Along the last two decades, sepsis-related mortality has shown a significant decrease in developed countries, where 5% to 20% mortality rates have been registered ^{(9–12, 21, 22)}. Despite improvements in the outcomes, the rise in the pediatric sepsis incidence has resulted in the increase of the total number of sepsis-related deaths in children ⁽⁹⁾. Similar improvements in the outcomes of pediatric sepsis, however, have not been observed in developing countries, where mortality rates as high as 40% have been reported ^{(14–16, 18)}.

The high sepsis mortality observed in low-income countries may be related to the late recognition and management of sepsis in children. Wang et al. ⁽¹⁶⁾ reported 34.6% mortality for severe sepsis and septic shock among Chinese children, with nearly two-thirds of the deaths occurring within the first 72 h of hospital admission. In that study, late hospital admission and delays in the diagnosis and treatment were possibly related to the high mortality. Similarly, Jaramillo-Bustamante et al. ⁽¹⁴⁾ reported that in their study nearly 50% of the children presented with septic shock and 40% had Multiple Organs Dysfunction Syndrome on PICU admission which, according to the authors, determined a high mortality (34%) due to septic shock. In the SPROUT study ⁽¹⁸⁾ overall hospital mortality resulting from sepsis was 25%. A wide variation in mortality rates was observed among different areas, but no differences in mortality were observed between developed and developing regions.

Taken together, these data suggest that sepsis-related mortality most strongly correlates with the precocity on diagnosis and treatment and with the adherence to sepsis treatment guidelines than with the use of advanced technological resources. Some authors suggest that educational rather than economical aspects are associated with sepsis outcomes ^{(29, 30)}.

A summary of the sepsis-related frequency and mortality reported on the epidemiologic studies published from 2005 to 2015 is shown in Table 1.

Table 1: Epidemiologic studies of pediatric sepsis

Pediatric sepsis: what is the problem?

The analysis of the medical literature published in the last decade indicates three main obstacles to a comprehensive understanding of the epidemiology and improvement in the outcomes of pediatric sepsis. The first hindrance relates to the limitations to the use of the IPSCC definition at the bedside, particularly in low-resources settings. Second, the little awareness of the subject by the general public, with consequent unrecognition of the severity of the symptoms and late search for medical assistance, and by the pediatricians, resulting in delays in the diagnosis and treatment of this life-threatening condition. Finally, the low adherence to the published guidelines for the diagnosis and treatment of pediatric sepsis, which is well recognized as a problem associated with worse outcomes.

Studies conducted in children have shown that, when compared with clinical diagnosis, both the administrative criteria (ICD-9-CM) and the 2005 IPSCC definitions underdiagnose sepsis cases ^{(19, 21, 22, 31)}. The limitations to the use of the IPSCC definition have been recent subject of debate ⁽³²⁾. Ideally, sepsis definition should be a sensitive tool, simple to be memorized, and easily applicable at the bedside. It should also take into account that sepsis is a complex and dynamic disease that may present with multiple, non-specific signs, and symptoms that evolve and change in a short time course. Disease definitions are, in the other hand, static and categorical. The classification of the sepsis patient into one single category may, therefore, be inaccurate. Additional difficulties in sepsis classification apply to childhood, since the predisposing factors, the physiologic response to infection, the frequency and nature of comorbidities, the infectious sources, and causative organisms exhibit large diversity among the different pediatric age groups.

Another limitation of the IPSCC definition relates to the complexity of the organ dysfunction criteria, which may hinder its use at the bedside and pose additional difficulties in drawing the subtle and artificial line that discriminates the different sepsis stages. In addition, differently from the adult population, normotensive septic shock is frequent in children, making it possible that in some cases, “severe sepsis” and “septic shock” represent a similar disease severity. Finally, the use of the core temperature and laboratory data may difficult its application in settings with limited resources.

Besides the pitfalls related to the definition, the lack of knowledge of the general population and healthcare professionals about the “sepsis problem” represents an additional barrier to improvements in sepsis outcomes. In a survey conducted with more than 6,000 subjects in Europe and in the United States, Rubulotta et al. ⁽³³⁾ observed that over 50% of the interviewees ignored the term “sepsis,” as well as its importance as a life-threatening condition. Similar results have been reported in studies conducted in Singapore and Brazil ^{(34, 35)}. The little awareness by the general public of the initial symptoms and the importance of the early treatment may contribute to delays in searching for medical attention and, consequently, worse outcomes. Also, the unrecognition of the early stages of sepsis by healthcare professionals and the unfamiliarity with the sepsis clinical practice treatment guidelines have been recognized as obstacles to the early diagnosis and aggressive sepsis treatment ^{(36, 37)}. The lack of resources and technical skills by physicians may also contribute to the low compliance with the treatment guidelines in developing countries ⁽³⁸⁾. Continuous medical education and specific training have proved effective in increasing the adherence to the treatment guidelines and to reduce sepsis-related mortality in children and adults ^{(29, 30)}.

Taken as a whole, the studies analyzed in this review suggest high prevalence and mortality of pediatric sepsis and identify opportunities to improvements in the outcomes. However, the narrative, non-systematic approach of this review and the use of a single-search database represent limitations of this analysis and must be kept in mind when interpreting the results presented herein.

CONCLUSION

The current available data demonstrate that sepsis is a frequent disease in childhood and represents a major cause of PICU admission and death. Accurate estimates of the frequency and mortality of pediatric sepsis among different geographic areas are still necessary. Although the reduction in sepsis burden represents a global challenge, the problem is even more unsettling in developing countries, where sepsis frequency and mortality are higher. Poor outcomes, however, do not seem to be related to the limitation of resources but to the delays in the recognition and early treatment of sepsis. Educational efforts aiming to increase the awareness on sepsis by the general public and the adherence to the treatment guidelines by healthcare providers may result in significant improvements in sepsis survival. The global attention to pediatric sepsis, however, can only be achieved with the standardization of the definitions and the use of simple and sensitive diagnostic criteria that incorporate the differences in the necessities among different settings and the availability of local resources.