Randolph, Adrienne G. MD, MSc
The Pediatric Acute Lung Injury and Sepsis Investigator’s (PALISI) Network is comprised of investigators at more than 50 pediatric intensive care units (ICU) across North America with links to adult and pediatric research networks in other countries (http://pedsccm.wustl.edu/research/palisi/palisi.html). Founded in 1999, PALISI has made considerable progress in studying the epidemiology and treatment of acute lung injury, but investigation in the area of sepsis has been more challenging. Although sepsis is one of the top causes of and contributors to mortality in the pediatric ICU, there has never been a well-defined, prospective epidemiologic study of sepsis in infants and children, except in the area of nosocomial infection surveillance. The lack of understanding of the disease process has hindered interventions to improve outcomes in critically ill neonates, infants, and children.
The mission of the International Sepsis Forum (ISF) is to improve the understanding and clinical management of patients with severe sepsis (www.sepsisforum.org). The ISF is headed by a steering committee of international experts and opinion leaders. One of the major goals of the ISF is to develop an international consensus on the latest understanding of key scientific and clinical issues in the area of severe sepsis and, in particular, septic shock. The International Sepsis Forum on Sepsis in Infants and Children was a joint initiative of the PALISI Network and ISF. The goals of the 3-day consensus meeting were as follows:
To review and make recommendations on current definitions of the systemic inflammatory response syndrome (SIRS), sepsis, severe sepsis, septic shock, and organ failure for neonates, infants, and children.
To carefully define the most common infections seen in critically ill neonates, infants, and children from three perspectives: a) identifying infection early to enroll children in sepsis trials; b) identifying infections for surveillance and epidemiologic studies; and c) identifying infections for diagnosis and optimal therapy.
To review some of the most common conditions predisposing infants and children to sepsis including a compromised immune system, age, gender, burns and other traumatic injuries, and genetic predisposition.
To debate whether predisposition and type of infection should influence clinical trials evaluating agents to prevent or treat sepsis in infants and children.
To discuss non-mortality outcome measures in clinical trials of therapies to treat or prevent sepsis in children.
An extensive process of expert recommendations and thorough PubMed searching took place to choose the panel of international experts invited to the meeting, with the goal of including expertise in neonatal, pediatric, and adult sepsis. The international panel came from the United States, United Kingdom, Canada, Australia, The Netherlands, Switzerland, South Africa, and Malaysia.
The conference started out with an introduction to the scope and epidemiology of pediatric sepsis by Drs. Scott Watson and Joe Carcillo. They reported that in the United States alone, there were an estimated 4,400 pediatric deaths from sepsis every year, with hospital costs of $1.7 billion. They found severe sepsis to be the fourth leading cause of hospital admissions in the United States. They described in detail the characteristics of the population of children with severe sepsis in the hospital. The worldwide burden of sepsis in children is even more tremendous, and they reviewed these disturbing numbers. They encouraged ISF to develop definitions that identify sepsis in its earliest stages, when simple interventions like antibiotics and intravenous or enteral fluids can save the greatest number of children worldwide.
When you ask pediatric residents if a 5-yr-old with pneumonia, high fever, elevated white blood count, tachypnea, and tachycardia is septic, they will usually ask you whether the blood pressure is stable. If you say yes, they will determine that the patient is not yet septic. This conclusion would be erroneous according to the current definitions of sepsis in infants and children in which SIRS plus infection equals sepsis. To most pediatric practitioners, however, the term sepsis in infants and children is equated with septic shock. Calling a patient with straightforward pneumonia “septic” or a patient with severe pneumonia who progresses to respiratory failure requiring mechanical ventilator support “severely septic” may not add much to their understanding of the disease process. Drs. Brilli and Goldstein presented the current definitions of pediatric SIRS and sepsis, and a vigorous discussion then ensued regarding the utility of these definitions.
A large portion of the meeting was devoted to defining infections in critically ill children. The experts were asked to define infections from the three perspectives listed above. Diagnoses were also to be categorized according to certainty, using the categories of definite, probable, and possible. The following types of infections were reviewed: pneumonia, meningitis, and other central nervous system infections, central venous catheter-related infections, urinary tract infections, intraabdominal infections, surgical wound infections, and other skin and soft tissue infections, bloodstream infections, and meningococcemia. Working groups were held to review current definitions and to make revisions specific to neonates, infants, and children when indicated. This information was then presented to the entire group for discussion. After the conference, manuscripts with the proposed definitions were sent to all members of the group for comment before arriving at the final definitions. We realize that these definitions are only a beginning. Our goal is for epidemiologic studies to be designed to test these definitions with feedback and ongoing refinement.
The next portion of the conference was devoted to predisposition to sepsis in neonates, infants, and children. Dr. Steven Opal presented the PIRO (predisposition, infection, response, and organ dysfunction) concept of understanding sepsis that is being continually refined by adult sepsis experts. Dr. Michael Quasney then presented an overview of what is known about genetic predisposition to sepsis, and Dr. Sally Vitali reviewed the limitations of these studies. A large portion of the predisposition section was devoted to the immunocompromised host. Dr. Greg Preibe’s editorial describes this in detail. Cancer, acquired immunodeficiencies, HIV, and congenital immunodeficiencies were reviewed regarding how they predispose children to sepsis. This section of the proceedings provides fascinating insights into those children we care for in the ICU who have the highest mortality from sepsis. Other factors that predispose to sepsis, such as trauma and burns, were then reviewed by Drs. Jeffrey Upperman and Robert Sheridan, respectively. Dr. Sheridan was also able to posit definitions for infectious processes that occur exclusively in burn patients who are included with the surgical wound infection portion in the skin and soft tissue infection definitions article in these proceedings.
The final portion of the meeting was devoted to issues relevant to the design of sepsis studies. Dr. Joachim Fischer presented a fascinating overview of issues hindering a physician’s ability to diagnose sepsis in children. Drs. Jacques Lacroix and Jacques Cotting reviewed the state of the art for severity of illness and organ dysfunction scoring in children. Dr. William Tarnow-Mordi described the use of severity of illness and organ dysfunction scoring in neonatal clinical trials. A general discussion was then held on the role that these scores should have in clinical trials in neonates, infants, and children. Dr. Tarnow-Mordi then reviewed what has been learned from designing sepsis studies in neonatal intensive care. He discussed the need for very large, simple, randomized, controlled trials and the benefit of preplanned prospective meta-analysis. Mortality in pediatric sepsis in the developed world is markedly lower than for adult sepsis. This makes mortality infeasible as an outcome measure for most clinical trials because the largest trials in pediatric critical care to date enrolled fewer than 350 children, and even then, obtaining that number of subjects was extremely challenging and took years. Drs. Martha Curley and Jerry Zimmerman reviewed in great depth the pros and cons of non-mortality outcome measures for clinical trials. Finally, Dr. Joe Carcillo concluded the meeting by facilitating an invigorating discussion on what the next ISF on sepsis in infants and children topic should be. His article on future directions addresses the global burden of sepsis and provides fascinating insights into sepsis in the developing world.
This consensus conference was funded through a generous donation from the Mannion Family Fund—Center for the Critically Ill Child, Division of Critical Care Medicine at Children’s Hospital Boston, with additional support from the PALISI Network and ISF. This funding has allowed us to publish the full proceedings of the conference and to allow free access to these articles from anywhere in the world through internet links through Pediatric Critical Care Medicine or the ISF Web sites. Our hope is that investigators and clinicians will put these definitions and ideas through rigorous testing so that the further revision, which we know is needed, is performed expeditiously.
Adrienne G. Randolph, MD, MSc
Children’s Hospital of Boston
©2005The Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies