SCCM Task Force Develops New Criteria to Identify Pediatric Sepsis

Children make up nearly half of patients with sepsis. Those younger than five years are at greatest risk, especially if they live in areas with few resources.¹ Sepsis is a global health problem, and researchers and clinicians have called for updated definitions of sepsis in pediatric patients. The Society of Critical Care Medicine (SCCM) responded by convening the SCCM Pediatric Sepsis Definition Task Force. The task force used an extensive approach to developing the most comprehensive and up-to-date criteria, which were published in JAMA in February 2024.² The article details the new Phoenix criteria for sepsis in children. Members of the task force conducted the session, “Announcement of the Novel Phoenix Pediatric Sepsis Criteria,” at SCCM’s 2024 Critical Care Congress. Speakers outlined the new criteria and the methods used to develop them.

The new criteria were developed to replace 2005 criteria written by members of the International Consensus Conference on Pediatric Sepsis. The 2005 criteria were based on expert opinion but lacked data-driven evidence as well as input from clinicians working in lower-resource countries.³ The criteria characterized sepsis as infection in the presence of systemic inflammatory response syndrome (SIRS) and defined severe sepsis as sepsis with cardiovascular or respiratory organ dysfunction or dysfunction of at least two other organ systems.

The limitations of these criteria became increasingly controversial, and they were inconsistently applied in clinical practice and research. The SCCM Pediatric Sepsis Definition Task Force set out to perform a systematic review and global survey, followed by a comprehensive data-driven derivation and validation process, to revise criteria for sepsis in children.

Focusing on identifying children at the highest risk for life-threatening infection, the SCCM task force developed a new assessment system for organ dysfunction called the Phoenix Sepsis Score. They determined that children with a Phoenix Sepsis Score of at least 2 points (indicating potentially life-threatening dysfunction of the respiratory, cardiovascular, coagulation and/or neurologic systems) had in-hospital mortality of 7.1% in higher-resource settings and 28.5% in lower-resource settings, which was more than eight times that of children with suspected infection who did not meet those criteria. The new criteria do not include the term SIRS (which has poor predictive value) or severe sepsis (which is redundant under the new criteria). The new criteria apply to children younger than 18 years but not to newborns or those born before 37 weeks.

A Real-World Approach to Defining Pediatric Sepsis
In 2016, the Third International Consensus Definitions for Sepsis and Septic Shock drew from data on more than 150,000 adults with suspected sepsis to define sepsis as life-threatening organ dysfunction caused by dysregulated host response to infection, but it did not include children.⁴ Sepsis in children varies from that in adults, including differences in vital signs based on age and developmental age-dependent immune function, as well as differences in comorbidities, epidemiology, and outcomes. There was clearly a need for new sepsis definitions and criteria in children based on updated, validated data.

The new definitions and criteria are based on objective, data-driven, and standardized information that identify sepsis and septic shock in children applicable in most settings worldwide. The SCCM task force comprises 35 members from 12 countries representing various specialties, including critical care, emergency medicine, pediatrics, and infectious disease. They used a three-pronged approach to develop the new criteria:

• A global survey of a diverse group of 2835 clinicians, which revealed that most pediatric clinicians used the term sepsis when referring to infection with life-threatening organ dysfunction rather than SIRS
• A systematic review and meta-analysis of data to synthesize the evidence and identify children who are at high risk of death from sepsis
• A data-driven derivation and validation study based on more than three million electronic health record encounters from 10 sites on four continents

“We recognized it would be highly valuable to better understand common practice around the world, so we asked clinicians how they cared for children with sepsis, including what criteria they use and the tools they have available,” said Luregn J. Schlapbach, MD, PhD, cochair of the task force and head of intensive care and neonatology at University Children’s Hospital in Zurich, Switzerland. “We asked them what they thought was most important for future criteria for sepsis, and it was nice to see that the final data approach actually confirmed the clinical perception of the signs of sepsis in children.”

The new criteria are geared toward saving lives by helping clinicians identify children at the highest risk. “The previous criteria focused on SIRS, which basically every child with a bad cold has,” said R. Scott Watson, MD, MPH, cochair of the task force and professor of pediatrics at the University of Washington Seattle Children’s Hospital in Seattle, Washington, USA. “They included many things that did not identify children with life-threatening organ dysfunction. As a result, clinicians often did not follow those criteria because they carried less weight in identifying the children who were at highest risk.”

The Phoenix Sepsis Score assigns points to the following variables. A score of 2 or more points indicates potentially life-threatening sepsis:

• Respiratory: 0-3 points based on various levels of respiratory function and need for support
• Cardiovascular: 0-6 points based on lactate level, age-specific thresholds for arterial hypotension, and need for vasoactive medication
• Coagulation: 0-2 points based on platelet, D-dimer, and fibrinogen levels
• Neurologic: 0-2 points based on Glasgow Coma Scale score and whether the patient’s pupils are reactive or fixed bilaterally

“The data-driven process, led by Tellen Bennett, MD, at Children’s Hospital of Colorado, and L. Nelson Sanchez-Pinto, MD, at Ann and Robert H. Lurie Children’s Hospital of Chicago, included the data collection, harmonization, and machine learning-based analyses that were unprecedented in pediatrics. It revealed that the organs involved with respiratory, cardiovascular, neurologic, and hematologic dysfunction were most associated with mortality,” said Dr. Watson. “Also, the previous sepsis criteria were ICU based. More than 80% of patients with sepsis first come through the emergency department, so these criteria were designed to also be useful for those clinicians.”

The New Criteria Are Valid in Lower-Resource Settings
Task force members set out to ensure that the Phoenix Sepsis Score system would be accurate in lower-resource settings. For example, they determined that restricting the criteria to four organ systems reduces the need for data collection and laboratory testing. They acknowledged that serum lactate testing and mechanical ventilation or vasoactive medications may not be available in lower-resource settings. However, they noted that built-in redundancy in the score allows other score items to be used instead to ensure high accuracy in identifying life-threatening sepsis, even in lower-resource settings.

“We realized early on that developing the perfect criteria that apply in every circumstance is probably not a realistic goal,” said Dr. Schlapbach. “However, the analytic approach did lead to high-performing criteria in terms of specificity and sensitivity that outperformed all previous approaches. These criteria can now inform the development of tools for early recognition of sepsis, which is certainly a need.”

Some members of the task force are planning to perform additional analyses to identify sepsis earlier, which will be crucial for clinicians at the bedside. The task force recommends that the criteria be used to measure the burden of disease and for benchmarking purposes. The criteria should also be used to help develop data-driven screening tools that are reasonably precise and highly sensitive and can be used in a variety of clinics, including those in lower-resource settings. The task force also calls for research on sepsis in infants who are preterm or have perinatal infections, as well as children with sepsis who manifest organ dysfunction in a different site from the original infection.


References

1. Rudd KE, Johnson SC, Agesa KM, et al. Global, regional, and national sepsis incidence and mortality, 1990-2017: analysis for the Global Burden of Disease Study. Lancet. 2020 Jan 18;395(10219):200-211.
2. Schlapbach LJWatson RSSorce LR, et al; Society of Critical Care Medicine Pediatric Sepsis Definition Task Force. International consensus criteria for pediatric sepsis and septic shock. JAMA. 2024 Feb 27;331(8):665-674.
3. Goldstein B, Giroir B, Randolph A; International Consensus Conference on Pediatric Sepsis. International pediatric sepsis consensus conference: definitions for sepsis and organ dysfunction in pediatrics. Pediatr Crit Care Med. 2005 Jan;6(1):2-8.
4. Singer M, Deutschman CS, Seymour CW, et al. The third international consensus definitions for sepsis and septic shock (Sepsis-3). JAMA. 2016 Feb 23;315(8):801-10.