New criteria for pediatric sepsis, based on a novel score that predicts mortality in children with suspected or confirmed infection, perform better than existing organ dysfunction scores and criteria and have the potential to improve clinical care globally, researchers say.
Current pediatric-specific criteria for sepsis were published in 2005, based on expert opinion. In 2016, sepsis was redefined for adults as life-threatening organ dysfunction caused by a dysregulated host response to infection, as opposed to an earlier focus on systemic inflammation. But the paradigm-shifting changes were not extended to children (< 18 years, but not newborns), setting the stage for the new initiative.
The new criteria, and their development and validation, were published in JAMA and presented the same day at the Society of Critical Care Medicine’s 2024 Critical Care Congress in Phoenix, Arizona.
International Consensus
“The new criteria we derived are based on data from electronic health records and analysis of more than 3 million pediatric healthcare encounters from 10 hospitals around the world, including in low-resource settings,” L. Nelson Sanchez-Pinto, MD, MBI, a critical care physician at the Ann and Robert H. Lurie Children’s Hospital of Chicago, told this news organization.
Dr. Sanchez-Pinto co-led the data group of the international expert task force convened by the Society of Critical Care Medicine (SCCM) to develop and validate the criteria, which are based on evidence from an international survey, systematic review and meta-analysis, a newly created organ dysfunction score (Phoenix Sepsis Score), and sites on four continents.
Based on the findings, the task force now suggests that pediatric sepsis be defined by a Phoenix Sepsis Score of at least 2 points in children with suspected infection, which indicates potentially life-threatening dysfunction of the respiratory, cardiovascular, coagulation, and/or neurological systems. Septic shock is defined as sepsis with at least 1 cardiovascular point in the score.
Disparities Across Settings
To derive and validate the new criteria across differently resourced settings, the researchers conducted a multicenter, international, retrospective cohort study involving 10 health systems in the United States, Colombia, Bangladesh, China, and Kenya, 3 of which were used as external validation sites.
Data were collected from pediatric emergency and inpatient encounters from 2010 to 2019. The development set comprised 3,049,699 children, and the external validation set included 581,317.
Stacked regression models to predict mortality in children with suspected infection were derived and validated using the best-performing organ dysfunction subscores from eight existing scores.
The final model was then translated into the integer-based Phoenix Sepsis Score and used to establish binary criteria for sepsis and septic shock.
Among 172,984 children with suspected infection in the first 24 hours (development set; 1.2% mortality), a four-organ-system model performed best. The Phoenix Sepsis Score — the integer version of the model — had areas under the precision recall curve of 0.23 to 0.38, and areas under the receiver operating characteristic curve of 0.71 to 0.92 to predict mortality in the validation sets.
A Phoenix Sepsis Score of 2 points or higher in children with suspected infection as criteria for sepsis, plus 1 or more cardiovascular points as criteria for septic shock, resulted in a higher positive predictive value and higher or similar sensitivity compared with the 2005 International Pediatric Sepsis Consensus Conference criteria across differently resourced settings.
Specifically, children with a Phoenix Sepsis Score of at least 2 points had in-hospital mortality of 7.1% in higher-resource settings and 28.5% in lower-resource settings — more than 8 times that of children with suspected infection not meeting these criteria.
Mortality also was higher in children who had organ dysfunction in at least one of four organ systems — respiratory, cardiovascular, coagulation, and/or neurological — that was not the primary site of infection.
Children with septic shock, indicated by at least 1 cardiovascular point in the Phoenix Sepsis Score, had severe hypotension for age, blood lactate exceeding 5 mmol/L, or need for vasoactive medication. These children had an in-hospital mortality rate of 10.8% in higher-resource settings and 33.5% in lower-resource settings.
A Better Score
Given the findings, the task force recommends that “the former criteria based on systemic inflammatory response syndrome should not be used to diagnose sepsis in children [and] the former term severe sepsis should no longer be used because sepsis is life-threatening organ dysfunction associated with infection and is thus indicative of a severe disease state.”
The task force cautions that although the four organs in the Phoenix Sepsis Score are most commonly involved in sepsis, “this does not diminish the crucial importance of the assessment and management of other organ dysfunction.”
Furthermore, they emphasize that the Phoenix score was designed to identify sepsis in children, not to screen children at risk for developing sepsis or early identification of children with suspected sepsis.
Additional Considerations
In related editorials, commentators noted some caveats and concerns with regard to the study design and the new criteria.
Roberto Jabornisky, MD, PhD, of National University of the Northeast, Corrientes, Argentina, and colleagues pointed out that “all the low-resource validation sites were institutions with electronic health records and most had PICUs [pediatric intensive care units], which does not adequately reflect conditions in most low-resource settings. These factors introduce a distinct bias favoring a ‘PICU-based consensus,’ potentially limiting the generalizability and adoption of the new criteria by health care practitioners in non-PICU and nonhospital settings responsible for recognizing and managing children with sepsis.” The editorialists called for additional prospective validation in differently resourced settings, especially those with the highest disease burdens.
“Until then,” they wrote, “it is essential to refrain from considering these criteria as an inflexible directive governing medical interventions for pediatric sepsis. No definition can fully substitute for the clinical judgment of an experienced, vigilant clinician caring for an unwell child.”
Erin F. Carlton, MD, MSc of the University of Michigan, Ann Arbor, and colleagues added in a separate editorial, “The Phoenix criteria identify a sicker subset of patients than prior SIRS [systemic inflammatory response syndrome]-based criteria. Some may worry this higher threshold could delay management of patients not meeting sepsis criteria. Just as patients with chest pain and a troponin leak warrant monitoring and treatment (but are not prioritized for immediate heart catheterization), patients with infection need monitoring and treatment. Improvements in care should thus be judged not only by improved outcomes among patients with sepsis but also by decreased progression to sepsis among patients with infection.”
The International Consensus Criteria paper was supported by the Society of Critical Care Medicine and a grant from the Eunice Kennedy Shriver National Institute of Child Health and Human Development to Tellen C. Bennett, MD, MS, and Nelson Sanchez-Pinto, MD. Data for the Kenya site were collected with support of the Wellcome Trust to the Kenya Major Overseas Programme. Dr. Jabornisky reported no conflicts of interest. Dr. Carlton reported serving on the Pediatric Surviving Sepsis Campaign Guideline committee and receiving grant support from the NIH.
Source link