(620) Prevalence of Sepsis by Phoenix Criteria in Children With COVID-19: An SCCM: VIRUS Registry Report

Introduction: The definition of pediatric sepsis was updated in January 2024 as Phoenix sepsis score PSS. This study’s specific aim was to calculate the prevalence of sepsis in children hospitalized with COVID-19 as defined by the PSS and to validate the PSS with respect to outcomes [mortality and hospital length of stay (HLOS)].

Methods: Study included children enrolled in the SCCM VIRUS: COVID-19 registry from 03/2020 to 04/2024. The performance of the PSS in discriminating mortality was assessed using the receiver operating characteristic (ROC) and the precision recall curve (PRC). Cross-validation was performed using leave-one-out cross-validation (LOOCV). Mixed linear effect models were used to examine the relationship between the PSS and HLOS.

Results: 1731 patients (58 hospitals;167 [9.6%] non-US patients) met the study criteria ([MIS-C]: 479 [27.7%], ICU admission: 809 [46.7%]). Of these, 159 (9.2%) met the criteria for sepsis, and 167 (9.7%) met the criteria for septic shock. The mortality was 1.4% (25/1731), with significant differences among the three groups (non-septic [10/1405, 0.7%], sepsis [6/159, 3.8%], septic shock [9/167, 5.4%], p< 0.001). After adjusting for confounding variables, both the sepsis group and the septic shock group were associated with significantly longer HLOS compared to the non-septic group. One unit increase in the numeric PSS led to a 70% increase in the risk of mortality (odds ratio 1.70, p< 0.001). There were significant differences in odds of mortality between the sepsis and non-septic groups as well as the septic shock and non-septic groups, but there was no significant difference in odds of mortality between the septic shock and sepsis groups. The area under the curve in the ROC model was 0.9, and the area under the PRC was 0.27. The threshold of ≥2 for detection of mortality had a sensitivity of 0.63, specificity of 0.82, positive predictive value of 0.05.

Conclusions: The PSS was able to identify most of the relatively small cohort of children with COVID-19 who developed sepsis and septic shock. Patients meeting the criteria for sepsis had higher mortality and longer lengths of stay. The Phoenix sepsis score had a good AUROC to predict mortality, but low precision (high false positive rates).