Evaluation of Phoenix Sepsis Score Criteria: Exploratory Analysis of Characteristics and Outcomes in an Emergency Transport PICU Cohort From the United Kingdom, 2014-2016.

OBJECTIVES

To assess characteristics and outcomes of children with suspected or confirmed infection requiring emergency transport and PICU admission and to explore the association between the 2024 Phoenix Sepsis Score (PSS) criteria and mortality.

DESIGN

Retrospective analysis of curated data from a 2014-2016 multicenter cohort study.

SETTING

PICU admission following emergency transport in South East England, United Kingdom, from April 2014 to December 2016.

PATIENTS

Children 0-16 years old ( n = 663) of whom 444 (67%) had suspected or confirmed infection.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

The PSS was calculated as a sum of four individual organ subscores (respiratory, cardiovascular, neurological, and coagulation) using the worst values during transport (i.e., from referral until the time of PICU admission). A score cutoff of greater than or equal to 2 points was used to define sepsis; and septic shock was defined as sepsis plus 1 or more cardiovascular subscore points. Sepsis occurred in 260 of 444 children (58.6%) with suspected or confirmed infection, with septic shock occurring in 177 of 260 (68.1%) of those with sepsis. A PSS score greater than or equal to 2 points occurred in 37 of 67 bronchiolitis cases, 19 of 35 meningoencephalitis cases, 30 of 47 pneumonia/empyema cases, 38 of 46 septic/toxic shock cases, nine of 15 severe sepsis cases, and 58 of 118 definite viral infections. Overall, 14 of 444 children died (3.2%). There were 12 deaths in the 260 children with PSS greater than or equal to 2, and two deaths in the 184 children with PSS less than 2 (4.6% vs. 1.1%; absolute difference, 3.5%; 95% CI, 0.1-6.9%; p = 0.04).

CONCLUSIONS

In 2014-2016, over half of the critically ill children undergoing emergency transport to PICU with presumed or confirmed infection, and meeting retrospectively applied PSS criteria for sepsis, had a range of clinical diagnoses including bronchiolitis, meningoencephalitis, and pneumonia/empyema. Furthermore, the PSS criteria for categorization of sepsis and septic shock were associated with outcome and may be of value in future risk-stratification in clinical trials.