Prognostic Accuracy of the Phoenix Sepsis Criteria for Mortality in Children With Suspected Infection: A Systematic Review and Meta-Analysis.

Summary

This systematic review and meta-analysis included 15 studies comprising 16 cohorts and 2,601,038 pediatric encounters. The Phoenix Sepsis Criteria achieved a pooled sensitivity of 0.77, specificity of 0.73, diagnostic odds ratio of 9.2, and area under the curve of 0.81, outperforming the IPSCC criteria, although specificity was substantially lower in intensive care unit cohorts than in emergency-department cohorts.

Key Findings

• Fifteen studies and 16 cohorts totaling 2,601,038 pediatric encounters were included.
• The Phoenix Sepsis Criteria had pooled sensitivity of 0.77, specificity of 0.73, and AUC of 0.81 for mortality.
• The Phoenix Sepsis Criteria outperformed IPSCC criteria, but specificity was lower in ICU cohorts than in emergency-department cohorts.

Clinical Implications

The Phoenix Sepsis Criteria can support pediatric mortality risk stratification and clinical decision-support development, but performance should be calibrated by care setting. They should not be used alone as a universal early sepsis screening tool, particularly in intensive care units or lower-resource settings.

Limitations

• The included studies were heterogeneous in design, setting, and implementation of the criteria.
• The pooled estimates describe prognostic discrimination and do not establish that use of the criteria improves patient outcomes.

Future Directions

Prospective implementation studies should evaluate whether Phoenix-based risk stratification improves treatment timeliness, resource allocation, and outcomes across emergency departments, wards, intensive care units, and lower-resource settings.