Piloting a registry for paediatric sepsis: The PoRPoiSe study.

AIM: To develop a model for a paediatric sepsis registry for use in emergency care settings. A regional study, in the UK, was undertaken to identify the most basic registry components which are desirable and feasible using the concept of a minimum viable product. METHODS: Two-round survey of clinicians using a modified Delphi methodology in conjunction with a regional data collection project in three paediatric emergency departments across London. RESULTS: The survey identified 34 desirable information items to be included in a registry. Fifteen of 34 items are currently feasible from our experience of data collection. CONCLUSION: The development of a multi-centre paediatric sepsis registry sepsis may have several benefits but is currently extremely limited primarily because of technological fragmentation within our Health Service. Our findings have important implications for researchers wishing to plan sepsis surveillance programmes, locally and internationally.

Evaluation of a novel approach to recognising community-acquired paediatric sepsis at ED triage by combining an electronic screening algorithm with clinician assessment.

OBJECTIVE: We report the utilisation and impact of a novel triage-based electronic screening tool (eST) combined with clinical assessment to recognise sepsis in paediatric ED. METHODS: An electronic sepsis screening tool was implemented in the paediatric EDs of two large UK secondary care hospitals between June 2018 and January 2019. Patients eligible for screening were children < 16 years of ages excluding those with minor injuries or who were brought directly to resuscitation. Subsequently, a retrospective evaluation was performed to determine the performance of the tool alone and in combination with clinical assessment after triage, to identify septic patients, using sensitivity, specificity, positive, negative predictive values (PPV and NPV) and likelihood ratios. RESULTS: 19 912 children were triaged during the study period, of whom 90 (0.45%) were classified as having sepsis. 99% of all eligible patients were screened. The eST alerted for 2651 (13.3%) patients. After immediate physician assessment, 151 were treated for sepsis in the ED, of whom 70 had a final diagnosis of sepsis. Eight patients who were not thought to be septic returned with sepsis within 24 hours. The eST showed a sensitivity of 86.7% (95% CI 77.5% to 92.6%), specificity 87.0% (95% CI 86.5% to 87.5%), PPV 2.94% (95% CI 2.35% to 3.68%), NPV 99.9% (95% CI 99.8% to 99.9%) which improved with combined clinical assessment to a sensitivity of 90.0% (95% CI 81.4% to 95.0%), specificity 99.4 (95% CI 99.3% to 99.5%), PPV 42.0 (95% CI 35.0% to 49.3%) and NPV 99.9% (95% CI 99.9% to 99.9%). CONCLUSION: Utilisation of a novel triage-based eST allowed sepsis screening in over 99% of eligible patients. The screening tool showed good accuracy to recognise sepsis at triage in the ED, which was augmented further by combining it with clinician assessment. The screening tool requires further refinement through multicentre evaluation to avoid missing sepsis cases.