Performance of influenza-specific triage tools in an H1N1-positive cohort: P/F ratio better predicts the need for mechanical ventilation and critical care admission

Background

Pandemic influenza presents a major threat to global health and socioeconomic well-being. Future demand for critical care may outstrip supply and force clinicians to triage patients for admission. We evaluated the Simple Triage Scoring System (STSS), Ontario Health Plan for an Influenza Epidemic (OHPIP) and PaO2/FIO2 (P/F) ratio to determine utility in predicting need for mechanical ventilation.

Methods

We conducted a retrospective case note review of patients admitted to two centres, Royal Liverpool University Hospital and Countess of Chester Hospital, during the UK influenza pandemic of 2010–11. Demand for critical care during this period forced hospitals in Cheshire and Merseyside to implement escalation policies and increase capacity. Inclusion criteria were polymerase chain reaction–confirmed H1N1 influenza and age >18 years. Exclusion criteria were no evidence of treatment for influenza, patient not admitted to hospital or the inability to locate case notes.

Results

One hundred and one patients were included, 29 were admitted to critical care and 23 required mechanical ventilation. The P/F ratio predicted the need for mechanical ventilation with a receiver operating characteristic area under the curve (ROC AUC) of 0.885 (CI 0.817–0.952). Predictive ability was not reduced when the P/F ratio had to be estimated using the Pandharipande tool. The STSS score predicted the need for mechanical ventilation [ROC AUC 0.798 (CI 0.704–0.891)]. The reverse triage component of the OHPIP tool was a poor predictor of patient outcome.

Conclusions

The P/F ratio was a better predictor of need for mechanical ventilation than STSS. The P/F ratio is a simple and accepted determinant of hypoxaemia and should be used if secondary triaging becomes necessary during future influenza pandemics.