Evaluating hyperoxia and its impact on clinical outcomes in paediatric intensive care: a systematic review and meta-analysis

Abstract Purpose To evaluate the adverse effects of hyperoxia on critically ill paediatric patients by systematically reviewing and conducting a meta-analysis of studies that employ variable definitions of hyperoxia. Methods We searched Ovid MEDLINE and EMBASE (Jan 1990-Dec 2024) for randomised control trials (RCTs) and observational studies comparing hyperoxia to normoxia/non-hyperoxia in Paediatric Intensive Care Unit (PICU) patients. The primary outcome was all-cause mortality at the longest follow-up. Secondary outcomes included incidence and duration of invasive mechanical ventilation, respiratory support, acute kidney injury, organ support, length of stay in PICU and hospital, neurological outcomes, long-term lung function, and overall functional status. Results 17 studies ( n = 43,391 patients) were included, comprising 2 RCTs and 15 observational studies. Hyperoxia (variably defined as PaO 2 > 193–300 mmHg) was associated with increased mortality (OR 2.00; 95% CI 1.42–2.82; I 2 = 89.3%). Excluding three outliers yielded a similar effect (OR 1.77; 95% CI 1.33–2.34; I² = 77.5%). Hyperoxia was also linked to increased reliance on organ support (OR 3.49; 95% CI 1.53–7.96; I² = 81.2%), whilst associations with acute kidney injury (OR 1.09; 95% CI 0.80–1.48; I² = 0%) and neurological sequelae (OR 0.75; 95% CI 0.49–1.14; I² = 0%) were not statistically significant. Conclusion This study demonstrates the deleterious effects of hyperoxia in paediatric intensive care, with increased mortality and organ support requirements. Whilst inconsistent study designs and definitions of hyperoxia preclude recommendations for future clinical practice, these findings support the adoption of conservative oxygenation strategies pending further trials.