A novel oxygenation distension index (ODI): a driving pressure–based metric compared with the oxygenation index in pediatric PARDS
摘要
The oxygenation index (OI) is widely used to assess disease severity in pediatric ARDS; however, it relies on mean airway pressure (Pmean), which does not reflect lung-distending forces relevant to ventilator-induced lung injury. Driving pressure (DP) is a determinant of lung strain. We developed and evaluated the oxygenation distension index (ODI) by substituting DP for Pmean in the OI formula. This observational study included children aged 1 month to 18 years with PARDS requiring ≥ 24 h of invasive mechanical ventilation in six centers. Lung ultrasound scores (LUS) and respiratory mechanics, including transpulmonary pressure (PL), transpulmonary driving pressure (DPL), mechanical power (MP), and transpulmonary mechanical power (MPL), were assessed 4–24 h after diagnosis according to PALICC-2 recommendations. Associations of OI and ODI with lung injury markers and clinical outcomes were analyzed. Among 56 patients, 41 (73%) had mild-to-moderate and 15 (27%) had severe PARDS. ODI demonstrated stronger correlations than OI with lung injury markers, including LUS (r = 0.818 vs. 0.501), PL (r = 0.779 vs. 0.697), DPL (r = 0.745 vs. 0.671), MP normalized to predicted body weight (r = 0.518 vs. 0.438), and MPL normalized to predicted body weight (r = 0.634 vs. 0.628). For 30-day mortality prediction, ODI showed higher discriminative performance than OI (AUC 0.740 [95% CI, 0.598–0.866] vs. 0.685 [95% CI, 0.535–0.835]; DeLong p = 0.040).
Conclusions: By incorporating driving pressure, ODI more closely reflects lung injury severity than the conventional oxygenation index and improves discrimination for mortality in PARDS.