A Multicenter Retrospective Study Predicting Early Noninvasive Ventilation Failure in Patients With Acute Hypoxic...

Volume OXygenation (VOX) index has good efficacy in predicting the failure of high-flow nasal cannula therapy. However, its predictive value for treatment failure in patients receiving noninvasive ventilation (NIV) remains uncertain.

Patients who underwent early NIV treatment were grouped based on their 2-h NIV VOX Youden index. The low-risk group consisted of patients with a VOX value > 20.45 (n = 188), while the high-risk group included those with a VOX value ≤ 20.45 (n = 200). Baseline data and arterial blood gas values were collected at 2, 12, and 24 h after NIV initiation.

Compared to the low-risk group, the high-risk group exhibited higher SOFA scores, respiratory rates, and heart rates, along with a lower oxygenation index (P/F) (all p < 0.05). Following NIV treatment, the low-risk group showed a more significant increase in P/F values at 2 h, 12 h, and 24 h after NIV initiation. The low-risk group showed a lower VT and MV (minute ventilation volume) at 2 h, 12 h, and 24 h of NIV (p < 0.05). Moreover, the low-risk group had a lower intubation rate (7.98% vs. 77%, p < 0.05) and mortality rate (4.79% vs. 17.5%, p < 0.05). At 2 h of NIV, the area under the receiver operating characteristic curve for predicting NIV failure using the VOX index was 0.843 (95% CI 0.805-0.882). Using a VOX value threshold of 20.45 to predict NIV failure, the sensitivity was 69.1%, and the specificity was 94.4%. Furthermore, a VOX value ≤ 20.45 was identified as an independent risk factor for tracheal intubation and death.

VOX index shows promise to serve as an effective evaluation index to predict early NIV efficacy in patients with AHRF; a VOX value > 20.45 after 2 h of NIV treatment can better predict improvements in hypoxia, respiratory drive, and NIV outcomes, guide early tracheal intubation in cases of NIV failure, and have a certain predictive effect on patient outcomes.