Retrospective cohort study finds that stagnant oxygenation index predicts percussive ventilation failure in children, suggesting early oxygen saturation tracking can guide critical transfers.
Background High-frequency percussive ventilation (HFPV) is an alternative mode of mechanical ventilation that combines time-cycled, pressure-controlled breaths with superimposed high-frequency percussions. Pediatric data suggest improved gas exchange with HFPV. Most reports originate from large academic centers with extracorporeal membrane oxygenation (ECMO) capability. We aimed to evaluate HFPV outcomes in children with acute respiratory failure managed in a community-based pediatric ICU without ECMO capability and to determine whether early changes in oxygen saturation index (OSI) are associated with HFPV failure. Methods We conducted a single-center retrospective cohort study of intubated children with acute respiratory failure who received HFPV for ≥12 h between 2018 and 2023 in a 10-bed PICU without ECMO capability. HFPV failure was defined as death or transfer for higher-level cardiopulmonary support. Changes in OSI (delta-OSI) were analyzed using receiver operating characteristics (ROC) curves to assess the association of HFPV failure. Results Thirty-one subjects met inclusion criteria; median age was 60 months (interquartile range [IQR] 4–180). Seven subjects (23%) met criteria for HFPV failure, including 5 deaths (16%) and 2 transfers (7%) for higher level of cardiopulmonary support. Oxygenation trajectories differed between groups, with higher OSI at 24 h in the failure group (21.7 [IQR 6.4–22.0] vs. 10.3 [IQR 8.0–15.2]). Discriminative ability of delta-OSI improved over time, with the highest area under the curve observed at 24 h. Conclusions In a resource-limited PICU without ECMO capability, lack of improvement in OSI within 12–24 h of HFPV initiation was associated with HFPV failure. These findings suggest that early OSI trajectories may aid clinical decision-making regarding escalation of care, though prospective, multicenter validation is needed.
No takes yet. Share an insight, caveat, or question.
Proaño et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: