Abstract Rationale We investigated whether pre-existing respiratory neuromuscular disease (NMD) is associated with prolonged weaning in mechanically ventilated pediatric ARDS (PARDS) patients. Understanding the magnitude of this relationship and which other clinical outcomes are modulated by NMD may help with risk stratification and PARDS research. We hypothesized that NMD would be strongly associated with poor weaning outcomes. Methods This was a secondary analysis of a single-center randomized controlled trial evaluating lung- and diaphragm-protective ventilation in PARDS (R01 HL134666). Daily standardized spontaneous breathing trials (SBTs) were performed after patients met prespecified weaning criteria, with failure defined by standardized parameters including esophageal manometry-derived effort of breathing. We identified patients with respiratory NMD during primary study screening and validated diagnoses through the EMR. The primary outcome was length of weaning, defined as time from first attempted SBT to time of extubation or first successful SBT. Secondary outcomes were first SBT failure, ICU and hospital length of stay, and ventilator free days (VFDs). Multivariable modeling with transformation (when necessary) adjusted for age, oxygenation index (OI) at randomization, and primary trial arm. Results Of 217 patients who entered the weaning stage of ventilation, 33 (17%) had respiratory NMD. Patients with NMD had a median age of 8.7 years versus 3.8 years in those without NMD, with median OI of 6.5 versus 7.6 respectively. The median time from randomization to first SBT was 2.0 days in patients with NMD versus 2.9 days in those without. Median length of weaning was 1.05 days in patients with NMD with 25% of patients having a weaning duration 4 days. In patients without NMD, the median length of weaning was 0.95 days, with only 10% of patients having a weaning duration 4 days. In multivariable analysis, NMD was independently associated with longer length of weaning (OR 3.14; 95% CI 1.56-6.31; p = 0.001), longer ICU LOS (77% increase 95% CI 35%-133%; p = 0.001), longer hospital LOS (47% increase 95% CI 0%-97%; p = 0.01) and fewer VFD (IRR 0.85; 95% CI 0.72-0.99; p = 0.048). NMD was not associated with differences in rates of first SBT success, and no significant interaction was observed between NMD and primary trial arm for any clinical outcome (all p 0.05). Conclusion Pre-existing respiratory neuromuscular disease is independently associated with prolonged weaning and other adverse clinical outcomes in PARDS. Future studies should investigate the physiologic drivers of this relationship and test whether targeted ventilator strategies can improve outcomes in this vulnerable population. This abstract is funded by: Department of Anesthesiology and Critical Care Medicine, Children’s Hospital Los Angeles
Lovern et al. (Fri,) studied this question.