(1321) POCUS for Prediction of Post-Extubation Stridor: A New Approach to an Old Problem?

Introduction: Post-extubation stridor (PES) is a major cause of extubation failure in pediatrics, increasing morbidity and mortality. Current tools used to assess these at risk patients have poor sensitivity and specificity. Airway point of care ultrasound (POCUS) has been proposed to assess edema and predict PES risk by measuring the laryngeal air column width difference (LACWD) and ratio (LACWR). We hypothesized that LACWD and LAWCR, measured by PICU physicians with laryngeal ultrasound (LUS), will provide a consistent measurement of upper airway patency to predict PES when compared to unit standard endotracheal tube (ETT) cuff pressure measurements.

Methods: We conducted a prospective, single-center observational study. LUS was performed on intubated children in the PICU, 0-18 yrs, within 6 hrs of extubation readiness. PICU operators were trained in LUS technique. LACWD was calculated: LACW deflated minus LACW inflated; LACWR was calculated: LACW deflated divided by LACW inflated. Each operator recorded 6 measurements. The presence of PES was noted. ETT cuff pressures were recorded per unit protocol. Interrater and intrarater reliability tests were conducted among 4 scanners.

Results: This interim analysis included 29 patients with 62% < 8yrs. PES developed in 14%, all < 8 yrs. None required reintubation. In the < 8 group, comparing those with PES and those without, there were no significant differences in mean LACWD, LACWR, or cuff pressures. Spearman's correlation showed a negative relationship between LACWD and cuff pressures in those < 8yrs (coefficient -0.25, p-value 0.36). Intrarater reliability was high-excellent (ICC 0.9, CI 0.942 to 0.995). Interrater reliability varied from poor-excellent (ICC 0.4 to 0.8, CI 0.008 to 0.981) among different scanners.

Conclusions: This study reinforces the need for a reproducible, sensitive and specific tool for PES prediction, particularly in those < 8yrs. LUS lacked sensitivity and specificity to predict PES in our small cohort. Although LUS demonstrated high intrarater reliability, the difficulty and notable variability in LUS technique among operators poses a significant obstacle to widespread adoption. Enhanced patient recruitment combined with improvement in operator technique could significantly increase the utility of LUS in future applications.