Assessment of vocal cord movement by ultrasound in the ICU.

Abstract

Ultrasound can be used to non-invasively and rapidly examine airway conditions, but vocal cord visualization with the traditional approaches is poor. Our aim was to compare the accuracies of front-side transverse-axis ultrasound (FTU), lateral-side longitudinal-axis ultrasound (LLU), and the combination of both approaches for vocal cord movement disorder diagnoses (e.g., vocal cord paralysis or arytenoid cartilage dislocation). We compared FTU, LLU, and the combination of both methods for patients in the intensive care unit (ICU). We used nasal fiber-optic endoscopy to confirm vocal cord injury. Among the 120 patients examined, 24 (20%) had vocal cord paralysis. The visualization rate of vocal cords for FTU was 71.7% (assessable, 86; non-assessable, 34), that for LLU was 88.3% (assessable, 106; non-assessable, 14), and that for the combined approach was 96.7% (assessable, 116; non-assessable, 4). The sensitivities and specificities were 58.3% (14/24) and 75% (72/96) for FTU, 91.7% (22/24) and 87.5% (84/96) for LLU, and 100% (24/24) and 95.8% (92/96) for the combined approach. Visualization rates for LLU were significantly higher than for FTU (P = 0.002); FTU + LLU rates were higher than those for FTU (P = 0.001). The difference between LLU and FTU + LLU was not statistically significant (P = 0.025). LLU can be used to evaluate arytenoid cartilage activity inICUs, and the results are highly correlated with the diagnosis of nasal fiber-optic endoscopy. The combination of FTU and LLU shows promise as a rapid primary screening method for vocal cord injury.