Cuff Leak Tests at the Time of Extubation Correlate With Voice Quality Assessment

Saba Radhi, Diana Guerra, Raed Alalawi*, Rishi Raj, Kenneth Nugent

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

1 Scopus citations

Abstract

Background. Endotracheal intubation frequently causes laryngeal injury, including edema, ulceration, and vocal cord immobility. These problems can complicate the immediate postextubation course. Methods. The authors measured the cuff leak and made bedside assessment of the voice in 51 patients requiring mechanical ventilation for acute respiratory failure. Results. The median tidal volume was 450 mL (range = 300-600 mL). The median duration of ventilation was 3 days (range = 0-19 days). The median cuff leak was 180 mL (range = -33 to 460 mL); the median percentage cuff leak was 40.9% (range = -7.75% to 99.2%). Women had significantly smaller cuff leaks (18.6% of tidal volume vs 50.3% in men, P =.01). Patients with hoarseness frequently had smaller cuff leaks (median 24.1% vs. 46.7% in nonhoarse patients). Eleven patients (84.6%, 11/13) with a cuff leak <15% had hoarseness; 14 patients (14/38) with a cuff leak >15% had hoarseness (P =.01). Four patients had stridor, and all had a cuff leak >15% of the tidal volume. Conclusion. Patients with low cuff leak volumes are more likely to have postextubation hoarseness. This suggests that these patients have some degree of laryngeal injury and/or edema even in the absence of stridor.

Original languageEnglish (US)
Pages (from-to)27-30
Number of pages4
JournalICU Director
Volume3
Issue number1
DOIs
StatePublished - Jan 1 2012

Keywords

  • cuff leak
  • endotracheal tube
  • hoarseness
  • laryngeal edema

ASJC Scopus subject areas

  • Critical Care
  • Critical Care and Intensive Care Medicine
  • Management Science and Operations Research

Fingerprint Dive into the research topics of 'Cuff Leak Tests at the Time of Extubation Correlate With Voice Quality Assessment'. Together they form a unique fingerprint.

Cite this