Importance: Tracheoesophageal voice prosthesis (TEP) is a commonly used device for alaryngeal voicing following total laryngectomy. Variability in TEP voice and speech production may be reflected in differences in intratracheal pressures.
Objective: To examine the association between intratracheal manometric pressure (IMP), speech function, and proximal esophageal abnormalities in patients with TEP.
Design, setting, and participants: This retrospective cohort study was conducted at a single academic institution and included all patients with a history of total laryngectomy and TEP who completed a radiographic esophagram and intratracheal manometry between September 9, 2019, and December 4, 2019.
Main outcomes and measures: Intratracheal manometric pressures during sustained phonation, conversational speech, and maximum loudness tasks were obtained for 22 patients. Pressure metrics, phonation duration time, and fluency of speech production were compared between patients with dysphagia with those without swallowing complaints. The association between pressure and speech metrics was also assessed.
Results: Of 22 patients, 17 (77.2%) were men, 5 (22.7%) were women, and the mean (SD) age was 66.6 (8.8) years. The mean (SD) sustained phonation duration time was 9.3 (5.2) seconds, and the mean (SD) IMP during sustained phonation was 50.5 (21.4) cm H20. The IMPs during sustained phonation were strongly correlated with conversational speech (r = 0.712; 95% CI, 0.384-1.039) and moderately correlated with maximum loudness tasks (r = 0.524; 95% CI, 0.127-0.921). The IMPs during conversational speech were moderately correlated with maximum loudness task (r = 0.538; 95% CI, 0.145-0.931). Increased IMP during sustained phonation was moderately correlated with decreased sustained phonation duration time (r = -0.450; 95% CI, -0.867 to -0.034) and conversational speech was moderately correlated with decreased sustained phonation duration time (r = -0.524; 95% CI, -0.921 to -0.127). Patients with subjective swallowing complaints and proximal esophagus abnormalities had a shorter maximum phonation time (mean [SD], 7.08 [5.03] seconds) compared with patients without subjective dysphagic complaints (mean [SD], 11.95 [4.40] seconds), with a large effect size (Cohen d = 1.031; 95% CI, 0.141-1.92). All patients with nonfluent TEP speech production had structural abnormalities of the proximal esophagus. No difference in pressure or speech metrics was observed for primary closure vs reconstructive flap type.
Conclusions and relevance: This cohort study found that increased IMP and abnormalities of the proximal esophagus were associated with worse TEP speech quality. Evaluation of the esophagus should be considered in patients who are experiencing difficulties with TEP voicing. Further investigation of intratracheal manometry as a biofeedback tool to improve TEP voicing is needed.