Douglas B Chepeha1,2, Matthew E Spector2, Steven B Chinn3, Keith A Casper4, Eric J P Chanowski5, Jeffrey S Moyer2, Robert Morrison2, Emily Carvill6, Teresa H Lyden7. 1. Department of Otolaryngology-Head and Neck Surgery, University of Toronto, Toronto, Ontario, Canada. 2. Department of Otolaryngology-Head and Neck Surgery, University of Michigan, Ann Arbor, Michigan. 3. Department of Head and Neck Surgery, University of Texas MD Anderson Cancer Center, Houston, Texas. 4. Department of Otolaryngology-Head and Neck Surgery, University of Cincinnati, Cincinnati, Ohio. 5. Department of Anesthesiology, University of Florida, Gainesville, Florida. 6. Department of Speech-Language Pathology, McLaren Flint Health Care System, Flint, Michigan. 7. Department of Speech-Language Pathology, University of Michigan, Ann Arbor, Michigan.
Abstract
BACKGROUND: The purpose of this study was to model >12 month speech and the oral phase of swallowing outcomes with the reconstructive metrics of tongue elevation and protrusion in patients reconstructed with the rectangle tongue template for a hemiglossectomy defect. METHODS: We conducted a study using 40 surviving patients (23 men, 17 women) treated between 2000 and 2012. Statistically significant correlations of elevation and protrusion with functional outcomes were modeled with receiver operator characteristic (ROC) curves to understand the performance and reliability of the rectangle tongue reconstruction. RESULTS: Tongue elevation (1.8-1.9 cm) reliably produces best outcomes in nutritional mode, range of liquids, and ≥4/6 for range of solids. Greater tongue elevation (2.1-2.2 cm) reliably produces best outcomes for eating and speaking in public and understandability of speech. Tongue protrusion (0.8-1.0 cm) reliably produces best scores across all assessed outcomes except ≥4/6 for range of solids and ≥4/5 understandability of speech. CONCLUSION: ROC curves are useful for assessing reliability and relating reconstructive objectives to functional outcomes.
BACKGROUND: The purpose of this study was to model >12 month speech and the oral phase of swallowing outcomes with the reconstructive metrics of tongue elevation and protrusion in patients reconstructed with the rectangle tongue template for a hemiglossectomy defect. METHODS: We conducted a study using 40 surviving patients (23 men, 17 women) treated between 2000 and 2012. Statistically significant correlations of elevation and protrusion with functional outcomes were modeled with receiver operator characteristic (ROC) curves to understand the performance and reliability of the rectangle tongue reconstruction. RESULTS: Tongue elevation (1.8-1.9 cm) reliably produces best outcomes in nutritional mode, range of liquids, and ≥4/6 for range of solids. Greater tongue elevation (2.1-2.2 cm) reliably produces best outcomes for eating and speaking in public and understandability of speech. Tongue protrusion (0.8-1.0 cm) reliably produces best scores across all assessed outcomes except ≥4/6 for range of solids and ≥4/5 understandability of speech. CONCLUSION: ROC curves are useful for assessing reliability and relating reconstructive objectives to functional outcomes.
Authors: Phil Weir-Mayta; Kristie A Spencer; Tanya L Eadie; Kathryn Yorkston; Sara Savaglio; Chris Woollcott Journal: Am J Speech Lang Pathol Date: 2017-06-22 Impact factor: 2.408