Longitudinal Patient-Reported Voice Quality in Early-Stage Glottic Cancer

Maarten C Dorr,Aniel Sewnaik,Elrozy Andrinopoulou,Diako Berzenji,Emilie A C Dronkers,Simone E Bernard,Arta Hoesseini,Lisa Tans,Dimitris Rizopoulos,Robert J Baatenburg de Jong,Marinella P J Offerman
DOI: https://doi.org/10.1002/ohn.263
Abstract:Objective: Patient-reported voice quality is an important outcome during counseling in early-stage glottic cancer. However, there is a paucity of adequate longitudinal studies concerning voice outcomes. This study aimed to investigate longitudinal trajectories for patient-reported voice quality and associated risk factors for treatment modalities such as transoral CO2 laser microsurgery, single vocal cord irradiation, and local radiotherapy. Study design: A longitudinal observational cohort study. Setting: Tertiary cancer center. Methods: Patients treated for Tcis-T1b, N0M0 glottic cancer were included in this study (N = 294). The Voice Handicap Index was obtained at baseline and during follow-up (N = 1944). Mixed-effects models were used for investigating the different trajectories for patient-reported voice quality. Results: The mean follow-up duration was 43.4 (SD 21.5) months. Patients received transoral CO2 laser microsurgery (57.8%), single vocal cord irradiation (24.5%), or local radiotherapy (17.5%). A steeper improvement during the first year after treatment for single vocal cord irradiation (-15.7) and local radiotherapy (-12.4) was seen, compared with a more stable trajectory for laser surgery (-6.1). All treatment modalities showed equivalent outcomes during long-term follow-up. Associated risk factors for different longitudinal trajectories were age, tumor stage, and comorbidity. Conclusion: Longitudinal patient-reported voice quality after treatment for early-stage glottic cancer is heterogeneous and nonlinear. Most improvement is seen during the first year of follow-up and differs between treatment modalities. No clinically significant differences in long-term trajectories were found. Insight into longitudinal trajectories can enhance individual patient counseling and provide the foundation for an individualized dynamic prediction model.
What problem does this paper attempt to address?