Purpose: To determine retrospectively the prognostic significance of airway compromise necessitating tracheostomy in carcinoma of the larynx managed with radical radiotherapy and surgery for salvage (RRSS).
Methods and materials: The charts of 270 patients managed with RRSS at the Toronto-Sunnybrook Regional Cancer Centre between June 1980 and December 1990 were reviewed. Airway compromise necessitating tracheostomy was documented in 26 patients prior to radiotherapy and 3 patients during radiotherapy. Of 29, 27 had T3T4 primaries. Patients have been followed for a median of 5 years.
Results: Patients managed without tracheostomy had a 2-year disease-free survival of 74% compared to 41% for those managed with tracheostomy. The adverse impact of airway compromise was more marked in patients with glottic primaries (78% vs. 32%, p = 0.0001) than those with supraglottic primaries (64% vs. 47%, p = 0.18). Tracheostomy was identified in univariate analysis, but not in multivariate analysis, as having a statistically significant impact on local control and local-regional control. Radiotherapy controlled disease above the clavicles in 185 of 267 (69%) evaluable patients. 83% of isolated local-regional failures underwent salvage surgery. Among those managed without tracheostomy, ultimate local-regional control (LRC) was achieved in 161 (94%) of 172 glottic primaries and 54 (81%) of 67 supraglottic primaries. Among those managed with tracheostomy, ultimate LRC was achieved in 9 (69%) of 13 glottic primaries and 12 (80%) of 15 supraglottic primaries. In a subset analysis of 76 patients with T3T4 primaries, there was no statistically significant difference in larynx preservation, disease-free survival, or cause-specific survival between those managed with and without tracheostomy.
Conclusion: Airway compromise necessitating tracheostomy is an adverse prognostic factor in patients with carcinoma of the larynx. However, larynx preservation is possible in over 40% of those undergoing tracheostomy and radiotherapy with no compromise of cause-specific survival. The need for pretreatment tracheostomy should not rule out the possibility of RRSS.