Impact of surveillance on survival after laryngeal cancer in the medicare population

David O. Francis, Bevan Yueh, Ernest A. Weymuller, Albert L. Merati

Research output: Contribution to journalArticlepeer-review

24 Scopus citations


Objectives/Hypothesis: Routine surveillance is advocated to detect recurrent disease after treatment for laryngeal cancer. This aim of this study was to determine the 1-and 5-year postrecurrence mortality for laryngeal cancers and evaluate whether more intensive surveillance improved survival. Study Design: Retrospective cohort study. Methods: Patients with recurrent cancers (1992-1999) were identified in a national cancer clinical database. Multivariate analysis was used to evaluate the effect of surveillance on postrecurrence survival. Results: Of 2,121 recurrent cancers identified, 913 were laryngeal. Patients with laryngeal cancer recurrence had 27% (P =.001) and 22% (P =.007) better odds of 1-and 5-year survival than other sites. The 1-and 5-year postrecurrence survival rates for laryngeal cancer patients were 72.4% and 41.3%, respectively. Glottic cancer cases had the best postrecurrence life expectancy. Multivariate regression revealed that clinical surveillance intensity had no independent impact on their survival (P <.05). However, patients with recurrent glottic cancer seen in surveillance had 23% improved odds of survival (P =.037). Conclusions: More frequent surveillance visits was not associated with a survival advantage in the overall population. Patients with glottic cancer had a postrecurrence survival advantage if seen during the surveillance period. Laryngeal cancer patients had better postrecurrence survival than other head and neck sites.

Original languageEnglish (US)
Pages (from-to)2337-2344
Number of pages8
Issue number12
StatePublished - Dec 1 2009


  • Laryngeal cancer
  • Recurrence
  • Surveillance
  • Survival


Dive into the research topics of 'Impact of surveillance on survival after laryngeal cancer in the medicare population'. Together they form a unique fingerprint.

Cite this