Treatment patterns of aging Americans with differentiated thyroid cancer.

BACKGROUND: The incidence of differentiated thyroid cancer (DTC) increases with age. Total thyroidectomy, often followed by radioactive iodine (RAI), is recommended for patients who have tumors that measure > or =1 cm in greatest dimension. In the current study, the authors assessed the use of thyroidectomy and RAI among elderly patients with DTC and the effects on survival.

METHODS: Adults aged > or =45 years with DTC > or =1 cm in the Surveillance, Epidemiology, and End Results database from 1988 to 2003 were included. Bivariate and multivariate analyses were used to measure associations between demographic, clinical, and pathologic characteristics and the likelihood of receiving treatment according to current practice guidelines.

RESULTS: Of 8899 patients who were identified, 26% were ages 65 years to 79 years, and 5% were aged > or =80 years. Compared with younger patients, patients aged > or = 65 years were more likely to have larger tumors, stage IV disease, extrathyroid extension, and nonpapillary histology. Elderly patients were less likely to undergo total thyroidectomy (74% vs 80%; P < .001) or to receive RAI (47% vs 54%; P < .001). These trends were most pronounced among those aged > or =80 years. Among the patients who did not undergo surgery, elderly patients did not report higher rates of contraindications to surgery. In multivariate analysis, the groups ages 65 years to 79 years and aged > or =80 years were associated with lower rates of total thyroidectomy (odds ratio, 0.77 and 0.43, respectively; P < .001) and RAI (odds ratio, 0.85 [P < .01] and 0.39 [P < .001], respectively). Among elderly patients, predictors of worse survival included no surgery (hazard ratio, 5.51; P < .001) and no RAI (hazard ratio, 1.36; P < .001).

CONCLUSIONS: Elderly patients with DTC received less aggressive surgical and RAI treatment than younger patients despite having more advanced disease and the improved survival associated with these treatments among elderly patients. Long-term outcomes should be measured to determine the impact of this apparent discrepancy in care.