Cancer mortality in women with thyroid disease.

A retrospective follow-up study of 7338 women with either nontoxic nodular goiter, thyroid adenoma, hyperthyroidism, hypothyroidism, Hashimoto's thyroiditis, or no thyroid disease was conducted. All women patients at the Massachusetts General Hospital Thyroid Clinic who were seen between 1925 and 1974 and who were treated for a minimum of 1 year were traced. A total of 2231 women (30.4%) were dead and 2012 women (27.4%) were alive as of December 31, 1978. Partial follow-up information was available for the remaining 3095 women (42.2%). The average length of follow-up was 15.2 years. When losses to follow-up were withdrawn at the time of their loss, the standardized mortality ratios (SMR) for all causes of death were 1.2 [95% confidence interval (CI), 1.1-1.3] for women with nontoxic nodular goiter, 1.2 (95% CI 1.0-1.3) for those with thyroid adenoma, 1.4 (95% CI 1.3-1.5) for women with hyperthyroidism, 1.5 (95% CI 1.3-1.7) for hypothyroid women, 1.2 (95% CI 0.9-1.5) for those with Hashimoto's thyroiditis, and 1.5 (95% CI 1.4-1.6) for those without thyroid disease. For deaths from all cancers, the standardized mortality ratios were 1.5 (95% CI 1.2-1.8) for women with nontoxic nodular goiter, 1.5 (95% CI 1.1-1.9) for those with thyroid adenoma, 1.2 (95% CI 1.0-1.4) for women with hyperthyroidism, 1.0 (95% CI 0.7-1.4) for the hypothyroid women, 1.2 (95% CI 0.7-2.1) for those with Hashimoto's thyroiditis, and 1.3 (95% CI 1.0-1.5) for those women without thyroid disease. When specific cancer sites were studied, excess numbers of deaths were observed from breast cancer in women with nontoxic nodular goiter (SMR = 1.6, 95% CI 1.0-2.6) and from lymphatic and hematopoietic cancer in women with nontoxic nodular goiter (SMR = 2.4, 95% CI 1.2-4.3) and thyroid adenoma (SMR = 2.7, 95% CI 1.1-5.2). An increase in thyroid cancer risk was observed in women with thyroid adenoma (SMR = 11.7, 95% CI 1.3-42.1) but was based on only two deaths. In hyperthyroid women, statistically significant increases in the number of deaths were observed from pancreatic cancer (SMR = 2.6, 95% CI 1.4-4.3) and respiratory cancer (SMR = 2.2, 95% CI 1.3-3.5), but not breast cancer (SMR = 1.3, 95% CI 0.8-1.8). When the data were stratified by the time between the onset of thyroid symptoms and death, a nonsignificant excess number of cancer deaths was observed in hyperthyroid women who died 20 or more years after their symptoms began.(ABSTRACT TRUNCATED AT 400 WORDS)