Autoantibodies in thyrotoxicosis: a quantitative study of their behavior in relation to the course and outcome of treatment.

Abstract

Many previous investigations have looked into the association between thyroid autoantibodies and the course of thyrotoxicosis. However, the interpretation of these studies is often difficult, owing to their largely retrospective nature, and comparability is limited because of differences in criteria and sampling schedules. The purpose of this study was to investigate prospectively the serial changes in autoantibodies under carefully controlled conditions during treatment of thyrotoxicosis, and to relate these changes to clinical outcome and surgical histology. Sixty-seven patients with Graves’ disease were studied for 24 months; 34 were treated medically and 33 surgically. Anti-thyroglobulin (anti-Tg) and anti-microsomal (anti-M) antibodies were measured on 16 occasions. The frequency of patients showing anti-Tg fell progressively irrespective of the type of treatment, while that showing anti-M fell initially but rose again. Weak associations were found between anti-Tg and suppressibility after 5 months of drug treatment, between the cumulative amount of anti-M and the final TSH levels in those who remained euthyroid after drugs, and between the circulating levels of anti-M just before surgery and the frequency of clinical hypothyroidism 18 months later. However, no significant correlation emerged between the cumulative amount of antibody during the two-year period and the final TSH level or clinical result after surgery, nor between the prevailing antibody level and surgical histology. Wide, rapid and apparently random fluctuations of antibody levels in individuals cast considerable doubt upon the value of single estimations in the distinction of antibody-positive from antibody-negative members of a thyrotoxic population. We conclude that the measurement of thyroid autoantibodies in Graves’ disease has little practical value for the clinician.