The first comprehensive theory about the development of multinodular goiter was proposed by David Marine and studied further by Selwyn Taylor, and can be considered one of the classics in this field. Nodular goiter may be the result of any chronic low-grade, intermittent stimulus to thyroid hyperplasia. Supporting evidence for this view is circumstantial. David Marine first developed the concept, that in response to iodide deficiency, the thyroid first goes through a period of hyperplasia as a consequence of the resulting TSH stimulation, but eventually, possibly because of iodide repletion or a decreased requirement for thyroid hormone, enters a resting phase characterized by colloid storage and the histologic picture of a colloid goiter. Marine believed that repetition of these two phases of the cycle would eventually result in the formation of nontoxic multinodular goiter.8 Studies by Taylor of thyroid glands removed at surgery led him to believe that the initial lesion is diffuse hyperplasia, but that with time discrete nodules develop.9

By the time the goiter is well developed, serum TSH levels and TSH production rates are usually normal or even suppressed.10 For example, Dige-Petersen and Hummer evaluated basal and TRH-stimulated serum TSH levels in 15 patients with diffuse goiter and 47 patients with nodular goiter.11 They found impairment of TRH-induced TSH release in 27% of the patients with nodular goiter, suggesting thyroid autonomy, but in only 1 of the 15 with diffuse goiter. Smeulers et al. 22, studied clinically euthyroid women with multinodular goiter and found that there was an inverse relationship between the increment of TSH after administration of TRH, and size of the thyroid gland (Figure 17-1). It was also found that, while being still within the normal range, the mean serum T3 concentration of the group with impaired TSH secretion was significantly higher than the normal mean, whereas the mean value of serum T4 level was not elevated.12 These and other (1) results12 are consistent with the hypothesis that a diffuse goiter may precede the development of nodules. They are also consistent with the clinical observation that, with time, autonomy may occur, with suppression of TSH release, even though such goiters were originally TSH dependent.

Comprehensive reviews about insights into the evolution of multinodular goiter have been published by Studer et al.13-16 An adapted summary of the major factors that are discussed by these authors is presented in Table 17-1 and will be referred to in the discussion that follows.