Objective: To study the prevalence of primary thyroid disorders in patients who underwent endogenous hypercortisolism. Design: Retrospective evaluation of 59 patients with Cushing's syndrome (CS) who had, at least, a record of thyroid palpation by expert endocrinologists and basal measurements of TSH by second generation assays. When available, tri-iodothyronine and thyroxine serum levels, TRH-TSH tests and anti-thyroid antibodies were also analyzed. There were two age-and gender-matched control groups. The 'goiter control group' comprised 118 healthy subjects who underwent thyroid palpation. The 'antibody control group' was composed of 40 individuals who attended the blood bank of our hospital. Antibodies against thyroperoxidase and measurements of TSH were analyzed in their blood samples. Methods: Available files of 83 CS patients admitted to our endocrine unit from 1985 to 1998 were examined. Fifty-nine patients (52 women and 7 men) with a mean age of 36.2 years (range 14 -61 years) met the above requirements. Diagnosis of hypercortisolism had been established by a standard 1-mg overnight dexamethasone suppression test and urinary free cortisol (UFC). Etiological diagnosis involved dynamic testing, measurements of ACTH levels and imaging techniques. After treatment, all but one of the patients were cured or controlled of their hypercortisolism. This was established by the finding of subnormal serum cortisol concentrations and/or subnormal 24-h UFC levels. Primary thyroid disorders were defined by the presence of one or more of the following diagnostic criteria: (i) goiter, (ii) positive anti-thyroid antibodies and/or (iii) primary thyroid function abnormalities. Results: Eighteen (30.5%) patients had goiter (diffuse in 78% and nodular in 22%), 14 (23.7%) had primary subclinical hypothyroidism and 5 (8.4%) had hyperthyroidism. In 41 patients evaluated for antithyroid antibodies, it was found that 23 (56.1%) had positive titers. In a group of patients in which thyroid autoantibodies were measured both before and after resolution of hypercortisolism, prevalences of positive titers were 26.7% and 86.7% respectively ðP ¼ 0:001Þ: The overall frequency of primary thyroid abnormalities in our patients with Cushing's syndrome was 55.9%. Conclusions: Patients with endogenous Cushing's syndrome exhibit a remarkably high prevalence of primary thyroid disease. Resolution of hypercortisolism seems to trigger the development of autoimmune thyroid disorders in presumably predisposed subjects.

We report a 49-year-old woman with Marfan syndrome who underwent total thyroidectomy for follicular carcinoma. The patient was given 100 mCi of radioactive iodine (131I) followed by levothyroxine (LT4) 0.2 mg/day after surgery. The subsequent five total body scans were negative and thyroglobulin (TG) measurements ranged between undetectable levels to 12 ng/mL. Nine years after thyroidectomy the patient developed bilateral exophthalmos with markedly positive thyroid-stimulating immunoglobulins (TSI), indicating the presence of Graves' disease. TG levels increased and concurrently pulmonary metastases that did not concentrate radioiodine at tracer doses, were diagnosed. Due to these metastatic lesions, the patient received a therapeutic dose of 150 mCi of 131I 1 month after LT4 withdrawal, and a total body scan was made 10 days later. Slight uptake of 131I was found in the right side of the neck, whereas predominant uptake occurred in the right lung base. We suggest that the elevated TSI played a role in the growth of metastases.

The empty sella turcica is defined as the extension of the subarachnoid space toward the intrasellar region with displacement of the pituitary towards the posteroinferior wall. By autopsy studies, the incidence in the general population is around 20%. The association of Cushing's disease (CD) and empty sella has been infrequently reported. In our group, from a total of 68 patients with CD studied by magnetic resonance imaging (MRI), we found the presence of a primary empty sella syndrome (ESS) in 11 (16%). Of these, 9 had partial and 2 total ESS, and in four of them a microadenoma could be identified. Remission, ascertained by subnormal postoperative cortisol levels in blood and/or urine was obtained in 5 of 6 patients operated on by the transphenoidal route. Following surgery, 2 patients presented cerebrospinal fluid (CSF) leakage, 2 diabetes insipidus, and 2 some form of hypopituitarism, figures apparently higher than in non-ESS patients with CD. Ketoconazole was indicated as second line treatment in 2 patients and as primary therapy in 4, resulting in normal urinary free cortisol (UFC) values, with no complications. The relationship of ESS and CD is probably fortuitous given that the frequency of ESS in the general population is similar. Although in empty sella both surgery and radiotherapy seem to have greater risk of complications, surgery remains the first line treatment. Nevertheless, chronic treatment with ketoconazole could be a useful first choice, particularly when no adenoma is seen, or in those who have contraindications for surgery.