A clinical and "blind" histologic review of 82 cases of bronchial carcinoid tumour is presented. The malignant potential of the tumours was only partly predictable from their histologic appearance. Histologically 65 of the tumours were typical benign carcinoids. Regional metastases were found at operation in two of these patients. Fatal carcinoid syndrome with regional and distant metastases appeared in two patients about 1 and 3 years postoperatively. One of the patients with regional metastases at operation is clinically free from carcinoid 12 years later. Malignancy was histologically suspected in 17 cases, in ten of which regional metastases were found at operation. Three of these ten patients are alive 6 to 16 years postoperatively, but two without regional metastases at operation died of local recurrence and distant metastases after 3 to 4 years. Carcinoid syndrome was not seen in these 17 patients. There was one peroperative death. Altogether ten patients (12%) died of recurrence. Among the cases judged at the "blind" histologic review to be suspectedly malignant, the corresponding figure was 50%. For typical carcinoids, conservative resection, including lymph-node metastases, is the treatment of choice. Wedge or sleeve resection with or without pulmonary resection were employed in ten cases. Suspectedly malignant carcinoid tumours may require more extensive surgery.

A virilizing adrenocortical adenoma was demonstrated in a young female. Urinary 17-ketosteroid excretion and subfractions, plasma testosterone level and urinary 17-ketogenic steroid excretion were markedly increased. Dehydroepiandrosterone was the main constituent of the androgen excess. Otherwise the adrenocortical function was found to be normal as evaluated from measurements of cortisol, corticosterone and their metabolites under basal conditions and during dynamic tests. The androgen excess showed an unexpected response to trophic hormones. Human chorionic gonadotrophin stimulation resulted in a pronounced increase in androgen production, whereas no gonadotrophin-dependency could be demonstrated by means of the oestrogen suppression test. Similarly, no corticotrophin-dependency could be demonstrated by corticotrophin stimulation and suppression tests. Removal of the tumour resulted in normalization of the androgen production and no abnormal response upon human chorionic gonadotrophin administration could now be found. The inappropriate response of tumours to trophic hormones is discussed. It is concluded that the reliability of stimulation and suppression tests in determining the site of excessive androgen production should be accepted with reservation.