IT HAS BEEN SUSPECTED THAT PATIENTS with adrenocortical insufficiency have been increasing in number because of increased therapeutic use of adrenal steroids and the performance of adrenalectomy or hypophysectomy for treatment of certain diseases. This has produced a growing probability of encountering iatrogenic diseases or disorders. Among them adrenal insufficiency during or after anaesthesia and surgery is the most dangerous complication, characterized mainly by arterial hypotension and tachycardia and terminating in shock or death. 1 Therefore, if we could find an appropriate method of predicting adrenocortical hypofunction preoperatively, it would improve the safety of anaesthetic management of steroid-treated patients. The present study was undertaken to investigate a possibility to predict and to prevent adrenal hypofunction during anaesthesia in patients treated with steroid. METHOD Seventy-nine subjects were involved; 30 subjects served as controls, and 39 patients previously treated with steroid who underwent various operations were studied, as shown in Table I These 39 patients had been treated with steroid at some period up to 8 years prior to operation. The total amount of glucocorticoids expressed as equivalent to cortisol ranged from 1.0 gm to 732.4 gm. None of them received steroids on the day before operation or on the day of surgery. In addition a group of ten patients who had no history of steroid therapy in the past and who had operations served as a control.Each patient in both groups was premedicated with pentobarbital 50-100 mg orally at night and 1% hours prior to the induction of anaesthesia. Meperidine 35 mg and atropine 0.5 mg were given by intramuscular injection 1& hours before the induction of anaesthesia. Due to the diurnal variations of the concentration of free cortisol in the plasma in man, induction of anaesthesia was started at 8:30 AM. Anaesthesia was induced with halothane (0.5-2 per cent), nitrous oxide (2 litres/min), and oxygen (2 litres/min) under a mask, followed by succinylcholine chloride (SCC) 40 mg intravenously for endotracheal intubation. Anaesthesia was maintained with halothane, nitrous oxide, and oxygen, combined with intermittent SCC injection whenever needed-Ventilation was controlled or assisted throughout the procedure. A moderate depth of anaesthesia was maintained during each procedure by clinical judgment and occasional electroencephalographic monitoring.Determination of free cortisol (17-hydroxycorticosteroid = 17-OHCS) concentration in 1 ml plasma by Rudd's fluorimetric method 2 and six blood sampling times were similar to that previously reported. 1 For the protection of adrenocor-