Background: The aims of this study were to evaluate the validity of preoperative basal serum cortisol levels measured in predicting preoperative adrenal insufficiency and also the validity of basal serum cortisol levels and early postoperative insulin tolerance test (ITT) in predicting postoperative adrenal insufficiency. Methods: The study was prospectively designed and included 64 patients who underwent pituitary surgery for conditions other than Cushing's disease. An ITT was performed preoperatively, on the 6th postoperative day and at the 1st postoperative month. Basal serum cortisol levels were measured on the 2nd, 3rd, 4th, 5th, and 6th postoperative days. Results: Patients with a preoperative basal cortisol level of !165 nmol/l (6 mg/dl) showed insufficient cortisol response and those with levels higher than 500 nmol/l (18 mg/dl) had sufficient cortisol response to the preoperative ITT. The positive predictive value of the ITT performed on the 6th postoperative day was 69.7%, and the negative predictive value in predicting adrenal insufficiency at the 1st postoperative month was 58%. Patients were considered to have an insufficient cortisol response to ITT at the 1st postoperative month if their basal cortisol levels were !193 nmol/l (7 mg/dl) or 220 nmol/l (8 mg/dl) or 193 nmol/l (7 mg/dl) or 165 nmol/l (6 mg/dl) or 83 nmol/l (3 mg/dl) on the 2nd-6th postoperative days respectively. Conclusion: Serum basal cortisol levels may be used as the first-line test in the assessment of the hypothalamic-pituitary-adrenal axis both preoperatively and postoperatively. Dynamic testing should be limited to the patients with indeterminate basal cortisol levels.