Clinical review: Early morning cortisol levels as a predictor of remission after transsphenoidal surgery for Cushing's disease.

INTRODUCTION We describe the use of serum cortisol and ACTH levels on postoperative d 1 and 2 as remission predictors after transsphenoidal surgery for Cushing's disease (CD). METHODS Morning cortisol and ACTH levels were drawn daily after surgery; glucocorticoids were withheld until evidence of hypocortisolemia. Early remission was defined retrospectively as a subnormal morning cortisol level [< or =140 nmol/liter (< or =5 microg/dl)] on postoperative d 1 or 2 and sustained remission as subsequent eucortisolemia. RESULTS Of 40 consecutive adults with CD (mean age 39 yr), 80% achieved early remission. Of 39 patients with a minimum follow-up of 14 months (mean 33 months), 31 (79.5%) achieved sustained remission at a mean follow-up of 32 months, including 30 of 31 (97%) with early remission and one of eight (12%) without early remission (P < 0.0001). Sustained remission was achieved in 26 of 28 (93%) patients having their first operation, compared with five of 11 (45%) with a prior unsuccessful operation (P < 0.001). For the 32 patients in early remission vs. the eight in nonremission, mean nadir cortisol levels were 57.6 +/- 33.0 (2.05 +/- 1.2 microg/dl) vs. 631.1 +/- 352.2 nmol/liter (22.9 +/- 12.8 microg/dl) (P < 0.0001), and nadir ACTH levels were 11.9 +/- 6.5 vs. 64.1 +/- 54.6 ng/liter (P < 0.001). Of 31 patients with sustained remission, 100% had subnormal morning cortisol levels, whereas 31% had subnormal ACTH levels (P < 0.0001). CONCLUSIONS Serum morning cortisol levels on postoperative d 1 and 2 without glucocorticoid replacement provide a safe, simple, and reliable measure of early remission for CD and are predictive of sustained remission. This method allows for consideration of a repeat operation during the same hospitalization in patients with persistent hypercortisolemia.