Quality of glucocorticoid replacement in adrenal insufficiency: clinical assessment vs. timed serum cortisol measurements

OBJECTIVE: Evaluation of glucocorticoid replacement quality in adrenal insufficiency (AI) relies primarily on clinical judgement and thus largely depends on the physician's expertise. It is a matter of debate whether cortisol day curves are of value in assessing glucocorticoid replacement quality. Here we compared the results of a structured clinical assessment to the outcome of repeated, timed serum cortisol measurements. DESIGN: Cross-sectional study in the outpatient department of a university teaching hospital. PATIENTS: Forty-six patients (19 men, 27 women, age range 16-76 years) with primary (n = 23) and secondary (n = 23) AI on stable replacement with a median dose of 37.5 mg cortisone acetate (range 25-50 mg) since 10 +/- 7 years (range 1-31 years). MEASUREMENTS: Clinical performance was scored by structured assessment of signs and symptoms, physical examination and routine biochemical tests. Serum cortisol was measured on two to three separate occasions in three timed samples after the morning glucocorticoid dose. Bone mineral density was measured in 15 patients with long-standing glucocorticoid replacement. RESULTS: Thirty-seven patients were considered well replaced, whereas clinical scores suggested over- or under-replacement in five and four, respectively. There was no correlation of the clinical score with total or body weight-adjusted glucocorticoid dose. The mean z score of serum cortisol differed significantly between under- and over-replaced patients (P < 0.05) but neither group differed significantly from well-replaced patients. Bone mineral density was normal in all patients studied. CONCLUSIONS: Our results suggest that serum cortisol day curves are of limited value in the monitoring of glucocorticoid replacement. Bone mineral density in AI is generally normal and does not require routine follow-up.