The Diagnosis of Hypercortisolism. Biochemical Criteria Differentiating Patients from Lean and Obese Normal Subjects and from Females on Oral Contraceptives
Hypercortisolism may be difficult to diagnose because the clinical features are sometimes unimpressive, or because current laboratory determinations may furnish misleading results. Studies have been performed on 40 lean subjects, 28 obese subjects, 15 females on oral contraceptives, 23 patients with long-standing paraplegia, 17 patients with undoubted features of Cushing's syndrome and chemical evidence of hypercortisolism, and 28 individuals with questionable hypercortisolism. The overlap between obese normal subjects and patients with hypercortisolism in urinary 17-OHCS excretion and cortisol secretion rates has been eliminated by expressing these measurements per gram of excreted creatinine. Patients with hypercortisolism invariably excreted more than the normal amounts of 17-OHCS (2.0–6.5 mg/day/g creatinine) and secreted more than normal amounts of cortisol (4.4–19.0 mg/day/g creatinine). Plasma 17-OHCS in patients with hypercortisolism was sometimes normal between 8 and 9 am [normal =19.2 ±7.2 (sd) μg/100 ml], but always exceeded the normal limits defined at 12 noon (normal =13.3 ±7.0 μg/100 ml), 4 pm (10.2 ±5.0 μg/100 ml), 8 pm (7.7 ±5.4 Mg/100 ml) and 12 midnight (6.3 ±8.6 μg/100 ml). While urinary 17-OHCS misleadingly fell below 4 mg/day on dexamethasone, 0.5 mg 6-hourly, in 3 of the patients with typical Cushing's syndrome the urinary 17-OHCS/g creatinine and plasma 17-OHCS invariably remained outside the defined limits of the normal responses to graded, submaximal doses of dexamethasone, and invariably failed to fall below 1 mg/g creatinine on the second day of dexamethasone, 20 μg/kg/day, given 6-hourly. The excretion of free cortisol revealed very little overlap between normal subjects and patients with hypercortisolism. Oral contraceptives preserved diurnal variations in plasma 17-OHCS but usually at levels higher than normal, while urinary 17-OHCS/g creatinine were low-normal, cortisol secretion rates normal, and suppressibility by dexamethasone was normal. Paraplegics had normal plasma 17-OHCS concentrations but frequently elevated urinary 17-OHCS and cortisol secretion rates per gram of creatinine. Among the 28 patients with “questionable hypercortisolism,” 16 were shown by the studies used to be clearly normal, 4 to have chemical evidence of hypercortisolism, and 8 to have conflicting steroid data, the interpretation of which is discussed. It is concluded that diagnostic interpretation of steroid measurements in patients with suspected hypercortisolism is greatly facilitated by taking into account the apparent fact that cortisol secretion rate is a function of body cell mass, except when significant losses of cell mass have taken place in adult life.