Dose-response relationship between plasma ACTH and serum cortisol in the insulin-hypoglycaemia test in 25 healthy subjects and 109 patients with pituitary disease.

OBJECTIVE

The insulin hypoglycaemia test (IHT) is believed to be the most reliable test for evaluating the entire hypothalamo-pituitary-adrenal (HPA) axis. The lower limit for the normal peak serum cortisol response has been reported to be between 500 and 580 nmol/l. Reference levels for a normal plasma ACTH response have not been reported recently.

DESIGN AND PATIENTS

We performed the IHT in 25 healthy subjects and in 109 patients with proven or suspected pituitary disorders with serial measurements of serum or plasma cortisol and of plasma ACTH, in order to establish reference levels and to study the dose-response relationship between ACTH and cortisol in this test. In most patients, other pituitary hormonal axes were evaluated in addition.

RESULTS

With the cortisol kit from Diagnostic Products Corporation (DPC), serum cortisol was about 13% lower than plasma (EDTA) levels with an excellent correlation between serum and plasma (r = 0.976; P<0.001). In the normals, the lower limit of the cortisol response (mean cortisol peak level minus 2 SD.) was 570 nmol/l for plasma and 500 nmol/l (calculated) for serum, while the lower limit of the ACTH response was 17.6 pmol/l (80 ng/l). In normals, the cortisol response was independent of the magnitude of the ACTH response. Seventeen out of 30 patients with ACTH responses to levels < 8.8 pmol/l (< 40 ng/l) had subnormal cortisol responses. However, 38 of the patients with pituitary disease had normal cortisol responses in spite of subnormal ACTH responses (group 2), while 47 patients had completely normal IHT results (group 1). Patients in group 2 had more often additional pituitary hormone deficiencies than those of group 1. The dose-response relationship between ACTH and cortisol in the patients resembled a dose-response curve that had been set up previously in normal subjects who received incremental doses of subcutaneous human ACTH (1-39).

CONCLUSIONS

The normal increment of plasma ACTH in the IHT is greater than necessary for stimulating serum cortisol to levels > 500 nmol/l. Patients with a subnormal ACTH but normal cortisol response in the IHT have a decreased ACTH secretory reserve. It is unlikely that they are at increased risk of developing an adrenal crisis perioperatively or in other stressful situations unless pituitary function deteriorates. The ACTH-cortisol relationship in the IHT performed in patients with pituitary disease shows no sharp dividing line between normality and disease, and whether a patient needs permanent glucocorticoid substitution is a discretionary decision.