Many previous studies have shown decreased bone mineral density (BMD) in patients with endogenous hypercortisolism, although data on the severity of bone disease are contradictory. There are very sparse data about the correlations between plasma cortisol concentrations and BMD.
The study included 188 patients with clinically inactive adrenal adenomas (IAA) and 30 patients with ACTH-independent Cushings syndrome (CS) due to adrenocortical adenomas. All patients underwent a detailed hormonal evaluation including measurements of plasma cortisol at 08 and 24 h, as well as low-dose dexamethasone testing (LDDT). BMD was measured by DEXA at the lumbar spine (LS), left femoral neck (FN), as well as at the total proximal femur (TF).
ROC analysis performed for plasma cortisol at 24 h and after LDDT in patients with IAA and CS indicated that the optimal cut-off value which discriminated between IAA and CS was 6.0 μg/dl for midnight plasma cortisol and 3.6 μg/dl for LDDT. Among patients with IAA, a subgroup of patients without overt CS had plasma cortisol levels >6.0 μg/dl at 24 h and >3.6 μg/dl after LDDT, and these patients were considered as having subclinical hypercortisolism (SH, n=9, 4.8% of total).
Patients with IAA had normal BMD at LS, FN and TF. Patients with SH had significantly lower (P<0.01) z-scores at FN (−0.57±0.81 vs. +0.27±1.00) and at TF (−0.48±1.43 vs 0.29±1.03) but not at LS compared to IAA patients without SH. Patients with CS had decreased BMD (P<0.01) at all region (−0.90±1.07 at LS; −0.43±1.07 at FN and −0.38±1.01 at TF) compared to those with IAA. Significant negative correlations were found between plasma cortisol concentrations at 08 h, 24 h as well as after LDDT and BMD z-scores at FN and at TF but not at the LS in the combined group of patients with IAA and CS.
These results indicate that even mild forms of endogenous hypercortisolism lead to decreased BMD at TF. Plasma cortisol concentrations are not directly correlated with BMD at LS.