MON-187 Prevalence and Assessment of Overnight Dexamethasone Suppression Tests for Screening Endogenous Hypercortisolism When Serum Dexamethasone Is Below Threshold

Abstract

Background: The 1mg-dexamethasone (DXA) suppression test (DST) is used to verify cortisol (F) autonomy in Cushing’s syndrome (CS) and subclinical hypercortisolism (SCH) of adrenal incidentalomas (AI), based on the ability of DXA to suppress F by inhibiting the HPA axis. A post-DST serum F >1.8 μg/dL usually defines autonomy. Aim: 1) determine the prevalence of invalid tests in a large series of DST used to investigate hypercortisolism during a 12y period in a single institution; 2) assess the percentage of negative tests (normal F suppression) among the subpopulation of invalid DST; 3) examine for interfering substances and clinical conditions that may justify and low DXA levels. Methods: DXA-controlled 1mg-DST was carried out in 162 control subjects (Cont; 107F/55M; 20-75y [median 50y]), to determine a valid threshold for serum DXA, and in 768 patients (80% female; 11-91y [median 53y]), investigated from 2007-19, for F autonomy (a total of 1,300 tests, with 41% repetitions). F and DXA were determined by specific RIA. We search our laboratory data bank for “invalid” DST, tests in which post-DST serum DXA values were below the cutoff established from the control subjects. Results: we set a cutoff for post-DST serum DXA at 140 ng/dL, the lowest value obtained from controls in whom post-DST serum F levels were < 2.5 µg/dL. From the 1,300 DST examined, 146 (11.2%) were considered invalid (DXA <140 ng/dL), and in 36 of them (25%), DXA was undetectable. Also, 35 DST (25%) gave F results below 1.8 µg/dL. Of all 146 invalid DST, 8 (6%) did not take DXA the night before as directed and 14 (10%) admitted being on glucocorticoids (GC), most of them from the undetectable DXA subgroup. Also, 21 patients (14%) were on anticonvulsants, another 21 (14%) were using other moderate or potent CPY3A4 inducers (phenytoin, rifampicin, efavirenz), and 15 (10%) subjects had gastrointestinal abnormalities (colectomy, colostomy, erosive gastritis and disabsorptive syndromes). Conclusions: Test accuracy depends on the serum levels reached by DXA, after adequate ingestion and absorption. Inappropriately low values may result in false-positive test results. One-fourth of “invalid” tests, presented DXA levels below the assay limit of sensitivity, suggesting patient noncompliance. Adherence to verbal and written recommendations and use of medication are essential to interpret the test. Conditions that alter gastric pH, as well as surgical procedures in the GI tract may reduce absorption or accelerate GI transit affecting DXA absorption. DXA metabolism by cytochrome p450 enzymes, mainly CYP3A4, can be enhanced by inducing drugs, making mandatory DXA dosage to validate the test. The wide individual variability in DXA metabolism may be associated not only with environmental and health factors, but also with CYP gene polymorphisms, which can modify drug clearance.