Is Subclinical Cushing’s Syndrome an Entity or a Statistical Fallout from Diagnostic Testing? Consensus Surrounding the Diagnosis Is Required before Optimal Treatment Can Be Defined

Abstract

Subclinical hypercortisolism (SH) is a commonly used term to define subtle cortisol hypersecretion but in the absence of symptoms or signs specific of overt hypercortisolism (1, 2). This subtle cortisol excess has been described in up to 20% of patients with an adrenal incidentaloma. With 7% of subjects aged 70 yr harboring such tumors, SH is therefore potentially a common disease (3, 4). When groups of such patients are evaluated, increased body mass index, hypertension, glucose intolerance, hyperlipidemia, and reduced bone mineral density are reported, in keeping with deleterious effects of hypercortisolism. On this background, several studies have now evaluated the effect of laparoscopic adrenalectomy in reversing the biochemical and clinical features of SH with conflicting results (5–7). The latest retrospective study from Chiodini et al. (8) evaluated 108 patients from four hospitals in Milan referred over a 5-yr period. Forty-one of108patientswerediagnosedashavingSH,andofthese, 25 underwent adrenalectomy in a nonrandomized fashion,and16weretreatedconservatively.Thepaperisnovel in that it also presents data on the 67 patients who did not have SH, 30 of whom had surgery on grounds of unacceptable tumor size. They showed that blood pressure, weight, and fasting glucose responded greatest in the treatedSHgroupwhencomparedwithboththeuntreated SH group and the treated non-SH group; conversely, weight, blood pressure, lipids, and glucose seemed to deteriorate in the nontreated SH group over a mean follow-up of 3 yr. The paper has limitations. It is not prospective or randomized and there was clearly a selection bias in that treatedpatientswereyoungerandhadbiggertumors.The absolutevaluesforoutcomemeasuresdonotalwaysalign with the percent numbers of subjects who had steady, increased, or decreased values from which the headline conclusions are drawn. Thus, 32% of treated SH subjects are stated to have a reduction in body weight after treatment, but the raw data show no overall change in body weight in this group (or in fact any of the other three groups after prolonged follow-up). Blood pressure and fasting glucose did fall in the treated SH group but also fell in the non-SH treated group, an interesting observation in its own right but one that suggests an adrenalectomy-induced effect that is independent of cortisol. The greatest area of concern, however, in this paper, and others of its kind, relates to the diagnosis of SH. First, there is a nomenclature issue here. If we agree that SH is more frequently associated with obesity, glucose intolerance,andreducedbonemass,howisthissubclinicalwhen these are established features of Cushing’s syndrome? As we debated when establishing Endocrine Society guidelines for the diagnosis of Cushing’s syndrome (9), the reality is that some of these patients will undoubtedly have mild Cushing’s syndrome, and the difficulty is securing a diagnosis at low levels of cortisol hypersecretion. Here, obesityanddiabetesarepotentialconfounders;activation ofthehypothalamic-pituitary-adrenalaxiswithincreased cortisol secretion is a hallmark of obesity (10), and demonstrating autonomous cortisol hypersecretion vs. increased secretion rate per se is crucial (9). Second, many patients labeled as having SH are likely to be victims of false-positive testing. No single test to