Cushing's syndrome (CS) is associated with an incidence of venous thromboembolism (VTE) about ten times higher than in the normal population. The aim of our study was to develop a model for identifying CS patients at higher risk of VTE. We considered clinical, hormonal, and coagulation data from 176 active CS patients and used a forward stepwise logistic multivariate regression analysis to select the major independent risk factors for thrombosis. The risk of VTE was calculated as a 'CS-VTE score' from the sum of points of present risk factors. VTE developed in 20 patients (4 pulmonary embolism). The group of CS patients with VTE were older (p < 0.001) and had more cardiovascular events (p < 0.05), infections and reduced mobility (both p < 0.001), higher midnight plasma cortisol levels (p < 0.05), and shorter APTT (p < 0.01) than those without. We identified six major independent risk factors for VTE: age ≥69 years and reduced mobility were given two points each, whereas acute severe infections, previous cardiovascular events, midnight plasma cortisol level >3.15 times the normality and shortened APTT were given one point each. A CS-VTE score <2 anticipated no risk of VTE; a CS-VTE score of two mild risk (10 %); a CS-VTE score of three moderate risk (46 %); a CS-VTE score ≥4 high risk (85 %). Considering a score ≥3 as predictive of VTE, 94 % of the patients were correctly classified. A simple score helps stratify the VTE risk in CS patients and identify those who could benefit from thromboprophylaxis.
Keywords: Cushing’s syndrome; Model; Risk; Score; Venous thromboembolism.