Endocrine Abstracts

Early diagnosis of CS could reduce the morbidity and mortality associated with endogenous hypercortisolism. Many clinical methods have been proposed to establish suspicion of CS, but they show a poor positive predictive value (PPV) and are sometimes difficult to assess. Assuming the well-known influence of hypercortisolism on white blood cell (WBC) count, we evaluated its potential usefulness as a screening test to trigger this suspicion. We analysed WBC count around the diagnosis of CS patients (cases) and compared them in a 1:2 ratio with age- and sex-matched controls who had undergone a negative Nugent’s test (plasma cortisol <1.8 μg/dl after 1 mg DXM). We evaluated the predictive value of haematological parameters by applying ROC curves and tried to find a reference value to establish the suspicion of CS. We collected 72 confirmed CS cases and 144 matched controls from three centres. Mean age was similar (48.5 vs 48 years), both groups included 82% women. Controls had a higher mean BMI than cases (36 vs 29.2; P<.001). There was no significant difference in the incidence of diabetes (27.8% in controls and 31.9% in cases) and there was a significant higher prevalence of hypertension in cases than in controls (61.1% vs 37.5%; P=.001). The rate of active smoking showed no difference between the two groups. Mean WBC count, neutrophil count and percentage were significantly higher in cases, while mean lymphocyte count and percentage were significantly lower in cases than in controls. ROC curves showed an AUC greater than 0.7 for total WBC count, neutrophil count and percentage, lymphocyte count and percentage, and for the difference of total WBC count minus lymphocytes, the difference of both WBC subtypes (neutrophils minus lymphocytes) and the neutrophil/lymphocyte ratio. The two highest levels of discrimination were obtained with the ratio of neutrophil to lymphocyte count (N/L) and the percentage of lymphocytes, which both had an AUC equal to 0.865. The combination of an N/L ratio greater than 2.73 and a lymphocyte percentage less than 25% gave a sensitivity of 77.8% and a specificity of 80.5% for detecting CS, with a PPV of 66.7%. In conclusion, a simple assessment of white blood cell count could be a valuable and inexpensive clue to the suspicion of Cushing’s syndrome. Its routine application in patients with metabolic syndrome could lead to increased detection of CS in endocrinology clinics and reduce the burden of late diagnosis of this disease.