Epidemiology and blood parameter changes in Cushing's syndrome - a population-based study.

Abstract

To provide complete epidemiological data on Cushing's syndrome (CS) with analysis and differentiation of biochemical parameters, including blood count indices and serum inflammation-based scores. Clinical records of 35 patients diagnosed with CS between 2008 and 2020 at Malta's only central National Health Service hospital were retrospectively analyzed. Detailed clinical and biochemical data were obtained for each patient. Correlation and receiver operator characteristics (ROC) curve analyses were used to establish a threshold value for different variables to predict malignant CS. Standardized incidence rate (SIR) (/million/year) of CS was 4.5, and SIR of Cushing's disease (CD) was 2.3, 0.5 for ectopic CS, 1.5 for cortisol secreting adrenal adenoma, and 0.3 cases for cortisol-producing ACC. Malignant cause of CS had statistically significantly higher cortisol levels and size of tumor and lower potassium at diagnosis (P < 0.001). Additionally, malignant causes had a higher neutrophil-to-lymphocyte ratio (NLR) (P = 0.001) and systemic immune inflammation index (P = 0.005) and a lower lymphocyte-to-monocyte ratio (P < 0.001). Using ROC curve analysis to predict malignant cause of CS, a potassium level of < 3.05 was 75% sensitive and 100% specific (ROC-AUC 0.907, P = 0.001), a post-ODST cortisol level of > 841nmol/L was 100% sensitive and 91% specific (ROC-AUC 0.981, P < 0.001), while a NLR ratio > 3.9 was 100% sensitive and 57.7% specific (ROC-AUC 0.885, P = 0.001). Biochemical and blood count indices and serum inflammatory-based scores differ remarkably between benign and malignant causes of endogenous CS. Such indices can help predict the severity of disease and prognosis.