BackgroundCushing's syndrome (CS) is a condition caused by an increase in blood cortisol levels, leading to elevated blood sugar, osteoporosis, neuropsychiatric disorders, and weight gain and abnormal fat distribution. Its diagnostic process is cumbersome, and the process of clinical confirmation is influenced by many factors. ObjectivesIn this study, we investigated the relationship between the 24-hour urinary cortisol/urinary creatinine ratio (24hUCor/Cr) and other relevant biological markers associated with CS in an attempt to find novel ways to diagnose the disease. MethodsThis cross-sectional study included 395 hospitalized patients. Urinary cortisol and creatinine were measured by retaining a 24-hour urine sample. Multivariate binary logistic regression analysis was used to test the correlation between the 24hUCor/Cr and CS. All patients diagnosed with CS underwent a standardized diagnostic process. ResultsOf the 395 subjects, 94 (23.8%) were diagnosed with CS. In the unadjusted model, the likelihood of subjects having CS increased with increasing 24hUCor/Cr (odds ratio: 2.05, 95% confidence interval: 1.56–2.71; P<0.001). In the fully adjusted model, the risk of patients developing CS increased by 74% for each standard deviation increase in the 24hUCor/Cr (95% confidence interval: 1.02–2.97; P<0.05). Based on the subject working characteristic curve analysis and Youden’s index, the sensitivity and specificity of the 24hUCor/Cr for predicting CS were 88.3% and 91.7%, respectively, in all subjects (cutoff value: 52.3554, AUC: 0.942; P<0.001) ConclusionA higher 24hUCor/Cr helped to identify CS with an optimal cutoff value of 52.3554.
Read full abstract