New test may help in early diagnosis of Cushing's Syndrome: Frontiers

The key findings of this study are:

1. The final estimation of the model performance revealed an average AUC=0.841, indicating a little overfitting impact.

2. The retrieved score was organized on a 17.5-point scale as follows: low-risk class (score value: 5.5, probability of disease=0.8%); intermediate-low-risk class (score value: 6-8.5, probability of disease=2.7%); intermediate-high-risk class (score value: 9-11.5, probability of disease=18.5%); and high-risk class (score value: 12, probability of disease=72.5%).

3. With an AUC of 0.873, clinical data demonstrated a great predictive value for distinguishing between participants with and without a final diagnosis of CS.

4. Obesity was maintained as a significant independent predictor of CS in this final multivariable model as well.

5. The accuracy of the Cushing score produced in our study is significantly higher than that of models.

In conclusion, this is the first research to describe an algorithm that is only based on clinical characteristics and can aid clinicians in differentiating instances with a low or high pre-test risk of CS. The derived Cushing score is a simple tool that could be widely adopted in clinical practice and could be of significant assistance in reducing the length and potential pitfalls in CS diagnostic work-up, with implications for patient health and health-care costs, particularly during the COVID-19 pandemic.

Reference:

Parasiliti-Caprino M, Bioletto F, Frigerio T, et al. A new clinical model to estimate the pre-test probability of Cushing's syndrome: The Cushing score. Front Endocrinol (Lausanne) Published Online October 5, 2021. doi:10.3389/fendo.2021.747549