Figure 2.

Decision curve analysis displaying the superior net benefit of using AMH <4 ng/ml for positive sperm retrieval at mTESE in men with idiopathic non-obstructive azoospermia. In decision curve analysis, the lines labelled ‘testing all’ and ‘testing none’ represent reference lines that help interpret the results of the analysis. These lines provide a benchmark against which the performance of a diagnostic or predictive model can be compare, allowing researchers, and decision-makers to evaluate the clinical utility and potential benefits of a test or intervention. The ‘testing all’ line represents a scenario where all individuals, regardless of their risk profile, are subjected to the test or intervention being evaluated. It assumes that the test has perfect accuracy and everyone benefits from it. In this scenario, the model’s net benefit is calculated by comparing the proportion of individuals who benefit from the test to those who are harmed. On the other hand, the ‘testing none’ line represents a scenario where no one undergoes the test or intervention. This line assumes that no one benefits from the test, and the net benefit is determined solely based on the proportion of individuals who are harmed by false positives or unnecessary interventions. AMH, Anti-Müllerian Hormone; mTESE, microdissection testicular sperm extraction.