Diffusion‐weighted magnetic resonance imaging and magnetic resonance spectroscopy for non‐invasive characterization of azoospermia: A prospective comparative single‐center study

Abstract Background Azoospermia affects about 15% of childless males. The differential diagnosis between subtypes of azoospermia is the initial step in its management. Objectives To investigate the role of diffusion‐weighted magnetic resonance imaging and proton magnetic resonance spectroscopy in distinguishing obstructive azoospermia from non‐obstructive azoospermia and predicting sperm retrieval together with histological alterations in men with non‐obstructive azoospermia. Materials and methods This prospective comparative study involved 60 men with obstructive azoospermia (group A) and 60 men with non‐obstructive azoospermia (group B). Scrotal proton magnetic resonance spectroscopy and diffusion‐weighted magnetic resonance imaging were conducted for all participants to respectively evaluate testicular metabolites and normalized apparent diffusion coefficient 1 week before sperm retrieval. Results Apparent diffusion coefficient was significantly higher in group B as compared to group A (0.47 ± 0.11 vs. 0.29 ± 0.05; and 0.46 ± 0.14 vs. 0.28 ± 0.02) for the right and left testis, respectively. Conversely, testicular choline and lipids were significantly higher in group A as compared to group B. Normalized apparent diffusion coefficient, choline, and lipids at cut‐off levels of 0.353, 0.31, and 0.725 could differentiate between obstructive azoospermia and non‐obstructive azoospermia (area under the curve = 0.963; confidence interval = 0.939–0.986, area under the curve = 0.985; confidence interval = 0.974–0.997, and area under the curve = 0.970; confidence interval = 0.940–0.999, respectively). Regarding the prediction of sperm retrieval in the non‐obstructive azoospermia group, choline levels had the highest area under the curve (0.923), and its cut‐off level was 0.195. The normalized apparent diffusion coefficient was significantly lower in men with positive sperm retrieval as compared to men with unsuccessful retrieval. Finally, it was revealed that all magnetic resonance imaging parameters except creatine could independently predict testicular histology in men with non‐obstructive azoospermia. The highest prediction was 95% in normal spermatogenesis, and the least prediction was 40% in spermatid arrest. Regression analysis was used to detect final predictors and extrapolate an equation that could be used to predict testicular pathology Conclusions Normalized apparent diffusion coefficient and proton magnetic resonance spectroscopy are helpful in differentiating obstructive azoospermia from non‐obstructive azoospermia and predicting sperm retrieval and related histological alterations in men with non‐obstructive azoospermia.