Non-invasive ultrasound localization microscopy (ULM) in azoospermia: connecting testicular microcirculation to spermatogenic functions

Abstract

Rationale: Azoospermia is a significant reproductive challenge. Differentiating between non-obstructive azoospermia (NOA) and obstructive azoospermia (OA) is crucial as each type requires distinct management strategies. Testicular microcirculation plays a profound role in spermatogenic functions. However, current diagnostic methods are limited in their ability to effectively elucidate this crucial connection. Methods: We employed ultrasound localization microscopy (ULM) to visualize testicular microcirculation in NOA and OA patients and quantified the testicular hemodynamic parameters. Pearson correlation analysis was conducted to investigate the inner connection between parameters of testicular microcirculation and clinical spermatogenic functions. We conducted multiple logistic regression analysis to establish a new diagnostic model that integrates follicle-stimulating hormone (FSH) and mean vascular diameter to distinguish NOA from OA. Results: Our findings demonstrated significant differences in vascular parameters between NOA and OA, with NOA characterized by lower mean vascular diameter (p < 0.001), vessel density (p < 0.001), and fractal number (p < 0.001). Testicular volume showed a moderate positive correlation with mean vascular diameter (r = 0.419, p < 0.01) and vessel density (r = 0.415, p < 0.01); Mean vascular diameter exhibited negative correlations with both FSH (r = -0.214, p < 0.05) and age (r = -0.240, p < 0.05); FSH (r = -0.202, p < 0.05) and luteinizing hormone (LH) (r = -0.235, p < 0.05) were negatively correlated with mean blood flow velocity. The diagnostic model demonstrated an area under the curve (AUC) of 0.968. We also reported a method to map the vascular pressure distribution derived from the blood flow velocity generated by ULM. Conclusions: ULM provides a non-invasive and detailed assessment of testicular microvascular dynamics. The ULM-derived vascular parameters are able to connect testicular microcirculation to spermatogenic functions. The combination of FSH and mean vascular diameter enhances diagnostic precision and holds potential for distinguishing NOA from OA.

Keywords: azoospermia; pressure distribution maps; spermatogenic functions; testicular microcirculation; ultrasound localization microscopy (ULM).

Figure 1

The processing pipeline for generating super-resolution images from contrast-enhanced ultrasound (CEUS) data. The process includes singular value decomposition (SVD), two-stage image registration, motion correction, and super localization and tracking to produce high-quality super-resolution images. MB: microbubble.

Figure 2

Comparative ultrasonographic assessment of human testes in azoospermia patients. The top row shows images from a patient with non-obstructive azoospermia (NOA), and the bottom row depicts images from a patient with obstructive azoospermia (OA). From left to right: (A) & (E) B-mode ultrasound, (B) & (F) color Doppler flow imaging (CDFI), (C) & (G) contrast-enhanced ultrasound (CEUS), and (D) & (H) maximum intensity projection (MIP) images. To facilitate an intuitive comparison, the NOA images were horizontally mirrored. Scale bars = 5 mm.

Figure 3

Ultrasound localization microscopy (ULM) of testicular microvasculature in patients with non-obstructive azoospermia (NOA) and obstructive azoospermia (OA). The top panel represents NOA and the bottom panel represents OA. Each panel includes a direction map on the left, displaying the orientation of blood flow (red indicates microbubble flow toward the transducer, while blue signifies flow away from the transducer), and a velocity map on the right, showing the speed of blood flow within the testicular tissue. Highlighted regions of interest (ROIs) within white boxes were magnified to provide a closer view of the vascular patterns and flow dynamics characteristic of each condition. MB: microbubble. Scale bars = 1 cm.

Figure 4

Comparison of testicular microcirculation manifestations in non-obstructive azoospermia (NOA) and obstructive azoospermia (OA) patients using ultrasound localization microscopy (ULM). (A) & (B) Normalized vessel intensity profiles along a selected distance for NOA and OA, respectively, with the selected distance marked by white lines in the magnified regions shown in Figure 3. (C) & (D) Frequency spectra generated using Fourier transform in NOA and OA. (E) & (F) Vascular pressure distribution maps derived from super-resolution velocity maps in NOA and OA testes. Scale bars = 5 mm.

Figure 5

Quantitative assessment of hemodynamics in testicular microcirculation of all patients in non-obstructive azoospermia (NOA) and obstructive azoospermia (OA) groups. (A) Five vascular-related quantitative parameters were extracted from the SR images of the testes. (B-F) Violin plots depicting the distribution of vessel density, fractal number, mean velocity, mean tortuosity, and mean diameter of testicular microvasculature. The dashed lines represent quartiles, the solid lines represent the median, and the dots represent the individual values of each sample. Asterisks indicate statistically significant difference in means (results calculated as shown in Table 1). ^*** indicates p < 0.001. (G-I) Histograms of the mean distribution of velocity, tortuosity, and diameter in the testes of all NOA and OA patients.

Figure 6

Schematic diagram of testicular regional division and quantitative comparison of microvascular parameters in each region for all non-obstructive azoospermia (NOA) and obstructive azoospermia (OA) patients. (A) Segmented regions of a testis, with different colors representing various zones (1-12) used for analysis, and the scale indicating the corresponding regions. (B-F) Graphs showing mean velocity, mean tortuosity, mean diameter, vessel density, and fractal number in different regions for NOA and OA.

Figure 7

Heatmap of Pearson correlation coefficient matrix representing the relationships among various parameters of spermatogenic functions and testicular microcirculation in azoospermia patients. Each cell displays the Pearson correlation coefficient between pairs of variables, where 1.0 indicates a perfect positive correlation, 0 indicates no correlation, and -1.0 indicates a perfect negative correlation. The colors range from red (positive correlation) to blue (negative correlation). FSH: follicle-stimulating hormone; LH: luteinizing hormone; T: testosterone. ^*** indicates p < 0.001, ^** indicates p < 0.01, ^* indicates p < 0.05.

Figure 8

Receiver Operating Characteristic (ROC) curves demonstrating the diagnostic performance of different parameters in distinguishing between NOA and OA. (A) Clinical parameters: FSH, testicular volume, age, LH, and T. (B) Ultrasound localization microscopy (ULM)-derived vascular parameters: mean diameter, vessel density, fractal number, mean tortuosity, and mean velocity. (C) A multiple logistic regression model combining FSH and mean diameter showed the highest accuracy, with an AUC of 0.968. FSH: follicle-stimulating hormone; LH: luteinizing hormone; T: testosterone; AUC: area under the curve. ^*** indicates p < 0.001, ^** indicates p < 0.01.