Background: Benign prostate syndrome (BPS) is the clinical manifestation of benign prostatic hyperplasia (BPH), a widespread disease of the male population. Prostate artery embolization (PAE) is a young, minimally invasive therapy method in which the vessels supplying the prostate are closed with embolic material. This cuts off the blood supply to the embolized tissue and subsequently necrotizes it. This infarction can be detected via a reduction in diffusion in magnetic resonance imaging (MRI). Due to the technically demanding procedure and the complexity of the prostatic vessels, magnetic resonance angiography (MRA) is attracting more and more attention for the radiation-free visualization of the course of the vessels. The aim of this doctoral thesis was to uncover possible predictive factors for the therapeutic success of PAE and to compare intervention planning with the help of MRA and without MRA. Material and Methods: In this retrospective analysis 259 patients who received PAE at the Institute for Diagnostic and Interventional Radiology of the University Hospital Frankfurt am Main were included. The collective was divided into two groups depending on the presence of a pre-interventional MRA. In 137 patients, the PAE was planned with the help of a preinterventional MRA reconstruction (= group A). In 122 patients, PAE was performed without MRA reconstruction (= group B). The data before and after PAE were evaluated using the International Prostate Symptom Score (IPSS), International Index of Erectile Function (IIEF) and Quality of Life (QoL) as well as MRI measurements (diffusion coefficient, prostate volume). Descriptive data analysis, the Wilcoxon-matched-pairs-test, the Wilcoxon-Mann-Whitney-U-test for non-parametric variables as well as Spearman's correlation matrix were used as statistical methods. Results: In group A, a significantly higher median prostate volume reduction (rho = 0.1827, p = 0.032448) and a reduction in radiation dose (p = 0,000000) was achieved compared to group B. In group A, a significantly higher decrease in the diffusion coefficient was observed than in group B (p = 0.036303). An infarcted area after PAE was reached in 59.3% (n = 67) of group A patients and in 40.7% (n = 46) of group B patients. Within group A, the initially evaluated diffusion coefficient correlated with the preinterventional IPSS (rho = 0.3259, p = 0.0213). Furthermore, when using MRA, the diffusion coefficient measured before treatment and the IPSS correlated with the decrease in symptoms (rho = 0.3138, p = 0.0285; rho = 0.6723, p = 0.0000) and the volume reduction (rho = 0.3690, p = 0.0024; rho = 0.4681, p = 0.0020). The reduction in the diffusion coefficient correlated with the severity of the initial symptoms (rho = 0.3703, p = 0.0127) and with the improvement in symptoms (rho = 0.3830, p = 0.0107). Conclusion: In a synopsis of all results, the pre-interventional measurement of the diffusion coefficient, IPSS survey and prostate volumetry in combination with the MRA are prognostically meaningful. What the present study achieves are initial findings in an area that has so far hardly been studied in humans. This radiation-saving prognostic approach could contribute to better patient selection and create further approaches for the use of artificial intelligence. This could minimize radiation dose, costs and risks.
