Sagar Yadav, Udit Khurana, Varun Sharma, Shubham Singh Bais, Aakash Yadav, Krishn Kant Bhardwaj
MRI-DWI shows good-to-excellent accuracy and can support localization, operative planning, and counselling. However, a negative MRI-DWI result does not safely exclude the presence of a viable or otherwise at-risk undescended testis and leaves a clinically meaningful residual probability of disease. Therefore, negative imaging should not be used to omit operative exploration. MRI-DWI should be regarded as an adjunctive localisation tool rather than a replacement for diagnostic laparoscopy, which remains necessary for definitive evaluation.
PURPOSE: Non-palpable undescended testes require accurate localisation, but the role of magnetic resonance imaging (MRI) combined with diffusion-weighted imaging (DWI) remains uncertain. We evaluated the diagnostic accuracy of MRI-DWI for detecting clinically non-palpable undescended testes.
METHODS: This PROSPERO (CRD420261411862) registered review searched PubMed, Embase, Web of Science, and CENTRAL to June 6, 2026. Eligible diagnostic-accuracy studies assessed MRI-DWI against diagnostic laparoscopy or operative exploration. Two reviewers extracted 2 × 2 data, and assessed QUADAS-2 risk of bias. Sensitivity and specificity were pooled using bivariate random-effects models, with hierarchical summary receiver operating characteristic (HSROC) analysis, likelihood ratios, diagnostic odds ratios, and Fagan nomograms used to summarise diagnostic performance.
RESULTS: Fifteen studies were included. In seven testis-level studies (340 testes), pooled sensitivity and specificity were 92.6% (95% CI 88.4-95.3%) and 94.3% (85.6-97.9%) and the HSROC area under the curve (AUC) was 0.957. In eight patient-level studies (1782 patients), pooled sensitivity and specificity were 86.7% (77.1-92.7%) and 85.8% (68.8-94.3%) and the HSROC AUC was 0.917. Positive MRI-DWI increased post-test probability on the Fagan nomogram, but a negative result left a residual probability of 21%. Heterogeneity was minimal at the testis level but substantial at the patient level analysis. Deeks' test suggested small-study effects only in the testis-level analysis.
CONCLUSIONS: MRI-DWI shows good-to-excellent accuracy and can support localization, operative planning, and counselling. However, a negative MRI-DWI result does not safely exclude the presence of a viable or otherwise at-risk undescended testis and leaves a clinically meaningful residual probability of disease. Therefore, negative imaging should not be used to omit operative exploration. MRI-DWI should be regarded as an adjunctive localisation tool rather than a replacement for diagnostic laparoscopy, which remains necessary for definitive evaluation.