Louis Pin1, Anne-Coline Monseau-Thiburce2, Caroline Ziade-Coularis3, Alice Benjamin4, Fanny Menut4, Jean-Luc Brun5, Benjamin Merlot2, Jean-François Chateil6. 1. Service d'imagerie anténatale, de l'enfant et de la femme, Hôpital Pellegrin, CHU de Bordeaux, Place Amélie Raba- Léon, 33076 Bordeaux Cedex, France. Electronic address: louispin@hotmail.fr. 2. Service d'imagerie médicale, Clinique Tivoli Ducos, 220 rue Mandron, 33000 Bordeaux Cedex, France. 3. Service d'imagerie anténatale, de l'enfant et de la femme, Hôpital Pellegrin, CHU de Bordeaux, Place Amélie Raba- Léon, 33076 Bordeaux Cedex, France. Electronic address: caroline.ziade@chu-bordeaux.fr. 4. Service d'imagerie anténatale, de l'enfant et de la femme, Hôpital Pellegrin, CHU de Bordeaux, Place Amélie Raba- Léon, 33076 Bordeaux Cedex, France. 5. Service de chirurgie gynécologique, Maternité Centre Aliénor d'Aquitaine, Place Amélie Raba-Léon, 33076 Bordeaux Cedex, France. Electronic address: jean-luc.brun@chu-bordeaux.fr. 6. Service d'imagerie anténatale, de l'enfant et de la femme, Hôpital Pellegrin, CHU de Bordeaux, Place Amélie Raba- Léon, 33076 Bordeaux Cedex, France; CRMSB, UMR 5536, CNRS/University of Bordeaux, 146 rue Léo Saignat, 33076 Bordeaux Cedex, France. Electronic address: jean-françcois.chateil@chu-bordeaux.fr.
Abstract
OBJECTIVE: To evaluate the performance of magnetic resonance imaging (MRI) with susceptibility-weighted imaging (SWI) in the assessment of endometriosis. MATERIAL AND METHODS: This prospective study was performed during the diagnostic step or the pre-operative assessment of endometriosis, between June 2017 and April 2018. The MRI was conducted with a 3T MRI device; protocol included T2W, T1W, with and without fat-saturation sequences completed with a SWI sequence: T2-star weighted angiography (SWAN). The diagnostic performance values of MRI and inter-observer agreement were first evaluated with a conventional MR protocol and then with the complementary SWAN sequence by 2 readers. MRI results were correlated with surgical findings in patients who underwent laparoscopy. RESULTS: 74 patients were included in the study, and among them 10 patients were treated by laparoscopy. 81% of the endometriosis lesions had signal losses on the SWAN sequence related to hemorrhagic character whereas only 52% of the lesions had T1-weighted hyperintense implants. Diagnostic performance of the MRI examination was improved by the use of the SWAN sequence compared to the conventional MR protocol (Se = 94% and Spe = 73% in complete protocol and Se = 88% and Spe = 69% in conventional protocol), especially for the involvement of torus uterinus, utero-sacral ligament and retro-cervical site. An excellent interobserver agreement (қ-value = 0,94) was noted between the two readers. CONCLUSION: SWI can improve the diagnostic accuracy of MRI by allowing the detection of hemorrhagic character of endometriosis lesions.
OBJECTIVE: To evaluate the performance of magnetic resonance imaging (MRI) with susceptibility-weighted imaging (SWI) in the assessment of endometriosis. MATERIAL AND METHODS: This prospective study was performed during the diagnostic step or the pre-operative assessment of endometriosis, between June 2017 and April 2018. The MRI was conducted with a 3T MRI device; protocol included T2W, T1W, with and without fat-saturation sequences completed with a SWI sequence: T2-star weighted angiography (SWAN). The diagnostic performance values of MRI and inter-observer agreement were first evaluated with a conventional MR protocol and then with the complementary SWAN sequence by 2 readers. MRI results were correlated with surgical findings in patients who underwent laparoscopy. RESULTS: 74 patients were included in the study, and among them 10 patients were treated by laparoscopy. 81% of the endometriosis lesions had signal losses on the SWAN sequence related to hemorrhagic character whereas only 52% of the lesions had T1-weighted hyperintense implants. Diagnostic performance of the MRI examination was improved by the use of the SWAN sequence compared to the conventional MR protocol (Se = 94% and Spe = 73% in complete protocol and Se = 88% and Spe = 69% in conventional protocol), especially for the involvement of torus uterinus, utero-sacral ligament and retro-cervical site. An excellent interobserver agreement (қ-value = 0,94) was noted between the two readers. CONCLUSION: SWI can improve the diagnostic accuracy of MRI by allowing the detection of hemorrhagic character of endometriosis lesions.