Added value of gadolinium-based contrast agents for magnetic resonance evaluation of adnexal torsion in girls

• Published in: Abdominal imaging Vol. 47; no. 11; pp. 3868 - 3882
• Main Authors: Elsingergy, Mohamed M., Oliver, Edward R., Otero, Hansel J., Back, Susan J.
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.11.2022; Springer Nature B.V
• Subjects: Adnexal Diseases - diagnostic imaging; Adnexal Diseases - surgery; Adolescent; Child; Contrast agents; Contrast Media; Cysts; Evaluation; Female; Gadolinium; Gastroenterology; Hepatology; Humans; Image contrast; Imaging; Intravenous administration; Lesions; Magnetic Resonance Imaging; Magnetic Resonance Spectroscopy; Medical imaging; Medicine; Medicine & Public Health; Ovarian Cysts; Ovarian Neoplasms; Ovarian Torsion; Ovaries; Patients; Pediatrics; Pelvis; Radiology; Retrospective Studies; Torsion Abnormality - diagnostic imaging; Torsion Abnormality - surgery; Ultrasonic imaging; Ultrasound; Ovary; MRI; Gadolinium; Children; Pelvis; Adnexa; Torsion; Girls
• ISSN: 2366-0058; 2366-004X; 2366-0058
• DOI: 10.1007/s00261-022-03642-x

Purpose Ultrasound is the first-line imaging modality to evaluate adnexa in girls with clinical suspicion of torsion. Patients with equivocal ultrasound findings can undergo MRI for better delineation of adnexal pathology. Here, we assess the utility of intravenous contrast in MRI evaluation of adnexal torsion in children. Methods Two pediatric radiologists (R1, R2) retrospectively reviewed 198 pelvic MRI exams in 172 girls (median age 15 years). Each MRI was reviewed twice. The first review included pre-contrast images only. A second review, at least 1 month later, included both pre- and post-contrast images. Readers concluded if findings were suspicious for torsion or not. Readers’ findings were compared to each other’s and to surgical and MRI reports and clinical course. Results 198 MRI exams yielded 354 evaluable ovaries. Surgical and pathological reports were available for 47 patients. 11 patients had adnexal torsion. Both readers accurately diagnosed acutely torsed ovaries during pre- and post-contrast reviews ( n  = 4). However, readers disagreed on torsed paraovarian cysts ( n  = 4) and chronically/intermittently torsed ovaries ( n  = 3). In 21 non-torsed ovaries that had lesions, one or both readers concluded that there were pre-contrast features of torsion. In this set with ovarian lesions, contrast helped readers to correctly conclude no torsion ( R 1 = 8, R 2 = 6) more commonly than to incorrectly conclude torsion (1 each), improving post-contrast specificity for each reader. Conclusions Post-contrast sequences did not provide additional benefit in evaluating acutely torsed ovaries but helped in excluding torsion in patients with adnexal lesions. Therefore, contrast administration should be individualized, potentially reserved only for those with abnormal ultrasound or pre-contrast images. Graphical abstract