Assessing 3D T2 FSE sequence for identification of the normal appendix: working toward a single-sequence MR appendicitis protocol

Purposes: To assess the ability of a single isotropic 3D T2 FSE sequence to identify the normal appendix in children with potential implication as alternative second-line modality in pediatric appendicitis. Materials and methods: Retrospective review of MR abdomino-pelvis or pelvis in children from Oct 2014–Dec 2016 was done. Only exams with 3D T2 FSE sequence performed on a single scanner type were selected. Patients with history of post appendectomy, studies in which field of view did not include ileocecal valve, with technical and protocol errors, or with fat-saturated acquisition were excluded. All images were blindly reviewed by three radiologists. The following questions were asked: (1) Is the appendix visualized?, (2) What is the most clearly demonstrable plane?, (3) Where is the appendix located?, (4) Are there any findings indicating appendicitis? Average pairwise percentage and Fleiss Kappa were used to assess the inter-rater agreement on the visualization and location of the appendix. Results: A total of 22 MRI studies were reviewed (F = 13:M = 9, Mean age = 11.45 yrs. [SD = 4.3]). Total MRI scan time was estimated at 6 mins. Readers saw appendix in 72.7–81.8%, and saw with high confidence in 12 cases (54.5%), low confidence in 4–6 case (18.2–27.3%), and did not see in 4–6 cases(18.2–27.3%). The average pairwise percentage agreement was 71.2%. Fleiss Kappa of overall rating of visualization was 0.52 (p < 0.001). Fleiss Kappa of high confidence of visualization was 0.76 (p < 0.001). Reader 1 and reader 2 saw the appendix best on the axial plane. Reader 3 saw the appendix best on the coronal plane. For location, deep pelvis was the most common identified location. The average pairwise percentage agreement was 68.2% and Fleiss Kappa was 0.76 (p < 0.001). There is no findings indicating appendicitis. Conclusion: A single 3D T2 FSE sequence had good ability to detect normal appendices and also allowed multiplanar reconstructions, which may use as a single-sequence protocol in cases of suspected appendicitis in children when ultrasound is inconclusive.