Sequential MRI Study of Graft Integrity and Signal Following Pediatric All-epiphyseal ACL Reconstruction: Does the “Sharp Turn” at the Socket of the Distal Femoral Aperture Matter?

Abstract

Objectives: To analyze graft structure and signal with particular emphasis on the distal femoral socket aperture following all-epiphyseal ACLR using hamstring autografts with sequential MRI in skeletally immature athletes. Methods: Retrospective cohort study of 23 skeletally immature patients who underwent ACLR by the same surgical team at a tertiary center during 2011-2013. Athletes had at least two follow-up MRIs, the first MRI 6-12 months after surgery and the second MRI >18 months, were included. Exclusion criteria included those athletes with inMRI follow-up (6) or with a failure of their reconstructions (1). All athletes were treated with an arthroscopic all-inside, allepiphyseal ACLR, using hamstring autograft, secured with adjustable loop cortical buttons on both tibia and femur. MRI images were analyzed independently and blinded by an orthopaedic surgery fellow and a musculoskeletal radiology fellow. Using a GE Functional Analysis Software, the signal intensity (SI) of the graft was measured in 5 different locations: 1) femoral tunnel, 2) intra-articular proximal turn, 3) midsubstance, 4) intra-articular distal turn, and 5) tibial tunnel. Values were normalized to cortical bone density. The amount of perigraft scarring and synovitis was analyzed. An intraclass correlation coefficient was used to quantify inter-rater reliability, non-parametric Wilcoxon test for perigraft scarring and synovitis, one-way ANOVA to test if significant differences of SI were seen between the different graft locations, and a 2-tailed student t-test for SI changes from 1st to 2nd MRI. Results: The study included 16 patients (5 girls and 11 boys), with an average age at surgery of 11.9 years (range 10-15). The first follow-up MRI was on average at 8.4 months (range 6-12 months), while the 2nd MRI was on average 30.7 months (range 18-40) after surgery. Intra-class correlation coefficients were above 0.7 for all measurements, indicating an excellent concordance between observers. Perigraft scarring tended to reduce with follow-up (p=0.057) though not significantly, while synovitis had a significant reduction over time (p=0.01). On average, normalized SI showed no significant differences between measurements taken in different regions of the graft (p=0.58). When comparing the graft SI from 1st to 2nd MRI, no significant differences were found in any of the locations: femoral tunnel (p=0.14), proximal turn (p=0.11), midsubstance (p=0,29), intra-articular distal (p=0.10), or tibial tunnel (p=0.15). All 16 athletes returned to their prior sport at the same level of performance without re-injury. Conclusion: ACL grafts in skeletally immature patients with all-epiphyseal reconstructions maintain a stable intensity signaling at long term MRI follow-up, with no significant signal reduction over time. Despite the sharp turn created at the distal femoral socket aperture in physeal-sparing reconstructions, no particular anatomic location of the graft presents significantly different signal intensity over others. This is the first sequential mri study in pediatric epiphyseal acl reconstructions demonstrating postoperative maintenance of graft integrity and graft signal.