The purpose of this study was to investigate the role of preoperative bone marrow lesion (BML) size and location on (1) postoperative patient reported outcomes and (2) postoperative failure and time to failure after osteochondral allograft (OCA) transplantation. Consecutive patients from 2 senior surgeons who underwent isolated OCA transplantation to the knee from 2009-2018 were identified for the case series. Preoperative magnetic resonance imaging (MRI) was evaluated for BMLs based on 2 classification systems (Welsch etal. and Costa-Paz etal.) by 2 independent graders. BMLs associated with minimum 1-year postoperative outcomes were evaluated, and the effect of BML classification on survivorship was investigated with Kaplan-Meier curves. The 77 patients who underwent isolated OCA transplantation (mean follow-up: 39.46 ± 22.67 months) and had preoperative MRIs were included. Within this cohort, 82% of patients demonstrated a BML. The preoperative Costa-Paz etal. classification was significantly positively correlated with the postoperative Visual Analog Scale, International Knee Documentation Committee and Veterans RAND 12-Item Health Survey raw scores for both graders (P < 0.05). Failure occurred in 5 of 65 (8%) patients at a mean of 22.86 ± 12.04 months postoperatively. The presence of BML alone did not significantly affect survival (P=0.780). However, for 1 grader, the Welsch etal. classification was associated with increased risk of graft failure (P=0.031). Preoperative subchondral BMLs were present in 82% of patients undergoing OCA transplantation. We found that more severe BMLs based on the Costa-Paz classification, with increasing involvement in the juxta-articular surface, were correlated with higher postoperative patient-reported functional outcomes after OCA. BMLs may be associated with an increase in graft failure, but their role in this remains unclear. IV, Retrospective Case Series.
Read full abstract