We read with great interest the study by Dr. Wang et al. and commend the authors for their efforts to further elucidate the utility of distal femoral fresh osteochondral allograft transplantation in the revision cartilage restoration setting. Few can deny the controversy that exists in selecting the optimal treatment approach after previous failed attempts at cartilage repair. The power of this study is that it is apparently the first conducted in patients undergoing revision to corroborate improvements in clinical findings with a novel, high-resolution magnetic resonance imaging (MRI) assessment of cartilage appearance and subchondral bone integration, among other important postoperative parameters1. Methodologically, this study is sound, given a critical appraisal of its clear objectives and protocols, explicit inclusion criteria, and thorough, prospective outcomes data collection and analyses of consecutive patients with respectable follow-up rates (i.e., 84% clinical and 70% imaging).
This study was not without limitations. First, the authors analyzed a patient population that was heterogenous in terms of skeletal maturity (age range, 14.6 to 61.9 years), number and type of previous failed cartilage surgical procedures, and proportion of patients undergoing concomitant procedures (44.2%) and/or fresh osteochondral allograft transplantation at multiple distal femoral locations (14.0%). Although we acknowledge the potential increased need for concomitant procedures in the revision setting, it would have been more appropriate to present and interpret the results as 2 cohorts, 1 cohort with isolated fresh osteochondral allograft transplantation (n = 24) and 1 cohort in which the transplantation was not an isolated procedure (n = 19). Moreover, this small sample size was increasingly subject to both the technical bias of 3 surgeons performing fresh osteochondral allograft transplantation procedures and the image-interpretation bias with unspecified combinations of 1.5-T and 3.0-T scans interpreted by only 1 musculoskeletal radiologist. Ideally, and especially for image-related scoring systems, a duplicate, independent assessment with interrater agreement on cartilage-specific MRI sequences (i.e., delayed gadolinium-enhanced MRI of cartilage [dGEMRIC], T1 rho, T2 mapping, and diffusion weighting) is preferred. In addition, despite an exhaustive battery of patient-reported outcomes measures being employed, those arguably most important to the typical fresh osteochondral allograft transplantation patient population (i.e., Lysholm, Tegner Activity, and Knee Injury and Osteoarthritis Outcome Scores) were omitted. Moreover, outcome differences from the cartilage-specific Marx Activity Scale failed to reach significance.
Nevertheless, the authors interpret and present their findings appropriately within the confines of these limitations and are accurate in their assessment that fresh osteochondral allograft transplantation remains a versatile, viable, and effective salvage strategy for revision of failed large cartilage repair attempts. Although short-term to intermediate-term postoperative outcomes were significantly better, with graft survival rates exceeding 90%, it is noted that patients undergoing fresh osteochondral allograft transplantation in the revision setting demonstrated a 39.5% rate of a subsequent surgical procedure and a 9.3% failure rate (i.e., removal of the osteochondral allograft). These rates are on par with the 41.4% reoperation rate and 18.9% failure rate reported by Gracitelli et al.2, the 67% reoperation rate and 39% failure rate reported by Horton et al.3, and the 37% reoperation rate and 13% failure rate reported by Frank et al.4.
Interestingly, this study by Wang et al. appears to be the first in the revision cartilage setting to utilize postoperative MRI scans in addition to clinical outcomes. We find many parallels with regard to current fresh osteochondral allograft transplantation controversies and our experience with patients after anterior cruciate ligament (ACL) reconstruction, particularly with regard to clearance for return to sport. In patients who undergo ACL reconstruction, return-to-sport clearance varies from 3 months to 1 year postoperatively and, ideally, should occur when the graft has fully matured and is able to tolerate sport-specific loads. At the University of Pittsburgh, we are the first, and one of the few globally, to perform postoperative MRI assessments to assess graft ligamentization and tunnel incorporation as part of the clearance assessment for return to elite sports5. Moreover, although 92.5% of patients in this study demonstrated persistent marrow edema patterns on 1-year postoperative MRI, we acknowledge that a lack of consensus exists with regard to optimal postoperative MRI intervals for fresh osteochondral allograft transplantation and that the long-term MRI response postoperatively is unknown1. As in the ACL, the presence of bone bruises or edema has not been associated with poorer outcomes6 and, in patients who have undergone fresh osteochondral allograft transplantation, may represent a normal finding1. Thus, we applaud the emphasis that these authors place on highlighting biological outcomes in addition to the clinical and kinematic ones.
However, the mean total Osteochondral Allograft MRI Scoring System (OCAMRISS) score did not correlate with clinical outcome scores. Although the OCAMRISS represents a valid and reliable measure, its applicability in the revision setting remains unknown. One must be cognizant that 4 of 13 items in OCAMRISS are of questionable relevance in the revision setting (i.e., opposing cartilage status, meniscal tears, synovitis, fat pad scarring), and their inclusion likely tempers any outcomes differences. This is evidenced by findings of superior clinical outcomes when analyzing patients solely by osseous and cartilaginous parameters. As such, this study highlights the need for both improved objective outcomes measurements and a more meticulous search for differences, achieved perhaps by incorporating the diagnostic arthroscopic evaluation data for the 40% of cases that underwent reoperation.
In summary, this is a well-designed case series that can inform surgeon practice and can generate further research efforts. We believe that the authors effectively demonstrate that fresh osteochondral allograft transplantation can and should be reliably offered as a salvage option for the revision of large failed cartilage repairs. Health-care providers and patients should understand that there is an approximate 40% reoperation risk in the short to intermediate term. As such, one should recognize that, although clinical improvements and high graft survival are expected, persistent MRI evidence of bone edema up to 2 years postoperatively may exist, the importance of which is unknown.
1. Meric G, Gracitelli GC, McCauley JC, Pulido PA, Chang EY, Chung CB, Bugbee WD. Osteochondral allograft MRI scoring system (OCAMRISS) in the knee: interobserver agreement and clinical application. Cartilage. 2015 Jul;6(3):142-9.
2. Gracitelli GC, Meric G, Pulido PA, McCauley JC, Bugbee WD. Osteochondral allograft transplantation for knee lesions after failure of cartilage repair surgery. Cartilage. 2015 Apr;6(2):98-105.
3. Horton MT, Pulido PA, McCauley JC, Bugbee WD. Revision osteochondral allograft transplantations: do they work? Am J Sports Med. 2013 Nov;41(11):2507-11. Epub 2013 Aug 27.
4. Frank RM, Lee S, Levy D, Poland S, Smith M, Scalise N, Cvetanovich GL, Cole BJ. Osteochondral allograft transplantation of the knee: analysis of failures at 5 years. Am J Sports Med. 2017 Mar;45(4):864-74. Epub 2017 Jan 5.
5. Rabuck SJ, Baraga MG, Fu FH. Anterior cruciate ligament healing and advances in imaging. Clin Sports Med. 2013 Jan;32(1):13-20. Epub 2012 Oct 4.
6. Lattermann C, Jacobs CA, Reinke EK, Scaramuzza EA, Huston LJ, Dunn WR, Spindler KP. Are bone bruise characteristics and articular cartilage pathology associated with inferior outcomes 2 and 6 years after anterior cruciate ligament reconstruction? Cartilage. 2017 Apr;8(2):139-45. Epub 2016 Jul 8.