2.1 Search strategy and eligibility criteria

Using Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) guidelines,²² PubMed and Embase searches were performed to include peer-reviewed articles from inception (1977) to February 2025 to identify observational clinical studies evaluating outcomes of donor-recipient matching practices for OCAT. Using the advanced search features, keywords of (1) “osteochondral allograft transplantation”, (2) “match OR donor OR recipient”, and (3) “outcomes OR survival OR failure OR revision” were combined. All titles and abstracts from the initial search were screened, eliminating articles irrelevant to this review. Following this, a full-text review of the remaining articles was performed by 2 independent reviewers (B.W.R. and J.P.) to determine eligibility for inclusion. Inclusion criteria included ≥2 years of follow-up data comparing OCAT outcomes between a cohort of tissue recipients with a matched donor variable to a cohort with the corresponding mismatched donor variable. Exclusion criteria included studies unavailable in English or those without standardized outcome measures. Following the database search, the bibliographies of eligible studies were screened by 2 independent reviewers (B.W.R. and J.P.) for identification of additional studies of relevance, which underwent eligibility assessment as previously mentioned. When eligibility of an article was in question following initial review, inclusion was determined by a third reviewer (K.R.).

2.2 Data Extraction

Eligible articles were broadly categorized as those evaluating donor-recipient matching practices according to demographic, immunologic, or morphometric variables. Following a generalized Population, Intervention, Comparison, Outcome, Time format, data extracted from eligible articles included:•Population: sample size, patient age, donor age, sex, BMI, smoking status•Intervention: graft size, defect diameter•Comparison: matched vs. mismatched donor-recipient variable•Outcome: graft survival rate, patient reported outcome measures (PROMs), radiographic outcome measures•Time: follow-up duration

2.3 Risk of bias assessment

Eligible articles underwent a risk of bias assessment using the Cochrane Library Risk Of Bias In Non-randomized Studies – of Interventions, Version 2 tool (ROBINS-I V2) by 2 independent reviewers (B.W.R. and J.P.). The ROBINS-I V2 tool assesses bias as low, moderate, serious, or critical according to 7 domains: confounding, selection of participants, classification of interventions, deviation from intended intervention, missing data, measurement of outcomes, and selection of reported results. Discrepancies in individual domains of bias as well as final scores was resolved by a third reviewer (K.R.).

3.1 Search results and risk of bias assessment

The primary search yielded 258 articles. Of these, 7 met inclusion criteria and after bibliography reviews, 2 additional articles were identified for inclusion, resulting in 9 total articles for systematic review (Fig. 1). These 9 articles were categorized as studies evaluating evidence of donor-recipient matching practices according to demographic variables (n = 3), immunologic variables (n = 3), or morphometric variables (n = 3).

Close

Fig. 1 PRISMA flow chart outlining the process and results of the systematic review search strategy.

Export to PPT

All included studies were determined to have a degree of bias with failure to control for confounding variables (Domain 1), and missing data (Domain 5) being denoted as moderate sources of bias across most studies. No study was determined to have a low risk of bias, 7 were determined to have moderate risk of bias, 2 had serious risk of bias, and none were considered to have a critical risk of bias (Table 1).

3.2 Demographic variables

Three clinical cohort studies evaluated donor-recipient demographic variables for OCAT matching with 2 assessing sex-matching and 1 assessing age- and sex-matching. In a study evaluating 372 patients undergoing OCAT involving single or multiple plug grafts in the same or different compartments of the knee, Hevesi et al.¹⁴ reported a range in 5-year survivorship among female-to-male (78.8%), male-to-male (87.9%), male-to-female (90.1%), and female-to-female (96.3%) transplants, with no significant differences among cohorts (p = 0.569). In addition, donor-recipient sex-mismatch was not predictive of achieving a Patient Acceptable Symptom State (PASS).²³ In contrast, Merkely et al.¹⁶ reported a significantly lower 5-year graft survival rate for male-to-female versus male-to-male transplants (64% vs. 44%, p = 0.04) in a study assessing single or multiple plug grafts in the same or different compartments of the knee in 154 patients. However, findings from Williams et al.¹⁵ supported the Hevesi study in reporting no significant differences (p = 0.324) in graft survivorship among the different sex-matched and -mismatched cohorts for single or multiple plug or shell OCATs in the knee in 170 patients. Williams et al.¹³ also assessed donor-recipient age-mismatches of 15 years or more in this same patient population and found no significant differences in outcomes (p = 0.398) between age-matched and -mismatched cohorts.

The mechanisms by which donor-recipient sex-mismatches may impact OCAT outcomes include anatomic and material property discrepancies, hormonal influences, and immune responses. While the role of minor histocompatibility antigens such as H-Y antigens encoded on the Y-chromosome has been cited as a risk factor for male-to-female mismatches in solid organ transplants, there is no such evidence for risk for OCAT.^24,25 Similarly, sex-based differences in anatomy, articular cartilage thickness, and joint loading characteristics have not been associated with OCAT treatment failures. Taken together, the existing literature does not support the need for donor-recipient sex-matching for safe and effective OCAT in the knee.

Similarly, donor-recipient age-mismatching does not appear to be a significant risk factor for OCAT treatment failure.^14,15,26 However, only 1 study directly assessed age-mismatching, and only included data for differences of recipients ≥15 years older than donors versus donor-recipient pairs with age differences <15 years.¹³ The paucity of data and impact for age-mismatching is likely related to the combination of tissue banks imposing a maximum OCA donor age of 35 years and most OCAT indications being limited to patients 55-years-old or younger.^23,27,28 While the current evidence suggests that donor age does not need to influence OCAT matching protocols, mismatching beyond 15 years, as well and changes to maximum donor age cut-offs, would need to be evaluated prior to clinical implementation.

3.3 Immunologic variables

Three clinical cohort studies evaluated donor-recipient immunologic variables for OCAT matching. Luk et al.¹⁷ evaluated blood antigen matching among 103 patients undergoing plug or shell OCAT. This study reported no significant differences in proportions of survival or failure based on donor-recipient mismatching of ABO blood type (p = 0.821), rhesus (Rh)-factor (p = 0.352), or both combined (p = 0.412). Additionally, among tissues recovered from 18 failed OCATs in this patient population, no significant differences were reported for the presence or absence of a histological immune response based on mismatching of ABO blood type (p = 0.312), Rh-factor (p = 1), or both combined (p = 0.312). However, immunohistochemical assessments suggested that T- and B-cell-mediated subrejection immune responses related to residual donor bone marrow elements and independent of donor–recipient blood type mismatching may be involved in OCAT treatment failures.¹⁷ In the study by Hunt et al.,¹⁸ 67 patients who underwent single or multiple plug or shell OCAT in 1 or more compartments of the knee were assessed for the presence or absence of anti-human leukocyte antigen (HLA) antibodies. While OCA size (>10 cm²) was strongly associated with the presence anti-HLA antibodies postoperatively, no significant differences (p = 0.152) in graft survival related to the presence of anti-HLA antibodies were noted. In addition, no significant differences (p = 0.482) in postoperative Knee Society function scores were found between positive and negative anti-HLA antibody cohorts. Sirlin et al.¹⁹ analyzed anti-HLA antibody status with regard to magnetic resonance imaging findings following OCAT in 36 patients receiving single or multiple shell grafts in 1 or more compartments of the knee. Scoring by 2 blinded radiologists indicated greater edema (p < 0.002), thicker interface (p < 0.03), more abnormal graft marrow (p < 0.04), and higher proportion of surface collapse (p < 0.03) in anti-HLA antibody positive patients compared to anti-HLA antibody negative patients in the study population.

Taken together, the current evidence regarding immunologic variables for OCAT donor-recipient matching suggest that donor bone marrow elements may be associated with cell-mediated and humoral subrejection responses that affect OCA osteointegration.¹⁷ The effects of ABO and/or Rh-factor incompatibility appear to be negligible in practice.^6,29–31 However, the impact of HLA mismatching in OCAT remains unclear. Hunt et al.¹⁸ documented a 15% higher OCAT survival rate in their anti-HLA antibody-negative cohort, which would certainly be considered clinically meaningful. In addition, the diagnostic imaging findings reported by Sirlin et al.¹⁹ suggests that anti-HLA antibody responses are associated with adverse characteristics of OCATs that may increase risk for treatment failure. Donor-recipient HLA mismatches can trigger production of class-I cytotoxic antibodies, extending the inflammatory phase and elevating degradation mediators that diminish the pro-osteogenic microenvironment necessary for osteointegration.^6,32–35 Importantly, these responses appear to be primarily driven by vascular endothelium, blood, and marrow cells in donor bone and are OCA volume-dependent.^18,29 Therefore, while HLA matching may require further investigation for the largest volume OCATs, preimplantation techniques including minimization of OCA bone thickness and thorough irrigation appear to be safe and effective for mitigating burden of immunologic mismatch risk.^36–38 As such, the synthesis of this evidence supports the current practice of exclusion of immunology-based matching while strictly adhering to evidence-based preimplantation and transplantation techniques that reduce donor “bioburden” and optimize OCA osteointegration.

3.4 Morphometric variables

Three clinical cohort studies evaluated donor-recipient morphometric variables for OCAT matching. Gursoy et al.¹¹ evaluated the effects of curvature matching in the sagittal plane on functional outcomes for single plug OCATs in one compartment of the knee for matched curvature, low convexity, and high convexity cohorts. Patient-reported outcome measures (PROMs) (Tegner, IKDC, KOOS and SF-12 Physical and Mental Health scores) were significantly lower (inferior) for the high convexity cohort through 2 years following OCAT. There were no significant differences in PROMs between the matched curvature and low convexity cohorts. Wang et al.¹² evaluated the effects of AP and ML mismatches between donors and recipients as continuous variables for single or multiple plug femoral condyle OCATs. Among 69 patients (50 non-failures, 19 failures), there were no significant differences in AP or ML mismatches between non-failures and failures. Additionally, the magnitude of AP mismatch was not associated with differences in functional outcomes. Wang et al.¹³ also assessed outcomes related to donor-recipient condyle laterality matching in 77 patients undergoing single or multiple plug OCATs in 1 compartment of the knee, reporting no significant differences in survival between condyle-matched (86%) and condyle-mismatched (91%) cohorts. In addition, reoperation rates and functional outcomes were not significantly different between cohorts.

At this time, the data regarding donor-recipient morphometric variable matching for OCAT are too sparse and disjointed to establish evidence-based guidelines. In general, major curvature mismatching appears to have importance with respect to functional outcomes, at least for femoral condyle OCAT.¹¹ Bernstein et al.³⁹ reported that use of the radius of curvature as a matching criterion for femoral condyle OCATs significantly increased the number of available matches and reduced wait times for patients when compared to anatomic, laterality, and dimensional methods of morphometric matching. However, this is only applicable for femoral condyle OCATs, it entails a more complicated method of matching between surgeons and tissue banks, and current evidence suggests that anatomic, laterality, and dimensional methods are sufficient for avoiding major mismatches that are not well-tolerated.^40–42 As such, morphometric matching protocols for OCAT should continue to focus on efficient and effective use of donated tissues while further research targets novel methods that are user-friendly for tissue banks and surgeons to further optimize functional outcomes for plug and shell grafts in all anatomic locations indicated for OCAT.