Outcomes of Restorative Treatment for Isolated Patellar Cartilage Lesions: A Systematic Review and Meta-analysis Authors: Jelle P van der List, Tarik Taoufik, David C Flanigan, Robert A Magnussen, Nicholas A Apseloff Published: 2026 ABSTRACT AbstractBackground:Patellar cartilage lesions often cause debilitating symptoms, and several treatments exist. However, most review studies have combined patellar and trochlear lesions.Purpose:To assess the outcomes of different treatments of only patellar cartilage lesions.Study Design:Systematic review; Level of evidence, 4.Methods:The PubMed, Embase, and Cochrane Library databases were searched in March 2025 for clinical studies reporting outcomes of advanced cartilage treatments beyond microfracture or chondroplasty. The inclusion criteria consisted of isolated patellar lesions (no trochlear or bipolar lesions), regardless of concomitant procedures, with a minimum 1-year follow-up and at least 10 patients. Relevant outcomes were failure, reoperation, and complication rates, and patient-reported outcome measures (PROMs). Random-effect models were used, and outcomes were reported with 95% CIs.Results:A total of 36 studies (1058 patients; mean age, 32 years; follow-up, 5.1 years) were included. Most patients underwent (matrix-induced) autologous chondrocyte implantation (MACI/ACI; mean lesion, 4.3 cm2), followed by osteochondral autograft transfer (OAT/mosaicplasty; mean lesion, 1.5 cm2) and autologous matrix-induced chondrogenesis (AMIC; mean lesion, 2.9 cm2). Most studies were level 3 or 4 evidence and published since 2010. MACI/ACI resulted in an overall failure risk of 8.8% (4.4%-14%), a Lysholm score of 72.8 (68.7-76.8), a Kujala score of 74.8 (66.9-82.8), and an International Knee Documentation Committee (IKDC) score of 65.3 (60.7-69.9). OAT/mosaicplasty resulted in an overall failure risk of 1.7% (0.4%-6.7%), a Lysholm score of 82.7(76.1-89.3), a Kujala score of 77.6 (75.3-79.8), and an IKDC score of 84.2 (65.4-103). AMIC resulted in a failure risk of 8% (3.9%-15.9%), a Lysholm score of 87.7 (85.1-90.4), a Kujala score of 78.7 (63.9-93.5), and an IKDC score of 84.3 (73.9-94.7). Sensitivity analysis of lesions <3.5 cm2 showed equivalent failure rates among treatments, and mean IKDC scores were 84.2 (65.4-103), 67.2 (63.1-73.1), and 89.5 (84.2-94.9) after OAT/mosaicplasty, MACI/ACI, and AMIC, respectively.Conclusion:In this systematic review with meta-analysis, overall good to excellent outcomes were observed for all identified treatments, with mean estimated failure rates <10% and PROMs ranging from 70% to 85% of the maximum score. Treatment with OAT/mosaicplasty and AMIC resulted in the highest IKDC score for patients with lesions <3.5 cm2.Keywords: autologous chondrocyte implantation, cartilage lesion, patella, osteochondral allograft, osteochondral autograft transfer, tibial tubercle osteotomy Patellar cartilage lesions commonly result in knee symptoms such as pain, joint effusions, catching or locking, and/or the inability to perform daily activities or sports. 54 The incidence of cartilage lesions during arthroscopy is approximately 60% at a mean age of 35 to 40 years, of which 10% to 25% have patellar lesions, and 5% to 10% of patients have full-thickness lesions.19,27,39 Patellar lesions often occur after patellar instability or patellar maltracking, and various underlying factors can contribute to the onset and persistence of these lesions, such as malalignment, lateral tuberosity, trochlear dysplasia, patellar instability, muscle imbalances, or patella alta.7,21,50,51,63,64,75 Initial treatment of cartilage lesions typically consists of nonoperative modalities, with microfracture or chondroplasty considered for those with persistent symptoms, but outcomes of this approach are disappointing for larger lesions and at longer-term follow-up.2,32,44,53,65,70 As a result, treatments have shifted toward more restorative procedures such as osteochondral autograft transplantation (OAT) or mosaicplasty, osteochondral allograft (OCA), or (matrix-associated) autologous chondrocyte implantation (MACI or ACI), which are often performed with concomitant procedures such as tibial tubercle osteotomy (TTO), trochleoplasty, medial patellofemoral ligament reconstruction (MPFLR), and lateral retinaculum lengthening or release (LRL). Given the various treatment options for these cartilage lesions, it is important to assess the overall outcomes in the literature to guide treatment for surgeons and patients. However, most systematic reviews on this topic combine patellar and trochlear lesions together,3,16,38 or only report on specific procedures. 17 One study reported on isolated patellar chondral treatment 30 but several new studies have been published since, indicating that an updated and complete systematic review is of additional value.6,20,22,52 Therefore, this study aimed to assess the outcomes of advanced cartilage treatments for isolated patellar lesions. The hypothesis was that the various treatments would lead to satisfactory outcomes regarding failure, reoperation, return to sports, and subjective outcomes. Methods This systematic review with meta-analysis followed the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, and the study was registered in PROSPERO (number blinded). There was no funding or conflict of interest for this study. Literature Search The electronic databases PubMed, Cochrane Library, and Embase were searched on March 26, 2025, for studies reporting on the outcomes of surgical treatment for patellar cartilage lesions. The search terms included “(patella OR patellar OR patellofemoral) AND (cartilage OR osteochondral or chondral) AND (((osteochondral) AND (allograft OR autograft)) OR transplantation OR implantation OR restoration OR repair) AND (outcomes OR reoperation OR results) AND knee” and were limited to English studies with human subjects. In the first round, titles and abstracts of all identified studies were reviewed for potential inclusion; in the second round, the full texts of the potential studies were reviewed for inclusion. In addition, the references of 7 systematic reviews and meta-analyses on patellofemoral cartilage injuries were reviewed to ensure no studies were missed.14,23,24,30,38,56,71 Agreement was ultimately reached on inclusion and exclusion for all studies, and no third reviewer was required. The inclusion criteria consisted of (1) isolated patellar cartilage lesion (ie, no bipolar or trochlear lesions); (2) advanced cartilage treatments (ie, no chondroplasty or microfracture); (3) outcomes at minimum 1 year follow-up; (4) minimum of 10 patients; (5) indication of chondral or osteochondral lesions; and (6) skeletal maturity. A minimum of 1 year and a minimum of 10 patients were chosen to ensure capturing all relevant studies on this topic. The exclusion criteria consisted of (1) animal or biomechanical studies; (2) concomitant other cartilage lesions treated; and (3) indication of osteoarthritis. Concomitant procedures such as TTO or MPFLR were not considered exclusion criteria, as they are frequently performed in patients with cartilage lesions (Figure 1). PRISMA flow diagram of the study. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses. Quality Assessment The level of evidence of the included studies was assessed using the adjusted Oxford Centre for Evidence-Based Medicine. 80 The methodological quality of the included studies was assessed using the Methodological Index for Non-Randomized Studies (MINORS) criteria, as no randomized studies were identified. 69 Finally, the quality of evidence was graded using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach. 35 Data Collection Extracted data were collected in Excel (Version 2021; Microsoft Corp). Baseline data consisted of author names, publication year, journal, number of knees, age, length of follow-up, sex, type of procedure, and concomitant procedures. Based on the available data, relevant outcomes consisted of failures (defined by authors and as revision of procedure or conversion to arthroplasty), reoperations, complications, return to sports (RTS), Lysholm score, Kujala, International Knee Documentation Committee (IKDC) score, Tegner activity level, and visual analog scale of pain (VAS). Statistical Analysis For statistical analysis, Excel and the R package (Version 4.2.3; R Foundation for Statistical Computing) were used. 72 Baseline characteristics were reported as weighted means, and means and standard deviations were calculated using predefined methods if not reported and possible.40,78 Given the lack of comparative studies, no comparative meta-analyses were performed. Heterogeneity was assessed using I2 values; values <25% were considered low. 10 Forest plots were generated for all outcomes by treatment modality, using random-effect models, and reported with 95% CIs. Additionally, plots were designed to assess the correlation between defect size and patient-reported outcome measures (PROMs) regardless of treatments. Finally, sensitivity analysis was performed by only including lesions <3.5 cm2 for IKDC and failure rates. Results Study Selection After duplicate removal, 1904 titles and abstracts were reviewed, and 56 full-text studies were ultimately reviewed. Ultimately, 36 studies were included (Figure 1). § Two studies34,48 reported on the outcomes of OCA, 9 studies1,4,5,18,20,25,28,36,58 on OAT or mosaicplasty, 18 studies MACI on ACI ‖ or 7 studies20,41,47,57,67,74,77 on autologous matrix-induced chondrogenesis (AMIC), and 2 studies52,73 on particulated juvenile articular cartilage (PJAC). Results of the procedures are presented with estimated failure rates (Figure 2), reoperations (Figure 3), complications (Figure 4), RTS (Appendix Figure 1), Lysholm (Appendix Figure 2), Kujala (Appendix Figure 3), IKDC (Appendix Figure 4), Tegner activity level (Appendix Figure 5), and VAS pain (Appendix Figure 6). Trends of PROMs with length of follow-up were also assessed (Figure 5). Forest plot of studies reporting failure rates grouped by the different treatments. ACI, autologous chondrocyte implantation; AMIC, autologous matrix-induced chondrogenesis; MACI, matrix-induced autologous chondrocyte implantation; OAT, osteochondral autograft transfer; OCA, osteochondral allograft; PJAC, particulated juvenile articular cartilage. Forest plot of studies reporting reoperation rates grouped by the different treatments. ACI, autologous chondrocyte implantation; AMIC, autologous matrix-induced chondrogenesis; MACI, matrix-induced autologous chondrocyte implantation; OAT, osteochondral autograft transfer; OCA, osteochondral allograft; PJAC, particulated juvenile articular cartilage. Forest plot of studies reporting complication rates grouped by the different treatments. ACI, autologous chondrocyte implantation; AMIC, autologous matrix-induced chondrogenesis; MACI, matrix-induced autologous chondrocyte implantation; OAT, osteochondral autograft transfer; OCA, osteochondral allograft; PJAC, particulated juvenile articular cartilage. Trends in the different PROMs are shown relative to defect size, regardless of surgical treatment, for (A) IKDC, (B) Lysholm, (C) VAS pain, and (D) Kujala score. IKDC, International Knee Documentation Committee score. Lysholm, Lysholm Knee Scoring Scale. VAS pain, visual analog scale for pain. Quality of Studies The included studies had limited level of evidence with only 1 level 2 prospective 26 study, 8 level 3 studies,6,20,28,37,49,57,59,61 and the majority of level 4 studies. ¶ Included noncomparative studies had a mean MINORS criteria score of 12 out of 16 points, corresponding to 75% of the maximum score. In contrast, comparative studies had a mean score of 17.6 out of 24, corresponding to 73% of the maximum score (Table 1). Given the heterogeneity of studies and the lack of level 1 studies, the GRADE recommendation was very low. Of the studies included, 81% were published since 2010 and 36% since 2020. Table 1Quality Assessment of the Included Studies Using the MINORS Criteria a AuthorsYearJournalLoEStudy Design123456789101112TotalGracitelli et al 34 2015Am J Sports Med4Case series2222022012Lin et al 48 2020Orthop J Sports Med4Case series2222220214Gaweda et al 28 2006Int Orthop3Cohort study2022020022119Nho et al 58 2008Am J Sports Med4Case series2222220012Hangody et al 36 2010Am J Sports Med4Case series212202009Figueroa et al 25 2011Knee4Case series2222022012Cohen et al 18 2012Rev Bras Ortop4Case series2222012011Astur et al 4 2014J Bone Joint Surg Am4Case series2222222014Astur et al 5 2017KSSTA4Case series2222022214Akgün and Akpolat 1 2019JOS (HK)4Case series2212022011Debieux et al 20 2025J Cartil Jt Preserv3Cohort study22120200222217Peterson et al 66 2002Am J Sports Med4Case series2002222010Henderson and Lavigne 37 2006Knee3Cohort study22220220222220Gobbi et al 31 2006Am J Sports Med4Case series2222222014Kreuz et al 43 2007Osteoarthritis Cartilage4Case series2222020010Niemeyer et al 60 2008AOTS4Case series2212221012Macmull et al 49 2012Int Orthop3Cohort study20220200222216Biant et al 8 2014Am J Sports Med4Case series2222222014Filardo et al 26 2014Am J Sports Med2Cohort study22220210222219Gillogly and Arnold 29 2014Am J Sports Med4Case series2222021011Kon et al 42 2016Am J Sports Med4Case series2222021011von Keudell et al 76 2017Cartilage4Case series2222221013Mehl et al 55 2019Knee4Case series2212220011Niemeyer et al 59 2020AOTS3Cohort study22220212222221Niethammer et al 61 2021Int Orthop3Cohort study21220202222219Baumann-Jungmann et al 6 2024Skeletal Radiol3Cohort study22020200222216Boffa et al 9 2024Am J Sports Med4Case series2222021011Ebert et al 22 2024Am J Sports Med4Case series2222121012Kusano et al 47 2012KSSTA4Case series2212222013Sadlik et al 67 2017J Knee Surg4Case series2222222014Tradati et al 74 2020J Clin Med4Case series2212022011Kaiser et al 41 2021AOTS4Case series211202109Migliorini et al 57 2021Life (Basel)3Cohort study22222210222221Waltenspul et al 77 2021Cartilage4Case series2222222216Tompkins et al 73 2013Arthroscopy4Case series2212022011Marmor et al 52 2024Orthop J Sports Med4Case series2222221013aThe criteria of MINORS 69 with 0 points when not reported, 1 when reported but inadequate, and 2 when reported and adequate; maximum score, 16 for noncomparative studies and 24 for comparative studies. (1) A clearly stated aim: the question addressed should be precise and relevant in the light of available literature. (2) Inclusion of consecutive patients: all patients potentially fit for inclusion (satisfying the criteria for inclusion) have been included in the study during the study period (no exclusion or details about the reasons for exclusion). (3) Prospective collection of data: data were collected according to a protocol established before the beginning of the study. (4) Endpoints appropriate to the aim of the study: an unambiguous explanation of the criteria used to evaluate the main outcome, which should be in accordance with the question addressed by the study. In addition, the endpoints should be assessed on an intention-to-treat basis. (5) Unbiased assessment of the study endpoint: blind evaluation of objective endpoints and double-masked evaluation of subjective endpoints. Otherwise, the reasons for not blinding should be stated. (6) Follow-up period appropriate to the aim of the study: the follow-up should be sufficiently long to allow the assessment of the main endpoint and possible adverse events. (7) Loss to follow-up, 5%: all patients should be included in the follow-up. Otherwise, the proportion lost to follow-up should not exceed the proportion experiencing the major endpoint. (8) Prospective calculation of the study size: information on the size of the detectable difference of interest with a calculation of 95% CI, according to the expected incidence of the outcome event, and information about the level of statistical significance and estimates of power when comparing the outcomes. (9) An adequate control group: having a gold standard diagnostic test or therapeutic intervention recognized as the optimal intervention according to the available published data. (10) Contemporary groups: the control and studies group should be managed during the same period. (11) Baseline equivalence of groups: the groups should be similar regarding the criteria other than the studied endpoint. Absence of confounding factors that could bias the interpretation of the results. (12) Adequate statistical analyses to determine whether statistics were in accordance with the type of study, with the calculation of confidence intervals, or relative risk. LoE, level of evidence; MINORS, Methodological Index for Non-Randomized Studies; RCT, randomized controlled trial. Quality Assessment of the Included Studies Using the MINORS Criteria a The criteria of MINORS 69 with 0 points when not reported, 1 when reported but inadequate, and 2 when reported and adequate; maximum score, 16 for noncomparative studies and 24 for comparative studies. (1) A clearly stated aim: the question addressed should be precise and relevant in the light of available literature. (2) Inclusion of consecutive patients: all patients potentially fit for inclusion (satisfying the criteria for inclusion) have been included in the study during the study period (no exclusion or details about the reasons for exclusion). (3) Prospective collection of data: data were collected according to a protocol established before the beginning of the study. (4) Endpoints appropriate to the aim of the study: an unambiguous explanation of the criteria used to evaluate the main outcome, which should be in accordance with the question addressed by the study. In addition, the endpoints should be assessed on an intention-to-treat basis. (5) Unbiased assessment of the study endpoint: blind evaluation of objective endpoints and double-masked evaluation of subjective endpoints. Otherwise, the reasons for not blinding should be stated. (6) Follow-up period appropriate to the aim of the study: the follow-up should be sufficiently long to allow the assessment of the main endpoint and possible adverse events. (7) Loss to follow-up, 5%: all patients should be included in the follow-up. Otherwise, the proportion lost to follow-up should not exceed the proportion experiencing the major endpoint. (8) Prospective calculation of the study size: information on the size of the detectable difference of interest with a calculation of 95% CI, according to the expected incidence of the outcome event, and information about the level of statistical significance and estimates of power when comparing the outcomes. (9) An adequate control group: having a gold standard diagnostic test or therapeutic intervention recognized as the optimal intervention according to the available published data. (10) Contemporary groups: the control and studies group should be managed during the same period. (11) Baseline equivalence of groups: the groups should be similar regarding the criteria other than the studied endpoint. Absence of confounding factors that could bias the interpretation of the results. (12) Adequate statistical analyses to determine whether statistics were in accordance with the type of study, with the calculation of confidence intervals, or relative risk. LoE, level of evidence; MINORS, Methodological Index for Non-Randomized Studies; RCT, randomized controlled trial. Baseline Characteristics A total of 36 studies were included, with 1058 patients with a mean age of 32 years (range, 13-64 years), a mean follow-up of 5.1 years (range, 1.7-17.6 years), and 54% were women. The overall mean treated location size was 3.6 cm2 (range, 1.1-6.3 cm2) (Table 2). A total of 27 studies reported concomitant procedures in 779 patients, with TTO most frequently performed (28%), followed by LRL or release (24%), MPFLR (13%), and trochleoplasty (4%) (Table 3). Table 2Study Characteristics of All Included Studies by Type of Surgical Technique a AuthorsYearNo. of KneesAge, yearsFollow-up, yearsFemale (%)Lesion SizeMeanRangeMeanRangeMeanRangeOCA Gracitelli et al 34 2015283414-649.71.8-30.154 Lin et al 48 20202042 b 3.9 b 2-7.1 b 3.1 Total483714-647.23.9-9.7543.1OAT/mosaicplasty Gaweda et al 28 200619262 Nho et al 58 2008223015-572.41.5-4.8451.7 Hangody et al 36 20101824 b 14-49 b 9.6 b 2-17 b 2.41-3 Figueroa et al 25 2011102015-383.12-5.801.20.9-2 Cohen et al 18 2012173816-591.71-2.8471-4 Astur et al 4 2014333815-592.92-4.5481-2.5 Astur et al 5 20172026-452551.21-1.8 Akgün and Akpolat 1 2019143019-493.72.7-6.7431.30.8-1.8 Debieux et al 20 2025353616-572561.10.7-2.2 Total1883214-593.11.7-9.6431.51.1-2.4ACI/MACI Peterson et al 66 200217277.45-114.41-12 Henderson and Lavigne 37 2006443414-562.30.8-4.6523.1 Gobbi et al 31 20062231 b 15-55 b 24.7 b 0.8-12 b Kreuz et al 43 2007182818-433725.73-12 Niemeyer et al 60 200870343.21.2-5.34.8 Macmull et al 49 2012253517-503.8804.71-8.8 Macmull et al 49 2012233521-462.9614.81.0-10.5 Biant et al 8 2014363010.410-12644.51.3-10 Filardo et al 26 201428295.0502.8 Gillogly and Arnold 29 201425317.65.1-11.4526.4 Kon et al 42 20162031 b 104.5 b von Keudell et al 76 2017303215-497.3604.7 Mehl et al 55 201978336.5594.3 Niemeyer et al 59 202045345.0365.4 Niethammer et al 61 2021253313-563.0604.62-12 Baumann-Jungmann et al 6 202435232.2492.2 Boffa et al 9 2024243017.6542.8 Ebert et al 22 2024413623-611210.5-15443.21.5-6 Total6063213-6162-17.6554.32.2-6.4AMIC Kusano et al 47 201220392.4504.4 Sadlik et al 67 2017123622-523.22-5.8592.50.8-4 Tradati et al 74 202014385.72.1-9.3534.5 Kaiser et al 41 202115409.3562.6 Migliorini et al 57 202127363.8422.71.1-3 Waltenspul et al 77 202131284.12-8422 Debieux et al 20 2025123716-542592.21-4 Total1313516-544.32.0-9.3502.92-4.5PJAC Tompkins et al 73 201315262.42.4 Marmor et al 52 2024702714-512.83-7762.80.8-8. Total852714-512.73-7762.72.4-2.8 Overall total10583213-645.11.7-17.6543.61.1-6.4aACI, autologous chondrocyte implantation; AMIC, autologous matrix-induced chondrogenesis; MACI, matrix-induced autologous chondrocyte implantation; OCA, osteochondral allograft; OAT, osteochondral autograft transfer; PJAC, particulated juvenile articular cartilage.bThese studies only reported criteria for all patients. Study Characteristics of All Included Studies by Type of Surgical Technique a ACI, autologous chondrocyte implantation; AMIC, autologous matrix-induced chondrogenesis; MACI, matrix-induced autologous chondrocyte implantation; OCA, osteochondral allograft; OAT, osteochondral autograft transfer; PJAC, particulated juvenile articular cartilage. These studies only reported criteria for all patients. Table 3Concomitant Procedures and Surgical Outcomes of All Included Studies by Type of Surgical Technique a AuthorsYearNo. of KneesConcomitant ProceduresTTO, %MPFLR, %TP, %LR, %Other, nOCA Gracitelli et al 34 2015281440251 Lin et al 48 202020 Total28144025-OAT/mosaicplasty Gaweda et al 28 2006191000010019 Nho et al 58 2008224100593 Hangody et al 36 2010181700612 Figueroa et al 25 2011101020004 Cohen et al 18 201217060350 Astur et al 4 20143300000 Astur et al 5 20172000000 Akgün and Akpolat 1 201914000140 Debieux et al 20 20253500000 Total187172027-ACI/MACI Peterson et al 66 200217 Henderson and Lavigne 37 2006445000500 Gobbi et al 31 200622 Kreuz et al 43 200718 Niemeyer et al 60 20087000000 Macmull et al 49 201225 Macmull et al 49 201223 Biant et al 8 201436 Filardo et al 26 201428 Gillogly and Arnold 29 2014251000161000 Kon et al 42 201620 von Keudell et al 76 201730630179326 Mehl et al 55 20197801201324 Niemeyer et al 59 202045 Niethammer et al 61 20212500009 Baumann-Jungmann et al 6 202435054000 Boffa et al 9 2024240 Ebert et al 22 20244132 Total346238328 - AMIC Kusano et al 47 201220900000 Sadlik et al 67 201712178001 Tradati et al 74 20201407070 Kaiser et al 41 202115000019 Migliorini et al 57 20212700000 Waltenspul et al 77 202131616158016 Debieux et al 20 20251200000 Total1293016141-PJAC Tompkins et al 73 2013155320002 Marmor et al 52 202470816604760 Total857658039- Overall total7792813424 - aACI, autologous chondrocyte implantation; AMIC, autologous matrix-induced chondrogenesis; LR, lateral release; MACI, matrix-induced autologous chondrocyte implantation; MPFLR indicates medial patellofemoral ligament reconstruction; OAT, osteochondral autograft transfer; OCA, osteochondral allograft; PJAC, particulated juvenile articular cartilage; TP, trochleoplasty; TTO, tibial tubercle osteotomy. Concomitant Procedures and Surgical Outcomes of All Included Studies by Type of Surgical Technique a ACI, autologous chondrocyte implantation; AMIC, autologous matrix-induced chondrogenesis; LR, lateral release; MACI, matrix-induced autologous chondrocyte implantation; MPFLR indicates medial patellofemoral ligament reconstruction; OAT, osteochondral autograft transfer; OCA, osteochondral allograft; PJAC, particulated juvenile articular cartilage; TP, trochleoplasty; TTO, tibial tubercle osteotomy. Eighteen studies reported on the outcomes of ACI/MACI with 606 patients, who had a mean age of 32 years (range, 13-61 years), a mean follow-up of 6 years (range, 2-17.6 years), and the largest mean lesion size of all treatments (mean, 4.3 cm2). The second most common treatment was OAT/mosaicplasty, with 9 studies and 188 patients with comparable baseline characteristics (mean age, 32 years; mean follow-up, 3.1 years), although they had the smallest mean lesion size (mean, 1.5 cm2 [range 1.1-2.4 cm2]). The third most reported treatment is AMIC, with 7 studies and 131 patients with a mean age of 35 years (range, 16-54 years), a mean follow-up of 4.3 years, and a mean defect size of 2.9 cm2. The 2 least frequently reported procedures were OCA (2 studies, 48 patients) and PJAC (2 studies, 85 patients). Those studies had similar baseline characteristics, with mean ages of 37 and 27 years, respectively, and mean lesion sizes of 3.1 and 2.7 cm2, respectively. The OCA cohorts had relatively long follow-up (3.9 and 9.7 years), whereas the PJAC cohorts did not have any studies beyond short-term follow-up (2.4 and 2.8 years). ACI and MACI Thirteen studies (476 patients) reported an overall estimated failure rate of ACI/MACI of 8.8% (95% CI, 4.4%-14%), and 8 studies (297 patients) reported an overall estimated reoperation rate of 9.7% (95% CI, 2.1%-34.5%). Ten studies (336 patients) reported an overall estimated complication rate was 16.4% (95% CI, 6%-37.7%). Only 2 studies (93 patients) reported an RTS rate of 71% (95% CI, 46.4%-87.5%), and 2 studies (105 patients) reported a Lysholm score of 72.8 (95% CI, 68.7%-76.8%), which was lower than that of other treatments. The estimated Kujala score was comparable to other treatments, with an estimated overall score of 74.8 (95% CI, 66.9%-82.8%) in 3 studies (141 patients). However, the IKDC was lower than that of other treatments, with an estimated overall score of only 65.3 (95% CI, 60.7%-69.9%) in 7 studies involving 289 patients. Similarly, the Tegner score was higher than the overall mean (3.5 [95% CI, 2.6%-4.4%]) in 2 studies with 52 patients, and the VAS pain was higher, although this was based on only 1 study with 25 patients 61 (Table 4). Sensitivity analysis with only lesions <3.5 cm2 revealed a failure rate of 2.6% (95% CI, 0.4%-13.9%) and IKDC of 67.2 (95% CI, 63.1%-71.3%). Table 4Reported Outcomes of All Included Studies by Type of Surgical Technique a AuthorsYearNo. KneesIKDCKujalaLysholmTegnerMod CincVASOCA Gracitelli et al 34 20152867 Lin et al 48 202020663.2OAT/Mosaicplasty Gaweda et al 28 200619 Nho et al 58 20082274 ± 12 Hangody et al 36 201018 Figueroa et al 25 20111094 ± 295 ± 4 Cohen et al 18 20121774 ± 1876 ± 19 Astur et al 4 20143375 ± 1381 ± 12 Astur et al 5 201720772.4 ± 2.6 Akgün and Akpolat 1 20191478 ± 780 ± 7 Debieux et al 20 20253580 ± 1280 ± 115.5 ± 2.2ACI/MACI Peterson et al 66 2002178.86.62.8 Henderson and Lavigne 37 20064468 ± 236.5 Gobbi et al 31 20062271 Kreuz et al 43 2007182.2 ± 0.7 Niemeyer et al 60 20087062 ± 2273 ± 226.2 ± 2.2 Macmull et al 49 2012254.95 Macmull et al 49 2012236.14 Biant et al 8 2014367.92 Filardo et al 26 20142870 ± 1882 ± 133.9 ± 1.7 Gillogly and Arnold 29 20142576797 Kon et al 42 20162076 ± 19 von Keudell et al 42 2017305.7 ± 1.5 Mehl et al 55 20197865 ± 2068 ± 20 Niemeyer et al 59 202045 Niethammer et al 61 20212555 ± 203.4 ± 2.0 Baumann-Jungmann et al 6 20243576 ± 1672 ± 207.1 ± 0.9 Boffa et al 9 20242462 ± 203 ± 2 Ebert et al 22 202441AMIC Kusano et al 47 20122074 ± 1785 ± 134 ± 12 ± 2 Sadlik et al 67 20171290 ± 182.3 ± 2.0 Tradati et al 74 20201480884.31.5 Kaiser et al 41 20211585 ± 142.3 ± 2.1 Migliorini et al 57 20212789 ± 1690 ± 115 ± 21.3 ± 0.7 Waltenspul et al 77 20213171 ± 174.2 ± 1.9 Debieux et al 20 20251286 ± 1388 ± 114.9 ± 1.3PJAC Tompkins et al 73 20131573 ± 185.51.9 ± 1.4 Marmor et al 52 2024707486aData are presented as mean ± SD, unless otherwise specified. ACI, autologous chondrocyte implantation; AMIC, autologous matrix-induced chondrogenesis; IKDC, International Knee Documentation Committee; MACI, matrix-induced autologous chondrocyte implantation; Mod Cinc, Modified Cincinnati; OCA, osteochondral allograft; OAT, osteochondral autograft transfer; PJAC, particulated juvenile articular cartilage; VAS, Visual analog scale. Reported Outcomes of All Included Studies by Type of Surgical Technique a Data are presented as mean ± SD, unless otherwise specified. ACI, autologous chondrocyte implantation; AMIC, autologous matrix-induced chondrogenesis; IKDC, International Knee Documentation Committee; MACI, matrix-induced autologous chondrocyte implantation; Mod Cinc, Modified Cincinnati; OCA, osteochondral allograft; OAT, osteochondral autograft transfer; PJAC, particulated juvenile articular cartilage; VAS, Visual analog scale. OAT/Mosaicplasty Six studies (115 patients) reported the lowest overall estimated failure rate for any treatment at 1.7% (95% CI, 0.4%-6.7%) and a comparable reoperation rate of 7.3% (95% CI, 2.8%-17.9%). Seven studies (133 patients) reported comparable complication rates with an estimated 8.8% (95% CI, 2%-31.9%). RTS was achieved in 100% of patients (2 studies, 29 patients). The Tegner level was higher than in other studies (5.5), but this was only reported in 1 study with 34 patients. 20 The Lysholm score (5 studies, 108 patients), Kujala score (4 studies, 98 patients) and VAS pain (1 study, 20 patients) were comparable to other treatments with estimated means of 82.7 (95% CI, 76.1%-89.3%), 77.6 (95% CI, 75.3%-79.8%), and 2.4 (95% CI, 1.2%-3.6%), respectively. The IKDC score (2 studies, 32 patients) was higher than ACI/MACI, with an estimated mean of 84.2 (95% CI, 65.4%-103%). AMIC Four studies (87 patients) reported an estimated failure rate of 8% (95% CI, 3.9%-15.9%) after AMIC, a reoperation rate of 0.4% (95% CI, 0%-75.1%), and 6 studies (119 patients) reported an estimated complication rate of 1.4% (95% CI, 0%-46.9%). Kujala and Tegner activity scores were equivalent to other treatments with 78.7 (95% CI, 63.9%-93.5%) and 4.4 (95% CI, 3.9%-4.9%), respectively, whereas RTS was lower compared with other treatments, with 2 studies (45 patients) reporting a rate of 35.6% (95% CI, 23.1%-50.4%). Finally, Lysholm and IKDC scores were relatively high compared with other treatments, and VAS pain was relatively low, with estimated means of 87.7 (95% CI, 85.1%-90.4%), 84.3 (95% CI, 73.9%-94.7%), and 1.8 (95% CI, 1.2%-2.4%), respectively. Sensitivity analysis with only lesions <3.5 cm2 revealed a failure rate of 9.6% (95% CI, 4.6%-18.8%) and an IKDC score of 89.5 (95% CI, 84.2%-94.9%). OCA and PJAC Only 2 studies reported on OCA and PJAC outcomes each, preventing a reliable meta-analysis of outcomes, except for PJAC failure rates. Two studies (85 patients) reported an overall estimated failure rate of 7.1% (95% CI, 3.2%-14.8%). Discussion Most studies of isolated patellar cartilage lesion treatment have been published on ACI/MACI procedures, followed by OAT/mosaicplasty and AMIC. Overall, good to excellent outcomes were observed across all treatments, with mean estimated failure rates <10% and PROMs ranging from 70% to 85% of the maximum score. AMIC had the lowest incidence of reoperations and complications, and OAT and AMIC reported the highest IKDC and Lysholm scores compared with ACI/MACI. However, OAT/mosaicplasty had the smallest defect size, and ACI/MACI had the largest defect size and the longest follow-up. This study identified that ACI/MACI procedures were most reported in the literature. This is perhaps not surprising as the first study on ACI was published in 1994, and the technique has undergone several developments to improve outcomes.11-13,33,79 The indications for the different treatments in this study vary significantly, which should be considered when interpreting the results. Not only were there differences in defect size between the different treatments (ie, OAT/mosaicplasty had the smallest defect and ACI/MACI the largest), but the follow-up of ACI/MACI was also longer (Table 2), and the incidence of concomitant procedures differed between the cohorts as well (Table 3). Regarding failure rates in this study, they were overall similar across treatments, without significant differences. However, the failure rate of OAT/mosaicplasty was lower (1.7%) than that of other treatments. This can likely be explained by the fact that patients in this cohort had smaller defects and shorter follow-up than those in other treatments. Nonetheless, these findings show that OAT/mosaicplasty can be an excellent treatment for patients with smaller lesions. These findings are consistent with the literature on tibiofemoral cartilage treatments. 62 Oeding et al 62 assessed the outcomes of OAT at 15-year follow-up in 12 studies with 495 patients and noted excellent outcomes and low rates of reoperation. Based on these studies, it seems that OAT for patellar lesions leads to excellent results and might be the preferred treatment for smaller lesions, even though cartilage thickness is different between the patella and other areas of the knee. 68 Interestingly, the PROMs (IKDC and Lysholm) were lowest in the ACI/MACI group and highest in the OAT/mosaicplasty and AMIC groups. For systematic reviews like this one, it is impossible to correct for baseline differences such as lesion size or length of follow-up. However, when reviewing the outcome scores by defect size (Figure 5), one can see that Kujala and Lysholm scorers did not differ between smaller and larger lesions, and that the IKDC score was only mildly worse in the larger group. In contrast, the VAS pain score was indeed worse at follow-up in patients with initially larger lesions. Therefore, it seems that defect size can only partially explain the inferior outcomes of MACI for patellar lesions. Among the AMIC studies, the study with the largest defect size (4.4 cm2) by Kusano et al 47 had indeed inferior IKDC outcomes compared with AMIC for smaller defect sizes (2.5 and 2.7 cm2), 57 but still had superior IKDC scores compared with 6 out of 7 ACI/MACI studies. Although only a limited number of studies assessed RTS, OAT/mosaicplasty was associated with the highest rate, with all 29 patients returning to sports in 2 studies. This is similar to studies focusing on RTS rates for all cartilage lesions in the knee (with the majority being tibiofemoral lesions), as various systematic reviews have shown that the highest RTS rates are achieved after OAT/mosaicplasty.15,45,46 In this study, complications reported in the included studies were somewhat higher in the MACI/ACI cohort than in other studies, with rates ranging from 0% to 60%. Five out of 10 studies reported a complication rate of 31% or higher. When reviewing these studies, hypertrophy of the graft was the most common complication after ACI/MACI, followed by chondroplasty, arthrofibrosis, and periosteal patch complications in the older studies.6,29,37,43 The issues of graft hypertrophy were not seen with other treatments. However, the AMIC study by Kusano et al 47 reported a 45% stiffness rate, which could be attributed to a slower rehabilitation protocol, and the study by Waltenspül et al 77 mainly reported complications with the TTO screws. Limitations Limitations are present in this study. First, the study design, as a systematic review with meta-analysis, prevents the correction for potential confounders, such as defect size and length of follow-up, and limits conclusions about the direct comparison of treatments. Nonetheless, this study provides an overview of the available evidence in the literature for the outcomes of all treatments for their indications, and the sensitivity analysis with lesions <3.5 cm2 enables assessment of smaller lesions. Furthermore, different reviews may have used different criteria for outcomes of reoperations and complications, which may be explained by procedures that are indirectly related, such as TTO screw removal. Finally, this study cannot control for the role of concomitant procedures, year of procedure, and treatment indications. Conclusion In this systematic review with meta-analysis, overall good to excellent outcomes were observed across all identified treatments, with mean estimated failure rates <10% and PROMs ranging from 70% to 85% of the maximum score. Treatment with OAT/mosaicplasty and AMIC resulted in the lowest reoperation rates and highest IKDC and Lysholm scores, although these lesions generally had smaller defect sizes. Appendix Forest plot of studies reporting return to sport rates grouped by the different treatments. Forest plot of studies reporting Lysholm scores grouped by the different treatments. Forest plot of studies reporting Kujala scores grouped by the different treatments. Forest plot of studies reporting subjective IKDC scores grouped by the different treatments. Forest plot of studies reporting Tegner scores grouped by the different treatments. Forest plot of studies reporting visual analog scale (VAS) pain grouped by the different treatments. References 1, 4-6, 8, 9, 18, 20, 22, 25, 26, 28, 29, 31, 34, 36, 37, 41-43, 47, 48, 49, 52, 55, 57-61, 66, 67, 73-77. References 6, 8, 9, 22, 26, 29, 31, 37, 42-43, 49, 55, 59-61, 66. References 1, 4, 5, 8, 9, 18, 22, 25, 29, 31, 34, 36, 41-43, 47, 48, 52, 55, 58, 60, 66, 67, 73-77. Footnotes Final revision submitted March 21, 2026; accepted March 28, 2026. The authors declared that they have no conflicts of interest in the authorship and publication of this contribution. ORCID iDs: Jelle P. van der List https://orcid.org/0000-0002-7940-5152 David C. Flanigan https://orcid.org/0000-0001-8876-7043 Robert A. Magnussen https://orcid.org/0000-0001-9864-1508 References