Outcomes Following Autologous Osteochondral Transplantation for Osteochondral Lesions of the Talus at 10-Year Follow-Up: A Retrospective Review

Abstract

Objective

The purpose of this study was to evaluate outcomes following autologous osteochondral transplantation (AOT) for the treatment of osteochondral lesions of the talus (OLT) at a minimum of 10-year follow-up.

Design

Retrospective chart review identified patients who underwent AOT for the treatment of OLT. Pre-operative magnetic resonance imaging (MRI) scans were obtained in all patients. Clinical outcomes assessed included: pre- and post-operative foot and ankle outcome score (FAOS), visual analog scale (VAS), patient satisfaction, complications, failures and secondary surgical procedures.

Results

Thirty-nine patients with a mean lesion size was 122.3 ± 64.1 mm² and mean follow-up time of 138.9 ± 16.9 months were included. The mean FAOS scores improved from a preoperative score of 51.9 ± 16.0 to 75.3 ± 21.9 (P < 0.001). Increasing lesion size was variable associated with inferior FAOS scores (R² = 0.2228). There was statistically significant higher mean T2 relaxation values at the superficial layer at the site of the AOT graft (42.9 ± 5.2 ms) compared to the superficial layer of the adjacent native cartilage (35.8 ± 3.8 ms) (P < 0.001). Seventeen complications (43.6%) were observed, the most common of which was anterior ankle impingement (25.6%). There were 2 failures (5.1%), both of which had a history of prior bone marrow stimulation via microfracture and post-operative cysts identified on MRI.

Conclusion

This retrospective review found that AOT for the treatment of large OLTs produced a 94.9% survival rate at a minimum of 10-year follow-up. Increasing lesion size was associated with inferior clinical outcomes. The findings of this study indicates that AOT is a viable long-term surgical strategy for the treatment of large OLTs.

Introduction

Osteochondral lesions of the talus (OLT) are a challenging pathology to treat due to the limited regenerative capacity of the native articular cartilage. ¹ This is primarily due to the avascular nature of the cartilage that lines the talar dome. ² Treatment is primarily dictated by lesion size. Smaller lesions (lesion diameter less than 10 mm in and/or area less than 100 mm²) can be managed with reparative procedures including arthroscopic debridement, bone marrow stimulation via microfracture, retrograde drilling and scaffold based therapies including extracellular matrix cartilage allograft.^{3 -5} Larger lesions (lesion diameter greater than 10 mm in and/or area greater than 100 mm²) require a more extensive surgical procedure including autologous osteochondral transplantation (AOT) or allogenic osteochondral transplantation. ⁶

AOT involves resection of the diseased cartilage from the talus with direct replacement with an autograft harvested from the ipsilateral non-weightbearing portion of the lateral femoral condyle. ⁶ AOT is not only indicated for large OLTs, but for uncontained OLTs, cystic OLTs and OLTs with a prior history of microfracture.^{6 -8} To date, numerous studies have demonstrated that AOT results in excellent functional outcomes and return to sporting activity data.^6,7,9,10 Furthermore, satisfactory quantitative and qualitative outcomes on postoperative magnetic resonance imaging (MRI) scans have been observed following AOT. ⁷ However, these studies have been limited by the short-to-mid-term follow-up, and little is known regarding outcomes following AOT in the long-term.

The purpose of this study was to retrospectively review the clinical and functional outcomes at a minimum of 10-year follow-up following AOT for the treatment of OLTs. In addition, we aimed to describe the complication rate, failure rate and secondary surgical procedure rate in this cohort. Finally, we sought to identify any potential predictors of inferior clinical outcomes. Our hypothesis was that AOT for the treatment of large OLTs would result in significant improvement in subjective clinical outcomes and high survival rate at long-term follow-up.

Methods

Patient Recruitment

This retrospective cohort study used chart review for retrieval of data on individual patients following approval by the institutional review board (i23-00823). All patients who underwent AOT for OLT by the senior author as part of standard clinical care from January 1, 2006, to October 31, 2013, were identified. The inclusion criteria were age greater than 18 years old at the time of AOT, with a minimum follow-up of 120 months. In order to establish a more homogenous patient cohort, patients with a history of diabetes and patients with inflammatory arthritic conditions such as rheumatoid arthritis, psoriatic arthritis and gout were excluded. Figure 1 illustrates the patient selection process. The indications for AOT included: patients who failed a minimum of 3 months of conservative management, in patients with a lesion size greater than 10 mm in diameter and/or 100 mm² in area. Other indications included revision procedures for failed BMS, uncontained lesions and/or cystic lesions. Preoperative data regarding patient demographic data, duration of symptoms, prior ankle surgeries, and concomitant procedures were also collected and extracted.

Figure 1.

Patient flow selection diagram.

Surgical Technique

The AOT procedure was performed by a senior orthopedic foot and ankle surgery attending as previously described. ⁶ The Osteochondral Autograft Transfer System (Arthrex, Naples, FL) was employed in all procedures. Before initiating the AOT procedure, bone marrow aspirate was harvested from the ipsilateral iliac crest in all patients. Subsequently, the aspirate underwent preparation using a commercially available concentrated bone marrow aspirate (cBMA) system, specifically the Magellan Autologous Platelet Separator System (Arteriocyte Medical Systems, MA). Medial OLTs were accessed using a chevron-type osteotomy of the medial malleolus, while a trapezoidal osteotomy of the anterolateral tibia was required to gain access to lateral OLTs. The osteochondral graft donor plug was extracted from the non-weightbearing portion of the ipsilateral lateral femoral condyle and then bathed in cBMA. Following this, the osteochondral graft plug was precisely transplanted in the recipient talar dome to align as closely as possible with the native cartilage. In instances where a lesion required two grafts, a figure-of-8 “nested” technique was applied to minimize the vacant space, preventing the accumulation of fibrocartilage. Finally, the osteotomy site was realigned and secured using titanium screws.

Postoperative Rehabilitation

Initially, patients were instructed to remain non-weightbearing on the operated limb and were placed into a short leg splint for the first 2 weeks postoperatively. Subsequently, the splint was removed, and patients transitioned to a CAM boot, allowing for touchdown weightbearing and engaging in dorsiflexion/plantarflexion exercises. At the fourth postoperative week, patients gradually initiated weightbearing, beginning with 10% of their body weight and progressively increasing until achieving full weightbearing approximately by the sixth postoperative week. Formal physical therapy commenced at the 6-week mark postoperatively, followed by the initiation of sports-specific physical therapy at the 10th week postoperatively.

Clinical Outcome Analysis

Subjective clinical outcomes were evaluated using the Foot and Ankle Outcome Score (FAOS) scale and the visual analog scale (VAS) for pain, which were collected pre-operatively and at final follow-up by the senior orthopedic foot and ankle surgery attending. The minimal clinically important difference (MCID) represents the smallest alteration in pre- and postoperative patient-reported outcome scores that indicates a clinically meaningful change in symptom improvement or deterioration. MCID was calculated for FAOS scores and VAS scores via distribution-based method of standard deviation (SD): MCID = 0.5*SD. Patient satisfaction with the AOT procedure was evaluated by a 5-point Likert scale at the final follow-up visit. The following options were provided to patients regarding their subjective opinion on the outcome of the procedure: “very satisfied,” “satisfied,” “neutral,” “unsatisfied” or “very unsatisfied.” Complications, failures and secondary surgical procedures were also recorded. Failure was defined as no improvement or worsening of symptoms without evidence of anterior ankle impingement, or requiring a joint sacrificing procedure such as a total ankle arthroplasty or ankle arthrodesis.

Postoperative MRI Analysis

Postoperative T2-mapping MRI scans were conducted on all patients, utilizing a 3-T clinical imaging system (GE Healthcare). Fast spin echo proton density sequences were employed for the assessment of articular cartilage. The image acquisition parameters included a 2.5 mm-thick slice and a 512 x 512 matrix in proton density, as well as a 3 mm-thick slice and a 512 x 512 matrix in fat-saturated proton density or short tau inversion recovery. The T2 relaxation values of the reparative cartilage tissue was assessed through linear least squares estimation (FuncTool 3.1; GE Healthcare) within a 0.2 mm² region of interest, both in the deep and superficial layers of the repair tissue. Similarly, T2 relaxation values for the neighboring, healthy cartilage were calculated using the identical 0.2 mm² region of interest. The assessment of the reparative cartilage encompassed various aspects, including the examination for cysts surrounding the osteochondral plug, which was carried out using fat-saturated proton density sequences in axial, coronal, and sagittal planes. A thorough evaluation of the cartilage around the graft site was conducted using the modified magnetic resonance observation of cartilage repair tissue (MOCART) scoring system. The analysis was conducted by a musculoskeletal radiologist who remained blinded to the surgical procedure and clinical outcome scoring, ensuring an unbiased assessment of the imaging results.

Statistical Analysis

All statistical analyses were performed using RStudio (software version 3.3.0). The Shapiro-Wilk tests were employed to assess the normality assumption for continuous variables. In instances where the assumption was not satisfied, nonparametric Wilcoxon signed rank tests were utilized to compare differences in pre-operative and post-operative continuous variables. Mann Whitney U tests were performed to compare differences in superficial and deep T2-relaxation times between the AOT graft and the adjacent native cartilage. For discrete variables, Chi-square or Fisher exact tests were applied to compare differences between different subgroups. For continuous variable, means and standard deviations (SDs) were reported. Multiple variable linear and logistic regression models were employed to investigate factors influencing postoperative FAOS and VAS scores. The factors considered in the model encompassed various clinical characteristics, such as age, sex, gender preoperative scores, lesion size, history of prior microfracture and graft type. A significance level of P < 0.05 was adopted to determine statistical significance.

Results

Patient Demographics

Forty-three patients who underwent AOT for the treatment of OLT were identified. Four patients were not contactable at final follow-up ( Fig. 1 ). In total, 39 patients (39 ankles) were included in the final analysis. Patient demographics are listed in Table 1 . There were 27 males (69.2%) and 12 females (31.8%). There were 15 right ankles (38.5%) and 24 left ankles (62.5%). The mean age at the time of surgery was 36.3 ± 13.3 years (range, 18-62 years) and the mean body mass index was 28.6 ± 3.7 kg/m² (range, 25.0-34.4 kg/m²). The mean follow-up time was 138.9 ± 16.9 months (range, 121-180 months). The mean duration of symptoms was 29.9 ± 21.3 months (range, 3–60 months). A traumatic etiology was identified in 30 patients (76.9%) which included ankle sprains (51.3%), ankle fracture (17.9%) and direct trauma to the ankle (7.7%).

Table 1.

Patient Demographics and Lesion Characteristics.

Patients (N)	39
Age (y)	36.3 ± 13.3 years
Sex (M/F)	27/12
Laterality (R/L)	15/24
Mean BMI (kg/m2)	28.6 ± 3.7 kg/m2
Mean follow-up (mo)	138.9 ± 16.9 months
Mean duration of symptoms (mo)	29.9 ± 21.3 months
“Single plug” procedure (N)	32 (82.1%)
“Double plug” procedure (N)	7 (17.9%)
Concomitant procedures (N)	12 (30.8%)
Mean lesion area (mm2)	122.3 ± 64.1 mm2
Cystic lesions (N, %)	17 (43.6%)
Shoulder lesions (N, %)	29 (74.4%)
Prior microfracture (N, %)	11 (28.2%)

N = number; y = years; R/L = right/left; kg/m² = kilogram per meter squared; mo = month; mm² = millimeter squared.

Thirty-two “single plug” procedures (82.1%) were performed and 7 “double plug” procedures (17.9%) were performed. Concomitant procedures were performed in 12 patients (30.8%) which included the modified Brostrom-Gould procedure in 7 patients (17.9%), peroneal tenosynovectomy in 2 patients (5.1%), posterior tibial tenosynovectomy in 1 patient (2.6%), pigmented villonodular synovitis resection in 1 patient (2.6%), and an os peroneum resection in 1 patient (2.6%).

Lesion Characteristics

Lesion characteristics are listed in Table 1. The mean lesion size was 122.3 ± 64.1 mm² (range, 52.1 – 202.2 mm²). There were 29 shoulder lesions (74.4%) and 17 cystic lesions (43.6%). Eleven patients (28.2%) had a prior history of bone marrow stimulation via microfracture.

Clinical Outcomes

Clinical outcome measurements are listed in Table 2 . The mean FAOS scores improved from a pre-operative score of 51.9 ± 16.0 to a postoperative score of 75.3 ± 21.9 (P < 0.001). The mean VAS improved from a preoperative score of 6.6 ± 1.8 to a postoperative score of 1.6 ± 1.6 (P < 0.001). The MCID for FAOS scores was 8.0 which was achieved in 32 patients (82.1%). The MCID for VAS scores was 0.9, which was achieved in 37 patients (94.9%). Multiple variable regression analysis found that pre-operative lesion size was the only variable associated with inferior FAOS scores (R² = 0.2228) ( Fig. 2 ).

Table 2.

Patient Reported Outcome Measurement.

PROM	Pre-operative score (M ± SD)	Post-operative score (M ± SD)	Significance	Achieved MCID (N, %)
FAOS	51.9 ± 16.0	75.3 ± 21.9	P < 0.001	32 (82.1%)
VAS	6.6 ± 1.8	1.6 ± 1.6	P < 0.001	37 (94.9%)

PROM = patient reported outcome measurement; SD = standard deviation; FAOS = foot and ankle outcome score; VAS = visual analog scale; N = number.

Figure 2.

Linear regression analysis demonstrating statistically significant correlation between foot and ankle outcome scores (FAOS) and increasing lesion size.

With regards to patient satisfaction, 12 patients were “very satisfied” (30.7%), 16 patients were “satisfied” (41.0%), 4 patients were “neutral” (10.2%), 6 patients were “unsatisfied” (15.4%) and 1 patient was “very unsatisfied” (2.6%).

Postoperative MRI Analysis

Postoperative MRI analysis is listed in Table 3 . Postoperative MRIs were obtained in 33 patients (84.6%) at a mean time of 40.6 ± 42.3 months (range, 6–136 months) following the index procedure ( Fig. 3 ). The mean postoperative MOCART score was 84.7 ± 11.5. There was no statistically significant difference in mean T2 relaxation values between the deep layer at the site of the AOT graft (33.9 ± 4.7 ms) and the deep layer of the adjacent native cartilage (31.4 ± 4.5 ms) (P = 0.17068). There was statistically significant higher mean T2 relaxation values at the superficial layer at the site of the AOT graft (42.9 ± 5.2 ms) compared to the superficial layer of the adjacent native cartilage (35.8 ± 3.8 ms) (P = 0.00424). Post-operative cysts surrounding the graft were identified in 16 patients (41.0%).

Table 3.

Postoperative Magnetic Resonance Imaging (MRI) Analysis.

N	33
Mean follow-up (mo)	40.6 ± 42.3 months
Number of patients with cysts (N, %)	16 (41.0%)
Mean MOCART score	84.7 ± 11.5
T2 relaxation times for superficial layer	AOT graft (42.9 ± 5.2 ms); adjacent native cartilage (35.8 ± 3.8 ms) (P = 0.00424).
T2 relaxation times for deep layer	AOT graft (33.9 ± 4.7 ms); adjacent native cartilage (31.4 ± 4.5 ms) (P = 0.17068).

MRI = magnetic resonance imaging; mo = months; N = number; ms = milliseconds.

Figure 3.

Magnetic Resonance Imaging (MRI) scan of the right ankle 12 years following autologous osteochondral transplantation. (A) Coronal MRI demonstrating excellent incorporation of the graft. (B) T2 mapping MRI demonstrating similar relaxation values between the graft cartilage and the adjacent native cartilage. (C) Axial MRI demonstrating excellent incorporation of the graft.

Complications, Failures and Secondary Surgical Procedures

Data regarding complications, failures and secondary surgical procedures are listed in Table 4 . In total, 17 complications (43.6%) were observed, the most common of which was anterior ankle impingement due to excessive scar tissue formation (25.6%). Nine patients (23.1%) received at least 1 post-operative corticosteroid injection to treat symptomatic anterior ankle impingement, with 7 patients (17.9%) undergoing anterior ankle arthroscopic scar tissue resection for refractory anterior ankle impingement. At final follow-up, 5 patients (12.8%) presented with recurrent symptoms of anterior ankle impingement and rated their satisfaction with the procedure as “unsatisfied.” Other complications included symptomatic hardware in 5 patients (12.8%), infection in 1 patient (2.6%) and transient donor site knee pain in 1 patient (2.6%) which resolved following physical therapy.

Table 4.

Complications, Failures and Secondary Surgical Procedures.

Complications (N, %)	17 (43.6%)
Excessive scar tissue formation (N, %)	10 (25.6%)
Symptomatic hardware (N, %)	4 (10.4%)
Wound infection (N, %)	2 (5.2%)
Transient donor site knee pain (N, %)	1 (2.6%)
Failures (N, %)	2 (5.1%)
Secondary surgical procedures (N, %)	15 (38.5%)
Anterior ankle arthroscopic scar tissue resection (N, %)	7 (17.9%)
Removal of symptomatic hardware (N, %)	2 (5.2%)
Retrograde drilling (N, %)	2 (5.2%)
Anterior talofibular ligament reconstruction (N, %)	1 (2.6%)
Peroneus longus tenosynovectomy (N, %)	1 (2.6%)
Ankle arthrodesis (N, %)	1 (2.6%)
Total ankle arthroplasty (N, %)	1 (2.6%)

N = number.

In total, there were 2 failures (5.1%) at final follow-up, for which 1 patient underwent a total ankle arthroplasty (2.6%) and 1 patient underwent an ankle arthrodesis (2.6%) at a mean time of 71.2 ± 14.3 months following the index AOT procedure. Both of these patients had a history of prior bone marrow stimulation via microfracture. Additionally, a cyst surrounding the graft on postoperative MRI at mid-term follow-up was identified in both patients.

In total, there were 15 secondary surgical procedures (38.5%), the most common of which was anterior ankle arthroscopic scar tissue resection in 7 patients (17.9%). Other procedures included removal of symptomatic hardware in 2 patients (5.2%), retrograde drilling in 2 patients (5.2%), anterior talofibular ligament reconstruction in 1 patient (2.6%), peroneus longus tenosynovectomy in 1 patient (2.6%), ankle arthrodesis in 1 patient (2.6%) and total ankle arthroplasty in 1 patient (2.6%).

Discussion

The most important finding of this retrospective review was that AOT for the treatment of large OLTs produced a 94.9% survival rate at a minimum of 10-year follow-up. Increasing lesion size was associated with inferior clinical outcomes. A history of prior microfracture procedure and the presence of a post-operative cyst were identified in the cohort of patients who required a joint sacrificing procedure. The findings of this study indicates that AOT is a viable long-term surgical strategy for the treatment of large OLTs.

An osteochondral lesion involves damage to the osteochondral unit, which is composed of hyaline cartilage, subchondral plate and subchondral bone. ¹ A large OLT (diameter > 10 mm or area > 100 mm²) typically has involvement that extends down to at least the subchondral plate, thus direct replacement of the osteochondral unit via procedures such as AOT is warranted. The AOT procedure was initially described by Kennedy et al. ⁶ in 2011. The authors performed a retrospective case series which included 72 patients who underwent AOT at a short-term follow-up of 28 months. This study reported excellent post-operative FAOS scores of 86.19 and short-form 12 scores of 88.63 at 28-months follow-up, together with a mean return to sport time of 12 weeks. Furthermore, Shimozono et al. ¹¹ retrospectively compared 26 patients who underwent AOT and 16 patients who underwent allogenous osteochondral transplantation at a mean follow-up of 24 months. The study found superior postoperative FAOS scores, superior MOCART scores, lower postoperative cyst formation and lower failure rates in the AOT cohort compared to the allograft cohort at short-term follow-up. There is scant literature evaluating outcomes following AOT for the treatment of large OLTs at long-term follow-up. Toker et al. ¹² reported excellent clinical outcomes and low failure rates following AOT in a cohort of 20 patients at a mean follow-up time of 143.5 months. This current retrospective review mirrors the findings of Toker et al.’s retrospective case series. Satisfactory mean postoperative functional scores were observed at final follow-up, as evidenced by mean FAOS scores of 75.3 ± 21.9 and mean VAS scores of 1.6 ± 1.6 at final follow-up. The FAOS score is the only validated patient reported outcome measurement for the treatment of OLTs, ¹³ which reinforces that AOT produces reliable functional outcomes at long-term follow-up. Although this study demonstrates satisfactory FAOS scores at long-term follow-up, these scores are notably inferior compared to the FAOS scores reported in the initial study by Kennedy et al. ⁶ at 28 months follow-up. There are numerous reasons which may elucidate the degradation of these clinical outcomes.

This current review found that of the 2 patients who were in the failure cohort, both patients underwent bone marrow stimulation via microfracture prior to AOT. Microfracture involves perforating the subchondral plate with awls or picks to stimulate the recruitment of local mesenchymal stem cells in an effort to promote chondrogenesis via differentiation of these stem cells into mature chondrocytes. ¹⁴ However, Shapiro et al. ¹⁵ evaluated the effects of microfracture in an in vivo rabbit osteochondral lesion model, where at 6 weeks, the reparative cartilage tissue was “hyaline-like” and was composed predominantly of type-2 collagen and proteoglycans. However, this “hyaline-like” cartilage eventually degraded to inferior fibrocartilage, with de-differentiation of the type-2 collagen to type-1 collagen and diminution of proteoglycans. In addition to the histological repair cartilage deficiencies associated with microfracture, microfracture leads to irrevocable damage to the subchondral plate, which is a critical component of the osteochondral unit. ¹⁶ The subchondral plate bears approximately 30% of the mechanical compressive load and facilitates cross-talk with the articular cartilage. ¹⁶ Thus, it is plausible that the destabilization of the integrity of the talar dome caused by microfracture may extend beyond the limits of the focal osteochondral lesion. Patients undergoing an AOT procedure following a prior microfracture procedure should be counseled pre-operatively regarding the potential inferior outcomes compared to patients undergoing a primary AOT procedure.

Postoperative MRIs with T2-mapping were obtained at mid-term follow-up in this patient cohort. T2-mapping MRI evaluates the integrity, orientation and concentration of the collagen fiber network, and the cartilaginous water content of the articular cartilage in a multi-echo spin-echo (MESE) sequence. ¹⁷ This specialized biochemical MRI technique aims to identify early cartilage degeneration prior to morphologic degradation. This current review demonstrated statistically significant prolongation of T2 relaxation times in the superficial layer of the graft compared to the superficial layer of the adjacent native cartilage. T2 relaxation times indicate the tissue’s anisotropy, the integrity of the collagen framework, and the water content within the tissue. ¹⁸ The discrepancy in the relaxation times between the graft and the adjacent cartilage may reflect damage to the donor graft at the time of implantation, incongruences at the graft’s edges or catabolic degradation of the graft due to the pro-inflammatory environment of joint in the early stages of repair following AOT. It has been hypothesized that a dysfunctional collagen structure at the superficial layer and incongruences at the periphery of the graft may be linked to the development of postoperative cysts, which may be a potential mode of failure. ¹⁹ This current review found that both patients who failed AOT presented with cysts identified on postoperative MRIs at mid-term follow-up. Although prior studies have demonstrated that postoperative cysts following AOT may not be clinically significant, these studies are limited by short-to-mid-term follow-up. Thus, these cystic changes may be a harbinger for failure of the graft. However, only 5 patients in this cohort obtained MRIs at greater than 10-years postoperatively, making it difficult to ascertain if cysts identified at long-term follow-up may also be associated with poor clinical outcomes.

Multiple variable regression analysis found that increasing lesion area was associated with inferior clinical outcomes. These findings contradict prior studies which examined the impact of lesion size following AOT. A retrospective case series of 52 patients by Kim et al. ⁹ demonstrated no correlation between lesion size and clinical outcomes following AOT. Furthermore, Woelfle et al. ¹⁰ evaluated 32 patients who underwent AOT for OLT at 29-months follow-up, and found that increasing lesion area was not associated with poor AOFAS and VAS scores. However, these studies are limited by the short-term follow-up in comparison to this current retrospective study. OLTs with larger areas treated by AOT may decline gradually over time leading to a degradation in clinical outcomes at long-term follow-up.

In total, 10 patients (25.6%) presented with anterior ankle impingement due to excessive scar tissue formation, of which 7 patients (17.9%) underwent arthroscopic debridement. Additionally, 5 patients (12.8%) presented with recurrent symptoms of anterior ankle impingement at final follow-up. Numerous mechanisms exist which can contribute to the high rates of scar tissue formation following AOT. The open, aggressive nature of the procedure in order to gain access to the lesion creates a pro-inflammatory, pro-fibrotic micro-environment in the joint. ²⁰ Prolonged periods of non-weightbearing in the early postoperative period further propagates fibrous tissue development. ²⁰ Additionally, cBMA, which was utilized in all patients in this series, has been demonstrated to have a proclivity for the development of excessive scar tissue formation due to the presence of the pro-fibrotic TGF-β1 isoform. ²¹ Prophylactic measures to prevent the development of anterior ankle impingement include early ankle plantarflexion and dorsiflexion exercises, aggressive formal physical therapy and early weightbearing. ²⁰ Intra-articular corticosteroids to breakdown the excessive scar tissue should be used sparingly given its catabolic degradative effects on the articular cartilage. ²² Following failure of conservative management, arthroscopic resection of the impinging scar tissue and osteophytes is warranted. In recent years, in-office needle arthroscopy (IONA) has been utilized as a powerful tool to address numerous ankle pathologies, including anterior ankle impingement.^{23 -25} This is performed in the office setting with patient wide awake without the use of a tourniquet, facilitating immediate weightbearing and rapid return to activity. Colasanti et al. ²³ demonstrated excellent postoperative functional outcomes and accelerated return to sports following IONA for the treatment of anterior ankle impingement. IONA also permits concomitant assessment of the AOT graft and delivery of orthobiologics such as PRP if necessary.

This current retrospective study must be interpreted in light of its numerous limitations and inherent biases. Firstly, this study is limited due to its retrospective nature. Secondly, postoperative MRIs were obtained in only 84.6% of patients with only 5 MRIs obtained at greater than 10-year follow-up. The 10% of patients lost to follow-up may have had poor outcomes, which has not been accounted for in the analysis of this study. Finally, concomitant procedures were performed in 30.8% of patients, adding additional heterogeneity to the patient population thus potentially reducing the validity of the results.

Conclusion

This retrospective review found that AOT for the treatment of large OLTs produced a 94.9% survival rate at a minimum of 10-year follow-up. Increasing lesion size was associated with inferior clinical outcomes, suggesting that large OLTs treated with AOT may degrade over time. A history of prior microfracture procedure and the presence of a post-operative cyst were identified in the cohort of patients who required a joint sacrificing procedure. The findings of this study indicates that AOT is a viable long-term surgical strategy for the treatment of large OLTs.