Outcomes Following Repeat Ankle Arthroscopy and Microfracture for Osteochondral Lesions of the Talus

Introduction

Osteochondral lesions of the talus (OLTs) are congenital defects or post-traumatic injuries of the talar articular cartilage and surrounding bone. They most commonly occur in the central-medial or central-lateral talar dome.^1,2 Although 24% of patients presenting with symptomatic OLTs deny a history of acute or repetitive trauma, such lesions are fairly common following acute injury.^3,4 Arthroscopic inspection of the talar dome in the setting of an ankle fracture has estimated a 28% to 70% incidence of OLTs,^5-8 although not all of these lesions may impact patient outcomes. ⁹ The OLTs are also common in the setting of sprains with chronic instability, with lesions identified in 17% to 40% of patients undergoing ankle stabilization procedures.^10-13

While some patients with symptomatic OLTs have success with nonoperative management, roughly 50% of patients have persistent symptoms.^14,15 When nonoperative modalities fail, arthroscopy with bone marrow stimulation (BMS) techniques (microfracture, abrasion chondroplasty, debridement and drilling) is considered the first-line treatment for smaller contained lesions (<150 mm² or <15 mm in diameter)^16-18 in both the presence and absence of cysts.^19,20 The BMS techniques have demonstrated good to excellent outcomes as a primary intervention and are an attractive treatment option given relatively low cost, low technical demands, and lower postoperative pain in comparison with other treatment strategies. ²¹

However, not all primary interventions with arthroscopy and BMS techniques are successful, and there is comparatively less available evidence to guide the next steps in secondary operative management. In particular, the efficacy of repeat arthroscopy is debated and there is mixed outcome data available in the literature in less than 80 patients. ¹⁷ ,^22-26 Collectively, these data are limited and inconclusive. Despite these limited data, if repeat arthroscopy was efficacious in most of the patients, this may be preferable as a second procedure before proceeding with other techniques which are more expensive and technically demanding.

The purpose of the current study was to report our institutional outcomes with repeat arthroscopy and microfracture in the management of OLTs. We hypothesized that revision arthroscopy with microfracture would be a safe procedure and result in modest patient satisfaction.

Materials and Methods

After obtaining Institutional Review Board (IRB) approval, we searched our department database for patients with the assigned CPT codes 29891-29899 (Procedures involving ankle arthroscopy). From 1997 to 2015, 611 arthroscopic ankle procedures were performed. Charts from these procedures were reviewed, identifying 96 patients undergoing a secondary procedure involving ankle arthroscopy. Of the 96 patients, 38 underwent repeat ankle arthroscopy and microfracture for an OLT by one of the 3 fellowship-trained foot and ankle surgeons at our institution (Figure 1). Patient charts were reviewed and patients were excluded from further analysis if they were less than 18 years of age at the time of their repeat surgery, if the operative note did not detail the size of the lesion, or if they had undergone concomitant corrective osteotomy, arthroscopic arthrodesis, or any other open procedure. Fourteen patients remained after applying these exclusion criteria (8 males and 6 females). In reviewing the operative notes for these patients, no biologics were used in this patient cohort.

OPEN IN VIEWER

Figure 1.

Representative case. (A) An OLT (osteochondral lesion of the talus) previously treated with microfracture has incomplete fibrocartilage fill. (B) Probing of the lesion reveals it has an unstable fibrocartilage flap. (C) The lesion was debrided to a stable rim using curettes. (D) Repeat microfracture procedure was performed with small joint awls.

All patients were successfully contacted by phone and consent was obtained via a standardized IRB-approved script (Figure 2). Patients were asked to rate their overall satisfaction with their surgery (graded from 1 to 10 with 10 being the most satisfied) and their current pain level (graded from 1 to 10 with 10 being the most painful). Patients were also questioned about their ability to return to normal sport or activity level (yes/no), whether they had undergone any subsequent surgeries on the involved ankle (yes/no), whether they would undergo this procedure again (yes/no), and last whether they required usage of assistive devices (yes/no).

OPEN IN VIEWER

Figure 2.

Patient questionnaire. Patients were contacted by phone, and the above questions were asked in a standardized format according to our IRB-approved script. Dr “X” refers to the treating surgeon for that patient. Abbreviation: IRB, Institutional Review Board.

In addition to these data, patients were grouped for analysis based on the documented size of their OLT in their operative note, with small and large lesions defined as ≤150 mm² or >150 mm², respectively. A separate analysis was also performed comparing patients with cystic and noncystic OLTs as documented on preoperative magnetic resonance imaging (MRI) or the operative note. Patient age, body mass index (BMI), follow-up time, satisfaction scores, pain rating, rate of return to sport, willingness to undergo the procedure again, and need for assistive devices were compared. Continuous data were analyzed using t tests and categorical data were compared using Fisher exact test.

Results

The mean age of the study population at the time of surgery was 39 years (range 19-59). Average follow-up time from surgery was 5.1 ± 2.9 years (range 0.5-10.0; Table 1). The BMI of the study population was 30.1 ± 6.7. Thirteen patients had medial lesions, and 2 patients had lateral lesions (1 patient with both medial and lateral lesions). Mean lesion size was 168 mm² (range 78-300 mm²). Seven patients had cystic lesions, and 7 had noncystic lesions. All lesions were found to be uncontained. Twelve of 14 patients reported they would undergo the procedure again. The average scores for patient satisfaction and residual pain were 7.6 ± 3.5 and 4.7 ± 3.4, respectively. Further surgery was performed in 3 of 14 patients (1 ankle fusion, 1 repeat arthroscopic debridement and loose body removal, and 1 unknown procedure). There was one postoperative complication (a superficial venous thrombus).

OPEN IN VIEWER

Table 1.

Study Demographics.

Age, y	39.4 ± 11.4
Male:Female	8:6
BMI, kg/m2	30.1 ± 6.7
Average follow-up, y	5.1 ± 2.9
Average lesion size, mm2	168 ± 57.5
Number of small lesions, ≤150 mm2	5
Number of large lesions, >150 mm2	9
Medial talar lesions	13
Lateral talar lesions	2
Cystic talar lesions	7
Noncystic talar lesions	7
Contained talar lesions	0

Abbreviation: BMI, body mass index.

Five patients had small lesions (≤150 mm²) and 9 patients had large lesions (>150 mm²). Patients with larger OLTs were younger at the time of repeat arthroscopy (mean age 34.4 years vs 48.6 years, P = .02). There were no significant differences in patient sex, BMI, or lesion location between these groups (P > .05). Patients with small and large lesions had similar postoperative pain scores (4.2 ± 4.1 vs 4.9 ± 3.2, P = .71), postoperative satisfaction levels (6.4 ± 4.9 vs 8.3 ± 2.5, P = .34), reoperation rates (20% vs 22%, P = 1.0), and willingness to repeat the procedure (80% vs 78%, P = 1.0) (Table 2). In total, 33% of patients with small OLTs and 40% of patients with large OLTs reported using assistive devices including ankle bracing and orthotics (P = 1.0). In total, 80% of patients with small OLTs were able to return to their sport or desired activity, compared with 33% of patients with larger lesions (P = .26).

OPEN IN VIEWER

Table 2.

Effect of Lesion Size and Subchondral Cysts.

	Residual pain score (0-10)	Satisfaction score (0-10)	Reoperation rate, %	Return to activity, %	Use of assistive devices, %	Willingness to repeat treatment, %
Small OLT (≤150 mm2)	4.2 ± 4.1	6.4 ± 4.9	20	80	40	80
Large OLT (>150 mm2)	4.9 ± 3.2	8.3 ± 2.5	22	33.3	33.3	77.7
P value	.71	.34	1.0	.26	1.0	1.0
Cystic OLT	3.7 ± 3.5	8.0 ± 3.6	14	43	14	86
Noncystic OLT	5.6 ± 3.3	7.3 ± 3.6	29	57	57	71
P value	.31	.72	1.0	1.0	.56	1.0

Abbreviation: OLT, osteochondral lesions of the talus.

Seven patients had cystic OLTs, and 7 patients had lesions that were noncystic. Patients in these groups had similar patient and lesion characteristics (P > .05 for age, sex, BMI, lesion location, and lesion size; Table 2). Similarly, there were no significant differences in postoperative pain, postoperative satisfaction, rates of return to sport, or reoperation rates in patients with cystic and noncystic lesions (P > .05).

Discussion

The management of symptomatic OLTs previously treated with arthroscopy remains controversial as published results are variable and the overall body of evidence is limited. Operative options include repeat arthroscopy with BMS techniques with or without biologic augmentation (platelet-rich plasma or hyaluronic acid,^27,28 concentrated bone marrow aspirate, ²⁹ mesenchymal stem cells, ³⁰ or micronized cartilage), ³¹ osteochondral autograft and allograft transplantation,^32-34 autologous chondrocyte implantation, ³⁵ and matrix-induced autologous chondrocyte implantation. ³⁶ Autografting/allografting and chondrocyte implantation techniques are costlier, more technically demanding, more invasive, and require a longer duration of recovery and return to sport in comparison with arthroscopy with BMS. ³⁷ Thus, it is important to better understand the efficacy of repeat arthroscopic BMS techniques after failed primary interventions.

In the current study, we evaluated the clinical outcomes of OLTs treated with repeat arthroscopy and microfracture following failed primary arthroscopy and microfracture. At an average of 5-year follow-up, our results demonstrate that repeat arthroscopy is safe and leads to a reasonable degree of patient satisfaction, with nearly 80% of patients responding that they would choose the same course of treatment if given the opportunity. However, patients on average reported a moderate level of persistent pain, and 21% of patients required subsequent operative intervention to address their symptoms. Overall, our results suggest patients should be educated about the frequency of residual pain and the potential need for a more invasive procedure despite an overall high satisfaction rate with repeat arthroscopy.

Furthermore, outcome measures in revision arthroscopy were similar in patients with large (>150 mm²) and small (<150 mm²) lesions, and in patients with and without subchondral cysts. More patients with smaller lesions (80%) returned to sport or their desired activity when compared with patients with larger lesions (33%). However, our small patient population lacked adequate power to detect statistically significant differences in these outcomes. In addition, it is important to consider that the ideal lesion size for these analyses is somewhat debatable. The most recent practice guidelines have suggested an ideal lesion size of <10 mm in diameter or <100 mm² for BMS techniques, ³⁸ an area/size smaller than originally described (150 mm²). ¹⁶ However, a recent systematic review highlights this conclusion has been reached without the adoption of a uniform measurement technique. ³⁹ The originally suggested ideal area of 150 mm² was developed from MRI measurements, using a formula to calculate the area of an ellipse (area = sagittal length × coronal length × 0.79). ¹⁶ Follow-up studies have used their own MRI measurement techniques or have described arthroscopic measurements, which have been shown to differ from imaging measurements given poor interobserver reliability and given measurements are often made after debriding unstable cartilage that secondarily increases lesion size. ³⁹ Furthermore, outcomes regarding lesion size have been analyzed without the use of a validated scoring system for OLTs. ³⁹ Given the lack of uniformity in published measurements and the absence of a validated outcome metric to determine a critical size, we feel using a critical cutoff of 150 mm² was appropriate for our comparisons.

The initial small-cohort studies assessing the efficacy of repeat arthroscopy found poor clinical outcomes. In studies by Schimmer et al ²³ and Robinson et al, ²⁶ 11 total patients were reported to have poor or fair results. Importantly, neither of these studies described the size of the lesions being treated, which is an important characteristic to consider given the increased rate of failure in treating larger lesions in the primary setting. ³⁹ Our results agree with 2 larger and more recent studies. Schuman et al found that 75% (12/16) of patients experienced good or excellent results at an average 5.5-year follow-up after revision arthroscopic debridement and drilling. However, this work did not report on lesion size. ²⁴ Similarly, Savva et al ²² demonstrated 92% satisfaction in 12 patients with small (<150 mm²), noncystic OLTs who underwent revision arthroscopy with microfracture or drilling at a mean 6.7 years. We found similar outcomes to these more recent studies, although we did not find measurable differences in patient outcomes based on the size of the lesions being treated. However, we recognize that our analysis of lesion size was underpowered.

While the results of our study demonstrate excellent safety and moderate efficacy for repeat arthroscopy, the high rate of residual pain and reoperation at midterm follow-up warrants consideration of and comparison with other available treatment options. Allografting, autografting, and chondrocyte implantation procedures have demonstrated durable midterm outcomes in the primary treatment of OLTs,^40-47 and there is a growing body of evidence indicating successful outcomes in revision surgery following failed arthroscopic treatment.^25,43 In addition, repeat arthroscopy has been directly compared with osteochondral autograft transplantation (OAT) in 44 patients having failed primary arthroscopy. ²⁵ At a mean follow-up of 50 months, Yoon et al found that 82% of patients had good or excellent outcomes with OAT in comparison with 32% of patients undergoing repeat arthroscopy. The OAT demonstrated significant advantages in long-term results for visual analog scale (VAS) pain, AOFAS (American Orthopaedic Foot and Ankle Society) score, and clinical survival (as defined by the absence of pain or persistent symptoms, no subsequent additional operative interventions, and AOFAS score >80). However, more than half of the study population was found to have subchondral cysts, which have been associated with poor outcomes in some studies ²⁶ but not in others.^19,20 In addition, 86% of the treated defects were uncontained, which has been shown to negatively affect the outcome of arthroscopy with BMS techniques. ⁴⁸ Finally, their results of OAT as a secondary intervention are more favorable than the published outcomes for primary OAT. At an average of 48 months after OAT following failed primary arthroscopy, Yoon et al ²⁵ demonstrated a VAS pain score of 1.9 and AOFAS score of 85.3, with 18.2% of patients experiencing fair or poor outcomes and 0% requiring further operative intervention. In comparison, at an average 54 months after a primary OAT procedure, Raikin ⁴⁹ demonstrated average VAS pain scores of 3.3 and AOFAS scores of 83.0, with 26% of patients experiencing fair or poor outcomes and 13% undergoing a revision of their primary OAT to an arthrodesis. Similarly, at an average 38 months after primary OAT, El-Rashidy et al ⁵⁰ demonstrated VAS pain scores of 3.3 and AOFAS scores of 79.0, with 26% of patients experiencing poor outcomes and 26% undergoing revision to arthrodesis or arthroplasty. Collectively, these findings challenge the only published direct comparison between repeat arthroscopy and OAT as treatment for OLTs failing an index arthroscopy with BMS techniques.

The present study has several limitations. Although our outcomes were similar in patients with small and large lesions as well as cystic and noncystic lesions, our population size was small and our study was not adequately powered to detect these differences. Our study and the existing published studies evaluating repeat arthroscopy for OLTs are small (all less than 25 patients), identifying the relatively low frequency of this procedure. Combining the available data in a meta-analysis would be of limited efficacy given the heterogeneity in available data and the lack of key information in some existing publications (lesion size, description of containment, presence of subchondral cysts). There is a need for a multicenter analysis in which an adequate number of patients can be treated and have their outcomes compared based on the various available treatment options. In addition, the design of our study as a retrospective phone interview introduced recall bias. However, we did not ask patients to rate their pain or outcomes immediately after surgery, but rather describe their current levels of pain and function. Finally, our analysis would have been strengthened by multiple assessments of patient-reported outcomes over time with more robust outcome measures and serial physical examinations. This is of particular interest as patient outcomes after BMS techniques are thought to deteriorate over time. ³⁷ Ideally, this should be performed with a validated outcome scoring system going forward, which to our knowledge has not yet been defined. ⁵¹ Finally, the average BMI of our study population was >30, which is above the threshold (25) associated with poorer clinical outcomes after primary arthroscopy at 10-year follow-up. ⁵² A larger multicenter study is needed to adequately power a similar analysis on the relationship between BMI and clinical outcomes after revision procedures.

In summary, the current study demonstrates that repeat arthroscopy with microfracture is a safe and viable option for patients who have failed primary arthroscopy for osteochondral talar defects. Although there is insufficient evidence to compare BMS techniques with transplantation/implantation procedures in the treatment OLTs failing primary arthroscopy, our work and the work of others highlight that patients can have successful outcomes with repeat arthroscopy. The BMS techniques warrant consideration in a revision setting given the lower cost, faster recovery, decreased operative morbidity, and lower risk of complications compared with transplantation/implantation procedures.