Introduction

Ankle sprains as a common joint injury and nearly 27,000 injuries per day occurred in the United States [1]. Moreover, nearly 70% of sprains and fractures involving the ankle could cause osteochondral lesions [2]. The entire body weight was supported, and stabilisation by the ankle and the small area of distribution caused the joint to be sensible to shearing stresses [3]. Osteochondral defects of the talus contained the lesion in the subchondral bone and its overlying cartilage, and mostly osteochondral defects of the talus occurred after an ankle fracture or lateral ankle ligament rupture [2, 4]. Moreover, osteochondral defects could progress to cystic lesion and induce deep ankle pain during activity, prolonged swelling, diminished range of motion, and synovitis [5, 6].

Recently, autologous chondrocyte implantation (ACI) is widely used to cover the anatomical defects for repairing osteochondral defects, which is based on two-time surgical procedures. The first procedure includes revision arthroscopy of the joint with the lesion area as well as a trephine of healthy cartilage tissue and then graft obtained by stimulating chondrocyte mitosis. The second procedure is conducted by implant matrix using arthroscopy or arthrotomy of the medial malleolus to expose the injury area [7, 8]. The effectiveness of ACI for patients with osteochondral defects of the talus has already been demonstrated, whereas, the treatment effectiveness was variable across studies [9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31]. Therefore, this systematic review and meta-analysis aimed to assess the effectiveness of ACI for patients with osteochondral defects of the talus.

Methods

Data sources, search strategy, and selection criteria

This systematic review and meta-analysis were performed and reported using the Preferred Reporting Items for Systematic Reviews and Meta-Analysis statement [32]. Studies that assessed the effectiveness of ACI for patients with osteochondral defects of the talus were eligible in our study. Electronic databases PubMed, Embase, and the Cochrane Library were systematically searched to identify eligible studies using the following search terms until November 2020: (“Chondrocytes” AND “Transplantation, Autologous” OR “autologous chondrocyte transplantation” OR “autologous chondrocyte implantation”) AND (“Ankle” OR “Ankle Joint” OR “Ankle injuries” OR “Talus” OR “talus” OR “talar”). Furthermore, the reference lists of potentially relevant reviews and original articles were also manually searched to identify any new eligible study.

The literature search and study selection were independently performed by two reviewers, and the inconsistency was settled by a group discussion. Studies were included if they met the following criteria: (1) patients: osteochondral defects of the talus; (2) intervention: ACI; (3) outcome: success rate (defined as American Orthopedic Foot and Ankle Society (AOFAS) score > 80) and AOFAS score; and (4) study design: case series and observational and randomised controlled trials.

Data collection and quality assessment

The data collection and quality assessment were independently conducted by two reviewers, and any disagreement between reviewers was settled by discussion mutually until a consensus was reached. The items collected from each study included the first author’s name, publication year, country, evidence level, sample size, age, number of males and females, technique, first-line or revision ACI, subchondral bone grafting, indication, size, follow-up, success rate, assessment tool, and reported outcomes. The modified Coleman methodology score was determined for each study to assess the study quality and the different types of detected bias [33].

Statistical analysis

After ACI, the success rate and AOFAS score were assigned as categorical and continuous data, respectively. Then, the random-effects model was used to calculate the pooled incidence of success rate and AOFAS score [34, 35]. After this, I2 and Q statistics were applied to assess the heterogeneity across the included studies, and significant heterogeneity was defined as I2 > 50.0% or P < 0.10 [36, 37]. Sensitivity analyses for success rate and AOFAS score were also performed to assess the impact of a single study on the overall conclusion [38]. Subgroup analyses for success rate and AOFAS score were also performed based on age, technique, indication, size, or follow-up duration, and the interaction P test was performed to assess the difference between subgroups [39]. Furthermore, Funnel plot, Egger, and Begg tests were performed to assess publication bias for success rate and AOFAS score [40, 41]. All reported P values are two-sided, and a significant difference was defined as P < 0.05. The STATA software (version 10.0; Stata Corporation, College Station, TX, USA) was used to perform all of the analyses in this study.

Results

Literature search

By initial electronic searches, a total of 786 articles were identified, and 381 studies were retained after exclusion of the duplicate articles. After this, 311 studies were removed because of irrelevant topics. The remaining 70 studies were retrieved for further full-text evaluations, and 47 studies were excluded because of osteochondral defects in knee (n = 21), no sufficient data (n = 15), and other interventions (n = 11). After reviewing the reference lists of relevant studies, three potentially included studies were found, and all of these studies were included in electronic searches. Therefore, the remaining 23 studies were selected for final meta-analysis [9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31] (Fig. 1).

Fig. 1
figure 1

The PRISMA flowchart for the literature search and study selection

Full size image

Study characteristics

Table 1 shows the characteristics of the included studies and patients. All of the 23 included studies were designed as case series, and a total of 458 patients with osteochondral defects of the talus were recruited. The sample size ranged from 7 to 46, and the follow-up duration ranged from 12.0 to 154.8 months. For all of the included studies, the evidence level was IV. Six studies applied periosteum-covered ACI, and the remaining 17 studies applied matrix-associated ACI. All of the included studies had a level of evidence of IV. The mean modified Coleman methodology score was 48.1, and the score in each study ranged from 35 to 65, which suggested that all of the included studies were of low to moderate quality.

Table 1 The baseline characteristics of identified studies and included patients
Full size table

Success rate

A total of 17 studies reported the incidence of success rate for patients after ACI. We noted that the pooled incidence for success rate was 89% (95% confidence interval (95%CI) 85%–92%; P < 0.001; Fig. 2), and unimportant heterogeneity was detected across the included studies (I2 = 30.1%; P = 0.117). Sensitivity analysis indicated that the incidence of success rate was robust after sequentially excluding individual study (Online Resource 1). Subgroup analysis indicated that the pooled incidence of success rate was > 80.0% in all subgroups, and age, technique, indication, size, or follow-up duration was not affecting the incidence of success rate (Table 2). Furthermore, a significant publication bias for success rate was found (P value for Egger: < 0.001; P value for Begg: 0.002; Online Resource 2), and the pooled incidence of success rate was not altered by adjusted using the trim and fill method [42].

Fig. 2
figure 2

Effect of ACI on the incidence of success rate

Full size image
Table 2 Subgroup analyses for success rate and American Orthopaedic Foot and Ankle Society
Full size table

AOFAS score

The AOFAS score for patients after ACI was reported in a total of 19 studies. We noted that the AOFAS score after ACI was 86.33 (95% CI 83.33–89.33; P < 0.001; Fig. 3), and no evidence of heterogeneity was seen among the included studies (I2 = 0.0%; P = 0.603). The pooled AOFAS score after ACI was stable after sequentially excluding single study (Online Resource 1). Subgroup analyses revealed that the pooled AOFAS score was lower when pooling studies for patients with chondral lesion (Table 2). Moreover, we noted that the AOFAS score could be affected after ACI for patients with osteochondral defects of the talus (P = 0.006). There was no significant publication bias for the AOFAS score after ACI (P value for Egger: 0.794; P value for Begg: 0.069; Online Resource 2).

Fig. 3
figure 3

Effect of ACI on AOFAS score

Full size image

Adverse events

Ten out of the included studies reported complications after ACI [11,12,13, 15,16,17,18, 23,24,25, 28]. Six studies indicated no intraoperative or postoperative complications [11, 13, 15, 16, 24, 28]. Whittaker et al. reported one patient (10.0%) who presented with superficial infection of the ankle [12]. Schneider et al. reported two patients (10.0%) with anterior graft impingement, two patients (10.0%) with recurrent pain associated with hardware, and two patients (10.0%) who presented with clear failures combined with persistent pain and synovitis [18]. Lee et al. showed that the prevalence of nonunion and delayed unions of the osteotomy sites was 2.6% and 5.3%, respectively. Moreover, nine ankles (29.0%) sustained damaged medial malleolar cartilage [23]. Finally, Buda et al. reported three patients (15.0%) who presented with adhesions or joint effusion [25].

Discussion

The treatment effectiveness of ACI for patients with osteochondral defects of the talus has already been illustrated in numerous studies, while the effect was variable and not confirmed to date. This systematic review and meta-analysis was performed and assessed the effectiveness of ACI on the incidence of success rate and AOFAS score. A total of 458 patients with osteochondral defects of the talus were identified from 23 case series studies, and the characteristics of patients were broad across the included studies. We noted that the pooled success rate was high, and the AOFAS score was improved after ACI. Moreover, the AOFAS score after ACI for patients with osteochondral defects of the talus could be affected by age.

In a previous systematic review, 16 studies were identified and revealed that the ACI should be considered as a promising treatment for osteochondral and chondral defects of the talus [43]. Erickson et al. conducted a systematic review of 19 studies and found that there were no significant differences among the combination of open or arthroscopic matrix-associated ACI and periosteum-covered ACI for talar osteochondral lesions less than 2.5 cm2 [44]. However, these two studies have just given the qualitative analysis for the included studies, and according to patients’ characteristics, the quantitative analysis was not illustrated. Therefore, this systematic review and meta-analysis were performed to assess the treatment effectiveness of ACI on success rate and AOFAS score for patients with osteochondral defects of the talus.

The summary success rate for the effect of ACI was 89% (95% CI 85%–92%; P < 0.001), and the success rate in each study ranged from 50 to 100%. In Giza et al.’s study, 10 patients with osteochondral defects of the talus were recruited and only five patients showed significant improvement in AOFAS score [19]. Five of the included studies presented 100% of success rate for patients treated with ACI [9,10,11, 25, 29]. Although subgroup analyses revealed that age, technique, indication, size, or follow-up duration did not affect the success rate for patients with osteochondral defects, we noted that the success rate was higher when the age of patients was < 35.0 years, patients were treated with periosteum-covered ACI, patients were with osteochondral defects, and the lesion size was ≥ 2.0 cm2. These results suggested the ACI might give a superior effect on the success rate in patients with specific characteristics.

We noted that the pooled AOFAS score after ACI was 86.33 (95% CI 83.33–89.33) for patients with osteochondral defects of the talus, and the AOFAS score in the individual study ranged from 74.7 to 92.7. Subgroup analysis suggested the AOFAS score after ACI was high in the subgroups of the age of patients < 35.0 years, patients treated with matrix-associated ACI, patients with osteochondral defects, lesion size ≥ 2.0 cm2, and follow-up duration ≥ 36.0 months. Moreover, after ACI, there was a significant difference in AOFAS score when stratified by age of patients. Studies have already revealed that patients’ age could affect cartilage repair and the clinical outcome after ACI [15, 45], while this result was not consistent [46]. The potential reason for a beneficial effect of ACI on younger patients could be the restore ability of younger patients was stronger than elderly patients.

Several shortcomings of this study should be discussed. First, all of the included studies were designed as case series, and the evidence level was lower (IV). Second, in a smaller number of studies, the comparisons of various treatment strategies were reported [47], and in this study, the lack of controlled treatment strategies and the superiority or inferiority effects of ACI compared with other techniques were not addressed. Third, the analysis based on crude data and the potential role of other characteristics was not adjusted. Fourth, the background therapies including physical therapy, bracing, casting, and nonsteroidal anti-inflammatory medication were not mentioned, which could affect the treatment effectiveness of ACI. Fifth, the current study was not registered, and the transparency was restricted. Sixth, the AOFAS score is not a validated score, which could affect the treatment effects of ACI. Finally, inherent limitations for meta-analysis based on pooled data, including inevitable publication bias, and the restricted detail analyses.

Conclusions

The pooled success rate and AOFAS score after ACI for patients with osteochondral defects of the talus were 89% (95% CI 85%–92%) and 86.33 (95% CI 83.33–89.33), respectively. Moreover, the treatment effectiveness of ACI on the AOFAS score could be affected by age of patients. Further controlled compared studies should be conducted to compare the efficacy and safety of ACI with other techniques for patients with osteochondral defects of the talus.