Autologous minced cartilage repair for chondral and osteochondral lesions of the knee joint demonstrates good postoperative outcomes and low reoperation rates at minimum five-year follow-up

The incidence of chondral and osteochondral lesions of the knee is increasing, but treatment remains challenging [13, 52]. Several different cartilage repair techniques have been described, each one aiming to maximize the amount of mature and organized hyaline or hyaline-like cartilage [7, 19]. However, so far, no technique has been able to regenerate normal hyaline cartilage in adults on a regularly basis and no technique has been proven to be superior. Osteochondral autograft transfer system (OATS), autologous chondrocyte implantation (ACI), matrix-assisted chondrocyte implantation (MACI) or several cartilage repair techniques using scaffolds (natural or synthetic) are contemporarily the most commonly therapies for medium to large defects [7, 22]. ACI and MACI are purported to generate hyaline or hyaline-like cartilage, with low associated reoperation rates and favourable clinical outcomes even in complex cases [5, 7, 18, 29, 49]. However, the two-step approach of ACI and MACI, which requires laboratory cell cultivation, results in a high financial and clinical burden. OATS results are reported to be similar to MACI, but allografts availability may be limited and cost intensive.

Recently, the minced cartilage procedure, a technique in which viable autologous cartilage is collected, sliced into small pieces, and reimplanted, has undergone renewed interest [51, 55]. Autologous thrombin and PRP, fibrin glue or a membrane might be additionally used for fragment fixation [42, 50‐52]. Short-term outcomes in literature have been promising, but mid- and long-term results are lacking but urgently needed [8, 10, 42].

The aim of the present single-cohort study is to demonstrate minimum five-year outcome data of a cohort that underwent minced cartilage for the treatment of chondral and osteochondral lesions in the knee. It has been hypothesized that minced cartilage may maintain favourable clinical results over longer term follow-up, both in terms of patient-reported outcome measures and reoperation rates.

Local ethical committee of the Canton of Zurich (KEK-ZH-Nr. 2015–0258) approval was obtained prior to study initiation. Informed consent was signed by all participants. The first 34 consecutive patients who underwent minced cartilage knee surgery between 2015 and 2016 were retrospectively analysed from a prospectively maintained database that routinely collected patient-related outcome data.

All patients were contacted personally by telephone for an interview and questionnaires were collected by mail. Electronic medical records were reviewed to obtain patient demographic, surgical, and imaging data.

A total of 34 consecutive patients treated with minced cartilage were included in the study. A detailed overview of patient characteristics and concomitant procedures is shown in Table 1. Cartilage defect characteristics are displayed in Table 2. The final follow-up after a mean of 65.5 ± 4.1 months was 82.4% (n = 28). Patients lost to follow-up did not differ in baseline characteristics or in intraoperative findings to those with complete mid-term follow-up.

The primary finding of the study was that second-generation minced cartilage procedure for the treatment of chondral and osteochondral lesions of the knee without the necessity for subchondral bone treatment is an effective and safe procedure with good mid-term results in terms of pain and knee function. The present 5-year data demonstrated no significant worsening of pain and function at mid-term compared to short-term follow-up.

The primary aim of minced cartilage is to recapitulate hyaline or hyaline-like cartilage at site of implantation from cell outgrowth, proliferation, and differentiation without the need for a two-step surgical process. Fragmentation of healthy cartilage has been shown to “activate” chondrocytes by increasing tissue surface and therefore promoting outgrowth [37, 51]. Outgrowth of chondrocytes results in proliferation and matrix production. This process is believed to be positively influenced by native joint physical-biomechanical inputs and the osteochondral microenvironment, as mechanical and biological stimuli have been shown to promote proliferation and chondrogenic differentiation [51, 56]. Several animal models have demonstrated the feasibility of minced cartilage showing better results compared to microfracturing, while demonstrating similar outcomes to two-stage autologous chondrocyte implantation [1, 9, 15, 26, 38, 41].

As it currently stands, there is only one prospective randomized clinical trial for single-stage autologous cartilage fragments procedure (CAIS, Cartilage Autograft Implantation System) [10] and three trials involving use of autologous minced cartilage with clinical follow-up of up to 24 months. [8, 12, 42]. Comparing two-year outcomes of patients randomly treated with either microfracturing (MFX) or CAIS shows significantly higher PROMs for CAIS [10]. There was no difference in the number of surgery-related complications, but a higher number of intralesional osteophyte formation in patients treated with MFX [10]. A statistical significant increase in MOCART scores and PROMs were also reported in eight patients with osteochondrosis dissecans treated with a combination of autologous bone and cartilage chips (ADTT, autologous dual-tissue transplantation) [8]. Similar, a statistically significant improvement between pre- and postoperative PROMs values were reported in fifteen patients treated with a novel autologous-made matrix, hyaline cartilage chips and platelet-rich growth factors [12]. In the present study, five-year outcomes indicate maintained low pain scores and knee function which are not statistically different from one- and two-year postoperative outcomes. Despite a slight increase in pain compared to one- and two-year results, the NAS pain values were within the limits of the minimal clinically important difference (MCID) and not statistically significantly different. Further long-term data are needed to assess the postoperative progression over time.

When comparing outcomes of different studies it is important to keep in mind, that several factors including surgical techniques, patient characteristics [48], previous or subsequent surgical interventions [48], defect size and location [23] as well as rehabilitation [20] have an influence on the postoperative outcome and limit therefore the direct comparability. Recently two-stage autologous chondrocyte implantation or osteochondral allograft transplantation (OATS), research showed promising mid-term results after MACI for medium to large defects [20, 24, 31]. A recent prospective randomized controlled trial (RCT) comparing MFX (61.8 ± 21.5) to MACI (68.5 ± 21.2) reported significant higher IKDC scores for the latter [4]. These findings are supported by recent meta-analysis showing no increased risk of clinical failure but superior improvements in PROMs for MACT compared to MFx at short- to mid-term [14] [20]. Contrary, other studies question the superiority of MACI over MFX [34, 35] or even report superiority of MFX especially in patients with only small chondral defects [47]. A prospective, controlled clinical trial studying the safety and efficacy of MACI with spheroid technology reported good and stable improvement of IKDC (74.6 ± 18.7) and KOOS (77.1 ± 18.6) scores after 48 months [46]. This is in accordance with several other mid- to long-term studies reporting stable improvements after ACI [2, 4, 11, 24, 36, 45, 57] When directly comparing ACI to OATS [43] fair to good mid-term IKDC outcomes (IKDC: 50 – 80) were reported [21]. Similarly, no significant differences at mid-term were found in PROM regarding outcomes between ACI and AMIC with VAS for pain scores ranging between 2.3 – 3- two years postoperative [25, 27, 54]. Patient-reported outcomes of the present study (IKDC: 71.6 ± 17.8) are comparable to MACI, OATS and AMIC results of the above-mentioned studies but appear to be higher than those reported for MFX [4]. Gender, concomitant injuries, defect localization, defect characteristics as well as fixation techniques for minced cartilage seem to have little influence on postoperative outcomes. Patients treated with one-staged minced cartilage for chondral and osteochondral knee defects can expect similar postoperative outcomes in the mid-term as compared to other established procedures.

Overall surgery-related complications and revision surgery are relatively low and comparable between different chondral regenerative techniques [18, 28, 44, 46]. Graft hypertrophy after MACI was reported in 12% of the patients after five years with a need for revision surgery in 8% [16]. Similar revision surgery rates were reported after five years in a prospective RCT rates comparing MACI (10.8%) and MFX (9.5%) [4]. In contrast, no graft hypertrophy was observed in a study using ACI (Spherox™, CO.DON AG, Germany; formerly known as chondrosphere) [45]. Mid-term outcome analysis of OATS revealed an 87% graft survival rate at 5 years with a 37% reoperation rate. These results were supported by a recent systematic review reporting similar survival rates (87%) and reoperation rates (30%) at mid- to long-term. In the present study, a total of four (14.2%) complications were primarily related to the minced cartilage procedure, one of which required revision surgery whereas the others resolved without surgical intervention. No revision operation was necessary due to graft hypertrophy or mechanical symptoms related to the chondral graft, although one patient reported still of occasional joint locking at 5 years, possibly related to graft hypertrophy. Altogether, this makes minced cartilage comparable to MFX and ACI in terms of surgery-related complication and revision surgery, and overall a safe and efficient procedure.

Despite the well-known lack of correlation between postoperative radiological outcomes and PROMs, radiological scoring remains relevant as a measure of structural change over time [17]. Two prospective randomized controlled trials reported mean MOCART scores of 76 ± 16 at two years and 75.5 ± 13.1 after 4 years, respectively [45, 46]. When comparing patellar and femoral condyle defects similar MOCART scores were observed [39, 58]. In the present study, only unscheduled follow-up MRIs within two and five years postoperatively were included, that were obtained for re-presentation due to recent trauma, pain, or revision surgery. Naturally, slightly lower MOCART 2.0 scores were therefore observed compared to the above-mentioned studies. Due to the nature of follow-up MRI indications in the present study, it can be assumed that MOCART 2.0 scores would be higher if the whole study cohort was included.

This work has some limitations. First, this study presents mid-term outcomes of the first series of patients undergoing minced cartilage of a single surgeon and therefore lacks a comparative group of patients treated with alternative osteochondral procedures such as ACI, OATS or MTX. Second, as consecutive patients were included and inclusion criteria were not limited to isolated cartilage defects, the heterogeneity of the study population might pose a limitation. However, a recent systematic review demonstrated good clinical outcomes after cartilage repair at the patellofemoral joint even in complex cases [5]. Third, the majority of the cartilage defects were located at the patella or femoral condyle and only a few at the trochlea or tibia. Therefore, the present results may not be applied without restriction to all cartilage defect locations in the knee joint. Moreover, overall outcomes might be lower, as less satisfactory results are for patients affected by cartilage lesions of the patella compared to other sides [23]. Due to the small sample size, statistical subgroup analyses were not possible, but the descriptive statistics appear to be comparable between the different defect localizations. Finally, while all patients obtained a preoperative and a six-month postoperative MRI scan, an additional, mid-term radiological examination for all patients would have been desirable; however, this was beyond the scope of the present work.

Minced cartilage procedure for the treatment of medium to large chondral and osteochondral defects of the knee, show good and promising results at mid-term. Minced cartilage procedure seems to be a viable treatment alternative in patients where a two-staged approach is not desired.

Patients treated with autologous minced cartilage procedure for medium to large chondral and osteochondral lesions of the knee without the necessity for subchondral bone treatment report good mid-term results in pain and knee function and low rates of postoperative adverse events. Pain and functional levels remain stable and within the MCID at mid-term.