Treatment of focal chondral lesions in the knee using a synthetic scaffold plug: Long-term clinical and radiological results

Abstract

The management of symptomatic articular cartilage lesions, especially in the young, fit individual remains an area of considerable controversy. Articular cartilage repair or reconstruction techniques may offer these patients alternatives to arthroplasty. The TruFit™ plug is a synthetic biphasic polymer scaffold that is designed for implantation at the site of a focal chondral defect. The aim of this study is to report the long-term clinical and radiological outcomes of patients treated with the TruFit™ plug for chondral defects within the knee. 11 patients underwent TruFit™ plug implantation. Long-term outcome scores were available for 6 patients at a mean follow up of 121 months (SD 12.0 months, 1 patient unavailable and 4 excluded after arthroplasty surgery). There was no statistically significant improvements in any score although all scores did improve. At a mean radiographic follow up of 70 months (17–113) of 9 patients, the mean MOCART score was 22.2 (SD 15.6). All patients had incomplete or no evidence of plug incorporation and persistent chondral loss. Based on these results, we do not recommend the use of the TruFit™ plug.

1. Introduction

Articular cartilage damage in the knee represents a common clinical problem. Aroen et al. found two-thirds of patients undergoing arthroscopy have chondral lesions. In 11%, these were focal partial to full thickness defects which were suitable for surgical intervention, with over half measuring greater than 2 cm² in size.¹

The management of symptomatic articular cartilage lesions, especially in the young, fit individual remains an area of considerable controversy. Articular cartilage repair or reconstruction techniques once mal-alignment, instability and meniscal lesions have been addressed may offer these patients alternatives to arthroplasty. Established operative options include acute repair of osteochondral lesions of sufficient size, abrasion chondroplasty, marrow stimulation techniques such as microfracture, osteochondral autograft transfer, and autologous or matrix-induced autologous chondrocyte implantation (ACI/MACI), and osteochondral allograft transplantation.^2,3 A systematic review of available techniques have failed to show a treatment modality with consistently superior results.⁴

The TruFit™ plug (Smith & Nephew, London, UK) is a synthetic biphasic polymer scaffold that is designed for implantation at the site of a focal chondral defect. It is marketed to resorb and allow tissue ingrowth 6–9 months following implantation, providing structural support during this period by matching the biomechanical properties of the adjacent bone and cartilage.⁵ The implant comes in a variety of sizes and may be placed either arthroscopically or via an open approach depending on the site of the lesion. It is used for full thickness chondral lesions of less than 2 cm.² The surgical technique has previously been described.⁶

The aim of this study is to report the long-term clinical and radiological outcomes of patients treated with the TruFit™ plug for chondral defects within the knee.

2. Methods

2.1. Patient selection

The TruFit™ plug was used selectively in patients presenting with symptomatic, full-thickness, chondral lesions in the knee between June 2007 and December 2009. All TruFit™ plugs were implanted by the senior author. Following appropriate counselling, patient selection fell into two broad categories. Firstly, those classed as ‘ideal’ patients with focal traumatic defects, but otherwise good chondral surfaces and menisci; and secondly in those where the implant was used as a salvage procedure following failed chondroplasty or microfracture.

TruFit™ plugs were implanted either by an arthroscopic or mini-open approach depending on the site of the chondral lesion, according to the operative technique described by Carey-Smith et al.⁷ Post-operative management entailed a period of non-weight bearing for six weeks, or restricted knee flexion in a brace for six weeks, or both, depending on the site of the lesion(s). Return to non-impact activity was permitted by three months and impact activity by six months.

Local institutional approval was given for the assessment and analysis of patient data to evaluate the service provided. This study adheres to the 1964 Helsinki declaration and its later amendments.

2.2. Outcome measures

All patients were followed up with clinical examination at six weeks, three months, six months, one year and annually thereafter. Primary outcome was defined as conversion to arthroplasty. Functional scores in the form of the Oxford knee score (OKS), Tegner activity scale and Lysholm scores as well as radiological analysis were secondary outcomes. Radiological evaluation was conducted in all patients using plain radiographs and MRI (if not contra-indicated), and with selective use of CT arthrogram. The post-operative time interval for obtaining MRI and CT arthrogram were not standardised. This was to allow for a broader understanding of the radiological characteristics of the implant to be evaluated over time, without subjecting patients to repeated investigations, both on the basis of cost and avoidance of ionising radiation. The latest MRI scans were reviewed separately by the authors (FS and AM) and scored using the Magnetic Resonance of Cartilage Repair Tissue (MOCART) score.⁸ In the presence of differing scores, the 2 authors re-reviewed the images together to agree the final score.

2.3. Statistical analysis

Descriptive statistics were reported as the mean with standard deviations. Analyses of scores was conducted using a Wilcoxon Signed Rank Test. All statistical analyses were performed using SPSS Version 13.0 (IBM Corporation, USA). A p-value of less than 0.05 was seen as significant.

3. Results

3.1. Demographics

11 patients (7 women, 4 men) underwent TruFit™ plug implantation. A total of 35 TruFit™ plugs were implanted. Mean clinical follow up was 93 months (range 24–144 months). In 5 patients TruFit™ implantation was as a primary procedure. In 6 patients it was used as a salvage option after a failed procedure (2 microfracture, 4 chondroplasty). One patient underwent surgery at two different time points. Mean age at the time of surgery was 46.1 years (SD 13.7 years).

The number and site of plugs used is shown in Table 1. One patient underwent simultaneous medial opening wedge high tibial osteotomy and another patient arthroscopic anterior cruciate ligament (ACL) reconstruction. 4 patients underwent subsequent conversion to arthroplasty surgery due to continuing symptoms.

Table 1.

Site of implantation of TruFit™ plugs used in patients, and additional procedures undertaken. MFC = medial femoral condyle, LFC = lateral femoral condyle.

Patient Number	Number and site of TruFit™ plugs used	Additional procedures	Length of follow up (months)	Conversion to arthroplasty (time)
1	3 MFC then 1 MFC (9 months later)	ACL reconstruction at first procedure	120	No
2	2 MFC, 2 LFC, 3 Trochlea		114	No
3	1 MFC		114	No
4	2 MFC, 3 Trochlea		111	No
5	3 MFC		60	Yes (58 months)
6	2 LFC, 5 Trochlea		24	Yes (19 months)
7	1 MFC	High tibial osteotomy	65	Yes (63 months)
8	3 MFC		112	No
9	2 Trochlea		144	No
10	1 MFC		124	No
11	1 Patella		38	Yes (36 months)

3.2. Outcome scores

Pre-operative and 1-year post-operative outcome scores were available for 10 patients. Long-term outcome scores were available for 6 patients at a mean follow up of 121 months (SD 12.0 months, 1 patient unavailable and 4 excluded after arthroplasty surgery). At both time points, there were no statistically significant improvements in any score although all scores did improve (Table 2).

Table 2.

Mean patient related outcome scores.

Score	Mean pre-op score (n = 10)	1-year post-op (n = 10)	Latest follow up (mean 121 months, n = 6a)
OKS	41.2	33.6	37.2
Lysholm	38.1	63	73.2
Tegner	1.8	2.5	2.6

^aPatients who had undergone arthroplasty surgery excluded (n = 4).

3.3. Radiology

All patients underwent plain radiographs preoperatively and at regular intervals post-operatively (see Table 3). 9 patients had post-operative MRI scans and the most recent were reviewed at a mean follow up of 70 months (17–113). 2 patients had CT scans. The mean MOCART score was 22.2 (0–45). All patients had incomplete or no evidence of plug incorporation and persistent chondral loss and residual oedema or cystic changes in the latter years. Typical images are displayed in Fig. 1, Fig. 2.

Table 3.

Long-term radiographic outcomes.

Patient Number	Imaging	Timing of imaging (months post-op)	MOCART Score	CT features
1	MRI	110	30
2	MRI	63	45
3	MRI	72	0
4	MRI	17	10
5	CT	21		3/3 plugs not incorporated
6	CT Arthrogram	11		2/2 LFC plugs incorporated with overlying cartilage 4/5 Trochlea plugs not incorporated.
7	MRI	49	20
8	MRI	91	45
9	MRI	91	10
10	MRI	113	25
11	MRI	24	15

Fig. 1.

Sagittal T2-weighted fat-supressed MRI image of the medial femoral condyle at 3 months (a), 12 months (b), and 8 years (c) post operatively following a single TruFit™ plug being implanted.

Fig. 2.

Axial T2-weighted fat-supressed MRI scan at 1-year (a) and 6.5 years (b) post-operatively showing no incorporation of the plug or evidence of cartilage regeneration.

4. Discussion

This paper reports the longest-term results published of a case series of patients undergoing TruFit™ plug implantation for full-thickness articular cartilage lesions in the knee. Our results show no significant improvement in patient related outcome scores and a high conversion rate to arthroplasty surgery. In addition, the plugs appear to take many years to incorporate and, in most cases, do not encourage a meaningful chondral surface to form. In some cases, persistent cavitations or bone oedema persist. The TruFit™ plug is a poly lactic-co-glycolic acid (PLGA) based structure that upon degradation releases acidic oligomers that have detrimental effects on surrounding bone. In vivo animal studies have confirmed increased subchondral bone cysts with the use of the TruFit™ plug when compared with alternatives, this is in keeping with our findings on imaging.⁹

Concerns have been raised over the post-operative MRI appearances of TruFit™ plugs. These have been further characterised by Bedi et al. using quantitative T2 mapping, who report 75% of plugs show favourable characteristics at early (<6 months) follow up. Appearances then deteriorate by 12 months, but subsequently improve at longer term (>16 months) follow up with 70% showing flush morphology and 90% showing complete or near-complete fill in.⁹ More recently, Bugelli et al. reviewed 5 patients with 8 plugs at a mean follow up of 71 months. 7 plugs incorporated with a mean MOCART of 67.85 and 1 plug did not.¹⁰

These results have not been replicated in other studies. Azam et al. reported partial incorporation in 7 of 12 plugs, 3 with complete incorporation, and 2 with no incorporation at 2 years.¹¹ D'Ambrosi et al. used the MOCART score to evaluate the integration of the plugs and had a mean score of 45.78 at mean 102 months post-op in 21 patients.¹² Gelber et al. reported a mean MOCART score of 43.2 in 57 patients and a lack of osseous incorporation of the plugs.¹³ Joshi et al. reported all 10 of their patients had a degree of incorporation failure and articular cartilage collapse or fissuring.¹⁴ Quarch et al. used TruFit™ plugs to fill large donor site defects (mean 5.5 cm²) and reported the filling of these defects took at least 2 years and was longer than leaving the defects alone.¹⁵ In a series of 9 patients investigated with plain CT at variable time intervals post operatively, the TruFit™ plug was shown to degrade over a period of 9 months, at which point it returned a density consistent with fibrous scar tissue formation with no evidence of osseous incorporation which persisted even at 63-months follow up.¹⁶

Our MRI findings reflect those seen in other published work as stated above. Due to our long follow up, we have also showed that the lack of restoration of suitable cartilage leads to continuing symptoms both in the short and long-term. Although outcome scores marginally improved post-operatively, these results were not significant and 4 of our patients subsequently underwent arthroplasty surgery. Joshi et al. also showed no significant improvement in scores at 2 years post op (KOOS, SF-36, VAS).¹⁴ Hindle et al. compared TruFit™ (35 patients) to mosaicplasty (31 patients) and found that Trufit patients had worse EQ-5D pain scores and a reduced rate of returning to sport.¹⁷

Although Gelber et al. showed poor MRI characteristics, they found an improvement in KOOS, SF-36, and VAS at 44.8 months.¹³ However, 59.6% of these patients had concomitant surgery including ligament reconstructions and meniscal surgery and so there is little evidence to suggest these improvements were solely due to TruFit™ implantation. D'Ambrosi et al. again reported an improvement in KOOS and HSS score, although 6 out of 21 patients had concomitant ACL surgery.¹² Azam et al. reported an improvement in IKDC at 2 years from 41.8 to 57.4, although the authors stated the clinical improvement was small.¹¹ Bugelli et al. reported an increase in Lysholm scores from 50.4 to 95.4 in 5 patients.¹⁰ There is no detail regarding concomitant procedures. It appears the literature is mixed with regard to the clinical effect of TruFit™ implantation. However, a systematic review (published prior to many of the previously mentioned studies) suggested there was inadequate evidence to suggest its effectiveness against conservative or other operative techniques.¹⁸

This study has its limitations. It is a small case series of heterogenous patients and therefore one would expect diverse clinical outcomes. There was no control group or blinding of the clinical or radiological scorers. The post-operative imaging was also not standardised and orthopaedic surgeons reviewed the images rather than radiologists, which may have compromised the scoring. However, the MRI's were taken at times where graft incorporation would be expected to have been completed and so the findings cannot be attributed due to inadequate follow up. The results were collected prospectively and the significant length of follow up of this novel implant appears to be the longest recorded in the literature. This long follow up sheds light on the debate of the potential incorporation of the plugs.

There are ongoing advances in magnetic resonance imaging which may be useful in improving the non-invasive assessment of chondral lesions and post-operative healing of such implants.¹⁹ Specifically, delayed gadolinium enhanced MRI for cartilage (dGEMRIC), can be considered the method of choice for quantitative evaluation and visualisation of proteoglycan content in articular cartilage.²⁰ Following intravenous administration of gadolinium, MRI can be used to quantitatively assess gylcosaminoglycan (GAG) concentrations on T1-weighted images. GAG concentrations are a feature of cartilage maturation, and the dGEMRIC index has been shown in vitro to have good correlation with the biomechanical properties of repair cartilage.²¹ However, in vivo studies in patients treated with ACI have not shown correlation to be as high, suggesting the addition of T2-mapping, which assesses the collagen content in the repair is a useful adjunct.¹⁹ These modalities require a three-teslar MRI scanner, and therefore, remain tools principally for the research setting.

In conclusion, this case series presents the longest clinical and radiographic follow up of patients undergoing TruFit™ plug implantation. There was no evidence of clinical improvement and a high conversion rate to arthroplasty. MRI and CT imaging showed a failure of graft incorporation in the majority of cases with limited evidence of chondral surface regeneration. In some cases, large osseous defects persisted long-term. Based on these results, we do not recommend the use of the TruFit™ plug.

CRediT authorship contribution statement

Faiz S. Shivji: Writing - original draft. Aadil Mumith: Writing - review & editing. Sam Yasen: Investigation, Writing - review & editing. Joel TK. Melton: Writing - review & editing. Adrian J. Wilson: Conceptualization, Supervision.

Declaration of competing interest

None declared.