Introduction

In recent years, there has been a steady rise in the volume of revision total knee arthroplasty (TKA) procedures [1] and the number of primary TKA procedures performed in the next two decades is expected to rise exponentially [2]. It is therefore critical to prolong focal defect management to preserve healthy articular surfaces, soft tissues and bone stock, and to delay the onset of conventional arthroplasty.

Full-thickness articular cartilage defects are commonly found in symptomatic knees [3, 4], the majority of which are observed in patients over the age of 40 years [5, 6]. Left untreated, these defects may result in progressive articular degeneration and predispose to the development of osteoarthritis [7–9]. Single or repeat repair procedures such as microfracture, OATS, allograft or cell-based technologies can provide relief from focal defects [10–15], but appear to be increasingly ineffective with increasing age [13, 16–18]. However, if these defects remain symptomatic, this patient population reflects a serious problem for the orthopaedic surgeon, as conventional joint arthroplasty does not provide joint preservation in middle-aged patients with isolated defects.

Focal defect management has been extended with the development of a reconstructive treatment option using a metallic inlay resurfacing prosthesis (HemiCAP®, Arthrosurface Inc., Franklin, MA, USA). The procedure allows the surgeon to precisely match the implant to the contour of the patient’s articular cartilage, thus filling the defect and restoring a smooth and continuous joint surface (Fig. 1). Healthy cartilage, bone and soft tissue structures are therefore preserved for conventional joint arthroplasty if it becomes necessary at a later stage.

Fig. 1
figure 1

HemiCAP® Implant for the femoral condyle composed of a cannulated tapered titanium alloy anchoring screw (fixation component) and attached Co–Cr alloy resurfacing implant (articular component)

Full size image

The purpose of this multicenter study was to evaluate the durability of the focal resurfacing procedure in middle-aged patients with a minimum of 5 years follow-up. Outcomes were evaluated with validated outcome questionnaires, standard radiographs and comparison to normative, age-matched reference data.

Materials and methods

Study design

This is a level 4, multicenter (5 centers), case series. Patients, who had received the resurfacing device with a minimum of 5 years follow-up, were screened for eligibility according to the inclusion and exclusion criteria (Table 1) using pre- and intraoperative patient records. To achieve a homogenous middle-aged cohort, patients 65 years and older were excluded. Patients with kissing lesions and advanced grade cartilage degeneration in other compartments were excluded as well to support conclusions toward the potential benefit of the procedure. Institutional review board approval was obtained prior to screening. All patients gave informed consent before inclusion into the study.

Table 1 Inclusion and exclusion criteria for the study
Full size table

Resurfacing prosthetic

The HemiCAP® contoured articular prosthetic (Fig. 1) for focal femoral condyle resurfacing consists of two components, a fixation and an articular component. The two are joined by a morse taper connection to provide a stable and immobile fixation of the implant and stress bearing contact at the bone/prosthetic interface. The fixation component is a cannulated cancellous screw with a tapering distal tip (titanium alloy). The articular dome-shaped component comes in 15- and 20-mm diameters, which are available in various off-set sizes corresponding to the radius of curvatures at the implant site (cobalt–chromium–molybdenum alloy with titanium plasma spray coverage on the underside).

Rehabilitation

Standard rehabilitation protocol included weight bearing as tolerated using crutches for 1–2 weeks with early functional rehabilitation and no restriction to range of motion.

Outcome measures

Demographic data and validated questionnaires were gathered at the follow-up visit. The Knee Injury and Osteoarthritis Outcome Score (KOOS) score [19] includes five domains: pain, symptoms, activities of daily living (ADL), sports and recreation ability, and knee-related quality of life (QoL) and has been used and validated for the treatment of focal cartilage defects [20, 21] (100 = best possible score, 0 = worst). Overall physical health assessment was obtained via the SF-36 Health Status Survey [22]. The SF-36 is a self-administered, patient-based scoring system assessing patient global health with 36 questions leading to a physical and mental health component score. The Tegner activity scale [23] is a numerical scale measuring the level of competitiveness of an individual sport’s participation. Each activity is measured from 0 to 10 where an elite professional athlete would compete in a sport at a level of 10, high-level recreational athletes would be at 6 and persons with knee problems or a disability would score at a level of 0. These scores were compared with their preoperative scores gathered at the latest follow-up to evaluate the effects of device implantation. Satisfaction was evaluated by a questionnaire that rated the general satisfaction with the treatment outcomes (1 = excellent, 2 = very good, 3 = good, 4 = fair, 5 = poor), if they would undergo the procedure again and if they would recommend this procedure to friends and family. Furthermore, results were compared to normative patient population data for KOOS, SF-36 and Tegner.

Radiographic assessment was evaluated comparing the preoperative radiographs to those obtained at the 5-year follow-up visit (AP and lateral standing X-rays). The radiographs were graded according to the Kellgren–Lawrence classification [24] and Osteoarthritis Research Society International (OARSI) Grading System [25, 26] to compare the stage of osteoarthritis with the preoperative status. Prosthetic connectivity, subsidence, migration, radiolucency, implant loosening and subchondral cyst formation were evaluated.

Secondary procedures following the index surgery were assessed according to the patient records and confirmed at the follow-up visit. Treatment failure was defined as removal of the prosthetic device and replacement by arthroplasty (uni-compartmental or total). All evaluations were carried out by the first and second author. Discrepancies in interpretation were settled by consensus.

Statistical analysis

Statistical analysis was performed using the software package SPSS 18.0 (SPSS Inc., Chicago, IL, USA). Mean and standard deviation values were recorded. To determine differences between the data of preoperative and follow-up scores, the Wilcoxon nonparametric test (significance, P ≤ 0.05) was used.

Results

Patient characteristics

Forty patients were implanted with the resurfacing device; 21 patients were included in this study (Table 1; Fig. 2). The average age at implantation was 54 years (range 38–63). There were 11 males and 10 females. The average follow-up was 5.3 years (64 months; range 60–71 months). Six patients were not available for radiographic examination at follow-up; however, they were contacted by telephone and returned their questionnaires by mail. One patient underwent conversion to unicondylar knee replacement at 2 years postoperatively leaving 20 patients available for a 5-year analysis. Patient demographics are summarized in Table 2. Defects in four patients were resurfaced with a 15-mm surface implant and in 17 patients with a 20-mm implant. Surgical history and joint status are summarized in Table 3.

Fig. 2
figure 2

Patient selection diagram

Full size image
Table 2 Patient demographics
Full size table
Table 3 Summary of the surgical history and joint status
Full size table

Score outcome measures

The results are summarized in Table 4. The average 5-year postoperative KOOS domains scores were 77.6 (pain), 79.5 (symptoms), 82.4 (ADL), 57.8 (sports/recreation), and 55.0 (QoL). The physical component score of the SF-36 was 46.9, the mental component score 51.2. The average Tegner Activity Scale was 4.0, meaning patients could engage in moderately heavy labor and recreational sports such as cycling. General satisfaction with the treatment outcomes was rated as 2.9 (SD 1.4, range 1–5). The result was rated excellent in 4/21 of the patients, very good in 6/21, good in 3/20, fair in 5/21 and in 3/21 as poor. Of the 21 patients, 16 (76%) patients indicated they would undergo the procedure again and would recommend it to friends and family.

Table 4 Results of different outcome scores
Full size table

Radiographic assessment

Fourteen patients were evaluated radiographically at a minimum of 5 years follow-up. No device disassembly, subsidence/migration, peri-prosthetic radiolucency, implant loosening or subchondral cyst formation was found. The average Kellgren–Lawrence score slightly increased from 2.2 to 2.5 over the course of the follow-up period. No clinically significant joint space narrowing was noted according to the OARSI Grading System. The average OARSI grade for osteophyte formation (tibial plateau and femoral condyles) remained unchanged when compared with baseline (Table 5).

Table 5 Preoperative and postoperative results of the radiographic evaluation
Full size table

Subsequent surgical procedures

One patient underwent arthroscopic debridement 4.9 years following the index procedure due to pain. Another patient underwent three further operations of the affected knee including arthroscopic debridement at 5 months due to persistent pain in the medial compartment, high tibial osteotomy for the correction of varus malalignment at 2 years and hardware removal/further debridement at 3 years. Satisfaction outcome for these two patients was rated fair and poor. One female patient underwent conversion from focal resurfacing to uni-compartmental knee arthroplasty at 2 years after the index procedure due to ongoing pain. This patient was regarded as a failure and was not included in the 5-year score analysis, but included in the general satisfaction analysis and rated as poor.

Comparison to normative data

When comparing the results of our patients to normative data for KOOS domains [27] and age 55–74 years, slightly better scores were found for our female patients for all domains except the Sports and Recreation domain, which was similar, and the QOL domain which was slightly worse (Fig. 3). When comparing the scores to those of normal scores for males, our patients were inferior across all domains (Fig. 4). Comparison to SF-36 normal scores [28] showed slightly inferior scores for the physical component (5 yr. = 46.9; normal = 52.1) and similar scores for the mental component (5 yr. = 51.2; normal = 52). When compared with a population of community volunteers 46–60 years of age [29], who considered their knee function normal and without any history of knee surgery or recent injury, the average Tegner activity level of our patients were again slightly inferior to the normal score published for this population (5 yr. = 4.0; normal = 4.6).

Fig. 3
figure 3

KOOS scores compared to normal female scores [27]. Population-based reference data for KOOS (females): 35–54 years; n = 80. 55–74 years; n = 85. Current study mean age at follow-up (female): 59 years (range 43–68); n = 9

Full size image
Fig. 4
figure 4

KOOS scores compared to normal male scores [27]. Population-based reference data for KOOS (males): 35–54 years; n = 78. 55–74 years; n = 88. Current study mean age at follow-up (male): 58 years (range 46–68), n = 11

Full size image

Discussion

Contoured metallic resurfacing is a novel reconstructive treatment option intended to extend the range of focal procedures in middle-aged patients who frequently find themselves between biological solutions and conventional arthroplasty. Unicondylar knee replacement has been shown to be a successful treatment option when larger surface reconstruction is warranted [30]. However, in middle-aged patients with isolated condylar defects, significant loss of healthy soft tissues and bone stock should be avoided.

The use of a metal implant, however, is of concern resulting in damage to the opposing articulating structures due to rigid fixation of the resurfacing prosthetic in the underlying bone. Basic science studies have previously investigated the validity of the resurfacing device: Kirker-Head et al. [31] assessed the functional and biological response in a goat model. One year after implantation, gross necropsy and histological data implied the biocompatibility and functionality of the implant, since no ongoing degenerative joint disease was apparent. Some chondral damage in the operated joint was found in several goats in opposing locations on the proximal tibia, including some focal meniscal injury. Similar results, however less frequently, were encountered in the un-operated controls [31]. This outcome was substantially better in comparison to other reported experimental animal studies in goats with untreated full-thickness defects [8, 32]. In vitro studies using fresh frozen human cadaver specimens to evaluate the effects of the prosthetic device on tibiofemoral contact pressure suggested biomechanical safety with no significant increase of contact pressures compared to untreated healthy knees, even under continuous dynamic loading [33, 34].

This clinical series represents the first report on patients with a minimum of 5 years of follow-up after focal prosthetic resurfacing of localized full-thickness cartilage and osteochondral condylar defects. All patients in our series had previously failed conservative treatment and the vast majority underwent single or repeat biological procedures prior to focal prosthetic resurfacing. The study looked specifically at middle-aged patients without kissing lesions and advanced grade cartilage degeneration in other compartments that would benefit the most from a limited resurfacing treatment. However, 42.5% of the implanted patients were operated outside the target population resulting in their exclusion from the study. Further weaknesses of the study were the small number of patients, the lack of prospective baseline data and that only 2/3 of the included patients were available for a radiographic follow-up.

Studies have shown that biological treatment measures are less suitable for middle-aged or active elderly patients resulting in inferior outcomes in older patients [13, 16–18]. Reports on microfracture found a strong and significant correlation between pain relief, functional results and age: better outcomes for patients under 35 years [16], less improvement for patients between 35 and 45 years compared to younger patients, [13] and significant increase in pain and decrease in function between 18 and 36 months after microfracture for patients over 40 years, while remaining relatively constant for those 40 and under at the same time points [17]. In a randomized trial comparing autologous chondrocyte implantation (ACI) to microfracture, Knutsen et al. [18] found that younger patients (less than 30 years old) had a significantly better clinical outcome at 5 years follow-up than older patients, regardless of their treatment group.

Recently, ACI as a primary procedure was reported not to result in inferior outcomes for patients aged older than 40 years [35] or 45 years [36] when compared to younger patients. ACI as a secondary procedure, in particular following microfracture, has shown a lower success rate [37]. High costs, high reoperation rates, complex rehabilitation and prolonged delay before return to sports [36, 38, 39] represent further limitations to the use of this technique.

Patients in our study rated their outcomes as: 62% excellent, very good or good; 24% fair; and 14% poor. As many as 76% would choose to undergo the procedure again. Slightly better rates were found in a study reporting results after ACI in patients over the age of 45 years. Patients rated their outcome at a mean follow-up of 4.7 years as: 72% good or excellent, 15% fair and 13% poor with 81% choosing to undergo the procedure again [36]. However, the failure rate requiring open repeat surgery was 14%, and even 43% of patients required second-look arthroscopy within the first year after ACI [36]. In a prospective study of ACI in patients with failed prior treatment (mean: 1.9 procedures), the subsequent surgical procedure rate observed was 49% overall and 40% related to ACI [39]. Results of the KOOS subscales at an average of 3.8 years follow-up were in all categories inferior compared to our results. Self-rating of the overall knee condition was 77% good or excellent [39]. In our study, one patient (1/21) was converted to a uni-compartmental knee arthroplasty due to ongoing pain and functional limitations. This failure rate is relatively low when compared with other treatments for focal defects [10, 18, 36, 39, 40].

In a study evaluating population-based reference data for the KOOS score, pain, physical function and knee-related quality of life varied with age and gender with a trend to lower normative levels with increasing age [27]. Although the average age of our patients at surgery was 54 years (range 38–63) and a higher number of women were treated compared to studies on ACI [36, 39], we showed superior results in the majority of KOOS subscales when compared to these reports.

Radiographic signs of osteoarthritis were common in our study. Knutsen et al. [18] reported that 34% of the patients had radiographic evidence of early osteoarthritis at 5 years after ACI or microfracture treatment. Our findings revealed a slight increase of osteoarthritis, but no progression in the grade of the Kellgren–Lawrence classification and OARSI Grading System. In an analytic literature synthesis on Kellgren–Lawrence progression in knee osteoarthritis, the annual progression by at least one grade ranged from 1.0 to 19.6% with an overall mean risk of progression of 5.6% at a mean follow-up of 5.3 years [41]. Thus, the osteoarthritic changes among our patients appear not to exceed the overall mean risk of progression.

In summary, by using validated outcome measures and standard radiographs, we have demonstrated clinically meaningful improvements in the majority of patients and radiological safety after focal prosthetic resurfacing for the treatment of full-thickness cartilage and osteochondral defects at the medial femoral condyle at a minimum follow-up of 5 years. Nevertheless, the results need to be confirmed in larger patient cohorts.