HemiCAP Knee Implants: Mid- to Long-Term Results

Abstract

Purpose

The aim of the study was to investigate the long-term outcomes of the Focal Femoral Condyle Resurfacing Prosthesis (HemiCAP) using clinical and radiographic assessments, and to evaluate the revision and survival rates.

Methods

Clinical evaluation was performed in those not revised and was able to participate. This was a prospective single-center cohort study of HemiCAP patients with 7 to 10 years of clinical and radiographic follow-up. The clinical examination included the Knee Society Score (KSS) and visual analogue scale (VAS) score. The radiographic examination included the Kellgren-Lawrence (KL) grade. Survival was estimated by Kaplan-Meier survival analysis, and potential risk factors for revision was evaluated by a regression analysis.

Results

Of the 62 patients with 64 HemiCAP prostheses, 37 were HemiCAP condyle, 11 HemiCAP PF, and 16 HemiCAP Wave; 27 (42%) were revised—HemiCAP condyle 17 (42%), HemiCAP PF 4 (36%), HemiCAP Wave 6 (37%), and 1 died. Examinations were performed on 31 patients (86%). When compared with the preoperative data, there were significant increases in the KSS objective (mean = 51.5, standard deviation [SD] = 5.9 vs. mean =94.2, SD = 5.0) and function (mean = 51.0, SD = 6.2 vs. mean = 93.7, SD = 4.8) scores, a decrease in the VAS score (mean = 7.1, SD = 0.7 vs. mean = 2.7, SD = 1.7) and a decrease in the KL lateral score (mean = 1.1, SD = 0.3 vs. mean = 0.6, SD = 0.6). The mean follow-up was 7.3 years (SD 1.4) with minimum 4.2 years and maximum 10.2 years. No failures occurred in the series beyond 5 years.

Conclusions

As hypothesized, we found good clinical and radiographic outcomes, and for those patients who did not require revisions, there were long-term improvements in disability and function. This suggests that patient selection is a key element to successfully applying these devices in clinical practice.

Introduction

Middle-aged patients with knee pain and disability caused by cartilage lesions or early osteoarthritis (OA) are challenging to treat, largely due to the progressive decrease in the healing capacity of the cartilage with advancing age.^1-5 For this patient population, several different treatment modalities are available,^6,7,28,29 but none of them are entirely favorable. Many necessitate a long rehabilitation period after the intervention, with a concomitant risk of leaving the labor force, resulting in enormous societal and personal costs.^8,9

Previous studies have demonstrated that biological treatments, such as marrow stimulation and chondrocyte transplantation, are preferred in younger patients (30-35 years old) and in patients with smaller lesions (less than 10 × 10 mm). However, these treatment modalities have unfavorable outcomes with an increasing patient age.^3-6,10,30 Until recently, such patients were left with an osteotomy, unicompartmental knee arthroplasty (UKA) and total knee arthroplasty (TKA) as their remaining surgical options. Although some patients might benefit from the operations, especially the UKA and TKA, they are rarely considered to be appropriate for middle-aged patients due to the insufficient treatment response in cases in which the OA is not fully developed clinically and radiographically with severe knee pain, daily disability, and progressing arthritic signs, as shown by the Kellgren-Lawrence (KL) grade. ¹¹ Thus, these patients may pursue nonoperative treatment modalities, such as physiotherapy, weight loss, analgesics, and activity modifications.^12,13

In order to fill this existing treatment gap, small implants were developed that leave the unaffected joint compartments alone. It is in this context that the mini-prostheses were developed, with a goal to manage only the isolated symptomatic cartilage lesions. The first were onlay mini-prostheses,^14-16 followed by the later development of inlay mini-prostheses, which addressed not only the size of the cartilage lesion but also the patient-specific curvatures of the knee.^{2,14,16-21,28}

In 2003, an anatomic metallic implant for femoral resurfacing called the Focal Femoral Condyle Resurfacing Prosthesis (HemiCAP) was introduced for full-thickness smaller condylar lesions (both femoral condyle and HemiCAP PF [trochlear], and later HemiCAP Wave for larger trocheal lesions or isolated OA in the patellofemoral joint. It was approved for use in Denmark in 2006 and 2009, respectively, with publications from 2015 and 2016 describing its specific indications.^21,22 Only limited evidence of the clinical outcomes and failure rates has been presented for the HemiCAP. Two case series of approximately 20 patients^15,18 with varying osteochondral pathologies demonstrated reduced pain and improved knee function. Additionally, 2 studies^21,22 from 2015 and 2016 demonstrated the same, but with a concerning 23% revision rate based on the medium-term follow-up results (6 years) for the HemiCAP condyle and HemiCAP PF and later for the subgroup of HemiCAPs—18 HemiCAP Waves ²² with 6 years revision rate at 28%. A systematic review by Malahias et al. ¹² from 2018 concludes the lack of mid- to long-term clinical outcomes in well-designed clinical studies. We present the first long-term clinical cohort-study of 64 HemiCAPs followed up to 10 years, and we hypothesized that the implants would reduce pain and improve knee function in long-term treatment offering a better treatment-option than biological treatment or even UKA and TKA.

The aim of this study was therefore to investigate the long-term treatment of Focal Femoral Condyle Resurfacing Prosthesis in middle-aged patients, with cartilage pathology in femoral condyles and trochlea to be treated with HemiCAP condyle, HemiCAP PF, or HemiCAP Wave. The objectives were to (1) evaluate the clinical and radiographic follow-up of the prosthesis, (2) investigate the revision and survival rates, and (3) evaluate potential risk-factors for revision.

Material and Methods

Study Design and Setting

This was a prospective single centre cohort study with follow-up of patients treated with femoral resurfacing between 2007 and 2013.^21,22 The follow-up period took place from October through December of 2016 at the Hospital of Southern Jutland. It was run through the Emergency Medicine research unit at the University of Southern Denmark, Institute of Regional Health Research. It was reported according to principles outlined in the Strengthening the Reporting of Observational Studies in Epidemiology statement. ²³

Participants

From 2007 to 2013, 62 patients (64 CAPs) aged 28 to 65 years were treated with femoral resurfacing using HemiCAP implants at the Hospital of Southern Jutland Orthopaedic Department. Two of the patients had both knees treated. In total, 27 trochlear (11 HemiCAP PF, 16 HemiCAP Wave), and 37 femoral condyle implants were inserted. The treatment indications for this implant included symptomatic cartilage lesions at the femoral condyle or trochlea as demonstrated using magnetic resonance imaging or arthroscopy, with an International Cartilage Repair Society (ICRS) grade of 3 to 4 and a lesion size of less than 400 mm² for the HemiCAP condyle and HemiCAP PF. For the HemiCAP Wave it was lesions larger than 400 mm² or isolated OA in the patellofemoral joint space. There were 28 males and 34 females, with a median age of 49 years (range 28-65 years).

Patient were not offered treatment, if they had a valgus or varus malalignment exceeding 5°, ligament instability, more than 50% meniscus removal or a body mass index of more than 40 kg/m². The 62 HemiCAP patients included in this study were followed for 7 to 10 years.

Procedure

The revised patients and the revision times were found using the protocol data from our local database and the Danish Knee Registry. ²⁴ Those patients who were not revised were invited to participate in this study. They were contacted electronically by the responsible surgeon. If written consent was obtained, they were clinically examined and radiographically evaluated by assessing the KL grade for the medial, lateral, and tibiofemoral compartments, ¹¹ the Knee Society Score (KSS) objective and function subscales, ²⁵ and the visual analogue scale (VAS) pain score using a numerical rank scale (0-10), with 10 being the worst possible pain.

Device Description

The HemiCAP resurfacing implant consists of 2 components, a fixation component and a modular articular component, connected by a Morse taper. It is available in sizes of 15 or 20 mm in diameter, and it comes in 16 different offset configurations that correspond to the superior/inferior and medial/lateral radii of the curvatures at the implant site. The HemiCAP PF (trochlear implant) is 15 or 20 mm in diameter, with a single concave shape fitting the trochlear sulcus with or without a patellar button, and a 35 × 35 mm HemiCAP Wave ( Fig. 1 ). A polyethylene inlay is available for the patella, and it is recommended to be used in combination with the HemiCAP Wave implant.

Figure 1.

Hemicap standard, Hemicap PF, and Hemicap PF-xl Wave (Artrosurface Inc., Franklin, MA): components—cobalt-chromium alloy (Co-Cr-Mo); undersurface coating—titanium (CP Ti); fixation stud—titanium alloy (Ti-6Al-4V).

Statistical Analysis

The demographics and baseline (preoperative) characteristics of the patients were presented as median and interquartile range (IQR) values. A Wilcoxon signed-rank test was used for the paired data comparisons. The Kaplan-Meier survival analysis was used with revision or death as the endpoints and a 95% confidence interval (CI). Cofactors for revision were analyzed with uni- and multivariate regression. For the statistical analysis, Stata: Data Analysis and Statistical Software for Professionals version 15.1 (StataCorp LLC, College Station, TX, USA) was used. P values of less than 0.05 were considered to be statistically significant.

All the data collected were stored in accordance with the Danish Data Protection Agency requirements. This study was approved by the regional data committee of the Region of South Jutland (#2008-58-0035).

Ethical Consideration

Written consent to participate in the study was obtained from all patients when enrolled in the initial protocol prior to operation. According to Danish law approval by ethical committee was not necessary for follow-up studies.

Results

Of the 64 HemiCAP procedures, 28 (44%) were excluded from the follow-up due to revisions or death ( Fig. 2 ). The revision causes included increasing pain, disability, and OA progression. In addition, 1 patient exhibited aseptic loosening and 1 had a deep infection. Five were unable to or refused to participate in the follow-up examinations, none of them revised, resulting in the examination of 31 HemiCAPs. Of those, 13 (42%) were males, and the preoperative median age was 49 years (IQR = 41-60 years). The mean follow-up time was 7.3 years (SD 1.4 years) and minimum time was 4.2 years and maximum time 10.2 years.

Figure 2.

Flowchart of up to 10 years follow-up on HemiCAPs split in different types.

The objective and subjective outcomes (KSS) and radiographic and OA evaluations (KL-OA) are shown in Table 1 and Figs. 3 , 4 , and 5 ). Both the KSS objective and function scores improved significantly from the preoperative scores to the follow-ups. However, the KL grade did not change significantly from the preoperative medial score, but it did change significantly from the preoperative lateral score ( Table 1 ). The VAS pain scores were reduced significantly ( Figs. 6 , 7 , and 8 ).

Table 1.

Up to 10 Years’ Follow-up on 31 HemiCAPs with High Significant Improvements in Pain and Function Scores.

	Preoperative			10 Years’ Follow-up			Comparison
	n	Median	IQR	n	Median	IQR	P a
BMI, kg/m2	31	28	25-30	31	27	23-31	0.88
KSS
Objective	30	50	45-55	31	95	90-100	<0.000
Function	30	50	45-55	31	95	90-95	<0.000
Pain score	30	7	7-8	29	2	2-4	<0.000
KL score
Medial	31	1	1-1	30	1	1-1	0.19
Lateral	30	1	1-1	31	1	0-1	0.0008

IQR = interquartile range; KSS = Knee Society Score; BMI = body mass index; KL = Kellgren-Lawrence.

^aWilcoxon signed-rank test.

Figure 3.

Knee Society Score (KSS) objective and function scores in 17 patients (HemiCAP femur-condyles): preoperative, at 1 year, at 2 years, and at follow-up till 10 years.

Figure 4.

Knee Society Score (KSS) objective and function scores in 7 patients with Hemicap PF: preoperative and at follow-up.

Figure 5.

Knee Society Score (KSS) objective and function scores in 7 patients -HemiCAP Wave: preoperative and at follow-up till 8 years (median 6 years).

Figure 6.

Visual analogue scale (VAS) score in 17 patients not revised with HemiCAP in femur condyles up to 10 years follow-up (median 7.3 years).

Figure 7.

Visual analogue scale (VAS) score in 7 HemiCAP PF patients: preoperative, at 2 years, and at follow-up to 9 years.

Figure 8.

Visual analogue scale (VAS) score in 7 patients: preoperative and at follow-up median at 6 years.

Table 2 presents the potential risk factors for revision analyzed with uni- and multivariate analysis with the only significant factor—physical active work. There were no significant factors among the 3 different CAP types (P = 0.6).

Table 2.

Potential Risk Factors for Revision of 64 HemiCAP Mini-Prostheses. ^a

	Not Revised (n = 37)	Revised (n = 27)	Univariate Analysis			Multivariate Analysis
			Odds Ratio	95% CI	P	Odds Ratio	95% CI	P
Age, n (%)					0.1			0.2
<39 years	8 (22)	1 (4)	1			1
40-49 years	11 (30)	13 (48)	9.5	(1.0-89)		7.0	(0.7-67)
>50 years	18 (49)	13 (48)	5.8	(0.6-53)		4.3	(0.46-41)
Sex, n (%)					0.9
Male	16 (43)	12 (44)	1			1
Female	21 (57)	15 (56)	0.95	(0.3-2.6)		0.96	(0.3-2.9)
BMI, kg/m2, median (IQR)	28 (25-30)	30 (26-34)	1.1	(0.98-1.3)	0.09
Work, n(%)					0.04
Sedentary	32 (86)	17 (63)	1			1
Physically active	5 (14)	10 (37)	3.8	(1.1-13)		3.0	(0.8-11)
Sick leave	31 (84)	21 (78)	0.7	(0.2-2.4)	0.5
Morphine, n (%)	8 (22)	7 (26)	1.2	(0.4-4.1)	0.7
Meniscus injury, n (5)	8 (22)	9 (33)	1.8	(0.6-5.6)	0.3
Cap type, n (%)					0.6
1	20 (54)	17 (63)	1
3	9 (24)	4 (15)	0.5	(0.1-2.0)
4	8 (22)	6 (22)	0.9	(0.4-1.6)

CI = confidence interval; BMI = body mass index; IQR = interquartile range.

^aAge, sex, BMI, work (sedentary or physically active, sick leave ≥6 weeks), preoperative use of morphine, meniscus injury (operated on together with the cap operation), cap type (1 = HemiCAP condyle, 3 = HemiCAP patellofemoral [PF] and 4 = HemiCAP Wave).

Of the 64 HemiCAP procedures, 27 (42%) were revised, 17 (46%) HemiCAP condyle, 4 (36%) HemiCAP PF, and 6 (37%) HemiCAP Wave. The Kaplan-Meier survival rate was overall 60% at 10 years, and no failures were seen if the prosthesis survived the first 5 years ( Figs. 9 , 10 , and 11 ).

Figure 9.

Thirty-seven HemiCAP femur condyles followed up 10 years: 1 died, 17 (46%) were revised, and 19 (51%) survived with medium follow-up at 7.3 years. Seven HemiCAPs were on the lateral femur condyle, of which 3 were revised. Two patients were unable to participate in follow-up.

Figure 10.

Eleven HemiCAP PF followed up 10 years. Four (36%) were revised and 7 (64%) survived with medium follow-up at 7.3 years.

Figure 11.

Sixteen HemiCAP Wave in patients followed up 7.5 years. Six (37%) were revised and 10 (63%) survived with medium follow-up at 6 years. (Totally, there were inserted 18 HemiCAP Wave)—1 patient, 70 years old, was operated bilateral with HemiCAP Wave because bilateral isolated PF arthrosis was not revised. One patient was unable to participate at the clinical follow-up and 2 patients refused.

Discussion

The most important and interesting findings were the prosthesis survival rates from 5 to 10 years, which were the subject of this article. No further revisions were seen in this group. The clinical and function scores remained positive, and the KL grade did not worsen significantly in the operated or not operated knee compartments. Although there were a high revision rates during the first 5 years after the prosthesis implantation, this is more comparable to cartilage procedures such as autologous chondrocyte implantation or mosaicplasty with 10-year survival rates around 60% to 70%.^2-6,10 We found no significant differences among the 3 different CAP types, but workload and high activity were influencing the revision rate negatively. These findings have not been previously reported, and they may indicate that, with proper patient selection, an inlay mini-prosthesis can serve as a long-term treatment modality for difficult-to-treat patients between the ages of 35 and 65 years.^17-19,21,22

At present, this study is the largest long-term cohort study, with 64 HemiCAP femoral resurfacing mini-prostheses and clinical and radiographic follow-ups now up to 10 years, even with regard to the revision and survival rates.

As reported by Laursen et al., ²¹ there was a relatively high revision rate of 42% in the first 5 years. The revised patients required either a UKA or a TKA. Such revision rates can be seen in the national registries in Australia and Denmark.^24,26 The most important finding at the time of the reassessment was that there was a lack of significant OA progression (KL grading). This suggests that if the initial patient selection is appropriate, and particularly, if the implicated lesion is small and deep (ICRS grade 3 to 4),^17-19,21 rather than an element of generalized OA, this HemiCAP treatment may provide long-term improvement. Perhaps, it may prevent the need for a UKA or a TKA in middle-aged patients (35-65 years old) with symptomatic cartilage lesions or early degenerative knee pathologies.^2,17,21,22

Overall for the HemiCAPs, the survival rate was 60% up to 10 years postoperatively, which is acceptable compared with cartilage procedures^2-6,10,30 but unacceptable when compared with the UKA or TKA.^24,26,27 Overall, the study suggests that for the subgroup of patients who do not require early revisions, there is the potential for long-lasting treatment effectiveness.

In the short- to mid-term follow-up studies of the HemiCAP implants, we focused on pain (VAS) and functional ability (AKSS objective and function scores) and revision and time to revision, but we did not go into other risk factors for revision. Now we even focused on cofactors for eventual revision, as they might tell us more about the risk of revision and identify patient groups with the highest probability of survival, even long term. The only conclusion to be drawn from these data was that a heavy labor burden means a significant higher risk of revision, but there was no association in the multivariate analysis when tested against age, sex, and BMI. The strengths of this study included its large sample size, the follow-up duration of up to 10 years and the comprehensive data concerning the revisions, which was a gain of having a national registry. The study was limited in that it was a single-center case-cohort study with only 1 operating surgeon, who, at the same time, was the clinical investigator, thereby weakening the external validity. There was a low clinical follow-up rate caused by the revisions, but among those not revised, 86% were followed clinically. A prospective case series such as this yields heterogeneous patient material with regard to the cartilage pathology and previous surgery. This resulted in a study population that might be typical for the patient population suffering significantly from symptomatic cartilage lesions.

Conclusions

As hypothesized, we found good clinical and radiographic outcomes, and for those patients who did not require revisions, there were long-term improvements in disability and function. This suggests that patient selection is a key element to successfully applying these devices in clinical practice.