Interventional procedure overview of synthetic cartilage implant insertion for first metatarsophalangeal joint osteoarthritis (hallux rigidus)
Closed for comments This consultation ended on at Request commenting lead permission
Summary of key evidence on synthetic cartilage implant insertion for first metatarsophalangeal joint osteoarthritis (hallux rigidus)
Study 1 Baumhauer JF (2016)
Study type | RCT |
|---|---|
Country | UK and Canada (12 centres) |
Recruitment period | 2009 to 2012 |
Study population and number | n=219 (132 randomised to hydrogel implant, 65 randomised to arthrodesis and 22 non-randomised hydrogel implant) Patients with advanced great toe arthritis |
Age and sex | Mean age: 57 years (hydrogel implant), 55 years (arthrodesis); 79% (142/180) female |
Patient selection criteria | Inclusion criteria included: 18 years or older, degenerative or post-traumatic arthritis of the first MTP joint and a candidate for arthrodesis with grade 2, 3 or 4 according to Coughlin and Shurnas, preoperative visual analogue scale pain score 40 or above, presence of good bone stock, with less than 1 cm osteochondral cyst and without need for bone graft. Exclusion criteria included: active bacterial infection of the foot, additional ipsilateral lower limb pathology that needs active treatment, bilateral arthritis of the first MTP joint that needs simultaneous treatment of both MTP joints, previous cheilectomy resulting in inadequate bone stock, inflammatory arthropathy, gout, any significant bone loss, avascular necrosis, or large osteochondral cyst of the first MTP joint, lesions greater than 10 mm in size, hallux varus to any degree or hallux valgus more than 20°, physical conditions that would tend to eliminate adequate implant support, chronic anticoagulation for a bleeding disorder or has taken anticoagulants within 10 days before surgery, cancer diagnosis in previous 2 years and chemotherapy treatment or radiation to the lower extremity to be treated, suspected allergic reaction to polyvinyl alcohol, muscular imbalance, peripheral vascular disease that prohibits adequate healing or a poor soft-tissue envelope in the surgical field, absence of musculoligamentous supporting structures or peripheral neuropathy, any medical condition that makes the patient unsuitable for inclusion in the study in the opinion of the investigator, comorbidity that reduces life expectancy to less than 36 months, pregnancy or planning to become pregnant during the course of the study, breastfeeding, or not using contraception and childbearing age, history of substance abuse, prisoner or ward of the state. |
Technique | Implant: Cartiva Synthetic Cartilage Implant, Cartiva Inc., US. The implant was seated to allow for 1 to 2 mm to extend beyond the adjacent native cartilage of the metatarsal head. The patient could bear weight immediately and begin range of motion exercises at 1 week as tolerated. Skin sutures were removed at 2 to 3 weeks, at which time the patient could return to wearing his or her regular shoes. Arthrodesis: the proximal phalanx was positioned in slight dorsiflexion or valgus and stabilised with crossed screws or plate and screws. The patient's foot was immobilised in a cast or boot. Weight bearing was delayed until 2 to 6 weeks after the procedure at the discretion of the surgeon. |
Follow-up | 2 years |
Conflict of interest/source of funding | The authors declared no potential conflicts of interest with respect to the research, authorship, and publication of the article. Two authors reported personal fees or grants from Cartiva Inc. during the conduct of the study, 1 of whom also reported consultancy fees from DJO, Nextremity Solutions, Ferring Pharm, and Wright Medical. Five authors report grants or fees or other funds from Carticept. One author reported consulting and grant fees from Integra and Depuy. |
Analysis
Follow-up issues: A total of 236 patients were originally enrolled in the study; 17 withdrew before randomisation (leaving 197 randomised patients and 22 non-randomised patients) and 17 withdrew after randomisation (2% [2/132] in the implant group and 23% [15/65] in the arthrodesis group).Of the 202 treated patients, 3% (5/152) who had the implant and 6% (3/50) who had arthrodesis were lost to follow-up. Of the 5 patients lost to follow-up in the implant group, 2 were in the randomised group and 3 were in the non-randomised group of training patients.
Study design issues: Prospective, randomised, multicentre, non-inferiority study. Patients were randomised within 72 hours of surgery (method of randomisation not described). Of the 152 patients who had the implant, 22 were training patients and were not randomised. The primary endpoint of the study was a composite of 3 outcomes (pain, function and safety). A successful outcome was defined as improvement (decrease) from baseline in VAS pain of 30% or more at 12 months, maintenance of function from baseline in Foot and Ankle Ability Measure (FAAM) sports subscore at 12 months and absence of major safety events. The sample size was based on an 80% effect size at a 1-sided significance level of p<0.05 (n=210). Non-inferiority of the implant to arthrodesis was considered statistically significant if the 1-sided 95% lower confidence interval was greater than the equivalence limit (<15%). An intent to treat (ITT) analysis was done for all randomised patients and a modified intent to treat (mITT) analysis was done for all randomised and treated patients using last observation carried forward for missing data.
Study population issues: There were no statistically significant differences in the baseline characteristics of age, gender, height, weight, body mass index, VAS, FAAM ADL, SF-36 physical functioning, or hallux rigidus grade between treatment groups. Overall 30% of patients had grade 2 osteoarthritis, 54% had grade 3 and 16% had grade 4 (on the Coughlin and Shurnas scale from 0 to 4, where 0 represents normal radiographic findings and no pain and 4 represents substantial joint-space narrowing and nearly constant pain, including definite pain at mid-range of passive motion, and substantial stiffness).
Other issues: It appears that clinical outcomes were only reported for patients who retained their implant at follow-up, although this is not explicitly stated in the paper.
Key efficacy findings
Number of patients analysed: 197 randomised (132 hydrogel implant, 65 arthrodesis) and 22 non-randomised (hydrogel implant)
Analysis population (n) | Implant (%) | Arthrodesis (%) | 1-sided 95% lower bound (%) | Non-inferiority p value |
ITT (I:132; A:65) | 79 | 62 | 5.52 | <0.001 |
mITT (I:130, A:50) | 80 | 80 | -10.50 | <0.0075 |
Alternate effectiveness endpoint analyses in mITT population
Effectiveness endpoint components | Implant (%) | Arthrodesis (%) | 1-sided 95% lower bound (%) | Non-inferiority p value |
VAS, FAAM ADL, and safety – 12 months | 84.5 | 85.1 | -10.63 | 1.00 |
VAS, FAAM sports, and safety – 24 months | 80.0 | 78.7 | -10.17 | 0.835 |
VAS, FAAM ADL, and safety – 24 months | 80.5 | 78.7 | -9.64 | 0.832 |
FAAM sports scores by treatment group over time in mITT population
Visit | Implant Mean (SD), n Med (min, max) | Arthrodesis Mean (SD), n Med (min, max) | t-test p value | Wilcoxon p value |
Baseline | 36.9 (20.9), 127 34.4 (0, 100) | 35.6 (20.5), 50 31.3 (0, 87.5) | 0.694 | 0.505 |
2 weeks | 18.4 (18.3), 127 12.5 (0, 75) | 7.8 (12.4), 47 3.1 (0, 46.9) | 0.000 | 0.000 |
6 weeks | 39.5 (26.3), 126 35.7 (0, 100) | 22.4 (22.5), 49 20.3 (0, 81.3) | <0.0001 | 0.000 |
6 months | 66.6 (26.3), 120 65.6 (0, 100) | 78.6 (23.8), 42 79.7 (0, 100) | 0.01 | 0.005 |
1 year | 75.8 (24.8), 120 81.2 (0, 100) | 84.1 (16.9), 43 90.6 (28.1, 100) | 0.043 | 0.098 |
2 years | 79.5 (24.6), 113 87.5 (0, 100) | 82.7 (20.5), 41 90.6 (28.1, 100) | 0.461 | 0.437 |
FAAM ADL scores by treatment group over time in mITT population
Visit | Implant Mean (SD), n Med (min, max) | Arthrodesis Mean (SD), n Med (min, max) | t-test p value | Wilcoxon p value |
Baseline | 59.4 (16.9), 129 58.3 (7.1, 100) | 56.0 (16.8), 50 54.9 (22.6, 95.2) | 0.222 | 0.152 |
2 weeks | 48.8 (21.6), 126 47.6 (2.4, 100) | 40.3 (20.7), 47 39.3 (7.5, 84.2) | 0.021 | 0.023 |
6 weeks | 69.0 (19.0), 126 69.6 (19.0, 100) | 59.6 (24.8), 48 63.1 (10.7, 100) | 0.008 | 0.032 |
6 months | 82.7 (17.5), 123 88.1 (22.6, 100) | 89.9 (12.4), 43 95.2 (50.0, 100) | 0.014 | 0.01 |
1 year | 88.6 (14.4), 123 95.0 (27.4, 100) | 94.1 (6.8), 43 95.2 (71.4, 100) | 0.0176 | 0.066 |
2 years | 90.4 (15.0), 116 96.4 (29.8, 100) | 94.6 (7.1), 41 96.4 (69.0, 100) | 0.082 | 0.524 |
SF-36 Physical Functioning scores by treatment group over time in mITT population
Visit | Implant Mean (SD), n Med (min, max) | Arthrodesis Mean (SD), n Med (min, max) | t-test p value | Wilcoxon p value |
Baseline | 52.4 (22.8), 130 50 (0, 100) | 49.8 (23.6), 50 40 (15, 100) | 0.499 | 0.352 |
6 weeks | 60.7 (23.7), 128 60 (10, 100) | 44.7 (26.8), 49 45 (0, 100) | 0.000 | 0.000 |
6 months | 72.3 (26.3), 124 80 (0, 100) | 82.8 (22.4), 43 90 (5, 100) | 0.021 | 0.014 |
1 year | 78.9 (22.7), 123 90 (5, 100) | 83.7 (24.9), 43 95 (0, 100) | 0.247 | 0.064 |
2 years | 83.2 (20.9), 116 95 (25, 100) | 85.1 (19.5), 41 95 (5, 100) | 0.613 | 0.597 |
VAS pain scores by treatment group over time in mITT population
Visit | Implant Mean (SD), n Med (min, max) | Arthrodesis Mean (SD), n Med (min, max) | t-test p value | Wilcoxon p value |
Baseline | 68.0 (13.9), 130 68.3 (27.8, 100) | 69.3 (143), 50 70 (38, 97.5) | 0.571 | 0.529 |
6 weeks | 33.2 (24.7), 128 27.4 (0, 96) | 17.2 (17.6), 48 11.5 (0, 71.8) | <0.0001 | 0.000 |
6 months | 28.9 (27.8), 124 20.5 (0, 97) | 11.7 (18.3), 43 4.3 (0, 74.8) | 0.000 | 0.000 |
1 year | 17.8 (23.0), 123 9.0 (0, 91) | 5.7 (8.5), 43 2.5 (0, 56.5) | 0.0011 | 0.000 |
2 years | 14.5 (22.1), 116 5.0 (0, 94) | 5.9 (12.1), 41 1.5 (0, 70) | 0.002 | 0.005 |
Active peak dorsiflexion angles by treatment group over time in mITT population
Visit | Implant Mean (SD), n Med (min, max) | Arthrodesis Mean (SD), n Med (min, max) | t-test p value | Wilcoxon p value |
Baseline | 22.7 (11.2), 130 20 (0, 58) | 22.9 (11.2), 50 20 (5, 50) | 0.910 | 0.965 |
6 weeks | 25.1 (10.8), 127 25 (5, 55) | 13.0 (9.0), 48 14.5 (0, 26) | <0.0001 | 0.000 |
6 months | 28.1 (9.8), 124 30 (5, 60) | 14.9 (8.6), 44 15 (0, 30) | <0.0001 | 0.000 |
1 year | 28.8 (11.2), 123 30 (5, 60) | 16.0 (7.3), 43 15 (0, 35) | <0.0001 | 0.000 |
2 years | 29 (11.9), 114 30 (5, 60) | 15.1 (8.4), 41 16 (0, 35) | <0.0001 | 0.000 |
Implant removal or reoperation in implant group=11.2% (17/152)
Implant removal and conversion to arthrodesis=9.2% (14/152) (conversion to arthrodesis was considered to be straightforward and resulted in 86% reduction in pain and 39 point increase in function)
Joint manipulation for motion=0.7% (1/152)
Debridement of the joint for scar and synovitis and implant repositioning=0.7% (1/152)
Moberg osteotomy of the proximal phalanx for improved toe positioning, motion, and pain relief=0.7% (1/152)
There were 7 (14%) procedures for hardware removal in the arthrodesis group.
Key safety findings
Adverse events in safety population
Implant (n=152) | Arthrodesis (n=50) | ||||||
Events | n | % | Events | n | % | p value | |
Any adverse event | 245 | 105 | 69.1 | 72 | 36 | 72.0 | 0.727 |
Treatment emergent event | 102 | 67 | 44.1 | 32 | 21 | 42.0 | 0.870 |
Nontreatment emergent event | 143 | 73 | 48.0 | 40 | 26 | 52.0 | 0.745 |
Any serious adverse event | 37 | 30 | 19.7 | 12 | 9 | 18.0 | 0.999 |
Treatment emergent event | 17 | 17 | 11.2 | 4 | 4 | 8.0 | 0.605 |
Nontreatment emergent event | 20 | 14 | 9.2 | 8 | 5 | 10.0 | 0.999 |
Study 2 Goldberg A (2017)
Study type | RCT (same trial as reported in Baumhauer JF, 2016) |
|---|---|
Country | UK and Canada (12 centres) |
Recruitment period | 2009 to 2012 |
Study population and number | n=219 (132 randomised to hydrogel implant, 65 randomised to arthrodesis and 22 non-randomised hydrogel implant) Patients with advanced great toe arthritis |
Age and sex | Mean age: 57 years (hydrogel implant), 55 years (arthrodesis); 79% (142/180) female |
Patient selection criteria | Inclusion criteria included: 18 years or older, degenerative or post-traumatic arthritis of the first MTP joint and a candidate for arthrodesis with grade 2, 3 or 4 according to Coughlin and Shurnas, preoperative visual analogue scale pain score 40 or above, presence of good bone stock, with less than 1 cm osteochondral cyst and without need for bone graft. Exclusion criteria included: active bacterial infection of the foot, additional ipsilateral lower limb pathology that needs active treatment, bilateral arthritis of the first MTP joint that needs simultaneous treatment of both MTP joints, previous cheilectomy resulting in inadequate bone stock, inflammatory arthropathy, gout, any significant bone loss, avascular necrosis, or large osteochondral cyst of the first MTP joint, lesions greater than 10 mm in size, hallux varus to any degree or hallux valgus more than 20°, physical conditions that would tend to eliminate adequate implant support, chronic anticoagulation for a bleeding disorder or has taken anticoagulants within 10 days before surgery, cancer diagnosis in previous 2 years and chemotherapy treatment or radiation to the lower extremity to be treated, suspected allergic reaction to polyvinyl alcohol, muscular imbalance, peripheral vascular disease that prohibits adequate healing or a poor soft-tissue envelope in the surgical field, absence of musculoligamentous supporting structures or peripheral neuropathy, any medical condition that makes the patient unsuitable for inclusion in the study in the opinion of the investigator, comorbidity that reduces life expectancy to less than 36 months, pregnancy or planning to become pregnant during the course of the study, breastfeeding, or not using contraception and childbearing age, history of substance abuse, prisoner or ward of the state. |
Technique | Implant: Cartiva Synthetic Cartilage Implant, Cartiva Inc., US. The implant was seated to allow for 1 to 2 mm to extend beyond the adjacent native cartilage of the metatarsal head. The patient could bear weight immediately and begin range of motion exercises at 1 week as tolerated. Skin sutures were removed at 2 to 3 weeks, at which time the patient could return to wearing his or her regular shoes. Arthrodesis: the proximal phalanx was positioned in slight dorsiflexion or valgus and stabilised with crossed screws or plate and screws. The patient's foot was immobilised in a cast or boot. Weight bearing was delayed until 2 to 6 weeks after the procedure at the discretion of the surgeon. |
Follow-up | 2 years |
Conflict of interest/source of funding | One or more of the authors has received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this article. The authors disclosed receipt of research and institutional support, consulting fees and support for travel expenses from Cartiva Inc. for the research, authorship, and/or publication of this article. |
Analysis
Follow-up issues: A total of 236 patients were originally enrolled in the study; 17 withdrew before randomisation (leaving 197 randomised patients and 22 non-randomised patients) and 17 withdrew after randomisation (2% [2/132] in the implant group and 23% [15/65] in the arthrodesis group).Of the 202 treated patients, 3% (4/152) who had the implant and 6% (3/50) who had arthrodesis were lost to follow-up. Of the 4 patients lost to follow-up in the implant group, 1 was in the randomised group and 3 were in the non-randomised group of training patients.
Study design issues: Prospective, randomised, multicentre, non-inferiority study. Patients were randomised within 72 hours of surgery (method of randomisation not described). Of the 152 patients who had the implant, 22 were training patients and were not randomised. The purpose of this study was to evaluate the longitudinal data from the original RCT, to determine the association between numerous patient factors and the success or failure of these procedures. The original trial was powered for non-inferiority to demonstrate equivalence of the 2 procedures, so it may not be sufficiently powered for some patient factors. A patient's outcome was deemed successful if composite primary endpoint criteria for clinical success were met at 24 months, namely: 1) VAS pain reduction ≥30%; 2) maintenance or improvement in function; 3) freedom from radiographic complications; and 4) no secondary surgical intervention. Final outcome data were assessed in the modified intent to treat population.
Study population issues: Patient demographics and baseline outcome measures were similar for both groups.
Key efficacy findings
Number of patients analysed: 176 (129 synthetic cartilage implant, 47 arthrodesis)
Success rates stratified by patient factors
Patient variable | Stratification | Synthetic cartilage implant % success | p value | Arthrodesis % success | p value | p value between groups |
Coughlin hallux rigidus grade | 2 | 72.2% (26/36) | 0.364 | 66.7% (12/18) | 0.331 | 0.759 |
3 | 83.6% (61/73) | 85.0% (17/20) | 0.999 | |||
4 | 80.0% (16/20) | 88.9% (8/9) | 0.999 | |||
Preoperative hallux valgus angle | 0 to <15° | 80.2% (81/101) | 0.797 | 82.4% (28/34) | 0.429 | 0.999 |
15° to 20° | 78.6% (22/28) | 69.2% (9/13) | 0.698 | |||
Preoperative active peak dorsiflexion | 40° to 60° | 70.0% (7/10) | 0.308 | 50.0% (2/4) | 0.357 | 0.580 |
30° to <40° | 77.3% (17/22) | 90.0% (9/10) | 0.637 | |||
>10° to <30° | 77.8% (56/72) | 80.1% (21/26) | 0.999 | |||
≤10° | 92.0% (23/25) | 71.4% (5/7) | 0.201 | |||
Gender | Female | 77.9% (81/104) | 0.405 | 80.6% (29/36) | 0.679 | 0.817 |
Male | 88.0% (22/25) | 72.7% (8/11) | 0.343 | |||
Age | ≥65 years | 90.9% (20/22) | 0.243 | 100% (9/9) | 0.172 | 0.999 |
<65 years | 77.6% (83/107) | 73.7% (28/38) | 0.659 | |||
Body mass index | <30 kg/m2 | 80.9% (76/94) | 0.629 | 76.9% (30/39) | 0.667 | 0.640 |
≥30 kg/m2 | 77.1% (27/35) | 87.5% (7/8) | 0.999 | |||
Duration of symptoms before surgery | <24 months | 66.7% (10/15) | 0.183 | 100% (3/3) | 1.00 | 0.522 |
≥24 months | 81.6% (93/114) | 77.3% (34/44) | 0.655 | |||
Previous first MTP joint surgery | Yes | 66.7% (8/12) | 0.259 | 100% (4/4) | 0.564 | 0.516 |
No | 81.2% (95/117) | 76.7% (33/43) | 0.513 | |||
Preoperative pain VAS score | Mild (0 to <40 mm) | 50.0% (1/2) | 0.196 | 50.0% (1/2) | 0.140 | 0.999 |
Moderate (40 to 58 mm) | 88.9% (24/27) | 100% (8/8) | 0.999 | |||
Severe (>58 to 100 mm) | 78.0% (78/100) | 75.7% (28/37) | 0.819 |
Study 3 Glazebrook M (2019)
Study type | Case series (patients who had implant insertion as part of Motion RCT) |
|---|---|
Country | UK and Canada (12 centres) |
Recruitment period | 2009 to 2013 |
Study population and number | n=119 Patients who had synthetic cartilage implant hemiarthroplasty for advanced hallux rigidus |
Age and sex | Mean 58 years (range 30 to 79); 78% (87/112) female |
Patient selection criteria | Patients included in the original study were at least 18 years old and had been diagnosed with Coughlin grade 2, 3 or 4 hallux rigidus. |
Technique | Implant: Cartiva Synthetic Cartilage Implant, Cartiva Inc., US. The implant was seated to allow for 1 to 2 mm to extend beyond the adjacent native cartilage of the metatarsal head. The patient could bear weight immediately and begin range of motion exercises at 1 week as tolerated. Skin sutures were removed at 2 to 3 weeks, at which time the patient could return to wearing regular shoes. |
Follow-up | Mean 5.8 years |
Conflict of interest/source of funding | The authors received research and institutional support, consulting fees and support for travel expenses from Cartiva Inc. Twelve authors reported personal fees and/or grants from Cartiva during the conduct of the study and 2 authors reported personal fees from Carticept. |
Analysis
Follow-up issues: Of the original 152 patients who had the implant inserted, 14 had the implant removed and conversion to arthrodesis and 3 were lost to follow-up during the first 24 months. Of the 135 patients eligible to be included in this 5-year follow-up study, 17 (13%) could not be contacted (the implant status was known for 7 of these patients); 5 patients declined to participate, and 1 patient died. The study population therefore included 74% (112/152) of the original cohort and 83% (112/135) of eligible patients. Clinical outcomes were available for 106 patients at last follow-up (6 patients who underwent implant removal and revision were excluded).
Study design issues: Prospective, multicentre case series of patients treated as part of an RCT. The primary outcome was implant survival at a minimum of 5 years. Primary endpoint analyses included the 112 patients in the analysis set plus the 7 patients not enrolled in the follow-up study but for whom the device status was known (which included 3 implant removals). A target sample size of 115 was calculated to give 80% power to reject the null hypothesis that the implant removal rate from 2 to 5 years is 23.5% or greater, versus the alternative hypothesis that the implant removal rate is less than 23.5%. Clinical outcomes were summarised for patients who retained their implant.
Study population issues: The patient cohort enrolled in the 5-year follow-up study was older (58.2±8.8 years) than the 40 patients who were not enrolled (54.3±7.4 years, p=0.004) and they had a lower mean VAS score at baseline (67.0±14.4 compared with 72.4±12.2, p=0.031).
Key efficacy findings
Number of patients analysed: 119
Outcome | n | mean±SD | range | 99% CI |
VAS pain, mm | 106 | -57.9±18.6 | -96.0, -3.0 | -65.2 to -53.5 |
FAAM ADL, points | 105 | 33.0±17.6 | -8.3, 92.9 | 28.6 to 37.5 |
FAAM sports, points | 104 | 47.9±27.1 | -25.0, 100 | 39.3 to 57.1 |
FAAM ADL, current level of function, % | 104 | 33.3±22.5 | -29.0, 90.0 | 25.0 to 40.0 |
FAAM sports, current level of function, % | 103 | 39.2±31.2 | -70.0, 100 | 25.0 to 50.0 |
Patients with a clinically significant improvement in outcome at mean follow-up of 5.8 years after implant insertion among patients free from implant removal
Number of patients (%) | 99% CI | |
Pain VAS ≥30% decrease | 103/106 (97.2%) | 90.8% to 100% |
FAAM ADL ≥8 points increase | 95/105 (90.5%) | 81.7% to 100% |
FAAM sports ≥9 points increase | 97/104 (93.3%) | 85.3% to 100% |
Active MTP joint peak dorsiflexion an active MTP joint natural dorsiflexion were maintained at 5.8 years. The postoperative axial MTP joint alignment was a mean of 8.2° (±5.7) (range 0 to 20°).
When patients were asked if their overall wellbeing had improved, 56.2% (59/105) strongly agreed and 31.4% (33/105) agreed.
When asked if they would have the procedure again, 93.4% (99/106) stated they would and 6.6% (7/106) stated they would not.
Implant survival
Implant survival rate at 24 months=90.8% (138/152)
Between years 2 and 5, 7.6% (9/119) patients had implant removal and conversion to arthrodesis (upper bound of 1-sided 95% confidence interval [CI] was 11.3%). In a sensitivity analysis, the upper bound of the1-sided CI remained below 23.5% (the null hypothesis) even when all missing values were defined as failures.
Of the 9 patients who had implant removal, 8 were successfully converted to arthrodesis with no additional complications.
Kaplan-Meier implant survivorship at 5.8 years follow-up=84.9%
Key safety findings
Adverse events
Serious pain at implant site=4.5% (5/112) (all patients had implant removal and conversion to arthrodesis, as described in efficacy section). One of these patients had a second operation for hardware removal after fusion.
One patient had implant removal, sinus debridement, and removal of tissue from the first MTP joint at about 36 months' follow-up and a Staphylococcus aureus infection was confirmed.
Bony reactions at 5.8 years:
Erosion=2.1% (2/97)
Cystic changes=20.6% (20/97)
Loss of cortical margins=24.7% (24/97)
Osteolysis=2.1% (2/97)
Any bony reaction=49.5% (48/97)
No bony reaction=50.5% (49/97)
No evidence of avascular necrosis, device migration or fragmentation was observed upon independent radiographic review.
1 patient developed a radiolucency of ≤2 mm at the bone implant surface between 2 and 5.8 years follow-up.
Study 4 Cassinelli S (2019)
Study type | Case series |
|---|---|
Country | US |
Recruitment period | 2016 to 2018 |
Study population and number | n=60 (64 implants) Patients who hallux rigidus that had not responded to conservative management |
Age and sex | Mean 62 years (range 38 to 86); 87% (52/60) female |
Patient selection criteria | Inclusion criteria: patients diagnosed with hallux rigidus that had failed nonoperative management, with a minimum 12-month clinical or telephone follow-up. Exclusion criteria: patients with hallux valgus angle >20°, concomitant bunion correction, peripheral vascular disease, prior fusion of the ankle, hindfoot or midfoot, inflammatory arthropathy, or peripheral neuropathy. No patient had concomitant bilateral synthetic cartilage implantation. |
Technique | Implant: Cartiva Synthetic Cartilage Implant, Cartiva Inc., US. An 8- or 10-mm implant was used and seated to allow the implant to sit about 2 to 2.5 mm proud relative to the articular surface. With the initial patients, plantar stripping of the metatarsal head and sesamoids was done when 70° of dorsiflexion was not achieved. This technique was abandoned because of early postoperative stiffness seen at follow-up. Subsequently, when necessary, a Moberg closing wedge osteotomy of the proximal phalanx was done. Additional forefoot or hindfoot procedures were done as necessary. Patients were placed in a postoperative shoe and were allowed weight bearing as tolerated. Sutures were removed 2 weeks after the procedure and the patient transitioned to a regular shoe as swelling and pain permitted. |
Follow-up | Mean 18.5 months (range 12 to 30) |
Conflict of interest/source of funding | No financial support was received for the research, authorship, or publication of the article. One author reported stock in Paragon 28, speaking for Stryker and Editor-in-Chief of Foot & Ankle International, outside the submitted work. One author reported personal fees from Extremity Medical, outside the submitted work. |
Analysis
Follow-up issues: Clinical and telephone follow-up were available for all patients, but some outcomes were only available at the most recent clinical follow-up for 66% (42/64) and 63% (40/64) of implants.
Study design issues: Retrospective, single centre case series of consecutive patients. Outcomes included the Patient Reported Outcomes Measurement Information System (PROMIS) physical function score, PROMIS pain interference score and patient satisfaction (measured on a 1 to 5 point satisfaction scale).
Study population issues: 23% of patients had previously had surgery on the hallux. Of the 64 implants, 18 (25%), 40 (60%) and 8 (13%) were classified as grade 2, 3 and 4 hallux rigidus respectively, according to the Coughlin and Shurnas grading system.
Other issues: 45% (29/64) of implant insertions had at least 1 additional procedure done at the same time (11 Moberg osteotomy, 3 medial eminence resection, 3 hindfoot or ankle procedures, 14 'other forefoot' procedures including lesser hammer toe correction, web space neurolysis, plantar fibroma excision and nail margin ablation).
Key efficacy findings
Number of patients analysed: 60 (64 implants)
Mean PROMIS pain interference score at follow-up (mean 15 months)=60 (mild pain); range 2 to 30; n=40
Mean PROMIS physical function score at follow-up (mean 15 months)=42 (mild dysfunction); range 2 to 30; n=42
Proportion of patients who would have the same surgery again=65.6% (42/64)
Proportion of patients who would rather have had arthrodesis=10.9% (7/64)
Reoperation=20% (13/64) (conversion to fusion [5 patients], lysis of adhesions [4 patients], Moberg osteotomy [1 patient], and implant exchange with bone grafting for impinging soft tissue or implant subsidence [3 patients]). Conversion to fusion was done at a mean of 16.4 months postoperatively (range 10 to 26 months).
Satisfaction rates
Very unsatisfied=26.6% (17/64)
Unsatisfied=10.9% (7/64)
Neutral=20.3% (13/64)
Satisfied=28.1% (18/64)
Very satisfied=14.1% (9/64)
Postoperative corticosteroid injection=52% (33/64) (mean 7.6 months after the procedure)
Patient-reported restricted hallux MTP range of motion needing the use of a dynamic splinting device after the procedure=14% (9/64)
Key safety findings
30% (19/64) of implants had advanced imaging workup after the procedure for continued pain, inflammation and further evaluation of the implant and surrounding structures (mean 12 months follow-up, range 3 to 30 months). MRIs showed residual capsular inflammation in all patients, bone marrow oedema of the proximal phalanx or metatarsal in 18 patients, and degenerative changes or inflammation at the metatarsosesamoid articulation in 7 patients.
In 1 patient who had implant removal and conversion to fusion, the implant had subsided below the cortical bone of the metatarsal head with resultant bony contact between the proximal phalanx and metatarsal head.
A second patient had revision for continued pain, inflammation, and suspected implant subsidence. The implant had subsided and was damaged and pathology was consistent with foreign body giant cell reaction and neovascularisation. A bone mineral density scan revealed the patient to have osteopenia.
Study 5 Brandao B (2020a)
Study type | Non-randomised comparative study |
|---|---|
Country | UK |
Recruitment period | Not reported |
Study population and number | n=72 (30 implant insertion, 42 arthrodesis) Adults with symptomatic hallux rigidus |
Age and sex | Mean age: 57 years (implant), 64 years (arthrodesis); 69% (50/72) female |
Patient selection criteria | Inclusion criteria included adult patients over 18 years old with symptomatic hallux rigidus who had primary synthetic cartilage implant hemiarthroplasty or arthrodesis. Patients with previous surgery on the first metatarsal and patients with traumatic osteoarthritis were excluded. |
Technique | Implant: Cartiva Synthetic Cartilage Implant, Cartiva Inc., US. |
Follow-up | Mean 18 months (range 12 to 30) for implant group; mean 19 months (range 14 to 36) for arthrodesis group |
Conflict of interest/ source of funding | None |
Analysis
Follow-up issues: Patients were only included if they had at least 1 year follow-up.
Study design issues: Single centre non-randomised comparative study. First MTP joint arthritis was graded preoperatively according to the Hattrup and Johnson classification (scale 1 to 3, where 1 is mild changes with minimal osteophytosis, grade 2 is moderate changes including narrowing of the joint with osteophytosis of the metatarsal head or phalanx and subchondral sclerosis or cysts, and grade 3 is severe arthritis with loss of joint space, marked osteophytosis and subchondral bone changes). The main outcome measure was sporting ability evaluated using the FAAM sports subscale.
Study population issues: Of the 30 patients in the implant group, 0 had grade 1 arthritis, 9 (30%) had grade 2 and 21 (70%) had grade 3. In the arthrodesis group, 3 (7%) had grade 1 arthritis, 17 (40%) had grade 2 and 22 (52%) had grade 3. In the implant group, most patients participated in walking (63%) followed by gym sports (26%) and running sports (11%). In the arthrodesis group, gym sports (48%) and walking (43%) were the most popular sporting activities followed by running sports (9%).
Other issues: Dorsiflexion was not evaluated because there was no funding for patients to return to the clinic.
Key efficacy findings
Number of patients analysed: 72 (30 implant, 42 arthrodesis)
Mean postoperative FAAM Sports scores
Implant=76.4% (SD ±16.6)
Arthrodesis=80.9% (SD ±21.9), p>0.3
Adjusting the results for age (<55 and >55) and gender (male or female) showed no statistically significant results.
Reoperation
Revision to arthrodesis in implant group=3.3% (1/30) (because of persistent pain at 17 months after implant insertion)
Key safety findings
No safety data were reported.
Study 6 Chrea B (2020)
Study type | Non-randomised comparative study |
|---|---|
Country | US |
Recruitment period | 2016 to 2018 |
Study population and number | n=133 (60 cheilectomy and Moberg osteotomy with synthetic cartilage implant insertion, 73 cheilectomy and Moberg osteotomy alone) Adults with moderate to advanced hallux rigidus |
Age and sex | Mean age: 56 years (range 25 to 75); 73% (97/133) female |
Patient selection criteria | Inclusion criteria: patients with moderate to advanced hallux rigidus. Patients were excluded if they had prior surgical treatment for their condition or if they had cheilectomy alone, with or without implant. Patients who had the implant alone were also excluded. Patients with little to no motion at the first MTP joint and with advanced arthritis on plain film radiographs had MTP fusion and so were not included. Patients who had insufficient follow-up or were missing baseline functional outcome scores were excluded. |
Technique | Implant: Cartiva Synthetic Cartilage Implant, Cartiva Inc., US. All patients had cheilectomy and Moberg osteotomy with or without synthetic cartilage implant insertion. In those patients who also had an implant, the procedure was modified to include a limited cheilectomy. Of the 60 patients who had an implant, 3 had an 8-mm implant and 57 had a 10-mm implant. The implant was placed in the metatarsal head 2 mm beyond the margin of the metatarsal head. Patients could bear weight immediately or were limited for the first 2 weeks to allow the incision to heal. Sutures were removed 2 to 3 weeks postoperatively and patients were transitioned into regular shoe wear. |
Follow-up | Mean 28 months (range 16 to 46) for patients with implants and mean 37 months (range 19 to 48) for patients with cheilectomy and osteotomy alone. |
Conflict of interest/ source of funding | 1 author reported 'other' potential conflict of interest from Wright Medical, outside the submitted work. |
Analysis
Follow-up issues: The cheilectomy and osteotomy alone group had a longer follow-up period than the group who also had an implant inserted.
Study design issues: Retrospective non-randomised comparative study. The severity of hallux rigidus was assessed using the Coughlin and Shurnas classification system. Patient-reported outcomes were assessed using Patient-Reported Outcomes Measurement Information System (PROMIS) scores. Higher scores indicate greater physical function, pain interference, pain intensity, global health, and depression.
Study population issues: Patients had grade 2, 3, or 4 osteoarthritis based on combined radiologic and clinical examination.
Key efficacy findings
Number of patients analysed: 133 (60 cheilectomy and Moberg osteotomy with implant, 73 cheilectomy and Moberg osteotomy without implant)
Comparison of PROMIS scores between treatment groups
Domain | Preoperative score (± SD) | Postoperative score (± SD) | p value | Score change (± SD) |
Physical function | ||||
Implant group | 44.6 (± 8.2) | 48.8 (± 8.0) | <0.01 | +3.6 (± 6.2) |
No implant group | 45.0 (± 6.3) | 51.8 (± 8.7) | <0.01 | +7.1 (± 8.5) |
p value | 0.763 | 0.043 | 0.011 | |
Pain interference | ||||
Implant group | 58.9 (± 6.5) | 51.2 (± 8.4) | <0.01 | -7.2 (± 8.6) |
No implant group | 58.1 (± 5.7) | 49.4 (± 9.6) | <0.01 | -9.1 (± 9.2) |
p value | 0.466 | 0.248 | 0.268 | |
Pain intensity | ||||
Implant group | 51.6 (± 6.1) | 43.4 (± 8.7) | <0.01 | -7.5 (± 7.7) |
No implant group | 49.9 (± 6.7) | 39.9 (± 8.3) | <0.01 | -10.2 (± 8.6) |
p value | 0.139 | 0.019 | 0.086 | |
Global physical health | ||||
Implant group | 46.9 (± 7.5) | 51.9 (± 8.7) | <0.01 | +5.1 (± 7.2) |
No implant group | 47.3 (± 7.4) | 53.8 (± 8.2) | <0.01 | +6.7 (± 7.2) |
p value | 0.765 | 0.216 | 0.229 | |
Global mental health | ||||
Implant group | 53.1 (± 8.1) | 54.0 (± 8.8) | 0.245 | +1.1 (± 5.3) |
No implant group | 54.1 (± 8.9) | 55.5 (± 8.8) | 0.074 | +1.0 (± 6.9) |
p value | 0.507 | 0.352 | 0.948 | |
Depression | ||||
Implant group | 46.9 (± 7.7) | 47.6 (± 6.4) | 0.509 | +0.4 (± 5.9) |
No implant group | 47.4 (± 8.5) | 46.9 (± 8.3) | 0.611 | -1.9 (± 10.0) |
p value | 0.750 | 0.606 | 0.148 |
Revision rate
Implant group=5.0% (3/60)
No implant group=1.4% (1/73)
In the implant group, revisions were needed at 14, 22 and 33 months respectively after the index surgery. In 1 patient, there was inflammation and fibrous tissue at the MTP joint with loosening of the implant; the implant was removed and a conversion to arthrodesis and bone grafting were done. A second patient presented with pain and there was wearing of the medial aspect of the implant; the implant was removed, and a revision hemiarthroplasty was done. The third patient presented with persistent pain and a hyperdorsiflexed hallux; the patient had revision to correct the hyperdorsiflexion.
In the group of patients who had cheilectomy and Moberg osteotomy without an implant, the patient who needed a revision was diagnosed with a metabolic bone disorder concurrently managed by a metabolic bone specialist and rheumatologist. She had revision cheilectomy with synthetic cartilage implant insertion at 21 months after the index procedure. The patient subsequently developed persistent pain and an MRI scan showed a stable implant with significant oedema 8 months after the revision procedure. She had 2 rounds of shockwave therapy as well as ultrasound-guided injection of the first MTP.
Key safety findings
Persistent pain (treated by steroid injections, orthotics or shockwave therapy):
Implant group=11.7% (7/60)
No implant group=11.0% (8/73), p=0.90
Postoperative infection needing antibiotics:
Implant group=5.0% (3/60)
No implant group=0% (0/73), p=0.05
Study 7 Eble S (2020)
Study type | Case series |
|---|---|
Country | US |
Recruitment period | 2017 to 2018 |
Study population and number | n=103 Patients with hallux rigidus |
Age and sex | Mean 58 years (range 26 to 76); 72% (74/103) female |
Patient selection criteria | Patients treated with or without concurrent Moberg osteotomy were eligible for inclusion. Patients who had a polyvinyl alcohol hydrogel implant for a condition other than hallux rigidus were excluded. |
Technique | Implant: Cartiva Synthetic Cartilage Implant, Wright Medical Group, US. A 10-mm implant was used in 66 patients, and an 8-mm implant in 7 patients. The implant was placed into the implant delivery system and seated in the metatarsal head protruding 2 mm above the metatarsal head. A soft dressing with postoperative shoe or splint was applied. Patients could immediately bear weight or were limited for the first 2 weeks to allow the incision to heal. Sutures were removed 2 to 3 weeks postoperatively and patients were transitioned into regular shoe wear. 71% (52/73) of patients had concurrent Moberg osteotomy. |
Follow-up | Mean clinical follow-up=26.2 months (range 14 to 36) |
Conflict of interest/ source of funding | None |
Analysis
Follow-up issues: Of the 103 patients, 90 (87.4%) had baseline PROMIS scores, and 81.1% (73/90) of these had minimum 1-year postoperative scores, 70.9% (73/103) of patients had both preoperative and postoperative scores.
Study design issues: Retrospective case series. Patient-reported outcome scores and clinical outcomes were assessed.
Population issues: Of the 73 patients with PROMIS scores, 10 had had a prior procedure of the first MTP, and 52 had concurrent Moberg osteotomy at the time of synthetic cartilage implant insertion. There may be some patient overlap with Chrea et al. (2020).
Key efficacy findings
Number of patients analysed: 103
Baseline and postoperative PROMIS scores (n=73); mean follow-up=14 months
PROMIS Domain | Mean baseline score | Clinical interpretation of baseline score | Mean postoperative score | Clinical interpretation of postoperative score | p value |
Physical function | 44.7 | Mild functional impairment | 48.2 | Normal range | 0.009 |
Pain interference | 58.0 | Mild pain symptoms | 52.5 | Normal range | <0.0001 |
Pain intensity | 50.9 | Normal range | 43.5 | Normal range | <0.0001 |
Global physical health | 46.9 | Normal range | 50.8 | Normal range | 0.006 |
Global mental health | 53.3 | Very good function | 54.2 | Very good function | 0.499 |
Depression | 47.1 | Normal range | 47.7 | Normal range | 0.632 |
Comparison of patients who had concurrent Moberg osteotomy with those who did not
PROMIS Domain | Timepoint | Moberg (n=46) | No Moberg (n=17) | p value |
Physical function | Mean baseline | 45.8 | 44.4 | 0.53 |
Mean postoperative | 50.0 | 46.5 | 0.12 | |
Mean change | +4.2 | +2.1 | 0.28 | |
Pain interference | Mean baseline | 58.1 | 57.3 | 0.65 |
Mean postoperative | 50.7 | 54.9 | 0.05 | |
Mean change | -7.3 | -2.4 | 0.03 | |
Pain intensity | Mean baseline | 48.7 | 50.8 | 0.43 |
Mean postoperative | 41.1 | 46.1 | 0.02 | |
Mean change | -10.2 | -4.9 | 0.05 | |
Global physical health | Mean baseline | 46.9 | 46.3 | 0.83 |
Mean postoperative | 52.7 | 48.3 | 0.09 | |
Mean change | +5.2 | +1.8 | 0.12 | |
Global mental health | Mean baseline | 52.5 | 53.5 | 0.73 |
Mean postoperative | 54.9 | 54.3 | 0.83 | |
Mean change | +1.7 | +0.7 | 0.57 | |
Depression | Mean baseline | 45.0 | 47.3 | 0.37 |
Mean postoperative | 46.8 | 48.7 | 0.35 | |
Mean change | -0.6 | +1.4 | 0.46 |
Revision surgery=1.9% (2/103)
Both revisions were for persistent pain. The first was a conversion to arthrodesis at 14 months after the index procedure; the implant was noted to be loose, with inflammation and fibrous tissue at the joint. The second revision was a hemiarthroplasty with synthetic cartilage implant, done at 21 months postoperatively. The implant was found to have marked loss of contour at the medial aspect of the toe.
Therapeutic steroid injection=5.8% (6/103); between 2 and 11 months postoperatively.
Symptom-specific orthotics=5.8% (6/103); between 3 and 6 months postoperatively.
Key safety findings
MRI scans done on 21.4% (22/103) of patients with persistent postoperative pain showed persistent oedema surrounding the implant (including the 2 patients who had a revision procedure).
One patient had an intraoperative metatarsal fracture during insertion of the implant, which
was treated at the time with open reduction and internal fixation.
Minor wound complications=2.9% (3/103)
Study 8 An T (2020)
Study type | Case series |
|---|---|
Country | US |
Recruitment period | 2016 to 2018 |
Study population and number | n=16 (18 implants) Patients who had MRI for persistent pain after synthetic cartilage insertion for hallux rigidus |
Age and sex | Mean 61 years; 81% (13/16) female |
Patient selection criteria | Inclusion criteria: patients who had synthetic cartilage implant insertion for hallux rigidus after nonoperative management had failed, and who had postoperative MRI for persistent pain or dysfunction at the metatarsophalangeal joint. Patients who had staged bilateral surgery were included in MRIs were available for both feet. Exclusion criteria: diagnosis other than hallux rigidus, hallux valgus angle greater than 20°, concomitant bunion correction, peripheral vascular disease, prior fusion procedures of the ankle, hindfoot, or midfoot, inflammatory disease, peripheral neuropathy. |
Technique | Implant: Cartiva Synthetic Cartilage Implant, Wright Medical Group, US. Of the 18 implants, 17 were 10 mm implants and 1 was an 8 mm implant. |
Follow-up | Mean 13 months for MRI imaging, mean 21 months for clinical follow-up data |
Conflict of interest/ source of funding | The authors received no financial support for the research, authorship and publication of the article. One author reported stock in Paragon 28, speaking for Stryker and Editor-in-Chief of Foot & Ankle International, outside the submitted work. One author reported personal fees from Extremity Medical, outside the submitted work. |
Analysis
Follow-up issues: There were 60 patients who met the inclusion and exclusion criteria. Of these, 16 patients had persistent symptoms at the surgical site that prompted advanced imaging and therefore qualified for study inclusion. Of the 16 patients, 14 had follow-up plain radiographs available for review (mean 13 months, range 2 to 25). Clinical follow-up data were also available for 14 of the 16 patients (16 out of 18 implants).
Study design issues: Retrospective single centre case series. The main aim of the study was to characterise radiological findings of the synthetic cartilage implant and surrounding tissues. Clinical outcomes included Patient Reported Outcome Measures Informational System (PROMIS) physical function and pain interference scores. Patient satisfaction and revision surgical procedures, including conversions to fusion, were collected by telephone interviews by a study member not involved with the synthetic cartilage implant insertion surgery. The single 8 mm implant was excluded from the MRI measurement analysis.
Key efficacy findings
Number of patients analysed: 16 (18 implants)
Mean PROMIS physical function score=41 (range 27 to 56) (reported as corresponding to moderate physical dysfunction)
Mean PROMIS pain interference score=63 (range 50 to 74) (reported as corresponding to a moderate level of pain interfering with daily activities)
Median satisfaction rating (5-point Likert scale)=2
Mean satisfaction rating (5-point Likert scale)=2.25
Reoperation within 2 years of implant insertion=38% (6/16):
Revision synthetic cartilage implant at 16 months
Lysis of adhesions and manipulation under anaesthesia at 6 months and lysis of adhesions and Moberg osteotomy at 12 months
Revision synthetic cartilage implant and bone grafting at 17 months
Conversion to fusion at 26 months
Revision synthetic cartilage implant and Moberg osteotomy at 12 months
Conversion to fusion at 10 months
Progression of osteoarthritis at the hallux MTP joint=100% (16/16)
Key safety findings
Mean medial joint space reduced from 2.3 mm immediately after the procedure to 0.4 mm at final follow-up (p<0.001).
Mean lateral joint space reduced from 2.1 mm immediately after the procedure to 0.6 mm at final follow-up (p<0.001).
78% (14/18) of implants had evidence of fluid around the implant and all implants had oedema in the bony proximal phalanx, metatarsal and soft tissues.
Of the 17 implants with an MRI, the mean implant diameter was 9.7 mm. There was a mismatch between the implant and the bony channel, which measured a mean of 11.2 mm in diameter. The mean implant height was 9.5 mm and the mean bony depth was 9.7 mm, so the implant had often subsided below the subchondral bone of the metatarsal head.
Study 9 Brandao B (2020b)
Study type | Case series |
|---|---|
Country | UK |
Recruitment period | Not reported |
Study population and number | n=55 Patients with symptomatic hallux rigidus |
Age and sex | Mean 56 years; 75% (41/55) female |
Patient selection criteria | Inclusion criteria included adult patients over 18 years old with symptomatic hallux rigidus who had a primary synthetic cartilage implant procedure. Patients with previous surgery to the 1st metatarsal or traumatic osteoarthritis were excluded. |
Technique | Device: Cartiva Synthetic Cartilage Implant Preparation of the metatarsal head included removal of dorsal osteophytes. The implant was inserted using press fit technique and was left at least 3 mm proud. Intraoperatively, the aim was to achieve 90° of dorsiflexion. |
Follow-up | Mean 21 months (range 12 to 38) |
Conflict of interest/ source of funding | None |
Analysis
Follow-up issues: No losses to follow-up were described.
Study design issues: Single centre prospective case series. First metatarsophalangeal joint arthritis was graded preoperatively according to the Hattrup and Johnson classification. The aim of the study was to analyse the efficacy of the procedure using patient reported outcome measures. Outcomes were evaluated using the FAAM ADL subscale and the MOXFQ.
Population issues: Of the 55 patients, 14 (25%) had grade 2 or moderate arthritis and 41 (75%) had grade 3 or severe arthritis at baseline.
Key efficacy findings
Number of patients analysed: 55
Mean FAAM ADL scores
Mean objective scores improved from 64% at baseline to 87% postoperatively (p<0.0001).
Subjective scores of functionality improved from 41% to 87%.
Mean MOXFQ scores
Domain | Baseline | Postoperative | p value |
MOXFQ Index | 58 | 24 | <0.0001 |
Walking or standing | 58 | 25 | <0.02 |
Pain | 66 | 25 | <0.02 |
Social interaction | 46 | 20 | <0.02 |
3 patients with psoriatic inflammatory arthritis and 1 patient with primary severe sesamoid osteoarthritis showed little improvement or deteriorated postoperatively.
Patient satisfaction=89.4%
The authors noted that there was some dissatisfaction among the patients who enjoyed wearing high heeled shoes, because they were unable to do so after the procedure.
Reoperation
There was 1 revision to another synthetic cartilage implant at 14 months postoperatively with use of calcaneal bone graft, as the implant had sunk into the metatarsal. The patient was a 30 year old white female ballet dancer with grade 2 osteoarthritis with evidence of osteopenia and a body mass index of 18.
There was 1 revision to arthrodesis at 17 months postoperatively in a 61 year old black male patient with grade 2 osteoarthritis. The patient had experienced increasing pain and reduced range of motion.
27.3% (15/55) of patients had manipulation under anaesthesia and steroid and local anaesthetic injection at 12 weeks postoperatively because of stiffness.
Key safety findings
There were no complications, including infection, wound breakdown or material failure.
Study 10 Harmer J (2020)
Study type | Case series |
|---|---|
Country | UK |
Recruitment period | 2016 to 2017 |
Study population and number | n=20 (17 first MTP joint, 2 second and 1 third) Patients with painful moderate to severe arthritic degeneration of a MTP joint |
Age and sex | Mean 51 years (range 35 to 72); 85% (17/20) female |
Patient selection criteria | Inclusion criteria: patients over the age of 18 and who had synthetic cartilage implant insertion for painful moderate to severe arthritic degeneration of a MTP joint. Patients with early arthritic degeneration with minimal cartilage loss or those who had not previously had conservative care or had marked transverse plane deformity were not offered the procedure. |
Technique | Implant: Cartiva Synthetic Cartilage Implant, Cartiva Inc., US. |
Follow-up | Mean 19 months (range 11 to 24) |
Conflict of interest/ source of funding | Not reported |
Analysis
Follow-up issues: 10% (2/20) of patients were lost to follow-up at 12 months. All patients had postoperative x-ray evaluation at 6 months.
Study design issues: Retrospective, single centre case series. All outcome data were collected using the PASCOM-10 audit database, an online resource for reporting clinical and patient reported outcomes. This includes a patient satisfaction questionnaire (PSQ-10). The Manchester Oxford Foot/Ankle Questionnaire (MOXFQ) was used for patient-reported outcomes.
Population issues: Of the 20 patients, 3 had arthritis in a lesser MTP joint. Of the 17 patients with first MTP joint arthritis, 8 (47%) had hallux rigidus stage 2, 8 (47%) had stage 3 and 1 (6%) had stage 4, according to the Coughlin and Shurnas classification system.
Other issues: the authors noted that they stopped doing the procedure because of cost and suboptimal results noted at early follow-up, which is why the sample size is so small.
Key efficacy findings
Number of patients analysed: 20
Summary of mean MOXFQ and PSQ-10 scores
Domain | Pre-op | 6 months post-op | Score change | 12 months post-op | Score change | Minimal clinical important difference |
Walking | 67 | 47 | 20 | 33 | 34 | 16 |
Pain | 80 | 45 | 35 | 32 | 48 | 12 |
Social | 60 | 33 | 27 | 20 | 40 | 24 |
PSQ-10 | 76 | 78 |
Further descriptive data from PSQ-10 questionnaires showed that at 6 months after the procedure, 65% of patients felt that their original complaint was now better or much better, while 4 (20%) patients felt their foot condition had deteriorated. At 12 months, 60% of patients felt better or much better and only 1 patient reported a deterioration in their foot condition.
20% of patients noticed an improvement in joint range of motion at 12 months, 80% had no improvement or a deterioration in joint range of motion.
At 6 months, 80% of patients felt that their original expectations had been met or partly met and 95% reported that they would be prepared to have the same surgery again. This reduced to 75% and 80% respectively by 12 months.
x-ray evaluation at 6 months typically showed marked narrowing of the joint space, proximal impaction of the synthetic cartilage implant into the head of the metatarsal and there was significant arthritic involvement of the sesamoid apparatus.
Revision at 6 months for implant failure=15% (3/20) (1 first MTP joint arthrodesis, 1 first MTP joint primus implant, 1 second MTP joint interplex Rod)
Revision at 12 months for implant failure=10% (2/20) (both first MTP joint arthrodesis)
Total revision rate=25%
Key safety findings
Complications at 6 months
Joint pain and stiffness=60% (12/20) (treated by manipulation under anaesthesia with an intra-articular corticosteroid injection)
Swelling=10% (2/20)
Transfer metatarsalgia=5% (1/20)
Complications at 12 months
Joint restriction=20% (4/20)
Joint pain and stiffness=25% (5/20)
Study 11 Metikala S (2020)
Study type | Case series - US Food and Drug Administration's (FDA) Manufacturer and User Facility Device Experience (MAUDE) database |
|---|---|
Country | US |
Recruitment period | 2016 to 2018 |
Study population and number | n=49 adverse events Reports of adverse events associated with the Cartiva device |
Age and sex | Not reported |
Patient selection criteria | The MAUDE database was retrospectively investigated to identify the adverse reports of Cartiva, which was registered with the product code: PNW (Prosthesis, Metatarsophalangeal Joint Cartilage Replacement Implant). Duplicate reports were excluded. |
Technique | Device: Cartiva or Synthetic Cartilage Implant (Wright Medical). The most common implant size was 10 mm (19 reports) followed by 8 mm in 2 but unknown in the remaining 28 events. |
Follow-up | Not reported |
Conflict of interest/source of funding | One of the 6 authors reported personal fees from Wright Medical, Stryker, and Kinos, outside the submitted work. The remaining authors declared no conflict of interest. |
Analysis
Follow-up issues: Follow-up period is not routinely reported on the MAUDE database.
Study design issues: The US FDA MAUDE database was used to review voluntary reported adverse event reports associated with synthetic cartilage implant insertion. The MAUDE database does not report the total incidence of implant insertion.
Other issues: Although the total number of procedures is unknown, the authors noted that nearly 22,400 devices (unpublished company data) have been implanted during the 3 years of the study period.
Key efficacy findings
No efficacy data were reported.
Key safety findings
Adverse reports (n=49)
35 events mentioned further surgery at a mean interval of 4.75 months.
Subsidence, n=16 (reported terms: subsided device, sunk prosthesis, recessed implant, receded with bone-on-bone); 6 were converted to first MTP joint fusion, 1 implant removal with bone grafting and interpositional arthroplasty, 1 resection of prominent bone edges with reimplantation of the same device, 2 revisions using a different arthroplasty device, 1 implant removal only. No information was available for the remaining 5 reports of implant subsidence.
Fragmentation, n=9 (reported terms: broken device, torn implant, fragmented device); 3 implants were removed, there were 2 revisions to an unspecified implant, 1 implant was removed and replaced with another Cartiva implant, 1 was converted to first MTP joint fusion, the fragmented portion was replaced and the same device reimplanted in 1 and the outcome was unknown for 1.
Erosion, n=3 (reported terms: eroded bone, fractured metatarsal); 2 were treated by bone grafting, 1 fracture was converted to an unspecified alternate procedure.
Infection, n=4; 1 was managed by oral antibiotics and 1 was treated by removal of the implant. The remaining 2 reports of infection were salvaged by implant removal with the placement of an antibiotic spacer, but no details were available on further care.
Foreign body reaction, n=1
Unspecified, n=16 (reports stated that postoperative symptoms occurred but with no additional detail on the status of the implant or bone. The most common symptoms was persistent pain. Other symptoms included lack of motion, deformity, postoperative discomfort, redness, skin sloughing, drainage, and chronic regional pain syndrome.)
Summary of secondary procedures cited in 35 of 49 reports
Secondary procedure | Number of reports | Mean interval (months) |
Conversion to first MTP joint fusion | 10 | 5.5 |
Removal of Cartiva implant | 7 | 4 |
Revision to another Cartiva implant | 4 | 1.5 |
Revision to an alternate implant | 5 | 8 |
Revision to an alternate procedure | 2 | Unspecified |
Debridement and reimplantation of same Cartiva | 2 | Unspecified |
Cartiva removal and bone grafting | 2 | 7 |
Cartiva removal and antibiotic spacer | 2 | 2.5 |
Bone grafting and interpositional arthroplasty | 1 | Unspecified |
Study 12 Brandao B (2020c)
Study type | Non-randomised comparative study |
|---|---|
Country | UK (single centre) |
Recruitment period | Not reported |
Study population and number | n=78 (55 implant insertion, 23 cheilectomy) Adults with symptomatic hallux rigidus |
Age and sex | Mean age: 56 years (implant), 58 years (cheilectomy); 77% (60/78) female |
Patient selection criteria | Adult patients over 18 years of age with symptomatic hallux rigidus who had primary synthetic cartilage implant interpositional arthroplasty or cheilectomy surgery were included. Patients with previous surgery to the first metatarsal or hallux valgus with underlying arthritis were excluded. |
Technique | Implant: Cartiva Synthetic Cartilage Implant, Cartiva Inc., US. |
Follow-up | Mean 21 months (range 12 to 38) for implant group; mean 69 months (range 50 to 90) for arthrodesis group |
Conflict of interest/ source of funding | None |
Analysis
Follow-up issues: Follow-up times varied between the 2 groups.
Study design issues: Prospective, single centre non-randomised comparative study. Patients were assigned to either group based on patient and surgeon preference. The 2 groups were operated on by the same 3 surgeons at the same centre. The aim was to analyse the efficacy of the synthetic cartilage interpositional arthroplasty compared with cheilectomy in the treatment of hallux rigidus using patient reported outcome measures. The MOXFQ was used to evaluate patient reported outcomes and the FAAM sports subscale was used to assess sporting ability outcomes at a minimum of 1 year after the procedure. No preoperative scores were taken so the relative improvements after surgery could not be compared. In both groups, a small percentage of patients did not routinely participate in sporting activities and were therefore excluded from the FAAM sports assessment.
Study population issues: In the implant group, 14 (25%) patients had Hattrup and Johnson grade 2 (moderate) arthritis and 41 (75%) patients had grade 3 (severe) arthritis. In the cheilectomy group, 5 (22%) patients had grade 1 (mild), 10 (44%) had grade 2 (moderate) and 8 (13%) patients had grade 3 (severe) arthritis.
Key efficacy findings
Number of patients analysed: 78 (55 implant, 23 cheilectomy)
Procedure | Pain | Walking or standing | Social interaction | MOXFQ index |
|---|---|---|---|---|
Synthetic cartilage implant | 27.52 | 28.76 | 23.88 | 27.00 |
Cheilectomy | 27.17 | 6.83 | 8.15 | 14.13 |
p value | 0.6818 | 0.00054 | 0.00308 | 0.01242 |
Mean postoperative FAAM Sports scores
Implant (n=30) = 74.9%
Cheilectomy (n=19) = 82.7%, p=0.11
Recovery and return to function times were comparable with no statistically significant difference.
Patient satisfaction with both techniques was high.
Key safety findings
There were no perioperative complications. None of the patients needed revision surgery.
Study 13 Reddy S (2021)
Study type | Case report |
|---|---|
Country | US |
Recruitment period | Not reported |
Study population and number | n=1 A patient who had early catastrophic wear of a synthetic cartilage implant |
Age and sex | 68 year old woman |
Patient selection criteria | Not applicable |
Technique | Bilateral 10-mm synthetic cartilage implant insertion. |
Follow-up | 2 years |
Conflict of interest/ source of funding | None |
Key safety findings
The patient presented with bilateral grade 3 hallux rigidus. Her past medical history included hypertension and scleroderma. She had tried conservative measures, including shoewear modifications, nonsteroidal medication, and cortisone injections. She wished to maintain joint mobility and was reluctant to have an arthrodesis. She chose to have a synthetic cartilage implant inserted in both feet concurrently.
Her initial postoperative course was unremarkable, except for a left second metatarsal shaft fracture sustained when she inadvertently stepped on a piece of wood. Radiographs taken at this time demonstrated joint space loss within both hallux MTP articulations. She began to notice increasing pain at her 1-year follow up visit and had intra-articular cortisone injections bilaterally, with limited relief. She tried to manage her symptoms conservatively but had progressive difficulty with activity and with shoewear, as well as increasing stiffness of both joints. Two year follow up radiographs demonstrated progressive deterioration of the left hallux MTP joint with shortening of the hallux and an elevatus deformity. The patient chose to have bilateral hallux MTP arthrodesis. Evaluation of the left hallux demonstrated severe erosive wear of the bearing surface of the synthetic cartilage implant, with bone loss of the medial aspect of the base of the proximal phalanx, erosion of the subchondral plate, and exposure of underlying trabecular bone. The implant was noted to be well fixed. There were no clinical signs of infection. A hallux MTP arthrodesis using a lengthening tricortical allograft and iliac crest bone marrow aspirate concentrate was done to restore length. The implant was noted to have subsided on the right without wear of the implant. Both sides had healed at the 3‑month follow-up, with improved overall pain and function.
The authors noted that the reason for the catastrophic wear observed in this case is unclear, though likely related to the bone loss observed within the base of the proximal phalanx resulting from osteolysis.
Study 14 Joo P (2021)
Study type | Non-randomised comparative study |
|---|---|
Country | US |
Recruitment period | 2015 to 2019 |
Study population and number | n=181 (59 implant insertion, 122 arthrodesis) Patients with advanced hallux rigidus |
Age and sex | Mean age: 57 years (implant), 61 years (arthrodesis), p<0.01; 77% (140/181) female |
Patient selection criteria | Both treatment options were offered to patients after August 2016, when the implant became available for use in the US. Synthetic cartilage implant insertion was generally not recommended for patients who had inflammatory arthropathy (gout or rheumatoid arthritis), severely limited range of motion (dorsiflexion less than 20 degrees), severe osteoporosis, presence of a large cyst, avascular necrosis, malalignment as with a bunion deformity, or peripheral neuropathy. As the decision was mutually made, patients who wanted to retain the great toe motion preferred the implant over arthrodesis. Patients who were unwilling to take the potential risk of implant failure elected to have arthrodesis. Patients who had concomitant multidigit hammertoe or hallux valgus correction, and history of extensive foot or ankle surgery were excluded. |
Technique | Implant: Cartiva Synthetic Cartilage Implant, Cartiva Inc., US. Among the patients who implant insertion, 12 (20%) had a concurrent Moberg osteotomy. 3 (5%) of the implant insertions and 46 (38%) of the arthrodesis procedures were done with other minor procedures of the lesser toes, including up to 2 concurrent hammertoe corrections, hardware removal, and calcaneal bone graft harvest. |
Follow-up | Overall mean time from surgery to final follow-up was 33 months (range 14 to 59). Mean follow up for implant group: 27 months (range 14 to 40) Mean follow up for arthrodesis group: 38 months (15 to 59), p<0.01 |
Conflict of interest/ source of funding | One author reported personal fees from Stryker, outside the submitted work. |
Analysis
Follow-up issues: Of the 181 patients called for final follow-up, 101 (56%) completed the final phone survey and were included for secondary postoperative analysis. The response rate in the implant group was 50% compared with 68% in the arthrodesis group.
Study design issues: Retrospective, non-randomised comparative study. Electronic medical charts of all patients who met the criteria were reviewed to identify operative complications and other concomitant procedures. Patient data was included if a preoperative PROMIS t score and at least 1 postoperative t score were available, and the patient was at least 12 months from surgery. The patients were then contacted by telephone to obtain final PROMIS Physical Function and Pain Interference t scores. Callers had training and used standardised scripts to avoid bias. All data obtained were secured directly with patients. Three calls were attempted, and only patients who answered within the 3 attempts were included in this study.
Study population issues: At baseline, PROMIS Physical Function t scores were statistically significantly higher in the implant cohort compared to the arthrodesis cohort, with average t scores of 47.1 and 43.9, respectively (p=0.01). Baseline PROMIS Pain Interference t scores were similar between groups, with an average of 55.6 in the implant cohort and 57.4 in the arthrodesis cohort (p=0.07).
Other issues:
Key efficacy findings
Number of patients analysed: 181 (59 synthetic cartilage implant insertion, 122 arthrodesis)
Synthetic cartilage implant insertion | Arthrodesis | ||||||
|---|---|---|---|---|---|---|---|
Follow up | Score | n | 95% CI | Score | n | 95% CI | p |
Preoperative | 47.1 | 59 | 45.1 to 49.1 | 43.9 | 122 | 42.5 to 45.3 | <0.01 |
0 to 15 days | 36.6 | 42 | 34.2 to 38.8 | 31.0 | 84 | 29.4 to 32.6 | <0.01 |
1 month | 40.6 | 30 | 38.0 to 43.2 | 35.2 | 62 | 33.4 to 37.0 | <0.01 |
3 months | 45.8 | 27 | 43.0 to 48.6 | 43.5 | 58 | 41.6 to 45.4 | 0.17 |
6 months | 49.6 | 9 | 44.6 to 54.7 | 42.0 | 26 | 39.1 to 44.9 | <0.01 |
Final follow up | 51.4 | 40 | 48.8 to 53.9 | 45.9 | 60 | 43.9 to 47.9 | <0.01 |
Synthetic cartilage implant insertion | Arthrodesis | ||||||
|---|---|---|---|---|---|---|---|
Follow up | Score | n | 95% CI | Score | n | 95% CI | p |
Preoperative | 55.6 | 59 | 53.6 to 57.6 | 57.4 | 122 | 56.0 to 58.8 | 0.07 |
0 to 15 days | 62.5 | 42 | 60.3 to 64.8 | 63.9 | 84 | 62.3 to 65.5 | 0.33 |
1 month | 56.2 | 30 | 53.7 to 58.8 | 57.4 | 62 | 55.6 to 59.1 | 0.49 |
3 months | 54.5 | 27 | 51.7 to 57.3 | 54.6 | 58 | 52.7 to 56.5 | 0.94 |
6 months | 49.5 | 9 | 44.4 to 54.6 | 56.5 | 26 | 53.6 to 59.3 | 0.02 |
Final follow up | 49.4 | 40 | 46.9 to 51.9 | 48.2 | 61 | 45.9 to 50.5 | 0.49 |
Key safety findings
Proportion of patients with 'significant' pain at follow up
Implant=10.0% (4/40)
Arthrodesis=8.2% (5/61), p=0.76
Complications
Implant=3.4% (2/59); both patients had the implant removed and conversion to arthrodesis after 12 and 21 months.
Arthrodesis=2.5% (3/122); there were 2 hardware failures with revision at 12 and 36 months after initial surgery and 1 hardware removal because of pain 22 months after surgery.
Study 15 Lee W (2021)
Study type | Case series |
|---|---|
Country | US |
Recruitment period | 2017 to 2019 |
Study population and number | n=90 (96 implants) Patients with hallux rigidus |
Age and sex | Mean 54.4 years (range 27 to 74); 88% (84/96) female |
Patient selection criteria | Inclusion criteria: at least 18 years old, diagnosed with hallux rigidus, had synthetic cartilage implant insertion procedure, and at least 1 year out from surgery at the time of study initiation. Patients who had a previous synthetic cartilage implant insertion procedure for hallux rigidus or polyvinyl alcohol hydrogel implantation for lesser MTP joints were excluded from the study. |
Technique | Implant: Cartiva Synthetic Cartilage Implant, Wright Medical, US. |
Follow-up | Mean 26.4 months (range 12 to 39) |
Conflict of interest/ source of funding | One author reported being a consultant for Wright Medical, outside the submitted work. |
Analysis
Follow-up issues: An additional 2 patients (4 implants) declined to join the study, and 24 patients (24 implants) could not be reached for the survey.
Study design issues: Single-centre, retrospective case series. Charts were reviewed for perioperative patient data. A questionnaire was designed to evaluate patient satisfaction, self-reported clinical improvement, and changes in sporting ability after the procedure. The questionnaire was administered by telephone survey by an author who was not involved in perioperative patient care. There may be some recall bias because patients were asked to evaluate their previous clinical status as well as their status at present. Preoperative and postoperative pain VAS score and PROMIS‑10 scores were also collected for each patient.
Population issues: Mean body mass index at baseline was 25.1 kg/m2 (range 19.4 to 37.6). Review of preoperative foot radiographs identified 14 cases of grade 1 (15.9%) hallux rigidus, 33 cases of grade 2 (34.4%), 33 cases of grade 3 (34.4%), and 8 cases of grade 4 (8.3%). The mean hallux valgus angle was 11.9 degrees (range 1.2 to 24.1).
Key efficacy findings
Number of patients analysed: 90 (96 implants)
At final follow-up, the mean PROMIS-10 physical health T-score was 54.2 ± 8.3 points (range 26.7 to 67.6), which corresponds to a label of "very good" and the mean PROMIS-10 mental health T-score was 57.4 ± 7.8 (range 31.3 to 67.6), corresponding to a label of "excellent."
Mean VAS pain scores
Baseline=7.9 ± 1.8 (range 2 to 10)
Postoperative=1.5 ± 2.5 (range 0 to 10), p<0.001
Range of motion, n (%)
Self-reported clinical improvement, n (%)
Much improved=55.2% (53/96)
Improved=26.0% (25/96)
Same=8.3% (8/96)
Worse=7.3% (7/96)
Much worse=3.1% (3/96)
Satisfaction, n (%)
Very satisfied=41.7% (40/96)
Satisfied=32.3% (31/96)
Neutral=8.3% (8/96)
Unsatisfied=8.3% (8/96)
Very unsatisfied=9.4% (9/96)
Time to new normal, n (%)
0 to 3 months=25.0% (24/96)
3 to 6 months=17.7% (17/96)
6 to 12 months=34.4% (33/96)
More than 12 months=10.4% (10/96)
Never=10.4% (10/96)
Not sure=2.1% (2/96)
Intention to have the same surgery again under the same circumstances, n (%)
Yes=75.0% (72/96)
No=22.9% (22/96)
Unsure=3.1% (3/96)
Would recommend the same surgery to family and friends, n (%)
Yes=69.8% (67/96)
No=17.7% (17/96)
Unsure=12.5% (12/96)
Statistically significant differences between the satisfied and unsatisfied subgroups were found in preoperative corticosteroid injection use (21.1% compared with 41.1%, p=0.029) and preoperative VAS pain score (8.2 compared with 7.1, p=0.036).
Sports activity level | Preoperative | Postoperative |
|---|---|---|
None | 7/96 (7.3%) | 3/96 (3.1%) |
Low impact | 57/96 (59.4%) | 55/96 (57.3%) |
Mid impact | 17/96 (17.7%) | 18/96 (18.8%) |
High impact | 15/96 (15.6%) | 20/96 (20.8%) |
Key safety findings
Postoperative complications and reoperations
Revision to arthrodesis=2.1% (2/96); at 13 months and 23 months after index procedure.
Reoperation other than revision=2.1% (2/96); both were tibial sesamoidectomy
Wound dehiscence=2.1% (2/96); managed with dressing changes only
Validity and generalisability of the studies
One randomised controlled trial (RCT) was identified, which included data from the UK.
One of the case series reported longer term follow-up of patients who were originally included in the RCT (Glazebrook, 2019).
Different systems were used to grade the degree of osteoarthritis. The RCT included patients with grade 2, 3 and 4 osteoarthritis according to the Coughlin and Shurnas system.
In some studies, patients had other procedures done on the first MTP joint at the same time as the implant insertion.
Some of the adverse events reported in the review of the FDA MAUDE database may also be included in the published literature.
The longest mean follow-up was 5.8 years.
How are you taking part in this consultation?
You will not be able to change how you comment later.
You must be signed in to answer questions