Health Technology Assessment (Mar 2023)
Total ankle replacement versus ankle arthrodesis for patients aged 50–85 years with end-stage ankle osteoarthritis: the TARVA RCT
Abstract
Background We aimed to compare the clinical effectiveness, cost-effectiveness and complication rates of total ankle replacement with those of arthrodesis (i.e. ankle fusion) in the treatment of end-stage ankle osteoarthritis. Methods This was a pragmatic, multicentre, parallel-group, non-blinded randomised controlled trial. Patients with end-stage ankle osteoarthritis who were aged 50–85 years and were suitable for both procedures were recruited from 17 UK hospitals and randomised using minimisation. The primary outcome was the change in the Manchester–Oxford Foot Questionnaire walking/standing domain scores between the preoperative baseline and 52 weeks post surgery. Results Between March 2015 and January 2019, 303 participants were randomised using a minimisation algorithm: 152 to total ankle replacement and 151 to ankle fusion. At 52 weeks, the mean (standard deviation) Manchester–Oxford Foot Questionnaire walking/standing domain score was 31.4 (30.4) in the total ankle replacement arm (n = 136) and 36.8 (30.6) in the ankle fusion arm (n = 140); the adjusted difference in the change was –5.6 (95% confidence interval –12.5 to 1.4; p = 0.12) in the intention-to-treat analysis. By week 52, one patient in the total ankle replacement arm required revision. Rates of wound-healing issues (13.4% vs. 5.7%) and nerve injuries (4.2% vs. < 1%) were higher and the rate of thromboembolic events was lower (2.9% vs. 4.9%) in the total ankle replacement arm than in the ankle fusion arm. The bone non-union rate (based on plain radiographs) in the ankle fusion arm was 12.1%, but only 7.1% of patients had symptoms. A post hoc analysis of fixed-bearing total ankle replacement showed a statistically significant improvement over ankle fusion in Manchester–Oxford Foot Questionnaire walking/standing domain score (–11.1, 95% confidence interval –19.3 to –2.9; p = 0.008). We estimate a 69% likelihood that total ankle replacement is cost-effective compared with ankle fusion at the National Institute for Health and Care Excellence’s cost-effectiveness threshold of £20,000 per quality-adjusted life-year gained over the patient’s lifetime. Limitations This initial report contains only 52-week data, which must therefore be interpreted with caution. In addition, the pragmatic nature of the study means that there was heterogeneity between surgical implants and techniques. The trial was run across 17 NHS centres to ensure that decision-making streams reflected the standard of care in the NHS as closely as possible. Conclusions Both total ankle replacement and ankle fusion improved patients’ quality of life at 1 year, and both appear to be safe. When total ankle replacement was compared with ankle fusion overall, we were unable to show a statistically significant difference between the two arms in terms of our primary outcome measure. The total ankle replacement versus ankle arthrodesis (TARVA) trial is inconclusive in terms of superiority of total ankle replacement, as the 95% confidence interval for the adjusted treatment effect includes both a difference of zero and the minimal important difference of 12, but it can rule out the superiority of ankle fusion. A post hoc analysis comparing fixed-bearing total ankle replacement with ankle fusion showed a statistically significant improvement of total ankle replacement over ankle fusion in Manchester–Oxford Foot Questionnaire walking/standing domain score. Total ankle replacement appears to be cost-effective compared with ankle fusion at the National Institute for Health and Care Excellence’s cost-effectiveness threshold of £20,000 per quality-adjusted life-year gained over a patient’s lifetime based on long-term economic modelling. Future work We recommend long-term follow-up of this important cohort, in particular radiological and clinical progress. We also recommend studies to explore the sensitivity of clinical scores to detect clinically important differences between arms when both have already achieved a significant improvement from baseline. Trial registration This trial is registered as ISRCTN60672307 and ClinicalTrials.gov NCT02128555. Funding This project was funded by the National Institute for Health and Care Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 27, No. 5. See the NIHR Journals Library website for further project information. Plain language summary Each year, over 29,000 patients with ankle osteoarthritis seek a specialist opinion, of whom 4000 undergo NHS surgical treatment. The main surgical treatments for severe ankle osteoarthritis are total ankle replacement or arthrodesis (i.e. ankle fusion). Both are known to be good treatments to relieve pain, and each has its advantages. Total ankle replacement is a more popular patient choice than ankle fusion. When deciding whether to undergo ankle replacement or fusion, patients consult various sources, but the majority of them rely on the advice of their surgeon to make a final decision. To the best of our knowledge, there has never been a high-quality randomised clinical trial comparing these two treatments and there are no published guidelines on the most suitable management. In this study, 303 patients were randomised to a type of ankle surgery: 138 in the total ankle replacement arm and 144 in the ankle fusion arm received surgery. We found that both total ankle replacement and ankle fusion improved patients’ walking ability, but we did not find a statistically significant difference between the treatment arms based on our primary outcome measure at 1 year. When we considered the type of total ankle replacement implant, we found that the implant most commonly used in the NHS (a fixed-bearing two-component implant) had better outcomes at 1 year than ankle fusion. Both total ankle replacement and ankle fusion appear to be safe. However, there were more wound-healing issues and nerve injuries in the total ankle replacement arm than in the ankle fusion arm. Twelve per cent of patients experienced bone non-union in the ankle fusion arm, but only 7.1% experienced symptoms. We estimate that there is a 69% chance that total ankle replacement would be cost-effective compared with ankle fusion at the National Institute for Health and Care Excellence’s cost-effectiveness threshold of £20,000 per quality-adjusted life-year gained over a patient’s lifetime. This study provides the NHS with important information that could help to obtain the best possible outcome for patients with severe ankle arthritis. Scientific summary Background More than 29,000 patients in the UK present to specialists each year with symptomatic end-stage ankle osteoarthritis, a condition in which the cartilage lining the ankle joint has worn away, causing progressive pain and stiffness. Ankle osteoarthritis causes major disability and has a similar impact on quality of life (QoL) as end-stage hip osteoarthritis and cardiac failure. The demand incidence for ankle osteoarthritis has been estimated at 47.7 per 100,000. The majority of this is secondary to trauma caused by fractures or severe sprains, both of which are increasing; hence, ankle osteoarthritis is likely to become an increasingly important health problem, especially among working adults. Other causes of ankle osteoarthritis include long-standing inflammatory arthropathies (e.g. rheumatoid arthritis, haemochromatosis and haemophiliac arthropathy). In the early stages of disease, non-operative measures such as a change in activity levels, weight loss, physiotherapy, painkillers and ankle braces should be used. When these conservative measures have failed, and a surgeon confirms the diagnosis of end-stage osteoarthritis on the basis of radiological and clinical evidence (i.e. plain radiographs and unrelenting symptoms, respectively), surgery might then be considered. Although arthrodesis (i.e. ankle fusion) is the most common surgical treatment for end-stage ankle osteoarthritis, in response to patient demand, more and more surgeons are performing total ankle replacement (TAR). At least 4000 patients are treated with ankle fusion or TAR each year in the NHS. The TARs implanted in England, Wales, Northern Ireland, the Isle of Man and Guernsey are captured on the National Joint Registry, which has revision surgery as its end point. The British Orthopaedic Foot & Ankle Society only recently started capturing outcome data on ankle fusion patients. All studies comparing TAR with ankle fusion to date have been observational and, to the best of our knowledge, there have been no high-quality prospective randomised trials reported. Objectives The total ankle replacement versus ankle arthrodesis (TARVA) trial aimed to compare the clinical effectiveness and cost-effectiveness of TAR with that of ankle fusion for the treatment of end-stage ankle osteoarthritis in patients aged 50–85 years. Clinical effectiveness was measured through self-reported pain-free function using a standardised questionnaire of walking and standing ability 52 weeks post surgery. We also aimed to determine whether or not there was a difference in physical function [measured using the Foot and Ankle Ability Measure – Activities of Daily Living (FAAM-ADL)], QoL [measured using the EuroQol 5-Dimensions, five-level version (EQ-5D-5L)] and range of ankle motion at 26 and 52 weeks post surgery. We investigated the cost-effectiveness and cost–utility of TAR and ankle fusion. Methods Design We conducted a pragmatic prospective, multicentre, parallel-group, non-blinded randomised controlled trial (RCT). Participants were randomised equally between two arms: TAR and ankle fusion. The study protocol [Goldberg AJ, Zaidi R, Thomson C, Doré CJ, Skene SS, Cro S, et al. Total ankle replacement versus arthrodesis (TARVA): protocol for a multicentre randomised controlled trial. BMJ Open 2016;6:e012716] was developed before recruitment commenced and detailed the design, interventions and study procedures. Setting and participants The trial was conducted across 17 participating UK sites. Patients with end-stage ankle osteoarthritis who were aged 50–85 years and who the surgeon believed were suitable for both TAR and ankle fusion were eligible to join the trial. Patients had to be able to read and understand the patient information sheet (PIS) and provide written consent on an informed consent form (ICF). Interventions and follow-up At randomisation patients were allocated to receive either TAR or ankle fusion. For TAR, the remaining damaged cartilage was removed and the joints resurfaced with metal implants and an intervening polyethylene liner, either fixed or mobile bearing, to act as a gliding surface. All prostheses were CE marked. For ankle fusion, the remaining damaged cartilage was removed from the ends of the bone and the two bones held together in compression using screws or plates to join them as one bone (bone fusion) so that there was no longer any movement at the tibiotalar joint. Participants were seen at recruitment, randomisation, surgery visit and at 2, 6, 12, 26 and 52 weeks post surgery. Blinding This was an open (non-blinded) trial. It was not possible to blind patients, surgeons, radiologists or clinical assessors. Randomisation Participants were randomised in a 1 : 1 ratio to either TAR or ankle fusion. Randomisation was carried out using minimisation incorporating a random element, with surgeon and whether osteoarthritis was present in the subtalar or the talonavicular joint as minimisation factors. A secure online service (Sealed Envelope™; Sealed Envelope Ltd, London, UK) provided the treatment arm allocations. Outcome measures The primary outcome was the change in the Manchester–Oxford Foot Questionnaire (MOXFQ) walking/standing domain scores (0–100, where low scores are better) between the preoperative baseline and 52 weeks post surgery. The secondary outcomes were change in MOXFQ walking/standing domain score from preoperative baseline to 26 weeks and change in MOXFQ pain and social interaction domain scores from baseline to 26 and 52 weeks. An additional measure of physical function, the Foot and Ankle Ability Measure (FAAM), was captured at baseline and at 26 and 52 weeks. The changes in FAAM-ADL (0–100, where higher scores are better) and FAAM sport subscale scores from baseline were compared between arms. We also compared changes in QoL from baseline to 12, 26 and 52 weeks using the EQ-5D-5L questionnaire. Longer-term follow-up at 2, 5 and 10 years post surgery is planned. Total range of motion (ROM) of the tibia to the floor was captured at baseline and 52 weeks. All adverse events (AEs), serious adverse events (SAEs) and complications reported from the date of consent until 52 weeks were compared between arms. Secondary outcomes that related to the economic evaluation included quality-adjusted life-years (QALYs), health-care resource use collected from patient files and a modified version of the Client Service Receipt Inventory (CSRI), and mean incremental cost per QALY gained. Sample size The sample size calculation for the primary outcome (change in MOXFQ walking/standing domain score by 52 weeks) was performed using Stata/IC®, version 12.1 (StataCorp LP, College Station, TX, USA). It was based on achieving 90% power to detect an estimated minimal important difference (MID) in the primary outcome at the 5% level of significance, accounting for expected loss to follow-up. The sample size calculation was based on previous observational studies and determined it was important to detect a difference of 12 in the change from baseline in the MOXFQ walking/standing domain score between the two treatment arms. The standard deviation (SD) of the MOXFQ walking/standing domain score was estimated to be 27, and loss to follow-up was estimated to be 10% (attrition in similar RCTs has been 5–7%). Based on these quantities, the required sample size was estimated to be 118 patients per arm. The sample size was adjusted to account for clustering by surgeon. The intraclass correlation coefficient (ICC) was estimated based on previous studies, and the initially computed sample size was inflated by a factor of f = 1 + (m – 1) × ICC. Assuming an average cluster size (m) of 14 (patients per surgeon) and an ICC of 0.03, an inflation factor of f = 1.39 was estimated, leading to a final required sample size of 164 per arm or 328 patients in total. Data collection and management Data were entered into a central MACRO v4 database (Elsevier, Amsterdam, the Netherlands) by sites, with internal validation checks to improve data quality; data queries were resolved by site staff before database lock and final analysis. Statistical methods As per the statistical analysis plan, all the analyses were conducted on an intention-to-treat (ITT) basis, meaning that all randomised participants with at least one postsurgery follow-up visit were included in the analysis, regardless of their adherence to treatment. In addition, a per-protocol analysis was performed for the primary outcome that included outcome data from only those patients who received their randomised surgical procedure within the time window specified in the protocol. The primary analysis involved fitting a multilevel repeated-measures linear regression model to estimate the difference between treatment arms in the change in MOXFQ walking/standing domain score from baseline to 52 weeks. This analysis model used all available visit data (from 26 weeks and 52 weeks) to strengthen confidence in the missing at random assumption and provide greater power to detect differences at individual visits. The model was adjusted for baseline MOXFQ walking/standing domain score and presence of osteoarthritis in each of the two adjacent joints. A random surgeon effect was also included in the model to account for clustering by surgeon. Similar models were used for other continuous secondary outcomes to estimate differences at 26 and 52 weeks post surgery. Economic evaluation The aim of the economic evaluation was to assess the cost-effectiveness of TAR compared with ankle fusion in patients with end-stage osteoarthritis. We compared the costs and outcomes of the two arms over the time horizon of 52 weeks. Outcomes were QALYs, calculated using utility index values obtained from the EQ-5D-5L. The primary within-trial analysis was conducted according to the ITT principle from the NHS and personal social services (PSS) perspective. Costs included cost of surgery, cost of health-care resource use (collected using the CSRI) and cost of concomitant medications. Sensitivity analyses included per-protocol analysis and analysis from a societal perspective. The societal perspective included additional out-of-pocket costs incurred by the participants and any productivity loss. The analytical approach is a cost–utility analysis as it estimates the mean incremental cost per QALY gained of TAR compared with ankle fusion. We built a decision model to extrapolate the trial results to a lifetime horizon. We constructed a simple Markov model, which simulates participants’ pathways after TAR or ankle fusion. Monte Carlo simulations were used to account for uncertainty. We estimated the probability of the intervention being cost-effective at the cost-effectiveness thresholds of £20,000–30,000 per QALY gained, recommended by the National Institute for Health and Care Excellence (NICE). Results Baseline characteristics Between March 2015 and January 2019, 303 participants were randomised; 282 participants had surgery and 281 attended at least one follow-up visit. The mean age was 68 years, 71% of participants were male and 43% had arthritis in one or more adjacent joints. The arms were well balanced at baseline, as observed from the baseline characteristics. Primary outcome The mean (SD) MOXFQ walking/standing domain score at 52 weeks was 31.4 (30.4) in the TAR arm and 36.8 (30.6) in the ankle fusion arm. Patients improved in both arms, but the adjusted mean [95% confidence interval (CI)] difference of –5.56 (–12.49 to 1.37) suggests that, on average, patients who received TAR had a MOXFQ walking/standing score 5.56 points lower than those who received ankle fusion at 52 weeks post surgery. This difference was not statistically significant (p = 0.12). The 95% CI included the MID of –12, so the trial was not able to exclude the MID. After 52 weeks, more patients achieved the MID in the TAR arm (82%) than in the ankle fusion arm (80%). In a post hoc analysis, when each of the two TAR subtypes (fixed- and mobile-bearing implants) were compared with ankle fusion, the mean (SD) MOXFQ walking/standing domain score at 52 weeks was 25.9 (28.3) in the fixed-bearing TAR arm and 36.8 (30.6) in the ankle fusion arm. The adjusted difference of –11.1 (95% CI –19.3 to –2.9) suggests that, on average, patients who received a fixed-bearing TAR had a MOXFQ walking/standing score 11.1 points lower than those who received ankle fusion at 52 weeks post surgery. This difference was statistically significant (p = 0.008). Secondary outcomes The MOXFQ pain and social interaction domain scores also suggested improvement in patients in both arms, but the adjusted difference of –4.20 (95% CI –9.80 to 1.39) for pain and –5.06 (95% CI –10.37 to 0.26) for social interaction at 52 weeks post surgery were not statistically significant (p = 0.14 and p = 0.06, respectively). The difference between the TAR and ankle fusion arms in the change in MOXFQ walking/standing domain score at 26 weeks was statistically significant (p = 0.02). The difference between the TAR and ankle fusion arms in the change in FAAM-ADL scores at 52 weeks was statistically significant (p = 0.01). There were improvements from baseline in both arms, but a difference of 6.16 (95% CI 1.54 to 10.78) between arms. The change in the EQ-5D-5L visual analogue scale value was statistically significant at 26 weeks (p = 0.03), but the change in the EQ-5D-5L index value was not significantly different at 26 weeks (p = 0.08) and 52 weeks (p = 0.32) between the two treatment arms. At 52 weeks from baseline, the range of motion (dorsiflexion and plantarflexion) improved for patients with TAR and decreased for those with ankle fusion; the difference between arms was statistically significant (p < 0.001). One or more SAE occurred in 17.8% of TAR and 23.8% of ankle fusion patients (p = 0.19). One or more AE occurred in 54.3% of TAR and 52.6% of ankle fusion patients (p = 0.84). The risks of patients experiencing any SAE or AE during the course of the trial were not statistically significantly different between the two arms. Economic evaluation Total ankle replacement generated more QALYs than ankle fusion, but this difference was not statistically significant (adjusted difference 0.02, 95% CI –0.008 to 0.05; p = 0.14). The CI was generated using the bootstrapping technique (1000 iterations). The total cost of TAR from the NHS and PSS perspective was £2576 higher than the total cost of ankle fusion (95% CI £1181 to £3988; p < 0.01). The difference was due to the difference in the cost of surgery (£2230, 95% CI £1024 to £3103; p < 0.01), as other differences in other cost components were not statistically significant. The incremental cost-effectiveness ratio (ICER) was £127,931 per QALY gained at 52 weeks. Model-based analysis suggested that TAR is cost-saving compared with ankle fusion when extrapolated to a lifetime horizon. As the population of interest is aged 50–85 years, the average life expectancy was 17 years; therefore, the model was run for 17 cycles. Over the lifetime horizon, there was a 69% probability that TAR would be cost-effective compared with ankle fusion at the cost-effectiveness threshold of £20,000 per QALY gained. Conclusions Both TAR and ankle fusion improved patients’ QoL at 1 year, but we did not show one to be superior in terms of clinical scores at 52 weeks when using either ITT or per-protocol analysis. The TARVA trial is inconclusive in terms of the superiority of TAR, as the 95% CI for the adjusted treatment effect includes both a difference of zero and the MID of 12. However, we can rule out the superiority of ankle fusion. Both operations appear to be safe. A post hoc analysis of the most common type of implant in the UK, the fixed-bearing TAR, did show a statistically significant improvement of TAR over ankle fusion, suggesting that fixed-bearing TAR may outperform ankle fusion. There is a 69% probability of TAR being cost-effective compared with ankle fusion at the NICE cost-effectiveness threshold £20,000 per QALY gained over patients’ lifetime. Future research There is a strong case for continuing follow-up, in particular to study the radiological and clinical progress of these patients, and the need for revision surgery. There is also a need for studies to explore the sensitivity of clinically important differences between arms when both have already improved significantly from their baseline scores. Trial registration This trial is registered as ISRCTN60672307 and ClinicalTrials.gov NCT02128555. Funding This project was funded by the National Institute for Health and Care Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 27, No. 5. See the NIHR Journals Library website for further project information.
Keywords
