Clinical outcomes of surgical treatment for end-stage ankle osteoarthritis in patients aged over 75 years: a multicenter, retrospective study

The purpose of this study was to clarify the clinical outcomes of surgical treatment for end-stage ankle osteoarthritis in patients aged ≥ 75 years and compare the outcomes with those of patients aged <75 years. A total of 148 patients, including 65 post-total ankle arthroplasty and 83 post-ankle arthrodesis patients, were surveyed retrospectively. Clinical outcomes were assessed using the Japanese Society for Surgery of the Foot hindfoot scale and the self-administered foot evaluation questionnaire preoperatively and at the last follow-up. Patient characteristics, including age, sex, body mass index, radiographic severity, and follow-up period, were also assessed. Patients were divided into the older age group ( ≥ 75 years) and the younger age group (<75 years). Improvements in outcomes were then compared between the age groups using univariate analysis and analysis of covariance adjusted for patient characteristics. Post-total ankle arthroplasty and post-ankle arthrodesis were analyzed separately.

End-stage ankle osteoarthritis (OA) is a disabling condition in which quality of life impairment is as severe as that associated with hip OA (1). Although the risk of developing ankle OA is lower than that of developing OA in the hip and knee joints, the number of patients with ankle OA has recently been expected to increase (2). Generally, surgical treatment for end-stage OA includes either total ankle arthroplasty (TAA) or ankle arthrodesis (AA) (3). Both techniques have advantages and disadvantages, and both have reported satisfactory clinical outcomes (3). Over the past decade, the performance of both surgical procedures has been increasing worldwide, even in Japan (4,5).
Life expectancy has increased worldwide. The world population is aging rapidly, especially in Japan, with an average life expectancy of >80 years in 2019 (6). The number of surgical treatments for degenerative joint disease in elderly patients has also been increasing accordingly; therefore, it is not uncommon for patients in their 70s and 80s to undergo joint arthroplasty (7). However, there is little data on whether the clinical outcomes after TAA and AA in elderly patients are comparable to those in younger patients. Elderly patients may have coexisting medical conditions and reduced physical functioning, making it di cult to achieve su cient functional recovery after surgery. Several studies have compared the clinical outcomes of TAA between older and younger age groups (8-13). However, these studies divided the patients using the 50-60 age group threshold to highlight outcomes in younger patients rather than older patients. Few studies have reported the outcomes of older patients in their 70s and 80s (14,15). Regarding AA, patients undergoing the procedure are generally young, and only a few studies have assessed the clinical outcomes of AA in elderly patients (16,17). The data on the outcomes of TAA and AA would provide important information on clinical decision-making for our aging population. Therefore, the purpose of this study was to clarify the clinical outcomes of surgical treatment for ankle OA in elderly patients and compare the results with those of younger patients.

Study design
This multicenter, retrospective study was approved by the Ethics Committee of St. Luke's International Hospital (17-R164) and the participating hospitals and was conducted following the Declaration of Helsinki. Since this was a retrospective study, each hospital provided opt-out consent with no written consent. Patients were screened using data extracted from the electronic medical records of eight hospitals. Patients who underwent TAA or AA for end-stage ankle OA between January 2015 and July 2019 were included in the study. The exclusion criteria were as follows: 1) diagnoses other than OA (e.g., rheumatoid arthritis and neuropathic arthritis); 2) revision surgery; 3) talocalcaneal joint arthrodesis; 4) history of TAA or AA on the contralateral side, and 5) <1-year follow-up. Overall, 169 ankles of 169 patients met the inclusion criteria. Among these patients, three had rheumatoid arthritis, ve underwent revision surgery, one had talocalcaneal joint arthrodesis, and one had a history of TAA or AA. Furthermore, 11 patients were lost to follow-up, including one patient who died. Therefore, the data of the remaining 148 patients, including 65 and 83 patients who underwent TAA and AA, respectively, were analyzed.

Patient background
Data on patient background, including age, sex, body mass index (BMI), and follow-up period, were recorded. Patients were classi ed into those aged ≥75 years or <75 years. We used a threshold of 75 years according to the de nition set for determining older senior citizens by the Ministry of Health, Labor and Welfare of Japan (18). Radiographic severity was assessed using the Tanaka-Takakura classi cation (19) on preoperative weight-bearing anteroposterior radiographs of the ankle. We divided the patients into those with stage 3b or less (joint space obliteration extended to the roof of the talar dome) and those with stage 4 (obliteration of the whole joint space) as the staging might affect the clinical outcomes (20).

Surgery
The types of surgery were classi ed as TAA or AA. Due to the retrospective nature of this study, the operative indications and the selection of surgical procedures were based on the patients' and surgeons' shared decisions. In general, TAA was performed in older, less obese, and sedentary patients. In addition, patients with moderate deformities tended to undergo TAA. In contrast, AA was performed in younger, obese, and physically active patients and those with severe deformities. Furthermore, the type of implant (TNK Ankle, Kyocera, Kyoto, Japan; Fine Ankle, Teijin Nakashima Medical, Okayama, Japan) was recorded for TAA patients. The surgical approach (arthroscopic or open) was recorded for patients with AA.

Clinical outcomes
The clinical outcomes were evaluated using the Japanese Society for Surgery of the Foot (JSSF) hindfoot scale and the Self-Administered Foot Evaluation Questionnaire (SAFE-Q) (21,22). The JSSF hindfoot scale is a validated objective evaluation with a 40-point subscale for pain, a 45-point subscale for function, and a 15-point subscale for alignment; higher scores indicate better clinical outcomes (22). The SAFE-Q is a validated questionnaire consisting of 34 questions on foot and ankle symptoms ( [21). It has ve subscale scores: pain and pain-related, physical functioning and daily living, social functioning, shoe-related, and general health and well-being-possible scores for each subscale range from 0 to 100, with higher scores indicating better clinical outcomes. We assessed the outcomes preoperatively and at the last follow-up, and the recorded improvement in scores was used for statistical analysis. Furthermore, we recorded the occurrence of revision surgery during the follow-up period. The major intraoperative complications investigated were fractures, and technical errors (iatrogenic complications). The postoperative complications included deep infection, peroneal tendon dislocation, and revision surgery.

Statistical analysis
Patients who underwent TAA and AA were analyzed separately because the indications for these surgical procedures were different, as discussed earlier. Patient characteristics were compared between patients aged ≥75 years and those aged <75 years using Student's t-tests and Chi-square tests, as appropriate. The pre-and postoperative JSSF hindfoot scale scores and SAFE-Q subscale scores were compared using paired t-tests. To compare the clinical outcomes between the older and younger age groups, the recorded improvements in the JSSF and SAFE-Q subscale scores were also compared between the age groups using Student's t-tests. Complications were compared using Fisher's exact test. Furthermore, an analysis of covariance was performed to determine the independent association between age group and clinical outcomes. The clinical outcome was entered as the objective variable, and the age group was entered as the explanatory variable, adjusting for sex, BMI, radiographic stage, and follow-up period. The statistical signi cance value was set at p < 0.05.
The mean age of the older age group was 78 years. Except for the age, there were no signi cant differences in patient backgrounds between the older and younger age groups (Table 1). In the older age group, the JSSF hindfoot scale score improved from 53 points preoperatively to 90 points at the last follow-up (p < 0.001, Table 2). Similarly, all SAFE-Q subscale scores improved signi cantly, with the improvement in scores ranging from 30 to 37 points (p < 0.001 for all subscales, Table 2). There were no signi cant differences in the improvement of any of the outcomes between the older and younger age groups in the univariate analysis (Table 4). Furthermore, in the analysis of covariance adjusted for patient background, patient age was not associated with either of the clinical outcomes (Table 4). Four patients (11%) had major intraoperative complications in the older group, including four fractures. Two patients (6%) had major intraoperative complications in the younger group, including two fractures. During the follow-up period, four patients (11%) had major complications in the older group, including one deep infection that required debridement surgery, one peroneal tendon dislocation, and two revision surgeries to 1 aseptic loosening and one infection. Three patients (10%) had major complications in the younger group, including one removal of impinging osseous overgrowth and two revision surgeries due to aseptic loosening. There was no signi cant difference in the occurrence of intra-and postoperative complications between the groups (p = 0.57).  Values are presented as mean ± standard deviation.
JSSF, Japanese Society for Surgery of the Foot; SAFE-Q, self-administered foot evaluation questionnaire.

Ankle arthrodesis
Overall, 83 patients, including 22 patients aged ≥75 years and 61 patients aged <75 years, were analyzed ( Table 1). The proportion of women was 55% (n=46), and the mean BMI was 25 kg/m 2, with ve (6%) patients recorded as obese. The mean age was 79 years in the older age group and 15 years in the younger group. In the older age group, the JSSF hindfoot scale and all SAFE-Q subscale scores improved postoperatively (Table 3). More speci cally, the SAFE-Q scores improved by 20-35 points (p < 0.001 for all subscales, Table 3). Similarly to the TAA patients, there were no signi cant differences in improving all clinical outcomes between the older and younger age groups after univariate analysis (Table 3). In the analysis of covariance adjusted for the patient background, patient age was not associated with any of the clinical outcomes except for the SAFE-Q physical functioning subscale score (p=0.04, Table 4). The recorded improvements in the physical functioning scores were 20 and 29 points in the older and younger age groups, respectively. In the older group, no patients had major intraoperative complications. Two patients (6%) had major intraoperative complications in the younger group, including iatrogenic complications (endosseous drill breakage). During the follow-up period, one (4%) patient in the older age group had major complications, including revision surgery due to nonunion. Also, three (5%) patients in the younger age group had major complications, including revision surgery due to nonunion, overgrowth, and two revision surgeries due to aseptic loosening. There was no signi cant difference in the occurrence of intra-and postoperative complications between the groups (p = 0.62). Values are presented as mean ± standard deviation.
JSSF, Japanese Society for Surgery of the Foot; SAFE-Q, self-administered foot evaluation questionnaire.

Discussion
This study showed that surgical treatment for end-stage ankle OA resulted in satisfactory clinical outcomes in patients aged ≥75 years; improvements in scores were observed in patients undergoing both TAA and AA. Furthermore, the improvements were comparable to those observed in patients aged <75 years. Thus, the results of our study suggest that surgical treatment of end-stage ankle OA can be a viable treatment option, even in elderly patients.
We showed that the JSSF hindfoot scale score and all SAFE-Q subscale scores improved signi cantly for TAA patients in the older age group with a mean age of 78 years. Speci cally, the improvement in the SAFE-Q subscale scores ranged from 13 to 38 points. Non-surgical treatment is the rst-line treatment for ankle OA; however, insu cient data support its e cacy (23). Moreover, elderly patients, such as those aged >75 years, tend to undergo non-surgical management because of their limited physical functioning and concern for postoperative complications (7). The minimal clinically important differences in the SAFE-Q scores for TAA patients were not determined. However, the recorded improvements in the scores found in this study were comparable to those of previous reports (24-26) and were clinically signi cant.
This study showed that the recorded improvements in the outcomes of the older age group were comparable to those of the younger age group for TAA patients. Our results were in line with those of Tenenbaum et al. (14) in which patients aged >70 years and those aged 50-60 years demonstrated equivalent improvement in the American Orthopedic Foot and Ankle Society (AOFAS) ankle/hindfoot scale and visual analog scale (VAS) pain scores. Demetracopoulos et al. (15) also found that most of the clinical outcomes in patients aged >70 years were comparable to those aged ≤70 years, although the AOFAS function and SF-36 vitality subscale scores were lower in older patients than in younger patients.
Other studies used a younger cut-off age of 50-65 years (8-13) and reported no difference in clinical outcomes depending on age. Thus, the results of our study suggest that TAA could be a treatment of choice even in patients aged ≥75 years. Furthermore, there was no difference in intraoperative and postoperative complications between the groups, which was a concern for the elderly. Post-AA, improvements in clinical scores were comparable between patients aged <75 years and those aged ≥75 years. Previous studies have shown that age at surgery was not associated with clinical outcomes (27)(28)(29). However, the patients in these studies were relatively young, with a mean age of 57-63 years. Therefore, the results of our study suggest that even older patients can expect a similar level of clinical improvement after undergoing AA. In contrast, Berlet et al. (30) reported that patients aged ≥60 years had a higher risk of nonunion after foot and ankle arthrodesis. However, we could not determine the effect of age on bone union because of the small study population. Furthermore, there was no difference in intraoperative and postoperative complications between the groups or within the TAA group.
This study has several limitations. First, the older age group may have consisted of selected patients with good health status; therefore, surgeons would have logically expected good postoperative outcomes. This limitation could be biased toward better results in the older age group. Second, several covariates that may affect the clinical outcomes, including comorbidities, mental status, and physical activity level, were not included in the multivariate analysis because of the retrospective nature of this study. Although our study can provide clinically meaningful information to patients and surgeons, further prospective studies are necessary to draw de nitive conclusions on the association between age and clinical outcomes. Third, the follow-up period was relatively short, with a mean period of 18-26 months. However, considering both short-term results and long-term outcomes is important for elderly patients who may have a shorter life expectancy than younger patients. Fourth, despite collecting patient data from eight hospitals, the study population was relatively small, especially for comparing the incidence of revision surgery between the age groups.

Conclusions
The JSSF scale and all SAFE-Q subscale scores improved postoperatively in patients with end-stage ankle OA aged ≥75 years. Furthermore, the recorded improvements in the scores were comparable to those in patients aged <75 years. Therefore, surgical treatment of end-stage ankle OA, using either TAA or AA, can be a viable treatment option, even in elderly patients.  (17-R164) and the participating hospitals and was conducted following the Declaration of Helsinki. Since this was a retrospective study, each hospital provided opt-out consent with no written consent.

Consent for publication
Availability of data and materials

Competing interests
The authors declare that they have no competing interests.

Funding
This research did not receive any speci c grant from funding agencies in the public, commercial, or notfor-pro t sectors.
Authors' contributions KA participated in the conception and design of the study, acquisition of data, analysis, and interpretation of the data, drafting the article, and critical revision of the article for important intellectual content. SY participated in the conception and design of the study, acquisition of data, analysis, and interpretation of the data. AT participated in the acquisition of data. TK participated in the acquisition of data. HS participated in the acquisition of data. YK participated in the analysis and interpretation of the data. YS participated in the analysis and interpretation of the data. TY participated in the acquisition of data. NK participated in the acquisition of data. KN participated in the acquisition of data. NK participated in the conception and design of the study, acquisition of data. All authors read and approved the nal manuscript