Correction of hallux abducto valgus by scarf osteotomy. A ten-year retrospective multicentre review of patient reported outcomes shows high satisfaction rates with podiatric surgery

Hallux abducto valgus (HAV), the bunion deformity, is one of the most common presentations in foot and ankle practice [1, 2]. It is a complicated multiplanar pathology and is characterised by lateral deviation and valgus rotation of the hallux and medial deviation of the first metatarsal [1, 3, 4]. It has been reported that approximately 23% of adults aged between 18 and 65 years have the deformity, which increases to 35.7% of the population above 65 years of age [3]. It has a higher female predilection and can often result in a decreased quality of life [5]. HAV is often intractable despite conservative care, ultimately requiring a surgical correction for many patients after a failure of conservative care [1, 2]. Over a hundred different surgical solutions have been proposed throughout the literature with varying outcomes [6, 7]. Schrier et al. [8] suggested that up to a third of patients operated on for HAV may be dissatisfied with the outcome of their surgery (though the paper they attribute this value to in fact demonstrated good patient outcomes). Patient reported outcomes are now crucial in capturing and evaluating treatment effectiveness.

A total of 1603 scarf osteotomies (all variants) were performed using 4557 pieces of orthopaedic hardware to fixate, of which of which 3491 (76.6%) were a standard AO screw or equivalent. There was a total of 1351 patients in the cohort with 1227 females (90.83%, average age 59.13, standard deviation [SD] 13.51) and 124 males (9.17%, average age 59.43 SD 15.37) who had undergone a scarf HAV correction with or without an Akin osteotomy. The greatest percentage of females who underwent surgery were in the 60–79 age group accounting for 47.38% of the total cohort. Males in the same age range accounted for 5.10% of the population. The average outcome data score was 89.68 out of 100 (SD 9.64). Regrettably, not all episodes were fully captured on the PASCOM-10 database: 1189 (88%) patients completed the PSQ-10 questionnaire but 162 (12%) were lost to follow up (Table 1).

Of those patients who completed PSQ-10, 1177 (98.9%) patients identified their expectations for the surgery in Q1 of the questionnaire: ‘briefly state what you expected to gain from treatment’. The responses were collated into Microsoft Excel (Microsoft 365) which could then be arranged into four common themes. Some patients mentioned more than one expectation but a reduction in pain (70%) was the biggest goal for patients, followed by a ‘better looking’ or ‘straighter foot’ (25.5%) and better footwear fit/choices (25%). Only 12% of the patients who answered were looking for a return to sports or the ability to function better on their foot following their bunion correction. This may be partly due to the age demographics with over 74% of patients aged over 55 who are perhaps less likely to partake in active sports. A summary of the PSQ-10 Q1 themes can be found in Table 2.

A summary of answers (as per PASCOM, to one decimal place, which gives rounding errors in questions 6 and 9) to other questions, condensed for the brevity of this article, is listed in Table 3.

Post-operative complication data was gathered from the in-built reporting tool. In 75.1% (964 of the cohort), no sequalae were observed. A detailed breakdown of post-surgical sequalae can be observed in Table 4, as per PASCOM, to one decimal place, which gives rounding errors.

PROMs are becoming an integral component of health care practice, and which should encompass items that are relevant to the patient [22]. Baumhauer et al [23] showed that outcome factors regarded by patients as important for foot/ankle complaints differ from factors judged by physicians. It was the intention at the conceptual stage of the PSQ-10 questionnaire to capture and explore the patient’s perception of their treatment rather than assessing the outcome from a clinician’s point of view [19]. Maher and Tollafield [18] proposed that a PSQ-10 threshold score of 75 or above can be for isolated HAV correction. In their paper, they presented a mean PSQ-10 score of 86.5 for patients who underwent a scarf osteotomy. The timing of obtaining a PROM is dependent on many variables [5]. Rudge & Tollafield [19] asked centres to send questionnaires to patients 6 months after surgery and thus the consensus within UK podiatric surgical units is that PROMs have continued to be gathered at this point following treatment. Further work is currently being undertaken by the Northampton team to identify if PASCOM-10 outcome scores change over time.

From this study it was found that a total of 1151 patients (96.8% of the cohort) felt that their foot condition was either ‘better’ (16.9%) or ‘much better’ (79.9%) following surgery. This is consistent with the findings of Crevoisier et al. [24], and Spruce et al [25], who each identified that 94% of their cohort felt they had improved following a scarf osteotomy. 86% of Crevoisier’s cohort said they would undergo the same procedure again whilst Kerr et al. [26] reported that only 74.3% of their patients who underwent a scarf osteotomy would. These findings are lower than the results of this study which showed 96.6% would proceed with the same surgery in the future. Importantly, over 92% of our patients felt that their expectations had been fully met following the scarf osteotomy with 6% of those answering feeling that they had been partly met. The reasons for this warrant further study. Less than 1% felt that their expectations for surgery had not been met following the surgical procedure. Managing expectations to surgery is crucial in a consent to surgery, and value may be found in analysing the outliers.

Metatarsal troughing, delayed union, rotational malunion, recurrence of deformity and infection were all identified as post-operative complications following a scarf osteotomy by Coetzee [27]. Metatarsal fracture was the most common post-operative complication affecting 4.6% of our cohort but may reflect differences in the recording of sequela by the five units. In discussion between authors, the Nottingham units included any minor osteotomy movement as a fracture. Fracture was not shown to be a factor in the study performed by Lipscombe et al. [28] who only had two complications: a traumatic neuroma and avascular necrosis requiring an arthrodesis of the first metatarsophalangeal joint. A midterm experience of scarf osteotomy by Samaras et al [11] highlighted 15 out of 70 ft with complications which included seven recurrences, one deep vein thrombosis (DVT), four cases of complex regional pain syndrome, three intra-operative fractures and three cases of failure of osteosynthesis. Smith and colleagues [10] reported metatarsal fractures: these were intraoperative in two cases whilst two stress fracture were seen at three and 7 months post operatively. They didn’t however identify in which metatarsal the stress fractures occurred. Barouk [29] observed 3% of his study to be at risk of metatarsal fracture.

Joint pain and stiffness were recorded in 53 (4.1%) of patients in this study which is much less than that found by Crevoisier et al [24] who found that first metatarsophalangeal joint motion worsened in 7% of their patients who underwent a scarf osteotomy in isolation, and in 6.7% of patients who underwent an additional Akin osteotomy [22]. Our analysis identified 26 patients (2%) who acknowledged a problem with their scar at their six-month review which differs from Ieong et al. [30] who found that 31% of their patients had experienced symptoms at the scar site when they were reviewed at a minimum of 1 year post operatively.

There were no reported incidences of complex regional pain syndrome (CRPS) in our study. Conversely, Samaras et al [11] and Deenik et al [31] noted rates of 5 and 10.6% respectively. Seven cases of DVT were highlighted in our study equating to 0.5% of the cohort. As noted above, Samaras et al. [11] documented only one case of DVT in their 70 patients undergoing a scarf osteotomy. Their patients were noted to have been prescribed low molecular weight heparin (LMWH) for 3 weeks despite being allowed to immediately weight bear in a heel wedge shoe. This seems to contradict the thoughts of Jameson et al. [32] who suggested that patients in the U.K. undergoing bunion surgery had no clear risk factors for venous thromboembolism (VTE) and VTE prophylaxis had no benefit in ambulatory cases. Harrison and Walker [33] also proposed that patients who are ambulatory after day case surgery were not considered to be at high risk of VTE and forefoot surgery should be considered low risk.

As with any audit and data collection, the results are only as good as the information inputted. Consequently, the post-operative sequalae shown in this study may be underestimated. Some post-operative dressings were performed by specialist nurses or health care assistants in the private sector who may not have had access to the PASCOM-10 database, and this may therefore represent an under reporting of incidences, such as stitch abscesses. Many of the common complications are seen early in the post-operative phase whilst the main post-operative concerns are inputted into the PASCOM-10 software at 6 months following surgery when the PSQ-10 questionnaire is completed. In the future it would be prudent to record all sequalae as and when they occur, rather than waiting until the final 6 month review. The link between sequela and final outcome is not well delineated using PASCOM-10 and it is unclear from the data set if those patients that experienced a complication of surgery (e.g., hallux varus) were or were not satisfied with their overall result. Future studies would need to build in systems to cross reference these factors.

Both NHS units are training centres for podiatric surgery trainees and registrars. It was not possible from the collated data to unpick whether the outcome varied between surgeon grades (and therefore by extrapolation, experience). This would be a useful area to unpick in future studies but would require separate data to be prospectively gathered using PASCOM-10.

Although the figures presented are mid-term results, the authors believe that the data gives a good indication of the outcomes and post-operative sequelae that may affect patients undergoing a scarf (with or without an Akin) osteotomy for HAV correction as performed by podiatric surgeons. This should enable a more informed consenting process with our patients in the future. The data suggests good outcomes for the procedure in a wide cohort of patients, across five podiatric centres, where much of the surgery is performed by non-consultant grades. This study has also highlighted the importance of inputting post-operative sequalae as they occur, rather than waiting until the final review, to capture a more accurate and comprehensive overview of any concerns raised by both patient and clinician.

Patient expectations of surgery were often achieved in the centres studies which may be attributed to the pre-operative work up of the patient. Further investigation into this correlation is suggested.