Foot & Ankle

Complications following ankle arthrodesis versus ankle arthroplasty

The comparison between arthrodesis and arthroplasty of the ankle continues. There is now ample research comparing the functional outcomes of the two, although the large randomized trials are ongoing and have yet to report. The state of play for most mainstream surgeons is that in selected patients who do not have too much deformity and whose functional demands are not too high, ankle arthroplasty can be considered. In the other patients who are young, are high-demand, or have significant preoperative deformity or bone loss, the majority of surgeons favour ankle arthrodesis. One of the things that does steer decision-making in conditions such as ankle arthritis, where there are two potentially effective treatments, is the side-effect and complication profile. This study from Charlotte, North Carolina (USA) sets out to describe the complication profiles of these two treatments in greater detail.¹ The authors sought to compare national rates of perioperative complications in the United States between a statistically matched cohort of patients who underwent either an ankle arthrodesis or arthroplasty. The authors identified 4192 ankle arthroplasty patients and compared them with 16 278 arthrodesis patients who underwent surgery during the same time period. The inpatient sample data were used to establish major and minor complications, and mortality was also recorded. There is a risk of selection bias and, to counter this, the cohorts were matched for age, sex, race, surgery year, hospital type, comorbidities, adjunctive procedures, and surgical indication. This resulted in a statistically matched cohort of 1574 patients. There was at least one major complication in 8.5% of the arthrodesis patients (n = 134) compared with 5.3% (n = 84) of the arthroplasty group, and this high rate of complication in the arthrodesis group was also reflected in minor complications (4.7% vs 5.9%). Once the authors had undertaken case-mix adjustment, it appeared that ankle arthrodesis was 1.8 times more likely to be followed by a major complication than ankle arthroplasty was, although the minor complication rate was 29% lower in the former group. This paper will do a lot to set the mind at rest with regard to perioperative complications in the total ankle arthroplasty group, which, in the largest study of its type, have been shown to be substantially fewer than for patients undergoing arthrodesis. This allays concerns about high rates of intraoperative and postoperative complications, which have been mostly reduced through the development of newer prostheses and refined surgical techniques. The remaining questions, of course, are functional, and we wonder what sort of longevity can be expected from the newer generation of ankle prosthesis.

Does supramalleolar osteotomy work?

One of the difficulties with joint preservation surgery for arthritis is establishing whether it actually has the desired effect. That said, realignment osteotomies of the knee and ankle are regaining widespread popularity while unicompartmental knee arthroplasty and ankle arthroplasty have not fully lived up to their early promise in terms of longevity in the younger population. However, there is little in the way of long-term outcome data for osteotomies and, specifically, there is little evidence to support the hypothesis that realignment results in a reduction in the progression of arthrosis. We were delighted to come across this report from Gyeonggi-do (South Korea), where the authors have undertaken supramalleolar osteotomy with a relook arthroscopy, radiological films, and clinical follow-up at an interval of at least a year.² Their series of 22 patients was reported including Visual Analogue Scale (VAS) pain scores and American Orthopaedic Foot and Ankle Society (AOFAS) ankle-hindfoot scores, along with the usual gamut of radiological measures. The arthroscopic findings were reported using the Outerbridge classification, and essentially complete results were available for 14 patients. From a clinical perspective, there was a significant improvement in outcome in both the AOFAS scores (improving from 60.7 to 87.1) and VAS pain scale (improving from 6.5 to 1.1). Perhaps most crucially, in the 14 patients who underwent second look ankle arthroscopy at one year, cartilage regeneration of the medial compartment of the tibiotalar joint was observed in 12 patients (85%), and there were no observed disease progressions. Although ultimately a report of a small number of patients, the results here are very encouraging. The authors present improved outcomes in a difficult-to-treat condition, and also report that perhaps, unlike in other joints, an offloading realignment osteotomy can result in improvement in the cartilage condition, and appears to prevent degeneration in this small series. Clearly, more work is needed here. Osteotomy is a technically challenging procedure with low tolerance for surgical errors, but offers a potentially robust solution undertaken without the need for fusion or arthroplasty.

Generalized ligamentous laxity and the Broström repair

The Broström procedure is widely recognized as an excellent approach for treating lateral ligamentous laxity around the ankle. The soft-tissue repair has been shown to result in improved symptoms and reduced instability in cohort series of patients, and is generally considered the ‘operation of choice’ in those patients who are suitable for it. However, there are a few caveats, and many surgeons would think twice about carrying out the procedure on patients with a generalized ligamentous laxity due to concerns about poor functional outcomes and ‘restretching’ of the soft tissues. In a similar vein to the last paper, it was reassuring to read this report from Seoul (South Korea), where the authors have undertaken a modified Broström repair and then compared the outcomes in patients with and without generalized ligamentous laxity.³ As the surgery itself was undertaken with arthroscopic assistance, the surgical team were also able to undertake a direct inspection of the joint at the time of surgery, and also report their outcomes clinically using the American Orthopaedic Foot and Ankle Society (AOFAS) ankle-hindfoot score, a Visual Analogue Scale (VAS) pain score, and radiologically, using the talar tilt angle. This is a large series of 99 patients, which is one of the key strengths of the paper, with 24 patients in the laxity group (Beighton score of 4+) and 75 patients without laxity. There was a significantly greater change in talar tilt angle from preoperative to 12 in the laxity group (-6.9) than in the normal group (-4.2). In terms of clinical outcomes, the reported final follow-up AOFAS ankle-hindfoot score and VAS pain scores were both significantly better than the preoperative scores. This is an important study in that, although surgeons will clearly still treat patients on an individual basis, generalized ligamentous laxity perhaps should not cause as much concern as it currently does to surgeons contemplating ankle instability surgery.

Do we need to operate on Achilles tendons?

Plates and nails equal in ankle fusion

It is an oft-quoted truism that when there is more than one described and widely used surgical technique for any operation then there is probably no single perfect operation, or differing indications yield different results for different operations. Two of the most widely described operative approaches to tibiotalocalcaneal (TCC) arthrodesis are the intramedullary (IM) nail and locked plating. It is, of course, not just the implants that differ here; so too does the operative approach. The plates are generally inserted through a lateral transfibular approach, giving the advantage of simultaneous joint preparation and plate insertion. The intramedullary nail is obviously percutaneously inserted and, as such, provides some flexibility in soft-tissue approach. In this series, the authors undertook a posterior Achilles tendon-sparing approach, giving a better soft-tissue envelope. This series from a study team based in Durham, North Carolina (USA) describes a comparative outcome series of 38 patients undertaken with an IM nailing and posterior Achilles tendon-sparing approach, compared with a lateral transfibular approach.⁵ Patients who underwent TCC fusion but did not have the implant and approach combination described were excluded from the series. The overall union rate was 71%, slightly higher in the IM nail group (76% vs 64%), and this was reflected in the revision for symptomatic nonunion rates (16% vs 7%). These differences, however (in this relatively small series), were not symptomatic. It seems that, overall, the rates of complication, revision, and nonunion are broadly similar across this series of patients. It is therefore a reasonable idea to choose the operative approach and implant choice based on individual patient-related factors such as preoperative deformity, soft-tissue envelope, and surgeon preferences to try to get the best result for each patient. This paper again highlights the high rates of nonunion seen in TCC fusions, but shows that many of these are asymptomatic or not symptomatic enough to warrant revision surgery.

PROMIS reliable in the foot and ankle

The dependability of evidence-based medicine and performance-related healthcare payments is entirely reliant on the collection of patient-reported outcome measures (PROMs). While this is a development we would wholeheartedly support here at 360, it does carry with it a substantial responder burden, with patients often required to fill out long questionnaires at multiple timepoints during their healthcare journey – a task that not only takes patient time, but can also encourage incomplete or inaccurate questionnaires. The Patient-Reported Outcome Measurement Information System (PROMIS) is an adaptive questioning system that homes in very quickly on relevant questions, and has been shown in many diagnoses and conditions to be as reliable as more traditional questionnaires. The difficulty with an adaptive system like this, of course, is that it needs to be validated against traditional questionnaires in almost every diagnosis. We were delighted to see that researchers in St Louis, Missouri (USA) have done just this for patients with hallux valgus.⁶ The study revolves around establishing the relationship between traditional PROMs scores (in this case, the Foot and Ankle Ability Measure (FAAM)) and the PROMIS system in hallux valgus diagnoses. The authors enrolled in the study, on a retrospective basis, 85 patients with hallux valgus diagnoses who had completed both the FAAM Activities of Daily Living (ADL) and PROMIS scores (physical function, pain interference, and depression). The authors analyzed the data using a rudimentary approach of an initial Spearman’s correlation coefficient and then a stepwise regression approach, in an attempt to establish which demographic and outcome variables were determining any observed variance in scores. At a top-down level, there was a strong correlation between the FAAM scores and PROMIS physical function (r = 0.70), while weaker correlations were seen with PROMIS pain interference (r = -0.65) and PROMIS depression (r = -0.35) indices. The authors conclude that a significant proportion of the variation in the PROMIS physical function and FAAM ADL scores can be accounted for by the PROMIS pain scale. Here at 360, we are not sure if, with an adaptive score like PROMIS, this sort of simplified approach to scoring variation can be taken, as it is likely that the PROMIS itself is non-linear. However, this paper does clearly show that, in hallux valgus patients, the efficiency of data capture allowed by the PROMIS can be easily harnessed.

Are multiple procedures in the diabetic foot just ‘creeping amputations’?

The diabetic foot and ankle is a fickle beast, and surgeons, diabetologists, podiatrists, and patients often input significant time, effort, healthcare expenditure, and emotion in multiple limb salvage procedures that may or may not eventually end up in amputation. The authors of this paper from Singapore (Singapore) ask the question, is a ‘creeping amputation’ really a good option in these patients? And given their time again, would the patients have opted primarily for an amputation? The authors report a series of 41 patients, all of whom underwent below-knee amputation (BKA) over a two-year period for the sequelae of diabetic infection. The cohort was divided into those who had primary amputation and those who had a creeping amputation (multiple failed salvage procedures). Outcomes were assessed using the Barthel Index (BI) and the Reintegration to Normal Living Index (RNLI). In addition, the authors asked the participants a series of structured questions with the aim of determining whether the patient would consider having the multiple attempts at limb salvage again if faced with the same problem. The eventual outcomes were good, with a mean BI of 14.2 and RNLI of 73.2. Between the two groups, there were no differences in likelihood of prosthesis usage or in either of the outcome scores. However, perhaps surprisingly, 16 of 17 patients with creeping amputation would undergo the same procedures if given a similar option; conversely, only 15 patients (62.5%) with primary amputation would do the same again, while the other nine patients (37.5%) would choose to do everything possible to save their leg if faced with a similar situation. The conclusions one can draw from this paper are fairly clear. If faced with a diabetic foot infection, there really aren’t any advantages in jumping straight to amputation, either for patient preferences or eventual outcomes, as those who fail salvage do just as well as those who have a primary amputation.