Subtalar Arthrodesis Union Rates With and Without Adjacent Ankle Arthrodesis

Introduction

Subtalar arthritis causes considerable morbidity and can be secondary to a range of pathologies including posttraumatic changes, inflammatory disease, or in those with a cavovarus or planovalgus deformity.^1,4,13 Treatment consists of initial nonoperative management with the use of orthotics, analgesia, and subtalar injections. ¹⁷ With severe arthritis, the majority of patients undergo surgical treatment with a subtalar arthrodesis.^1,9

Subtalar arthrodesis has been shown to decrease pain and give good functional outcomes and satisfaction. ⁸ These have traditionally been performed open, but there is an increasing trend for arthroscopic arthrodesis with promising results.^1,9,11 Several risk factors for nonunion of subtalar arthrodesis have been reported including high alcohol intake, diabetes, neuropathy, smoking, and infection.^1,8,11,12,20 Nonunion rates of up to 40% have been reported, but most modern studies report much lower rates.^1-3,8

Subtalar arthritis may develop in patients with ankle arthritis. The surgical treatment of end-stage ankle arthritis includes ankle arthrodesis or ankle replacement. ⁶ Both have demonstrated improved functional outcomes. It is thought there may be an increased risk of developing adjacent joint arthritis following an ankle arthrodesis. ⁵ When a subtalar arthrodesis is performed below a previous ankle arthrodesis, there is concern that there are higher rates of nonunion. However, there is limited evidence on the union rates of subtalar arthrodesis below an adjacent ankle arthrodesis. ¹⁹

The aim of this retrospective cohort study was to compare the subtalar arthrodesis union rate of patients with native tibiotalar joints to that of patients with prior tibiotalar arthrodesis. Secondary outcomes were to assess risk factors for nonunion.

Methods

This was a retrospective cohort study of consecutive patients who underwent a subtalar arthrodesis over a 10-year period performed in a single center between January 2010 and December 2020. Five fellowship-trained orthopaedic foot and ankle surgeons performed the procedures.

The surgical technique was determined by the operating surgeon. All were performed by an open lateral approach. Screw placement was 1 or 2 screws either antegrade or retrograde. No patients who had already had an ankle arthrodesis underwent subtalar arthrodesis with a hindfoot nail. Patients were put into a cast for 6 weeks following surgery and kept nonweightbearing. At the 6-week follow-up, patients were allowed to fully weightbear if radiographs were satisfactory. All patients were followed to union, and none were lost to follow-up.

The inclusion criteria were (1) isolated subtalar arthrodesis with or without an adjacent ankle arthrodesis and (2) patients were assessed by 2 authors working independently. The exclusion criteria were a subtalar arthrodesis performed with (1) a talonavicular or calcaneocuboid arthrodesis, (2) a tibiotalocalcaneal arthrodesis, (3) previous talonavicular or calcaneocuboid arthrodesis, (4) major deformity, (5) neuromuscular condition, or (6) Charcot arthropathy.

Demographics and patient comorbidities were collected and analyzed including age, alcohol intake, smoking, peripheral vascular disease, diabetes mellitus, and chronic kidney disease. Union was defined as a combination of radiologic union with bridging callus on 2 views, and the patient had clinical symptoms of union including the ability to fully weightbear without pain. Radiographs were reviewed by 2 reviewers working independently (1 fellowship-trained surgeon and 1 specialist trainee). If there was uncertainty based on radiographs or the patients had clinical symptoms, then a nonweightbearing CT was performed during routine follow-up and this was used to determine union.

Statistical analysis was undertaken in Stata (version 16). Chi-square test and Mann-Whitney tests compared differences in demographics and risk factors for nonunion between groups. A logistical regression model was performed to determine risk factors for nonunion, initially undertaking crude and subsequent multivariable analysis. Results are expressed with odds ratios, 95% CIs, and P values.

Results

A total of 71 patients were included, of which 18 patients had an adjacent ankle arthrodesis and 53 patients had a native ankle (Figure 1). No patients were lost to follow-up. Those with an adjacent ankle fusion were significantly older (57 years, SD 14.7) than those with a native ankle joint (47.7 years, SD 17.1) (P = .042). Overall, 77.7% of those with an adjacent ankle fusion were male compared with 50.9% of those with a native ankle (P = .046). There were higher rates of diabetes mellitus in those with an adjacent ankle arthrodesis (P = .095), but there were no other differences in demographics between groups (Table 1).

OPEN IN VIEWER

Figure 1.

Flow diagram of patient identification.

OPEN IN VIEWER

Table 1.

Differences Between Patients With and Without an Adjacent Ankle Arthrodesis.

	Native Tibiotalar Joint, n (%)(n = 53)	Prior Ankle Arthrodesis, n (%)(n = 18)	P Value
Diabetic	6 (11.3)	5 (27.8)	.095
Rheumatoid arthritis	2 (3.8)	0	.403
Other inflammatory arthritis	0	0
Smoker	13 (24.5)	4 (22.2)	.843
Peripheral vascular disease	(1.9)	1 (5.6)	.416
Steroids	1 (1.9)	0	.557
Alcohol excess	5 (9.4)	2 (11.1)	.837
Chronic kidney disease	2 (3.8)	1 (5.6)	.745
Immunosuppressants	2 (3.8)	0	.403
1 screw	41 (77.4)	2 (11.1)	<.001

Patients with an adjacent ankle arthrodesis had surgery with 1 screw in 2 (11.1%) and 2 screws in 16 (88.9%). Those without an adjacent ankle arthrodesis had surgery with 1 screw in 41 (77.4%) and 2 screws in 12 (22.6%) (P < .001).

The successful union rate in those with a preexisting ankle joint arthrodesis was 44.4% compared with 86.8% in those without an ankle joint arthrodesis (P < .001).

On multivariable logistic regression modeling, an adjacent ankle arthrodesis was the only statistically significant risk factor for nonunion (odds ratio 4.90, 95% CI 1.02-23.56; P = .047) (Table 2). In addition, 9.4% of patients without an ankle arthrodesis underwent a revision subtalar arthrodesis compared with 44.4% of those with an ankle arthrodesis (P = .001). The odds ratio of requiring revision surgery for those with an adjacent ankle arthrodesis was 7.68 (95% CI 2.07-28.43) (P = .002)

OPEN IN VIEWER

Table 2.

Logistic regression modelling.

Variable	Crude		Multivariable
	Odds Ratio	P Value	Odds Ratio	P Value
Adjacent ankle arthrodesis	8.21 (2.42-27.91)	.001	4.90 (1.02-23.56)	.047
Age	1.04 (1.00-1.08)	.044	1.05 (0.99-1.10)	.204
Diabetic	2.07 (0.52-8.16)	.300	0.99 (0.17-5.91)	.994
Smoker	1.46 (0.43-4.96)	.546	3.95 (0.62-24.96)	.144
Peripheral vascular disease	3.31 (0.20-56.00)	.406	2.96 (0.11-82.44)	.523
Alcohol excess	2.68 (0.54-13.40)	.230	2.10 (0.27-16.18)	.475
Chronic kidney disease	1.63 (0.14-19.12)	.699	1.82 (0.08-40.17)	.704
1 or 2 screws	5.70 (1.72-18.84)	.004	1.81 (0.38-8.70)	.457

Discussion

This study demonstrates that subtalar arthrodesis performed in patients with an adjacent ankle arthrodesis have significantly higher rates of nonunion compared with those with a native ankle after accounting for potential confounding factors.

Studies have found good union rates with both open and arthroscopic subtalar arthrodesis techniques, and the only comparative study demonstrated no significant differences in union, complications, and function between either technique.^3,7,15 All procedures in this study were performed by the open technique. These findings are similar in this study. Previous studies have demonstrated risk factors for subtalar nonunion to include a prior ankle arthrodesis, as well as an increased body mass index, diabetes mellitus, and neuropathy.^10,18

There has been limited published data on the union rates of a subtalar arthrodesis below an ankle. Zanolli et al ¹⁹ undertook 151 subtalar arthrodeses, of which only 13 were below an ankle. They demonstrated a fusion rate of 61.5% in those with an adjacent ankle arthrodesis, and a significantly higher fusion rate of 91.3% (P = .007) in a native ankle. Additionally, this study analyses other risk factors for nonunion and performs logistic regression taking into account other variables that may influence the development of nonunion in subtalar arthrodesis. This determines that it is the presence of an adjacent ankle fusion, and no other risk factors, that account for the increased rate of nonunion.

It is difficult to be certain why the nonunion rates of subtalar joints are higher with an adjacent ankle arthrodesis. A possibility is that because of the lack of movement in the ankle joint, there is still some mobility in the subtalar joint when mobilizing, and increased forces are transmitted through this joint compared with when there is a native ankle or an ankle replacement. This has the potential to create increased strain and therefore preventing union. Another hypothesis may be that the previous ankle arthrodesis surgery has disrupted the blood supply within the talus, leading to a higher nonunion rate.

The treatment of end-stage ankle arthritis is either ankle arthrodesis or ankle replacement. There is still no consensus on the difference in outcomes following these alternative procedures. ⁶ A concern with an ankle arthrodesis is the potential development of adjacent joint arthritis requiring surgery, and this has been shown to be lower in ankle replacements.^5,14 This study demonstrates that those that then develop subtalar arthritis and require a subtalar arthrodesis have significantly higher rates of nonunion than those with a native joint. Fusion rates of the subtalar joint following ankle replacement in contrast have been found to be high. ¹⁶

Supplemental Material