Outcomes of Joint Preservation Surgery in Valgus Ankle Arthritis Without Deltoid Ligament Insufficiency

Introduction

The fundamental principle of joint preservation surgery for asymmetric ankle arthritis is to transfer weightbearing load from the degenerated area to the uninvolved side of the joint. ⁸ It is known that malalignment of the hip-knee-ankle-hindfoot axis can lead to altered load distribution within the tibiotalar joint, and this is considered a potential cause of asymmetric ankle arthritis.^4,19,25 In this regard, realignment of the weightbearing axis through supramalleolar correction has been a commonly adopted method in treating ankles with mild to moderate arthritis.^16,19,26

In the setting of varus ankle arthritis, both varus lower limb mechanical axis and varus distal tibial plafond are well-established causes of eccentric load on the medial tibiotalar joint.^18,19 Thus, joint preservation surgery often utilizes valgus angulating supramalleolar osteotomies as the primary procedure, with additional inframalleolar correction for residual talar tilt and/or hindfoot realignment.^19,27

In contrast, an association between supramalleolar malalignment and valgus ankle arthritis development is largely unexplored, especially in the case of primary origin. The main etiology of primary valgus ankle arthritis described in the literature is deltoid ligament insufficiency from advanced flatfoot deformity.^2,3 Consequently, joint preservation methods have primarily focused on inframalleolar realignment procedures, such as deltoid ligament reconstruction with concomitant hindfoot alignment correction.^6,9,22

In recent radiographic studies, a distinct relationship was observed between valgus ankle arthritis and specific lower limb alignment or subtalar joint orientation.^4,5,10,14 Anecdotally, we have observed a subset of valgus arthritic ankles with varus malalignment of the mechanical axis and/or tibial plafond without evidence of deltoid ligament insufficiency. This led us to hypothesize that, similar to the joint preservation techniques in varus ankle arthritis, there may be a role for supramalleolar realignment procedures in redistributing eccentric load at the tibiotalar joint.

In the present study, we hypothesized that both clinical and radiographic outcomes would improve with realignment surgery with supramalleolar osteotomies in valgus arthritic ankles without deltoid ligament insufficiency. This study aimed to investigate (1) the radiographic characteristics of the valgus ankle arthritis without deltoid ligament insufficiency and (2) the clinical and radiographic results following realignment surgery with supramalleolar correction.

Methods

Establishment of the Study Cohort

A retrospective analysis of consecutive patients who underwent joint preservation surgery for primary valgus ankle arthritis between 2015 and 2018 was performed. The etiology of ankle arthritis was determined utilizing a previously defined method by Valderrabano and colleagues. ²⁸ Therefore, primary arthritis included ankles with no documented history of posttraumatic deformity or arthritis from secondary causes such as rheumatoid, diabetes, or infection. A diagnosis of valgus ankle arthritis was made clinically and radiographically by a single senior surgeon (WCL) who performed all surgical procedures in the present study. Clinically, patients complained of pain and discomfort with impairment of physical activities. The diagnosis was made radiographically with the presence of lateral tibiotalar joint narrowing and a talar tilt angle (TT) larger than 2 degrees. All patients had failed conservative treatment modalities such as pain medication, physical therapy, and braces prior to surgery. Patients were considered surgical candidates for joint preservation procedures, regardless of age, when radiographs demonstrated preservation of more than half of the tibiotalar joint space. Joint preservation surgery was not performed if the preoperative TT was greater than 10 degrees, as larger talar tilt is associated with insufficient radiographic correction in the treatment of varus ankle arthritis. ¹⁹

The study cohort included patients who were over 18 years of age and completed a minimum 2-year follow-up. Patients with (1) a history of total knee arthroplasty (TKA) or high tibial osteotomy and (2) incomplete 2-year follow-up were excluded from the study (Figure 1). Ultimately, 13 ankles in 13 patients (11 males and 2 females) remained for analysis after applying the inclusion and exclusion criteria. The mean age of patients was 51.2 (range, 29-69) years, and the mean body mass index (BMI) was 25.2 (range, 21-31.6) kg/m². The mean follow-up was 31 (range, 24-60) months. Demographic data of the study cohort are tabulated in Table 1.

OPEN IN VIEWER

Figure 1.

Flowchart showing inclusion and exclusion of patients. HTO, high tibial osteotomy; TKA, total knee arthroplasty.

OPEN IN VIEWER

Table 1.

Demographic Data of the Patients. ^a

No.	Sex	Age(y)	BMI(kg/m2)	Side	Concurrent procedures	Preopgrade b	Postopgrade b	FU(mo)
1	F	37	23	L	SMO, MDCO, AD	2	0	60
2	M	69	25.6	L	SMO, MDCO, SE	2	1	42
3	M	67	25.7	L	SMO, SA, MCS, GR, SE	3	2	33.6
4	M	33	23.1	L	SMO, MDCO, SE	2	0	24
5	F	60	28.7	R	SMO, SA, GR, SE	2	0	24
6	M	60	26	R	SMO, MDCO, SE, GR	2	1	28.8
7	M	53	31.6	L	Dome, SA, GR, SE, DCMO	3	3	24
8	M	52	26.4	L	SMO, MDCO, SE	2	1	27.6
9	M	65	25.4	R	SMO, SA, GR, SE	2	1	30
10	M	53	24.1	R	SMO, MDCO, GR, SE	2	1	26.4
11	M	58	21	L	SMO, SA, GR, SE	3	1	34.8
12	M	29	23.8	R	SMO, MDCO, SE	2	0	24
13	M	30	23.3	R	Dome, MDCO, GR	2	0	24

Abbreviations: AD, arthroscopic debridement; BMI, body mass index; DCMO, distal chevron osteotomy for hallux valgus correction; Dome, dome supramalleolar osteotomy; F, female; FU, follow-up; GR, gastrocnemius recession; L, left; M, male; MCS, medial column stabilization using flexor hallucis tendon transfer; MDCO, medial displacement calcaneal osteotomy; R, right; SMO, medial open wedge supramalleolar osteotomy; SA, subtalar arthrodesis; SE, spur excision.

a

All patients included in the study cohort did not have a history of ankle fracture or syndesmotic injury.

b

Grade represents the radiographic arthritis grade as described in the Figure 3.

Radiographic Assessment

All patients completed the preoperative and minimum 2-year postoperative radiographic evaluation, including weightbearing ankle anteroposterior (AP), lateral, hindfoot alignment view, and whole limb radiographs as described by Haraguchi and colleagues ⁷ (Figure 2). In the ankle AP view, the TT, medial distal tibial angle (MDTA), talar center migration (TCM), and angle between the tibial plafond and ground surface (TP-GS) were measured.^1,19,29 The TCM was defined as the distance between the tibial axis and the center point of the talus, as described previously. ²⁹ This was considered positive if the center of the talus was medial relative to the tibial axis. The TP-GS angle was defined as an angle between the tibial plafond and the weightbearing ground surface. It was considered positive when the tibial plafond slope was angled in varus relative to the ground surface. In the lateral ankle view, the anterior distal tibial angle (ADTA) and Meary angle (lateral talo–first metatarsal angle) were measured.^1,30 In the hindfoot alignment view, the hindfoot alignment angle (HAA) and hindfoot moment arm (HMA) were measured. ²³ Both HAA and HMA were considered negative when the hindfoot alignment was in valgus, determined when the lowest point of the calcaneus lies lateral to the tibial axis. In the whole limb radiograph, mechanical axis deviation (MAD) was measured as previously described ²⁰ and was considered positive with varus lower limb alignment.

OPEN IN VIEWER

Figure 2.

Demonstration of radiographic measurements. (A) TT (talar tilt angle), MDTA (medial distal tibial angle), TCM (talar center migration), and TP-GS (angle between the tibial plafond and ground surface) were measured in weightbearing anteroposterior ankle view. Measurement of the TT_WBCT and TP-GS_WBCT is shown in the box. (B) HAA (hindfoot alignment angle) and HMA (hindfoot moment arm) were measured in hindfoot alignment view. (C) ADTA (anterior distal tibial angle) and the Meary angle (lateral talo–first metatarsal angle) were measured in weightbearing lateral ankle view. (D) MAD (mechanical axis deviation) was measured in the whole lower limb radiographs. WBCT, weightbearing computed tomography.

In addition, a radiographic grading system for ankle arthritis was used to evaluate the effectiveness of joint preservation surgery. The stage of arthritis and talar tilt was graded using weightbearing ankle AP radiographs according to the modification of the Takakura classification system for varus ankle arthritis ²⁵ : grade 0, parallel joint, no tibiotalar tilt, and no signs of arthritis; grade 1, parallel joint, no tibiotalar tilt, but signs of subchondral sclerosis or osteophyte formation; grade 2, tibiotalar tilt with valgus alignment without subchondral bone contact; grade 3, tibiotalar tilt with valgus alignment with subchondral bone contact; and grade 4, loss of total joint space with total subchondral bone contact (Figure 3).

OPEN IN VIEWER

Figure 3.

Radiographic arthritis grade using the modified Takakura classification. Grades 0 to 3 are demonstrated, as no patients in the present study had grade 4.

Surgical Procedures

Realignment surgery aimed to shift the weightbearing load from the lateral to the medial aspect of the ankle joint by straightening the hip-knee-ankle-hindfoot axis. To accomplish this, preoperative radiographic and clinical evaluation was performed to determine the type of procedure indicated. The morphology of deformity was determined by referencing radiographic parameters, including the lower limb mechanical axis (MAD), angulation of distal tibial plafond (MDTA), and hindfoot alignment (HAA and HMA). Intraoperatively, the integrity of the deltoid ligament was reconfirmed by a manual valgus stress test under fluoroscopy. There were no ankles with further widening of the medial clear space as seen in the preoperative valgus stress test (Figure 4).

OPEN IN VIEWER

Figure 4.

Varus and valgus ankle stress test. (A) A preoperative standing anteroposterior radiograph shows valgus ankle arthritis. (B) A preoperative varus stress revealed the reducibility of the tibiotalar joint. (C) A valgus stress test did not show further widening of the medial joint space, and the talar tilt angle in the stressed condition was similar to that of the standing radiograph in (A).

A medial opening wedge supramalleolar osteotomy of the tibia and fibula was performed in 11 ankles with a varus lower limb mechanical axis and varus distal tibial plafond. In 2 ankles with a varus lower limb mechanical axis and valgus tibial plafond, a varization supramalleolar dome osteotomy was performed to neutralize the distal tibial plafond. Concomitant hindfoot correction was performed in all ankles, with 8 medial displacement calcaneal osteotomies (MDCOs) and 5 subtalar arthrodeses. The choice of hindfoot procedure was determined by the rigidity and severity of the deformity, and WBCT findings such as sinus tarsi bony impingement or subfibular impingement favored subtalar arthrodesis. The types of concurrent procedures performed at the time of realignment surgery are listed in Table 1. Postoperatively, the foot was placed in a short-leg cast for 6 weeks, and progressive weightbearing was permitted. In general, full weightbearing without any external support was permitted at the 11th week postoperatively.

Results

Pre- and postoperative radiographic and WBCT measurements are tabulated in Table 2.

OPEN IN VIEWER

Table 2.

Comparison of Pre- and Postoperative Radiographic Parameters.

	Preoperative		Postoperative		P value
	Mean ± SD	95% CI	Mean ± SD	95% CI
Ankle AP view
TT	5.5 ± 2.4	4.1-6.9	1.7 ± 2.0	0.4-2.9	<.001
MDTA	88.4 ± 3.1	86.5-90.3	92.2 ± 3.2	90.3-94.2	<.05
TCM	2.0 ± 2.4	0.6-3.5	–0.1 ± 2.9	–1.9 to 1.6	<.05
TP-GS	4.5 ± 2.6	2.9-6.1	0.3 ± 2.6	–1.3 to 1.9	<.05
Ankle lateral view
ADTA	77.0 ± 3.6	74.8-79.2	77.6 ± 4.1	75.1-80.1	.338
Meary angle	10.4 ± 12	3.1-17.6	3.8 ± 5.9	0.3-7.3	.123
Hindfoot alignment view
HAA	–5.5 ± 5.9	–9.1 to 2.0	2.8 ± 3.0	1.0-4.6	<.001
HMA	–7.0 ± 7.0	–11.2 to 2.7	–1.0 ± 4.9	–3.9 to 1.9	<.05
Whole lower limb view
MAD a	1.1 ± 1.2	0.3-1.8	0.8 ± 1.3	–0.2 to 1.7	.343
WBCT
TTWBCT	7.2 ± 1.8	6.1-8.2	3.7 ± 2.3	2.3-5.1	<.001
TP-GSWBCT	9.1 ± 4.0	6.7-11.5	1.7 ± 2.3	0.4-3.1	<.001

Abbreviations: ADTA, anterior distal tibial angle; AP, anteroposterior; HAA, hindfoot alignment angle; HMA, hindfoot moment arm; MAD, mechanical axis deviation; MDTA, medial distal tibial angle; Meary angle, lateral talo–first metatarsal angle; TCM, talar center migration; TP-GS, angle between the tibial plafond and the ground surfaces; TT, talar tilt angle; WBCT, weightbearing computed tomography.

a

Postoperative whole lower limb radiographs were available in 10 patients.

Overall, the preoperative TT was 5.5 ± 2.4 (95% CI, 4.1-6.9) degrees. The MDTA and TCM of the study cohort were 88.4 ± 3.1 (95% CI, 86.5-90.3) degrees and 2.0 ± 2.4 (95% CI, 0.6-3.5) mm, respectively, indicating a varus distal tibial plafond and medial position of the talus relative to the tibial axis. The preoperative MAD was 1.1 ± 1.2 (95% CI, 0.3-1.8) degrees, indicating the varus lower limb mechanical axis. The Meary angle, HAA, and HMA were 10.4 ± 12 (95% CI, 3.1-17.6) degrees, –5.5 ± 5.9 (95% CI, −9.1 to 2.0) degrees, and −7 ± 7 (95% CI, −11.2 to 2.7) mm, respectively, indicating a lower medial longitudinal arch with valgus hindfoot alignment in the cohort.

Postoperatively, all except 1 patient demonstrated improvement in the TT. The TT improved to 1.7 ± 2 (95% CI, 0.4-2.9) degrees (P < .001). There were significant improvements in the MDTA, TCM, HAA, HMA, and TP-GS (P < .05) (Table 2). A 2-year postoperative whole limb alignment radiograph was available in 10 patients, and the MAD decreased to 0.8 ± 1.3 (95% CI, −0.2 to 1.7) degrees, but this improvement did not reach statistical significance.

The preoperative radiographic arthritis grade was grade 2 in 10 patients (77%) and grade 3 in 3 patients (23%). Postoperatively, this changed to grade 0 in 4 patients (31%), grade 1 in 6 (46%), grade 2 in 2 (15%), and grade 3 in 1 (8%). The arthritis grade improved in 12 patients (92%) (Figures 5 and 6), while 1 (8%) remained unchanged in the same grade.

OPEN IN VIEWER

Figure 5.

(A) Whole limb radiograph and (B) pre- and (C) postoperative radiographs of a 65-year-old male. (A) The whole limb radiograph shows varus lower limb mechanical axis. (B and C) Radiographs demonstrate improved radiographic arthritis grade following supramalleolar osteotomy and subtalar arthrodesis. Note restoration of the tibiotalar joint space in the weightbearing computed tomography images (box in B and C).

OPEN IN VIEWER

Figure 6.

Anteroposterior and lateral radiographs of a 69-year-old male with valgus ankle arthritis. (A) Preoperative and (B) 30 months postoperative ankle standing radiographs show an improved talar tilt angle and restoration of the joint space in the weightbearing computed tomography (WBCT) images (box in each image). (A) Note that the arthritis appreciated in the WBCT is more severe than that seen in the plain radiograph.

In WBCT analysis, preoperative TT_WBCT and TP-GS_WBCT were 7.2 ± 1.8 (95% CI, 6.1-8.2) and 9.1 ± 4 (95% CI, 6.7-11.5) degrees, respectively, which improved to 3.7 ± 2.3 (95% CI, 2.3-5.1) and 1.7 ± 2.3 (95% CI, 0.4-3.1) degrees postoperatively with statistical significance (P < .001). There were significant differences between the TT and TP-GS measured in plain radiographs and WBCT (P < .05).

The intraobserver reliability ranged from 0.93 to 0.96, and the interobserver reliability ranged from 0.98 to 0.99 (Table 3). The weighted kappa coefficient of the interobserver reliability for the radiographic arthritis grade was 0.89 (95% CI, 0.75-1).

OPEN IN VIEWER

Table 3.

Interobserver and Intraobserver Reliability of the Radiographic Parameters.

	Reliability (95% CI)
Measurement	Interobserver	Intraobserver
Ankle anterior view
TP-GS	0.92 (0.82-0.96)	0.98 (0.96-0.99)
TP-GSWBCT	0.96 (0.89-0.98)	0.99 (0.97-0.99)
TTWBCT	0.93 (0.84-0.97)	0.98 (0.91-0.99)

Abbreviations: TP-GS, angle between the tibial plafond and ground surface; TT, talar tilt angle; WBCT, weightbearing computed tomography.

Clinical scores improved in all patients. The mean VAS improved from 7.3 ± 1 (95% CI, 6.7-7.9) to 2.5 ± 1.5 (95% CI, 1.6-3.5; P < .05), and the mean FFI improved from 57.7 ± 9.4 (95% CI, 52-63.4) to 18.6 ± 15.7 (95% CI, 9.1-28.2; P < .001). However, in the 1 patient with an unimproved radiographic grade (case 7, grade 3) (Figure 7), the VAS and FFI at the final follow-up were 7 and 56.5, respectively. This patient continued to be monitored to identify any worsening outcomes. There were no notable complications related to the surgery, and 6 patients underwent hardware removal due to irritation symptoms over the medial malleolus. None of the patients underwent joint-sacrificing procedures such as total ankle arthroplasty or ankle arthrodesis during the study period.

OPEN IN VIEWER

Figure 7.

(A) Pre- and (B) postoperative radiographs of a 53-year-old male, whose radiographic arthritis grade did not improve following supramalleolar dome osteotomy and subtalar arthrodesis.

Discussion

This study investigated the radiographic characteristics of valgus ankle arthritis that are not associated with deltoid insufficiency and presented outcomes of realignment surgery. Interestingly, most ankles demonstrated varus distal tibial plafond or varus lower limb mechanical axis. In contrast, valgus hindfoot alignment was observed in all patients. Realignment surgery, including supramalleolar osteotomies and concurrent heel alignment correction, was performed to straighten the hip-knee-ankle-hindfoot axis. All except 1 patient demonstrated clinical and radiographic improvements without significant procedure-related complications at a minimum 2-year follow-up.

While supramalleolar realignment has been regarded as the principal method of joint preservation surgery in asymmetric varus ankle arthritis, the treatment of primary valgus ankle arthritis described in the literature has focused mainly on the correction of inframalleolar causes, specifically deltoid ligament insufficiency.^2,6,22 Furthermore, the utility of supramalleolar correction has only been described in the correction of posttraumatic valgus ankles. ¹⁵ The rationale behind this approach lies in the long-held belief that insufficiency of the deltoid ligament from hindfoot valgus or posterior tibial tendon dysfunction may be a leading cause of primary valgus ankle arthritis as the deltoid ligament is the main structure resisting valgus talar tilt.^2,3 However, emerging evidence has shown that malalignment of the lower limb mechanical axis is associated with talar tilt of the ankle joint.^10,17 Nonetheless, there has been a lack of literature analyzing the morphology of lower limb alignment and the results of coronal plane alignment correction in the primary valgus ankle arthritis.

In our series, the mean Meary angle was 10.4 degrees, with only 2 patients with a Meary angle greater than 30 degrees, suggesting severe flatfoot deformity. On preoperative examination, the inversion power was well preserved in all patients. Only 3 patients had tenderness over the posterior tibialis tendon, which was revealed to be relatively healthy in direct visualization. Furthermore, no patients demonstrated opening of the medial joint space in valgus stress radiographs, which led the authors of the current study to believe that there may be another mechanism other than deltoid ligament insufficiency in the development of valgus ankle arthritis in these patients. The improvement in the TT further supports this hypothesis as the realignment surgery was performed without concurrent deltoid ligament reconstruction.

Lower limb alignment analysis revealed varus lower limb mechanical axis and varus distal tibial plafond in most ankles. Several studies investigating radiographic changes TKA have described a causal relationship between changes in the lower limb mechanical axis and ankle or hindfoot alignment.^10,17,21 In their radiographic analysis of lower limb and hindfoot alignment in TKA patients, Norton and colleagues ²¹ found that valgus hindfoot alignment was associated with varus knee deformity, possibly from compensation at the subtalar joint. Furthermore, Jeong and colleagues ¹⁰ demonstrated that correction of varus deformity of the lower extremity with TKA shifted both ankle and hindfoot alignment into varus. In light of these findings, the authors of the current study deduced that valgus ankle arthritis in the study cohort may be associated with a compensatory phenomenon to varus angulation of the mechanical axis and the distal tibial plafond. In order to restore alignment, a medial open wedge supramalleolar osteotomy was performed to correct the varus mechanical axis. The distal tibial plafond was corrected into either neutral or slightly valgus to avoid eccentric lateral side loading from overcorrection.

While the majority of patients exhibited varus lower limb mechanical axis and varus distal tibial plafond, 2 patients had valgus distal tibial plafond with either varus or neutral lower limb mechanical axis. In these patients, varization supramalleolar dome osteotomy was performed. Performing medial closing wedge supramalleolar osteotomy can be advocated when only angulation of the distal tibia is considered. However, because the medial closing wedge can translate the ankle medially, it can aggravate or shift lower limb alignment into varus. Thus, dome osteotomy was performed in those patients to correct deformity at the distal tibial plafond while not increasing the MAD.

In addition to supramalleolar correction, heel alignment correction was performed in all patients with either MDCO or subtalar arthrodesis, depending on the severity of the deformity. Dynamic medial column stabilization by transfer of the FHL to the base of the first metatarsal was performed for flatfoot correction in 2 patients with a severe deformity in addition to the coronal plane realignment. ¹¹

Radiographic analysis revealed that the deformity observed in WBCT was greater than that observed in plain radiographs. This stands to reason why the authors included ankles with TTs larger than 2 degrees in this study, while some previous studies have set a higher threshold of 4 degrees for asymmetric arthritis. For example, 1 ankle in this series had a TT of 3 degrees in the plain radiograph, while the TT_WBCT was measured to be 7 degrees (Figure 8). An arthroscopic investigation revealed narrowed lateral joint space and cartilage degeneration. Given these findings, we believe the actual eccentric narrowing at the tibiotalar joint can be greater than is observed in plain radiographs and therefore recommend evaluating patients with WBCT to capture the degree of deformity more accurately.

OPEN IN VIEWER

Figure 8.

Simple anteroposterior (AP) ankle radiograph and weightbearing computed tomography (WBCT) images. (A) In the preoperative simple AP radiograph, the talar tilt angle is found to be 3 degrees, while WBCT images reveal eccentric narrowing of the joint together with cyst formation. (B) The joint space was restored following realignment surgery.

One ankle with unimproved radiographic arthritis grade had a large preoperative TT of 9.5 degrees. Although the threshold for our approach with joint-preserving supramalleolar correction cannot be determined from the current study, unsatisfactory outcomes or the possibility of conversion to joint-sacrificing procedures may be a risk and should be explained to patients at the preoperative visit, especially when treating ankles with a large preoperative TT.

This study is not without limitations. First, this was a retrospective analysis with a relatively small number of patients. Therefore, this study could not provide an optimal target angle for correction nor determine a cutoff value for the TT when adopting this approach. However, this study was the first clinical series describing outcomes of realignment surgery in primary valgus ankle arthritis without deltoid ligament insufficiency. Second, it is difficult to know the cause-and-result relationship or the outcome of specific surgical techniques due to inhomogeneity of the treatment and different morphologies in the cohort. However, we believe that this study will stimulate conversations about the role of lower limb malalignment in the development of valgus ankle arthritis, which does not appear to be caused by deltoid ligament insufficiency. Lastly, although the authors included ankles with a minimum 2-year clinical and radiographic follow-up, future studies should focus on collecting longer-term outcomes and survival analysis to determine the efficacy of these joint-preserving procedures.

Conclusion

This study evaluated outcomes of realignment surgery in valgus ankle arthritis that is not associated with deltoid ligament insufficiency. Radiographic analysis revealed varus mechanical axis and varus distal tibial plafond in most ankles. Realignment surgery, including supramalleolar osteotomies, which straightens the mechanical axis and decreases the slope of the tibial plafond, may be a reasonable approach in joint preservation of valgus ankle arthritis without deltoid ligament insufficiency.

Supplemental Material