2.1

The institutional review board of our hospital approved this study, and all patients signed written informed consent for inclusion in the review. This retrospective study prospectively followed 14 consecutive patients (15 affected ankles) who underwent SMO/AA for treating end-stage ankle arthritis with large varus deformity from April 2019 to June 2023. The first author (H.K.) with more than 25 years of experience in foot and ankle reconstruction performed these procedures.

2.2

The combined procedure was considered suitable for patients with (1) intractable ankle pain and/or instability refractory to conservative treatment, (2) end-stage varus ankle osteoarthritis, (3) varus deformity of the hindfoot of ≥ 15° on the weightbearing hindfoot alignment (HA) radiographs, (4) relatively undamaged subtalar joint, and (5) preserved leg-foot range of motion. Conservative treatment, including medication, lateral wedge plantar orthosis, and an ankle brace, failed to relieve the symptoms for ≥ 6 months. The analysis excluded patients who underwent other concurrent hindfoot surgeries or followed for < 2 years. Importantly, patients who smoke were excluded to prevent the increased risk of postoperative infection and delayed bone healing.

This study followed up five male and nine female patients with a mean age of 68.3 (52–79) years for a mean duration of 3.4 (2–6.2) years. The mean body mass index was 24.8 (18–36.4). Of the 14 patients, 12 had ≥ 1 comorbidities, including hypertension in 9, type 2 diabetes in 3, obesity in 3, Parkinson disease in 1, and putaminal hemorrhage in 1 patient. The etiologies of ankle arthritis consisted of chronic ankle sprain and instability in seven, plafond fracture in two, primary osteoarthritis without a history of trauma in two, malleolar fracture in one, paralytic foot in one, and foot dystonia in one patient. No patient had undergone previous operative treatment in the affected lower limb; however, one patient underwent contralateral total knee arthroplasty for medial gonarthrosis three years before SMO/AA. One patient underwent bilateral surgeries at a 4-month interval.

Clinical outcome measures included a 10-cm visual analog scale (VAS) for pain, the Japanese Society for Surgery of the Foot (JSSF) ankle–hindfoot scale ^, , and the Self-Administered Foot Evaluation Questionnaire (SAFE-Q) . The validity, reliability, and responsiveness of the JSSF scale and SAFE-Q have been verified ^,, . The main body of the SAFE-Q consists of 34 questionnaire items, providing five subscale scores of pain and pain-related, physical functioning and daily living, social functioning, shoe-related, and general health and wellbeing. Each subscale score ranges from 0 to 100 points. Other clinical results included leg-foot range of motion, osteotomy wedge height, time to full weightbearing, and possible postoperative complications. Leg-foot range of motion was measured with a goniometer applied along the lateral border of the leg and foot. These clinical assessments were conducted preoperatively, 6 months and 12 months postoperatively, and annually thereafter. In the present study, the preoperative and most recent follow-up clinical outcome variables were compared.

Weightbearing anteroposterior, lateral, and HA radiographs were obtained in each case, and all ankles exhibited joint space obliteration with partial or complete tibiotalar contact. The severity of ankle osteoarthritis was assessed with the Takakura–Tanaka classification system ^, . Arthritic changes in the subtalar joint were assessed with the Morrey–Wiedeman classification, graded as zero (normal ankle), one (small osteophytes and minimal joint narrowing), two (moderate osteophytes and moderate joint narrowing), and three (significant joint narrowing with joint deformation or fusion) . Further, multiplanar reconstruction computed tomography (CT) scans and magnetic resonance images were used to aid preoperative assessment of the subtalar joint ( Fig. 1 ). We believe that patients with preoperative Grade 3 arthritis may be beyond the limits of this subtalar joint–sparing procedure. The radiographic outcome variables included tibial anterior surface (TAS) angle, talar tilt (TT) angle, and medial talar center migration (MTCM) in weightbearing anteroposterior radiograph; tibial lateral surface (TLS) angle, talo-first metatarsal (TMT) angle, calcaneal pitch (CP) angle, and anterior talar center migration (ATCM) in weightbearing lateral radiograph. HA angle and hindfoot moment arm (HMA) were measured on the HA radiograph and were represented as positive numbers in varus-oriented hindfoot ^, . HA angle of ≥ 15° is considered a large varus deformity of the ankle/hindfoot. These radiographic assessments were conducted preoperatively, monthly postoperatively before bone union confirmation, 6 months and 12 months postoperatively, and annually thereafter. The most recent follow-up radiographs were analyzed for preoperative ankle/hindfoot deformity correction. As a special exception, preoperative Morrey–Wiedeman grade assessment and postoperative MTCM and ATCM measurements were conducted 1–2 months after SMO/AA for more accuracy and certitude ( Fig. 2 ). Rapideye Core software (Toshiba Medical Systems, Tokyo, Japan) was employed as the image measurement device. The absence of local pain during ambulation indicates a clinically successful bone union and the presence of bridging bone across the operative site and intact osteosynthesis material denotes radiographical success. Delayed union is a bone union time of > 6 months, whereas nonunion is a bone union time of > 9 months . To avoid potential bias, an independent observer (Y.S.), who was blinded to the study objects, performed all clinical and radiographical assessments.

Coronal (A) T1-weighted and (B) T2-weighted magnetic resonance images, and (C) computed tomography scan showing end-stage ankle arthritis with a large varus deformity. Note the relatively undamaged subtalar joint.

(A) Anteroposterior, (B) lateral, and (C) hindfoot alignment radiographs demonstrating radiographic variables. Hindfoot alignment angle and hindfoot moment arm are presented as positive numbers when the hindfoot is varus-oriented. Abbreviations: MTCM: medial talar center migration; TMT: talo-first metatarsal (angle); CP: calcaneal pitch (angle); ATCM: anterior talar center migration; HA: hindfoot alignment (angle); HMA: hindfoot moment arm.

2.3

The previously described operative technique was modified . Surgery was performed with the patient under general anesthesia, and intravenous patient-controlled analgesia was administered for postoperative pain control. The patient was placed in the lateral decubitus position, with a pneumatic tourniquet applied to the thigh. The surgeon stood on the dorsal side and the fluoroscopy unit was positioned on the ventral side of the patient. A 10-cm longitudinal incision was created along the fibula from the tip of the lateral malleolus. An oscillating saw was used to remove an 8-cm section of the distal fibula for a bone graft. The anterior and posterior capsules were arthrotomized, and the joint was exposed with a small pin-based distractor. The articular cartilage was removed utilizing a metal cutting burr and curettage, and multiple holes were drilled using a diamond burr while irrigating frequently to prevent thermal necrosis. During AA, we tried to improve bone fusion by increasing the contact area between the tibia and talus, rather than correcting varus deformity. Coronal plane deformity was mainly corrected at the osteotomy site, whereas sagittal and transverse plane deformities were mainly corrected at the arthrodesis site. An autogenous cancellous bone graft from the resected fibula was packed into the arthrodesis site. A 3.0-mm Kirschner wire (K-wire) was inserted via the plantar aspect of the calcaneus, passing through the talus, and into the tibia, for temporal fixation.

The hemiwedge SMO aimed to correct the varus angulation in the coronal plane and, thus, obtain a parallel ankle joint line (i.e., TAS angle of 90°). Under fluoroscopy, the first proximal 2.0-mm K-wire was inserted obliquely, 2.5 cm above the ankle joint, while leaving space to accommodate the arthrodesis and osteotomy sites for plate fixation. The hinge point was identified by dividing the proximal osteotomy line by approximately 2–1 from lateral to medial. Our goal is a TAS angle of 90°. The distal osteotomy line was determined with a goniometer according to the preoperative planning, and the second distal 2.0-mm K-wire was inserted ( Fig. 3 A). A lateral-based bone wedge was removed with an oscillating saw and thin chisels, and the proximal osteotomy was completed ( Fig. 3 B). The osteotomy gap was then closed, and a 3.0-mm K-wire was advanced proximally to the osteotomy site. A single lateral locking plate was applied to secure both osteotomy and arthrodesis sites ( Fig. 3 C). TomoFix anatomical plate (for the ipsilateral side; DePuy Synthes, West Chester, PA) was employed to treat 12 ankles, and Lateral Tibiotalar Fusion Plate (Arthrex, Naples, Florida) was applied to 3 ankles in this study. Proper ankle alignment and plate position were confirmed with intraoperative radiographs ( Fig. 3 D). Four locking screws were inserted proximal to the osteotomy, and three screws were inserted into the talus to stabilize the plate ( Fig. 4 ). A closed suction drain was placed on the plate and removed the following day. The lateral malleolus periosteum was sutured to cover the plate, and the soft tissues were restored.

(A) Intraoperative radiographs demonstrating the first proximal 2.0-mm Kirschner wire (K-wire) inserted obliquely 2.5 cm above the ankle joint while leaving a space to accommodate the arthrodesis and osteotomy sites for plate fixation. The hinge point was identified by dividing the proximal osteotomy line by approximately 2–1 from lateral to medial. The distal osteotomy line was determined with a goniometer according to the preoperative planning, and the second distal K-wire was inserted. (B) A lateral-based bone wedge was removed with an oscillating saw and thin chisels, and the proximal osteotomy was completed. (C) The osteotomy gap was closed, and a 3.0-mm K-wire was advanced proximally to the osteotomy site. Both the osteotomy and arthrodesis sites were secured with a single lateral locking plate. (D) Proper alignment of the ankle and plate position was confirmed with intraoperative radiographs.

Intraoperative photograph demonstrating four locking screws inserted proximal to the osteotomy and three screws inserted into the talus to stabilize the plate. Note the osteotomy site (arrow) and the arthrodesis site (arrowhead).

2.4

The patient’s leg was immobilized using a nonweightbearing short cast for 4 weeks postoperatively. Subsequently, they initiated progressively increased weightbearing in a fracture boot for another 2 weeks. Full-weight double-crutch walking was frequently permitted as tolerated after 6 weeks. The fracture boot was removed and stair exercises with a small cane were prescribed after 15 weeks. The patients resumed normal activities with regular shoes approximately 4 months postoperatively.

2.5

Nonparametric statistics were used because of the small number of patients and skewed distribution of most variables. Statistical analysis of the descriptive data was conducted using the Wilcoxon signed-rank test, with a p-value of < 0.05 indicating statistical significance. Statcel5 package (OMS Publishing Inc., Tokyo, Japan) was used for raw data analysis.