Study design and participants
This study was approved by the Medical Ethics Committee of the University of Tsukuba (no. R05-093) and was conducted in accordance with the Declaration of Helsinki. The requirement for informed consent was waived because of the retrospective study design.
All patients who underwent hybrid closed-wedge HTO for varus knee osteoarthritis performed by the same knee surgeon between March 2019 and March 2022 were considered for inclusion in this study. Patients who could not be followed up or did not undergo postoperative radiological evaluation were excluded. Information on patient characteristics, including sex, age at the time of surgery, height, body weight, body mass index, smoking status, and diabetes mellitus, was obtained from medical records. Surgical findings, specifically correction distance, were recorded. In addition, the occurrence of reoperation was investigated.
Surgical procedure and postoperative rehabilitation
Hybrid closed-wedge HTO was performed according to the method described by Takeuchi et al. [3]. Preoperative planning was performed using full-limb standing radiography. The mechanical axis was determined by drawing a line from the center of the femoral head to the center of the tibial plafond. The ratio of the distance between the point where the mechanical axis passed and the medial edge of the tibia to the distance of the width of the tibial plateau was defined as the mechanical axis percentage (%MA) [6]. The target %MA for surgical planning was 62.5% [7].
A pneumatic thigh tourniquet was applied to the operative leg. Arthroscopic examination was performed to verify the status of the articular cartilage and menisci before HTO in all cases. After arthroscopy, fibular osteotomy was performed, removing approximately 10–20 mm in length of bone to allow for tibial correction through a separate incision. Subsequently, tibial osteotomy was performed. The hinge point was set at three-quarters of the osteotomy line from the lateral side, dividing the proximal tibial osteotomy line by approximately 1:3. Osteotomy was performed along the proximal and distal osteotomy lines, from the proximal lateral cortex to the hinge point, using a bone saw and chisel. A separate ascending biplanar osteotomy cut was made behind the patellar tendon insertion in the frontal plane, retaining the tibial tuberosity at a thickness of approximately 10 mm. After removing the lateral closed-wedge bone block, the medial cortex was cut according to the proximal osteotomy line. Finally, the medial side was opened, the lateral side was closed, and plate fixation was performed using a TriS locking plate (Olympus Terumo Biomaterials, Japan).
All patients underwent the same rehabilitation protocol. Joint range of motion exercises were started on the day after surgery. One-third partial weight-bearing was started at 3 weeks postoperatively, one-half partial weight-bearing was started at 6 weeks postoperatively, two-thirds of partial weight-bearing was started at 9 weeks postoperatively, and full weight-bearing was started at 3 months postoperatively. Thereafter, the patient was instructed to walk with a T-cane if pain persisted while walking.
Radiographic and clinical evaluations
%MA was measured preoperatively and at 1 year postoperatively on full-length standing lower extremity radiographs. The medial proximal tibial angle (MPTA) [8] was measured in the anteroposterior view of the knee preoperatively and at 1 week and 1 year postoperatively. The posterior tibial slope (PTS) [9] was measured in the lateral view of the knee preoperatively and at 1 week and 1 year postoperatively.
Loss in radiographic variables was defined as the numerical difference between values obtained at 1 week and 1 year postoperatively (specifically, the value at 1 year postoperatively minus the value at 1 week postoperatively). Correction failure was defined as a loss of ≥ ± 3° in the MPTA or PTS [9, 10].
CT images obtained at 3 months, 6 months, and 1 year postoperatively were evaluated to assess bone union using Osirix imaging software (Pixmeo, Geneva, Switzerland).
Coronal and sagittal slices were defined as those connecting the midpoint of the respective osteotomy line to the midpoint of the trabecular bone width at the height of the most distal screw (Fig. 1). Medial, hinge, and lateral sites were evaluated using coronal images, whereas the anterior flange and posterior sites were evaluated using sagittal images. Bone union was investigated using the Tomographic Union Score, as previously described [11]. Based on this score, the callus was classified into the following four categories: no callus (Fig. 2a,c), discontinuous callus (Fig. 2d), continuous bridging immature callus (Fig. 2e), and continuous remodeled callus (Fig. 2b,f). Continuous bridging immature callus and continuous remodeled callus were classified as bone union [12]. The absence of callus or discontinuous callus at any of the above five sites at three months postoperatively was defined as delayed bone union.
Two orthopedic surgeons (N. K. and N. A.) with specialized training assessed all radiographs on two separate occasions, with a time interval of 6 weeks between evaluations. The inter- and intra-observer reliabilities of the assessments were determined using the intraclass correlation coefficient, and the 95% confidence intervals for both inter- and intra-observer reliabilities exceeded 0.85.
As a clinical evaluation, the Japanese Orthopedic Association (JOA) score was assessed preoperatively and at 1 year postoperatively [13].
Statistical analysis
Continuous variables are presented as the mean ± standard deviation. The normality of data distribution was assessed using the Shapiro-Wilk test. For continuous variables, the paired t-test and Wilcoxon signed-rank test were used to compare variables before and after surgery when parametric assumptions were met and not met, respectively; the Pearson’s chi-squared test was used to compare qualitative variables.
Group differences (delayed bone union group vs bone union group) were evaluated using the Student's t-test for normally distributed variables, and the Mann-Whitney U test for non-normally distributed continuous variables. To investigate the factors associated with delayed bone union, a binomial logistic regression analysis was performed, with delayed bone union as the dependent variable and sex, age, height, body weight, body mass index, correction distance, smoking, and diabetes mellitus as independent variables.
Statistical analyses were conducted using SPSS software (version 26.0; IBM Corp., Armonk, NY, USA). Statistical significance was defined as p < 0.05.