Comparison of Radiological and Clinical E cacy of Normal and Overcorrected Varus Deformity After Medial Open Wedge High Tibial Osteotomy: A Two- Year Follow-up Investigation

Kui-shuai XU The A liated Hospital of Qingdao University Liang ZHANG The A liated Hospital of Qingdao University Rui SHEN A liated hospital of Qingdao University Jie SUN Qilu Hospital of Shandong University Qingdao Jian WANG qingdao 971 Naval hospital Xia ZHAO the a liated hospital of qingdao university Yi ZHANG a liated hospital of qingdao university Teng-bo YU (  ytb8912@hotmail.com ) The A liated Hospital of Qingdao University https://orcid.org/0000-0003-1134-0134


Introduction
Open wedge high tibial osteotomy (OWHTO) has become a common treatment for medial compartment arthritis of the knee in recent years [1][2][3][4].The key factor for the long-term success after OWHTO is the transfer of the lower limb line and load-bearing axis to the outside, so as to achieve the uniform distribution of knee mechanical load [2,11,12].The selection of force line correction point is the key to this operation. It has been reported that the force line after surgical correction can obtain the best effect by Fujisawa point, which is 62.5% point on the lateral tibial plateau [3,9,26]. For patients with severe varus deformity of the knee, intraoperative osteotomy Angle needs a lot of correction, according to the multiple stages of knee osteoarthritis, according to different degrees of degeneration of the medial interventricular cartilage injury of the knee, the implementation of different force line orthopedic programs. After the follow-up found that some patients OWHTO exists in lower limb power line corrective postoperative WBLR (67% or higher), to evaluate the short-term clinical e cacy and postoperative complications in patients with such and adjacent joint compensatory changes, therefore, this study by imaging and related clinical function score, to explore the diagnosis and treatment OWHTO excessive correct varus deformity the short-term curative effect of the medial compartment knee osteoarthritis. Until now, compensatory changes in hip, patellofemoral and ankle joints after OWHTO overcorrected varus deformity have been rarely studied.
The purpose of this study was to assess whether overcorrection of varus malformation affects early clinical outcomes after OWHTO and to assess compensatory changes in the hip, patellofemoral, and ankle joints. We hypothesized that due to compensatory changes in hip and patella positions and ankle joints, some degree of overcorrection would not adversely affect clinical outcomes after OWHTO.

Materials And Methods
A retrospective analysis was performed on 92 patients with osteoarthritis who underwent initial unilateral OWHTO in A liated Hospital of Qingdao University from February 2016 to August 2019. Inclusion criteria were as follows: Clinical diagnosis was consistent with early anterior medial compartment osteoarthritis of knee; The follow-up time was at least 1 year. No neuromuscular disease; X-ray examination of the knee joint showed clear structure; Accompanied by varus genius deformity <15°(proximal tibia deformity mainly), exion contracture <10°;The range of motion of the knee is >110°.Exclusion criteria: osteoarthritis or in ammatory arthritis in the lateral compartment of the knee; The follow-up time was less than 1 year.OWHTO was performed at the same time as other joint operations; OWHTO was performed on both sides. Incomplete imaging data and missing persons; Knee ligament injury or functional insu ciency; Corrective insu ciency (WBLR≤ 50%); A total of 22 patients were excluded from the study according to the exclusion criteria. Among them, there were 5 patients who received bilateral OWHTO, 7 patients with postoperative loss, 8 patients with incomplete imaging data, and 9 patients with corrective insu ciency (WBLR < 50%). Therefore, the nal study consisted of 63 patients with an average age of (58.32±5.26) years old, including 12 males and 51 females. According to postoperative WBLT, the patients were divided into the following two groups: 52 patients in the normal correction group (50% Page 4/17 ≤WBLR < 67%) and 11 patients in the over-correction group (WBLR≥ 67%). The radiological index measurement, complications and functional recovery of affected limb were compared between the two groups after OWHTO operation. Preoperative and postoperative imaging indexes of all patients were measured by the same person for 3 times at different time periods to reduce errors, and then the mean value was taken. All operations were performed by the same senior medical o cer. All patients agreed to participate in this study and signed informed consent, and the study plan was approved by the ethics Committee of our hospital.

Operative technique
All patients were in supine position during the operation under general anesthesia. Tourniquet was routinely used at the root of the operative side thigh. A longitudinal incision of 6-8cm was made along the medial proximal tibia and gradually exposed to the foot of the goose. Two 3.0mm kirschner needles were inserted horizontally toward the upper tibio bular joint on the outer edge of the tibia, and the direction of kirschner needles was the same as the retrograde direction of the tibia. The ascending and horizontal planes of biplane osteotomy were at an Angle of about 110°. After the completion of osteotomy, the osteotomy Angle needed to be corrected was determined according to the preoperative lower extremity line, and the osteotomy space was opened with bone knife and retainer, so as to avoid lateral cortical fracture by careful operation. After c-arm uoroscopy showed satisfactory force lines, TomoFix locking plate was used for xation, and tranexamic acid injection was administered 20min before surgery. The tourniquet is also applied locally before completion of suturing.

Postoperative rehabilitation
After surgery, the affected limb was lifted, all patients were pressurized with elastic bandage, the drainage tube was clamped for 4 h after surgery, and then drainage was opened. Drainage could be removed if the postoperative drainage volume was less than 50mL.Antibiotics and anticoagulant drugs were used in a standardized manner within 24 hours after the operation. Quadriceps femur exercise and ankle pump exercise were performed within 24 hours after the operation. Partial weight-bearing walking with crutches was performed within 6 weeks after the operation, and full weight-bearing walking was performed after gradually abandoning crutches. The patient was followed up on time after discharge.

Clinical Assessment
All patients received preoperative and nal knee functional follow-up: The American Hospital for Special Surgery score (100 points) : pain 30 points, function 22 points, activity 18 points, muscle strength 10 points, exion deformity 10 points, stability 10 points and subtractive items; the higher the score, the better the function; The Western Ontario and McMaster University Osteoarthritis Index: a 0-4 grading method is used to evaluate 5 aspects of pain, 2 aspects of stiffness, and 17 aspects of di culty in daily activities. A higher score means a worse function. The motion range of the knee was measured using a long arm goniometer. All patients were followed up for at least 1 year, and postoperative complications including subcutaneous congestion, neurovascular injury, lateral hinge fracture, infection, deep venous thrombosis of the lower extremity, delayed or nonunion, and joint stiffness were obtained through outpatient review and telephone follow-up.

Radiological assessment
Radiological measurements included Hip-knee-ankle angle (HKA), Weight-bearing line ratio (WBLR) medial proximal tibial angle (MPTA) lateral distal femoral angle (LDFA), joint line convergence Angle (JLCA), Posterior tibial slope (PTS) Hip abduction angle (HAA), tibial plafond inclination TPI , Talar inclination Angle (TIA) ,Carton index, Lateral patellar tilt(LPT), Lateral patellar shift (LPS), medial patellofemoral space and lateral patellofemoral space PTS were measured using knee lateral radiographs. LPT, LPS, Carton index, and medial patellofemoral space and lateral patellofemoral space were measured using patella 30° axial radiographs, and other radiological parameters were measured using full-length loaded radiographs of the lower extremities. HKA is the included Angle (varus, negative value) formed between the mechanical shaft of the femur and the mechanical shaft of the tibia ( Fig.  2A).WBLR was de ned as the horizontal distance between the load bearing line and the medial edge of the tibial plateau divided by the width of the tibial plateau ( Fig 2B).PTS is de ned as the Angle between the proximal tibial central anatomical axis and the tangent of the tibial plateau ( Fig 2C).JLCA was de ned as the Angle between the tangent of the femoral condyle and the tibial plateau, and the positive value of JLCA represented the opening of the lateral joint (Fig. 3A).MLDFA is de ned as the lateral Angle formed between the mechanical shaft of the femur and the tangent of the femoral condyle (Fig. 3B).MPTA is the medial Angle formed between the mechanical shaft of the tibia and the tangent of the tibial plateau ( Fig  3C).TIA is de ned as the included Angle between the articular surface of the talus and the horizontal line, and the varus is positive [11] (Fig 4A). TPI is the Angle between the tangent and the horizontal of the distal tibia, and varus is positive (Fig.4b).HAA is the included Angle between the mechanical shaft of femur and the straight line perpendicular to the ground, and abduction is de ned as positive value [14] ( Fig. 4C).LPT is the Angle formed by the line between the highest point of the internal and external condyles of the femur and the extension line of the maximum transverse diameter of the patella (Fig.5A).LPS was de ned as a straight line drawn from the highest point of the internal and external condyle of the femur, the measured distance was B, and the lateral edge of the patella was perpendicular to the line (Fig.5B).Make a vertical line through the highest point of the lateral condyle of femur, and the distance between the two perpendicular lines is a, then LPS=A/B. Medial and lateral interspaces of patellofemoral joints [16] are the distance from the center of the joint surface to the patella by drawing vertical lines to measure the medial and lateral interspaces of patellofemoral joints ( Fig 5C).The Carton index is the ratio of the distance between the lower pole of the patella joint surface and the anterior upper Angle of the tibia plateau to the length of the patella joint surface. The normal value is 0.6~1.2 (Fig. 5D).

Statistical analysis
Statistical data were analyzed by SPSS 25.0 software. Normal distribution was detected for the two groups of data, and the measurement data in line with normal distribution was expressed as mean ± standard deviation. Comparison of measurement data using T test or non-parametric test; Statistical data were measured by chi-square test, and the difference was statistically signi cant at P≤0.05.

General patient information and radiological parameters
According to the inclusion and exclusion criteria, 63 patients nally participated in this study. According to postoperative WBLT, 52 patients in the normal correction group (50% ≤WBLR < 67%) and 11 patients in the over-correction group (WBLR≥ 67%) were divided into the following two groups, as shown in Figure   1.63 patients for at least 12 months of follow-up, excessive correct group and normal correction for age, sex, side, body mass index, duration, operation time, bone cutting Angle, and follow-up time knee K -L grade, speci c data are shown in table 1, 2 set of preoperative tibiofemoral Angle, proximal tibial medial Angle, the U.S. special surgical hospital ratings, WOMAC osteoarthritis index there was no signi cant difference compared with signi cance (P & gt;0.05), which is comparable. See Table 2-4 and Table 6 for the speci c data of each group.

Comparison of postoperative radiological parameters between the two groups
After OWHTO, the last HKA was compared between the two groups. The overcorrected group was (7.39±1.61)°, and the normal corrected group was (3.28±1.59)°, with statistically signi cant difference (P=0.000).The last MPTA of the two groups was (94.95±1.01)° in the overcorrected group and (91.1±2.87)° in the normal corrected group, and the difference was statistically signi cant (P=0.001).Comparison of the last HAA between the two groups showed that the overcorrected group was (-2.96±1.58)° and the normal corrected group was (-0.97±2.31)°, with statistically signi cant difference (P=0.026).The degree of postoperative correction of TPI and TIA in the over-corrected group was higher than that in the normal corrected group, but the difference was not statistically signi cant (p > 0.05). The comparison of other imaging indicators showed no statistically signi cant difference. The results are shown in Table 2-4. 3.3Comparison of postoperative complications between the two groups Excessive correct set of postoperative complications occurred in 4 cases, including 1 case with subcutaneous ecchymosis large area, delayed union of fracture in 1 case 1 case incision infection,1 cases of lateral hinge fracture, the incidence of 36.36% Normal correct set of 6 cases of postoperative complications, including 1 case with subcutaneous ecchymosis large area, stiff joints in 1 case, fracture healing, 1 case of incision infection, 1 case, the lateral hinge fractures in 2 cases, the incidence of 11.54%, compared two groups of postoperative complications, difference was statistically signi cant (p = 0.041).For complications, none of the patients underwent a second operation, and after active conservative treatment, all of the patients recovered from these complications. Therefore, we believe that these complications are not related to the nal clinical e cacy, as shown in Table 5. 3.4 Postoperative clinical results were compared between the two groups At the last follow-up, the hospital score of American Special Surgery in the 2 groups was signi cantly increased compared with that before surgery, with signi cant difference (P < 0.05), and WOMAC osteoarthritis index was signi cantly decreased compared with that before surgery, with signi cant difference (P < 0.05).There was no signi cant difference between the hospital score of special surgery and WOMAC osteoarthritis index in the 2 groups (P > 0.05).Shown in table 6.

Discussion
Medial open wedge high tibial osteotomy is generally suitable for patients with medial compartment osteoarthritis of the knee and is an effective joint conservative treatment method [3,4,9]. New alignment of lower limb weight-bearing machinery axle weight is an important determinant of clinical e cacy after medial tibial high osteotomy [5]. Accurate line-of-force correction is the key to the success of OWHTO, and the control of orthopedic Angle is an important risk factor affecting prognosis [6]. Fujisawa et al. [3] believed that when the lower limb line passed through the tibial plateau 60% to 70% after surgery, cartilage damage would not progress, and cartilage growth was the best at 62.5%. Dugdale et al. [7] suggested that the force line after osteotomy should pass through the area near 60% of the lateral tibial plateau on the coronal plane. In order to obtain a better long-term effect, Myrnerts [8] found in the study on the optimal correction of varus deformity in high tibial osteotomy that between 3 and 7 degrees of orthotopic valgus (three-point technique) can lead to the optimal subjective evaluation of the results by patients. Recently, scholars have also con rmed through the nite element analysis of knee joint that the lower limb force line near 62.5% of the medial side of knee joint can obtain the minimum difference in area and pressure value of stress concentration area in the medial and lateral compartment of knee joint, and can obtain a more uniform distribution of force on the tibial surface [9]. Lee analyzed that in 34 patients undergoing medial open wedge high tibial osteotomy, the changes in JLCA may be greater in patients with underlying medial relaxation or severe varus deformity requiring greater correction, which may lead to accidental overcorrection of HTO [10].In this study, the imaging changes and clinical e cacy of OWHTO in overcorrection of varus deformity were studied in patients with excessive lateral displacement of lower limb bearing axis after surgery.
Respectively from two groups of patients in this study TPI (6.23-3.9) ° and preoperative (5.29 + 5.75) °d own to the last follow-up (0.74 + 5.4) ° and (2.46-4.33) °, TIA, respectively from the preoperative (8.3-4.3)° and the (5.75 + 6.05) ° down to the last follow-up (1.95 + 5.71) ° and (2.81-4.62) °, differences were statistically signi cant (p < 0.05), and further veri ed the OWHTO exhibit a lateral ankle positioning levels tend to be more even outer, This conclusion is also consistent with previous literature [11,12]. As for the relationship between HTO's posterior ankle joint and knee joint, Takeuchi et al. also reported the successful treatment of ipsilateral knee and ankle osteoarthritis by HTO [13].In this study, it was found that the degree of postoperative correction of TPI and TIA in the over-corrected group was higher than that in the normal corrected group, but the difference was not statistically signi cant (p > 0.05), and there were few relevant literatures. Therefore, future studies need to expand the sample size and conduct longer follow-up. Previous similar studies found that patients in the over-corrected MPTA group (MPTA > 95° after surgery) showed no statistically signi cant difference in postoperative talus inclination Angle compared with those in the normal correction group (p=0.777) [14].
The main ndings of this study were that varus deformity was overcorrected after OWHTO, and patients with lower extremity linear load bearing axis greater than 67% were followed up at the last time, with a high incidence of postoperative complications, but the postoperative clinical e cacy was not affected. After measurement of lower extremity load position tablets, compensatory adduction of the affected side hip joint was found to be increased. Comparison of the last HAA between the two groups showed that the overcorrected group was (-2.96±1.58) ° and the normal corrected group was (-0.97±2.31) °, with statistically signi cant difference (P=0.026). Previous studies have shown that compared with normal patients, patients with varus after high tibial osteotomy had signi cantly higher adduction moments of the hip and knee before surgery, and after surgery, the adduction moments decreased to below the knee level and the hip joint to the normal level. When the knee becomes slightly everted, the frontal plane torque around the hip can be reduced to a normal level, and the reduced load may be bene cial to patients with ipsilateral osteoarthritis of the hip and knee [15]. Goshima found that KJLO was maintained parallel to the ground after OWHTO due to compensatory movement of the hip joint, and that a certain degree of overcorrection for MPTA (postoperative MPTA > 95°) did not affect clinical outcomes after OWHTO [14].Ishimatsu et al. also reported that the intraoperative and lateral space of patellofemoral joint after OWHTO was reduced compared with that before OWHTO, with statistically signi cant differences (P < 0.05) [16]. Due to the excessive displacement of the weight-bearing axis of the affected limb, it is reasonable to believe that compensatory adduction of the hip joint is increased to maintain the stability of the lower limb and re-remodeling the line of force.
In this study, according to imaging analysis, no changes in the relative position of patellofemoral joint after OWHTO were found between the overcorrection group and the normal correction group, and patellar height decreased to varying degrees in both groups, but the difference between the two groups was not statistically signi cant (P =0.139), which may be related to the short follow-up time. Previous studies have also been controversial. It was found that 69 patients (69 knees) who received OWHTO treatment did not change patella height when the knee axial correction was less than 15 degrees and had good short-term clinical results [22]. Otsuki study found that OWHTO could induce patella depression and decrease Caton-Deschamps index by 1.7%. Preoperative Caton-Deschamps index < 0.8 should be considered as a risk factor [17]. In other studies, 16 patients with 18 knees were followed up for an average of 54.2 months (25 to 96 months), and patella height was reduced by an average of 15% (P <0.05) [18]. Correction of a major deformity of the medial wedge high tibial osteotomy may result in degeneration of patellofemoral articular cartilage [19,20]. Yang treated 62 cases of knee joint (61 patients) with MOWHTO for single intervenors osteoarthritis of knee joint. Bone index was measured by postoperative radiological assessment. The average changes of patella height and tibial posterolateral Angle were statistically signi cant, but the tilt and displacement of patella were not signi cantly changed [21]. However, recent studies have found that the lateral inclination of patella (LPT) decreased from 8.67±2.60 degrees before surgery to 6.13±2.30 degrees after surgery, respectively. Patella still has lateral displacement, and internal rotation of the distal tibia is an important reason for LPT reduction [22].
The incidence of postoperative complications between the two groups was statistically signi cant (P =0.041). The average osteotomy gap was 11.16±1.81 mm in the over-corrected group and 8.86±2.07 mm in the normal corrected group, with statistically signi cant differences (P =0.008). The enlargement of the osteotomy space and the increase of bone defect mean that the incidence of non-union or delayed union of the postoperative osteotomy is higher. Previous studies have also con rmed [23] that open clearance (opening width exceeding 13.0mm, OR=1.61, P=0.02) can lead to delayed bone healing after open wedge high tibial osteotomy. In addition, with an increased osteotomy gap, there is an increased risk of hinge fracture. This is consistent with the conclusion of Seo [24] that the size of the medial opening (>11mm) is an important risk factor for lateral cortical fractures (OR=4.98). In addition, the increase of osteotomy space, the increase of intraoperative uoroscopy, and bone grafting in some patients all increased the operation time. The prolonged operation time, along with the increase of intraoperative bleeding, and the prolonged exposure of the operative area to air, also increased the risk of infection. Combined with the above factors, the incidence of postoperative complications in the over-corrected group was higher than that in the normal corrected group.
This study time at the end of the two groups of patients with postoperative follow-up clinical curative effect comparison, excessive correct set of HSS score and WOMAC index average (87.63 + / -3.62) and (9.50-1.60) points, normal correction HSS score and WOMAC score averaged (89.77-3.63) and (8.91-3.31) points, excessive correction HSS score below the normal orthodontic group, but there was no statistically signi cant difference (p = 0.139), excessive correction WOMAC score higher than normal correction group, but there was no statistically signi cant difference (p = 0.501). There is still controversy about the in uence of degree of correction on clinical e cacy. Myrnerts [8] believed that the e cacy of patients with overcorrection was signi cantly better than that of normal orthodontists. Dugdale et al. [25] believed that when the lower extremity negative gravity line passed 50%~75% of the width of the tibial plateau (average 62.5%) after medial tibial high osteotomy, the postoperative clinical e cacy was the most satisfactory. Another study evaluated the hip, knee, and ankle angles, and corrected the affected limb to an average of -4.3° to 3.7°, enabling patients to regain excellent function and daily physical activity ability through micro-orthosis [26]. At present, it is generally believed that setting different intraoperative force lines according to different knee wear conditions can reduce the pressure in the medial compartment of the knee to the greatest extent, thus reducing knee pain [27,28].There was no statistical signi cance in HSS score between the two groups, which may be because knee HSS score was affected by factors such as joint range of motion and muscle strength, and the pain improvement and daily motor function were more affected by lower limb line and cartilage repair, so the improvement in HSS score was not statistically signi cant [29,30]. In this study, there was no statistically signi cant difference between THE HSS score and WOMAC score in patients with OWHTO posterior extremity line over-migration compared with the normal correction group so far.If the orthopedic Angle is too large and the lower limb force line is overcorrected, it is bound to cause excessive knee eversion, resulting in the wear and degeneration of the lateral compartment cartilage, and increasing the probability of postoperative lateral compartment degeneration of the knee. Therefore, the long-term clinical e cacy of postoperative knee arthroplasty and whether to increase the possibility of knee replacement should be followed up for a longer period of time.

Limitations
There are some limitations in our study, which should be considered as follows: 1. The full-length X-ray images of lower extremities in our hospital were taken by using conventional photographic parameters in 3 neutral positions of the whole lower extremities and were imaged, which resulted in the errors of the Xray images themselves.2. The absolute neutral position cannot be guaranteed by any knee joint X-ray, and minor rotation, tilt, and pronation will lead to huge errors in the subsequent measurement. 3. The Angle measurement of ankle joint stability was not based on the anteroposterior and lateral ankle radiographs, but based on the preoperative and postoperative X-ray radiographs, which had some errors.4. The follow-up time was short, and the degree of correction after TPI and TIA in the over-corrected group was higher than that in the normal corrected group, but the difference was not statistically signi cant, requiring long-term follow-up results. 5. As a single-center retrospective study, this study still has limitations such as a small sample size, so it still needs to further expand the sample size and conduct a prospective multi-center study.6. This study did not carry out preoperative and postoperative arthroscopic repair of medial cartilage damage and wear of lateral cartilage.

Conclusions
Overcorrection of varus deformity (postoperative WBLR≥ 67%) did not affect the clinical e cacy of OWHTO within two years, but increased the incidence of postoperative complications and resulted in increased compensatory adduction of the hip joint.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. tables.docx