Patient-specific instrumentation for medial opening wedge high tibial osteotomies in the management of medial compartment osteoarthritis yields high accuracy and low complication rates: A systematic review

Importance: There has been growing interest in the use of patient-specific instrumentation (PSI) tomaximise accuracy and minimise the risk of major complications for medial opening-wedge high tibial osteotomies (MOW-HTOs). Numerous studies have reported the efficacy and safety of implementing this technology into clinical practice, yet no systematic review summarising the clinical literature on PSI for MOW-HTOs has been performed to date. Aim: The aim of this investigation was to perform a systematic review summarising the evidence surrounding the use of PSI for MOW-HTOs in the management of medial compartment osteoarthritis. Evidence review: PubMed, Scopus, and the Cochrane Library were queried in October 2021 for studies that used PSI for MOW-HTOs when managing medial compartment knee osteoarthritis. Primary outcomes included accuracy in coronal plane correction (mechanical medial proximal tibial angle), sagittal plane correction (posterior tibial slope), and mechanical axis correction (hip-knee-ankle angle [HKA], mechanical femorotibial angle, and weightbearing line). Accuracy was defined as error between post-operative measurements relative to the planned preoperative correction. A secondary outcome was the incidence of major complications. Findings: This review included eight different techniques among the 14 included studies. There was a weighted mean error of 0.5 (range: 0.1 –1.3 ) for the mechanical medial proximal tibial angle, 0.6 (range: 0.3 –2.7 ) for the posterior tibial slope, and 0.8 (range: 0.1 –1.0 ) for the hip-knee-ankle angle. Four studies compared the correctional error of the mechanical axis between conventional techniques and PSI techniques. The comparative difference between the two techniques favoured the use of PSI for MOW-HTOs (standardised mean difference 1⁄4 0.52; 95% confidence interval, 0.16 to 0.87; p 1⁄4 0.004). Among the 14 studies evaluated, four studies explicitly reported no major complications, while five studies reported a non-zero incidence of major complications. Among these nine studies, the weighted mean major complication rate was 7.1% (range: 0.0–13.0%). Conclusions and relevance: The findings of this present systematic review suggest that the use of PSI for MOW-HTOs leads to high accuracy relative to the planned corrections in the coronal plane, sagittal plane, and mechanical axis. Furthermore, these findings would suggest there is a low risk of major complications when implementing PSI for MOW-HTOs. Level of evidence: Systematic review; IV. on; MOW, medial opening wedge; HTO, high tibial osteotomy; mMPTA, mechanical medial proximal tibial angle; gle; WBL, weight bearing line; DFO, distal femoral osteotomy; CAN, computer assisted navigation; SMD, stanFTA, mechanical femorotibial angle; OA, osteoarthritis; UCLA, University of California at Los Angeles; KOOS, Knee dimensional; 2D, two dimensional; MINORS, Methodology Index for Non-Randomized Studies; aPPTA, anatomic Reporting Items for Systematic Reviews and Meta-analyses; IV, inverse-variance. Orthopaedic Surgery, Rush University Medical Center, Chicago, IL, USA ic Surgery, Rush University Medical Center, 1611 W. Harrison St, Suite 300, Chicago, IL, USA. Tel.: (303)931-1921 Chahla). 5 January 2023; Accepted 25 February 2023 evier Inc. on behalf of International Society of Arthroscopy, Knee Surgery and Orthopedic Sports Medicine. This is p://creativecommons.org/licenses/by/4.0/). What is already known 1 of malalignment in the coronal plane can result in an additional 12% more body weight distributed to the medial compartment. Consequently, pre-operative planning and accurate, precise intraoperative correction are critical for satisfactory high tibial osteotomy outcomes The utilisation of patient-specific instrumentation (PSI) in medial opening-wedge high tibial osteotomies has since been proposed as a solution to improve accuracy, precision, and safety of the procedure as it allows for the implementation of threedimensional pre-operative planning as well as intraoperative guidance while performing the procedure Clinical studies have also demonstrated several benefits related to PSI, including decreased operative times, reduced fluoroscopy exposure, fast learning curves, and a decreased risk of major complications such as hinge fractures and non-unions What are the new Findings Four studies compared the correctional error of the mechanical axis between conventional techniques and PSI techniques, which showed that a comparative difference between the two techniques favoured the use of PSI for medial opening-wedge high tibial osteotomies (standardised mean difference 1⁄4 0.52; 95% confidence interval, 0.16 to 0.87; p 1⁄4 0.004) Operative error relative to pre-operative planning in the hipknee-ankle angle measurement was reported in eight studies (279 patients). The weighted mean of the hip-knee-ankle angle error was 0.8 (range: 0.1 –1.0 ) In nine of the included studies (253 patients), there were 18 total major complications and a 7.1% (range: 0–13.0%) major complication rate S.P. Dasari et al. Journal of ISAKOS 8 (2023) 163–176


Introduction
Medial opening wedge high tibial osteotomies (MOW-HTOs) are effective procedures for redistributing load within the knee away from the medial compartment to offload articular cartilage and minimise pain associated with medial compartmental osteoarthritis (OA) [1]. Over the past three decades, osteotomies about the knee have become increasingly used for unicompartmental arthritis in young and active patients [2]. While effective, high tibial osteotomies (HTOs) can be challenging due to a relatively high published rate of complications, including excessive unintended tibial slope changes, hinge fractures, infections, delayed union, and non-union [3]. For patients to have successful HTO outcomes, achieving proper mechanical alignment is crucial given the substantial changes in load distribution with even modest coronal corrections [4][5][6]. For example, a study by Hsu et al. demonstrated that just 1 of malalignment in the coronal plane can result in an additional 12% more body weight distributed to the medial compartment [7]. Consequently, pre-operative planning and accurate, precise intraoperative correction are critical for satisfactory HTO outcomes.
Historically, surgeons utilised two-dimensional (2D) radiographic planning to calculate the required correction needed to satisfactorily redistribute load. These views were an anteroposterior view of the knee, a weightbearing posteroanterior view (Rosenberg/skiers view), a lateral view of the knee, an axial view of the patellofemoral joint, and a weightbearing alignment view of both lower extremity limbs from hip to ankle [8]. Calculating the degrees of correction using traditional methods is intrinsically challenging as the surgeon must use 2D planning for complex three-dimensional (3D) anatomy. For example, a study by Kawakami et al. reported significant differences in the measured mechanical femorotibial angle (mFTA) and hip-knee-ankle (HKA) angle with slight changes in the axial rotation of the limb [9]. Additionally, rudimentary instruments like rulers, callipers, and protractors have intrinsic limitations when planning these complex procedures [10]. This has led to the implementation of computer assisted navigation (CAN) and patient-specific instrumentation (PSI) in an attempt to minimise intraoperative error when performing a MOW-HTO.
CAN has been used to enhance the precision and accuracy of osteotomy cuts. It utilises real-time feedback of correction angles in multiple planes at the time of surgery in order to aid the surgeon [11]. A major benefit of this technology is that it facilitates 3D evaluation and allows the surgeon to assess rotational components that are often overlooked when using traditional 2D planning [12]. While effective, CAN still has limitations and can often lead to unintended changes of the mechanical medial proximal tibial angle (mMPTA) in the coronal plane or posterior tibial slope (PTS) in the sagittal plane during the procedure [13].
The utilisation of PSI in MOW-HTOs has since been proposed as a solution to improve accuracy, precision, and safety of the procedure as it allows for the implementation of 3D pre-operative planning as well as intraoperative guidance while performing the procedure. This theoretically minimises challenges associated with conventional and CAN techniques. Initial results examining the use of PSI technology in MOW-HTOs have been promising [14]. Several clinical studies have demonstrated exceptional accuracy and precision of the technique when implemented in patient care [3,12,[15][16][17][18]. These clinical studies have also demonstrated several other benefits related to PSI, including decreased operative times, reduced fluoroscopy exposure, fast learning curves, and a decreased risk of major complications such as hinge fractures and non-unions.
Despite the growing evidence supporting the efficacy of PSI for MOW-HTOs, there has not been a systematic review that summarises the clinical literature regarding the implementation of this technology. Therefore, the purpose of this study was to 1) systematically summarise the current clinical literature regarding PSI for MOW-HTOs and approximate the correctional accuracy of this technique in the coronal plane, sagittal plane, and mechanical axis, and 2) assess the published major complication rate of this technique. The authors hypothesised that the implementation of PSI will lead to accuracy within 1 of the pre-operative planning for all primary outcome measures and lead to low complication rates.

Article identification and selection
This systematic review was conducted in accordance with the 2020 Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines [19]. A search was performed in October of 2021 using PubMed, Scopus, and the Cochrane Library. The following search terms were used: "opening wedge high tibial osteotomy," "opening wedge proximal tibial osteotomy," and "patient specific tibial osteotomy." The inclusion criteria were as follows: utilised patient specific instrumentation, medial opening wedge HTO, primarily managing medial compartment OA with a varus axis, and clinical study. Exclusion criteria were as follows: distal femoral osteotomy (DFO), medial tibial closing wedge osteotomy, management of lateral compartment or valgus/varus deformity without OA, studies with fewer than four patients, any cadaveric/animal/in vitro study, any editorial article, any survey, any letter to the editor, any special topics, and any expert reviews. Two independent reviewers (R.F. and A.W.) performed a review of the abstracts from all identified articles. For studies that were included based on abstracts, a full-text review was performed. Any disagreements were discussed until a consensus decision could be made.
What is already known 1 of malalignment in the coronal plane can result in an additional 12% more body weight distributed to the medial compartment. Consequently, pre-operative planning and accurate, precise intraoperative correction are critical for satisfactory high tibial osteotomy outcomes The utilisation of patient-specific instrumentation (PSI) in medial opening-wedge high tibial osteotomies has since been proposed as a solution to improve accuracy, precision, and safety of the procedure as it allows for the implementation of threedimensional pre-operative planning as well as intraoperative guidance while performing the procedure Clinical studies have also demonstrated several benefits related to PSI, including decreased operative times, reduced fluoroscopy exposure, fast learning curves, and a decreased risk of major complications such as hinge fractures and non-unions

What are the new Findings
Four studies compared the correctional error of the mechanical axis between conventional techniques and PSI techniques, which showed that a comparative difference between the two techniques favoured the use of PSI for medial opening-wedge high tibial osteotomies (standardised mean difference ¼ 0.52; 95% confidence interval, 0.16 to 0.87; p ¼ 0.004) Operative error relative to pre-operative planning in the hipknee-ankle angle measurement was reported in eight studies (279 patients). The weighted mean of the hip-knee-ankle angle error was 0.8 (range: 0.1 -1.0 ) In nine of the included studies (253 patients), there were 18 total major complications and a 7.1% (range: 0-13.0%) major complication rate

Data extraction
Data were extracted in a standardised fashion into a customised spreadsheet. Data that were extracted for each study included first author, year, study design, total number of patients, number of male patients, number of female patients, average age in years at time of surgery, average final follow-up in months, primary indication for surgery, patient specific surgical planning system/technique, plate used, use of grafting, error of correction (HKA angle, weightbearing line [WBL], mFTA, mMPTA, PTS), and major complications such as hinge fracture, non-union, deep wound infection, or as defined by the respective authors. For studies that included data from patients who received osteotomies other than a MOW-HTO, only data related to the MOW-HTO were extracted.
Studies were designated a level of evidence by the classification system described by Wright et al. [20]. A bias analysis was performed by a single author (A.A.W.) on studies included for data extraction. The Methodology Index for Non-Randomised Studies (MINORSs) score was utilised [21]. The MINORS criterion is a validated scoring tool used for non-randomised studies. It involves 12 items to assess quality, four of which are only applicable to comparative studies. This creates a 16-point scale for non-comparative studies and a 24-point scale for comparative studies. Each item is scored 0 to 2 as follows: 0, not reported; 1, reported but poorly done and/or inadequate; and 2, reported well and adequately done. In order to best represent the data, weighted means with a range were utilised to summarise the accuracy and complication rate of the included studies. This was calculated using IBM SPSS Statistics for Macintosh (Version 28.0. Armonk, NY: IBM Corp).

Data analysis
Accuracy of PSI was defined as the difference in degrees between post-operative measurements relative to the planned preoperative correction. For this study, this was addressed with three primary outcomes. It was measured in the coronal plane as the mMPTA. In the sagittal plane, the accuracy was measured using the PTS. Finally, overall alignment was measured with the HKA angle. The primary alignment outcome measures for this study are depicted in Fig. 1.
Four studies examined the error in the mechanical axis correction relative to pre-operative planning and compared this outcome between conventional techniques and 3D printed PSI techniques [3,15,22,23]. For this outcome, a pooled estimate of the effect size was calculated and a standardised mean difference (SMD) with a 95% confidence interval was used to compare the correctional error between the two groups. The magnitude of the SMD was assessed according to Cohen's d estimate, where <0.5, 0.5-0.8, and >0.8 correspond to small, medium, or large effect sizes, respectively [24]. For one study, values were reported as a mean with a range, so the method described by Hozo et al. was utilised to estimate the standard deviation for their mechanical axis error outcomes [25]. An inverse-variance random effects model was implemented. The variance in the true effect value (T 2 ) and the percentage of variance from the sampling error were determined using I-squared tests (I 2 ). A statistical analysis was performed using Review Manager 5 (The Nordic Cochrane Centre; Copenhagen, Denmark).

Fig. 1. Key Alignment Measurements
The weight bearing radiographs (A-C) and lateral radiograph (D) demonstrating primary outcome measures for this systematic review. Panels A-C demonstrate the weight bearing line, the hip-kneeankle angle, and the medial mechanical proximal tibial angle. Panel D depicts the posterior tibial slope angle and the anatomic posterior proximal tibial angle.

Study characteristics
The database query yielded a total of 1,025 studies after duplicates were removed (Fig. 2). Fourteen studies satisfied all the pre-specified inclusion criteria. Study characteristics of these included studies are presented in Table 1. Five studies had a LoE of II, one study had an LoE of III, and eight studies had a LoE of IV. A total of 382 patients were included in this systematic review with a weighted age of 46.2 years (range: 44.0-67.2 years).
A bias analysis was performed for all the 14 included studies utilising the MINORS score to evaluate the quality level. Four included studies were comparative and had an average MINORS score of 18.3 (range: 16 to 20). The remaining 12 studies were non-comparative. These included non-comparative studies had a mean score of 12.2 (range: 7 to 16). The results of this bias analysis are summarised in Table 1 and presented in greater detail in Supplementary Table 1.

Patient specific instrumentation surgical techniques
When summarising a surgical technique based on the planning software, eight different techniques were described among the 14 included studies. These are summarised in Table 2. Bone graft augmentation was used in eight studies, while injectable cement was used in three studies. The two most commonly used plates were the TomoFix plate (Synthes GmbH, Solothurn, Switzerland) and the Activmotion plate (Newclip Technics, Haute-Goulaine, France). These plates were used in six and five studies, respectively.

Fig. 2. Preferred Reporting Items for Systematic Reviews and Meta-analyses Flow Diagram
The Preferred Reporting Items for Systematic Reviews and Meta-analyses flow diagram representing the search and screening process of studies implementing patientspecific instrumentation for medial opening wedge high tibial osteotomies.
The technique described using a Newclip Technics patient-specific cutting guide system (Haute-Goulaine, France) and corresponding Activmotion plate was the most common and implemented in five studies ( Fig. 3) [15,17,18,26,27]. A 2020 case series by Fucentese et al. implemented CASPA pre-operative planning software (Balgrist CARD AG, Zurich, Switzerland) to design PSI for a MOW-HTO [12]. They utilised the TomoFix Medial High Tibial Plate. The authors used a computer algorithm that they described in detail through a prior publication to calculate the optimal correction plane. A 2020 study by Mao et al. reconstructed an intact model with an osteotomy simulation software (OsteoMaster) that allowed them to generate a virtual 3D model [3]. Using this model, the authors were able to optimise the sagittal and coronal correction angles as well as the depth, width, height, slope, and position of the osteotomy before printing a custom-made cutting guide. A 2018 case series by Yang et al. used MatLab software (MathWorks Inc., Natick, MA, USA) to calculate the correction angle and utilised 3D models constructed from CT scans using Amira 4.0 software (Mercury Computer Systems, Inc., Berlin, Germany) and SolidWorks Ed. 2015 software (SolidWorks Corporation, Concord, MA, USA) [28]. The authors shifted the WBL to Fujisawa's point and achieved fixation utilising the TomoFix system. Gao et al. used Mimics software (Materialise, Leuven, Belgium) and implemented a calibrated connecting rod that was designed to expand and open the wedge based on pre-operative planning [16]. In two separate studies, Kim et al. implemented 3D slicer software (Brigham and Women's Hospital, Boston, MA, USA) to create a wedge that would open and hold the osteotomy at the ideal angle until the OhtoFix locking plate (Ohtomedical Co. Ltd., Goyang, Korea) was used for fixation [23,29]. Victor and Premanathan utilised Mimics software to design a patient-specific cutting guide that generated pre-specified drill holes [10]. Similar to the Newclip Technics technique, the osteotomy was wedged open until the pre-drilled holes matched the screw holes on the fixation plate. P erez-Mañanes et al. implemented a patient-specific 3D printed osteotomy guide as well as wedges; the authors utilised Meshmixer 2.4 software for their planning [22]. Van Genechten et al. utilised 3-matic (Materialise, Heverlee, Belgium) to create patient-specific wedges to execute their pre-operative plan [30].

Accuracy of patient specific instrumentation
Five studies (213 patients) reported mMPTA and the associated correctional error (Table 3). For the included studies, the weighted mean of the mMPTA correction error was 0.5 (range: 0.1 -1.3 ). For PTS, seven studies (242 patients) reported on the accuracy in this plane ( Table 3). The weighted mean error in the PTS for PSI in a MOW-HTO was 0.6 (range: 0.3 -2.7 ). Finally, for alignment accuracy, the HKA angle was evaluated (Table 3). Operative error relative to pre-operative planning in the HKA measurement was reported in eight studies (279 patients). The weighted mean of the HKA angle error was 0.8 (range: Four studies compared the error in the correction of the mechanical axis of the traditional 2D planning techniques against the implementation of 3D printed PSI (Fig. 4) [3,15,22,23]. All four studies reported on the error of the mechanical axis correction relative to the planned correction for both techniques. More specifically, two studies reported mechanical axis correction using the HKA measurement, one utilised the error in correction of the WBL to Fujisawa's point in millimetres and the other utilised mFTA as an additional measure. The pooled SMD suggested that there is a medium effect that favoured the use of PSI to maximise accuracy and minimise error when correcting a patient's mechanical axis using a MOW-HTO (SMD ¼ 0.52; 95% confidence interval, 0.16 to 0.87; p ¼ 0.004). For this set of data, the variance of true effects (T 2 ) was 0.04 and the I 2 was 32%, indicating that there is little heterogeneity in the estimates of the effect sizes. The standard deviation of true effects (T) was 0.2, and the resulting prediction interval was 0.12-0.92. This would suggest that 95% of patients sampled would have a more accurate correction of their mechanical axis if PSI were implemented at the time of their MOW-HTO procedure instead of traditional (non-PSI) osteotomy techniques.

Complication rate
The final outcome extracted in this systematic review was the incidence of a major complication (Table 3). For this study, major complications were defined as non-unions, hinge factures, deep tissue infections, and those defined by the respective authors. For this systematic review, five studies reported a major complication, four studies explicitly reported that no major complications had occurred, and five studies did not report the incidence of major complications. The five studies that did not explicitly state that there were zero complications were not included in this analysis of major complications. When examining the remaining nine studies (253 patients), there were 18 total major complications and a 7.1% (range: 0-13.0%) major complication rate.      None None WBL, PTS None A 4-cm incision was made in the proximal and anteromedial portion of the tibia, then dissection of the pes anserinus and superficial MCL was performed. The PSI guide was attached, and Kwires were drilled to fix the guide. Using the cutting slot and guiding plane, a biplanar cut was performed to create the desired hinge. After two wedges had been created, the guide was separated into proximal and distal parts from the cutting slot. An osteotome was inserted until the planned length reached the sawing depth. This was followed by the insertion of another osteotome for distraction. A spreader was used to hold the distracted wedge. The distal K-wires were removed before taking the radiographs. Finally, the TomoFix plate and locking screws were used to fix the osteotomized tibia.  The medial opening wedge high tibial osteotomy using the NewClip Technics (Haute-Goulaine, France) patient-specific instrumentation system. Panel A depicts the patient-specific cutting guide fitting against the anteromedial cortex. Panel B depicts the cutting guide held in place by two unicortical pins proximal to the slotted capture and two bicortical pins distal to the slotted capture; the cut pin, an additional guide pin, and the hinge pin are also placed. Panel C depicts the high tibial osteotomy plate being secured with a proximal and distal locking screw, while the osteotomy is wedged open to the optimal correction angle. Panel D depicts the final construct for the medial opening wedge high tibial osteotomyusing the New-Clip Technics (Haute-Goulaine, France) patientspecific instrumentation system. Table 3 Correctional error for the hip-knee-ankle angle, mechanical medial proximal angle, posterior tibial slope, and rate of major complications.

Discussion
The findings from this systematic review support the implementation of PSI for MOW-HTOs in clinical practice and demonstrated an average operative accuracy within 0.6 of the pre-operative plan in both the coronal (mMPTA) and sagittal planes (PTS). Additionally, this study would suggest that there is a high degree of accuracy in the correction of the mechanical axis with an average error that is within 0.8 of the planned correction in the HKA angle. The SMD would suggest that there is reduced error in the correction of alignment when using PSI for MOW-HTOs relative to techniques using conventional pre-operative planning. Furthermore, when examining secondary outcomes, the use of PSI led to a low incidence of major complications (7.1%).
The largest included study in this systematic review was a prospective observational study by Chaouche et al. [17]. In this study, 100 patients were treated with a MOW-HTO using PSI. The authors noted excellent accuracy in all planes/axis of correction as well as a low rate of major complications (4%). Furthermore, they noted substantial clinical improvements in the Knee Injury and Osteoarthritis Outcome Score and University of California at Los Angelesoutcome scores at two-year follow-up. A separate study by Jacquet et al. sought to evaluate the learning curve of implementing PSI into a surgeon's practice [18]. The authors examined the use of PSI in 71 total cases performed by three separate surgeons. They noted a fast-learning curve: it took the surgeons 10 cases to optimise their operative time to a mean of 26.3 min per case. It took the surgeons eight cases to lessen their anxiety levels and nine cases to decrease the number of fluoroscopic images utilised. Additionally, the authors noted highly accurate coronal plane, sagittal plane, and mechanical axis outcomes in all the included cases, and the accuracy of the initial 10 cases was not affected by the learning curve for the procedure.
Four studies sought to directly compare PSI to conventional techniques. A 2020 study by Mao et al. sought to directly compare outcomes related to 19 patients treated with a freehand conventional MOW-HTO to 18 patients treated with a PSI MOW-HTO [3]. The authors noted significantly improved accuracy in the correction of the mFTA and mMPTA. The authors also noted significantly shorter operative times by approximately 16.8 min per case and significantly decreased radiation exposure for the PSI group. This led Mao et al. to conclude that PSI is readily implementable and has superior accuracy relative to conventional MOW-HTO techniques.
A separate 2020 multicentre non-randomised prospective study by Tardy et al. allocated 126 patients into three different treatment arms: conventional technique, CAN, and PSI [15]. The authors performed pre-operative planning and assessed the HKA angle accuracy relative to pre-operative planning for all three groups. They reported an error of 1.1 in the conventional group, 2.1 in the CAN group, and 0.3 in the PSI group. In 2016, P erez-Mañanes et al. compared error in HKA correction relative to pre-operative planning and found a difference in 0.6 that favoured the PSI group over the conventional group [22]. These results were further supported by a 2018 study by Kim et al. that reported increased accuracy in the correction of the weight-bearing line when using PSI relative to more conventional MOW-HTO techniques [23]. When these results were aggregated in the present systematic review, the SMD also suggested that the implementation of PSI in MOW-HTOs would lead to a reduced error in the mechanical axis correction relative to conventional osteotomy techniques. Thus, in the context of the current clinical literature, the results reported in this systematic review support the implementation of PSI for accurate, safe, and precise MOW-HTOs.
A recently published systematic review by Aman et al. demonstrated that PSI improves coronal correctional accuracy and minimises complications for osteotomies around the knee [31]. Like our systematic review, their study supported the implementation of PSI; however, the authors did not assess sagittal accuracy and included opening and closing wedge HTOs as well as opening and closing wedge DFOs. Our present study assessed the coronal accuracy, sagittal accuracy, mechanical axis correction accuracy, and complication rate for MOW-HTOs using PSI when treating medial compartment OA. It demonstrated a low complication rate and a high degree of accuracy for all included primary outcomes. Furthermore, a direct comparison of PSI relative to a traditional (gold standard) technique for MOW-HTOs was feasible in our study by calculating relative effect sizes and a SMD. Our study suggested that 95% of included patients would have a more accurate mechanical axis correction if PSI was implemented for their MOW-HTO. Finally, given the chronological nature of systematic reviews, our study was able to include 14 studies examining PSI for MOW-HTOs, while the study by Aman et al. included 14 total studies when combining opening/closing wedge HTOs and DFOs.
The improved accuracy outcomes with the PSI noted in this study are clinically relevant given that error in either the coronal or sagittal planes can lead to subsequent suboptimal patient outcomes. Additionally, unintended changes in the sagittal plane can also lead to impaired knee kinematics, instability, and excessive strain on the cruciate ligaments [12]. This is relevant in the field of osteotomies given that LaPrade et al. demonstrated that a conventional MOW-HTO leads to an average unintended PTS change of 2.9 when using a freehand technique [32]. The findings of the present systematic review suggest that such unintended changes in sagittal deformity could potentially be prevented through the use of PSI.
This study is not without limitations. First, the analysis was limited by the overall currently available evidence on this topic, with substantial variations noted in the PSI technique as well as planning software. Nevertheless, 14 studies were available for inclusion, with four studies available for the calculation of a SMD between PSI and traditional techniques. Second, the study was limited by variability in imaging the outcome reporting as it relates to which coronal and sagittal measurements were provided in each study. Additionally, the definition of major/ reported complications among the included studies varied on the basis of the study design. Finally, long-term studies are needed to establish and quantify the clinical benefit of incremental increases in HTOs and define how they relate to long term patient reported outcomes and reoperation/ arthroplasty rates. The above limitations highlight the need for future well-designed prospective, randomised clinical studies in order to better evaluate the true efficacy, safety, and benefits of PSI for MOW-HTOs. Fig. 4. Comparison of Error Relative to Pre-operative Planning of Mechanical Axis Correction A forest plot demonstrating the standardised mean difference for error in the correction of the mechanical axis relative to pre-operative planning. This includes a summary estimate (centre of the diamond) and a 95% confidence interval (width of the diamond) for the standardised mean difference. The size of each square represents the relative weight given to each respective study. Legend: Weightbearing Line (WBL); Mechanical Femorotibial Angle (mFTA); Hip-Knee-Ankle Angle (HKA); Inverse-Variance Model (IV); Confidence Interval (CI).

Conclusion
The findings of this present systematic review suggest that the use of PSI for MOW-HTOs leads to high accuracy relative to the planned corrections in the coronal plane, sagittal plane, and the mechanical axis. Furthermore, these findings would suggest that there is a low risk of major complications when implementing PSI for MOW-HTOs.

Funding
No funding was received for this study, no informed consent/ethical approval was required for this study design, and the authors have no relevant conflicts of interest to declare.

Authors contribution
All the authors have fulfilled the following ICJME requirements for authorship: substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; drafting the work or revising it critically for important intellectual content; final approval of the version to be published; and agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. Suhas Dasari (Contribution: substantial conception/design of the work, performed measurements, data collection, statistical analysis, interpretation of data, drafting the work, critically revising the work, manuscript preparation, approving final version for publication, and agreement for accountability of all aspects of work. Mario Hevesi (Contribution: substantial conception/design of the work, performed measurements, data collection, statistical analysis, interpretation of data, drafting the work, critically revising the work, manuscript preparation, approving final version for publication, and agreement for accountability of all aspects of work. Luc Fortier (Contribution: substantial conception/design of the work, interpretation of data for the work, revising of the work for important intellectual content, final approval of the version for publication, and agreement to accountability of all aspects of the work). Robert Ferrer-Rivero (Contribution: substantial conception/design of the work, interpretation of data for the work, revising of the work for important intellectual content, final approval of the version for publication, and agreement to accountability of all aspects of the work). Alec Warrier (Contribution: substantial conception/design of the work, revising of the work for important intellectual content, final approval of the version for publication, and agreement to accountability of all aspects of the work). Bhargavi Maheshwer (Contribution: substantial conception/design of the work, revising of the work for important intellectual content, final approval of the version for publication, and agreement to accountability of all aspects of the work). Garrett Jackson (Contribution: substantial conception/ design of the work, revising of the work for important intellectual content, manuscript preparation, final approval of the version for publication, and agreement to accountability of all aspects of the work). Harkirat Jawanda (Contribution: substantial conception/design of the work, revising of the work for important intellectual content, manuscript preparation, final approval of the version for publication, and agreement to accountability of all aspects of the work). Enzo Mameri (Contribution: substantial conception/design of the work, revising of the work for important intellectual content, manuscript preparation, final approval of the version for publication, and agreement to accountability of all aspects of the work). Zeeshan Khan (Contribution: substantial conception/design of the work, revising of the work for important intellectual content, manuscript preparation, final approval of the version for publication, and agreement to accountability of all aspects of the work). Benjamin Kerzner (Contribution: substantial conception/design of the work, revising of the work for important intellectual content, manuscript preparation, final approval of the version for publication, and agreement to accountability of all aspects of the work). Robert Browning (Contribution: substantial conception/design of the work, revising of the work for important intellectual content, manuscript preparation, final approval of the version for publication, and agreement to accountability of all aspects of the work). Safa Gursoy (Contribution: substantial conception/design of the work, revising of the work for important intellectual content, final approval of the version for publication, and agreement to accountability of all aspects of the work). Jorge Chahla (Contribution: substantial conception/design of the work, revising of the work for important intellectual content, manuscript preparation, final approval of the version for publication, and agreement to accountability of all aspects of the work).