Changes in Balance Outcomes Following High Tibial Osteotomy in Patients With Knee Osteoarthritis: A Systematic Review

ABSTRACT

Background and Aim

Knee osteoarthritis (KOA) is often accompanied by impaired postural balance, which may persist despite surgical correction. High tibial osteotomy (HTO) is a standard joint‐preserving procedure for medial KOA, but its influence on postural balance is unclear. This systematic review aimed to synthesize evidence on balance changes following HTO.

Methods

A systematic search was conducted in PubMed, Scopus, Web of Science, ProQuest, Science Direct, and Google Scholar (1990–October 2025) using the PECO framework. Eligible studies were observational designs including adults with KOA undergoing HTO and reporting balance‐related outcomes. Methodological quality was assessed using the Joanna Briggs Institute (JBI) critical appraisal tools, with separate checklists applied for longitudinal pre–post studies and cross‐sectional studies.

Results

Nine observational studies met the inclusion criteria which included 572 participants. Balance outcomes included force‐plate analyses of center of pressure (COP) during bipedal or single‐leg stance, indices of weight distribution and steadiness, dynamic COP trajectory during gait, and functional tests such as the Timed Up and Go and Tinetti assessment. Short‐term postoperative (within the first 3–6 months after HTO) findings were inconsistent, with some studies showing selective improvements and others no change. At mid‐ to long‐term follow‐up (≥ 12 months postoperatively), more consistent improvements were observed in weight distribution symmetry and functional performance. However, global steadiness often remained inferior to healthy reference group. Methodological quality assessment indicated that all longitudinal pre–post studies were of moderate quality, whereas the cross‐sectional study was rated as high quality based on JBI criteria.

Conclusion

Balance outcomes tended to improve following HTO, particularly in load distribution and certain functional measures; however, global postural stability often remained inferior to that of healthy individuals. These findings suggest that alignment correction alone may be insufficient to fully restore balance. Therefore, structured rehabilitation and more rigorous, standardized future studies are warranted.

1. Introduction

Knee osteoarthritis (KOA), particularly affecting the medial compartment often accompanied by varus malalignment, can lead to functional deficits in postural balance and increased risk of falls in affected individuals. High tibial osteotomy (HTO) has long been employed as a joint‐preserving surgical intervention, aiming to realign the mechanical axis of the lower limb, unload the medial compartment, and alleviate pain while potentially restoring neuromuscular control and balance mechanisms [1, 2].

Despite its widespread use and demonstrated improvements in alignment and pain relief, the impact of HTO on postural stability remains less extensively documented. Recent evidence from static posturography has shown that, compared with healthy group, patients with KOA exhibit increased postural sway preoperatively and at early postoperative stages; however, by 1 year after HTO, these sway parameters converge with those of controls, suggesting significant normalization of balance control [3]. Similarly, short‐term assessments following inverted V‐shaped HTO revealed early improvements in mediolateral postural balance—specifically reductions in the center of pressure (COP) mediolateral range—alongside enhanced clinical outcomes such as decreased pain within 6 weeks post‐surgery [4]. Additionally, rehabilitation studies employing an interactive balance system indicated progressive improvements in weight distribution indices and postural stability over a 6‐month postoperative period, although full normalization relative to asymptomatic individuals was not completely achieved [5].

Collectively, these findings underscore that HTO may impart beneficial effects on postural regulation and static stability among patients with KOA. Nevertheless, variability in follow‐up durations, osteotomy techniques, balance assessment modalities, and rehabilitation protocols confounds direct comparisons of findings across studies.

Several systematic reviews have previously examined the outcomes of HTO from different perspectives. For instance, recent reviews have addressed compensatory mechanisms of proximal and distal joint alignment and gait in varus KOA following HTO [6], radiographic changes and clinical outcomes after open‐ and closed‐wedge HTO [7], the relationships between coronal plane alignment and patient‐reported outcomes [8], and outcomes of HTO for valgus knee deformity [9]. Despite growing evidence on the clinical and biomechanical outcomes of HTO, its effect on postural balance remains unclear. Since postural control is fundamental to functional recovery and fall prevention, clarifying balance outcomes after HTO has direct clinical relevance. To date, no systematic review has specifically addressed this issue. Therefore, the present review aimed to synthesize available evidence on changes in postural balance following HTO in patients with KOA.

2. Methods

2.1. Design

The protocol of this systematic review was already registered in the PROSPERO database with the No. CRD420251144879. After studies' identification, the process was followed by the PRISMA guideline.

2.2. Evidence Acquisition

This systematic review was conducted to evaluate changes in postural balance following HTO in patients with KOA. A comprehensive literature search was performed in PubMed, Scopus, Web of Science, ProQuest, Science Direct, and Google Scholar for studies published from inception to October 2025 (Supplementary Appendix). The search strategy was based on the Population, Exposure, Comparator, Outcome (PECO) framework, focusing on the Population and Outcome components. The Population component included the keywords “high tibial osteotomy” and its synonyms (“HTO,” “tibial re‐alignment surgery,” “tibial osteotomy,” “medial opening wedge osteotomy,” “inverted V‐shaped osteotomy,” “valgus‐producing osteotomy”), and the Outcome component included “balance” and its synonyms (“postural balance,” “postural control,” “postural stability,” “standing balance,” “balance performance,” “center of pressure,” “postural sway,” “weight distribution” “timed up & go”). Boolean combinations of Keyword‐I AND Keyword‐II were applied. All records were imported into EndNote software, and duplicates were removed. Two independent reviewers (Hiva Farhadabadi and Aydin Mahmoud Alilou) screened titles and abstracts for eligibility, and full texts were retrieved for detailed evaluation, with disagreements resolved by a third reviewer (Aliyeh Daryabor) or, if necessary, by contacting study authors.

Inclusion criteria were observational studies (including longitudinal pre–post designs, prospective or retrospective cohort, cross‐sectional, or case‐control) involving adults with KOA undergoing HTO (any surgical technique) and reporting balance‐related outcomes such as postural control, postural stability, COP parameters, or functional balance tests (e.g., Timed Up & Go (TUG) test, single‐leg standing (SLS) test), with either a healthy group or a within‐subject preoperative comparison, and with no language restriction. Studies including healthy individuals, primarily used as reference groups (hereafter referred to as “healthy reference group”). Exclusion criteria included reviews, meta‐analyses, case reports, conference abstracts, and letters; studies on non‐KOA populations; and studies that exclusively reported radiographic, clinical, or gait outcomes without balance assessment.

2.3. Assessment of Methodological Quality

The methodological quality of the included studies was assessed using the Joanna Briggs Institute (JBI) critical appraisal tools. Separate checklists were applied according to study design, with the JBI checklist for quasi‐experimental studies used for longitudinal pre–post studies and the JBI checklist for analytical cross‐sectional studies used for the cross‐sectional study [10]. As the JBI tools do not provide a standardized scoring system, overall study quality was determined based on the proportion of criteria met. Studies fulfilling ≥ 70% of the applicable items were classified as high quality, those meeting 50%–69% as moderate quality, and those below 50% as low quality. Two independent reviewers (Farnaz Farshchi and Maryam Mohammadi) performed the quality assessment, and discrepancies were resolved through discussion or by consultation with a third reviewer (Rasul Gharekhan Maleki).

2.4. Evidence Synthesis

Data extracted from each eligible study included the study design, characteristics of participants in the study group, healthy reference group (where applicable), surgical procedure performed, rehabilitation protocols following surgery (when reported), balance outcomes and assessment tools used, time points of outcome evaluation, and main study findings. The interpretation of statistical significance followed the criteria applied in the original reports, with p < 0.05 generally considered as the cutoff. A meta‐analysis could not be performed due to heterogeneity in study designs and outcome measures; therefore, the findings were synthesized narratively. These details are summarized in Table 1.

Table 1.

Studies characteristics.

Author (year)	Study design	Participants of study group	Healthy reference group	Surgical procedure of HTO	Rehabilitation following surgery	Balance outcomes (tools)	Assessment times of outcomes	Main results
Kurz (2022) [11], Follow‐up study of Brehme et al. (2020)	Prospective longitudinal pre–post study (with healthy reference group)	24 male individuals with medial tibiofemoral OA and genu varum of tibia (varus degree: ns), KL grade: 3, mean age: 56.6 years, BMI: 30.8 kg/m2, operated side: 12 their dominant	25 healthy matched individuals (sex, age, height), age: 54.9 years, BMI: 27.4 kg/m2	Bi‐plane medial‐based openingwedge	ns	Percentage weight under the foot at one side, stability index, and weight distribution index while bipedal standing (InteractiveBalance System including four independent force plate)	‐Preoperation− 6 weeks postoperation− 1 year postoperation− 2 years postoperation	Following HTO, the proportion of body weight borne on the operated limb increased significantly (p < 0.001), while standing asymmetry was also reduced over time (p = 0.003). In contrast, standing steadiness assessed by the stability index showed no significant change across the follow‐up periods (p = 0.71). For evaluating outcomes, so, percentage weight appears to be a suitable indicator of short‐term effects, whereas both percentage weight and the weight distribution index provide more insight into long‐term adaptations
Zhang (2023)[3]	Prospective longitudinal pre–post study (with healthy reference group)	32 (12 F, 20 M) individuals with unilateral medial knee KOA and genu varum of tibia, KL grade ≥ 3, mean age: 52.2 years, BMI: 28.9 kg/m2, operated side: ns	40 healthy matched individuals, age: 53.9, BMI: 27.5 kg/m2,	Bi‐plane medial‐based openingwedge	Standard rehab with straight‐leg raises, quadriceps isometrics, and two weeks of continuous passive motion from day one post‐operation.	COP trajectory while bipedal standing with two visual conditions (eyes open and closed) along with and foam support) (force plate)	‐One month postoperation‐ Three months postoperation‐ One year postoperation	Relative to healthy reference group, patients demonstrated significantly greater postural sway both one month prior to surgery and three months after the procedure (p < 0.05). By the one‐year follow‐up, however, no significant difference in sway measures was detected between the two groups (p > 0.05).
Kim (2023)[12]	Single‐group Prospective longitudinal pre–post study	90 (30 F, 30 M) individuals with medial knee KOA and genu varum of tibia, KL grade:ns, mean age: 61.2 years, BMI: 26.6 kg/m2, operated side: 34 left; 56 right, correction angle after surgery: 12.1 degrees, Participants classified at 1 month after operation by HAA: HAA (–) < 0° (n:52), HAA (+) ≥ 0° (n:38)	—	Medial opening wedge	ns	‐Functional balance (TUG and SLS)	‐Preoperation− 2 years postoperation	Two years after HTO, patients demonstrated significant improvement in TUG performance compared with baseline (p < 0.05), while SLS outcomes did not change significantly (p > 0.05). Moreover, participants in the HAA (−) group had markedly lower TUG scores than those in the HAA (+) group (p = 0.011).
Li (2023)[13]	Prospective longitudinal pre–post study (with healthy reference group)	66 (41 F, 25 M) individuals with unicompartmental medial knee KOA and genu varum of tibia (≥ 5°), KL grade≥ 2, mean age: 59.1 years, BMI ≤ 35 kg/m2, operated side: ns	34 healthy matched individuals, age: 57.8	Medial opening wedge	‐ Quadriceps isometric exercises started day one post‐ operation.‐ Partial weight bearing progressed over six weeks; full weight bearing allowed thereafter.	Medial‐lateral pressure ratio, foot progression angle, COP trajectory including anteroposterior and lateral symmetry of COP while bipedal standing (force plate)	Preoperation‐ Average 18 months postoperation	Preoperatively, lateral symmetry of the COP was significantly greater compared with postoperative and healthy reference group (p = 0.031), while no significant difference was found between the latter two. Conversely, the forefoot medial–lateral pressure ratio, anteroposterior COP, and foot progression angle did not show significant variation among the three groups (p > 0.05).
Sabashi (2023)[4]	Single‐group Prospective longitudinal pre–post study	15 (9 F, 6 M) individuals with medial knee KOA and genu varum of tibia (genu varum degree:8‐12 degree), KL grade: 3 or 4, mean age: 55.9 years, BMI: 29.3 kg/m2, operated side: 9 right, 6 lest	—	Inverted V‐shaped	ns	Maximum range, mean velocity, and area of COP movements in anteroposterior and mediolateral directions during single‐leg standing (force plate)	Preoperation− 6 weeks postoperation	Six weeks after inverted V‐shaped HTO, the mediolateral COP range was significantly reduced (p = 0.017), while the anteroposterior COP velocity increased (p = 0.011). These findings indicate that valgus correction through this surgical technique enhances mediolateral balance control and provides favorable short‐term clinical benefits.
Yuan (2022)[14]	Retrospective observational cross‐section	278 (165 F, 113 M) individuals with medial knee KOA and genu varum of tibia (varum degree: ns), KL grade: ns, mean age: 56.51 years, BMI: 27.01 kg/m2, operated side: 135 right, 143 left	—	Medial opening wedge	ns	Tinetti assessment score (3‐point ordinal scale of 0, 1 and 2 on 28 points)	ns	The highest Tinetti balance scores were observed when the mechanical axis correction reached approximately 53.7% (p < 0.05). Moreover, the extent of lower‐limb alignment adjustment showed a significant correlation with postoperative functional knee status (p < 0.05).
Whatling (2020)[15]	Prospective longitudinal pre–post study (with healthy reference group)	18 (2 F, 16 M) individuals with unicompartmental medial knee KOA and genu varum of tibia (8°), KL grade: 2 KL2,12 KL3,5 KL4, mean age: 51.2 years, BMI: 29.5 kg/m2, operated side: 9 right, 9 left	180healthy individuals (not matched forage, gender, or weigh), age: 34.6 years, BMI: 24.5	Medial opening wedge	ns	COP trajectory relative to the foot axis and relative to the knee during walking (force plate)	Preoperation (average 1.7 ± 1.8 months)‐ Average 13.8 ± 4.5 months postoperation	Compared with controls, COP relative to the foot showed only minor postoperative differences without a surgical effect. In contrast, COP relative to the knee shifted laterally after HTO at both EKAM peak 1 and 2 (p < 0.001). At peak 1, it was more anterior than controls preoperatively (p = 0.027) and postoperatively (p = 0.025), while at peak 2 it shifted posteriorly after surgery (p = 0.012) but did not differ significantly from controls.
Brehme (2020)[5]	Prospective longitudinal pre–post study (with healthy reference group)	32 male individuals with symptomaticmedial tibiofemoral OA, KL: 3, mean age: 55.3 years, BMI: 30.7 kg/m2, operated side: 12 left; 20 right,	32 healthy matched individuals (sex, age, height)	Bi‐plane medial‐based opening wedge	Standard rehabilitation with weight bearing, isometrics, and strength/endurance physical therapy from week 1–16 onward.	‐Stability testing (InteractiveBalance System including four independent force plate) on the side of injury/correction	‐Preoperation− 6 weeks postoperation− 12 weeks postoperation− 12 months postoperation	Patients who underwent HTO showed no significant improvement in postural performance after surgery and subsequent rehabilitation (p > 0.05). However, mediolateral weight distribution on the operated side improved significantly following HTO (p < 0.05; ηp² = 0.163), although postural stability remained consistently lower than that of asymptomatic controls (p > 0.05).
Hunt (2009)[16]	Single‐group Prospective longitudinal pre–post study	49 (5 F, 44 M) individuals with medial knee KOA and genu varum of tibia (genu varum degree: ns), KL grade: ns, mean age: 46 years, BMI: 28.6 kg/m2, operated side: ns	—	Medial opening wedge	‐Weight bearing allowed once bony union confirmed (~3 months).‐No standardized rehab provided post‐opetation	Pathlength, SD, and range of COP while standing single‐limb balance (force plate)	‐Preoperation −12 months postoperation	Though all COP measures were lower following HTO, none reached statistical significance compared with pre‐operative stage (p > 0.05). So, standing balance in individuals with KOA is not significantly different following HTO surgery

Abbreviations: BMI, body mass index; COP, center of pressure; EKAM, external knee adduction moment; HAA, hip abduction angle; HTO, high tibial osteotomy; KL, Kellgren–Lawrence; KOA, knee osteoarthritis; ns, no stated; SD, standard deviation; SLS, single‐leg standing; TUG, Timed Up & Go.

3. Results

3.1. Study Selection

The database search initially yielded 746 citations. After eliminating 403 duplicates using EndNote, 343 records underwent title and abstract screening. From these, 33 articles were retrieved for full‐text assessment, of which 24 were excluded due to non‐eligible populations, interventions other than HTO, or unrelated outcomes. Ultimately, 9 studies fulfilled the inclusion criteria and were incorporated into this review (Figure 1), all of which were published in English.

Figure 1.

Flow diagram of the study selection based on PRISMA.

3.2. Study Designs and Participants

All nine included studies employed observational designs. Of these, eight studies were longitudinal pre–post observational studies [3, 4, 5, 11, 12, 13, 15, 16], and one study was an analytical cross‐sectional study [14]. Among the longitudinal studies, five studies also included healthy reference groups [3, 5, 11, 13, 15], allowing for additional comparisons with normative values, while the remaining three studies [4, 12, 16] were limited to within‐subject (pre–post) comparisons in the surgical group. In studies that included healthy individuals, these participants were primarily used as reference groups to provide normative values for postural balance, rather than as independent comparator groups for causal inference. Accordingly, the main analytical focus across the included studies was on within‐subject changes over time following HTO. Overall, the included populations consisted of middle‐aged to older adults with varus KOA, consistent with the typical clinical indications for HTO.

Across the nine included studies, a total of 572 unique patients with medial KOA accompanied by varus deformity of the tibia underwent HTO. Two included studies [5, 11] were based on the same group of patients assessed at different follow‐up time points (short‐term and long‐term); therefore, to avoid double‐counting, these studies were considered as a single sample when estimating the total sample size. Sample sizes ranged from 15 participants [4] to 278 participants [14]. The mean age of patients varied between 46 years [16] and 61.2 years [12]. Both sexes were represented, although some studies included predominantly male participants [5, 11, 15]. Reported body mass index (BMI) values generally ranged from 26 to 31 kg/m². Disease severity was most frequently classified as Kellgren–Lawrence grade ≥ 2 or ≥ 3; however, in some studies, specific radiographic grades or degrees of varus deformity were not stated.

3.3. Surgical Procedures and Rehabilitation Following Surgery

All studies applied medial opening wedge techniques, except Sabashi et al. [4] who used inverted V‐shaped osteotomy. Rehabilitation protocols following surgery were heterogeneous. Some studies described structured programs including early weight‐bearing, isometric quadriceps strengthening, continuous passive motion, and progressive physiotherapy [3, 5, 13, 16], while in other studies, rehabilitation was not described or only general clinical guidance was provided [4, 11, 12, 14, 15]. Overall, while the surgical approaches were relatively consistent, considerable variability existed in the reporting and application of rehabilitation protocols across studies.

3.4. Outcome Measures

The included studies assessed a range of outcomes related to postural balance in patients with KOA undergoing HTO. Several studies employed force plate–based posturography to capture static or dynamic balance performance, including measures such as COP trajectory, sway velocity, sway area, mediolateral and anteroposterior COP displacement, stability index, and weight distribution index [3, 4, 5, 11, 13, 15, 16]. Others utilized functional clinical balance tests, particularly the TUG test and SLS test, to evaluate functional aspects of balance relevant to daily activity [12]. In addition, the Tinetti Performance‐Oriented Mobility Assessment was applied as a clinical rating scale to quantify mobility and fall risk [14]. Notably, one study assessed COP trajectory during walking relative to the foot axis and knee joint using force plate analysis, thereby extending the assessment of balance beyond static standing tasks [15].

The timing of balance assessments varied widely across studies, ranging from the early postoperative period (e.g., 6 weeks and 3 months) to long‐term follow‐up at 12 months and up to 2 years [4, 5, 11, 15]. While some investigations focused on short‐term changes such as mediolateral sway reduction or early improvements in COP parameters [4], others highlighted sustained adaptations in weight distribution and symmetry that persisted beyond 1 year [11, 13, 15]. Collectively, these outcome measures provided complementary perspectives on postural balance recovery, spanning both laboratory‐based instrumented assessments and clinically meaningful functional tests.

3.5. Methodological Quality

Overall, the longitudinal pre–post studies [3, 4, 5, 11, 12, 13, 15, 16] demonstrated moderate methodological quality (Table 2). The most common limitations were the absence of independent control groups, limited control of potential confounding factors, and insufficient reporting regarding follow‐up completeness. Although some studies included healthy groups, these were primarily used as reference values rather than as true comparative controls, and therefore did not fulfill the criteria for independent control groups within the appraisal framework. In contrast, the cross‐sectional study [14] was rated as high methodological quality (Table 3), as it met the majority of the JBI criteria, including clearly defined inclusion criteria, appropriate measurement of outcomes, and adequate statistical analysis.

Table 2.

Evaluation of the quality of studies for longitudinal pre–post studies.

Author (year)	Q1: Cause–effect	Q2: Similar participants	Q3: Similar care	Q4: Control group	Q5: Multiple measurements	Q6: Follow‐up completeness	Q7: Same measurement	Q8: Reliable measurement	Q9: Statistical analysis	Overall	Quality status
Kurz (2022) [11]	Yes	Yes	No	No	No	Yes	Yes	Unclear	Yes	5	Moderate
Zhang (2023) [3]	Yes	Yes	Yes	No	No	Unclear	Yes	Unclear	Yes	5	Moderate
Kim (2023) [12]	Unclear	Yes	Unclear	No	No	Yes	Yes	Yes	Yes	5	Moderate
Li (2023) [13]	Yes	Yes	Yes	No	No	Unclear	Yes	Yes	Yes	6	Moderate
Sabashi (2023) [4]	Yes	Yes	Yes	No	No	Unclear	Yes	No	Yes	5	Moderate
Whatling (2020) [15]	Yes	Yes	Unclear	No	No	Yes	Yes	No	Yes	5	Moderate
Brehme (2020) [5]	Yes	Yes	Yes	No	No	Yes	Yes	No	Yes	6	Moderate
Hunt (2009) [16]	Yes	Yes	Yes	No	No	Unclear	Yes	No	Yes	5	Moderate

Table 3.

Evaluation of the quality of studies for cross‐section studies.

Author (year)	Q1	Q2	Q3	Q4	Q5	Q6	Q7	Q8	Overall	Quality status
Yuan (2022) [14]	Yes	Yes	Unclear	Yes	Yes	Yes	Yes	Yes	7	High

3.6. Main Findings

3.6.1. Static Postural Control in Quiet Standing

Short‐term (≤ 3 months postoperatively): In the early postoperative period, Sabashi et al. (2023) [4] assessed single‐leg stance and reported that 6 weeks after HTO the mediolateral COP range decreased significantly compared with preoperative values (p = 0.017), while the mean velocity of COP in the anteroposterior direction increased (p = 0.011). Zhang et al. (2023) [3], using bipedal stance with eyes open/closed and on foam, found greater postural sway in patients compared with matched healthy reference group at 1 month and 3 months after surgery.

Mid‐term (≈12 months postoperatively): Hunt et al. (2009) [16] measured single‐leg stance and found that COP trajectory and variability were lower 12 months postoperatively compared with baseline, but the differences did not reach statistical significance (p > 0.05). Brehme et al. (2020) [5], during bipedal standing, observed no significant improvements in postural performance at 6 or 12 weeks compared with baseline; however, by 12 months mediolateral weight distribution improved compared with preoperative values, although stability indices remained consistently lower than in matched healthy reference group.

Long‐term (> 12 months postoperatively): Li et al. (2023) [13] assessed bipedal stance and showed that after ~18 months, lateral COP symmetry improved significantly compared with baseline (p = 0.031) and reached levels comparable with healthy reference group. Other COP parameters, including anteroposterior displacement and foot progression angle, showed no significant changes across groups.

3.6.2. Weight Distribution and Symmetry Indices

Short‐term (≤ 3 months postoperatively): Brehme et al. (2020) [5] demonstrated some significant improvements in weight distribution parameters at 6–12 weeks compared with baseline (p < 0.05); however, these changes were not uniformly observed across all assessed variables.

Mid‐term (≈12 months postoperatively): The same study [5] reported improved mediolateral weight distribution at 12 months compared with preoperative values (p = 0.04), although overall stability remained inferior to controls.

Long‐term (≈24 months postoperatively): Kurz (2022) et al. [11], using bipedal standing tests, found that the percentage of body weight borne on the operated side increased significantly compared with preoperative measures (p < 0.001) and standing asymmetry decreased (p = 0.003) up to 2 years after surgery. However, standing steadiness (stability index) did not change significantly across time points (p = 0.71).

3.6.3. Functional Balance Tests

Mid‐term (≈2 years postoperatively): Kim et al. (2023) [12] showed that TUG scores improved significantly at 2 years compared with preoperative values, while SLS time did not. Patients with residual varus alignment (hip abduction angle ≥ 0°) performed worse on TUG compared with those with neutral or valgus correction (p = 0.011).

Long‐term (> 12 months postoperatively): Yuan et al. (2022) [14] evaluated functional balance using the Tinetti scale and reported that the highest balance scores were observed when the mechanical axis correction reached ~53.7%. This indicates that surgical correction was directly associated with postoperative functional balance recovery.

3.6.4. Dynamic Balance during Gait

Mid‐term (~14 months postoperatively): Whatling et al. (2020) [15] analyzed gait using a force plate and reported that COP relative to the foot axis showed small differences compared with healthy reference group both pre‐ and postoperatively, with no clear surgical effect. In contrast, COP relative to the knee shifted laterally at both peaks external knee adduction moment (EKAM) after HTO (p < 0.001) and posteriorly at EKAM peak 2 (p = 0.012). Compared with controls, the COP remained more anterior relative to the knee center at both time points, indicating altered dynamic balance strategies during walking.

3.7. Brief Synthesis

Overall, short‐term static postural measures after HTO show heterogeneous results (some studies report early selective improvements, others no change), whereas mid‐ to long‐term follow‐up more consistently indicates improvement in weight distribution/symmetry and some functional measures (e.g., TUG), especially when adequate mechanical axis correction is achieved. Persistent deficits in general postural steadiness (stability index or many COP parameters) are reported in several studies despite improved load sharing, suggesting that alignment correction alone may not fully restore all aspects of postural control. Heterogeneity in surgical technique, rehabilitation reporting, healthy reference group matching, and outcome timing likely contributes to inconsistent findings across studies.

4. Discussion

This systematic review synthesized nine observational studies evaluating postural balance after HTO in patients with KOA. KOA itself is strongly associated with impaired balance due to pain, quadriceps weakness, altered proprioception, and maladaptive neuromuscular control, and several of the included studies confirmed that patients demonstrated greater postural sway than healthy reference group preoperatively [5, 11, 13, 14]. Previous research has established that impaired proprioception [17] and quadriceps weakness [18] are major contributors to instability in KOA, further explaining why balance deficits are evident even before surgery.

HTO aims to correct varus malalignment and redistribute mechanical load from the diseased medial compartment, thereby reducing pain and slowing progression [19]. However, this primarily structural intervention cannot directly reverse proprioceptive deficits or restore neuromuscular control, which likely explains why alignment correction improves some balance‐related outcomes but not others. These findings should be interpreted with caution, as most included studies were of moderate methodological quality and were limited by factors such as small sample sizes, heterogeneity in rehabilitation protocols, and lack of independent control groups. Reviews of HTO confirm that mechanical realignment reduces joint load and pain [1, 20], but neuromuscular deficits require separate interventions.

The temporal pattern of outcomes across studies is particularly informative. In the short term (up to 3 months), results were inconsistent: some studies reported small improvements in mediolateral sway or weight‐bearing symmetry [3, 4], while others found no significant changes or even increased sway velocity in specific planes [5]. By contrast, mid‐ to long‐term follow‐up (≥ 12 months) more consistently showed improvement in weight distribution between limbs and in certain functional tests such as the TUG [5, 13], yet overall postural steadiness often remained below that of healthy reference group [11, 13]. These findings suggest that mechanical correction and pain relief translate relatively quickly into more symmetric loading, but that the recovery of steady postural control is slower, incomplete, and possibly dependent on rehabilitation.

Closer inspection of the outcome measures underscores this point. The COP‐based parameters measured during quiet standing highlight different aspects of stability: while mediolateral weight distribution and symmetry improved after HTO in several studies [5, 11, 13], global sway measures such as COP trajectory, sway velocity, and stability index often showed limited or no recovery [5, 16]. Moreover, studies using more demanding tasks such as single‐leg stance reported persistent instability, even when bipedal stance measures had normalized [4, 16]. Similarly, dynamic COP trajectories recorded during gait revealed altered patterns even after surgery: Whatling et al. [14] showed that although COP shifted laterally relative to the knee joint after HTO, it remained different from controls, indicating that realignment may alter gait mechanics without fully restoring normal dynamic balance. Thus, the interpretation of “improvement” depends heavily on which COP metric is used, under what task conditions, and at what time point.

Functional balance assessments provided additional insight. The TUG improved significantly at follow‐up in some studies [16], and the Tinetti score correlated with the degree of mechanical axis correction [12], suggesting that functional performance is sensitive to successful alignment. However, these improvements did not always coincide with normalization of postural sway indices, supporting the idea that balance recovery involves both biomechanical and neuromuscular components.

The observation that balance outcomes often improved compared with preoperative values but still lagged behind healthy reference group is consistent across studies. Several explanations are plausible. First, alignment correction and pain reduction allow patients to load the operated limb more symmetrically, explaining within‐subject gains [21, 22], but this does not restore the proprioceptive acuity or neuromuscular efficiency of healthy individuals. Second, in some studies, healthy reference groups were younger or not matched for key variables such as sex and BMI (e.g., Whatling [14]), which may have exaggerated the observed differences between patients and controls. Finally, rehabilitation protocols varied markedly and were poorly reported; while some studies received structured physiotherapy, others had minimal or unstandardized rehabilitation. This heterogeneity likely contributed to discrepancies in outcomes and highlights that surgery alone is insufficient to fully restore balance without systematic postoperative neuromuscular training. Evidence from systematic reviews confirms that proprioceptive and balance training significantly improves function and postural control in KOA and after lower‐limb surgery [23, 24].

Third, postoperative rehabilitation protocols varied considerably and were often insufficiently reported. While some studies received structured physiotherapy, others underwent minimal or non‐standardized rehabilitation. This heterogeneity likely contributed to variability in outcomes and may confound the isolated effect of HTO on postural balance. Finally, the limited availability of long‐term follow‐up data (≥ 2 years) restricts the interpretation of whether these improvements are sustained over time, making it difficult to draw firm conclusions about the long‐term impact of HTO on postural stability. Overall, these findings suggest that although HTO may contribute to improvements in load distribution and functional performance, surgery alone is unlikely to fully restore postural control without targeted neuromuscular rehabilitation. Consistent with this interpretation, evidence from systematic reviews indicates that proprioceptive and balance training can significantly improve function and postural control in individuals with KOA and following lower‐limb surgery [23, 24].

Given the limitations of surgical correction in fully restoring proprioceptive input and neuromuscular control, emerging evidence indicates that virtual reality–based exercises can significantly enhance muscle excitability and improve balance in individuals with KOA [25, 26]. This highlights the potential of integrating innovative, technology‐driven neuromuscular interventions with surgical treatment to optimize recovery. Such approaches may help translate the mechanical advantages achieved through HTO into more effective functional postural stability.

Previous systematic reviews on HTO have mainly focused on gait biomechanics [6], radiographic outcomes [7], clinical function [27], or valgus knee correction [9]. None specifically examined balance outcomes. The present review therefore adds new knowledge by demonstrating that although HTO effectively corrects varus deformity and improves load distribution, postural stability is only partially restored, with persistent deficits evident in many COP‐based parameters. These findings emphasize the need to combine HTO with targeted rehabilitation focusing on proprioception, dynamic stability, and neuromuscular re‐education. Integrating such interventions into routine postoperative care may help translate biomechanical corrections into more complete functional recovery and potentially reduce fall risk in this population.

4.1. Limitations and Future Research Directions

This review has several limitations that should be acknowledged. First, the number of available studies investigating postural balance after HTO remains limited, and most are based on non‐randomized longitudinal (pre–post) observational designs with relatively small sample sizes, which may restrict the generalizability of the findings. Second, substantial methodological heterogeneity across studies—including differences in surgical techniques, follow‐up durations, rehabilitation protocols, and balance outcome measures—complicated direct comparisons and precluded the conduct of a meta‐analysis. Consequently, the evidence synthesis was limited to a narrative approach, and quantitative estimation of effect sizes was not feasible. Third, variability in postoperative rehabilitation protocols may act as a confounding factor, potentially obscuring the isolated effect of HTO on balance outcomes and contributing to inconsistencies in reported findings, particularly in the short term. Fourth, although the overall methodological quality of the included studies was rated as moderate to high based on JBI critical appraisal tools, important limitations remain. These include the lack of independent control groups in several studies, limited control of confounding variables, and insufficient reporting of sample size justification, all of which may affect the internal validity of the findings. Finally, the predominance of observational study designs inherently limits causal inference. Therefore, the observed improvements in balance outcomes should be interpreted as associations rather than definitive treatment effects.

Future studies should address these limitations by adopting adequately powered prospective cohort or randomized controlled designs with standardized rehabilitation programs and long‐term follow‐up. Greater uniformity in the choice and reporting of balance assessment methods—including static and dynamic COP parameters, as well as validated functional balance tests—will improve comparability across studies and facilitate evidence synthesis. In addition, careful matching of control groups and transparent reporting of methodological details will strengthen the rigor of future work. Given that HTO primarily restores alignment rather than directly addressing neuromuscular deficits, future trials should also evaluate the added value of targeted neuromuscular and proprioceptive training programs in enhancing postural control and functional recovery after surgery.

5. Conclusion

In summary, the available evidence suggests that balance outcomes may improve following HTO, particularly in weight distribution and certain functional measures at mid‐ to long‐term follow‐up; however, global postural stability often remains inferior to that of healthy individuals. These findings indicate that correction of alignment alone may be insufficient to fully restore postural control in patients with KOA. Incorporating structured rehabilitation strategies alongside surgical correction and improving methodological rigor in future studies are essential steps toward achieving a clearer understanding of the role of HTO in restoring postural balance.

Author Contributions

Hiva Farhadabadi: methodology, investigation. Rasul Gharekhan Maleki: data curation, methodology. Aydin Mahmoud Alilou: methodology. Farnaz Farshchi: methodology, investigation. Maryam Mohammadi: methodology, investigation. Aliyeh Daryabor: conceptualization, supervision, writing – review and editing, writing – original draft, project administration.

Funding

The authors have nothing to report.

Disclosure

The lead author Aliyeh Daryabor affirms that this manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained.

Ethics Statement

This manuscript is a systematic review was already registered in the PROSPERO database with the No. CRD420251144879. After studies' identification, the process was followed by the PRISMA guideline.

Consent

It has not been published elsewhere, and that it has not been submitted simultaneously for publication elsewhere.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File

Data Availability Statement

Data sharing not applicable to this article as no datasets were generated or analysed during the current study.