Khalifa AA, Lemma S, Fadle AA. In knee osteoarthritis, what is the knee deformity threshold after which the ankle realigns beyond the neutral zone? World J Orthop 2026; 17(5): 118547 [DOI: 10.5312/wjo.v17.i5.118547]
Corresponding Author of This Article
Ahmed A Khalifa, MD, Assistant Professor, FRCS, Department of Orthopaedics, Qena Faculty of Medicine and University Hospital, South Valley University, Kilo 6 Qena-Safaga Highway, Qena 83523, Egypt. ahmed_adel0391@med.svu.edu.eg
Research Domain of This Article
Orthopedics
Article-Type of This Article
Observational Study
Open-Access Policy of This Article
This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
Baishideng Publishing Group Inc, 7041 Koll Center Parkway, Suite 160, Pleasanton, CA 94566, USA
Share the Article
Khalifa AA, Lemma S, Fadle AA. In knee osteoarthritis, what is the knee deformity threshold after which the ankle realigns beyond the neutral zone? World J Orthop 2026; 17(5): 118547 [DOI: 10.5312/wjo.v17.i5.118547]
Author contributions: Khalifa AA conceived and designed the study, carried out a statistical analysis, and did the critical revision; Lemma S and Fadle AA performed data acquisition, assessment, and literature searches, and prepared the images and tables; Fadle AA performed the measurements; Khalifa AA and Lemma S drafted the manuscript; all authors read, discussed, and approved the final manuscript.
Institutional review board statement: This article does not contain any experimental studies with human participants or animals performed by any of the authors, and it was approved by Assiut Faculty of Medicine Institutional Review Board (Approval No. 4-2024-300470).
Informed consent statement: All study participants, or their legal guardian, provided informed written consent prior to study enrollment.
Conflict-of-interest statement: The authors declare no conflict of interest.
STROBE statement: The authors have read the STROBE Statement—checklist of items, and the manuscript was prepared and revised according to the STROBE Statement—checklist of items.
Corresponding author: Ahmed A Khalifa, MD, Assistant Professor, FRCS, Department of Orthopaedics, Qena Faculty of Medicine and University Hospital, South Valley University, Kilo 6 Qena-Safaga Highway, Qena 83523, Egypt. ahmed_adel0391@med.svu.edu.eg
Received: January 5, 2026 Revised: January 29, 2026 Accepted: February 27, 2026 Published online: May 18, 2026 Processing time: 133 Days and 13.7 Hours
Abstract
BACKGROUND
Coronal plane malalignment of the knee secondary to knee osteoarthritis (OA) is known to influence ankle joint alignment. Although the direction and magnitude of ankle joint realignment are well studied, the knee deformity threshold beyond which the ankle will align outside the accepted neutral coronal plane alignment remains incompletely defined.
AIM
To investigate the relationship between knee and ankle joint alignments in the coronal plane in patients with primary knee OA, and to identify knee deformity thresholds beyond which ankle alignment exceeds the accepted neutral range.
METHODS
This cross-sectional radiographic study included 845 lower limbs from 523 patients with primary knee OA. Coronal alignment was measured on long-leg standing radiographs as the hip to knee to ankle angle (HKAA) to assess knee deformity and the tibiotalar angle (TTA) to assess ankle joint alignment, with neutral alignment defined as 180° ± 2° for HKAA and 88.9° ± 3.0° for TTA. Associations between knee and ankle alignment were assessed using correlation analysis, categorical association testing, and multivariable regression. Direction-specific linear and logistic regression models were used to identify knee deformity thresholds associated with ankle malalignment. Analyses were adjusted for age, sex, and Kellgren-Lawrence grade.
RESULTS
Varus knee alignment was present in 74.8%, while valgus alignment was uncommon (5.0%). A significantly small-to-moderate association was observed between knee and ankle alignment categories (Cramér’s V = 0.19, P < 0.001), where varus ankle alignment was commonly observed with knee varus deformity and vice versa. HKAA and TTA demonstrated a moderate positive correlation (Spearman’s ρ = 0.396, P < 0.001), with an average increase of approximately 0.5° in TTA per 1° increase in HKAA. In multivariable linear regression, HKAA was the strongest independent predictor of TTA (β = 0.49, P < 0.001), while OA severity showed a smaller but significant independent association with varus ankle alignment. In varus knees, the HKAA threshold corresponding to a 50% probability of varus ankle malalignment was approximately 175.7°, with higher probabilities occurring at greater varus deformities. Valgus ankle malalignment occurred only at marked valgus alignment (approximately 189.2° HKAA).
CONCLUSION
Coronal knee alignment is the primary predictor of ankle alignment adaptation in knee deformity secondary to primary OA, with varus deformity exerting a stronger and more predictable influence than valgus deformity. Clinically relevant ankle malalignment occurs at about four degrees of varus knee deformity, whereas valgus ankle malalignment requires about eight degrees of valgus deviation.
Core Tip: We investigated 845 lower limbs with primary knee osteoarthritis (OA) and detected an association between knee deformity and ankle joint alignment in the coronal plane. Varus knee alignment was the most commonly occurring and was strongly associated with varus ankle alignment, whereas valgus knees showed heterogeneous ankle adaptation. The hip to knee to ankle angle (HKAA) and the tibiotalar angle were moderately correlated, with knee alignment as the significant determinant of ankle alignment, independent of age and sex. OA severity (Kellgren-Lawrence grades) was associated with progressively greater varus alignment at the knee and ankle. The ankle joint will align beyond the accepted neutral range at specific HKAA thresholds of approximately 175.7° and 189.2° for varus and valgus knee deformity, respectively. These thresholds were consistent across sex and OA severity, underscoring the primacy of coronal knee alignment in driving ankle compensation.
Citation: Khalifa AA, Lemma S, Fadle AA. In knee osteoarthritis, what is the knee deformity threshold after which the ankle realigns beyond the neutral zone? World J Orthop 2026; 17(5): 118547
Knee osteoarthritis (OA), whether primary or secondary, is one of the most common forms of joint arthritis, with a prevalence reaching up to 13% and 10% in women and men aged 60 years, respectively[1]. Although in its early stages, conservative management is adequate; however, if the disease progresses or a deformity develops, various surgical interventions are available, leading to deformity correction and possible alteration of lower limb biomechanics[2-4].
Besides preoperative planning and knee deformity analysis, nearby joints such as the hip and ankle should be evaluated, which is amenable as it is a common practice for surgeons to obtain long-leg [hip to knee to ankle (HKA)] radiographs preoperatively[5-8]. Such an evaluation is crucial, given that biomechanical relationships exist between lower limb joints, and it is well documented that deformity at one joint can affect the whole lower limb biomechanical unit, depending on the magnitude of the deformity and the resilience of the other joints[9-13].
The correlation between the knee and foot and ankle complex is an example of where deformity at one joint will force the other joints to compensate by a certain degree; however, if such compensation and realignment exceeded the accepted neutral alignment ranges, it might result in joint overload, pain, with possible OA development and progression[14-18]. Furthermore, the trauma literature showed that long-standing angular tibial deformity resulting from fracture malunion might induce ankle complications such as degeneration and OA[19,20].
Understanding and anticipating compensatory ankle realignment is beneficial for preoperative planning and patient counseling during knee surgery[21-24]. Furthermore, the extent of ankle malalignment associated with knee deformities is paramount, as suboptimal ankle realignment after knee deformity correction has been associated with less-than-optimal postoperative knee alignment, possible persistent postoperative ankle pain, and patient dissatisfaction[16,25-27].
Therefore, our main aim was to evaluate ankle joint alignment variation in association with knee deformity in the coronal plane in patients with primary knee OA and to define the knee deformity threshold beyond which the ankle will align outside the accepted neutral coronal alignment range. We hypothesized that the ankle joint would align differently depending on the direction of the knee deformity in the coronal plane (varus vs valgus), and at a certain degree of knee deformity, the ankle would align beyond its accepted neutral coronal alignment zone.
MATERIALS AND METHODS
Study design
This was a cross-sectional radiographic study conducted after obtaining Assiut Faculty of Medicine Institutional Review Board approval (Approval No. 4-2024-300470). We followed the ethical principles outlined in the Helsinki Declaration, and all patient data were kept anonymous. The study was conducted according to the STROBE guidelines[28].
Research participants, inclusion, and exclusion criteria
Between January 2021 and March 2025, we reviewed the radiographic records of skeletally mature patients who were diagnosed with primary knee OA graded from 1 to 4 per Kellgren-Lawrence (KL) grading scale, who were scheduled for or had either high tibial osteotomy (HTO), unicompartmental knee arthroplasty, or total knee arthroplasty (TKA). The radiographs were reviewed for eligibility criteria, which included a properly obtained standing long-leg anteroposterior plain radiograph showing the HKA, and the ankle joints appearing normal (free of arthritis, apparent deformities, previous fractures, or retained hardware). We excluded inappropriately obtained radiographs, patients who had extraarticular deformities, and those who had any lower limb interventions (such as hip surgeries or fracture fixation). Of the 640 patients’ radiographs evaluated, 523 were eligible, and 845 limbs were included in the final assessment.
Radiographic assessment
Two angles were measured (Figure 1). The HKA angle (HKAA), which assesses lower-limb alignment and knee deformity in the coronal plane, measured as the medial angle between the femoral and tibial mechanical axes. Where neutral alignment was considered as 180° ± 2° (178°-182°), and values below were considered varus knee alignment, while values above were considered as valgus alignment[5,29].
Figure 1 Long leg anteroposterior plain radiograph of the hip, knee, and ankle joints, which was used to measure the knee and ankle coronal alignment.
The hip to knee to ankle angle (HKAA) in orange is the medial angle between the mechanical axis of the tibia and femur; a neutral HKAA was defined as 180° ± 2°. The tibiotalar angle (TTA) in blue is the medial angle between the distal tibial axis and the line tangential to the upper talar surface; a neutral TTA was defined as 88.9° ± 3°. HKAA: Hip to knee to ankle angle; TTA: Tibiotalar angle.
The other angle was the tibiotalar angle (TTA) to assess ankle joint coronal alignment, which was the medial angle between the anatomical tibial axis and a line tangential to the superior talar articular surface, where the neutral TTA was considered as 88.9° ± 3° (85.9°-91.9°) according to Najefi et al[30], values above were considered valgus ankle, and values below were considered varus ankle.
Statistical analysis
These were performed using IBM SPSS Statistics (version 27.0; IBM Corp., Armonk, NY, United States), and generative artificial intelligence (ChatGPT 5.2, OpenAI) was used to assist in verifying statistical workflows, checking the internal consistency of results, and generating graphical visualizations using raw data provided by the authors. Continuous variables were assessed for normal distribution using the Shapiro-Wilk test. Associations between categorical variables were evaluated by the Pearson χ2 test, with effect size quantified using Cramér’s V (< 0.10 negligible, 0.10-0.19 small, 0.20-0.39 moderate, ≥ 0.40 strong). The relationship between the HKAA and the TTA was examined using Spearman’s rank correlation, and Pearson correlation was used as a supplementary analysis. To assess whether the knee-ankle relationship was independent of demographic and disease-related factors, multivariable linear regression was performed with TTA as the dependent variable and HKAA, age, sex, and KL grade as independent variables. Binary logistic regression was used to derive HKAA thresholds corresponding to 50%, 70%, and 80% predicted probabilities of ankle malalignment. In contrast, linear regression was used to quantify the average directional effect and not for threshold determination. Threshold robustness was assessed using non-parametric bootstrap resampling (1000 iterations) to generate bias-corrected 95% confidence intervals. Locally weighted scatterplot smoothing was used for exploratory visualization of non-linear relationships between HKAA and TTA. All statistical tests were two-tailed, and P < 0.05 was considered statistically significant.
RESULTS
Basic demographics and alignment distribution
A total of 845 lower limbs (523 patients) were analyzed, including 418 females (79.9%) and 105 males (20.1%), with a mean age of 56.8 ± 8.1 years. Age did not differ significantly between females (56.8 ± 7.9 years) and males (56.7 ± 8.7 years, P > 0.05). OA KL grades were distributed as follows: KL grade 1 in 18.9%, grade 2 in 34.6%, grade 3 in 32.1%, and grade 4 in 14.4% of limbs. Based on the HKAA, 632 knees (74.8%) were varus, 171 (20.2%) were neutral, and 42 (5.0%) were valgus. According to TTA, 459 ankles (54.3%) were varus, 310 (36.7%) were neutral, and 76 (9.0%) were valgus (Table 1).
Table 1 Basic demographics of the included patients and their knee and ankle alignment distribution.
A statistically significant small-to-moderate association was observed between knee alignment category and ankle alignment category (P < 0.001; Cramér’s V = 0.19). Among varus knees, varus ankle alignment predominated (386 of 632, 61.1%). In contrast, valgus knees exhibited a heterogeneous ankle alignment pattern, with 28.6% valgus, 28.6% varus, and 42.8% neutral (Table 2).
Table 2 The association between ankle alignment and each knee deformity category.
When analyzed as continuous variables, HKAA and TTA demonstrated a moderate positive monotonic association (Spearman’s ρ = 0.396, P < 0.001), indicating that increasing knee valgus alignment was associated with increasing ankle valgus alignment (Figure 2A-C). Pearson's correlation analysis yielded similar results (r = 0.38, P < 0.001). Linear regression analysis confirmed a significant positive directional effect, with an average increase of approximately 0.5° in TTA for each 1° increase in HKAA (β1 = approximately 0.48-0.56, P < 0.001).
Figure 2 Scatter plots illustrating the relationship between the hip to knee to ankle angle and the tibiotalar angle with shaded neutral zones (hip to knee to ankle angle: 178-182°, tibiotalar angle: 85.9-91.9°).
A: All the patients; B: Varus knees; C: Valgus knees; D: Stratified based on patients’ sex; E: Stratified based on the knee osteoarthritis Kellgren-Lawrence grades. The solid curve represents locally weighted scatterplot smoothing, used to explore non-linear trends without assuming a predefined functional form. A steeper and more consistent trend is observed in varus knees, whereas greater dispersion is seen in valgus knees. HKAA: Hip to knee to ankle angle; TTA: Tibiotalar angle; KL: Kellgren-Lawrence; LOWESS: Locally weighted scatterplot smoothing.
Sex-based interaction analyses
Males demonstrated slightly more varus HKAA than females (P = 0.018). Ankle alignment (TTA) did not differ between sexes (P = 0.692). The distributions of knee alignment categories (P = 0.384) and ankle alignment categories (P = 0.626) were also similar between sexes (Figure 2D and Table 3).
Table 3 Sex based differences in the demographics and alignment.
KL grade was significantly associated with both knee and ankle alignment (Figure 2E). Increasing KL grade was associated with progressively more varus knee alignment (P < 0.001; Spearman’s ρ = -0.298, P < 0.001) and more varus ankle alignment (P < 0.001; Spearman’s ρ = -0.258, P < 0.001). Categorical analyses demonstrated significant but small-to-moderate associations between OA grade and knee alignment category (Cramér’s V = 0.118) and ankle alignment category (Cramér’s V = 0.159; Table 4).
Table 4 Knee osteoarthritis severity-based differences in demographics and alignment.
Multivariable regression analysis for factors affecting ankle alignment
In multivariable linear regression analysis adjusting for age, sex, and KL grade, the HKAA remained the strongest independent predictor of TTA (β = 0.49, 95%CI: 0.36-0.61; P < 0.001). KL grade was also independently associated with TTA, with advancing KL grade corresponding to greater varus ankle alignment (β = -0.66, 95%CI: -1.29 to -0.02; P = 0.04). Age and sex were not independently associated with ankle alignment. In multivariable logistic regression analysis restricted to varus knees, increasing HKAA was associated with a significantly lower probability of varus ankle malalignment (OR = 0.79 per degree, 95%CI: 0.71-0.87; P < 0.001), reflecting a reduced likelihood of varus ankle compensation with decreasing knee varus deformity. Sex was independently associated with varus ankle malalignment, with males having lower odds than females (OR = 0.32, 95%CI: 0.17-0.62; P < 0.001). In contrast, KL grade and age were not independently associated with the probability of varus ankle malalignment (Table 5).
Table 5 Multivariable regression analyses of ankle alignment.
Using the accepted neutral TTA range (85.9°-91.9°), direction-specific modeling demonstrated marked asymmetry between varus and valgus deformities. Based on linear regression, the predicted mean TTA crossed into varus ankle malalignment (TTA < 85.9°) occurred at approximately HKAA = approximately 176.0° (bootstrap 95%CI: 175.1°-177.3°). In contrast, valgus ankle malalignment (TTA > 91.9°) occurred only at marked knee valgus alignment (HKAA = approximately 188.4°), with wide confidence intervals (bootstrap 95%CI: 185.1°-202.8°). Probability-based logistic regression yielded concordant results. In varus knees, the HKAA value corresponding to a 50% probability of varus ankle malalignment was approximately 175.7° (bootstrap 95%CI: 174.6°-177.4°), with higher probabilities (70%-80%) occurring at progressively greater varus deformities. In valgus knees, the 50% probability threshold was approximately 189.2°, with wide confidence intervals (185.6°-208.8°) due to the small valgus subgroup (n = 42); these findings should therefore be considered exploratory (Figure 3). Sex-specific analyses demonstrated nearly identical varus thresholds in females (175.6°) and males (175.8°). Furthermore, when stratified by OA severity, knee deformity thresholds for ankle malalignment showed minimal variation across KL grades (< 1.5°; Table 6).
Figure 3 Scatterplot showing the relationship between hip to knee to ankle angle and tibiotalar angle with locally weighted scatterplot smoothing.
Blue dashed vertical lines indicate hip to knee to ankle angle (HKAA) thresholds corresponding to 50%, 70%, and 80% predicted probabilities of varus ankle malalignment [tibiotalar angle (TTA)< 85.9°], while red dashed lines indicate corresponding thresholds for valgus ankle malalignment (TTA > 91.9°), derived from direction-specific logistic regression models. Shaded regions represent neutral knee HKAA (178°-182°) and ankle neutral TTA (85.9°-91.9°) alignment ranges. HKAA: Hip to knee to ankle angle; TTA: Tibiotalar angle.
Table 6 Knee deformity thresholds for ankle malalignment probability (°).
The crucial finding of the current study was that, among our patients diagnosed with primary knee OA, varus knee deformity predominated and was strongly associated with coronal-plane varus ankle alignment. Furthermore, the linear regression analysis confirmed a significant positive directional effect where an average increase of approximately 0.5° in TTA for each 1° increase in HKAA. The HKAA and TTA had a moderate correlation, with knee alignment as the dominant determinant of ankle alignment regardless of patients’ age or sex; however, knee OA severity, according to KL grading, was associated with progressively more varus knee and ankle alignment but did not materially alter the knee-ankle compensatory relationship. To identify the ankle alignment thresholds outside the acceptable neutral range, we determined HKA thresholds of 176.0° and 188.4° for varus and valgus knees, respectively.
Before discussing the results we obtained, several limitations of the current study should be highlighted. First, this was a radiological study, and no clinical correlation of the findings was proposed, which would have been beneficial for determining whether the ankle alignment changes were reflected in clinical complaints such as pain and discomfort. Second, the lower limb alignment, including the knee and ankle, was evaluated only in the coronal plane; however, assessment in the sagittal and axial planes is equally important, and it is worth noting that this relationship has been reported in previous studies[21]. Third, part of the compensation at the ankle level is attributed to realignment of the subtalar and hindfoot joints, which were not considered in the current study due to a lack of specific foot-assessment radiographic views. Fourth, we considered the literature-based ranges for neutral knee and ankle alignment, as we do not have data from our own population, and these ranges may be subject to ethnic variability. Lastly, the lack of valgus knees compromised the statistical analysis of this particular group, making the current study results more applicable to varus knees and less trustworthy for valgus knees.
Given that obtaining long-leg lower-limb radiographs during preoperative planning for knee osteotomy or arthroplasty is a routine practice in some institutes, including our institution, which facilitated the current study, evaluating compensatory realignment at the hip or ankle levels would be feasible and beneficial[5-8]. However, care should be taken when obtaining and assessing these radiographs, as inappropriate radiographs, such as those showing malrotation, may affect measurement accuracy[31].
A positive correlation between knee and ankle deformity is well established in the literature, where the knee varus deformity increases the ankle’s alignment into varus, and vice versa. Furthermore, the extent to which the ankle will realign can be calculated[17,23,32,33]. What we tried to offer in the current study was an approximate knee deformity value, which we estimated to be an HKAA of 176° for varus knees and 188.4° for valgus knees, beyond which the ankle will mostly align outside the defined ankle neutral alignment zone; furthermore, we found that the probability of such malalignment increases as the knee deformity increases.
What we obtained in the current study, when it comes to the significant correlation between the direction of knee deformity and ankle joint realignment, and the positive correlation between HKAA and TTA, where a one-degree increase in HKA will lead to a 0.5° increase in TTA, were similar to the results reported by Hodel et al[21] who evaluated 60 lower limbs in normal individuals having a mean age of 27.1 ± 10 (20-67) years, the authors reported a significant correlation between HKAA [mean of 1.7 ± 3.7 (-8.0 to 9.0)] and ankle joint line orientation in the coronal plane [mean of 0.5 ± 4.4 (-11.7 to 10.0)], furthermore, their multivariate regression analysis showed that HKAA valgus increase by one degree was associated with an increase in ankle joint valgus orientation by 0.5° (95%CI: 0.2°-0.7°; P < 0.001). Gao et al[12] evaluated the relationship between knee malalignment (measured as HKAA) secondary to OA and the effect on ankle alignment (measured as TTA), they included 149 patients, most of them were females, for varus knees, they found a significant positive correlation between the HKAA (mean of 4.8° ± 0.6°) and TTA (mean of 87.9° ± 4.8°; r = 0.243, P = 0.028). The same significant correlation was observed in valgus knees, where mean HKAA and TTA were 4.5° ± 0.5° and 1.3° ± 5.2°, respectively (r = -0.513, P = 0.010). The authors reported that this correlation might enhance ankle joint degeneration.
According to Gao et al[12], their patients did not complain of ankle pain, although ankle deformity and OA were observed on their radiographs; they hypothesized that this could be attributed to the masking effect of knee pain, combined with the limited physical activity of patients with advanced knee OA. Moreover, compensation occurs not only at the ankle joint but also at the subtalar joint, which might enhance the ability to compensate for knee deformity and distribute load[34]. According to our study, these compensatory mechanisms occur; however, once a knee deformity threshold is exceeded, the ankle will align outside accepted normal ranges, and, according to Gao et al[12], persistent lower limb deformity will inevitably lead to ankle degeneration.
Although we included patients with apparently normal ankle joints, subclinical early-stage degenerative arthritis might be present, which can be aggravated by further ankle deformity and unbalanced loads[35,36]. Moreover, patients who complain of ankle pain preoperatively, in association with or secondary to knee deformity, might experience improvement or worsening after deformity correction[9,37]. According to Xie et al[14] who evaluated 106 knees that underwent TKA; 36.8% had radiological signs of ankle OA, which were significantly associated with HKA (OR = 1.13, P = 0.017).
Kim et al[9] evaluated changes in ankle joint alignment after TKA or HTO and found a significant correlation toward a more valgus orientation after correction of knee varus deformity; furthermore, there was a medial-to-lateral shift in the weight-bearing line through the ankle. The authors concluded that such a shift might improve symptoms of medial ankle pain, but, conversely, it could worsen lateral ankle arthritis, further highlighting the importance of evaluating the ankle joint and its orientation relative to knee deformity. Furthermore, Wu et al[37] identified knee varus deformity > 6 degrees and preoperative ankle OA as strong predictors of persistent post-TKA ankle pain. It is worth noting that the knee deformity threshold for persistent postoperative ankle pain reported by Wu et al[37] was near the HKAA threshold of 4 degrees, which we defined as the HKAA at which the ankle aligns beyond the accepted neutral range.
Interestingly, Hodel et al[21] did not find a significant difference in ankle joint orientation by gender, which was similar to what we observed in our patients, where no difference was found in alignment categories or ankle joint alignment; however, slightly more varus knees were reported in male patients. On the contrary, Xie et al[38], who evaluated 95 knees (KL grades 3 and 4), did not observe a difference in ankle alignment secondary to knee deformity between males and females. However, they found no correlation between the HKAA and TTA. Furthermore, they found that changes in ankle alignment were significantly correlated with knee deformity in females but not in males.
In the current study, as knee OA severity increased (toward a higher KL grade), knee and ankle deformities increased. Furthermore, there was a significant but small association between the OA degree and the knee and ankle alignment categories, as shown in Table 4. Furthermore, when evaluating the threshold for knee deformity for the ankle to align outside the neutral alignment zone, there was a very minimal difference between KL grades (< 1.5°), which is clinically negligible. We believe that the effect of KL grades is secondary and closely related to the increasing knee deformity with advancing KL grades.
Changes at the foot and ankle levels secondary to the severity of knee OA, as graded by the KL scale, were reported by Hakukawa et al[39] where the authors assessed computed tomography in an upright position of 45 limbs. They found various realignments at the foot and ankle levels across different KL grades, with the varus knee deformity associated with ankle joint internal rotation in KL 1 knees (r = 0.76, P < 0.05). In KL 2 knees, with eversion at the subtalar and ankle joints (r = 0.63 and -0.65, respectively, P < 0.05). In KL 3 knees, with subtalar joint internal rotation (r = - 0.62, P < 0.05), and in KL 4 knees, with subtalar joint internal rotation (r = -0.58, P < 0.05). Interestingly, as they did not find a significant association between hindfoot alignment and knee deformity in nearly all planes, they concluded that as the knee OA becomes more severe, the compensatory realignment is dependent on the ankle rather than the subtalar joint, which further highlights the importance of assessing the ankle joint status during preoperative planning.
Hakukawa et al[39] study findings partially aligned with the findings we obtained, as they found progression of the knee varus deformity as the KL grade became more severe; however, they reported no significant difference in ankle joint coronal plane alignment across KL grades. Xie et al[38] reported a significant difference in some ankle parameters according to the KL grade; however, the TTA showed no difference between the two grades (mean of 88.86 ± 3.34 and 89.69 ± 3.09 for KL grades 3 and 4, respectively; P = 0.296). On the contrary, we found a significant difference in TTA across different KL grades, with TTA decreasing as the grade increased.
One last clinically relevant point should be highlighted regarding the observed ankle varus or valgus deformities on long-leg radiographs and the possibility of exiting the neutral alignment zone as the knee deformity progresses, as noted by Shih et al[40] who reported that preoperative valgus ankle deformity might not be corrected after correction of the knee deformity, compared with varus ankle alignment, is a crucial point to consider during preoperative planning.
CONCLUSION
Varus knee deformity is associated with early and consistent varus ankle alignment, whereas valgus knee deformity demonstrates heterogeneous and delayed ankle compensation. Furthermore, once a specific knee deformity value is reached, the ankle will align outside the accepted neutral coronal alignment range, which may be beneficial during preoperative planning. Although minor sex-related differences in knee alignment exist, age and sex do not meaningfully modify the relationship between knee and ankle alignment; however, the severity of knee OA has a significant but minor effect. Future studies should focus on including more valgus knees, evaluating compensation at the subtalar and hindfoot levels, and reporting the clinical effect of such alignment changes.
Van Oevelen A, Burssens A, Krähenbühl N, Barg A, Devos Bevernage B, Audenaert E, Hintermann B, Victor J. Osteotomies around the knee alter alignment of the ankle and hindfoot: a systematic review of biomechanical and clinical studies.EFORT Open Rev. 2023;8:818-829.
[RCA] [PubMed] [DOI] [Full Text][Cited by in RCA: 9][Reference Citation Analysis (0)]
van der Schoot DK, Den Outer AJ, Bode PJ, Obermann WR, van Vugt AB. Degenerative changes at the knee and ankle related to malunion of tibial fractures. 15-year follow-up of 88 patients.J Bone Joint Surg Br. 1996;78:722-725.
[PubMed] [DOI]
