Association of coronal knee phenotype and joint line obliquity with proximal tibiofibular joint morphology in advanced osteoarthritis

Abstract

Purpose

The proximal tibiofibular joint (PTFJ) is biomechanically related to the tibiofemoral joint and may influence load transmission across the knee. However, it remains unclear whether PTFJ morphology is associated with coronal knee phenotype, joint line obliquity, or both. This study aimed to radiologically evaluate the relationship between coronal knee alignment, joint line obliquity, and PTFJ morphology in patients with advanced knee osteoarthritis.

Methods

This retrospective study included 324 patients with Kellgren–Lawrence grade 3–4 knee osteoarthritis who underwent standing long-leg radiographs and knee computed tomography. Coronal alignment parameters (mHKA, aHKA) and joint line obliquity were assessed on radiographs. PTFJ morphology—including coronal and sagittal inclination angles, joint obliquity, axial morphology, joint contour, fibular height, and fibula/tibia moment arm ratio—was evaluated using CT images. Associations between coronal knee phenotype, joint line obliquity, and PTFJ parameters were analyzed using non-parametric tests and categorical analyses.

Results

PTFJ morphological classifications differed significantly across coronal knee phenotypes. Flat PTFJ morphology was more frequent in varus-aligned knees, whereas convex axial morphology predominated in valgus alignment. Sagittal PTFJ inclination angle, tibial plateau–fibular axis angle, and fibula/tibia moment arm ratio differed significantly among alignment groups. Joint line obliquity was associated with differences in sagittal PTFJ inclination, fibular height, and related angular parameters, while coronal PTFJ slope and joint obliquity did not differ significantly.

Conclusion

In advanced knee osteoarthritis, PTFJ morphology is associated with coronal knee phenotype and joint line obliquity. These findings suggest that PTFJ morphology reflects the relative spatial relationship between the tibia and fibula rather than being solely a consequence of degenerative change, supporting a combined constitutional and load-related influence.

Introduction

The proximal tibiofibular joint (PTFJ) is a joint that plays a role in load transmission between the knee and the ankle and is functionally and biomechanically related to the tibiofemoral joint. Previous studies have demonstrated that morphological characteristics of the PTFJ are associated with tibiofemoral joint loading patterns, ¹ and a significant relationship between the severity of PTFJ osteoarthritis and tibiofemoral osteoarthritis has also been reported. ² Furthermore, certain PTFJ morphological parameters have been linked to osteoarthritic changes, while specific PTFJ morphological types have been shown to be particularly associated with lateral tibiofemoral compartment osteoarthritis.^3,4

In total knee arthroplasty, mechanical alignment (MA) and kinematic alignment (KA) approaches lead to distinct morphological and biomechanical consequences at the tibiofemoral joint. In varus knees, MA tibial resection typically requires greater bone removal from the lateral tibial plateau, which has been associated with a relative increase in lateral compartment laxity and increased loading of the medial compartment on radiographic assessment. ⁵ However, in certain CPAK phenotypes—particularly Types I and IV—unexpectedly lower lateral compartment contact pressures have been reported following MA, findings that cannot be fully explained by standard tibial resection techniques alone. Therefore, lateral load distribution may be associated with coronal knee alignment and PTFJ morphology—especially measurable radiological parameters such as relative joint height, sagittal inclination, and anteroposterior position—since these variables may influence the tension patterns of the lateral ligamentous complex and consequently alter lateral compartment loading across different degrees of knee flexion. ⁶

Current alignment concepts, including mechanical alignment strategies, largely disregard the tibiofibular relationship and assume that the PTFJ passively adapts to tibial realignment. ⁷ This assumption may be overly simplistic. There is insufficient evidence regarding whether PTFJ morphology is specific to coronal knee phenotypes or whether it is influenced by joint line obliquity. The aim of this study was to radiologically evaluate the relationship between PTFJ morphology and coronal knee alignment. We hypothesized that PTFJ morphological parameters differ significantly according to coronal knee phenotype and joint line obliquity.

Methods

The study was designed retrospectively. Approval was obtained from our institutional ethics committee under decision number 2025-351, and the study was conducted in accordance with the principles of the Declaration of Helsinki. Measurements were taken using version 4.3 of the Extreme XDS PACS software program, utilizing our hospital’s Image Archiving and Communication Systems (PACS). Patients who had undergone lower limb length radiographs and knee computed tomography (CT) between January 2023 and January 2025 for the evaluation of coronal knee alignment and proximal tibiofibular joint (PTFJ) morphology were included. CT scans had originally been performed for clinical indications unrelated to this study. Patients older than 18 years with appropriately acquired lower limb radiographs and knee CT scans, and who had Kellgren–Lawrence grade 3–4 osteoarthritis ⁸ were enrolled. Exclusion criteria were prior surgery around the knee (osteotomy, ligament reconstruction, arthroplasty, or arthroscopy), hip surgery due to osteoarthritis, trauma-related treatment, inflammatory or infectious arthritis, tumoral lesions, congenital disorders, or insufficient image quality.

For coronal alignment assessment, long-leg radiographs were obtained. ⁹ All images were acquired with patients standing, knees and hips in extension, and os patella centered. Radiographs were performed by experienced radiology technicians with a film–focus distance of 130 cm, 60 kV, 200 mA, and 50 ms exposure.

The mechanical hip–knee–ankle angle (mHKA), mechanical lateral distal femoral angle (mLDFA), and mechanical medial proximal tibial angle (MPTA) were measured. The arithmetic HKA (aHKA) was calculated using the formula aHKA = MPTA – mLDFA. aHKA value less than −2° was considered varus alignment, and a value greater than +2° was considered valgus alignment. The joint line obliquity (JLO) was defined as MPTA + LDFA. A neutral JLO was considered to be 180° ± 3°; values greater than 183° were considered proximal apex, and values less than 177° were considered distal apex ⁵ (Figure 1).

Figure 1.

Measurement of LDFA and MPTA angles for aHKA.

For PTFJ evaluation, previously acquired knee CT scans of the same patients were used. Coronal and sagittal tibiofibular joint inclination angles were measured using reconstructed CT images. Simulated CT scans were used to evaluate coronal (Figure 2(a)) and sagittal proximal tibiofibular joint (PTFJ) inclination angles (Figure 2(b)). In the simulated sagittal and coronal planes, the sagittal and coronal inclination angles were measured using a line drawn perpendicular to the longitudinal axis of the fibula.^10,11 In simulated sagittal CT scans, angles less than 20° were classified as horizontal, whereas angles greater than 20° were classified as oblique. ¹² The distance between the proximal articular surface of the tibia and the most proximal point of the fibular head was defined as the fibular head–tibia joint distance (Figure 3). The cPLATO–cPTFJ angle ¹³ (Figure 4(a)), moment arm ratio (fibula/tibia) ¹⁴ (Figure 4(b)) and Tibia plateau-Fibular axis angle ¹⁵ (Figure 4(c)) were measured on knee AP radiographs. The joint contour was categorized into three types: type A, with straight articular lines of both tibia and fibula; type B, with a convex fibular articular line; and type C, with a convex tibial articular line^10,16 (Figure 5). On axial CT images, PTFJ morphology was classified into four groups: flat, convex, concave, and atypical (Figure 6). PTFJ Joint obliquity was defined on axial CT images as the angle formed between the PTFJ joint line and the posterior tangent line of the tibia^10,16 (Figure 7).

Figure 2.

Coronal and sagittal PTFJ angles.

Figure 3.

Fibula head-Tibia joint distance.

Figure 4.

(a)-cPLATO – cPTFJ angle (b)-Moment arm ratio (fibula/tibia) (c)-Tibia plateau-Fibular axis angle.

Figure 5.

Proximal tibiofibular joint contour morphology types. Type A demonstrates straight tibial and fibular articular contours; Type B demonstrates a convex fibular articular contour; and Type C demonstrates a concave fibular articular countour.

Figure 6.

Schematic illustration of axial proximal tibiofibular joint (PTFJ) morphology types on axial computed tomography images, including flat, convex, concave, and atypical configurations.

Figure 7.

PTF Joint obliquity.

The relationship between tibiofemoral joint obliquity (neutral, distal apex, or proximal apex), coronal alignment phenotype (varus, neutral, or valgus), and PTFJ parameters was analyzed statistically.

Statistical analysis

The normality of continuous variables was assessed using visual methods (histograms and Q–Q plots) and analytical testing (Shapiro–Wilk test). Normally distributed variables were presented as mean ± standard deviation, whereas non-normally distributed variables were reported as median (minimum–maximum). Differences in continuous PTFJ morphological parameters (coronal and sagittal inclination angles and joint obliquity) across coronal knee phenotypes were evaluated using the Kruskal–Wallis test; for variables demonstrating significant overall group differences, pairwise comparisons were performed using Dunn’s test with Bonferroni correction. In addition, comparisons of continuous PTFJ morphological parameters between joint line obliquity subgroups (apex-neutral vs apex-distal) were conducted using the Mann–Whitney U test. Patients with proximally oriented joint line obliquity were identified; however, due to the limited sample size, this subgroup was not included in comparative statistical analyses to avoid insufficient statistical power. Associations between coronal knee phenotypes and categorical PTFJ morphological classifications (horizontal/oblique type, joint contour types, and axial appearance variations) were analyzed using the chi-square test or Fisher’s exact test when expected cell counts were low. All radiological measurements were independently performed by two experienced observers who were blinded to clinical and alignment data. Intra-observer and inter-observer reliability were assessed using the intraclass correlation coefficient (ICC). A p value <0.05 was considered statistically significant for all analyses.

Results

Of the total 324 patients, 230 (71%) were female and 94 (29%) were male. The mean age was 67.77 ± 8.72 years, with a maximum of 88 and a minimum of 47 years. Inter-observer and intra-observer reliability analyses demonstrated excellent agreement for all radiological measurements. The intra-observer intraclass correlation coefficient (ICC) values were 0.931 for the first observer and 0.914 for the second observer, while the inter-observer ICC value was 0.909. All reliability analyses were statistically significant (p < 0.001).

The distribution of PTFJ variations showed significant differences among coronal alignment groups (p < 0.001) (Table 1). Although the concave type was the most common variation in all groups, the flat type was observed at a higher rate in varus knees. Joint contour types also demonstrated significant differences according to coronal alignment (p = 0.011); while Type C was predominant across all groups, Type A was more frequently observed in varus knees. Regarding PTFJ type, a significant difference was found among groups, with the oblique type being markedly more prevalent in all alignment groups (p < 0.001) (Table 1). In classification based on joint obliquity, PTFJ variations and joint contour distribution showed significant differences (p = 0.04 and p < 0.001, respectively), whereas the distribution of PTFJ type was similar among groups (p = 0.149) (Table 1).

Table 1.

The relationship of PTFJ morphology with coronal alignment and joint obliquity.

​	Coronal alignment			p value*	Joint obliquity			p value*
	Varus n (%)	Neutrol n (%)	Valgus n (%)		Distal n (%)	Neutral n (%)	Proximal n (%)
PTFJ variations	​	​	​	<.001	​	​	​	.04
Flat	76 (%23.5)	12 (%3.7)	5 (%1.5)		72 (%22.)	20 (%6.2)	1 (%0.3)
Concave	143 (%44.1)	23 (%7.1)	27 (%8.3)		134 (%41.7)	44 (%13.7)	15 (%3.7)
Convex	20 (%6.2)	1 (%0.3)	8 (%2.5)		21 (%6.5)	8 (%2.5)	0 (%0.0)
Atypical	5 (%1.5)	4 (%1.2)	0 (%)		5 (%1.6)	4 (%1.2)	0 (%0.0)
Joint contour	​	​	​	.011	​	​	​	<.001
Type A	80 (%24.7)	5 (%1.5)	8 (%2.5)		53 (%16.5)	36 (%11.2)	4 (1.2%)
Type B	31 (%9.6)	8 (%2.5)	4 (%1.2)		39 (%12.1)	4 (%1.2)	0 (%0.0)
Type C	133 (%41)	27 (%8.3)	28 (%8.6)		140 (%43.6)	36 (%11.2)	9 (%4.9)
PTFJ type	​	​	​	<.001	​	​	​	.149
Horizontal	15 (%4.6)	4 (%1.2)	12 (%3.7)		19 (%5.9)	8 (%2.5)	3 (%0.9)
Oblique	229 (%70.7)	36 (%11.1)	28 (%8.6)		213 (%66.4)	68 (%21.2)	10 (%3.1)

Significant differences were observed in Sagittal PTFJ inclination angle, cPLATO–cPTFJ angle, and fibula/tibia moment arm ratio (p = 0.002; p = 0.002; p < 0.001, respectively), while coronal PTFJ slope angle (p = 0.585), PTFJ joint obliquity (p = 0.688), and fibular head–tibial joint distance (p = 0.434) did not differ significantly among groups. In valgus-aligned knees, the fibula/tibia moment arm ratio decreased (p < 0.001), whereas in varus knees the tibial plateau–fibular axis angle was lower (p < 0.001). In knees with distal apex orientation, sagittal PTFJ inclination angle and cPLATO–cPTFJ angle were higher (p < 0.001 and p = 0.008, respectively), while tibial plateau–fibular axis angle and PTFJ joint obliquity angle were lower (p < 0.001 and p = 0.002, respectively). Furthermore, fibular height was found to be significantly greater in knees with distal apex orientation compared to those with neutral orientation (p = 0.005) (Table 2).

Table 2.

Relationship between PTFJ measurements and coronal alignment and joint obliquity.

​	Coronal alignment			p* value	η2	Joint obliquity		p** value	r
	Varus median [IQR]	Neutrol median [IQR]	Valgus median [IQR]			Distal	Neutral
Coronal PTFJ inclination angle (degree)	19 [11.6]	18.05 [10.1]	20.05 [24.7]	.585	−0.00	19.45 [12]	17.3 [11]	.089	0.17
Sagittal PTFJ inclination angle (degree)	29.1 [9.9]	31.6 [7.57]	24.6 [11.7]	.002	0.03	29.85 [11.2]	26.8 [9.8]	<.001	0.206
PTF joint obliquity (degree)	52.6 [12.4]	49.5 [7.8]	54.7 [13.3]	.688	−0.00	51.2 [12.7]	54.8 [9.9]	.002	0.18
cPLATO – cPTFJ angle	44 [9.9]	46.85 [4.3]	41.15 [8]	.002	0.034	45.75 [9.8]	41.9 [9.4]	.008	0.151
Moment arm ratio (fibula/tibia)	2.08 [0.81]	2.25 [0.27]	1.62 [0.81]	<.001	0.08	2.14 [0.83]	2.07 [0.75]	.877	​
Fibula head-tibia joint distance (mm)	10.6 [4.5]	11.25 [6.8]	10.3 [4.1]	.434	−0.00	11.05 [4.4]	10.5 [6.4]	.005	0.159
Tibia plateau-fibular axis angle (degree)	82.7 [4.2]	85.1 [3.5]	86.06 [2.4]	<.001	0.09	82.8 [3.9]	85.3 [4.32]	<.001	0.205

*Kruskal Wallis -Overall group differences were assessed using the Kruskal–Wallis test. Pairwise comparisons were performed using Dunn’s test with Bonferroni correction for multiple comparisons.

**Mann-Whitney U.

Discussion

Previous studies investigating the relationship between proximal tibiofibular joint morphology and knee osteoarthritis have demonstrated that certain parametric PTFJ measurements may influence medial tibiofemoral compartment loading, particularly variables such as joint surface area, load-bearing surface, and sagittal plane inclination, which have been associated with medial compartment osteoarthritis. ³ Xin-Zheng Qi et al. ⁶ reported that morphological parameters of the PTFJ, including relative articular height and declination, were associated with coronal knee alignment in osteoarthritic populations. However, in the present study, varus or valgus alignment alone could not be associated with a significant change in fibular head distance. In contrast, when stratifying patients according to tibiofemoral joint line obliquity defined on anteroposterior radiographs (apex neutral or distally), significant differences were observed in both fibular head distance and sagittal obliquity of the PTFJ. Specifically, groups with a distally oriented joint line apex demonstrated greater fibular head distance and increased sagittal PTFJ inclination, whereas a Type A (flat) joint morphology predominated in neutral-apex groups. These findings suggest that the spatial configuration of the PTFJ may vary not only according to the presence of gonarthrosis but also in relation to joint line obliquity. Accordingly, the geometric positioning of the PTFJ may be associated with joint line orientation independently of conventional medial–lateral osteoarthritis determinants.

Qi et al. suggested that variations in proximal fibula height may contribute to knee osteoarthritis through changes in the fibula moment arm. ⁶ The study by Qi et al. reflects an indirect radiological estimation of the moment arms. In our study, no difference in proximal fibula height was observed in the Kellgren-Lawrence grade 3–4 patient cohort according to varus or valgus alignment. This discrepancy can be explained by differences between early and late stages of the disease, as previously suggested. Furthermore, no significant difference in the fibula moment arm was observed between groups classified according to tibiofemoral joint inclination (neutral and distal apex). Our findings suggest that PTFJ morphology may be more closely related to the relative coronal alignment between the tibia and fibula rather than joint line inclination.

Previous reports have demonstrated associations between knee osteoarthritis and more oblique PTFJ configurations, ⁴ as well as increased sagittal PTFJ angles. ⁶ In our cohort of patients with advanced knee osteoarthritis (KL grades 3–4), simulated sagittal CT images similarly revealed a predominance of more oblique PTFJ configurations, with mean sagittal and coronal angles of 29.2 ± 7.2° and 21.2 ± 10.2°, respectively. Importantly, this increased PTFJ obliquity was not influenced by coronal knee alignment (varus vs valgus), despite advanced degenerative changes. In addition, the association reported by Norifumi Suga et al. ¹² between a more horizontal PTFJ type and discoid lateral meniscus supports the notion that PTFJ morphology may be at least partially determined by constitutional or structural factors and may remain relatively stable independent of advanced degenerative changes.

Previous studies have also shown that PTFJ morphology may influence joint translation and tibiofemoral loading patterns, and that different PTFJ types can alter mechanical load distribution across the knee. ⁴ Jun Chang et al. reported that increased joint surface area and greater sagittal PTFJ obliquity may be associated with increased medial compartment loading. ² Similarly, Xin-Zheng Qi demonstrated that PTFJ morphology differs according to coronal knee alignment in advanced osteoarthritis. ⁶ In the present study, PTFJ types were associated with coronal knee alignment, with flat PTFJ morphology more frequently observed in varus knees and convex axial morphology predominating in valgus knees. These findings suggest that PTFJ morphology and knee alignment may be shaped through load-dependent adaptive changes. However, given the retrospective and cross-sectional nature of this study, causal inferences cannot be established, and these associations should be interpreted as hypothesis-generating rather than definitive evidence of adaptation.

The morphology of the proximal tibiofibular joint (PTFJ) appears to be influenced by degenerative changes, ¹⁷ tibiofemoral loading patterns, and lateral compartment biomechanics. In contrast, previous studies have associated certain PTFJ configurations with conditions such as discoid lateral meniscus, ¹² supporting the possibility of underlying constitutional anatomy. Therefore, the angles and morphological characteristics of the PTFJ may represent a hybrid phenomenon reflecting both intrinsic anatomical variation and long-term adaptive remodeling related to knee loading patterns.

These findings may have clinical relevance for alignment strategies used in total knee arthroplasty. Because the proximal tibiofibular joint is anatomically and biomechanically related to the lateral and posterolateral stabilizing structures of the knee, variations in PTFJ morphology may influence soft-tissue tension following coronal realignment. ¹⁷ In the study describing the CPAK classification system, differences in compartmental pressure distribution and dynamic pressure changes during flexion–extension were demonstrated among various knee phenotypes. ⁵ Although the present study does not establish a direct causal relationship, our findings suggest that PTFJ morphology may represent an additional anatomical factor relevant to individualized alignment strategies in knee arthroplasty.

The primary limitations of this study include its single-center, retrospective design and the inclusion of only patients with advanced knee osteoarthritis (KL grades 3–4). The lack of three-dimensional analysis software for assessing joint surface area represents another limitation. Nevertheless, the combined radiological evaluation of PTFJ morphology and biomechanical parameters in advanced knee osteoarthritis provides complementary data to the existing literature. Another limitation is the exclusion of patients with proximally oriented joint line obliquity from subgroup analyses due to the small number of cases, which limited our ability to evaluate PTFJ morphology across the full spectrum of joint line orientations.

In conclusion, in advanced knee osteoarthritis, PTFJ morphology should not be considered solely a consequence of degenerative changes but rather a structure shaped by coronal knee alignment and the relative spatial relationship between the tibia and fibula.