Abstract
Background:
Accurate radiographic assessment of syndesmotic alignment is critical in the management of ankle injuries. Given limitations in traditional radiographic parameters, sagittal tibiofibular relationship (STFR) has been increasingly utilized. However, subtle variations in lower-limb rotation during imaging may alter STFR measurements, potentially leading to misinterpretation. This cadaveric study investigates the effect of ankle positioning on STFR in a cadaver model with intact syndesmotic ligaments.
Methods:
Twelve knee-disarticulated cadaver lower limbs with intact syndesmotic ligaments were used. True lateral fluoroscopic images were obtained using the talar dome double-shadow superimposition method, followed by incremental internal rotation (IR), external rotation (ER), abduction (ABD), and adduction (ADD) to simulate minor positional changes. Two foot and ankle fellowship-trained orthopaedic surgeons confirmed image acceptability. STFR was measured using the anterior tibiofibular interval (ATFI), posterior tibiofibular interval (PTFI), tibial/fibular width ratio (TW/FW), and anterior-posterior tibial-fibular distances (AF-AT, PF-PT). Pearson correlation analysis assessed associations between positioning and each measurement.
Results:
All STFR measurements were significantly correlated with axial plane rotation (P < .001) demonstrating that slight rotational malalignment in limb positioning was associated with apparent changes in tibiofibular alignment. Inter-rater reliability ranged from 0.75 to 0.95 and intrarater reliability from 0.63 to 0.98. ABD/ADD variation showed no significant effect on STFR measurements (all P > .05).
Conclusion:
Although increasing in utilization to assess syndesmotic alignment, STFR is sensitive to minor rotational deviations in the axial plane. Awareness of this limitation should be considered to avoid misinterpretation of syndesmotic reduction during intraoperative or postoperative imaging.
Clinical Relevance:
Radiographic assessment of syndesmotic alignment is important for guiding intraoperative reduction and postoperative evaluation in ankle injuries. This study suggested that STFR could be relatively sensitive to minor rotational limb malpositioning, which alter apparent tibiofibular positioning. These findings caution that small rotational deviations may significantly change STFR measurements and may lead to misinterpretation of syndesmotic reduction.
Introduction
Ankle syndesmotic injuries are relatively common injuries that occur in approximately 20% of malleolar fractures.1 -5 Accurate diagnosis and reduction is important as an untreated syndesmotic injury may lead to chronic ankle instability, persistent pain, post-traumatic osteoarthritis and poor function. 6 Previous works have demonstrated that sensitivity and specificity for syndesmosis injury on radiographs is 0.53 and 0.98, respectively, and computed tomography (CT) with 0.67 and 0.87, respectively. 7 The ability to detect syndesmotic malalignment or malreduction is critical as missed syndesmotic injury is the most common reason for malleolar fracture revision. 8
Malreduction of the syndesmosis is reported to occur in up to 16% of malleolar fractures. 9 Although CT and magnetic resonance imaging may be more sensitive and specific, it is not routinely feasible for intra-operative evaluation. 7 Consequently, most surgeons rely on intraoperative fluoroscopy. Although traditional evaluation of syndesmotic alignment used anterior and mortise views, there has been a recent emphasis on evaluating the sagittal tibiofibular alignment for assessing the syndesmosis on the lateral view.10 -12
Several measurements have been proposed to evaluate sagittal tibiofibular relationships (STFRs) and are typically assessed using a lateral view demonstrating talar dome superimposition. However, in our experience, even minor changes in internal/external rotation (IR/ER) or abduction/adduction (ABD/ADD) of the lower leg relative to the C-arm x-ray beam can preserve talar dome superimposition (thus producing an acceptable lateral view) while potentially altering the tibiofibular relationship. Given the limitations of fluoroscopic resolution and the complex contour of the talus, it remains unclear how sensitive STFR measurements are to minor malpositioning of the limb relative to the x-ray beam.
The purpose of this cadaveric study was to examine the effect of slight axial and coronal malpositioning of the limb relative to the x-ray beam on STFR using fluoroscopy. We hypothesized that small rotational deviations in ankle positioning would significantly alter STFR parameters, potentially mimicking syndesmotic malreduction and rendering these STFRs clinically not useful.
Methods
Experimental Design
Twelve fresh frozen, knee-disarticulated lower limb specimens with intact collateral and syndesmotic ligaments were used. Each specimen was mounted in a custom-designed apparatus that allowed controlled sequential rotation in both the axial (IR, ER) and coronal (ABD, ADD) planes. ABD and ADD of the ankle specimens (as would occur at the level of the hip) was created by translating the entire cadaver specimen. Fluoroscopic imaging was performed using a Siemens surgical C-arm Cios Select (Siemens AG), with rotational adjustments monitored by a digital inclinometer. The qualitatively determined midpoint of fluoroscopic images where the cadaver had complete superimposition of the talar dome was deemed the reference true lateral ankle fluoroscopic image. Starting from this position, each specimen was sequentially rotated in 2° increments of IR and ER up to 20° in each direction (Figure 1). The specimen was then returned to the neutral position and rotated in 5° increments of ABD and ADD up to 15° (Figure 1). Finer increments were used for axial rotation because of greater sensitivity of STFR measurements to anterior-posterior displacement. For statistical analyses, ER and ADD were coded as positive rotations (Figure 1).
Diagram of the experimental design of axial plane rotation (IR/ER) of the lower leg up to 20° (left) and coronal plane rotation (ABD/ADD) up to 15° (right). ABD, abduction; ADD, adduction; ER, external rotation; IR, internal rotation.
After completion of data collection, images were randomized. Two independent fellowship-trained foot and ankle orthopaedic surgeons, blinded to foot position, independently evaluated each image for acceptability as a true lateral view. Only images independently rated as acceptable by both reviewers were included in the analysis. The midpoint of the accepted rotational range was defined as the neutral (baseline) position. Nonconsensus images were excluded to ensure clinical relevance by limiting analyses to views that would plausibly be used for intraoperative syndesmosis assessment.
Radiographic Measurements
All measurements were completed using a free, open-source medical image viewer (version v3.3.6, Horos Project, https://horosproject.org/). As described by Croft et al, 11 we measured the tibial and fibular widths (TW, FW), the anterior tibiofibular interval (ATFI; anterior fibular cortex to anterior tibial cortex), and the posterior tibiofibular interval (PTFI; posterior tibial cortex to posterior fibular cortex) at 1 cm above the tibial plafond (Figure 2). The distance between the anterior fibula to the anterior tibia plafond (AF-AT) was measured as described by Xenos et al 12 (Figure 3). The distance from the posterior fibula to the posterior tibial plafond (PF-PT) was measured as described by Cogan et al 10 (Figure 4).
Diagram demonstrating the STFR ratio measurements as described by Croft et al. 11 Measurements are performed 1 cm above the midpoint of the tibial plafond. STFR, sagittal tibiofibular relationship.
AF-AT measurement of STFR as described by Xenos et al. 12 The perpendicular measurement tool is aligned along the tibial plafond. The line B is placed on the anterior tibial plafond, and the line C is placed at the intersection of the tibial plafond with the anterior fibula. The distance BC is the AF-AT. AF-AT, distance between the anterior fibula to the anterior tibia plafond; STFR, sagittal tibiofibular relationship.
PF-PT measurement of STFR as described by Cogan et al. 10 The perpendicular measurement tool is aligned along the tibial plafond. The line B is placed on the posterior tibial plafond, and the line C is placed at the intersection of the posterior fibula with the tibial plafond line. The distance BC is the PF-PT. PF-PT, distance from the posterior fibula to the posterior tibial plafond; STFR, sagittal tibiofibular relationship
Statistical Analysis
All continuous variables were summarized as mean ± SD. Pearson correlation coefficients were used to assess relationships between rotation magnitude (IR/ER and ABD/ADD) and each STFR parameter (α = 0.05). Normality of STFR measures was tested using the Shapiro-Wilk test. To assess reproducibility, each observer repeated 30 random measurements more than 1 month apart. Inter- and intrarater reliability were calculated using intraclass correlation coefficients (ICC), with ICC (2, 1) for inter-rater and ICC (3, 1) for intrarater agreement. All analyses were conducted using Pingouin (version 0.5.5) in Python (version 3.10.18) and reviewed by a statistician.
Results
Consensus was achieved for all cadavers on 132 of 324 radiographs (40.7%) as clinically acceptable lateral views, corresponding to limb positions between 8° of IR/ER and 5° of ABD/ADD. Although the proportion of clinically acceptable radiographs seems low, the experiment was designed to conservatively collect a wide range of angles and a low acceptance rate for acceptable lateral views was expected. All STFR parameters demonstrated normal distributions [ATFI/TW: P = .16; PTFI/TW: P = .07; ATFI/(ATFI + FW): P = .14; PTFI/(PTFI + FW): P = .10; AF-AT: P = .05; PF-PT: P = .41]. The baseline mean ± SD values for STFR parameters were 0.48 ± 0.09 for ATFI/TW, 0.20 ± 0.06 for PTFI/TW, 0.44 ± 0.06 for ATFI/(ATFI + FW), and 0.25 ± 0.06 for PTFI/(PTFI + FW). Mean ± SD distances for AF-AT and PF-PT were 8.10 ± 2.27 mm and 1.17 ± 2.89 mm, respectively. Tables 1 to 3 report median (IQR) values for each rotation angle. In 9 of 12 cadavers (75%), the posterior fibular border intersected within ±2 mm of the tibial plafond edge (Tables 1-3).
Abbreviations: ABD, abduction; ADD, adduction; ATFI, anterior tibiofibular interval; ER, external rotation; FW, fibular width; IR, internal rotation; N/A, not applicable; PTFI, posterior tibiofibular interval; STFR, sagittal tibiofibular relationship; TW, tibial width.
ER and ADD are the positive rotational directions. All measurements are recorded on true lateral radiographs agreed by foot and ankle orthopaedic surgeons.
Abbreviations: ABD, abduction; ADD, adduction; AF-AT, anterior fibula to anterior tibia plafond; ER, external rotation; IR, internal rotation; N/A, not applicable; STFR, sagittal tibiofibular relationship.
ER and ADD are the positive rotational directions. All measurements are recorded on true lateral radiographs agreed by foot and ankle orthopaedic surgeons.
Summary of the STFR Measurement as Described by Cogan et al. a
Abbreviations: ABD, abduction; ADD, adduction; ER, external rotation; IR, internal rotation; N/A, not applicable; PF-PT, posterior fibula to posterior tibial plafond; STFR, sagittal tibiofibular relationship.
ER and ADD are the positive rotational directions. All measurements are recorded on true lateral radiographs agreed by foot and ankle orthopaedic surgeons.
There was a significant correlation in all STFR measurement methods with axial rotation from −8° to 8° of rotation (all P < .001). Conversely along a perpendicular anatomical plane, there was no significant correlation with all the STFR measurements with ABD and ADD (ATFI/TW: P = .82; PTFI/TW: P = .40; ATFI/(ATFI+FW): P = .21; PTFI/(PTFI+FW): P = .16; AF-AT: P = .15; PF-PT: P = .43).
Intrarater reliability was documented for each measurement (Table 4). The inter-rater reliability was good to high for all measurements methods (range 0.75-0.95), and the intrarater reliability was good to high for both raters (range 0.63-0.98).
Intrarater and Inter-rater Reliability for the STFR Measurements. a
Abbreviations: AF-AT, anterior fibula to anterior tibia plafond; ATFI, anterior tibiofibular interval; FW, fibular width; PF-PT, posterior fibula to posterior tibial plafond; PTFI, posterior tibiofibular interval; STFR, sagittal tibiofibular relationship; TW, tibial width.
All measurements show good to high agreement between raters. Each observer repeated 30 measurements >1 month apart.
Discussion
Accurate coronal plane radiographic assessment of syndesmotic alignment is important in the management of ankle injuries. Given limitations in traditional radiographic parameters, STFR has been increasingly used. While increasing in popularity to assess syndesmotic alignment, the current investigation provides evidence that sagittal tibiofibular alignment measurements are sensitive to rotational variations in limb position, which may render these STFRs unreliable. The main clinical significance is that variations in ankle positioning may result in misinterpretation of syndesmotic tibiofibular relationships.
Tibiofibular overlap and clear space, as visualized on AP/mortise views, have traditionally been used to evaluate syndesmosis integrity. Typically, measurements on AP and mortise plain film radiographs are used to assess talocrural joint alignment and identify potential diastasis between the fibula and tibia under static and dynamic conditions. Radiographic assessments of syndesmosis injury has a sensitivity and specificity of 0.53 (95% CI: 0.38-0.67) and 0.98 (95% CI: 0.89-0.998), respectively. 7 Recent work suggests that the lateral views may improve radiographic syndesmosis assessments.10,11 Our investigation evaluated the consistency and reliability of lateral radiographic measurements with differential ankle positioning and found a significant correlation between axial plane rotation and all STFR measurement methods. Furthermore, we observed that surgeons frequently deemed imaging as true lateral ankle views despite the presence of up to 8° of ER/IR or 5° of ABD/ADD. Although this effect may have been seen with lesser degrees of ABD/ADD, finer increments of ABD/ADD angles were not examined because of constraints of our experimental rig. Although a limited sample size of surgeon observers (with inherent questions of generalizability to all surgeons), this suggests that intraoperative fluoroscopy may not be granular enough to detect minor variations in talar dome superimposition. Rather, clinically acceptable lateral views may still include substantial rotational variability, which has important implications for assessing sagittal tibiofibular alignment.
Croft et al 11 examined lateral radiographs of 72 disease-free ankles and reported an ATFI/TW ratio of 0.39 ± 0.09 with strong to very strong intrarater reliability and moderate inter-rater reliability without significant difference between age, sex, or laterality. We found a mean baseline ATFI/TW of 0.48 ± 0.09 for our cadaveric ankles. Under different axial rotations, they were significantly correlated with ATFI/TW measurements, with the most similar measurement to Croft et al being the acceptable lateral radiographs with 8° of IR at 0.41 ± 0.07. Intuitively, as the ankle was internally rotated, the fibular was projected more anteriorly on a lateral radiograph, which aligns with our findings of decreasing ATFI/TW with IR. Differences between this investigation and the current literature may also be attributable to selection criteria of included radiographs, anatomic variation of our cadaver specimens, and relatively smaller sample size.
Xenos et al 12 investigated the tibiofibular diastasis on 25 cadavers on mortise and lateral views. Although they did not report baseline AF-AT measurements, the current investigation demonstrated that AF-AT measurements significantly correlated with axial rotation. Cogan et al 10 studied adults undergoing ankle fracture open reduction internal fixation and used lateral views of their contralateral limb to assess for baseline syndesmotic alignment. They reported that 58% of normal ankles had their fibular posterior border intersect the tibial plafond within ±2 mm of the plafond edge. In our study we found that this radiographic relationship existed in 75% of our cadavers. When comparing the PF-PT measurements across the different ankle rotations, there was a significant correlation with axial plane rotation but not coronal plane rotations, similar to the Croft et al and Xenos et al methods.
All the STFR measurements on true ankle lateral radiograph significantly correlated with axial rotational positioning. To our knowledge, this is the first study exploring the correlation of STFR methods, which are candidate measures of syndesmosis integrity, to different ankle rotational positioning for acceptable true lateral radiographs.
Although variability of axial rotation within 8° can result in acceptable lateral imaging, rotational malalignment will affect measurements of lateral tibiofibular intervals. As STFR measurements may be used in assessing ankle alignment in various conditions including ankle instability and total ankle arthroplasty, attention should be paid towards standardizing limb positioning. Lateral imaging should continue to be interpreted in the setting of direct observation through open or arthroscopic exposure and AP and mortise radiographic views. Additional comparisons with the uninjured contralateral limb may control for anatomical variance. 13 However, care should be taken towards matching axial rotation relative to the x-ray beam since even with acceptable lateral views, ER and IR malpositioning may reduce the reliability of measurement and affect judgement of syndesmotic alignment.
This study had several limitations. First, we used 12 cadaveric specimens. Although our specimens were morphologically normal and this number is well within the sample size of many previously published anatomic studies, anatomic variation, and its influence on study findings (and generalizability to larger populations) is always of concern. Second, determination of a true lateral radiograph was performed by consensus of 2 fellowship-trained foot and ankle orthopaedic surgeons. Generalizability to a wider range of orthopaedic surgeons is unknown as there likely exists variation in what is considered an acceptable lateral view for either clinical or research purposes. Additionally, only 40.7% of available radiographs met consensus criteria introducing the possibility of selection bias. Although the extremes of the malpositioning tested were designed to be consistently rejected by orthopaedic surgeons, it is plausible that the true range of axial and coronal malpositioning may differ from our results. Finally, malpositioning of the ankle specimens was performed in 2 unidirectional manners to simplify our methodology. In clinical practice, the ankle may be malpositioned in multiple planes when obtaining intraoperative fluoroscopic imaging. However, although multiplanar malpositioning may have led to differential results, it is likely that similar correlations in lateral tibiofibular overlap would have been demonstrated. Future work should expand to in vivo or simulated intraoperative settings to validate these findings and consider including bilateral imaging.
Conclusion
Although increasing in utilization to assess syndesmotic alignment, STFR measurements are likely sensitive to rotational variations in axial limb position. Based on these cadaveric findings, surgeons should exercise caution during fluoroscopic acquisition and interpretation of lateral ankle views to avoid misdiagnosis of syndesmotic malalignment particularly when utilizing STFR.
Supplemental Material
sj-pdf-1-fao-10.1177_24730114261440555 – Supplemental material for Exploratory Analysis of Subtle Ankle Malpositioning on Sagittal Tibiofibular Alignment: Implications for Assessing Syndesmotic Malreduction
Supplemental material, sj-pdf-1-fao-10.1177_24730114261440555 for Exploratory Analysis of Subtle Ankle Malpositioning on Sagittal Tibiofibular Alignment: Implications for Assessing Syndesmotic Malreduction by Nicholas J. Yee, Mohammad Hossein Chegini-Kord, Ronit Avadhuta, Carl Habib Rai, Ivy Lee, Jaeyoung Kim, Soheil Ashkani-Esfahani and John Y. Kwon in Foot & Ankle Orthopaedics
Footnotes
Acknowledgements
Thank you to the team at the Foot & Ankle Research and Innovation Lab at Massachusetts General Hospital (Harvard Medical School), Orthopaedic Biomechanics Laboratory in the Holland Bone and Joint Program at the Sunnybrook Research Institute (University of Toronto), Institute of Biomedical Engineering (University of Toronto), and the Division of Orthopaedic Surgery (University of Toronto) for their support on this project. Figure 1 was created in BioRender by Yee, N. (2025; license: 
).
ORCID iDs
Ethical considerations
This is a cadaveric study. There are no human participants in this article and informed consent is not required.
Consent to participate
Not applicable
Consent for publication
Not applicable
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Nicholas J. Yee, MD, received research scholarship support from the Canadian Institutes of Health Research (236-2024-2025-Q2-00454 and 236-2024-2025-Q3-00103) and Ontario Ministry of Health and Long-Term Care, and salary support from the University of Toronto Department of Surgery.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. Disclosure forms for all authors are available online.
Data Availability Statement
Data generated or analyzed during the study are available from the corresponding author by request.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
