No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Comparative Analysis of Anatomic Coordinate Systems to Calculate Hindfoot Kinematics Using Biplane Fluoroscopy
Abstract
Accurate measurement of hindfoot kinematics during walking and other functional activities is essential for understanding pathologic foot conditions, tracking progression of deformity, evaluating impact of injury, and determining effectiveness of interventions. Multi-segmental foot models are now clinically used to evaluate hindfoot motion in a variety of populations; however, they are unable to isolate talocrural or subtalar motion because the talus lacks reliable palpable landmarks to place external markers. Biplane fluoroscopy provides a means to track motion of these joints by directly measuring the motion of each bone but there is currently no standard method for embedding local coordinate systems for measurement of talocrural and subtalar motion. Previous biplane work has assigned talar and calcaneal bone coordinate systems to be parallel to the tibia, though this may not accurately reflect the true orientation of these bones, particularly in cases of malalignment and deformation. The purpose to this work was to compare kinematic differences between a model with this coordinate system (tibia-aligned) and one in which each bone was assigned a local coordinate system based on anatomic landmarks (bone-independent). Biplane fluoroscopy data were collected for eight healthy subjects during over-ground walking. Bone-based coordinate systems were defined using anatomic landmarks visible on bone surfaces. Model-based tracking was used to align the digitally reconstructed radiographs of each bone with the fluoroscopic images from each time point. Kinematics were calculated using the tibia-aligned and bone-independent coordinate systems and talocrural and subtalar joint positions and excursions were compared. Results showed shifts in kinematic curves of up to 45° and range of motion differences of up to 6° in all three planes of motion. These results are important in recognizing the differences between these approaches to establishing coordinate systems, and the potential of each to be able to identify atypical joint motion and alignment. In addition, this work illustrates the importance of establishing a consensus for coordinate systems as biplane fluoroscopy assessment becomes more wide-spread.
... As a result, there are many individualized hindfoot bone coordinate system definitions and it is sometimes unclear what effect these numerous strategies have on the resulting kinematics. Recent work presented a comparison of the kinematics from the tibia-aligned and boneindependent coordinate systems showed variability among subjects, joints, and plane of motion (Kruger et al., 2019b). Comparison between these coordinate systems showed mean offsets in kinematic curves of up to 45°and excursion differences of up to 6°in all three planes of motion. ...
... In this review, several articles defined coordinate systems of individual bones based on their individual anatomy while others defined the axes of the talus and calcaneus as parallel to the tibia axes during standing. A previous comparison of these two methods reported shifts in kinematic curves of up to 45°and range of motion differences of up to 6°in all three planes of motion (Kruger et al., 2019b). This finding is consistent with a previous parametric study by Long et at. ...
The introduction of biplane fluoroscopy has created the ability to evaluate in vivo motion, enabling six degree-of-freedom measurement of the tibiotalar and subtalar joints. Although the International Society of Biomechanics defines a standard method of assigning local coordinate systems for the ankle joint complex, standards for the tibiotalar and subtalar joints are lacking. The objective of this systematic review was to summarize and appraise the existing literature that 1) defined coordinate systems for the tibia, talus, and/or calcaneus or 2) assigned kinematic definitions for the tibiotalar and/or subtalar joints. A systematic literature search was developed with search results limited to English Language from 2006 through 2020. Articles were screened by two independent reviewers based on title and abstract. Methodological quality was evaluated using a modified assessment tool. Following screening, 52 articles were identified as having met inclusion criteria. Methodological assessment of these articles varied in quality from 61 to 97. Included articles adopted primary methods for defining coordinate systems that included: 1) anatomical coordinate system (ACS) based on individual bone landmarks and/or geometric shapes, 2) orthogonal principal axes, and 3) interactive closest point (ICP) registration. Common methods for calculating kinematics included: 1) joint coordinate system (JCS) to calculate rotation and translation, 2) Cardan/Euler sequences, and 3) inclination and deviation angles for helical angles. The methods each have strengths and weaknesses. This summarized knowledge should provide the basis for the foot and ankle biomechanics community to create an accepted standard for calculating and reporting tibiotalar and subtalar kinematics.
Purpose:
The ankle rearfoot complex consists of the ankle and subtalar joints. This is an observational study on two test conditions of the rearfoot complex. Using high-speed biplane fluoroscopy, we present a method to measure rearfoot kinematics during normal gait and compare rearfoot kinematics between barefoot and shod gait.
Methods:
Six male subjects completed a walking trial while biplane fluoroscopy images were acquired during stance phase. Bone models of the calcaneus and tibia were reconstructed from computed tomography images and aligned with the biplane fluoroscopy images. An optimization algorithm was used to determine the three-dimensional position of the bones and calculate rearfoot kinematics.
Results:
Peak plantarflexion was higher (barefoot: 9.1°; 95 % CI 5.2:13.0; shod: 5.7°; 95 % CI 3.6:7.8; p = 0.015) and neutral plantar/dorsiflexion occurred later in the stance phase (barefoot: 31.1 %; 95 % CI 23.6:38.6; shod: 17.7 %; 95 % CI 14.4:21.0; p = 0.019) during barefoot walking compared to shod walking. An eversion peak of 8.7° (95 % CI 1.9:15.5) occurred at 27.8 % (95 % CI 18.4:37.2) of stance during barefoot walking, while during shod walking a brief inversion to 1.2° (95 % CI -2.1:4.5; p = 0.021) occurred earlier (11.5 % of stance; 95 % CI 0.2:22.8; p = 0.008) during stance phase. The tibia was internally rotated relative to the calcaneus throughout stance phase in both conditions (barefoot: 5.1° (95 % CI -1.4:11.6); shod: 3.6° (95 % CI -0.4:7.6); ns.).
Conclusions:
Biplane fluoroscopy can allow for detailed quantification of dynamic in vivo ankle kinematics during barefoot and shod walking conditions. This methodology could be used in the future to study hindfoot pathology after trauma, for congenital disease and after sports injuries such as instability. LEVEL OF EVIDENCE: II.
The purpose of this study was to quantify the accuracy and precision of a biplane fluoroscopy system for model-based tracking of in vivo hindfoot motion during over-ground gait. Gait was simulated by manually manipulating a cadaver foot specimen through a biplane fluoroscopy system attached to a walkway. Three 1.6-mm diameter steel beads were implanted into the specimen to provide marker-based tracking measurements for comparison to model-based tracking. A CT scan was acquired to define a gold standard of implanted bead positions and to create 3D models for model-based tracking. Static and dynamic trials manipulating the specimen through the capture volume were performed. Marker-based tracking error was calculated relative to the gold standard implanted bead positions. The bias, precision, and root-mean-squared (RMS) error of model-based tracking was calculated relative to the marker-based measurements. The overall RMS error of the model-based tracking method averaged 0.43 ± 0.22 mm and 0.66 ± 0.43° for static and 0.59 ± 0.10 mm and 0.71 ± 0.12° for dynamic trials. The model-based tracking approach represents a non-invasive technique for accurately measuring dynamic hindfoot joint motion during in vivo, weight bearing conditions. The model-based tracking method is recommended for application on the basis of the study results.
Measurements of joint kinematics are essential to understand the pathomechanics of ankle disease and the effects of treatment. Traditional motion capture techniques do not provide measurements of independent tibiotalar and subtalar joint motion. In this study, high-speed dual fluoroscopy images of ten asymptomatic adults were acquired during treadmill walking at 0.5 m/s and 1.0 m/s and a single-leg, balanced heel-rise. Three-dimensional (3D) CT models of each bone and dual fluoroscopy images were used to quantify in vivo kinematics for the tibiotalar and subtalar joints. Dynamic tibiotalar and subtalar mean joint angles often exhibited opposing trends during captured stance. During both speeds of walking, the tibiotalar joint had significantly greater dorsi/plantarflexion (D/P) angular ROM than the subtalar joint while the subtalar joint demonstrated greater inversion/eversion (In/Ev) and internal/external rotation (IR/ER) than the tibiotalar joint. During balanced heel-rise, only D/P and In/Ev were significantly different between the tibiotalar and subtalar joints. Translational ROM in the anterior/posterior (AP) direction was significantly greater in the subtalar than the tibiotalar joint during walking at 0.5 m/s. Overall, our results support the long-held belief that the tibiotalar joint is primarily responsible for D/P, while the subtalar joint facilitates In/Ev and IR/ER. However, the subtalar joint provided considerable D/P rotation, and the tibiotalar joint rotated about all three axes, which, along with translational motion, suggests that each joint undergoes complex, 3D motion.
Background:
Differences in foot structure are thought to be associated with differences in foot function during movement. Many foot pathologies are of a biomechanical nature and often associated with foot type. Fundamental to the understanding of foot pathomechanics is the question: do different foot types have distinctly different structure and function?
Aim:
To determine if objective measures of foot structure and function differ between planus, rectus and cavus foot types in asymptomatic individuals.
Methods:
Sixty-one asymptomatic healthy adults between 18 and 77 years old, that had the same foot type bilaterally (44 planus feet, 54 rectus feet, and 24 cavus feet), were recruited. Structural and functional measurements were taken using custom equipment, an emed-x plantar pressure measuring device, a GaitMat II gait pattern measurement system, and a goniometer. Generalized Estimation Equation modeling was employed to determine if each dependent variable of foot structure and function was significantly different across foot type while accounting for potential dependencies between sides. Post hoc testing was performed to assess pair wise comparisons.
Results:
Several measures of foot structure (malleolar valgus index and arch height index) were significantly different between foot types. Gait pattern parameters were invariant across foot types. Peak pressure, maximum force, pressure-time-integral, force-time-integral and contact area were significantly different in several medial forefoot and arch locations between foot types. Planus feet exhibited significantly different center of pressure excursion indices compared to rectus and cavus feet.
Conclusions:
Planus, rectus and cavus feet exhibited significantly different measures of foot structure and function.
Accurate clinical evaluation of the alignment of the calcaneus relative to the tibia in the coronal plane is essential in the evaluation and treatment of hindfoot pathologic condition. Previously described radiographic views of the foot and ankle do not demonstrate the true coronal alignment of the calcaneus relative to the tibia. Some of these views impose on the patient an unnatural posture that itself changes hindfoot alignment, whereas other methods distort the coronal alignment by the angle of the x-ray beam. Our purpose was to develop a modified radiographic view and measurement method for determining an angular measurement of hindfoot coronal alignment based on a cadaver study of the radiographic characteristics of the calcaneus and motion analysis of standing subjects.
The view was obtained by having the subject stand on a piece of cardboard to create a foot template. The template was then positioned so that each foot was x-rayed perpendicular to the cassette while still maintaining the natural base of support. A method using multiple ellipses was developed to determine more accurately the coronal axis of the posterior calcaneus.
A study using cadavers was performed in which radio-opaque markers were placed on multiple bony landmarks on the calcaneus. The tibia was held fixed in a vertical position, and the foot was x-rayed using the above techniques in different degrees of rotation without changing the relation of the calcaneus to the tibia. The radiographs of the modified Cobey and our view were examined to verify which markers were visible at different angles of rotation and how the hindfoot alignment measurements changed with foot rotation.
To define further the differences between the views, an analysis of postural stability was conducted while the subjects were standing with the feet in the positions for imaging both the Buck modification of the Cobey view and our hindfoot alignment view.
The combined results of the cadaver, radiographic measurement, and postural stability segments of the study reveal that this coronal hindfoot alignment view and measurement method is reproducible, more closely measures “true” coronal hindfoot alignment, and is more clinically applicable because the alignment is measured while the patient is standing with a normal angle and base of stance. The modified radiographic measurement method relies on posterior calcaneal anatomic landmarks, is less affected by rotation of the foot and ankle, and is reproducible between observers.
Three-dimensional data is required to have advanced knowledge of foot and ankle kinematics and morphology. However, studies have been difficult to compare due to a lack of a common coordinate system. Therefore, we present a means to define a coordinate frame in the foot and ankle and its clinical application.
We carried out ten CT scans in anatomically normal feet and segmented them in a general purpose segmentation program for grey value images. 3D binary formatted stereolithography files were then create and imported to a shape analysis program for biomechanics which was used to define a coordinate frame and carry out morphological analysis of the forefoot.
The coordinate frame had axes standard deviations of 2.36 which are comparable to axes variability of other joint coordinate systems. We showed a strong correlation between the lengths of the metatarsals within and between the columns of the foot and also among the lesser metatarsal lengths.
We present a reproducible method for construction of a coordinate system for the foot and ankle with low axes variability.
To conduct meaningful comparison between multiple subjects the coordinate system must be constant. This system enables such comparison and therefore will aid morphological data collection and improve preoperative planning accuracy.
We present a new kinematic model measuring the three-dimensional orientation of multiple segments of the foot and ankle. The model defines neutral alignments based on the alignments of the underlying bony segments, and indexes the orientation of skin-mounted markers to the bony anatomy using measures from weightbearing x-rays. The sensitivity of the model to these radiographic input parameters was analyzed using data from walking trials. Kinematic output in each plane was found to be most sensitive to perturbations of radiographic measurements in that same plane; however, perturbations in the coronal and transverse planes demonstrated significant carry-over into other planes. The analysis highlights the importance of accurately accounting for the underlying anatomy in measuring intersegmental kinematics.
A five-camera Vicon (Oxford Metrics, Oxford, England) motion analysis system was used to acquire foot and ankle motion data. Static resolution and accuracy were computed as 0.86 +/- 0.13 mm and 98.9%, while dynamic resolution and accuracy were 0.1 +/- 0.89 and 99.4% (sagittal plane). Spectral analysis revealed high frequency noise and the need for a filter (6 Hz Butterworth low-pass) as used in similar clinical situations. A four-segment rigid body model of the foot and ankle was developed. The four rigid body foot model segments were 1) tibia and fibula, 2) calcaneus, talus, and navicular, 3) cuneiforms, cuboid, and metatarsals, and 4) hallux. The Euler method for describing relative foot and ankle segment orientation was utilized in order to maintain accuracy and ease of clinical application. Kinematic data from a single test subject are presented.
Variations among joints in the initiation and progression of degeneration may be explained, in part, by metabolic, biochemical and biomechanical differences. Compared to the cartilage in the knee joint, ankle cartilage has a higher content of proteoglycans and water, as well as an increased rate of proteoglycan turnover and synthesis, all of which are responsible for its increased stiffness and reduced permeability. Chondrocytes within ankle cartilage have a decreased response to catabolic factors such as interleukin-1 and fibronectin fragments, compared to the chondrocytes of knee cartilage. Moreover, in response to damage, ankle chondrocytes synthesize proteoglycans at a higher rate than that found in knee cartilage chondrocytes, which suggests a greater capacity for repair. In addition to the cartilages of the two joints, the underlying bones also respond differently to degenerative changes. Taken together, these metabolic, biochemical and biomechanical differences may provide protection to the ankle.
Accurate knowledge of in vivo ankle joint complex (AJC) biomechanics is critical for understanding AJC disease states and for improvement of surgical treatments. This study investigated 6 degrees-of-freedom (DOF) in vivo kinematics of the human AJC using a combined dual-orthogonal fluoroscopic and magnetic resonance imaging (MRI) technique. Five healthy ankles of living subjects were studied during three in vivo activities of the foot, including maximum plantarflexion and dorsiflexion, maximum supination and pronation, and three weight-bearing positions in simulated stance phases of walking. A three-dimensional (3D) computer model of the AJC (including tibia, fibula, talus, and calcaneus) was constructed using 3D MR images of the foot. The in vivo AJC position at each selected position of the foot was captured using two orthogonally positioned fluoroscopes. In vivo AJC motion could then be reproduced by coupling the orthogonal images with the 3D AJC model in a virtual dual-orthogonal fluoroscopic system. From maximum dorsiflexion to plantarflexion, the arc of motion of the talocrural joint (47.5 +/- 2.2 degrees) was significantly larger than that of the subtalar joint (3.1 +/- 6.8 degrees). Both joints showed similar degrees of internal-external and inversion-eversion rotation. From maximum supination to pronation, all rotations and translations of the subtalar joint were significantly larger than those of the talocrural joint. From heel strike to midstance, the plantarflexion contribution from the talocrural joint (9.1 +/- 5.3 degrees) was significantly larger than that of the subtalar joint (-0.9 +/- 1.2 degrees). From midstance to toe off, internal rotation and inversion of the subtalar joint (12.3 +/- 8.3 degrees and -10.7 +/- 3.8 degrees, respectively) were significantly larger than those of the talocrural joint (-1.6 +/- 5.9 degrees and -1.7 +/- 2.7 degrees). Strong kinematic coupling between the talocrural and subtalar joints was observed during in vivo AJC activities. The contribution of the talocrural joint to active dorsi-plantarflexion was higher than that of the subtalar joint, whereas the contribution of the subtalar joint to active supination-pronation was higher than that of the talocrural joint. In addition, the talocrural joint demonstrated larger motion during the early part of stance phase while the subtalar joint contributes more motion during the later part of stance phase. The results add quantitative data to an in vivo database of normals that can be used in clinical diagnosis, treatment, and evaluation of the AJC after injuries.