Alignment of Adult-Acquired Flatfoot Deformity: A Comparison of Clinical Assessment and Weight Bearing Cone Beam CT Examinations

, Francois Lintz2, Cesar de Cesar Netto3, Alexej Barg4, Arne Burssens5 and Scott Ellis6

Department for Foot and Ankle Surgery, Hospital Rummelsberg, Schwarzenbruck, Germany

Foot and Ankle Surgery Centre, Clinique de l’Union, Toulouse, France

Department of Orthopedics and Rehab, University of Iowa, Iowa City, IA, USA

University Orthopedic Center, University of Utah, Salt Lake City, UT, USA

Department of Orthopedics and Trauma, University Hospital of Ghent, Ghent, OVL, Belgium

Department of Orthopedic Surgery, Hospital for Special Surgery, New York, NY, USA



Adult-acquired flatfoot deformityFlatfootWeight bearingWeight bearing CTCone beam CTHindfoot alignment


Adult-acquired flatfoot deformity (AFFD) represents a progressive and complex structural deformity of foot [1, 2]. Although posterior tibial tendon (PTT) dysfunction has historically been recognized as the principal culprit leading to collapse of the medial longitudinal arch (MLA) [3], further soft tissue insufficiency and underlying bony deformities have been implicated in the development of AFFD [4, 5]. Equinus contracture, spring ligament, interosseous and deltoid ligament attenuation, as well as an increased innate valgus angulation of the subtalar joint can predispose to and eventually lead to subtalar joint eversion and subsequent hindfoot valgus [4, 6].

Currently, a set of measurements based on clinical and radiographic examinations are employed to evaluate hindfoot alignment [7]. Although these measurements have been extensively described, substantial disagreement remains in clinical judgment and radiographic measures to define an accurate method for the evaluation of hindfoot alignment [8, 9]. Clinical assessment including visual evaluation and Harris mat footprint and quantitative measures such as valgus index have been defined; however they have shown to be highly unreliable due to significant interobserver variation even among experienced orthopedic surgeons [7, 8, 10].

Furthermore, radiographic assessment of hindfoot alignment is quite cumbersome. The two-dimensional nature of plain radiographs limits their accuracy, and optimal evaluation of hindfoot alignment is hampered by complex anatomy of subtalar joint [5, 11]. Besides, lack of standardized methods to evaluate the alignment is another source of disagreement [9]. Previous cohorts used distinct set of reference points as well as different hindfoot specific views including long axial view [1214]. Some studies used angular measurements, while others employed linear measurements. Prior reports have also indicated that malpositioning during image acquisition, inconsistent angulation, or superimposition could generate considerable measurement errors [7, 15]. Therefore, radiographic measures of hindfoot alignment are associated with major fundamental flaws due to several anatomical and observer-related bias.

Cross-sectional imaging modalities including computed tomography (CT) provide enhanced, detailed visualization of hindfoot with simultaneous demonstration of different structures; however, they are only able to provide images obtained with the patient supine [16, 17]. Additionally, in patients with AFFD, hindfoot instability has been observed when a weight bearing condition is applied [18]. Therefore, due to the major impact of loading on hindfoot alignment, it is crucial to assess suspected cases in the standing position [19]. Recent developments in CT scan design have contributed to the advent of cone beam computed tomography (CBCT). This novel technique allows imaging of lower extremity in a normal upright weight bearing state. Initial studies reported excellent image quality with sufficient contrast resolution to visualize soft tissue and bone exceeding conventional radiography and multiple detector computed tomography (MDCT) [20, 21].

Considering the ability of WB CBCT to demonstrate three-dimensional deformities in a standing physiologic setup with an enhanced visualization of bony landmarks and soft tissue structures, the application of this modality in patients with AFFD has recently been demonstrated to accurately reflect the effect of body weight in this dynamic deformity [22]. Measurements used in the staging and evaluation of the deformity were also reported to be repeatable and reliable when performed not only by experts [22] but also by in-training medical personnel [23]. Also recently, significant correlation between clinical and conventional radiographic hindfoot alignment was demonstrated in patients with flexible AAFD, but the radiographic measurements of hindfoot valgus were found to be significantly more pronounced valgus alignment than the standardized clinical evaluation [23]. Thus, in this study, we intended to compare clinical assessment of hindfoot valgus alignment with different possible hindfoot alignment measurements performed on WB CBCT images, in patients with AAFD. Our hypothesis was that measurements would correlate but different degrees of valgus alignment would be found, depending on the anatomical landmarks used.

Materials and Methods

Study Design

This dual-center IRB-approved prospective study complied with the Health Insurance Portability and Accountability Act (HIPAA) and the Declaration of Helsinki. Informed consent was signed by all study participants.


In the two involved tertiary hospital clinics, consecutive patients with clinical diagnosis of symptomatic flexible AAFD from October 2014 till June 2016 were recruited. We excluded patients who were younger than 18 years old, were not able to communicate efficiently with clinical study personnel, or were not able to stand still independently for at least 40 seconds. Individuals with the inability to bear weight and a rigid deformity or those who had standard contraindications for standard CT scans including pregnancy or those with serious medical or psychiatric issues were also not enrolled in this prospective study.

Coronal plane clinical hindfoot alignment of all study participants was measured in the physiologic WB position by the most experienced of the senior authors (LCS). Patients were instructed to stand in a comfortable and natural stance position, and the medial border of each foot was positioned over two parallel lines that were drawn into the floor, controlling for rotational misalignment. The measurement obtained here, named clinical hindfoot alignment angle (CHAA) (Fig. 16.1a), was similar to the standing tibiocalcaneal angle (STCA) and was based on clinical expertise of the senior author and evaluation of anatomical surface landmarks [8, 23].


Fig. 16.1

Example of measurements performed: (a) Clinical hindfoot alignment angle (CHAA); (b) weight bearing computed tomography clinical hindfoot alignment angle (WBCT CHAA); (c) Achilles tendon/calcaneal tuberosity angle (ATCTA); (d) tibial axis/calcaneal tuberosity angle (TACTA); (e) tibial axis/subtalar joint angle (TASJA); (f) hindfoot alignment angle (HAA); (g) hindfoot moment arm (HMA)

Following clinical examination, all patients underwent WB CBCT examinations.

CBCT Imaging Technique

All CT studies were conducted using a CBCT extremity scanner (Generation II, Carestream Health Inc., Rochester, NY). All imaging studies were performed under physiological upright WB position with the patients standing with their feet almost at shoulder width, distributing body weight evenly between their both legs. We employed the same scan protocol that was described in prior studies [22, 24, 25].

WB CBCT Measurements

To develop computer-based measurements, the raw 3D data were used to generate and create axial, coronal, and sagittal image slices and were digitally transferred into our dedicated software (Vue PACS, Carestream Health, Inc., Rochester, NY). Image annotations were deleted, and a unique random number was allocated to each imaging study.

Following a mentored training protocol entailing five AAFD cases, three fellowship trained foot and ankle surgeons applied different hindfoot alignment measurements independently using the dedicated software. All observers were blinded, and the order of images was randomized. One month following the first assessment, a second set of measurements for intraobserver reliability was performed by one of the observers. The measurements were performed on the clinical reconstructed 3D images. Rotational position control was assured with the use of images where the medial aspect of the heel and the most medial aspect of forefoot and the medial eminence of the first metatarsal were in line with each other. The images used were also tangential to the floor.

The first measurement performed aimed to mimic the clinical hindfoot alignment evaluation and was named WBCT clinical hindfoot alignment angle (WBCT CHAA). It was obtained on 3D images where the windowing was set to maintain the surface anatomical landmarks, including the skin (Fig. 16.1b).

The second measurement was performed in an image with the same positioning, but the windowing was changed, removing the skin and subcutaneous, but keeping some of the overlying soft tissue structures including the Achilles tendon. That image also allowed a better evaluation of the calcaneal tuberosity, used as a bony anatomical landmark. The angle measured was formed by the longitudinal axis of the Achilles tendon and the longitudinal axis of the calcaneal tuberosity and was named Achilles tendon/calcaneal tuberosity angle (ATCTA) (Fig. 16.1c). To adequately assess the alignment of the calcaneal tuberosity, we used similar technique drawing ellipses as described by Johnson et al. [13].

The last four measurements were performed in images with the same positioning, however with different windowing that removed all the soft tissue structures, maintaining only the bony anatomy. The third measurement obtained was the tibial axis/calcaneal tuberosity angle (TACTA), which was formed by the intersection of axes of calcaneal tuberosity and the tibia (Fig. 16.1d). The fourth measurement was the tibial axis/subtalar joint angle (TASJA) formed by the intersection of tibial axis and the line connecting midpoint of the posterior facet of the subtalar joint and most inferior point of the calcaneal tuberosity (Fig. 16.1e). The fifth measurement was the hindfoot alignment angle (HAA), as described by Williamson et al. [26] (Fig. 16.1f), and the sixth measurement was the hindfoot moment arm (HMA) (Fig. 16.1g), as described by Saltzman et al. [12].

Positive values were considered valgus alignment.

Statistical Analysis

Data analysis was performed with JMP Pro version 12.2.0 (SAS Institute, Marlow-Buckinghamshire, UK). We used Shapiro-Wilk W test to evaluate the normal distribution of each set of measurements. The intraobserver and interobserver reliability was assessed by intraclass correlation coefficient (ICCs). Correlations between 0.81 and 1 were regarded almost perfect, 0.61–0.8 were considered as considered as substantial, 0.41–0.6 were considered as moderate, 0.21–0.4 were regarded as fair, 0.1–0.2 were considered as slight agreement, and less than 0.1 were regarded as poor agreement [22, 27]. Measurements obtained from clinical examination and WB CBCT images were compared by one-way ANOVA and paired Student’s t-tests or the Wilcoxon rank-sum tests and nonparametric comparison for each pair by the Wilcoxon method. We used a linear regression model to evaluate the relation between values of hindfoot moment arm and measurements obtained from clinical examinations as well as WB CBCT images. P-values of less than .05 were considered significant.


Twenty patients (12 men and 8 women) with mean age of 52.2 (range, 20–88) years old and mean body mass index value of 30.35 (range, 19–46) kg/m2 were included in this cohort.

We observed almost perfect intraobserver agreement for all WB CBCT 3D measurements, with ICC ranging from 0.87 to 0.97. Interobserver agreement, measured by ICC, ranged from 0.51 to 0.88. A summary of the agreements is presented in Table 16.1.

Table 16.1

Intra- and interobserver reliability for three-dimensional (3D) weightbearing (WB) cone-beam computed tomography (CBCT) hindfoot valgus measures


Intraobserver agreement

Interobserver agreement





Soft tissue windowing level

WBCT clinical hindfoot alignment angle (WBCT CHAA)


Almost perfect



Achilles tendon/calcaneal tuberosity angle (ATCTA)


Almost perfect



Bone windowing level

Tibial axis/calcaneal tuberosity angle (TACTA)


Almost perfect



Tibial axis/subtalar joint angle (TASJA)


Almost perfect



Hindfoot alignment angle (HAA)


Almost perfect



Hindfoot moment arm (HMA)


Almost perfect


Almost perfect

Abbreviations: ICC intraclass correlation coefficient

A summary with the mean values and 95% confidence interval (CI) for all hindfoot valgus measures performed on the 3D WB CBCT images is outlined in Table 16.2, and a graphical plot demonstrating the mean values of each measurement is shown in Fig. 16.2.

Table 16.2

Summary of three-dimensional (3D) weightbearing (WB) cone beam computed tomography (CBCT) hindfoot valgus measures


3D WB CBCT hindfoot alignment

Mean value

Standard error of the mean

Lower 95% CI

Upper 95% CI

Soft tissue windowing level

WBCT clinical hindfoot alignment angle (WBCT CHAA)





Achilles tendon/calcaneal tuberosity angle (ATCTA)





Bone windowing level

Tibial axis/calcaneal tuberosity angle (TACTA)





Tibial axis/subtalar joint angle (TASJA)





Hindfoot alignment angle (HAA)





Hindfoot moment arm (HMA)

15.1 mm


13.4 mm

16.9 mm

Abbreviations: CI confidence interval

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Apr 25, 2020 | Posted by in MUSCULOSKELETAL MEDICINE | Comments Off on Alignment of Adult-Acquired Flatfoot Deformity: A Comparison of Clinical Assessment and Weight Bearing Cone Beam CT Examinations

Full access? Get Clinical Tree

Get Clinical Tree app for offline access