Talar Center of Migration Ratio (TCM-R) as a Reliable Ratio-Based Method for Measuring Talar Translation on Noncalibrated Radiographs: A Short Report

Abstract

Background:

Coronal talar translation is an important marker of deformity in ankle arthritis, and its correction is vital in total ankle replacement (TAR). The current method for measuring this deformity is talar center of migration (TCM). We developed a novel method, TCM-Ratio (TCM-R), to measure talar translation accurately across all talus sizes using noncalibrated radiographs. Furthermore, we aimed to describe correlations between TCM-R and concomitant procedures during TAR.

Methods:

Patients were collected from a single-institution prospective TAR registry from June 2023 to May 2024. Inclusion criteria were preoperative calibrated weight-bearing radiographs. For each patient, 4 raters measured the TCM (distance from center of talus to mechanical axis of tibia) and the TCM-R (talar displacement divided by talar radius). Intraclass correlation coefficients (ICCs) were used to assess inter-rater agreement. Pearson correlations were used to assess agreement between TCM-R and TCM.

Results:

Fifty-eight patients were included, 60.3% male, with mean age 64.8 ± 9.4 years, and mean BMI 30.0 ± 4.7. Inter-rater agreement was excellent for TCM (ICC = 0.914) and TCM-R (ICC = 0.944). TCM-R was highly correlated with TCM (R = 0.95, CI 0.91-0.97, P < .001). The mean absolute value of TCM-R was 13.89% ± 11.29% (range, −34.57% to +47.53%).

Conclusion:

We described a new ratio-based method, TCM-R, to measure coronal talar translation, obviating the need for calibrated radiographs while accounting for variations in talar size. TCM-R had excellent inter-rater reliability and strongly correlated with TCM.

Level of Evidence:

Level IV, case series.

Introduction

Total ankle replacement (TAR) is an increasingly common procedure used to relieve symptoms of ankle osteoarthritis. ¹ In addition to replacing the joint, one goal of TAR is to correct deformity associated with ankle osteoarthritis. One parameter of deformity is coronal talar translation, wherein the talus shifts in the medial-lateral plane underneath the tibial axis. Restoration of talus position in alignment with the tibia is considered vital to a successful TAR. ²

The current method for measuring talar translation is the talar center of migration (TCM), the distance from the center of the talus to the tibial axis.^3-5 Previous studies demonstrated the importance of talar translation in TAR, supporting a need for intraoperative procedures to balance the ankle in the coronal plane in severe deformity. ⁵ Although TCM provides valuable insight on coronal alignment and preoperative planning, limitations exist. Because an absolute measurement is used, calibrated radiographs are required. Additionally, TCM does not contextualize deformity with talar size.

To address these limitations, we developed a new method to measure talar translation, the talar center of migration ratio (TCM-R), and evaluated reliability and agreement between TCM-R and TCM. Next, we sought to describe correlations between TCM-R and frequency of intraoperative procedures during TAR that may be performed to correct this deformity. We hypothesized that TCM-R would have high reliability and that patients with more medial talar translation and greater TCM-R values would undergo procedures used to balance the ankle intraoperatively—tendo-Achilles lengthening (TAL), lateral ligament reconstruction, and/or first ray dorsiflexion osteotomy—at greater frequencies.

Methods

Patients

Of 228 patients from a single-institution TAR registry between June 2023 and May 2024, 64 had calibrated preoperative weight-bearing radiographs. Calibrated radiographs were used to appreciate the relationship between TCM, which requires measurement on calibrated radiographs, and TCM-R. Six patients were excluded because of hardware that interfered with measurements or lack of anterior-posterior (AP) radiographs. This patient registry collects demographic information, concomitant procedures performed during TAR, and radiographic data.

Radiographic Analysis

For each patient, 4 raters measured talar translation using TCM and TCM-R as shown in Figure 1.

Figure 1.

Representative measurement of talar translation using (A) TCM (7.7 mm) and (B) TCM-R on 1 patient (42.58%). TCM is the shortest distance between the center of the talus, defined along the bimalleolar axis, and the mechanical axis of the tibia. In the TCM-R method, the center of the talus is defined as the center of a best-fit circle drawn over the talus. The shortest distance from this point to the mechanical of the tibia is measured, and the TCM-R is a ratio of this distance divided by the radius of the best-fit circle. TCM, talar center of migration; TCM-R, talar center of migration ratio.

TCM defines the center of the talus along the bimalleolar axis and measures the distance from this point to the mechanical axis of the tibia. The tibial axis was determined by centering 2 circles in the tibial diaphysis, 50 mm and 100 mm above the tibial plafond. The mechanical axis of the tibia intersected the center of each circle, extending below the talus.

To measure TCM-R, a circle was fitted to the overall talar articular contour, prioritizing medial and lateral talar articular surfaces while incorporating the talar dome as much as possible. When it was not possible to achieve a single circle that fit the medial and lateral articular surfaces and the talar dome because of complex 3-dimensional talar geometry, the circle was optimized to best represent global talar contour. The distance from the center of this best-fit circle to the mechanical axis of the tibia was then measured. TCM-R was calculated by dividing this distance by the best-fit circle’s radius.

Both measurements were recorded using calibrated preoperative weight-bearing AP radiographs. Positive values indicated medial translation; negative values indicated lateral translation. Raters were blinded to surgery outcome, patient information, and each other’s measurements.

Statistical Analysis

Descriptive statistics characterized patient demographic data. Mean, SD, and range were reported for continuous variables. Frequency and percentage were reported for discrete variables. Given limited radiographs, 4 raters reviewed radiographs to maximize statistical calculations. An intraclass correlation coefficient (ICC) assessed interrater agreement. An ICC greater than 0.90 indicates excellent reliability. To compare TCM-R with TCM, Pearson correlations were applied.

We used t tests to compare TCM-R and TCM with the frequency of certain intraoperative procedures during TAR. An alpha of .05 was used to establish statistical significance. Analysis was performed by a biostatistician using R: A Language and Environment for Statistical Computing, version 4.1.0. ⁶

Results

Fifty-eight patients, 60.3% male, with mean age 64.8 ± 9.4 years and mean BMI 30.0 ± 4.7, were included in analysis. Additional demographic information is shown in Table 1.

Table 1.

Patient Demographics for the Cohort of Primary TAR Patients With Calibrated Radiographs Included in This Study (N = 58). ^a

Characteristic	Patients
Age, y, mean ± SD	64.8 ± 9.4
Gender, n (%)
Male	35 (60.3)
Female	23 (39.7)
BMI, mean ± SD	30.0 ± 4.7
History of hindfoot fusion, n (%)	3 (5.2)
Ankle arthritis etiology, n (%)
Post-traumatic	19 (32.8)
Instability	27 (46.6)
Primary arthritis	3 (5.2)
Inflammatory arthritis	3 (5.2)
Progressive collapsing foot deformity	5 (8.6)
Adjacent joint arthritis after subtalar fusion	1 (1.7)
Concomitant procedures at the time of TAR, n (%)
Tendo-Achilles lengthening	8 (13.8)
Gastrocnemius recession	7 (12.1)
Medial malleolus prophylactic fixation	56 (96.6)
Medializing calcaneal osteotomy	5 (8.6)
Lateral ligament reconstruction	16 (27.6)
Cotton medial cuneiform osteotomy	4 (6.9)
First ray dorsiflexion osteotomy	1 (1.7)
First tarsal-metatarsal fusion	3 (5.2)
Subtalar fusion	1 (1.7)
Removal of hardware	6 (10.3)
Lateralizing calcaneal osteotomy	1 (1.7)
Prophylactic fibular fixation	1 (1.7)

Abbreviations: SD, standard deviation; TAR, total ankle replacement.

a

Categorical variables are expressed as frequencies with percentages. Continuous variables are expressed as means ± SD.

Agreement between 4 raters was excellent for TCM and TCM-R, with ICCs of 0.914 and 0.944, respectively. As demonstrated in Figure 2, TCM-R was highly correlated with TCM (R = 0.95, CI 0.91-0.97, P < .001).

Figure 2.

TCM-R as a function of TCM, demonstrating a high degree of correlation across the range of values. TCM, talar center of migration; TCM-R, talar center of migration ratio.

The mean absolute value of TCM was 2.71 ± 2.11 mm (range, −8.625 to 8.625), indicating that the talus was, on average, translated nearly 3 mm medially or laterally beneath the tibia. The mean absolute value of TCM-R was 13.89% ± 11.29% (range, −34.57% to +47.53%), indicating that the talus was, on average, 14% medially or laterally translated underneath the tibia. Medial translation was present in 25 of 58 patients (43.1%).

During TAR, 8 patients (14%) underwent TAL, 16 patients (28%) underwent lateral ligament stabilization, and 1 patient (2%) underwent first ray dorsiflexion osteotomy. With the numbers available, no significant difference in the mean absolute values of TCM and TCM-R could be detected between patients who did and did not undergo TAL. Patients who received lateral ligament stabilization had a similar average TCM magnitude, but a slightly higher average TCM-R magnitude (P = .127), compared to patients who did not.

Discussion

We developed a new method, TCM-R, to measure talar translation, a deformity observed in ankle osteoarthritis where the talus is shifted medially or laterally underneath the tibia. TCM-R uses a ratio to account for talar translation relative to talar size without requiring calibrated radiographs. Our first aim was to evaluate TCM-R reliability and assess agreement with TCM, the current method of quantifying talar translation. We also sought to identify the relationship between TCM-R and likelihood of certain intraoperative procedures during TAR.

We observed strong correlation between TCM-R and TCM. High inter-rater reliability demonstrated consistent measurements across different raters. Because TCM-R uses a ratio of the talar displacement relative to the radius of the talus while TCM is the displacement itself, we expected that the same distance of talar displacement would represent more significant deformity in a smaller talus than a larger one. However, we did not observe significant differences in TCM and TCM-R at extremes of talus size; both methods remained highly correlated across all talus sizes. Still, the ratio-based method used in TCM-R provides a method to measure talar translation without calibrated radiographs.

Similarly, with the numbers available, no significant differences in TCM-R could be detected for patients who did and did not receive TAL, lateral ligament stabilization, or first ray dorsiflexion osteotomy concomitant with TAR. Anecdotally, these procedures are considered intraoperatively in patients with severe medial translation talar deformities. Although not statistically significant, a slight difference was observed in TCM-R for patients who did and did not receive lateral ligament stabilization concomitant with TAR. Patients with medial talar translation who underwent this procedure had increased coronal translation deformity compared to those who did not. This relationship may indicate that medial translation represents a form of ankle arthritis secondary to chronic instability, specifically insufficiency of the lateral ligaments, hence the increased frequency in procedures to stabilize these ligaments. ⁷ Future investigation may explore this relationship with larger patient cohorts.

This study has several limitations—notably, small sample size because of the limited number of calibrated radiographs. Expanding the sample size may characterize the range and distribution of TCM-R values more precisely while also providing power to clarify the relationship between TCM-R and intraoperative procedures. ⁸ Additi-onally, full-length lower-leg radiographs were not available for this cohort. Although the distal tibial segment reflects standard practice in studies relying on ankle-level imaging, full-length lower-leg radiographs may better characterize the tibial mechanical axis.

Future directions may expand on this reliability and agreement study to evaluate TCM-R validity. Distinct analyses would be needed to validate TCM-R with a 3-dimensional measurement (criterion validity), demonstrate clinically distinct deformity groups based on TCM-R (construct validity), and reveal correlations between preoperative TCM-R with postoperative outcomes (predictive validity).

Conclusions

We described a new ratio-based method, TCM-R, to measure talar translation on non-calibrated radiographs while also contextualizing talar translation with talus size. TCM-R had excellent inter-rater reliability and strong correlation with TCM. With the numbers available, a statistically nonsignificant trend was observed in which patients who underwent lateral ligament stabilization concomitant with TAR had slightly higher medial talar translation compared to patients who did not, inviting further investigation with increased sample size. Future analysis may evaluate validity of TCM-R, going beyond reliability and agreement demonstrated here.

Supplemental Material

sj-pdf-1-fao-10.1177_24730114261451259 – Supplemental material for Talar Center of Migration Ratio (TCM-R) as a Reliable Ratio-Based Method for Measuring Talar Translation on Noncalibrated Radiographs: A Short Report

Supplemental material, sj-pdf-1-fao-10.1177_24730114261451259 for Talar Center of Migration Ratio (TCM-R) as a Reliable Ratio-Based Method for Measuring Talar Translation on Noncalibrated Radiographs: A Short Report by Kira Lu, Ricardo Ummen de Almeida Tenório Villar, Samantha Morton, Ranqing Lan, Constantine Demetracopoulos and Jensen Henry in Foot & Ankle Orthopaedics