Ankle alignment on lateral radiographs. Part 2: reliability and validity of measures

Abstract

Background: In ankles with end-stage osteoarthritis or after total ankle replacement (TAR), radiographic landmarks based on joint surface morphology usually are obscured and inadequate for measurement. Two methods for quantifying anteroposterior tibial-talar alignment without relying on those landmarks were identified in a corollary cadaver-based study. This study aimed to verify reliability and validity of those candidate measures.

Methods: On clinical radiographs of 33 nonarthritic and 35 arthritic ankles, the anteroposterior tibial-talar alignment was quantified by the two methods; the tibial-axis-to-talus ratio (T-T ratio: the ratio into which the midlongitudinal axis of the tibial shaft divides the longitudinal talar length) and the posterior-tibial-line-to-talus ratio (P-T ratio: a similar ratio, but using the posterior longitudinal line along the tibial shaft). Two observers performed every measurement twice to evaluate intraobserver and interobserver reliability of the candidate measures. For nonarthritic ankles, the anteroposterior tibial-talar alignment was further determined by a control measure that directly quantified orientation of the talar dome relative to the tibial shaft. Correlation of the T-T and P-T ratios with the control measure was then evaluated for validity.

Results: Measurement of the T-T ratio with arthritic ankles was highly reproducible with the coefficients of determination (R(2)) greater than 0.95, for either interobserver or intraobserver. Correlation between this measure and the control measure was supported (R(2) = 0.60, p < 0.0001). Reliability of the P-T ratio also was strong (R(2) > 0.91), although both reliability and validity of this measure were relatively inferior to the T-T ratio.

Conclusions: The T-T ratio reliably and validly described the anteroposterior tibial-talar alignment on clinical radiographs, regardless of the condition of ankle joint surface. This measure appears to be a reliable radiographic measure for determining the magnitude of anteroposterior talar subluxation in ankles with articular degeneration or after TAR and can facilitate clinical investigations.

Fig. 1

Radiographic measurement of T-T ratio (a) and P-T ratio (b) on example OA-ankle radiographs. Posterior talar point (point A) is identified as the intersection between the contours of the posterior subtalar articular surface and the postero-superior calcaneal cortex. Talar reference line (TRL) is a line through point A parallel to the floor. Anterior talar point (point B) is the vertical projection of the most anterior aspect of the talus onto the TRL. Longitudinal talar length is measured as a length AB. a) The distal tibial axis (DTA) is a longitudinal mid-bisecting line of the distal tibial shaft determined at 5 and 10 cm above the ankle. The intersection of the DTA with the TRL is point C. The T-T ratio in this case is 55.3%. b) The posterior tibial line (PTL) is a longitudinal line along the posterior tibial shaft surface, determined at the same heights. The intersection of the PTL with the TRL is point D. The P-T ratio in this case is −18.7% (Note: this measure is negative when point D is posterior to point A).

Fig. 2

Histogram of AP tibial-talar alignment quantified by A) T-T ratio and B) P-T ratio, in thirty-three non-arthritic ankles (non-OA ankles) and thirty-two arthritic ankles (OA ankles). The height of each bar represents number of cases for which the measure is within the corresponding 5% interval range.

Fig. 3

Correlation with the control measure, with each candidate measure; A) the T-T ratio or B) the P-T ratio. On each graph, a plotted point indicates the output of the candidate measure (on the x-axis) and that of the control measure (on the y-axis), for a single non-OA ankle. The trend line by a linear regression analysis is represented as a black line, accompanied by the corresponding equation and by the coefficient of determination (R²).