Foot & Ankle International - June 2017 - 679

Wilson et al

Figure 1. T2* mapping scans of the peroneal tendons in the
(A) sagittal and (B) axial planes.

Figure 2. Subregional analysis of the peroneus brevis (gray)
and longus (black) tendons. A 6-cm-long section of each tendon,
centered on the inferior tip of lateral malleolus landmark, was
divided into (A) 1-cm regions of interest and (B) 2-cm regions of
interest for comparison.

for comparison. The 1-cm analysis included 3 subregions
proximal to the inferior tip of the lateral malleolus and 3
subregions distal (Figure 2A). The 2-cm analysis thus
included 1 subregion centered on the lateral malleolus, 1
located proximal and 1 distal (Figure 2B). T2* mapping
value means and standard deviations were calculated for
each tendon as a 6-cm-long section, and within each region
of interest (1- and 2-cm analyses), using both the sagittal
and axial planes.

Tendon Excursion Investigation
To determine the value of analyzing subregions around
the inferior tip of the lateral malleolus, an investigation of
7 asymptomatic volunteers was performed to measure the
excursion of the peroneus brevis tendon about the fulcrum of the lateral malleolus throughout the ankle range

679

Figure 3. As part of the tendon excursion investigation, the
distance between the inferior tip of the lateral malleolus and
tendon insertion on the fifth metatarsal was measured in each
ankle flexion position. The distance was measured by drawing a
straight line between the 2 landmarks on a central sagittal slice
of the tendon. Example slices are shown for a subject in (A)
dorsiflexion and (B) plantarflexion.

of motion. A clinical test of peroneal tendon excursion
has been described previously.31 Sagittal MR images were
acquired of the ankle in 3 positions stressed against a
resistance band: maximal dorsiflexion, neutral position,
and maximal plantarflexion with the volunteer in the
supine position. The length of tendon between the inferior
tip of the lateral malleolus and tendon insertion on the
fifth metatarsal was estimated in each ankle flexion position using imaging software (Mimics). Using a distance
measurement tool, a straight line was drawn between the
2 landmarks on a central sagittal slice of the tendon
(Figure 3).

Statistical Analysis
All statistical analyses were performed using a statistical
programming language (R, version 3.2.3; R Development
Core Team, Vienna, Austria). A repeated measures analysis
was conducted via random-intercepts mixed-effects models
(nlme: Linear and Nonlinear Mixed Effects Models; R
Development Core Team). An exponential spatial correlation structure was assumed for peroneal subregions, and
equal variance among subregions was not assumed. Tukey
post hoc comparisons were used to make all pairwise comparisons among subregions. Residual diagnostics were performed to confirm that the model fits well and that
assumptions were met. Significance was defined as P <.05.
Intra- and interrater measurement reliability was assessed
for mean T2* values. For each subregion, a 2-way random
effects model was used to calculate the single measures,
absolute agreement version of the intraclass correlation
coefficient (ICC). The ICC values were interpreted as follows: ICC <0.40 was defined as poor agreement; 0.40< ICC
<0.75 was defined as fair to good agreement; ICC >0.75
was defined as excellent agreement.10

Table of Contents for the Digital Edition of Foot & Ankle International - June 2017