Category: Midfoot/Forefoot; Basic Sciences/Biologics Introduction/Purpose: Charcot Marie Tooth disease (CMT) is a progressive genetic neurological condition that, despite heterogeneity in genotype and expression, presents with characteristic cavovarus foot deformity consisting of a varus hindfoot with a high arch and valgus forefoot. Bony morphology differences have been identified in CMT but have not been studied between demyelinating and axonal subtypes. Existing work has predominantly utilized automated angle measures from weight-bearing computed tomography (WBCT), which has limitations for describing a complex 3-dimensional deformity. Statistical shape modeling (SSM) from WBCT images is a mathematical tool that can be used to quantify morphology and alignment in foot pathologies. The objective of this study is to use SSM to analyze differences in osseous morphology between demyelinating and axonal subtypes of CMT and controls. Methods: Retrospective chart review identified patients with CMT who had received WBCT scans before any foot or ankle surgery and healthy controls (IRB: 00154634). This yielded 11 WBCT scans in the demyelinating group, 11 scans in the axonal group, and 15 controls. Images were automatically segmented then manually edited and verified. 3-dimensional parts were aligned using an iterative closest point algorithm. A 14 bone SSM was created for the tibia through metatarsals and single-bone SSMs were created for each individual bone. Principal component analysis was used to calculate modes of variation and parallel analysis was used to identify statistically significant modes across the population. Along each mode of variation, PCA component scores between demyelinating, axonal, and control groups were tested for normality then compared with a t-test, Wilcoxon rank sum test, or ANOVA with a Holm-Sidak correction with a significance value of α=0.05. Results: Differences along at least one PCA mode were identified between axonal and demyelinating groups in the tibia, fibula, cuneiforms, and metatarsals (Figure 1). The tibia and fibula showed a shortened medial malleolus and alteration of the attachment sites of the deep transverse faschia and posterior tibio-fibular ligament in the demyelinating group as compared to axonal and control groups. In the cuneiforms, different patterns of variation were evident between subtypes primarily around the attachment sites of cuneo-navicular and cuneo-cuboid ligaments and the plantar Lisfranc ligament. In the first and second metatarsals, differences between subtypes are present in the morphology of the metatarsal heads and at the attachment of the peroneus longus tendon. Conclusion: There are differences in osseous morphology between subtypes of CMT. In the tibia and fibula, these support axonal CMT as a milder presentation that demyelinating CMT. However, different patterns of morphology variation between subtypes in the midfoot suggest a difference in bone adaptation between groups. Studying demyelinating and axonal groups is a simplification of the genetic variability in CMT, and additional work is necessary to characterize differences based on specific mutations. These findings are relevant to surgical planning as soft tissue surgeries may be less effective with bony deformity. Improved characterization of osseous morphology may improve surgical treatment of CMT.
Read full abstract