Quantifying shortening of the fractured clavicle assuming clavicular symmetry is unreliable

Abstract

Background: One of the more commonly used methods of determining the amount of shortening of the fractured clavicle is by comparing the length of the fractured side to the length of contralateral unfractured clavicle. A pre-existing natural asymmetry can make quantification of shortening using this method unreliable. The goal of this study is to assess the side-to-side variation in clavicle length in 100 uninjured, skeletally mature adults.

Materials and methods: To assess the side-to-side difference in clavicle length the length of both clavicles of 100 patients on thoracic computed tomography (CT) scans were measured. Patients without a history of pre-CT clavicular injury were included. The measurements were allocated into three groups based on the amount of asymmetry (< 5, ≥ 5-10 and > 10 mm). Dominant side and sex were analyzed to determine influence on the length of the clavicle.

Results: In 30 patients (30%), an asymmetry of 5 mm or more was found. 2% of the patients had a side-to-side difference of more than 10 mm. The absolute side-to-side length difference (LD) was 3.74 mm (95% CI 3.15-4.32; p < 0.001). A significant association between clavicle length and dominant side or sex was found (p < 0.001).

Conclusion: These results show that by utilizing a treatment algorithm based upon clavicular symmetry has a potential for error and can lead over- or under-treatment of the fractured clavicle. A significant association between clavicle length and dominant side or sex was found (p < 0.001).

Level of evidence: 2.

Keywords: Clavicle; Imaging; Interobserver agreement; Length; Symmetry.

Fig. 1

Example of measurements on 3D reconstruction of a CT scan showing both clavicles

Fig. 2

Scatterplot of right versus left clavicle lengths. Dots in the dark grey area represent length differences < 5 mm. Length differences between 5 and 10 mm are within the light grey area