Rotational analysis of varus-type proximal humerus fractures: implications for fracture reduction.

Abstract

Background: Varus-displaced proximal humerus fractures can be treated with closed reduction with or without internal fixation. However, the rationale behind the subsequent maneuvers of the closed reduction has not been evaluated. The aim of this study was to analyze the rotational displacement of the humeral diaphysis compared to the humeral head in 3 dimensions for varus-displaced proximal humerus fractures.

Methods: One hundred fifty-one 3D reconstructions of computed tomography scans of proximal humerus fractures were obtained in a retrospective cohort set-up. An axis system was defined in the humeral head as well as the diaphysis using the Mimics and 3-Matic software (Materialise, 21.0 and 13.0 respectively, Leuven, Belgium). Rotational analysis around the 3 anatomical axes (valgus-varus, internal-external rotation, and flexion-extension) of these 2 axis systems was performed using MATLAB software (MathWorks, R2021a, Natick, MA, USA). These were compared to normal values obtained on 87 3D reconstructed CT images of intact shoulders and divided into varus and valgus type fractures (one standard deviation below and above the average of intact humeri, respectively), yielding 81 varus fractures. Statistical analysis was done using the Mann-Whitney U Test. After performing normality analysis, these rotations were correlated using the spearman's correlation coefficient.

Results: Eighty-one varus-displaced proximal humerus fractures were analyzed. This showed that these fractures had a significantly more retroflexed diaphysis compared to intact humeri, and this for 2-part, 3-part and 4-part fractures (P < .001, P < .001 and P = .007, respectively). Three-part fractures showed a significantly more externally rotated diaphysis compared to intact humeri (P < .001). Furthermore, 3-part fractures showed a statistically significant positive correlation between displacement of the greater tuberosity and relative external rotation of the humeral diaphysis.

Conclusions: Understanding fracture patterns allows for better treatment and reduction strategies. The findings of this study show that to obtain a good closed reduction of varus type proximal humerus fractures, the practitioner has to reduce any translation and the varus displacement of the humeral shaft, but should also take the relative retroflexion of the humeral shaft into account. For 3-part fractures, one should also reduce the external rotation of the diaphysis compared to the head.