[Computed tomographic analysis of humeral retrotorsion and glenoid retroversion]

Analysis of the anatomic relations of the humeral head and the glenoid cavity is particularly important for clinical study of shoulder arthroplasty and glenohumeral instability. Analysis of humeral retroversion and glenoid retroversion is quite difficult and data in the literature are scarse. We conducted a computed tomography (CT) analysis of stable shoulders to detail retroversion of the entire height of the glenoid cavity and to measure humeral retrotorsion using two comparative methods. We also compared glenoid retroversion and humeral retrotorsion observed in individual subjects.

This prospective study used a standardized CT analysis method. Both shoulders of 30 persons free of glenohumeral instability were studied. Two methods, described by Dähnert and Bernageau, were used to analyze humeral retrotorsion. The Benageau method was used to analyze glenoid retroversion.

According to the Dähnert method, humeral retrotorsion was 10 degrees +/- 13 degrees; it was 24 degrees +/- 13 degrees with the Bernageau method; data dispersion was 60 degrees and 65 degrees respectively. According to the Dähnert method, retrotorsion was more pronounced on the dominant side compared with the non-dominant side. There was a significant correlation between retrotorsion values for the two sides. For 95% of the shoulders, glenoid retorversion decreased progressively from the superior part of the glenoid cavity (12.8 degrees +/- 6.4 degrees ) to the lower part (3.1 degrees +/- 4.4 degrees ). Glenoid retroversion was greater on the dominant side. For 21 of the 30 persons (70%), there was a significant correlation between retroversion for the two sides. Correlation coefficients between glenoid retroversion and humeral retrotorsion were negative. Greater humeral retrotorsion was thus related with less pronounced glenoid retroversion and vice versa.

This study allowed quantification of glenoid retroversion and humeral retrotorsion. There is a spiral twist in the joint surface of the glenoid cavity with progressive decrease in glenoid retorversion from the upper to the lower part of the cavity for 95% of the shoulders. To our knowledge, this spiral twist in the glenoid cavity is not taken into consideration in any of the currently available implants. The correlation for both parameters between the right and left side is probably determined genetically. The influence of dominance could be explained by adaptation to more or less pronounced stress. The negative correlation between humeral retrotorsion and glenoid retroversion would improve glenohumeral stability. A comparative study with unstable shoulders would be required to verify this hypothesis. The validity of the Dähnert method for assessing humeral retorversion is, in our opinion, insufficiently established. The Bernageau method, which provides a direct measurement, appears to be preferable despite the difficulty in identifying anatomic landmarks.