CT scan method accurately assesses humeral head retroversion

Abstract

Humeral head retroversion is not well described with the literature controversial regarding accuracy of measurement methods and ranges of normal values. We therefore determined normal humeral head retroversion and assessed the measurement methods. We measured retroversion in 65 cadaveric humeri, including 52 paired specimens, using four methods: radiographic, computed tomography (CT) scan, computer-assisted, and direct methods. We also assessed the distance between the humeral head central axis and the bicipital groove. CT scan methods accurately measure humeral head retroversion, while radiographic methods do not. The retroversion with respect to the transepicondylar axis was 17.9 degrees and 21.5 degrees with respect to the trochlear tangent axis. The difference between the right and left humeri was 8.9 degrees. The distance between the central axis of the humeral head and the bicipital groove was 7.0 mm and was consistent between right and left humeri. Humeral head retroversion may be most accurately obtained using the patient's own anatomic landmarks or, if not, identifiable retroversion as measured by those landmarks on contralateral side or the bicipital groove.

Fig. 1

The radiographic method is illustrated. Line A-B marks the boundaries of the humeral head articular surface. Line C-D is perpendicular to A-B and defines the humeral head axis. Line E-F marks the elbow trochlear tangent axis. The humeral head retroversion angle (α) is shown.

Fig. 2A–B

The computed tomography scan method is shown. (A) Line A-B marks the boundaries of the humeral head articular surface using the limits of the subchondral bone (Lauman and Kramps [34]), whereas line C-D marks the anterior and posterior depressions corresponding to the insertion of the articular capsule (Bernageau [2]). These two lines are approximately parallel. (B) Line 1 defines the epicondylar axis. Line 2 defines the humeral head axis.

Fig. 3A–B

The computer-assisted method is shown. (A) The industrial coordinate measuring machine connected with a computer is used to determine humeral head retroversion. (B) The retroversion angle can be measured in relation to the transepicondylar axis A-B (digitized method 1) or in relation to the trochlear tangent axis C-D (digitized method 2).

Fig. 4A–B

The direct method is shown. (A) The special device constructed and used after osteotomizing the humeral head articular surface is illustrated. (B) The “tropometer” of Broca, which inspired our apparatus, is illustrated.

Fig. 5

The mean retroversion angle obtained with the computer-assisted method for the 65 humeri was 17.9° (95% confidence limit, 14.5–21.3) when measured with respect to the transepicondylar axis.

Fig. 6

The mean difference between the right and left humeri was 8.9 (95% confidence limit, 7.1–10.7; p = 0.0008). The equation of the regression line is y = 0.79x + 5.27, in which y is the right retroversion and x is the left retroversion for the same individual (r = 0.79, p = 0.0007).

Fig. 7

Contrary to humeral head retroversion, the distance between the central axis of the humeral head and the bicipital groove is essentially constant (average, 7.0 mm, 95% confidence limit, 6.1–7.9) and the difference between sides is less than 1 mm.