The purpose of this study was to determine the landmarks important for evaluating torsional deviations of the humerus when minimally invasive surgical techniques are used for humeral fracture fixation. The results obtained showed that differences in the angle between the transepicondylar axis line and the connecting line at the vertices of the intertubercular sulcus ranged from 1.0° to 73.1°, with a mean angle of 41.1°. This study identified the profile of the intertubercular sulcus line of the humeral head and the axis of the distal humerus as an accurate tool for assessing the precision of torsion during MISs for humeral shaft fractures without fluoroscopic assistance. The line drawn through the vertices of the intertubercular sulcus line of the humeral head was externally rotated by approximately 41.1° compared with the axis of the distal humeral condyles.
In practice, surgeons may palpate the bony prominence of the intertubercular sulcus of the proximal humeral fragment as a landmark to evaluate the tangent line. Surgeons may rotate the distal humeral fragment, making the transepicondylar axis line internally rotated by 41.1° compared with the tangent line of the intertubercular sulcus. After the fixation is completed and stable, a clinical exam of internal rotation and external rotation of the shoulder joint can be performed to determine whether the forearm is oriented in the proper direction [22]. Thus, the fracture will undergo proper torsional reduction. Additionally, the technique may be effective for unstable comminuted or segmental humeral fractures (AO type 12C) if it is difficult to maintain the fracture in adequate alignment during the operation. A clinical trial should be conducted in the future as a final test. This technique needs to be demonstrated to prevent an increase the numbers of torsional malalignments and complications.
If a preoperative CT scan is available, the fracture can be virtually reduced using 3D system software [23]. The reconstructed model can be used to restore the physiological magnitude of humeral torsion and measure the alpha angle of the humerus.
In this study, the parameters of the humerus were measured using CT. In all 28 cases, the connecting line at the two vertices of the intertubercular sulcus was found to be externally rotated compared with the transepicondylar axis line.
Measurements were made on serial images of the humerus with 5 mm between each section. Errors may occur because the deepest sulcus and the longest axis of condyles of the sections taken for measurement may not be the actual deepest and longest ones.
Previous research studies considered a humeral malrotation of 15° in fracture alignment acceptable [24]. Although the standard deviation of the α angle obtained in this study was 17.1°, the mean angle may still be used as a reference for reduction. Statistically, an extreme outlier in the present data was identified. The standard deviation without the outlier was 15.5°, which is close to the abovementioned acceptable degree (15°) of humeral malrotation.
The proximal incision during MIPO is made with the deltopectoral approach [25]. The surgeon can directly touch the biceps sulcus as a landmark. The soft tissue around the elbow is thin in most cases. Epicondyles can be easily touched to identify the transepicondylar axis. The axis identified by the surgeon is not very different from the actual direction.
The palpable proximal and distal osseous landmarks (intertubercular sulcus, medial and lateral epicondyles) located by orthopaedic surgeons are slightly different from the imaging landmarks. For example, the biceps groove is located at the proximal humerus and becomes shallower toward the inferior end. The proximal landmark over the intertubercular sulcus in this study was obtained from the deepest site. Surgeons may not be able to locate the deepest point of the bicipital groove. However, extensive surgical experience and good judgement can increase the accuracy of identifying this groove. Highly trained and specialized orthopaedic surgeons can precisely locate the deepest groove and other anatomical landmarks.
However, the landmarks are not applicable in some situations, such as when the humeral head or the distal humerus is severely deformed due to acquired or congenital disorders.
In past research, a strong relationship has been shown to exist between humeral torsion variables obtained with ultrasound and CT [26]. If ultrasound equipment is available in the operating room, an ultrasonographic assessment of the humeral retroversion method can be used as a secondary confirmation [22]. In addition, when the patient’s soft tissue layer is thick and it is difficult to palpate the bony landmarks, ultrasonic positioning can be used to identify the transepicondylar axis.
Various techniques have been employed to measure torsional parameters of the humeral bone. Retroversion of the humeral head is most commonly used for defining the angular difference between the orientation of the proximal humeral head and the axis of the elbow at the distal humerus [19]. However, the results are highly variable, ranging in some case series from − 6° to 50° [27,28,29,30].
Nevertheless, obtaining the retroversion angle in operation requires fluoroscopic assistance, and it is difficult to confirm whether the proximal line is perpendicular to the articular surface.
A previous study used the bicipital groove of the humeral head to predict the torsional state of the humerus for intraoperative evaluations, but fluoroscopy was still needed with this approach [31].
To the best of our knowledge, no other study has used CT to measure humeral torsion with the tangent line of the intertubercular sulcus. A similar study measured humeral head retroversion with lateralization of the intertubercular groove using CT [32], which seems to be valuable for anatomical imaging but unsuitable for clinical orthopaedic surgeries.
This study also identified the correlations of the α angle with the humeral length and patient age. The correlation between the α angle and humeral length was statistically low, while that between the α angle and patient age was moderate. In the absence of a strong correlation with the α angle, age and humeral length need not be considered in clinical-surgical evaluations.
There were 13 right and 15 left humerus bones included in the research study. The mean alpha angle in the present study was 37.4 ± 18.6 degrees on the right side and 44.3 ± 15.7 degrees on the left side. The sample size is small, so the statistical significance of the results is unclear.
Previous studies have shown that the dominant arm of patients has a higher retroversion angle than the contralateral arm. On average, the degree of retroversion is 10.6 degrees larger in the dominant arm compared with the nondominant arm in overhead throwing athletes [33, 34]. Although the alpha angle is not an exact measure of humeral head retroversion, perhaps the angles on the bilateral humerus are different for people who have participated in throwing sports. However, whether the participants in our study practised throwing sports was not recorded. In a future study, we can add this factor to determine whether it has statistical significance.
Finally, only 28 extremities from 28 participants were included in the analysis. The volume of data in our imaging system limited our sample size. While more patients should be included in future prospective research studies, the costs and radiation exposure associated with CT scans should be taken into consideration when designing these studies.