Long Biceps Tendon: Normal Position, Shape, and Orientation in Its Groove in Neutral Position and External and Internal Rotation

Introduction

The tendon of the long head of the biceps brachii muscle—the long biceps tendon (LBT)—attaches at its anchor to the superior glenoid tubercle (1). From this position, it courses intraarticularly to the entrance of the bicipital groove between the major and lesser tubercles of the proximal humerus, runs caudally inside the groove, and eventually leaves the groove to end in the proximal musculotendinous junction (2,3). Along its intraarticular course, the LBT is suspended by the coracohumeral ligament, the superior glenohumeral ligament, and the pulley fibers, which arise from the aforementioned ligaments and the subscapularis tendon (4–10). Additionally, it may be stabilized by a variety of sporadic developmental interconnections with the joint capsule, ranging from pulleylike slings to complete adherence of the LBT to the joint capsule (11,12). This suspension, especially by the pulley fibers, is thought to keep the LBT in place and centered at the entrance to the bicipital groove (3–5,13–15).

Instability of the LBT can be observed arthroscopically in about 45% of patients with rotator cuff tears, and injury to the pulley fibers has an incidence of about 7% at diagnostic arthroscopy (16). These lesions are thought to be responsible for medial dislocation of the LBT and could possibly lead to a subset of partial LBT tears at the entrance to the bicipital groove (17–21). In addition to abnormalities in the ligamentous suspension of the LBT, there are abnormalities in the subscapularis tendon in most patients with LBT dislocation. Typically, partial tears at the superior margin of the subscapularis tendon are present (9,21).

An eccentric position of the LBT, especially at the level of the entrance to the bicipital groove, has been proposed as a diagnostic criterion for the detection of pulley lesions that may not be otherwise apparent (22–24). So far, eccentricity of the LBT at the entrance to the groove has not been measured but merely described qualitatively. Furthermore, the criteria for LBT instability, dislocation, and subluxation are not uniform, and these terms are used differently (16,25–27).

Thus, the purpose of this study was to characterize the position, shape, and orientation of the LBT on transverse magnetic resonance (MR) images acquired in the neutral position and in maximal external and internal rotation of the shoulder in asymptomatic volunteers.

Materials and Methods

The institutional review board of the University of Zurich approved this study, and informed consent was obtained from all volunteers.

Volunteers

Fifty-three asymptomatic volunteers (mean age, 33 years; age range, 21–58 years) were included in the study. There were 23 men (mean age, 34 years; age range, 21–56 years) and 30 women (mean age, 33 years; age range, 21–58 years). The volunteers had never experienced shoulder pain, could not recollect any shoulder trauma, had not undergone shoulder surgery of any kind, and reported no limitation in range of motion in the shoulder joint during daily activity. Twenty-nine right and 24 left shoulders were examined. Twenty-four shoulders were of the dominant side, and 29 were of the nondominant side. For the first volunteer, the shoulder to image was determined randomly. For the following volunteers, the shoulder to image was chosen so that we could obtain the same amount of right and left shoulders and the same amount of shoulders of the dominant and nondominant sides, if possible. Because of a preponderance of right-handed volunteers, however, this was not entirely possible.

MR Imaging

Two 1.5-T MR imaging systems (Avanto or Espree; Siemens Medical Solutions, Erlangen, Germany) were used with a dedicated four-element shoulder array coil.

The shoulder was imaged in neutral position with the arm flanking the body and the thumb pointing up. First, the standard protocol for shoulder examinations at our institution was performed. This protocol consisted of a coronal fat-saturated intermediate-weighted turbo spin-echo sequence (repetition time msec/echo time msec, 3000/13; field of view, 16 × 16 cm; matrix, 512 × 512 pixels; number of signals acquired, one; section thickness, 4 mm; and echo train length, seven), a coronal fat-saturated T2-weighted turbo spin-echo sequence (3130/72; field of view, 16 × 16 cm; matrix, 512 × 512 pixels; number of signals acquired, one; section thickness, 4 mm; and echo train length, 11), a sagittal non–fat-saturated T1-weighted turbo spin-echo sequence (525/12; field of view, 16 × 16 cm; matrix, 512 × 512 pixels; number of signals acquired, one; and section thickness, 4 mm), a sagittal short inversion time inversion-recovery sequence (5500/18; inversion time, 150 msec; field of view, 16 × 16 cm; matrix, 512 × 512 pixels; number of signals acquired, one; section thickness, 4 mm; and echo train length, seven), and an axial fat-suppressed steady-state gradient-echo (true fast imaging with steady-state precession) sequence (12.43/5.31; field of view, 18 × 18 cm; matrix, 512 × 512 pixels; number of signals acquired, one; and section thickness, 1.7 mm). Additionally, the last sequence (the axial steady-state gradient-echo sequence) was repeated with the shoulder in positions of maximal external and maximal internal rotation. External rotation of the shoulder was achieved by maximal external rotation of the shoulder joint and forearm, resulting in the palm of the hand facing upward; this position was stabilized by placing a heavy 2-kg bag filled with sand over the wrist. The whole arm was then externally rotated once again until a soft “stop” was felt; the arm was stabilized in this position. Maximal internal rotation was achieved by placing the volunteer’s hand palm-backward on the back of his or her lumbar spine with the elbow flexed 90°.

Measurements and Evaluations on MR Images

All MR images were evaluated by a musculoskeletal radiologist (F.M.B., with 5 years of experience in musculoskeletal radiology) before the measurements. Volunteers with lesions or any kind of degenerative changes to the rotator cuff tendons, coracohumeral ligament, superior glenohumeral ligament, and/or pulley fibers would have been excluded from the study. However, no volunteers had to be excluded.

The LBT was evaluated by two independent musculoskeletal radiologists (F.M.B. [reader 1] and T.J.D. [reader 2, with 2 years of experience in musculoskeletal radiology]) at three levels of the bicipital groove (superior, middle, and inferior) with the shoulder in neutral position, in external rotation, and in internal rotation. The superior level, at the entrance of the LBT into the bicipital groove, was defined as being shown on the most superior image that included the subscapularis tendon; the inferior level, where the LBT exited from the bicipital groove, was defined as being shown on the most inferior image that included the subscapularis tendon; and the middle level represented the midway point between those levels (Fig 1). All three levels and all three positions were evaluated at the same time in a standardized order: first, on the images obtained in neutral position, then, on the images obtained in maximal external rotation, and finally, on the images obtained in maximal internal rotation.

Figure 1a:

Download as PowerPointOpen in Image Viewer

Figure 1b:

Download as PowerPointOpen in Image Viewer

Figure 1c:

Download as PowerPointOpen in Image Viewer

Figure 1d:

Download as PowerPointOpen in Image Viewer

Position.—The position of the center of the LBT was evaluated with respect to the bicipital groove. The center of the LBT was defined as the point in the middle of the LBT (Fig 2). In flat or comma-shaped tendons, the center of the LBT was defined as the point in the middle of the LBT on a line drawn perpendicular to the greatest tendon diameter (Fig 1). The distance from the center of the LBT to a reference line connecting the anterior contours of the greater and lesser tubercles was measured (Fig 2). This measurement quantified the LBT position in the anteroposterior direction; a positive measurement indicated that the LBT was located outside the bicipital groove, and a negative measurement indicated how deep the LBT was located in the bicipital groove.

Figure 2a:

Download as PowerPointOpen in Image Viewer

Figure 2b:

Download as PowerPointOpen in Image Viewer

So that we could evaluate the position of the LBT in the mediolateral direction, a perpendicular line to the above-mentioned reference line (Fig 2) was drawn through the deepest point of the bicipital groove. The distance of the center of the LBT to this line was determined (Fig 2); a positive measurement indicated a position of the LBT medial to the deepest point of the bicipital groove, and a negative measurement indicated a position of the LBT lateral to the deepest point of the bicipital groove.

Any changes in the position of the LBT in the neutral position and in external and internal rotation of the shoulder were then determined at the three measurement levels.

Shape.—The cross-sectional shape of the LBT was categorized as round, oval, flat, or comma shaped (Fig 3). For a round LBT, the cross-sectional area was a circle with equal perpendicular diameters. For an oval LBT, the dimension of the smallest cross-sectional tendon diameter was greater than (but not equal to) half of the greatest tendon diameter. For a flat LBT, the smallest cross-sectional tendon diameter was smaller than half of the greatest diameter, and the lateral and medial contours of the tendon were roundish and obtusely angled. For a comma-shaped LBT, the same criteria applied as for a flat LBT, but the medial contour of the tendon was pointed and acutely angled.

Figure 3a:

Download as PowerPointOpen in Image Viewer

Figure 3b:

Download as PowerPointOpen in Image Viewer

Figure 3c:

Download as PowerPointOpen in Image Viewer

Figure 3d:

Download as PowerPointOpen in Image Viewer

Orientation.—The angle of the direction of the greatest LBT diameter with respect to the previously described reference line drawn between the anterior contour of the major and lesser tubercles was defined as the LBT orientation (Figs 4, 5). The orientation of the LBT along its course in the bicipital groove, as well as any changes in its orientation during external and internal rotation of the shoulder, were then described at the three measurement levels.

Figure 4a:

Download as PowerPointOpen in Image Viewer

Figure 4b:

Download as PowerPointOpen in Image Viewer

Figure 5a:

Download as PowerPointOpen in Image Viewer

Figure 5b:

Download as PowerPointOpen in Image Viewer

Figure 5c:

Download as PowerPointOpen in Image Viewer

Figure 5d:

Download as PowerPointOpen in Image Viewer

Figure 5e:

Download as PowerPointOpen in Image Viewer

Figure 5f:

Download as PowerPointOpen in Image Viewer

Statistical Analysis

Because LBT position and orientation are continuous variables and LBT shape is categoric, interreader agreement was quantified by calculating the intraclass correlation coefficient (ICC) for the evaluation of the biceps tendon position and orientation, and the weighted κ statistic was used to determine interreader agreement of the evaluation of the biceps tendon shape. The Wilcoxon signed-rank test was used to find significant differences in LBT position, shape, and orientation at the different measurement levels and in the different positions of the shoulder. Statistical software (PASW, version 18; SPSS, Chicago, Ill) was used. P < .05 was considered to indicate a significant difference.

Results

Position

The position of the LBT changed significantly in both directions (anteroposterior and mediolateral) from the neutral position to external rotation and from the neutral position to internal rotation at all measurement levels, except for some measurements in the anteroposterior direction on the middle and inferior levels. Detailed measurement and statistical results are provided in Table 1.

Table 1 Position and Orientation of LBT in Different Rotational Positions of the Humerus

Note.—Data are means ± standard deviations, with ranges in parentheses. P values were calculated with the Wilcoxon signed-rank test.

However, the changes in LBT position in external rotation and internal rotation compared with its position at neutral were markedly small (Fig 6). On the superior and middle measurement levels, these differences were all less than 1 mm for both readers; on the inferior measurement level, they were less than 1.5 mm.

Figure 6a:

Download as PowerPointOpen in Image Viewer

Figure 6b:

Download as PowerPointOpen in Image Viewer

Figure 6c:

Download as PowerPointOpen in Image Viewer

Figure 6d:

Download as PowerPointOpen in Image Viewer

Figure 6e:

Download as PowerPointOpen in Image Viewer

Figure 6f:

Download as PowerPointOpen in Image Viewer

Figure 6g:

Download as PowerPointOpen in Image Viewer

Figure 6h:

Download as PowerPointOpen in Image Viewer

Figure 6i:

Download as PowerPointOpen in Image Viewer

In neutral position of the shoulder joint, the averaged (between readers 1 and 2) position of the LBT was 2 mm medial to the deepest point of the bicipital groove at the superior level, 1.7 mm medial to the deepest point of the bicipital groove at the middle level, and 1.4 mm medial to the deepest point of the bicipital groove at the inferior level.

Interreader agreement was substantial (ICC > 0.6) to almost perfect (ICC > 0.8) and highly significant (P < .005), except for those measurements at the superior level in internal rotation and at the inferior level in the neutral position in the anteroposterior direction. Detailed results are provided in Table 2.

Table 2 Interreader Agreement for Position and Orientation of LBT

Surprisingly, LBT eccentricity in the mediolateral direction was most pronounced in the neutral position and was less pronounced in external rotation and internal rotation at all measurement levels.

Shape

At the superior level, the LBT was flat to comma shaped in almost all volunteers independent of the shoulder position. At the middle level, the LBT was flat to comma shaped, whereas in external rotation, the tendon had an oval shape at times. At the inferior level, the LBT was predominantly oval to flat, except in internal rotation, when there was a tendency for it to be round or oval in shape.

In all joint positions (neutral, external rotation, and internal rotation), there was a significant difference in LBT shape between the superior and middle levels (Table 3). For readers 1 and 2, respectively, for the neutral position, P = .002 and .002; for external rotation, P = .003 and < .001; and for internal rotation, P = .010 and .003. The same applied to the difference in LBT shape between the middle and inferior levels. For readers 1 and 2, respectively, for the neutral position, P = .001 and .009; for external rotation, P = .001 and .002; and for internal rotation, P < .001 and < .001. More results of LBT shape classification are provided in Table 3.

Table 3 Shape of 52 LBTs in Different Rotational Positions of the Humerus

Note.—Data are numbers of tendons, with percentages in parentheses and P values (calculated with the Wilcoxon signed-rank test) in brackets. Change = numbers of tendons that changed their shape between the position to the left and the position to the right (external rotation vs neutral, neutral vs internal rotation).

Interreader agreement was moderate (weighted κ > 0.5) to substantial (weighted κ > 0.6). Specifically, weighted κ values were as follows: 0.72 for external rotation, 0.72 for the neutral position, and 0.71 for internal rotation at the superior level; 0.81 for external rotation, 0.73 for the neutral position, and 0.62 for internal rotation at the middle level; and 0.68 for external rotation, 0.55 for the neutral position, and 0.59 for internal rotation at the inferior level.

Orientation

In all shoulder joint positions, the differences in LBT orientation between the measurements at the superior and middle levels and between the middle and inferior levels were both highly significant (P < .001), except for the comparison of measurements in external rotation between the superior and middle levels (reader 1: P = .192; reader 2: P = .167) and for the comparison of measurements in internal rotation between the middle and inferior levels (reader 1: P = .206; reader 2: P = .920).

Compared with the orientation of the LBT in neutral position, the orientation of the LBT changed significantly (P < .001) in external and internal rotation of the shoulder at the middle and inferior levels but not at the superior measurement level (Table 1).

Results concerning the interreader agreement of the LBT orientation measurements are provided in Table 4. Interreader agreement was mostly statistically significant (P < .05) and was substantial to almost perfect. Except at the superior level, there was no significant interreader agreement in internal rotation.

Table 4 Interreader Agreement for Evaluation of LBT Orientation

Discussion

Our data describe the physiologic position, shape, and orientation of the LBT in the neutral position, in external rotation, and in internal rotation of the shoulder.

In our volunteers, the position of the LBT in the bicipital groove was slightly dependent on the rotation of the shoulder. The changes in LBT position in external and internal rotation compared with the neutral position were less than 1.5 mm at all measurements levels. Surprisingly, medial eccentricity of the LBT was greatest in the neutral position of the shoulder joint at all measurement levels. Therefore, rotational misplacement in the shoulder joint at the time of image acquisition does not increase LBT eccentricity to the medial side and should not lead to erroneously diagnosed LBT subluxation or dislocation.

Superiorly, the LBT was situated eccentrically in the mediolateral direction with respect to the bicipital groove, irrespective of the shoulder joint position. With regard to the normal LBT position in the mediolateral direction at the entrance of the bicipital groove, physiologic values were found to be −1.0 to 7.5 mm medial to the deepest point of the bicipital groove. Because of this wide variability, a far medial LBT position of 8 mm medial to the deepest point of the bicipital groove would be required if the diagnosis of LBT subluxation was being considered. However, because our data are based on measurements in healthy volunteers, we cannot provide criteria to diagnose LBT instability but emphasize the variability of the LBT position in physiologic conditions.

The shape and orientation of the LBT changed depending on the level of evaluation and the position of the shoulder. Superiorly, the tendon was almost exclusively flat or comma shaped. In the middle of the bicipital groove, the tendon was generally flat to comma shaped, with a tendency to be more oval in external rotation. Inferiorly, the tendon was generally oval to flat, with a tendency to be more rounded in internal rotation.

Interreader agreement of the LBT orientation was considerably weaker at the inferior evaluation level. This may relate to a preponderance of round to oval tendon shapes and, hence, greater difficulty in the precise identification of the greatest tendon diameter at this level. Also, interreader agreement of the evaluation of images acquired in internal rotation was poorer compared with the interreader agreement of measurements on images acquired in the neutral position or in external rotation. We believe that this may relate to the crowding of the subscapularis tendon, LBT, and joint capsule between the humeral head, coracoid process, and anterior rim of the glenoid in maximal internal rotation. As a consequence, these structures were much more difficult to identify. Despite the fact that there is a change of shape and orientation of the LBT depending on the level of evaluation and the position of the shoulder, we have no evidence of any direct correlation between the shape of the LBT and its position with respect to the intertubercular groove.

Lafosse et al (16) observed LBT instability, defined as subluxation or dislocation of the LBT, in 45% of 200 consecutive patients undergoing arthroscopic rotator cuff repair. In these patients, LBT instability was anterior in 16%, posterior in 19%, and both in 10%. However, frank dislocation was observed only in the anterior direction. To the best of our knowledge, there have been no criteria published that allow for the diagnosis of posterior LBT instability on MR images. Our results describe the physiologic variation of LBT position in asymptomatic volunteers. It remains unanswered if there is a reasonable threshold that enables the identification of a pathologically eccentric LBT.

Spritzer et al (28) examined a small group of nine patients with arthroscopically proven LBT instability (average age, 42 years) and compared MR imaging findings with those in a small control group of 10 patients (average age, 47 years) with shoulder disease that was different from LBT instability. The authors concluded that “in the clinical setting of pain or symptoms referable to the biceps, the presence of a flat, degenerated biceps tendon perched on the lesser tubercle with an obtusely angled bicipital groove should raise the suspicion of instability of the long head of the biceps tendon.” Our data, although partly in agreement with these findings, indicate that a flat LBT perched on the lesser tubercle can also be seen in the neutral position or with external rotation of the shoulder in asymptomatic persons, suggesting that this position and shape of the LBT can be physiologic (Fig 7).

Figure 7a:

Download as PowerPointOpen in Image Viewer

Figure 7b:

Download as PowerPointOpen in Image Viewer

Surprisingly, the LBT was more centered in external rotation of the shoulder joint. The data acquired for this study do not allow for an explanation of this finding. It seems that an uninjured pulley system and rotator cuff tendons center the LBT better in external rotation of the shoulder joint than in the neutral position.

There were some limitations of this study. The position, shape, and orientation of the LBT in asymptomatic volunteers with no degenerative changes on MR images are described. Despite the evaluation of the suspension of the LBT to the best of our knowledge, it is still possible that a volunteer had minor degeneration of these structures. Because there was no control group consisting of patients with proven LBT instability, it was not possible to identify reliable criteria for LBT instability.

In conclusion, the position of the LBT is only slightly dependent on shoulder rotation. LBT eccentricity is maximal in the neutral position. Rotational misplacement during image acquisition does not increase LBT eccentricity and should not lead to erroneously diagnosed LBT subluxation or dislocation.

Disclosures of Potential Conflicts of Interest: F.M.B. No potential conflicts of interest to disclose. T.J.D. No potential conflicts of interest to disclose. D.R. No potential conflicts of interest to disclose. B.J. Financial activities related to the present article: none to disclose. Financial activities not related to the present article: is a consultant with Zimmer and with Karl Storz Endoscopy. Other relationships: none to disclose. C.W.A.P. No potential conflicts of interest to disclose.

Author Contributions

Author contributions: Guarantor of integrity of entire study, F.M.B.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, F.M.B., T.J.D., C.W.A.P.; clinical studies, F.M.B., T.J.D., B.J., C.W.A.P.; statistical analysis, F.M.B., C.W.A.P.; and manuscript editing, all authors