Keywords
biceps - rotator cuff - tenodesis - tenotomy
Introduction
When the long head of the biceps tendon (LHBT) is ruptured or subjected to tenotomy
or tenodesis, retraction of the muscle belly may result in changes in the strength
and/or aesthetics of the arm.[1] However, the amount of retraction that is able of causing clinical repercussions
is still not well-defined. Muscle atrophy, interstitial fibrosis and fatty infiltration
(FI) are frequently associated with chronic rotator cuff injuries. These irreversible
structural changes favor the loss of muscle strength and elasticity and are major
obstacles to the success of surgical repair of muscles such as the subscapularis,
supraspinatus and infraspinatus muscles.[2] Although often studied, it is not well-established whether chronic shortening of
the biceps muscle mass can cause a change in the muscle belly similar to that caused
by FI in the belly of the rotator cuff muscles in cases of chronic and retracted ruptures.[3]
[4]
In the mid-1900s, the LHBT was seen as the main source of shoulder pain, and treatment
of that pain focused on tenotomy and tenodesis.[5]
[6]
[7] Currently, these procedures for the treatment of LHBT-related injuries are widely
performed during shoulder arthroscopies.[8]
[9] There is no clear consensus regarding the best surgical treatment for LHBT injuries.[10] There is also no consistent method to ensure that appropriate biceps tension has
been restored postoperatively.[11]
Our objective was to determine whether FI is present in the muscle of the biceps brachii
after the treatment of LHBT injuries (tenodesis or tenotomy) and to establish a relationship
between this possible FI with changes in muscle fiber length, the presence of deformities
and strength.
Materials and Methods
This was an observational, cross-sectional, case-control study consisting of the clinical
and imaging analysis of two groups: one composed of patients undergoing biceps tenodesis
in the bicipital groove with an interference screw and another composed of patients
undergoing biceps tenotomy. Patients were randomly separated according to sealed envelopes
that were opened before each surgery. In both groups, a comparison was made with the
findings on the contralateral side of each patient (control group).
The present study was submitted to the ethics committee under the number CAAE 92179218.8.0000.5505,
and all patients included in the study signed an informed consent form.
Patient Selection
The patients were selected based on a simple search of the medical records in the
private clinic of the investigator. The inclusion criteria were: agreeing and signing
the informed consent form, having undergone unilateral biceps tenotomy or tenodesis,
and postoperative follow-up of > 1 year.
The exclusion criteria were the presence of rotator cuff or LHBT injury evidenced
during the study.
A total of 34 patients agreed to participate in the study. There were 17 patients
from the tenodesis group and 17 patients from the tenotomy group. Of these 34 patients,
3 were excluded from the study after the MRI exam (2 in the tenotomy group and 1 in
the tenodesis group) because they had clinical and radiological signs of rotator cuff
injury (RCI) in the contralateral shoulder, with complaint of arm pain and loss of
elbow flexion strength. After exclusions, 16 patients from the tenodesis group and
15 patients from the tenotomy group were evaluated, totaling 31 patients (31 operated
arms and 31 contralateral arms for control). All patients had undergone surgery for
RCI treatment as the main diagnosis.
Magnetic resonance imaging
The MRI exams were all performed using a 1.5 Tesla GE Brivo 355 (General Electric,
Boston, MA, USA). All patients were imaged in the supine position, with the arms positioned
along the body in neutral rotation. T1-weighted coronal and axial images were obtained
of both arms of each patient following a specific protocol. The T1-weighted coronal
sequences were aligned with the body of the scapula with 3-mm spacing. This sequence
identified the apex of the humeral head and the length of the studied humerus, and
then half of the length was calculated. A double sequence of T1-weighted axial slices
was started from the lower edge of the acromion, with the first sequence with 25 slices
at a spacing of 3 mm followed by a second sequence with 35 more slices at a spacing
of 4 mm. With this sequence, from a triangulation with the images of the coronal slices,
it was possible to evaluate, with the thinner slices (3 mm), the distance between
the apex of the humeral head and the first muscle fiber of the long portion of the
biceps at the musculotendinous junction (AH-MTJ); that distance was compared with
the contralateral side, resulting in the difference in AH-MTJ distance ([Fig. 1]). In the group of patients subjected to tenodesis, we also assessed the distance
between the apex of the humeral head and the interference screw (AH-IS).
Fig. 1 Distance between the apex of the humeral head and the first muscle fiber of the long
portion of the biceps at the musculotendinous junction.
Individualizing the muscle belly of the long and short heads of the biceps and separating
them from the other muscles of the anterior compartment of the arm cannot be done
with MRI, as previously reported.[3]
[4] For this reason, we comparatively investigated the presence or absence of FI in
the muscle mass of the anterior compartment of the arm using the T1-weighted axial
section performed as close as possible to half the length of the humerus in both groups.
We adapted the existing FI classification,[12] comparing both arms of each patient. This classification suggests 5 FI grades (grade
0: normal; grade 1: some fatty streaks present; grade 2: < 50% FI; grade 3: 50% FI;
grade 4: > 50% FI),[13] and although this classification is not validated for use with MRI images of the
biceps, it was used for this purpose in a recent study.[4] The images were evaluated by an experienced radiologist, trained in the musculoskeletal
system and with 15 years of clinical practice.
Clinical Evaluation
The presence of arm deformities was classified as suggested by Scheibel et al.[14] as none, mild, moderate and severe; to facilitate statistical analysis, we subdivided
the presence of deformities into 2 groups: satisfactory (none or mild) and unsatisfactory
(moderate or severe) ([Figs. 2] and [3]). Elbow flexion strength was measured with a manual dynamometer (Fabrication Enterprises,
White Plains, New York, USA) (Baseline Mechanical Spring Push/Pull Dynamometer-60
Pounds) in both arms of each patient. Pain on palpation of the bicipital groove was
evaluated by the examiner by finger compression 5 cm distal to the anterolateral corner
of the acromion with the arm in neutral rotation.
Fig. 2 Satisfactory deformity.
Fig. 3 Unsatisfactory deformity.
Statistical Analysis
A descriptive analysis of the variables was performed by constructing box plots and
calculating the mean, standard error, standard deviation (SD), variance, coefficient
of variation, minimum, 1st quartile, median, 3rd quartile and maximum for all study
variables.
Due to the small sample size, the Fisher exact test was used to determine independence
between pairs of variables. When the assumptions of a normal distribution of the mean
were not met for the two samples or when one of the samples was very small (size equal
to two or three), the nonparametric Mann-Whitney test of equality of means was employed
for the two groups. If one of the sample sizes was smaller than four, the Mann-Whitney
test for the difference of means may not be valid because very small sample sizes
are not conducive to rejecting the tested hypothesis.
We calculated the respective descriptive levels (p-value) for all the hypothesis tests performed and rejected the hypotheses with descriptive
levels lower than the level of significance adopted, which was equal to 0.05. Data
analysis was performed using Minitab v.18 (Minitab, LLC., State College, PA, USA).
Results
For the 31 patients evaluated, the mean postoperative period at the time the MRI exam
was performed was 5 years: 5.8 years (range, 2 to 11 years) for the tenotomy group
and 4.2 years (range, 1 to 9 years) for the tenodesis group. The mean age of the participants
was 60 years old (53 to 77 years old) for the tenotomy group and 50.8 years old (33
to 69 years old) for the tenodesis group ([Table 1]).
Table 1
|
Group
|
Number of patients
|
Mean postoperative period
|
Indication for surgery
|
Mean age at surgery
|
Average Distance
|
Men
|
Women
|
|
Tenotomy
|
15
|
5.8 years (2–5 years)
|
Rotator cuff injury
|
63 years (53–77 years)
|
5
|
3 (20%)
|
12 (80%)
|
|
Tenodesis
|
16
|
4.2 years (1–9 years)
|
Rotator cuff injury
|
50.8 years (33–69 years)
|
1
|
15 (93.5%)
|
1 (6.5%)
|
Arm Deformity
Among all individuals studied, 7 had unsatisfactory (moderate or severe) deformities
in the operated arm. The difference in AH-MTJ distance ranged from - 0.6 cm to 3.1 cm
(mean of 0.34 cm). We found no significant relationship between satisfactory deformities
and the difference in AH-MTJ distance (p = 0.124) or between the incidence of unsatisfactory deformities and the difference
in AH-MTJ distance (p = 0.077). The means of the differences in the AH-MTJ distance were similar for patients
who presented satisfactory (0.175 cm) and unsatisfactory (0.171 cm) deformities (p = 0.984). Both had normal distribution ([Fig. 4]).
Fig. 4 Individual values of difference in length versus deformity.
Fatty Infiltration
No patient in either group exhibited changes in muscle mass of the anterior compartment
of the arm, and no case of FI was evident in the T1-weighted axial section at the
level of half the total length of the arm. All cases were classified as grade 0. ([Fig. 5]).
Fig. 5 Fatty infiltration.
Flexion Strength
The percent flexion strength of the operated arms did not follow a linear relationship
with the difference in AH-MTJ distance (p = 0.070) ([Fig. 6]). There was no difference in the mean percent strength between the groups (p = 0.399): 106.7% (range, 92.86 to 137.57%) in the tenodesis group and 102.4% (range,
75 to 130.77%) in the tenotomy group. Both groups presented a normal distribution.
Fig. 6 Percent force of the operated arm versus difference in length.
Pain on Palpation of the Intertubercular Groove
We did not find a significant relationship between pain on palpation of the bicipital
groove and the difference in AH-MTJ distance (p = 0.103) ([Fig. 7]). There was no significant correlation between the incidence of pain in the bicipital
groove and the mean percent elbow flexion strength of the operated arm (p = 0.062).
Fig. 7 Graph of individual values of no pain and pain.
Discussion
We found no evidence of FI in the muscle mass of the anterior compartment of the arms
subjected to procedures for the treatment of the LHBT injuries at a mean postoperative
follow-up of 5 years. There is no difference in muscle appearance on the MRI after
biceps tenotomy or tenodesis. Thus, the variations in the length of the biceps muscle
mass found in the present study (difference in AH-MTJ distance) did not contribute
to the onset of FI in the muscle belly. It was also not possible to establish a correlation
between the discrepancy in biceps length measured on MRI images and residual pain,
loss of flexion strength or arm deformity.
Our MRI images were collected from both arms and from all 31 patients, and the muscle
mass on the operated side was compared with the muscle mass of the contralateral side,
which was free of significant changes in the biceps and in the rotator cuff. We quantitatively
investigated the retraction of the muscle belly evaluated and its possible relationship
with the presence of FI.
There were seven patients with unsatisfactory (moderate or severe) deformities in
the operated arm and the means of the differences in the AH-MTJ distance were similar
for patients who presented satisfactory and unsatisfactory deformities. In conclusion,
the AH-MTJ distance is not the only aspect that determines the satisfaction of the
patient.
In the present study, we performed MRI examinations in both arms of the 31 patients
so that we could compare the muscle mass of the anterior compartment of both arms
and compare the variation in the length of the biceps muscle mass based on the difference
in the AH-MTJ distance. There was no relationship between the difference in the AH-MTJ
distance and FI and there was also no relationship between FI and arm deformity. Similarly,
in the 31 evaluated patients, there were no signs of FI after a mean postoperative
period of 5 years. This finding suggests that chronic biceps injuries can be repaired
while maintaining good muscle power and function because the musculature is preserved
over the years, even when there is retraction. We also did not find a correlation
between FI and arm deformity.
The chronic lack of load and simple tenotomy of the rotator cuff muscles can result
in FI and in a significant reduction in muscle volume.[2]
[15] The long head of the biceps originates in the supraglenoid tubercle proximally,
merges with the short head and inserts distally in the tuberosity of the radius, crossing
2 joints; therefore, even if its proximal end is detached, the distal end remains
inserted, generating load and activating muscle fibers.
In the case of tenotomy, residual pain may occur because some of the force generated
by muscle contraction is not transmitted to a tendon attached to the bone,[3] and an increase in the incidence of cramping pain in the biceps has been reported.[16] In tenodesis, pain may occur due to the permanence of synovitis in the bicipital
groove, especially if the fixation is performed at the joint margin or in the most
proximal portion of the intertubercular groove.[9] For any technique, when there is excessive retraction of the muscle belly, eventual
chronic tension in the common branch of the MCN could be a cause of pain. We did not
find a significant relationship between pain on palpation of the intertubercular groove
and the difference in AH-MTJ distance (p = 0.103). Residual pain related to the biceps may be found in 19% of cases of tenotomy
and in 24% of cases of tenodesis.[9] In our study, pain on palpation of the intertubercular groove was present in 10
of the 31 patients, 3 in the tenodesis group (18.75%) and 7 in the tenotomy group
(46.6%), but this difference was not significant (p = 0.135), possibly due to the sample size.
Our study compared percent elbow flexion strength with the difference in the AH-MTJ
distance, and there was no significant relationship (p = 0.070). It remains controversial whether there is a clinical difference in the
loss of strength after tenotomy and tenodesis.[9] Shank et al.[17] found no difference in strength when comparing the postoperative period after tenodesis
and tenotomy. Another study found a 20% loss of elbow flexion strength after tenotomy.[16] Although it was not the objective of our study, we individually compared the results
of the 2 groups and found no significant difference in percent elbow flexion strength
(p = 0.311). The tenotomy group had a mean strength of 102.4% compared with the contralateral
side, while the tenodesis group had a mean strength of 106.7%, which was slightly
higher.
Our study has some limitations. Initially, considering that the MTJ may extend for
∼ 8 cm,[11] we stipulated as the standard measure the first muscle fibers of the biceps identified
in the T1-weighted axial section; however, this measure may be subject to individual
anatomical variations. To minimize these effects, all MRI exams were performed on
the same machine, following the same protocol, and evaluated by the same radiologist.
A second limitation was not having performed a histological evaluation of the muscle
fibers. Although a histological evaluation is perhaps the gold standard for assessing
FI, it would be ethically impossible to perform; in addition, MRI evaluation has been
widely used to investigate muscle trophism and is well-established in shoulder and
elbow surgery. A third limitation is the use of a manual dynamometer, which is not
the perfect tool, but it gives a good idea of the elbow flexion strength. The number
of individuals may have been insufficient for some secondary analyses; however, it
was absolutely sufficient for the initial and main purpose of evaluating FI of the
biceps brachii muscle.
Conclusion
At an average follow-up of 5 years, none of the evaluated patients had evidence of
FI in the muscle mass of the anterior compartment of the arm after biceps tenodesis
or tenotomy.
There was no difference between tenotomy and tenodesis regarding the muscle mass pattern
observed by MRI.
It was not possible to establish a correlation between the biceps muscle length discrepancy
measured on MRI, the presence of deformities, strength and the presence of FI in the
anterior compartment of the arm.
