Seventeen embalmed upper limbs (eight left and nine right limbs) from the
Ray Last Anatomy Laboratory, University of Adelaide, were examined. The
cadavers were perfused and fixed with a mixture of ethanol, glycerol,
formalin, and phenol. The biceps muscle was identified and dissected with use
of loupe (×2.5) magnification. The muscle bellies of the long and short
heads were dissected, and their relationships to each other and the
proportions of each relating to the formation of the distal tendon were
observed and recorded. The bicipital aponeurosis, the bicipitoradial
bursa, and the insertion pattern of the distal tendon were dissected
and recorded. The size of the tendon was measured with use of a ruler
(accuracy, 0.5 mm). The bicipital aponeurosis was followed from proximal to
distal, and its components were recorded. In the dissected specimens, the
actions of the biceps muscle were assessed by pulling on the musculotendinous
junctions and observing the movements of the forearm.
The results are described in relation to three zones, from proximal to
distal, as preaponeurosis, aponeurosis, and postaponeurosis.
Zone 1: Preaponeurosis
All of the specimens had two muscle bellies, a short head originating from
the coracoid process of the scapula and a long head originating from the
superior lip of the glenoid. In ten specimens (Group 1), these two muscle
bellies continued along their entire length as separate muscles
(Fig. 1). The two muscle
bellies were surrounded by loose epimysial tissue. The short head remained on
the ulnar side of the arm throughout its course. The long head ran parallel to
the short head on the radial side of the arm.
The remaining seven specimens (Group 2) showed varying amounts of
interdigitation of muscle into a raphe in the distal third of the muscle
bellies. The maximum interdigitation occurred 5 cm proximal to the distal
biceps tendon. In these seven specimens, the two bellies of the muscle could
be easily separated with blunt dissection, by peeling them apart.
Zone 2: Aponeurosis (Lacertus Fibrosus)
In Group 1, each muscle continued as a separate tendon distally
(Fig. 2). The tendon of the
long head continued on the radial side of the tendon of the short head. The
cross section of the long head was crescentic in shape, and the short head was
oval. The cross-sectional areas of the separate tendons appeared equal
(Fig. 3).
In Group 2, the distal tendons continued in line with the respective muscle
bellies and could be easily separated with blunt dissection. The tendons
continued as for the other ten specimens described above.
The aponeurosis commenced at the level of the musculotendinous junction.
All specimens demonstrated that the aponeurosis consisted of three layers
(Fig. 3). These layers may be
important in stabilizing the tendons distally. The superficial layer
originated from the anterior radial aspect of the long head of the biceps just
proximal to the commencement of the distal biceps tendon. This superficial
layer macroscopically was the thickest layer in all specimens, and it passed
in a distal and ulnar direction anterior to the musculotendinous junction of
the short head. In some specimens, a rudimentary middle layer, which acted as
a mesentery, was present. It was the only layer to attach to the short head.
This middle layer passed in an ulnar direction to merge anteriorly with the
superficial layer. The deep layer originated from the deep radial side of the
musculotendinous area of the long head of the biceps. This layer passed in an
ulnar direction deep to the tendon of the short head to merge with the other
two layers.
These three layers merged and continued distally, superficial to the ulnar
flexor muscles of the forearm. There were several strong fascial adhesions to
the ulnar flexor muscles, tethering the aponeurosis. The aponeurosis also
continued radially to the forearm flexor muscles as well as the median nerve
and brachial artery. The aponeurosis was attached to both the radial and ulnar
aspects of the proximal part of the ulna, completely encircling the forearm
flexor muscles (Fig. 4). It
inserted into the antebrachial fascia and reinforced it. There were several
perforating holes in the radial side of the aponeurosis for the recurrent
radial vessels.
The two distal tendons in the majority of the specimens (Group 1) were able
to move separately from one another in a sliding action. The tendons in Group
2 followed the same line as those in the other group, but they did not have
the ability to glide independently.
Zone 3: Postaponeurosis
The two tendons continued distal to the aponeurosis and inserted into the
proximal part of the radius. In both groups, the tendon of the long head
passed deep to the tendon of the short head to insert more proximally. The
insertion of the tendon of the long head was oval in shape, occupying most of
the radial tuberosity. The tendon of the short head curved anterior to the
tendon of the long head, to insert in a fan-like fashion into the distal
portion of the radial tuberosity, and extended distal to it
(Fig. 5). The attachments of
the two distal tendons were surrounded by the bicipitoradial bursa
(Fig. 6). This bursa completely
encircled the distal tendons in all specimens. The bursa could be easily
distended by injection of =7 mL of saline solution or latex on its deep
radial side. The bursal membrane continued around the ulnar side of the
tendons, where it was adherent to the tendon and would not distend. The bursa
was attached proximally to the biceps tendon on the radial aspect. From this
point, it draped down over the tendons, adhering to both tendons on the ulnar
aspect. The bursa was attached along the proximal deep edge of the tendon of
the long head to create a teardrop shape. Thus, the bursa lay between the
groove in the brachialis muscle and the distal biceps tendons with the elbow
extended, and between the proximal part of the radius and the biceps tendons
during pronation of the forearm.
The insertion of the long head was at a point farthest away from the
rotation of the radius, potentially providing a greater lever arm to increase
supination power (Fig. 7).
Conversely, the tendon of the short head was attached more distally, providing
it with the potential for greater flexion power.
We demonstrated that, in most individuals, the biceps muscles are two
independent muscle bellies of the two heads, with two separate tendon areas.
The remaining individuals had several interdigitations between both muscle
bellies and again two easily defined tendons. No biomechanical or histological
investigations were performed, and this is a potential limitation of the
study.
The distinct pattern found in the majority of patients was described
recently in a case
report4. The authors
reported a duplicated biceps tendon and failed to identify any evidence of
fusion between the muscle bellies in the distal 8 to 10
cm4. This was an
uncommon finding. It has been our clinical experience that an acute rupture of
the biceps tendon often occurs with avulsion of the tendon from the bone as
one unit, with the two heads often held together with loose areolar tissue,
and with the lacertus fibrosus usually remaining intact.
The biceps tendon is controlled by the lacertus fibrosus, which is a
fixed-length structure. As the forearm muscles contract, the flexor muscle
mass migrates proximally, increasing its cross-sectional area. This tenses the
aponeurosis, pulling the biceps tendon medially. This increased force on the
biceps tendon may contribute to the etiology of rupture of the distal biceps
tendon (Fig. 8).
In light of our findings in repairs of acute rupture of the distal biceps
tendon, we use an Endobutton (Smith and Nephew, Memphis, Tennessee) and place
Bunnell sutures in each of the two tendon bundles, as we originally described
in 20005 and as has
been subsequently reported by
others6-9.
Both tendon components are secured to the proximal part of the radius in their
correct orientation. If there is tendon retraction of 2 cm, we may release the
lacertus fibrosus to allow the tendon to be advanced onto the radial
tuberosity. This ensures that no neurovascular structures become entrapped by
the tight lacertus in the pronated position. With the arm in pronation and
extension, the lacertus is then repaired to the biceps tendon, to reconstitute
it without impinging on the neurovascular bundle.
The current study has had further direct implications on our clinical
practice. In a delayed rupture with substantially greater retraction, we use a
hamstring tendon graft to reconstitute the length of the biceps
tendon10. The
tendon unit is often scarred together to a single mass. Since performing this
study, we have modified our technique so that the proximal end of each tendon
graft is woven into each of the two separate muscle bellies. As it passes
distally, the normal rotation of each tendon is recreated so that the short
head inserts more distally when it is locked into the radial tuberosity.
The concept of two individual muscle bellies driving separate parts of the
distal biceps tendon unit with active motion of the forearm has not been fully
explored, and its clinical benefits are yet to be determined. However, the
anatomical data from the present study encourage awareness and further
development of this concept. ?