A radiographic evaluation of birotational head motion in
the bipolar shoulder hemiarthroplasty
Peter Stavrou, MBBS,aJohn Slavotinek, FRACR,band Jegan Krishnan, FRACS,aBedford Park, South Australia
The concept of bipolar hemiarthroplasty has been de-
scribed in the hip for more than 20 years, its role be-
ing to decrease acetabular wear. Shoulder bipolar
hemiarthroplasty was described shortly afterward. The
purpose of this study was to determine if the prosthesis
acted as a bipolar device moving primarily at the in-
ner metal on polyethylene bearing as intended, or as
a unipolar hemiarthroplasty moving at the outer metal
on glenoid surface articulation. Eleven bipolar shoul-
der hemiarthroplasties, with a minimum follow-up of
22 months, were examined fluoroscopically. The pro-
portion of arm abduction occurring at the scapulotho-
racic articulation and between the two components of
the bipolar hemiarthroplasty was assessed and com-
pared with that of normal patients and those with total
shoulder replacements previously reported in the litera-
ture. The results of this study show that most of the
movement in active arm abduction occurred at the
scapulothoracic articulation and that the bipolar hemi-
arthroplasty acted predominantly as a unipolar device.
(J Shoulder Elbow Surg 2006;15:399-401.)
The concept of bipolar hemiarthroplasty was origi-
nally described for the hip by Bateman2and Gil-
berty5independently in 1974. The objective was to
allow preferential movement at the metal-on-polyeth-
ylene inner bearing surface rather than the outer
metal-on-articular cartilage surfaces, thus minimizing
damage to articular cartilage in the acetabulum. The
determinants of whether a femoral bipolar hemiar-
throplasty functions as a bipolar hemiarthroplasty are
varied. Phillips8found that the condition of the ace-
tabular cartilage determined outcome in osteoarthritis
of the hip joint, 80% of prostheses functioned as a
bipolar hemiarthroplasty, and in subcapital fractures
(normal cartilage), 75% functioned as a unipolar
hemiarthroplasty. These results were independent of
Conversely, Mess and Barmada7found weight-
bearing status to be a significant factor when exam-
ining function. The loaded (weight-bearing) prosthesis
has a ratio of inner-to-outer bearing surface move-
ment of 3:1 compared with the ratio of 1:1 in the
non-weight-bearing situation, and Verberne12found
no difference when comparing active and passive
movements. He also found that, by 3 months postop-
eratively, the total amount of inner bearing surface
movement was less than 20% of that seen in the
immediate postoperative period.
Swanson10,11designed a bipolar hemiarthroplasty
for the shoulder in 1975. This was for use in both
osteoarthritis and rheumatoid arthritis of the shoulder. Its
main purpose was to prevent superior subluxation of the
humeral head due to rotator cuff deficiency and the
development of cuff arthropathy by articulating with
both the glenoid articular surface and the coracoacro-
mial arch. This provided several potential advantages
over the unipolar hemiarthroplasty and total shoulder
replacement. The bipolar hemiarthroplasty eliminated
the problems of glenoid fixation and loosening seen in
total shoulder replacement. There is also a reduction in
the potential for abutment of the greater tuberosity
against the acromion because of its action as a spacer.
The increase in lateral offset allows greater lever arms
for the rotator cuff and deltoid in abduction of the arm.
Furthermore, it was postulated that glenoid wear would
be lessened owing to preferential movement at the inner
metal-on-polyethylene bearing surface rather than the
outer metal-on-glenoid articulation. This may lead to
preservation of glenoid articular cartilage and a reduc-
tion in polyethylene stress concentrations and wear. The
aims of this study were to examine bipolar shoulder
hemiarthroplasty and to determine whether the motion
elements at the articulating interfaces are those of a
unipolar or bipolar prosthesis.
MATERIALS AND METHODS
Between November 1994 and September 1996, 11
shoulders in 11women were operated on. Their average
age was 75.6 years old (range, 59-86 years). The indica-
aFrom the Departments of Orthopaedic Surgery and
Imaging, Flinders Medical Centre and Repatriation General
Hospital and Flinders University.
Reprint requests: Associate Professor J Krishnan, FRACS, Senior
Clinical Director, Department of Orthopaedic Surgery, Flinders
Medical Centre, Flinders Drive, Bedford Park, South Australia
Copyright © 2006 by Journal of Shoulder and Elbow Surgery
Board of Trustees.
tion for surgery was degenerative osteoarthritis in 7 and
trauma (4-part neck of humerus fractures) in 4. The average
time between fracture and hemiarthroplasty was 26 days
(range, 22-34 days). Two patients had previously had
rotator cuff surgery. The rotator cuff was intact in all shoul-
ders at operation. Nine shoulders were on the right side, the
other two were on the left. All patients had a deltopectoral
approach to the shoulder. The same prosthesis (Biangular
Bipolar Shoulder Hemiarthroplasty, Biomet, Warsaw, IN)
was used in all patients. In 7 patients, the prosthesis was
cemented, and the remaining 4 received porous, coated
implants. Patients were followed up for a minimum of 22
months (range, 22-42 months), with an average of 31.9
Patients were assessed fluoroscopically in active abduc-
tion with the arm in the scapular plane. The position of the
Bipolar Shoulder Hemiarthroplasty was assessed with the
arm in 5 positions of abduction: 0° 30°, 45°, 60° and
maximal abduction. Both arms were abducted simulta-
neously, and arm angles were measured with a goniometer
to which a spirit level was attached, in a means similar to
that described by Freedman and Munro.3Spot films were
taken and the radiographic angles were measured (Figure
1). Angle A indicates the angle of the arm to the vertical,
this angle having two components: S, the angle of the
scapular articular surface to the vertical, and GH, the
glenohumeral angle as described by Freedman and
Two equations can be determined from Figure 1. The
arm angle (A) is equal to the scapular angle (S) and
glenohumeral angle (GH) added together; thus, A ? S ?
GH as described by Freedman and Munro.3Second, the
glenohumeral angle is equal to the addition of its two
component angles, inner bearing (I) and outer bearing
surface (O); thus, GH ? I ? O.
We can thus determine two ratios: (1) a ratio for the arm
angle component movements, ie, the glenohumeral-to-scap-
ular movement (GH/S), which has been described by
Freedman and Munro,3and (2) a new ratio, inner bearing-
to-outer bearing (I/O), which is derived from the compo-
nents of glenohumeral movement and demonstrates the
ratio of movement at the inner metal/polyethylene bearing
compared with the outer metal/glenoid articulation. A high
ratio would indicate the component is functioning as a
bipolar hemiarthroplasty. These ratios were calculated from
the total change in average angles of our measured param-
The results of the radiographic measurements are
presented in Table I. Two points need to be made with
regard to Table I. A negative value for S (scapular
articular surface angle) indicates that the articular
surface is facing inferiorly. A negative value for O
(outer bearing surface angle) indicates that the lines
from the glenoid and outer bearing converge inferior
to the glenoid (this was usually the case).
The results of this study demonstrate that, in the
adducted position (0°), the average angle of the
scapular articular surface (S) is pointing inferiorly
(?4.6°); this is similar to the Freedman and Munro
finding of ?5.29°. One can also see that only a small
portion of change in glenohumeral angle occurs at
Figure 1 Diagram of measured angles: A, angle of the arm to the vertical; S, angle of the scapular articular surface
to the vertical; GH, glenohumeral angle; I, angle of the inner bearing surface of the bipolar prosthesis to the arm;
O, angle between the outer bearing surface and the glenoid.
400 Stavrou, Slavotinek, and Krishnan
J Shoulder Elbow Surg
the inner bearing surface, 5.1° of 22.2°. I/O ?
0.30, which is consistent with an I:O ratio of 1:3.4.
Thus, most of the glenohumeral change is between the
outer bearing surface and the glenoid. Furthermore,
most of the change in arm angle is scapular (40.4°)
rather than glenohumeral (22.2°); thus, GH/S ?
0.55, which is equivalent to a GH:S ratio of 1:1.8.
When the etiologies of the shoulders are com-
pared, one can see that in the 7 shoulders with
osteoarthritis, the mean GH/S ? 0.52, and in the 4
with fractures, the mean GH/S ? 0.60. Although
these values are similar, the mean I/O differed be-
tween the two groups. In the osteoarthritis group, the
I/O ? 0.39 and in the fracture group, GH/S ? 0.18.
When the literature is examined, the ratio of gle-
nohumeral to scapulothoracic movement is quite var-
ied. Inman et al6determined that the normal gleno-
humeral-to-scapulothoracic ratio was 2:1. More
recent studies by Freedman and Munro3have found
this ratio to be 1.35:1, and Poppen and Walker9
found the ratio to be 1.25:1.
Friedman4examined patients before and after to-
tal shoulder replacement and found that, preopera-
tively, the glenohumeral-to-scapulothoracic ratio was
reversed (1:2) and that, postoperatively, this ratio
was closer to normal (1:1.3). Friedman concluded
that biomechanics in the arthritic shoulder are abnor-
mal both before and after operation. These last ratios
are similar to those found in our study (GH:S ?
1:1.8), a reversal of the normal ratio after shoulder
The literature is sparse with regard to radiologic
reports of bipolar shoulder hemiarthroplasty. Wor-
land et al13examined 22 patients (average follow-
up, 28.4 months), all with rotator cuff arthropathy,
who had bipolar shoulder hemiarthroplasty in situ
with plain radiographs in adduction and maximal
abduction and found that motion persisted postoper-
atively between the head and the shell. Arredondo
and Worland1examined 48 patients with osteoarthri-
tis (average follow-up, 3.1 years) with plain radiog-
raphy in adduction and full abduction and found an
average increase in the humerus-to-shell angle of 19°.
This study compares inner bearing movement in
those with osteoarthritis and those with fractures. In
neither group did the bipolar shoulder hemiarthro-
plasty act as a bipolar prosthesis. The osteoarthritis
group had a slightly better ratio of inner-to-outer bear-
ing movement than the fracture group, which is in
keeping with the published reports on the bipolar
hemiarthroplasty in the hip.8There was little differ-
ence in GH:S ratios between the groups.
In conclusion, this study demonstrates that, in ab-
duction, most average arm angle changes occurred
at the scapulothoracic articulation, 40.4° of 62.6°,
rather than at the glenohumeral articulation. We also
demonstrate that only a small portion of the glenohu-
meral angle change occurs at the inner bearing sur-
face (5.1° of 22.2°), the mean I:O ratio being 1:3.4.
This study indicates that, in radiologic examination of
bipolar shoulder hemiarthroplasty, the bipolar com-
ponent functions principally as a unipolar component.
1. Arredondo J, Worland RL. Bipolar shoulder arthroplasty in pa-
tients with osteoarthritis: short-term clinical results and evaluation
of birotational head motion. J Shoulder Elbow Surg 1999;8:
2. Bateman JE. Single-assembly total hip prosthesis—preliminary
report. Orthop Dig 1974;2:15-22.
3. Freedman L, Munro RR. Abduction of the arm in the scapular
plane: scapular and glenohumeral movements. J Bone Joint Surg
4. Friedman RJ. Prospective analysis of total shoulder arthroplasty
biomechanics. Am J Orthop 1997;26:265-270.
5. Gilberty RP. Hemiarthroplasty of the hip using a low-friction
bipolar endoprosthesis. Clin Orthop 1983;175:86-92.
6. Inman VT, Sanders JB, Abbott LC. Observations of the function of
the shoulder joint. J Bone Joint Surg Am 1944:26:1-31.
7. Mess D, Barmada R. Clinical and motion studies of the Bateman
bipolar prosthesis in osteonecrosis of the hip. Clin Orthop 1990;
8. Phillips TW. The Bateman bipolar femoral head replacement.
J Bone Joint Surg Br 1987;69:761-4.
9. Poppen NK, Walker PS. Normal and abnormal motion of the
shoulder. J Bone Joint Surg Am 1976;58:195-201.
10. Swanson AB, de Groot Swanson G, Maupin BK, Wei JN, Khalil
MA. Bipolar implant shoulder arthroplasty. Orthopedics 1986;
11. Swanson AB, de Groot Swanson G, Sattel AB, Cendo RD, Hynes
D, Wei JN. Bipolar implant shoulder arthroplasty. Clin Orthop
12. Verberne GHM. A femoral head prosthesis with a built-in joint. A
radiological study of the movements of the two components.
J Bone Joint Surg Br 1983;65:544-547.
13. Worland RL, Jessup DE, Arredondo J, Warburton KJ. Bipolar
shoulder arthroplasty for rotator cuff arthropathy. J Shoulder Elbow
Table I Average angles for the measured parameters
Average arm angle
A, Angle of the arm to the vertical; S, angle of the scapular articular surface
to the vertical; GH, glenohumeral angle; I, angle of the inner bearing surface
of the bipolar prosthesis to the arm; O, angle between the outer bearing
surface and the glenoid
J Shoulder Elbow Surg
Volume 15, Number 4
Stavrou, Slavotinek, and Krishnan401