Boileau, P., Watkinson, D., Hatzidakis, A. M. & Hovorka, I. Neer Award 2005: The Grammont reverse shoulder prosthesis: results in cuff tear arthritis, fracture sequelae, and revision arthroplasty. J. Shoulder Elbow Surg. 15, 527-540
This clinical study was performed to analyze the midterm results and potential complications of the reverse prosthesis in different diagnosis. Forty-five consecutive patients with Grammont prosthesis were evaluated clinically and radiographically with a mean follow-up of 40 months (range, 24-72 months). The indication was a massive and irreparable cuff tear associated with arthrosis (CTA) in 21 cases, fracture sequelae (FS) with arthritis in 5 cases, and failure of a revision arthroplasty (revision) in 19 cases. Fourteen complications occurred in 11 patients. 3 dislocations, 3 deep infections (all 3 in the revision group), 1 case of aseptic humeral loosening, 2 periprosthetic humeral fractures, 1 intraoperative glenoid fracture, 1 wound hematoma, 2 late acromial fractures, and 1 axillary nerve palsy. Of the patients, 10 (22%) required further surgery: 4 reoperations, 4 prosthesis revisions, and 2 prosthesis removals. Complications were higher in revision than in CTA (47% vs. 5%). All 3 groups showed a significant increase in active elevation (from 55 degrees preoperatively to 121 degrees postoperatively) and Constant score (from 17 to 58 points) but no significant change in active external rotation (from 7 degrees to 11 degrees ) or internal rotation (S1 preoperatively and postoperatively). Of the patients, 78% were satisfied or very satisfied with the result and 67% had no or slight pain. However, the postoperative Constant score, adjusted Constant score, and American Shoulder and Elbow Surgeons shoulder score were all significantly higher in the CTA group with as compared with the revision group (P = .01, .004, and .002, respectively). Scapular notching was seen in 24 cases (68%). No glenoid loosening was observed at current follow-up, even when the notch extended beyond the inferior screw (28% of cases). Atrophy of severe fatty infiltration of the teres minor was associated with lower external rotation (15 degrees vs 0 degrees , P = .02) and lower functional results (Constant score of 46 points vs 66 points, P < .007). The Grammont reverse prosthesis can improve function and restore active elevation in patients with incongruent cuff-deficient shoulders; active rotation is usually unchanged. Results are less predictable and complication and revision rates are higher in patients undergoing revision surgery as compared with those in patients with CTA. Results of the reverse prosthesis depend on the diagnosis and on the remaining cuff muscles, specifically the teres minor. Surgeons should be vigilant with regard to low-grade infection in revision surgery.
Neer Award 2005: The Grammont reverse shoulder
prosthesis: Results in cuff tear arthritis, fracture sequelae,
and revision arthroplasty
Pascal Boileau, MD, Duncan Watkinson, FRCS, Armodios M. Hatzidakis, MD, and Istvan Hovorka, MD, Nice,
This clinical study was performed to analyze the mid-
term results and potential complications of the reverse
prosthesis in different diagonosis. Forty-ﬁve consecutive
patients with Grammont prosthesis were evaluated clin-
ically and radiographically with a mean follow-up of
40 months (range, 24-72 months). The indication was
a massive and irreparable cuff tear associated with
arthrosis (CTA) in 21 cases, fracture sequelae (FS) with
arthritis in 5 cases, and failure of a revision arthro-
plasty (revision) in 19 cases. Fourteen complications
occurred in 11 patients. 3 dislocations, 3 deep infec-
tions (all 3 in the revision group), 1 case of aseptic
humeral loosening, 2 periprosthetic humeral fractures,
1 intraoperative glenoid fracture, 1 wound hematoma,
2 late acromial fractures, and 1 axillary nerve palsy.
Of the patients, 10 (22%) required further surgery: 4
reoperations, 4 prosthesis revisions, and 2 prosthesis
removals. Complications were higher in revision than
in CTA (47% vs. 5%). All 3 groups showed a signiﬁ-
cant increase in active elevation (from 55° preopera-
tively to 121° postoperatively) and Constant score
(from 17 to 58 points) but no signiﬁcant change in
active external rotation (from 7° to 11°) or internal ro-
tation (S1 preoperatively and postoperatively). Of the
patients, 78% were satisﬁed or very satisﬁed with the
result and 67% had no or slight pain. However, the
postoperative Constant score, adjusted Constant score,
and American Shoulder and Elbow Surgeons shoulder
score were all signiﬁcantly higher in the CTA group
with as compared with the revision group (P ! .01,
.004, and .002, respectively). Scapular notching was
seen in 24 cases (68%). No glenoid loosening was
observed at current follow-up, even when the notch
extended beyond the inferior screw (28% of cases).
Atrophy of severe fatty inﬁltration of the teres minor
was associated with lower external rotation (15° vs
0°, P ! .02) and lower functional results (Constant
score of 46 points vs 66 points, P " .007). The Gram-
mont reverse prosthesis can improve function and re-
store active elevation in patients with incongruent cuff-
deﬁcient shoulders; active rotation is usually
unchanged. Results are less predictable and complica-
tion and revision rates are higher in patients undergo-
ing revision surgery as compared with those in pa-
tients with CTA. Results of the reverse prosthesis
depend on the diagnosis and on the remaining cuff
muscles, speciﬁcally the teres minor. Surgeons should
be vigilant with regard to low-grade infection in revi-
sion surgery. (J Shoulder Elbow Surg 2006;15:
Shoulder arthroplasty via an unconstrained prosthe-
sis is a well-tested procedure and, in most cases,
provides excellent pain relief and improved func-
However, in situations in which a dam-
aged and incongruent glenohumeral joint is associ-
ated with a deﬁcient or nonfunctional rotator cuff, the
results are often much less satisfactory.
From the Department of Orthopaedic Surgery and Sports Trauma-
tology, Hôpital de L’Archet, University of Nice.
This study received the Charles Neer Award at the annual Ameri-
can Shoulder and Elbow Surgeons meeting, Washington, DC,
No beneﬁts in any form have been or will be received from a
commercial party related directly or indirectly to the subject of
this article. No funds were received in support of this study.
Reprint requests: Pascal Boileau, MD, Professor and Chairman,
Department of Orthopaedic Surgery and Sports Traumatology,
Hôpital de L’Archet, University of Nice, 151, Route de St Antoine
de Ginestière, 06202 Nice, France (E-mail: boileau.p@chu-
Copyright © 2006 by Journal of Shoulder and Elbow Surgery
Board of Trustees.
there are at least 3 situations in which surgeons must
face this challenging combination: ﬁrst, in cases of
arthritis associated with a massive and irreparable cuff
tear; second, in cases of fracture sequelae in which
distortion of the proximal humeral anatomy around the
cuff insertion results in a functionally useless cuff; and
third, in cases of revision after failure of a previous
arthroplasty in which the cuff is deﬁcient, scarred, or
retracted or has undergone fatty inﬁltration.
In such incongruent and rotator cuff–deﬁcient
shoulders, the forces that normally counteract the
upward component of the deltoid and stabilize the
center of rotation of the shoulder are lost; as a result,
deltoid contraction tends to cause the ascension of
the humerus, rather than the rotatory movement
necessary for elevation of the limb.
When a shoulder arthroplasty is performed in this
difﬁcult biomechanical environment, the standard
of care is an unconstrained hemiarthroplasty,
accepting Neer’s “limited goals” (pain relief but
strained prostheses have been introduced to exceed
these limited goals, without success. Some provided
good pain relief, but active elevation was usually
Furthermore, none of these
designs have been able to withstand the increased
stresses that their ﬁxed fulcrum imposed on the bone-
prosthesis interface, leading to early failure and
abandonment of these designs.
The Grammont reverse prosthesis was developed
to provide a solution to this dilemma, restoring mobil-
ity around a stable center of rotation and avoiding
The Grammont prosthesis is not
fully constrained, but it does have congruent joint
surfaces, making it a semiconstrained prosthesis. Un-
like previous reverse ball-and-socket designs, the
Grammont uses a large hemispherical glenoid com-
ponent with no neck. This means that the center of
rotation of the shoulder joint (the center of the sphere)
is actually situated at the glenoid bone–prosthesis
interface, minimizing torque on the component.
addition, the Grammont has a congruent polyethyl-
ene humeral cup, implanted with a nonanatomic,
more horizontal inclination of 155°. This has the
advantage of lowering the humerus, thereby placing
the deltoid muscle under tension to provide a stable
and biomechanically stronger fulcrum.
fulcrum is essential for active elevation in a shoulder
with a severely deﬁcient unbalanced rotator cuff.
This clinical study was performed to analyze the
midterm results and potential complications of the
Delta prosthesis in 3 groups of patients: those with
massive and irreparable cuff tear arthrosis (CTA),
those with fracture sequelae (FS), and those who
needed revision surgery after failure of a previous
MATERIALS AND METHODS
The Delta shoulder prosthesis (Depuy-International Ltd) is
a reverse ball-and-socket prosthesis that was designed in
1985 by Professor Paul Grammont in Dijon, France.
The glenoid component was, initially, a cemented barrel
(two thirds of a sphere) that ﬁt around the glenoid. This
design was changed to the current hemispherical conﬁgu-
ration in 1991 because of initial problems with loosening.
The original hemispheric component was ﬁxed to the gle-
noid baseplate via a peripheral thread. This design was
changed after unscrewing of the glenosphere from the
metaglene occurred, mainly in right shoulders, as active
forward ﬂexion produced a ratchet-like effect. This problem
was solved in 1996 when the peripheral thread was
changed to a Morse taper with a central, countersunk
screw. Our experience began in 1997, after these initial
problems were addressed, and the design has since re-
The glenoid component consists of 2 parts: a baseplate
(the metaglene), 29 mm in diameter, to which a polished
hemisphere (the glenosphere) is secured. The metaglene,
which has a roughened hydroxyapatite-coated surface, is
ﬁxed to the glenoid via a central peg and 4 divergent
screws. The superior and inferior screws are aimed at the
coracoid base and the inferior pillar of the scapula, respec-
tively, for optimal hold. Their heads are threaded to lock
into the metaglene at an angle of 120° relative to each
other. The anterior and posterior screws are nonlocking and
are inserted at a free angle. The glenosphere is available in
a diameter of either 36 mm or 42 mm.
The humeral component consists of a circular stem with a
modular neck component and a polyethylene cup. The
inclination of the neck of the humeral prosthesis is 155°. The
stem and neck are available with either a polished surface
for cementing or a hydroxyapatite-coated surface for press-
ﬁt application. The stem is available in 3 lengths (100, 150,
or 180 mm) and 4 diameters (6, 9, 12, or 15 mm). The
neck diameter is either 36 mm or 42 mm, corresponding to
the humeral cup. The polyethylene cup is 6 mm thick and is
also available in a more constrained conﬁguration for
instability problems; other thicknesses are now available (3,
6, and 9 mm). Finally, a neck extension may be added
between the neck and the cup to further increase the offset
The No. 3 in the name of the prosthesis refers to the
reverse design. The Delta 1 prosthesis is a hemiarthroplasty
with a metal humeral head inserted onto the humeral neck
component instead of the humeral cup. The Delta 2 prosthe-
sis is a standard-conﬁguration total shoulder prosthesis with
the ball on the humerus and a polyethylene cup inserted onto
the metaglene. Because they share the same stem, a Delta 3
prosthesis can be converted to a Delta 1 hemiarthroplasty
should a major problem occur with the glenoid component,
either at the time of operation or subsequently.
Grammont originally described a transacromial ap-
proach but subsequently considered using a deltopectoral
We used a superior, transdeltoid approach
(cuff approach) for our ﬁrst 4 cases but then changed to a
classic deltopectoral approach for 2 reasons.
528 Boileau et al J Shoulder Elbow Surg
concerned about possible damage to the deltoid muscle,
the sole motor for this prosthesis. Furthermore, access to the
humeral diaphysis was sometimes necessary, especially in
the revision and FS groups.
Even with a massive cuff tear, the inferior third of the
subscapularis tendon was often found to be partially intact.
This was, therefore, released from the lesser tuberosity, as in
a standard shoulder arthroplasty, and preserved for rein-
sertion at the end of the procedure by use of transosseous,
nonabsorbable No. 5 Ethibond sutures (Ethicon, Somer-
Revision cases and FS cases were more complex. The
superior cuff sometimes appeared to be in continuity, but
closer inspection invariably showed that this was nothing
more than a thin layer of nonfunctional ﬁbrous tissue. This
superior cuff remnant was excised to obtain adequate
exposure. In FS cases, the tuberosities were either malunited
or nonunited. Indeed, in many cases, all that remained was
a thin shell of cortical bone attached to a retracted and
scarred cuff. The tuberosities’ abnormal position often con-
stituted a physical block to joint movement, necessitating
excision. In 3 cases, it was possible either to preserve the
tendons or to reattach the tuberosities to the prosthetic neck
at the end of the procedure.
A jig was used to make the humeral cut with an inclina-
tion of 155° and approximately 20 to 30° of retroversion,
by use of the forearm as a reference. Although it has been
recommended to implant this prosthesis with 0° to 10° of
retroversion, we chose 20° to 30° of retroversion to alter the
native anatomy of the proximal humerus as little as possible;
more speciﬁcally, we strove to avoid destruction of the
lesser tuberosity during humeral preparation (for reattach-
ment of the subscapularis on its footprint). If the bone
appeared fragile, further humeral preparation was de-
ferred until the glenoid components were in place. Other-
wise, sequential reamers were used to open and prepare
the humeral canal; next, a special conical reamer was used
to shape the metaphysis. The trial prosthesis (minus the trial
cup) was left in place to protect the proximal humerus
during glenoid preparation. We cemented the stems and
used a cement restrictor in all cases. The humeral polyeth-
ylene liner was then impacted on the humeral component. A
relatively tight trial reduction was sought to ensure ade-
quate tension of the deltoid. In cases of severe bone loss of
the proximal humerus (those with FS or after revision arthro-
plasty), bilateral, scaled radiographs of both humeri were
taken. A tracing was made to assess the length of the bone
loss of the deﬁcient humerus and to estimate the height at
which the prosthesis should be cemented to maintain stability.
Adequate exposure of the glenoid is essential because
the Delta reamers and components are more bulky than
those used in a standard total shoulder arthroplasty. To
achieve the necessary release, an anterior and inferior
release of the capsule from the glenoid was performed to
allow posterior dislocation of the humerus.
After placement of 2 fork retractors on the glenoid, 1
anterior and 1 posterior, a central hole was drilled for the
central peg of the metaglene; this also served as a pilot hole
for the reamers. After the remaining subchondral bone was
reamed to a ﬂat and uniform surface, holes were drilled at
the periphery of the glenoid surface with a 3.2-mm drill to
obtain bone bleeding and to aid secondary ﬁxation of the
hydroxyapatite-coated metaglene. The metaglene was then
impacted into position. The anterior and posterior screws
were placed ﬁrst, to obtain compression between the meta-
glene and the glenoid bone. Fixation was augmented with
the upper and lower divergent locking screws.
Finally, the glenosphere was ﬁxed to the metaglene, by
ﬁrst impacting the Morse taper and then tightening the
central screw. A 36-mm glenosphere was used in all but 4
cases (large men) in which a 42-mm glenosphere was used.
Intraoperative determination of deltoid tension may be
difﬁcult, guided mostly by surgical experience. A tight
reduction is the only rubric; with the arm at the side and the
elbow extended, the conjoined tendon should feel ten-
sioned after reduction.
In CTA cases, the prosthesis was
usually stable, as a minimal bone cut had been made on the
humerus and the glenosphere had been implanted low on
the glenoid surface. In revision or FS cases, stability was
often more difﬁcult to attain: the humeral prosthesis was
cemented proud, and additional cup thickness (3, 6, or 9
mm) was added if needed. We avoided the more con-
strained cup, as we believed that the additional constraint
on the glenoid would increase the shear forces and, there-
fore, the risk of glenoid loosening. If the prosthesis was felt
to be too tight (ie, not possible to reduce), the humerus was
recut as needed.
The subscapularis tendon, or at least part of it, was
reinserted with transosseous sutures in most cases. A biceps
tenodesis was systematically performed by suturing the
tendon to the pectoralis major tendon after resection of the
intraarticular portion of the biceps.
Postoperatively, the patient’s arm was placed in a simple
sling. Patients were encouraged to perform self-rehabilita-
tion with pendulum exercises for 5 minutes, 5 times a day.
After 3 weeks, rehabilitation was started with a physiother-
apist who was instructed to mobilize the arm above the
horizontal level in the plane of the scapula. The combina-
tion of abduction at 90° with external rotation (throwing
position) was not allowed, as it placed the prosthesis at risk
for anterior dislocation. In patients who were at high risk for
prosthetic instability (revision), immobilization with an ab-
duction splint at 60° was used for 4 weeks.
Over a 5-year period, starting in 1997, 50 consecutive
Grammont shoulder replacements were performed either by
the senior author (P.B.) or under his direct supervision. All
patients were prospectively followed up clinically and ra-
diographically on a regular basis: at 3, 6, and 12 months
and then yearly after the procedure. Of the patients, 5 were
excluded: 2 had died, 1 had had a debilitating cerebro-
vascular accident (unrelated to surgery), and severe demen-
tia (Alzheimer’s disease) developed in 1. The last excluded
patient underwent reconstruction with a custom reverse
prosthesis after tumor excision.
Forty-ﬁve patients were
available for clinical analysis. The mean follow-up was 40
months (range, 24-72 months).
The patients were divided into 3 groups, according to
etiology: massive and irreparable CTA, sequelae of a prox-
imal humeral fracture (FS), and revision prosthesis after
failure of a previous arthroplasty (revision). Pain, associ-
ated with stiffness or pseudoparalysis of the shoulder (inef-
fective shrug because of anterosuperior subluxation of the
J Shoulder Elbow Surg
Boileau et al 529
Volume 15, Number 5
humeral head), was the primary indication for surgery in
each of the 3 subgroups.
The CTA group consisted of 21 patients with a mean age
of 77 years (range, 67-86 years). They were signiﬁcantly
older (P ! .0002) than those in the revision and FS groups.
Of the patients in the CTA group, 19 were women (90%),
and the dominant side was involved in 86%. All patients
had persistent pain and pseudoparalysis of the shoulder. All
had been treated with pain medications, antiinﬂammatory
medications, and physical rehabilitation for at least
6 months. Two had lost active elevation after extensive
subacromial decompression surgery: one overaggressive
acromioplasty and one anterior acromionectomy compli-
cated by postoperative avulsion of the anterior deltoid.
According to the radiologic classiﬁcation of Hamada et
, there were 3 patients with grade II changes (upward
migration of the humeral head but no signiﬁcant acetabu-
larization), 6 with grade III changes (concavity of the acro-
mial undersurface with acetabularization), 10 with grade
IV changes (grade III changes plus narrowing of the gleno-
humeral joint), and 2 with grade V changes (humeral head
There were 5 patients in the FS group. The mean age
was 72 years (range, 66-79 years), 3 were women (80%),
and the dominant side was involved in 2 cases (40%). The
failed initial treatment was percutaneous pinning in 2 cases
and conservative treatment in 3. The 5 patients had a
painful and stiff shoulder. According to the classiﬁcation of
FS described by Boileau et al,
there were 3 type 4 se
quelae (severe malunion or nonunion of the tuberosities), 1
type 3 (surgical neck nonunion) with osteonecrosis of the
humeral head, and 1 type 1 (valgus impacted malunion)
with previous tuberculous arthritis and a massive cuff tear.
There were 19 patients in the revision group. The mean
age was 67 years (range, 50-87 years), and 14 were
women (70%). The dominant side was involved in 53%.
Three patients had persistent pain and pseudoparalysis of
the shoulder after previous surgery for cuff tear arthritis:
failed Bi-Polar hemiarthroplasties (Biomet, Warsaw, IN) in
two and failed arthrodesis (excessive abduction and inter-
nal rotation) in one. In the other 16 patients, shoulder
replacement performed for a displaced proximal humeral
fracture had failed. The reason for failure was a tuberosity
migration or nonunion in 14 cases and prosthesis malposition
with a cuff tear in 2. Revision surgery was performed as a
2-stage procedure in 4 of these patients, because they had a
concomitant deep infection.
Patient ﬁles were reviewed to determine preoperative
range of motion (ROM) and Constant score.
Constant score was calculated as a percentage of normal
reference values matched for age and sex.
computed tomography (CT) arthrograms and operation
notes were examined to determine the condition of the cuff
muscles and the glenoid bone stock. ROM and Constant
score were recorded at each follow-up visit, and the ﬁnal
examination was performed by 2 independent observers
(D.W. and A.M.H.). Patients were then asked to complete
the American Shoulder and Elbow Surgeons (ASES) shoul-
der questionnaire (which includes a visual analog scale for
pain) and to rate their level of satisfaction on a 4-point scale
(very satisﬁed, satisﬁed, no better and no worse, or worse).
Fluoroscopically controlled radiographs with a minimum
2-year follow-up were available for 42 patients. Radiologic
assessment included an anteroposterior view tangential to
the baseplate, a scapular lateral view, and an axillary
view. The radiographs were examined for notching at the
inferior margin of the scapular neck and for glenoid and
humeral radiolucent lines.
Humeral radiolucent lines were classiﬁed according to
width ("2 mm or !2 mm) and number of zones involved.
The humerus was divided into 7 zones around the prosthe-
sis as shown in Figure 1. Zones 1 and 7 were at the level of
the neck component, zones 2 and 6 were at the proximal
half of the stem component, and zones 3 and 5 were at the
distal half of the stem component.
Glenoid radiolucent lines were classiﬁed according to
their width ("2 mm or !2 mm), and the involvement of
individual zones was analyzed. Glenoid zones were de-
ﬁned as follows: 1, superior baseplate; 2, inferior base-
plate (independent from a notch if present); 3, central pillar;
and 4, screws.
Figure 1 Classiﬁcation of humeral zones: the humerus was di-
vided into 7 zones around the prosthesis.
530 Boileau et al J Shoulder Elbow Surg
Glenoid notching is a documented feature of the Delta 3
prosthesis and is thought to be due to impingement of the
humeral cup against the inferior scapular neck.
extent of the notch was classiﬁed in relation to the inferior
screw as shown in Figure 2.
Preoperative cuff status
A preoperative CT arthrogram was obtained to evaluate
the trophicity and fatty inﬁltration of the cuff muscles and to
assess the potential bone loss of the glenoid. A preoperative
CT arthrogram was available in 19 patients in the CTA
group, 13 in the revision group (although a metal artifact
limited the interpretation in 2 cases), and 3 in the FS group.
Fatty inﬁltration of the subscapularis, infraspinatus, and
teres minor was classiﬁed by use of a simpliﬁed version of
the system of Goutallier et al.
Grades 0, 1, and 2 were
grouped together ("50% fatty inﬁltration), as were grades
3 and 4 (!50% fatty inﬁltration).
The subscapularis was intact in 19 cases, partially torn
in 9, and completely torn in 11; fatty inﬁltration was less
than 50% in 16 cases and greater than 50% in 19. The
supraspinatus was intact in 3 cases and completely torn in
34; fatty inﬁltration was less than 50% in 6 cases and
greater than 50% in 27. The infraspinatus was intact in 4
cases, partially torn in 9, and completely torn in 24; fatty
inﬁltration was less than 50% in 8 cases and greater than
50% in 26. The teres minor was intact in 35 cases and torn
in 2; fatty inﬁltration was less than 50% in 23 cases and
greater than 50% in 12.
Statistical analysis was performed on preoperative and
postoperative data by use of the Student t test for continuous
test for nominal data.
In total, 14 complications occurred in 11 patients
(24%). These are summarized in Table I. Of these, 9
occurred in the revision group, 4 in the CTA group,
and 1 in the FS group.
There were 2 acromial fractures that appeared as
incidental ﬁndings on the 3-month postoperative ra-
diographs. Neither patient could recall any trauma;
both were completely asymptomatic, and there did
not appear to be any detrimental effect on function
(Constant scores of 69 and 77 points and active
elevation of 140° and 130°).
There was 1 axillary nerve palsy, representing the
only serious complication in the CTA group. After
3 years, recovery remained limited. Both ROM and
function were poor, with active elevation limited to
50° and a Constant score of 9 points. The nerve was
identiﬁed during the procedure and was intact. We
are unsure of the etiology of the palsy, but we suspect
that the nerve may have been stretched by the lower-
ing of the humerus. The patient declined any further
Of the patients, 10 (22%) needed further surgery:
4 reoperations without a humeral stem or glenoid
revision, 4 prosthetic revisions, and 2 prosthesis re-
movals. These are broken down per group in Table II.
A reoperation was deﬁned as surgery without re-
vision of the prosthesis. Exchange of the interface was
not considered as a revision of the prosthesis. Four
patients underwent reoperation with or without a
change of the interface components: one in the CTA
group and three in the revision group.
One patient underwent evacuation of a wound
hematoma after the procedure. She had no further
Three prosthetic dislocations occurred. Two of
them occurred around 1 month when the brace was
removed. They were both in the revision group and
were successfully treated by adding an extension to
the humeral neck component to increase the offset
and tension. The third dislocation occurred in the CTA
group 6 months after the operation. At reoperation, a
damaged polyethylene humeral cup was exchanged.
The patient underwent immobilization in neutral rota-
tion for 6 weeks and had no further dislocations.
These 4 patients were all included for analysis of the
results, because the initial prosthesis was still in place
at review and they had more than 2 years of follow-up
after the reoperation.
Prosthesis revision and removal
A revision was deﬁned as surgery with a change of
the prosthesis. There were no revisions or prosthesis
removals in the CTA group.
One prosthesis was revised in the FS group be-
cause of an intraoperative glenoid fracture. The gle-
noid fracture occurred during reaming of the glenoid.
Intraoperatively, the ﬁxation of the baseplate seemed
to be sufﬁcient, but the postoperative radiographs
showed malpositioning and migration of the glenoid
component. The patient underwent revision the next
day, and the reverse prosthesis was converted to a
hemiarthroplasty (Delta 1). At the last follow-up, the
result remained poor, with 40° of active elevation and
persistent shoulder pain.
Three deep infections occurred in the months after
Figure 2 Classiﬁcation of extent of notch: 0, no notch; 1, small
notch stopping short of inferior screw; 2, medium notch reaching
inferior screw; 3, large notch extending beyond inferior screw.
J Shoulder Elbow Surg Boileau et al 531
Volume 15, Number 5
the reverse prosthesis was implanted. All 3 occurred
in the revision group. The ﬁrst patient had a persis-
tently inﬂamed wound from the time of operation. In
the second, whose index surgery was a staged pro-
cedure because of previous infection, recurrent infec-
tion developed 12 months postoperatively, after a
good initial recovery. These 2 patients with overt
infection underwent resection arthroplasty at 16 and
18 months. After this, active elevation was 40° and
50°, respectively, and the Constant score was 13 and
22 points, respectively. In the third patient, septic
loosening of the humeral component developed, also
around the 12-month mark. This prosthesis was re-
vised in 1 stage to change the reverse prosthesis.
Two other patients, also in the revision group,
underwent revision. In 1 patient with a cemented
stem, early, aseptic humeral loosening developed
and revision was performed 1 year postoperatively.
Intraoperative cultures were negative. At 28 months,
the prosthesis remained well ﬁxed; active elevation
was 100°, and the Constant score was 34 points.
Another patient underwent revision for a comminuted
periprosthetic humeral fracture after a motorcycle ac-
cident. This patient underwent revision to a long-stem
reverse prosthesis with cerclage of the bone frag-
ments. The fracture healed, but the results have re-
mained poor, with a Constant score of 36 points and
active elevation of 70°.
From 1997 to 2002, the Grammont prosthesis was
implanted in 45 patients. Three patients required
revision surgery but retained their prostheses. Fol-
low-up for these patients was longer than 2 years, so
their results were included in the following sections.
Two patients eventually required resection arthro-
plasty, and one patient was converted to a Delta 1
prosthesis (see “Complications” section). These cases
were counted as clinical failures and have been ex-
cluded from further clinical or radiologic analysis,
leaving 42 patients with a mean follow-up of 40
months. There was no signiﬁcant difference in fol-
low-up between the CTA and revision groups. The FS
group was too small to allow any meaningful com-
parison to the other 2 groups.
Preoperative and most recent active ROM and
pain data are shown in Table III. There was a signif-
icant gain in active anterior elevation in all 3 groups
(P ! .01 in the FS group and P " .001 in the other 2
groups). There was a small increase in active external
rotation in the CTA and FS groups and a small de-
crease in the revision group, but none of these were
signiﬁcant, given the numbers available.
The proportion of patients with no or slight pain
(score on visual analog scale "3/10) was 81% in the
CTA group, 36% in the revision group, and 50% in
the FS group. The pain score was signiﬁcantly better
in the CTA group when compared with the revision
group (P ! .01).
The Constant score demonstrated a signiﬁcant in-
crease in all 3 groups (Table IV). However, the in-
crease was signiﬁcantly greater in the CTA group
when compared with the revision group (P ! .01).
The adjusted ASES shoulder scores and Constant
scores were signiﬁcantly higher in the CTA group as
compared with the revision group (P ! .004 and
.002, respectively) (Figures 3 and 4).
All but 4 of 45 patients believed that they had
beneﬁted from the operation. In the CTA group, 17
were very satisﬁed, 3 were satisﬁed, and 1 was
worse. The patient who stated that she was worse had
an axillary nerve palsy postoperatively that has not
Table I Complications and treatment
Complications (N ! 14) No. Treatment
Axillary nerve palsy 1 —
Late acromial fracture 2 —
Hematoma 1 Evacuation
Dislocation 3 Reoperated: cup extension in 2 and change of polyethylene cup in 1
Intraoperative glenoid fracture 1 Revised to hemiarthroplasty (Delta 1 prosthesis)
Deep infection 3 Revised: prosthesis removed in 2 and exchanged in 1
Aseptic humeral loosening 1 Revised to cemented long stem
Periprosthetic humeral fracture 2 (1 perioperative and 1 late traumatic) Immobilization in 1 and revised to long stem in 1
Table II Number of reoperations, prostheses revised, and removals
in the 3 groups and overall series
CTA (n ! 21) 1 — — 1 (5%)
FS (n ! 5) — 1 — 1 (20%)
(n ! 19) 3 3 2 8 (45%)
(N ! 45) 4 (9%) 4 (9%) 2 (4%) 10 (22%)
532 Boileau et al J Shoulder Elbow Surg
recovered. In the revision group, 11 were very satis-
ﬁed, 3 were satisﬁed, and 3 were no better and no
worse. In the FS group, 3 were very satisﬁed and 2
were no better and no worse.
Although there was no signiﬁcant difference in
external rotation among the groups, the preoperative
status of the teres minor did appear to be an impor-
tant factor. Mean active external rotation was 15° in
patients with less than 50% fatty inﬁltration, as op-
posed to 0° in those with more than 50% (P ! .02).
The mean Constant score was also signiﬁcantly better
in the former group (66 points vs 46 points, P !
We were unable to demonstrate any signiﬁcant
effect of the preoperative condition of the infraspina-
tus (tear or fatty inﬁltration) on postoperative external
rotation or function. However, only 4 patients had a
completely intact tendon, and only 7 had less than
50% fatty inﬁltration, so this ﬁnding should be inter-
preted with caution.
The belly-press test was positive in 22 patients
postoperatively, but neither this nor the preoperative
state of the subscapularis had any signiﬁcant effect on
either movement or functional scores.
After exclusion of the 4 cases of failure, radiologic
analysis was possible for 38 patients.
Glenoid components. With regard to glenoid com-
ponents, a radiolucent line was seen in zone 1 (supe-
rior part of the baseplate) in 17 cases (45%), but only
Table III Active ROM (preoperatively and at latest follow-up) for the 3 groups and overall series
Anterior elevation (°)
[mean (95% CI)]
External rotation (°)
[mean (95% CI)] Internal rotation
scale)Preoperative Follow-up Preoperative Follow-up Preoperative Follow-up
CTA (n ! 21) 53 (41 to 65) 123 (108 to 139) 9 (1 to 16) 14 (7 to 21) S1 L3 1.7/10 (0.4 to 2.9)
FS (n ! 4) 56 (44 to 68) 122 (96 to 148) #2(#12 to 8) 9 (#10 to 28) GT D12 2.6/10 (0.4 to 4.8)
Revision (n ! 17) 56 (44 to 68) 113 (100 to 126) 8 (#2 to 19) 1 (#6 to 7) S1 L5 4.5/10 (3.1 to 6.0)
Overall series (N ! 42) 55 (47 to 63) 121 (111 to 131) 7 (1 to 13) 11 (5 to 16) S1 S1 3.2/10 (1.6 to 4.8)
CI, Conﬁdence interval.
Table IV Constant score for the 3 groups and overall series preoperatively and at latest follow-up
Gain P valuePreoperative Follow-up
CTA (n ! 21) 18 (14-22) 66 (58-74) 49 (41-56) P " .001
FS (n ! 4) 15 (9-21) 61 (44-78) 42 (26-59) P ! .008
Revision (n ! 17) 15 (11-19) 46 (37-55) 32 (22-42) P " .001
Overall series (N ! 42) 17 (14-19) 58 (51-64) 41 (35-47) P " .001
CI, Conﬁdence interval.
Figure 3 Mean ASES shoulder score by group with 95% conﬁ-
Figure 4 Mean adjusted Constant score by group with 95%
J Shoulder Elbow Surg Boileau et al 533
Volume 15, Number 5
one of these exceeded 2 mm and none appeared to
be progressive. No radiolucent lines were seen in any
Scapular notching. Scapular notching was seen in
26 cases (68%). Of these, 18 were grade 1, 5 were
grade 2, and 5 were grade 3. There was notching
extending to or beyond the inferior screw in 26% of
cases. However, neither the presence nor the size of
the notch had a negative effect on the Constant score,
the adjusted Constant score, or the ASES score.
Humeral components. With regard to humeral
components, radiolucent lines less than 2 mm wide
were seen in 23 cases (60%); these involved 1 zone
in 10 cases, 2 zones in 8, 3 zones in 3, 4 zones in 1,
and 5 zones in 1. Radiolucent lines greater than 2
mm were seen in 6 cases (16%); 5 were progressive.
One zone was involved in 2 cases, and two zones
were involved in 4.
Heterotopic ossiﬁcation. Heterotopic ossiﬁcation
was seen in 17 cases (45%). It occurred at the lower
margin of the glenoid and was usually minor; how-
ever, we also noticed a bony spur at the inferior
margin of the scapular neck in 24 cases (63%). This
always occurred in association with a notch and often
tended to make the notch appear deeper than it really
was. The spur was always located at the medial
border of the notch, adjacent to the point of impinge-
ment between the humeral cup and the scapular neck
Constrained prostheses, both anatomic and re-
verse ball-and-socket, have a reputation for early
failure.* The Grammont prosthesis is not fully con-
strained, but it does have congruent joint surfaces,
making it a semiconstrained prosthesis.
mont initially designed the Delta prosthesis for pa-
tients with arthritis due to a massive cuff tear (ie, “cuff
tear arthritis”) (Figure 6).
Another potential indica-
tion for a reverse prosthesis is FS, where distortion of
the proximal humeral anatomy around the cuff inser-
tion results in a functionally useless cuff (Figure 7).
Results of conventional, unconstrained arthroplasty in
these cases are often unpredictable because a
greater tuberosity osteotomy and reﬁxation (needed
to implant a conventional arthroplasty) often lead to
unreliable functional healing.
The third indication for
reverse arthroplasty in our series was revision of a
previous arthroplasty, where the cuff is deﬁcient,
scarred, or retracted or has undergone fatty inﬁltra-
tion (Figure 8).
In this study, we report the midterm
results and complications of the Delta reverse prosthe-
sis for these 3 etiologies. This study represents one of
the largest Delta series to date, and it is the ﬁrst to
analyze the different underlying pathologies sepa-
rately. Other strengths include examination by inde-
pendent observers, minimal loss to follow-up (despite
the advanced age of many patients), and detailed
radiologic analysis of the state of both the cuff preop-
eratively and the prosthesis postoperatively.
Our study shows that the Delta reverse prosthesis
can improve function and restore active elevation
beyond the horizontal level in patients with severely
cuff-deﬁcient shoulders (Figure 6). Active rotation,
however, is not improved. Thus far, the few published
reports on the Delta are mostly small series, limited to
a single pathology, usually CTA (Table V). Compari-
sons should be made cautiously, but the ﬁndings do
appear to be consistent with ours. The mean values
for the Constant score and active elevation in the 6
previous CTA series are all within the limits of our
95% conﬁdence intervals for this group. In our expe-
rience, this prosthesis yields better results than tradi-
tional unconstrained shoulder arthroplasty in patients
with CTA, but longer follow-up is required to deter-
mine the results’ durability. Our reoperation rate of
5% for the CTA group also compares favorably with
As one might expect, the functional results are not
as good when the Grammont prosthesis is used in
revision surgery, mainly because of the high rate of
complications, reoperations, or revisions (Tables I
and II). Both the Constant score and the ASES score
were signiﬁcantly lower in the revision group as com-
pared with the CTA group, but there was no signiﬁ-
cant difference in active elevation in those who had
no complications or revision (Figures 3 and 4). The
vast majority of the complications in the series oc-
curred in the revision group, and this was reﬂected in
may not have been as good as in the CTA group, those
patients who did not require further revision had a
References 4, 8, 11, 29, 30, 31, 33, 39, 45, 51.
Figure 5 Notching (bone lysis) of scapular neck with formation of
bony spur (arrow).
534 Boileau et al J Shoulder Elbow Surg
signiﬁcant increase in function and active elevation after
the operation; furthermore, 13 of 16 believed that they
had beneﬁted from the procedure (Figure 8). Schnee-
berger and Gerber
have reported a similar experi
ence with Grammont replacement for revision cases,
with a high complication and revision rate of 50%.
Figure 6 Results of reverse prosthesis for CTA. A, Preoperative anteroposterior view. B, Placement of Grammont
reverse prosthesis. The patient had complete active elevation (C), but external rotation (D) and internal rotation (E)
were not improved.
Figure 7 Results of reverse prosthesis for FS. A, Preoperative anteroposterior radiograph showing CTA. B,
Placement of Grammont reverse prosthesis. The patient had active elevation to about 120° (C). External rotation
was not improved (D), whereas internal rotation was satisfactory (E).
J Shoulder Elbow Surg Boileau et al 535
Volume 15, Number 5
Figure 8 Revision of failed shoulder replacement (Bi-Polar) for fracture with infection. A, Tuberosity migration and
infection. B, First step (prosthesis removal and spacer with antibiotics). C, Second step (placement of reverse
prosthesis, which was complicated by dislocation after 2 months). D, Re-revision (addition of neck extension). The
patient had active elevation to about 90° (E), despite an almost absent anterior deltoid because of the previous
operation (F). External rotation (G) and internal rotation (H) remained limited.
536 Boileau et al J Shoulder Elbow Surg
Although the Grammont prosthesis is able to re-
store active elevation, active external rotation remains
an unsolved problem (Figure 6). A potential explana-
tion for the weak external rotation after Grammont
reverse prosthesis implantation is related to the medi-
alization of the center of rotation and of the humerus.
Because of this medialization, the amount of posterior
deltoid that can be used to compensate for the absent
external rotators is decreased.
The posterior deltoid
does, theoretically, provide some external rotation
power but only coupled with abduction.
the retroversion of the humeral component should
improve external rotation, in theory, but only at the
expense of internal rotation. In addition, the design of
the prosthesis itself limits rotation: the humeral cup is
limited in its rotation around the glenosphere and can
impinge against the posterior neck of the scapula
when the arm is at the side. Like other investigators,
we have found that absence of severe fatty inﬁltration
of the teres minor does confer superior external rota-
tion, but even when this was the case, it still averaged
Absence of active external rotation
may be obvious clinically because of a persistent
Moreover, we found that severe
fatty inﬁltration of the teres minor was associated with
lower functional results (Constant score of 46 points
vs 66 points, P " .007).
A positive belly-press test was found in 22 patients
postoperatively. Of these, 10 had a subscapularis
tear preoperatively and 2 had grade 3 fatty inﬁltra-
tion of the subscapularis. The other 10, however, had
an intact subscapularis muscle preoperatively, sug-
gesting either failure of the subscapularis to heal or a
limit inherent in the prosthetic ROM (because of the
medialization). The subscapularis may be prone to
failure of healing, as the Grammont prosthesis dis-
places the humerus inferiorly, relative to the scapula,
theoretically increasing the tension on the subscapu-
laris tendon repair. This can be a signiﬁcant compli-
cation in unconstrained shoulder arthroplasties, but
we have not shown any signiﬁcant effect on either
ROM or functional results with the Grammont.
The glenoid component is usually the main site of
loosening in total shoulder arthroplasty, particularly
with a constrained design.
There was only 1
case of early glenoid loosening in our series (the
week after implantation), and it was associated with
an intraoperative glenoid fracture, which compro-
mised implant ﬁxation. Although a longer follow-up is
mandatory, glenoid loosening has not been encoun-
tered thus far. This conﬁrms the value of Grammont’s
concept of medialization of the center of rota-
A thin radiolucent line was seen in the
superior part of the baseplate (zone 1) in 44% of
glenoids, but none of these progressed, and this
probably represents incomplete bony in-growth, a
zone of ﬁbrous tissue. In unconstrained total shoulder
arthroplasties, such radiolucent lines have been re-
ported in up to 80% of cemented glenoid compo-
We encountered more difﬁculties with hu
meral loosening, although most of these cases were
associated with low-grade infection.
Most of the complications observed are common to
any shoulder prosthesis.
The Delta 3 prosthesis,
however, seems prone to 3 speciﬁc problems related
to its design: acromial fracture, dislocation, and scap-
To compensate for the deﬁcient cuff,
adequate tension must be restored to the deltoid.
This is conﬁrmed by a tight reduction and palpation of
a taut conjoined tendon. The bone in these elderly
patients is often osteoporotic, and excessive deltoid
tension can result in acromial fracture (Figure 9).
Fortunately, the 2 acromial fractures that we observed
were both asymptomatic fatigue fractures, which did
not compromise active elevation or abduction. At the
other end of the spectrum, insufﬁcient tension in the
Table V Published series for Grammont prosthesis
Grammont et al (1996)
16 CTA 27 NA 14/69 13%
De Buttet et al (1997)
71 CTA 24 NA/120 19.4/59.9 4.2%
De Wilde et al (2001)
5 Revision 30 “Fair” 14/62 20%
Rittmeister et al (2001)
8 RA 54 NA 17/63 37.5%
Jacobs et al (2001)
7 CTA 16 NA 17.9/56.7 0%
Sirveaux et al (2001)
80 CTA 44 73/138 22.6/65.6 5%
Valenti et al (2001)
39 CTA 84 60/120 21/63 15%
Boulahia et al (2002)
16 CTA and FS 35 70/138 31/59 12.5%
Delloye et al (2002)
5 Revision 81 NA/72 NA/40 60%
De Wilde et al (2002)
6 Tumors 12 NA/106 NA 0%
Our series (2005) 45 CTA, FS, and Revision 40 55/121 17/58 22%
NA, Not available; RA, rheumatoid arthritis.
J Shoulder Elbow Surg Boileau et al 537
Volume 15, Number 5
deltoid can result in prosthetic instability (Figure 8, C).
Restoring deltoid tension may be difﬁcult because the
proximal epiphysis is often missing, especially after a
failed prosthetic replacement for fracture with tuber-
osity migration and lysis. For these cases, we have
learned to cement the prostheses proud from the
remaining bone, so tension on the deltoid can be
restored within available neck-lengthening options
The scapular notch is another cause for concern,
although very little has been published about it in
Two fundamental design prin
ciples of the Grammont prosthesis are (1) the absence
of a neck on the reverse glenoid component, minimiz-
ing torque at the prosthesis-bone interface, and (2) a
nonanatomic inclination of 155° of the humeral im-
plant. These design aspects, crucial to its excellent
function, are also responsible for impingement of the
humeral cup on the inferior margin of the scapular
The bony spur that we observed at the medial
border of the notch (present in 68% of cases) may be
a reactive process analogous to osteophyte formation
in osteoarthritis (Figure 5). Although impingement
may explain grade 1 or even grade 2 notches, it
seems unlikely to be the direct cause of a notch
extending beyond the inferior screw. A possible ex-
planation for such large notches is osteolysis induced
by polyethylene debris released by the impingement.
This phenomenon would also likely result in prosthetic
loosening in the long term, as suggested recently by
Nyffeler et al.
Nonetheless, we have not observed
glenoid osteolysis with up to 7 years’ follow-up, ex-
cept in the region of the notch. In a longer-term report
on the Grammont prosthesis, Valenti et al
observed no progression of notches to glenoid loos-
ening at a mean of 7 years postoperatively. To mini-
mize notching, our current practice is to insert the
glenoid component as low as possible.
chanical studies and reﬁnement in prosthetic tech-
nique and design hopefully will decrease the rate of
Finally, low-grade, indolent infection is another
serious complication that is associated with revision
surgery. It occurred in 3 of our patients and resulted in
resection arthroplasty in 2 of them. When revising a
failed prosthesis, surgeons must be aware that low-
grade infection can be present without overt clinical
symptoms or signs.
Our preoperative evaluation of
these patients always includes assessment of the com-
plete blood count with differential, erythrocyte sedi-
mentation rate, and C-reactive protein level, as well
as a 3-phase bone scan. Intraoperatively, specimens
are sent for culture and frozen section analysis. If an
infection is discovered or highly suspected, our cur-
rent practice is to perform a 2-step procedure. The ﬁrst
operation entails removal of the prosthesis, aggres-
sive irrigation and debridement, and placement of an
antibiotic cement spacer. This is followed by implan-
tation of a reverse prosthesis in 6 weeks, only if
laboratory tests and intraoperative frozen sections
suggest that the infection has been cleared. Four
patients had a 2-stage procedure to revise an infected
conventional arthroplasty to a reverse arthroplasty;
none of them had a recurrent infection at the latest
In summary, the reverse prosthesis can improve
function and restore active elevation in patients with
incongruent cuff-deﬁcient shoulders. Improvement in
active rotation, however, does not usually occur. At
midterm follow-up, glenoid loosening has not oc-
curred. Our clinical results of the Grammont reverse
prosthesis differentiate it from the reverse ball-and-
socket designs of the past and live up to the biome-
chanical concept of Paul Grammont. These results
are, furthermore, an improvement on Neer’s limited
goals in patients with cuff-deﬁcient arthritis. Our study
also shows that this prosthesis offers a solution for
other difﬁcult clinical situations: failure of previous
prostheses and severe FS with an absent or nonfunc-
tional cuff. However, results are clearly less predict-
able and complication and revision rates are higher
in those patients as compared with CTA patients.
These clinical results support the use of this prosthesis
Figure 9 Acromial “fatigue” fracture related to overtensioning of
deltoid in a patient with osteoporotic bone.
538 Boileau et al J Shoulder Elbow Surg
for patients with severe pain from signiﬁcant unrepair-
able cuff deﬁciency, arthritis, and a lack of a stable
fulcrum for shoulder elevation. In revision surgery, we
are very wary of occult infection, even in the absence
of obvious clinical symptoms or signs. Caution is
required, as such patients are often younger, and
informed consent must obviously cover the high com-
plication rate in this group, as well as the unknown
longer-term outcome. On the basis of the current
design and results, the reverse prosthesis should be
considered a salvage procedure: its use should be
limited to elderly patients, arguably those aged over
70 years, with poor function and severe pain related
to cuff deﬁciency. A reverse prosthesis should not be
offered to a young individual who desires to have a
normal shoulder and will demand more of the pros-
thesis than it is designed to do.
Results of the reverse
prosthesis depend on the diagnosis and the remain-
ing cuff muscles, speciﬁcally the teres minor.
1. Arntz CT, Jackins S, Matsen FA III. Prosthetic replacement of the
shoulder for the treatment of defects in the rotator cuff and the
surface of the glenohumeral joint. J Bone Joint Surg Am 1993;
2. Barrett WP, Franklin JL, Jackins SE, Wyss CR, Matsen FA. Total
shoulder arthroplasty. J Bone Joint Surg Am 1987;69:865-72.
3. Baulot E, Chabernaud D, Grammont P. Results of Grammont’s
inverted prosthesis in omarthritis associated with major cuff de-
struction. Apropos of 16 cases [in French]. Acta Orthop Belg
4. Bayley JIL, Kessel L. The Kessel total shoulder replacement. In:
Bayley J, Kessel L, editors. Shoulder surgery. Berlin: Springer;
5. Boileau P, Trojani C, Walch G, Krishnan SG, Romeo A, Sinner-
ton R. Shoulder arthroplasty for the treatment of the sequelae of
fractures of the proximal humerus. J Shoulder Elbow Surg 2001;
6. Boileau P, Watkinson DJ, Hatzidakis AM, Balg F. Grammont
reverse prosthesis: design, rationale, and biomechanics. J Shoul-
der Elbow Surg 2005;14(Suppl S):147S-61S.
7. Boulahia A, Edwards TB, Walch G, Baratta RV. Early results of a
reverse design prosthesis in the treatment of arthritis of the shoul-
der in elderly patients with a large rotator cuff tear. Orthopedics
8. Broström LÅ, Wallenstein R, Olsson E, Anderson D. The Kessel
prosthesis in total shoulder arthroplasty. A ﬁve-year experience.
Clin Orthop Relat Res 1992:155-60.
9. Coﬁeld RH. Total shoulder arthroplasty with the Neer prosthesis.
J Bone Joint Surg Am 1984;66:899-906.
10. Constant CR, Murley AHG. A clinical method of functional
assessment of the shoulder. Clin Orthop Relat Res 1987:160-4.
11. Coughlin MJ, Morris JM, West WF. The semiconstrained total
shoulder arthroplasty. J Bone Joint Surg Am 1979;61:574-81.
12. Coste JS, Reig S, Trojani C, Berg M, Walch G, Boileau P. The
management of infection in arthroplasty of the shoulder. J Bone
Joint Surg Br 2004;86:65-9.
13. De Buttet A, Bouchon Y, Capon D, Delfosse J. Grammont shoul-
der arthroplasty for osteoarthritis with massive rotator cuff tears—
report of 71 cases [abstract]. J Shoulder Elbow Surg 1997;6:
14. De Wilde L, Mombert M, Van Petegem P, Verdonk R. Revision of
shoulder replacement with a reversed shoulder prosthesis (Delta
III). Report of ﬁve cases. Acta Orthop Belg 2001;67:348-53.
15. De Wilde LF, Van Ovost E, Uyttendaele D, Verdonk R. Results of
an inverted shoulder prosthesis after resection for tumor of the
proximal humerus [in French]. Rev Chir Orthop Reparatrice Appar
16. Delloye C, Joris D, Colette A, Eudier A, Dubuc J. Mechanical
complications of total shoulder inverted prosthesis [in French]. Rev
Chir Orthop Reparatrice Appar Mot 2002;88:410-4.
17. Edwards TB, Boulahia A, Kempf JF, Boileau P, Nemoz C, Walch
G. The inﬂuence of the rotator cuff on the results of shoulder
arthroplasty for primary osteoarthritis: results of a multicenter
study. J Bone Joint Surg Am 2002;84:2240-8.
18. Favard L, Lautmann S, Sirveaux F, Oudet D, Kerjean Y, Huguet D.
Hemi arthroplasty versus reverse arthroplasty in the treatment of
osteoarthritis with massive rotator cuff tear. In: Walch G, Boileau
P, Molé D, editors. 2000 Shoulder prosthesis . . . two to ten year
follow-up. Montpellier, France: Sauramps Medical; 2001. p.
19. Fenlin JM. Total glenohumeral joint replacement. Orthop Clin
North Am 1975;6:565-83.
20. Field LD, Dines DM, Zabinski SJ, Warren RF. Hemiarthroplasty of
the shoulder for rotator cuff arthropathy. J Shoulder Elbow Surg
21. Franklin JL, Barrett WP, Jackins SE, Matsen FA III. Glenoid loos-
ening in total shoulder arthroplasty. Association with rotator cuff
deﬁciency. J Arthroplasty 1988;3:39-46.
22. Goutallier D, Postel JM, Bernageau J, Lavau L, Voisin MC. Fatty
muscle degeneration in cuff ruptures. Pre- and postoperative
evaluation by CT scan. Clin Orthop Relat Res 1994:78-83.
23. Grammont P, Trouilloud P, Laffay JP, Deries X. Etude et réalisation
d’une nouvelle prothèse d’épaule. Rhumatologie 1987;39:17-22.
24. Grammont PM, Baulot E. Delta shoulder prosthesis for rotator cuff
rupture. Orthopaedics 1993;16:65-8.
25. Grammont PM, Baulot E, Chabernaud D. Résultats des 16
premiers cas d’arthroplastie totale d’épaule inversée sans ciment
pour des omarthroses avec grande rupture de coiffe [abstract].
Rev Chir Orthop Reparatrice Appar Mot 1996;82(Suppl I):169.
26. Hadzidakis A, Norris T, Boileau P. Reverse shoulder arthroplasty
indications, technique, and results. Tech Shoulder Elbow Surg
27. Hamada K, Fukuda H, Mikasa M, Kobayashi Y. Roentgeno-
graphic ﬁndings in massive rotator cuff tears. A long-term obser-
vation. Clin Orthop Relat Res 1990:92-6.
28. Jacobs R, Debeer P, De Smet L. Treatment of rotator cuff arthrop-
athy with a reversed Delta shoulder prosthesis. Acta Orthop Belg
29. Kölbel R, Friedebold G. Stabilization of shoulders with bone and
muscle defects using joint replacement implants. In: Bateman J,
Welsh P, editors. Shoulder surgery. At Louis: Mosby. p. 281-7.
30. Lee DH, Niemann KM. Bipolar shoulder arthroplasty. Clin Orthop
Relat Res 1994:97-107.
31. Lettin A, Copeland S, Scales J. The Stanmore total shoulder
replacement. J Bone Joint Surg Br 1982;64:47-51.
32. Levine WN, Djurasovic M, Glasson JM, Pollock RG, Flatow EL,
Bigliani LU. Hemiarthroplasty for glenohumeral osteoarthritis: re-
sults correlated to degree of glenoid wear. J Shoulder Elbow Surg
33. McElwain J, English E. The early results of porous coated total
shoulder arthroplasty. Clin Orthop Relat Res 1987:217-24.
34. Neer CS, Watson KC, Stanton FJ. Recent experience in total
shoulder replacement. J Bone Joint Surg Am 1982;64:319-37.
35. Nyffeler RW, Werner CM, Simmen BR, Gerber C. Analysis of a
retrieved Delta III total shoulder prosthesis. J Bone Joint Surg Br
36. Nyffeler RW, Werner CM, Gerber C. Biomechanical relevance
of glenoid component positioning in the reverse Delta III total
shoulder prosthesis. J Shoulder Elbow Surg 2005;14:524-8.
37. Pollock RG, Deliz ED, McIlveen SJ, Flatow EL, Bigliani LU. Pros-
thetic replacement in rotator cuff deﬁcient shoulders. J Shoulder
Elbow Surg 1992;1:173-86.
J Shoulder Elbow Surg Boileau et al 539
Volume 15, Number 5
38. Poppen NK, Walker PS. Forces at the glenohumeral joint in
abduction. Clin Orthop Relat Res 1978:165-70.
39. Post M, Haskell SS, Jablon M. Total shoulder replacement with a
constrained prosthesis. J Bone Joint Surg Am 1980;62:327-35.
40. Rittmeister M, Kerschbaumer F. Grammont reverse total shoulder
arthroplasty in patients with rheumatoid arthritis and nonrecon-
structible rotator cuff lesions. J Shoulder Elbow Surg 2001;10:
41. Sanchez-Sotelo J, O’Driscoll SW, Torchia ME, Coﬁeld RH, Row-
land CM. Radiographic assessment of cemented humeral com-
ponents in shoulder arthroplasty. J Shoulder Elbow Surg 2001;
42. Sanchez-Sotelo J, Coﬁeld RH, Rowland CM. Shoulder hemiar-
throplasty for glenohumeral arthritis associated with severe rotator
cuff deﬁciency. J Bone Joint Surg Am 2001;83:1814-22.
43. Schneeberger A, Gerber C. Complications with the Delta re-
versed prosthesis [abstract]. Presented at the ASES Closed Meet-
44. Sirveaux F, Favard L, Oudet D, Huguet D, Lautman S. Grammont
inverted total shoulder arthroplasty in the treatment of glenohu-
meral osteoarthritis with massive and non repairable cuff rupture.
In: Walch G, Boileau P, Molé D, editors. 2000 Shoulder pros-
thesis . . . two to ten year follow-up. Montpellier: Sauramps
Medical; 2001. p. 247-52.
45. Swanson AB, de Groot Swanson G, Sattel AB, Cendo RD, Hynes
D, Jar-Ning W. Bipolar implant shoulder arthroplasty. Long-term
results. Clin Orthop Relat Res 1989:227-47.
46. Torchia ME, Coﬁeld RH, Settergren CR. Total shoulder arthro-
plasty with the Neer prosthesis: long-term results. J Shoulder
Elbow Surg 1997;6:495-505.
47. Valenti PH, Boutens D, Nerot C. Delta 3 reversed prosthesis for
arthritis with massive rotator cuff tear: long term results ($ 5
years). In: Walch G, Boileau P, Molé D, editors. 2000 Shoulder
prosthesis . . . two to ten year follow-up. Montpellier: Sauramps
Medical; 2001. p. 253-9.
48. Walch G, Madonia G, Pozzi I, Riand N. Lévigne C. Arthro-
scopic tenotomy of the tendon of the long head of the biceps in
rotator cuff ruptures. In: Gazielly D, Gleyze P, Thomas T, editors.
The cuff. Amsterdam: Elsevier; 1997. p. 350-5.
49. Walch G, Boulahia A, Calderone S, Robinson AHN. The drop-
ping and hornblower’s signs in evaluation of rotator cuff tears.
J Bone Joint Surg Br 1998;80:624-8.
50. Williams GR Jr, Rockwood CA Jr. Hemiarthroplasty in rotator
cuff-deﬁcient shoulders. J Shoulder Elbow Surg 1996;5:362-7.
51. Wirth MA, Rockwood CA Jr. Current concepts review. Compli-
cations of total shoulder arthroplasty. J Bone Joint Surg Am
52. Worland RL, Jessup DE, Arredondo J, Warburton KJ. Bipolar
shoulder arthroplasty for rotator cuff arthropathy. J Shoulder Elbow
540 Boileau et al J Shoulder Elbow Surg