Abstract
Background: The reverse Delta III shoulder prosthesis can relieve
pain and restore function in patients with cuff tear arthropathy. The most
frequently reported radiographic complication is inferior scapular notching.
The purpose of the present study was to evaluate the clinical relevance of
notching and to determine the anatomic and radiographic parameters that
predispose to its occurrence.
Methods: Seventy-seven consecutive shoulders in seventy-six patients
with an irreparable rotator cuff deficiency were managed with a reverse Delta
III shoulder arthroplasty and were followed clinically and radiographically
for a minimum of twenty-four months. The effects of cranial-caudal glenoid
component positioning and the prosthesis-scapular neck angle on the
development of inferior scapular notching and clinical outcome were
assessed.
Results: All shoulders that had development of notching did so in
the first fourteen months. Of the seventy-seven shoulders that were studied,
thirty-four (44%) had inferior scapular notching, twenty-three (30%) had
posterior notching, and six (8%) had anterior notching. Osteophytes along the
inferior part of the scapula occurred in twenty-one (27%) of the seventy-seven
shoulders. The angle between the glenosphere and the scapular neck (r = 0.667)
as well as the craniocaudal position of the glenosphere (r = 0.654) were
highly correlated with inferior notching (p < 0.001). A notching index was
calculated with use of the height of implantation of the glenosphere and the
postoperative prosthesis-scapular neck angle. This allowed prediction of the
occurrence of notching with a sensitivity of 91% and specificity of 88%. The
height of implantation of the glenosphere had approximately an eight times
greater influence on inferior notching than the prosthesis-scapular neck angle
did. Inferior scapular notching was associated with a significantly poorer
clinical outcome.
Conclusions: Inferior scapular notching after reverse total shoulder
arthroplasty adversely affects the intermediate-term clinical outcome. It can
be prevented by optimal positioning of the glenoid component.
Level of Evidence: Prognostic Level II. See Instructions
to Authors for a complete description of levels of evidence.
Early attempts to treat glenohumeral arthritis associated with advanced
rotator cuff disease were not proven to be clinically
satisfactory1,2.
Conventional total shoulder arthroplasty was associated with early loosening
and subsequent failure of the glenoid component due to the rocking horse
mechanism3,4.
In an attempt to improve function and prosthetic longevity, constrained and
semiconstrained prostheses were introduced, but instability, unsatisfactory
active mobility, and early glenoid component
loosening5,6
led surgeons to abandon constrained prostheses and to return to
hemiarthroplasty, which, however, did not yield the desired functional
results3,7,8.
In 1985, the Delta III reverse prosthesis was introduced by
Grammont9. This
prosthesis features a semiconstrained inverted design with a fixed, lowered,
and medialized center of rotation. By medializing and lowering the center of
rotation, the reverse prosthesis decreases shear forces at the glenoid
component-bone interface, improves the lever arm of the deltoid, and results
in improved survival compared with other constrained and semiconstrained
designs9-12.
While the medialized center of rotation decreases shear forces and improves
the deltoid lever arm, it also tends to result in mechanical impingement
between the superomedial aspect of the humeral polyethylene insert and the
scapular neck, resulting in so-called inferior scapular notching. To date, the
clinical relevance of scapular notching has been disputed, although it has
been associated with a poorer clinical outcome, polyethylene wear, chronic
inflammation of the joint capsule, and local
osteolysis13.
Furthermore, there is concern that scapular notching could be progressive and
could lead to late glenoid loosening and catastrophic glenoid component
failure.
Nyffeler et
al.14 conducted a
cadaveric biomechanical study in which different glenosphere positions were
examined to determine which position yielded the least mechanical conflict
during adduction. Inferior overhang of the glenosphere beyond the glenoid rim
resulted in the greatest adduction before impingement of the humeral
prosthesis on the inferior scapular neck occurred. This concept suggests that
there would be less inferior notching with the glenosphere in a more inferior
position. The goals of the present investigation were to determine the
clinical relevance of inferior glenoid notching and to verify the hypothesis
that inferior notching is related to the exact position of the glenosphere on
the scapula.
Patient Selection
Between 1995 and 2003, 186 reverse Delta III prostheses were implanted by a
single surgeon (C.G.). For the purpose of the present study, Delta III
arthroplasties that were performed for the treatment of painful cuff tear
arthropathy or irreparable rotator cuff tears with pseudoparesis (defined as
the inability to actively elevate the shoulder >90° in the presence of
free passive anterior elevation) were included. All reverse prostheses that
were implanted for the revision of a previous arthroplasty or for the
treatment of acute fracture, posttraumatic deformity, or posttraumatic
arthritis were excluded. In addition, all patients who had a previous or
concurrent glenoid fracture or glenoid bone-grafting were excluded. These
criteria yielded a consecutive series of 107 shoulders in 105 patients.
Patients with less than two years of complete, strictly standardized
radiographic and clinical follow-up were also excluded. Patients with any
complication that resulted in a resection arthroplasty or revision to a second
prosthesis prior to the twenty-four months were also excluded from the study.
This left seventy-seven shoulders in seventy-six patients available for the
final evaluation.
The study group included twenty-seven men and forty-nine women with an
average age of seventy-one years (range, fifty-four to eighty-five years). All
patients were reviewed clinically and radiographically after an average
duration of follow-up of forty-four months (range, twenty-four to ninety-six
months). Fifty-six operations involved the dominant extremity. Nineteen
shoulders had painful cuff tear arthropathy, and fifty-eight shoulders had an
irreparable rotator cuff tear with pseudoparesis. Thirty-five shoulders had
undergone previous procedures, including twenty-eight rotator cuff repairs,
six biceps tenodeses or tenotomies, ten acromioplasties, one arthroscopic
rotator cuff débridement, one Weber osteotomy, two internal fixation
procedures for the treatment of an unstable os acromiale, one bone allograft
procedure for the treatment of a reverse Hill-Sachs lesion, and one latissimus
dorsi tendon transfer.
Surgical Technique
All procedures were performed by or under the direct supervision of the
senior author (C.G.). A variable combination of regional and general
anesthesia was used in all cases. The procedures were performed with the
patient in the beach-chair position and with use of a deltopectoral approach
without detachment of the deltoid origin. In the present report, the terms
metaglene and glenosphere refer to the trademark names of the glenoid
baseplate and the glenoid hemisphere, respectively, of the Delta III reverse
shoulder system (DePuy Orthopaedics, Warsaw, Indiana). The subscapularis, if
intact, was taken down from the lesser tuberosity, and the axillary nerve was
protected. If the long biceps tendon was intact, it was tenotomized at its
origin. All osteophytes were removed from the humeral head-neck junction. With
the assistance of the humeral resection guide, an oscillating saw was used to
resect the humeral head in a targeted retroversion of between 0° and
20°. Soft tissue was then released from around the entire 360° of the
glenoid. The triceps origin was released along the scapular neck. Releasing
the triceps allowed the precise location of the inferior glenoid rim and the
lateral scapular border to be identified, which aids in choosing the correct
orientation of the metaglene and the inferior screw. In addition, release of
the long head of the triceps eliminates the impingement of the humeral
prosthesis against this soft-tissue component and facilitates adduction of the
humerus. The glenoid-centering drill-guide was then aligned approximately in
the center of the glenoid from anterior to posterior and approximately 2 to 3
mm inferior from the center of the glenoid. Given the findings of a previous
study by Nyffeler et
al.14, the glenoid
baseplate (metaglene) was placed more distally over the course of the present
study beginning in 2002. The glenoid was then reamed with a flat reamer to
remove articular cartilage and sclerotic bone. The metaglene was implanted in
a standard fashion with four 4.5-mm screws ranging in length from 18 to 48 mm,
depending on measurement with a depth gauge. A 36-mm glenosphere was inserted
in seventy-six shoulders, and a 42-mm glenosphere was inserted in one
shoulder. The humeral stem was implanted with gentamicin-impregnated cement
(Palacos; Essex Chemie, Lucerne, Switzerland) in all shoulders.
Antibiotic-impregnated cement is utilized routinely at our institution to
comply with local regulations. We do not have evidence that it reduces the
infection rate for this particular procedure, but we are also not aware of any
evidence that antibiotic-impregnated cement is inferior in terms of long-term
clinical performance. A standard lateralized humeral polyethylene cup was used
in all cases. All incisions were closed over one or two suction drains.
Postoperative Rehabilitation
All patients were managed with a simple sling postoperatively, with the arm
at the side and the shoulder internally rotated. Immediate passive and active
mobilization was begun in a therapeutic pool under the direction of physical
therapists after the suction drains were removed at twenty-four to seventy-two
hours postoperatively. Passive and active-assisted exercises were continued
with a gradual progression to independent activities of daily living at six
weeks. The arm was protected in a sling for a total of six weeks after
surgery.
Radiographic Evaluation
Radiographic assessment, including the evaluation of true anteroposterior
and axillary lateral radiographs, was performed under fluoroscopic control
preoperatively and postoperatively at three months, six months, one year, and
annually thereafter, coincident with each clinical evaluation. For reasons of
compliance or scheduling, the one-year follow-up examination was carried out
between twelve and fourteen months after the operation. The final radiographic
follow-up examination for the patients in the present study was performed at a
minimum of twenty-four months (mean, forty-four months; range, twenty-four to
ninety-six months). In order to standardize the scapular anteroposterior
radiograph, the scapula and the central x-ray beam were positioned under
fluoroscopy to maintain the base of the coracoid as a geometric circle and to
visualize the glenosphere in profile. Standardization of the axillary lateral
radiograph was achieved by maintaining the glenosphere in profile and
maintaining the scapular spine and the entire length of the coracoid in view.
Magnification was also controlled. Radiographs were reviewed by two
orthopaedic surgeons and one orthopaedic resident (R.W.S., N.H., and M.A.Z.)
for inferior, anterior, and posterior scapular notching as well as for
inferior scapular osteophytes and component position without knowledge of the
clinical outcome. Images were interpreted and measurements were made with use
of ProVision Web 4.0 SP2 software (Cerner, Idstein, Germany).
Inferior notching, consistent with previous reports, was defined as an
erosion of the scapular neck in proximity to the glenoid prosthesis. Inferior
scapular notching was classified both descriptively as well as with use of the
Nerot15 grading
system (Table I). The presence
of anterior or posterior notching was recorded by reviewing the axillary
lateral radiographs. Anterior and posterior notching was defined as any
glenoid radiolucency disrupting the normal contour of the lateral bone
adjacent to the glenosphere anteriorly and posteriorly, respectively
(Fig. 1).
Osteophytes were recorded on the anteroposterior radiographs by tracing the
scapular neck from medial to lateral and noting any osseous excrescences
(Fig. 2). An osseous
excrescence was only considered to be an osteophyte if it had been absent
preoperatively and if it was in continuity with the bone of the scapular neck
on postoperative radiographs, thereby differentiating it from heterotopic
bone. A distinction was made between osteophytes and notching, and, in cases
in which both were present simultaneously, this finding was recorded. The
distance of the most lateral aspect of the osteophyte was measured from the
medial surface of the glenosphere.
The angle between the scapular neck and the glenoid face (the scapular neck
angle) was recorded on the preoperative anteroposterior radiograph (Figs.
3-A and
3-B). A line was drawn on the
glenoid face between point A, marking the most lateral and superior glenoid
bone along the superior glenoid rim, and point B, marking the most lateral and
inferior glenoid bone along the inferior glenoid rim. A second line was drawn
to connect point B to point C, which was located 1 cm medial to point B along
the inferior glenoid rim or scapular neck. A distance of 1 cm was chosen
because all inferior notching occurred within this distance from the lateral
aspect of the inferior glenoid rim. The angle subtended by the intersection of
these two lines was recorded as the scapular neck angle.
The angle projected by the glenosphere and the scapular neck, referred to
as the prosthesis-scapular neck angle, as well as the distance from the top of
the metaglene central peg to the inferior glenoid rim, referred to as the
peg-glenoid rim distance (Figs. 4-A and
4-B), were recorded on the immediate postoperative anteroposterior
radiograph. To determine the prosthesis-scapular neck angle, two lines were
constructed in an analogous fashion as for the scapular neck angle: on a
strict radiographic profile of the glenosphere, a line was drawn connecting
point A at the most proximal medial aspect of the glenosphere to point B at
the junction of the most distal medial aspect of the glenosphere and the
inferior lateral bone of the glenoid rim. A second line was constructed
connecting point B to point C, which was 1 cm medial to point B along the
inferior glenoid rim or scapular neck. The intersection of these two lines was
recorded as the prosthesis-scapular neck angle. The peg-glenoid rim distance
was recorded by calculating the distance along a line between point D, which
marked the radiographic superior intersection of the metaglene central peg and
the glenosphere, and point B, which referenced the most inferior bone of the
inferior glenoid rim adjacent to the medial surface of the glenosphere.
Finally, the delta scapular neck angle was recorded as the difference between
the scapular neck angle and the prosthesis-scapular neck angle.
Intraclass coefficients for interobserver and intraobserver reliability
were calculated for the scapular neck angle, the prosthesis-scapular neck
angle, and the peg-glenoid rim distance with use of the independent
measurements of the two orthopaedists and one orthopaedic resident. Intraclass
coefficients for interobservational reliability were calculated for the
scapular neck angle, the prosthesis-scapular neck angle, and the peg-glenoid
rim distance with use of anteroposterior radiographs that were made
immediately postoperatively and at six weeks. This measurement was indicative
of how well our radiographs were standardized for position and
magnification.
Clinical Evaluation
Standardized clinical assessment was performed preoperatively and then
postoperatively at three months, six months, one year, and annually
thereafter, coincident with radiographic follow-up by an orthopaedist who was
different from the operating surgeon. The minimum duration of follow-up was
twenty-four months (mean, forty-four months; range, twenty-four to ninety-six
months). This assessment always included a thorough history and physical
examination, including scoring according to the system of Constant and
Murley16-18,
and obtaining a subjective shoulder value, which was the patient's estimation
of his or her shoulder as a percentage of a completely normal shoulder.
Strength was measured in kilograms with use of a validated electronic
dynamometer (Isobex; Cursor, Bern, Switzerland) with the shoulder in neutral
rotation and 90° of abduction in the scapular plane. Satisfaction was
measured on an analog scale from 1 to 4 by allowing the patient to position a
slide ruler. Active range of motion, including flexion, abduction, and
external and internal rotation at 0° of abduction, was recorded with use
of a manual goniometer. The patient was seated to avoid hyperlordosis for
elevation. Elevation was measured as the angle between the proximal aspect of
the arm and the trunk. For abduction, both arms were abducted simultaneously
to prevent bending of the trunk, and the measured abduction angle was the
angle subtended by the affected humerus with the vertical. Internal and
external rotation were measured with the patient seated with the arm at the
side.
Statistical Analysis
Statistical analysis of the results was performed by two of the authors
(R.W.S. and M.A.Z.) in conjunction with a professional statistical consultant
with use of SPSS for Windows (version 11.5; SPSS, Chicago, Illinois). Analysis
of outcomes was done with Spearman rank correlation and the Mann-Whitney U
test for the correlation and association of notching and osteophyte formation
with the scapular neck angle, the prosthesis-scapular neck angle, and the
peg-glenoid rim distance as well as preoperative and postoperative clinical
parameters. The level of significance was set at p < 0.05. In order to
determine a formula that was predictive of inferior notching (a notching
index), several methods were employed. Receiver operating characteristic
curves were defined with use of 95% confidence intervals for the variables of
prosthesisscapular neck angle, peg-glenoid rim distance, and notching index.
Sensitivity and specificity were ascertained for the values of the
prosthesis-scapular neck angle, the peg-glenoid rim distance, and the notching
index in shoulders with inferior notching. Estimated coefficients with their
standard errors for the variables of prosthesis-scapular neck angle and
pegglenoid rim distance were combined, and a formula for the notching index
was calculated by means of logistic regression analysis with use of the Hosmer
and Lemeshow
test19.
Interobserver, intraobserver, and interobservational reliability for the
scapular neck angle, prosthesis-scapular neck angle, and peg-glenoid rim
distance were assessed by determining intraclass correlation coefficients with
95% confidence intervals with use of two-way analysis of variance with mixed
effects for a single rater.
Thirty-four (44%) of the seventy-seven shoulders had inferior notching. All
of the cases of inferior notching presented by fourteen months after surgery.
No inferior notching occurred later. Six inferior notches were classified as
Nerot15 grade 1,
fourteen were classified as grade 2, twelve were classified as grade 3, and
two were classified as grade 4. Six shoulders (8%) had anterior notching, and
twenty-three shoulders (30%) had posterior notching. Twenty-one (27%) of the
seventy-seven shoulders had inferior scapular osteophytes ranging in size from
4 to 22 mm. On the average, the osteophytes ceased to increase in size at two
years (range, one to four years) and consistently occurred within 2 cm of the
medial surface of the glenosphere.
The group of shoulders with inferior notching had an average duration of
follow-up (and standard deviation) of 47 ± 18 months, whereas the group
of shoulders without notching had an average duration of follow-up of 42
± 21 months (p = 0.047). However, inferior notches were present
radiographically at an average of 4.5 months and no later than fourteen months
and ceased to progress in size or grade at an average of eighteen months
(range, twelve to forty-eight months). All inferior notches occurred within
fourteen months, although the minimum duration of follow-up for all shoulders
was twenty-four months. No significant difference in the duration of follow-up
for the shoulders with and without osteophytes was demonstrated. Likewise, no
significant difference in the frequency of inferior notching or osteophyte
formation was seen in association with extremity dominance, age, gender, or
history of previous operations.
Correlation of Glenosphere Position and Inferior Notching
The intraclass correlation coefficients representing interobserver,
intraobserver, and interobservational reliability for the measurement of the
scapular neck angle were calculated to be 0.90 (p < 0.001), 0.89 (p <
0.001), and 0.92 (p < 0.001), respectively. The intraclass correlation
coefficients representing interobserver, intraobserver, and interobservational
reliability for the measurement of the prosthesis-scapular neck angle were
calculated to be 0.97 (p < 0.001), 0.98 (p < 0.001), and 0.96 (p <
0.001), respectively. The intraclass correlation coefficients representing
interobserver, intraobserver, and interobservational reliability for the
measurement of the peg-glenoid rim distance were calculated to be 0.96 (p <
0.001), 0.97 (p < 0.001), and 0.95 (p < 0.001), respectively.
The mean preoperative anatomic scapular neck angle was 88° ±
17° (range, 61° to 140°), and the mean postoperative
prosthesis-scapular neck angle was 106° ± 23° (range, 67°
to 156°). This represented an average change in the scapular neck angle
from preoperatively to postoperatively (that is, a delta scapular neck angle)
of 18° ± 22°. The mean distance between the inferior glenoid
rim and the superior border of the central peg of the metaglene (the
peg-glenoid rim distance) was 22 ± 4 mm (range, 12.9 to 30 mm), with a
smaller value representing a lower component position on the glenoid.
A higher prosthesis-scapular neck angle was correlated with an increased
frequency of inferior notching (r = 0.667, p < 0.001). The average
prosthesis-scapular neck angle was 93° for shoulders without inferior
notching, compared with 124° for shoulders with inferior notching (p <
0.001). There was no significant correlation between the preoperative scapular
neck angle and notching (r = 0.179, p = 0.118). However, the difference
between the preoperative and postoperative scapular neck angles (the delta
scapular neck angle) was positively correlated with inferior notching (r =
0.334, p = 0.003). The mean delta scapular neck angle was 9° for shoulders
without inferior notching, compared with 31° for shoulders with inferior
notching (p < 0.001). A larger peg-glenoid rim distance, reflecting higher
positioning of the metaglene and thus the glenosphere, was correlated with an
increased frequency of inferior notching (r = 0.654, p < 0.001). The mean
peg-glenoid rim distance was 20.1 mm for shoulders without inferior notching,
compared to 24.7 mm for those with notch inferior notching (p < 0.001).
Table II illustrates the mean
values for the prosthesis-scapular neck angle, the delta scapular neck angle,
and the peg-glenoid rim distance, with standard deviations and ranges, for
shoulders with and without inferior notching.
When the peg-glenoid rim distance and the prosthesis-scapular neck angle
initially were observed independently, the receiver operating characteristic
curves analyzing the area of the standard error, including the 95% confidence
intervals, had high areas under the curve of 0.88 and 0.89, respectively. From
these curves, the sensitivity and specificity and critical values of both the
peg-glenoid rim distance and the prosthesis-scapular neck angle were
determined to maximize both the sensitivity and the specificity for the
detection of inferior notching. A peg-glenoid rim distance of =22 mm
yielded a sensitivity of 82% and a specificity of 79% for inferior notching,
and a prosthesis-scapular neck angle of =109° yielded a sensitivity of
82% and a specificity of 81% for inferior notching. The influences of the
prosthesis-scapular neck angle and the peg-glenoid rim distance on inferior
notching were then analyzed together. A scatter plot of inferior notching was
constructed with the variables of peg-glenoid rim distance and
prosthesis-scapular neck angle being plotted against each other
(Fig. 5). An increased
frequency of inferior scapular notching was noted when the values of both the
peg-glenoid rim distance and the prosthesis-scapular neck angle were higher.
With use of logistic regression and the Hosmer and Lemeshow test, the
estimated coefficients of the prosthesis-scapular neck angle and peg-glenoid
rim distance, representing the relative contributions of the
prosthesis-scapular neck angle and peg-glenoid rim distance to inferior
notching, were determined to be 0.13 and 1.0, respectively, with standard
errors of 0.04 and 0.32, respectively. The Wald statistic coefficients for the
prosthesis-scapular neck angle (12.9) and peg-glenoid rim distance (10.3) were
highly significant (p < 0.001). On the basis of this finding, we calculated
a notching index with use of the formula: notching index =
(prosthesis-scapular neck angle × 0.13) + (peg-glenoid rim
distance).
Receiver operating characteristic curves
(Fig. 6) were then constructed
to compare the notching index with the prosthesis-scapular neck angle and the
peg-glenoid rim distance. The area under the receiver operating characteristic
curve for the notching index was 0.964 (p < 0.001), exceeding those of both
prosthesis-scapular neck angle and the peg-glenoid rim distance alone. On the
basis of a highly significant receiver operating characteristic curve
calculation, a notching index value of 35 showed the highest specificity and
sensitivity together (p = 0.0001; 95% confidence interval, 0.923 and 1.0006,
respectively). A notching index of >35 demonstrated a sensitivity of 91%
and a specificity of 88% for the affected shoulder to manifest inferior
notching. Thus, a notching index of 35 would be expected to avoid inferior
notching 91% of the time. Table
III was constructed with use of the above formula to determine an
acceptable peg-glenoid rim distance for each 5° increment of the
prosthesis-scapular neck angle between 60° and 160° with use of a
notching index value of 35. In addition, the peg-glenoid rim distance value
was then subtracted from 21.9 mm, the distance from the top of the metaglene
central peg to the inferior tip of the Delta III 36-mm glenosphere. This value
represents the amount of overhang, in millimeters, of the inferior tip of a
36-mm glenosphere over the inferior glenoid rim, corresponding to the specific
peg-glenoid rim distance value. The distance of insertion of the central
drill-guide pin from the inferior glenoid rim was calculated by subtracting
3.9 mm, the radius of the Delta III metaglene central peg, from the value of
peg-glenoid rim distance.
Inferior Notching and Clinical Outcome
All clinical parameters except for external rotation improved in comparison
with the preoperative states. In the entire series, the relative Constant
score increased from an average of 38% preoperatively to 78% at the time of
the final follow-up (p < 0.001). The subjective shoulder value increased
from 28% to 67% (p < 0.001). Active flexion improved from an average of
65° to 115° (p < 0.001), whereas abduction improved from 63° to
111° (p < 0.001). At the time of the latest follow-up, the shoulders
with measurable inferior notching were associated with a lower mean relative
Constant score (72% compared with 83%; p = 0.028), a lower subjective shoulder
value (62% compared with 71%; p = 0.032), and lower postoperative active
flexion (110° compared with 127°; p = 0.004) and abduction (102°
compared with 118°; p = 0.033) in comparison with the values for the
shoulders without inferior notching (Table
IV). There was a significant negative correlation between an
increasing inferior notch size and the postoperative relative Constant score
(p = 0.02; r = —0.270), the subjective shoulder value (p = 0.025; r =
—0.261), active flexion (p = 0.004; r = —0.332), and active
abduction (p = 0.022; r = —0.266). The variables that were associated
with notching, namely, a higher prosthesis-scapular neck angle, a higher delta
scapular neck angle, and a greater peg-glenoid rim distance, were correlated
with inferior clinical results as well
(Table V).
Posterior and Anterior Notching
There was a significant positive correlation between greater external
rotation at 0° of abduction preoperatively (r = 0.292; p = 0.014) and
postoperatively (r = 0.403; p = 0.001) and the development of posterior
notching. Shoulders with posterior notching had an average of 31° and
34° of external rotation postoperatively and preoperatively, respectively,
compared with 17° (p = 0.01) and 15° (p < 0.001) for shoulders
without posterior notching. With the numbers studied, there was no significant
effect of posterior or anterior notching on clinical outcome.
Osteophytes
Higher values for the prosthesis-scapular neck angle (p = 0.001) and the
delta scapular neck angle (p < 0.05) were associated with a lower frequency
of osteophyte formation. The peg-glenoid rim distance did not influence
osteophyte formation. The presence of osteophytes was associated with
decreased abduction strength (p = 0.014) and a decreased range of active
elevation (p = 0.016) but higher satisfaction (p = 0.029) and a higher
relative Constant score (p = 0.020) at the time of the most recent
follow-up.
Contemporary reverse ball-and-socket prostheses have yielded good pain
relief and restoration of function in patients with irreparable rotator cuff
disease associated with
arthritis11,12,20,21.
Nonetheless, surgeons have remained cautious about long-term glenoid component
failure. The major complication associated with the reverse Delta III
prosthesis that has been noted radiographically and reported so far has been
inferior scapular notching. This process is poorly understood but is
associated with substantial concern that it might lead to glenoid failure and
ultimately to humeral component loosening due to the polyethylene
wear13.
We found inferior notching in association with thirty-four (44%) of the
seventy-seven Delta III reverse prostheses that were evaluated. The process of
inferior notch formation was rapid, occurring within fourteen months and
without further progression after an average of eighteen months. The majority
of these notches were classified as grade 1 and 2 according to the system of
Nerot15 and
therefore were minor. In only two cases did notches extend under the baseplate
(at an average of seventeen months after implantation), but none of these
notches were associated with loosening or catastrophic failure of the
metaglene and glenosphere after an additional average follow-up of sixteen
months. Our findings are consistent with previous studies in the literature,
which have demonstrated inferior notching in association with 50% to
96%11,12,15,20,21
of the Delta III prostheses and relatively infrequent glenoid component
loosening and failure after intermediate-term
follow-up20,21.
Our results indicate that the craniocaudal positioning of the glenosphere
as well as the angular relationship between the glenosphere and the scapular
neck are highly correlated with clinical outcome and inferior scapular
notching. The effects of the prosthesis-scapular neck angle and the
peg-glenoid rim distance on inferior scapular notching are tightly coupled.
However, the peg-glenoid rim distance exerts more of an effect than the
prosthesis-scapular neck angle by a factor of approximately eight. A lower
frequency of inferior scapular notching is predicated on reaching critical
values of the peg-glenoid rim distance as well as the prosthesis-scapular neck
angle in each patient (Figs. 7-A,
7-B, and 7-C). Both
variables can be controlled intraoperatively. On the basis of our
measurements, the optimal position of the inferior tip of a 36-mm glenosphere
as well as the entry point for the central drill-guide pin relative to the
inferior glenoid rim can be calculated. These values are reported in
Table III and can be used as a
guide for planning and implanting a reverse shoulder replacement. Higher
prosthesis-scapular neck angle values require more inferior positioning of the
glenosphere, demanding a more inferior placement of the central drill-guide
(Figs. 8-A, 8-B, and
8-C). These data are in
contradiction to the recommendations of the manufacturer of the Delta III
reverse total shoulder system (DePuy), which suggest placement of the central
guide-pin just posterior and inferior to the intersection of the longest
superoinferior and anteroposterior axes of the glenoid, regardless of osseous
scapular anatomy. Our results are consistent with and expand on the
experimental findings of Nyffeler et
al.14, who reported
that a more inferior position of the glenosphere results in less mechanical
impingement of the inferior scapular neck in addition to better range of
motion. We recognize that there may be additional factors that contribute to
the development of inferior notching that were not analyzed in the scope of
the present study. One such factor may be body habitus, whereby obese patients
may physically be unable to adduct the arms to the extent of mechanical
impingement. The precision of the prediction with the described formula
suggests, however, that the two identified factors are of paramount
importance.
Superior or inferior orientation of the central guide-pin directly
influences the prosthesis-scapular neck angle. Whereas some authors believe
that an inferior tilt of the glenosphere is desirable for long-term stability
of the prosthesis, this inferior inclination results in an increase in the
prosthesis-scapular neck angle, which increases the probability of inferior
scapular
notching22. The
surgeon must weigh the potential benefits and risks in determining how to
orient the glenosphere. Medial erosion will also bring the glenosphere closer
to the lateral pillar of the scapula, thereby increasing the
prosthesis-scapular neck angle, a phenomenon that would be exacerbated by
further glenoid reaming, especially with a flat reamer. Therefore, excessive
medial reaming and medialization of the glenoid component should be avoided to
prevent inferior scapular notching.
Previous studies have been somewhat inconclusive with regard to the effect
of inferior notching on the clinical outcome after reverse total shoulder
arthroplasty6,21.
However, we found that the prevention of inferior notching is not only
mechanically but also clinically relevant. In the present study, after an
average duration of follow-up of forty-four months, patients with inferior
scapular notching had a significantly lower mean relative Constant score and
subjective shoulder value as well as a more limited range of motion and lower
strength compared with those without inferior scapular notching.
In addition to the well described inferior scapular notching, we also
documented posterior and anterior notching. Posterior notching was frequent
and was associated with good active external rotation while anterior notching
was more rare. These forms of scapular notching had no influence on the
clinical results, but additional studies are necessary to elucidate the
biomechanical and clinical relevance of these forms of notching.
Inferior scapular osteophytes were seen in twenty-one (27%) of the
seventy-seven shoulders that were studied. We routinely release the origin of
the long head of the triceps along the scapular neck intraoperatively, and we
believe that these inferior scapular osteophytes were traction osteophytes
resulting from incomplete release of this tendon as they were in continuity
with the scapular cortex. Their effect on clinical outcome was unclear in the
present study. Valenti et
al.15 reported that
osteophytes exist along the inferior aspect of the scapular neck, but no study
has characterized their etiology or effect. Clearly, future studies must
distinguish osteophytes from inferior notches in order to assess the effect of
either.
In the present study, radiographic analysis involved the use of fluoroscopy
to align the central radiographic beam, allowing for the systematic evaluation
of the peg-glenoid rim distance and the prosthesis-scapular neck angle.
Extension of our conclusions to other practitioners and patient cohorts would
require a similarly rigorous radiographic technique. A change in inclination
or obliquity during imaging would lead to nonstandardization and,
subsequently, to a change in the values for the prosthesis-scapular neck angle
and peg-glenoid rim distance, thereby not allowing direct comparison among
patients and x-ray series.
In summary, the present study confirms that inferior scapular notching
represents a substantial radiographic phenomenon related to the implantation
of the Delta III reverse ball-and-socket prosthesis that is associated with
worse clinical outcomes. Preoperative planning and proper selection of glenoid
implants as well as very precise surgical positioning of the glenoid component
may assist in preventing inferior scapular notching. ?
Note: The authors thank Professor Burkhardt Seifert, Department
of Biostatistics, University of Zürich, for his statistical support.
Sanchez-Sotelo J, Cofield RH, Rowland
CM. Shoulder hemiarthroplasty for glenohumeral arthritis associated with
severe rotator cuff deficiency. J Bone Joint Surg Am.2001;83:
1814-22.831814
2001
[PubMed]
Williams GR Jr, Rockwood CA Jr.
Hemiarthroplasty in rotator cuff-deficient shoulders. J Shoulder Elbow
Surg.1996;5:
362-7.5362
1996
[CrossRef]
Franklin JL, Barrett WP, Jackins SE,
Matsen FA 3rd. Glenoid loosening in total shoulder arthroplasty. Association
with rotator cuff deficiency. J Arthroplasty.1988;3:
39-46.339
1988
[PubMed][CrossRef]
Neer CS 2nd, Craig EV, Fukuda H.
Cuff-tear arthropathy. J Bone Joint Surg Am.1983;65:
1232-44.651232
1983
[PubMed]
Post M, Haskell SS, Jablon M. Total
shoulder replacement with a constrained prosthesis. J Bone Joint Surg
Am.1980;62:
327-35.62327
1980
Sirveaux F, Favard L, Oudet D, Huguet D,
Lautman S. Grammont inverted total shoulder arthroplasty in the treatment of
glenohumeral osteoarthritis with massive and non repairable cuff rupture. In:
Walch G, Boileau P, Molé D, editors. 2000 Prosthèses
d'épaule... recul de 2 à 10 ans. Paris: Sauramps
Médical; 2001. p
247-52.247
2001
Arntz CT, Jackins S, Matsen FA 3rd.
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;75:
485-91. Erratum in: J Bone Joint Surg Am.
1993;75:1112.75485
1993
Pollock RG, Deliz ED, McIlveen SJ,
Flatow EL, Bigliani LU. Prosthetic replacement in rotator cuff-deficient
shoulders. J Shoulder Elbow Surg.1992;1:
173-86.1173
1992
[CrossRef]
Boileau P, Watkinson DJ, Hatzidakis AM,
Balg F. Grammont reverse prosthesis: design, rationale, and biomechanics.
J Shoulder Elbow Surg.2005;14(1 Suppl S):
147S-61S.14147S
2005
[PubMed][CrossRef]
Seebauer L, Walter W, Keyl W. Reverse
total shoulder arthroplasty for the treatment of defect arthropathy.
Oper Orthop Traumatol.2005;17:
1-24.171
2005
[PubMed][CrossRef]
Sirveaux F, Favard L, Oudet D, Huguet D,
Walch G, Molé D. Grammont inverted total shoulder arthroplasty in the
treatment of glenohumeral osteoarthritis with massive rupture of the cuff.
Results of a multicentre study of 80 shoulders. J Bone Joint Surg
Br.2004;86:
388-95.86388
2004
[CrossRef]
Vanhove B, Beugnies A. Grammont's
reverse shoulder prosthesis for rotator cuff arthropathy. A retrospective
study of 32 cases. Acta Orthop Belg.2004;70:
219-25.70219
2004
[PubMed]
Nyffeler RW, Werner CM, Simmen BR,
Gerber C. Analysis of a retrieved Delta III total shoulder prosthesis.
J Bone Joint Surg Br.2004;86:
1187-91.861187
2004
[PubMed][CrossRef]
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:
524-8.14524
2005
[PubMed][CrossRef]
Valenti PH, Boutens D, Nerot C. Delta 3
reversed prosthesis for osteoarthritis with massive rotator cuff tear: long
term results (>5 years). In: Walch G, Boileau P, Molé D, editors.
2000 Prosthèses d'épaule... recul de 2 à 10
ans. Paris: Sauramps Médical; 2001. p
253-9.253
2001
Constant CR, Murley AH. A clinical
method of functional assessment of the shoulder. Clin Orthop Relat
Res.1987;214:
160-4.214160
1987
Constant CR. Age related recovery
of shoulder function after injury [thesis]. Cork, Ireland: University
College; 1986.
1986
Gerber C. Latissimus dorsi transfer for
the treatment of irreparable tears of the rotator cuff. Clin Orthop
Relat Res.1992;275:
152-60.275152
1992
Stokes ME, Davis CS, Koch GG.
Categorical data analysis using the SAS system. Great Britain:
John Wiley and Sons; 2001:
626.626
2001
Rittmeister M, Kerschbaumer F. Grammont
reverse total shoulder arthroplasty in patients with rheumatoid arthritis and
nonreconstructible rotator cuff lesions. J Shoulder Elbow Surg.2001;10:
17-22.1017
2001
[PubMed][CrossRef]
Werner CM, Steinmann PA, Gilbart M,
Gerber C. Treatment of painful pseudoparesis due to irreparable rotator cuff
dysfunction with the Delta III reverse-ball-and-socket total shoulder
prosthesis. J Bone Joint Surg Am.2005;87:
1476-86.871476
2005
[PubMed][CrossRef]
Gutierrez S, Greiwe RM, Frankle MA,
Siegal S, Lee WE 3rd. Biomechanical comparison of component position and
hardware failure in the reverse shoulder prosthesis. J Shoulder Elbow
Surg.2006Sep19;[Epub ahead
of print].
2006