60 The Iowa Orthopaedic Journal
AbstrAct
the purpose of this study was to analyze pre-
operative and intra-operative factors that affect the
outcome of shoulder arthroplasty. We undertook a
retrospective review of all shoulder arthroplasties
performed at our institution between 1986 and
2003. Patients were contacted and outcomes were
assessed using the simple shoulder test and the
Western Ontario Osteoarthritis of the shoulder
Index questionnaires. One hundred six patients
(126 shoulders) participated in the study. the
average length of follow-up was 6 years 9 months
(range 2 to 20 years). revision arthroplasty sur-
gery and female gender were associated with worse
outcomes. Age, the number of medical comorbidi-
ties, obesity, pre-operative range of motion, prior
non-arthroplasty surgery, smoking, and alcohol
abuse did not correlate with outcome. Patients who
had shoulder arthroplasty for osteoarthritis had
better outcome scores than those with rheumatoid
arthritis. For intra-operative variables, significantly
worse outcomes were found both with the use of
hemiarthroplasty and in patients with a rotator
cuff tear identified at the time of surgery. these
findings may help to optimize patient and surgery
selection in shoulder arthroplasty and assist in
preoperative patient counseling.
IntrOductIOn
The outcome of shoulder arthroplasty can be affected
by several factors. These include patient related vari-
ables, such as the underlying etiology for glenohumeral
degeneration, comorbid conditions and demographics,
as well as intra-operative findings. Pre-operatively, the
diagnoses most frequently encountered in advanced
glenohumeral degeration include osteoarthritis (OA),
rheumatoid arthritis (RA), severe proximal humerus
fractures, post-traumatic degenerative arthritis, avascular
necrosis and cuff-tear arthropathy. For the diagnosis
of OA, multiple studies have shown that arthroplasty
reliably improves pain and ROM.2,6,17 Likewise, several
studies have demonstrated that RA patients benefit from
shoulder arthroplasty.8,13,14,29 There is less information on
the impact of demographics and comorbid conditions.11
Intra-operative findings and decisions may impact pa-
tient outcomes in shoulder arthroplasty as well, including
the presence of a rotator cuff tear and whether a HA or
TSA is performed. Multiple studies have suggested that
a rotator cuff tear is associated with worse outcomes by
both subjective and objective scores,3,7,10,11 although other
studies have not validated this finding in OA5,12 or RA.8
Controversy still exists regarding the superiority of TSA
versus HA for the treatment of glenohumeral arthrosis
from OA and RA.
Although prior studies have examined the effect of
some preoperative and intra-operative factors,4,5,7,11,12 none
has simultaneously examined the effect of multiple pre-
operative and intra-operative variables on outcomes in
shoulder arthroplasty patients. The goal of the present
study was to validate previous findings of preoperative
and intraoperative factors that have been shown to affect
outcome as well as attempt to delineate other character-
istics of patients whose outcomes are better (or worse)
after shoulder arthroplasty in a group of patients with
varying etiologies of shoulder degeneration.
MAterIAls And MethOds
Institutional Review Board Approval was obtained for
the study. We searched hospital records between 1986
and 2003 for current procedural terminology (CPT)
codes involving shoulder arthroplasty (23470 and 23472).
A chart review was performed and underlying diagno-
sis, patient demographics, length of follow up, medical
comorbidities, pre-operative range of motion, and prior
ipsilateral shoulder surgery were recorded. Significant
medical comorbidities were recorded as present or ab-
sent in a binary fashion and then summed. The impact
of the total number of comorbidities on outcome was
Department of Orthopaedics and Rehabilitation
University of Iowa Hospitals and Clinics
Conflict of interest statement:
None of the authors received any financial support for this work.
None of the authors has financial or other relationships that might
lead to a conflict of interest in relationship to this publication.
Correspondence:
Brian R. Wolf, MD, MS
University of Iowa Hospitals and Clinics
200 Hawkins Drive
Iowa City, Iowa 52242
319-353-7954
319-384-9306 (facsimile)
brian-wolf@uiowa.edu
Pre-OPerAtIve And IntrA-OPerAtIve FActOrs relAted
tO shOulder ArthrOPlAsty OutcOMes
Jonathan A. Donigan, MD, W. Anthony Frisella, MD, Daniel Haase, MD, Lori Dolan, PhD, Brian Wolf, MD, MS

Volume 29 61
Factors in Shoulder Arthroplasty Outcomes
analyzed. The following comorbidities were recorded:
chronic obstructive pulmonary disease, hypertension,
heart disease (CAD, arrythmia, congestive heart failure),
diabetes mellitus, tobacco abuse, alcohol abuse, Axis I
psychiatric disease, and other rheumatologic disease.
Patient weight, as measured by body mass index, was
assessed as well, and patients were divided into normal
weight (BMI less than 25), overweight (BMI greater
than or equal to 25), and obese (BMI greater than or
equal to 30) based on standard BMI cutoffs. For each
case a chart review of the operative report was done to
identify those patients who had a rotator cuff tear identi-
fied at the time of surgery. When no specific mention
was made of a cuff tear the patient was recorded in the
database as having an intact rotator cuff. The use of a
TSA or HA was also documented.
Questionnaires and informed consent documents
were sent by mail to each patient. Patients who had
had bilateral shoulder arthroplasties returned separate
outcomes measures for each shoulder. Each patient re-
ceived and returned two outcomes questionnaires, the
Simple Shoulder Test (SST) and the Western Ontario Os-
teoarthritis of the Shoulder Questionnaire (WOOS).15,16
The Simple Shoulder Test consists of 12 simple ‘yes’
or ‘no’ questions and is easy to administer and under-
stand. It has been tested in populations of patients with
multiple shoulder pathologies, including OA, RA, avascu-
lar necrosis, and rotator cuff tears.15 It has been shown
to be able to distinguish between normal shoulders and
those with the previously listed conditions. In the present
study SST scores were expressed as a percent of tasks
that the patient could complete. A score of 50 meant
that the patient could complete half of the tasks, while
a score of 100 meant that the patient could complete all
of the tasks with that shoulder. The WOOS is a validated
questionnaire designed specifically for use in patients
with OA.16 It has been correlated with multiple other
measures of shoulder function. The WOOS was chosen
for these reasons and because the plurality of patients
studied had OA as their primary diagnosis. As for the
SST, the WOOS score has been expressed as a percent
of a total best score (i.e., 100 is best possible score, 0 is
lowest score possible).
Data was analyzed using the SAS statistical analysis
package (v. 9.0 Cary, NC). For comparisons between
two groups, we used t-test for univariate variables and
Pearson’s correlation for continuous variables. T-tests
were used for the analysis of revision surgery, patient
sex, and individual comorbidities. T-tests were also used
for paired comparison of variables after ANOVA analy-
sis. Pearson’s correlation was used for analysis of SST
and WOOS scores and patient age, length of follow-up,
and range of motion. We used ANOVA for data where
multiple comparisons were made, including underlying
diagnosis, comorbidities, and BMI.
results
We identified 169 patients who had undergone either
a HA or TSA during the study period who were still
living. Six patients were mentally or physically unable
to participate, leaving 163 patients. We were unable to
contact 21 patients after a thorough search using hos-
pital records and internet search databases. This left
a total of 142 patients. Of those we were able to locate
and contact, 106 participated in the study. Of the 106
participants, 20 had bilateral shoulder arthroplasty, for
a total of 126 shoulders.
There were 43 men and 63 women. Average time to
follow-up was 6.8 years (range 2-20 years).
Pre-operative Factors
The underlying diagnoses were as follows: 61 patients
had primary OA, 23 had RA, 9 had acute fractures, 10
had revision of a failed HA to a TSA, 8 had cuff tear
arthropathy, 5 had post-traumatic OA, 4 had osteone-
crosis, 5 had revision of a prosthesis for other reasons,
and one had a recurrent giant cell tumor. Outcomes
were compared for patients with a diagnosis of primary
OA (n=60) to patients with RA (n=23). Both WOOS and
SST scores were significantly better for patients with
primary OA (WOOS 78 versus 66, p=0.05; SST 70 ver-
sus 42, p<0.0001) (Figure 1). Multivariate regression of
outcomes based on underlying diagnosis revealed OA
patients had the best outcomes overall.
The effect of patient demographics on outcome was
examined. The 54 shoulders in male patients had better
outcomes than the 72 shoulders in females (WOOS 68
versus 53, p=0.0065, SST 75 versus 52, p=0.0001) (Figure
2). Neither age nor length of follow-up correlated with
outcome (p>0.05, WOOS and SST).
18
396
397
398
399
400
Figures: 401
Figure 1: Osteoarthritis patients has significantly higher WOOS and SST scores 402
compared to patients with Rheumatoid arthritis. 403
404
405
Figure 2: Males had statistically significant higher WOOS and SST scores compared with 406
females. 407
Figure 1. Osteoarthritis patients has significantly higher WOOs and
sst scores compared to patients with rheumatoid arthritis.