Summary
Our investigation of pre-operative diagnostic injection shows that it accurately predicts SANE scores after shoulder arthoplasty at 3 months and underestimates clincal outcomes at 6, 12 and 24 months.
Abstract
Background
Shoulder arthroplasty is highly effective in improving shoulder pain and function.
However, it is difficult to predict how much a patient will improve after a shoulder
replacement. Because injections are frequently used to diagnose the cause of a
patient’s pain, this study sought to evaluate whether a pre-operative diagnostic
injection accurately predicts a patient’s post-operative outcome after a shoulder
arthroplasty.
Methods
Using a single-surgeon shoulder arthroplasty registry, patients who underwent total
shoulder arthroplasty (TSA) or reverse total shoulder (RSA) for a radiographically-
confirmed diagnosis of glenohumeral arthritis (GHOA) or rotator cuff arthropathy
(RCA), respectively, between January 1, 2015 and December 31, 2021 were identified.
All patients who received a pre-operative injection of 0.5% marcaine (ultrasound-
guided intra-articular injection for GHOA and subacromial injection for RCA), and had
their pain improvement immediately after the injection recorded as the change in their
Single Assessment Numeric Evaluation (SANE) score were included. Post-operatively,
SANE scores were obtained at 3 months, 6 months, 12 months, and 24 months.
Repeated Measures ANOVA was used to assess the association between their
response to the pre-operative diagnostic injection and post-operative response to their
shoulder arthroplasty.
Results
A total of 63 patients met the inclusion criteria. Patient demographics, co-morbidities
and surgical factors demonstrated no significant association with their pre-operative
response to a diagnostic injection or their post-operative SANE scores. The response
to a pre-operative diagnostic injection was not significantly different from the 3-month
SANE score (0.4% [95% CI: -7.6, 8.4] (p = 0.23)). However, the pre-operative
response to the diagnostic injection significantly under-predicted the response to
arthroplasty at 6 months (11.0% [95% CI: 5.1, 16.9] (p < 0.05)), 12 months (10.7%
[95% CI: 4.8, 16.7] (p <0.05)), and 24 months (18.4% [95% CI: 11.9, 24.9] (p<0.05)).
Post-operative patients continued to experience an improvement in their SANE score
through 24 months, as observed in a mixed effect linear regression.
Discussion
A patient’s improvement after a pre-operative, diagnostic injection accurately predicts
their SANE score three months after a shoulder arthroplasty. However, this intervention
underestimates their clinical outcome at 6 months, 12 months and 24 months. In this
way, a diagnostic injection may best emulate the pain relief provided by a shoulder
replacement, but may not effectively anticipate their improvement in function and
quality of life.