Gartsman and colleagues compared total shoulder arthroplasty with a cemented polyethylene glenoid against humeral head replacement alone for osteoarthritis of the shoulder.
Randomized trial of 51 shoulders in 47 patients at the Texas Orthopedic Hospital, Houston, followed a mean of 35 months (24 to 72).
Methods
Design
Randomized controlled trial
Allocation by random‑numbers table
Allocation revealed intraoperatively after glenoid exposure
Examiner not blinded to the procedure
Setting
Texas Orthopedic Hospital, Houston
Patients
47 patients (51 shoulders)
Degenerative osteoarthritis scheduled for arthroplasty
Intact rotator cuff and concentric glenoid without bone loss
4 patients lost to follow‑up
Groups
Total shoulder arthroplasty with cemented polyethylene glenoid (27 shoulders)
Hemiarthroplasty without glenoid resurfacing (24 shoulders)
Same humeral component and same operating or supervising surgeon in both groups
Follow-up
Mean 35 months (range 24 to 72)
Outcomes
UCLA shoulder score
ASES shoulder score
Pain
Range of motion
Satisfaction, function and strength
Cost, operative time and blood loss
Reoperations
Key Findings
Total shoulder arthroplasty gave significantly better pain relief (p = 0.002 on both pain scales) and internal rotation (p = 0.003) than hemiarthroplasty, while elevation and external rotation were similar.
Total ASES and UCLA scores favored total shoulder arthroplasty, but the UCLA difference (p = 0.008) missed the threshold of p = 0.006 set to adjust for 9 comparisons.
The trial enrolled fewer shoulders than its planned 35 per group, and the authors note it may have been too small to detect a difference in total scores.
On radiographs, 4 of 27 glenoid components had a radiolucent line wider than 2 mm and 2 had shifted position, both shifted components without symptoms.
Three of 24 hemiarthroplasties were revised to add a glenoid component for pain and joint‑space loss (at 19, 39 and 48 months), and no total shoulder arthroplasty was revised.
Total shoulder arthroplasty took longer (98 vs 63 minutes, p = 0.01), lost more blood (300 vs 150 mL, p = 0.01) and raised hospital charges by at least $1177 per case.
Board pearl
After hemiarthroplasty, the joint space narrowed progressively in 14 of 24 shoulders, and 7 of these had increasing pain.
Authors' conclusion
Total shoulder arthroplasty provided superior pain relief compared with hemiarthroplasty in patients who had glenohumeral osteoarthritis, but it was associated with an increased cost of $1177 per patient.
Gartsman and colleagues compared total shoulder arthroplasty with a cemented polyethylene glenoid against humeral head replacement alone for osteoarthritis of the shoulder.
Randomized trial of 51 shoulders in 47 patients at the Texas Orthopedic Hospital, Houston, followed a mean of 35 months (24 to 72).
Methods
Design
Randomized controlled trial
Allocation by random‑numbers table
Allocation revealed intraoperatively after glenoid exposure
Examiner not blinded to the procedure
Setting
Texas Orthopedic Hospital, Houston
Patients
47 patients (51 shoulders)
Degenerative osteoarthritis scheduled for arthroplasty
Intact rotator cuff and concentric glenoid without bone loss
4 patients lost to follow‑up
Groups
Total shoulder arthroplasty with cemented polyethylene glenoid (27 shoulders)
Hemiarthroplasty without glenoid resurfacing (24 shoulders)
Same humeral component and same operating or supervising surgeon in both groups
Follow-up
Mean 35 months (range 24 to 72)
Outcomes
UCLA shoulder score
ASES shoulder score
Pain
Range of motion
Satisfaction, function and strength
Cost, operative time and blood loss
Reoperations
Key Findings
Total shoulder arthroplasty gave significantly better pain relief (p = 0.002 on both pain scales) and internal rotation (p = 0.003) than hemiarthroplasty, while elevation and external rotation were similar.
Total ASES and UCLA scores favored total shoulder arthroplasty, but the UCLA difference (p = 0.008) missed the threshold of p = 0.006 set to adjust for 9 comparisons.
The trial enrolled fewer shoulders than its planned 35 per group, and the authors note it may have been too small to detect a difference in total scores.
On radiographs, 4 of 27 glenoid components had a radiolucent line wider than 2 mm and 2 had shifted position, both shifted components without symptoms.
Three of 24 hemiarthroplasties were revised to add a glenoid component for pain and joint‑space loss (at 19, 39 and 48 months), and no total shoulder arthroplasty was revised.
Total shoulder arthroplasty took longer (98 vs 63 minutes, p = 0.01), lost more blood (300 vs 150 mL, p = 0.01) and raised hospital charges by at least $1177 per case.
Board pearl
After hemiarthroplasty, the joint space narrowed progressively in 14 of 24 shoulders, and 7 of these had increasing pain.
Authors' conclusion
Total shoulder arthroplasty provided superior pain relief compared with hemiarthroplasty in patients who had glenohumeral osteoarthritis, but it was associated with an increased cost of $1177 per patient.