| Methods |
Randomised controlled trial.
Blinding: both participants and outcome assessors were blinded.
Loss to follow‐up: 1 patient (11.1%) in steroid alone group and 1 patient (7.7%) in steroid + distension group.
Appropriate statistical analysis: Based upon completers only analysis (20/22) |
| Participants |
22 patients.
Inclusion criteria:
1) age between 18 and 70 yrs; 2) frozen shoulder of more than 6 weeks duration; 3) nocturnal accentuation of pain; 4) passive range of external rotation in shoulder less than 50% of opposite shoulder; 5) no effusion in glenohumeral joint; 6) normal x‐ray of affected shoulder; 7) normal ESR, haemoglobin, leucocytes, alkalic phosphates and negative IgM rheumatoid factor; 8) no trauma to shoulder in last 6 months that caused pain or restricted movement of the shoulder within one week (acceptance of trivial minor injuries); 9) no diabetes; 10) no other treatment for frozen shoulder except analgesics in study period. |
| Interventions |
Group 1(13 patients): distension with 19ml of 0.5% lidocaine and 20mg triamcinolone hexacetonid.
Group 2 (9 patients): 20mg triamcinolone hexacetonid injection alone.
The intraarticular injection in the glenohumeral joint was carried out by a posterior approach and confirmed by ultrasound.
The treatment was repeated once a week for a maximum of 6 weeks or until no symptoms. |
| Outcomes |
Assessed at baseline, 3, 6 and 12 weeks.
1) physician judgement of severity of disorder (severe = 1, moderate = 2, light pain = 3) from patients' verbal expression of pain and function (undressing);
2) passive flexion, extension, abduction, external rotation, and elevation of affected shoulder in comparison with opposite shoulder (i.e. 0‐25% = 1, 25‐50% = 2, 50‐75% = 3, 75‐100% = 4). Elevation performed with fixed scapulae;
3) Pain at rest and on function using VAS (where 0 = no pain and 10 = unbearable pain). Patients recorded their average pain every day in study period and mean score of each week was used for evaluations; 4) Daily usage of analgesics;
5) type and number of side effects. |
| Notes |
This article met the inclusion criteria for this review but the data was not presented in a format which allowed meta‐analysis.
Randomisation according to the envelope method.
Results:
Signficant improvement in range of motion and analgesic use in group treated with distension with local anesthetis and steroid vs steroid alone. There was no difference in pain at rest but a trend favouring the distension group for pain with activity. |
| Risk of bias |
| Bias |
Authors' judgement |
Support for judgement |
| Allocation concealment? |
Unclear risk |
B ‐ Unclear |
