Sabeti 2007.

Methods	Study design: single‐centre, parallel‐group, two‐arm, single‐blind, RCT Setting: outpatient clinic in Department of Orthopedics, Vienna Medical School, Vienna, Austria Trial time period: not reported Interventions: low‐dose ESWT vs high‐dose ESWT Sample size calculation: not performed Analysis: not ITT
Participants	Number of participants: screened: 50 enrolled: 47 (3 excluded no reasons given) randomised: 47 (22 in low‐dose ESWT group; 25 in high‐dose ESWT group) included in analyses: 44 (21 in low‐dose ESWT group; 23 in high‐dose ESWT group) Inclusion criteria: aged ≥ 18 years pain refractive to therapy for > 6 months calcifying tendinitis of the supraspinatus tendon verified by X‐rays ≥ 2 trials of different conservative treatments had to be attempted before recruitment Exclusion criteria: history of malignant tumours local skin conditions radiologically verified osteoarthritis of the shoulder joint cardiac pacemakers cervicobrachial syndrome pregnant women Baseline characteristics: Low‐dose ESWT (22 participants): mean (SD) age: 49.38 (8.37) years mean (SD) pain, VAS 0–100: 69.95 (14.47) mean (SD) function, Constant score 0–100: 49.71 (14.47) calcification size: not reported High‐dose ESWT (25 participants): mean (SD) age: 53.57 (8.80) years mean (SD) pain, VAS 0–100: 65.57 (22.37) mean (SD) function, Constant score 0–100: 48.04 (11.54) calcification size: not reported Pretreatment group differences: none
Interventions	Low‐dose ESWT: description of modality used: navigated and X‐ray‐assisted, focused shock wave treatment delivered by a lithotripter (Storz Modulith SLK, Storz Medical Products Kreuzlingen, Switzerland). The calcium deposit was localised with a 3‐dimensional localisation device using X‐rays Lithotrack (Storz Medical Products Kreuzlingen, Switzerland) method of administration: the calcium deposit was located in the centre of a crosshair by fluoroscopy in 2 planes. The computer calculated the angle and distance for achieving maximum precision, and the distance from the shock wave focus to the deposit was stated in millimetres on a monitor on the navigation device. A contact gel (Gerosonic, Geropharmazeutica Vienna, Austria) was used at the interface of lithotripter and skin. The therapy was delivered without local anaesthesia dose: 0.08 mJ/mm² by 1000 impulses frequency: 3 sessions at weekly intervals co‐interventions: none High‐dose ESWT: description of modality used: as above method of administration: delivered with subacromial anaesthesia of Xyloneural 5mL (Gebro Pharma GmbH, Austria) was given under sterile conditions. The anaesthetic was infiltrated dorsally to keep the puncture area at a safe distance from the interface of skin and lithotripter dose: 0.2 mJ/mm² by 2000 impulses frequency: 2 sessions at weekly intervals co‐interventions: none
Outcomes	Measured at baseline and 12 weeks Outcomes included in review: function assessed by CMS 0–100 with a higher score indicating better function pain assessed by VAS 0–100, higher score indicating more pain calcification size: number of participants with complete resorption of calcification on X‐ray (grade I on a 4‐point grading scale) treatment success: proportion of participants with function Constant score > 85 (i.e. excellent result) adverse events withdrawals due to adverse events, intolerance to treatment or other reasons Outcomes excluded from review: calcification size: proportion of participants with grades 2, 3 or 4 on a 4‐point scale treatment success: proportion of participants with pain VAS < 15 (i.e. excellent result)
Source of funding	Not reported
Notes	Trial registration: not registered Time points included in review: 12 weeks Data analysis: pain, function, calcification resorption and treatment success extracted at 12 weeks. The study contact was e‐mailed to request further information on the methods of allocation concealment, and their response was used to guide the risk of bias assessment Withdrawals: 4/22 in low‐dose group (2 excluded due to strong pain during therapy, 1 had urgent personal reasons, 1 was lost to follow‐up) and 2/25 in high‐dose group due to loss to follow‐up Adverse events: Low‐dose ESWT: serious adverse events: 0/22 other adverse events: 0/22 High‐dose ESWT serious adverse events: 0/25 other adverse events: 0/25
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "The randomization into two groups was performed after every ten consecutive patients were enrolled, thus a total of five randomization procedures were carried out. Patients' names were written on cards that were put into envelopes, mixed and randomised."
Allocation concealment (selection bias)	Unclear risk	Response from study team: "One of the nurses, working with us in the treatment rooms wrote the names on cards, which were put in envelopes and sealed and put in a cup. As noted in the paper, as soon as ten patients were collected the nurse chose randomly 5 envelopes which were assigned to Group one, and the remaining 5 to Group two. The patients were recruited by the outpatient clinics and were consecutively included, meaning: the first eligible patient´s name was put in the envelope, – put in the Cup, – Cup with ten names,‐ random Distribution 5 vs 5." Comment: no information was provided on whether the envelopes used were opaque and sealed.
Blinding of participants and personnel (performance bias) All outcomes	High risk	The study described itself as an "observer‐blinded" study and participants were not blinded to treatment allocation. Quote: "The treatment room was the same for both groups, but patients were scheduled at different times so that individuals within the groups would not contact each other." Comment: as the 2 study groups differed in session number, dose and presence of anaesthesia, it was difficult to assess how the treatment results would have been affected.
Blinding of outcome assessment (detection bias) Self‐reported outcomes	Unclear risk	As there was no report of participant blinding, there was a risk of bias in self‐reported outcomes of pain, function and treatment success.
Blinding of outcome assessment (detection bias) Assessor‐reported outcomes	Low risk	Quote: "The clinical follow‐up examination was carried out by an independent observer who had no information about the treatment protocol. X‐rays were evaluated by an independent observer."
Incomplete outcome data (attrition bias) All outcomes	Low risk	6/50; 4/22 (18%) in low‐dose group (2 excluded due to strong pain during therapy,1 urgent personal reason, 1 loss to follow‐up) and 2/25 (8%) in high‐dose group due to loss to follow‐up.
Selective reporting (reporting bias)	Low risk	Comment: no published study protocol, but results were reported for all outcomes as mentioned in methods. There is, therefore, a low risk of reporting bias.
Other bias	Low risk	Comment: no other biases apparent.