Table 1

Methodological characteristics of included studies

First author— tendinopathyStudy typeRandomisation methodBlinding methodAllocation concealmentStatistical power calculationBaseline comparisonInclusion criteriaExclusion criteriaFollow-up completion
Brox36
Shoulder
Randomised controlled trialRandom permuted blocksPatients wearing t-shirts at follow-up to hide scar; patients asked not to talk to assessor about treatmentYes, 90%Less women in surgery groupAge 18–66 y, shoulder pain for >3 m, resistant to physio and drugs, dysfunction/pain on abduction, normal passive ROM, pain during two of the three isometric-eccentric tests, positive impingement testsAcromioclavicular joint arthritis, cervical syndrome, rotator cuff rupture, glenohumeral instability, bilateral muscular pain with tenderness and severely decreased ability to relax the shoulder/neck/temporomandibular joint on examination, reluctant to accept one or more of the treatment regimens of the study99%
Rahme37
—Shoulder
Randomised controlled trialBlocked randomisationIndependent assessorNoNo comparisonIsolated shoulder disease, working age, shoulder pain >1 y at rest accentuated by movements involving elevation, positive Hawkins (impingement) sign, positive impingement test (relief of symptoms within 15 min of injection of local anaesthetic)Glenohumeral osteoarthritis, those requiring resection of the lateral end of the clavicle93%
Rompe30
Shoulder
Randomised trial (non-controlled)Based on whether reimbursement for ESWT was approved by insurance companyNot blindedNo differencesCalcareous deposit on standard AP radiographs of a diameter of at least 10 mm; the morphological features of the deposit had to be homogeneous in appearance and with well-defined borders, or inhomogeneous in structure with sharp outline or homogeneous in structure with no defined border, shoulder pain for more than 12 m, clinical signs of subacromial impingement, unsuccessful conservative therapy in the previous 6 m, no evidence of bone-related anatomic outlet impingement or functional outlet impingement as seen on radiographs or MRICloudy and transparent appearance of deposit, radiological signs of spontaneous resorption, evidence of type III acromial morphological feature according to Bigliani et al on the outlet view of the acromion, evidence of subacute acromial bursitis, evidence of acromial spur or acromioclavicular osteophytes on AP radiographs, evidence of rotator cuff tears on MRI, evidence of functional impingement of rotator cuff on USS or arthroMRI, tears of the glenohumeral ligaments of the labrum, hypertrophy of supraspinatus muscle, dysfunction in neck or thoracic region, prior shoulder surgery, local degenerative disease of shoulder, RA, neurological abnormalities of the upper limb with calcifying tendinitis, pregnancy, infection, tumour75%
Haahr35
Shoulder
Randomised controlled trialComputer-generated random sequence generationNot blindedSealed envelopesYes, 80%More sick leave due to more severe shoulder pain in surgery groupFulfilment of all diagnostic criteria (shoulder pain, pain on abduction with painful arc, Hawkins sign, positive impingement test (relief of symptoms within 15 min of injection of local anaesthetic), symptoms 6 m–-3 y, age 18–55 y, normal passive glenohumeeal movementsImpaired glenohumeral rotation, history of acute trauma, previous surgery of fracture near affected shoulder, known OA in glenohumeral or acromioclavicular joint, clacifications >2 cm in rotator cuff tendons, rupture of the cuff, cervical root syndromes93%
Ketola29
Shoulder
Randomised controlled trialComputer-generated numbersIndependent physiotherapist conducted 5-year assessment; participants wearing t-shirts to cover scars and asked not to indicate treatment groupRandomisation by independent statistician;
sealed envelopes
Yes, 80%No difference in outcome measures, no comparison of demographicsClinical symptoms of shoulder impingement, positive Neer’s test, symptoms for at least 3 m, failed management with rest, NSAIDs, steroid injections and regular physiotherapy, age 18–60 y, no previous shoulder operations, willingness and capacity to comply with study protocolOA, glenohumeral instability, penetrating rotator cuff rupture, cervical radiculopathy, adhesive capsulitis, shoulder neuropathy78%
Farfaras28—ShoulderRandomised controlled trialEnvelopes divided in boxes based on sex and ageIndependent physiotherapist conducted assessment; participants encouraged to wear a t-shirt at follow-up to conceal their scarYes, sample size not enough for 80% powerNo differencePositive Neer and Hawkins tests, failed conservative management, subacromial pain for more than 6 mDiabetes mellitus, neurological or spine disorders, radiographic OA, chronic joint disorders, full-thickness rotator cuff tear, subacromial impingement syndrome stage 363%
Beard34
Shoulder
Randomised controlled trialAutomated computer-generated minimisation systemDouble blinded except ‘no treatment’ group;
masked assessment
Centralised telephone randomisation centre usedYes, 90%No differenceSubacromial pain for at least 3 m, consultant’s clinical diagnosis, eligible for arthroscopic surgery, completion of conservative management programme including physiotherapy and at least one steroid injectionFull-thickness rotator cuff tear, other shoulder pathology identified on MRI or USS, previous shoulder surgery on affected side, RA or other inflammatory joint conditions, cervical spine pathology, previous septic arthritis in shoulder, radiotherapy in same side as affected shoulder, lacking consent, cognitive impairment or language issues, unable to perform clinical assessments, >75 y of age81%
Bahr27
Patella
Randomised controlled trialRandomisation sequence in blocks of four created by statistician; those who had failed eccentric strengthening were allocated to secondary surgery groupNot blindedSealed envelopes;
randomisation sequence created by statistician
Yes, 90%No differencesHistory of exercise-related pain in proximal patellar tendon or patellar insertion and tenderness to palpation, pain during and after activity and unable to participate in sports at same level as before onset of pain, thickening and increased signal intensity on MRIHistory of knee/patellar tendon surgery, inflammatory or degenerative joint condition, less than 18 y of age, inability to understand oral and written Norwegian88%
Alfredson33
Achilles
Randomised trial (non-controlled)Box with envelopesNot blindedOpaque envelopesNoNo comparison as information not provided95%
Radwan31
Lateral elbow
Randomised trial (non-controlled)Closed envelopesNot blindedClosed envelopesYes, 80%Characteristics of two groups presented but comparison not performedEstablished diagnosis of lateral epicondylitis with failure of conservative Tx for 6 m (NSAIDs, steroid injections, physical therapy, exercise programme, elbow brace)
Pain induced by >1 of palpation of lateral epicondyle, resisted wrist extension, chair test
Younger than 18 y, local infection, malignancy, elbow arthritis, generalised polyarthritis, ipsilateral shoulder dysfunction, neurological abnormalities, radial nerve entrapment, cardiac arrhythmia, pacemaker, steroid injection last 6 w, pregnancy89%
Kroslak22
Lateral elbow
Randomised controlled trialComputer-generated codeDouble blindedSealed, unmarked envelopesYes, 90% but not enough participant recruited>18 y of age, clinical diagnosis lateral epicondylitis (point tenderness over lateral epicondyle and worse pain with chair pick-up test and maximal hand grip), failed conservative therapy for 6 m (including injections)Previous surgery or dislocation of affected elbow, steroid injection in last 3 m, inadequate skin coverage over elbow, sensory/motor changes distal to elbow, unwillingness/inability to attend follow-up or enter either treatment arm85%
(42% for clinical assessment)
  • AP, antero-posterior;ESWT, extracorporeal shock wave therapy;m, months; MRI; magnetic resonance imaging; NSAIDs, non-steroidal anti-inflammatory drugs;OA, osteoarthritis;RA, rheumatoid arthritis;ROM, range of movement;USS, ultrasound scan;w, weeks;y, years.