Yadav 2015.

Study characteristics
Methods	Design: single‐centre 2‐arm parallel randomised controlled trial Setting: Safdarjang Hospital, New Delhi, India Timing: between October 2012 and April 2014 Interventions: platelet‐rich plasma (PRP) injection vs glucocorticoid injection Sample size: sample size calculation not done Analysis: data for those who completed the study were analysed
Participants	Number of participants Number screened: not reported Number excluded: not reported Number randomised: 65 (numbers to PRP group and glucocorticoid group not reported) Numbers included in analysis at 3 months: 60 (30 in PRP group, 30 in glucocorticoid group) Inclusion criteria Aged > 18 years Diagnosed lateral epicondylitis based on clinical signs and symptoms Only on conservative treatment (2‐week washout period was given to all those on analgesics and anti‐inflammatory drugs) or no treatment for elbow pain Exclusion criteria History of arthritis, trauma, or fracture Nerve entrapment around elbow Bleeding disorder Psychiatric disorder Baseline characteristics PRP group Mean age (years): 36.6 No. male:female: 10:20 Mean duration of symptoms (months): 2.26 Mean VAS score for pain: 7.6 Mean maximum grip strength: 74.6 Mean quick‐DASH score for function: 88 Glucocorticoid group Mean age (years): 36.7 No. male:female: 7:23 Mean duration of symptoms (months): 1.93 Mean VAS score for pain: 7.7 Mean maximum grip strength: 74.5 Mean quick‐DASH score for function: 88 Pre‐treatment group differences: no baseline differences were found between the 2 groups
Interventions	Platelet‐rich plasma (PRP) injection Participants received a single injection of PRP (1 mL), with absolute platelet count of 1 million platelets/mm³ as confirmed by manual counting. PRP was prepared under aseptic conditions as per the procedure standardised in the departmental laboratory. A 9001:2000 ISO certified R‐23 centrifuge was used for the purpose of platelet concentration. PRP was injected into the common extensor origin at the lateral epicondyle of the humerus under aseptic conditions Glucocorticoid injection Participants received a single injection of glucocorticoid (methylprednisolone, 40 mg in 1 mL). The site of injection and the technique used were the same in both groups Post intervention Only paracetamol (500 mg) tablets were allowed as rescue medication for a maximum period of 1 week in both groups
Outcomes	Outcomes were measured at baseline and at 15 days, 1 month, and 3 months Study outcomes Elbow pain measured on VAS numerical scale ranging from 0 to 10, where 0 = no pain and 10 = maximum possible pain Elbow function measured on the quick‐DASH, a shortened version of the DASH Outcome Measure. Instead of 30 items, the quick‐DASH uses 11 items to measure physical function Grip strength measured by the Jamar hydraulic dynamometer Outcomes used in this review VAS pain scores DASH function scores Proportion of participants experiencing adverse events Time pointsused in this review 1 month and 3 months
Notes	Funding: not reported Trial registration: not done Withdrawals: 5/65 were lost to follow‐up; however group‐wise numbers were not reported Adverse events PRP injection group Total adverse events No. of events = 0 Serious adverse events No. of events = 0 Glucocorticoid injection group Total adverse events No. of events = 0 Serious adverse events No. of events = 0 Data analysis: 1‐month data were entered for the 6‐week time point analysis. Study authors did not report standard deviation for the groups. We calculated SD for pain, function, and grip strength using reported P values. At 3 months, P values were reported as P < 0.0001; thus we could not calculate SD, therefore we imputed it using SD from 4‐week analysis
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Method of randomisation was not reported
Allocation concealment (selection bias)	Unclear risk	Method of allocation concealment was not reported
Blinding of participants and personnel (performance bias) All outcomes	High risk	Study authors do not describe any method of blinding, and blood was likely not drawn in the glucocorticoid group. Thus, participants must have been aware of the intervention they received. There was no report of interventionists being blinded
Blinding of outcome assessment (detection bias) for self reported subjective outcomes (pain, function, treatment success, quality of life)	High risk	As participants most likely were not blinded, there is high risk of bias in the measurement of self‐reported outcomes
Blinding of outcome assessment (detection bias) objective outcomes	Unclear risk	It is unclear whether assessors measuring grip strength were blinded
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	5/65 (7.6%) participants were lost to follow‐up, but study authors did not report group‐wise missing data. Data from these 5 participants were excluded from the analysis
Selective reporting (reporting bias)	Unclear risk	Study protocol was not available and the trial was not registered; measures of variance for study outcomes were not reported
Other bias	Low risk	None is apparent