Olujohungbe 2011

Methods	Double‐blind, placebo controlled parallel group trial with four arms. The first phase of the trial was an observational phase lasting a minimum of three weeks and a maximum of 13 weeks to give a baseline before participants were randomised to intervention or placebo. The intervention phase ran for 26 weeks.
Participants	131 participants were enrolled onto the trial. 86 participants completed the observational phase of the trial. 78 participants were randomised to 1 of the 4 treatment arms. Inclusion criteria: males aged 12 or over with confirmed sickle cell anaemia or sickle cell beta‐thalassaemia on haemoglobin electrophoresis and stuttering priapism. Participants had to be in active attendance at a designated care centre (1 visit in last 6 months). Participant characteristics were only reported for the complete group of 86 not specifically for the 46 randomised participants. Age (participants who completed phase A n = 86): mean age was 23.6 (range 14.5 ‐ 46.1). Median weekly number of attacks in phase A ‐ 1.5 (range 0 – 7). Median weekly duration of attacks in phase A – 2.0h (range 0 – 42 h). Number (%) of weekly attacks > 4h duration in phase A – 22 (26%). Average attack pain score (mean (range)) in phase A – 3.4 (0 – 8.8). Participants were excluded if there was a history of an attack lasting longer than 4 hours or who needed acute hospital. admission and intervention or history of cerebrovascular accident or hypertension.
Interventions	Group 1: participants received 15 mg ephedrine per day. Group 2: participants received 30 mg ephedrine per day. Group 3: participants received 50 mg etilefrine per day. Group 4: participants received a placebo. All treatments were taken once a day at bedtime for 6 months.
Outcomes	Weekly total number of attacks expressed as a ratio of phase B to phase A. Mean difference in pain score from phase A to phase B. Number of participants who had an attack lasting more than four hours. Adverse events (palpitations, tachycardia, lack of sleep, hand shaking, anxiety, dry mouth) experienced during the trial period. Participants were reviewed on a 6‐weekly basis when diaries were monitored and BP was taken.
Notes	None of the participants were on a chronic hypertransfusion program. One participant had been on hydroxyurea for 18 months prior to the trial.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Method of randomisation not stated.
Allocation concealment (selection bias)	Unclear risk	No description of allocation concealment.
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Blinding of the trial drug was carried out by Bilcare Ltd clinical trial supplies although the authors do not state who was blinded and what methods were used. It is a double‐blind trial.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Blinding of the trial drug was carried out by Bilcare Ltd clinical trial supplies although the authors do not state who was blinded and what methods were used. It is a double‐blind trial.
Incomplete outcome data (attrition bias) All outcomes	High risk	86 participants completed the observational phase and of these, 59% entered the intervention phase. 78 participants were randomised to the intervention phase but data was only available for 46. The remainder were lost to follow‐up. 46 participants entered the intervention phase and 25 participants (54%) completed 26 weeks.
Selective reporting (reporting bias)	Low risk	No protocol available therefore outcomes reported in the results section against the methods section of the paper; no discrepancies found.
Other bias	Low risk	None identified.