| Methods |
Randomised controlled trial
Prednisone versus azathioprine |
| Participants |
10 participants
2 participants with idiosyncratic reaction to azathioprine crossed over to prednisone in the first 4 weeks. 2 participants without improvement under azathioprine crossed over to the prednisone group after one year |
| Interventions |
60 mg prednisone for 6 weeks, tapering 20 mg on alternate days.
Azathioprine 50 mg, then increased to 2.5 mg/kg over several weeks, reduced if white blood cell count < 4000 |
| Outcomes |
4 participants in the prednisone group improved in strength, impairment and decrement.
The 2 participants in the azathioprine group who crossed over after one year improved under prednisone, one participant stable on azathioprine alone |
| Notes |
No statistical evaluation, no obvious differences in thymectomised (2 in each group) and non‐ thymectomised participants in both groups. Two participants in the prednisone group needed plasma exchange for relapses.
Endpoints not defined |
| Risk of bias |
| Bias |
Authors' judgement |
Support for judgement |
| Random sequence generation (selection bias) |
Unclear risk |
Method of randomisation not described |
| Allocation concealment (selection bias) |
Unclear risk |
Method of concealment not described |
| Blinding (performance bias and detection bias)
Observer |
Unclear risk |
It is not possible to know from the report if there was blinding or not |
| Blinding (performance bias and detection bias)
Patient |
Unclear risk |
It is not possible to know from the report if there was a blinding or not |
| Incomplete outcome data (attrition bias)
All outcomes |
High risk |
2 out of 5 participants in the azathioprine group were in fact treated with prednisone |
| Selective reporting (reporting bias) |
High risk |
Neither primary nor secondary outcomes are prespecified |
| Other bias |
High risk |
Number of participants to be included not planned and only 10 participants included. Statistical power weak |
