What are authors afraid of?

The authors discuss some of the difficulties with recent randomized controlled trials (RCTs) that have not demonstrated the good outcomes our interventions were purported to deliver. The title of the editorial, that RCTs can be a ‘double-edged sword’, seems to warn the reader against something, but against what exactly?

1. Could it be that we are designing and participating in too many trials? In fact, we are collectively responsible for our field delivering poor (if any) evidence regarding the merits of our daily interventions. We need more trials, preferably trials designed and conducted by neurointerventionists. We must regain control of how to evaluate the merits of our own practice. Most importantly, if we are to offer patients care that they can trust, we must be constantly working to validate our still unvalidated interventions. What is the best way for us to do this? A trial, of course, but not just any type of trial. We will get back to this point.

2. Should we distrust the disappointing results of recent RCTs because they have ‘limitations’, as the citation from Concato1 suggests? What are we to do about trials that have design shortcomings? Should we stubbornly practice interventions that have now been shown to be harmful, albeit in trials ‘with limitations’, claiming them to be standard of care, just like an intervention that has been proven beneficial? Of course not; disappointing trial results simply mean that such interventions should only be offered within the context of better designed trials.

3. Should the readers of the editorial be warned against …