Should I give digoxin to patients with heart failure if they are in sinus rhythm?
If so, to whom? Are there dangers to stopping it once started?

This is a very contentious issue. It is well known that the only prospective trial that was powered for mortality, failed to show that digoxin could lessen deaths.1 On the other hand, hospitalisation from all causes, including cardiovascular, was reduced by 6%. Personally, bearing in mind all the hazards of digoxin, I would rather add to the basic diuretic-ACE inhibitor therapy, spironolactone in a low dose (25mg daily). The latter improves mortality substantially, as shown in the RALES study.

Or, if I had the patience and skill, and the patient is haemodynamically stable, I would add a beta blocker such as bisoprolol, metoprolol or carvedilol, starting in a very low dose given to a haemodynamically stable patient and working up the dose over 2 to 3 months. Any doubt about the mortality benefit of beta blockade has been removed by the recent CIBIS study.3 If after all this I was still looking for further improvement, I would certainly add digoxin but take great care to avoid overdosing, which can be fatal, especially in the presence of a low plasma potassium level. Once I had started digoxin, I would not hesitate to stop it if toxicity were suspected. But if the patient came to me already taking digoxin with a low therapeutic blood level, and seemed to be doing well, then I would not stop the drug. The problems with digoxin withdrawal suggested by the withdrawal trials such as RADIANCE is that they merely show that patients who do well while on digoxin, should not have it withdrawn.4 These are nonrandomised trials and give no information on how the patients reacted to the addition of digoxin. For example, to take an extreme case, if digoxin had potentially adverse effects, and actually killed patients, such an increase of mortality could not be detected by assessing the effects of withdrawal of the drug from the survivors.

No comments: