The question came up at stroke conference today, and here’s a link to the American College of Chest Physicians guideline that addresses the issue. On the basis of a meta-analysis of relatively low-quality research reports, they recommend, for patients at low risk of bleeding complications, employing dual therapy:
Beyond just providing this reference, I wanted to highlight the relative uses of references such as Up to Date, clinical practice guidelines, and the primary literature. For example, if you enter “Dual Antithrombotic Mechanical Valve” into a Pub Med search box, here’s what you get:
Just 6 hits, none of which address our question. If you make the wording broader, you’ll get more hits–“Aspirin Warfarin Mechanical Valve” delivers 91 of them–but again, not the specific information we were seeking. If we’re going to use Pub Med to research this question, we’ll need to construct the query much more carefully, using Medical Subject Headings and filters.
On the other end of the spectrum are general references such as Up to Date. These are much better places to start such a query, but I was troubled a bit by the suggestion that we had found the “right” answer because Up to Date recommended dual therapy. Remember that Up to Date is essentially an online textbook. A very good resource, but really a starting place and not the final authority when it comes to clinical questions.
One of the best things about Up to Date and similar resources is that it provides the references for us to go look up the source literature. These references include clinical practice guidelines such as the one from ACCP quoted above, as well as the individual studies and meta-analyses on which the guidelines are based. When you make a clinical decision based on a clinical practice guidelines published by a recognized authority, you’re on firmer ground than when just relying on Up to Date. The ACCP guidelines on antithrombotic therapy are indeed quite reputable.
Even better is to at least take a look at the underlying studies, references to which will be in the guideline. In this case, the guideline cites (references 137-145) 9 papers addressing the issue of dual therapy and combined their results in what appears to be an ad hoc meta analysis. Importantly, table 12 is very up front that the quality of these papers is so-so, due to bias, unclear outcome definitions, etc., and gives the recommendation a grade of 1B. So, when we decide to employ dual therapy, we should do so with the knowledge that the strength of the evidence is relatively weak.