Aspirin has long been known to prevent strokes and heart attacks in patients with a previous stroke or heart attack.  But aspirin has potentially serious side-effects.  Aspirin can cause stomach ulcers, and it inhibits blood clotting raising the risk of life-threatening bleeding.

If we knew in advance that a patient was going to be in a car accident or have a bleeding stomach ulcer, we would discontinue the aspirin a week before the event and minimize the bleeding risk.  (This is exactly what we do in anticipation of routine surgery.)  But of course such events don’t herald themselves, so doctors are left reacting to adverse events after they occur.  If a patient has life-threatening bleeding we stop the aspirin and consider that the risk may outweigh the benefits.  If the patient months or years later has a stroke we reconsider restarting the aspirin.  But this strategy is irrational.  What is needed is a way to balance risks and benefits of aspirin based on the likelihood of future events, regardless of which event happened most recently.

A study published in this issue of Annals of Internal Medicine examines the wisdom of the current practice of discontinuing aspirin after bleeding from stomach ulcers.  The study followed 156 patients who had been taking aspirin for stroke or heart attack prevention and developed bleeding stomach ulcers.  All patients had endoscopy to determine the site of bleeding and to stop the bleeding.  They were all then started on acid suppressing medication to decrease the risk of future bleeding.  Half the patients were randomized to continue 80 mg of aspirin daily and the other half to placebo.

The patients were followed for 8 weeks to test whether the patients on aspirin had more recurrent bleeding than those on placebo.  The hope was that the acid blocking medication would make the aspirin safe.  It didn’t.  Significantly more patients had bleeding on aspirin than on placebo.  But surprisingly, more patients died on placebo than on aspirin.  The reason was that more patients had strokes and heart attacks on placebo.

This study is too small to reach definitive conclusions, but its results should rattle our current thinking.  Rather than stop aspirin because an adverse effect occurs, the right course may be to remember why we recommended aspirin in the first place.  After all, strokes and heart attacks are much harder to fix than bleeding ulcers.

Learn more:

Annals of Internal Medicine article:  Continuation of Low-Dose Aspirin Therapy in Peptic Ulcer Bleeding

Annals of Internal Medicine editorial:  Aspirin Withdrawal in Acute Peptic Ulcer Bleeding: Are We Harming Patients?

Tangential miscellany:

My post last week, Antidepressants for Mild Depression May Not Help Much, generated many interesting comments, some from the handful of psychiatrists and psychologists who are my patients.  Many comments pointed out important limitations of the study I wrote about.  Two psychiatrists pointed me to the following New York Times articles which make the case that antidepressants are more beneficial than the study I cited suggests.  I’m grateful to all who wrote to me.

New York Times article:  Before You Quit Antidepressants …

New York Times Op-Ed:  The Wrong Story About Depression