SSRI Bleeding Problems Are Real, But Rare

While selective serotonin reuptake inhibitors (SSRIs) remain the most commonly prescribed class of antidepressants, many physicians and patients express concerns about the risks of bleeding associated with these medications. It is thought that this risk is due to reduction in platelets’ serotonin, which impairs their role in clotting (Anglin et al. 2014). There is also evidence that SSRI use increases gastric acidity, promoting gastritis and peptic ulcers and associated gastrointestinal (GI) bleeding (Andrade and Sharma 2016).

Concerns about SSRIs increasing the risk for bleeding have focused on GI bleeding, perioperative bleeding, hemorrhagic stroke, and postpartum bleeding. While studies have varied in their conclusions, the overall evidence suggests that in most patients, SSRIs elevate the risk of bleeding, but the overall chance that a bleeding event occurs is still small, and these events are rarely serious. A recent meta-analysis concluded that the overall bleeding risk is increased at least 36 percent (Laporte et al. 2017). It is important to be aware of those patients who may be at the greatest risk of bleeding while taking SSRIs.

The most commonly reported bleeding problem in patients taking SSRIs is GI bleeding. While some studies have found that patients who take SSRIs are not at increased risk of GI bleeding compared with controls (Vidal et al. 2008), others have suggested SSRIs double the risk of GI bleeding (Loke et al. 2008). Meta-analyses that have combined individual studies suggest that the people taking SSRIs have about a 1.5-fold increased risk of a GI bleeding event. This rate is on par with the risk of bleeding posed by a non-steroidal anti-inflammatory drug (NSAID), such as ibuprofen. Similarly, some but not all studies suggest that people taking SSRIs may be at some increased risk of perioperative bleeding, but the absolute risk is small and unlikely to be clinically significant except in high risk patients. Whether SSRIs increase risk for hemorrhagic stroke is less clear, as depression and anxiety are both independent risk factors for hypertension and stroke.

As noted, the absolute risk of a serious bleeding event is small. One study estimated the number needed to harm for upper GI bleeding due to SSRIs as 3,177 for low-risk patients (Anglin et al. 2014). That means that over 3,000 people would have to be prescribed SSRIs before one extra case of harmful GI bleeding develops. In addition, no published studies have demonstrated any increased mortality due to SSRI-associated bleeding. For most individuals with depression, bleeding concerns should not be a factor when considering SSRIs as a treatment.

Patients with a history of abnormal bleeding, people with peptic ulcers or cirrhosis, and the elderly may be more likely to experience bleeding problems while taking SSRIs (Andrade and Sharma 2016; Anglin et al. 2014; Dall et al. 2009; de Abajo and Garcia-Rodriguez 2008). For these patients, the use of a proton-pump inhibitor (such as omeprazole) in combination with an SSRI can counteract the bleeding risk. Alternative antidepressants can also be considered. There have been fewer research studies conducted on serotonin and norepinephrine reuptake inhibitors (SNRIs), but available evidence suggests these medications do not increase the risk of GI bleeding (Cheng et al. 2015).

When prescribing, it is important to be aware of those who are at a significantly increased risk for bleeding, such as those with significant thrombocytopenia (extremely low platelet counts) and patients with disorders impairing platelet function such as von Willebrand disease. SSRIs should be used with extreme caution in patients with platelet counts below 25,000, and alternative antidepressants are recommended.

The risk of SSRI-associated GI bleeding has been shown to be even greater when SSRIs are given to patients taking other medications known to impair platelets, including aspirin, NSAIDs, and antiplatelet agents such as clopidogrel. It is probably safe to prescribe SSRIs with warfarin (Dong et al. 2017), though I would avoid prescribing fluvoxamine because of its inhibition of multiple CYP450 enzymes. There is no evidence to date of SSRIs increasing bleeding risk in patients taking the newer anticoagulants such as apixaban, dabigatran etexilate, or rivaroxaban.

One other population that deserves special mention is pregnant women, as there is the potential for SSRIs to cause excess bleeding after delivery. The evidence to date is inconclusive on whether SSRIs increase risk for postpartum hemorrhage (Kim et al. 2016), and those studies that have found an increased risk also found similar increased risk with nonserotonergic psychotropics (Heller et al. 2017, Jiang et al. 2016). As with GI and perioperative bleeding, the risk of clinically significant postpartum hemorrhage is small and should not be a reason for foregoing antidepressant prescription for a woman who is pregnant or thinking about pregnancy. For pregnant women who are taking SSRIs, I recommend against stopping SSRIs prior to delivery, as this can precipitate postpartum depression.

In conclusion, the concerns linking SSRIs to bleeding are valid, but the absolute risk of life-threatening bleeding is extremely small, except in patients already at high risk of bleeding. Before prescribing an SSRI, it is important to do a thorough review of a patient’s medical history, especially in those with diseases affecting platelets and coagulation (for example, cirrhosis) and those who may be on multiple antiplatelet agents. ■