Clinical Psychopharmacology · Paper A
SSRI bleeding risk
SSRIs (selective serotonin reuptake inhibitors) raise bleeding risk through platelets, a completely different route from how anticoagulants work, which matters most when the two are combined. A patient already on warfarin who starts an SSRISelective serotonin reuptake inhibitor, a common type of antidepressant.Read more → for depression faces a compounded bleeding risk, not because the SSRI thins blood the way warfarin does, but because it separately impairs platelet function. Clinically, aspirin, NSAIDs, anticoagulants, age and any prior gastrointestinal bleed all need reviewing before an SSRI is prescribed.
SSRIs deplete platelet serotonin and impair platelet aggregation (the clumping platelets need to form a clot), a mechanism entirely separate from anticoagulants, which act on the clotting cascade instead. Combining an SSRISelective serotonin reuptake inhibitor, a common type of antidepressant.Read more → with an NSAID or an anticoagulant multiplies gastrointestinal bleeding risk, because the two mechanisms stack rather than cancel out. E.g. an older patient taking regular ibuprofen who starts sertraline now has platelet impairment added on top of NSAID-related gastric injury, raising bleed risk from two directions at once. Gastroprotection with a PPI (proton pump inhibitor) is specifically recommended when an SSRI is combined with an NSAID, rather than for SSRIs used alone.
Imagine a patient established on warfarin for atrial fibrillation who is then started on an for depression. Their (a blood test measuring how anticoagulated a patient is) hasn't changed at all: so is the combination actually more dangerous? Yes, and understanding why requires knowing that SSRIs raise bleeding risk through a mechanism that has nothing to do with the clotting cascade that INR measures. SSRIs (selective serotonin reuptake inhibitors), the first-line antidepressant class recommended by NICE (National Institute for Health and Care Excellence) for most cases of moderate-to-severe depression, are generally well tolerated, but this one risk catches many prescribers off guard, particularly in older patients or anyone already taking blood-thinning or gastric-irritant medication.
Platelets, the small blood cells responsible for forming an initial clot at a site of vessel injury, rely on stored serotonin to help them aggregate (clump together) effectively. Unlike neurons, platelets cannot make their own serotonin: they take it up from the bloodstream via the same serotonin transporter that SSRIs block in the brain.
Because SSRIs block this transporter throughout the body, not just centrally, platelets become progressively depleted of stored serotonin over time on treatment. With less serotonin available, platelet aggregation is less effective, and the normal clotting response to vessel injury is blunted. This is a completely different mechanism from how anticoagulants such as warfarin or the direct oral anticoagulants (DOACs) work: those drugs act directly on the clotting cascade, the sequence of clotting-factor activation that stabilises a platelet plug into a durable clot. An SSRI does not "thin the blood" in that sense at all: it simply makes the first step of clot formation, platelet aggregation, less efficient, which is exactly why it will never show up as a change in .
Because an SSRI's effect on bleeding is mechanistically separate from an anticoagulant's effect, the two risks do not overlap: they add together. The warfarin patient from the opening example now has two independent reasons for impaired clotting, and the combined bleeding risk is meaningfully higher than either drug alone, even though nothing shows up on a routine clotting test to flag it. The same logic applies to NSAIDs (nonsteroidal anti-inflammatory drugs such as ibuprofen), which directly irritate the gastric mucosa and increase gastrointestinal bleeding risk on their own; adding an on top compounds that risk through the separate platelet mechanism, particularly in older patients or anyone with a prior gastrointestinal bleed.
For this reason, gastroprotection with a PPI (proton pump inhibitor, a drug that reduces stomach acid) is specifically recommended when an is prescribed alongside an NSAID, rather than being needed for SSRIs used alone in a low-risk patient. Working through it step by step: an SSRI alone carries a small absolute increase in bleeding risk; an SSRI plus an anticoagulant adds two independent mechanisms; an SSRI plus an NSAID adds a platelet effect to a direct mucosal effect; and any of these combinations in an older patient or one with a prior gastrointestinal bleed raises the baseline risk further still, which is when a PPI or an alternative antidepressant becomes a real consideration rather than a routine step.
The single most important distinction here is mechanism: -related bleeding is platelet-mediated and will not show up as an abnormal clotting screen or a raised , whereas warfarin- and DOAC-related bleeding is clotting-cascade-mediated and will show up on exactly those tests. NSAID-related gastrointestinal bleeding is a third, again separate mechanism: direct chemical irritation of the gastric mucosa rather than any effect on platelets or clotting factors. A patient who bleeds while on an SSRI alone, with normal clotting studies and no NSAID or anticoagulant on board, is showing the platelet mechanism in isolation; a patient who bleeds while on an SSRI plus warfarin is showing both mechanisms stacked, which is why the combination is flagged specifically rather than either drug alone.
There is no single diagnostic test for " bleeding risk" as such: assessment is clinical, based on a careful medication history covering anticoagulants, antiplatelet drugs (such as aspirin or clopidogrel) and NSAIDs, plus age, any history of peptic ulcer disease, and prior gastrointestinal bleeding. In a patient who presents with unexplained bruising, melaena (black, tarry stools indicating upper gastrointestinal bleeding) or haematemesis (vomiting blood) while on an SSRI, investigation follows the usual pathway for suspected gastrointestinal bleeding (full blood count, clotting screen, and endoscopy if indicated) with the SSRI considered a contributing rather than a sole cause, and the clotting screen typically returning normal precisely because the SSRI mechanism does not touch the clotting cascade.
Management is largely about prevention. Before starting an , it is worth reviewing concurrent aspirin, NSAID, or anticoagulant use, and considering gastroprotection with a PPI when an SSRI and an NSAID must be combined. In patients where bleeding risk is a particular concern (for example, someone already on aspirin for cardiovascular prevention) mirtazapine is a reasonable alternative antidepressant, since it has much less effect on platelets than SSRIs and therefore carries a lower added bleeding risk. This does not mean SSRIs should be avoided in every patient on antiplatelet or anticoagulant treatment; it means the combination should prompt a specific, documented risk discussion rather than an automatic switch, in line with Maudsley Prescribing Guidelines' general approach to interacting drug combinations.
If a vignette gives you a normal or normal clotting screen alongside unexplained bleeding in a patient on an , that normal result is the clue, not a reason to dismiss the SSRI as the cause: SSRI bleeding is platelet-mediated and simply does not alter clotting-cascade tests.
SSRIs raise bleeding risk by depleting platelet serotonin and impairing platelet aggregation: a mechanism entirely separate from, and additive with, the clotting-cascade effect of anticoagulants and the mucosal effect of NSAIDs, which is exactly why the combination matters more than any drug used alone.