Toxicology, Hematology

Bleeding Risk after NSAID Use

A 27-year-old male presents to the emergency department with abdominal pain and rib pain after colliding with a tree at the local ski resort.

The patient is tachycardic at 115, his other vital signs are stable. The patient has no significant past medical history or allergies. Point-of-care ultrasound reveals an unremarkable E-FAST exam. He also endorses a headache and a minor laceration to the left arm. You want to address his pain and consider non-steroidal anti-inflammatory drugs (NSAIDs) as an option.

NSAIDs are the most commonly used class of drugs in the world. They possess analgesic, antipyretic, and anti-inflammatory properties. Most NSAIDs inhibit COX-1 and COX-2 activity, although some are selective for COX-2. COX-1 and COX-2 regulate production of certain prostaglandins, COX-1 regulates prostaglandins that protect the gastrointestinal lining and activate platelets. COX-2 regulates prostaglandins geared more towards mediating the body’s inflammatory response to injury and protecting against vascular ischemia.1 It is estimated that between 5-7% of hospital admissions are related to medication side effects. Among those, 11-12% of admissions are from gastrointestinal, nervous system, renal, or allergic effects of non-aspirin NSAIDs.2 The GI side effects are most apparent and well documented and can increase risk of bleeding.

Anti-platelet Effect
Aspirin, one of the most widely used NSAIDs, is commonly prescribed by physicians for its anti-platelet activity through irreversible inhibition of COX-1. The COX-1 enzyme is found within platelets, it helps to produce prostaglandin H2 which is then converted into thromboxane A2, a necessary platelet adhesion factor. When COX-1 is inhibited, so is the creation of thromboxane A2 and thus platelet aggregation.3 Low dose aspirin irreversibly acetylates and inhibits COX-1 and only weakly inhibits COX-2. Since platelets cannot make new COX-1, the inhibition of TxA2 is permanent for the life of the platelet.4 Because of this phenomenon, aspirin is most commonly prescribed for vascular disease including coronary artery disease and cerebrovascular disease to prevent thrombosis. It has been shown to decrease risk of non-fatal MI by 21%, but on the other hand, has shown to increase the risk of bleeding by 48% due to its potent anti-platelet activity.5 Aspirin reversal time is roughly 7 days due to average lifespan of platelets being between 7 and 10 days. When the patient stops taking aspirin, there will still be irreversibly inhibited platelets in the body until they have all been replaced 7-10 days after discontinuation. By 3-4 days after discontinuation of aspirin, there are enough functional platelets to guarantee suitable hemostasis.6

Among non-aspirin NSAIDs, the complete reversal time to get full platelet function back to normal varies. For example, ketorolac is 2 days and ibuprofen is 1 day for complete reversal; these are shorter intervals due to the reversibility of COX-1 inhibition. This allows the platelets to more quickly return to normal functionality, as they don’t have to be completely recycled.6 Note that paracetamol, also known as acetaminophen, is listed even though it is not an NSAID. It inactivates a site different than the aspirin-binding site on COX-1 and there is some controversy over whether this drug has an effect on platelets or not. Generally, it is accepted that its effect is not well established and if there is an effect, it is much less potent.7

NSAID Bleeding Risk: Nonselective vs Selective NSAIDs

Nonselective COX-1 and COX-2 Inhibitors
Aside from the anti-platelet effects previously discussed, NSAID enteropathy can also increase bleeding risk. This side effect may not be recognized to its full extent, as some researchers claim that as much as 53-80% healthy short term NSAID users and a similar percentage in long term users for > 3 months8 are affected. Anyone with gastrointestinal tissue damage from NSAID use is at a higher risk of bleeding. COX-1 and COX-2 enzymes reside in different areas of the body, and because of this, the prostanoids they each help to create exhibit different effects. COX-1 enzymes reside in the gastrointestinal epithelial cells, when inhibited, this blocks the production of prostaglandins in these epithelial cells that are shown to protect the mucosal lining of the GI tract, when their protective effects are diminished, this leaves the intestinal epithelium susceptible to damage. Ulcers and erosions can form and eventually damage small capillaries causing a slow but steady bleed that can continue for some time if not treated.1

In a study on GI bleeds, 3 groups of elderly participants received a 2-week NSAID regimen for arthritis. Each group received a different NSAID: diclofenac, naproxen, and piroxicam. Fecal blood loss was then measured.

"The mean 24-hour fecal blood loss with diclofenac (0.53 mL±0.21) was not significantly different from control (0.28 mL±0.06), whereas it was significantly increased with naproxen (2.76 mL±2.22) and piroxicam (1.16 mL±0.62), p= 0.0013"9

Multiple hypotheses have been considered based on these data: COX-1 selective NSAIDs (naproxen, piroxicam) may pose a higher bleeding risk than the more COX-2 selective diclofenac, and longer reversal times correlated with increased bleeding risk.

Diclofenac has been shown to have comparable COX-2 selectivity to celecoxib. However, multiple studies have shown that celecoxib and etoricoxib are associated with less GI toxicity than diclofenac and thus would be a much better comparison to the above NSAIDs in the study to demonstrate the correlation between COX-2 selectivity and decreased bleeding risk.10 Diclofenac, although selective, is associated with even more GI toxicity than some nonselective NSAIDs.11

There is a correlation between longer reversal time and increased bleeding risk. Both naproxen and piroxicam have longer reversal times than diclofenac and resulted in more fecal blood loss. Two of the patients from the above study that were on the piroxicam trial were withdrawn due to extensive GI bleeding.

Selective COX-2 inhibition
The high majority of COX-2 enzyme expression resides in the kidneys, brain, monocytes, and macrophages,12 thus they pose a decreased risk of GI bleeding from mucosal injury and platelet malfunction.13 Since the anti-platelet effect and enteropathy is only from the inhibition of COX-1 within the platelet and GI epithelial cells, the development of selective COX-2 inhibitors was meant to diminish these effects while still providing anti-pyretic and analgesic benefits.3 One meta-analysis of 35 studies regarding the correlation between COX-2 inhibitors and risk of bleeding concluded that, "Highly selective COX-2 inhibitors did not significantly increase the risk of intraoperative, postoperative bleeding, or blood loss. They also had no significant effect on platelet function. Therefore, perioperative, single dose, or short course of COX-2 inhibitors can be safely used in individuals who are undergoing surgery."14 This meta-analysis included mostly studies on short term or single dose COX-2 inhibitors thus the evidence on bleeding risk and long-term COX-2 is not concrete. This analysis also reported that high dose COX-2 inhibitors administered with aspirin increase the risk of bleeding postoperatively and this combination is not recommended. The American College of Surgeons does not include selective COX-2 inhibitors in their guidelines on perioperative cessation of anti-platelet agents, the only NSAID with any preoperative guidelines is aspirin.15 However, although COX-2 inhibitors have fewer anti-platelet effects, research has also shown that major coronary events are increased by about one third when using coxib drugs or diclofenac.16

Mitigating GI Bleeding Risk
All NSAIDs carry varying degrees of bleeding risk. The decision of what to use should be based on each unique patient. Increased bleeding risk is seen in patients with the following: age > 60, history of peptic ulcer disease or GI bleeds, diabetes, or hypercholesterolemia. Use of ASPIRING, clopidogrel, ticlopidine, alendronate, or corticoids have also shown to increase risk.17 Interestingly, SSRIs have also shown to increase risk.18 If a patient has any of these risk factors, consider co-administration of mitigating medications and/or use of non-NSAID analgesia.

One study done on mitigating NSAID related GI damage, studied the use of misoprostol and proton pump inhibitor (PPI) medications concurrently with NSAIDs to reduce risk. They found that 51% of patients who took a placebo with NSAID remained ulcer free after 12 weeks, 93% who took Misoprostol were ulcer free, and 80% who took lansoprazole were ulcer free. The side effects of Misoprostol, namely diarrhea, caused a significant number of participants to withdraw from the study and when this was taken into consideration, misoprostol and lansoprazole were comparable in their effectiveness at decreasing ulcer risk.19 It is proposed that decreasing ulcer risk will decrease risk of bleeding.

Can you give NSAIDs to preoperative patients?
Research is conflicting, there are studies that show an increased bleeding risk, and others that show no significant correlation. One such study found "perioperative ASPIRING use does not increase hemorrhagic complications in elective intracranial tumor surgery."20 This study had 3 groups of patients, including those who were not taking aspirin, those who were taking 81 mg, and those who were taking 325 mg. Another study found that preoperative aspirin continuation reduced venous thromboembolism risk at the expense of increased preoperative major bleeding.21 These, and many other studies emphasize the fact that the answer is unclear on whether or not to continue anti-platelet agents such as aspirin, and that further research is needed. The general consensus is that these decisions should be made by the surgical team with careful consideration of each individual patient’s bleeding risks based on medical history, age, type of surgery, medications and more. Even with these conflicting studies, the American College of Surgeons still recommends discontinuing aspirin 7 days prior to surgery.15

Case Conclusion
Generally speaking, the use of NSAIDs increase the risk of bleeding. Aspirin possesses the greatest platelet inhibition for the longest period of time (7 days) while ibuprofen and diclofenac (1 day) and celecoxib (0 days) have significantly less. In patients presenting to the ED with acute pain, NSAIDs are the most commonly used analgesic and often appropriate. In patients with active bleeding, NSAIDs should be avoided. In patients with gastrointestinal disease or risk factors, careful consideration should be made whether or not to use NSAIDs. If the physician elects to use NSAIDs, then they should consider co-administering a PPI, H2-blocker or misoprostol. Alternatively, selective COX-2 inhibitors should be considered. In patients presenting to the ED with an acute surgical emergency, the use of NSAIDs should be avoided unless risk-benefits are discussed with the surgical team. In all cases, when prescribing or using NSAIDs, physicians should use the "safest" NSAID, at the lowest dose, for the shortest period of time.

The 27-year-old male was administered ketorolac for pain relief and an appropriate improvement was noted. CT scan of the head revealed no acute findings and it was assumed that his acute symptoms were due to multiple contusions to the head and torso on initial impact. The small laceration was repaired. Upon questioning, it was noted that the patient had no additional risk factors or significant medical history. After several hours under hospital care, the patient was discharged. He was given ibuprofen 400 mg PRN for pain and told to follow up with his primary care physician if his symptoms did not continue to improve.


  1. Ornelas A, Zacharias-Millward N, Menter DG, et al. Beyond COX-1: the effects of aspirin on platelet biology and potential mechanisms of chemoprevention. Cancer Metastasis Rev. 2017;36 (2):289-303.
  1. Pirmohamed M, James S, Meakin S, et al. Adverse drug reactions as cause of admission to hospital: prospective analysis of 18 820 patients. BMJ. 2004;329:15.
  2. Vitale P, Panella A, Scilimati A, Perrone MG. COX-1 Inhibitors: Beyond Structure Toward Therapy.Med Res Rev. 2016;36(4):641-671.
  3. Phillips DR, Conley PB, Sinha U, Andre P. Therapeutic approaches in arterial thrombosis. J Thromb Haemost. 2005 Aug;3(8):1577-89.
  4. Christiansen M, Grove EL, Hvas AM. Primary Prevention of Cardiovascular Events with Aspirin: Toward More Harm than Benefit-A Systematic Review and Meta-Analysis.Semin Thromb Hemost. 2019;45(5):478-489.
  5. Llau JV, Lopez-Forte C, Sapena L, Ferrandis R. Perioperative management of antiplatelet agents in noncardiac surgery.Eur J Anaesthesiol. 2009;26(3):181-187.
  6. Driver B, Marks DC, van der Wal DE. Not all (N)SAID and done: Effects of nonsteroidal anti-inflammatory drugs and paracetamol intake on platelets. Res Pract Thromb Haemost. 2019 Nov 26;4(1):36-45.
  7. Srinivasan A, De Cruz P. Review article: a practical approach to the clinical management of NSAID enteropathy.Scand J Gastroenterol. 2017;52(9):941-947.
  8. Scharf S, Kwiatek R, Ugoni A, Christophidis N. NSAIDs and faecal blood loss in elderly patients with osteoarthritis: is plasma half-life relevant?.Aust N Z J Med. 1998;28(4):436-439.
  9. Walker C. Are All Oral COX-2 Selective Inhibitors the Same? A Consideration of Celecoxib, Etoricoxib, and Diclofenac [published correction appears in Int J Rheumatol. 2019 Jun 2;2019:8635073].Int J Rheumatol. 2018;2018:1302835. Published 2018 Dec 9.
  10. Massó González EL, Patrignani P, Tacconelli S, García Rodríguez LA. Variability among nonsteroidal antiinflammatory drugs in risk of upper gastrointestinal bleeding.Arthritis Rheum. 2010;62(6):1592-1601.
  11. Patrono C. Cardiovascular effects of cyclooxygenase-2 inhibitors: a mechanistic and clinical perspective.Br J Clin Pharmacol. 2016;82(4):957-964.
  12. Brune K, Patrignani P. New insights into the use of currently available non-steroidal anti-inflammatory drugs.J Pain Res. 2015;8:105-118. Published 2015 Feb 20.
  13. Teerawattananon C, Tantayakom P, Suwanawiboon B, Katchamart W. Risk of perioperative bleeding related to highly selective cyclooxygenase-2 inhibitors: A systematic review and meta-analysis.Semin Arthritis Rheum. 2017;46(4):520-528.
  14. Hornor MA, Duane TM, Ehlers AP, et al. American College of Surgeons' Guidelines for the Perioperative Management of Antithrombotic Medication.J Am Coll Surg. 2018;227(5):521-536.e1.
  15. Coxib and traditional NSAID Trialists' (CNT) Collaboration, Bhala N, Emberson J, et al. Vascular and upper gastrointestinal effects of non-steroidal anti-inflammatory drugs: meta-analyses of individual participant data from randomised trials.Lancet. 2013;382(9894):769-779.
  16. Chi TY, Zhu HM, Zhang M. Risk factors associated with nonsteroidal anti-inflammatory drugs (NSAIDs)-induced gastrointestinal bleeding resulting on people over 60 years old in Beijing.Medicine (Baltimore). 2018;97(18):e0665.
  17. de Abajo FJ, García-Rodríguez LA. Risk of upper gastrointestinal tract bleeding associated with selective serotonin reuptake inhibitors and venlafaxine therapy: interaction with nonsteroidal anti-inflammatory drugs and effect of acid-suppressing agents.Arch Gen Psychiatry. 2008;65(7):795-803.
  18. Graham DY, Agrawal NM, Campbell DR, et al. Ulcer prevention in long-term users of nonsteroidal anti-inflammatory drugs: results of a double-blind, randomized, multicenter, active- and placebo-controlled study of misoprostol vs lansoprazole.Arch Intern Med. 2002;162(2):169-175.
  19. Hanalioglu S, Sahin B, Sahin OS, et al. Effect of perioperative aspirin use on hemorrhagic complications in elective craniotomy for brain tumors: results of a single-center, retrospective cohort studyJ Neurosurg. 2019;132(5):1529-1538. Published 2019 Apr 5.
  20. Wolff G, Navarese EP, Brockmeyer M, et al. Perioperative aspirin therapy in non-cardiac surgery: A systematic review and meta-analysis of randomized controlled trials.Int J Cardiol. 2018;258:59-67.

Related Articles

Small, but Deadly: Part 2

Small, but Deadly: Part 2 Remember that 22-month-old girl from the last issue? Well, armed with the information from the last article, you now  feel ready to treat any of her potential medication ing

Bleach Ingestion: To Scope or Not to Scope?

Bleach can be found in most homes and workplaces across America. This broad, easy availability makes it a candidate for potentially toxic ingestions. When should you assess for structural damage?