NSAIDs and the Elderly

NSAIDs and the Geriatric Patient: A Cautionary Tale

John J. Liantonio, MD; ŸB. Brent Simmons, MD, FAAFP


Department of Family, Community, and Preventive Medicine, Drexel University College of Medicine, Philadelphia, PA

Abstract: Nonsteroidal anti-inflammatory drugs (NSAIDs) are known to complicate many common medical problems, particularly among older adults. The risks associated with NSAIDs have been equally observed in both men and women, and several studies have shown increased hospital admissions in the elderly population because of NSAID-related complications. Some of the adverse effects reported with NSAIDs, many of which are commonly used, include an increased risk of reinfarction, gastrointestinal bleeding, atrial fibrillation, stroke, and cardiovascular death. To avoid these complications, healthcare providers need to be cautious in prescribing NSAIDs and counsel their patients regarding their risks, particularly because many of these agents can be purchased over the counter. What follows is a review of some of the most common severe complications caused by NSAIDs, with a focus on what has been reported with regard to the geriatric patient.  

Key words: NSAIDs, NSAID complications, NSAID risks, gastrointestinal bleeding, naproxen, ibuprofen.

The incidence of self-reported chronic pain seems to increase up to, but not beyond, the seventh decade of life,1 and nonsteroidal anti-inflammatory drugs (NSAIDs) are among the most commonly written prescriptions for its treatment. In the United States alone, approximately 70 million prescriptions are written annually in addition to the 30 billion tablets sold over the counter.2 These prescribed NSAIDs belong to one of two classes: nonselective and selective (Table). Nonselective NSAIDs are drugs that inhibit both types of the cyclooxygenase (COX) enzyme: COX-1, which is present in most tissues and helps regulate kidney and platelet function and protects the gastric mucosa; and COX-2, which is active in the cells that line blood vessels and is also expressed during inflammation, fever, and wound healing. Selective NSAIDs inhibit only COX-2, and these NSAIDs were developed to provide relief of pain and fever, while protecting the gastric mucosa.

NSAIDs are most frequently used to treat pain and inflammatory conditions, including osteoarthritis, rheumatoid arthritis, gout, and chronic pain. Although NSAIDs are effective pain relievers, the high risk of adverse effects on multiple organ systems, including cardiovascular, gastrointestinal, and renal systems, limit their use in the older population. They should also be avoided in patients with certain comorbidities, including hypertension, kidney disease, and cardiovascular disease. Nevertheless, they are still being prescribed to and used by such patients. In an Australian population-based study that assessed chronic NSAID use, 60.8% of users had hypertension, 30.8% had stage 3 or higher chronic kidney disease, 20.7% had a 10-year cardiovascular disease risk of more than 15%, and 17.2% had cardiovascular disease.3 These data indicate that healthcare providers need to be more conscientious in considering these comorbidities both when prescribing and counseling patients on the use of NSAIDs. 

In this article, we briefly review the epidemiology of NSAID use among older adults. We also outline the research that has demonstrated the need for caution before prescribing or recommending these agents to patients with any of the aforementioned comorbidities, all of which are more prevalent in elderly persons. 


Geriatric patients take more medications than their younger counterparts, with as many as 95% of older patients taking at least one medication4 and 46% of nursing home patients being prescribed more than nine medications.5 One study found that more than 30% of elderly patients seen by general practitioners were taking an NSAID regularly, with one third of these patients classified as chronic users (ie, they took NSAIDs for more than 1-month).6

NSAIDs have been shown to increase the risk of hospital admissions. One study examining admissions to a geriatric unit in a hospital in Italy found that NSAIDs were responsible for 23.5% of admissions related to ADRs.4 Of these events, 45.1% were defined as “definitely avoidable.” In the general population, NSAIDs are cited as being responsible for 8% of ADR-related hospitalizations, the same rate as observed for anticoagulants and analgesics.7 What follows is a review of the ADRs that can lead to hospitalization and other adverse outcomes. 

Cardiovascular System Effects 

Although the effects of NSAIDs on the cardiovascular system have been debated for many years, these agents are now widely recognized to increase cardiovascular risks. Studies have indicated that these agents can worsen hypertension and increase the risk of both myocardial infarction (MI) and atrial fibrillation, with some studies indicating that they pose a cardiovascular hazard of similar magnitude to that resulting from being a smoker or having diabetes.8

Worsening Hypertension 
One in three adults in the United States have hypertension, putting them at an increased risk for stroke and heart disease.9 The elderly population is at especially high risk of developing complications secondary to hypertension, yet 12% to 15% of these patients take both an NSAID and an antihypertensive despite this combination having been shown to increase supine blood pressure (BP) by approximately 5 mm Hg.10

In a 2012 study, investigators compared the effects of various pain relievers on BP in a group of hypertensive patients.11 Compared with patients using acetaminophen, NSAID users had a 2 mm Hg increase in systolic BP (95% confidence interval [CI], 0.7-3.3). When examining the effect of specific NSAIDs on BP, ibuprofen was associated with a 3 mm Hg increase in systolic BP compared with naproxen (95% CI, 0.5-4.6), and a 5 mm Hg increase compared with celecoxib (95% CI, 0.4-10). In addition, certain antihypertensive agents were found to exacerbate these effects, with a 3 mm Hg increase in systolic BP observed in a subgroup of patients concomitantly prescribed angiotensin-converting enzyme inhibitors or calcium channel blockers, and a 6 mm Hg increase among those prescribed a beta-adrenergic blocker. BP changes in patients prescribed diuretics or multiple antihypertensive agents did not reach statistical significance.11 

The relationship between NSAID use and hypertension has been shown in both men and women. A large prospective cohort study that included 16,031 nonhypertensive male health professionals who used NSAIDs six to seven times weekly had a relative risk (RR) of 1.38 for developing hypertension over the 4-year study period.12 A similar study that observed 51,630 nonhypertensive women aged 44 to 69 years showed a comparable RR of 1.35 for developing hypertension upon using NSAIDs for more than 22 days a month.13 

Myocardial Infarction 
In 2004, the US Food and Drug Administration pulled rofecoxib (Vioxx), a selective NSAID, from the market after evidence began showing it increased the risk of MI. Nonselective NSAIDs have also been linked to MI. When compared with no NSAID use, the hazard ratio (HR) of MI in NSAID users was 0.97 for naproxen, 1.01 for ibuprofen, 1.63 for diclofenac, and 2.01 for celecoxib.14 

In a large cohort study that investigated NSAID use after an MI, the HR for death or reinfarction was related to treatment duration.15 During the first week following an MI, the HR was 1.34 and then increased to 1.82 for days 7 to 14, 1.58 for days 14 to 30, and 1.86 for days 30 to 90. After 90 days, the HR was 1.56. A subgroup analysis stratified HR for individual NSAIDs. This analysis showed diclofenac to have the highest risk of reinfarction, which posed an HR of 3.52 during the first 7 days post-MI, decreased to 2.57 for days 7 to 14; 2.08 for days 14 to 30; 2.61 for days 30 to 90; and 2.02 for more than 90 days. Of particular note are the worsened side effect profiles in persons older than 80 years: in this subgroup, the HR increased to 5.49 during the first week. Similarly, during the first week, the HR was 2.46 for persons aged 70 to 79 years and 2.56 for persons aged 60 to 69 years. Drugs like ibuprofen and celecoxib had less dramatic results. Ibuprofen showed no statistically significant risk in week 1, but had an HR of 1.57 at days 7 to 14; 1.43 at days 14 to 30; 1.91 at days 30 to 90; and 1.52 after 90 days post MI. Celecoxib showed no statistical risk for weeks 1 and 2; however, the HR increased to 2.33 on days 14 to 30 following an MI. On days 30 to 90, HR was 1.74, and after 90 days post-MI, it increased to 1.71. Naproxen showed no statistical increase in risk.15 

A meta-analysis that reviewed the available evidence on the cardiovascular safety of NSAIDs found naproxen to be the least harmful of the assessed agents, which also included ibuprofen, diclofenac, celecoxib, etoricoxib, rofecoxib, lumiracoxib, or placebo.16 Of these agents, rofecoxib posed the greatest risk of MI, with an RR of 2.12, followed by ibuprofen with an RR of 1.61, celecoxib with an RR of 1.35, and naproxen and diclofenac with an RR of 0.82. Ibuprofen showed the greatest risk of causing a stroke, with an RR of 3.36, followed by diclofenac, naproxen, celecoxib, and rofecoxib, which had an RR of 2.86, 1.76, 1.12, and 1.07, respectively. Finally, ibuprofen and diclofenac pose an equivalent risk of cardiovascular death, with an RR of 2.39, followed by celecoxib and rofecoxib, which had an RR of 2.07 and 1.58, respectively. Naproxen showed no increase in the risk of cardiovascular death, with an RR of 0.98.16 

Atrial Fibrillation 
Another cardiac condition affected by NSAID use is atrial fibrillation, which is the most common cardiac arrhythmia seen in clinical practice.17 The prevalence of this condition is reported to double every decade of life, with 70% of patients being between the ages of 65 and 85 years.18 According to a population-based, case-controlled study that included hospitalized individuals with a new diagnosis of atrial fibrillation or flutter, use of nonaspirin NSAIDs increased the risk of these conditions, particularly among new users of these drugs. Compared with non-NSAID users, individuals taking nonselective NSAIDs before hospital admission had an RR of 1.33 for developing atrial fibrillation and those taking selective NSAIDs had an RR of 1.50. Persons with chronic renal insufficiency were also at higher risk of developing atrial fibrillation upon using NSAIDs, with an RR of 1.75 and 2.87 for nonselective and selective NSAIDs, respectively.19 

Gastrointestinal System 

One study found that more than 30% of elderly patients seen by general practitioners were taking an NSAID regularly, with one third of these patients classified as chronic users (ie, they took NSAIDs for more than 1-month).6 The study also found that the use of these drugs was associated with a greater number of upper gastrointestinal (GI) symptoms, and, subsequently, additional prescriptions for medications to treat these symptoms. 

The gastrointestinal system is directly affected by NSAIDs through several mechanisms, including the gastric mucosa, which can lead to ulcerations, gastrointestinal bleeding, and diverticular disease if destroyed. Of persons taking NSAIDs, 10% to 20% report dyspepsia. Estimates indicate there are approximately 103,000 gastrointestinal-related NSAID hospitalizations annually, with costs reaching as high as $2 billion.2 

Upper Gastrointestinal Complications 
One retrospective meta-analysis of case-control studies that examined the risks of upper gastrointestinal bleeding among NSAID users found that the RR of upper gastrointestinal complications decreased with length of time to NSAID exposure.20 Current users (ie, used an NSAID within 0-7 days of the indexed date) had an RR of 4.1, intermediate users (ie, used an NSAID within 7-30 days of the indexed date) had an RR of 1.9, recent users (ie, used an NSAID within 31-90 days of the indexed date) had an RR of 1.6, and past users (ie, used an NSAID within 91-180 days of the indexed date) had an RR of 1.0. The study found that current NSAID users who took a single NSAID had an RR of 3.9, whereas those using more than one NSAID had an RR of 11.20 

When looking at specific NSAIDs, a meta-analysis indicated that not all pose the same risk of serious gastrointestinal bleeding.21 In this study, serious upper gastrointestinal bleeding included cases of acute upper gastrointestinal hemorrhage with hematemesis or melena, which was confirmed by endoscopy, radiology, or surgery. Odds ratios (OR) of several NSAIDs were calculated, with ibuprofen being the least dangerous with an OR of 1.7, followed by diclofenac at 4.9, indomethacin at 6.0, naproxen at 9.1, piroxicam at 13.1, and ketoprofen at 34.9. The OR of every one of these aforementioned NSAIDs increased linearly with dose when looking at low, medium, and high doses. This study also examined the risk of bleeding in relation to how recently users had taken an NSAID. Recent NSAID users (ie, those taking an NSAID in the week leading up to an event) had an OR of upper gastrointestinal bleeding of 11.7. Long-term users (ie, those taking an NSAID within the week of an event and within 2 to 4 weeks of a bleeding event) had an OR of 5.6. Finally, recent users (ie, those not taking an NSAID within the week of an event, but within 2 to 4 weeks of the event) had an OR of 3.2.21 

Diverticular Disease 
Often recognized for their role in upper gastrointestinal disease, NSAIDs also play a role in the lower gastrointestinal tract, with 30% to 50% of serious gastrointestinal events associated with the colon. The most common of these colonic complications is diverticular disease.22 

The relationship between diverticular complications, such as diverticulitis or diverticular bleeding, and NSAIDs is not completely understood. One proposed relationship is NSAID-related damage to the lower colonic mucosa, similar to the effect seen on the upper gastrointestinal tract. With direct mucosal involvement and degradation of the mucosal wall due to decreased prostaglandin synthesis, NSAID use can result in diverticular complications.23 

One large prospective cohort study worked to prove this relationship by following 47,210 men for 22 years and observing both NSAID use and resulting diverticulitis and diverticular bleeding. The results showed aspirin users having an HR of 1.25 for diverticulitis and 1.70 for diverticular bleeding, compared with nonusers. Regular users of nonaspirin NSAIDs had an HR of 1.72 for diverticulitis and 1.74 for diverticular bleeding compared with nonusers.22 

In addition, patients taking NSAIDs have been reported to have a greater chance of developing complications related to diverticular disease than those not taking NSAIDs.24 In a group of hospitalized patients with known diverticular disease and perforation, there was an RR of 2.961 for those on NSAIDs versus non-NSAID users. This study also showed that in similarly matched persons, the RR of diverticular disease was 1.869 in NSAID users versus non-NSAID users.24 

Renal Disease 

NSAIDs are known to play a role in renal homeostasis. This becomes clinically significant in those with already reduced kidney function. NSAID-induced changes include sodium retention, peripheral edema, BP elevation, congestive heart failure, hyperkalemia, and acute renal failure.25 

Acute Renal Failure 
A large study that included 1799 hospitalized elderly patients aged 65 years or older found that prescription NSAID use resulted in 25 excess renal-failure–associated hospitalizations per 10,000 years of use.26 In the study, ibuprofen accounted for 35% of NSAID use, and increasing doses of these agents were associated with increased ORs, with doses less than 1200 mg/day having an OR of 0.94, doses of 1200 to 2400 mg/day having an OR of 1.89, and doses of 2400 mg/day having an OR of 2.32. Individual NSAIDs have different risks of developing acute renal failure. Naproxen has an overall OR of 1.03, diclofenac of 1.47, ibuprofen of 1.63, piroxicam of 1.95, and indomethacin of 2.40.26 

Important Guidelines 

The 2012 American Geriatrics Society Beers Criteria also cautions against use of NSAIDs by elderly patients.27 These guidelines specifically recognize the increased risk of gastrointestinal bleeding and state that chronic use of most NSAIDs should be avoided. The Beers Criteria also cautions against using indomethacin and ketorolac, citing indomethacin as being the most dangerous NSAID; thus, avoidance of both is strongly recommended.27 

In a review article by the American College of Gastroenterology,28 the role of NSAIDs and aspirin in gastrointestinal bleeding and ulceration was discussed at length. These guidelines recommend taking a thorough history before prescribing NSAIDs to assess for risk factors, including history of gastrointestinal bleeding or peptic ulcers, age older than 60 years, and use of dual antiplatelet therapy. In the case of peptic ulcers, once Helicobacter pylori is ruled out as a possible source of the ulcers, concomitant treatment with a proton pump inhibitor may be indicated, as it has been shown to reduce the risk of bleeding events in those on NSAIDs.28 The addition of misoprostol in patients on long-term aspirin therapy is also thought to be beneficial; however, given the side effect profile of this prostaglandin, it is not routinely recommended and little used.28 

What Pain Relief Options Do Practitioners Have? 

NSAIDs are commonly used to treat rheumatologic conditions like osteoarthritis, but when treating osteoarthritis in patients older than 75 years who have significant cardiac, gastrointestinal, or renal disease, the American College of Rheumatology (ACR) advises physicians to use caution.29 When treating osteoarthritis, initial recommendations include physical and occupational therapy in patients who are able to comply and participate. For pharmacologic therapy, acetaminophen is recommended with a daily dose no greater than 4000 mg. The next step would be the use topical or oral NSAIDs and/or intra-articular injectable corticosteroids. According to the American College of Rheumatology, for persons 75 years of age and older, topical NSAIDs are preferred. For those younger than 75 years, no preference for either topical or oral NSAIDs is described. Alternatively, tramadol, an injectable hyaluronic acid, should be considered. In persons with refractory pain, especially those who are poor surgical candidates or who do not want to undergo arthroplasty, opioid agents should be considered.29 

Topical NSAIDs are thought to be less dangerous when compared with oral agents. According to a double-blinded study, there was no statistical difference in adverse events in patients given topical diclofenac versus a placebo. This study also assessed for events in patients with comorbidities, such as hypertension, diabetes, and other cardiovascular disease, and saw no increase in risk associated with diclofenac.30 In a 2012 Cochrane review,31 topical diclofenac was found to be a useful alternative to oral NSAIDs for the treatment of osteoarthritis with local irritation cited as the only major adverse event. 


NSAIDs and their risks have been documented extensively in the literature. Collective analyses of these data indicate that the elderly population is at particularly high risk. Hypertension, atrial fibrillation, cardiovascular disease, upper and lower gastrointestinal disease, and kidney problems have been described. Although some NSAIDs are less harmful on some systems, they can negatively affect others, indicating there are no safe NSAIDs. For example, naproxen has been shown to have a relatively low risk of causing an MI, but a high risk of causing upper gastrointestinal bleeding. Therefore, no matter which NSAID is prescribed, caution must be used to prevent the risk of adverse events. 


1. Helme RD, Gibson SJ. The epidemiology of pain in elderly people. Clin Geriatr Med. 2001;17(3):417-431. 

2. Wolfe MM, Lichtenstein DR, Singh G. Gastrointestinal toxicity of nonsteroidal antiinflammatory drugs. N Engl J Med. 1999;340(24):1888-1899. 

3. Adams RJ, Appleton SL, Gill TK, Taylor AW, Wilson DH, Hill CL. Cause for concern in the use of non-steroidal anti-inflammatory medications in the community—a population-based study. BMC Fam Pract. 2011;12:70. 

4. Franceschi M, Scarcelli C, Niro V, et al. Prevalence, clinical features and avoidability of adverse drug reactions as cause of admission to a geriatric unit: a prospective study of 1756 patients. Drug Saf. 2008;31(6):545-556. 

5. Tamura BK, Bell CL, Lubimir K, Iwasaki WN, Ziegler LA, Masaki KH. Physicians intervention for medication reduction in a nursing home: the polypharmacy outcomes project. J Am Med Dir Assoc. 2011;12(5):326-330. 

6. Pilotto A, Franceschi M, Leandro G, Di Mario F. NSAID and aspirin use by the elderly in general practice: effect on gastrointestinal symptoms and therapies. Drugs Aging. 2003;20(9):701-710. 

7. Wasserfallen J, Livio F, Buclin T, Tillet L, Yersin B, Biollaz J. Rate, type, and cost of adverse drug reactions in emergency department admissions. Eur J Intern Med. 2001;12(5):442-447. 

8. Science Daily. NSAIDs and cardiovascular risk explained. Published May 2, 2013. Accessed May 19, 2013. 

9. US Department of Health and Human Services. Health, United States, 2008 with special feature on the health of young adults. Published March 2009. Accessed February 20, 2013. 

10. Johnson AG, Nguyen TV, Day RO. Do nonsteroidal anti-inflammatory drugs affect blood pressure? A meta-analysis. Ann Intern Med. 1994;121(4):289-300.

11. Aljadhey H, Tu W, Hansen RA, Blalock SJ, Brater DC, Murray MD. Comparative effects of non-steroidal anti-inflammatory drugs (NSAIDs) on blood pressure in patients with hypertension. BMC Cardiovasc Disord. 2012;12:93.

12. Forman JP, Rimm EB, Curhan GC. Frequency of analgesic use and risk of hypertension among men. Arch Inter Med. 2007;167(4):394-399.  

13. Dedier J, Stampfer MJ, Hanskinson SE, Willett WC, Speizer FE, Curhan GC. Nonnarcotic analgesic use and the risk of hypertension in US women. Hypertension. 2002;40(5):604-608. 

14. Fosbøl EL, Gislason GH, Jacobsen S, et al. Risk of myocardial infarction and death associated with the use of nonsteroidal anti-inflammatory drugs (NSAIDs) among healthy individuals: a nationwide cohort study. Clin Pharmacol Ther. 2009;85(2):1190-197. 

15. Schjerning Olsen AM, Fosbøl EL, Lindhardsen J, et al. Duration of treatment with nonsteroidal anti-inflammatory drugs and impact on risk of death and recurrent myocardial infarction in patients with prior myocardial infarction: a nationwide cohort study. Circulation. 2011;123(20):2226-2235. 

16. Trelle S, Reichenbach S, Wandal S, et al. Cardiovascular safety of non-steroidal anti-inflammatory drugs: network meta-analysis. BMJ. 2011;342:c7086. 

17. Heeringa J, can der Kulp DA, Hofman A, et al. Prevalence, incidence and lifetime risk of atrial fibrillation: the Rotterdam study. Eur Heart J. 2006;27(8):949-953. 

18. Feinberg WM, Blackshear JL, Laupacis A, Kronmal R, Hart RG. Prevalence, age distribution and gender of patients with atrial fibrillation. Analysis and implications. Arch Intern Med. 1995;155(5):469-473. 

19. Schmidt M, Christiansen CF, Mehnert F, Rothman KJ, Sørensen HT. Non-steroidal anti-inflammatory drug use and risk of atrial fibrillation or flutter: population based case-control study. BMJ. 2011;343:d3450. 

20. Garcia Rodriguez LA, Hernández-Díaz S. Relative risk of upper gastrointestinal complications among users of acetaminophen and nonsteroidal anti-inflammatory drugs. Epidemiology. 2001;12(5):570-576.

21. Lewis SC, Langman MJ, Laporte JR, Matthews JN, Rawlins MD, Wiholm BE. Dose-response relationships between individual nonaspirin nonsteroidal anti-inflammatory drugs (NANSAIDs) and serious upper gastrointestinal bleeding: a meta-analysis based on individual patient data. Br J Clin Pharmacol. 2002;54(3):320-326. 

22. Strate LL Liu YL, Huang ES, Giovannucci EL, Chan AT. Use of aspirin or nonsteroidal anti-inflammatory drugs increase risk for diverticulitis and diverticular bleeding. Gastroenterology. 2011;140(5):1427-1433. 

23. Lanas A, Sopeña F. Nonsteroidal anti-inflammatory drugs and lower gastrointestinal complications. Gastroenterol Clin North Am. 2009;38(2):333-352. 

24. Goh H, Bourne R. Non-steroidal anti-inflammatory drugs and perforated diverticular disease: a case-control study. Ann R Coll Surg Engl. 2002;84(2):93-96. 

25. Weir MR. Renal effects of nonselective NSAIDs and coxibs. Cleve Clin J Med. 2002;69(Suppl 1):SI53-SI58. 

26. Griffin MR, Yared R, Ray WA. Nonsteroidal antiinflammatory drugs and acute renal failure in elderly persons. Am J Epidemiol. 2000;151(5):488-496. 

27. American Geriatrics Society 2012 Beers Criteria Update Expert Panel. American Geriatrics Society updated Beers Criteria for potentially inappropriate medication use in older adults. J Am Geriatr Soc. 2012;(60):616-631. 

28. Bhatt DL, Scheiman J, Abraham NS, et al. ACCF/ACG/AHA 2008 expert consensus document on reducing the gastrointestinal risks of antiplatelet therapy and NSAID use. Am J Gastroenterol. 2008;103(11):2890-2907. 

29. Hochberg MC, Altman RD, April KT, et al. American College of Rheumatology 2012 recommendations for the use of nonpharmacologic and pharmacologic therapies in osteoarthritis of the hand, hip, and knee. Arthritis Care Res (Hoboken). 2012;64(4):465-474. 

30. Peniston JH, Gold MS, Wieman MS, Alwine LK. Long-term tolerability of topical diclofenac sodium 1% gel for osteoarthritis in seniors and patients with comorbidities. Clin Interv Aging. 2012;7:517-523. 

31. Derry S, Moore RA, Rabbie R. Topical NSAIDs for chronic musculoskeletal pain in adults. Cochrane Database Syst Rev. 2012;9:CD007400.

DisclosuresThe authors report no relevant financial relationships.

Address correspondence to: John J. Liantonio, MD; Drexel University College of Medicine; Department of Family, Community, and Preventive Medicine; 10 Shurs Lane, Suite 301; Philadelphia, PA 19127;