American Journal of Therapeutics 7, 115—121 (2000)
`
`Gastrointestinal Complications of Prescription and
`Over-the-Counter Nonsteroidal Anti-inflammatory Drugs:
`A View from the ARAMIS Database
`
`
`
`Gurkirpal Singh
`
`More than 30 million people worldwide consume prescription nonsteroidal anti-inflammatory
`drugs (NSAIDs) on a daily basis. Gastrointestinal (GI) toxicity owing to the use of NSAIDs is a
`well-recognized clinical problem, with approximately 25% of all reported adverse drug reactions
`being attributed to prescription NSAID use. In addition to prescription NSAIDs, the use of over-
`the-counter (OTC) formulations of these products is common. Although it has been suggested that
`OTC doses of NSAIDs may not lead to significant GI toxicity, the data confirming this have been
`lacking. Data on the GI risks of OTC doses of aspirin, ibuprofen, naproxen, paracetamol, and no
`drug from 4164 consecutively diagnosed patients with rheumatoid arthritis from eight ARAMIS
`(Arthritis, Rheumatism, and Aging Medical information System) centers in North America are
`presented. Serious GI events were defined as G] bleeds and other clinically significant GI events
`requiring hospitalization. Relative risks were standardized for potential demographic confounders
`using Cox proportional hazard models. Although the relative risk of OTC doses of NSAIDs (3 to 4)
`is less than the previously published risk of prescription doses (6 to 7), it remains clinically signifi-
`cant and a matter of serious concern because of the widespread use of these medications and an
`underappreciation of the true risk. Paracetamol was not associated with increased risk of G1 com-
`plications and should be considered first—line therapy.
`
`Keywords: paracetamol, acetaminophen, gastrointestinal toxicity, over-the-counter dose, nonsteroi-
`dal anti-inflammatory drug.
`
`
`|NTRODUCT|ON
`
`Aspirin and the other nonsteroidal anti-inflammatory
`drugs (NSAIDs) are considered to be one of the most
`frequently used class of drugs, with more than 30 mil-
`lion people using them on a daily basis. Despite their
`popularity, these drugs do carry risks. It is now well
`recognized that they are associated with significant
`adverse effects on the gastrointestinal (GI) tract, that
`these are the most prevalent category of adverse drug
`reactions,1 and that they lead to significant morbidity
`
`and mortality.2 In recent years, there has been much
`research interest in the area of Gl-related complica—
`tions. The aim of this paper is to put these GI compli—
`cations into clinical perspective using information
`gained from the studies of the ARAMIS (Arthritis,
`Rheumatism, and Aging Medical Information System)
`database. To this end, we briefly review the literature
`and update previously published ARAMIS informa—
`tion from the prescription setting as well as new in-
`formation relating to the use of over-the—counter
`(OTC) analgesics.
`
`The Department DfML’tllClilL’, Division of Innninn)iog_1/ and Rimn-
`Inatology, Stanford University School of Medicine, Pain Alto,
`California, LISA.
`*
`.
`.
`Address for corrrspomirnce: ARAMIS PosMiller/ruin“7 Snrve/l—
`lance Prugrmn, Sillllfol'il University, 1000 WelcllRoml, Snitr203,
`Pnlo Alto CA LISA.
`‘
`
`THE ARAMIS DATABASE
`
`.
`.
`.
`.
`ARAMIS .‘
`ros vective observation il nomnterven-
`is "I
`‘
`1'
`‘ P
`tlonal cohort study that has systematlcally collected
`data on 1nd1v1duals With chronic rheumatlc dlseases.
`It provides a means with which physician- and pa—
`
`1075—2765 © 2000 Lippincott Williams £7 Wilkins, Inc.
`
`Patent Owners' Ex. 2046
`
`lPR2018-00272
`
`Page 1 of 8
`
`Patent Owners' Ex. 2046
`IPR2018-00272
`Page 1 of 8
`
`

`

`116
`
`SINGH
`
`tient—reported data can be systematically collected and
`analyzed. Funded primarily by grants from the Na-
`tional Institutes of Health, the ARAMIS database has
`now amassed detailed clinical outcome information
`
`on more than 36,000 patients with rheumatic diseases
`from 17 centers in the United States and Canada who
`
`have been followed for more than 300,000 patient—
`years. The ARAMIS Post—Marketing Surveillance
`(PMS) Program has prospectively followed patient
`outcome status, drug side effects, and economic im-
`pact of illness in a cohort of more than 12,000 consec—
`utively enrolled osteoarthritis (OA) and rheumatoid
`arthritis (RA) patients from eight patient populations
`(Stanford, CA; Santa Clara County, CA; Wichita, KS;
`Phoenix, AZ; Cincinnati, OH; Baltimore, MD; Saska—
`toon, Saskatchewan, Canada; and Montreal, Quebec,
`Canada).3 These data provide a large body of infor-
`mation that can be used to evaluate various aspects of
`the epidemiology of NSAID—related GI side effects.
`Patient characteristics, study design, and data collec-
`tion methods are described in detail e]sewhere.4”6
`
`NSAID—RELATED
`
`GASTROINTESTINAL
`
`COMPLICATIONS
`
`NSAID-related GI complications cover a number of
`different disorders that can be divided into three main
`categories ranging from nuisance symptoms to endo—
`scopically determined mucosa] lesions and serious GI
`complications.3
`
`Nuisance symptoms, including dyspepsia
`
`Nuisance symptoms, such as heartburn, nausea, dys-
`pepsia, and abdominal pain, affect approximately 10%
`to 20% of patients after taking an NSAID,"‘8 although
`this may range from 5% to 50%, depending on factors
`
`such as study design, patient populations, drugs, dos-
`ages, and duration of use.8 Such symptoms can have a
`great impact on NSAID compliance. Indeed,
`it has
`been noted that, among patients with CA on NSAID
`treatment after 1 year, only 15% to 20% were still tak-
`ing the same drug"
`
`Endoscopically determined mucosa] lesions
`
`NSAID—induced mucosa] lesions can range from mi—
`nor gastroduodena] lesions to serious GI complica-
`tions such as bleeding and perforation. Acute hemor-
`rhages and erosions occur within several hours to
`minutes of taking NSAIDs and can be seen in as many
`as 80% of patients treated with NSAIDs. 1” These ulcers
`are usually asymptomatic and will heal and redevelop
`over time. Endoscopic evidence of mucosal damage
`can be seen without the patient reporting any symp-
`toms; similarly a patient may report symptoms with-
`out any evidence of mucosa] damage; and serious G1
`complications can occur without the patient reporting
`any symptoms or there being any endoscopic evi-
`dence of mucosal damage.H
`
`Serious gastrointestinal complications
`
`Serious GI complications resulting in hospitalization
`occur annually in 1% to 2% of individuals taking
`NSAle regularly.12 We recently reported the data on
`the incidence of serious GI complications resulting in
`hospitalization among patients with RA and CA
`(Table 1),3 showing that, when compared with previ-
`ously published data,""3‘]5 the rate of hospitalization
`for NSAID-related serious GI complications may have
`decreased in recent years. This decline may be ac-
`counted for by two factors, namely, the establishment
`of more stringent criteria for hospitalization in a man-
`aged-care setting and physician-education campaigns,
`which emphasize increased use of newer, less Gl-toxic
`
`Table 1. Gastrointestinal complications in osteoarthritis and rheumatoid arthritis.
`“M—
`OA
`RA
`
`
`
`Hospitalizations
`Hospitalizations Deathsm“
`Number of patients
`2921
`3883
`1283
`Person—years of observation
`12,224
`19,961
`3234
`Person—years taking NSAIDs
`8471
`15,638
`2199
`Number of G] events
`25
`228
`19
`Number of GI events while taking NSAle
`19
`205
`16
`Rate per year while taking NSAiDs (%)
`0.22
`1.31
`0.73
`Rate per year while not taking NSAIDs (%)
`0.05
`0.19
`0.29
`Relative risk while taking NSAle
`
`
`6.772.51 4.21m
`Adapted from ref. 3.
`
`Anwricmz journal of TIIL'I'IIpUIIfics (2000) 7(2)
`
`Patent Owners' Ex. 2046
`
`lPR2018-00272
`
`Page 2 of 8
`
`Patent Owners' Ex. 2046
`IPR2018-00272
`Page 2 of 8
`
`

`

`GI COMPLICATIONS 0F PRESCRIPTION AND OTC NSAIDS.’ A VIEW FROM THE ARAMIS DATABASE
`
`117
`
`NSAIDs and non-NSAID analgesics, such as paracet-
`amol, in high-risk populations?"3
`
`MANY CONSUMERS DO NOT
`
`RECOGNIZE THE SERIOUSNESS
`
`THE BURDEN OF NSAlD-RELATED
`
`GASTROINTESTINAL
`
`COMPLICATIONS
`
`Approximately 10% to 15% of cases of hospitalizations
`for upper GI bleeding results in death.16 In our previ-
`ously published ARAMIS data, we reported 26 CI
`deaths in 12,224 patient-years of exposure to NSAIDs.3
`Of these deaths, 19 could be attributed to NSAle,
`giving a GI death rate of 0.22% per year, and a relative
`risk of 4.21.
`
`It could be argued that this may not seem signifi—
`cant. However, given the high usage of these prod-
`ucts, often on a chronic basis, the lifetime risk and
`health care burden may be substantial. Based on these
`conservative figures and ARAMIS data, the number of
`hospitalizations for serious GI complications per year
`is estimated to be 103,000, which, at a conservative
`estimated cost of US$15,000 to 20,000 per hospitaliza-
`tion,qamounts to an annual cost in excess of US$32 bil-
`lion:
`
`OF THE PROBLEM
`
`A survey conducted in 1998 by Roper—Starch, the
`American Gastroenterological Association, and GD.
`Searle 8: Co. revealed that limited consumer knowl-
`edge plus a lack of concern contribute to the problem
`of NSAID-related G1 complications. Some 807 indi-
`viduals were included in the analysis, which revealed
`that almost half (45%) took NSAIDs for 5 days or
`longer at least once a month.3 Furthermore, nearly
`40% of individuals combined OTC and prescription
`NSAIDs. Of great concern was the revelation that of
`those consumers who regularly used NSAIDs, nearly
`75% did not know about or were unconcerned about
`
`NSAID-related GI complications (Fig. 2). In addition,
`almost two thirds believed they would get warning
`signs in advance of a serious NSAID—induced compli-
`cation (Fig. 3). Yet other studies showed that only 20%
`of people who have a serious GI complication will
`have any warning sign.6
`
`ARE THERE WARNING SIGNS FOR
`
`COMPLICATIONS?
`
`SERIOUS GASTROINTESTINAL
`
`Among RA and 0A patients, data are even more
`startling. It has been conservatively estimated that
`16,500 NSAID-related deaths occur in these patients
`every year in the United States (Fig. 1). When com-
`Whereas dyspeptic symptoms are common in patients
`pared with some other common causes of death in the
`on NSAle, there is little correlation between these
`United States during 1997,17 it becomes apparent that
`symptoms and serious GI complications. For example,
`NSAIDs may be responsible for almost as many
`we previously reported that in a prospective cohort
`deaths as AIDS and significantly more deaths than
`evaluation of 1921 patients, the overall incidence of
`asthma, cervical cancer, and Hodgkin’s disease.3
`
`Number of deaths from selected causes. US population (1997)
`
`25,000
`
`20,000
`
`15,000
`
`10.000
`
`5,000
`
`Number
`oi deaths
`
`
`
`
`
`o ,
`
`Leukemia
`
`HIV
`
`NSAle GI Multiple
`myelorna
`Cause of Death
`
`Asthma
`
`Cervical
`cancer
`
`Hodgkln's
`dleeaee
`
`Fig. 1. Number of deaths
`associated with NSAlD-in-
`duced gastrointestinal dam-
`age compared with other
`causes (US. population).17
`From ref. 3, with permis-
`sion.
`
`American journal of TIwrnpeutics (2000) 7(2)
`
`Patent Owners' Ex. 2046
`
`lPR2018-00272
`
`Page 3 of 8
`
`Patent Owners' Ex. 2046
`IPR2018-00272
`Page 3 of 8
`
`

`

`cause Rx
`
`users
`
`OTC
`users
`
`Duni
`users
`
`lAwara of
`complications
`but unconcerned
`
`I Unaware of
`complications
`NSAIDs can
`
`SINGH
`
`constant over time. Some believe that if patients are
`going to bleed, they will do so early in the course of
`therapy. Therefore, it is postulated that those patients
`who continue with NSAID therapy may have become
`tolerant and thus will have a reduced risk of develop-
`ing serious GI complications. Our studies, which have
`followed cohorts for as long as 15 years, suggest that
`the risk of GI bleed (the hazard function) remains con-
`stant.“ One of the ways of understanding this is to use
`the analogy of a game of darts. The probability of
`hitting the center of a dartboard is dependent on the
`number of darts thrown—the more darts that are
`
`thrown, the greater the odds. Similarly, the longer a
`patient takes NSAIDs, the greater the odds of a G1
`bleed. But, based on chance alone, the first dart can hit
`the center, or the second one can, or the third one can
`and similarly, a patient can bleed from just one tablet,
`or the second tablet, or the third one.
`The single most important risk factor for an NSAID-
`related bleed is the duration of NSAID therapy; how-
`ever, currently available risk factor models do not con-
`sider this exposure. Through ARAMIS it has been pos-
`sible to develop Cox proportional hazard models
`based on life—table analysis that enable us to estimate
`the true risks associated with various patient charac-
`teristics. These have now evolved into a simple point-
`based, risk-factor algorithm called the GI SCORE
`(Standardized Calculator of Risks for Events) that may
`be used to assess a patient 5 risks from NSAID
`therapy. The SCORE program has been well vali—
`dated and is available from the authoi should more
`information be neededmm It is hoped that this simple
`scoring system will aid physicians in making in-
`formed decisions regarding the prescription of NSAID
`therapy, the use of prophylactic therapy, and the fre-
`quency of patient monitoring, all of which would help
`to reduce the incidence of serious NSAID-induced GI
`complications.3
`
`Fig. 2. Percentage of people who are unaware of NSAID-
`related complications in a cohort of regular NSAID users.
`Rx, prescription. From ref. 3, with permission.
`
`NSAID-related GI side effects was 15%, with a 2. 2%
`incidence of serious GI complications requiring hospi-
`talization3 and that as many as 81% of patients who
`had serious GI complications had no prior (31 symp-
`toms.
`
`WHO IS AT GREATEST RISK FOR
`
`SERIOUS GASTROINTESTINAL
`
`COMPLICATIONS?
`
`Owing to the largely asymptomatic nature of GI side
`effects, determining risk factors is important. Data
`from many studies established a variety of risk factors
`including advanced age, ”‘22 higher NSAID doses,2223
`histozry of GI problems (eg, peptic ulcer and GI bleed-
`1ng),2" concomitant corticosteroid use24272“ dura-
`tion of therapy,1323''29 and concomitant anticoagulant
`use .243” Despite the consistent results from these stud-
`ies, many are based on univariate analyses and do not
`take into account the interaction among multiple fac-
`tors (eg, an older patient is also likely to have a longer
`duration of disease and is perhaps sicker with more
`comorbidities).
`One important issue in this respect is the question of
`whether the risk for NSAID—related gastropathy is
`
`Rx users
`
`OTC users
`
`OVER-THE-COUNTER NSAIDS: AN
`
`O
`
`Don’t Expect Warning Signs
`Expect Warning Signs
`
`Fig. 3. Percentage of regular NSAID users who expect to
`experience a warning sign before a serious gastrointes~
`tinai complication. From ref. 3, with permission.
`
`American journal of Therapeutics (2000) 7(2)
`
`UNDERESTIMATED PROBLEM
`
`ln the United States in 1998, 16.] billion tablets of OTC
`NSAle were sold compared with only 2.9 billion of
`prescription NSAIDs. Of the more than 30 million
`people who regularly take NSAIDS, 40% take both
`prescription and OTC NSAle at the same time, and
`nearly one third of people who take prescription
`NSAIDs consume more than the maximum recom-
`mended doses.” Yet, as demonstrated earlier, there is a
`perception among the general public that because
`these drugs are available OTC, they do not carry any
`significant risk.
`
`Patent Owners' Ex. 2046
`
`lPR2018-00272
`
`Page 4 of 8
`
`Patent Owners' Ex. 2046
`IPR2018-00272
`Page 4 of 8
`
`

`

`GI COMPLICATIONS OF PRESCRIPTION AND OTC NSAIDS: A VIEW FROM THE ARAMIS DATABASE
`
`119
`
`The association of prescription NSAIDS with serious
`GI complications has been established primarily from
`case-control studies. Most of these studies are based
`
`on analysis of computer databases of Medicare, Med-
`icaid, or managed care records. Although these retro-
`spective studies document that prescriptions have
`been dispensed, actual consumption cannot be deter-
`mined, and the prevalence of OTC NSAID use cannot
`be assessed.
`
`There are several case series demonstrating the as-
`sociation of prescription NSAIDS with upper GI hem-
`orrhage, but few have specifically looked at OTC
`NSAIDS. In a prospective study, Wilcox and col-
`leagues showed that in a large US. county hospital,
`56% of 421 upper GI bleeds during 1990 to 1992 were
`associated with the use of NSAIDS.32 Of the NSAID-
`related bleeds, 63% were associated with OTC aspirin
`use and 16% with OTC nonaspirin NSAID use. They
`state:
`
`In our experience, when a patient is asked generically
`concerning medication use, consumption of OTC
`preparations is often not volunteered as patients con-
`sider these OTC agents as not a drug per 53. In addi—
`tion, some patients consume non—aspirin NSAIDS as
`an "aspirin substitute” but are unaware of their po-
`tential gastrointestinal
`toxicity. Given the apparent
`frequency with which these OTC NSAIDS are used,
`these products may represent an important health
`hazard, particularly in patients with ulcer disease.
`
`How important the health hazard is was brought
`out clearly in a recent study conducted by the Ameri-
`can College of Gastroenterology (AC6)33 to assess de-
`mographics, management strategies, and outcomes
`for patients with GI bleeding. All ACG members and
`Fellows were requested to send information on as
`many as ten bleeding patients and for ten procedure-
`matched controls. A total of 635 bleeders and 600 con-
`
`trols was studied. Among the 635 bleeders, 56.9"”
`were taking NSAIDs and of these, 84% were taking
`OTC NSAIDS, with the vast majority using aspirin.
`Alcohol intake was also recorded. Overall, 47.6% of
`bleeders were taking OTC aspirin or NSAIDS com-
`pared with only 19.4% of controls. Of note is the fact
`that 84% of the NSAID-related GI bleeds occurred
`
`among patients taking OTC NSAIDS. These authors
`estimated that the odds ratio for use of OTC NSAIDS
`
`was 2.76 (confidence interval [CI] 2.03—3.74) and that
`of alcohol use was 2.07 (CI 1.48—2.88). When OTC
`NSAIDS and alcohol were used together, the relative
`risk increased dramatically to 4.47 (CI 2.73—7.32),
`showing a significant interaction. In this study, the
`authors showed that paracetamol did not have any
`significant risk, being used by only 4.4% of bleeders
`
`and 6.2% of controls, nor was there an increased GI
`risk for those using paracetamol and alcohol.
`Blot and McLaughlin recently presented a detailed
`analysis of the ACG bleed registry and association
`with OTC NSAIDS.34 Overall, 27% of bleeders were
`taking OTC dose aspirin compared with 12.0% of con—
`trols. The corresponding numbers with OTC dose ibu-
`profen were 10.1% and 5.8%, respectively, and those
`with OTC dose paracetamol were 4.5% and 6.3%, re-
`spectively. The odds ratios for a serious GI bleed with
`OTC dose aspirin and ibuprofen were 2.7 and 2.3—
`both representing a statistically significant increase in
`risk. The odds ratio for paracetamol was 0.9, repre-
`senting a risk not statistically different from back—
`ground. The odds ratios were related to the dose: ibu-
`profen doses of <600 mg/day had a ratio of 1.7. This
`increased to 3.4 in patients who took 600 to 1200 mg/
`day, and to 3.5 in those who took >1200 mg/day.
`There was a strong interaction between alcohol con—
`sumption and OTC NSAID use. Although alcohol use
`doubled the risk for a GI bleed, when alcohol was
`combined with aspirin, the risk increased to fourfold.
`In patients who took OTC ibuprofen and alcohol to-
`gether, the risk increased to 7.1.
`There are several studies on the risk of aspirin, al—
`though many do not differentiate between prescrip-
`tion and OTC use. In a case-control study, Levy and
`colleagues reported that the relative risk of occasional
`aspirin use (presumably OTC) in causing major upper
`GI bleeds was 5.6 (CI 2.7—12.0).35 There was no in-
`creased risk with the use of paracetamol. Weil and
`colleagues estimate that approximately two thirds of
`the 10,000 episodes of ulcer bleeding in people aged 60
`years and older every year in England and Wales are
`related to the use of NSAIDS.“ Of these, 46% are re—
`lated to aspirin use, two thirds of which is given for
`prophylactic purposes. According to the authors, no
`close of aspirin was safe: the relative risks for daily
`doses of 75 mg, 105 mg, and 300 mg were 2.3, 3.2, and
`3.9, respectively. In a recent study, Cryer and Feldman
`reported that even 10 mg aspirin substantially re-
`duced gastric mucosal prostaglandin levels to ap-
`proximately 40"u of the baseline value and caused sig-
`nificant gastric injury.”
`
`ARAMIS |NVEST|GAT|ON OF
`
`OVER-THE—COUNTER
`
`DOSE NSAIDS
`
`Using the ARAMlS database, we conducted a pro—
`spective observational 1mninterventiona1 study in a
`large group of RA patients taking OTC doses of
`
`American IournnI of Therapeutics (2000) 7(2)
`
`Patent Owners' Ex. 2046
`
`lPR2018-00272
`
`Page 5 of 8
`
`Patent Owners' Ex. 2046
`IPR2018-00272
`Page 5 of 8
`
`

`

`120
`
`SINGH
`
`toxicity of aspirin,
`NSAIDs. We studied serious CI
`ibuprofen, and naproxen in OTC doses, comparing
`these drugs with the non—NSAII) analgesic paracetav
`mol and no drug therapy in 4164 patients with 6279
`patient-years of observation. Of these, 2064 patients
`were taking OTC NSAIDs, 72]
`taking paracetamol,
`and 'I379 were not taking any medication during the
`study. Serious GI events were defined as GI bleeds
`and other clinically significant CI events requiring
`hospitalization.
`
`ARAMIS RESULTS
`
`AND CONCLUSIONS
`
`There was no difference in the incidence rates of seri-
`
`ous Cl complications among those patients taking
`paracetamol and those not on any drug therapy. How-
`ever, patients taking OTC doses of NSAIDs had an
`incidence rate of 0.58 per ‘100 patient-years, statisti—
`cally significantly higher than the rate, among patients
`on paracetamol (0.16; P < .05) or not on any therapy
`(0.'l7; P < .05). Relative risks were standardized for
`potential demographic confounders using Cox pro-
`portional hazard models. Because this was an obser—
`vational study, there were systematic differences in
`people on different analgesics. For example, patients
`on paracetamol tended to be older and had more se-
`vere disease (and thus more likely to have a CI ad—
`verse event) compared with patients on OTC NSAIDs.
`Cox proportional hazard models were used to control
`for confounders such as age, duration of therapy, and
`disease severity. Standardized relative risks of differ-
`ent therapies were compared with patients not on any
`therapy. The relative risk of paracetamol did not differ
`significantly from background, whereas the ()TC
`doses of NSAIDs had a relative risk of 3.92 (l’ < .01).
`The individual NSAII) risks were as folloWs: aspirin,
`4.14 (I’ < .0'I); ibuprofen, 3.5 (P < .05); and naproxen,
`3.42 (p < .05).
`Although the relative risks observed for the OTC
`doses of NSAIDs are, as expected, lower than those for
`prescription doses, they are clinically meaningful and
`statistically significant. These data lead us to conclude
`that even the lower doses of NSAIDs available O'IC
`are not without GI risk. However, it should be noted
`that these data were derived from patients with RA,
`and two thirds of the patients in our study were taking
`OTC NSAIDs regularly and not in an occasional fash—
`ion.
`
`CONCLUSIONS
`
`The seriousness of NSAlD—related GI ulcer complica—
`tions (eg, bleeding and perforation) are well docu-
`
`Aiiii'ricim lam‘iml of'I‘lii'mpi'mics (2000) 7(2)
`
`mentecl. Unfortunately and of great concern, many pa—
`tients are unaware of these potential complications. In
`the vast majority of cases, serious GI complications are
`not preceded by any symptoms or signs. Patients and
`physicians need to be aware of the risk factor profiles
`that predispose patients to serious (31 problems, using
`tools such as the GI SCORE to their advantage in this
`respect.
`The risk of OTC NSAIDs remains a matter of serious
`
`concern because of the widespread use of these inedi—
`cations and an uncIerappreciation of the true risk. Al-
`though consistent data are beginning to emerge on
`problems caused by OTC doses, there is still a need for
`further information on the risks involved.
`
`REFERENCES
`
`I. Butt JH, Barthel JS, Moore RA: Clinical spectrum of the
`upper gastrointestinal effects of nonsteroidal anti—
`inflammatory drugs. Natural history, symptomatology,
`and significance. Am J Med 1988;84:5—l-l.
`Wolfe MM, Lichtenstein DR, Singh (1: Gastrointestinal
`toxicity of nonsteroidal antiinflammatory drugs. N IanI
`J Med '1 999340: |888¥l 89‘).
`
`[0
`
`'l‘riadafilopoulos (i: [Epidemiology of NSAID
`3. Singh (7,
`induced gastrointestiiml complications.
`J Rheumatol
`l999;2(7:'18—24.
`
`4. Singh G, [Tries JF, Spill. 1’, Williams CA: 'I'oxic effects of
`azathioprine in rheumatoid arthritis. A national post—
`marketing perspective. Arthritis Rheum “$932183?
`843.
`
`toxicity profiles of
`5. Singh G, [tries Jli, Williams CA, et al:
`disease modifying antirheunmtic drugs in rheumatoid
`arthritis] Rheumatol I99I;18:I88—IL)4.
`0. Singh G, Ramey DR, Morfeld D, et al: (iastrointestirml
`tract complications of nonsteroidal anti—inflammatory
`drug treatment in rheumatoid arthritis. A prospective
`observational cohort study. Arch Intern Med 'l9‘)o;l5b:
`1530—1536.
`
`7. Hardin JG, l,.ongenecker (it; Handbook of! )l'll‘g’ ’I'Iii'i‘api/ iii
`R/icuiiiiilic Disease:
`l’lmrmrico/ogy um! Clinical Aspi'cts.
`Little, Brown, Boston, 1992.
`8. Coldenberg DL, Cohen AS: Drugs in ”11' Rheumatic Dis-
`ciiscs. (irune and Stratton, New York, 1986.
`L). Scholes D, Stergachis A, Penna I‘M, et at: Nonsteroidal
`antiinflammatory drug discontinuation in patients with
`osteoarthritis. J Rheumatol
`[995,22z7tIS—7l2.
`It). Ehsanullah RS, Page MC, 'l'ildesley (i, Wood IR: I’reren-
`tion of gastroduodenal damage induced by non-
`steroidal anti-inflammatory drugs: controlled trial of ra-
`nit‘idine. BM] 1988;2‘J711tII7—102L
`I I. (.iraham DY, Smith IL: (iastroduodenal complications of
`chronic NSAID therapy. Am .I Liaslroenterol
`I‘)88;83:
`’l081—‘I084.
`
`Patent Owners' Ex. 2046
`
`lPR2018-00272
`
`Page 6 of 8
`
`Patent Owners' Ex. 2046
`IPR2018-00272
`Page 6 of 8
`
`

`

`GI COMPLICATIONS OF PRESCRIPTION AND OTC NSAIDS: A VIEW FROM THE ARAMIS DATABASE
`
`121
`
`19.
`
`12. Cryer B: Nonsteroidal anti-inflammatory drug gastroin-
`testinal toxicity. Curr Opin Gastroenterol 1999;15:473—
`480.
`13~ Singh G, Ramcy DR, Balise R, Triadatilopolous G: Epi-
`demiology of NSAlD—related GI complications in pa-
`tients with osteoarthritis (0A): a 13 year prosPective
`study [abstract]. Arthritis Rheum 1998;41:5180.
`14- Friest, Ramey DR; SinghG, 9‘ “Ii A reevaluation 0f
`aspirin therapy 1“ rheumatmd arthritis. Arch Intern Med
`1993;153:2465—2471
`15. Singh G: Recent considerations in nonsteroida] anti-
`inflammatory drug gastropathy. Am I MEd 19987105:
`315—383
`.
`.
`16. Armstrong CP, Blower AL: Non-sterorda-l antl-
`inflammatory drugs and life threatening complications
`0f PePt‘C ulceration. Gut 19875281527632
`17. National Center for Health Statistics: 1998. Hyattsville,
`MD‘
`,
`_
`.
`t'd7
`18Bk D]~N‘t=d'lt-fl
`. [or man ‘
`'. 0115.”? .d an 1 m amma‘oiy IUF
`'
`:gdUCCd gastrointestinal injury. Am] Med 19)6,101.25S—
`‘Ls‘
`‘
`1 C .
`.
`.
`f
`W
`.
`,
`-.
`,
`f
`ontgstreti
`1F.tEpi:lerti.iioli)]gy o :ospitalizatiTnt. or
`acu e upper gas rom es ina
`iemorr age: a popu a ion-
`based study. Am] Gastroenterol 1995;90:206—210.
`,
`.
`.
`.
`,
`20. Greene JM, Winickoff RN: Cost-conscmus prescribing of
`.
`.
`.
`,
`nonsterordal anti-inflammatory drugs for adults With ar—
`.
`.
`.
`.
`thritis. A reView and suggestions. Arch Intern Med 1992;
`152,1995_2002
`21. Gabriel SE, Jaakkimainen L, Bombardier C: Risk for se-
`rious rastrointestinal corn lications related to use of
`. E".
`.
`.
`.
`P .
`.
`nonsteroidal anti-inflammatory drugs. A meta—analysis.
`Ann Intern Med 1991;115:787—796.
`22. Griffin MR, Piper JM, Daugherty JR, et al: Nonsteroidal
`anti-inflammatory drug use and increased risk for PGP'
`tic ulcer disease in elderly persons. Ann Intern Med
`1991;114:257—263.
`.
`23. Langman M], Wei] ], Wainwright P, et al: Risks of bleed—
`ing peptic ulcer associated with individualnon-steroidal
`anti-inflan‘imatory drugs. Lancet 1994;343:10754073
`24. Garcia RL, Jick H: Risk of upper gastrointestinal bleed-
`ing and perforation associated with individual non-
`steroidal anti-inflammatory drugs. Lancet 1994;343:769—
`772.
`'
`25. Hallas J, Lauritsen I, Villadsen HD, Gram LF: Nonste-
`roidal anti-inflammatory drugs and upper gastrointesti-
`nal bleeding, identifying high-risk groups by excess risk
`estimates.Scand]Gastroenterol1995;30:438—444.
`26. Nielsen JC, Bjerring P, Arendt—Nielsen L: A comparison
`
`of the hypoalgesic effect of paracetamol in slow—release
`and plain tablets on laser-induced pain. Br I Clin Phar-
`macol 1991 31:267—270.
`27. Hochain P, Berkelmans I, Czernichow P, et al: Which
`patients taking non—aspirin non-steroidal anti-
`inflammatory drugs bleed? A case-control study. Eur]
`Gastroenterol Hepatol 1995;72419—426.
`28. Piper JM, Ray WA, Daugherty JR, Griffin MR: Cortico-
`steroid use and peptic ulcer disease: role of nonsteroidal
`anti-inflammatory drugs. Ann Intern Med 1991;114:735—
`740
`29. Silverstein FE, Graham DY, Senior JR, et a1: Misoprostol
`reduces serious gastrointestinal complications in pa-
`tients with rheumatoid arthritis receiving nonsteroidal
`anti-inflammatory drugs. A randomized, double-blind,
`placebo-controlled trial. Ann Intern Med 1995;123:241—
`249.
`JR, Griffin MR: Concur-
`30. Shorr RI, Ra WA, Dauvhert
`y
`5y
`rent use of nonsteroidal anti-inflammatory drugs and
`oral anticoagulants places elderly persons at high risk
`for hemorrhagic peptic ulcer disease. Arch Intern Med
`1993;153:1665—1670.
`.5.
`.
`7
`(
`,
`T 1
`r
`.
`.
`7‘
`‘
`.
`s
`‘ 1' smbh G lecy DR’
`‘T‘qfhf‘lomlous G' Lt 11' “I
`SCORE. A Simple self—assessment instrument to quan-
`.
`.‘
`‘
`.
`‘
`.
`.
`‘
`tin the risk of serious NSAID-related G1 complications
`.
`.
`.
`in RA and 0A [abstract]. Arthritis Rheum 1998;41:875.
`,
`,
`,
`.
`32. Wilcox CM, Shalek KA, Cotsonis G: Striking prevalence
`of over—the-counter nonsterOidal anti-inflammatory
`drug use m patients WIth upper gastromtestinal hemor-
`rhage. Arch intern Med 1994;154:42—46.
`_
`33. Peura DA, Lanza FL, Gostout C], Foutch PG: The Ameri-
`C,““,C0“°$e of G?Stroflftel"’lt’gy 3199mm; REE-W)" 1?ij
`{ministry findings [abstract]. Am] Gastroenteiol 1997,)2.
`24—92%]
`34. Blot W, McLaughlin]: GI bleeding in relation to analge—
`sic use [abstract]. Digestion 1998;59:207-
`.
`35» LEVY M, Mllle DR, Kaufman DWI Gt al: MaJOF UPPE‘I‘
`gastrointestinal tract bleeding. Relation to the use of as-
`pirin and other nonnarcotic analgesics. Arch Intern Med
`19882148281435
`36. Weil ], Colin-Jones D, Langman M. et al: Prophylactic
`aspirin and risk of peptic ulcer bleeding. BM] 1995;310:
`827-830-
`37. Cryer B, Felclman M: Effects of very low dose daily,
`long-term aspirin therapy on gastric, duodenal, and rec-
`tal prostaglandin levels and on mucosal
`injury in
`healthy humans. Gastroenterology 1999;117:17—25.
`
`AIIIUI‘IL‘IHI [LIIII‘IIIII of Therapeutics (2000) 7(2)
`
`Patent Owners' Ex. 2046
`
`lPR2018-00272
`
`Page 7 of 8
`
`Patent Owners' Ex. 2046
`IPR2018-00272
`Page 7 of 8
`
`

`

`Patent Owners' Ex. 2046
`
`lPR2018-00272
`
`Page 8 of 8
`
`Patent Owners' Ex. 2046
`IPR2018-00272
`Page 8 of 8
`
`