Eur J Epidemiol DOI 10.1007/s10654-014-9971-7

REVIEW

Prophylactic use of aspirin: systematic review of harms and approaches to mitigation in the general population Mangesh A. Thorat • Jack Cuzick

Received: 26 March 2014 / Accepted: 30 October 2014 Ó Springer Science+Business Media Dordrecht 2014

Abstract A careful assessment of benefits and harms is required to assess suitability of aspirin as a prophylactic public health measure. However, comprehensive population-level data on harms are lacking. We collected and synthesized age and sex-specific data on harms relevant to aspirin use in average-risk individuals aged 50 years or older. We conducted systematic literature searches to identify baseline rates of gastrointestinal (GI) bleeding, peptic ulcer, major extra-cranial bleeding, and case-fatality rates due to GI bleeding or peptic ulcer in general population. The magnitude of aspirin-associated increase, the prevalence and attributable risk of Helicobacter pylori infection on these events in aspirin users was also assessed. Baseline rates of major extracranial bleeding events and GI complications increase with age; an almost threefold to fourfold increase is observed from age 50–54 to 70–74 years. Low or standard-dose aspirin use increases GI bleeding events by 60 % leading to an annual excess of Disclaimer: The findings and conclusions in this report are those of the authors and do not represent the official position of the authors’ institutions.

Electronic supplementary material The online version of this article (doi:10.1007/s10654-014-9971-7) contains supplementary material, which is available to authorized users. M. A. Thorat (&)  J. Cuzick Centre for Cancer Prevention, Wolfson Institute of Preventive Medicine, Queen Mary University of London, Charterhouse Square, London EC1M 6BQ, UK e-mail: [email protected] J. Cuzick e-mail: [email protected] M. A. Thorat Division of Surgery and Interventional Science, Whittington Hospital, Magdala Avenue, London N19 5NF, UK

0.45 and 0.79 GI bleeding events per 1,000 women and men aged 50–54 years respectively. 5–10 % of major GI complications are fatal; a clear age dependence—higher fatality in older individuals, is seen. Eradication of H. pylori infection before aspirin use could reduce the incidence of upper GI complications by 25–30 %. GI complications are increased by about 60 % due to aspirin use but are fatal only in a very small proportion of individuals younger than 70 years of age. Major bleeding events that are comparable in severity to cancer or CVD, are infrequent. Screening and eradication of H. pylori infection could substantially lower aspirin-related GI harms. Keywords Aspirin  Gastrointestinal bleeding  Peptic ulcer  Helicobacter pylori  Prevention  Cancer Abbreviations ACS American Cancer Society ATT Antithrombotic Trialists BDT British Doctors’ Trial BHF British Heart Foundation CRUK Cancer Research UK CVD Cardiovascular disease HPFS Health Professionals Follow-up Study HTA Health Technology Assessment GI Gastrointestinal ISCaP International Society of Cancer Prevention LGIB Lower gastrointestinal bleeding NANSAIDs Non-aspirin NSAIDs NHS Nurses’ Health Study ONS Office of National Statistics PHS Physicians’ Health Study SAH Sub-arachnoid haemorrhage UGIB Upper gastrointestinal bleeding UGIC Upper gastrointestinal complications

123

M. A. Thorat, J. Cuzick

USPSTF WHS 1,000-PY

U.S. Preventive Services Task Force Women’s Health Study 1,000 Person years

Background Aspirin or medicinal acetylsalicylic acid was synthesised by Bayer AG in 1897 [1, 2]. Apart from analgesic, antipyretic and anti-inflammatory properties, several other medicinal roles of aspirin have become apparent since then. These include its use as an anti-thrombotic agent in the prevention of cardiovascular diseases (CVD) due to anti-platelet effects [3] and more recently as an agent for cancer prevention and treatment [4, 5]. Cancer and CVD comprise a substantial proportion of the global disease burden, and are the leading causes of disability and death in the developed world [6]. Aspirin has considerable potential to reduce these if used prophylactically in the general population. However, aspirin is also associated with excess bleeding, particularly gastrointestinal (GI) bleeding and haemorrhagic stroke being the most significant concerns. Evaluation of aspirin as a prophylactic measure for the general population requires a careful assessment of both benefits and harms. It is important to recognise that although relatively common, the vast majority of aspirinrelated adverse effects (excluding intracranial bleeding) do not have long-term sequelae and are rarely fatal, especially for individuals under the age of 70 years [7]. To make a balanced comparison of benefits and harms, it is important to focus on harms that are comparable in severity to the benefits viz. prevention of cancer or myocardial infarction (MI). For example, an MI, stroke or a cancer results in a substantially greater disability than epistaxis, dyspeptic pain or a single episode of haematemesis not requiring blood transfusion. On the other hand, preventing non-fatal MI would not be considered to provide net benefit if it caused an equal number of cases of haematemesis requiring hospitalisation and multiple blood transfusions. Therefore benefit-harm evaluation of prophylactic aspirin needs to consider not only the incidence of benefits and harms but also their severity. It is also important to investigate adverse effects in a general population as these may very well be lower than those for unwell individuals. Benefit-harm evaluation also needs to consider if harms can be avoided. GI complications are the most common harms of aspirin use. Helicobacter pylori infection plays a causal role in a substantial proportion of upper gastrointestinal bleeding (UGIB) events and peptic ulcers [8]. The proportion of upper gastrointestinal complications (UGIC) attributable to H. pylori infection in a contemporary western developed nation is not well documented. These data

123

are needed to estimate the potential value of eradication before beginning long term aspirin prophylaxis. We recently reported on a comprehensive benefit- harm analysis of use of aspirin in the general population [9]. Space limitations precluded a full analysis of the potential harms of aspirin use. We address that deficiency here. The specific aims of this study were to 1.

2. 3. 4.

estimate age and sex-specific baseline rates and aspirin-related excess events of GI bleeding, peptic ulcer, and major extra-cranial bleeding; estimate age and sex-specific baseline rates for fatal GI bleeding or peptic ulcer; estimate the magnitude of the harms caused by aspirin; estimate the proportion of UGIC attributable to H. pylori infection.

Although the rates in older individuals are also estimated, this study focussed mainly on average-risk individuals aged 50–70 years, as they are the most likely candidates for prophylactic aspirin.

Methods Search strategy and selection criteria A systematic search of the literature using PubMed/Medline was conducted according to PRISMA guidelines [10]. Specific details of search terms and criteria for each search are given below and respective PRISMA flowcharts are provided in the supplementary material (Figs. 1, 2, 3, 4). Only studies reported in English language and conducted on humans were considered. In order to obtain contemporary estimates, all searches (except for case-fatality rates) were restricted to reports published from 1st January 2000 to 30th June 2013. For case-fatality rates of GI bleeding or peptic ulcer, the search period was extended to start on 1st January 1996 to obtain sufficient data. Formal assessment of quality by assigning scores to studies was not done and no study was excluded for weakness of design or data quality. In case of multiple reports, the most recent report was considered. Rates of GI bleeding, peptic ulcer, and major extra-cranial bleeding The objective of this search was to identify studies reporting age and sex-specific rates of GI bleeding and/or peptic ulcer and/or major extra-cranial bleeding in the UK general population. The following keywords were searched in the title or abstract: [gastrointestinal bleeding] or [peptic ulcer] or [major extracranial bleeding] and [incidence] and [population]. Studies reporting only on hospitalised

Prophylactic use of aspirin

patients or on specific subpopulations or populations other than the UK general population were excluded. Case-fatality rates of GI bleeding or peptic ulcer The objective of this search was to identify studies reporting case-fatality rates due to GI bleeding and/or peptic ulcer in any general population. The following keywords were used: [gastrointestinal bleeding] or [peptic ulcer] and [fatal*]. Studies reporting on specific interventions or subpopulations were excluded. The harms due to aspirin use The objective of this search was to identify studies reporting on harms associated with aspirin use. The search used the following keywords: [aspirin] and [bleeding]. Studies reporting on specific conditions or subpopulations e.g. atrial fibrillation, intermittent claudication, diabetes mellitus, surgical patients were excluded. Studies reporting use of other drugs or drug combinations with aspirin were also excluded. Prevalence of H. pylori infection The objective of this search was to identify studies reporting prevalence of H. pylori infection in the UK general population that included middle-aged individuals. The following keywords were searched: [H. pylori infection] and [epidemiology*] and [population] and [middle age]. Studies reporting only on hospitalised patients or on specific subpopulations or populations other than the UK general population were excluded. Estimation of rates The objective was to obtain sex-specific rates for 5-year age-groups. If the selected studies reported rates that satisfied these conditions, these rates were used directly (either from tables or accurately measured from the figures/ plots using pixel count on a magnified image where required). If the studies reported rates for different agegroups (for example 45–54 or 60–69 years) than those required, age and sex-specific rates were derived by plotting reported rates and fitting a polynomial trendline with three knots. The rates were not extrapolated outside the range reported in the respective study. Where only a cumulative rate for a broad group was available, e.g. major extracranial bleeding, an age-incidence curve was created and data were extrapolated to fit the curve with the scale estimated for the dataset in question (details given in the supplementary material).

The rates of fatal GI bleeding or peptic ulcer were estimated using two methods: 1.

2.

Multiplying baseline rates of GI bleeding or peptic ulcer by age-specific case-fatality ratios; estimated rates labelled as FGI-CF. Using the number of age and sex-specific deaths (ICD10 codes K25-29, K62.5 and K92.0-92.2) in year 2008 as reported by the Office of National Statistics (ONS) for England and Wales [11]; estimated rates labelled as FGI-ONS

Estimation of the fraction of events attributable to a risk factor The proportion of excess events (PAR) attributable to a risk factor (for example, H. pylori infection or NSAID use) was calculated as PAR = [P 9 (RR-1)]/[1 ? P 9 (RR-1)] where P = prevalence of exposure and RR = Risk (RR or Odds Ratio) associated with the exposure. Pooled analyses Review Manager 5.2 software (Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration 2012) was used for pooled analyses. Analyses used random effects model wherever significant heterogeneity among studies existed.

Results Rates of GI bleeding, peptic ulcer, and major extracranial bleeding The search to identify studies reporting age and sex-specific rates of GI bleeding and/or peptic ulcer in the UK general population produced 201 results (supplementary Fig. 1); 4 full-text articles were assessed, one of these was excluded as it did not provide sufficient data to derive age and sex-specific rates, 3 reports were included in our study [12–14]. All studies were based on general practice research databases in the UK, Hippisley-Cox et al. [14] used the QRESEARCH database (2000–2004); Cai et al. [12] used The Health Improvement Network (THIN) database (1997–2005); and Garcia-Rodriguez et al. [13] used the General Practice Research Database (GPRD) (1995–1999). Hippisley-Cox et al. [14] reported rates of adverse upper gastrointestinal event (peptic ulcer or haematemesis) while Cai et al. [12] and Garcia-Rodriguez et al. [13] reported rates of peptic ulcer only (Supplementary tables 3, 4 and 5). Garcia-Rodriguez et al. [13] reported slightly higher ulcer rates as compared to Cai et al.

123

M. A. Thorat, J. Cuzick Table 1 Baseline rates of bleeding and peptic ulcer events in the UK general population Age-group (years)

Major extracranial bleeding

Any GI bleeding

Peptic ulcer

GI bleeding and peptic ulcer

Men

Women

Men

Women

Men

Women

Men

50–54

0.44

0.22

1.46

0.84

0.67

0.58

2.13

1.42

55–59

0.78

0.39

1.16

0.96

0.85

0.71

2.01

1.67

60–64

1.12

0.56

2.68

1.42

1.06

0.87

3.74

2.29

65–69

1.46

0.74

3.54

2.52

1.3

1.05

4.84

3.57

70–74

1.81

0.92

4.87

2.96

1.42

1.16

6.29

4.12

75–79

2.27

1.21

7.78

4.96

1.44

1.21

9.22

6.17

80–84

2.95

1.75

9.12

7.12

1.44

1.21

10.56

8.33

Women

Baseline age and sex specific major extracranial bleeding [15], any GI Bleeding [12, 14], peptic ulcer [12] and any GI complication (GI bleed or peptic ulcer) events (per 1,000 person years) estimated in the UK general population GI Bleeding and peptic ulcer rates are not adjusted for baseline NSAID use in the studies [12, 14]. Adjusted rates would be about 10 % lower in non-users Peptic ulcers refer to ulcers that are neither bleeding nor perforated Any GI bleeding comprises of both upper and lower GI bleeding, including bleeding from peptic ulcer

[12]. Garcia-Rodriguez et al. [13] also had higher rates of aspirin and NSAID use (25 vs. 18.5 %) and reported data from an earlier period, when there was a higher prevalence of H. pylori infection (see below), resulting in higher ulcer rates than the more contemporary cohort of Cai et al. [12]. To better reflect current risks, we used, data from Hippisley-Cox et al. [14] and Cai et al. [12] to estimate age and sex-specific rates of GI bleeding and peptic ulcer risk in the UK general population (any GI bleeding). These two studies were based on a similar time period, and had similar rates of drug use; aspirin in 11 and 15 % of individuals and non-aspirin NSAIDs (NANSAIDs) in 7.5 and 10 % of individuals in Cai et al. [12] and Hippisley-Cox et al. [14] respectively. Uncomplicated peptic ulcer events (per 1,000 person years) reported in Cai et al. [12] were subtracted from all upper GI events reported in HippisleyCox et al. [14] to obtain UGIB events. These were then multiplied by 2 (see ‘‘Discussion’’) to derive total GI bleeding events as none of the studies provided information on lower GI bleeding (LGIB) events. Total GI bleeding rates and peptic ulcer rates thus derived are given in Table 1. These rates are not adjusted for baseline aspirin or NSAID use of 18.5 and 25 % in Cai et al. [12] and Hippisley-Cox et al. [14] respectively. Adjusted rates in nonusers of NSAIDs would be 10–13 % lower (excess events attributable to aspirin or NANSAID use by 18.5 to 25 % individuals at the relative risk of 1.6). No study reporting age and sex-specific rates of major extra-cranial bleeding in the UK general population was identified. Therefore we used data from the Antithrombotic Trialists’ (ATT) Collaborators’ meta-analysis [15] to estimate these rates as described in methods. Majority of the patients in the ATT Collaborators’ meta-analysis were aged 50–69 years (mean 56 years) and the rate of major

123

extracranial bleeding was 0.7 per 1,000 person years in the control group. The estimated rates of major extracranial bleeding are given in Table 1. These range from 0.22 per 1,000 person years (1,000-PY) in women aged 50–54 to 1.81 per 1,000-PY in men aged 70–74. Case-fatality rates of GI bleeding or peptic ulcer The search to identify studies reporting on case-fatality rates due to GI bleeding and/or peptic ulcer produced 207 results (supplementary figure 2); 6 additional papers were identified through reference lists, 14 full-text articles were assessed, 3 were excluded, and 11 reports were included in this analysis [16–26]. The study by Kang et al. [19] reported age and sex-specific case-fatality rates due to peptic ulcer (Supplementary table 2), these varied from 2 to 10 % among persons aged 45–74. Case-fatality rates due to UGIB were 5.5, 5.6, 9.5 and 10 % in studies by Lanas et al. [20], Paspatis et al. [22], Morales-Uribe et al. [21], Button et al. [16] respectively. A UK audit in 2007 reported an overall fatality rate of 10 % in acute UGIB in 6,750 patients with a median age of 68 years [23]. A similar Spanish audit reported a 5.6 % mortality rate associated with hospital admission due to major GI complications (upper or lower tract) [26]. Case-fatality rates due to bleeding peptic ulcer were 4.4, 5.5, and 10.7 % in studies by Garcia-Rodriguez et al. [17], Hasselgren et al. [18] and Mose et al. [24] respectively. Risk of death was high when co-morbidities were present [16, 21, 23]. Risk of death also increased with increasing age [17, 19], Garcia-Rodriguez et al. [17] reported 2.8-fold, 4.2-fold and 6.2-fold risks of mortality in those aged 60–69, 70–79 and 80–89 years respectively as compared to persons aged 30–59 years. In an overview of studies (any study type) reporting mortality

Prophylactic use of aspirin

associated with UGIB or ulcer perforation, Straube et al. [25] reported that in studies conducted after 1997, 7.4 % of cases were fatal; the case-fatality rate declined to this level from 11.6 % in pre-1997 studies [25]. Based on these studies we estimate that the case-fatality rates in GI bleeding and peptic ulcer are 5 and 10 % in those aged 50–70 and those above 70 years of age respectively.

figure 3); 29 full-text articles were assessed, 5 were excluded, 24 reports were included in this study. Ten of these studies were meta-analyses or systematic reviews of RCTs [15, 27–36], whereas 14 observational studies were identified [37–50]. None of the overviews synthesized data from observational studies. Overviews of RCTs

Rates of fatal GI bleeding or peptic ulcer The age and sex-specific rates of fatal GI bleeding or peptic ulcer (labelled as FGI-CF) were estimated by multiplying baseline rates of GI bleeding or peptic ulcer by 0.05 and 0.10 in age-groups below and above 70 years respectively. The rates were also calculated using ONS cause-specific mortality data (labelled as FGI-ONS) for year 2008 [11] Both these rates are shown in Fig. 1. The rates estimated by both methods show a strong and consistent effect of age. The FGI-ONS rates are 50–60 % lower than the FGI-CF rates. This is likely due to two factors; overestimation in FGI-CF rates as these are based on all GI complications (not just major complications), and underestimation in FGI-ONS rates due to under-ascertainment of cause-of-death for the specific ICD codes. Excess bleeding risks due to aspirin use The search to identify studies reporting on harms associated with aspirin use produced 3514 results (supplementary Fatal GI complicaon rates by age 1 0.9

per 1000 person years

0.8 0.7 0.6 0.5 0.4 0.3

Overviews by Derry et al. [28], McQuaid et al. [31] Serebruany et al. [34] and Lanas et al. [30] included a large number of trials including post-surgery trials. Other overviews were mainly based on 9 RCTs [51–59] conducted using either low or standard-dose aspirin (Supplementary table 1). Two of these trials, JPAD [56] and POPADAD [51] were conducted in diabetic individuals while HOT [55] was conducted in hypertensive individuals. ATT Collaborators’ meta-analysis [15] included 6 primary prevention RCTs comprising 95,000 patients, 5 of 6 trials used effective daily dose of \162.5 mg (325 mg alternate day in the Physicians’ Health Study—PHS); British Doctors trial [57] of 5,139 participants, which used 300 and 500 mg/day doses was the only exception. The results of these ATT Collaborators’ analyses of primary prevention RCTs are most relevant for considerations in average-risk individuals. The relative increase in the risk of bleeding associated with aspirin use ranged from 31 to 96 % for GI bleeding, from 31 to 73 % for major bleeding, from 22 to 30 % for fatal bleeding and from 32 to 36 % for haemorrhagic stroke (Table 2). Association between the increase in the risk of bleeding and aspirin dose was not significant in the overviews by Raju et al. [32], Berger et al. [27], and Seshasai et al. [35]; only Serebruany et al. [34] reported a significant relationship between dose and increase in the risk. Derry et al. [28] reported the largest relative increase in the risk of bleeding, a 96 % increase in GI bleeding for doses above 162.5 mg/day, however, the trials included in this dosegroup were relatively smaller trials that used high doses of aspirin, the majority using [900 mg/day. Observational studies

0.2 0.1 0 50-54

55-59

60-64

65-69

70-74

Men (CF)

Women (CF)

Men (ONS)

Women (ONS)

75-79

80-84

Fig. 1 Estimated age and sex specific fatal GI event rates in the UK general population. Age and sex-specific rates (per 1,000 person years) of fatal GI complications—estimated from baseline GI event rates and calculated using mortality data for year 2008 from the Office of National Statistics (ONS) for England and Wales. CF: rates estimated using baseline GI event rates and case-fatality rates. ONS: rates calculated using ONS mortality data for year 2008

Study selection Of the 14 observational studies identified, 12 reported on the risk of GI bleeding; the study by Schmidt et al. [47] reported on the risk of sub-arachnoid haemorrhage (SAH), and the study by De Berardis et al. [48] reported on the risk of major bleeding, these 2 studies were not included in pooled analyses. Additionally, the study by Pilotto et al. [46] was not included in quantitative synthesis since it reported only on selected elderly individuals undergoing endoscopy. A large case–control study by De Berardis et al. [48] investigated the risk of major bleeding, they reported that low-dose aspirin use was

123

M. A. Thorat, J. Cuzick Table 2 Adverse events associated with the use of aspirin in overviews of RCTs Study

Studies (n)

Dose (mg/day)

Aspirin (n)

Control (n)

OR/RR

Events

At risk

Events

At risk

Any GI bleeding Derry [28]

24

\162.5

574

24,964

362

24,963

1.59 (1.40–1.81)

Derry [28]

24

[162.5

258

8,658

100

7,402

1.96 (1.58–2.43)

McQuaid [31]

22

Raju [32]

9

Lanas [30]

35





25,036



25,024

1.62 (1.25–2.09)

\162.5a

1,947

47,439

1,560

47,498

1.37 (1.15–1.62)

B325









1.31 (1.21–1.42)

Major bleeding/non-trivial bleeding/fatal bleeding Major bleeding (extra-cranial) ATT Collab. [15]

6

\162.5b

335

47,293

219

45,618

1.54 (1.30–1.82)

Sanmuganathan [33]

4





25,133



23,407

1.73 (1.14–2.63)





26,673



26,712

1.71 (1.41–2.08)

\162.5a

406

48,968

234

47,244

1.66 (1.41–1.95)

Major bleeding McQuaid [31] Raju [32]

22 9

Berger [27]

9



458

52,145

278

50,476

1.62 (1.31-2.00)

Seshasaic [35]

9



5337

50,868

4,712

49,208

1.31 (1.14–1.50)

Fatal bleeding McQuaid [31]

22





25,497



25,550

1.30 (0.70–2.42)

Lanas [30]

35

B325

62



52



1.22 (0.78–1.89)

Haemorrhagic stroke/intracranial bleeding

a

Hart [29]

5



52

25,133

33

23,407

1.36 (0.88–2.10)

ATT Collab. [15]

6

\162.5b

116

47,293

89

45,618

1.32 (1.00–1.75)

Raju [32] Berger [27]

9 9

\162.5a –

112 113

41,469 42,746

76 78

39,817 41,085

1.36 (1.01–1.82) 1.35 (1.01–1.81)

Connollyd [36]

9

B325

18

50,247

12

47,007

1.60 (0.80–3.50)

8 of 9 trials used effective daily dose of B162.5 mg/day (325 mg alternate day); specific trials are listed in Supplementary Table 1

b

5 of 6 trials used effective daily dose of \162.5 mg/day (325 mg alternate day)

c

Nontrivial bleeding

d

Subdural haematoma

associated with a 55 % increase in the risk of major bleeding (IRR 1.55; 95 % CI 1.48–1.63). Pooled analyses Results of our pooled analyses of 3 cohort and 8 case–control studies are shown in Figs. 2 and 3 respectively. Significant heterogeneity existed among cohort studies (I2 = 92 %), hospital-based case–control studies (I2 = 76 %) and also among the three types (cohort, hospital-based case–control and nested case–control) of studies (I2 = 79 %). In particular, hospital-based case–control studies show a threefold increase in the risk of GI bleeding (OR 2.98; 95 % CI 2.22–3.99), whereas cohort studies (RR 1.88; 95 % CI 1.30–2.74) and case–control studies nested within general practice research databases (OR 1.85; 95 % CI 1.70–2.01), show a much smaller 85–88 % increase in the risk. This is similar to that seen in the meta-analyses of RCTs and it appears to be a more reliable estimate.

123

Hospital-based case–control studies showed a much higher risk of excess GI bleeding associated with aspirin use than that reported in the RCTs, cohort studies and nested case–control studies. Hospital-based case–control studies are prone to recall bias and all but one [43] reported exclusively on UGIB where the risk of aspirin-induced excess bleeding appears to be higher than LGIB according to the data from the Health Professionals Follow-up Study (HPFS) [49] and the Nurses’ Health Study (NHS) [37]. Therefore, the estimates from hospital-based case–control studies are likely to be too high and low or standard-dose aspirin appears to be associated with a 60–70 % increase in the risk of GI bleeding (UGIB ? LGIB) when the results of RCTs and observational studies are considered together. The effect of aspirin dose Two cohort studies, the HPFS [49] and the NHS [37] observed a consistent trend of increasing bleeding risk associated with increasing dose.

Prophylactic use of aspirin

Fig. 2 Risk of GI bleeding due to aspirin—pooled analyses of cohort studies

Fig. 3 Risk of GI bleeding due to aspirin—pooled analyses of case–control studies

The cohort study by Sorensen et al. [50] investigated use of only low-dose aspirin. Two case–control studies by de Abajo et al. [38, 44] and one case–control study by Lanas et al. [41] reported separate risk estimates for low-dose and standard or higher dose aspirin. Although the dose categories are not uniform to permit pooling of data, all these studies report higher estimates of risk of bleeding with higher aspirin doses. The effect of duration of aspirin use The HPFS and NHS explored the effect of duration of aspirin use and both concluded that increasing duration of use does not confer greater risk. On the contrary, both studies observed smaller risk estimates for more than 5 years of aspirin use as compared to short-term use; HPFS short-term use RR 1.52 (95 % CI 1.20–1.93) versus long duration RR 1.39 (95 % CI 1.09–1.78) and NHS short-term use RR 2.17 (95 % CI

1.51–3.12) versus long duration RR 1.52 (95 % CI 1.18–1.95). The site of GI bleeding The HPFS and NHS also reported risk estimates for UGIB and LGIB separately. The relative risk of aspirin-induced UGIB was larger than that of LGIB in both studies; HPFS UGIB RR 1.49 (95 % CI 1.16–1.92) versus LGIB RR 1.22 (95 % CI 0.95–1.56) and NHS UGIB RR 1.70 (95 % CI 1.45–2.00) versus LGIB RR 1.21 (95 % CI 1.03–1.41). Excess peptic ulcer risks due to aspirin use None of the overviews reported specifically on the increase in the risk of peptic ulcers due to aspirin use. Three of the primary prevention trials [57–59] provided relevant data. These show an increase of 20–60 % in the risk of peptic ulcer

123

M. A. Thorat, J. Cuzick

(Table 3). The Women’s health study (WHS) and the PHS using alternate day dosing (325 and 100 mg respectively) showed an approximately 30 % increase in risk as compared to around 60 % increase estimated from the British Doctors’ Trial (BDT), which used 300 and 500 mg daily doses. Excess events attributable to aspirin use Excess events attributable to aspirin use were calculated using an RR of 1.6 for excess GI bleeding and peptic ulcers and an RR of 1.54 for excess major extracranial bleeding events. Estimates for 5-year age-groups are given in Table 4 and shown in Fig. 4. Excess GI complications range from 0.77 per 1,000-PY in women aged 50–54 to 3.40 per 1,000-PY in men aged 70–74. Men have a 1.5-fold higher risk as compared to women and the risk increases threefold from 50–54 to 70–74 years. The numbers of major extra-cranial bleeding events are approximately quarter to third of all GI complications in all age-groups and the risk increases fourfold from 50–54 to 70–74 years. Prevalence of H. pylori infection and potential impact on UGIB or peptic ulcer in aspirin users The search to identify studies reporting the burden of H. pylori infection in the UK produced 513 results

(supplementary figure 4); 5 full-text articles were assessed, 2 were excluded. Three population-based studies have reported the burden of H. pylori infection in the UK [60– 62]. A cohort effect with declining prevalence for each more recent decade of birth was observed in all 3 studies. Vyse et al. [60] investigated prevalence in different parts of England and Wales at 2 different time points, years 1986 and 1996 (Supplementary table 6). Harvey et al. [61] reported very similar rates from the Bristol Helicobacter project (Supplementary table 7). Jackson et al. [62] observed a 26 % prevalence of H. pylori infection overall. Prevalence of H. pylori infection in the UK in those currently aged 50-70 years ranges from 17 to 25 % and is declining in more recent birth cohorts. Two meta-analyses [63, 64] have investigated the separate roles of H. pylori and NSAIDs and the interaction between the two in causing peptic ulcer disease (PUD) whereas Fletcher et al. [65] deemed data in aspirin users to be insufficient to allow meta-analyses. The results of metaanalyses (Supplementary table 8) by Huang et al. [63] and Papatheodoridis et al. [64] suggest that in NSAID users, H. pylori leads to a 3.5-fold increase in the risk of uncomplicated peptic ulcer and a 2.5-fold increase in the risk of bleeding peptic ulcer. The evidence for interaction between H. pylori infection and NSAIDs is lacking. In view of limited data on aspirin, RRs of NSAIDs are used.

Table 3 The relative risks of peptic ulcer associated with aspirin use in primary prevention trials Study

Dose (mg/day)

Aspirin (n)

Control (n)

Events

At risk

RR (95 % CI)

Events

At risk

PHS [59]

162.5a

169

11,037

138

11,034

WHS [58]

50a

542

19,934

413

19,942

1.32 (1.16–1.50)

BDT [57]

300/500

88c

18,820b

28c

9,470b

1.58 (1.03–2.42)c

a

Alternate day dose

b

Person years, patients in each arm were 3,429 (aspirin) and 1,710 (control)

c

Calculated from the data given in the report, RR was not separately reported

1.22 (0.98–1.53)

Table 4 Aspirin-associated excess bleeding and peptic ulcer events in the UK general population Age-group (years)

Major extracranial bleeding

Any GI bleeding

Peptic ulcer

GI bleeding and peptic ulcer

Men

Women

Men

Women

Men

Women

Men

Women

50–54

0.24

0.12

0.79

0.45

0.36

0.31

1.15

0.77

55–59

0.42

0.21

0.63

0.52

0.46

0.38

1.09

0.90

60–64

0.60

0.30

1.45

0.77

0.57

0.47

2.02

1.24

65–69

0.79

0.40

1.91

1.36

0.70

0.57

2.61

1.93

70–74

0.98

0.50

2.63

1.60

0.77

0.63

3.40

2.22

75–79

1.23

0.65

4.20

2.68

0.78

0.65

4.98

3.33

80–84

1.59

0.95

4.92

3.84

0.78

0.65

5.70

4.50

Excess age and sex specific major extracranial bleeding, any GI bleeding, peptic ulcer and any GI complication (GI bleed or peptic ulcer) events (per 1,000 person years) associated with low or standard-dose aspirin in the UK general population Peptic ulcers refer to ulcers that are neither bleeding nor perforated

123

Prophylactic use of aspirin Aspirin-related excess events by age 6

per 1000 person years

5

4

3

2

1

0 50-54

55-59

60-64

65-69

70-74

75-79

80-84

Age Major extracranial bleeding (Men) Major extracranial bleeding (Women) Any GI complicaon (Men) Any GI complicaon (Women)

Fig. 4 Aspirin-associated excess event rates in the UK general population. Excess major extracranial bleeding and any GI complication events (per 1,000 person years) associated with low or standard-dose aspirin in the UK general population

Assuming an average prevalence of 21 % in individuals aged 50–70 years in the UK, H. pylori infection is estimated to account for 34 % of peptic ulcers and 24 % of all bleeding peptic ulcers in NSAID users. H. pylori infection does not have effect on LGIB.

Discussion To ascertain the harms of any intervention, the baseline level of these events, their severity and the extent to which these events are increased by the intervention under study need to be determined. We conducted this study to determine potential harms of prophylactic aspirin use in average-risk general population. We focussed on data from the United Kingdom since the UK general population is similar to most of the countries in Western Europe and North America [66]; and universal healthcare in the UK also results in a reasonable uniformity in treatment, outcomes and data collection. Aspirin appears to have a dose response effect on bleeding above doses of 325 mg/day but there is no clear evidence for dose effect below this level. If the results from higher dose aspirin [28] are not used, low or standard-dose aspirin is associated with a 60–70 % increase in the risk of GI bleeding or major bleeding (Table 2). On the basis of these data, we used a 60 % increase in the risk of GI bleeding and 54 % increase (as per ATT data) in the risk of major extracranial bleeding as our best-

estimates for estimating excess of these events due to aspirin use in the UK general population (Table 4). Increase in the risk of peptic ulcer is between 30 and 60 % for low or standard-dose aspirin; we used a conservative 60 % increase for estimating excess of peptic ulcers (Table 4). The relative increase in the risk of haemorrhagic stroke is lower, at around 35 %. We did not separately estimate excess of haemorrhagic strokes as we believe that it is more appropriate to consider all strokes together and overall effects of aspirin on all strokes since aspirin prevents ischemic strokes resulting in an overall reduction in strokes. The aspirin-associated relative increase in the risk of fatal bleeding is lower (20–30 %) than the increase in the risk of bleeding overall [15, 30, 31, 67]. Further follow-up and analyses of major trials is needed to confirm or refute whether aspirin-associated bleeding has a lower probability of being fatal, and to suggest possible reasons for such an effect. It has been observed that excess bleeding events are more frequent during initial period after starting aspirin [67, 68] and excess risk disappears after 5 years of aspirin use [67]. Data from the HPFS and the NHS [37, 49] also show that the excess bleeding risk is lower in long-term users ([5 years), probably due to a dilution of early effects with lower late effects. Gastric adaptation to aspirin use is well known, and it may be partially responsible for this normalisation of the risk of bleeding [69–71]. How much does stopping aspirin use or use of gastro-protective drugs after a bleeding event contribute to such normalisation of the risk is not clear and further information is needed on risk as a function of duration of use. Of the three studies reporting age and sex-specific rates of GI bleeding and/or peptic ulcer in the UK general population, we used studies by Cai et al. [12] and Hippisley-Cox et al. [14] to estimate rates as these reported data from most recent period and were comparable. In addition to upper GI mucosal damage, aspirin also causes mucosal injury in the small bowel [72–76] and it also interferes with the coagulation pathway through antiplatelet effects. Therefore, it is important to consider aspirin’s effects on LGIB events as well, although the magnitude of such effect is likely to be smaller than that on UGIB as seen in the HPFS [49] and the NHS [37]. None of the studies reporting on baseline rates provided separate information on LGIB events. Overviews reporting effects of aspirin on GI bleeding also do not report upper and lower GI bleeding events separately [28, 30–32]. Data from the HPFS and the NHS however show that the number of UGIB and LGIB events in these cohorts are similar [37, 49], whereas other data [77] indicate that LGIB event rates are lower than UGIB rates. We multiplied UGIB rates by 2 to obtain total GI bleeding rates to avoid any possible underestimation of total GI bleeding rates. Further research

123

M. A. Thorat, J. Cuzick

to carefully assess aspirin’s effects on UGIB and LGIB separately and to ascertain these baseline rates in different populations is merited. Herna´ndez-Dı´az et al. [78] estimated the baseline risk of UGICs in the general populations from the UK and Spain, stratified according to known major gastrointestinal risk factors. They also calculated the risk attributable to aspirin using various RRs from 1.5 to 3.0. These estimates were used in the U.S. Preventive Services Task Force (USPSTF) review [79, 80]. We believe that consideration of only UGICs, ignoring other bleeding events, notably LGIB and not taking into account the severity of harm (e.g. major versus trivial bleeding) resulted in an analysis which was too limited to form a basis for a benefit-harm analysis. Definitions of major bleeding vary across studies; from bleeding requiring hospitalisation irrespective of duration to bleeding requiring transfusion [35, 48]. The trials within the ATT Collaborators’ meta-analysis [15] also used variable definitions. In spite of the variations in definitions, we believe that the results from primary prevention RCTs are the most robust and relevant for making estimates in average-risk individuals and therefore we have used these data to estimate rates of major extracranial bleeding. In particular, the design and participant profiles of the 6 primary prevention RCTs included in the ATT Collaborators’ meta-analysis [15] also indicate that these data are likely to be representative of the general population of Europe and North America. De Berardis et al. [48] reported major bleeding rates in Italy, but their age-specific rates in nonusers were approximately twice the rates we estimated from ATT data. The differences in their rates and our estimates are likely to be due to differences in populations and definitions. The population in the study by De Berardis et al. [48] had higher prevalence of diabetes mellitus (15 %) and hypertension (57 %), both of which are risk factors for bleeding [15] and the proportion of the population using other NSAIDs, another risk factor, was also higher (35 %). Therefore, our estimates of major extracranial bleeding are likely to be more representative of average-risk population in the developed western nations. Case-fatality due to GI bleeding and/or peptic ulcer is difficult to measure since it is affected by numerous factors like age, co-morbidity, underlying pathologies, treatment standards etc. For these reasons, we estimated rates for fatal GI bleeding or peptic ulcer by two methods; first calculated based on data reported in literature and second directly using data from the ONS. For the estimation using data reported in literature, we attempted to collect a larger amount of information by using a longer period in this search and also included studies irrespective of region. However, since casefatality rates are improving over time [19, 25], we chose not to include studies reported more than 15 years ago. The studies we identified provide a consistent picture and

123

suggest that the case-fatality rates in hospitalised GI bleeding and peptic ulcer are approximately 5 % in individuals younger than 70 years of age and 10 % in older individuals (Supplementary table 2). As reported by others [7, 78], the rates of fatal GI events rise sharply in older individuals (Fig. 1), due to higher baseline rates [26] and higher fatality [17] of GI events in older individuals. The case-fatality rates used here relate to hospitalised patients i.e. fatality in major bleeding, the overall rates in all GI bleeding or peptic ulcers (hospitalised and non-hospitalised) are likely to be lower. Therefore, it is likely that our fatal GI event rates (FGI-CF) are an overestimation of the true rates. Indeed, the rates (FGI-ONS) calculated using ONS mortality data for year 2008 [11] are 50–60 % lower (Fig. 1). Patients with major co-morbidities and already hospitalised patients constitute a substantial proportion of fatal GI events as the fatality rates are higher in them [23]. Cause-of-death ascertainment in these patients may not always be accurate due to pre-existing major illness or due to the illness necessitating hospital admission. Therefore, the possibility of under-ascertainment in ONS data for the specific ICD codes we used cannot be ruled out, and it is possible that the FGI-ONS rates are an underestimation of the true rates. The prevalence of H. pylori infection in the UK in individuals aged 50-70 years is substantial, with 17–25 % of individuals being infected. Similar or higher levels of prevalence have been reported from other western developed countries like Canada [81], USA [82] and Germany [83–85]. Prevalence is much lower at younger ages (10 % or lower in individuals \40 years) [60, 61]. However, in developing countries like China, high prevalence has been reported even in these age-groups [86, 87]. Due to limited data on aspirin, we used RRs for all NSAID users. We estimate that H. pylori infection is responsible for a quarter to a third of UGIB events or peptic ulcers in NSAID users but has no impact on LGIB. RCTs of H. pylori eradication in patients with [88] or without [89] dyspepsia or previous ulcer history before starting NSAID showed lower ulcer rates in those with successful eradication. H. pylori eradication [90] has also been shown to be equivalent to omeprazole maintenance therapy in aspirin users with healed, but previously bleeding H. pylori positive peptic ulcers. Although it is unclear whether ulcer healing is impaired due to antibiotic therapy [91–93], H. pylori screening and eradication has a significant potential to minimise aspirinrelated UGIB events and peptic ulcers, and it has been deemed cost-effective by the UK Health Technology Assessment (HTA) to reduce mortality and morbidity from gastric cancer and peptic ulcer disease [94]. The HEAT trial (ClinicalTrials.gov Identifier: NCT01506986) is evaluating such a strategy in current aspirin users. Concomitant use of proton pump inhibitors (PPIs) has also been shown to reduce adverse GI events by 66 % (OR

Prophylactic use of aspirin

0.34; 95 % CI 0.21–0.57) in a meta-analysis of 35 trials [30]. It has been reported to be cost-effective in secondary prevention setting [95]. A combined approach of concomitant PPI use and H. pylori eradication has been proposed in individuals starting aspirin and at high-risk of peptic ulcer [96]. Optimal control of blood pressure will also help in minimising intracranial bleeding events [55].

Conclusions We have estimated age and sex-specific baseline rates of GI bleeding, peptic ulcer, major extra-cranial bleeding and fatal GI bleeding or peptic ulcer in the UK general population. We also estimated the excess events likely to be caused by prophylactic aspirin use and the proportion of events attributable to easily modifiable factors like H. pylori infection. In average-risk individuals aged 50–70 years, who are the most likely candidates, prophylactic use of aspirin can result in an excess of 0.1–0.8 major extracranial bleeding events per 1,000 per year. The excess of GI complications for aspirin use between 50 and 70 years of age is estimated to be 0.8–2.6 per 1,000 per year depending on age and sex. 5–10 % of major GI complications can be fatal. The bleeding and ulcer event rates rise steeply after 70 years of age. Results from RCTs and observational studies suggest that the risk of aspirin-related excess bleeding decreases with long-term use, and raise the possibility that aspirin-associated bleeding events may have a lower likelihood of being fatal. Our results also show that H. pylori eradication can have a substantial effect in reducing aspirin-induced upper GI complications. Our estimates of aspirin-related adverse events allow a balanced analysis of benefits and harms of prophylactic aspirin use. Acknowledgments This study was a part of broader review of prophylactic use of aspirin. The review was sponsored by International Society of Cancer Prevention (ISCaP), Cancer Research UK (CRUK), British Heart Foundation (BHF) and American Cancer Society (ACS) and received funding from CRUK, BHF and ACS. Sponsors and funding sources had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Conflict of interest JC: Member of the Bayer advisory board. MAT: None. Ethical standard or patient data.

The manuscript does not contain clinical studies

References 1. Mahdi JG, Mahdi AJ, Mahdi AJ, Bowen ID. The historical analysis of aspirin discovery, its relation to the willow tree and antiproliferative and anticancer potential. Cell Prolif. 2006;39(2):147–55. doi:10.1111/j.1365-2184.2006.00377.x.

2. Sneader W. The discovery of aspirin: a reappraisal. BMJ. 2000;321(7276):1591–4. doi:10.1136/bmj.321.7276.1591. 3. Miner J, Hoffhines A. The discovery of aspirin’s antithrombotic effects. Tex Heart Inst J. 2007;34(2):179–86. 4. Cuzick J, Otto F, Baron JA, et al. Aspirin and non-steroidal antiinflammatory drugs for cancer prevention: an international consensus statement. Lancet Oncol. 2009;10(5):501–7. doi:10.1016/ S1470-2045(09)70035-X. 5. Thorat MA, Cuzick J. Role of aspirin in cancer prevention. Curr Oncol Rep. 2013;15(6):533–40. doi:10.1007/s11912-013-0351-3. 6. Hunter DJ, Reddy KS. Noncommunicable diseases. N Engl J Med. 2013;369(14):1336–43. doi:10.1056/NEJMra1109345. 7. Patrono C, Garcia Rodriguez LA, Landolfi R, Baigent C. Lowdose aspirin for the prevention of atherothrombosis. N Engl J Med. 2005;353(22):2373–83. doi:10.1056/NEJMra052717. 8. Suerbaum S, Michetti P. Helicobacter pylori infection. N Engl J Med. 2002;347(15):1175–86. doi:10.1056/NEJMra020542. 9. Cuzick J, Thorat MA, Bosetti C, et al. Estimates of benefits and harms of prophylactic use of aspirin in the general population. Ann Oncol. 2014;. doi:10.1093/annonc/mdu225. 10. Moher D, Liberati A, Tetzlaff J, Altman DG, Group P. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ. 2009;339:b2535. doi:10.1136/bmj. b2535. 11. Office for National Statistics. Mortality statistics: deaths registered in 2008. National Statistics London; 2009. 12. Cai S, Garcia Rodriguez LA, Masso-Gonzalez EL, HernandezDiaz S. Uncomplicated peptic ulcer in the UK: trends from 1997 to 2005. Aliment Pharmacol Ther. 2009;30(10):1039–48. doi:10. 1111/j.1365-2036.2009.04131.x. 13. Garcia Rodriguez LA, Hernandez-Diaz S. Risk of uncomplicated peptic ulcer among users of aspirin and nonaspirin nonsteroidal antiinflammatory drugs. Am J Epidemiol. 2004;159(1):23–31. doi:10.1093/aje/kwh005. 14. Hippisley-Cox J, Coupland C, Logan R. Risk of adverse gastrointestinal outcomes in patients taking cyclo-oxygenase-2 inhibitors or conventional non-steroidal anti-inflammatory drugs: population based nested case-control analysis. BMJ. 2005;331(7528):1310–6. doi:10.1136/bmj.331.7528.1310. 15. Baigent C, Blackwell L, Collins R, et al. Aspirin in the primary and secondary prevention of vascular disease: collaborative metaanalysis of individual participant data from randomised trials. Lancet. 2009;373(9678):1849–60. doi:10.1016/S0140-6736(09) 60503-1. 16. Button LA, Roberts SE, Evans PA, et al. Hospitalized incidence and case fatality for upper gastrointestinal bleeding from 1999 to 2007: a record linkage study. Aliment Pharmacol Ther. 2010;33(1):64–76. doi:10.1111/j.1365-2036.2010.04495.x. 17. Garcia Rodriguez LA, Ruigomez A, Hasselgren G, Wallander MA, Johansson S. Comparison of mortality from peptic ulcer bleed between patients with or without peptic ulcer antecedents. Epidemiology. 1998;9(4):452–6. 18. Hasselgren G, Blomqvist A, Eriksson S, Henningsson A, Lundell L. Short and long term course of elderly patients with peptic ulcer bleeding—analysis of factors influencing fatal outcome. Eur J Surg. 1998;164(9):685–91. doi:10.1080/110241598750005570. 19. Kang JY, Elders A, Majeed A, Maxwell JD, Bardhan KD. Recent trends in hospital admissions and mortality rates for peptic ulcer in Scotland 1982–2002. Aliment Pharmacol Ther. 2006;24(1): 65–79. doi:10.1111/j.1365-2036.2006.02960.x. 20. Lanas A, Garcia-Rodriguez LA, Polo-Tomas M, et al. Time trends and impact of upper and lower gastrointestinal bleeding and perforation in clinical practice. Am J Gastroenterol. 2009;104(7):1633–41. doi:10.1038/ajg.2009.164. 21. Morales Uribe CH, Sierra Sierra S, Hernandez Hernandez AM, Arango Durango AF, Lopez GA. Upper gastrointestinal bleeding:

123

M. A. Thorat, J. Cuzick

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

risk factors for mortality in two urban centres in Latin America. Revista espanola de enfermedades digestivas: organo oficial de la Sociedad Espanola de Patologia Digestiva. 2011;103(1):20–4. Paspatis GA, Matrella E, Kapsoritakis A, et al. An epidemiological study of acute upper gastrointestinal bleeding in Crete, Greece. Eur J Gastroenterol Hepatol. 2000;12(11):1215–20. Hearnshaw SA, Logan RF, Lowe D, Travis SP, Murphy MF, Palmer KR. Acute upper gastrointestinal bleeding in the UK: patient characteristics, diagnoses and outcomes in the 2007 UK audit. Gut. 2011;60(10):1327–35. doi:10.1136/gut.2010.228437. Mose H, Larsen M, Riis A, Johnsen SP, Thomsen RW, Sorensen HT. Thirty-day mortality after peptic ulcer bleeding in hospitalized patients receiving low-dose aspirin at time of admission. Am J Geriatr Pharmacother. 2006;4(3):244–50. doi:10.1016/j.amjo pharm.2006.09.006. Straube S, Tramer MR, Moore RA, Derry S, McQuay HJ. Mortality with upper gastrointestinal bleeding and perforation: effects of time and NSAID use. BMC Gastroenterol. 2009;9:41. doi:10. 1186/1471-230X-9-41. Lanas A, Perez-Aisa MA, Feu F, et al. A nationwide study of mortality associated with hospital admission due to severe gastrointestinal events and those associated with nonsteroidal antiinflammatory drug use. Am J Gastroenterol. 2005;100(8):1685–93. doi:10.1111/j.1572-0241.2005.41833.x. Berger JS, Lala A, Krantz MJ, Baker GS, Hiatt WR. Aspirin for the prevention of cardiovascular events in patients without clinical cardiovascular disease: a meta-analysis of randomized trials. Am Heart J. 2011;162(1):115–24. doi:10.1016/j.ahj.2011. 04.006. Derry S, Loke YK. Risk of gastrointestinal haemorrhage with long term use of aspirin: meta-analysis. BMJ. 2000;321(7270):1183–7. doi:10.1136/bmj.321.7270.1183. Hart RG, Halperin JL, McBride R, Benavente O, Man-Son-Hing M, Kronmal RA. Aspirin for the primary prevention of stroke and other major vascular events: meta-analysis and hypotheses. Arch Neurol. 2000;57(3):326–32. doi:10.1001/archneur.57.3.326. Lanas A, Wu P, Medin J, Mills EJ. Low doses of acetylsalicylic acid increase risk of gastrointestinal bleeding in a meta-analysis. Clin Gastroenterol Hepatol. 2011;9(9):762–8. doi:10.1016/j.cgh. 2011.05.020. McQuaid KR, Laine L. Systematic review and meta-analysis of adverse events of low-dose aspirin and clopidogrel in randomized controlled trials. Am J Med. 2006;119(8):624–38. doi:10.1016/j. amjmed.2005.10.039. Raju N, Sobieraj-Teague M, Hirsh J, O’Donnell M, Eikelboom J. Effect of aspirin on mortality in the primary prevention of cardiovascular disease. Am J Med. 2011;124(7):621–9. doi:10.1016/ j.amjmed.2011.01.018. Sanmuganathan PS, Ghahramani P, Jackson PR, Wallis EJ, Ramsay LE. Aspirin for primary prevention of coronary heart disease: safety and absolute benefit related to coronary risk derived from meta-analysis of randomised trials. Heart. 2001;85(3):265–71. doi:10.1136/heart.85.3.265. Serebruany VL, Steinhubl SR, Berger PB, et al. Analysis of risk of bleeding complications after different doses of aspirin in 192,036 patients enrolled in 31 randomized controlled trials. Am J Cardiol. 2005;95(10):1218–22. doi:10.1016/j.amjcard. 2005.01.049. Seshasai SR, Wijesuriya S, Sivakumaran R, et al. Effect of aspirin on vascular and nonvascular outcomes: meta-analysis of randomized controlled trials. Arch Intern Med. 2012;172(3):209–16. doi:10.1001/archinternmed.2011.628. Connolly BJ, Pearce LA, Kurth T, Kase CS, Hart RG. Aspirin therapy and risk of subdural hematoma: meta-analysis of randomized clinical trials. J Stroke Cerebrovasc Dis. 2013;22(4):444–8. doi:10.1016/j.jstrokecerebrovasdis.2013.01.007.

123

37. Huang ES, Strate LL, Ho WW, Lee SS, Chan AT. Long-term use of aspirin and the risk of gastrointestinal bleeding. Am J Med. 2011;124(5):426–33. doi:10.1016/j.amjmed.2010.12.022. 38. de Abajo FJ, Gil MJ, Bryant V, Timoner J, Oliva B, GarciaRodriguez LA. Upper gastrointestinal bleeding associated with NSAIDs, other drugs and interactions: a nested case-control study in a new general practice database. Eur J Clin Pharmacol. 2013;69(3):691–701. doi:10.1007/s00228-012-1386-3. 39. Garcia Rodriguez LA, Lin KJ, Hernandez-Diaz S, Johansson S. Risk of upper gastrointestinal bleeding with low-dose acetylsalicylic acid alone and in combination with clopidogrel and other medications. Circulation. 2011;123(10):1108–15. doi:10.1161/ CIRCULATIONAHA.110.973008. 40. Sakamoto C, Sugano K, Ota S, et al. Case-control study on the association of upper gastrointestinal bleeding and nonsteroidal anti-inflammatory drugs in Japan. Eur J Clin Pharmacol. 2006;62(9):765–72. doi:10.1007/s00228-006-0171-6. 41. Lanas A, Garcia-Rodriguez LA, Arroyo MT, et al. Risk of upper gastrointestinal ulcer bleeding associated with selective cyclooxygenase-2 inhibitors, traditional non-aspirin non-steroidal antiinflammatory drugs, aspirin and combinations. Gut. 2006;55(12):1731–8. doi:10.1136/gut.2005.080754. 42. Ibanez L, Vidal X, Vendrell L, Moretti U, Laporte JR, SpanishItalian Collaborative Group for the Epidemiology of Gastrointestinal B. Upper gastrointestinal bleeding associated with antiplatelet drugs. Aliment Pharmacol Ther. 2006;23(2):235–42. doi:10.1111/j.1365-2036.2006.02759.x. 43. Sapoznikov B, Vilkin A, Hershkovici M, Fishman M, Eliakim R, Niv Y. Minidose aspirin and gastrointestinal bleeding—a retrospective, case–control study in hospitalized patients. Dig Dis Sci. 2005;50(9):1621–4. doi:10.1007/s10620-005-2906-0. 44. de Abajo FJ, Garcia Rodriguez LA. Risk of upper gastrointestinal bleeding and perforation associated with low-dose aspirin as plain and enteric-coated formulations. BMC Clin Pharmacol. 2001;1:1. doi:10.1186/1472-6904-1-1. 45. Lanas A, Bajador E, Serrano P, et al. Nitrovasodilators, low-dose aspirin, other nonsteroidal antiinflammatory drugs, and the risk of upper gastrointestinal bleeding. N Engl J Med. 2000;343(12):834–9. doi:10.1056/NEJM200009213431202. 46. Pilotto A, Franceschi M, Leandro G, et al. The risk of upper gastrointestinal bleeding in elderly users of aspirin and other nonsteroidal anti-inflammatory drugs: the role of gastroprotective drugs. Aging Clin Exp Res. 2003;15(6):494–9. 47. Schmidt M, Johansen MB, Lash TL, Christiansen CF, Christensen S, Sorensen HT. Antiplatelet drugs and risk of subarachnoid hemorrhage: a population-based case-control study. J Thromb Haemost. 2010;8(7):1468–74. doi:10.1111/j.1538-7836.2010. 03856.x. 48. De Berardis G, Lucisano G, D’Ettorre A, et al. Association of aspirin use with major bleeding in patients with and without diabetes. JAMA. 2012;307(21):2286–94. doi:10.1001/jama.2012. 5034. 49. Huang ES, Strate LL, Ho WW, Lee SS, Chan AT. A prospective study of aspirin use and the risk of gastrointestinal bleeding in men. PLoS ONE. 2010;5(12):e15721. doi:10.1371/journal.pone. 0015721. 50. Sorensen HT, Mellemkjaer L, Blot WJ, et al. Risk of upper gastrointestinal bleeding associated with use of low-dose aspirin. Am J Gastroenterol. 2000;95(9):2218–24. doi:10.1111/j.15720241.2000.02248.x. 51. Belch J, MacCuish A, Campbell I, et al. The prevention of progression of arterial disease and diabetes (POPADAD) trial: factorial randomised placebo controlled trial of aspirin and antioxidants in patients with diabetes and asymptomatic peripheral arterial disease. BMJ. 2008;337:a1840. doi:10.1136/bmj. a1840.

Prophylactic use of aspirin 52. de Gaetano G. Low-dose aspirin and vitamin E in people at cardiovascular risk: a randomised trial in general practice. Collaborative group of the primary prevention project. Lancet. 2001;357(9250):89–95. doi:10.1016/S0140-6736(00)03539-X. 53. Fowkes FG, Price JF, Stewart MC, et al. Aspirin for prevention of cardiovascular events in a general population screened for a low ankle brachial index: a randomized controlled trial. JAMA. 2010;303(9):841–8. doi:10.1001/jama.2010.221. 54. The Medical Research Council’s General Practice Research Framework. Thrombosis prevention trial: randomised trial of low-intensity oral anticoagulation with warfarin and low-dose aspirin in the primary prevention of ischaemic heart disease in men at increased risk. The medical research council’s general practice research framework. Lancet. 1998;351(9098):233–41. doi:10.1016/S0140-6736(97)11475-1. 55. Hansson L, Zanchetti A, Carruthers SG, et al. Effects of intensive blood-pressure lowering and low-dose aspirin in patients with hypertension: principal results of the hypertension optimal treatment (HOT) randomised trial. HOT study group. Lancet. 1998;351(9118):1755–62. doi:10.1016/S0140-6736(98)04311-6. 56. Ogawa H, Nakayama M, Morimoto T, et al. Low-dose aspirin for primary prevention of atherosclerotic events in patients with type 2 diabetes: a randomized controlled trial. JAMA. 2008;300(18): 2134–41. doi:10.1001/jama.2008.623. 57. Peto R, Gray R, Collins R, et al. Randomised trial of prophylactic daily aspirin in British male doctors. Br Med J (Clin Res Ed). 1988;296(6618):313–6. 58. Ridker PM, Cook NR, Lee IM, et al. A randomized trial of lowdose aspirin in the primary prevention of cardiovascular disease in women. N Engl J Med. 2005;352(13):1293–304. doi:10.1056/ NEJMoa050613. 59. Steering Committee of the Physicians’ Health Study Research Group. Final report on the aspirin component of the ongoing physicians’ health study. N Engl J Med. 1989;321(3):129–35. doi:10.1056/NEJM198907203210301. 60. Vyse AJ, Gay NJ, Hesketh LM, et al. The burden of Helicobacter pylori infection in England and Wales. Epidemiol Infect. 2002;128(3):411–7. 61. Harvey RF, Spence RW, Lane JA, et al. Relationship between the birth cohort pattern of Helicobacter pylori infection and the epidemiology of duodenal ulcer. QJM Mon J Assoc Phys. 2002;95(8):519–25. 62. Jackson L, Britton J, Lewis SA, et al. A population-based epidemiologic study of Helicobacter pylori infection and its association with systemic inflammation. Helicobacter. 2009;14(5): 108–13. doi:10.1111/j.1523-5378.2009.00711.x. 63. Huang JQ, Sridhar S, Hunt RH. Role of Helicobacter pylori infection and non-steroidal anti-inflammatory drugs in pepticulcer disease: a meta-analysis. Lancet. 2002;359(9300):14–22. doi:10.1016/S0140-6736(02)07273-2. 64. Papatheodoridis GV, Sougioultzis S, Archimandritis AJ. Effects of Helicobacter pylori and nonsteroidal anti-inflammatory drugs on peptic ulcer disease: a systematic review. Clin Gastroenterol Hepatol. 2006;4(2):130–42. doi:10.1016/j.cgh.2005.10.006. 65. Fletcher EH, Johnston DE, Fisher CR, Koerner RJ, Newton JL, Gray CS. Systematic review: Helicobacter pylori and the risk of upper gastrointestinal bleeding risk in patients taking aspirin. Aliment Pharmacol Ther. 2010;32(7):831–9. doi:10.1111/j.13652036.2010.04415.x. 66. Mathers C, Fat DM, Boerma JT, World Health Organization. The global burden of disease: 2004 update. Geneva: World Health Organization; 2008. 67. Rothwell PM, Price JF, Fowkes FG, et al. Short-term effects of daily aspirin on cancer incidence, mortality, and non-vascular death: analysis of the time course of risks and benefits in 51

68.

69.

70. 71.

72.

73.

74.

75.

76.

77.

78.

79.

80.

81.

82.

83.

randomised controlled trials. Lancet. 2012;379(9826):1602–12. doi:10.1016/S0140-6736(11)61720-0. Garcia Rodriguez LA, Hernandez-Diaz S, de Abajo FJ. Association between aspirin and upper gastrointestinal complications: systematic review of epidemiologic studies. Br J Clin Pharmacol. 2001;52(5):563–71. doi:10.1046/j.0306-5251.2001.01476.x. Graham DY, Smith JL, Spjut HJ, Torres E. Gastric adaptation. Studies in humans during continuous aspirin administration. Gastroenterology. 1988;95(2):327–33. Graham DY, Smith JL, Dobbs SM. Gastric adaptation occurs with aspirin administration in man. Dig Dis Sci. 1983;28(1):1–6. Kawai T, Yamagishi T, Goto S. Circadian variations of gastrointestinal mucosal damage detected with transnasal endoscopy in apparently healthy subjects treated with low-dose aspirin (ASA) for a short period. J Atheroscler Thromb. 2009;16(3):155–63. doi:10.5551/jat.E615. Watanabe T, Sugimori S, Kameda N, et al. Small bowel injury by low-dose enteric-coated aspirin and treatment with misoprostol: a pilot study. Clin Gastroenterol Hepatol. 2008;6(11):1279–82. doi:10.1016/j.cgh.2008.06.021. Shiotani A, Haruma K, Nishi R, et al. Randomized, double-blind, pilot study of geranylgeranylacetone versus placebo in patients taking low-dose enteric-coated aspirin. Low-dose aspirin-induced small bowel damage. Scand J Gastroenterol. 2010;45(3):292–8. doi:10.3109/00365520903453182. Endo H, Sakai E, Higurashi T, et al. Differences in the severity of small bowel mucosal injury based on the type of aspirin as evaluated by capsule endoscopy. Dig Liver Dis Off J Italian Soc Gastroenterol Italian Assoc Study Liver. 2012;44(10):833–8. doi:10.1016/j.dld.2012.05.016. Matsumura T, Arai M, Sazuka S, et al. Negative capsule endoscopy for obscure gastrointestinal bleeding is closely associated with the use of low-dose aspirin. Scand J Gastroenterol. 2011;46(5):621–6. doi:10.3109/00365521.2010.545833. Endo H, Hosono K, Inamori M, et al. Characteristics of small bowel injury in symptomatic chronic low-dose aspirin users: the experience of two medical centers in capsule endoscopy. J Gastroenterol. 2009;44(6):544–9. doi:10.1007/s00535-009-0040-z. Rahme E, Roussy JP, Woolcott J, Nedjar H, Barkun A. Mortality and readmission rates after hospitalization for upper and lower gastrointestinal events in Quebec, Canada. J Clin Gastroenterol. 2013;47(7):586–92. doi:10.1097/MCG.0b013e318282a1d7. Hernandez-Diaz S, Garcia Rodriguez LA. Cardioprotective aspirin users and their excess risk of upper gastrointestinal complications. BMC Med. 2006;4:22. doi:10.1186/1741-7015-422. U.S. Preventive Services Task Force (USPSTF). Aspirin for the prevention of cardiovascular disease: U.S. Preventive Services task force recommendation statement. Ann Intern Med. 2009;150(6):396–404. doi:10.7326/0003-4819-150-6-20090317000008. Wolff T, Miller T, Ko S. Aspirin for the primary prevention of cardiovascular events: an update of the evidence for the U.S. preventive services task force. Ann Intern Med. 2009;150(6):405–10. doi:10.7326/0003-4819-150-6-200903170-00009. Naja F, Kreiger N, Sullivan T. Helicobacter pylori infection in Ontario: prevalence and risk factors. Canadian journal of gastroenterology. J Can de gastroenterol. 2007;21(8):501–6. Zajacova A, Dowd JB, Aiello AE. Socioeconomic and race/ethnic patterns in persistent infection burden among U.S. adults. J Gerontol Ser A Biol Sci Med Sci. 2009;64(2):272–9. doi:10. 1093/gerona/gln012. Gao L, Weck MN, Stegmaier C, Rothenbacher D, Brenner H. Alcohol consumption, serum gamma-glutamyltransferase, and Helicobacter pylori infection in a population-based study among

123

M. A. Thorat, J. Cuzick

84.

85.

86.

87.

88.

89.

90.

9733 older adults. Ann Epidemiol. 2010;20(2):122–8. doi:10. 1016/j.annepidem.2009.11.009. Schottker B, Adamu MA, Weck MN, Muller H, Brenner H. Helicobacter pylori infection, chronic atrophic gastritis and major cardiovascular events: a population-based cohort study. Atherosclerosis. 2012;220(2):569–74. doi:10.1016/j.atherosclerosis. 2011.11.029. Bode G, Rothenbacher D, Brenner H. Helicobacter pylori colonization and diarrhoeal illness: results of a population-based cross-sectional study in adults. Eur J Epidemiol. 2001;17(9): 823–7. doi:10.1023/A:1015618112695. Xia P, Ma MF, Wang W. Status of Helicobacter pylori infection among migrant workers in Shijiazhuang, China. Asian Pac J Cancer Prev. 2012;13(4):1167–70. Zhang DH, Zhou LY, Lin SR, et al. Recent changes in the prevalence of Helicobacter pylori infection among children and adults in high- or low-incidence regions of gastric cancer in China. Chin Med J. 2009;122(15):1759–63. Chan FK, To KF, Wu JC, et al. Eradication of Helicobacter pylori and risk of peptic ulcers in patients starting long-term treatment with non-steroidal anti-inflammatory drugs: a randomised trial. Lancet. 2002;359(9300):9–13. doi:10.1016/S01406736(02)07272-0. Chan FK, Sung JJ, Chung SC, et al. Randomised trial of eradication of Helicobacter pylori before non-steroidal anti-inflammatory drug therapy to prevent peptic ulcers. Lancet. 1997;350(9083):975–9. doi:10.1016/S0140-6736(97)04523-6. Chan FK, Chung SC, Suen BY, et al. Preventing recurrent upper gastrointestinal bleeding in patients with Helicobacter pylori infection who are taking low-dose aspirin or naproxen. N Engl J

123

91.

92.

93.

94.

95.

96.

Med. 2001;344(13):967–73. doi:10.1056/NEJM200103293441 304. Bianchi Porro G, Parente F, Imbesi V, Montrone F, Caruso I. Role of Helicobacter pylori in ulcer healing and recurrence of gastric and duodenal ulcers in longterm NSAID users. Response to omeprazole dual therapy. Gut. 1996;39(1):22–6. Chan FK, Sung JJ, Suen R, et al. Does eradication of Helicobacter pylori impair healing of nonsteroidal anti-inflammatory drug associated bleeding peptic ulcers? A prospective randomized study. Aliment Pharmacol Ther. 1998;12(12):1201–5. doi:10.1046/j.1365-2036.1998.00434.x. Hawkey CJ, Tulassay Z, Szczepanski L, et al. Randomised controlled trial of Helicobacter pylori eradication in patients on nonsteroidal anti-inflammatory drugs: HELP NSAIDs study. Helicobacter eradication for lesion prevention. Lancet. 1998;352(9133):1016–21. doi:10.1016/S0140-6736(98)04206-8. Roderick P, Davies R, Raftery J, et al. The cost-effectiveness of screening for Helicobacter pylori to reduce mortality and morbidity from gastric cancer and peptic ulcer disease: a discreteevent simulation model. Health Technol Assess. 2003;7(6):1–86. doi:10.3310/hta7060. Saini SD, Schoenfeld P, Fendrick AM, Scheiman J. Cost-effectiveness of proton pump inhibitor cotherapy in patients taking long-term, low-dose aspirin for secondary cardiovascular prevention. Arch Intern Med. 2008;168(15):1684–90. doi:10.1001/ archinte.168.15.1684. Lanas A. Prevention of aspirin-induced gastroduodenal damage: H. pylori infection eradication versus proton pump inhibitors or both. Dig Liver Dis. 2004;36(10):655–7. doi:10.1016/j.dld.2004. 06.010.

Prophylactic use of aspirin: systematic review of harms and approaches to mitigation in the general population.

A careful assessment of benefits and harms is required to assess suitability of aspirin as a prophylactic public health measure. However, comprehensiv...
662KB Sizes 0 Downloads 5 Views