Been Taking Baby Aspirin. For 10 Days Is It Safe to Stop Taking Them

Introduction

Clinical Perspective

What Is New?

Discontinuing long-term low-dose aspirin treatment in the absenteeism of major surgery or bleeding may exist hazardous.
Nosotros investigated that question among 601 527 users of low-dose aspirin for principal or secondary prevention identified in the Swedish drug prescription register.
Patients who discontinued aspirin had a 37% higher rate of cardiovascular events than those who continued, respective to an additional cardiovascular result observed per twelvemonth in one of every 74 patients who discontinue aspirin.
The risk increased shortly after discontinuation and did not appear to diminish over time.

What Are the Clinical Implications?

Adherence to depression-dose aspirin handling in the absence of major surgery or bleeding is likely an important treatment goal.
Millions of patients worldwide take aspirin on a daily basis and might consider stopping at some time during their lives.
This article may help physicians and patients to make an informed decision on whether to stop aspirin apply.

Low-dose aspirin has potent evidence for use in the secondary prevention of cardiovascular disease and is uniformly recommended in guidelines.^one–iii Its utility in primary prevention is under investigation.^four–6 In patients with a recent myocardial infarction, ten% to 20% have been reported to discontinue aspirin use during the first 1^7–9 to three¹⁰ years after the infarction. In broader patient settings, discontinuation rates of upward to 30% have been reported, and poor aspirin compliance has been noted in up to 50%.¹¹ The public wellness effects of discontinuing long-term aspirin treatment may be substantial just are not well known.

Discontinuation of secondary prevention with aspirin has been associated with higher risk of cardiovascular events in some studies,^12,13 with indications of an increased risk shortly subsequently discontinuation.^14–xviii Aspirin is often withdrawn because of surgery^{fourteen–17,19} or bleeding,^14,18 factors that per se may stimulate platelet assemblage and increase the take a chance of cardiovascular events. Effects of discontinuation in settings other than surgery or bleeding are unknown.

Aspirin can be bought over the counter in many countries, and previous studies have typically relied on cocky-reported utilise data. Registers with complete coverage in a land where low-dose aspirin is available only by prescription accept a unique potential to shed light on the issue.

We hypothesized that patients with aspirin treatment gaps and those who discontinue long-term aspirin treatment in the absence of major surgery or bleeding are at college run a risk of cardiovascular events than adherent patients without treatment gaps. We aimed to investigate the associations of aspirin treatment persistence patterns and aspirin discontinuation with risk of cardiovascular events using a big nationwide cohort of patients on long-term low-dose aspirin therapy for master and secondary prevention.

Methods

Report Sample

Using the unique civil registration number allocated to all Swedish citizens, we linked the mandatory nationwide Swedish prescribed drug register with the mandatory inpatient and cause-of-death registers with the help of the Swedish National Lath of Health and Welfare. In Sweden, low-dose aspirin cannot be purchased over the counter without a prescription.

For this report, we considered all individuals >40 years of age who had collected prescribed aspirin during the study period between July 1, 2005, and December 31, 2009, as identified in the prescribed drug register by the detection of an aspirin dispense of 75 to 160 mg. We also included aspirin 75 to 160 mg dispensed as part of pharmacy-prepacked multidose prescriptions used for patients unable to safely self-administer their medication. Aspirin tablet strengths other than 75 or 160 mg are not available for prevention of cardiovascular events in Sweden; neither is carbasalate calcium (B01AC08). Nosotros considered all 889 655 aspirin-treated patients who were free from previous cancer (International Classification of Diseases, 10th Revision codes C00–C99) at baseline. We were interested in treatment breaks and permanent discontinuation after a stable (1-year) period of continuous aspirin treatment. Therefore, nosotros excluded 207 180 patients who had <292 divers daily doses of aspirin dispensed during the get-go year after their beginning aspirin prescription (which equals 1 twelvemonth of aspirin treatment with ≥80% adherence, or medication possession ratio, which is normally viewed as acceptable²⁰). We besides excluded 80 621 patients who had a cardiovascular consequence (because our inquiry question involved long-term use) or died during this first twelvemonth of aspirin handling, too as 327 patients without whatsoever exposure time after the outset year, rendering a terminal study sample of 601 527 long-term low-dose aspirin users who entered the follow-upward stage.

To study treatment patterns in aspirin-naïve patients, we besides studied a sample of 227 135 patients who collected their first low-dose aspirin dispense at to the lowest degree one year later the start of the study period. This subsample was not subjected to the 1-year definition of long-term users and was non used in any statistical modeling.

In a second set of analyses, the timing of cardiovascular events in people who discontinued aspirin was investigated. We divers inclusion criteria on the dispense level and defined a timely dispense as ane that was collected between 0 and 10 days earlier the cease of the previous timely dispense, and it had to exist the final in a chain of 4 consecutive timely dispenses (typically equal to 1 year of treatment, chosen to balance strictness of the inclusion criteria with statistical power). This strict definition, although rendering a pocket-size sample of patients with presumed high adherence, was used to decide the fourth dimension of discontinuation every bit precisely as possible and to avoid periods of accumulated medication during and after which drug utilize status is more uncertain. This sample included the 38 736 patients who had four consecutive timely dispenses.

The construction of the samples is displayed in Figure one.

Figure 1. — **Effigy one.** **Flowchart for construction of the report samples.**

Ideals approval was obtained from the Ethics Review Board in Uppsala, Sweden. No informed consent was required.

Clinical Characteristics

All definitions are described in Table I in the online-but Data Supplement. Age, sex activity, and diabetes mellitus were determined at the inclusion appointment—that is, the first day of the first stable 1-year menstruation of continuous aspirin handling. Use of nonsteroidal anti-inflammatory drugs, oral steroids, and antiplatelet (other than aspirin) or oral anticoagulant drugs was divers every bit the fraction of an aspirin treatment period covered past the corresponding drug and was time updated during follow-up. Concomitant cardiovascular affliction was defined as a prior hospitalization for myocardial infarction (International Classification of Diseases, 10th Revision codes I21 and I22) or stroke (International Classification of Diseases, tenth Revision codes I63 and I64). Patients with concomitant cardiovascular disease were defined every bit users of aspirin for secondary prevention. Patients without concomitant cardiovascular were presumed to use their aspirin for primary prevention. Major bleeding was defined every bit a hospital admission acquired by gastrointestinal bleeding, intracranial bleeding, hemopericardium, postoperative bleeding, bleeding from venous varices, bleeding events in the respiratory organisation, hematuria, and other signs of blood loss. Nosotros determined the occurrence of surgery as all surgical procedures except pocket-sized diagnostic procedures such every bit gastroscopy and coronary angiography.

Exposures

We investigated ii sets of exposures. In the offset set of comparisons, we constructed groups based on time-updated aspirin treatment persistence patterns. In the 2nd set, we constructed groups that were on aspirin versus those that had but discontinued aspirin treatment. For both exposures, time spent in the hospital for diagnoses other than the consequence was added to the time at risk for the particular exposure group at the time of access considering patients are provided all in-hospital medications gratis of accuse and hence do not use their dispensed prescribed medications during hospitalization. We were well equipped to report persistence (staying on a drug long term or discontinuing) but not adherence (taking the prescribed number of pills).

Aspirin Handling Persistence Patterns

To tape time on aspirin and off aspirin as accurately equally possible, accounting for the fact that the exact time of discontinuation is uncertain, we constructed 4 fourth dimension-updated aspirin persistence groups, between which patients could move freely over time during follow-up:

On aspirin: the nominal duration of each aspirin manipulate.
Accumulated period: the time after the nominal duration of each aspirin dispense when the patient has tablets left over from previous dispenses in a consecutively dispensed flow.
Grace period: from the stop of the accumulated period plus 25% of the total elapsing on aspirin in the last consecutively dispensed period plus 20 days. This corresponds to lxxx% adherence with up to a week's gap between dispenses.
Off aspirin: from the end of the grace flow.

Although the hypothesis is answered with the utilize of groups 1 and iv, the other 2 groups are included for transparency. Persistence patterns were described for both aspirin-naïve patients and long-term aspirin users.

Timing of Events Later Aspirin Discontinuation

For this ready of analyses, nosotros sought to decide the risk of cardiovascular events in relation to the time of the aspirin discontinuation among patients who collected their dispenses in a timely fashion (as defined in a higher place). Nosotros compared patients who nerveless a timely fifth dispense with patients who did not after a series of 4 timely dispenses. A patient could contribute several times at risk to these analyses.

Follow-Up and Outcomes

Outcomes were defined with the Swedish inpatient and cause-of-death registers, which include all hospitalizations and deaths, respectively, classified with the International Classification of Diseases (codes in Tabular array I in the online-only Information Supplement). The outcome investigated was a get-go incidence of cardiovascular disease subsequently the showtime of follow-up, divers as a hospitalization for myocardial infarction, stroke, or cardiovascular death. Only the main diagnoses in the discharge letter or death certificates were used for classification of the outcome. The accuracy in the Swedish registers is loftier for the diagnoses examined in this report (positive predictive values, 98%–100% for myocardial infarction and 69%–98% for stroke).²¹

Because there may be a risk of reverse causation (aspirin withdrawn from people well-nigh to die), nosotros also investigated a secondary nonfatal cardiovascular events outcome like to the main upshot but without cardiovascular decease. A similar bias may arise if patients discontinue aspirin because they start using an oral anticoagulant equally a event of alter to a college risk stratum; nosotros therefore also investigated models censoring participants at the time of first of oral anticoagulant treatment.

Follow-upward started afterwards 1 year of aspirin treatment with high adherence, as defined above. Patients were followed upwards until the beginning case of the cardiovascular effect, a new diagnosis of cancer, noncardiovascular death, or the stop of follow-up on December 31, 2009. In the timing of events afterward aspirin discontinuation analyses, patients were besides censored after 100 days considering that is the maximum duration of the fifth manipulate. Patients were followed upwards from the day afterwards dispense (or start of a gap) until a cardiovascular disease event or censoring, whichever came first.

Major bleeding and surgical procedures may affect thrombogenicity, cause aspirin discontinuation, and be related to cardiovascular disease incidence for an unknown duration of time that nosotros assumed to be less than a few months. Therefore, in this study, a major bleeding or a surgical procedure during the study incurred a 3-month refractory catamenia from the time at risk, during which person-fourth dimension was not counted and outcomes were non considered.

Statistical Analysis

Baseline characteristics were presented per patient equally observed numbers (percentages) for categorical variables and as means (SDs) for continuous variables. Absolute risks of events were illustrated with Nelson-Aalen cumulative incidence plots, which were also used to appraise proportionality of hazards.

Cox proportional hazards models were used to investigate associations of the 4 aspirin persistence groups with chance of cardiovascular events. Models for minimizing bias were identified with directed acyclic graphs (Figure I in the online-only Information Supplement).²² All models were adjusted for age, sex activity, previous cardiovascular disease, diabetes mellitus, antiplatelet or oral anticoagulant drugs, nonsteroidal anti-inflammatory drugs, and oral steroids. Interactions in the form of difference from multiplicativity were investigated between the aspirin persistence groups and age, sex, prior cardiovascular disease, diabetes mellitus, nonsteroidal anti-inflammatory drugs, oral steroids, other antiplatelet or oral anticoagulant drugs, aspirin-naïve/not-naïve, previous major bleeding, and pharmacy-prepacked multidose dispenses. Considering of potential interaction signals, the main analyses were also performed in subgroups with and without prior cardiovascular illness.

For the timing of events after aspirin discontinuation analysis to which patients could contribute >1 observation period, a model with shared frailty for participant identity was used. Considering we causeless a priori that the gamble role associated with aspirin discontinuation may initially exist nonmonotonic and because nosotros wanted to compute fourth dimension quantiles, we investigated parametric regression models with exponential, Gompertz, Weibull, log logistic, and log normal parameterizations. The log normal distribution had the highest log likelihood and lowest Akaike information criterion and was used, with results displayed graphically and presented in the accelerated failure time metric. This assay was adjusted for the same covariates as the Cox models, and the same prepare of covariates was time updated at each manipulate.

Because this report used only official registers mandatory for all citizens, we assumed that no information were missing. The data were managed and analyzed at both an independent statistical contract company (Statisticon) and Uppsala University, and all authors had total access to the data. The statistical packages R version 3.0.one and Stata version 14 were used.

Results

Clinical characteristics of the aspirin handling persistence groups are displayed in Table 1. Notably, half of the sample were female, the hateful age was 73 years, 16% had diabetes mellitus, and one-half of the sample were on long-term aspirin treatment without having had a prior hospitalization for cardiovascular affliction. During a median of 3.0 years of follow-up (range, 0.002–three.5 years), corresponding to i 491 369 person-years at risk, 62 690 cardiovascular events occurred (incidence rate, 42.0 per 1000 person-years at gamble; carve up outcomes presented in Table Ii in the online-merely Information Supplement). A full of 73 636 people died during this time; 19 978 person-years were excluded from the analyses because of surgical procedures and major bleeding events as defined above.

**Table 1.** Clinical Characteristics by Aspirin Persistence Patterns
	On Aspirin (north=577 578)	Accumulated (n=521 799)	Grace Flow (n=492 074)	Off Aspirin (n=179 548)
Age, y	72.7 (xi.5)	72.7 (11.5)	72.7 (11.3)	72.vi (11.four)
Women, n (%)	311 844 (52)	308 576 (52)	279 886 (52)	266 479 (52)
Alphabetize yr, 2005/2006/2007/2008, %	lxx/12/9/9	71/12/nine/8	74/12/9/5	75/12/9/5
Diabetes mellitus, due north (%)	95 768 (16)	94 982 (16)	89 974 (17)	83 893 (xvi)
Prior CVD, n (%)	329 941 (55)	327 258 (55)	301 675 (56)	281 388 (55)
Major bleeding, due north (%)	15 644 (3)	15 313 (3)	14 774 (3)	xiv 336 (3)
Other antiplatelets, n (%)	23 111 (4)	23 698 (4)	9582 (2)	9992 (2)
Oral anticoagulants, n (%)	1200 (0)	1127 (0)	1462 (0)	1656 (0)
NSAIDs, northward (%)	35 870 (6)	37 544 (six)	23 184 (4)	22 992 (5)
Oral steroids, n (%)	14 191 (two)	fourteen 659 (2)	9286 (two)	ten 626 (2)

Treatment patterns are displayed in Effigy ii and Figure Ii in the online-simply Information Supplement. Among the long-term low-dose aspirin users (Figure two), persistence gradually merely slowly tapered off; 3 years later on inclusion, 3 of 4 patients collected their aspirin dispenses before the expiration of the previous dispense, and 4 of v had access to accumulated aspirin. Approximately xv% were off long-term aspirin handling after three years. Among the 227 135 aspirin-naïve patients, ≈20% did not collect a 2nd aspirin prescription (Figure II in the online-only Data Supplement). After the first twelvemonth of inconsistent prescription collections, those who remained on treatment had approximately the same persistence pattern as those treated with long-term aspirin.

Figure 2. — **Effigy 2.** **Persistence patterns in patients on stable long-term aspirin treatment (n=601527).** *On aspirin* is the nominal duration of each aspirin dispense. *Accumulated* is the time after the nominal duration of each aspirin dispense when the patient has tablets left over from previous dispenses in a consecutively dispensed period. *Grace menses* is from the stop of the accumulated menses plus 25% of the total elapsing on aspirin in the last consecutively dispensed flow plus 20 days. This corresponds to 80% adherence with upwards to a calendar week'due south gap between dispenses. *Off aspirin* is from the end of the grace period.

Adapted cumulative incidence of cardiovascular events according to aspirin treatment persistence groups is presented in Figure 3. Patients on persistent aspirin treatment had the lowest incidence of cardiovascular events. Patients who had discontinued aspirin had a 37% higher charge per unit of cardiovascular events (Table 2), corresponding to an absolute risk increase of 13.five events per 1000 person-years at risk. Put another style, on average, 1 of every 74 patients who discontinued aspirin had an additional cardiovascular consequence in 1 year.

**Table 2.** Risk of Cardiovascular Events by Aspirin Persistence Patterns
	Cardiovascular Events, n	Time at Hazard, y	Take a chance Ratio	95% Confidence Interval
Total sample
On aspirin	49 521	1 209 905	1
Accumulated	3957	94 643	ane.xvi	1.12–i.xx
Grace period	3667	84 932	1.xiv	1.x–1.18
Off aspirin	5545	101 889	i.37	1.34–ane.41
Master prevention
On aspirin	xiv 730	564 408	i
Accumulated	1202	44 966	1.13	1.07–one.20
Grace menstruation	1170	41 920	1.13	one.07–1.20
Off aspirin	1885	57 194	one.28	ane.22–1.34
Secondary prevention
On aspirin	34 791	645 498	1
Accumulated	2755	49 677	1.17	1.12–one.21
Grace period	2497	43 012	ane.15	1.10–1.20
Off aspirin	3660	44 695	1.46	1.41–i.51

Subgroup analyses further revealed that patients with higher age and prior cardiovascular disease were at higher risk increase for cardiovascular events when off aspirin, whereas treatment with oral anticoagulant or other antiplatelet drugs was associated with lower take chances increase for cardiovascular events when off aspirin (Figure 4). The majority (54%) of the study sample used aspirin for secondary prevention. Amongst those, discontinuing aspirin was associated with a 46% college rate of cardiovascular events than continuing on aspirin (Table 2), respective to an absolute risk increase of 28.0 per thousand person-years at risk or an additional cardiovascular result per yr in ane of every 36 patients who discontinued aspirin. Among the 46% who probably used aspirin as part of primary prevention, discontinuing aspirin was associated with a 28% higher charge per unit of cardiovascular events than standing on aspirin (Tabular array 2), an absolute risk increase of 6.9 per one thousand person-years at risk or an boosted cardiovascular event per yr in i of every 146 patients who discontinued aspirin.

Figure 4. — **Effigy 4.** **Risk of cardiovascular events by aspirin persistence patterns in subgroups.** P values are for multiplicative interactions of the subgroup gene with aspirin persistence pattern. CI indicates confidence interval; CVD, cardiovascular disease; and NSAID, nonsteroidal anti-inflammatory drug.

Aspirin discontinuation was also associated with nonfatal cardiovascular events, with a 10% higher risk of nonfatal cardiovascular events among people off versus on aspirin (Table Iii in the online-just Information Supplement). Censoring participants at the start of oral anticoagulant therapy produced results very similar to those of the main models, with a 43% college risk of cardiovascular events if discontinuing rather than continuing on aspirin (Table Four in the online-only Data Supplement).

The timing of events after aspirin discontinuation analyses included 38 736 patients with twoscore 355 times at risk during which 216 cardiovascular events occurred. Patients who stopped taking aspirin after a period of 4 timely dispenses had an early risk increase for cardiovascular events compared with those who nerveless their 5th timely dispense (Figure 5). The median time to the first cardiovascular issue in those who did not collect their fifth manipulate on schedule was one-third the time of those who collected their dispense on schedule (time ratio, 0.31; 95% confidence interval, 0.22–0.43).

Figure 5. — **Effigy 5.** **Timing of cardiovascular events after aspirin discontinuation in patients who stopped (red line) vs continued (green line) taking aspirin after 4 timely aspirin dispenses.** Parametric regression models with log normal parameterization and shared frailty on the patient level. Models adapted for the baseline variables of age, sex, and diabetes mellitus and the time-updated variables of previous cardiovascular affliction, antiplatelet or oral anticoagulant drugs, nonsteroidal anti-inflammatory drugs, and oral steroids.

Give-and-take

In this large nationwide patient accomplice, discontinuation of long-term low-dose aspirin was associated with a >thirty% higher risk of cardiovascular events, corresponding to an additional cardiovascular event observed per yr in i of every 74 patients who discontinue aspirin. The run a risk appeared to increment every bit soon as the patients discontinued aspirin, with no condom interval. In this sample, only one-half of the low-dose aspirin users had been hospitalized for cardiovascular disease before baseline; most of the remaining one-half were presumably treated, for example, for angina pectoris or stroke prevention in atrial fibrillation or treated as part of master prevention. Aspirin discontinuation appeared especially perilous among patients with previous cardiovascular disease, with an additional cardiovascular issue per twelvemonth in i of every 36 secondary prevention patients who discontinued aspirin compared with an additional cardiovascular effect per twelvemonth in 1 of every 146 primary prevention patients who discontinued aspirin.

We confirm a high initial discontinuation rate, every bit seen in prior studies^7–11: 1 of 5 aspirin-naïve patients did non collect the second aspirin dispense, and the main persistence drop was during the commencement year after aspirin initiation. In contrast, those who picked up their second prescription had a modest discontinuation rate over fourth dimension.

Our observations of the risks associated with low-dose aspirin discontinuation are of a magnitude very similar to those of previous randomized trials of aspirin initiation.^ane,v However, the timing of cardiovascular events after aspirin discontinuation remains uncertain. Although those analyses were based on a pocket-sized number of events, the adventure in this study appeared to increase soon after discontinuation. An acutely increased risk of cerebrovascular events,^14,15 ischemic events,^15–17 and bloodshed¹⁶ has been observed in case-control studies,^14,sixteen in perioperative studies,^15,17 and in individuals with bleeding ulcers.¹⁸ In those studies, the chief reasons for aspirin withdrawal were surgery^xiv–17 or haemorrhage,^xiv,xviii which per se may stimulate platelet assemblage. No increase in cardiovascular events was observed in a recent perioperative aspirin discontinuation trial,¹⁹ although aspirin was stopped inside 24 hours before the surgery in that study. Notably, the present report investigated aspirin discontinuation that was unrelated to surgery or bleeding events.

Experimental studies have suggested a rebound effect afterward aspirin discontinuation, involving increased thromboxane levels^23,24 possibly resulting from the prothrombotic effects of remainder very low levels of aspirin.²⁵ The clinical importance of a rebound effect may be substantial because of the large number of aspirin patients and the high discontinuation rates. For patients undergoing planned surgery or other procedures, it is unknown whether treatment gaps >seven days or <24 hours earlier the procedure are condom.¹⁹ For patients discontinuing aspirin therapy, it is unknown whether or when the rebound effect happens. In addition, for patients with poor adherence, any rebound effects may be in play more or less continuously. The possibility of such mechanisms is supported past the ascertainment in this study that aspirin discontinuation was not associated with cardiovascular events in patients protected by other antiplatelet or oral anticoagulant drugs (Figure 4), although those patients were probable at college absolute risk of such events.

Some limitations of this report are worth mentioning. Most important, there is a risk of confounding, as in all observational studies. Nosotros did not have access to data on socioeconomic status; physical examinations, including blood pressures and lipids; or lifestyle measures such as smoking. However, nosotros used directed acyclic graphs to place bias-minimized models and included merely people who had qualified as long-term users of aspirin. Furthermore, any confounding past indication would bias toward a null result, assuming that people with the highest risk of cardiovascular events would be the ones least probable to discontinue aspirin treatment. At that place is also a risk of opposite causation—that is, patients about to dice stop taking aspirin so die anyway. Associations of aspirin discontinuation with the secondary nonfatal cardiovascular events outcome were like but weaker, which may signify some reverse causation, a protective effect of aspirin confronting fatal events, or lower statistical power in those analyses. Models censoring at the time of start of oral anticoagulant treatment produced results similar to the main results, indicating low take chances of reverse causation by patients discontinuing aspirin treatment because of moving to a higher run a risk stratum and switching to oral anticoagulant therapy. An of import limitation is the imprecision in determining the exposure condition, which also would bias the results toward the null hypothesis. This is confirmed past the results amidst people with prepacked dispenses, for which nosotros have good precision in the exposure. These patients appear to exist at clearly higher take a chance of aspirin discontinuation than those who collect dispenses themselves. Thus, the true issue is likely higher than the primary observed result. Another limitation is the end of follow-upwardly in 2009, which implies lack of data on more than recent handling patterns but, on the other hand, avoids cohort effect bias because aspirin guidelines were stable during the study period.

Strengths include the large contemporary sample rendering >60 000 cardiovascular events, universal coverage of the prescription annals and hence inclusion of all long-term low-dose aspirin users nationwide, the universal coverage of the high-precision²¹ registers for determining the outcomes, and minimal loss to follow-up.

Conclusions

Amid long-term users of low-dose aspirin, discontinuation of aspirin in the absenteeism of major surgery or bleeding was associated with a >30% increased risk of cardiovascular events. The risk increased shortly after discontinuation. These findings can help policymakers focus on simple measures to ensure treatment persistence with a cheap medication like aspirin with substantial public wellness gains.

Acknowledgments

Editorial assistance was provided by Dr Ebba Bergman and Dr Vendela Roos, Uppsala Clinical Enquiry Centre, Sweden.

Sources of Funding

This study was partly funded by AstraZeneca by funds to Statisticon for the statistical analyses.

Disclosures

Drs Sundström, Hedberg, and Oldgren have served on an advisory board for AstraZeneca. P. Aarskog is a electric current employee and K.M. Johannesen is a former employee of AstraZeneca. Dr Thuresson is employed by Statisticon, of which AstraZeneca is a client. Dr Sundström is on an advisory board for Itrim. Dr Oldgren has received consultancy and lecture fees from Bayer, Boehringer-Ingelheim, BristolMyers Squibb and Pfizer, outside the submitted work.

Footnotes

References

1. Antithrombotic Trialists' Collaboration. Collaborative meta-assay of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high adventure patients. BMJ . 2002; 324:71–86.CrossrefMedlineGoogle Scholar
2. Smith SC, Allen J, Blair SN, Bonow RO, Contumely LM, Fonarow GC, Grundy SM, Hiratzka L, Jones D, Krumholz HM, Mosca L, Pasternak RC, Pearson T, Pfeffer MA, Taubert KA; AHA/ACC; National Eye, Lung, and Blood Constitute. AHA/ACC guidelines for secondary prevention for patients with coronary and other atherosclerotic vascular affliction: 2006 update: endorsed by the National Center, Lung, and Blood Institute. Circulation . 2006; 113:2363–2372. doi: 10.1161/CIRCULATIONAHA.106.174516.LinkGoogle Scholar
three. Graham I, Atar D, Borch-Johnsen One thousand, Boysen 1000, Burell Thou, Cifkova R, Dallongeville J, De Backer G, Ebrahim S, Gjelsvik B, Herrmann-Lingen C, Hoes A, Humphries S, Knapton M, Perk J, Priori SG, Pyorala K, Reiner Z, Ruilope L, Sans-Menendez Due south, Scholte op Reimer W, Weissberg P, Wood D, Yarnell J, Zamorano JL, Walma Eastward, Fitzgerald T, Cooney MT, Dudina A; European Order of Cardiology (ESC) Committee for Practice Guidelines (CPG). European guidelines on cardiovascular illness prevention in clinical practice: executive summary: Fourth Articulation Task Forcefulness of the European Society of Cardiology and Other Societies on Cardiovascular Disease Prevention in Clinical Practice (constituted by representatives of nine societies and by invited experts). Eur Centre J . 2007; 28:2375–2414. doi: 10.1093/eurheartj/ehm316.CrossrefMedlineGoogle Scholar
4. Seshasai SR, Wijesuriya Southward, Sivakumaran R, Nethercott S, Erqou S, Sattar N, Ray KK. Event of aspirin on vascular and nonvascular outcomes: meta-analysis of randomized controlled trials. Arch Intern Med . 2012; 172:209–216. doi: 10.1001/archinternmed.2011.628.CrossrefMedlineGoogle Scholar
5. Antithrombotic Trialists' Collaboration, Baigent C, Blackwell 50, Collins R, Emberson J, Godwin J, Peto R, Buring J, Hennekens C, Kearney P, Meade T, Patrono C, Roncaglioni MC, Zanchetti A. Aspirin in the primary and secondary prevention of vascular illness: collaborative meta-analysis of individual participant data from randomised trials. Lancet . 2009; 373:1849–1860.CrossrefMedlineGoogle Scholar
6. Ikeda Y, Shimada K, Teramoto T, Uchiyama Due south, Yamazaki T, Oikawa Southward, Sugawara M, Ando K, Murata M, Yokoyama K, Ishizuka Northward. Low-dose aspirin for primary prevention of cardiovascular events in Japanese patients 60 years or older with atherosclerotic risk factors: a randomized clinical trial. JAMA . 2014; 312:2510–2520. doi: 10.1001/jama.2014.15690.CrossrefMedlineGoogle Scholar
7. Eagle KA, Kline-Rogers East, Goodman SG, Gurfinkel EP, Avezum A, Flather Doc, Granger CB, Erickson S, White G, Steg PG. Adherence to evidence-based therapies after belch for acute coronary syndromes: an ongoing prospective, observational study. Am J Med . 2004; 117:73–81. doi: 10.1016/j.amjmed.2003.12.041.CrossrefMedlineGoogle Scholar
eight. Sud A, Kline-Rogers EM, Eagle KA, Fang J, Armstrong DF, Rangarajan K, Otten RF, Stafkey-Mailey DR, Taylor SD, Erickson SR. Adherence to medications by patients after astute coronary syndromes. Ann Pharmacother . 2005; 39:1792–1797. doi: 10.1345/aph.1G249.CrossrefMedlineGoogle Scholar
9. Kulkarni SP, Alexander KP, Lytle B, Heiss G, Peterson ED. Long-term adherence with cardiovascular drug regimens. Am Middle J . 2006; 151:185–191.CrossrefMedlineGoogle Scholar
10. Mostaza JM, Lahoz C, Martín-Jadraque R, Sanmartín MA, Vicente I, Tranche S, Taboada Thou, Mantilla T, Monteiro B, Sanchez-Zamorano MA; PRESENAP Study. Factors associated with the discontinuation of show-based cardiovascular therapies in patients with stable coronary artery disease: a primary care perspective. Eur J Cardiovasc Prev Rehabil . 2009; sixteen:34–38. doi: 10.1097/HJR.0b013e32831a47f3.CrossrefMedlineGoogle Scholar
eleven. Herlitz J, Tóth PP, Naesdal J. Low-dose aspirin therapy for cardiovascular prevention: quantification and consequences of poor compliance or discontinuation. Am J Cardiovasc Drugs . 2010; 10:125–141. doi: ten.2165/11318440-000000000-00000.CrossrefMedlineGoogle Scholar
12. Rodríguez LA, Cea-Soriano 50, Martín-Merino E, Johansson Due south. Discontinuation of low dose aspirin and risk of myocardial infarction: case-control study in United kingdom of great britain and northern ireland primary intendance. BMJ . 2011; 343:d4094.CrossrefMedlineGoogle Scholar
13. García Rodríguez LA, Cea Soriano 50, Loma C, Johansson S. Increased risk of stroke later discontinuation of acetylsalicylic acrid: a United kingdom of great britain and northern ireland primary intendance study. Neurology . 2011; 76:740–746. doi: x.1212/WNL.0b013e31820d62b5.CrossrefMedlineGoogle Scholar
fourteen. Maulaz AB, Bezerra DC, Michel P, Bogousslavsky J. Effect of discontinuing aspirin therapy on the run a risk of brain ischemic stroke. Arch Neurol . 2005; 62:1217–1220. doi: x.1001/archneur.62.8.1217.CrossrefMedlineGoogle Scholar
15. Burger W, Chemnitius JM, Kneissl GD, Rücker Yard. Low-dose aspirin for secondary cardiovascular prevention: cardiovascular risks afterward its perioperative withdrawal versus bleeding risks with its continuation: review and meta-assay. J Intern Med . 2005; 257:399–414. doi: 10.1111/j.1365-2796.2005.01477.ten.CrossrefMedlineGoogle Scholar
16. Collet JP, Montalescot G, Blanchet B, Tanguy ML, Golmard JL, Choussat R, Beygui F, Payot 50, Vignolles N, Metzger JP, Thomas D. Bear upon of prior use or recent withdrawal of oral antiplatelet agents on astute coronary syndromes. Circulation . 2004; 110:2361–2367. doi: x.1161/01.CIR.0000145171.89690.B4.LinkGoogle Scholar
17. Oscarsson A, Gupta A, Fredrikson Thou, Järhult J, Nyström Thousand, Pettersson E, Darvish B, Krook H, Swahn E, Eintrei C. To go on or discontinue aspirin in the perioperative period: a randomized, controlled clinical trial. Br J Anaesth . 2010; 104:305–312. doi: x.1093/bja/aeq003.CrossrefMedlineGoogle Scholar
18. Sung JJ, Lau JY, Ching JY, Wu JC, Lee YT, Chiu Pow, Leung VK, Wong VW, Chan FK. Continuation of low-dose aspirin therapy in peptic ulcer haemorrhage: a randomized trial. Ann Intern Med . 2010; 152:1–9. doi: ten.7326/0003-4819-152-1-201001050-00179.CrossrefMedlineGoogle Scholar
19. Devereaux PJ, Mrkobrada M, Sessler DI, Leslie K, Alonso-Coello P, Kurz A, Villar JC, Sigamani A, Biccard BM, Meyhoff CS, Parlow JL, Guyatt K, Robinson A, Garg AX, Rodseth RN, Botto F, Lurati Buse Yard, Xavier D, Chan MT, Tiboni Yard, Cook D, Kumar PA, Forget P, Malaga G, Fleischmann Eastward, Amir Grand, Eikelboom J, Mizera R, Torres D, Wang CY, VanHelder T, Paniagua P, Berwanger O, Srinathan Southward, Graham M, Pasin L, Le Manach Y, Gao P, Pogue J, Whitlock R, Lamy A, Kearon C, Baigent C, Chow C, Pettit S, Chrolavicius Due south, Yusuf S; POISE-ii Investigators. Aspirin in patients undergoing noncardiac surgery. North Engl J Med . 2014; 370:1494–1503. doi: x.1056/NEJMoa1401105.CrossrefMedlineGoogle Scholar
20. Osterberg 50, Blaschke T. Adherence to medication. North Engl J Med . 2005; 353:487–497. doi: ten.1056/NEJMra050100.CrossrefMedlineGoogle Scholar
21. Ludvigsson JF, Andersson East, Ekbom A, Feychting M, Kim JL, Reuterwall C, Heurgren M, Olausson PO. External review and validation of the Swedish national inpatient annals. BMC Public Health . 2011; 11:450. doi: ten.1186/1471-2458-11-450.CrossrefMedlineGoogle Scholar
22. Shrier I, Platt RW. Reducing bias through directed acyclic graphs. BMC Med Res Methodol . 2008; 8:70. doi: 10.1186/1471-2288-eight-seventy.CrossrefMedlineGoogle Scholar
23. McDonald JW, Ali Chiliad. Recovery of cyclooxygenase activeness after aspirin in populations of platelets separated on stractan density gradients. Prostaglandins Leukot Med . 1983; 12:245–252.CrossrefMedlineGoogle Scholar
24. Vial JH, McLeod LJ, Roberts MS. Rebound summit in urinary thromboxane B2 and 6-keto-PGF1 alpha excretion subsequently aspirin withdrawal. Adv Prostaglandin Thromboxane Leukot Res . 1991; 21A:157–160.MedlineGoogle Scholar
25. Doutremepuich C, Aguejouf O, Desplat V, Eizayaga FX. Paradoxical thrombotic effects of aspirin: experimental study on chiliad animals. Cardiovasc Hematol Disord Drug Targets . 2010; x:103–110.CrossrefMedlineGoogle Scholar

Been Taking Baby Aspirin. For 10 Days Is It Safe to Stop Taking Them

Source: https://www.ahajournals.org/doi/10.1161/circulationaha.117.028321