aspirin in healthy adults: harm without benefit??
there was a trio of articles in NEJM on the utility of aspirin in healthy elderly patients (first time i've seen a trio in NEJM...), finding not much benefit and some significant harm, thereby questioning the role of aspirin in primary prevention in the healthy elderly. this blog is a bit longer than usual, since it reviews 3 articles, but the one-liner conclusions in the press (including the medical press) that we should reconsider using aspirin in healthy older people should be a tad more nuanced.
1. Aspirin effects on cardiovascular events and bleeding, the ASPREE trial (see aspirin healthy not dec cardiovasc NEJM2018 in dropbox. or DOI: 10.1056/NEJMoa1805819)
Details:
-- 19,114 community-dwelling men and women in Australia and the United States who were 70 years old or older (65 or older for black and Hispanic people in the United States). All were initially free of overt coronary heart disease, overt cerebrovascular disease, atrial fibrillation, a clinical diagnosis dementia, clinically significant physical disability, high risk of bleeding, anemia, blood pressure >180/105, and those with a physician-estimated life expectancy <5yrs. Study was from 2010-2014.
-- 44% male, median age 74, 5% black, 30% with a BMI>30, 4% current smoker/41% former/55% never smoker, 11% diabetes, 74% hypertension, 65% dyslipidemia, 26% CKD, 31% 0-1 cardiovascular risk factors/42% with 2 risk factors/28% with 3-4 risk factors, 34% with statin use that entry, 14% NSAIDs, 25% PPIs
-- randomly assigned to enteric-coated aspirin 100 mg daily vs placebo, followed 4.7 years (trial terminated early with assessment that there would be no benefit from continued aspirin)
-- in the final 12 months of the trial, 62% on aspirin and 64% on placebo were still taking their assigned treatments
-- primary endpoint: composite of death, dementia, persistent physical disability
-- secondary endpoints were major hemorrhage (composite of hemorrhagic stroke, symptomatic intracranial bleeding, or clinically significant extracranial bleeding) and cardiovascular disease (fatal coronary heart disease, nonfatal MI, fatal or nonfatal stroke, hospitalization for heart failure)
Results:
-- rate of cardiovascular disease was 10.7 events per 1000 person-yrs on aspirin, 11.3 events per 1000 person-yrs on placebo, nonsignificant with HR 0.95 (0.83-1.08). Individual rates for MI, ischemic stroke, fatal cardiovascular disease, and hospitalization for heart failure were similar in the 2 groups
-- a review of the cumulative incidence of major cardiovascular events showed that after 2-3 years the curves for aspirin vs placebo were parallel, suggesting that follow-up beyond 6 years was unlikely to show more of a difference
-- no difference by prespecified subgroups on the risk of cardiovascular disease, including by sex, age, country of residence, race or ethnic group, smoking, BMI, prior use of aspirin, or the presence of diabetes/hypertension/dyslipidemia at baseline.
-- Also no difference by post hoc analyses of the subgroups, including: the number of cardiovascular risk factors that were present, CKD, use of statins or NSAIDs
-- rate of major hemorrhage was 8.6 events per 1000 person-yrs vs 6.2 events per 1000 person-yrs on placebo, a significant 38% increase with HR 38 (1.18-1.62), p<0.001.
-- Major hemorrhagic events were primarily in the upper G.I. tract and intracranial bleeding.
-- there was a continuing increase in the cumulative incidence of major hemorrhagic events across the trial follow-up period (vs no further protection for cardiovascular events after 2-3 years)
-- no difference by the non-prespecified subgroups (eg by use of PPI at randomization), other than the possibility of a less harmful effect with increasing age. However the incidence of bleeding overall was higher than in younger patients (the rate and those 65-73 was 74% increased risk, with HR 1.74, vs 23% for those> 73-year-old, HR 1.23, p=0.05 for the interaction)
Commentary:
-- it is important to remember that this trial involved patients without known cardiovascular disease (primary prevention). It does not apply to patients who have known cardiovascular disease (secondary prevention) where the overall benefit of aspirin therapy has been established.
-- In this context, it was interesting that there was no cardiovascular benefit apparent, given that the age of these patients as well as the fact that the majority had at least 2 cardiovascular risk factors suggested that there was likely underlying but not clinically manifest cardiovascular disease.
-- it was perhaps notable that the observed rate of cardiovascular events was about half of what was anticipated (around 11 events, vs the anticipated 22.4 events per thousand person-yrs), though they also found a somewhat decreasing incidence of hemorrhagic events in the much older people (also an unexpected finding)
-- one big concern with the trial was the low rate of aspirin continuation by the end of the study (fewer than 2/3 of the patients). The authors feel this should not have affected the results statistically, but there may in fact be significant differences between those patients who decided to stop taking the aspirin and those who did not. And this could really have altered the results of the study!
--one other concern is though this study was done in 2 different countries with lots of patients, the demographics noted not lots of racial diversity (86% white/5% black/3% Hispanic; and this diversity was confined to the relatively small US cohort: 1203 vs 8322 patients from Australia (53% of the US participants were black/Hispanic)
--also lacking: they did not have any information on the LDL cholesterol levels of the different groups, though they did look at "statin use" and found no difference between groups. But given the profound effect of the achieved LDL level, that would be a useful item to know (ie, that could be a really important bias in interpreting their results: see http://gmodestmedblogs.blogspot.com/2018/08/very-low-ldl-levels-benefit-without-harm.html
--the USPSTF gives a "B" recommendation to initiating low-dose aspirin for primary prevention of cardiovasc disease and colorectal center in adults 50-59 with a >=10% CVD risk, not at increased risk for bleeding, life expectancy at least 10 yrs, and willing to tak aspirin for at least 10 yrs. And a "C" recommendation for those 60-69yo. this current set of studies is helpful in looking at those more elderly (median age was 74)
-- this study was terminated after only 4.7 years. Reviewing the graphs of events over time, it appeared that the potential benefit of aspirin was not increasing over time, but the incidence of major bleeding continued to increase, justifying the early termination of the trial. But the potential benefit of aspirin on colon cancer typically takes long-term aspirin use (on the order of decades). so, this analysis cannot include this potential benefit in its risk/benefit calculation.
2. Aspirin effects on disability-free survival (see aspirin disabil-free survival nejm2018 in dropbox, or DOI: 10.1056/NEJMoa1800722)
Details:
--as above
--additional demographics: 59% not frail/39% prefrail/2%frail [prefrail means having met only 1 or 2 of the Fried frailty criteria: body weight, handgrip strength, self-reported exhaustion, walking speed, and self-reported physical activity]
--primary endpoint: composite of death, dementia, or persistent physical disability
Results:
--rate of composite of death, dementia, or persistent physical disability was 21.5 events per 1000 person-yrs in the aspirin group and 21.2 per 1000 person-yrs in the placebo group, HR 1.01 (0.92 - 1.11); p=0.79.
--50% of the combined endpoints were for deaths, 30% for dementia, and 20% for persistent physical disability
-- Differences between the aspirin group and the placebo group were not substantial with regard to the secondary individual end points;
-- death from any cause (12.7 events per 1000 person-yrs in the aspirin group and 11.1 events per 1000 person-yrs in the placebo group), HR 1.14 (1.01-1.29) favoring placebo
-- dementia 6.7/1000 person-yrs aspirin vs 6.9/1000 person-yrs placebo, non-significant
-- persistent physical disability 4.9/1000 person-yrs aspirin vs 5.8 person-yrs placebo, non-significant
--no significant differences in prespecified subgroup analyses, except marginally significant benefit in primary endpoint in those who were prefrail [HF 0.89 (0.79-1.00)] and marginally significant placebo benefit in those who were not frail [HR 1.17 (1.01-1.36)], and no difference in non-prespecified endpoints (eg statin use, CKD, quality of life questionnaires)
--rate of major hemorrhage, as above
3. Aspirin effects on all-cause mortality and its association with increased cancer-related mortality(see aspirin all-cause mortality healthy nejm2018 in dropbox, or DOI: 10.1056/NEJMoa1803955 )
Details:
--as per above
Results:
--1052 deaths occurred over the 4.7 years of follow-up
--risk of death from any cause: 12.7 events per 1000 person-yrs on aspirin, vs 11.1/1000 person-yrs with placebo, a 14% increase with aspirin: HR 1.14 (1.01-1.29)
--cancer-related deaths was the major contributor to the higher aspirin-related mortality, occuring in 3.1% on aspirin vs 2.3% placebo: HR 1.31 (1.10-1.56)
--no significant difference for cardiovasc causes (including ischemic stroke), major hemorrhage (including hemorrhagic stroke), or "other" (eg, sepsis, chronic lung disease, dementia, heart failure)
--of the cancer-related deaths (50% of all deaths, vs 19% for cardiovasc causes and 5% for major hemorrhage):
--there was a broad increase in all of the cancers they assessed
--the only one reaching statistical significance was colorectal cancers, with 35 deaths, a 77% increase in group on aspirin (HR 1.77 (1.02-3.06); this was the second most common cancer-related death, after lung cancer in 55
--there was a progressive divergence in both all-causes of death and cancer-related deaths after about 3-4 years
--subgroup analysis: trend to doing better with placebo in both those with and without a personal history of cancer, though a trend to doing better with aspirin in black and Hispanic patients (ie, in the US but not in the Australian cohorts)
Commentary:
--really not sure how to put interpret the cancer death increases with aspirin
--many studies have shown lower colorectal cancers with prolonged use of aspirin (this study was <5 yrs, so not surprising that they did not find benefit, BUT finding a statistically higher association???). eg see http://gmodestmedblogs.blogspot.com/2018/01/aspirin-prevents-colon-cancer.html
--an impressive meta-analysis of 17K people (see Rothwell PM. Lancet 2012; 379: 1591, or aspirin cancer mets lancet 2012 in dropbox) found that aspirin reduced the risk of:
--cancer with distant metastases: 36% decrease in all cancers, 46% decrease in adenocarcinoma, both highly statistically significant, and found after only 6.5 years (though curves showed benefit after only 2-4 years)
--31% decreased risk of adenocarcinoma with metastases at initial diagnosis
--55% decreased risk on subsequent followup
--50% reduction in death from cancer in those who developed adenocarcinoma, particularly in those with initial diagnosis but without mets
--and there were significant numbers of cancers: 1101 incident cancers and 563 deaths due to incident cancers
--results were independent of age > vs <60yo
--and another by Rothwell PM. Lancet 2011; 377:41, of individual patient data from randomized trials, finding:
--25570 patients having 674 cancer deaths
--35% decrease in all cancers, 54% decrease in GI cancers, and benefit found after 5 years' follow-up
--20% decrease in all solid cancers, 35% in GI cancers, over 20 years' follow-up, also significant decrease in adenocarcinomas of the lung and esophagus
--so, how can one explain the differences:
--?age (the current study patients were older, BUT the Rothwell metanalysis did find age at randomization did not affect the overall results, and aspirin's benefit on death from non-GI cancers was the most profound in the group >65yo, though was also decreased in those <55 and 55-64yo.
-- was it the short followup in the ASPREE trial that led to these divergent results? (nothing was apparent in the above Lancet assessments until about 5 years out).
-- was it the pretty high dropout rate in the ASPREE trial?? (though they comment that their dropout rate was pretty similar to other trials, but for some reason was there a different bias in ASPREE than the other trials??)
-- perhaps the differences in the Australian group (quite large, with more evidence of adverse cancer risk associated with aspirin) vs the smaller US group (trend to benefit with aspirin) mean something. is there a different exposure to environmental carcinogens in the US?? (which might mean that an Australian study has limited generalizability to the US population?). does it have to do with the large differences in demographic diversity?? (was the more diverse US population different in other ways? perhaps other types of potential carcinogenic exposures such as to different stressors or other lifestyle issues: stress-mediated cortisol levels might increase risk of cancer, diet/exercise can affect cancer risk.....)
-----------------------------------------------------
so, how does one put this all together (and what is to be done with these results)?? i really am not sure how to process all of this divergent information. it seems to make sense not to routinely include colorectal cancer prevention in the decision analysis to suggest aspirin in those more than 75 years old (as in the ASPREE group), since most studies do suggest benefit only after 1-2 decades of taking aspirin (and the bleeding risks are clear). but, the increased risk of cancer in the last trial seems so different from prior results that i am hesitant to include it in my decision analysis on aspirin therapy (as per above). the utility of aspirin is still clear for secondary prevention, though even there it would be really interesting to know if there were added benefit to aspirin over more aggressive lipid control with statins, which also have anti-inflammatory effects and improve endothelial function (ie, would be great to have an RCT with aspirin vs statin vs both vs neither, with a low statin-induced achieved LDL level). i personally still think that aspirin has a role in younger people who have high cardiovascular risk but no evident heart disease, given the additional colon cancer benefit. but these new studies have at least imprinted questions to aspirin's role in light of the evident harms, pending further data....
relevant recent blogs on aspirin:
http://gmodestmedblogs.blogspot.com/2018/09/aspirin-in-low-risk-diabetics.html which reviews recent study on low-risk diabetic patients,finding a significant decrease in serious vascular events with aspirin, though this was largely outweighed by an increase in major bleeding events
http://gmodestmedblogs.blogspot.com/2018/07/aspirin-one-dose-does-not-fit-all.html which suggests that weight-based aspirin dosing was appropriate
http://gmodestmedblogs.blogspot.com/2018/07/aspirin-one-dose-does-not-fit-all.html which found an increase in cardiovascular events in those who stopped their aspirin
1. Aspirin effects on cardiovascular events and bleeding, the ASPREE trial (see aspirin healthy not dec cardiovasc NEJM2018 in dropbox. or DOI: 10.1056/NEJMoa1805819)
Details:
-- 19,114 community-dwelling men and women in Australia and the United States who were 70 years old or older (65 or older for black and Hispanic people in the United States). All were initially free of overt coronary heart disease, overt cerebrovascular disease, atrial fibrillation, a clinical diagnosis dementia, clinically significant physical disability, high risk of bleeding, anemia, blood pressure >180/105, and those with a physician-estimated life expectancy <5yrs. Study was from 2010-2014.
-- 44% male, median age 74, 5% black, 30% with a BMI>30, 4% current smoker/41% former/55% never smoker, 11% diabetes, 74% hypertension, 65% dyslipidemia, 26% CKD, 31% 0-1 cardiovascular risk factors/42% with 2 risk factors/28% with 3-4 risk factors, 34% with statin use that entry, 14% NSAIDs, 25% PPIs
-- randomly assigned to enteric-coated aspirin 100 mg daily vs placebo, followed 4.7 years (trial terminated early with assessment that there would be no benefit from continued aspirin)
-- in the final 12 months of the trial, 62% on aspirin and 64% on placebo were still taking their assigned treatments
-- primary endpoint: composite of death, dementia, persistent physical disability
-- secondary endpoints were major hemorrhage (composite of hemorrhagic stroke, symptomatic intracranial bleeding, or clinically significant extracranial bleeding) and cardiovascular disease (fatal coronary heart disease, nonfatal MI, fatal or nonfatal stroke, hospitalization for heart failure)
Results:
-- rate of cardiovascular disease was 10.7 events per 1000 person-yrs on aspirin, 11.3 events per 1000 person-yrs on placebo, nonsignificant with HR 0.95 (0.83-1.08). Individual rates for MI, ischemic stroke, fatal cardiovascular disease, and hospitalization for heart failure were similar in the 2 groups
-- a review of the cumulative incidence of major cardiovascular events showed that after 2-3 years the curves for aspirin vs placebo were parallel, suggesting that follow-up beyond 6 years was unlikely to show more of a difference
-- no difference by prespecified subgroups on the risk of cardiovascular disease, including by sex, age, country of residence, race or ethnic group, smoking, BMI, prior use of aspirin, or the presence of diabetes/hypertension/dyslipidemia at baseline.
-- Also no difference by post hoc analyses of the subgroups, including: the number of cardiovascular risk factors that were present, CKD, use of statins or NSAIDs
-- rate of major hemorrhage was 8.6 events per 1000 person-yrs vs 6.2 events per 1000 person-yrs on placebo, a significant 38% increase with HR 38 (1.18-1.62), p<0.001.
-- Major hemorrhagic events were primarily in the upper G.I. tract and intracranial bleeding.
-- there was a continuing increase in the cumulative incidence of major hemorrhagic events across the trial follow-up period (vs no further protection for cardiovascular events after 2-3 years)
-- no difference by the non-prespecified subgroups (eg by use of PPI at randomization), other than the possibility of a less harmful effect with increasing age. However the incidence of bleeding overall was higher than in younger patients (the rate and those 65-73 was 74% increased risk, with HR 1.74, vs 23% for those> 73-year-old, HR 1.23, p=0.05 for the interaction)
Commentary:
-- it is important to remember that this trial involved patients without known cardiovascular disease (primary prevention). It does not apply to patients who have known cardiovascular disease (secondary prevention) where the overall benefit of aspirin therapy has been established.
-- In this context, it was interesting that there was no cardiovascular benefit apparent, given that the age of these patients as well as the fact that the majority had at least 2 cardiovascular risk factors suggested that there was likely underlying but not clinically manifest cardiovascular disease.
-- it was perhaps notable that the observed rate of cardiovascular events was about half of what was anticipated (around 11 events, vs the anticipated 22.4 events per thousand person-yrs), though they also found a somewhat decreasing incidence of hemorrhagic events in the much older people (also an unexpected finding)
-- one big concern with the trial was the low rate of aspirin continuation by the end of the study (fewer than 2/3 of the patients). The authors feel this should not have affected the results statistically, but there may in fact be significant differences between those patients who decided to stop taking the aspirin and those who did not. And this could really have altered the results of the study!
--one other concern is though this study was done in 2 different countries with lots of patients, the demographics noted not lots of racial diversity (86% white/5% black/3% Hispanic; and this diversity was confined to the relatively small US cohort: 1203 vs 8322 patients from Australia (53% of the US participants were black/Hispanic)
--also lacking: they did not have any information on the LDL cholesterol levels of the different groups, though they did look at "statin use" and found no difference between groups. But given the profound effect of the achieved LDL level, that would be a useful item to know (ie, that could be a really important bias in interpreting their results: see http://gmodestmedblogs.blogspot.com/2018/08/very-low-ldl-levels-benefit-without-harm.html
--the USPSTF gives a "B" recommendation to initiating low-dose aspirin for primary prevention of cardiovasc disease and colorectal center in adults 50-59 with a >=10% CVD risk, not at increased risk for bleeding, life expectancy at least 10 yrs, and willing to tak aspirin for at least 10 yrs. And a "C" recommendation for those 60-69yo. this current set of studies is helpful in looking at those more elderly (median age was 74)
-- this study was terminated after only 4.7 years. Reviewing the graphs of events over time, it appeared that the potential benefit of aspirin was not increasing over time, but the incidence of major bleeding continued to increase, justifying the early termination of the trial. But the potential benefit of aspirin on colon cancer typically takes long-term aspirin use (on the order of decades). so, this analysis cannot include this potential benefit in its risk/benefit calculation.
2. Aspirin effects on disability-free survival (see aspirin disabil-free survival nejm2018 in dropbox, or DOI: 10.1056/NEJMoa1800722)
Details:
--as above
--additional demographics: 59% not frail/39% prefrail/2%frail [prefrail means having met only 1 or 2 of the Fried frailty criteria: body weight, handgrip strength, self-reported exhaustion, walking speed, and self-reported physical activity]
--primary endpoint: composite of death, dementia, or persistent physical disability
Results:
--rate of composite of death, dementia, or persistent physical disability was 21.5 events per 1000 person-yrs in the aspirin group and 21.2 per 1000 person-yrs in the placebo group, HR 1.01 (0.92 - 1.11); p=0.79.
--50% of the combined endpoints were for deaths, 30% for dementia, and 20% for persistent physical disability
-- Differences between the aspirin group and the placebo group were not substantial with regard to the secondary individual end points;
-- death from any cause (12.7 events per 1000 person-yrs in the aspirin group and 11.1 events per 1000 person-yrs in the placebo group), HR 1.14 (1.01-1.29) favoring placebo
-- dementia 6.7/1000 person-yrs aspirin vs 6.9/1000 person-yrs placebo, non-significant
-- persistent physical disability 4.9/1000 person-yrs aspirin vs 5.8 person-yrs placebo, non-significant
--no significant differences in prespecified subgroup analyses, except marginally significant benefit in primary endpoint in those who were prefrail [HF 0.89 (0.79-1.00)] and marginally significant placebo benefit in those who were not frail [HR 1.17 (1.01-1.36)], and no difference in non-prespecified endpoints (eg statin use, CKD, quality of life questionnaires)
--rate of major hemorrhage, as above
3. Aspirin effects on all-cause mortality and its association with increased cancer-related mortality(see aspirin all-cause mortality healthy nejm2018 in dropbox, or DOI: 10.1056/NEJMoa1803955 )
Details:
--as per above
Results:
--1052 deaths occurred over the 4.7 years of follow-up
--risk of death from any cause: 12.7 events per 1000 person-yrs on aspirin, vs 11.1/1000 person-yrs with placebo, a 14% increase with aspirin: HR 1.14 (1.01-1.29)
--cancer-related deaths was the major contributor to the higher aspirin-related mortality, occuring in 3.1% on aspirin vs 2.3% placebo: HR 1.31 (1.10-1.56)
--no significant difference for cardiovasc causes (including ischemic stroke), major hemorrhage (including hemorrhagic stroke), or "other" (eg, sepsis, chronic lung disease, dementia, heart failure)
--of the cancer-related deaths (50% of all deaths, vs 19% for cardiovasc causes and 5% for major hemorrhage):
--there was a broad increase in all of the cancers they assessed
--the only one reaching statistical significance was colorectal cancers, with 35 deaths, a 77% increase in group on aspirin (HR 1.77 (1.02-3.06); this was the second most common cancer-related death, after lung cancer in 55
--there was a progressive divergence in both all-causes of death and cancer-related deaths after about 3-4 years
--subgroup analysis: trend to doing better with placebo in both those with and without a personal history of cancer, though a trend to doing better with aspirin in black and Hispanic patients (ie, in the US but not in the Australian cohorts)
Commentary:
--really not sure how to put interpret the cancer death increases with aspirin
--many studies have shown lower colorectal cancers with prolonged use of aspirin (this study was <5 yrs, so not surprising that they did not find benefit, BUT finding a statistically higher association???). eg see http://gmodestmedblogs.blogspot.com/2018/01/aspirin-prevents-colon-cancer.html
--an impressive meta-analysis of 17K people (see Rothwell PM. Lancet 2012; 379: 1591, or aspirin cancer mets lancet 2012 in dropbox) found that aspirin reduced the risk of:
--cancer with distant metastases: 36% decrease in all cancers, 46% decrease in adenocarcinoma, both highly statistically significant, and found after only 6.5 years (though curves showed benefit after only 2-4 years)
--31% decreased risk of adenocarcinoma with metastases at initial diagnosis
--55% decreased risk on subsequent followup
--50% reduction in death from cancer in those who developed adenocarcinoma, particularly in those with initial diagnosis but without mets
--and there were significant numbers of cancers: 1101 incident cancers and 563 deaths due to incident cancers
--results were independent of age > vs <60yo
--and another by Rothwell PM. Lancet 2011; 377:41, of individual patient data from randomized trials, finding:
--25570 patients having 674 cancer deaths
--35% decrease in all cancers, 54% decrease in GI cancers, and benefit found after 5 years' follow-up
--20% decrease in all solid cancers, 35% in GI cancers, over 20 years' follow-up, also significant decrease in adenocarcinomas of the lung and esophagus
--so, how can one explain the differences:
--?age (the current study patients were older, BUT the Rothwell metanalysis did find age at randomization did not affect the overall results, and aspirin's benefit on death from non-GI cancers was the most profound in the group >65yo, though was also decreased in those <55 and 55-64yo.
-- was it the short followup in the ASPREE trial that led to these divergent results? (nothing was apparent in the above Lancet assessments until about 5 years out).
-- was it the pretty high dropout rate in the ASPREE trial?? (though they comment that their dropout rate was pretty similar to other trials, but for some reason was there a different bias in ASPREE than the other trials??)
-- perhaps the differences in the Australian group (quite large, with more evidence of adverse cancer risk associated with aspirin) vs the smaller US group (trend to benefit with aspirin) mean something. is there a different exposure to environmental carcinogens in the US?? (which might mean that an Australian study has limited generalizability to the US population?). does it have to do with the large differences in demographic diversity?? (was the more diverse US population different in other ways? perhaps other types of potential carcinogenic exposures such as to different stressors or other lifestyle issues: stress-mediated cortisol levels might increase risk of cancer, diet/exercise can affect cancer risk.....)
-----------------------------------------------------
so, how does one put this all together (and what is to be done with these results)?? i really am not sure how to process all of this divergent information. it seems to make sense not to routinely include colorectal cancer prevention in the decision analysis to suggest aspirin in those more than 75 years old (as in the ASPREE group), since most studies do suggest benefit only after 1-2 decades of taking aspirin (and the bleeding risks are clear). but, the increased risk of cancer in the last trial seems so different from prior results that i am hesitant to include it in my decision analysis on aspirin therapy (as per above). the utility of aspirin is still clear for secondary prevention, though even there it would be really interesting to know if there were added benefit to aspirin over more aggressive lipid control with statins, which also have anti-inflammatory effects and improve endothelial function (ie, would be great to have an RCT with aspirin vs statin vs both vs neither, with a low statin-induced achieved LDL level). i personally still think that aspirin has a role in younger people who have high cardiovascular risk but no evident heart disease, given the additional colon cancer benefit. but these new studies have at least imprinted questions to aspirin's role in light of the evident harms, pending further data....
relevant recent blogs on aspirin:
http://gmodestmedblogs.blogspot.com/2018/09/aspirin-in-low-risk-diabetics.html which reviews recent study on low-risk diabetic patients,finding a significant decrease in serious vascular events with aspirin, though this was largely outweighed by an increase in major bleeding events
http://gmodestmedblogs.blogspot.com/2018/07/aspirin-one-dose-does-not-fit-all.html which suggests that weight-based aspirin dosing was appropriate
http://gmodestmedblogs.blogspot.com/2018/07/aspirin-one-dose-does-not-fit-all.html which found an increase in cardiovascular events in those who stopped their aspirin
geoff
If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.org
to get access to blogs since 8/15/17:
2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category
3. or you can just type in a name in the search box and get all the blogs with that name in them
to access older blogs from the BMJ website, from October 2013 until 8/15/17: go to http://blogs.bmj.com/bmjebmspotlight/category/archive/
please feel free to circulate this to others. also, if you send me their emails, i can add them to the list
Comments
Post a Comment
if you would like to receive the near-daily emails regularly, please email me at gmodest@uphams.org