In the JournalsPerspective

Moderate alcohol intake increases BP, stroke risk in men

In a genetic epidemiological study, the assumed protective effect of moderate alcohol consumption on CV events was noncausal, and any level of alcohol consumption was associated with increased BP and stroke risk in men.

“Using genetics is a novel way to assess the health effects of alcohol, and to sort out whether moderate drinking really is protective, or whether it’s slightly harmful,” Iona Y. Millwood, DPhil, from the Medical Research Council Population Health Research Unit at the University of Oxford, U.K., said in a press release. “Our genetic analyses have helped us understand the cause-and-effect relationships.”

The researchers analyzed 512,715 Chinese adults with documented levels of alcohol consumption from the China Kadoorie Biobank and followed them for approximately 10 years. Outcomes of interest included ischemic stroke, intracerebral hemorrhage and MI.

In addition, the researchers genotyped 161,498 participants for two variants that alter alcohol metabolism, ALDH2-rs671 and ADH1B-rs1229984.

Among the cohort, 33% of men and 2% of women reported drinking alcohol most weeks.

Drinking not protective

In men, according to self-reported alcohol intake, those who had one or two drinks per day (approximately 100 g per week) had lower risk for ischemic stroke, intracerebral hemorrhage and MI than those who drank more or those who drank not at all. However, genotype-predicted alcohol intake in men was not associated with a similar pattern, Millwood and colleagues wrote.

In a genetic epidemiological study, the assumed protective effect of moderate alcohol consumption on CV events was noncausal, and any level of alcohol consumption was associated with increased BP and stroke risk in men.
Source: Adobe Stock

Rather, as genotype-predicted alcohol intake in men rose, so did risk for ischemic stroke (RR per 280 g per week = 1.27; 95% CI, 1.13-1.43) and especially for intracerebral hemorrhage (RR per 280 g per week = 1.58; 95% CI, 1.36-1.84), according to the researchers.

There was no relationship between genotype-predicted alcohol intake in men and MI risk (RR per 280 g per week = 0.96; 95% CI, 0.78-1.18).

Also in men, increases in usual alcohol intake and genotype-predicted alcohol intake corresponded with increases in BP (P < .0001 for all), according to the researchers.

In women, the genotypes did not predict elevated alcohol intake and were not associated with BP, stroke or MI, Millwood and colleagues wrote.

“There are no protective effects of moderate alcohol intake against stroke. Even moderate alcohol consumption increases the chances of having a stroke,” Zhengming Chen, DPhil, from the Clinical Trial Studies Unit and Epidemiological Studies Unit, Nuffield Department of Population Health, University of Oxford, said in the release. “The findings for heart attack were less clear-cut, so we plan to collect more evidence.”

Further trials unethical

In a related editorial, Shiu Lun Au Yeung, PhD, and Tai Hing Lam, PhD, both from the School of Public Health, Li Ka Shing Faculty of Medicine, University of Hong Kong, wrote, “Considering that this study by Millwood and colleagues and a previous analysis cast doubts about the net protective effects of moderate drinking and the 2016 Global Burden of Disease Study showed harms at all levels of alcohol use, further trials would be unethical.” – by Erik Swain

Disclosures: The study was funded in part by GlaxoSmithKline. Chen and Millwood report receiving funding from GlaxoSmithKline during the conduct of the study. Please see the study for the other authors’ relevant financial disclosures. Lam and Yeung report no relevant financial disclosures.

In a genetic epidemiological study, the assumed protective effect of moderate alcohol consumption on CV events was noncausal, and any level of alcohol consumption was associated with increased BP and stroke risk in men.

“Using genetics is a novel way to assess the health effects of alcohol, and to sort out whether moderate drinking really is protective, or whether it’s slightly harmful,” Iona Y. Millwood, DPhil, from the Medical Research Council Population Health Research Unit at the University of Oxford, U.K., said in a press release. “Our genetic analyses have helped us understand the cause-and-effect relationships.”

The researchers analyzed 512,715 Chinese adults with documented levels of alcohol consumption from the China Kadoorie Biobank and followed them for approximately 10 years. Outcomes of interest included ischemic stroke, intracerebral hemorrhage and MI.

In addition, the researchers genotyped 161,498 participants for two variants that alter alcohol metabolism, ALDH2-rs671 and ADH1B-rs1229984.

Among the cohort, 33% of men and 2% of women reported drinking alcohol most weeks.

Drinking not protective

In men, according to self-reported alcohol intake, those who had one or two drinks per day (approximately 100 g per week) had lower risk for ischemic stroke, intracerebral hemorrhage and MI than those who drank more or those who drank not at all. However, genotype-predicted alcohol intake in men was not associated with a similar pattern, Millwood and colleagues wrote.

In a genetic epidemiological study, the assumed protective effect of moderate alcohol consumption on CV events was noncausal, and any level of alcohol consumption was associated with increased BP and stroke risk in men.
Source: Adobe Stock

Rather, as genotype-predicted alcohol intake in men rose, so did risk for ischemic stroke (RR per 280 g per week = 1.27; 95% CI, 1.13-1.43) and especially for intracerebral hemorrhage (RR per 280 g per week = 1.58; 95% CI, 1.36-1.84), according to the researchers.

There was no relationship between genotype-predicted alcohol intake in men and MI risk (RR per 280 g per week = 0.96; 95% CI, 0.78-1.18).

Also in men, increases in usual alcohol intake and genotype-predicted alcohol intake corresponded with increases in BP (P < .0001 for all), according to the researchers.

In women, the genotypes did not predict elevated alcohol intake and were not associated with BP, stroke or MI, Millwood and colleagues wrote.

“There are no protective effects of moderate alcohol intake against stroke. Even moderate alcohol consumption increases the chances of having a stroke,” Zhengming Chen, DPhil, from the Clinical Trial Studies Unit and Epidemiological Studies Unit, Nuffield Department of Population Health, University of Oxford, said in the release. “The findings for heart attack were less clear-cut, so we plan to collect more evidence.”

PAGE BREAK

Further trials unethical

In a related editorial, Shiu Lun Au Yeung, PhD, and Tai Hing Lam, PhD, both from the School of Public Health, Li Ka Shing Faculty of Medicine, University of Hong Kong, wrote, “Considering that this study by Millwood and colleagues and a previous analysis cast doubts about the net protective effects of moderate drinking and the 2016 Global Burden of Disease Study showed harms at all levels of alcohol use, further trials would be unethical.” – by Erik Swain

Disclosures: The study was funded in part by GlaxoSmithKline. Chen and Millwood report receiving funding from GlaxoSmithKline during the conduct of the study. Please see the study for the other authors’ relevant financial disclosures. Lam and Yeung report no relevant financial disclosures.

    Perspective
    Matthew F. Muldoon

    Matthew F. Muldoon

    This study adds information about whether alcohol might be protective in smaller amounts. The study found that it was not.

    The researchers did the study in China for a reason. There is a particular research strategy that they had using certain genetic polymorphisms that predict alcohol intake that are more robust in Chinese people. It nonetheless speaks generally to the effect of alcohol and CVD.

    This study is not without significant limitations. At the end of the day, it is an observational study. It is not a clinical trial and it is only through experimental studies such as a clinical trial that we can really show cause and effect. This study employs a newer technique, Mendelian randomization, to improve upon our ability in observational studies to demonstrate cause and effect. It is a step forward in terms of methodologies for observational studies.

    Nonetheless, this study was rigorously done and the findings are important. They are probably as important in China as they are around the world.

    We have long thought that alcohol is protective at low levels, but we have always known that it is detrimental at high levels for CVD. Whether this protective effect in low amounts has not been proven. The suspicion has been based upon observational studies. It is not as if we have trials that show that low amounts of alcohol are protective. This “new and improved” observational study does not find a protective effect in low amounts. That is important.

    Clinicians, by and large, do not encourage alcohol intake. What we often do is when we encounter someone with moderate intake, we say, "That is fine." We sometimes go so far as to say that what the patient is doing might be good. This study casts some doubt upon us continuing that last statement. The new study did not find that low amounts are bad (one drink per day for women and two drinks per day for men). Certainly, once you get beyond that, it is clearly bad no matter how you look at the data.

    We may need to back away from saying or implying that mild consumption is a good idea and simply say that alcohol in excess is bad, rather than sometimes give some implicit encouragement to continue two drinks per day. The public has heard that message, but we have never known it to be a valid recommendation.

    There is another subtlety in this study related to the fact that it was done in China. Asians have a lot more stroke than heart disease, and they have a lot more hemorrhagic stroke than ischemic stroke. These data show that high amounts of alcohol are particularly bad for stroke, yet not related much at all to heart disease. In the United States, we are burdened more so by ischemic heart disease. So that does affect the applicability of this study to Americans. The researchers found that alcohol was most strongly associated with hemorrhagic stroke, which is not as prominent of an issue here as it is in China.

    • Matthew F. Muldoon, MD, MPH
    • Professor of Medicine
      University of Pittsburgh

    Disclosures: Muldoon reports no relevant financial disclosures.