Drs Miyawaki and Maesaka perceptively point out1 that our data2,3 show that, independent of alcohol intake, systemic hypertension leads to increased risks of cardiovascular events. Thus, for several outcomes the increased relative risks of persons with elevated blood pressure (vs optimal pressures) were similar in abstainers, light drinkers, and heavy drinkers. In the abbreviated article for , I was not able to present other aspects of our data clearly enough to preclude misinterpretations. First, these findings do not negate the inverse association of alcohol with some cardiovascular conditions. Second, the inference that the cardiovascular effects of alcohol are primarily due to hypertension is incorrect.
The complex situation is composed of at least 3 independent relationships. One is the increased prevalence and risk of systemic hypertension in heavier drinkers. Hypertension prevalence is slightly increased even at reported intake of 1-2 drinks per day, although some of this appears related to underreporting of intake.4 Second is the lowered risk in light-moderate drinkers of atherothrombotic conditions, primarily coronary artery disease. Third is the increased risk of cardiovascular sequelae in persons with hypertension compared with those with lower blood pressure, a blood pressure effect that has nothing to do with the alcohol drinking or absence of same.
It is thus correct that persons reporting 1-2 drinks per day simultaneously enjoy lower coronary risk and have slightly higher risk of hypertension. Among moderate drinkers, the cardiovascular benefits dominate, so that the J-shaped alcohol-coronary curve (from abstinence to heavy drinking) is not obliterated. As Miyawaki and Maesaka state,1 hypertension is only 1 of the coronary artery disease risk factors and moderate drinking favorably affects several others. Furthermore, only a small proportion of persons with blood pressure ≥ 140/90 mm Hg reporting 1-2 drinks per day have hypertension. The increased risk of hypertension among these moderate drinkers was ~ 10%, making it likely that ~ 90% had hypertension for other reasons. As we pointed out,2,3 we had no way to tell which persons with hypertension had hypertension.
We agree with Miyawaki and Maesaka that the important point is that, even at moderate drinking levels where the cardiovascular benefits dominate, hypertension should not be ignored. While medical advice about alcohol drinking needs to be individualized,5 it seems clear that a typical middle-aged or older moderate drinker with hypertension would not be better off to quit drinking. What would almost surely be of benefit would be to control the high blood pressure.
Arthur L. Klatsky, MD
Kaiser Permanente Medical Center
Drs Maesaka and Miyawaki appreciate the author's clarifying comments.
1. Miyawaki N, Maesaka JK. Uncorking new evidence in the alcohol-hypertension relationship. Cardiol Rev. 2007;24(7):39-40.
2. Klatsky AL. Alcohol and hypertension. Cardiol Rev. 2007;24(7):35-38.
3. Klatsky AL, Koplik S, Gunderson E, et al. Sequelae of systemic hypertension in alcohol abstainers, light drinkers, and heavy drinkers. Am J Cardiol. 2006;98(7):1063-1068.
4. Klatsky AL, Gunderson EP, Kipp H, et al. Higher prevalence of systemic hypertension among moderate alcohol drinkers: an exploration of the role of underreporting. J Stud Alcohol. 2006;67(3):421-428.
5. Klatsky AL. Drink to your health? Sci Am. 2006;16(4):22-29.
Drs Schuijf and Bax1 shed light on one of the more puzzling questions raised with the emergence of a new, pervasive technology (multislice computed tomography [MSCT]) and its role in the assessment of cardiovascular disease. It is true that both MSCT and perfusion imaging can act in concert to determine the plaque burden and the functional assessment of the coronary arteries, respectively. However, it is important to distinguish those candidates who have a high pretest probability of having significant coronary disease from those who may have flow-limiting disease. Multislice computed tomography has proven to be an excellent tool for determining the former. But, given a candidate with multiple risk factors who presents with atypical or typical symptoms, this tool may not be the ideal in assessing whether that same stenosis seen by MSCT is causing a perfusion defect (because artifact generated by calcified plaques may [partially] obscure the lumen of the coronary arteries, thereby precluding adequate assessment of the severity of coronary stenosis).
Saadi Siddiqi, DO
St. Francis Hospital