Environmental tobacco smoke, cardiovascular disease, and the nonlinear dose-response hypothesis

Two recent government reports have focused attention on the hypothesis that environmental tobacco smoke (ETS) exposure increases the risk of cardiovascular disease (CVD) in nonsmokers. The first report was published by the California Environmental Protection Agency (CAEPA) in 1997. The second report was issued in 1998 by the Scientific Committee on Tobacco and Health (SCOTH) in the United Kingdom. A meta-analysis of five large prospective epidemiology studies reports that the relative risk for actively smoking 20 cigarettes per day is 1.78. Active smoking exposes the smoker to approximately 16 times the ETS concentration, and 100- to 300-fold the total smoke dose experienced by a nonsmoker (Smith and Ogden, 1998, JAMA 280, 32-33.). Despite the much lower smoke exposure, these government reports estimate the relative risk for ischemic heart disease in ETS-exposed nonsmokers at 1.30 (CAEPA) and 1.23 (SCOTH). As an explanation for this nonlinear dose-response anomaly, platelet aggregation is proposed to be a plausible and quantitatively consistent mechanism. Herein, evidence is presented suggesting that this low-dose hypothesis is inconsistent with the biochemistry and physiology of platelets and with the literature on the cardiovascular pathology of active smoking. In addition, several important biases and confounders are ignored. These epidemiologic biases and confounders include the following: misclassification of smokers as nonsmokers; improper use of death certificates as surrogates for mortality statistics; underreporting of diabetes and hypertension in the relatives of smokers; and additional atherogenic risk factors in smoking households. Future field studies on ETS and CVD should emphasize proximal markers of risk for thrombosis in exposed nonsmokers. Proximal thrombogenic risk markers identified in field studies should be mechanistically examined under controlled exposure conditions.