Drinking-water hardness and cardiovascular diseases: A review of epidemiological studies 1979-2004
Introduction
Cardiovascular diseases (CVDs) are among the main causes of mortality and morbidity in the industrialised countries and their main risk factors are hypertension, dyslipidemia, smoking, alcohol abuse, dietary habits and physical inactivity (Hornstra et al. 1998; Wilson 1999). However, these classic factors do not entirely explain the variability of CVD mortality in different countries. In order to better understand the determinants of CVD, particular attention has been paid to environmental factors, such as weather, air pollution or the mineral content of drinking water (DW). Since the 1950s a causal relation between DW hardness and some CVD has been hypothesized. The relationship between cardiovascular mortality and the mineral content of DW was first described by Kobayashi (1957) in Japan and by Schroeder (1960) in the United States. Since then, many epidemiological studies have been conducted worldwide, most of them describing a protective relationship between CVD mortality rates and DW hardness. A first series of studies was performed in the 1960s and 1970s. As noted in the previous chapter (Calderon and Craun), most of them had an ecological design and used mortality data from national registers and geographical areas defined on the basis of administrative boundaries as units of analysis. Critical reviews of these early studies emphasized that these studies had a major weakness. Because they considered average values of DW parameters, such as total hardness, calcium (Ca), or magnesium (Mg), they could induce considerable non-differential misclassification. Moreover, the temporal sequence of exposure and supposed effect (i.e., exposure to the DW risk factor precedes CVD mortality) was not always verified. In fact, the mineral content of DW was often determined at the time of the study and thus, may not represent the quality of the water the subjects had ingested during their lives. However, it should be pointed out that DW hardness is usually quite stable in time and that the health effects related to Ca and Mg concentrations may be both long- and short-term. Moreover, the main risk factors for CVD, which may be confounding factors of the relationship between DW hardness and CVD mortality, had often not been taken into account in the analysis.
More recently, several ecological studies were performed with more attention paid to exposure assessment and confounding factors. Epidemiological studies performed with a cohort design or with a case-control design were also carried out, but these studies are less numerous than the ecological studies even though they offer a greater potential for understanding the relationship between DW hardness and CVD mortality.
The hypothesized beneficial effect of DW hardness on CVD mortality may be due to: (1) the higher intake of calcium (Ca) and/or magnesium (Mg) itself; (2) the protective effect of other trace elements possibly present in hard water (e.g., selenium, lithium, silicon, zinc, vanadium); (3) the reduced adverse effect of toxics such as lead, which may be present at a higher concentration in soft, low pH corrosive water (Eisenberg 1992; Rylander 1996; Marx and Neutra 1997). In fact, the presence of lead, even at low blood levels, has long been associated with hypertension and also with stroke (Pirkle et al. 1985; Perry and Roccella 1998).
In this review we evaluated all of the epidemiological studies published on this issue since the early 1980s.