Fish, mercury, selenium and cardiovascular risk: current evidence and unanswered questions

Abstract

Controversy has arisen among the public and in the media regarding the health effects of fish intake in adults. Substantial evidence indicates that fish consumption reduces coronary heart disease mortality, the leading cause of death in developed and most developing nations. Conversely, concerns have grown regarding potential effects of exposure to mercury found in some fish. Seafood species are also rich in selenium, an essential trace element that may protect against both cardiovascular disease and toxic effects of mercury. Such protective effects would have direct implications for recommendations regarding optimal selenium intake and for assessing the potential impact of mercury exposure from fish intake in different populations. Because fish consumption appears to have important health benefits in adults, elucidating the relationships between fish intake, mercury and selenium exposure, and health risk is of considerable scientific and public health relevance. The evidence for health effects of fish consumption in adults is reviewed, focusing on the strength and consistency of evidence and relative magnitudes of effects of omega-3 fatty acids, mercury, and selenium. Given the preponderance of evidence, the focus is on cardiovascular effects, but other potential health effects, as well as potential effects of polychlorinated biphenyls and dioxins in fish, are also briefly reviewed. The relevant current unanswered questions and directions of further research are summarized.

Keywords: cardiovascular disease; fish; mercury; review; selenium.

Figure 1.

Schema of physiologic effects of n-3 PUFA consumption. The strength of effect denotes the relative impact of n-3 PUFA intake on clinical risk by means of the physiologic effect (e.g., triglyceride-lowering), whereas the time course to alter clinical events denotes the expected duration of intake for the benefits on the physiologic effect to be manifested in improved disease outcomes. For instance, the dose-response for anti-arrhythmic effects appears to be initially steep with a subsequent plateau, and effects on disease outcomes may be seen in weeks to months, whereas the dose-response for triglyceride-lowering is more gradual and monotonic, and effects on disease outcomes may require months to years of intake. Potentially important effects of n-3 PUFA on endothelial, autonomic, anti-inflammatory, and diastolic responses are not shown because dose- and time-responses of these effects are less well-established. Physiologic effects are not necessarily exclusive: e.g., anti-arrhythmic effects may be partly mediated by effects on blood pressure (BP) or heart rate. Reproduced with permission from Mozaffarian and Rimm [1].

Figure 2.

The number of 3.5 oz (100 g) fish servings per week needed to provide an average of 250 mg/day of the marine n-3 polyunsaturated fatty acids eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). Based on data from Mozaffarian and Rimm [1].

Figure 3.

Meta-analysis of studies of mercury exposure and risk of coronary heart disease (CHD). Relative risk (▪) and 95% CIs (–) are shown comparing the highest to the lowest quantile of mercury exposure after adjustment for other risk factors. Adapted from Mozaffarian and Rimm [1].

Figure 4.

The interaction between exposure to n-3 PUFA and mercury from fish consumption. Serum docosahexaenoic acid (DHA) and docosopentaenoic acid (DPA) and hair mercury were measured in 1,871 participants who were prospectively followed for fatal or nonfatal acute coronary heart disease (CHD) [77]. Greater fish consumption (reflected by higher serum n-3 PUFA levels) was associated with lower CHD risk whether mercury exposure was high (> 2.0 ug/g, representing the highest tertile of exposure) or low (< 2.0 ug/g, representing the lowest two tertiles of exposure). However, the slope of this benefit was different: the higher relative risk of CHD (RR = 1.66) seen with higher mercury exposure in this study reflects, in essence, the difference in slope between these two lines. Thus, those with higher mercury exposure had less relative benefit - but not net harm – from higher fish consumption.

Figure 5.

Calculated cancer risks vs. cardiac benefits from a lifetime of regular consumption of farmed or wild salmon, based on contents of PCBs/dioxins and n-3 PUFA in farmed and wild salmon [114]. The estimated cancer risks - 24 vs. 8 deaths for farmed vs. wild salmon, respectively – are orders of magnitude smaller than the calculated cardiac benefits of 7,125 fewer cardiac deaths for either farmed or wild salmon consumption.