Dietary magnesium does not predict ischaemic heart disease in the Caerphilly cohort.

OBJECTIVES

To examine the association between dietary magnesium intake and the risk of an ischaemic heart disease (IHD) event.

DESIGN

Estimates were made of dietary magnesium intake from food frequency records, supplemented by seven-day weighted intake records. The subsequent incidence of ischaemic heart disease events was recorded. The relative odds of an IHD event was related to base-line magnesium intake.

SETTING

Data on dietary magnesium intakes are available for 2172 men aged 45-59 y in the Caerphilly cohort. These have now been followed for ten years since base-line dietary data were collected, and during this time a total of 269 IHD events occurred. Of these, 96 were acute deaths (ICD 410) and 136 were non-fatal myocardial infarctions.

RESULTS

The overall mean dietary intake of magnesium was estimated to be 279 (s.d. 83) mg/day. The daily intake of those men who later experienced any IHD event was 266 (s.d. 84) mg/day and this differs from that in men who experienced no IHD event during this time (281 mg, P < 0.05). Men who suffered an acute IHD death had even lower intakes (mean 253 (s.d. 79); P < 0.005). Age, smoking habit, energy intake and alcohol consumption are all significantly associated with both Mg intake and IHD risk and are therefore possible confounding factors. Standardisation for these factors reduces the difference for all IHD events to 2.9 (s.e.m. 3.6) mg Mg/day, P > 0.05, and to 0.9 (s.e.m. 5.8) mg for acute IHD death. Similarly, when the men are ranked into fifths by their daily Mg intake, 70 of the 434 men with the lowest intakes went on to experience an IHD event, compared with only 41 of the 434 men with the highest Mg intakes. The relative odds (RO) for the fifth of men with the lowest intakes, compared with the fifth with the highest intakes, is 1.86 (P < 0.005), but standardisation for the confounding factors leads again to a loss of significance (RO 1.52, P > 0.05).

CONCLUSION

Although trends in the data are suggestive, data from the Caerphilly cohort give no certain evidence that dietary magnesium intake is independently predictive of ischaemic heart disease in the population studied.