Stroke is the third most common cause of death in developed countries. There are 125,000 cases of stroke in the UK each year and 60,000 deaths. Because treatment is limited, prevention through modification of risk factors such as high blood pressure and smoking is important.
Observational studies measuring homocysteine concentrations in healthy individuals have found that, on average, those with high levels of homocysteine in their blood are more likely to have a stroke. But other factors that increase homocysteine concentration and stroke risk--such as smoking and socioeconomic class--may be responsible for the relationship observed in these studies, confounding the result.
Uncovering whether or not homocysteine concentration is linked to stroke is important because folic acid and B vitamins can lower its concentration, although the results of large randomised trials to examine the efficacy and safety of this approach are still awaited.
Aroon Hingorani (University College London, UK) and colleagues used an approach known as Mendelian randomisation to overcome some of the problems of confounding factors seen in observational studies. The investigators analysed published data on the association between stroke and common variation in a gene called MTHFR, which is known to influence homocysteine concentration. People randomly inherit variant (TT) of the gene, which gives them a higher concentration of blood homocysteine, or variant (CC), which results in lower levels of the amino acid in their blood.
The authors compared risk estimates from observational studies of homocysteine and stroke with those obtained from genetic studies of MTHFR and stroke. They found individuals who carried the TT genotype of MTHFR had, on average, both a higher concentration of homocysteine, and a small increase in the risk of stroke. The effect of the variant on stroke risk was close to that expected from its effect on homocysteine concentration.
Dr Hingorani comments: "Because of the random allocation of the gene variant in advance of disease development this concordance of risk estimates implies that the relation between homocysteine and stroke seen in observational studies is not substantially confounded by other factors. Our study therefore provides evidence for a role of homocysteine in the development of stroke, though it must be emphasized that a systematic review of published studies such as this might be affected to some degree by reporting bias.
He continues: "Although it is already known that folic acid can lower people's homocysteine, our study does not provide any evidence that taking folic acid is beneficial or safe. Large randomised clinical trials of supplementation with folic acid, with or without vitamin B, will be necessary to test the therapeutic approach of lowering homocysteine concentrations to prevent stroke. Also, because many strokes still occur in people without the gene variant and with near-average levels of homocysteine, the study does not indicate that testing for homocysteine or variations in the MTHFR gene would necessarily be helpful in risk prediction". (Quote by e-mail; does not appear in published paper)
In an accompanying commentary (p,194) Graeme Hankey and John Eikelboom (Royal Perth Hospital, Australia) write that the results do not provide conclusive evidence of a causal association between homocysteine concentration and stroke because the MTHFR gene variations might affect behavioural or socioeconomic factors. Unhealthy behaviors or low socioeconomic status may therefore be alternative explanations for the finding. The authors comment that even if it is established that homocysteine concentration increases risk of stroke, it may not be wise for doctors to prescribe long-term vitamin B therapy to a broad group.
Contact: Dr Aroon D Hingorani, Centre for Clinical Pharmacology, Division of Medicine, British Heart Foundation Laboratories, University College London, Rayne Building, 5 University Street, London, WC1E 6JJ, UK. T) 44-207-679-6612 a.hingorani@ucl.ac.uk
(Commentary) Contact: Professor Graeme J Hankey, Department of Neurology, Royal Perth Hospital, Stroke Unit, Level 8, "A" Block, 197 Wellington Street, Perth 6001, Australia. T) 61-8-9224-1758 gjhankey@cyllene.uwa.edu.au
Journal
The Lancet