Published online before print August 21, 2003, doi: 10.1161/01.STR.0000089495.71068.90
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Stroke. 2003;34:e143.)
© 2003 American Heart Association, Inc.
Letters to the Editor |
Homocysteine and the Risk of Stroke and Thromboembolism in Atrial Fibrillation: An Uncertain Role
Haemostasis, Thrombosis, and Vascular Biology Unit, University Department of Medicine, City Hospital, Birmingham, UK
To the Editor:
We read with great interest the article by Ay et al1 on the role of homocysteine in stroke and thromboembolic complications in patients with atrial fibrillation. Certainly, homocysteine has previously been shown to be associated with an increased risk of stroke,2,3 thrombosis,4 and impaired fibrinolysis5; it is therefore important that its association with atrial fibrillation and the risk of left atrial thrombus and subsequent strokes be clarified.
Nonetheless, we have issues with certain aspects of the study by Ay et al.1 First, the numbers in groups studied are small (22 and 20, respectively). We agree that recruiting patients with an acute stroke and nonvalvular atrial fibrillation may not be easy, but in order to draw meaningful conclusions, we need to have larger numbers to avoid type II errors. A power calculation may be needed to convince the skeptics. Indeed, to show a (marginal) difference in mean/median homocysteine levels in stroke or ischemic heart disease between patents and controls, very large numbers (ranging from 110 to 
400) were previously required in each group.6,7
Furthermore, a majority of the patients in the study by Ay et al1 had some form of vascular disease. Hypertension, which in itself is a risk factor for high homocysteine levels,8,9 was present in 70% in each group. Similarly, diabetes and ischemic heart disease were also present in many patients. Although there was no statistical difference between the groups, the patient numbers studied are probably not large enough to account for the many confounders and associated comorbidity that would influence homocysteine levels—indeed, statistical adjustments cannot fully account for all biological interactions and physiological processes.
Similarly, other important factors influencing the formation of left atrial thrombus—such as duration of atrial fibrillation, mean left atrial size for the 2 groups, and the use of antithrombotic agents or anticoagulants prior to the stroke—were not specified. Also, as the authors have rightly pointed out, homocysteine levels do rise in an acute stroke, and this may have some bearing on the levels measured.
Finally, is the association biologically plausible? We know that high homocysteine levels are associated with an increased risk of thrombosis in vascular disease, and atrial fibrillation is commonly associated with vascular disease. While the evidence is clear that atrial fibrillation confers a prothrombotic or hypercoagulable state,10 the possibility that increased homocysteine levels predict left atrial thrombosis in atrial fibrillation per se, independent of associated vascular disease and other "thrombogenic" factors, seems extremely remote. Indeed, as the authors have pointed out, Friedman11 did not show any difference in homocysteine levels between patients with atrial fibrillation and those in sinus rhythm. In this study, the patients who had a previous stroke did have a higher level of homocysteine than those who did not, but they were significantly older, and the authors did show a strong correlation between age and homocysteine levels. This study therefore refutes rather than corroborates the findings of Ay et al.1
Thus, we respectfully disagree with the authors’ conclusion that their data are suggestive that hyperhomocysteinemia may be accepted as a risk factor for stroke and thromboembolism in patients with nonvalvular atrial fibrillation.
References
1. Ay H, Arsava EM, Tokgozoglu SL, Ozer N, Saribas O. Hyperhomocysteinemia is associated with the presence of left atrial thrombus in stroke patients with nonvalvular atrial fibrillation. Stroke. 2003; 34: 909–912.
2. Sacco RL, Roberts JK, Jacobs BS. Homocysteine as a risk factor for ischemic stroke: an epidemiological story in evolution. Neuroepidemiology. 1998; 17: 167–173.[CrossRef][Medline] [Order article via Infotrieve]
3. Homocysteine Studies Collaboration. Homocysteine and risk of ischemic heart disease and stroke: a meta-analysis. JAMA. 2002; 288: 2015–2022.
4. Hainaut P, Jaumotte C, Verhelst D, Wallemacq P, Gala JL, Lavenne E, Heusterspreute M, Zech F, Moriau M. Hyperhomocysteinemia and venous thromboembolism: a risk factor more prevalent in the elderly and in idiopathic cases. Thromb Res. 2002; 106: 121–125.[CrossRef][Medline] [Order article via Infotrieve]
5. Tofler GH, D’Agostino RB, Jacques PF, Bostom AG, Wilson PW, Lipinska I, Mittleman MA, Selhub J. Association between increased homocysteine levels and impaired fibrinolytic potential: potential mechanism for cardiovascular risk. Thromb Haemost. 2002; 88: 799–804.[Medline] [Order article via Infotrieve]
6. Perry IJ, Refsum H, Morris RW, Ebrahim SB, Ueland PM, Shaper AG. Prospective study of serum total homocysteine concentration and risk of stroke in middle-aged British men. Lancet. 1995; 346: 1395–1398.[CrossRef][Medline] [Order article via Infotrieve]
7. Whincup PH, Refsum H, Perry IJ, Morris R, Walker M, Lennon L, Thomson A, Ueland PM, Ebrahim SB. Serum total homocysteine and coronary heart disease: prospective study in middle aged men. Heart. 1999; 82: 448–454.
8. Lip GYH, Edmunds E, Martin SC, Jones AF, Blann AD, Beevers DG. A pilot study of homocysteine levels in essential hypertension: relationship to von Willebrand factor, an index of endothelial damage. Am J Hypertens. 2001; 14: 627–631.[CrossRef][Medline] [Order article via Infotrieve]
9. Sheu WH, Lee WJ, Chen YT. Plasma homocysteine concentrations and insulin sensitivity in hypertensive subjects. Am J Hypertens. 2000; 13: 14–20.[CrossRef][Medline] [Order article via Infotrieve]
10. Lip GYH. Does atrial fibrillation confer a hypercoagulable state? Lancet. 1995; 346: 1313–1314.[CrossRef][Medline] [Order article via Infotrieve]
11. Friedman HS. Serum homocysteine and stroke in atrial fibrillation. Ann Intern Med. 2001; 134: 253–254.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||