Plasma Homocyst(e)ine Concentrations in Cerebrovascular Disease
To the Editor:
We read with interest the article, “Plasma Homocyst(e)ine Concentration, but Not MTHFR Genotype, Is Associated With Variation in Carotid Plaque Area.”1 The authors showed that compared with reference control subjects, homocysteine concentrations were increased (13.7±0.73 versus 10.6±5.6 μmol/L) in 307 patients (148 women, 159 men) with stenotic artery disease (as indicated by carotid plaque area in 2D ultrasound). They observed a strong association between extent of carotid artery disease and homocysteine concentrations but not with the termolabile mutation (677T) in the gene encoding the methylenetetrahydrofolate reductase (MTHFR). Thus, their observations are in accordance with earlier reports.2
We would like to comment on 2 points. First, recently we differentially investigated homocysteine concentrations in 144 patients (55 females, 89 males) with cerebral large-vessel disease (plaques or stenoses of the extracranial arteries and/or increase of cerebral blood flow velocity of ≥140 cm/s in the large intracranial vessels) and small-vessel disease (ie, subcortical vascular encephalopathy).3 The latter disease is clinically characterized by stepwise progressive mnestic deficits and cognitive decline, typical gait disorders, and incontinence and neuroradiologically by diffuse periventricular white matter abnormalities and central lacunar lesions.4 We observed that patients with this cerebral microangiopathy exhibited surprisingly high concentrations of homocysteine (18.2±8.5 μmol/L).3 Logistic regression analysis revealed that hyperhomocysteinemia is an independent risk factor for subcortical vascular encephalopathy (OR=5.7). Surprisingly, we did not find significantly increased levels in patients with cerebral large-vessel disease, after the exclusion of patients with cerebral microangiopathy. Therefore, elevated homocysteine concentrations in carotid artery disease observed in the study of Spence et al1 and in earlier studies2 could be, in part, due to the presence of some patients with undiagnosed cerebral microangiopathy in their study populations, because microangiopathy and macroangiopathy often occur together. Without exclusion of cerebral microangiopathy by exact neurological, neuropsychological, and neuroradiological examinations, it is difficult to attribute elevated homocysteine concentration to cerebral large-vessel disease.
Second, regarding possible causes of homocysteinemia, Spence et al could not find an association between carotid artery disease and MTHFR genotype. They observed that the majority of subjects with elevated plasma homocysteine concentrations did not have the MTHFR 677T genotype and suggested further causes contributing to hyperhomocysteinemia. This is in accordance with our observation of decreased plasma concentrations of vitamins B6 (9.9±5.2 versus 15.2±14.1 μg/L) and B12 (371.2±226.9 versus 451.3±270.0 ng/L) in cerebral microangiopathy and a significant correlation between concentrations of vitamin B12 (r=−0.24, P<0.05) and folate (r=0.29, P=0.01). These associations suggest—although they do not demonstrate—that hypovitaminosis rather than genetical factors could have contributed to hyperhomocysteinemia in cerebrovascular disease.
- Copyright © 1999 by American Heart Association
Spence JD, Malinow MR, Barnett PA, Marian AJ, Freeman D, Hegele RA. Plasma homocyst(e)ine concentrations, but not MTHFR genotype, is associated with variation in carotid plaque area. Stroke.. 1999;30:969–973.
Ghika J, Bogousslavsky J. Subcortical arteriosclerotic encephalopathy (Binswanger’s disease). In: Ginsberg MD, Bogousslavsky J, eds. Cerebrovascular Disease: Pathophysiology, Diagnosis, and Management. Malden, Mass: Blackwell Science; 1998:1755–1771.
Dr Fassbender and colleagues correctly suggest that plasma homocyst(e)ine should be studied for its association with other well-characterized cerebrovascular phenotypes, such as cerebral microangiopathy. However, their assertion that it is “difficult” to associate an elevated plasma homocysteine concentration with cerebral large-vessel disease without determining cerebral microangiopathy is incongruous with our results. Our data strongly suggested that carotid arterial changes were associated with elevated plasma homocyst(e)ine.R1 Whether or not some of our study subjects might also have had cerebral microangiopathy is immaterial to the observed association of homocyst(e)ine with carotid plaque area. Because we could measure carotid plaque area with a high degree of accuracy, we used carotid plaque area as a dependent quantitative trait in a linear regression model to determine the sources of its variation. This analysis led us to conclude that plasma homocyst(e)ine was a significant determinant of carotid plaque area whereas MTHFR genotypes were not.R1 Some subjects might have been ascertained because their presenting symptoms had resulted from cerebral microangiopathy. However, our subsequent phenotypic analysis indicated that almost 80% of study subjects had detectable carotid plaque area.R2 We were not interested in, nor can we now exclude, a possible association between plasma homocyst(e)ine and unmeasured cerebral microangiopathy in some subjects. However, the existence of such a possibility does not affect our interpretation that plasma homocyst(e)ine is significantly associated with variation in carotid plaque area.
We also agree that an inadequate intake of vitamins is probably an important contributor to hyperhomocysteinemia. Most published studies to date have little or no information on the vitamin status of the study subjects. This data gap is notable, because gene–environment interactions certainly play a role in homocyst(e)ine metabolism.R2 Future studies will need to carefully document dietary intake of vitamins and also possibly assay from blood and other tissues the concentrations of those vitamins that play a role in homocysteine metabolism.