Professor Ming-Ching Shen, Department of Laboratory Medicine, No.7 Chung-Shan South Road, Taipei 100, Taiwan. E-mail: email@example.com
Summary. The biological effects of age, sex and vitamin status on plasma total homocysteine (tHcy), and association of hyperhomocysteinaemia with venous thromboembolism in Taiwanese Chinese individuals, were investigated. Eighty patients (16–85 years) with venous thrombophilia and 123 healthy subjects (15–85 years) without history of vascular thrombosis were studied for plasma levels of tHcy, folate and vitamin B12. A multivariate analysis in healthy subjects revealed that plasma tHcy levels tended to increase with age (P < 0·001) and with decreasing plasma levels of folate (P=0·001) or vitamin B12 (P < 0·029); men tended to have higher plasma tHcy levels than women (P=0·006). Thrombotic risk assessment in a case–control study demonstrated that neither plasma level of tHcy [odds ratio (OR), 1·07; 95% confidence interval (CI), 0·96–1·18; P=0·210] nor hyperhomocysteinaemia (OR, 1·65; 95% CI, 0·50–5·49; P=0·415) was significantly associated with venous thrombophilia. The relationship between hyperhomocysteinaemia and recurrence of episode remained insignificant (P=0·560). We conclude that age, sex and vitamin status affect plasma tHcy but hyperhomocysteinaemia is possibly not an important risk factor for venous thrombophilia in Taiwanese Chinese.
There are two pathways, i.e. trans-sulphuration and re-methylation, involved in homocysteine metabolism in vivo (Selhub & Miller, 1991). Both genetic and environmental factors that participate in these pathways can regulate plasma homocysteine levels. The rare homozygous deficiency of cystathionine beta synthetase (CBS), which is needed in a trans-sulphuration pathway, is the most frequent cause of severe hyperhomocysteinaemia (Mudd et al, 1995). Another rare homozygous deficiency of methylenetetrahydrofolate reductase (MTHFR), which catalyses the reduction of methylenetetrahydrofolate to methyltetrahydrofolate, may also cause severe hyperhomocysteinaemia and is the most common inherited defect of the remethylation pathway (Rosenblatt et al, 1995). Mild to moderate hyperhomocysteinaemia is associated with heterozygous CBS or MTHFR deficiencies or a thermolabile MTHFR; the latter is due to a homozygous C to T substitution at nucleotide 677 of the encoding gene, the prevalence of which varies between 5% and 12% in Caucasians and Chinese (Frosst et al, 1995; Motulsky, 1996; Lin et al, 2000; Zheng et al, 2000). Apart from the genetic factors, there are other biological and nutritional factors, e.g. age, sex and vitamin status, which may affect plasma homocysteine level (Selhub et al, 1993; Jacques et al, 1999).
Hyperhomocysteinaemia is now recognized as an independent risk factor for occlusive arterial diseases in Caucasians (Clarke et al, 1991; Ueland et al, 1992; Boushey et al, 1995). Two prospective studies in Caucasians failed to demonstrate that hyperhomocysteinaemia was a predictor of the risk for venous thromboembolism (Petri et al, 1996; Ridker et al, 1997). However, two recent meta-analyses had supported hyperhomocysteinaemia as a risk factor for venous thrombosis in the Caucasian population (den Heijer et al, 1998; Ray, 1998). In addition, hyperhomocysteinaemia has been reported to be associated with recurrent venous thrombosis (Eichinger et al, 1998). The reported prevalence of hyperhomocysteinaemia in Caucasians ranged from 5·7% to 25·0% and 0% to 11·5% in patients with venous thrombosis and control subjects respectively (D'Angelo & Selhub, 1997). Whether hyperhomocysteinaemia is associated with increased risk of venous thrombosis is unknown in the Taiwanese Chinese. In addition, there seems to exist a racial difference in the cause of venous thrombosis (Shen et al, 1997; Lin et al, 1998) between Caucasians and Chinese. The effects of age, sex and vitamin status on plasma homocysteine level in Chinese population appeared to be controversial (Lolin et al, 1996; Ho, 2000). The aim of this study was to investigate the association between hyperhomocysteinaemia and venous thrombosis, and to understand any racial difference in the biological effects of age, sex and vitamin status on homocysteine level in Taiwanese Chinese.
Patients and methods
Study population. From 1998 to March 2000, patients with a history of venous thrombosis who were consecutively referred to the Coagulation Laboratory for investigating the causes of venous thrombophilia at National Taiwan University Hospital were evaluated. Laboratory tests evaluated included deficiencies in antithrombin III, protein C, protein S or plasminogen, dysfibrinogenaemia and antiphospholipid antibodies (anticardiolipin antibody and lupus anticoagulant). Factor V Leiden and prothrombin G20210A mutation were not included because of the extremely low prevalence in the Chinese population (Shen et al, 1997; Lin et al, 1998). Patients who met the inclusion criteria were recruited for this study after acquiring informed consent. These criteria included objectively documented venous thrombosis (deep vein thrombosis, pulmonary embolism or cerebral sinus thrombosis) using image studies, and no thrombotic episode could be related to hepatic, renal or myeloproliferative disorders, malignancy, surgery and pregnancy. Healthy Chinese subjects without any history of vascular thrombosis were randomly selected from hospital employees, medical students and individuals requesting a routine health examination.
Blood sampling and laboratory procedures. All studied subjects were instructed to stop taking over-the-counter vitamin supplements for 2 or more weeks and to fast for at least 10 h before collection of whole blood in the morning. Blood samples of patients were collected at the time after thrombosis when oral anticoagulation was stable. EDTA-anticoagulated blood was placed on ice if immediate processing was not available after sampling. The plasma was separated by centrifugation (1500 g, 10 min) within 1 h of venepuncture. Aliquoted plasma was stored at −70°C until analysis. All samples were immediately subjected to assays for total homocysteine (tHcy), folate and vitamin B12 after thawing at 37°C. The measurement of tHcy concentration was based on fluorescence polarization immunoassay on an Abbott IMX analyser (Abbott Laboratories, TX, USA). Folate and vitamin B12 levels were assayed on an Abbott AxSYM System based on microparticle enzyme immunoassay (Abbott Laboratories). These tests were performed according to instructions recommended by the manufacturer.
Statistical analysis. Hyperhomocysteinaemia was arbitrarily defined as a value above the 95th percentile of healthy subjects. First, differences between patients and healthy subjects for age, plasma levels of tHcy, folate and vitamin B12 (Student's t-test), sex and prevalence of hyperhomocysteinaemia (chi-square test) were examined. Second, the associations between plasma tHcy levels and the potential variables, age, sex, folate and vitamin B12 levels, were examined using univariate and multivariate regression analyses in the healthy group. Third, sex difference in plasma tHcy levels as well as vitamin status was evaluated in all healthy subjects and subjects consisting of age-matched (± 2 years) men and women using the Mann–Whitney rank-sum test and multivariate regression analysis respectively. Finally, both the thrombotic risk of hyperhomocysteinaemia and the association of plasma tHcy level with venous thrombosis were examined in a case–control study enrolling patients and healthy subjects by multivariate regression analysis in which the odds ratio (OR) and 95% confidence interval (CI) were calculated. The association of hyperhomocysteinaemia with recurrence of episode was also examined in the patient group by multivariate regression analysis.
Demographic data of study population
Eighty patients (41 men and 39 women) and 123 healthy subjects (71 men and 52 women) were recruited. Mean age ± SD (range) in patients and healthy subjects was 50·8 ± 17·8 (16–85) years and 49·0 ± 17·4 (15–85) years respectively. There was no significant difference in mean age or sex between patients and healthy subjects (P=0·477 and 0·466 respectively). All patients were regularly monitored for anticoagulation therapy in outpatient clinics and none of them were prescribed any vitamin supplements before the study. Mean age of patients at the first thrombotic event was 44·4 ± 17·2 (14–80) years. A thrombophilic cause, as defined in Materials and methods, could be found in 51 patients (63·8%), i.e. deficiencies of antithrombin III, protein C or protein S (n=45), presence of lupus anticoagulant or anticardiolipin antibody (n=4), plasminogen deficiency (n=1) and hypodysfibrinogenaemia (n=1). Thirty (37·5%) of all patients had recurrent episodes. None of the episodes in 39 female patients was related to oral contraceptive use.
Plasma levels of total homocysteine, folate and vitamin B12
Mean fasting plasma concentrations of tHcy, folate and vitamin B12 in patients and healthy subjects are listed in Table I. The mean values of plasma tHcy, folate and vitamin B12 levels were not statistically different between these two groups (P=0·129, 0·072 and 0·096 respectively). The prevalence of hyperhomocysteinaemia (> 15·5 µmol/l) in patients (8·9%) was not statistically different (P=0·397) from that found in healthy individuals (4·8%).
Table I. Fasting plasma levels of total homocysteine (tHcy), folate and vitamin B12 in 80 patients with venous thrombophilia and 123 healthy subjects.
Influence of age, sex and vitamin status on homocysteinaemia
Univariate regression analyses in healthy subjects revealed that plasma tHcy levels tended to increase with age (P=0·001) and with decreasing plasma levels of folate (P=0·001) or vitamin B12 (P=0·007); men had higher plasma tHcy levels than women (P < 0·001). These significant associations were consistently present in a multivariate regression analysis (Table II). Even when log-transformed plasma levels of folate, vitamin B12 and tHcy were analysed, these relationships did not change (data not shown). Sex difference in plasma levels of tHcy, folate and vitamin B12 are listed in Table III, which showed that men had a higher mean tHcy concentration (P < 0·001) but a lower mean folate level (P=0·040) than women, although the mean age was also higher (P=0·014) in men than in women. In addition, median homocysteine levels related to age were consistently higher in men than in women in all age subgroups (Fig 1). There was no sex difference in mean vitamin B12 level (P=0·392). Significantly higher plasma tHcy levels (P=0·018) in men than in women remained after adjusting for plasma levels of folate and vitamin B12 when age-matched (± 2 years) men (n=41) and women (n=41) were evaluated together by multivariate analysis.
Table II. Univariate and multivariate determinations* of fasting plasma level of total homocysteine in 123 healthy subjects.
Linear regression analysis.
Vitamin B12 (ng/l)
Table III. Sex difference in fasting plasma levels of total homocysteine (tHcy) and vitamin status in 123 healthy subjects.
Association of homocysteinaemia with venous thrombosis
A logistic regression analysis (enter mode) on 80 patients and 123 healthy subjects did not show a significant association of plasma tHcy level (OR, 1·07; 95%CI, 0·96–1·18; P=0·210) or hyperhomocysteinaemia (OR, 1·65; 95%CI, 0·50–5·49; P=0·415) with occurrence of venous thrombosis. A further analysis using conditional regression model on 74 patients and 74 age- (± 2 years) and sex-matched healthy subjects showed an insignificant association between venous thrombosis and plasma tHcy level (OR, 1·13; 95% CI, 0·98–1·24; P=0·116) or hyperhomocysteinaemia (OR, 0·51; 95% CI, 0·14–1·91; P=0·321). A multivariate analysis in the patient group (n=80) also failed to demonstrate an association of hyperhomocysteinaemia with recurrence of episode (P=0·560) for venous thrombosis after adjusting for age, sex, plasma levels of folate and vitamin B12.
The patients seen at National Taiwan University Hospital come from all locations nationwide but are more likely to be from the northern part of the country. The prevalence of hyperhomocysteinaemia in this study (patients, 8·9%; healthy subjects, 4·8%) seemed similar to those found in Caucasians (D'Angelo & Selhub, 1997). Our study revealed that plasma tHcy concentration tended to increase with age and with decreasing plasma levels of folate or vitamin B12; men tended to have higher plasma tHcy level than women (Table II). Therefore, age, sex and vitamin status were determining factors for plasma tHcy level in the Chinese. These results were consistent with those found in Caucasians (Selhub et al, 1993; Jacques et al, 1999; Morris et al, 2000). Our findings were in contrast to those found in a pilot study on Hong Kong Chinese, which demonstrated that plasma tHcy concentration was independent of a deficiency of folate and/or vitamin B12 (Lolin et al, 1996). The current data was also quite different from a recent study in Taiwanese Chinese subjects, which demonstrated that both sex and plasma folate did not affect plasma tHcy using both univariate and multivariate analyses (Ho, 2000). The possible reason for these discrepant findings in contrast to those found in Caucasians was not explained. Higher plasma tHcy concentrations in the elderly may be caused by insufficient intake of vitamins in this population (Selhub et al, 1993). This study had demonstrated that men had higher tHcy levels than women in all age subgroups. Morris et al (2000) suggested that higher oestrogen status was associated with a decreased mean serum tHcy concentration, independent of nutritional status and muscle mass, and that oestrogen might explain the male–female difference in tHcy concentration. Our very preliminary evaluation on a small number of Taiwanese Chinese consisting of age-matched men and women suggested that a lower mean folate level in men might play a part, in addition to hormone effect, in the sex difference in plasma tHcy level. However, this premature finding needs to be clarified before drawing any conclusion.
The prevalence of hyperhomocysteinaemia (> 15·5 µmol/l) in controls (4·8%) was not significantly different (P=0·397) from that in patients (8·9%). The cut-off value for defining hyperhomocysteinaemia was not similar among different studies and was dependent on both sampling and assay procedures for plasma tHcy measurement. This case–control study on age- and sex-matched subjects neither supported hyperhomocysteinaemia as an important risk factor for venous thrombophilia nor showed a positive association of hyperhomocysteinaemia with recurrence of venous thrombosis in Taiwanese Chinese. However, the number of individuals studied was too low to allow a conclusion to be drawn. These results seemed similar to our previous findings that MTHFR C677T homozygous mutation was not a significant risk factor for Chinese venous thrombophilia (Lin et al, 2000; Shen et al, 2000).
In summary, age, sex and vitamin status are important determinants of plasma homocysteine levels but the mechanism for the sex difference in homocysteinaemia is unknown in Taiwanese Chinese. Finally, hyperhomocysteinaemia is possibly not an important risk factor for venous thrombophilia in Taiwanese Chinese.
This study was granted by National Taiwan University, College of Medicine (1999). We are grateful for technical help from our colleagues (Biochemistry Laboratory, National Taiwan University Hospital) and Mr Abel Chien (Abbott Laboratories, Taiwan).