Hyperhomocyst(e)inemia and a Common Methylenetetrahydrofolate Reductase Mutation (Ala223Val MTHFR) in Patients With Inherited Thrombophilic Coagulation Defects
Abstract To assess whether certain abnormalities of the sulfated amino acid metabolism are associated with the occurrence of thromboembolic events in patients with inherited thrombophilic conditions, the levels of homocyst(e)ine, before or after methionine load, and the presence of the Ala223Val substitution in the 5,10-methylenetetrahydrofolate reductase (MTHFR) were evaluated in 119 subjects with a congenital single thrombophilic condition (type I deficiency of antithrombin n=10, protein C n=24, protein S n=16; activated protein C resistance due to factor V Leiden mutation n=69). Sixty-three subjects had experienced at least one documented thrombotic event, while the remaining 56 subjects were still free from any thrombotic symptom. Our results show that (1) high homocyst(e)ine levels, either in fasting condition or after methionine load, were not more frequent in subjects with inherited thrombophilic alterations (14.4%) than in normal control subjects (10% by definition) and (2) the frequency of hyperhomocyst(e)inemia was similar in thrombophilic subjects, who already have (14.3%) or have not (14.6%) experienced thrombotic events. As regards the MTHFR mutation, the homozygous condition was present in 23.2% of the thrombophilic patients versus 17.5% in the control subjects, a nonsignificant difference. The mutation was slightly more frequent in those thrombophilic subjects who had suffered a thrombotic episode (25.5%) versus those with no thrombosis (20.8%), with odds ratios of 1.61 (confidence interval (CI)=0.58-4.52) and 1.24 (CI=0.42-3.43), respectively. These differences were also nonsignificant. It is concluded that in subjects with inherited thrombophilias, a condition of hyperhomocyst(e)inemia “per se” is not a factor increasing the risk of thrombosis. The risk enhancement conferred by the MTHFR mutation, if any, seems to be slight or limited, and its significance could be ascertained only in a large multicenter trial.
- Received December 10, 1996.
- Accepted March 18, 1997.
It is well established that inherited blood-clotting alterations impairing physiological anticoagulant activity−ATIII, PC and PS deficiencies, and presence of APCR (caused by the known factor V Leiden mutation)−are associated with thrombophilic states.1 However, the great clinical variability occurring in affected patients still remains unclear. It has been estimated that only about 50% of patients become symptomatic for venous/arterial thrombotic disease.1 The association of multiple deficiencies or the presence of other congenital/acquired risk factors is currently being widely investigated to explain why thrombotic events seem to affect only some of these patients.2 3 4 5 6 7 8
A number of clinical studies (reviewed in Reference 99 ) have indicated that hyperhomocyst(e)inemia is a risk factor for premature vascular disease (both arterial and venous). In particular, a condition of mild/moderate hyperhomocyst(e)inemia has recently been shown to have a high prevalence in patients with juvenile venous thrombosis with no other thrombophilic conditions10 11 12 13 14 and to be a risk factor for recurrence of venous thrombosis.15 Various congenital or acquired alterations of homocyst(e)ine transsulfuration or remethylation pathways may lead to elevated homocyst(e)ine levels. Moreover, a common genetic mutation (Ala223Val) of MTHFR has recently been described.16 This enzyme regulates the remethylation pathway of homocyst(e)ine, leading to thermolability and reduced activity of the enzyme with subsequent higher homocyst(e)ine levels.
The aims of the present study were (1) to investigate the prevalence of hyperhomocyst(e)inemia, before or after methionine load, and of the above-mentioned MTHFR mutation in patients with ascertained thrombophilic conditions (ATIII, PC and PS deficiency, or APCR) and (2) to assess whether or not a condition of hyperhomocyst(e)inemia is associated with the history of thromboembolic events in these patients.
As detailed in Table 1⇓, the study investigated 119 subjects (44 males), belonging to 64 families, with confirmed (at least two subjects with the same alteration in a family) congenital single thrombophilic condition, ie, type I deficiency of ATIII (10 cases), PC (24 cases), PS (16 cases), or presence of APCR due to factor V Leiden mutation (69 cases). Of these subjects, 63 (56 propositi) had experienced at least one documented thrombotic event, mainly involving the venous system: deep vein thrombosis and/or pulmonary embolism in 56 cases; repeated episodes of superficial thrombophlebitis in 6 cases. Involvement of the arterial vascular system was reported in only 1 patient. At the time of the present study, the remaining 56 subjects, though affected by a congenital thrombophilic abnormality, were still free from any thrombotic symptoms. In these latter subjects, diagnosis of a thrombophilic defect had usually been performed within 3 to 4 months of that carried out in the relevant propositus. There was no difference in age (mean and range) between those who had experienced thrombotic episodes versus the others (see Table 1⇓).
In 38 of 63 (60.3%) subjects who had thrombosis, the following risk/trigger factors were found to be associated with occurrence of the first event: use of oral contraceptives (n=14), surgery or prolonged bed rest,10 trauma and/or plasters,8 pregnancy or puerperium,5 hematologic disease1 ; no known predisposing disease or risk/trigger factors could be detected in the remaining 25 persons. The median (range) age of occurrence of the first thrombosis was 30 years (15 to 68). More than one thrombotic event was present in the history of 28 (44.4%) subjects. Other cases of thrombosis were detected among the family members of 25 (44.6%) of the 56 propositi investigated.
The subjects with thrombotic events were examined for the purposes of the present study at least 3 months after their last/only thrombotic event. At the time of the present study, oral anticoagulants were given to 16 of them. Subjects with multiple congenital alterations or with other rarer causes of congenital thrombophilia were excluded from the study, as were subjects with lupus anticoagulant, high anticardiolipin antibody levels, abnormal liver or renal function, or evidence of neoplastic or autoimmune diseases.
Fasting and postmethionine load homocyst(e)ine levels were measured in 83 (28 males) of these 119 subjects: 42 (ATIII deficiency n=6; PC n=10; PS n=10; factor V Leiden mutation n=16) had a personal history of thrombosis, while the remaining 41 (ATIII deficiency n=4; PC n=14; PS n=6; factor V Leiden mutation n=17) were still asymptomatic when tested. A total of 34 subjects (19 males, age 32.8 ± 10.9 y) with no overt clinical manifestations were examined as control group for fasting and post methionine load homocyst(e)ine measurements.
The presence of the genetic mutation (Ala223Val) of the MTHFR was evaluated in 99 (51 males) among the 119 subjects; a previous thrombotic event was present in 51 subjects (ATIII deficiency n=4; PC n=10; PS n=7; factor V Leiden mutation n=30), while 48 (ATIII deficiency n=4; PC n=12; PS n=6; factor V Leiden mutation n=26) were still free from thrombosis. As control group, 40 apparently healthy subjects were investigated for the presence of this mutation.
All individuals examined as control subjects, either for homocyst(e)ine level or MTHFR mutation determination, were not investigated for the possible presence of congenital thrombophilic alterations.
This study was approved by the ethical committee of our institution, and informed consent to participate in the study was obtained from all the examined subjects.
Blood sampling and methionine loading test. Venous blood samples were taken between 7:30 and 9 am (after overnight fast) from subjects in supine position for at least 10 minutes, to measure plasma homocyst(e)ine levels and serum levels of vitamin B12 and folate. l-Methionine (Sigma) at a dose of 100 mg/kg body weight was then administered orally diluted in about 200 mL of fruit juice. A light and standardized breakfast consisting of coffee or tea and/or fruit juice and aproteic biscuits was then allowed. Four hours after methionine loading, a second blood sample was drawn for plasma homocyst(e)ine measurement.
Sample preparation and analysis. Blood samples for plasma homocyst(e)ine determination were collected in tubes containing 4.4 mmol/L EDTA 2K+, immediately placed on ice in the dark, and plasma was separated within 1 hour. Plasma was stored at −80°C until analysis. Blood samples for serum folate and vitamin B12 measurement were collected in empty tubes and sent to the biochemical laboratory for routine radioimmunoassay analysis (Ciba-Corning Diagnostic Corporation).
Total plasma homocyst(e)ine levels were determined by HPLC according to a modification of the Araki and Sako method,17 which entails complete reduction of homocystine and the mixed disulfide [cysteine-homocyst(e)ine] and the release of protein-bound homocyst(e)ine. The method thus measures total (free + protein-bound) plasma homocyst(e)ine concentrations.
The original method for chromatographic separation17 was modified by applying an ion-pair reversed-phase HPLC. In brief, the procedure was the following: 10 μL of a 10% solution of tri-n-butylphosphine (Sigma) in dimethylformamide (Carlo Erba) was added to 100 μL of plasma sample or homocystine standard solution. After 30 minutes’ incubation at 4°C, 100 μL of a 0.6 mol/L solution of trichloroacetic acid in 1 mmol/L EDTA 2Na+ was added and mixed, kept 10 minutes at room temperature, and centrifuged 15 minutes at 3000g. A 75-μL aliquot of clear supernatant was vigorously mixed with 150 μL of 2 mol/L borate buffer in 4 mmol/L EDTA 2Na+, pH 10.5, and derivatized with 75 μL of 7-fluorobenzo-2-oxa-1,3-diazole-4-sulfonate (1 mg/mL, Sigma) in 2.0 mol/L borate buffer in 4 mmol/L EDTA 2Na+, pH 9.5 (freshly prepared). After 60 minutes’ incubation in 60°C water bath, aliquots were cooled to room temperature; 10-μL aliquots were then injected in the HPLC. Separation was carried out at room temperature, at a flow rate of 1.2 mL/min, using an analytical column (Bakerbond C18, 3-μm particle size) protected by an inlet filter model 7315, Rheodyne. The eluted peaks were monitored by a fluorometric detector set.
Standard calibration solutions at different concentrations of homocystine and cystine were obtained by further diluting a stock solution containing homocystine (1 μmol/L) and cystine (0.5 μmol/L) prepared in 0.01 mol/L HCl.
The intraday (interday) coefficient of variation of the analytic method was 1.5% (4.0%), 1.4% (2.9%), and 1.3% (3.9%) for plasma samples containing 6.5, 15.5, and 32.6 μmol/L of homocyst(e)ine, respectively.
Criteria for diagnosis of hyperhomocyst(e)inemia. Both fasting plasma homocyst(e)ine levels and postmethionine-load absolute increment of homocyst(e)ine were analyzed to identify subjects with hyperhomocyst(e)inemia. The following values, corresponding to the 90th percentile of the value distribution in the control subjects, were used as cutoff levels: 17 μmol/L and 29 μmol/L for fasting and postmethionine-load absolute increment homocyst(e)ine levels, respectively.
MTHFR Gene Mutation
The Ala223Val substitution, created by a C-to-T transition at nucleotide 677 and producing an additional HinfI site, was detected by restriction analysis of PCR products. Primers for amplification, described by Frosst et al,16 were 5′TGAAGGAGAAGGTGTCTGCGGGA3′ and 5′AGGACGGTGCGGTGAGAGTG3′. Thirty PCR amplification cycles were run as follows: 20 seconds of denaturation at 94°C, 20 seconds of annealing at 62°C, and 20 seconds of extension at 72°C.
Laboratory Thrombophilia Diagnosis
Diagnosis of type I ATIII, PC, or PS deficiencies had been established using conventional functional and immunological tests. Diagnosis of APCR condition was made by using a clotting method18 ; in all cases the presence of factor V Leiden mutation (in heterozygous condition) was confirmed by DNA analysis.19 The presence of the same mutation was excluded in the other subjects.
Differences between groups were assessed by ANOVA and also by the χ2 test or by the two-sample proportion test as appropriate. ORs using the approximation of Woolf, and 95% CIs were also calculated. The SOLO statistical software package (version 4.0, BMDP) was used for data processing.
As shown in Table 2⇓, total plasma homocyst(e)ine levels in fasting state and their absolute increments after methionine load were not different in the whole group of subjects with inherited thrombophilic condition versus the control group, as well as between thrombophilic subjects who had or had not previously experienced thrombotic events. Furthermore, no differences were recorded as regards folic acid and vitamin B12 levels between the different groups investigated. Abnormally high homocyst(e)ine levels, either fasting or after methionine load (in no subject were both alterations present contemporaneously), were recorded in 12 (14.4%) of the 83 subjects examined, with no difference versus the control group (10% by definition) (OR=1.26, 95% CI=0.38-4.25, NS). As shown in the Figure⇓, hyperhomocyst(e)inemia was present in 6 of 42 (14.3%, 5 propositi, 4 females) and 6 of 41 (14.6%, 6 females) subjects who had or had not experienced thrombotic events, respectively; the ORs were 1.25 (95% CI=0.32-4.84, NS) and 1.29 (95% CI 0.33-4.99, NS), respectively.
Table 3⇓ reports the number (and percent) of cases with hyperhomocyst(e)inemia in subjects according to the congenital thrombophilic alterations. The frequency of hyperhomocyst(e)inemia was higher (though not significantly) in subjects carrying factor V Leiden mutation. In each thrombophilic defect group, the distribution of hyperhomocyst(e)inemia was similar among the subjects who had or had not suffered from thrombotic events.
Clinical Details on the Subjects With Hyperhomocyst(e)inemia
The median age (31.5 years) of onset of the first thrombotic event in the six cases with hyperhomocyst(e)inemia was no different from that of all the cases with thrombosis. In four of these six cases, risk/trigger factors were present at onset of the event (two plasters, one surgery and one pill); one case had a transient (lasting less than 1 year) lupus anticoagulant phenomenon with high IgG anticardiolipin antibody level. Only one subject with hyperhomocyst(e)inemia had recurrent thrombotic events (versus 22 of 42 in the whole group with thrombosis). Other cases of thrombosis in the family were present in three of the five propositi with hyperhomocyst(e)inemia (frequency not different from that of the whole thrombotic group).
MTHFR Gene Mutation
The prevalence of homozygous mutation was not significantly different in thrombophilic and control subjects (see Table 4⇓) (OR=1.43, 95% CI=0.56-3.65, NS). Among thrombophilic subjects, the prevalence of the mutation was higher in the subjects who had experienced thrombotic events (25.5%) than in those who had not (20.8%), but this difference was not significant; the ORs were 1.61 (95% CI=0.58-4.52, NS) and 1.24 (95% CI=0.42-3.63, NS), respectively.
The number (and percent) of cases with homozygous MTHFR mutation according to the congenital thrombophilic alterations are reported in Table 5⇓. The frequency of MTHFR gene mutation in homozygous condition was higher (though not significantly) in subjects with PS deficiency (30.8%) and in those carrying factor V Leiden mutation (25.0%). In each thrombophilic defect group, the distribution of homozygous MTHFR gene mutation was not different among the subjects who had or had not suffered from thrombotic events.
The presence of homozygous MTHFR mutation in the 63 thrombophilic subjects whose homocyst(e)ine levels were measured was associated with significantly higher fasting homocyst(e)ine levels, while postmethionine-load levels were not affected by the mutation (Table 6⇓).
Clinical Details on the Subjects With Ala223Val MTHFR Gene Mutation
The median age of the first thrombotic event onset in the 13 cases with homozygous Ala223Val MTHFR gene mutation was no different (39.0 years) from that of the heterozygous (n=19, 27.0 years) and normal (n=19, 32 years) cases with thrombosis. Risk/trigger factors were present at onset of thrombotic event in 62.5%, 63.1%, and 47.4% of homozygous, heterozygous, and normal cases, respectively. Four of 13 (30.8%) homozygous subjects had recurrent thrombotic events (versus 57.9% and 47.4% of heterozygous and normal cases). Thrombotic events in other family members were recorded in 6 of 13 homozygous cases (46.1%), a rate that did not differ significantly from that observed in heterozygous (52.6%) and normal (36.8%) cases.
In the last decade, a number of clinical studies have proved that there is a relationship between congenital or acquired conditions of mild hyperhomocyst(e)inemia and early onset of arterial thrombotic disease (for a review see Reference 99 ). Only more recently has the relationship with occurrence of venous thromboembolism also been investigated. High prevalence of moderate hyperhomocyst(e)inemia in patients with juvenile venous thrombosis has been found in some10 11 12 13 14 though not all studies.20 21 Hyperhomocyst(e)inemia has been shown to be a risk factor for first venous thrombotic episode in patients of any age14 and has also been postulated to be a risk factor for recurrence of thrombotic events.15
The present study examined subjects with one of the ascertained conditions of inherited thrombophilia to detect a possible association with hyperhomocyst(e)inemia. Our results show that (1) high plasma homocyst(e)ine levels, either in fasting condition or after methionine load, are no more frequent in these subjects than in normal control subjects and (2) the frequency of hyperhomocyst(e)inemia is similar in thrombophilic subjects who had or had not experienced thrombotic events. Among the different thrombophilic alterations, a condition of hyperhomocyst(e)inemia was more frequent, though not significantly, in subjects with APCR due to factor V Leiden mutation. No differences in the age of onset of the first thrombotic event, presence of risk/trigger factors, frequency of recurrences, and rate of thrombotic events in other family members could be detected between all the thrombophilic subjects with thrombosis and the six who also had hyperhomocyst(e)inemia.
A common mutation (Ala223Val) in MTHFR, which results in thermolability of the enzyme, reduction in its activity, and subsequent higher plasma homocyst(e)ine levels, has recently been identified and suggested as a candidate genetic risk factor for vascular disease.16 In our study, the presence of the mutation was determined by HinfI restriction analysis in thrombophilic subjects and in a control group. In the latter, the recorded frequency of genotypes carrying the Ala223Val substitution was higher than that reported in other populations.16 22 This feature has been confirmed in additional studies of normal Italian subjects (unpublished data, G. Marchetti, 1997). Unlike the results reported by Frosst et al,16 though in line with findings of other authors,23 our study detected increased plasma homocyst(e)ine levels associated with homozygous MTHFR mutation only in fasting condition and not after methionine load.
In our series of patients with classical thrombophilic defects, the probability that the presence of the homozygous mutant MTHFR genotype might be associated with clinical thrombosis (OR=1.61) was only slightly higher than that of no such association (OR=1.24), and neither of the two ORs was statistically significant. These data do not allow the conclusion, but do not exclude completely, that the presence of the mutant thermolabile MTHFR enzyme is a factor enhancing the risk of thrombosis in subjects with inherited thrombophilic defects. In fact, since the relative thrombotic risk conferred by a condition of hyperhomocyst(e)inemia seems to be rather small, as reported in a previous study14 and the MTHFR mutation is a cause of mild hyperhomocyst(e)inemia, it cannot be excluded that with a larger cohort of patients in a multicenter trial, the relevant OR may reach significance. Interestingly, as found for hyperhomocyst(e)inemia, the MTHFR mutation was detected more frequently, though not significantly, in subjects with factor V Leiden mutation (and also in those with PS deficiency). These results may suggest a positive association between hyperhomocyst(e)inemia and factor V Leiden mutation.
In conclusion, our data show that blood levels of hyperhomocyst(e)ine, either basal or after methionine load, do not enhance the risk of thrombosis in inherited thrombophilias and suggest that the risk conferred to these patients by the MTHFR mutation, if any, is probably slight, although statistical significance should be assessed in a larger population.
Selected Abbreviations and Acronyms
|APCR||=||resistance to activated protein C|
This work was supported by funds from CNR (1995) and the Emilia-Romagna Region to Prof Coccheri and grant NE 125 from Telethon Italy to Prof Bernardi.
De Stefano V, Finazzi G, Mannucci PM. Inherited thrombophilia: pathogenesis, clinical syndromes, and management. Blood. 1996;87:3531-3544.
Koeleman BPC, Reitsma PH, Allaart CF, Bertina RM. Activated protein C resistance as an additional risk factor for thrombosis in protein C-deficient families. Blood. 1994;84:1031-1035.
Martinelli I, Magatelli R, Cattaneo M, Mannucci PM. Prevalence of mutant factor V in Italian patients with hereditary deficiencies of antithrombin, protein C or protein S. Thromb Haemost. 1996;75:694-695.
Palareti G, Legnani C, Coccheri S. Hyperhomocysteinemia and vascular disease. In: Seghatchian MJ, Samama MM, Hecker SP, eds. Hypercoagulable States: Fundamental Aspects, Acquired Disorders, and Congenital Thrombophilia. Boca Raton, Fla: CRC Press, Inc; 1996:395-407.
Bienvenu T, Ankri A, Chadefaux B, Kamoun P. Dosage de l’homocysteine plasmatique dans l’exploration des thromboses du sujet jeune. Presse Med. 1991;20:985-988.
Falcon CR, Cattaneo M, Panzeri D, Martinelli I, Mannucci PM. High prevalence of hyperhomocyst(e)inemia in patients with juvenile venous thrombosis. Arterioscler Thromb. 1994;14:1080-1083.
Amundsen T, Ueland PM, Waage A. Plasma homocysteine levels in patients with deep venous thrombosis. Arterioscler Thromb Vasc Biol. 1995;15:1321-1323.
Kluijtmans LAJ, Vandenheuvel LPWJ, Boers GHJ, Frosst P, Stevens EMB, Vanoost BA, den Heijer M, Trijbels FJM, Rozen R, Blom HJ. Molecular genetic analysis in mild hyperhomocysteinemia: a common mutation in the methylenetetrahydrofolate reductase gene is a genetic risk factor for cardiovascular disease. Am J Hum Genet. 1996;58:35-41.
Jacques PF, Bostom AG, Williams RR, Ellison RC, Eckfeldt JH, Rosenberg IH, Selhub J, Rozen R. Relation between folate status, a common mutation in methylenetetrahydrofolate reductase, and plasma homocysteine concentrations. Circulation. 1996;93:7-9.