UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)
Familial hypercholesterolemia. The determination of phenotype
Jansen, A.C.M.
Publication date
2003
Link to publication
Citation for published version (APA):
Jansen, A. C. M. (2003). Familial hypercholesterolemia. The determination of phenotype.
General rights
It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s)
and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open
content license (like Creative Commons).
Disclaimer/Complaints regulations
If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please
let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material
inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter
to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You
will be contacted as soon as possible.
Thee risk of hyperhomocysteinemia
inn patients with familial
hypercholesterolemia a
vngeliquee CM Jansen
1, Pernette RW de Sauvage Nolting
1, Mieke D Trip
1,
iekee J Smilde
2, Marianne E Wittekoek
1, Joep C Defesche
1, Erik de Groot
1,
Aeilkoo H Zwinderman
3, Anton FH Stalenhoef
2and John JP Kastelein
1'Departmentt of Vascular Medicine and 3 Department of Clinical Epidemiology and Biostatistics,, Academic Medical Center, University of Amsterdam, department of
Internall Medicine, University Medical Center Nijmegen, the Netherlands.
Acceptedd for publication by Atherosclerosis
edd with Young Investigators Award, DALM congress, New York,
September,, 2001
Abstract t
Inn order to identify patients with Familial Hypercholesterolemia (FH) at the highest risk, continuouss efforts are being made to detect cardiovascular risk factors that influence the occurrencee of cardiovascular disease (CVD) in these patients. In this regard it was our objectivee to determine the relationship between hyperhomocysteinemiar the MTHFR
TT-genotypee (a genetic cause of life-long hyperhomocysteinennia) and the development of atherosclerosis.. In a cross-sectional design, we included 981 FH patients. FH patients with CVDD (36.5%) had increased median plasma homocysteine levels compared to disease-free patientss (12.6 vs, 11.9 mmol/L; p=0.01). The accompanying odds ratio was significant: 1.711 (95% Ci 1.11 - 2 64), and suggested that hyperhomocysteinemia could indeed be an independentt predictor of atherosclerotic cardiovascular disease in FH. However, whereas thosee patients with the MTHFR TT-genotype showed significantly increased homocysteine levelss (13.2 vs. 12.0 in CC-genotype and 12.3 mmol/L in CT-genotype; p<0.001), this was neitherr reflected in CVD rates (p=0.98), nor in intima media thickness (IMT) (p=0.47). Even moree remarkable was the fact that homocysteine levels were not related to IMT (r=0.064;p=0.1).. These latter findings weaken the concept of risk conferred by hyperhomocysteinemiaa in FH and support a reserved attitude towards aggressive diagnosis andd treatment of hyperhomocysteinemia in these patients.
Introduction n
Familiall hypercholesterolemia (FH), a common, dominant disorder characterized by severely elevatedd LDL-cholesterol levels, is caused by a plethora of mutations in the gene encoding the loww density lipoprotein (LDL) receptor.' As a consequence, many FH patients carry a high risk off premature ischemic events. However, a large variability has been noted in the onset and severityy of cardiovascular symptoms in these patients.2 In order to identify those patients at thee highest risk, an intensive search for cardiovascular risk factors that influence the occurrence e.. cardiovascular uisease y^vw/ m FH patients nas been earned uuL over the iast decades. Inn the general population, elevated homocysteine levels were known to be associated with cardiovascularr disease, but whether this relationship was causal was uncertain. In addition, studiess on the relationship between the methylenetetrahydrofolate reductase (MTHFR)-genotypee and the occurrence of CVD have yielded equivocal results. The C677T mutation in thee MTHFR gene, which renders the enzyme thermolabile and less active, is a well-known geneticc cause of hyperhomocysteinemia. Recently, three large meta-analyses studying the associationn between homocysteine levels and the MTHFR-genotype and the occurrence of ischemicc heart disease and stroke have strengthened the evidence that the relationship is indeedd causal.35 In the general population, due to the small risk conferred by elevated homocysteinee and the unfavourable MTHFR-genotype, these meta-analyses have demonstrated thee need to study very large numbers of cases and controls to provide conclusive evidence for ann association between homocysteine or the MTHFR-genotype and disease.
Wee hypothesized that the effect of elevated homocysteine levels and the MTHFR-genotype mightt be more pronounced in patients with Familial Hypercholesterolemia since the combinationn of increased homocysteine and veryy high LDL-cholesterol levels in these patients couldd potentially lead to remarkably increased odds ratios (OR) for vascular events. So far, onlyy a few authors focused on the association between homocysteine or MTHFR-genotype andd atherosclerotic disease in FH with conflicting outcomes.615 The contradicting data are likelyy due to limited number of patients involved in these studies (n=21-526).
Inn general, nor screening for the MTHFR-genotype, nor lowering homocysteine levels with folic-acidd based vitamin supplementation is recommended as of yet it is advised to await the resultss of large randomized controlled trials. Provided the risk of hyperhomocysteinemia is moree pronounced in patients with FH, vitamin therapy could already form an effective, inexpensivee and safe therapy to treat FH patients at the highest risk and MTHFR-genotyping mightt form an easy additional test to identify patients at increased risk.
Thee objective of our study was therefore to determine the relationship between hyperhomocysteinemia,, the MTHFR-genotype and the development of atherosclerosis in a large numberr of patients with FH. We studied these relationships with intima media thickness (IMT), aa well-known surrogate marker for atherosclerosis, as well as the occurrence rate of CVD.
Methods s
Studyy population
Patientss with Familial Hypercholesterolemia, referred to t w o university Lipid Clinics {Amsterdamm and Nijmegen) between 1996 and 1998, were retrospectively selected for inclusionn in this study. Inclusion criteria were a diagnosis of heterozygous FH based upon well-knownn criteria.1 Secondary causes of hypercholesterolemia were all excluded.
Baselinee medical data on clinical manifestations of CVD, laboratory parameters, IMT measurementss and risk factors were obtained from existing databases. Since many patients participatedd in clinical trials in which different measurements were performed (e.g. vitamin levelss and IMT measurements), the availability of data varies among the patients.
Alll data were verified for the current study by diligent checking of medical records and originall source documents.
Thee Institutional Review Boards approved the protocol and written informed consent was obtained.. All procedures were performed in accordance with the Declaration of Helsinki.
Clinicall manifestations of CVD
Patientss were classified as having clinical manifestations of CVD if the medical history included att least one of the following anomalies: coronary artery disease, cerebral arterial disease andd peripheral arterial disease.
CoronaryCoronary artery disease: myocardial infarction documented by characteristic ischemic pain
inn combination with (i) ECG abnormalities (new pathologic Q waves, or loss of R-wave voltage)) or (ii) documented release of cardiac markers; percutaneous transluminal coronary angioplasty;; coronary artery bypass grafting; anginal complaints with (i) myocardial ischemia documentedd by objective testing, i.e. positive exercise test, nuclear scintigram or dobutamine stresss ultrasound, or (ii) at least one > 50% diameter stenosis (coronary angiogram).
CerebralCerebral arterial disease: ischemic stroke documented by CT-scan.
PeripheralPeripheral arterial disease: intervention by either (i) balloon angioplasty or (ii) bypass surgery;
intermittentt claudication documented by an objective test, i.e. (i) ankle-arm index < 0.9, (ii) duplexx Doppler examination or (iii) > 50% stenosis on angiogram.
Laboratoryy parameters
Lipidd levels {8 weeks after discontinuation of lipid-lowering medication) and homocysteine levelss were measured after overnight fasting. The LDL cholesterol was calculated with the Friedewaldd formula.
Plasmaa homocysteine was measured by high-performance liquid chromotography. Folate wass measured with a commercially available RIA (Dualcount, Solid Phase No Boil Assay, DPC,, Los Angeles, CA, USA) (n=181), and with an IMX analyzer (Abbott) (n=78).
MTHFRR g e n o t y p i n g
Inn 806 patients genomic DNA was available that was extracted from peripheral blood leukocytess by standard procedures. All samples were genotyped in a single procedure. The PCRR fragments were digested with Hinf I restriction enzyme according to the instructions of thee manufacturer (New England Biolabs, Beverly,MA). The digests were analyzed by electrophoresiss in a 2% agarose gel in TBE-buffer. The MTHFR genotype frequencies were comparedd to those of Dutch controls, reported in an earlier manuscript.16
Non-invasivee measurements of atherosclerosis: Intima M e d i a Thickness
B-modee ultrasound imaged the near and far walls of the left and right common carotid, carotidd bulb and internal carotid arterial segments (n= 607). Acuson 128 XP™ (Mountain View,, CA, USA) and Biosound Phase-2™ (BiosoundEsaote, USA) ultrasound machines, equippedd with high resolution near field transducers, were used. Intima-media thickness wass measured off-line from digitally stored images. Sonographies and image analysts were nott aware of the clinical status of the patients.
Statisticall analysis
Alll data were analyzed in a cross-sectional design. Differences between CVD and disease-freee groups were analyzed by Student's t test for continuous variables and Chi-square test forr nominal variables. Logistic regression analysis was performed with demonstrated CVD ass dependent variable and total homocysteine as independent variable. Additional adjustment wass performed for potential confounders (Table 2). Differences between the three MTFHR-genotypess were analyzed with ANOVA for continuous variables and Chi-square test for nominall variables.
Too study the influence of folate status, we additionally performed this analysis in two subgroups off patients; those with folate levels below the 50Ih percentile and those with levels equal and above,, in line with literature.* Pearson's correlation was used to assess the univariate association betweenn different variables and common carotid IMT. To study the relationship between the variationn in common carotid IMT and homocysteine and the MTHFR-genotype, multiple regressionn analysis was carried out, with additional adjustment for age and gender (Table 5; Modell 1 and 2). In Model 3, the influence of age and gender, together with all variables from Tablee 4 were studied, except for the vitamins (too low number), and total cholesterol and triglyceridess (close relationship with LDL-cholesterol and HDL-cholesterol, respectively). Inn all statistical analyses, tests were carried out on log-transformed data in case of skewed distributedd variables (fasting total homocysteine, triglycerides and folate), while medians aree presented. Statistical analysis was performed using SPSS 10.1. All P values were 2-tailed (<0.055 statistically significant).
Results s
Fromm a total of 1110 FH patients, total homocysteine plasma levels were available in 981 patients.. The baseline characteristics of these 981 patients are summarized in Table 1; 358 patientss (36.5%) had clinical manifestations of CVD. CVD rates of both in- and excluded FH patientss were similar (36.5% vs. 42.1%; p=0.2), as well as the CVD rates between the two Lipidd Clinics (36.9% vs. 38.4%; p=0.7).
Tablee 1. Clinical characteristics of FH patients w i t h and w i t h o u t CVD N N Demographics s Malee gender, % Age,, years Riskk factors Currentt smoking, %
Systolicc blood pressure, mmHg Diastolicc blood pressure, mmHg Bodyy mass index, kg/m 2
Laboratoryy Parameters
;; Total homocysteine, micromol/L Totall cholesterol, m m o l / L LDL-cholesterol,, mmol/L HDL-cholesterol,, mmol/L Triglycerides,, mmol/L :: Folate, nmol/L Vitaminn B6, nmol/L Vitaminn B12, pmol/l CVD++ / CVD-3 5 8 / 6 2 CVD-3 CVD-3 346/617 7 3 4 5 / 5 8 6 6 2 0 9 / 4 4 0 0 2 0 8 / 4 4 0 0 353/619 9 358/623 3 3 4 7 / 6 1 5 5 3 4 6 / 6 1 2 2 347/613 3 347/615 5 9 8 / 1 6 1 1 7 8 / 1 3 1 1 9 9 / 1 6 1 1 Measurementt of atherosclerosis
Commonn carotid IMT, mm Nominall variables are expressed exceptt for total homocysteine, tr
192/415 5 ass percentages, CVDD + (36.5%) ) 51.4 4 52.99 9 29.0 0 1311 15 811 8 26.44 + 3.5 12.66 [ 6 . 5 - 4 2 . 4 ] 10.299 1 8.211 2.27 1.155 9 1.900 [ 0 . 3 0 - 8 . 2 0 ] 16.22 [ 5 . 0 - 52.9 ] 555 [ 6 - 596] 3066 [ 7 9 - 9 0 2 ] 0.944 1 continuouss variables C V D --(63.5%) ) 49.6 6 44.22 4 35.2 2 1266 14 799 9 25.33 6 11.99 [ 4 . 8 - 117.1] 10.099 7 8.055 2.02 1.211 3 1.655 [ 0 . 2 7 - 17.5] 17.77 [ 3 . 9 - 5 5 . 2 ] 555 [22 - 499] 2944 [103 - 9 1 5 ] 0.877 8 P P 0.6 6 << 0.001 0.055 I << 0.001 0.011 | << 0.001 0.01 1 0.2 2 0.3 3 0.004 4 << 0.001 0.4 4 0.6 6 0.2 2 << 0.001
aree expressed as mean 5.D., glycerides,, folate, vitamin B6 and vitamin B12 (median
;; CVD+ indicates cardiovascular disease present; CVD-, cardiovascular disease absent; LDL ir densityy lipoprotein; HDL, high density lipoprotein; IMT,, intima media thickness.
andd range). dicatess low
FHH patients with CVD were older, more often hypertensive, more often obese and less current smokerss than patients without CVD. Both groups contained equal proportions of males and females.. Patients with CVD had significantly higher median levels of homocysteine than disease-freee patients: 12.6 mmol/L versus 11.9 mmol/L (p = 0.01). LDL-and total cholesterol levels weree increased in all FH patients <n=981), these lipid levels did not significantly differ between CVDD and non-CVD patients. HDL-cholesterol was lower and triglyceride levels higher in patients withh CVD (HDL 1.15 mmol/L vs. 1.21 mmol/L; p=0.004, triglycerides 1.90 mmol/L vs. 1.65 mmol/L;; pO.001). Folate, vitamin B6 and vitamin B12 levels, available in subsets of patients
only,, were comparable in both groups. Data of mean carotid IMT measurements w e r e available inn 607 patients. Compared t o the disease-free g r o u p , IMT's were significantly larger in case of CVD;; 0.94 1 m m versus 0.87 8 m m (p < 0.001).
Too assess the risk of hyperhomocysteinemia associated w i t h CVD, odds ratios w e r e calculated inn a logistic regression model (Table 2) w i t h CVD as the dependent variable and homocysteine ass the independent variable. The calculated odds ratio associated w i t h hyperhomocysteinemia w a ss significant: 1.71 ( 9 5 % CI 1.11 - 2 64). A d j u s t m e n t f o r a g e , gender and other generally k n o w nn confounders; smoking, systolic blood pressure, BMI and total cholesterol, did not
Tablee 2. Odds ratios for CVD associated with hyperhomocysteinemia N N 981 1 963 3 916 6 627 7 OR R 1.71 1 1.65 5 1.63 3 1.97 7 95%% confidence interval l 1.111 -2.64 1.011 -2.67 1.00-2.67 7 1.10-3.54 4 P P 0.02 2 0.04 4 0.05 5 0.02 2 Totall homocysteine*
Adjustedd for age and gender
Adjustedd for age, gender and smoking
Adjustedd for age, gender, smoking, systoiic blood pressure,, body mass index and TC
** Due to a skewed distribution total homocysteine was entered into the model as natural logtransformed variable.. TC indicates total cholesterol.
Tablee 3. Total homocysteine, intima media thickness and event rate according to MTHFR-genotype MTHFR-genotype e
CCC CT TT
N== 382 (47%) N= 368 (46%) N- 56 (7%)
Totall homocysteine, micromol/L 12,0 12,3 13.2* <0.001 IMT,, mm 0,91 0.89 0.88 0.5 Cardiovascularr events, % 36.6 36.1 37.5 1.0
Totall homocysteine is expressed as median, IMT as mean and event rate as percentage. * Value significantly differentt from other two homocysteine values (Bonferroni). IMT indicates intima media thickness.
Inn 8 0 6 patients, MTHFR genotype frequencies w e r e evaluated. Table 3 shows the distribution off the different genotypes. The prevalence of genotypes was; C C 4 7 % , C T 4 6 % and T T 7 % ; f o rr Dutch controls (n=1250) the g e n o t y p e distribution was 4 9 , 42 and 9 % respectively.15 Thee frequencies in FH patients and controls w e r e similar (X2=0.048, d f = 2 , p= 0.9). The FH
heterozygotess carrying the MTHFR TT-genotype showed significantly higher median levels off homocysteine than carriers of the other t w o genotypes: 13.2 versus 12.0 versus 12.3 (p << 0.001). The frequencies of CVD (37.5 % vs. 3 6 . 6 % and 3 6 . 1 % ; p= 1.0), and IMT (0.88 vs. 0.911 and 0.89; p= 0.5) were similar for the genotypes.
Tablee 4. Correlates of intima media thickness (IMT) Variable e
SignificantlySignificantly associated:
Age e
Smoking g Bodyy mass index Systolicc blood pressure Totall cholesterol LDL<holesterol l HDl-cholesterol l Triglycerides s Non-significantiyNon-significantiy associa Malee gender Totall homocysteine Vitaminn B6 Vitaminn B12 Folate e ted: ted: r r 0.44 4 0.10 0 0.11 1 0.17 7 0.20 0 0.19 9 -0.084 4 0.12 2 0.035 5 0.064 4 0.014 4 0.11 1 -0.10 0 P P << 0.001 0.02 2 0.006 6 << 0.001 << 0.001 << 0.001 0.04 4 0.003 3 0.4 4 0.1 1 0.9 9 0.1 1 0.1 1 r=Pearson'ss correlation coefficient. LDL indicates low densityy lipoprotein, HDL high h densityy lipoprotein.
Inn a subgroup of patients with folate levels below the 50lh percentile median homocysteine significantlyy increased in the TT-genotype compared to the CC- and CT-genotype (13.5 vs. 12.99 vs. 12.1 mmol/L, p=0.02). Patients with folate levels of the 50lh percentile and higher showedd no difference in median homocysteine levels between the three genotypes (10.8 vs.. 10.5 vs. 10.6 mmol/L; p=0.9). In patients of both subgroups the frequency of CVD and IMTT were similar for the three genotypes.
Onn univariate analysis, Pearson correlation coefficients between all different risk factors and commonn carotid IMT were evaluated. The variables which positively correlated with IMT; age, smoking,, BMI, systolic blood pressure, total cholesterol, LDL-cholesterol and triglycerides are summarizedd in Table 4. HDL-C was negatively correlated with IMT. No correlation was found betweenn IMT and homocysteine, nor with folate, vitamin B6 or vitamin B12.
Tablee 5 shows the results of different multiple regression analyses in which common carotid IMTT was entered as outcome variable. First, the relationship between IMT and homocysteine (Modell 1) and MTHFR-genotype (Model 2) was studied, adjusted for age and gender. Neitherr homocysteine, nor MTHFR contributed significantly to the variation in IMT. In Model 33 all potential risk factors were combined. Male gender, age, current smoking, LDL-C and HDL-CC significantly influenced the variation of IMT; together they explained 25.7% of the variationn of IMT.
Tablee 5. Multiple regression analyses. Outcome variable: Modell 1 st.. Beta Totall homocysteine 0.06 Malee gender 0.13 Agee 0.47 MTHFR-genotypee (TT- vs. CC- and CT-- genotype) Currentt smoking LDL-cholesterol l HDL-cholesterol l Systolicc blood pressure Bodyy mass index
R22 0.217 P P 0.1 1 0.001 1 << 0.001 << 0.001
common n carotidd intima
Modell 2 st.. Beta 0.14 4 0.48 8 0.00 0 .218 8 P P << 0.001 << 0.001 1.0 0 << 0.001 mediall thickness Modell 3 st.. Beta 0.026 6 0.11 1 0.46 6 0.11 1 0.17 7 -0.098 8 0.004 4 0.95 5 .257 7
St.. Beta indicates standardized Beta; MTHFR, methylenetetrahydrofolate reductase; LDL, lipoprotein;; HDL, high density lipoprotein.
P P 0.5 5 0.004 4 << 0.001 0.003 3 << 0.001 0.01 1 0.9 9 0.3 3 << 0.001 loww density
Discussion n
Wee determined the relationship between hyperhomocysteinemia, MTHFR-genotype and thee development of atherosclerosis in a large number of FH patients. FH patients with establishedd CVD had higher homocysteine levels than those without CVD. The accompanying oddss ratios suggested that hyperhomocysteinemia could indeed be an independent predictor off atherosclerotic cardiovascular disease in these patients. However, whereas those with thee MTHFR TT-genotype, a cause life-long of hyperhomocysteinemia, exhibited significantly increasedd homocysteine levels, this was neither reflected in CVD rates nor in carotid IMT. Evenn more remarkable was the fact that homocysteine levels were not related to IMT measurementss in any way. These latter findings weaken the concept of risk conferred by hyperhomocysteinemiaa in FH and questions the role of elevated homocysteine levels in thesee patients.
Onlyy few studies have assessed the role of homocysteine in FH and unfortunately, no prospectivee studies are available. In cross-sectional approaches, and in line with our findings, threee studies found increased homocysteine levels in FH patients with established CVD7,9Ab,
whereass in one study no such increase was found.6 In addition, our negative results regardingg the relationship between homocysteine levels and IMT were supported by three previouss FH studies on this topic.81013 In contrast, in one study in FH children a significant relationshipp became evident.1'
studiedd FH children and found that carriers of the MTHFR TT- genotype had higher homocysteinee levels, and in addition, they had more often parents with CVD.12'14 Kawashiri ett al. studied the role of the MTHFR-genotype in 199 FH adults. In contrast to our findings, thee MTHFR-TT genotype was more often found in patients with CAD, although the difference wass only marginal and just reached statistical significance.7
Whenn our findings are considered in the light of these other studies several possible explanationss arise. Firstly, and in our view most importantly, homocysteine might not be a riskk factor for the development of atherosclerosis in FH patients. This is concordant with ourr findings that the MTHFR-genotype is not related to IMT or CVD rate, and the fact that we,, and most others, have found that homocysteine is not associated with IMT in FH. However,, we and others did find elevated homocysteine levels in patients with established CVD.. A possible explanation for this latter finding could be the fact that homocysteine levelss in these CVD patients were elevated as a consequence of the actual event, rather thann before. Indeed, studies have shown homocysteine to increase during the acute or convalescentt periods after CAD or stroke.1719 Another explanation might be the renal impairmentt that is often caused by atherosclerosis and which leads to increased homocysteine levels.. This might also explain the simultaneous occurrence of CVD and elevated homocysteine levels.200 Unfortunately, no data were collected regarding the renal function in our patients. Conversely,, hyperhomocysteinemia might indeed be a risk factor in FH patients, but oniy in certainn circumstances, such as in the presence of low folate levels. Recently, a large meta-analysiss confirmed that the association between the MTHFR genotype and CVD is mainly presentt in patients with a low folate status.4 Indeed, in patients with the MTHFR TT genotype, wee only found significantly increased homocysteine levels in case of a low folate status. The CVDD risk in these patients was not increased, but we must keep in mind that these analyses weree only performed in small subsets of the cohort and consequently might lack the required power.. This might also explain the differences between the study of Kawashiri et al. and ours.. Both found increased levels of homocysteine in the MTHFR TT group, but only in the Japanesee population this was translated into a higher CAD risk; unfortunately, no vitamin levelss were available in this cohort.
Ourr study differs from others by the large number of patients included. By collecting data off FH patients from two large Lipid Clinics, we are the first to present data on the role of homocysteinee and the MTFHR-genotype in such large numbers of FH patients. We feel our resultss are supported by earlier reports; we did find increased homocysteine levels in patients withh CAD and patients with the MTHFR-TT genotype. Furthermore, we demonstrated several riskk factors to be associated with CVD and IMT, in line with earlier studies.6'91013
thee relation between the MTHFR genotype, homocysteine levels and CVD as outcome parameter,, while taking potential confounders into account, would have been preferable. Furthermore,, although only a minority of patients was on lipid-lowering drugs at the moment off data collection, we must take into account that the use of these drugs might have lessenedd the impact of the association between homocysteine and atherosclerosis. Wee started this study with the hypothesis that elevated homocysteine levels would have a pronouncedd untoward effect in patients with FH, who, due to their severely increased LDL-cholesteroll levels, develop atherosclerosis and CVD more progressively. However, we rnuld nott demonstrate a significant relationship between the MTHFR-TT genotype, IMT and CVD rate,, although these patients had significantly increased homocysteine levels. Lastly, homocysteinee levels were not associated with IMT. These results support a reserved attitude towardss aggressive diagnosis and treatment of hyperhomocysteinemia in patients with FH. Wee think that in FH, as well as in the general population, the relationship between homocysteine andd CVD can only definitively be assessed in large numbers of patients, unlikely to be collected byy one study group. Different studies should be ascertained in a meta-analysis, just as has beenn done recently in the general population for homocysteine studies.
Inn clinical practice, routine measurement of homocysteine and the MTHFR-genotype is premature,, since the consequences are not clear. While awaiting the results of the large endd point trials, individual physicians might want to consider treating FH patients with severe hyperhomocysteinemiaa with folic acid. In contrast, widespread use of folic acid should not bee advocated for FH patients, since the role of hyperhomocysteinemia per se and the treatmentt thereof in these patients has not been established to date.
Acknowledgments s
Wee thank Jorge Peter for his excellent technical assistance performing the MTHFR-genotyping.. John J.P. Kastelein is an established investigator of the Netherlands Heart Foundationn (grant D039/66510).
References s
1.. Goldstein JL, Hobbs HH, Brown MS. Familial hypercholesterolemia.. In: Scriver, C. R. et al., eds.. The metabolic basis of inherited disease. Neww York, NY: McGraw-Hill, 2001: 2863-2913. 2.. Sijbrands EJ, Westendorp RG, Defesche JC, de
Meierr PH, Smelt AH, Kastelein JJ. Mortality over twoo centuries in large pedigree with familial hypercholesterolemia:: family tree mortality study.. BMJ 2001;322:1019-1023.
3.. The homocysteine studies collaboration. Homocysteinee and risk of ischemic heart diseasee and stroke: a meta- analysis. JAMA 2002;288:2015-2022. .
4.. Klerk M, Verhoef P, Clarke R, Blom HJ, Kok FJ, Schoutenn EG. MTHFR 677C—>T polymorphism andd risk of coronary heart disease: a meta-analysis.. JAMA 2002;288:2023-2031.
55 Wald DS, Law M, Morris JK. Homocysteine and cardiovascularr disease: evidence on causality fromm a meta-analysis. BMJ 2002;325:1202-1208. .
6.. Hopkins PN, Stephenson S, Wu LL, Riley WA, Xinn Y, Hunt SC. Evaluation of coronary risk factorss in patients with heterozygous familial hypercholesterolemia.. Am J Cardiol 2001;87:547-553. .
7.. Kawashiri M, Kajinami K, Nohara A, Yagi K, Inazuu A, Koizumi J, Mabuchi H. Effect of commonn methylenetetrahydrofolate reductase genee mutation on coronary artery disease in familiall hypercholesterolemia. Am J Cardiol 2000;86:84OS45. .
8.. Raal FJ, Pilcher GJ, Waisberg R, Buthelezi EP, Vellerr MG, Joffe Bl. Low-density lipoprotein cholesteroll bulk is the pivotal determinant of atherosclerosiss in familial hypercholesterolemia. AmJCardioll 1999;83:1330-1333.
9.. Sauvage Nolting PR, Defesche JC, Buirma RJ, Huttenn BA, Lansberg PJ, Kastelein JJ. Prevalence andd significance of cardiovascular risk factors in aa large cohort of patients with familial hypercholesterolemia.. J Intern Med 2003;253:161-168. .
10.. Smilde TJ, van den Berkmortel FW, Boers GH, Wollersheimm H, de Boo T, van Langen H, Stalenhoeff AF. Carotid and femoral artery wall thicknesss and stiffness in patients at risk for cardiovascularr disease, with special emphasis onn hyperhomocysteinemia. Arterioscler Thromb Vasee Biol 1998;18:1958-1963
111 Tonstad S, Joakimsen O, Stensland-Bugge E, Lerenn TP, Ose L, Russell D, Bonaa KH. Risk factorss related to carotid intima-media thicknesss and plaque in children with familial hypercholesterolemiaa and control subjects. Arteriosclerr Thromb Vase Biol 1996;16:984-991 12.. Tonstad S, Refsum H, Ueland PM. Association
betweenn plasma total homocysteine and parentall history of cardiovascular disease in childrenn with familial hypercholesterolemia. Circulationn 1997,96:1803-1808.
13.. Tonstad S, Joakimsen O, Stensland-Bugge E, Ose L,, Bonaa KH, Leren TP. Carotid intima-media thicknesss and plaque in patients with familial hypercholesterolemiaa mutations and control subjects.. Eur J Clin Invest 1998;28:971-979. 14.. Tonstad S, Refsum H, Ose L, Ueland PM. The
C677TT mutation in the
methylenetetrahydrofolatee reductase gene predisposess to hyperhomocysteinemia in childrenn with familial hypercholesterolemia treatedd with cholestyramine. J Pediatr 1998;132:365-368. .
11 5. Tonstad S and Hjermann I. Cardiovascular risk factorss and testing of relatives amongst patientss with familial hypertipidaemia one decadee after a clinical trial. J Intern Med 2000;248:111-118. .
16.. Kluijtmans LA, Kastelein JJ, Lindemans J, Boers GH,, Heil SG, Bruschke AV, Jukema JW, van den Heuvell LP, Trijbels FJ, Boerma GJ, Verheugt FW, Willemss F, Blom HJ. Thermolabile
methylenetetrahydrofolatee reductase in coronaryy artery disease. Circulation 1997;96:2573-2577. .
17.. Meiklejohn DJ, Vickers MA, Dijkhuisen R, Greavess M. Plasma homocysteine
concentrationss in the acute and convalescent periodss of atherothrombotic stroke. Stroke 2001;32:57-62. .
18.. Qujeq D, Omran TS, Hosini L. Correlation betweenn total homocysteine, low-density lipoproteinn cholesterol and high-density lipoproteinn cholesterol in the serum of patients withh myocardial infarction. Clin Biochem 2001;34:97-101. .
19.. Senaratne MP, Griffiths J, Nagendran J. Elevationn of plasma homocysteine levels associatedd with acute myocardial infarction Clinn Invest Med 2000;23:220-226. 20.. Brattstrom L and Wikken DE. Homocysteine
andd cardiovascular disease: cause or effect? Am JJ Clin Nutr 2000;72:315-323.
