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 contribution of classical risk
factorss to cardiovascular disease in
familiall hypercholesterolemia
Dataa in 2400 patients
Angeliquee CM Jansen
1, MD; Emily S van Aalst-Cohen
1, MD; Michael W
Tanck
2,, PhD; Mieke D Trip
1, MD, PhD; Peter J Lansberg
3, MD, PhD; An Ho
Liem
4,, MD; Henk WO Roeters van Lennep
4, MD; Eric JG Sijbrands
5, MD,
PhD;; John JP Kastelein
1, MD, PhD
Departmentss of Vascular Medicine1 and Clinical Epidemiology and Biostatistics2, :ademicc Medical Center, University of Amsterdam, Amsterdam, Department of Internal
Medicine3,, Slotervaart Training Hospital, Amsterdam, Department of Cardiology4, Oosterscheldee Hospital, Goes, Department of Internal Medicine5, Erasmus Medical
Center,, University of Rotterdam, the Netherlands.
Abstract t
Background d
Familiall hypercholesterolemia (FH) is characterized by severely elevated LDL cholesterol levelss and premature cardiovascular disease (CVD). Although FH is a monogenic disorder, thee CVD risk varies to a large extent, possibly due to differences in risk factor profiles. We thereforee determined the contribution of classical risk factors to CVD in patients with heterozygouss FH.
Methods s
Life-timee data were collected from a cohort of 2400 FH patients of 27 different hospitals by scrutinizingg medical records and the use of questionnaires. Multivariate Cox regression was usedd to study the relationship between potential risk factors and the occurrence of CVD.
Results s
Duringg 112.943 person-years, 782 (32.6%) patients had had at least one cardiovascular event. Malee gender (RR 2.82, 95% CI 2.37-3.36), smoking (RR 1.67, 95% CI 1.40-1.99), hypertension (RRR 1.36, 95% CI 1.06-1.75), diabetes meltitus (RR 2.19, 95% CI 1.36-3.54), low HDL-C {RR 1.37,, 95% Ci 1.15-1.63) and elevated lipoprotein(a) levels {RR 1.50, 95% CI 1.20-1.79) proved too be independent CVD risk factors. These six risk factors explained 18.7 % o f the variation in thee occurrence of CVD.
Conclusions s
Malee gender, smoking, hypertension, diabetes mellitus, HDL-cholesterol and lipoprotein{a) levelss proved to be important risk factors for CVD in FH patients. In addition to the routine institutionn of statin therapy, controlling these factors needs special attention in the managementt of this disorder.
Introduction n
Heterozygouss familial hypercholesterolemia (FH) is a common hereditary disorder off lipoprotein metabolismm (prevalence 1:400). FH is caused by mutations in the low-density lipoprotein receptorr (LDL-R) gene. Untreated FH patients have severely elevated LDL cholesterol (LDL-C) levelss and develop substantial atherosclerotic vascular disease at a young age.1
Althoughh FH is a monogenic disorder, the cardiovascular (CVD) risk varies considerably. In earlierr studies, approximately 45% of male and 20% of female FH patients had documented
C A / HH h\y th<= a n o n f "^0 ^ R A r o n t l w t h o m r * r t ; a l i t w f r n m FW h a c h o o n s n a l v / v o r J \r\ ^ w a r n h m o
~ 11 — ~z>~ ~~ . . - _ - . . . _ , , ^ . . ^ ' . . « . ^ „ ^ , , ^ ^.j^^v.i.uHuiji.tu I I I U .v-ij, luiyi.
pedigreee without selection on CVD.3 Strikingly, this family-tree study showed that approximatelyy 40% of untreated FH patients had a normal life expectancy, whereas 60% sufferedd from premature death. Unfortunately, very few data are as yet available on the factorss that contribute to these large differences in atherosclerotic burden.
Wee hypothesized that the occurrence of CVD in FH is determined by the contributions of additionall metabolic, environmental and genetic risk factors, acting in conjunction with severee hypercholesterolemia. Unfortunately, previous studies had yielded conflicting data, butt small study sizes might explain the equivocal findings with regards to hypertension47, diabetes633 and lipoprotein(a)911 in FH patients. Since a number of these factors are open to intervention,, more in depth knowledge of the factors concerned is of clinical importance. Therefore,, we estimated the contribution of classical risk factors to CVD in a large cohort of heterozygouss FH patients.
Methods s
Design n
Thee present investigation was a retrospective, multicenter, cohort study. The central DNA andd Biobank at the Academic Medical Centre contains DNA samples and demographics of moree than 9300 hypercholesterolemic patients collected from Lipid Clinics throughout the Netherlands.. We randomly selected a cohort of 4000 hypercholesterolemic patients {from 277 different hospitals) out of this database.
Pilott study
Inn a pilot study, the feasibility of the study was assessed by reviewing 100 medical records att four different hospitals. The availability and quality of the medical record data were testedd and subsequently the case-record forms were improved where necessary. Data on riskk factors were present in more than 80% of the records and this was considered to be sufficient.. Only smoking habits proved to be not adequately documented (56%), and
additionall questionnaires were designed for this topic.
Dataa collection
AA team of 13 especially trained data-collectors visited 27 hospitals and reviewed 4000 patients'' records. Only patients that fulfilled the clinical criteria for FH (see below) were includedd in the study. Subsequently, extensive information on demographics, classical risk factors,, laboratory parameters and CVD endpoints was acquired. To ensure data completeness,, additional information was sought from general practitioners, patients, and hospitalss that patients had visited formerly. To obtain consistent datasets, quality guidelines weree implemented. All data collectors underwent extensive training; handbooks with clinical definitionss were used and interobserver studies were carried out. Questionnaires and study informationn were sent by mail. Written informed consent was obtained from all patients (exceptt deceased patients). The Ethics Institutional Review Board of each participating hospital approvedd the protocol.
FHH diagnostic criteria
Wee used a combination of established diagnostic criteria to identify patients with FH.1-2'214. Malee and female FH patients of 18 years and older were included if they met the following criteria:: (I) the presence of a documented LDL-receptor mutation, or (II) an LDL-C level above thee 95th percentile for gender and age, in combination with (a) the presence of tendon xanthomass in the patient or in a first-degree relative, or (b) an LDL-C level above the 95th
percentilee for age and gender in a first-degree relative, or (c) proven CAD in the patient or in aa first-degree relative under the age of 60 years. Only one index case per family was included.
Classicall risk factors
Malee gender, age, smoking, body mass index and the presence of hypertension and diabetes mellituss were considered classical risk factors. The lifetime consumption of cigarettes was definedd by start and stop dates (as retrieved from medical records and from questionnaires). Hypertensionn was defined when the diagnosis had been made and when anti-hypertensive medicationn was prescribed, or if three consecutive measurements of blood pressure were >1400 mmHg systolic or > 90 mmHg diastolic. Diabetes mellitus was defined when the diagnosiss had been made and medication (insulin or oral anti-diabetics) was prescribed, or byy a fasting plasma glucose of > 6.9 mmol/L.
Laboratoryy p a r a m e t e r s
Lipidd levels, as stated in the medical record, were determined in fasting patients not using lipid-loweringg medication for at least 6 weeks. Total cholesterol (TC), high-density lipoprotein cholesteroll (HDL-C) and triglycerides were measured by standard methods. LDL-C was
calculatedd with the Friedewald formula. The DNA samples had been screened for the presencee of LDL-R mutations, as published previously.lb
Cardiovascularr disease endpoints
Thee combination of cardiovascular mortality and CVD was the primary measure of outcome (totall CVD). CVD was defined by the presence of at least one of the following: (I) myocardial infarction,, proven by at least two of the following: (a) classical symptoms (>15 minutes), (b) specificc EKG abnormalities, (c) elevated cardiac enzymes (> 2x upper limit of normal); (II) percutaneouss coronary intervention or other invasive procedures: (III) coronary artery bypass grafting;; (IV) angina pectoris, diagnosed as classical symptoms in combination with at least onee unequivocal result of one of the following; (a) exercise test, (b) nuclear scintigram, (c) dobutaminee stress ultrasound, (d) a more than 70% stenosis on a coronary angiogram; (V) ischemicc stroke, demonstrated by CT- or MRI scan (VI) documented transient ischemic attack; (VII)) peripheral arterial bypass graft; (VIII) peripheral percutaneous transluminal angioplasty orr other percutaneous invasive intervention; (IX) intermittent claudication defined as classical symptomss in combination with at least one unequivocal result of one of the following: (a) ankle/armm index<0.9 or (b) a stenosis (>50%) on an angiogram or duplex scan.
Iff information on CVD did not strictly fulfill the above-mentioned criteria, or if any suspect history,, symptoms or diagnostic evaluations were found in the record, the case was presented too an independent adjudication committee.
Statisticall analyses
Differencess in clinical characteristics between patients with and without CVD were tested withh chi-square statistics or independent sample t-test. In case of a skewed distribution, the t-testt was performed on logtransformed data (total Lipid Clinic follow-up, triglycerides, Lp(a),, homocysteine), while medians and interquartile ranges are presented. To adjust for thee effects of age and gender we used multiple logistic regression and univariate general linearr modeling.
Coxx proportional hazard regression analysis was used to model the association of potential riskk factors and the occurrence of CVD in univariate and multivariate analyses. Follow-up startedd at birth and ended for each individual at the date of the first occurrence of established CVD.. Patients without CVD were censored at the date of the last Lipid Clinic visit or at the datee of death attributable to other causes. The following variables were entered into the analyses:: gender, smoking (time-dependent), hypertension (time-dependent), diabetes mellitus (time-dependent),, BMI, lipid levels, lipoprotein(a) and homocysteine levels. For smoking we implementedd a linearly decreasing risk effect for the three years after cessation.1Ê For practical interpretations,, HDL-C, Lp(a) and homocysteine were classified according to pre-defined risk categoriess (1/0): HDL-C risk category according to our normal laboratory values: for males <
0.99 mmol/L, for females < 1.1 mmol/L; Lp(a) risk category: > 300 mg/L 17; homocysteine risk category:: > 15 umol/L.,s To evaluate the effect of missing values in this retrospective study design,, we additionally used the statistical technique of multiple imputation.19 By using this technique,, missing values are replaced with plausible values estimated under a statistical modell that incorporates covariate information observed for these cases. It has been demonstratedd that this technique is useful and major improvements have been shown whenn using multiple (3-10) rather than single imputation, therefore, we randomly decided too create five complete data sets. The same multivariate Cox model was used on these five setss and average relative risks for the five sets combined were calculated.
Too determine the percentage of the contribution of the combined risk factors to the occurrencee of CVD, the proportion of explained variation was calculated.20 Analyses were performedd using SPSS (version 10.1, Chicago, Illinois) and SAS software (version 8.02, Cary,, North Carolina). A p-value of less than 0.05 was considered to be statistically significant.
Results s
Uponn review of 4000 medical records of hypercholesterolemic patients, 2400 patients met ourr diagnostic criteria of FH. Sixty-two percent of these patients had been referred by generall practitioners, 16% and 6% were secondary referrals from cardiologists and internists, respectively.. The most frequent reason for referral was dysiipidemia (61%), followed by CVDD (16%). A total of 200 patients (8%) were referred for both dysiipidemia and signs possiblyy related to CVD. The remaining 349 (1 5%) patients were referred for miscellaneous reasons.. All first Lipid Clinic visits took place between March 1969 and November 2002 (90%% after 1988).
Duringg 112.943 person-years, 782 (32.6%) patients had at least one cardiovascular event. Patientss suffered either from coronary artery disease (n=693, 28.9%), cerebrovascular disease (n=89,, 3.7%), or peripheral ischemic disease (n=111, 4.6%). A total of 35 patients (20 men andd 15 women) died during the Lipid Clinic follow-up. Twenty-seven (77.1 %) of these patients diedd from a CVD event, in four of them this was the first occurrence of cardiovascular disease. Meann age of onset of CVD was 45.9 years in men and 52.5 years in women (pO.001). Demographicss and clinical characteristics of FH patients with and without CVD and controls aree given in Table 1. Patients with CVD were older, were more often males and smokers, andd had a higher prevalence of hypertension, diabetes mellitus and obesity than patients withoutt CVD. A higher TC level was also observed in CVD patients, but this was not significant afterr adjustment for age and gender. LDL-C levels did not differ between CVD and non-CVDD patients; HDL-C levels were significantly lower (1.14 0.33 vs. 1.24 + 0.36 mmol/L; p<0.001)) and median triglyceride levels significantly higher (1.80 vs. 1.47 mmol/L; p<0.001)
TABLEE 1 . Clinical characteristics of FH patients w i t h and w i t h o u t CVD
Demographics s
Malee gender {%)
Agee at first visit Lipid Clinic Agee at last visit Lipid Clinic Totall Lipid Clinic follow-up
Classicall risk factors
Smoking,, ever (%) Hypertensionn (%) Diabetess mellitus {%)
Physicall examination
Bodyy mass index ( k g / m 2 ) (years) ) {years) ) (years) )
Systolicc blood pressure (mmHg) Diastolicc blood pressure {mmHg) Tendonn xanthomas (%)
Laboratoryy parameters
Totall cholesterol (mmol/L) LDLL cholesterol (mmol/L) HDLL cholesterol {mmol/L) Triglyceridess (mmol/L) Lp{a)) (mg/L) Homocysteinee (u.mol/L) LDL-rr mutation k n o w n {%) CVDD + N=7822 (32.6%) 62.3 3 50.44 ( 11.3) 56.66 ( 11.4) 4.77 [2.4-9.0] 82.8 8 17.3 3 11.1 1 25.88 ) 1388 ) 833 ( 11) 42.2 2 9.699 } 7.455 ) 1.144 ) 1.800 [1.30-2.39] 2300 [77-630] 12.00 [10.0-15.0] 42.7 7
Valuess are given as mean levels standard deviation, except where clinicc follow-up, triglycerid ?s,, Lp(a)
betweenn brackets. CVD+ indicates Lp(a),, lipoprotein{a);LDL-r, C V D --N = 1 6 1 8 ( 6 7 . 4 % ) ) 42.8 8 42.00 { 12.5) 46.44 { 12.7) 3.22 [1.2-6.5] 69.3 3 6.0 0 3.2 2 24.88 ( 3.6) 1333 ( 18) 811 ( 10) 40.5 5 9.466 < 1.91} 7.377 ( 1.86) 1.244 } 1.477 [1.01-2.09] 1500 [60-390] 10.77 [8.8-13.0] 53.9 9 givenn as percentages. pp value << 0.001 << 0.001 << 0.001 << 0.001 << 0.001 << 0.001 << 0.001 << 0.001 << 0.001 << 0.001 0.4 4 0.02 2 0.4 4 << 0.001 << 0.001 << 0.001 << 0.001 << 0.001 Totall outpatient andd homocysteine are given as median w i t h the interquartile range cardiovascularr disease present
loww density lipoprotein receptor.
CVD-,, cardiovascular diseasee absent;
inn patients with CVD. Moreover, patients with CVD had higher Lp(a) and homocysteine levelss (230 vs. 150 mg/L p<0.001 and 12.0 vs. 10.7 ^mol/L p<0.001, respectively). Tablee 2 provides the relative risks and 95% confidence intervals of the risk factors in univariate andd multivariate analyses. In univariate Cox survival analysis, male gender, smoking, hypertension,, diabetes mellitus, BMI, HDL-C, triglycerides, Lp(a) and homocysteine significantlyy increased CVD risk. In the multivariate Cox regression model, we included all co-variabless concomitantly to analyze the effects of single risk factors, adjusted for all other co-variables.. Table 2 shows the significant results of the multivariate model for patients withh complete data sets (n=1956). Male gender, smoking, hypertension, diabetes mellitus, loww HDL-C and high Lp{a) appeared to be significant risk factors for CVD. Compared to the resultss of univariate analysis, triglycerides, BMI and homocysteine were not significant. Lp(a)) levels were only available in 1698 patients and therefore Lp(a) was analyzed separately. Too evaluate the potential influence of missing values in this retrospective study design, the completee multivariate model was additionally run on five complete data sets (n=2400).
TABLEE 2 Relative risk (RR) and 9 5 %
cardiovascularr disease
Malee gender
Smokingg (time dependent ++ lag effect)
Hypertensionn (time dependent) Diabetess (time dependent) Bodyy mass index (kg/m2) HDLL risk (males<0.9 mmol/L; femaless <1.1 mmol/L) Triglyceridess (mmol/L) Lp(a)) risk (>300 mg/L)
Homocysteinee risk ( >15 umol/L) RR R 2.95 5 1.79 9 1.42 2 1.96 6 1.U3 3 1.36 6 1.12 2 1.46 6 1.57 7
confidencee interval (95% CI)
UNIVARIATE E 9 5 %% CI 2 . 5 4 -- 3.43 1 . 5 5 -- 2.08 1 . 1 5 -- 1.75 1.288 - 3.01 1.011 - 1.05 1.155 - 1.62 1.055 - 1.20 1.233 - 1.73 1 . 2 4 -- 1.99 pp value <0.0001 1 <0.0001 1 0.001 1 0.002 2 0.01 1 0.0004 4 0.001 1 <0.0001 1 0.0002 2 forr the RR R 2.82 2 1.67 7 1.36 6 2.19 9 1.37 7 1.50 0
presencee of a riskk factor fot
MULTIVARIATE E n=1956 6 9 5 %% CI 2.377 - 3.36 1.400 - 1.99 1 . 0 6 -- 1.75 1.366 - 3.54 1 . 1 5 - 1 . 6 3 3 1.200 - 1.7 pp value <0.0001 1 <0.0001 1 0.02 2 0.001 1 0.0004 4 0 . 0 0 0 1 * * ** n=1698 ËË 20 00 10 20 30 40 50 60 7C Agee (years) 100 20 30 40 50 60 70 80 90 100 Agee (years) 100 20 30 40 50 60 70 : - ! . 1
-FIGURESS 1a-1d. Cumulative survival for total CVD according to four different risk factors.
Thesee data sets were created using multiple imputation as described above. The average relativee risks calculated for these five data sets yielded identical results (data not shown). Univariatee and multivariate analyses for males and females separately yielded similar relative
riskss for all risk factors (data not shown).
Moreover,, both models were used for a subgroup of patients who had experienced their first eventt before 1989, the year the first statins became available to our patients. The same six variabless were found to be of significant influence. In this analysis, the point estimates suggest thatt (without statins) male gender, hypertension and low HDL-C had larger effects.
Inn Figures 1 a-1 d, Kaplan-Meier survival curves for total CVD according to discrete risk-factor levelss are depicted. While hypertension and diabetes meiütus worsened prognosis from the agee of 60 onwards, male gender and smoking, as well as low HDL-C and high Lp(a) (data
n o tt c h n v A i n l h;aH t h i c oiiort ^ I r o ^ H v / f r r \ m rji-to 3 H T h o n r r \ r i d r t i n n nf t h a w a r i a t i n n i n t h i s
occurrencee of CVD could be explained for 18.7 % by the combined risk factors.
Discussion n
Inn a retrospective study, we assessed the contribution of classical risk factors to the risk of CVDD in a very large cohort of heterozygous FH patients, representing the current Lipid Clinic populationn in the Netherlands. Male gender, smoking, hypertension, diabetes mellitus, low HDL-C,, and Lp(a) proved to be significant and independent risk factors, while homocysteine didd not. Since earlier reports had yielded equivocal results, the influence of these parameters, althoughh expected to be important, was not dear at the start of this study. Between 1976 andd 2001, apart from a score of small studies, only one prospective follow-up study21 and five case-controll studies of sufficient size4"722 were reported (including 96 to 526 patients).
inn the present study, we defined FH on the basis of strict criteria and applied a meticulous methodd to data collection to ensure reliable information. In fact, we applied similar methodology ass in intervention trials: standardized history and physical examination was documented in a testedd case record form; registration was centrally monitored; additional questionnaires were usedd when data collection was incomplete; and dubious endpoints were presented anonymouslyy to an independent adjudication committee. Also, the large size of the cohort allowedd analyses of multiple risk factors. Lastly, the patients were recruited from all over the countryy and selection on large families or on genetically isolated populations could be avoided. Nevertheless,, our results are derived from patients referred to Lipid Clinics and therefore cautionn is required when interpreting the results. Our data may not apply to asymptomatic FHH patients who are still in the general population. Conversely, patients at the highest risk mightt have died before visiting a Lipid Clinic, which might have caused underestimation of ourr results. However, in mortality analyses we rarely observed such early deaths.3
Ninety-fivee percent of our patients received statin therapy. Therefore, our results apply to thee current Lipid Clinic patient who receives statin therapy. A separate analysis for patients
withh events before 1989, the year in which statins were introduced in The Netherlands, suggestedd an even larger effect of some risk factors at that time. However, these results shouldd be interpreted with caution, since it is unknown whether the observed differences aree caused by statin treatment or in addition by secular trends.
Whilee smoking is an important risk factor for CVD, earlier reports on its role in FH were not unanimous.. An early study prospectively observed 96 patients during a 15-year follow-up, andd showed smoking to be a significant predictor of coronary mortality, which was recently confirmed.6-211 However, some of the other larger studies reported a sex-specific or no effect.4 5"" These studies had in common small numbers of smokers and CVD events, and ass a conseguence, lacked statistical power. In our study, smoking appeared to be a significant riskk factor in males as well as in females.
Dataa on hypertension47 and diabetes mellitus68 in FH were also dubious in previous reports. Recently,, a cross-sectional study in 526 FH patients, of whom 37.2% was diagnosed with CVD,, yielded hypertension but not diabetes mellitus as an independent risk factor.7 In contrast,, two other recent studies were negative for these risk factors4-6, while again other studiess found an effect of hypertension in females only.5" In our present study, we were ablee to demonstrate that hypertension and diabetes mellitus were both significant and independentt risk factors. It is noteworthy that the vast majority of our hypertensive and diabeticc patients, as a consequence of our definition, were on hypertensive or anti-diabeticc medication. While anti-diabetic drugs are known to reduce the risk for microvascular disease,, they do not influence the occurrence of macrovascular disease.23 Consequently, thee use of these drugs in our patients should not influence the detected relative risk for CVDD associated with diabetes. In contrast, the use of anti-hypertensive drugs, known to reducee the CVD risk, might have weakened the relative risk calculated for this risk factor. Thee age of onset of diabetes and hypertension in our cohort was relatively late and from thee survival curves it can be deduced that the risk of diabetes and hypertension starts to appearr at older age. Therefore, the previous results could possibly be explained by differences inn study design, as patients were often studied at a younger age, or by insufficient statistical power. .
Loww HDL-cholesterol was a strong and independent risk factor in our study, as in many earlierr reports.4 Novel pharmacological modalities that raise HDL-C levels have been developed andd might be useful in the future management of FH.24
Lipoprotein(a)) proved to be a risk factor for the presence of CVD in our current study, while manyy earlier studies in FH patients yielded conflicting results, again probably due to small samplee sizes.91' In our study, Lp(a) levels were available in 531 patients with CVD and in 11677 without and were assessed after the first diagnosis of CVD in about 80% of cases. Therefore,, we cannot conclude that Lp{a) is a causal risk factor, since it is not sure whether
elevatedd levels of Lp(a) were pre-existent. However, plasma lipoprotein(a) concentrations aree strongly genetically determined.25 Furthermore, a recent meta-analysis of prospective studiess showed a moderate-to-strong influence of elevated Lp{a) on CAD risk amongst the generall population.26 Taken together, one might conclude that Lp(a) is indeed a risk factor forr the development of CVD in FH.
Thee role of homocysteine in atherogenesis is heavily debated. A recent meta-analysis showed thatt in the genera! population elevated homocysteine is at most a modest independent predictorr of CAD in prospective studies.27 In FH, conflicting data regarding the role of
n w r i C r h n m n r u c t n i n o r T n ll 3 K ^ w o k o a n r o n n r + Q r l ^ ? K11+ h i ^ r v - i ^ r - w c - t i ^ i n ^ u n r t-*^+ -ir* ! r \ ^ ^ ^ f l ^ ^ A n t
.j r^ . . . V ^ . . . ^ ^ jr Jl ^ | . . ^ llI . U , , ^ , ^ U ^ ^ , , , ^ ^ W , L^ ^ . , ^ U L , iUi u u t J J l l . | l l t Ï Ï U J J I W L U U ^ p ^ l . U L I I l
riskk factor in our study.
Thee CVD burden from FH differs considerably from patient to patient. This was already recognizedd in 1966 when Harlan et al. reported normal survival in a large FH pedigree.28 Recently,, this was confirmed by three different mortality studies, carried out in the United Kingdomm and The Netherlands.2 329 In addition to patients with serious excess CVD mortality, manyy patients were observed having a normal lifespan. The results of the present study are complementaryy to these mortality studies by providing insight into what additional risk factorss may actually be of importance. We think that the results of our study have a number off clinical implications. The finding that classical risk factors still play a role in FH patients, whoo are on lipid-lowering medication, underscores the importance of vigorous screening andd modulation of these factors in the statin era.
Interestingly,, classical risk factors contributed 18.7 % to the occurrence of CVD in our FH patients.. Therefore, other, still unknown and possibly genetic, factors must have contributed too atherosclerotic disease in these patients. Future research should be aimed at the elucidation off such mechanisms.
Acknowledgments s
Thiss study was supported by a grant from the Netherlands Heart Foundation (98/165). J.J.P.. Kastelein is an established investigator of the Netherlands Heart Foundation (grant D039/66510).. We acknowledge the members of the independent adjudication committee; Dr.. R.J.G. Peters, cardiologist, Prof. Dr. J. Stam, neurologist and Prof. Dr. D. Legemate, vascularr surgeon, all from the Academic Medical Center, Amsterdam, the Netherlands. We thankk all the patients who participated and the specialists of the Lipid Clinics throughout thee Netherlands. We are indebted to Mrs. G.E.E. van Noppen for her valuable assistance withh manuscript preparation.
References s
1.. Goldstein JL, Hobbs HH, Brown MS. Familial hypercholesterolemia.. In: Scriver CR, et al., editors.. The metabolic basis of inherited disease.. New York, NY, McGraw-Hill. 2001; 2863-2913. .
2.. Risk of fatal coronary heart disease in familiall hypercholesterolemia. Scientific Steeringg Committee on behalf of the Simon Broomee Register Group. BMJ 1991; 303:
onoo one
3.. Sijbrands EJ, Westendorp RG, Defesche JC, dee Meier PH, Smelt AH, Kastelein JJ. Mortalityy over two centuries in large pedigreee with familial
hypercholesterolaemia:: family tree mortality study.. BMJ 2001; 322: 1019-1023.
4.. Ferrieres J, Lambert J, Lussier-Cacan S, Davignonn J. Coronary artery disease in heterozygouss familial hypercholesterolemia patientss with the same LDL receptor gene mutation.. Circulation 1995; 92: 290-295. 5.. Hill JS, Hayden MR, Frohlich J, Pritchard PH.
Geneticc and environmental factors affecting thee incidence of coronary artery disease in heterozygouss familial hypercholesterolemia. Arterioscler.Thromb.. 1991; 11:290-297. 6.. Hopkins PN, Stephenson S, Wu LL, Riley
WA,, Xin Y, Hunt SC. Evaluation of coronary riskk factors in patients with heterozygous familiall hypercholesterolemia. Am.J.Cardiol. 2001;; 87: 547-553.
7.. Sauvage Noiting PR, Defesche JC, Buirma RJ,, Hutten BA, Lansberg PJ, Kastelein JJ. Prevalencee and significance of
cardiovascularr risk factors in a large cohort off patients with familial
hypercholesterolaemia.. J.Intern.Med. 2003; 253:: 161-168.
8.. Yanagi K, Yamashita S, Kihara S et al. Characteristicss of coronary artery disease andd lipoprotein abnormalities in patients withh heterozygous familial
hypercholesterolemiaa associated with diabetess mellitus or impaired glucose tolerance.. Atherosclerosis 1997; 132: 43-51. 99 Carmena R, Lussier-Cacan S, Roy M et al.
Lp(a)) levels and atherosclerotic vascular diseasee in a sample of patients with familial hypercholesterolemiaa sharing the same genee defect Arterioscler.Thromb.Vase.Biol. 1996;; 16: 129-136.
10.. Seed M, Hoppichler F, Reaveley D et al. Relationn of serum lipoprotein(a) concentrationn and apolipoprotein{a) phenotypee to coronary heart disease in patientss with familial hypercholesterolemia. N.Engl.J.Med.. 1990; 322: 1494-1499. 11.. Wiklund O, Angelin B, Olofsson SO et al
Apolipoprotein(a)) and ischaemic heart diseasee in familial hypercholesterolaemia. Lancett 1990; 335: 136Q-1363.
12.. US MedPed program, www.medped.org. 2003. .
13.. Defesche J. Familial Hypercholesterolemia. In: Betteridgee DJ, editor. Lipids and Vascular Disease.. London, UK, Martin Dunitz Ltd. 2000.. 65-76.
14.. Williams RR, Hunt SC, Schumacher MC et al. Diagnosingg heterozygous familial
hypercholesterolemiaa using new practical criteriaa validated by molecular genetics. Am.J.Cardiol.. 1993; 72: 171-176. 15.. Fouchier SW, Defesche JC,
Umans-Eckenhausenn MW, Kastelein JP. The molecularr basis of familial
hypercholesterolemiaa in The Netherlands. Hum.Genet.. 2001; 109:602-615.
16.. Rea TD, Heckbert SR, Kaplan RC, Smith NL, Lemaitree RN, Psaty BM. Smoking status and riskk for recurrent coronary events after myocardiall infarction. Ann.Intern.Med. 2002; 137:494-500. .
17.. Hobbs HH, White AL. Lipoprotein(a): intriguess and insights. Curr.Opin.Lipidol. 1999;; 10:225-236.
18.. Graham IM, Daly LE, Refsum HM et al. Plasmaa homocysteine as a risk factor for vascularr disease. The European Concerted Actionn Project. JAMA 1997; 277: 1775-1781. 19.. Rubin DB, Schenker N. Multiple imputation in health-caree databases: an overview and some applications.. Stat Med. 1991; 10: 585-598. 20.. Schemper M, Henderson R. Predictive
accuracyy and explained variation in Cox regression.. Biometrics 2000; 56: 249-255. 21.. Miettinen TA, Gylling H. Mortality and
cholesteroll metabolism in familial
hypercholesterolemia.. Long-term follow-up of 966 patients. Arteriosclerosis 1988; 8: 163-167.
22.. Beaumont V, Jacotot B, Beaumont JL. ischaemicc disease in men and women with familiall hypercholesteroiaemia and xanthomatosis.. A comparative study of geneticc and environmental factors in 274 heterozygouss cases. Atherosclerosis 1976; 24:441-450. .
23.. Intensive blood-glucose control with sulphonyll ureas or insulin compared with conventionall treatment and risk of complicationss in patients with type 2 diabetess (UKPDS 33). UK Prospective Diabetess Stud" 'UKPDS^ Groun. Lancet 1998;; 352: 837-853.
24.. de Grooth GJ, Kuivenhoven JA, Stalenhoef AFF et al. Efficacy and safety of a novel cholesteryll ester transfer protein inhibitor, JTT-705,, in humans: a randomized phase II dose-responsee study. Circulation 2002; 105: 2159-2165. .
25.. Boerwinkle E, Leffert CC, Lin J, Lackner C, Chiesaa G, Hobbs HH. Apolipoprotein(a) genee accounts for greater than 90% of the variationn in plasma lipoprotein(a)
concentrations.. J.Clin.Invest 1992; 90: 52-60. .
26.. Danesh J, Collins R, Peto R. Lipoprotein(a) andd coronary heart disease. Meta-analysis of prospectivee studies. Circulation 2000; 102: 1082-1085. .
27.. Homocysteine and risk of ischemic heart diseasee and stroke: a meta- analysis. JAMA 2002;288:2015-2022. .
28.. Harlan WR, Graham JB, Estes EH. Familial hypercholesterolemia:: a genetic and metabolicc study. Medicine 1966; 45: 77-110. .
29.. Sijbrands EJ, Westendorp RG, Paola LM et al.. Additional risk factors influence excess mortalityy in heterozygous familial hypercholesteroiaemia.. Atherosclerosis 2000;; 149:421-425.
