tg 2014

8/12/2019 tg 2014

http://slidepdf.com/reader/full/tg-2014 1/12www.thelancet.com/diabetes-endocrinology Published online December 23, 2013 http://dx.doi.org/10.1016/S2213-8587(13)70191-8 1

Review

The polygenic nature of hypertriglyceridaemia: implicationsfor denition, diagnosis, and managementRobert A Hegele, Henry N Ginsberg, M John Chapman, Børge G Nordestgaard, Jan Albert Kuivenhoven, Maurizio Averna, Jan Borén, Eric Bruckert, Alberico L Catapano, Olivier S Descamps, G Kees Hovingh, Steve E Humphries, Petri T Kovanen, Luis Masana, Päivi Pajukanta, Klaus G Parhofer,Frederick J Raal, Kausik K Ray, Raul D Santos, Anton F H Stalenhoef, Erik Stroes, Marja-Riitta Taskinen, Anne Tybjærg-Hansen, Gerald F Watts,Olov Wiklund, on behalf of the European Atherosclerosis Society Consensus Panel

Plasma triglyceride concentration is a biomarker for circulating triglyceride-rich lipoproteins and their metabolicremnants. Common mild-to-moderate hypertriglyceridaemia is typically multigenic, and results from the cumulativeburden of common and rare variants in more than 30 genes, as quantied by genetic risk scores. Rare autosomalrecessive monogenic hypertriglyceridaemia can result from large-effect mutations in six different genes.Hypertriglyceridaemia is exacerbated by non-genetic factors. On the basis of recent genetic data, we redene thedisorder into two states: severe (triglyceride concentration >10 mmol/L), which is more likely to have a monogeniccause; and mild-to-moderate (triglyceride concentration 2–10 mmol/L). Because of clustering of susceptibility allelesand secondary factors in families, biochemical screening and counselling for family members is essential, but routinegenetic testing is not warranted. Treatment includes management of lifestyle and secondary factors, andpharmacotherapy. In severe hypertriglyceridaemia, intervention is indicated because of pancreatitis risk; in mild-to-moderate hypertriglyceridaemia, intervention can be indicated to prevent cardiovascular disease, dependent ontriglyceride concentration, concomitant lipoprotein disturbances, and overall cardiovascular risk.

IntroductionThe complex causes and classication of hyper tri-glyceridaemia frequently make diagnosis and manage-ment a challenge to many clinicians of diverse specialties.Hyper triglyceridaemia is usually diagnosed when thefasting plasma concentration of triglyceride exceeds a

threshold value (eg, >1·7 mmol/L [>150 mg/dL]). Severehypertriglyceridaemia is often diagnosed when plasmatriglyceride concentration is >10 mmol/L (>885 mg/dL). 1–7 Proposed denitions of hyper triglyceridaemia vary(table 1), and none predominates in clinical use.Traditional classication schemes for hypertri-glyceridaemia have used terms such as familial hypertri-glyceridaemia and familial combined hyperlipidaemia,which imply a single gene or monogenic cause. However,most cases of hyper triglyceridaemia are the result ofmany genetic factors—ie, they are multigenic orpolygenic, with accumulations of both common DNAvariants with small effect size, and rare DNA variantswith large effect size. 4 Hyper triglyceridaemia in

susceptible individuals is further exacerbated by exposureto non-genetic secondary factors, 4 including lifestylefactors such as being over weight and alcohol use.

Although prospective and case-control studies haveidentied high plasma concentration of triglyceride asan independent risk factor for cardiovascular disease, 8,9 uncertainty remains about the specic role oftriglyceride-rich lipoproteins in atherogenesis. 1–3 Furthermore, ndings from intervention studies aimedat reducing triglyceride concentrations have showninconsistent effects for cardiovascular disease outcomes,and no effect on stroke and all-cause mortality. 3 Therefore, mild-to-moderate hypertriglyceridaemia isoften viewed as a mere marker of cardiovascular diseaserisk, whereas severe hypertriglyceridaemia remains a

well known risk factor for acute pancreatitis. 4 Althoughthe need to intervene in an individual with severe hyper-triglyceridaemia is un disputed, the appropriate responsefor mild-to-moderate hypertriglyceridaemia is less clear.In this Review, we recommend redenition of hyper tri-glyceridaemia using a two-group classication to

simplify the diagnosis and clinical management ofhypertriglyceridaemia states.

Plasma triglycerideconcentration(mmol/L)

2011 ESC/EAS guidelines6,7

Normal <1·7

Hypertriglyceridaemia 1·7–9·9

Severe hypertriglyceridaemia ≥10

2001 NCEP ATP III guidelines5

Normal <1·7

Hypertriglyceridaemia

Borderline high 1·7–2·3

High 2·3–5·6

Very high >5·6

2012 Endocrine Society guidelines 1

Normal <1·7

Hypertriglyceridaemia

Mild 1·7–2·3

Moderate 2·3–11·2

Severe hypertriglyceridaemia

Severe 11·2–22·4

Very severe >22·4

ESC=European Society of Cardiology. EAS=European Atherosclerosis Society.NCEP ATP III=National Cholesterol Education Program Adult Treatment Panel III.

Table :Clinical denitions for hypertriglyceridaemia

Lancet Diabetes Endocrinol 2013

PublishedOnlineDecember 23, 2013http://dx.doi.org/10.1016/S2213-8587(13)70191-8

Department of Medicine,Western University, London,ON, Canada (Prof R A Hegele MD); IrvingInstitute for Clinical andTranslational Research,Columbia University, New York,NY, USA (Prof H N Ginsberg MD);Dyslipidaemia andAtherosclerosis Research Unit,INSERM U939, Pitié-SalpêtrièreUniversity Hospital, Paris,France (Prof M J Chapman DSc);Department of DiagnosticSciences, Herlev Hospital,University of Copenhagen,Denmark(Prof B G Nordestgaard DMSc);Department of MolecularGenetics, University MedicalCenter Groningen, University ofGroningen, Netherlands (J A Kuivenhoven PhD);Department of InternalMedicine, University ofPalermo, Palermo, Italy (Prof M Averna MD); StrategicResearch Center, SahlgrenskaCenter for Cardiovascular andMetabolic Research, Universityof Gothenburg, Gothenburg,Sweden (Prof J Borén PhD);Department of Endocrinologyand Metabolism,Endocrinology andCardiovascular DiseasePrevention, Hôpital Pitié-Salpêtrière, Paris, France (Prof E Bruckert MD);Department ofPharmacological Sciences,University of Milan andMultimedica IRCSS, Milan, Italy (Prof A L Catapano PhD); Centrede Recherche Médicale, LipidClinic, Hopital de Jolimont,Haine Saint-Paul, Belgium (O S Descamps PhD);Department of VascularMedicine, Academic MedicalCenter, University ofAmsterdam, Amsterdam,Netherlands (G K Hovingh PhD,Prof E Stroes MD); Centre for

http://crossmark.crossref.org/dialog/?doi=10.1016/S2213-8587(13)70191-8&domain=pdf

8/12/2019 tg 2014

http://slidepdf.com/reader/full/tg-2014 2/122 www.thelancet.com/diabetes-endocrinology Published online December 23, 2013 http://dx.doi.org/10.1016/S2213-8587(13)70191-8

Review

Cardiovascular Genetics,Institute of CardiovascularScience, University College

London, London, UK (Prof S E Humphries PhD);

Wihuri Research Institute,Helsinki, Finland

(Prof P T Kovanen MD); VascularMedicine and Metabolism Unit,

Sant Joan University Hospital,Universitat Rovira & Virgili,

IISPV, CIBERDEM, Reus, Spain (Prof L Masana PhD);

Department of HumanGenetics, David Geffen School

of Medicine, University ofCalifornia, Los Angeles, CA, USA (P Pajukanta PhD); Department

of Endocrinology andMetabolism, University ofMunich, Munich, Germany

(Prof K G Parhofer MD); Divisionof Endocrinology and

Metabolism, Director of theCarbohydrate and Lipid

Metabolism Research Unit,University of the

Witwatersrand, Johannesburg,South Africa (Prof F J Raal PhD);

Cardiovascular SciencesResearch Centre, St George’s

Hospital NHS Trust, London, UK(Prof K K Ray MD); Lipid Clinic

Heart Institute (InCor),University of São Paulo Medical

School Hospital, São Paulo,Brazil (R D Santos PhD);Department of Internal

Medicine, Radboud UniversityMedical Center, Nijmegen,

Netherlands (Prof A F H Stalenhoef PhD);

Cardiovascular Research Group,Heart and Lung Centre, HelsinkiUniversity Central Hospital and

Research Programs Unit,Diabetes and Obesity,

University of Helsinki, Helsinki,Finland (Prof M-R Taskinen PhD);

Department of ClinicalBiochemistry, Rigshospitalet,

University of Copenhagen,Copenhagen, Denmark

(Prof A Tybjærg-Hansen DMSc);School of Medicine and

Pharmacology, Royal PerthHospital Unit, The University of

Western Australia, Perth, WA,Australia (Prof G F Watts DSc);

and Department of Cardiology,Wallenberg Laboratory,Sahlgrenska University

Hospital, Gothenburg, Sweden (Prof O Wiklund PhD)

Correspondence to:Prof Robert A Hegele, Blackburn

Cardiovascular GeneticsLaboratory, Robarts Research

Institute, London, ON, Canada

N6A [email protected]

Considerations for measurement of triglyceride

concentrationsIn most countries, triglyceride concentration isestablished by direct laboratory analysis of plasma(usually) or serum after a 10–12 h fasting period. Indeed,clinicians routinely measure plasma triglyceride, becauseit is usually required for the Friedewald calculation ofLDL cholesterol concentration. Modern methods formeasurement of plasma triglyceride establish the freeglycerol concentration after specic lipase action, whichis the sum of the glycerol formed from the triglycerideplus the original free glycerol. However, the value for freeglycerol is usually ignored because of the low plasmaconcentration of this molecule. Therefore, hyper-triglyceridaemia can be incorrectly diagnosed in rarepatients with glycerol kinase deciency who have highbaseline concentrations of plasma glycerol. 10 The onlyprocedure that reliably differentiates the specictriglyceride-rich lipoprotein fractions is ultracentri-fugation followed by electrophoresis, which is done insome specialised lipid centres.

Most people—certainly in high-income countries—arein the non-fasting or postprandial state for most of theday. Although recent guidelines 1–3,5–7 unequivocallyrecommend measurement of fasting triglycerideconcentrations, the importance of measurement of non-fasting triglyceride and remnant cholesterol is anemerging aspect of stratication for cardiovascular diseaserisk, because these measures partly show the capacity ofthe individual to clear postprandial lipids. Findings frompopulation studies show that despite postprandialincreases in triglyceride, quantitative changes in otherlipids, lipoproteins, and apolipoproteins seem to benegligible in response to the average meal intake in mostindividuals.11,12 By contrast, for patients with dyslipidaemiawith or without insulin resistance, the postprandial areaunder the curve for triglyceride-rich lipoproteins can beup to four times larger than for patients withoutdyslipidaemia, with pronounced modication oflipoprotein remodelling leading to an increase in thepotentially atherogenic cholesterol load. 12

High concentration of non-fasting triglyceride is also

strongly associated with increased risk of myocardialinfarction, ischaemic stroke, and early death. 13,14 Evidencesuggests that non-fasting triglyceride-rich lipoproteins ortheir remnants predispose to ischaemic heart disease andmyocardial infarction. 15,16 Non-fasting concentrations oftotal cholesterol, triglyceride, HDL cholesterol, non-HDLcholesterol, LDL cholesterol, APOB, APOA1, the ratio oftotal cholesterol to HDL cholesterol, and the ratio ofAPOB to APOA1, are also associated with increased riskof cardiovascular disease. 11 These ndings suggest that,compared with proles for fasting lipids, proles for non-fasting lipids are not only useful but also perhaps equallyor more informative for risk prediction of cardiovasculardisease. This approach has already been used clinically insome Scandinavian countries (eg, Denmark). 17

Triglycerides and risk of cardiovascular disease

High plasma concentrations of triglyceride andtriglyceride-rich lipoproteins have a role in cardiovasculardisease. 1–3 The magnitude of the contribution of triglycerideto cardiovascular disease risk and the exact mechanismsby which triglyceride-rich lipoproteins exert their effectson the vascular wall are incompletely established. Thatnon-fasting triglyceride concentrations are relevant tocardiovascular disease risk is evident from large, long-termprospective studies 13,18 in the general population, therebycorroborating the original hypothesis proposed byZilversmit 19 that atherosclerosis, at least partly, occurspostprandially. Indeed, triglyceride-rich lipoproteins (eg,intermediate-density lipoprotein and very low-densitylipoprotein) could be particularly prone to entrapmentwithin the arterial wall, whereas nascent chylomicrons andvery large particles of very low-density lipoprotein are toolarge to penetrate. 19–23 Consistent with this notion, ndingsfrom Mendelian randomisation studies suggest thatlifelong high plasma concentrations of triglyceride-richlipoproteins or their remnants are causally associated withincreased risk of ischaemic heart disease, 15,16 independentof subnormal concentrations of HDL cholesterol. 13

The relative risk of cardiovascular disease from anincrease of 1 mmol/L in plasma triglyceride concentrationsranges from 1·14 to 1·80, dependent on sex and race, afteradjustment for established risk factors (eg, HDLcholesterol). 3 Other studies in various cohorts comparedthe top versus the bottom tertile or quintile for triglycerideconcentrations, and reported adjusted odds ratios (ORs) of1·2–4·0 for increased risk of cardiovascular disease. 2 The Emerging Risk Factors Collaboration 24 assessed302 430 people from Europe and North America withoutcardiovascular disease at baseline in 68 prospectivestudies. The hazard ratio (HR) for cardiovascular diseasewas 1·37 per standard deviation of triglyceride (95% CI1·31–1·42), after adjustment for non-lipid risk factors.However, this risk was essentially lost after adjustment forboth HDL and non-HDL cholesterol (HR 0·99, 95% CI0·94–1·05); 24 HDL cholesterol alone weakened, but didnot abolish, the association (Ray KK, unpublished).Importantly, even for non-fasting samples, triglyceride

was not independently associated with cardiovasculardisease risk after adjustment for non-HDL cholesterol andHDL cholesterol. Although this nding suggests thatHDL cholesterol drives the association with cardiovasculardisease, re-examination of the putative atheroprotectiverole of HDL25 suggests that triglyceride-relatedmechanisms should be reconsidered as part of thepathophysiology of cardiovascular disease. Finally,ndings from Mendelian randomisation analyses suggesta fairly direct causal association between triglyceride andtriglyceride-rich lipoproteins, and risk of coronary heartdisease, with similar ORs to those reported for prospectivestudies. 15,16,26 Findings from a recent genetic study 27 thataccounted for effects on multiple components of the lipidprole similarly supported a causal role for triglycerides in

8/12/2019 tg 2014


Review

development of coronary artery disease. On the basis of

the available epidemiological and genetic data, arandomised clinical trial of a specic triglyceride-loweringdrug should show a causal association between plasmatriglyceride concentrations and morbidity and mortalityfrom cardiovascular disease. However, ndings fromclinical trials of available drugs to reduce plasmatriglycerides, which also affect other components of thelipid prole, have shown little effect for cardiovasculardisease outcomes. 28–31 One reason for this nding mightbe that trials with cardiovascular outcomes have mainlyenrolled individuals without clinically relevanthypertriglyceridaemia, making the assessment of theeffects of specic interventions challenging.

Historical classication of hypertriglyceridaemiaphenotypesPhenotypic heterogeneity among patients withhypertriglyceridaemia was dened in the past by qualitativeand quantitative differences in plasma lipoproteins. In thepre-genomic era, the Fredrickson classication ofhyperlipoproteinaemia phenotypes (included in the WHOInternational Classication of Diseases 32) was based on theelectrophoretic patterns of lipoprotein fractions (table 2).Five of the six phenotypes described by this classicationinclude hyper triglyceridaemia in their denitions, the onlyexception being familial hypercholesterolaemia (hyper-lipoproteinaemia type 2A). 33,34

The different hypertriglyceridaemia-associated pheno-types are dened by the specic class or classes ofaccumulated triglyceride-rich lipoprotein particles,including chylomicrons, and very low-density lipoproteinsand their remnants (table 2). 33 Often, excess of triglyceride-

rich lipoproteins coexists with other lipoprotein

disturbances; for instance, patients with all forms ofhypertriglyceridaemia often have decreased concentrationsof HDL cholesterol. Implicit in this classication systemwas the idea that differences between hyper-triglyceridaemia-associated phenotypes were due togenetic differences; however, recent data suggest that thisscenario is typically not the case. 35–39 As a result, thisclassication system has neither improved scienticinsight nor been clinically useful to direct therapy orpredict hard outcomes (eg, cardiovascular mortality). Wesuggest that triglyceride concentration itself (gure 1),together with the presence of other risk factors, should bethe main driver of clinical management.

Complex genetic basis for hypertriglyceridaemiaFor several decades, the word familial has been used in thedenitions and classication of disorders of plasmatriglyceride metabolism. However, the constant re-inforcement of this terminology has a misleading effect.Colloquially, familial often implies a single-gene problem,as in the case of familial hypercholesterolaemia, amonogenic disorder characterised by increasedconcentrations of LDL cholesterol, xanthelasmapalpebrarum, arcus cornealis, tendon xanthomata, andaccelerated atherosclerosis. 34 Familial hyperchol ester-olaemia is usually due to loss-of-function mutations inLDLR, which encodes the LDL receptor, and in other genesencoding proteins that interact with LDLR such as APOB or PCSK9 . A clear monogenic cause can be established inmore than 80% of patients with a strong clinical diagnosisof familial hypercholesterolaemia, whereas in theremainder, high LDL cholesterol is a polygenic trait due to

WHO ICDnumber

Fredrickson hyper-lipoproteinaemiaphenotype

OMIM number Main lipid change Primarylipoproteinchange

Genetics

Familial hyperchylomicronaemia E78.3 Type 1 238600 ↑Triglyceride ↑Chylomicrons Monogenic; autosomal recessive due to two mutant alleles ofLPL, APOC , APOA, LMF, GPIHBP , or GPD ; presentationmainly paediatric or early adulthood

Familial hypercholesterolaemia E78.0 Type 2A 143890 ↑Total cholesterol ↑LDL Monogenic; autosomal codominant; heterozygous form resultsfrom one mutant allele of LDLR, APOB, orPCSK ; homozygousform results from two mutant alleles of these genes or ofLDLRAP

Combinedhyperlipoproteinaemia

E78.2,E78.4

Type 2B 144250 ↑Total cholesterol,↑triglyceride

↑VLDL,↑LDL

Polygenic; high GRS for hypertriglyceridaemia; excess of rarevariants in hypertriglyceridaemia-associated genes; high GRS forLDL cholesterol

Dysbetalipoproteinaemia E78.2 Type 3 107741 ↑Total cholesterol,↑triglyceride

↑IDL Polygenic; high GRS for hypertriglyceridaemia; excess of rarevariants in hypertriglyceridaemia-associated genes; APOE ε2/ε2homozygosity, or heterozygous rare mutation in APOE

Primary or simplehypertriglyceridaemia

E78.1 Type 4 144600 and145750

↑Triglyceride ↑VLDL Polygenic; high GRS for hypertr iglyceridaemia; excess of rarevariants in hypertriglyceridaemia-associated genes

Mixed hypertriglyceridaemia E78.3 Type 5 144650 ↑Total cholesterol,↑triglyceride

↑VLDL,↑chylomicrons

Polygenic; high GRS for hypertriglyceridaemia; excess of rarevariants in hypertriglyceridaemia-associated genes, with higherburden of risk alleles than for hyperlipoproteinaemia type 4

GRS was created by unweighted tallying of risk alleles from single nucleotide polymorphisms associated with increased plasma concentrations of triglyceride and hypertriglyceridaemia. Adapted from Hegele(2009). 33 ICD=International Classication of Diseases. OMIM=Online Mendelian Inheritance in Man database. VLDL=very low-density lipoprotein. GRS=polygenic genetic risk score. IDL=intermediate-densitylipoprotiein.

Table : Summary of classic hyperlipoproteinaemia phenotypes

8/12/2019 tg 2014


Review

an increased burden of common risk variants. 41 By starkcontrast, more than 95% of patients with hyper trigly-ceridaemia have a multigenic susceptibility component. 3,35–39

Multigenic hypertriglyceridaemia has a complex cause,consisting of an excess burden of common small-effectvariants (appendix), in addition to rare heterozygous large-effect variants in genes either directly or indirectlyassociated with plasma triglyceride concentration. Familialshould not be thought of as synonymous with monogenic;most cases of hypertriglyceridaemia are familial orinherited, but they are not monogenic. 35,36

Monogenic hypertriglyceridaemiaMonogenic hypertriglyceridaemia in patients with severehypertriglyceridaemia (triglyceride concentrations

>10 mmol/L) displays classic autosomal recessiveinheritance, with a population prevalence of about one in1 000 000. Typically, the disorder is rst evident in childhoodand adolescence. Affected individuals are oftenhomozygous or compound heterozygous for large-effectloss-of-function mutations in genes that regulatecatabolism of triglyceride-rich lipoproteins (eg, LPL,APOC2 , APOA5 , LMF1, GPIHBP1, and GPD1; table 2).37–39 Patients with monogenic disorders have substantiallyincreased fasting concentrations of chylomicrons, butusually do not develop premature atherosclerosis, probablybecause of size exclusion that limits the ability ofchylomicrons to traverse the vascular endothelialbarrier. 19–23 In the 20th century, a diagnosis of LPL deciencywas established biochemically by the absence of LPL

activity in plasma obtained after intravenous injection of

heparin.42

At present, the diagnosis can be made by DNAsequence analysis, which shows mutations in both LPL alleles causing complete LPL deciency; mutations in theother genes can also be detected by resequencing. 43

Multigenic hypertriglyceridaemiaRole of rare variantsHypertriglyceridaemia, with or without concomitantlipid or lipoprotein disturbances, tends to cluster infamilies. Although hypertriglyceridaemia usually doesnot result from strong single-gene effects that showMendelian inheritance, it still has a genetic basis, albeitone that is more complex in nature. This complexity wassuggested by ndings from pre-molecular-era familystudies of obligate heterozygous parents of patients withcomplete LPL or APOC2 deciency44,45 and unrelatedheterozygous carriers of disease-causing mutations fromthe general population. 46 These study ndings showedthat heterozygous carriers of disease-causing mutationshad a very wide range of triglyceride phenotypes, rangingfrom normotriglyceridaemia to severe hypertrigly-ceridaemia, 44–46 probably because of chance co-inheritanceof different numbers of common triglyceride-raisingvariants. Similarly, triglyceride concentrations rangefrom normal to very high in heterozygous carriers ofAPOA5 mutations. 47 Overall, mean triglycerideconcentrations in carriers of the various heterozygousmutations are higher than in normal family-based orpopulation-based controls, but many mutation carriershave normal triglyceride concentrations.

DNA resequencing has shown that individuals withtriglyceride concentrations of more than 3·3 mmol/L(>95th percentile in the USA and Canada) as a grouphave a clinically signicantly increased—about 2·5 timeshigher—frequency of rare, heterozygous loss-of-functionmutation in one of several genes governing triglyceridemetabolism compared with normotriglyceridaemiccontrols. 48,49 Most of these variants have conrmed loss-of-function effects in vitro or predicted harmful effects insilico.48,49 Although these rare mutations are stronglyassociated with hypertriglyceridaemia in patient groups,

they are not necessarily associated with hyper-triglyceridaemia in individual patients. Even withinfamilies, carriers of the same mutation show a widerange of triglyceride concentrations from normal tosevere hypertriglyceridaemia, with inconsistent verticaltransmission of triglyceride concentrations in mutationcarriers across generations. 50 Such ndings emphasisethat hypertriglyceridaemia is not a dominantly inheritedtrait in most families with a hypertriglyceridaemiaproband.

Role of common variantsFindings from genome-wide association studies ofhypertriglyceridaemia show that common variants inseveral genes (eg, APOA5 , GCKR, LPL, and APOB) are

• Severe triglycerides increase• Chylomicronaemia• Pancreatitis risk• Increased CVD risk likely

• Mild-to-moderate triglycerides increase• Increased CVD risk

28%

Multigenic:LPL, APOA , GCKR, APOB, LMF , GPIHBP ,CREBH , APOC , APOE, and small-effect variants

Can be monogenic:LPL, APOC , APOA ,LMF , GPIHBP , and GPD

0·1%

N u m

b e r o

f p e o p

l e

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Non-fasting plasma triglycerides (mmol/L)

26% 46%

Figure : Redenition of hypertriglyceridaemic states on the basis of new genetic dataTriglyceride concentrations of more than 10 mmol/L, especial ly in young patients, are more likely to be due tomonogenic causes combined with secondary factors, whereas patients with triglyceride concentrations of2–10 mmol/L represent a single group, based on the interplay of several genes (both heterozygous mutations oflarge effect, and the cumulative burden of small-effect variants, causing a high genetic risk score; gure 2),together with secondary factors. Plasma triglyceride concentrations and approximate population percentages arebased on data for more than 70 000 adults (>20 years of age) from the Copenhagen General Population Study.40

See Online for appendix

8/12/2019 tg 2014


Review

strongly associated with susceptibility to hyper tri-

glyceridaemia.51

In fact, common variations in32 triglyceride-associated genes, identied by the GlobalLipids Genetics Consortium, 37–39 are robustly associatedwith hypertriglyceridaemia; the same loci are alsoassociated with small variations in plasma triglycerideconcentrations within the normal range in healthypeople (appendix). A genetic risk score, constructed byunweighted tallying of carrier status for triglyceride-raising alleles at the 32 triglyceride-associated loci, wason average higher for patients with hypertriglyceridaemiathan for healthy controls (gure 2). 36,51 Thus, as formutation-negative patients with familial hyper-cholesterolaemia and LDL-cholesterol raising variants, 41 an increased burden of triglyceride-raising allelescontributes to hypertriglyceridaemia susceptibili ty. 48,49

Common plus rare variantsHypertriglyceridaemia susceptibility is thus established bycombinations of common small-effect and rare large-effectvariants in genes governing production or catabolism, orboth, of triglyceride-rich lipoproteins. 35–38,48,49 People withaverage triglyceride concentrations could have a balance ofprotective and harmful alleles. On the basis of studies of765 individuals for whom nine hypertriglyceridaemia-associated genes were resequenced, common and raregenetic variants together accounted for about 25% of totalvariation (and about 50% of attributable variation) inhypertriglyceridaemia susceptibility. 44,45 Because of thewide range of triglyceride concentrations and severity ofhypertriglyceridaemia phenotypes within families andamong carriers of the same genotype, genetic testing is notrecommended. Finally, the classic Fredrickson phenotypescharacterised by hypertriglyceridaemia closely resembleeach other at the genetic level, with similar accumulationsof common and rare genetic variants despite differentbiochemical phenotypes. 33,35–38,48,49 Among these phenotypes,hyperlipoproteinaemia type 3 (dysbetalipoproteinaemia) isunique in that a single gene ( APOE ) can force theexpression of hypertriglyceridaemia and hypercholester-olaemia because of accumulation of remnant particles; thecumulative effects of polygenic predisposition are

compounded by either homozygosity for the binding-defective E2 isoform of APOE, or heterozygosity for a raredysfunctional APOE mutation. 36

Secondary causesHypertriglyceridaemia is often associated with otherdisorders that independently increase plasma triglycerideconcentrations, such as type 2 diabetes, obesity, alcoholoveruse, hypothyroidism, pregnancy, hepatosteatosis,renal failure, or concomitant drug use (panel 1). 1,4,6,7 Whenone of these factors is present, hypertriglyceridaemia istermed secondary. However, secondary hyper-triglyceridaemia often also has a genetic component,because some secondary factors are frequently, but notuniversally, associated with hypertriglyceridaemia. This

genetic component suggests that people who developdyslipidaemia might carry inherited defects that confersusceptibility, which becomes clinically expressed in thepresence of an external or secondary stress. 4 For example,abdominal obesity, metabolic syndrome, and non-alcoholic fatty liver disease are associated with increasedrisk of hypertriglyceridaemia and are becomingincreasingly common in adults, adolescents, and evenchildren. Whether there is a strong secondary factor

Figure : Genetic risk scores for triglyceride-associated risk allelesUnweighted risk scores composed of risk alleles at 32 triglyceride-associated loci were summed across individualsand compared between patients with hypertriglyceridaemia and controls. The minimum unweighted risk scoreis 0, whereas the maximum unweighted risk score is 64, but most scores in the population range between 22 and46. Compared with healthy controls, the relative frequency distribution of triglyceride genetic risk scores wassignicantly increased in 504 patients with hypertriglyceridaemia (p=1·6×10− ). Figure reproduced from Johansenand colleagues39 by permission of Elsevier.

22 24 26 28 30 32 34 36 38 40 42 44 46

R e l a t i v e

f r e q u e n c y

Risk index (alleles)

0

0·05

0·10

0·15

0·20

0·25 Hypertriglyceridaemia

Healthy controls

Panel :Secondary causes of hypertriglyceridaemia

• Obesity

• Metabolic syndrome• Diet with high positive energy-intake balance, and high fat

or high glycaemic index• Increased alcohol consumption*• Diabetes (mainly type 2 diabetes)• Hypothyroidism• Renal disease (proteinuria, uraemia, or glomerulonephritis)• Pregnancy (particularly in the third trimester)• Paraproteinaemia• Systemic lupus erythematosus• Drugs including corticosteroids, oral oestrogen,

tamoxifen, thiazides, non-cardioselective β blockers andbile acid sequestrants, cyclophosphamide, asparaginase,protease inhibitors, and second-generation antipsychoticdrugs (eg, clozapine and olanzapine)

*Although the range is variable, clinically the risk of hypertriglyceridaemia is generallythought to increase with more than two units daily for men, and more than one unitdaily for women.

8/12/2019 tg 2014


Review

underlying dyslipidaemia needs to be established,because this knowledge would guide intervention.Furthermore, the severity of secondary hypertrigly-ceridaemia in an individual is probably determined bytheir genetic susceptibility component. Finally, somesecondary causes, such as obesity, metabolic syndrome,non-alcoholic fatty liver disease, and diabetes, have theirown genetic susceptibility components.

Redenition of hypertriglyceridaemic statesOn the basis of new genetic data, we recommend aredenition of hypertriglyceridaemia states (panel 2). First,triglyceride concentrations of more than 10 mmol/L arelikely to have a monogenic basis (especially in youngpatients), together with interacting secondary factors.However, even in this group, in many instances(particularly in adults) no monogenic cause can beestablished; in these cases, there is marked polygenicsusceptibility compounded by signicant exposure tosecondary factors. Thus, except in children and adolescentswith severe hypertriglyceridaemia, we do not recommendroutine genetic testing, even for adults with triglyceride

concentrations of more than 10 mmol/L. Second, peoplewith triglyceride concentrations of 2–10 mmol/L should beconsidered as a single group, irrespective of concomitantlipoprotein disturbances (eg, increased LDL cholesterol),with increased triglyceride caused by interaction of severalgenetic effects and secondary factors (gure 1). Forexample, individuals with hyper lipoproteinaemia type 2B(often called familial combined hyperlipidaemia; table 2)have the same genetic risk score as do individuals withsimilar triglyceride concentrations who have isolated orhyperlipoproteinaemia type 4 hypertriglyceridaemia; theydiffer in that individuals with hyperlipoproteinaemiatype 2B have a higher genetic burden of alleles associatedwith hypercholesterolaemia. 36 Again, we do notrecommend routine genetic testing in any individuals with

triglyceride concentrations of 2–10 mmol/L, and

recommend testing in severe triglyceridaemia only forpaediatric and adolescent patients.

Desirable concentrations of triglyceride andrelated variablesHypertriglyceridaemia is arbitrarily dened as a plasmatriglyceride concentration of more than 2 mmol/L(>175 mg/dL) on the basis of large prospectiveobservational studies, although plasma triglyceride canstart to confer risk or become a marker for cardiovasculardisease at even lower concentrations. 4,6,7,13,14 Triglycerideconcentrations rising above this threshold, due toincreased production or decreased clearance oftriglyceride-rich lipoproteins from the circulation, areaccompanied by changes in the metabolism andcomposition of other lipoprotein fractions such as LDLand HDL, which might partly explain the increasedcardiovascular disease risk. 3

According to the joint guidelines from the EuropeanSociety of Cardiology and the European AtherosclerosisSociety for management of dyslipidaemia, andConsensus Panel recommendations from the EuropeanAtherosclerosis Society, a triglyceride concentration ofless than 1·7 mmol/L (<150 mg/dL) is desirable,especially if HDL cholesterol is less than 1·0 mmol/L(<40 mg/dL) in men or 1·2 mmol/L (<45 mg/dL) inwomen. 3,6,7 In post-hoc subgroup analyses of clinicalendpoint studies of brates, clinical benet was shownin people with triglyceride concentrations of more than2·3 mmol/L (>200 mg/dL) and low HDL cholesterol. 28,29 Thus, when lifestyle measures are insuffi cient,individuals with high risk of cardiovascular disease andincreased plasma triglyceride could be considered fordrug treatment if triglyceride concentrations exceed2·3 mmol/L. 3,6,7 However, there is inadequate evidence todene treatment targets for plasma triglyceride. Even thetriglyceride threshold for diagnosis of hyper-triglyceridaemia is not irrefutable; no high-gradeevidence exists to suggest that 2·0 mmol/L is better than1·7 or 2·3 mmol/L. In this Review, we use 2·0 mmol/Las the diagnostic threshold for hypertriglyceridaemia

(panel 3), but expert opinion suggests values of±0·3 mmol/L around this cutpoint. 1,3–7

An emerging focus for reduction of cardiovasculardisease risk in patients with hypertriglyceridaemia is theconcentration of non-HDL cholesterol (comprisingcholesterol in LDL and in remnant triglyceride-richlipoproteins), which represents the total mass ofcholesterol in circulating atherogenic lipoproteinparticles. 3,13,15,16 This variable has been advocated becauseLDL cholesterol cannot be estimated by the Friedewaldequation when triglyceride concentrations are morethan 4·5 mmol/L; additionally, standardised directmeasurement of LDL cholesterol is not routinelyavailable in most centres. The desirable concentration ofnon-HDL cholesterol is less than 2·6 mmol/L

Panel : Proposed simplied redenition ofhypertriglyceridaemia• Normal: triglyceride concentration less than 2·0 mmol/L

(175 mg/dL)• Mild-to-moderate: triglyceride concentration between

2·0 and 10·0 mmol/L (175–885 mg/dL)• Severe: triglyceride concentration more than 10·0 mmol/L

(885 mg/dL)

Panel : Desirable concentrations of lipids and APOB inpatients at high risk of cardiovascular disease

• Triglycerides: concentration less than 1·7 mmol/L (150 mg/dL)• Non-HDL cholesterol: concentration less than 2·6 mmol/L

(100 mg/dL)• APOB: concentration less than 0·8 g/L in high-risk patients,

and less than 0·7 g/L in very high-risk patients

8/12/2019 tg 2014


Review

(<100 mg/dL) in high-risk individuals, and less than

3·4 mmol/L (<130 mg/dL) in low-risk individuals(panel 3). Again, there is insuffi cient high-grade evidenceto dene specic targets for any of these alternativevariables, and treatment should be individuallytailored. 3,6,7

An alternative estimation of atherogenic lipoproteinconcentrations uses APOB as a substitute for non-HDLcholesterol. APOB represents the total number ofatherogenic APOB-containing lipoprotein particles, andpredicts cardiovascular disease risk at least as well asnon-HDL cholesterol. 6,7 APOB can be reliably measuredin the presence of hypertriglyceridaemia and under non-fasting conditions. Some expert panels have thereforerecommended APOB as a secondary target in individualswith hypertriglyceridaemia. 3,6,7,52 Accordingly, APOBconcentrations of more than 1·2 g/L identify individualsat high risk of cardiovascular disease, and the desirableconcentration is less than 0·8 g/L. 3,6,7 For individuals atvery high risk, an APOB target of less than 0·7 g/L mightbe appropriate, corresponding to a non-HDL cholesterolconcentration of less than 2·6 mmol/L (100 mg/dL;panel 3).3,6,7

Management of hypertriglyceridaemiaTreatment for short-term and long-term risksTreatment of hypertriglyceridaemia has two distinctobjectives: immediate prevention of pancreatitis inpatients with severe hypertriglyceridaemia (triglycerideconcentration >10 mmol/L), and reduction of globalcardiovascular disease risk. Because hypertrigly-ceridaemia is characterised by increased concentrationsof remnant triglyceride-rich lipoproteins, concentrationsof non-HDL cholesterol or APOB are secondarytreatment targets, after LDL cholesterol. 53

After secondary causes have been treated, themanagement of mild-to-moderate hypertriglyceridaemiashould follow guideline recommendations, 3,6,7 with initialemphasis on diet and exercise. Non-pharmacological

therapy is recommended for individuals with triglyceride

concentrations of more than 2 mmol/L. The decision toinitiate pharmacological therapy depends on the amountof triglyceride elevation. Individuals with triglycerideconcentrations of more than 10 mmol/L warrantimmediate and aggressive triglyceride reduction tominimise the risk of acute pancreatitis, with use of astrict fat-reduced diet and avoidance of simplecarbohydrates; use of brates, nicotinic acid, oromega-3 fatty acids could also be considered. In thecontext of abdominal pain, treatment of severe hyper tri-glyceridaemia includes hospitalisation, with cessation oforal intake, and supportive measures including uidreplacement, avoidance of glucose infusions, and controlof obvious precipitating factors (eg, diabetes). Drugs areless effective in this situation, and substantialinterventions—eg, infusions of insulin or heparin, high-dose antioxidants, or plasma exchange—are alsoprobably of little value for most patients. 4 As notedearlier, because of the uncertain clinical benet, practiceguidelines are not universal or consistent regarding themanagement of individuals with triglycerideconcentrations of 2–10 mmol/L.

Patients with hypertriglyceridaemia should be assessedand managed for their global risk of cardiovasculardisease (table 3), which does not necessarily implymanagement of their triglyceride concentrations. Apositive family history of cardiovascular disease (denedas at least one rst-degree relative or at least two second-degree relatives with cardiovascular disease) should betaken into account, even if it is independent ofdyslipidaemia. Because many susceptibility alleles,environmental factors, and secondary factors tend to beshared within families, other family members might alsohave a lipid disorder and assessment for dyslipidaemiaand related cardiometabolic risk should be considered.This situation is analogous to that for type 2 diabetes,which clusters in families but is usually not associatedwith a single monogenic cause.

Moderately high (2–9·9 mmol/L) High (≥10 mmol/L)

Treatment priority Prevent cardiovascular disease Prevent acute pancreatitis

Primary therapeutic goal Achieve LDL cholesterol target Reduce triglyceride concentrations

Secondary therapeutic goals Achieve non-HDL cholesterol target, which is 0·8 mmol/L higher than LDLcholesterol goal, or APOB concentration <0·8 g/L; rule out and treatsecondary factors

Goals: achieve LDL cholesterol and non-HDL cholesterol goals once pancreatitisrisk is decreased, as described above; rule out and treat secondary factors

Non-pharmacologicaltherapeutic strategies

Reduce bodyweight, reduce alcohol intake, reduce simple sugar intake, increaseaerobic activity, reduce total carbohydrate intake, replace trans and saturatedfats with monounsaturated fats, increase dietary omega-3 fatty acids

Eliminate oral intake during acute pancreatitis with intravenous rehydration,then slowly re-introduce foods with small frequent meals, then longer-termstrict fat-reduced diet (<20% of calories as fat), reduce bodyweight, reducealcohol intake, reduce simple sugar intake, reduce total carbohydrate intake,replace trans and saturated fats with monounsaturated fats; increase dietaryomega-3 fatty acids; increase aerobic activity

Pharmacological therapeuticstrategies

Statins if necessary to control LDL cholesterol; if LDL cholesterol is close togoal, titrate statin dose to achieve both LDL and non-HDL cholesterol targets;if LDL cholesterol is at goal, but non-HDL cholesterol is still high, titrate statindose or add brate, nicotinic acid, or omega-3 fatty acids

Consider brate, nicotinic acid, and omega-3 fatty acids

Table : Treatment strategies for hypertriglyceridaemia by triglyceride concentration

8/12/2019 tg 2014


Review

Increased LDL cholesterol is part of the phenotype of

combined hyperlipidaemia, and amplies cardiovasculardisease risk. Thus, family members (particularly rst-degree relatives of such patients) should be screened.Irrespective of clinical designation, individuals withcombined hyperlipidaemia in particular, and hyper-triglyceridaemia in general, still need to be managed.Risk assessment and continuing care needs baseline andfollow-up lipid proling, especially because hyper-triglyceridaemia can obscure calculation of LDLcholesterol in these instances. Furthermore, measure-ments of non-HDL cholesterol (or, if available, APOB)concentrations can be helpful for both risk assessmentand monitoring of treatment if LDL cholesterolconcentrations cannot be measured.

StatinsStatins decrease LDL cholesterol concentrations by up to55%, leading to a reduction in cardiovascular disease riskof 23% per mmol/L of LDL cholesterol lowered, irrespectiveof baseline concentrations of LDL cholesterol, triglyceride,or HDL cholesterol. 54 The use of these drugs in patientswith hypertriglyceridaemia is justiable because of theirproven ability to reduce cardiovascular disease. They alsovariably reduce plasma triglyceride concentrations by up to30%,55 with reductions dependent on baseline triglycerideconcentration and dose of statin used. To achieverecommended targets, the choice of statin should be basedon effi cacy for LDL cholesterol reduction, taking intoaccount safety considerations. 6,7 In hyper triglyceridaemia,because LDL cholesterol often cannot be established,achievement of non-HDL cholesterol or APOB targetsshould also be a goal of treatment. 56

FibratesWith a triglyceride-lowering effect of 40%,57 dependent onbaseline triglyceride concentrations, brates are the rst-line treatment to decrease risk of pancreatitis for patientswith triglyceride concentrations of more than 10 mmol/L.Although controversial, ndings from a meta-analysis 28 including more than 45 000 individuals suggested thatbrates could reduce non-fatal acute coronary events and

revascularisation by about 9% (together with an absenceof overall effect for total and cardiovascular mortality and anon-signicant increase in non-cardiovascular deaths),particularly in people with triglyceride concentrations ofmore than 2·3 mmol/L and HDL cholesterolconcentrations of less than 1·0 mmol/L. 29 Therefore,brates can be used as additional therapy for individualswith high triglyceride and low HDL cholesterol. 3 However,in monogenic hypertriglyceridaemia due to LPL deciencyand triglyceride concentrations of more than 20 mmol/L,brates have little to no clinical benet.

Nicotinic acidTreatment with nicotinic acid (also known as niacin) at adose of 2–3 g/day is associated with up to 30% reduction

in triglyceride concentration, 20% increase in HDL

cholesterol concentration, up to 20% lowering of LDLcholesterol concentration, and up to 25% reduction inlipoprotein(a) concentration. Findings from studies of thecardiovascular disease benets of nicotinic acid areconicting. The 2011 Consensus Panel recommendationsfrom the European Atherosclerosis Society 3 supported theaddition of nicotinic acid to statin therapy for individualsnot at target concentrations of LDL cholesterol or non-HDL cholesterol, particularly if triglyceride remains highand HDL cholesterol is low. Combination therapy thatincludes nicotinic acid is a therapeutic option for statin-intolerant patients. However, nicotinic acid is no longer atherapeutic option in Europe, because of the withdrawalof extended-release nicotinic acid combined withlaropiprant after the announcement of negative resultsfrom the HPS-2 THRIVE study. 58 Extended-releasenicotinic acid remains available in North America underthe trade name Niaspan. This formulation was used in theAIM-HIGH study, 30 ndings from which also showed noclinical benet in reducing primary endpoints of deathfrom coronary heart disease, non-fatal myocardialinfarction, and ischaemic stroke.

Bile acid sequestrantsIn patients with hypertriglyceridaemia, bile acid sequestrantscan often cause a further increase in triglycerideconcentrations, so these drugs should be used with cautionin this patient group. Colesevelam can reduce LDLcholesterol concentrations by 15–20% in addition to thereduction achieved with statin therapy, 59 and might be anoption in the context of very mild hypertriglyceridaemia forindividuals whose LDL cholesterol, APOB, or non-HDLcholesterol are not at target concentrations, or in statin-intolerant people.

Omega-3 fatty acidsOmega-3 polyunsaturated fatty acids at doses of up to 4 gdaily reduce triglyceride concentrations by up to 30%,dependent on baseline concentrations, and mighttherefore be useful for prevention of pancreatitis. 4 Findings from a meta-analysis 31 showed that

omega-3 supplementation was not signicantlyassociated with reductions in all-cause mortality,myocardial infarction, or stroke.

Future research directions forhypertriglyceridaemiaRecent meta-analyses of gene-centric genome-wideassociation studies, and resequencing studies, havebegun to further expand and elucidate the geneticunderpinnings of different forms of hyper tri-glyceridaemia. 60 Incorporation of this knowledge intofuture exome and genome sequencing studies mightenable identication of new candidate genes. Forexample, patients with hypertriglyceridaemia andfamilies with a high genetic risk score could be sequenced

8/12/2019 tg 2014


Review

for known triglyceride loci to identify the full range of

hypertriglyceridaemia-associated variants in theseregions, whereas patients with hypertriglyceridaemiaand families with a low hypertriglyceridaemia geneticrisk score could be sequenced at the exome or genomelevel to identify new variants and genes forhypertriglyceridaemia.

Such approaches might offer the possibility ofpersonalised medicine, in which individuals withhypertriglyceridaemia are assessed, diagnosed, andtreated according to their individual genetic compositionand molecular phenotype. 60 To address the complexity ofthis task, systems approaches—integrating genomic,transcriptomic, proteomic, and epigenomic data withmetabolic and clinical phenotypes—are underdevelopment. 61 One example is weighted coexpressionnetwork analysis that correlates gene expression andmethylation networks with variants and phenotypes. 62 This analysis could provide a functionally orientedmethod to identify additional new hypertriglyceridaemia-associated genes and pathways in tissues relevant to lipidmetabolism. However, for many individuals withhypertriglyceridaemia, the usual treatment options willprobably be equally effective, irrespective of theunderlying combinations of predisposing alleles; thishypothesis needs to be formally studied.

Additionally, gene therapy is being studied in individualswith familial hyperchylomicronaemia. Specically,expression of a recombinant virus containing the humanhyperfunctional LPL*S447X variant showed promise inanimals, 63 and early clinical trials in people withintramuscular injections of alipogene tiparvovec (anadeno-associated virus carrying LPL) mediated local LPL expression, and was associated with a transient reductionin plasma triglyceride concentrations. 64 This treatment,which is also known by the trade name Glybera, wasrecently approved by the European Medicines Agency forthe treatment of classic hyperlipoproteinaemia type 1 (LPLdeciency).

Finally, new treatments for hypertriglyceridaemia havebeen developed that are based on genetic studies thatidentied rare causative mutations in families with

phenotypes of severely diminished triglycerideconcentrations. For example, lomitapide, an inhibitor ofMTTP that reduces triglyceride in addition to all APOB-containing lipoproteins, was developed becauseindividuals with homozygous mutations in MTTP causing abetalipoproteinaemia have depressed trigly-ceride concentrations. 65 Similarly, low triglycerideconcentrations in other families with monogenictriglyceride deciency prompted the development of newbiological agents targeting APOB (the recently approveddrug mipomersen), 66 APOC3,67 and ANGPTL3.68

ConclusionsDiagnosis of hypertriglyceridaemia is relevant becauseeven slight increases in triglyceride concentrations are

usually associated with increased risk of cardiovasculardisease, severely increased triglyceride is associated withincreased risk of pancreatitis, and hyper triglyceridaemiaoften coexists with other metabolic disturbances that areassociated with increased cardiometabolic risk.Epidemiological, genetic, and clinical trial evidence hasled us to recommend a simplied denition ofhypertriglyceridaemia (panel 1), with severehypertriglyceridaemia (triglyceride concentrations ofmore than 10 mmol/L, especially in the paediatric agegroup) more likely to be related to monogenic causes, andmild-to-moderate hyper triglyceridaemia (triglycerideconcentrations of 2–10 mmol/L) more likely to have apolygenic basis with secondary factors. The presence ofconcomitant lipid disturbances depends on additionalgenetic factors. Knowledge of the precise moleculardefect might be helpful to guide therapy for monogenichypertriglyceridaemia disorders, particularly in childrenand adolescents with severe hyper triglyceridaemia due toLPL deciency and related disorders. However, inpolygenic hypertriglyceridaemia, no evidence suggeststhat genotyping improves diagnosis or management.Non-fasting lipid measurements might improve theeffi ciency of screening and diagnosis of hyper-triglyceridaemia, whereas related variables (eg, non-HDLcholesterol and APOB) can provide guidance for therapy,especially when hypertriglyceridaemia is moderate to

severe. The present mainstay of treatment for all types ofhypertriglyceridaemia focuses on risk-factor control, diet,and lifestyle choice to ensure greatest health forindividuals with hyper triglyceridaemia. Pharmaco therapycan also be useful in selected subgroups, provided that itis in line with guideline recommendations. Finally,research in progress, both genetic and non-genetic, mightidentify new therapeutic targets that could lead tooptimisation of clinical management in individuals withhypertriglyceridaemia.ContributorsMJC and HNG are cochairs of the European Atherosclerosis SocietyConsensus Panel. MA, JB, EB, ALC, MJC, HNG, RAH, GKH, JAK, PP,KKR, AFHS, ES, M-RT, and AT-H are members of the Consensus Panel

writing committee. OSD, SEH, PTK, LM, BGN, KGP, FJR, RDS, GFW,and OW are members of the Consensus Panel. The Panel met twice in

Search strategy and selection criteria

We searched Medline, Current Contents, PubMed, andrelevant references with the terms “triglyceride”,“hypertriglyceridaemia”, “hyperlipidaemia”, “familial”,“monogenic”, “polygenic”, “polymorphism”, “mutation”, and“pharmacogenetics”. Articles published in English between2000 and 2013 were included. This Review was based ondiscussions at two meetings of the European AtherosclerosisSociety Consensus Panel organised and chaired by MJC andHNG, where the search results and drafts of the Review werecritically appraised; most of the Review results from aconsensus of expert opinions.

8/12/2019 tg 2014


Review

Paris and London at meetings organised and chaired by MJC and HNG.The rst meeting critically reviewed the literature whereas the secondmeeting scrutinised the rst draft of the Review. RAH, MA, JB, EB, ALC,JAK, PP, KKR, AFHS, ES, M-RT, AT-H, MJC, and HNG each draftedsections or outlines for the rst version, and the complete draft wasrevised by RAH, MJC, and HNG. All Panel members agreed to theconception and design, contributed to interpretation of available data,and suggested revisions to this Review. All Panel members approved thenal document before submission.

Conicts of interestRAH has received lecture honoraria, consultancy fees, or researchfunding from Aegerion, Amgen, Merck/Schering Plough, and Valeant.HNG has received lecture honoraria, consultancy fees, or researchfunding from Abbott, Amgen, AstraZeneca, Boehringer Ingelheim,Bristol-Myers Squibb, Genzyme, Hoffman-La Roche, Janssen, Kowa,Merck/Schering Plough, Novartis, Pzer, and Sano-Aventis/Regeneron.MJC has received lecture honoraria, consultancy fees, or researchfunding from Aegerion, Amgen, AstraZeneca, Danone, Genzyme,

Hoffman-La Roche, Kowa, Merck/Schering Plough, Pzer, and Sano-Aventis/Regeneron. BGN has received lecture honoraria, consultancyfees, or research funding from Aegerion, AstraZeneca, ISISPharmaceuticals, Merck/Schering Plough, Pzer, and Sano-Aventis/Regeneron. MA has received lecture honoraria, consultancyfees, or research funding from Aegerion, Genzyme, Hoffman-La Roche,Merck/Schering Plough, Pzer, and Sano-Aventis/Regeneron. JB hasreceived lecture honoraria, consultancy fees, or research funding fromAstraZeneca, Merck/Schering Plough, Pzer, and Sano-Aventis/Regeneron. EB has received lecture honoraria, consultancy fees,or research funding f rom Aegerion, AstraZeneca, Danone, Gent,Genzyme, Hoffman-La Roche, Kraft, Merck/Schering Plough, Sano-Aventis/Regeneron, and Unilever. ALC has received lecture honoraria,consultancy fees, or research funding from Aegerion, Amgen,AstraZeneca, Genzyme, Kowa, Lilly, Merck/Schering Plough, Pzer, andSano-Aventis/Regeneron. OSD has received lecture honoraria,consultancy fees, or research funding from AstraZeneca,

Merck/Schering Plough, Pzer, Sano-Aventis/Regeneron, and Solvay.GKH has received lecture honoraria, consultancy fees, or researchfunding from Genzyme, Merck/Schering Plough, and Pzer. SEH hasreceived lecture honoraria, consultancy fees, or research funding fromGenzyme. LM has received lecture honoraria, consultancy fees, orresearch funding from Amgen, AstraZeneca, Danone, Kowa,Merck/Schering Plough, Novartis, and Sano-Aventis/Regeneron. KGPhas received lecture honoraria, consultancy fees, or research fundingfrom Abbott, Aegerion, Amgen, AstraZeneca, Boehringer Ingelheim,Bristol-Myers Squibb, Genzyme, ISIS Pharmaceuticals, Lilly,Merck/Schering Plough, Novartis, and Sano-Aventis/Regeneron. FJRhas received lecture honoraria, consultancy fees, or research fundingfrom Amgen, ISIS Pharmaceuticals, and Sano-Aventis/Regeneron.KKR has received lecture honoraria, consultancy fees, or researchfunding from Abbott, Aegerion, Amgen, AstraZeneca, BoehringerIngelheim, Bristol-Myers Squibb, Genzyme, Hoffman-La Roche, Kowa,Merck/Schering Plough, Novartis, Novo-Nordisk, Pzer, Sano-

Aventis/Regeneron, Solvay, and Takeda. RDS has received lecturehonoraria, consultancy fees, or research funding from Aegerion, Amgen,AstraZeneca, Bristol-Myers Squibb, Genzyme, ISIS Pharmaceuticals,Merck/Schering Plough, Novo-Nordisk, Pzer, and Sano-Aventis/Regeneron. AFHS has received lecture honoraria, consultancyfees, or research funding from Genzyme and Hoffman-La Roche. ES hasreceived lecture honoraria, consultancy fees, or research funding fromBristol-Myers Squibb, Genzyme, ISIS Pharmaceuticals, and Sano-Aventis/Regeneron. M-RT has received lecture honoraria, consultancyfees, or research funding from AstraZeneca, Boehringer Ingelheim,Genzyme, Hoffman-La Roche, Kowa, Lilly, Merck/Schering Plough,Novartis, Novo-Nordisk, Pzer, and Sano-Aventis/Regeneron. GFW hasreceived lecture honoraria, consultancy fees, or research funding fromAbbott, Amgen, AstraZeneca, Boehringer Ingelheim, Gent,Merck/Schering Plough, Pzer, and Sano-Aventis/Regeneron. OW hasreceived lecture honoraria, consultancy fees, or research funding fromAstraZeneca, Merck/Schering Plough, Pzer, and

Sano-Aventis/Regeneron. JAK, PTK, PP, and AT-H declare that theyhave no conicts of interest.

AcknowledgmentsThe European Atherosclerosis Society is supported by unrestrictededucational grants from Amgen, Aegerion, AstraZeneca, Genzyme,Hoffman-La Roche, Kowa Europe, Novartis, and Sano-Aventis/Regeneron. These companies were not present at the Consensus Panelmeetings, had no role in the design or content of the Review, and had noright to approve or disapprove the nal document. RAH is supported bythe Jacob J Wolfe Distinguished Medical Research Chair at the WesternUniversity, the Edith Schulich Vinet Canada Research Chair in HumanGenetics (Tier I), the Martha G Blackburn Chair in CardiovascularResearch, and operating grants from the CIHR (MOP-13430, MOP-79523, CTP-79853), and the Heart and Stroke Foundation of Ontario(NA-6059, T-6018, PRG-4854). We thank Jane Stock (EuropeanAtherosclerosis Society Consensus Panel Administrat ion Offi ce, London,UK) for editorial and administrative support.

References1 Berglund L, Brunzell JD, Goldberg AC, et al, and the Endocrine

Society. Evaluation and treatment of hypertriglyceridemia:an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab 2012;97: 2969–89.

2 Boullart AC, de Graaf J, Stalenhoef AF. Serum triglycerides andrisk of cardiovascular disease. Biochim Biophys Acta 2012;1821: 867–75.

3 Chapman MJ, Ginsberg HN, Amarenco P, et al, and the EuropeanAtherosclerosis Society Consensus Panel. Triglyceride-richlipoproteins and high-density lipoprotein cholesterol in patients athigh risk of cardiovascular disease: evidence and guidance formanagement. Eur Heart J 2011;32: 1345–61.

4 Yuan G, Al-Shali KZ, Hegele RA. Hypertriglyceridemia: its etiology,effects and treatment. CMAJ 2007;176: 1113–20.

5 Expert Panel on Detection, Evaluation, and Treatment of HighBlood Cholesterol in Adults. Executive Summary of the ThirdReport of The National Cholesterol Education Program (NCEP)Expert Panel on Detection, Evaluation, and Treatment of HighBlood Cholesterol in Adults (Adult Treatment Panel III). JAMA 2001;285: 2486–97.

6 Catapano AL, Reiner Z, De Backer G, et al, and the Task Force forthe management of dyslipidaemias of the European Society ofCardiology (ESC) and the European Atherosclerosis Society (EAS),and the ESC Committee for Practice Guidelines 2008–2010 and2010–2012 Committees. ESC/EAS Guidelines for the managementof dyslipidaemias: the Task Force for the management ofdyslipidaemias of the European Society of Cardiology (ESC) and theEuropean Atherosclerosis Society (EAS). Atherosclerosis 2011;217 (suppl 1): S1–44.

7 Reiner Z, Catapano AL, De Backer G, et al, and the EuropeanAssociation for Cardiovascular Prevention & Rehabilitation, and theESC Committee for Practice Guidelines (CPG) 2008–2010 and2010–2012 Committees. ESC/EAS Guidelines for the management ofdyslipidaemias: the Task Force for the management of dyslipidaemiasof the European Society of Cardiology (ESC) and the EuropeanAtherosclerosis Society (EAS). Eur Heart J 2011;32: 1769–818.

8 Sarwar N, Danesh J, Eiriksdottir G, et al. Triglycerides and therisk of coronary heart disease: 10,158 incident cases among

262 525 participants in 29 Western prospective studies. Circulation2007;115:450–58.9 Sarwar N, Sattar N. Triglycerides and coronary heart disease: have

recent insights yielded conclusive answers? Curr Opin Lipidol 2009;20: 275–81.

10 Goussault Y, Turpin E, Neel D, et al. ‘Pseudohypertriglyceridemia’caused by hyperglycerolemia due to congenital enzyme deciency.Clin Chim Acta 1982;123: 269–74.

11 Langsted A, Nordestgaard BG. Nonfasting lipids, lipoproteins, andapolipoproteins in individuals with and without diabetes:58 434 individuals from the Copenhagen General Population Study.Clin Chem 2011;57: 482–89.

12 Guerin M, Egger P, Soudant C, et al. Cholesteryl ester ux from HDLto VLDL-1 is preferentially enhanced in type IIB hyperlipidemia inthe postprandial state. J Lipid Res 2002; 43: 1652–60.

13 Nordestgaard BG, Benn M, Schnohr P, Tybjaerg-Hansen A.Nonfasting triglycerides and risk of myocardial infarction, ischemicheart disease, and death in men and women. JAMA 2007;

298: 299–308.

8/12/2019 tg 2014


Review

14 Freiberg JJ, Tybjaerg-Hansen A, Jensen JS, Nordestgaard BG.Nonfasting triglycerides and risk of ischemic stroke in the generalpopulation. JAMA 2008; 300: 2142–52.

15 Varbo A, Benn M, Tybjærg-Hansen A, Jørgensen AB,Frikke-Schmidt R, Nordestgaard BG. Remnant cholesterol as acausal risk factor for ischemic heart disease. J Am Coll Cardiol 2013;61: 427–36.

16 Jørgensen AB, Frikke-Schmidt R, West AS, Grande P,Nordestgaard BG, Tybjærg-Hansen A. Genetically elevatednon-fasting triglycerides and calculated remnant cholesterol ascausal risk factors for myocardial infarction. Eur Heart J 2013;34: 1826–33.

17 Nordestgaard BG, Benn M. Fasting and nonfasting LDL cholesterol:to measure or calculate? Clin Chem 2009; 55: 845–47.

18 Bansal S, Buring JE, Rifai N, Mora S, Sacks FM, Ridker PM. Fastingcompared with nonfasting triglycerides and risk of cardiovascularevents in women. JAMA 2007;298: 309–16.

19 Zilversmit DB. Atherogenesis: a postprandial phenomenon.Circulation 1979;60: 473–85.

20 Nordestgaard BG, Wootton R, Lewis B. Selective retention of VLDL,IDL, and LDL in the arterial intima of genetically hyperlipidemicrabbits in vivo. Molecular size as a determinant of fractional lossfrom the intima-inner media. Arterioscler Thromb Vasc Biol 1995;15: 534–42.

21 Norata GD, Grigore L, Raselli S, et al. Post-prandial endothelialdysfunction in hypertriglyceridemic subjects: molecularmechanisms and gene expression studies. Atherosclerosis 2007;193: 321–27.

22 Malloy MJ, Kane JP. A risk factor for atherosclerosis:triglyceride-rich lipoproteins. Adv Intern Med 2001;47: 111–36.

23 Mamo JC, Proctor SD, Smith D. Retention of chylomicronremnants by arterial tissue; importance of an effi cient clearancemechanism from plasma. Atherosclerosis 1998;141 (suppl 1): S63–69.

24 Di Angelantonio E, Sarwar N, Perry P, et al, and the Emerging RiskFactors Collaboration. Major lipids, apolipoproteins, and risk ofvascular disease. JAMA 2009; 302: 1993–2000.

25 Ng DS, Wong NC, Hegele RA. HDL—is it too big to fail?Nat Rev Endocrinol 2013;9: 308–12.26 The Triglyceride Coronary Disease Genetics Consortium and

Emerging Risk Factors Collaboration. Triglyceride-mediatedpathways and coronary disease: collaborative analysis of 101 studies.Lancet 2010;375: 1634–39.

27 Do R, Willer CJ, Schmidt EM, et al. Common variants associatedwith plasma triglycerides and risk for coronary artery disease.Nat Genet 2013;45: 1345–52.

28 Jun M, Foote C, Lv J, et al. Effects of brates on cardiovascularoutcomes: a systematic review and meta-analysis. Lancet 2010;375: 1875–84.

29 Sacks FM, Carey VJ, Fruchart JC. Combination lipid therapy intype 2 diabetes. N Engl J Med 2010;363: 692–94.

30 Boden WE, Probsteld JL, Anderson T, et al, and theAIM-HIGH Investigators. Niacin in patients with low HDLcholesterol levels receiving intensive statin therapy. N Engl J Med 2011;365: 2255–67.

31 Rizos EC, Ntzani EE, Bika E, Kostapanos MS, Elisaf MS.Association between omega-3 fatty acid supplementation and riskof major cardiovascular disease events: a systematic review andmeta-analysis. JAMA 2012;308: 1024–33.

32 Fredrickson DS, Lees RS. A system for phenotypinghyperlipoproteinemia. Circulation 1965;31: 321–27.

33 Hegele RA. Plasma lipoproteins: genetic inuences and clinicalimplications. Nat Rev Genet 2009; 10: 109–21.

34 Soutar AK, Naoumova RP. Mechanisms of disease: genetic causesof familial hypercholesterolemia. Nat Clin Pract Cardiovasc Med 2007;4: 214–25.

35 Hegele RA, Ban MR, Hsueh N, et al. A polygenic basis for fourclassical Fredrickson hyperlipoproteinemia phenotypes that arecharacterized by hypertriglyceridemia. Hum Mol Genet 2009;18: 4189–94.

36 Johansen CT, Wang J, Lanktree MB, et al. An increased burden ofcommon and rare lipid-associated risk alleles contributes to the

phenotypic spectrum of hypertriglyceridemia.Arterioscler Thromb Vasc Biol 2011;31: 1916–26.

37 Johansen CT, Kathiresan S, Hegele RA. Genetic determinants ofplasma triglycerides. J Lipid Res 2011;52: 189–206.

38 Johansen CT, Hegele RA. Genetic bases of hypertriglyceridemicphenotypes. Curr Opin Lipidol 2011;22: 247–53.

39 Johansen CT, Hegele RA. Allelic and phenotypic spectrum ofplasma triglycerides. Biochim Biophys Acta 2012;1821: 833–42.

40 Nordestgaard BG, Freiberg JJ. Clinical relevance of non-fasting andpostprandial hypertriglyceridemia and remnant cholesterol.Curr Vasc Pharmacol 2011;9: 281–86.

41 Talmud PJ, Shah S, Whittall R, et al. Use of low-density lipoproteincholesterol gene score to distinguish patients with polygenic andmonogenic familial hypercholesterolaemia: a case-control study.Lancet 2013;381: 1293–301.

42 Beil U, Grundy SM, Crouse JR, Zech L. Triglyceride and cholesterolmetabolism in primary hypertriglyceridemia. Arteriosclerosis 1982;2: 44–57.

43 Rahalkar AR, Giffen F, Har B, et al. Novel LPL mutations associatedwith lipoprotein lipase deciency: two case reports and a literaturereview. Can J Physiol Pharmacol 2009; 87: 151–60.

44 Babirak SP, Iverius PH, Fujimoto WY, Brunzell JD. Detection andcharacterization of the heterozygote state for lipoprotein lipasedeciency. Arteriosclerosis 1989;9: 326–34.

45 Hegele RA, Breckenridge WC, Cox DW, Maguire GF, Little JA,Connelly PW. Interaction between variant apolipoproteins C-II andE that affects plasma lipoprotein concentrations. Arterioscler Thromb 1991;11: 1303–09.

46 Nordestgaard BG, Abildgaard S, Wittrup HH, Steffensen R,Jensen G, Tybjaerg-Hansen A. Heterozygous lipoprotein lipasedeciency: frequency in the general population, effect on plasmalipid levels, and risk of ischemic heart disease. Circulation 1997;96: 1737–44.

47 Priore Oliva C, Pisciotta L, Li Volti G, et al. Inherited apolipoproteinA-V deciency in severe hypertriglyceridemia.Arterioscler Thromb Vasc Biol 2005; 25: 411–17.

48 Johansen CT, Wang J, Lanktree MB, et al. Excess of rare variants ingenes identied by genome-wide association study ofhypertriglyceridemia. Nat Genet 2010;42: 684–87.

49 Johansen CT, Wang J, McIntyre AD, et al. Excess of rare variantsin non-genome-wide association study candidate genes inpatients with hypertriglyceridemia. Circ Cardiovasc Genet 2012;5: 66–72.

50 Lee JH, Giannikopoulos P, Duncan SA, et al. The transcriptionfactor cyclic AMP-responsive element-binding protein H regulatestriglyceride metabolism. Nat Med 2011;17: 812–15.

51 Teslovich TM, Musunuru K, Smith AV, et al. Biological, clinical andpopulation relevance of 95 loci for blood lipids. Nature 2010;466: 707–13.

52 Anderson TJ, Grégoire J, Hegele RA, et al. 2012 update of theCanadian Cardiovascular Society guidelines for the diagnosis andtreatment of dyslipidemia for the prevention of cardiovasculardisease in the adult. Can J Cardiol 2013;29: 151–67.

53 Sniderman AD, Williams K, Contois JH, et al. A meta-analysis oflow-density lipoprotein cholesterol, non-high-density lipoproteincholesterol, and apolipoprotein B as markers of cardiovascular risk.Circ Cardiovasc Qual Outcomes 2011;4: 337–45.

54 Baigent C, Keech A, Kearney PM, et al, and the CholesterolTreatment Trialists’ (CTT) Collaborators. Effi cacy and safety ofcholesterol-lowering treatment: prospective meta-analysis of datafrom 90,056 participants in 14 randomised trials of statins. Lancet 2005; 366: 1267–78.

55 Morrone D, Weintraub WS, Toth PP, et al. Lipid-altering effi cacy ofezetimibe plus statin and statin monotherapy and identication offactors associated with treatment response: a pooled analysis of over21,000 subjects from 27 clinical trials. Atherosclerosis2012;223: 251–61.

56 Boekholdt SM, Arsenault BJ, Mora S, et al. Association of LDLcholesterol, non-HDL cholesterol, and apolipoprotein B levels withrisk of cardiovascular events among patients treated with statins:a meta-analysis. JAMA 2012;307: 1302–09.

57 Staels B, Dallongeville J, Auwerx J, Schoonjans K, Leitersdorf E,Fruchart JC. Mechanism of action of brates on lipid andlipoprotein metabolism. Circulation 1998;98: 2088–93.

58 University of Oxford Clinical Trial Service Unit and EpidemiologicalStudies Unit. HPS2-THRIVE. http://www.ctsu.ox.ac.uk/research/mega-trials/hps2-thrive (accessed Nov 28, 2013).

8/12/2019 tg 2014

http://slidepdf.com/reader/full/tg-2014 12/12

Review

59 Huijgen R, Abbink EJ, Bruckert E, et al, and the Triple Study Group.Colesevelam added to combination therapy with a statin andezetimibe in patients with familial hypercholesterolemia: a 12-week,multicenter, randomized, double-blind, controlled trial. Clin Ther 2010;32: 615–25.

60 Asselbergs FW, Guo Y, van Iperen EP, et al, and the LifeLinesCohort Study. Large-scale gene-centric meta-analysis across32 studies identies multiple lipid loci. Am J Hum Genet 2012;91: 823–38.

61 Chen R, Mias GI, Li-Pook-Than J, et al. Personal omics prolingreveals dynamic molecular and medical phenotypes. Cell 2012;148: 1293–307.

62 Hawrylycz MJ, Lein ES, Guillozet-Bongaarts AL, et al.An anatomically comprehensive atlas of the adult human braintranscriptome. Nature 2012;489: 391–99.

63 Gaudet D, de Wal J, Tremblay K, et al. Review of the clinicaldevelopment of alipogene tiparvovec gene therapy for lipoproteinlipase deciency. Atheroscler Suppl 2010;11: 55–60.

64 Stroes ES, Nierman MC, Meulenberg JJ, et al. Intramuscularadministration of AAV1-lipoprotein lipase S447X lowerstriglycerides in lipoprotein lipase-decient patients.Arterioscler Thromb Vasc Biol 2008; 28: 2303–04.

65 Cuchel M, Meagher EA, du Toit Theron H, et al, and the Phase 3HoFH Lomitapide Study investigators. Effi cacy and safety of amicrosomal triglyceride transfer protein inhibitor in patients withhomozygous familial hypercholesterolaemia: a single-arm,open-label, phase 3 study. Lancet 2013;381: 40–46.

66 Raal FJ, Santos RD, Blom DJ, et al. Mipomersen, an apolipoproteinB synthesis inhibitor, for lowering of LDL cholesterolconcentrations in patients with homozygous familialhypercholesterolaemia: a randomised, double-blind, placebo-controlled trial. Lancet 2010;375: 998–1006.

67 Pollin TI, Damcott CM, Shen H, et al. A null mutation in humanAPOC3 confers a favorable plasma lipid prole and apparentcardioprotection. Science 2008; 322: 1702–05.

68 Musunuru K, Pirruccello JP, Do R, et al. Exome sequencing,ANGPTL3 mutations, and familial combined hypolipidemia.N Engl J Med 2010;363: 2220–27.

tg 2014

Documents