Postgrad Med J (1993) 69, 359- 369 i) The Fellowship of Postgraduate Medicine, 1993 Postgrad Med J: first published as 10.1136/pgmj.69.811.359 on 1 May 1993. Downloaded from

Special Article Management ofhyperlipidaemia: guidelines ofthe British Hyperlipidaemia Association D.J. Betteridge, P.M. Dodson', P.N. Durrington2, E.A. Hughes3, M.F. Laker4, D.P. Nicholls5, J.A.E. Rees6, C.A. Seymour7, G.R. Thompson8, A.F. Winder9, P.H. Winocour'° and R. Wray" Department ofMedicine, University College London Medical School London; 'East Birmingham NHS Trust, Birmingham; 2Department ofMedicine, University ofManchester; 3Sandwell District General Hospital, West Bromwich; 4Department ofMedicine, University ofNewcastle upon Tyne; 'Royal Victoria Hospital, Belfast; 6University Hospital of Wales, Cardiff; 7St George's Hospital Medical School, London; 8MRC Lipoprotein Team, Hammersmith Hospital, London; 9Department ofChemical Pathology, Royal Free NHS Trust and School ofMedicine, London; '0Department ofMedicine, University ofNewcastle upon Tyne; and "'Department ofMedicine, Hastings andRother NHS Trust, UK

Introduction There have been considerable advances in the Justification for attributing a causal role for understanding of plasma cholesterol and tri- LDL cholesterol in atherosclerosis has received glyceride metabolism and the lipoprotein particles considerable support from the outcome of both which transport these important lipids. Epidemi- primary and secondary CHD prevention trials ological evidence linking plasma cholesterol to employing lipid modifying therapy.6'7 Although copyright. coronary heart disease (CHD) risk has been streng- indirect evidence also points to the benefits of thened by the findings ofthe prospective part ofthe increasing plasma HDL concentrations and Multiple Risk Factor Intervention study.' Further- decreasing plasma triglyceride concentration, the more, the study of populations with plasma bulk of evidence implicates LDL as the major cholzsterol concentrations much below those seen atherogenic particle. Recent angiographic studies in Western industrialized societies has pointed to a suggest that radical alteration ofplasma lipids and continuous gradient of CHD risk with increasing lipoproteins by lifestyle, pharmacological and sur- plasma cholesterol.2 The value of measurement of gical measures can retard the progression of http://pmj.bmj.com/ individual plasma lipoproteins in defining CHD atherosclerotic lesions, delay the appearance of risk more closely is now well established. Low new lesions and in some patients lead to lesion density lipoprotein (LDL) cholesterol, is the major regression8'-2 (Table I). transporter of plasma cholesterol, and largely Advances in the cell and molecular biology of accounts for the relationship between total lipoprotein cell interactions have provided new cholesterol and CHD.3 In contrast, high density insights into the mechanisms ofatherosclerosis and lipoprotein (HDL), the other major cholesterol- much is now known about the regulation ofplasma on September 24, 2021 by guest. Protected rich lipoprotein, is inversely related to CHD.4 The lipoprotein levels. The central role of the liver in role of plasma triglyceride as an independent risk these processes has been identified. The seminal factor for CHD remains controversial,' although work of Goldstein and Brown,'3 which identified there is no doubt of the atherogenicity of some the LDL receptor pathway, provided the found- triglyceride-rich lipoproteins, namely chylomicron ation for major advances in this field. This work remnant particles and intermediate density lipo- has not only enabled the identification of genetic proteins (IDL). Very high concentrations of abnormalities underlying some of the primary plasma triglycerides, reflecting the accumulation of hyperlipidaemias but also has led to an under- chylomicrons which transport exogenous fat from standing of the mechanism of action of some the intestine, are associated with risk of pan- lipid-modifying drugs and the logical development creatitis. of new therapeutic agents. Correspondence: D.J. Betteridge, B.Sc., Ph.D., M.D., Various consensus committees have reviewed F.R.C.P., Department of Medicine, University College recent developments and have furnished guidelines London Medical School, The Middlesex Hospital, for optimal lipid and lipoprotein concentrations Mortimer Street, London WIN 8AA, UK. for the population as a whole in an attempt to Accepted: 11 January 1993 reduce the burden of CHD.4" 5 In addition, 360 D.J. BETTERIDGE et al. Postgrad Med J: first published as 10.1136/pgmj.69.811.359 on 1 May 1993. Downloaded from

Table I Differences between changes in patients in treatment and control groups in angiographic trials of lipid-lowering therapy LDL-C HDL-C Prog Reg A% A% A% A% NHLBI Type II8 (Ch) -20 + 6 -17 0 CLAS9 Co+NA -36 +35 -22 + 14 FATSI' L+Co -39 + 11 -25 +21 NA + Co -25 + 38 -21 + 28 UCSF" Co+NA±L -28 +28 -21 +20 STARS'2 Diet - 13 0 - 31 + 34 Diet + Ch -33 - 4 -34 + 29 Ch = cholestyramine; Co = colestipol; NA = nicotinic acid; L = lovastatin; NHLBI = National Heart Lung and Blood Institute; CLAS = Cholesterol Lowering Atherosclerosis Study; FATS = Familial Atherosclerosis Treatment Study; UCSF = University College San Francisco; STARS = St Thomas' Atherosclerosis Regression Study; Prog = progression; Reg = regression; LDL-C = LDL cholesterol; HDL-C = HDL cholesterol. Reproduced from Thompson, G.R. Progression and regression of coronary artery disease. Curr Opinion Lipidol 1992, 4, 263-267, with permission.

guidelines for the identification by screening and with the clinical stigmata of hyperlipidaemia; withcopyright. treatment of hyperlipidaemia have been pub- a strong family history of premature CHD or lished.6"7 This review provides the current consen- hyperlipidaemia; and those with other risk factors sus of the views of the British Hyperlipidaemia for CHD such as hypertension, diabetes mellitus Association, which consists ofphysicians and other and obesity. It is recognized that not all individuals health professionals involved in the management of with significant hyperlipidaemia will be identified lipid disorders. Hopefully general physicians, by these measures and more general screening may whether in hospital or the will find it of community, be appropriate on an opportunistic basis, as http://pmj.bmj.com/ benefit in guiding their own practice. resources permit. For general screening purposes a random, non- fasting plasma cholesterol measurement is appro- Screening for hyperlipidaemia priate. If this falls outside our previously published guidelines,'8 a full lipid profile including plasma Hyperlipidaemic individuals may be identified by cholesterol, plasma triglyceride and HDL-chol- plasma lipid measurements performed because esterol should be obtained following an overnight they have clinically overt vascular disease. Others fast (12-14 hours). LDL-cholesterol can then be on September 24, 2021 by guest. Protected may have the clinical stigmata of hyperlipidaemia calculated using the Friedewald equation:'9 (premature arcus, xanthelasma, xanthomata) and many will be identified in screening programmes. LDL cholesterol (mmol/l) = The question of which individuals should have Total cholesterol - HDL cholesterol trgl2ceride plasma lipid measurements performed has led to (applicable when total triglyceride < 4.5 mmol/l). much debate within the profession. Certain coun- tries such as the United States of America have Measurements are best performed in hospital instigated well-publicised cholesterol education chemical pathology departments. Care is required programmes designed to ensure that every adult in venepuncture techniques as prolonged venous has their cholesterol measured by opportunistic occlusion may increase cholesterol concentrations screening.'7 Although it is possible that other significantly. Posture also affects cholesterol levels countries will adopt similar strategies in the future, with a significant fall on sitting and a further fall on a strategy of selectively screening only high risk lying down. It is recommended therefore that blood individuals seems to prevail in the UK. The should be taken with minimal venous occlusion presence of hyperlipidaemia should be sought in after the subject has been sitting for 10 minutes. individuals with clinically overt vascular disease; Office-based machines employing dry chemistry GUIDELINES IN MANAGEMENT OF HYPERLIPIDAEMIA 361 Postgrad Med J: first published as 10.1136/pgmj.69.811.359 on 1 May 1993. Downloaded from technology have a role in screening as long as the disorders by genotype as opposed to laboratory instrument operator has received proper training phenotype. There is no doubt that lipoprotein and appropriate quality assurance is performed. phenotyping introduced in the early 1970s and However, it is inappropriate to make management adopted by the WHO2' gave a major impetus to the decisions based solely on these tests. understanding of the hyperlipidaemias. However, In high risk groups it is advisable to perform a this was a laboratory classification and has obvious full fasting profile in the first instance as abnor- drawbacks in the sense that it does not point to malities of triglyceride and HDL cholesterol con- aetiology; for example, type Ila hyperlipidaemia centrations may otherwise be missed. Long-term (increased LDL cholesterol) can be the result of plans should not be made without at least two hypothyroidism or familial hypercholesterolaemia. measurements. An important practice point is that For convenience the different lipoprotein pheno- many intercurrent illnesses, including acute types are shown in Table III which lists the more myocardial infarction and operative procedures, common primary hyperlipidaemias. Diagnosis may depress plasma cholesterol levels and lead to requires consideration of the clinical symptoms falsely low readings. Some patients with hyper- and signs, family history of CHD and hyper- lipidaemia will require tests beyond the scope of lipidaemia, and laboratory assessment. This diag- routine laboratories. These are available in nostic process is important as primary disorders are specialist lipid centres.20 associated with varying prognoses and differing treatments. Diagnosis ofhyperlipidaemia Treatment Once significant hyperlipidaemia has been identified it is important to diagnose the cause The main aim of treatment in the majority of where possible. It is convenient to determine hyperlipidaemic patients is to reduce the risk of whether hyperlipidaemia is secondary to other development of premature vascular disease conditions or is a primary abnormality. The prin- (primary prevention) or the occurrence of further cipal secondary causes are shown in Table II. vascular events in those with clinical vascular copyright. Diagnosis will require attention to clinical symp- disease (secondary prevention). Treatment is toms and signs and additional biochemical tests indicated in individuals with severe hypertri- such as blood glucose, thyroid function, renal and glyceridaemia primarily to prevent pancreatitis. liver function should be performed routinely. It cannot be overemphasized that treatment of If secondary causes are excluded it is presumed hyperlipidaemia in the context ofCHD prevention, that a primary hyperlipidaemia is present. As either primary or secondary, should be part of the

knowledge of the metabolic and genetic abnor- management of overall vascular risk. It is well http://pmj.bmj.com/ malities underlying these conditions has accumu- established from prospective studies that risk fac- lated it has become possible to classify lipid tors interact to multiply the risk of CHD. It is

Table II Secondary hyperlipidaemias Hormonalfactors Liver disease

Pregnancy Primary biliary cirrhosis on September 24, 2021 by guest. Protected Diabetes mellitus Extrahepatic biliary obstruction Hypothyroidism Nutritional factors Iatrogenic Obesity High dose thiazide diuretics* Anorexia nervosa P-Adrenergic receptor antagonists* which lack a-blocking effects, intrinsic sympathetic activity or vasodilator properties Alcohol abuse Corticosteroids Exogenous sex hormones Renal dysfunction Retinoids Nephrotic syndrome Chronic renal failure *For a short resume of antihypertensive drugs and lipids, see Betteridge, D.J. Questions about cholesterol. 'Lipid-friendly' antihypertensives. Postgrad Med J 1992, 68: 881-882. 362 D.J. BETTERIDGE et al. Postgrad Med J: first published as 10.1136/pgmj.69.811.359 on 1 May 1993. Downloaded from

therefore mandatory not to regard a particular CHD risk factor in isolation and attention should be given to the modification, where possible, of other coexisting factors in each patient, particular- ly smoking and raised blood pressure. The treatment of secondary hyperlipidaemia co. involves attention to the underlying cause (see 0 0 Cd Cd m 0 Table II). However, in certain causes (U secondary of CdI cd cd cu E 0- dyslipidaemia, particularly diabetes mellitus and 0 0 0 0 Cd x renal disease, significant lipid abnormalities may +-- cts -4--- Cd 4-A IV) 4= m cd cd persist best at cd E despite endeavours management of rA Cd r. X= x 0E x 4-b Cd Cd cd the primary disorder. The risk ofvascular disease is .0 W., - X 'A E 0 0. o o rA Cd cd cd high in these individuals and specific lipid-lowering -a o E 04 0 Cr-d O 2 Ps therapy may be indicated. A detailed discussion of X Cd m cd $20 W the treatment of these patients is beyond the scope of this article and the reader is referred to recent reviews of the topic.22'23 4- W A 0 The goals oftherapy in terms ofplasma lipid and 4- 4- 0 0 lipoprotein concentrations differ according to whether primary or secondary prevention is the 04 0 0 objective. In the former context a total cholesterol < 5.2 mmol/l and LDL cholesterol < 4.1 mmol/l > u u rA are appropriate. In patients with clinical vascular CIS disease data from regression studies suggest that CIS lower target values are necessary. It has been suggested and is the view of A:L, C.4 the authors that in this C> important patient group the appropriate target

V V V copyright. as >b 00 Ob Ob Ob Ob Ob Ob C-i C5 LDL cholesterol concentration should be <3.4 A V mmol/l.`7 Cd 0 0 rA rA Lifestyle measures W,) W') tfj I 0 0 0-0 W" .6 r-: r-: -6 a, Z v 0 0 0 0 0 0 0 The cornerstones of treatment of hyperlipidaemia are u u u u u u u nutritional and other lifestyle measures. These http://pmj.bmj.com/ measures are often successful in individuals with moderate polygenic hyperlipidaemia where hypo- lipidaemic drug therapy is seldom indicated. Life- style management should include advice on weight 3t > reduction in the obese, reduction of excessive qJ Cd C13 Cd O Z 0-4 0--4 1-. *_4 > Q, 0--4 P.-O 0 alcohol consumption, cessation ofsmoking and the (A encouragement ofappropriate physical activity. In 0 addition to the overall reduction of calories where on September 24, 2021 by guest. Protected Cd appropriate, dietary intake of fat is reduced to no Cd more than 30% oftotal calories. Dietary saturated Cd Cd co Cd -,a Cd fat raises plasma cholesterol levels and Cd rA potentiates Cd 0 0 Cd the hypercholesterolaemic effect of dietary choles- W 0 CO W r 0 terol. It is therefore 0 important to reduce the tn 0 O.. 0 W 0 saturated fat content of 0 0 Cd W the diet to less than 10% of x 0 00 04 0. 0 u 0 0 W calories and the dietary intake ofcholesterol below W = > Cd 0 W 300 (see Table IV). U. cd mg/day'6 Cd = cd O. 0 Nutritional counselling should be individualized 4-- Cd Cd where possible and success is more likely C-d C-d C-d 0 0 where Cd counselling ofthe patient together with the patient's 0 0 0 Cd Cd Cd 0 Cd partner is undertaken. The provision of adequate 0.4 4-0 WL, "4 04 "O u dietetic support, particularly in the primary care situation, remains a problem and the delivery of appropriate nutritional and lifestyle counselling will require optimum use of the limited profes- GUIDELINES IN MANAGEMENT OF HYPERLIPIDAEMIA 363 Postgrad Med J: first published as 10.1136/pgmj.69.811.359 on 1 May 1993. Downloaded from

Table IV Cholesterol-lowering diet resins are not only inconvenient to take but pro- duce gastrointestinal side effects of bloating, % total calories flatulence and constipation. These side effects can be ameliorated by a low dosage regimen of one Total fat < 30 sachet twice Saturated fat < 10 daily. Resins may interfere with the Monosaturated fat 10 absorption ofother drugs and there is a theoretical Polyunsaturated fat 10 risk of malabsorption of fat-soluble vitamins but Carbohydrate 50-60 clinically this is rarely a problem. Folate levels Protein 10-20 should be monitored in pregnant women and in Fibre 35 g/day children. Cholesterol < 300 mg/day Probucol Probucol is a lipophilic bis-phenol compound sional dietetic advice available. In primary care a structurally unrelated to the other lipid-lowering suitably trained practice nurse may help fulfil this drugs. It produces moderate reductions in plasma important role. The effectiveness of the delivery of cholesterol (15%) and its maximal effect may nutritional advice and patient motivation will take up to 3 months to occur. Probucol reduces largely determine the success of treatment and HDL cholesterol as well as LDL cholesterol and obviate the inappropriate prescribing of hypo- for this reason has not been considered as first-line lipidaemic drugs. Dietary measures should be therapy. However, clinically, probucol is pursued for at least 3-6 months before contem- associated with mobilization of tissue cholesterol plating lipid-lowering drug therapy. from skin and tendon xanthomata. There is increasing interest in the role ofoxidized LDL in atherogenesis. Probucol has antioxidant Lipid-modifying drugs effects which in vitro and in animal studies inhibit LDL oxidation and

atherogenesis.25 However, copyright. It is convenient to consider these agents under three there is as yet no evidence in man of an anti- main categories; drugs which lower plasma atherogenic effect of the drug. cholesterol alone, drugs which lower both plasma Probucol is generally well tolerated. Prolonga- cholesterol and triglyceride, and drugs which lower tion of the Q-T interval on the electrocardiogram plasma triglyceride alone. A detailed review of the has been described but the clinical significance of mechanisms ofaction ofhypolipidaemic agents has this is unclear. The drug accumulates in adipose been published recently.24 tissue and may persist for many weeks after the last

dose. For this reason the drug should be stopped in http://pmj.bmj.com/ women at least 6 months prior to attempted Drugs which lower plasma cholesterol alone conception. Anion-exchange resins Drugs which lower plasma cholesterol and These compounds which have been available for triglyceride clinical use for over 20 years are non-absorbable basic anion-exchange resins which are hydrophilic Fibric acidderivatives on September 24, 2021 by guest. Protected but insoluble in water. After oral administration the resins bind to bile acids in the intestine preven- Several fibrates are available for clinical use in ting their reabsorption. In response more bile acids Britain: bezafibrate, ciprofibrate, fenofibrate and are synthesized by the liver from cholesterol, which gemfibrozil. On the whole these agents are more is acquired from the plasma through increased effective than the parent compound, clofibrate, LDL receptor activity. Cholesterol synthesis is also which is now obsolete owing to the associated increased which partially offsets the cholesterol- increased risk of gallstones. The exact mode(s) of lowering effect ofresins. Nevertheless, in compliant action of the fibrates remain to be determined but patients these drugs can reduce LDL cholesterol the main effect is to stimulate clearance of levels by 20-30%, with a slight increase in HDL triglyceride-rich lipoproteins. cholesterol. Plasma triglycerides actually increase The major effect of fibrate drugs is to reduce with resin therapy, however. plasma triglyceride levels (; 50%). HDL Currently two resins are available, choles- cholesterol levels increase (t 15-20%) with tyramine and colestipol. They are in granular form therapy but effects on plasma cholesterol and LDL and need to be mixed with water or other fluid prior cholesterol vary depending on the type of hyper- to ingestion. Compliance can be a problem as the lipidaemia. In patients with isolated hyper- Postgrad Med J: first published as 10.1136/pgmj.69.811.359 on 1 May 1993. Downloaded from 364 D.J. BETTERIDGE et al. cholesterolaemia or mixed hyperlipidaemia fibrate specific competitive inhibitors of the microsomal therapy is associated with reductions of plasma enzyme HMG-CoA reductase which catalyses the LDL cholesterol of 10-25%. On the other hand, in conversion of HMG-CoA to mevalonate, a major patients with isolated hypertriglyceridaemia (type rate-determining step in cholesterol synthesis. They IV), where LDL concentrations tend to be low, are very potent inhibitors of the enzyme with K, LDL cholesterol may increase with fibrate therapy. values ranging from 0.2 to 2.2 x 10-9 M compared Apart from minor gastrointestinal symptoms the to the Km of the natural substrate which is fibrates are generally well tolerated and more 4 x 1O0-M. The liver is the major organ for serious side effects are rare. Occasionally, abnor- catabolism of LDL and the activity of specific malities of liver function occur and myopathy has hepatic receptors which bind and take up LDL been reported especially in subjects with impaired largely determines plasma LDL concentrations. renal function. The fibrates interact with The inhibition of hepatic cholesterol synthesis by anticoagulants increasing anticoagulant activity. these drugs stimulates the expression of LDL The long-term safety of fibrates has been ques- receptors resulting in increased uptake of LDL and tioned following the findings of the WHO primary a reduction in its plasma concentration. LDL prevention trial of cholesterol-lowering in the production is also decreased. Very low density 1970s which used clofibrate.26 The mortality rate lipoprotein (VLDL) is catabolised by the enzyme from non-cardiac causes was increased in the lipoprotein lipase resulting in VLDL remnants clofibrate-treated group due to an increased which can either be taken up by hepatic LDL incidence of malignancy and gallstone related receptors or further catabolised to LDL. Increased disorders, including cholecystectomy. Such effects LDL receptor activity as a result of HMG-CoA were not observed in the Helsinki Heart Study reductase inhibition will therefore result in in- where the drug used was gemfibrozil." creased direct hepatic uptake of VLDL remnants and consequent decreased conversion to LDL. Nicotinic acid As expected from their mode of action, these drugs have a major impact on plasma cholesterol In pharmacological doses nicotinic acid (3-6 g/ levels through the reduction of LDL cholesterol.copyright. day) lowers plasma cholesterol and plasma tri- This effect is dose-dependent and maximal after 4 glyceride and increases HDL cholesterol. The weeks of therapy. Reductions of 30-40% in mechanism of action of the drug is not fully plasma LDL cholesterol are commonly achieved. understood. It is unlikely that the well- M'odest reductions in plasma triglycerides demonstrated inhibition of the breakdown of (10-20%) are also observed together with a small adipose tissue triglyceride fully explains its effect. increase in HDL. The reduction in LDLcholesterol Nicotinic acid is poorly tolerated because of is paralleled by a reduction in apoprotein B, the flushing and gastrointestinal symptoms; these can major protein component of LDL. http://pmj.bmj.com/ be overcome to a certain extent by starting the drug The HMG-CoA reductase inhibitors are con- in low dosage. Tachyphylaxis to the flush develops venient to take as once-daily dosage and are but is lost ifa dose is missed. The flush appears to be generally well tolerated with few adverse effects. prostaglandin mediated as it is partly blocked by Unlike triparanol, an inhibitor of cholesterol syn- aspirin. Other side effects include a pruritic rash thesis on trial in the 1960s, there is no evidence of and hyperpigmentation. Nicotinic acid may any deleterious effects ofthe HMG-CoA inhibitors exacerbate gout and glucose intolerance and pro- on the lens. Rarely, elevated liver transaminases are duce abnormalities ofliver function. Two nicotinic observed but these abnormalities are rapidly rever- on September 24, 2021 by guest. Protected acid analogues, acipimox and nicofuranole, sible when the drugs are discontinued. Potentially Bradilan, are available for clinical use in the UK. the most important side effect is myositis with These agents are generally better tolerated than marked elevation of plasma creatine kinase. Most nicotinic acid, both in terms of symptoms and of the reported cases have been in patients taking metabolic side effects, and acipimox does not the drugs with other agents particularly cyclo- impair glucose tolerance. The analogues are not as sporine, nicotinic acid and gemfibrozil or in effective in modifying plasma lipid levels as patients with significant hepatic disease. The fre- nicotinic acid itself.28 quency of myositis is very rare. There is no evidence of clinically significant HMG-CoA reductase inhibitors effects of the drugs on other products of the cholesterol synthetic pathway such as steroid hor- The introduction ofthis new, exciting class ofdrugs mones, ubiquinones and dolichols. into clinical practice represents a major advance in the therapy of hyperlipidaemia.29 There are cur- rently two members of this class available in the UK, pravastatin and simvastatin. These agents are Postgrad Med J: first published as 10.1136/pgmj.69.811.359 on 1 May 1993. Downloaded from GUIDELINES IN MANAGEMENT OF HYPERLIPIDAEMIA 365

Drugs which lower plasma triglyceride alone The choice of hypolipidaemic drug will depend mainly upon the nature and severity of the lipid Omega-3fatty acids disorder to be treated, as shown in Table VI. For isolated moderate hypercholesterolaemia first line Omega-3 fatty acids when taken in large amounts agents would be the anion exchange resins. In this appear to reduce hepatic VLDL synthesis. Maxepa situation relatively small doses of the resins will be (a preparation containing eicosapentaenoic acid effective. Fibrates can be tried if patients are and decosahexaenoic acid) is licenced in the UK intolerant of resins. In severe hypercholestero- and is useful for the treatment of severe hypertri- laemia due to increased LDL cholesterol, as seen in glyceridaemia. In the recommended dosage (10 g/ familial hypercholesterolaemia, the HMG-CoA day) Maxepa does not reduce plasma cholesterol reductase inhibitors are effective first line agents. concentrations and in more modest hypertrigly- Despite their excellent efficacy in reducing LDL ceridaemia LDL concentrations may even increase cholesterol (Z 30-40%) the LDL concentration with treatment. may not reach acceptable levels because ofthe very high pretreatment value. In this situation con- comitant low dose resin therapy will have addi- When to use drugs and which drugs to use tional benefit. In children and women of child-bearing age Lipid-lowering drug therapy should not be under- resins are very useful agents because they are not taken lightly as therapy is usually lifelong and its absorbed. Women treated with other lipid- risk/benefit ratio should be carefully considered. lowering drugs should be advised to stop the The important clinical question is does the medication prior to conception. More information anticipated benefit of drug therapy in terms of is required on pharmacological therapy ofchildren reduction of CHD risk outweigh the imposition of with heterozygous familial hypercholesterolaemia. long-term drug therapy with the potential for Until more is known, resins remain the drugs of unpredicted adverse effects? Drug therapy should choice in this situation.

be reserved for those patients at high risk where In the patient with raised cholesterol and tri- copyright. lifestyle measures have failed to result in acceptable glycerides where the predominant abnormality is plasma lipid levels. In general terms, those patients hypercholesterolaemia, the fibrates can be used as with clinical vascular disease, those with severe first line agents. In some patients cholesterol may hyperlipidaemias (familial hypercholesterolaemia, remain high despite a satisfactory reduction in familial combined hyperlipidaemia and remnant plasma triglyceride. The addition oflow dose resin particle disease) and those with other major risk therapy is a useful option here. The HMG-CoA factors are more likely to require drug therapy. A reductase inhibitors are also useful in these patients guide to priorities for therapy is given in Table V. as single therapy. However, plasma triglycerides http://pmj.bmj.com/

Table V Priorities and action limits for lipid-lowering drug therapy in diet-resistant subjects* Total cholesterol LDL cholesterol Priority Subject category (mmol/l) (mmol/l) on September 24, 2021 by guest. Protected First Patients with existing CHD, >5.2 >3.4 or post-CABG, angioplasty or cardiac transplant Second Patients with multiple risk > 6.5 > 5.0 factors or genetically determined hyperlipidaemia, e.g. FH Third Males with asymptomatic > 7.8 > 6.0 hypercholesterolaemia Fourth Postmenopausal females > 7.8 > 6.0 with asymptomatic and HDL ratio hypercholesterolaemia < 0.2* *Aim of cholesterol lowering should be an LDL cholesterol < 3.4 mmol/l in the presence of CHD and < 4.1 mmol/l in the absence of CHD. 366 D.J. BETTERIDGE et al. Postgrad Med J: first published as 10.1136/pgmj.69.811.359 on 1 May 1993. Downloaded from

Table VI Drug therapy of hyperlipidaemia Lipid LDL Degree of Drug regimen abnormality cholesterol hyperlipidaemia 1st choice 2nd choice Type Ila Cholesterol + 4 Moderate Resin Fibrate Severe Statin Probucol Resin + statin Type IIb Cholesterol + 4 Moderate Fibrate NA Triglyceride t Severe Resin + fibrate or statin Type III Cholesterol 4 N or 4, Usually severe Fibrate Statin Triglyceride 4 Type IV Triglyceride 4 N Moderate Fibrate NA Cholesterol N or 4 Severe Fibrate + NA Type V Triglyceride 4 N Always severe Fibrate + NA Fish oil Cholesterol 4 Severe hyperlipidaemia is defined as a serum cholesterol > 7.8 or fasting triglyceride >4.5 mmol/l, especially if they occur together or are accompanied by an HDL cholesterol < I mmol/l. copyright. Resin = cholestyramine or colestipol; Fibrate = bezafibrate, ciprofibrate, fenofibrate, gemfibrozil; Statin (HMG-CoA reductase inhibitor) = pravastatin, simvastatin; NA (nicotinic acid or derivative) = acipimox; N = normal.

may remain high. The addition of a fibrate or Fibrates are the first line treatment for severe http://pmj.bmj.com/ nicotinic acid to a statin to reduce plasma tri- hypertriglyceridaemia with chylomicronaemia. glycerides in these patients cannot be generally Omega-3 fatty acids may also be useful in this recommended because of the potential for adverse situation and can be used in conjunction with reactions, particularly myositis. fibrates. Nicotinic acid or its derivatives can also be Fibrates remain the treatment of first choice in usefully added to fibrate therapy. remnant particle disease (type III). In addition, In chylomicronaemia of childhood due to lipo- there are favourable reports of the use of HMG- protein lipase deficiency or apoprotein C-II

CoA reductase inhibitors in this condition. Resins deficiency, no pharmacological treatment is on September 24, 2021 by guest. Protected aggravate the hypertriglyceridaemia and are contra- available. However, the very low fat diet necessary indicated. in these individuals may be supplemented by Pharmacological treatment ofisolated moderate medium chain triglycerides to improve tolerability. hypertriglyceridaemia remains controversial. However, some patient groups may be considered Monitoring drug therapy for drug therapy iflifestyle measures are ineffective, including patients with known CHD, familial com- The response to lipid-lowering drug therapy is not bined hyperlipidaemia, familial hypertri- always predictable and plasma lipid and lipo- glyceridaemia, a family history of CHD, remnant protein concentrations should be monitored to particle disease, diabetes mellitus and a low HDL- ensure adequate response. If one fibrate is cholesterol.30 Fibrates and nicotinic acid or its ineffective another member of the fibrate class can derivatives are the drugs of choice for isolated be tried. The response to HMG-CoA reductase hypertriglyceridaemia. LDL cholesterol tends to be inhibitors is more predictable and the majority of low in these subjects and may rise with fibrate patients will show the expected response. Failure to therapy in some individuals. For this reason, LDL respond to resins usually indicates poor com- concentrations should be monitored. pliance. Postgrad Med J: first published as 10.1136/pgmj.69.811.359 on 1 May 1993. Downloaded from GUIDELINES IN MANAGEMENT OF HYPERLIPIDAEMIA 367

Patients should be questioned with regard to the precipitation of LDL with heparin - the possible side effects of the drugs and intermittent Heparin Extracorporal LDL Precipitation (HELP) measurements of liver function tests are indicated system.36 during therapy with fibrates and HMG-CoA reductase inhibitors. If patients develop muscle pain then measurement of creatine kinase is Controversial issues indicated. Myositis is more likely to occur when HMG-CoA reductase inhibitors are combined Overall mortality with fibrates or nicotinic acid and when other drugs are used concurrently such as cyclosporine. Very It is generally accepted that primary prevention careful monitoring is required in these patients. trials of cholesterol-lowering reduce the risk of Patients taking nicotinic acid need screening for non-fatal and fatal CHD.37 However, concern has potential metabolic effects including liver function been expressed in some quarters that these trials tests, uric acid and plasma glucose. Ophthal- have failed to show a reduction in total mortality. mological surveillance is also required in patients Further, the potential adverse effects of lipid- taking nicotinic acid as a reversible macular lowering therapy in increasing non-cardiac deaths oedema may occur rarely. Initial concerns about has been emphasized by some authors.38'39 It should cataracts with the HMG-CoA reductase inhibitors be remembered that these trials have not had the have not materialized and there is no need for statistical power to answer the question as to routine ophthalmological surveillance. whether or not cholesterol-lowering alters total mortality. It is highly unlikely that a trial with such statistical power will ever be carried out because of Non-pharmacological therapy the cost. For the time being the apparent increase in non-cardiovascular mortality observed in some Surgicalprocedures drug trials remains unexplained and needs to be investigated further before its clinical significance

Partial ileal bypass3' involving the terminal one can be assessed. The variety ofdrugs used, some of copyright. third ofthe ileum has been performed in the past in them now obsolete, the very wide range ofcauses of heterozygous familial hypercholesterolaemic indi- non-CHD death and the lack of increase in the viduals intolerant of resin therapy. Reductions in latter when cholesterol was lowered in diet trials LDL cholesterol ofabout 30% have been achieved. renders a single causal mechanism most unlikely. Postoperative diarrhoea, mild steatorrhoea and the In the meantime responsible use of cholesterol- need for vitamin B12 replacement are disadvantages lowering drugs in diet-resistant patients with CHD of this procedure. The advent of the HMG-CoA or at high risk of developing CHD remains good reductase inhibitors means that few patients will clinical practice. http://pmj.bmj.com/ now be candidates for this operation despite recent evidence of its effectiveness in decreasing coronary The elderly mortality and morbidity.32 In homozygous familial hypercholesterolaemia, The gradient of risk between plasma cholesterol liver transplantation is the most definitive treat- levels and CHD risk is less steep in the elderly. ment providing a source of LDL receptor activity However, the number of CHD events is much but at the cost of life-long immunosuppression. A greater and the risk attributable to cholesterol is combined heart/liver transplant was originally per- therefore high. Currently there is a view that on September 24, 2021 by guest. Protected formed in 1984 in a 7-year-old patient. The primary prevention with hypolipidaemic agents cholesterol fell from a preoperative concentration should not be undertaken after the age of65. Trials of 25 mmol/l to 7 mmol/1.33 beginning in the United States of lipid-lowering in the elderly will help to clarify the position with Plasma exchange andrelated techniques regard to this increasing sector of the population. Plasma exchange was first used in 1974 for the Women treatment of homozygous familial hyperchol- esterolaemia.34 Regular exchanges (weekly or bi- Premenopausal women are at lower risk of CHD weekly) led to regression ofxanthomata and slow- than men and the major trials of lipid-lowering ing of progression of atherosclerosis. Recently therapy have been confined to men. More research more selective procedures have been developed is needed in more clearly defining individual risk in including LDL apheresis with disposable affinity women, particularly as caution is required with the columns. These columns remove apoprotein B prescription of most hypolipidaemic drugs ifpreg- containing lipoproteins including Lp(a), but HDL nancy is likely. is conserved.35 An alternative technique involves In postmenopausal women the risk of CHD 368 D.J. BETTERIDGE et al. Postgrad Med J: first published as 10.1136/pgmj.69.811.359 on 1 May 1993. Downloaded from increases steeply, possibly related to increasing reduce the risk of premature CHD, i.e. before the plasma LDL cholesterol levels. Changes in plasma age of 65. Diagnosis of the cause of raised plasma lipoprotein concentrations are ameliorated by hor- lipid levels will enable appropriate decisions to be mone replacement therapy. More information is taken with regard to management. The cornerstone required on the effects of combined hormone of treatment is nutritional counselling and atten- therapy (progestogen and oestrogen) on vascular tion to other major risk factors for CHD, partic- risk and on whether the progestogen, necessary for ularly smoking and hypertension. For a small the prevention ofendometrial hyperplasia, reduces percentage of patients with severe hyperlipidaemia the benefit conferred by the oestrogen. drug therapy is indicated. Appropriate drug choices need to be made based on the particular Children lipid abnormality to be treated. In general those patients with clinical vascular disease are treated Children from families carrying the gene for more aggressively than those where the aim is familial hypercholesterolaemia (FH) should be primary prevention. screened for the presence ofhypercholesterolaemia More research is needed to determine individual at an early age. It is generally accepted that a risk more precisely and to allow proper targeting of low-fat, low-cholesterol diet can be prescribed after therapy. Genetic factors, qualitative changes in the age of 5 years. However, more information is lipoproteins, lipoprotein (a), fibrinogen, and other needed on the safety of systemically active coagulation and thrombotic factors are likely to be hypolipidaemic drug therapy in children. In FH important in individual risk assessment. families where the clinical onset of CHD is early it There is no doubt that more information is may be necessary to prescribe drug therapy in needed from prospective studies of lipid-lowering children but most physicians would wait until after therapy in terms ofrisk benefit for affected individ- puberty. uals. Hopefully the major studies currently under- way will fill some of the gaps in our knowledge. Until then aggressive therapy with drugs should be Summary reserved for those at highest risk where the benefit is likely to be greatest. copyright. There is considerable evidence to suggest that the identification and treatment of dyslipidaemia will

References 1. Martin, M.J., Hulley, S.B., Browner, W.S., Kuller, L.H. & 9. Blankenhorn, D.H., Nessim, S.A., Johnson, R.L., Sanmarco, Wentworth, D. Serum cholesterol, blood pressure and mor- M.E., Azen, S.P. & Cashin-Hemphill, L. Beneficial effects of http://pmj.bmj.com/ tality: implications from a cohort of 361,662 men. Lancet combined colestipol-niacin therapy on coronary atherosc- 1986, ii: 933-936. lerosis and coronary venous bypass grafts. JAMA 1987, 257: 2. Chen, Z., Peto, R., Collins, R., MacMahon, S., Lu, J. & Li, 3233-3240. W. Serum cholesterol concentration and coronary heart 10. Brown, G., Albers, J.J. & Fisher, L.D. Regression of disease in population with low cholesterol concentrations. Br coronary artery disease as a result of intensive lipid lowering Med J 1991, 303: 276-282. therapy in men with high levels ofapolipoprotein B. N EnglJ 3. Kannel, W.B., Gordon, T. & Castelli, W.P. Role of lipid and Med 1990, 323: 1289-1298. lipoprotein fractions in assessing atherogenesis. The Fram- 11. Kane, J.P., Malloy, M.J., Ports, T.A., Phillips, N.R., Diehl,

ingham Study. Progr Lipid Res 1981, 20: 339-348. J.C. & Havel, R.J. Regression of coronary atherosclerosis on September 24, 2021 by guest. Protected 4. Castelli, W.P., Garrison, R.J., Wilson, P.W.F., Abbott, R.D., during treatment of familial hypercholesterolaemia with Kalousdian, S. & Kennel, W.B. Incidence of coronary heart combined drug regimens. JAMA 1990, 264: 3007-3012. disease and lipoprotein cholesterol levels. JAMA 1986, 256: 12. Watts, G.F., Lewis, B., Brunt, J.N.H. et al. Effects on 2835-2838. coronary artery disease of lipid-lowering diet, or diet plus 5. Hulley, S.B. & Avins, A.L. Asymptomatic hypertri- cholestyramine, in the St. Thomas' Atherosclerosis Regres- glyceridaemia. Br Med J 1992, 304: 394-396. sion Study (STARS). Lancet 1992, 339: 563-569. 6. Rossouw, J.E. & Rifkind, B.M. Does lowering serum 13. Brown, M.S. & Goldstein, J.L. Receptor-mediated control of cholesterol levels lower coronary heart disease risk? In: cholesterol metabolism. Science 1986, 191: 150-154. LaRosa, J.C. (ed) Endocrinology and Metabolism Clinics of 14. Consensus Conference. Lowering blood cholesterol to pre- North America. Lipid Disorders, Vol. 19, no. 2. W.B. vent heart disease. JAMA 1985, 253: 2080-2086. Saunders, Philadelphia, 1990, pp. 279-297. 15. Study Group of the European Atherosclerosis Society. 7. Rossouw, J.E., Lewis, B. & Rifkind, B.M. The value of Strategies for the prevention of coronary heart disease, a lowering cholesterol after myocardial infarction. N Engl J policy statement of the European Atherosclerosis Society. Med 1990, 323: 1112-1119. Eur Heart J 1987, 8: 77-88. 8. Levy, R.I., Brensike, J.F., Epstein, S.E. et al. The influence of 16. Study Group of the European Atherosclerosis Society. The changes in lipid values induced by cholestyramine and diet on recognition and management of hyperlipidaemia in adults: a progression ofcoronary artery disease: results ofthe NHLBI policy statement of the European Atherosclerosis Society. type II coronary intervention study. Circulation 1984, 69: Eur Heart J 1988, 9: 571-600. 325-337. GUIDELINES IN MANAGEMENT OF HYPERLIPIDAEMIA 369 Postgrad Med J: first published as 10.1136/pgmj.69.811.359 on 1 May 1993. Downloaded from

17. Report of the National Cholesterol Education Program 29. Grundy, S.M. HMG-CoA reductase inhibitors for treatment Expert Panel on Detection, Evaluation and Treatment of of hypercholesterolaemia. N Engl J Med 1988, 319: 24-33. High Blood Cholesterol in Adults. Arch Intern Med 1988,148: 30. Assmann, G., Betteridge, D.J., Gotto, A.M. & Stainer, G. 36-69. Management of hypertriglyceridaemic patients. Treatment 18. Shepherd, J., Betteridge, D.J., Durrington, P.N. et al. classifications and goals. Am J Cardiol 1991, 68: 30A-34A. Strategies for reducing coronary heart disease and desirable 31. Buchwald, H. Lowering ofcholesterol absorption and blood limits for blood lipid concentrations: guidelines of the British levels by ileal exclusion. Circulation 1964, 29: 713-720. Hyperlipidaemia Association. Br Med J 1987, 295: 32. Buchwald, H., Varco, R.L., Matts, J.P. et al. Effect ofpartial 1245-1246. ileal bypass surgery on mortality and morbidity from cor- 19. Friedewald, W.T., Levy, R.I. & Fredrickson, D.S. Estimation onary heart disease in patients with hypercholesterolaemia. of the concentration of low density lipoprotein cholesterol in Report of the Program on the Surgical Control of the plasma, without use of the preparative ultracentrifuge. Clin Hyperlipidaemias (POSCH). N Engl J Med 1990, 323: Chem 1972, 18: 499-502. 946-955. 20. Laker, M.F., Reckless, J.P.D., Betteridge, D.J. et al. 33. Starzl, T.E., Bilheimer, D.W., Bahnson, H.T. et al. Laboratory facilities for investigating lipid disorders in the Heart-liver transplantation in a patient with familial hyper- United Kingdom: results of the British Hyperlipidaemia cholesterolaemia. Lancet 1984, i: 1382-1383. Association Survey. J Clin Pathol 1992, 45: 102-105. 34. Thompson, G.R., Lowenthal, R. & Myant, N.B. Plasma 21. Beaumont, J.L., Carlson, L.A., Cooper, G.R., Fejfar, Z., exchange in the management of homozygous familial hyper- Fredrickson, D.S. & Strasser, T. Classification of hyper- cholesterolaemia. Lancet 1975, 1: 1208-1211. lipidaemias and hyperlipoproteinaemias. Bull World Health 35. Yokoyama, S., Hayashi, R., Santani, M. & Yamamoto, A. Organisation 1970, 43: 891-915. Selective removal of low density lipoprotein by plas- 22. Betteridge, D.J. Lipids, diabetes and vascular disease: the mapheresis in familial hypercholesterolaemia. Arterio- time to act. Diabet Med 1989, 6: 195-218. sclerosis 1985, 5: 613-622. 23. Short, C.D. & Durrington, P.N. Hyperlipidaemia and renal 36. Fuchs, C., Windisch, M., Wieland, H. et al. Selective disease. In: Betteridge, D.J. (ed) Clinical Endocrinology and continuous extracorporeal elimination of low density lipo- Metabolism. Lipid and Lipoprotein Disorders, Vol. 4, no. 4, proteins from plasma by heparin precipitation without Bailliere Tindall, London, 1990, pp. 777-806. cations. In: Plasma Separation and Plasma Fractionation. 24. O'Connor, P., Feely, J. & Shepherd, J. Lipid lowering drugs. Karger, Basel, 1983, pp. 272-280. Br Med J 1990, 300, 667-672. 37. Holme, I. An analysis of randomized trials evaluating the 25. Kita, T., Nagano, Y., Yokode, M. et al. Probucol prevents the effect ofcholesterol reduction on total mortality and coronary progession of atherosclerosis in Watanabe heritable hyper- heart disease incidence. Circulation 1990, 82: 1916-1924. lipidaemic rabbit, an animal model for familial hyper- 38. Muldoon, M.F., Manuck, S.B. & Matthews, K.A. Lowering cholesterolaemia. Proc Natl Acad Sci USA 1987, 84: cholesterol concentrations and mortality: a quantitative

5928-5231. review of primary prevention trials. Br Med J 1990, 301: copyright. 26. Report from the Committee of Principal Investigators. A 309-314. cooperative trial in the primary prevention ofischaemic heart 39. Smith, G.D. & Pekkanen, J. Should there be a moratorium on disease using clofibrate. Br Heart J 1978, 40: 1069-1118. the use of cholesterol lowering drugs? Br Med J 1992, 304: 27. Frick, M.H., Elo, O., Haapa, K. et al. Helsinki Heart Study: 431-434. primary prevention trial with gemfibrozil in middle-aged men with dyslipidaemia. N Engl J Med 1987, 317: 1237-1245. 28. O'Kane, M.J., Trinick, T.R., Tynan, M.B., Trimble, E.R. & Nicholls, D.P. A comparison of acipimox and nicotonic acid in type 2b hyperlipidaemia. Br J Clin Pharmac 1992, 33:

451-453. http://pmj.bmj.com/ on September 24, 2021 by guest. Protected