SEMINAR Seminar Essential hypertension Jan A Staessen, Jiguang Wang, Giuseppe Bianchi, Willem H Birkenhäger Hypertension is a frequent, chronic, age-related disorder, which often entails debilitating cardiovascular and renal complications. Blood pressure is usually noted in combination with other cardiovascular risk factors. Diagnosis of hypertension increasingly relies on automated techniques of blood pressure measurement. The pathophysiology of essential hypertension depends on the primary or secondary inability of the kidney to excrete sodium at a normal blood pressure. The central nervous system, endocrine factors, the large arteries, and the microcirculation also have roles in the disorder. Although monogenic forms of blood pressure dysregulation exist, hypertension mostly arises as a complex quantitative trait that is affected by varying combinations of genetic and environmental factors. Non- pharmacological strategies can reduce blood pressure. Antihypertensive drug treatment diminishes the complications of hypertension. The concept that a few major genes will provide the final clue to the pathogenesis of essential hypertension is an oversimplification that contradicts the heterogeneous nature of this disorder. Further integration of genetic, molecular, clinical, and epidemiological research could disclose subsets of patients in whom specific combinations of genetic and environmental factors raise blood pressure, and might lead to more individualised treatment. Essential hypertension refers to a lasting increase in blood Epidemiology pressure with heterogeneous genetic and environmental In cross-sectional and longitudinal population studies, causes. Its prevalence rises with age, irrespective of the systolic blood pressure increases with age until the eighth type of blood pressure measurement and the operational decade of life (figure 1). By contrast, diastolic blood thresholds used for diagnosis (panel 1). In developed and pressure rises only until 50 years of age, after which it developing countries alike, essential hypertension affects either becomes constant or even decreases slightly. In the 25–35% of the adult population, and up to 60–70% of Framingham Heart Study,1 increasing age entailed a shift those beyond the seventh decade of life. Hypertension from diastolic pressure to systolic pressure and then to aggregates with other cardiovascular risk factors, such as pulse pressure as the main predictor of cardiovascular abdominal obesity, dyslipidaemia, glucose intolerance, risk. Below age 50 years, diastolic pressure was the hyperinsulinaemia, and hyperuricaemia, possibly because strongest predictor. Age 50–59 years was a transition of a common underlying cause. period when all three blood pressure indices were similar We will review developments in the epidemiology, predictors, and from 60 years on, diastolic pressure was diagnosis, pathophysiology, genetics, and treatment of negatively related to the risk of coronary events, so that essential hypertension. During the past decade, these pulse pressure became a superior predictor to systolic disciplines have progressed towards substantial new pressure.1 The finding that in middle-aged and older insights. Nevertheless, despite relentless research efforts, patients cardiovascular outlook gets worse with raised the specific causes of essential hypertension remain pulse pressure—rather than mean pressure—is consistent incompletely understood. We will therefore emphasise the in men and women, in treated and untreated need for improved integration of multidisciplinary hypertensive individuals,2 and in patients with a history of research to clarify the pathophysiology of this complex myocardial infarction3 or renal failure.4 Furthermore, in and multifaceted disorder. older patients with isolated systolic hypertension, ambulatory pulse pressure is a stronger predictor of cardiovascular risk than is pulse pressure measured by conventional sphygmomanometry, whereas mean pressure regardless of the type of blood pressure measurement is not predictive.5 Lancet 2003; 361: 1629–41 Search strategy Studiecoördinatiecentrum, Hypertensie en Cardiovasculaire We identified original research papers, reviews, and editorial Revalidatie Eenheid, Departement voor Moleculair en comments by going through leading journals that publish Cardiovasculair Onderzoek, Katholieke Universiteit Leuven, Leuven, basic and clinical research in the field of hypertension, and Belgium (J A Staessen MD, J Wang MD); Divisione di Nefrologia by electronically searching the Medline database. We Dialisi e Ipertensione, Ospedale San Raffaele, Dipartimento di directed special attention to papers published since 1997. Scienze e Techologie Biomediche, Universitá Vita-Salute, Milano, Italy (Prof G Bianchi MD); and Erasmus University, Rotterdam, We reviewed consensus documents on automated blood Netherlands (Prof W H Birkenhäger MD) pressure measurement as well as guidelines for the management of hypertension. We also looked for research Correspondence to: Dr Jan A Staessen, Study Coordinating Centre, developments by discussions with our collaborators and with Laboratory of Hypertension, University of Leuven, Campus specialists in the clinical or fundamental aspects of Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium cardiovascular medicine (e-mail: email@example.com) THE LANCET • Vol 361 • May 10, 2003 • www.thelancet.com 1629 For personal use. Only reproduce with permission from The Lancet Publishing Group. SEMINAR Panel 1: Systolic and diastolic blood pressure thresholds used in clinical medicine Conventional blood pressure measurement* Automated blood pressure measurement† Category Boundaries Type of measurement P95‡ Normotension Hypertension Normotension Ambulatory Optimum <120/<80 24-h 132/82 <130/<80 >135/85 Normal 120–129/80–84 Daytime 138/87 <135/<85 >140/90 High-normal 130–139/85–89 Night-time 123/74 <120/<70 >125/75 Hypertension Self-recorded Grade I (mild) 140–159/90–99 Morning 136/85 <135/<85 >140/>90 Subgroup borderline 140–149/90–94 Evening 138/86 <135/<85 >140/>90 Grade II (moderate) 160–179/100–109 Both 137/85 <135/<85 >140/>90 Grade III (severe) >180/>110 Isolated systolic hypertension >140/<90 Subgroup borderline 140–149/<90 Blood pressure thresholds are in mm Hg. *Consensus classification proposed by European Society of Hypertension/European Society of Cardiology guidelines.10 †Classification proposed at the 8th International Consensus Conference on Ambulatory Blood Pressure measurement (Sendai, Japan, October 2001). ‡Mean of 95th percentiles in subjects who on conventional blood pressure measurement were normotensive in large-scale studies. Most epidemiologists6,7 share the point of view that the European and American people is coronary heart disease, relation between risk of cardiovascular complications and whereas in Asian and older people it is stroke.8 Black blood pressure is continuous, without a threshold above people tend to have higher blood pressure and which the rate of disease would suddenly increase. The hypertension-related mortality rates than other main complication of hypertension in middle-aged individuals.9 The Framingham investigators reported that Men (n=833) Women (n=859) 150 Systolic Conventional BP at home Systolic Daytime ambulatory BP 140 130 120 Blood pressure (mm Hg) 110 100 Diastolic Diastolic 90 80 70 60 50 4 9 4 9 4 9 4 9 4 9 4 9 4 4 4 9 4 9 4 9 4 9 4 9 4 9 4 4 –1 –1 –2 –2 –3 –3 –4 –4 –5 –5 –6 –6 –7 –8 –1 –1 –2 –2 –3 –3 –4 –4 –5 –5 –6 –6 –7 –8 10 15 20 25 30 35 40 45 50 55 60 65 70 75 10 15 20 25 30 35 40 45 50 55 60 65 70 75 Age group (years) n=28 44 38 62 78 75 108 84 66 79 61 50 33 27 n=30 45 28 67 67 113 105 82 63 86 64 44 38 27 Figure 1: Systolic and diastolic blood pressures (BP) in 5-year age groups in a representative sample of the population of Noord Limburg, Belgium n=number of patients. Closed symbols show the mean of ten conventional blood pressure readings obtained at two visits by the study nurses in the patients’ homes. Open symbols are mean daytime (10–20 h) ambulatory blood pressure values recorded every 20 min by oscillometric recorders. 1630 THE LANCET • Vol 361 • May 10, 2003 • www.thelancet.com For personal use. Only reproduce with permission from The Lancet Publishing Group. SEMINAR predictive superiority of ambulatory over conventional GLOSSARY blood pressure. Additionally, a blunted fall in nocturnal EPIGENETIC MECHANISM blood pressure is a harbinger of an unfavourable outcome, A term used to describe non-mutational occurences, such as DNA irrespective of whether the decrease is assessed as a methylation and histone modification, which change the expression of a continuous18 or a dichotomous characteristic.19 In clinical gene. Histones are proteins around which double-stranded DNA is coiled practice, the commonly used definition of white-coat and make up the core of a nucleosome, the most fundamental unit of hypertension is raised clinic blood pressure in the DNA packaging in the nucleus. presence of a normal daytime ambulatory blood IMPRINTS pressure.20,21 Results of event-based studies16,22 have clearly Genetic imprinting is any mechanism by which individual cells express shown that the risk of cardiovascular disease is lower in the maternal or paternal allele of a gene, but not both. Although patients with white-coat hypertension than in those with commonly used in relation to autosomal genes, X chromosome raised ambulatory blood pressure, even after controlling inactivation is a form of imprinting. for concomitant risk factors. When daytime ambulatory PENETRANCE blood pressure was below 130 mm Hg systolic and The extent to which a person carrying a defined allele has an altered 80 mm Hg diastolic, the incidence of cardiovascular phenotype. disorders did not differ between normotensive people and RECOMBINATION MAXIMA those with white-coat hypertension.16,20 The counterpart of Recombination is the crossover of chromosome fragments between the white-coat hypertension is masked hypertension, a disorder maternal and paternal homologues of each chromosome pair during the characterised by a normal conventional blood pressure meiosis, the specialised cell division giving rise to sperm and egg cells. confirmed on repeated clinic visits, but a raised daytime Recombination maximum refers to the location along the chromosome ambulatory blood pressure.23,24 Preliminary observations at which recombination is most likely to occur. suggest that this condition is associated with more target SYNTENY organ damage than that noted in patients with sustained The property of genes to be localised on the same chromosome across normotension.23,24 species. Conservation of synteny shows that a group of genes which is Self-measurement of blood pressure at home is cheaper on one chromosome in one species are similarly linked in a second species. than ambulatory monitoring. To pass validation,14 monitors for self-measurement must conform to the same quality standards as devices for ambulatory monitoring. Use of high-to-normal blood pressure, as defined in panel 1, is wrist devices is not recommended.14 To ensure a reliable associated with an increased risk of cardiovascular disease6 report of home blood pressure readings as instructed by a and incidence of hypertension.11 Compared with optimum professional, values must be downloaded from the memory blood pressure, high-normal blood pressure was of a device or printed. If applied correctly, self- associated with a risk-factor-adjusted hazard ratio for measurement accomplishes several of the advantages of cardiovascular disease of 2·5 in women and 1·6 in men.6 ambulatory monitoring, including the increased number of Similarly, the incidence of hypertension rose stepwise readings, the absence of the white-coat syndrome, and across the non-hypertensive blood pressure categories. (when automated devices are used) the absence of observer Below age 65, the 4-year rates were 5·3% in patients with bias. Furthermore, self-measurement could increase optimum blood pressure, 17·6% in those with normal compliance to prescribed drugs and reduce the number of blood pressure, and 37·3% in those with high-normal clinic visits required for the diagnosis and the treatment of blood pressure.11 Corresponding 4-year rates of hypertension.25 Diagnostic thresholds have been proposed progression to hypertension in subjects of 65 years and (panel 1),26 but they have not yet been widely validated by older were 16·0%, 25·5%, and 49·5%, respectively.11 prospective outcome studies.27 Conventional measurement remains the standard Diagnosis method for assessment of blood pressure. In patients with For the diagnosis of hypertension, doctors rely mainly on raised clinic blood pressure and either target organ damage Riva-Rocci’s technique and the auscultation of the or a high cardiovascular risk profile, treatment can be Korotkoff sounds. Raised clinic blood pressure readings started without confirmation of the increased clinic have to be confirmed on at least two visits with intervals readings by automated measurement techniques. However, dependent on the height of the blood pressure, presence when raised blood pressure is the only detectable of target organ damage, and other cardiovascular risk abnormality or when patients with a normal clinic blood factors.10,12,13 Nowadays, automated blood pressure pressure show unexplained target-organ damage, self- measurement is increasingly used in clinical practice, measurement, ambulatory monitoring, or both must be mainly for diagnosis of the white-coat syndrome, which used to exclude white-coat hypertension or masked consists of a transient rise of a patient’s blood pressure in hypertension, respectively (figure 2). In the absence of response to the medical environment or the observer other cardiovascular risk factors and signs of target organ recording the blood pressure. The accuracy of devices for damage, treatment of patients with white-coat measurement of blood pressure should be assessed in hypertension can be restricted to lifestyle measures and validation studies according to internationally recognised yearly follow-up of ambulatory blood pressure.28 protocols14 and published in peer-reviewed journals. Diagnostic thresholds are available for different types of Pathophysiology automated blood pressure measurement in adults Sodium and fluid balance and vasomotor tone are (panel 1), whereas those for children and adolescents are cornerstones in blood pressure regulation. Both still under discussion.15 mechanisms are affected by numerous genetic and Ambulatory monitoring substantially refines the risk environmental factors, and are controlled by hormonal, stratification provided by conventional sphygmomanom- nervous system, paracrine, and intracellular feedback etry, both in hypertensive patients16 and in the general loops. The interactions between these factors change with population.17 The greater number of measurements, the age, and account for the heterogeneous pattern of the absence of digit preference and observer bias, and the haemodynamic alterations that lead to and sustain high reduction of the white-coat effect contribute to the blood pressure throughout life. THE LANCET • Vol 361 • May 10, 2003 • www.thelancet.com 1631 For personal use. Only reproduce with permission from The Lancet Publishing Group. SEMINAR White-coat hypertension Masked hypertension exchange and, through calcium- (Isolated clinic hypertension) (Isolated ambulatory dependent pathways, stimulate hypertension) excitation-contraction coupling and the expression of growth-related genes.34,38 At very low concentrations Persistently raised Persistently normal within the nanomolar range, clinic blood pressure clinic blood pressure endogenous ouabain may increase the size of the membrane pool of Yes active sodium pumps43 and by this or Target organ damage? Target organ damage? other unknown mechanisms lead to renal sodium conservation rather than No Yes loss of salt.39 High High Home Start 24-h ambulatory Stimulation of the renal sympathetic Low treatment blood pressure nerves and activation of the renin- angiotensin system promote sodium 24-h ambulatory High and fluid retention. Interactions blood pressure between both pathways of sodium handling occur intrarenally as well as Low extrarenally. Circulating angiotensin II Continue to monitor Continue to monitor facilitates the presynaptic release of clinic and home clinic and ambulatory noradrenaline.40 It can also directly blood pressure blood pressure increase sympathetic outflow from the subfornical organ and the area Figure 2: Guidelines for assessment of patients by use of clinic, home, and ambulatory postrema, two specialised areas within monitoring of blood pressure the central nervous system that lack a Modified from reference 21. For definitions of normotension and hypertension, see panel 1. blood-brain barrier.44 Furthermore, angiotensin II itself acts as a Sodium and fluid balance peptidergic neurotransmitter in the brain.45 Angiotensin II, According to pioneering research29 and subsequent originating from neurons in the paraventricular nucleus, evidence,30 the kidneys play a central part in the attenuates the reflex modulation of sympathetic outflow pathophysiology of essential hypertension. Blood pressure from the brain and activates sympathetic preganglionic starts to rise when the kidney requires a higher than usual neurons. These central effects of angiotensin II are greatest blood pressure to maintain extracellular fluid volume during a low sodium diet (high renin activity), least during within normal limits.29 Models of essential hypertension in high sodium intake (low renin activity), and tonic during rats share the characteristic that renal sodium excretion is normal sodium diet.40 impaired at any degree of blood pressure.31 In rats32 and man,33 transplantation of a kidney from a normotensive Microvascular and macrovascular mechanisms donor reduces the blood pressure. Such findings32,33 do not Increased peripheral arterial resistance is the hallmark of suggest that the disturbance causing essential essential hypertension. The active component of arteriolar hypertension only affects the kidneys. Indeed, essential resistance depends on the contraction of vascular smooth hypertension is characterised by generalised membrane muscle cells. Structural changes in the wall-to-lumen ratio abnormalities,34 which could affect the function of many are termed vascular remodelling.46 Histopathological organs in various ways. studies of gluteal arterioles mounted on wire myographs The central role of the kidney in the pathogenesis of have shown that resistance vessels of hypertensive patients hypertension should be differentiated from salt sensitivity. have a reduced lumen, an increased media-to-lumen ratio, Secondary forms of hypertension, such as but a normal medial cross-sectional area.46 Experimental pheochromocytoma or hyperaldosteronism, can also be evidence favours the notion that these morphological associated with a reduced ability of the kidneys to excrete alterations arise as an adaptation to the increased blood a sodium load. In such cases, removal of the primary pressure through changes in the neurohumoral milieu, cause of hypertension corrects the enhanced renal sodium and that they amplify rather than cause increase of blood reabsorption. In genetic forms of hypertension, salt- pressure.47 Vascular hypertrophy associated with sensitivity may result from various mutations affecting hypertension can also lead to closure of small vessels.48 cytoskeleton proteins,35,36 ion transporters,34 or endocrine Vascular rarefaction could therefore contribute to the factors37,38 that control renal sodium handling. In other increased vascular resistance in long-standing forms of hypertension, salt-sensitivity originates in an hypertension. imbalance between the hormonal,39 nervous,40 or Isolated systolic hypertension in older patients is a haemodynamic29 regulations that affect sodium balance separate disease entity due to stiffening of the large through changes in glomerular filtration or tubular arteries with advancing age. Aortic pulse wave velocity, an reabsorption. index of arterial stiffness, doubles between 15 and The hormone ouabain is released endogenously from 70 years of age.49 The arterial pressure wave consists of a the adrenal glands and possibly from the midbrain41 in forward component generated by the heart and reflected response to hypoxia41 or expansion of the extracellular waves returning to the heart from peripheral sites.4,49 In fluid volume induced by aldosterone.42 Evidence suggests healthy young adults, the reflected waves coincide with that, dependent on its concentration, this hormone diastole, raise diastolic pressure, and boost coronary behaves as a versatile modulator of the ubiquitously perfusion. With arterial stiffening, which arises with expressed sodium pump.39 In renal tubular cells, ageing or in advanced hypertension, the reflected waves inhibition of Na /K ATPase activity by ouabain move faster, reach the proximal aorta during systole, and promotes natriuresis. In cardiac and vascular myocytes, cause an augmentation of late systolic pressure, whereas this process could slow the sarcolemmal Na /Ca2 diastolic pressure decreases.4,49 Lowered diastolic pressure 1632 THE LANCET • Vol 361 • May 10, 2003 • www.thelancet.com For personal use. Only reproduce with permission from The Lancet Publishing Group. SEMINAR can impair coronary blood flow and predispose to coefficients between relatives, usually in the range myocardial ischaemia. Increased systolic pressure 0·1–0·3. Parent-offspring correlations tend to be smaller augments cardiac work and can lead to heart failure. The than those among siblings. Heritability estimates, age-related changes in the elastic properties of the large generally based on complex biometrical assumptions and arteries and in the wave reflections also account for the Mendelian mechanisms of inheritance, cluster around greater brachial artery pressure than central aortic 20% in family studies, but in twin studies they vary pressure in young patients, whereas in the elderly or in around 60%.50 patients with end-stage renal disease4 both values tend to be similar. Monogenic forms of blood pressure dysregulation The last decade witnessed tremendous progress in high- Genetics of human hypertension throughput genomics and proteomics, which allow many More than 40 years ago, at the time of the controversy samples to be processed in a short time, as well as in cell between Sir George Pickering and Robert Platt, and molecular biology. Technological advances led to the researchers had already noticed that hypertension runs in discovery of 17 human genes (panel 2)52 that cause families. Population-based studies show correlation Mendelian forms of either hypertension or hypotension. Panel 2: Mendelian forms of blood pressure dysregulation Syndrome* Phenotype† Mutation‡ Mechanism Glucocorticoid remediable HT, serum K ↓, alkalosis, PRA↓, Chimerical gene containing Chimerical enzyme is aldosteronism (GRA) aldosterone suppressible by promoter area of AS and regulated by ACTH dexamethasone coding region of 11 H (dominant) Apparent mineralocorticoid HT, aldosterone↓, serum K ↓, Absence of 11 HSD Impaired conversion of excess (AME) alkalosis, PRA↓, suppressible by (recessive) cortisol (activates MR) to spironolactone cortisone (inactive at MR) Hypertension exacerbated by HT accelerated in pregnancy, PRA↓ Missense mutation in MR Steroids without 21-hydroxyl pregnancy37 receptor (dominant) group (progesterone) and spironolactone stimulate mutant MR receptor Dominant pseudohypo- BP↓, aldosterone↑, neonatal salt Mutant MR (usually dominant, Early in life, normal salt aldosteronism type 1 wasting, serum K ↑, acidosis, but recessive variant homoeostasis requires two (PHA 1)51 renal phenotype disappears in identified) normal copies of MR adulthood Recessive pseudohypo- BP↓, aldosterone↑, neonatal salt Mutated , , or subunit Loss of function of ENaC aldosteronism type 1 wasting, serum K ↑, acidosis, of ENaC (recessive) (PHA 1) phenotype does not improve with ageing, impaired clearance of lung water (respiratory failure) Pseudohypoaldosteronism HT, serum K ↑, PRA↓, normal Mutations in at least one of Volume-expanded HT, type 2 (PHA 2) renal function 3 genes mapped to 1q31–42, probably due increased renal 12p13, and 17p11–q21 tubular sodium reabsorption (dominant) Liddle syndrome Early onset HT, aldosterone↓, Mutations of or subunit of Increased ENaC activity due serum K ↓, alkalosis, PRA↓ ENaC (dominant) to reduced ENaC clearance Hypertension with Severe HT, abnormal skeletal Mutations mapped to Unknown brachydactyly development of hand, stroke 12p11.2–12.2 (dominant) Peroxisome proliferator- HT associated with insulin Missense mutation (dominant) Desensitisation of PPAR activated receptor resistance or diabetes, responsive receptor gamma (PPAR ) to thiazolidinediones Defective aldosterone BP↓, aldosterone↓, Na1loss, Deficient AS or 21 hydroxylase Aldosterone deficiency synthesis K retention (recessive) Gitelman’s syndrome BP↓, serum Mg2 ↓, serum K ↓, Mutation of the thiazide- Loss of function alkalosis, hypocalciuria, PRA↑, sensitive NaCl cotransporter aldosterone↑ in the distal collecting duct (recessive) Bartter’s syndrome BP↓, serum K ↓, alkalosis, Mutation in ion transporters Loss of function hypercalciuria, PRA↑, aldosterone↑ in thick ascending limb of Henle: Na -K -2Cl , ATP-sensitive K channel, or Cl channel (recessive) AS=aldosterone synthase. BP=blood pressure. ENaC=epithelial sodium channel. 11 H=11 -hydroxylase. 11 HSD=11 -hydroxysteroid dehydrogenase. HT=hypertension. MR=mineralocorticoid receptor. PRA=plasma renin activity. *Name of the syndrome and commonly used acronym. Original articles are listed in reference 52. †Main characteristics of the phenotype. ‡Change in the gene or gene product. Dominant or recessive refer to the pattern of autosomal inheritance. THE LANCET • Vol 361 • May 10, 2003 • www.thelancet.com 1633 For personal use. Only reproduce with permission from The Lancet Publishing Group. SEMINAR Phenotype Effect A second layer of complexity in the study of blood pressure as a quantitative trait originates from the Nucleus Multiple genes inconsistency of the relationship between phenotype and genotype. Many genes might influence the same phenotype, or a sole mutation could be associated with Proteins and grossly discordant phenotypes.64 Even in monogenic subcellular IP disorders with typical Mendelian inheritance (panel 2), organelles PENETRANCE may vary according to race,55 age,51,55 dietary salt intake,51 hormonal milieu,37,38 or just from one case to Host factors another.64 Mendelian principles or the biometrical Cell IP (race, sex, age, etc) assumptions underlying the decomposition of variance into genetic, environmental, and random components, might not apply to multigenic quantitative traits. Indeed, Tissue IP cardiovascular phenotypes are affected by distinct maternal and paternal genomic IMPRINTS that result in a divergent expression of parental alleles during fetal Whole BP Environment development and postnatal growth into adulthood.65 organism HT Lifestyle Offspring inherit their mitochondrial DNA mainly66 but Figure 3: Schematic representation of blood pressure as a not exclusively67 from their mother. Maternal and paternal complex multigenic trait, affected by host and environmental chromosomes show differences in location of factors and feedback control RECOMBINATION MAXIMA.68 The intrauterine environment IP=intermediate phenotypes. BP=blood pressure. HT=hypertension. might affect the occurrence of cardiovascular disease later Curved arrows represent feedback loops. in life and is largely controlled by genetic and environmental factors linked to the mother.69 Finally, However, monogenic disorders of blood pressure epigenetic mechanisms may lead to transdifferentiation, regulation are rare and do not explain blood pressure whereby specialised cells can undergo profound changes variability in the population at large.53 in their function.70 How, then, can researchers cope with the complexity of Blood pressure as a polygenic quantitative trait human essential hypertension? Some experts favour the Until now, geneticists have failed to identify common whole genome approach71–74 and exploit SYNTENY between genes with large effects on human hypertension. It is species75 in the search for major genes that could lead to conceivable that such genes do not exist and that blood hypertension. Inbred rats and genetically-engineered pressure is dependent on a mosaic of many loci, each mice, in which gene activities have been specifically with a small influence or with a contribution differing manipulated, produced new insights in the long-term according to sex,54 race,55,56 age,51,55 or lifestyle. However, regulation of blood pressure.31,72 However, these rodent many studies57 have limitations such that the existence of models31,72 do not stand comparison with the complexity major genes controlling blood pressure cannot yet be and heterogeneous nature of the human disease. Other completely ruled out. Lack of standardisation and investigators verified prespecified hypotheses about arbitrary dichotomisation of a continuous and variable known candidate genes,76 such as those listed in phenotype, inappropriate selection of cases and controls, panel 33,57,71,72,74,77-91 and tried to reconstruct the population admixture and stratification, insufficient pathogenesis of hypertension from cells up to the systemic sample size, and failure to account for confounders, level. Under controlled laboratory conditions, they environmental factors, or EPIGENETIC MECHANISMS, measured intermediate phenotypes less distant from the remain key obstacles to substantial progress. DNA than blood pressure, such as modulation status.92 Other reasons could also account for the shortfall, but Non-modulators show blunted renovascular and adrenal they are generally underappreciated. First, in most responses to infused angiotensin II in sodium repleted and people, blood pressure behaves as an age-related sodium depleted conditions, respectively.92 Selective quantitative trait (figure 1). Advancing age increases salt blockade of molecular mechanisms in pharmacogenomic sensitivity,58 steepens the relationship between blood experiments may constitute the final leg of the long pressure and exchangeable body sodium,59 decreases the journey from bench to bedside.93 For instance, in patients gain of the baroreceptor reflex,60 reduces renal with a genetic predisposition to salt retention, diuretics perfusion,61 and compromises the buffering effects of the may lower blood pressure more than other treatments.93 large arteries on both systolic and diastolic blood However, when used in isolation, the pharmacogenomic pressure.62 Throughout life, genetically determined host approach also has weakness, because a drug rarely factors continuously interact with environmental possesses the necessary selectivity. Moreover, influences, including lifestyle (figure 3). Any resulting interindividual differences in pharmacokinetics could alter change in blood pressure is initially counteracted by self- drug effects irrespective of specific interference with organising homoeostatic mechanisms, which encompass mechanisms that regulate blood pressure.94 intracellular signalling, metabolic and hormonal The discovery of the role of adducin in human regulation at cell and tissue levels, as well as systemic hypertension illustrates the potential of multidisciplinary feedback loops involving the whole body. For instance, research along the lines of a predefined pathophysiological acutely induced hypertension causes a rapid shift in hypothesis. Adducin is a cytoskeleton protein.81,82 sodium pump activity from proximal to distal sites of the Investigations in rats,81,95 in-vitro transfection studies,35 nephron.63 Since findings obtained under controlled interventions in never-treated hypertensive patients,36 and experimental conditions or in defined populations cannot epidemiological studies96,97 have shown a logical sequence be easily generalised, this innate plasticity of living of events leading from a point mutation in the subunit95 organisms (figure 3) should be incorporated into the to a cellular dysfunction characterised by higher activity of search for genes that may cause essential hypertension in the sodium pump,35 hence increased tubular sodium man. reabsorption in the kidney,36 and ultimately 1634 THE LANCET • Vol 361 • May 10, 2003 • www.thelancet.com For personal use. Only reproduce with permission from The Lancet Publishing Group. SEMINAR hypertension.96 In caucasian people, the mutated 70 Hypertension 900 Femoral intima-media -adducin allele and the deletion polymorphism of the thickness ACE gene jointly predicted the incidence of hypertension 60 DD (n=63) 850 Incidence per 100 patients (figure 4).96 In cross-sectional analyses of the same ID+II (n=195) population, these two polymorphisms combined were 50 associated with raised serum creatinine concentration97, 800 increased femoral intima-media thickness,98 and 24-h 40 +59% proteinuria97 (figure 4). Such effects of mutated adducin µm p=0·004 750 are in keeping with renal cross-transplantation 30 experiments.32,33 The adducin hypothesis was further substantiated by the finding that the blood pressure was 700 20 lowered by a specific inhibitor of the sodium pump99 and by the observation that hypertensive carriers of the 650 p=0·03 mutated -adducin have reduced cardiovascular 10 (vs DD) complications if they are treated with diuretics instead of drugs that do not antagonise the enhanced tubular 0 600 sodium reabsorption.100 4 6 8 10 12 n=44 79 44 Years of follow-up II ID DD Intervention through lifestyle In chimpanzees given a vegetarian diet with very low sodium and high potassium content, salt repletion Proteinuria 95 Serum creatinine 0·12 (10–15 g per day for 20 months) caused a rise in blood pressure by 33 mm Hg systolic and 10 mm Hg diastolic.101 94 Intervention studies in man102 produced the most 0·11 93 convincing evidence for the role of salt in hypertension. A meta-analysis of 56 trials accounted for measurement 92 0·10 error of urinary sodium excretion. For a reduction of the 91 µmol /L daily sodium excretion by 100 mmol (about 6 g of salt), g/day the decline in blood pressure in hypertensive patients 90 0·09 averaged 3·7 mm Hg (95% CI 2·4–5·0) systolic and 89 0·9 mm Hg (–0·1 to 1·8) diastolic.102 The corresponding 0·08 decreases in normotensive subjects were small: 88 p<0·008 p=0·02 1·0 mm Hg (95% CI 0·5–1·6) and 0·1 mm Hg ( –0·3 to (vs ID or DD) 87 (vs DD) 0·5), respectively. In the Dietary Approaches to Stop 0·07 Hypertension (DASH) trial,103 participants were fed meals 86 with varying salt levels for more than 4 weeks. The DASH 85 0·06 diet was rich in fresh fruits, vegetables, and low-fat diary products. For both the DASH and traditional diets, the n=139 299 155 n=51 117 53 lower the salt intake, the lower was the blood pressure.103 II ID DD II ID DD In line with the concept of pressure-natriuresis,29 some studies demonstrated the restoration of a dipping diurnal Figure 4: Association between a complex phenotype and the blood pressure profile in non-dipping hypertensive ACE I/D polymorphisms in carriers of the mutated -adducin patients given a low-salt diet.104 However, high sodium Trp allele intake105-107 associated or not with sodium sensitivity,105,107 Individuals were recruited from a Flemish population in Belgium. The did not uniformly predict a worse cardiovascular outcome complex phenotype consisted of hypertension, intima-media thickness of the femoral artery, serum creatinine concentration, and proteinuria. The in prospective studies. Furthermore, sodium restriction total number of patients investigated for each phenotype amounted to does not normalise peripheral vascular resistance, the 678,96 380,98 1454,97 and 556,97 respectively. The corresponding main haemodynamic disturbance in essential hyper- associations in homozygous carriers of the wild-type -adducin were not tension.108 significant. Stopping excessive alcohol consumption (>30 mL ethanol per day)109 and restriction of caloric intake110 are by pressure. However, the main goal of antihypertensive far the most effective lifestyle measures that consistently treatment is not to reduce blood pressure per se, but to reduce blood pressure. In an overview of intervention prevent the cardiovascular and renal complications studies, a 1 kg loss of weight entailed an average blood associated with raised blood pressure, extend longevity, pressure decrease by 1·6 mm Hg systolic and 1·3 mm Hg and improve quality of life. Several quantitative diastolic.110 Other lifestyle measures that have the potential overviews2,113–115 provide a detailed account of the outcome to slightly diminish blood pressure are regular dynamic trials in hypertension, including study design, study exercise (30–45 min for at least 4 days per week)111 and quality, and the effects of treatments on primary and abstaining from smoking.112 These lifestyle measures are secondary endpoints. recommendable because they reduce not only blood pressure but also cardiovascular risk. Primary prevention Placebo-controlled trials of antihypertensive drug Antihypertensive drug treatment treatment in middle-aged and older hypertensive patients Agencies currently approve new drugs for with predominantly diastolic hypertension showed that a antihypertensive treatment on the basis of the treatment’s 5–6 mm Hg mean decline in diastolic pressure sustained potential to decrease blood pressure as an intermediate over 5 years reduced incidence of stroke by nearly 40% endpoint. Despite substantial evidence of the risk of and that of coronary events by 15%.113 In older patients isolated systolic hypertension in older people,2 regulations with isolated systolic hypertension, treatment diminished require lowering of both systolic and diastolic blood the stroke rate by one third and the incidence of coronary THE LANCET • Vol 361 • May 10, 2003 • www.thelancet.com 1635 For personal use. Only reproduce with permission from The Lancet Publishing Group. SEMINAR Panel 3: Putative candidate genes or quantitative trait loci implicated in blood pressure regulation Receptors Cellular or paracrine agents Hormones or vasoactive Other proteins or QTLs substances Angiotensin receptors 1b and , , subunits of adducin81,82 Renin, angiotensinogen, angiotensin- SA locus78 257,72 converting enzyme57 Atrial natriuretic peptide 1, 2, 1 isomers of Na K Atrial and brain natriuretic peptides QTLs associated with X- and receptor/guanylyl cyclase A72 ATPase, Na,K,2Cl-cotransporter83 Y-chromosomes72 Glucocorticoid receptor77 , , subunits of epithelial 11 -hydroxylase, aldosterone QTL on human 12q21.3374 sodium channel85 synthase Dopamine receptors 1, 2 and Renal epithelial potassium Endothelin 2 and 372 QTLs associated with Lewis 372,74 channel72 blood group and complement C3F89 1b, 2, and 2 adrenergic 3 subunit of G-protein79 Endothelial nitric oxide synthase84 QTL associated with human receptor72 pancreatic phospholipase A288 Insulin receptor91 Calmodulin-dependent protein Prostacyclin synthase87 Neuropeptide Y72 kinase II72 Leptin receptor71 Adipsin71 Growth hormone86 Tyrosine hydroxylase80 Interleukin 672 Corticotropin releasing hormone71 Lipoprotein lipase90 Transforming growth factor 74 QTL=quantitative trait locus. Candidate genes were identified on the basis of rare Mendelian diseases (panel 2), genome scans, positional mapping, or potential functional relevance. events, including sudden death, by nearly 20%.2 Nifedipine Trial on Antiatherosclerotic Therapy, short- Furthermore, antihypertensive treatment substantially acting nifedipine (n=173) compared with placebo reduced the risk of heart failure, although estimates of (n=175) reduced the number of new coronary lesions by benefit varied across trials because of differences in the 28%, but the drug was associated with higher all-cause clinical criteria used for diagnosis. (14 vs 2) and cardiac mortality (10 vs 2), whereas the In the Systolic Hypertension in Europe (Syst-Eur) incidence of non-fatal cardiac events was similar in both trial,116 dementia was a pre-specified secondary endpoint. groups (52 vs 44).134 After a median follow-up of 3·9 years, treatment starting with the dihydropyridine calcium-channel blocker Comparative trials nitrendipine reduced the rate of degenerative dementia by In nine trials, 62 605 hypertensive patients were ran- 55% (p<0·001).116 In view of the null results observed in domised to initial treatment with old drugs (diuretics and other placebo-controlled studies testing thiazides,117,118 blockers), or new agents (calcium-channel blockers and blockers,117,118 or monotherapy with the ACE inhibitor converting-enzyme inhibitors). Compared with old drugs, perindopril,119 the Syst-Eur findings might be due to new agents offered similar overall cardiovascular chance rather than to blood pressure lowering or the protection, but calcium-channel blockers provided more inhibition of calcium influx into ischaemic brain reduction in the risk of stroke (13·5%, 95% CI neurons.120 Thus, the concept that antihypertensive 1·3–24·2%, p=0·03) and less reduction in the risk of treatment with long-acting dihydropyridines might myocardial infarction (19·2%, 3·5–37·3%, p=0·01).114 protect against dementia needs further testing in These cause-specific outcome results confirm findings of controlled clinical trials. other investigators,135 but have wide confidence intervals, and therefore must be interpreted with caution. Secondary prevention The researchers of the Losartan Intervention for Several trials119,121-125 have tested the hypothesis that Endpoint Reduction in Hypertension Study (LIFE)136 inhibitors of the renin-angiotensin system might be more reported that first-line treatment with losartan, compared effective than placebo in the secondary prevention of with atenolol, improved outcome over and beyond blood stroke and the cardiovascular-renal complications of pressure control. They measured achieved blood pressure hypertension. These drugs diminished the frequency of at the end of follow-up (mean 4·8 years) or just before an cardiovascular events123,124 and stroke recurrence.119 In event. This definition did not account for the differences diabetic patients with normoalbuminuria123 or in systolic pressure (higher on atenolol), diastolic pressure microalbuminuria,125 they also decreased the incidence of (lower on atenolol), or pulse pressure (higher on overt nephropathy, but they did not reduce the rate of atenolol), which arose during the first year of follow-up. progression of coronary122 or carotid121 atherosclerosis. In addition, in older patients such as those enrolled in the Furthermore, in patients with overt nephropathy due to LIFE study (age range 55–80 years, mean age 66·9 years), diabetes126,127 or hypertension,128 ACE inhibitors128 or blockers improve outcome less than diuretics.137 angiotensin receptor blockers126,127 were more effective The Antihypertensive and Lipid-Lowering Treatment than conventional therapy126 or amlodipine127 in to Prevent Heart Attack Trial (ALLHAT)138,139 had 42 424 postponing a doubling of the serum creatinine participants aged 55 years or older with hypertension and concentration or preventing end-stage renal disease. at least one other cardiovascular risk factor. They were However, in line with Fleckenstein’s investigations in randomised to first-line antihypertensive therapy animals,129 results of several trials have shown that long- with chlorthalidone (12·5 to –25·0 mg/day, n=15 255), acting calcium-channel blockers compared with amlodipine (2·5–10·0 mg/day, n=9067), lisinopril placebo,130 diuretics131,132 or atenolol133 slowed the (10–40 mg/day, n=9048), or doxazosin (2–8 mg/day, progression of carotid atherosclerosis. In the International n=9054) with an intended follow-up of 4–8 years. The 1636 THE LANCET • Vol 361 • May 10, 2003 • www.thelancet.com For personal use. Only reproduce with permission from The Lancet Publishing Group. SEMINAR Group Relative risk (95%) versus chlorthalidone Doxazosin139 Amlodipine138 Lisinopril138 Outcome CHD* All 1·03 (0·90–1·17) 0·98 (0·90–1·07) 0·99 (0·91–1·08) Combined CHD† All 1·10 (1·00–1·12) 1·00 (0·94–1·07) 1·05 (0·98–1·11) Total mortality All 1·03 (0·90–1·15) 0·96 (0·89–1·02) 1·00 (0·94–1·08) Non-CV mortality All .. 0·88 (0·79–0·97) 0·93 (0·84–1·03) Combined CVD‡ All 1·25 (1·17–1·13) 1·04 (0·99–1·09) 1·10 (1·05–1·16) 65 years 1·29 (1·20–1·40) 1·05 (0·99–1·12) 1·13 (1·06–1·20) Female .. 1·04 (0·96–1·13) 1·12 (1·03–1·21) Black 1·40 (1·25–1·57) 1·06 (0·96–1·16) 1·19 (1·09–1·30) Diabetic 1·24 (1·12–1·38) 1·06 (0·98–1·15) 1·08 (1·00–1·17) Stroke All 1·19 (1·01–1·40) 0·93 (0·82–1·06) 1·15 (1·02–1·30) 65 years .. 0·93 (0·81–1·08) 1·13 (0·98–1·30) Female .. 0·84 (0·69–1·03) 1·22 (1·01–1·46) Black .. 0·93 (0·76–1·14) 1·40 (1·17–1·68) Diabetic .. 0·90 (0·75–1·08) 1·07 (0·90–1·28) Heart failure All 2·04 (1·79–2·32) 1·38 (1·25–1·52) 1·19 (1·07–1·31) 65 years 2·07 (1·78–2·41) 1·33 (1·18–1·49) 1·20 (1·06–1·35) Female .. 1·33 (1·14–1·55) 1·23 (1·05–1·43) Black 2·18 (1·73–2·74) 1·47 (1·24–1·74) 1·32 (1·11–1·58) Diabetic 2·14 (1·76–2·58) 1·42 (1·23–1·64) 1·22 (1·05–1·42) ESRD All .. 1·12 (0·89–1·40) 1·11 (0·88–1·38) Main outcome results of the ALLHAT trial138,139 in all patients and prespecified subgroups CHD=coronary heart disease. Non-CV mortality=non-cardiovascular mortality. CVD=cardiovascular disease. ESRD=end-stage renal disease. *Primary endpoint including coronary mortality and non-fatal myocardial infarction. †Combined CHD includes coronary mortality, non-fatal myocardial infarction, coronary revascularisation, and hospitalised angina pectoris. ‡Combined CVD consists of coronary mortality, non-fatal myocardial infarction, coronary revascularisation, hospitalised or treated angina pectoris, stroke, hospitalised or treated congestive heart failure, and revascularisation for peripheral artery disease. primary outcome, a composite of coronary mortality and complications of raised blood pressure. Accordingly, the non-fatal myocardial infarction, occurred in 3321 need to start treatment increases in the presence of other patients.138,139 Because of the twofold higher risk of heart cardiovascular risk factors or when absolute risk reaches a failure on doxazosin compared with chlorthalidone, the specified threshold (for example in the British population, doxazosin arm was stopped prematurely after a median a 10-year risk of coronary heart disease of 15% or follow-up of 3·3 years.139 Follow-up of the other groups greater12). lasted an average of 4·9 years.138 The key finding was that On the basis of evidence from trials, most10,12,13 but not there were no differences between randomised patients in all guidelines10 suggest that low-dose thiazides might be the primary outcome and total mortality. Compared with the most cost-effective way to start pharmacological chlorthalidone, treatment starting with amlodipine was treatment in most patients. However, for each class of associated with a higher risk of heart failure (table). antihypertensive drugs, compelling indications or Furthermore, heart failure, stroke, and combined contraindications exist.10,12,13 For instance, blockers or cardiovascular disease arose more frequently in patients calcium-channel blockers can be used in patients with allocated doxazosin or lisinopril than in those of the angina pectoris. Stable heart failure is an indication for chlorthalidone group (table). thiazides, aldosterone receptor blockers,140 blockers,141 or Interpretation of the ALLHAT results is difficult, ACE inhibitors,124,142 but not for angiotensin receptor because at randomisation 90% of the patients were blockers.143 A history of myocardial infarction favours the already on antihypertensive treatment, most often use of blockers144 or ACE inhibitors.124,142 Renal diuretics. Thus, this trial tested continuation of a diuretic impairment, microalbuminuria, or proteinuria is an versus switching drug classes. Patients on diuretics with indication for ACE inhibitors123,128 or angiotensin receptor latent or compensated heart failure were deprived of their blockers.125-128 Systolic hypertension warrants the use of treatment when they were not randomised to thiazides117 or long-acting dihydropyridines.115 In black chlorthalidone. Moreover, the achieved systolic pressure people, hypertension responds better to thiazides or was higher with doxazosin (2·0 mm Hg),139 amlodipine calcium-channel blockers than to inhibitors of the renin (1·1 mm Hg),138 and lisinopril (2·3 mm Hg) than with system, in terms of both blood pressure reduction138,145 and chlorthalidone.138,139 Presumably, these factors explain why prevention of complications.138,142 In the absence of renal the Kaplan-Meier curves started to diverge immediately impairment at the initiation of antihypertensive after randomisation for heart failure and approximately treatment, all major drug classes equally prevent end- 6 months later also for stroke.115 Furthermore, the stage renal disease.138 sympatholytic agents used for step-up treatment (atenolol, The discussion about which drug class is better suited clonidine, or reserpine, at the physician’s discretion) led to start antihypertensive therapy is largely obsolete. In to a somewhat artificial treatment regimen, which does most patients, several drug classes and combinations need not accord with modern clinical practice.138 to be rotated to optimise treatment at acceptable Notwithstanding these limitations, ALLHAT stands out tolerance. The blood-pressure lowering activities of ACE as the largest double-blind trial ever undertaken in inhibitors and blockers are additive to those of thiazides hypertensive patients. The study was managed and and calcium-channel blockers, and vice-versa.146 Patients analysed independent of the pharmaceutical industry. younger than 50 years may be started on ACE inhibitors or blockers and switched to thiazides or calcium- Guidelines for antihypertensive drug therapy channel blockers if blood pressure remains uncontrolled, In view of the evidence from clinical trials, guidelines for whereas the reverse order can be used in those older than the treatment of hypertension10,12,13 are under revision. 50 years.146 However, most experts recommend an integrated The target of antihypertensive treatment is to achieve approach of risk management to prevent the normal levels of systolic and diastolic blood pressure as THE LANCET • Vol 361 • May 10, 2003 • www.thelancet.com 1637 For personal use. Only reproduce with permission from The Lancet Publishing Group. SEMINAR defined in panel 1. In young people and patients with pressure on the risk of cardiovascular disease. N Engl J Med 2001; diabetes mellitus, previous cardiovascular disorders, 345: 1291–97. multiple risk factors, or renal insufficiency, blood pressure 7 Thomas F, Bean K, Guize L, Quentzel S, Argyriadis P, Benetos A. Combined effects of systolic blood pressure and serum cholesterol must be lowered to within the optimum or normal range, on cardiovascular mortality in young (<55 years) men and women. if possible. To achieve blood pressure control, more than Eur Heart J 2002; 23: 528–35. two-thirds of patients need a combination of two or more 8 Wu K, Xie L, Chen D, Chen J. The natural history of borderline drugs.138 A meta-regression analysis115 included 149 407 hypertension in a Chinese population. J Hum Hypertens 1997; 11: 95–100. patients randomised in 30 trials. In contrast to recent 9 Gillum RF. The epidemiology of cardiovascular disease in black suggestions of benefit beyond blood pressure Americans. N Engl J Med 1996; 335: 1597–98. lowering,124-127,136 this overview115 suggested that most of the 10 Guidelines Committee. 2003 European Society of Hypertension/ benefit of antihypertensive treatment could be explained European Society of Cardiology Guidelines for the Management of by the decrease in blood pressure rather than by ancillary Arterial Hypertension. J Hypertens 2003 (in press). drug properties, and therefore shows the need for tight 11 Vasan RS, Larson MG, Leip EP, Kannel WB, Levy D. Assessment of frequency to progression to hypertension in non-hypertensive blood pressure control. participants in the Framingham Heart Study: a cohort study. Lancet 2001; 358: 1682–86. Conclusions 12 Ramsay L, Williams B, Johnston GD, et al. Guidelines for The causes of essential hypertension are still uncertain. management of hypertension: report of the third working party of the British Hypertension Society. J Hum Hypertens 1999; 13: For now, with few exceptions, our therapeutic approach 569–92. to essential hypertension and its cardiovascular 13 Hypertension Society of Southern Africa endorsed by the Medical complications remains generic rather than specifically Association of South Africa and the Medical Research Council. targeted. The notion that the discovery of major genes or Guidelines for the management of hypertension at primary health interference with a sole pathophysiological mechanism care level. S Afr Med J 1995; 85: 1321–25. 14 O’Brien E, Pickering T, Asmar R, et al. Working Group on Blood will substantially advance prevention and treatment is an Pressure Monitoring of the European Society of Hypertension oversimplification that ignores the heterogeneous nature International Protocol for validation of blood pressure measuring of this disorder. Integration of genetic, molecular, clinical, devices in adults. Blood Press Monit 2002; 7: 3–18. and epidemiological research could disclose the way in 15 Sorof JM, Portman RJ. Ambulatory blood pressure monitoring in the which genetic and environmental factors lead to different pediatric patient. J Pediatr 2000; 136: 578–86. 16 Verdecchia P, Schillaci G, Borgioni C, Ciucci A, Pede S, Porcellati expression of proteins controlling the processes that cause C. Ambulatory pulse pressure. A potent predictor of total hypertension. This integrated approach may identify cardiovascular risk in hypertension. Hypertension 1998; 32: 983–88. subsets of patients in whom specific combinations of 17 Ohkubo T, Hozawa A, Nagai K, et al. Prediction of stroke by genetic and environmental factors raise blood pressure, ambulatory blood pressure monitoring versus screening blood and could lead to individualised treatment. Blind pressure measurements in a general population: the Ohasama study. J Hypertens 2000; 18: 847–54. treatment by trial and error could then change into 18 Staessen JA, Thijs L, Fagard R, et al. Predicting cardiovascular risk comprehensive and specific intervention with the using conventional vs ambulatory blood pressure in older patients pathophysiological processes at work in every patient. with systolic hypertension. JAMA 1999; 282: 539–46. 19 Kario K, Pickering TG, Matsuo T, Hoshide S, Schwartz JE, Contributors Shimada K. Stroke prognosis and abnormal nocturnal blood All authors conceived and wrote the seminar. pressure falls in older hypertensives. Hypertension 2001; 38: 852–57. 20 Verdecchia P, Staessen JA, White WB, Imai Y, O’Brien ET. Conflict of interest statement Properly defining white coat hypertension. Eur Heart J 2002; 23: JAS and WHB did ad-hoc consultancies for pharmaceutical companies 106–09. with commercial interests in the cardiovascular field. During their 21 Pickering TG. Blood pressure measurement and detection of lifetimes, they have received funding for studies, seminars, and travel from hypertension. Lancet 1994; 344: 31–35. such companies. GB is an adviser to the Prassis Sigma Tau Research Institute (Milan, Italy). 22 Fagard RH, Staessen JA, Thijs L, et al. Response to antihypertensive therapy in older patients with sustained and nonsustained systolic hypertension. Circulation 2000; 102: 1139–44. Acknowledgments 23 Sega R, Trocino G, Lanzarotti A, et al. Alterations of cardiac JAS thanks Hilde Celis, Elly Den Hond, Tatiana Kuznetsova, structure in patients with isolated office, ambulatory, or home Tim Nawrot, Agnieszka Olszanecka, Katarzyna Stolarz, Lutgarde Thijs, hypertension: data from the general population (Pressione Arteriose and Valérie Tikhonoff. Research cited in this review was partly sponsored Monitorate E Loro Associazioni [PAMELA] Study). Circulation by the European Union (contracts IC15-CT98-0329-EPOGH and 2002; 104: 1385–92. QLG1-CT-2000-01137-EURNETGEN). JGW was funded by the 24 Liu JE, Roman MJ, Pini R, Schwartz JE, Pickering TG, Bilaterale Scientific and Technological Collaboration between the Devereux RB. Cardiac and arterial target organ damage in adults People’s Republic of China and Flanders (project BIL02/10). The with elevated ambulatory and normal office blood pressure. Ann funding sources had no role in the writing of the article. Intern Med 1999; 131: 564–72. 25 Chatellier G, Dutrey-Dupagne C, Vaur L, et al. Home self blood pressure measurement in general practice. The SMART Study. References Am J Hypertens 1996; 9: 644–52. 1 Franklin SS, Larson MG, Khan SA, et al. Does the relation 26 Thijs L, Staessen JA, Celis H, et al. The international database of of blood pressure to coronary heart disease change with aging? self-recorded blood pressures in normotensive and untreated The Framingham Heart Study. Circulation 2001; 103: 1245–49. hypertensive subjects. Blood Press Monit 1999; 4: 77–86. 2 a Staessen JA, G¸ sowski J, Wang JG, et al. Risks of untreated and 27 Ohkubo T, Imai Y, Tsuji I, et al. Home blood pressure measurement treated isolated systolic hypertension in the elderly: meta-analysis has a stronger predictive power for mortality than does screening of outcome trials. Lancet 2000; 355: 865–72. blood pressure measurement: a population-based observation in 3 Mitchel GF, Moyé LA, Braunwald E, et al. Sphygmomanometrically Ohasama, Japan. J Hypertens 1998; 16: 971–75. determined pulse pressure is a powerful independent predictor 28 Staessen JA, Byttebier G, Buntinx F, et al. Antihypertensive of recurrent events after myocardial infarction in patients with treatment based on conventional or ambulatory blood pressure impaired left ventricular function. Circulation 1997; 96: 4254–60. measurement. A randomized controlled trial. JAMA 1997; 278: 4 Safar ME, Blacher J, Pannier B, et al. Central pulse pressure 1065–72. and mortality in end-stage renal disease. Hypertension 2002; 39: 29 Guyton AC. Long-term arterial pressure control: an analysis from 735–38. animal experiments and computer and graphic models. Am J Physiol 5 Staessen JA, Thijs L, O’Brien ET, et al. Ambulatory pulse pressure as 1990; 259: R865–77. predictor of outcome in older patients with systolic hypertension. 30 Keller G, Zimmer G, Mall G, Ritz E, Amann K. Nephron number Am J Hypertens 2002; 15 (part 1): 835–43. in patients with primary hypertension. N Engl J Med 2003; 348: 6 Vasan RS, Larson MG, Leip EP, et al. Impact of high-normal blood 101–08. 1638 THE LANCET • Vol 361 • May 10, 2003 • www.thelancet.com For personal use. Only reproduce with permission from The Lancet Publishing Group. SEMINAR 31 Bianchi G, Ferrari P. Renal factors involved in the pathogenesis of 57 Wang JG, Staessen JA. Genetic polymorphisms in the renin- genetic forms of hypertension. In: Sassard J, ed. Genetic angiotensin system: relevance for susceptibility to cardiovascular Hypertension. Colloque INSERM, Montrouge, France: John Libbey disease. Eur J Pharmacol 2000; 410: 289–302. Eurotext, 1992: 447–58. 58 Whelton PK, Appel LJ, Espeland MA, et al. Sodium reduction and 32 Rettig R, Folberth CG, Stauss H, et al. Hypertension in rats induced weight loss in the treatment of hypertension in older persons. A by renal grafts from renovascular hypertensive donors. Hypertension randomized controlled trial of nonpharmacologic interventions in the 1990; 15: 429–35. elderly (TONE). JAMA 1998; 279: 839–46. 33 Guidi E, Menghetti D, Milani S, Montagnino G, Palazzi P, 59 Beretta-Piccoli C, Weidmann P, Brown JJ, Davies DL, Lever AF, Bianchi G. Hypertension may be transplanted with the kidney in Robertson JIS. Body sodium and blood volume state in essential humans: a long-term historical perspective follow-up of recipients hypertension: abnormal relation of exchangeable sodium to age grafted with kidneys from donors with or without hypertension in and blood pressure in male patients. J Cardiovasc Pharmacol 1984; their families. J Am Soc Nephrol 1996; 7: 1131–38. 6 (suppl 1): S134–42. 34 Orlov SN, Adragna NC, Adarichev VA, Hamet P. Genetic and 60 Mancia G, Grassi G, Giannattasio C, Seravalle G. Sympathetic biochemical determinants of abnormal monovalent ion transport in activation in the pathogenesis of hypertension and progression of primary hypertension. Am J Physiol 1999; 276: C511–36. organ damage. Hypertension 1999; 34 (part 2): 724–28. 35 Tripodi G, Valtorta F, Torielli L, et al. Hypertension-associated 61 Schmieder RE, Schächinger H, Messerli FH. Accelerated decline in point mutations in the adducin alpha and beta subunits affect actin renal perfusion with aging in essential hypertension. Hypertension cytoskeleton and ion transport. J Clin Invest 1996; 97: 2815–22. 1994; 23: 351–57. 36 Barlassina C, Schork NJ, Manuta P, et al. Synergistic effect of 62 Aviv A. Chronobiology versus biology. Telomeres, essential -adducin and ACE genes causes blood pressure changes with body hypertension, and vascular aging. Hypertension 2002; 40: 229–32. sodium and volume expansion. Kidney Int 2000; 57: 1083–90. 63 Magyar CE, Zhang Y, Holstein-Rathlou NH, McDonough AA. 37 Geller DS, Farhi A, Pinkerton N, et al. Activating mineralocorticoid Downstream shift in sodium pump activity along the nephron during receptor mutation in hypertension exacerbated by pregnancy. Science acute hypertension. J Am Soc Nephrol 2001; 12: 2231–40. 2000; 289: 119–23. 64 Muntau AC, Roschinger W, Habich M, et al. Tetrahydrobiopterin as 38 Schoner W. Endogenous cardiac glycosides, a new class of steroid an alternative treatment for mild phenylketonuria. N Engl J Med hormones. Eur J Biochem 2002; 269: 2440–48. 2002; 347: 2122–32. 39 Manunta P, Messaggio E, Ballabeni C, et al. Plasma ouabain-like 65 Young LE. Imprinting of genes and the Barker hypothesis. Twin Res factor during acute and chronic changes in sodium balance in 2002; 4: 307–17. essential hypertension. Hypertension 2001; 38: 198–203. 66 Morris AA, Lightowlers RN. Can paternal mtDNA be inherited? 40 DiBona GF. Nervous kidney. Interaction between renal sympathetic Lancet 2000; 355: 1290–91. nerves and the renin-angiotensin system in the control of renal 67 Schwartz M, Vissing J. Paternal inheritance of mitochondrial DNA. function. Hypertension 2000; 36: 1083–88. N Engl J Med 2002; 347: 576–80. 41 De Angelis C, Haupert GT Jr. Hypoxia triggers release of an 68 Kong A, Gudbjartsson DE, Sainz J, et al. A high-resolution endogenous inhibitor of Na+-K+-ATPase from midbrain and adrenal. recombination map of the human genome. Nat Genet 2002; 31: Am J Physiol 1998; 274: F182–88. 241–47. 42 Rossi G, Manunta P, Hamlyn JM, et al. Immunoreactive endogenous 69 Barker DJ, Clark PM. Fetal undernutrition and disease later in life. ouabain in primary aldosteronism and essential hypertension: Rev Reprod 1997; 2: 105–12. relationship with plasma renin, aldosterone and blood pressure levels. 70 Surani MZ. Reprogramming of genome function through epigenetic J Hypertens 1995; 13: 1181–91. inheritance. Nature 2001; 414: 122–27. 43 Ferrari P, Ferrandi M, Torielli L, et al. Antihypertensive compounds 71 Rice T, Rankinen T, Province MA, et al. Genome-wide linkage that modulate the Na-K pump. In: Jørgensen PL, Karlish SJD, analysis of systolic and diastolic blood pressure: the Québec Family Maunsbach AB, eds. Na, K-ATPase and related cation pumps: Study. Circulation 2001; 102: 1956–63. structure, function, and regulatory mechanisms. Ann NY Acad Sci 72 Rapp JP. Genetic analysis of inherited hypertension in the rat. (in press). Physiol Rev 2000; 80: 135–72. 44 Fink GD. Long-term sympatho-excitatory effect of angiotensin II: a 73 Cooper RS, Luke A, Zhu X, et al. Genome scan among Nigerians mechanism of spontaneous and renovascular hypertension. linking blood pressure to chromosomes 2, 3, and 19. Hypertension Clin Exp Pharmacol Physiol 1997; 24: 91–95. 2002; 40: 629–33. 45 Ferguson AV, Washburn DL. Angiotensin II: a peptidergic 74 Rice T, Rankinen T, Chagnon YC, et al. Genomewide linkage scan neurotransmitter in central autonomic pathways. Prog Neurobiol 1998; of resting blood pressure. HERITAGE Family Study. Hypertension 54: 169–92. 2002; 39: 1037–43. 46 Mulvany MJ. Vascular remodelling of resistance vessels: can we 75 Zimdahl H, Kreitler T, Gösele C, Ganten D, Hübner N. Conserved define this? Cardiovasc Res 1999; 44: 9–13. synteny in rat and mouse for a blood pressure QTL on human 47 Kvist S, Mulvany MJ. Reduced medication and normalization of chromosome 17. Hypertension 2002; 39: 1050–52. vascular structure, but continued hypertension in renovascular 76 Tabor HK, Risch NJ, Myers RM. Candidate-gene approaches for patients after revascularization. Cardiovasc Res 2003; 52: 136–42. studying complex genetic traits: practical considerations. Nat Rev 48 Serné EH, Gans RO, ter Maaten JC, Tangelder GJ, Donker AJ, 2002; 3: 391–97. Stehouwer CD. Impaired skin capillary recruitment in essential 77 Watt GCM, Harrap SB, Foy CJW, et al. Abnormalities of hypertension is caused by both functional and structural capillary glucocorticoid metabolism and the renin-angiotensin system: a four- rarefaction. Hypertension 2001; 38: 238–42. corners approach to the identification of genetic determinants of 49 O’Rourke MF, Staessen JA, Vlachopoulos C, Duprez D, Plante GE. blood pressure. J Hypertens 1992; 10: 473–82. Clinical applications of arterial stiffness: definitions and reference 78 Iwai N, Ohmichi N, Hanai K, Nakamura Y, Kinoshita M. Human values. Am J Hypertens 2002; 15: 426–44. SA gene locus as a candidate gene locus for essential hypertension. 50 Luft FC. Twins in cardiovascular genetic research. Hypertension Hypertension 1994; 23: 375–80. 2001; 37 (part 2): 350–56. 79 Siffert W, Rosskopf D, Siffert G, et al. Association of a human G- 51 Geller DS, Rodriguez-Soriano J, Vallo Boado A, et al. Mutations in protein 3 subunit variant with hypertension. Nat Genet 1998; 18: the mineralocorticoid receptor gene cause autosomal dominant 45–48. pseudohypoaldosteronism type I. Nat Genet 1998; 19: 279–81. 80 Sharma P, Hingorani A, Jia H, et al. Positive association of tyrosine 52 Lifton RP, Gharavi AG, Geller DS. Molecular mechanisms of human hydroxylase microsatellite marker to essential hypertension. hypertension. Cell 2001; 104: 545–56. Hypertension 1998; 32: 676–82. 53 Lander E, Kruglyak L. Genetic dissection of complex traits: 81 Ferrandi M, Bianchi G. Genetic mechanisms underlying the guidelines for interpreting and reporting linkage results. Nat Genet regulation of urinary sodium excretion and arterial blood pressure: 1995; 11: 241–49. the role of adducin. Acta Physiol Scand 2000; 168: 187–93. 54 Higaki J, Baba S, Katsuya T, et al. Deletion allele of angiotensin- 82 Matsuoka Y, Li X, Bennett V. Adducin: structure, function and converting enzyme gene increases risk of essential hypertension in regulation. Cell Mol Life Sci 2000; 57: 884–95. Japanese men. The Suita Study. Circulation 2000; 101: 2060–65. 83 Glorioso N, Filigheddu F, Troffa C, et al. Interaction of the 1- 55 Luft FC, Miller JZ, Grim CE, et al. Salt sensitivity and resistance of Na,K-ATPase and Na,K,2Cl-cotransporter genes in human essential blood pressure: age and race as factors in physiological responses. hypertension. Hypertension 2001; 38: 204–09. Hypertension 1991; 17 (suppl I): I·102–08. 84 Tsujita Y, Baba S, Yamauchi R, et al. Association analyses between 56 Price DA, Fisher NDL, Lansang MC, Stevanovic R, Williams GH, genetic polymorphisms of endothelial nitric oxide synthase gene and Hollenberg NK. Renal perfusion in blacks. Alterations caused by hypertension in Japanese: The Suita Study. J Hypertens 2001; 19: insuppressibility of intrarenal renin with salt. Hypertension 2002; 40: 1941–48. 186–89. 85 Iwai N, Baba S, Mannami T, Ogihara T, Ogata J. Association of a THE LANCET • Vol 361 • May 10, 2003 • www.thelancet.com 1639 For personal use. Only reproduce with permission from The Lancet Publishing Group. SEMINAR sodium channel alpha subunit promoter variant with blood pressure. 24-h ambulatory blood pressure and heart rate: a study in 352 J Am Soc Nephrol 2002; 13: 80–85. normotensive Danish subjects. Am J Hypertens 1997; 10: 483–91. 86 Julier C, Delépine M, Keavney B, et al. Genetic susceptibility for 113 Collins R, MacMahon S. Blood pressure, antihypertensive drug human familial essential hypertension in a region of homology with treatment and the risks of stroke and of coronary heart disease. blood pressure linkage on rat chromosome 10. Hum Mol Genet 1997; Br Med Bull 1994; 50: 272–98. 6: 2077–85. 114 Staessen JA, Wang JG, Thijs L. Cardiovascular prevention and 87 Nakayama T, Soma M, Kanmatsuse K. Organization of the human blood pressure reduction: a meta-analysis. Lancet 2001; 358: prostacyclin synthase gene and association analysis of a novel CA 1305–15. repeat in essential hypertension. Adv Exp Med Biol 1997; 433: 115 Staessen JA, Wang JG, Thijs L. Calcium-channel blockade and 127–30. cardiovascular prognosis: recent evidence from clinical outcome 88 Frossard PM, Lestringant GG. Association between a dimorphic site trials. Am J Hypertens 2002; 15: 85S–93S. on chromosome 12 and clinical diagnosis of hypertension in three 116 Forette F, Seux ML, Staessen JA, et al. The prevention of dementia independent populations. Clin Genet 1995; 48: 284–87. with antihypertensive treatment. New evidence from the Systolic 89 Wilson AF, Elston RC, Tran LD, Siervogel RM. Use of robust sib- Hypertension in Europe (Syst-Eur) Study. Arch Intern Med 2002; pair method to screen for single-locus, multiple-locus, and 162: 2046–52. pleiotrophic effects: application to traits related to hypertension. 117 SHEP Cooperative Research Group. Prevention of stroke by Am J Hum Genet 1991; 48: 862–72. antihypertensive drug treatment in older persons with isolated systolic 90 Wu DA, Bu X, Warden CH, et al. Quantitative trait locus mapping hypertension. Final results of the Systolic Hypertension in the Elderly of human blood pressure to a genetic region at or near the Program (SHEP). JAMA 1991; 265: 3255–64. lipoproteinase gene locus on chromosome 8p22. J Clin Invest 1996; 118 Prince MJ, Bird AS, Blizard RA, Mann AH. Is the cognitive function 97: 2111–18. of older patients affected by antihypertensive treatment? Results from 91 Ying LH, Zee RY, Griffiths LR, Morris BJ. Association of a RFLP 54 months of the Medical Research Council’s treatment trial of for the insulin receptor gene, but not insulin, with essential hypertension in older adults. BMJ 1996; 312: 801–05. hypertension. Biochem Biophys Res Comm 1991; 181: 486–92. 119 PROGRESS Collaborative Group. Randomised trial of a perindopril- 92 Hopkins PN, Hunt SC, Jeunemaitre X, et al. Angiotensinogen based blood-pressure-lowering regimen among 6105 individuals with genotype affects renal and adrenal responses to angiotensin II in prior stroke or transient ischaemic attack. Lancet 2001; 358: essential hypertension. Circulation 2002; 105: 1921–27. 1033–41. 93 Turner ST, Schwartz GL, Chapman AB, Dallas Hall D, 120 Zipfel GJ, Lee JM, Choi DW. Reducing calcium overload in the Boerwinkle E. Antihypertensive pharmacogenetics: getting the right ischemic brain. N Engl J Med 1999; 341: 1543–44. drug into the right patient. J Hypertens 2001; 19: 1–12. 121 MacMahon S, Sharpe N, Gamble G, et al. Randomized, placebo- 94 Tribut O, Lessard Y, Reymann JM, Allain H, Bentué-Ferrer D. controlled trial of the angiotensin-converting enzyme inhibitor, Pharmacogenomics. Med Sci Monit 2002; 8: RA152–63. ramipril, in patients with coronary or other occlusive arterial disease. 95 Bianchi G, Tripodi G, Casari G, et al. Two point mutations within J Am Coll Cardiol 2000; 36: 438–43. the adducin genes are involved in blood pressure variation. 122 Teo KK, Burton JR, Buller CE, et al. Long-term effects of Proc Natl Acad Sci USA 1994; 91: 3999–4003. cholesterol lowering and angiotensin-converting enzyme inhibition on 96 Staessen JA, Wang JG, Brand E, et al. Effects of three candidate coronary atherosclerosis. The Simvastatin/Enalapril Coronary genes on prevalence and incidence of hypertension in a Caucasian Atherosclerosis Trial (SCAT). Circulation 2000; 102: 1748–54. population. J Hypertens 2001; 19: 1349–58. 123 Heart Outcomes Prevention Evaluation Study Investigators. Effects 97 Wang JG, Staessen JA, Tizzoni L, et al. Renal function in relation to of ramipril on cardiovascular and microvascular outcomes in people three candidate genes. Am J Kidney Dis 2001; 38: 1158–68. with diabetes mellitus: results of the HOPE study and MICRO- HOPE substudy. Lancet 2000; 355: 253–59. 98 Balkestein EJ, Wang JG, Struijker Boudier HAJ, et al. Carotid and femoral intima-media thickness in relation to three candidate genes in 124 The Heart Outcomes Prevention Evaluation Study Investigators. a Caucasian population. J Hypertens 2002; 20: 1551–61. Effects of an angiotensin-converting-enzyme inhibitor, ramipril, on cardiovascular events in high-risk patients. N Engl J Med 2000; 342: 99 Ferrari P, Ferrandi M, Tripodi G, et al. PST 2238: a new 145–53. antihypertensive compound that modulates Na,K-ATPase in genetic hypertension. J Pharmacol Exp Ther 1999; 288: 1074–83. 125 Parving HH, Lehnert H, Bröchner-Mortensen J, et al. The effect of irbesartan on the development of diabetic nephropathy in patients 100 Psaty BM, Smith NL, Hekbert SR, et al. Diuretic therapy, the - with type 2 diabetes. N Engl J Med 2001; 345: 870–78. adducin gene variant, and the risk of myocardial infarction or stroke in persons with treated hypertension. JAMA 2002; 287: 1680–90. 126 Brenner BM, Cooper ME, de Zeeuw D, et al. Effects of losartan on renal and cardiovascular outcomes in patients with type 2 diabetes 101 Denton D, Weisinger R, Mundy NI, et al. The effect of increased salt and nephropathy. N Engl J Med 2001; 345: 861–69. intake on blood pressure of chimpanzees. Nat Med 1995; 1: 1009–16. 127 Lewis EJ, Hunsicker LG, Clarke WR, et al. Renoprotective effect of 102 Graudal N, Galløe AM, Garred P. Effects of sodium restriction on the angiotensin-receptor antagonist irbesartan in patients with blood pressure, renin, aldosterone, catecholamines, cholesterols, and nephropathy due to type 2 diabetes. N Engl J Med 2001; 345: triglycerides: a meta-analysis. JAMA 1998; 279: 1383–91. 851–60. 103 Sacks FM, Svetkey LP, Vollmer WM, et al. Effects on blood pressure 128 Agodoa LY, Appel L, Bakris GL, et al. Effect of ramipril vs of reduced dietary sodium and the dietary approaches to stop amlodipine on renal outcomes in hypertensive nephrosclerosis. hypertension (DASH) diet. N Engl J Med 2001; 344: 3–10. JAMA 2001; 285: 2719–28. 104 Uzu T, Ishikawa K, Fujii T, Nakamura S, Inenaga T, Kimura G. 129 Fleckenstein A, Frey M, Thimm F, Fleckenstein-Grün G. Excessive Sodium restriction shifts circadian rhythm of blood pressure from mural calcium overload—a predominant causal factor in the nondipper to dipper in essential hypertension. Circulation 1997; 96: development of stenosing coronary plaques in humans. 1859–62. Cardiovasc Drugs Ther 1990; 4: 1005–14. 105 Weinberger MC, Fineberg NS, Fineberg SE, Weinberger M. Salt 130 Pitt B, Byington RP, Furberg CD, et al. Effect of amlodipine on the sensitivity, pulse pressure, and death in normal and hypertensive progression of atherosclerosis and the occurrence of clinical events. humans. Hypertension 2001; 27 (part 2): 429–32. Circulation 2000; 102: 1503–10. 106 Tuomilehto J, Jousilahti P, Rastenyte D, et al. Urinary sodium 131 Zanchetti A, Agabiti Rosei E, Dal Palù C, et al. The Verapamil in excretion and cardiovascular mortality in Finland: a prospective Hypertension and Atherosclerosis Study (VHAS): results of long- study. Lancet 2001; 357: 848–51. term randomized treatment with either verapamil or chlorthalidone 107 Morimoto A, Uzu T, Fujii T, et al. Sodium sensitivity and on carotid intima-media thickness. J Hypertens 1998; 16: 1667–76. cardiovascular events in patients with essential hypertension. Lancet 132 Simon A, Gariépy J, Moyse D, Levenson J. Differential effects of 1997; 350: 1734–37. nifedipine and co-amilozide on the progression of early carotid wall 108 Omvik P, Lund-Johansen P. Is sodium restriction effective treatment changes. Circulation 2001; 103: 2949–54. of borderline and mild essential hypertension? A long-term 133 Zanchetti A, Bond G, Hennig M, et al. Calcium antagonist lacidipine haemodynamic study at rest and during exercise. J Hypertens 1986; 4: slows down progression of asymptomatic carotid atherosclerosis. 535–41. Principal results of the European Lacidipine Study on Atherosclerosis 109 Beilin LJ. Alcohol, hypertension and cardiovascular disease. (ELZA), a randomized, double-blind, long-term trial. Circulation J Hypertens 1995; 13: 939–42. 2002; 106: r47–r52. 110 Staessen J, Fagard R, Lijnen P, Amery A. Body weight, sodium 134 Lichtlen PR, Hugenholtz PG, Rafflenbeul W, et al. Retardation of intake and blood pressure. J Hypertens 1989; 7 (suppl 1): S19–23. coronary artery disease in humans by the calcium-channel blocker 111 Fagard R. Exercise characteristics and the blood pressure response to nifedipine: results of the INTACT Study (International Trial on dynamic physical training. Med Sci Sports Exercise 2001; 33 (suppl): Antiatherosclerotic Therapy). Cardiovasc Drugs Ther 1990; 4: S484–92. 1047–68. 112 Mikkelsen KL, Wiinberg N, Hoegholm A, et al. Smoking related to 135 Blood Pressure Lowering Treatment Trialists’ Collaboration. 1640 THE LANCET • Vol 361 • May 10, 2003 • www.thelancet.com For personal use. Only reproduce with permission from The Lancet Publishing Group. SEMINAR Effect of ACE inhibitors, calcium antagonists, and other blood morbidity and mortality in patients with sever heart failure. pressure-lowering drugs: results of prospectively designed N Engl J Med 1999; 341: 709–17. overviews of randomised trials. Lancet 2000; 356: 1955–64. 141 Stevenson LW. Beta-blockers for stable heart failure. N Engl J Med 136 Dahlöf B, Devereux RB, Kjeldsen SE, et al. Cardiovascular 2002; 346: 1346–47. morbidity and mortality in the Losartan Intervention For Endpoint 142 Exner DV, Dries DL, Domanski MJ, Cohn JN. Lesser response to reduction in hypertension study (LIFE): a randomised trial against angiotensin-converting-enzyme inhibitor therapy in black as atenolol. Lancet 2002; 359: 995–1003. compared with white patients with left ventricular dysfunction. 137 Messerli FH, Grossman E, Goldbourt U. Are beta-blockers N Engl J Med 2001; 344: 1351–57. efficacious as first-line therapy for hypertension in the elderly? 143 Dickstein K, Kjekshus J, and the OPTIMAAL Steering Committee, A systematic review. JAMA 1998; 279: 1903–07. for the OPTIMAAL Study Group. Effects of losartan and captopril 138 The ALLHAT Officers and Coordinators for the ALLHAT on mortality and morbidity in high-risk patients after acute Collaborative Research Group. Major outcomes in high-risk myocardial infarction: the OPTIMAAL randomised trial. Lancet hypertensive patients randomized to angiotensin-converting 2002; 360: 752–60. enzyme inhibitor or calcium channel blocker vs diuretic. The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart 144 Gheorghiade M, Goldstein S. -blockers in the post-myocardial Attack Trial (ALLHAT). JAMA 2003; 288: 2981–97. infarction patient. Circulation 2002; 106: 394–98. 139 The ALLHAT Officers and Coordinators for the ALLHAT 145 Sareli P, Radevski IV, Valtchanova ZP, et al. Efficacy of different Collaborative Research Group. Major cardiovascular events drug classes used to initiate antihypertensive treatment in black in hypertensive patients randomized to doxazosin vs chlorthalidone. subjects. Results of a randomized trial in Johannesburg, South The Antihypertensive and Lipid-Lowering Treatment to Africa. Arch Intern Med 2001; 161: 965–71. Prevent Heart Attack Trial (ALLHAT). JAMA 2000; 283: 146 Dickerson JEC, Hingorani AD, Ashby MJ, Palmer CR, Brown MJ. 1967–75. Optimisation of antihypertensive treatment by crossover rotation of 140 Pitt B, Zannad F, Remme WJ, et al. The effect of spironolactone on four major classes. Lancet 1999; 353: 2008-2013. Clinical picture Homozygous familial hypercholesterolaemia in identical twins J Zschocke, J R Schaefer At the age of 2 years, two identical twin boys of Turkish origin (figure) were referred for investigation of symmetrical skin lesions on ellbows, knees, hands, and ankles. Lipid analyses revealed massive hypercholesterolaemia, with total cholesterol of 1250/1140 mg/dL, LDL cholesterol of 1180/1080 mg/dL, and Lp(a) of 130/130 mg/dL, respectively, in the two children. The parents, who were consanguineous, both suffered from previously undiagnosed familial hyperc- holesterolaemia. Family history was surprisingly unremarkable with regard to cardiovascular disease. Conservative treatment with diet and various lipid- lowering drugs in the children failed to sufficiently reduce plasma cholesterol levels. At the age of three years, the xanthomas were markedly increased, and LDL apheresis therapy was initiated. To our knowledge these are the youngest children with LDL- receptor deficiency reported in the literature. The occurrence of identical skin lesions in the children was an important factor for early diagnosis— Department of Paediatrics and Institute of Human Genetics, Ruprecht-Karls-University, being an identical twin may sometimes 69120 Heidelberg, Germany (J Zschockel MD); Department of Internal Medicine, have unexpected advantages. Cardiology, Philipps-University, 35033 Marburg, Germany (J R Schaefer MD) THE LANCET • Vol 361 • May 10, 2003 • www.thelancet.com 1641 For personal use. Only reproduce with permission from The Lancet Publishing Group.