- Email: [email protected]
Abstracts - Heart UK 19th annual medical and scientific meeting
Atherosclerosis 182 (2005) S1–S8
Abstracts
H·E·A·R·T UK 19th Annual Medical and Scientific Meeting, University of Glamorgan, Pontypridd, Wales, UK, 30 June–1 July 2005 Available online 14 July 2005
0021-9150/$ – see front matter © 2005 Published by Elsevier Ireland Ltd. doi:10.1016/j.atherosclerosis.2005.05.015
Abstracts / Atherosclerosis 182 (2005) S1–S8
RELATIONSHIP OF THE METABOLIC SYNDROME AS ASSESSED BY DIFFERENT DEFINITIONS WITH CORONARY ARTERIAL DISEASE IN PAKISTANIS A.S. Wierzbicki1 , S. Nishtar2,3 , P.J. Lumb1 , M. LambertHammill1 , J. Gill2 1 Department
of Chemical Pathology, St. Thomas’ Hospital, London SE1 7EH, UK; 2 Department of Cardiology, St. Thomas’ Hospital, London SE1 7EH, UK; 3 Department of Cardiology, PIMS, Pakistan Metabolic syndrome is associated with coronary artery disease (CAD). This study investigated the significance of metabolic syndrome (M-IRS) as a CAD risk factor based on different definitions of abdominal obesity. This prospective case–control study recruited 400 individuals with chest pain; 200 with angiographic disease matched with 200 without occlusive disease from a tertiary referral cardiology unit in Pakistan. Metabolic syndrome was present in 37% of CAD(+) versus 27% in CAD(−) by caucasian criteria or 47% versus 42% by asian criteria (p < 0.001). After adjustment for other risk factors, M-IRS was not a significant predictor for CAD or angiographic disease. Age (p = 0.03), smoking (p < 0.001), sialic acid (p = 0.01) and creatinine (p = 0.008) accounted for the excess risk of CAD. Similarly age (p = 0.005), creatinine (p < 0.001), cigarette pack-years (p = 0.02), diabetes (p = 0.008) and sialic acid (p = 0.08) were predictors of greater angiographic disease. M-IRS differed in Pakistanis from caucasians as waist circumference showed weak correlations (r = −0.03 to 0.08; p = 0.45–0.52) with triglycerides, HDL-C, blood pressure or glucose. Sialic acid was the only inflammatory marker associated with M-IRS. Despite strong associations of individual risk factors with M-IRS and a univariate association of M-IRS with CAD, the principal discriminant risk factors in this group are age, smoking, inflammation, diabetes and impaired renal function. The poor sensitivity of M-IRS for CAD reflects either its high underlying prevalence reducing sensitivity, confounding by other urban lifestyle traits and/or a lack of association of waist circumference with M-IRS risk factors. The definition of M-IRS may have to be revised to increase its power as a discriminant risk factor for CHD in Pakistani populations. EXTENDING THE METABOLIC SYNDROME: INSULIN SENSITIVITY, LIVER FUNCTION, SUBCLINICAL INFLAMMATION AND HAEMATOLOGICAL MEASURES I.F. Godsland1,2 , D.G. Johnston1 for the HDDRISC Group2 1 Endocrinology
and Metabolic Medicine; 2 Wynn Department of Metabolic Medicine, Faculty of Medicine, Imperial College London, London, UK To refine the definition of the metabolic syndrome, large studies are needed, using sophisticated measures of insulin sensitivity with a broad range of potential correlates. Four
S3
hundred and seventy-two men (age 26–78 years) without evidence of diabetes or cardiovascular disease and not taking blood pressure-, lipid- or uric acid-lowering agents underwent an intravenous glucose tolerance test (IVGTT), with accompanying measurements of liver and renal function and haematology. Insulin sensitivity, SI , was measured by minimal model analysis of IVGTT glucose and insulin concentrations. Clustering of potential components of the metabolic syndrome was assessed by factor analysis. There were significant associations, independent of age, body mass index, alcohol intake, exercise and cigarette smoking, between SI and serum gamma glutamyl transferase (GGT), aspartate transaminase and alkaline phosphatase (ALP) activities, white blood cell count (WBC), erythrocyte sedimentation rate (ESR) and serum globulin concentration (all p < 0.01). On factor analysis, the principal factor (explaining 49.4% of variance) included in decreasing order of importance, body mass index, SI (negative), the HDL cholesterol/triglyceride concentration ratio (negative), fasting insulin, GGT activity, serum uric acid concentration, mean arterial pressure, ALP activity, blood haemoglobin concentration and fasting glucose. Although they are independent correlates of insulin sensitivity, indices of subclinical inflammation (WBC, ESR and globulin) may not cluster with other metabolic syndrome components. Raised GGT activity appears to be a prominent feature of the metabolic syndrome. HYPERINSULINAEMIA IS ASSOCIATED WITH ENDOTOXIN-MEDIATED INFLAMMATION IN OBESITY AND TYPE 2 DIABETES MELLITUS S.J. Creely1,2 , P.G. McTernan1 , F.F.M. M. Khanolkar3 , M. Evans3 , M. Kumar1,2
Fisher1 ,
1 Diabetes
and Metabolism Research Laboratories, Clinical Sciences Research Institute, Warwick Medical School, Coventry; 2 University Hospital of Coventry and Warwickshire, Coventry; 3 Heart Research Institute, Cardiff University, UK Obesity and type 2 diabetes (T2DM) are associated inflammation. Hyperinsulinaemia associates with bowel stasis, bacterial overgrowth and potentially increased gut permeability, suggesting gut as a source of endotoxin. We hypothesised that serum endotoxin is increased in association with hyperinsulinaemia and may cause the sub-clinical inflammation in T2DM. Further, the reduction of insulin achieved through the use of an insulin sensitizer leads to a reduction in serum endotoxin. Fasting serum samples were obtained from 51 T2DM patients with no complications or inflammatory conditions. 23 patients were on diet alone, 20 on oral agents, 7 on oral agents and insulin and 1 on insulin alone. Serum was also obtained from 33 age matched non-diabetic control subjects. A sub-group of previously untreated T2DM patients received rosiglitazone with fasting serum collected at baseline and 8 weeks. Endotoxin, insulin,
S4
Abstracts / Atherosclerosis 182 (2005) S1–S8
IL-6 and TNF-␣ were analysed using ELISA kits. Isolated adipocytes were cultured with or without 100 ng/mL LPS (14 h), IL-6 and TNF-␣ were measured in the conditioned media. Serum bacterial endotoxin levels were 43% higher in T2DM patients, and correlated with fasting insulin in the control subjects (r = 0·604, p = 0·0001). Treatment with rosiglitazone, reduced fasting serum insulin levels (↓89%, p = 0·018) and serum endotoxin (↓42%, p = 0·0229). Change in endotoxin correlated with change in fasting insulin. Endotoxin stimulated the secretion of pro-inflammatory cytokines from adipose tissue. Hyperinsulinaemia-associated inflammation in obesity and diabetes may be partially mediated by endotoxin. VASCULAR BENEFITS OF ROSIGLTAZONE AND METFORMIN COMPARED WITH SULPHONYLUREA AND METFORMIN COMBINATION THERAPY M. Khanolkar2 , A.W. Thomas1 , S.K. Jackson2 , A. Roberts2 , L. George2 , J. Geen3 , A. Rees2 , R.H.K. Morris1 , L.M. Evans2 1 School
of Applied Sciences, UWIC, Cardiff; 2 Department of Medicine, Cardiff University; 3 Clinical Biochemistry Department, Prince Charles Hospital, Merthyr, UK Hypetension is a major factor in the increased cardiovascular risk associated with insulin resistance (IR) and type 2 diabetes (T2D). Rosiglitazone is an insulin sensitizing oral hypoglycaemic agent with putative cardiovascular benefits. We conducted a randomised double blind study assessing the vascular and anti-hypertensive effects of rosiglitazone. 50 T2D patients on metformin (HbA1c > 6.5%) received rosiglitazone (R), 8 mg daily or gliclazide (G) 160 mg daily. 24 h ambulatory blood pressure, aortic compliance (augmentation index (Aix)), HbA1c, lipid profile, HOMA-IR, hs-CRP, oxidative stress (TBARS) and soluble markers of endothelial activation (e-selectin) were measured at baseline and at 3 months. Baseline parameters were similar in both groups, while Hba1c fell to a similar extent in both groups. HOMA-IR (1.12 ± 0.3 (R) versus 0.45 ± 0.1 (G)), hs-CRP (0.44 ± 0.4 (R) versus 0.13 ± 0.09 ng/l (G)), e-selectin (64.5 ± 21.1 (R) versus 22.9 ± 12.9 mol/l(G)) and Aix (19.3% (R) versus 3.1% (G)) and TBARS all fell to a greater extent in the rosiglitazone group (p < 0.05). There were also significant reductions in mean 24 h BP (131.3/78.9 ± 10.1/5.4 to 133.7/80.4 ± 8.8/5.9 (G) versus 133.2/79.4 ± 9.2/6.4 to 126/75.4 ± 8.6/4.5 (R), p < 0.05) and mean arterial pressure in the rosiglitazone group (88.3 ± 3.9 to 90.3 ± 7.8 (G) versus 88.9 ± 5.9 to 83.5 ± 4.5 (R)) compared to baseline and the gliclazide group. These BP changes correlated with the change in augmentation index and insulin resistance. Compared with glicalzide, rosiglitazone appears to exert greater vascular benefits including blood pressure lowering properties. Outcome studies are awaited to confirm the impact of these effects on vascular events.
CORONARY CALCIFICATION, HOMOCYSTEINE, c-REACTIVE PROTEIN AND THE METABOLIC SYNDROME IN TYPE 2 DIABETES: THE PREDICT STUDY I.F. Godsland1 , R.S. Elkeles1 , M.D. Feher3 , M.D. Flather2 , F. Nugara2 , M. Rubens2 , W. Richmond1 , M. Khan2 , J. Donovan2 , V. Anyaoku1 , for the PREDICT Study Group 1 Imperial College London and St Mary’s; 2 Royal Brompton,
London, UK; 3 Chelsea and Westminster Hospitals, London, UK The PREDICT Study aims to determine: (i) the association between cardiovascular risk factors and coronary calcification score (CACS) obtained by electron beam tomography and (ii) the predictive value of CACS for CHD events in type 2 diabetes. Having previously investigated relationships between CACS and conventional risk factors, we have now studied the novel risk factors, plasma high sensitivity C-reactive protein (CRP) and homocysteine, HOMA insulin resistance, serum apoproteins A1 and B concentrations, and metabolic syndrome (WHO and NCEP) in 573 subjects of the PREDICT (type 2) diabetic cohort. In univariate analyses significant correlates of CACS included age, waist–hip ratio, male gender, duration of diabetes (all p < 0.0001), homocysteine (p = 0.0007), systolic blood pressure (p = 0.002), use of antihypertensive drugs (p = 0.002) and CRP (negative, p = 0.006). Lipid- and insulin resistance-related variables, including the Metabolic Syndrome were unrelated to CACS. In a multivariate model that included all significant univariate correlates, CACS was independently associated with age (p < 0.0001), WHR (p < 0.02), male gender (p < 0.05) and duration of diabetes (p < 0.05). The negative association with CRP also remained, and was independent of statin use. The association with homocysteine, was reduced to borderline significance (p = 0.10). Age was, therefore, the principal factor affecting CACS in Type 2 diabetes, with further independent contributions from WHR, male gender and diabetes per se and a possible independent effect of homocysteine. Other cardiovascular risk factors appeared to have little positive effect. INCREASED RISK OF CARDIOVASCULAR DISEASE IN WOMEN WITH RHEUMATOID ARTHRITIS A.P. Yates1 , P.W. Pemberton1 , I. Laing2 , Y. Ahmad3 , D. Lokko4 , I.N. Bruce3 1 Clinical Research; 2 Clinical Biochemistry; 3 ARC Epidemi-
ology Unit; 4 Rheumatism Research Centre, Manchester Royal Infirmary, Oxford Road, Manchester M13 9WL, UK Background: Women with rheumatoid arthritis (RA) have increased morbidity and mortality due to coronary heart disease. Chronic systemic inflammation promotes atherosclerosis and arterial stiffness in patients, but other mechanisms may also be involved.
Abstracts / Atherosclerosis 182 (2005) S1–S8
Methods: Markers of lipid peroxidation, antioxidant status, inflammation and insulinaemia were measured in blood and urine from 46 RA patients and 48 age-matched controls. Plaque formation was measured using B-mode carotid Doppler scan. Current RA activity was assessed using disease activity score (DAS28). Results: Plaque was significantly increased (p = 0.042) in RA patients between 50–59 years old compared to agematched controls. Levels of 8-isoprostane, C-reactive protein (CRP), IL-6, insulin and adiponectin were all significantly increased (p = 0.004, p < 0.001, p < 0.001, p = 0.035, p = 0.012, respectively) in RA patients. Levels of selenium and LDL-cholesterol were both significantly decreased (p = 0.003, p = 0.025, respectively) in RA patients. 8isoprostane was positively associated with 10-year cardiac risk (r = 0.55, p < 0.001) in RA patients and more weakly with plaque (r = 0.26, p = 0.011) in all subjects. IL-6 was more strongly associated (r = 0.64, p < 0.001) with DAS28 than CRP (r = 0.32, p = 0.041), but neither marker correlated with plaque in RA patients. Insulin correlated positively with body mass index (r = 0.59, p < 0.001), systolic blood pressure (r = 0.37, p = 0.013) and glucose (r = 0.31, p = 0.034), and negatively with adiponectin (r = −0.43, p = 0.003). In the RA group, multivariate analysis showed that no single variable predicted plaque. Conclusions: The burden of atherosclerosis is increased in middle-aged women with RA. Oxidative stress and developing insulin resistance may work synergistically with inflammation to contribute to the atherogenic process. ACCURACY OF LIPID MEASUREMENTS IN UK LABORATORIES REVIEWED BY EXTERNAL QUALITY ASSESSMENT: DOES THIS MATCH THE EXPECTATIONS REQUIRED BY GUIDELINES FOR RISK CALCULATION AND GP PERFORMANCE TARGETS? R. Cramb1 , F. MacKenzie2 on behalf of the HEART UK Laboratory Sub-committee 1 Department
of Clinical Biochemistry, University Hospital Birmingham NHS Foundation Trust, Queen Elizabeth Hospital, Birmingham B15 2TH; 2 UK NEQAS Wolfson EQA Laboratory, P.O. Box 3909, Birmingham B15 2UE, UK The accurate measurement of lipids has taken a high priority with the need to assess primary risk for CHD and to meet targets in the new GP contracts. In the USA, since 1987, the National Cholesterol Education Program (NCEP) has set standards for lipid measurements. No such targets have been set in the UK. HEART UK with the UK National External Quality Assessment Scheme (UK NEQAS) for Lipid Investigations has circulated three single donation samples for lipid measurement each having values assigned by the CDC secondary reference laboratory in Glasgow. Our review shows that cholesterol measurements are accurate when triglyceride concentrations are low. However, increased triglycerides produce a negative bias in measured
S5
cholesterol of up to −20% dependent on the method used. HDL cholesterol methods showed a general positive bias that varied between 5 and 10%, with some methods showing a 30% bias. Triglyceride concentrations were negatively biased at low levels, but as their concentrations increased a method-dependent bias of anywhere between +12 and −15% was shown. These data support previous observations seen with pooled samples, distributed by UK NEQAS. Lipid results obtained from UK laboratories may adversely influence the assessment and treatment of patients using Framingham based risk scores and some patients in general practice may be denied access to treatment on the basis of current clinical targets. UK laboratory lipid measurements are not universally accurate and require re-examination. THE UTILITY OF DNA-BASED TESTING IN PATIENTS WITH FAMILIAL HYPERCHOLESTEROLAEMIA IN THE UK S.E. Humphries1 , R. Whittall1 , C. Hubbard1 , J.A. Cooper1 , A. Soutar2 , R. Naoumova2 , G.R. Thompson2 , M. Seed2 , P.N. Durrington4 , J.P. Miller5 , D.J.B. Betteridge1 , A. Neil6 1 RF&UCL; 2 Imperial
College; 3 University of Manchester;
5 Wythenshawe Hospital, Manchester; 6 University of Oxford,
UK Aims: To determine the impact of mutations in the genes for LDLR, APOB, PCSK9 in UK patients with definite familial hypercholesterolaemia (FH) in determining risk of coronary heart disease (CHD). Subjects: Four hundred and ten FH patients recruited from the UK Simon Broome Register, 158 with documented CHD. Methods: The LDLR was partially screened for mutations with SSCP (exons 3, 4, 6–10 and 14) and using a commercial kit for gross deletions and rearrangements. Common mutations in APOB (R3500Q) and PCSK9 (D374Y) were tested by specific assays. Results: Mutations were detected in 253 (61.9%) patients; 236 (57.7%) in LDLR; 10 (2.4%) in APOB and 7 (1.7%) in PCSK9. After adjusting for age, sex, smoking (ever versus never) and systolic blood pressure, the odds ratio of having CHD in those with an LDLR mutation was 1.84 (95% CI 1.10–3.06), for APOB 3.40 (0.71–16.36), and for PCSK9 19.96 (1.88–211.5), (p = 0.0003 overall). This high risk in LDLR and PCSK9-carrying subjects was partly explained by their poorer lipid profile than other groups (LDLR versus PCSK9 versus no-mutation, pre-treatment cholesterol 10.29 ± 1.85 mmol/l versus 13.12 ± 1.69 mmol/l versus 9.85 ± 1.90 mmol/l, (p = 0.0002). Conclusions: The higher CHD risk in patients carrying a detected LDLR mutation supports the usefulness of DNA testing in the management of FH patients. The relatively high prevalence of the D374Y mutation in PCSK9 and its high risk of CHD suggests that testing for this mutation will be clinically useful.
S6
Abstracts / Atherosclerosis 182 (2005) S1–S8
COMPARISON OF A NOVEL AUTOMATED ASSAY WITH TRADITIONAL TRIOLEIN ASSAY FOR DETERMINATION OF LIPOPROTEIN AND HEPATIC LIPASE ACTIVITIES A.S. Wierzbicki, S. Suryanarayanan, K. Lyttle, P.J. Lumb Department of Chemical Pathology, St. Thomas’ Hospital, London SE1 7EH, UK Previous methods to measure heparinised plasma total lipase (TL), lipoprotein lipase (LPL) and hepatic lipase (HL) activities have been time-consuming or imprecise. An automated multichannel analyser fluorescence-emission method using a commercial substrate was developed for assessment of lipase activities and compared with results obtained from a radioactive 14 C-triolein suspension assay. The automated assay was linear y = 0.199 + 12.42x; r2 = 0.985 with a detection limit of 0.50 pmol/ml/min. In males, the average and standard deviation for LPL and HL activity were 20.5 (7.2) pmol/ml/min and 7.6 (5.7) pmol/ml/min respectively. In females the ranges were 7.4(3.9) pmol/ml/min for LPL and 0.8(0.5) pmol/ml/min for HL. Assays in male patients (n = 15) with severe hypertriglyceridemia gave a median LPL of 3.1 (0.0–8.55) pmol/ml/min and HL 0.5 (0.1–2.5) pmol/ml/min while in women patients (n = 4) values were LPL 0.7 (0.0–1.1) pmol/ml/min and HL 1.0 (0.1–1.4) pmol/ml/min. All patients had significantly lower total and lipoprotein lipase activities. Hepatic lipase activities were non-significantly reduced compared to healthy controls. The new assay showed positive bias’ for TL and LPL and a profound negative bias for HL. The assay suffered interference from bilirubin, urea and creatinine and also from added plasma. The new automated method gives more precise results compared to the previous assay technique with considerable advantages in requiring less plasma, being more reproducible, non-radioactive and considerably faster and also in not giving negative results for LPL activity calculated by difference in high versus low salt conditions. DOES NIEMANN PICK C1-LIKE-1 PROTEIN REGULATES CHYLOMICRON ASSEMBLY THROUGH ITS EFFECT ON MTP? S. Lally, Tun T. Kyaw, D. Owens, G. Tomkin Department of Endocrinology, Trinity College Dublin, UK The Niemann Pick C1-like 1 (NP C1-L1) protein regulates cholesterol absorption but intestinal expression has not been described in humans. Aim: to investigate the relationship between intestinal mRNA levels of the NP C1-L1 and microsomal triglyceride transfer protein (MTP), factors influencing chylomicron composition. D2 gastroscopy biopsies were examined. Ethics committee approval and informed consent were obtained. Subjects (21) with normal duodenal biopsies were examined. The subjects were given an 1100 kcal fat meal and blood taken at 6 h. NP C1-L1 and MTP mRNA was measured using
real-time RT-PCR. Chylomicrons (density <0.006 g/l) were isolated and apo B48 and B100 determined. Results: Fasting serum cholesterol was 4.65 ± 1.1 mmol/l and serum triglyceride median 1.18 mmol/l. There was a significant positive correlation between NP C1-L1 mRNA and chylomicron apo B48 at 6h (r = 0.64, p < 0.05) and apo B100 (r = 0.75, p < 0.01). There was a positive correlation between NP C1-L1 mRNA and chylomicron cholesterol (r = 0.79, p < 0.01). There was a significant correlation between NP C1-L1 mRNA and microsomal triglyceride transfer protein (MTP) mRNA (r = 0.84, p < 0.01) but no correlation between NP C1-L1 mRNA and ABCG5 or G8 mRNA. Conclusion: This study suggests that NP C1-L1 plays a pivitol role in regulation of MTP and thus in the assembly of large triglyceride-rich lipoproteins. Since these particles are potentially atherogenic, inhibition of NP C1-L1 may have an important role to play in the reduction atherosclerosis. HUMAN HEPATITIS c VIRUS RNA CIRCULATES PREDOMINANTLY IN VLDL-LIKE LIPO-VIRALPARTICLES Soren U. Nielsen1 , Alastair D. Burt1,2 , Caroline Martin1 , Margaret F. Bassendine1 , Geoffrey L. Toms1 1 School of Clinical Medical Sciences, University of Newcastle, Newcastle upon Tyne, UK; 2 School of Clinical and Laboratory Sciences, University of Newcastle, Newcastle upon Tyne, UK
Introduction: Hepatitis C virus (HCV) RNA circulates in the blood of infected patients in heterogeneous particles associated with lipoprotein. The majority of HCV RNA is found in lipo-viro-particles (LVP) with density below 1.10 g/ml and these are the infectious particle of HCV. We have characterized these HCV-LVPs and sought to release virion particles from association with host lipoprotein by treatment with deoxycholic acid. Methods: Serum was fractionated in isotonic iodixanol density gradients and HCV RNA content of the fractions was assessed by quantitative real-time RT-PCR. Peak fractions containing HCV were further analysed by rate-zonal centrifugation to determine their sedimentation coefficients, and thereby, their size according to Stokes law. HCV RNA containing particles in gradient fractions were immunoprecipitated with polyclonal anti-apolipoprotein antibodies bound to protein G Sepharose. Results: Serum derived HCV RNA with a density below 1.10 g/ml could be precipitated with antibodies against apoB or apoE, suggesting an association of HCV RNA with host VLDL. Treatment with deoxycholic acid released HCV RNA containing putative virion particles with density 1.12 g/ml, a sedimentation coefficient of 215 S, a diameter of 57 nm and which remained precipitable with anti-apoB but not with anti-apoE. Conclusion: Blood derived HCV RNA was found to be predominantly associated with host lipoproteins containing
Abstracts / Atherosclerosis 182 (2005) S1–S8
apoB-100 and apoE. Thus the HCV-LVP appears to be a VLDL-like particle. HCV may utilise the lipid transport function of VLDL to promote the spread of infectious particles. RESPONSE TO ATORVASTATIN IN A CASE OF IDIOPATHIC HYPOALBUMINAEMIA COMPLICATED BY REMNANT HYPERLIPIDAEMIA C.S. Arun1 , H.K. Datta1,3 , R.D.G. Neely1,2 1 Lipid
and Metabolic Clinic; 2 Department of Clinical Biochemistry, Royal Victoria Infirmary; 3 School of Clinical and Laboratory Sciences, University of Newcastle, Newcastle upon Tyne, UK Idiopathic hypoalbuminaemia is a rare inherited disorder of primary albumin deficiency, usually associated with lipodstrophy, hyperlipoproteinaemia and compensatory increases in the plasma levels of other transport proteins. Although usually considered benign, the impact of this condition on the long-term risk of cardiovascular disease and other health outcomes is less certain. We report the case of an otherwise healthy, 27-year-old man with no family history of hyperlipidaemia who presented with subcutaneous lipomata, unexplained hypoalbuminaemia and severe mixed hyperlipidaemia (total cholesterol 15.4 mmol/L triglyceride 3.2 mmol/L). Following initial treatment with atorvastatin, lipid levels improved (total cholesterol 6.8 mmol/L, triglyceride 1.8 mmol/L) but elevated serum alphafetoprotein, liver enzymes and creatine kinase were noted. Despite the absence of oedema, serum albumin, measured using an immunonephelometric method, was 9 g/L. Albumin was undetectable in the urine (<5 mg/L). Following withdrawal of atorvastatin for 6 weeks, HDL-cholesterol and apolipoprotein A1 decreased, while apolipoprotein B, total cholesterol, LDL-, VLDLcholesterol and the VLDL-cholesterol/triglyceride ratio increased and beta-VLDL appeared in plasma. However, serum enzyme levels remained elevated. Apolipoprotein E was also increased and apoE genotying revealed homozygosity for ApoE2, consistent with remnant hyperlipidaemia. Remnant hyperlipidaemia has not previously been reported in association with idiopathic hypoalbuminaemia and may confer a greatly increased cardiovascular risk in this case, requiring treatment even in the absence of conventional cardiovascular risk factors.
S7
laemia (FH) and to compare the prevalence of clinical FH (elevated LDL-C) in relatives of probands with and without a detected mutation. Methods: Patients with a Simon Broome diagnosis of definite FH were recruited by a genetic nurse from the Lipid Clinic at the Royal Free Hospital. Patients were given information on the inheritance of FH, a pedigree was drawn, and a DNA test offered to try and identify the FH mutation, and where found, DNA testing offered to relatives. Results: Thirty three probands tested to date included 17 men and 16 women, ages ranging from 25 to 77 years (mean 54.2 years, median 56 years), with 13 (39%) with tendon xanthomas. Mutations were detected in 64% (21) of probands (4 APOB, and 17 LDLR). Fifteen different mutations were identified. The detailed pedigrees of 31 of the probands identified a total of 209 1st degree relatives, (3–15 per proband) of which 162 were living. Of these, 60.8% (127) were from probands with identified mutations, and 39.2% (82) from those with no identified mutation. Conclusion: The majority of FH patients have significant numbers of family members who could be identified by cascade testing. On average this should lead to contacting five 1st degree relatives per proband, of whom 50% should also have FH. These will become new probands for further cascading. The relative prevalence of clinical FH in mutation carriers and non-carriers will be compared. RELATIVE EFFICACY OF COMMERCIALLY AVAILABLE PLANT STEROL-CONTAINING FOODS ON PLASMA CHOLESTEROL AND HDL SUBCLASSES Nicola Harman, Kate Tabbernor, Bruce A. Griffin School of Biomedical and Molecular Sciences, University of Surrey, Guildford, Surrey GU2 7XH, UK
J.M. Farrer1 , D. Nair2 , G. Norbury3 , A. Taylor3 , S.E. Humphries1
Plant sterols have been shown to lower total and LDL cholesterol by 10–14% (∼0.45 mmol/l). Their effects on HDL cholesterol are neutral, and on HDL subclasses unknown. A study was designed to examine the relative efficacy of a range of sterol-containing food products (Flora Procactiv) on lowering cholesterol and to facilitate the development of a new technique for the separation of HDL subclasses. Three groups of normal, healthy subjects consumed the recommended daily intake of either spread (S, n = 6), milk (M, n = 7) or yogurt (Y, n = 7) for 2 weeks. Blood samples were taken at the beginning and end of the intervention and after a further 2 weeks for the analysis of plasma cholesterol (TC), triglycerides (TAG), HDL cholesterol and HDL subclasses. The latter were measured by iodixanol gradient ultracentrifugation.
1 London
%
DOES DNA INFORMATION COMPLEMENT LIPID MEASURES IN CASCADE TESTING IN FAMILIAL HYPERCHOLESTEROLAEMIA PATIENTS?
IDEAS Genetics Knowledge Park; 2 Royal Free Hospital; 3 Regional Molecular Genetics Laboratory, GOSH, London, UK Aim: To evaluate the utility of DNA testing in the diagnosis of patients and their relatives with familial hypercholestero-
TC
2 weeks +4.0Y , −18.7M* , −8.6S* 4 weeks −2.7Y , +6.9M , +7.2S
LDL-C
HDL-C
−0.5Y , −22.6M* , 11.5S +1.2Y , +10.0M , +11.0S
+6.3Y , −2.5M , −1.0S −3.0Y , −1.3M , −1.3S
Iodixanol centrifugation was shown to provide a more cost effective and rapid alternative to our existing techniques
S8
Abstracts / Atherosclerosis 182 (2005) S1–S8
for the separation of HDL2 and HDL3 subclasses. Total cholesterol was significantly decreased after 2 weeks on milk and spread (* p < 0.05) and returned to baseline levels 2 weeks after the intervention. In contrast, yogurt had no effect on plasma cholesterol. There was no overall effect on plasma TAG, HDL or HDL subclasses. In conclusion, while manufacturers recommend that the daily intake of plant sterols is achieved through a combination of all three products, this study demonstrates disparity in their individual cholesterol-lowering efficacy. CARDIOVASCULAR EFFECTS OF CONJUGATED LINOLEIC ACID AND FIBRATES: EVIDENCE FOR AN ANTIPLATELET EFFECT MEDIATED BY PLATELET PPAR␣ M. Khanolkar2 , A.W. Thomas1 , A. Roberts2 , L. George2 , J. Geen3 , A. Rees2 , R.H.K. Morris1 , L.M. Evans2 1 School
of Applied Sciences, UWIC, Cardiff; 2 Department of Medicine, Cardiff University; 3 Clinical Biochemistry Department, Prince Charles Hospital, Merthyr, UK Increased platelet activation is a recognised feature of insulin resistance (IR) and type 2 diabetes (T2D) and is an important factor in atherothrombosis. Recent studies have shown presence of functional PPAR␥ in human platelets capable of inhibiting platelet function. Conjugated linoleic acid (CLA), a ligand for both PPAR␣ and ␥ has been shown to inhibit platelet aggregation. We therefore hypothesised the presence of PPAR␣ in human platelets. Flow cytometry followed by western blotting confirmed the presence of PPAR␣ in human platelets, while Electrphoretic mobility shift assay (EMSA) confirmed the DNA binding ability of platelet PPAR␣. We also conducted in vitro studies on platelet rich plasma to study direct effects of Clofibrate (PPAR␣ agonist) and CLA at physiological concentrations on ADP induced platelet aggregation. A significant reduction platelet aggregation was observed with both compounds (Clofibrate:16.23 ± 4.26 ⇒ 6.14 ± 2.87 , p < 0.01; CLA:15.64 ± 3.97 ⇒ 3.11 ± 1.76 , p < 0.01). Pre-treating platelets with a PPAR␣-antagonist (MK-886) reduced the anti-platelet aggregatory effects of clofibrate and CLA, while pre-treatment with PPAR-␥ antagonist (GW-9662) did not alter the effects of CLA on platelet aggregation. This is the first study to demonstrate the presence of functionally active PPAR␣ with preserved DNA binding ability in human platelets and that PPAR␣ activation in platelets inhibits platelet activation. This study suggests that the potential cardiovascular benefits of both PPAR␣ and PPAR ␥ agonists may be partly mediated through modulating platelet function.
PULSE WAVE INTENSITY AND THE POTENTIAL IDENTIFICATION OF SUBCLINICAL CARDIOVASCULAR DISEASE IN TREATMENT NA¨ıVE TYPE 2 DIABETIC PATIENTS Aled Wyn Roberts1,2 , Frank Rakebrandt2 , Janice Swampillai2 , Laila Jones1 , Marc Evans1 , Alan Rees1 , Alan Fraser2 1 Department
of Diabetes and Endocrinology, University Hospital of Wales, Cardiff; 2 Wales Heart Research Institute, University Hospital of Wales, Cardiff, UK Background: Type 2 diabetes (T2D) is associated with increased risk of cardiovascular disease (CVD), which may be present prior to diagnosis of T2D. Diabetic cardiomyopathy in patients without coronary disease may be related to pathological changes in ventriculo-arterial coupling, with augmented central afterload caused by increased wave reflections from stiff conduit arteries. Pulse wave intensity has been developed as a non-invasive method of assessing the interaction of left ventricular contractility and relaxation with waves reflected from the periphery. We measured wave intensity in the common carotid artery (CCA) using prototype hardware (Aloka SSD-5500, 7.5 MHz lineararray probe) and analysed data using bespoke software (Matlab). Methods: We studied 18 subjects with T2D on diet control with no history of CHD. The right CCA was imaged, and arterial diameter measured by automated wall tracking was calibrated for pressure. Velocity was derived from Doppler colour flow. Wave intensity is the product of instantaneous rate of change of velocity and pressure. Local wave speed was calculated during early systole from a pressure-velocity loop, and separated forward and backwards travelling waves derived. Data from T2D subjects were compared with agematched controls, and statistical analyses performed using unpaired t-tests. Results: The ages of the T2D and control groups were comparable (means 67 versus 63 years). Systolic BP and MAP were higher in the T2D group. Epsilon and Beta, both measures of arterial stiffness, were significantly elevated in the T2D group (p = 0.02 and p < 0.01, respectively). The peak of the backward travelling compression wave, which augments central systolic pressure, was significantly higher in T2D subjects (p = 0.05). Conclusion: Analysis of our study group reveals reduced elasticity of the vasculature, in patients in the earliest stages of T2D. Non-invasive assessment of pulse wave intensity may provide evidence of subclinical vascular disease in high-risk patients, allowing the targeting of cardiovascular risk reduction therapy to those likely to most benefit.
Abstracts
H·E·A·R·T UK 19th Annual Medical and Scientific Meeting, University of Glamorgan, Pontypridd, Wales, UK, 30 June–1 July 2005 Available online 14 July 2005
0021-9150/$ – see front matter © 2005 Published by Elsevier Ireland Ltd. doi:10.1016/j.atherosclerosis.2005.05.015
Abstracts / Atherosclerosis 182 (2005) S1–S8
RELATIONSHIP OF THE METABOLIC SYNDROME AS ASSESSED BY DIFFERENT DEFINITIONS WITH CORONARY ARTERIAL DISEASE IN PAKISTANIS A.S. Wierzbicki1 , S. Nishtar2,3 , P.J. Lumb1 , M. LambertHammill1 , J. Gill2 1 Department
of Chemical Pathology, St. Thomas’ Hospital, London SE1 7EH, UK; 2 Department of Cardiology, St. Thomas’ Hospital, London SE1 7EH, UK; 3 Department of Cardiology, PIMS, Pakistan Metabolic syndrome is associated with coronary artery disease (CAD). This study investigated the significance of metabolic syndrome (M-IRS) as a CAD risk factor based on different definitions of abdominal obesity. This prospective case–control study recruited 400 individuals with chest pain; 200 with angiographic disease matched with 200 without occlusive disease from a tertiary referral cardiology unit in Pakistan. Metabolic syndrome was present in 37% of CAD(+) versus 27% in CAD(−) by caucasian criteria or 47% versus 42% by asian criteria (p < 0.001). After adjustment for other risk factors, M-IRS was not a significant predictor for CAD or angiographic disease. Age (p = 0.03), smoking (p < 0.001), sialic acid (p = 0.01) and creatinine (p = 0.008) accounted for the excess risk of CAD. Similarly age (p = 0.005), creatinine (p < 0.001), cigarette pack-years (p = 0.02), diabetes (p = 0.008) and sialic acid (p = 0.08) were predictors of greater angiographic disease. M-IRS differed in Pakistanis from caucasians as waist circumference showed weak correlations (r = −0.03 to 0.08; p = 0.45–0.52) with triglycerides, HDL-C, blood pressure or glucose. Sialic acid was the only inflammatory marker associated with M-IRS. Despite strong associations of individual risk factors with M-IRS and a univariate association of M-IRS with CAD, the principal discriminant risk factors in this group are age, smoking, inflammation, diabetes and impaired renal function. The poor sensitivity of M-IRS for CAD reflects either its high underlying prevalence reducing sensitivity, confounding by other urban lifestyle traits and/or a lack of association of waist circumference with M-IRS risk factors. The definition of M-IRS may have to be revised to increase its power as a discriminant risk factor for CHD in Pakistani populations. EXTENDING THE METABOLIC SYNDROME: INSULIN SENSITIVITY, LIVER FUNCTION, SUBCLINICAL INFLAMMATION AND HAEMATOLOGICAL MEASURES I.F. Godsland1,2 , D.G. Johnston1 for the HDDRISC Group2 1 Endocrinology
and Metabolic Medicine; 2 Wynn Department of Metabolic Medicine, Faculty of Medicine, Imperial College London, London, UK To refine the definition of the metabolic syndrome, large studies are needed, using sophisticated measures of insulin sensitivity with a broad range of potential correlates. Four
S3
hundred and seventy-two men (age 26–78 years) without evidence of diabetes or cardiovascular disease and not taking blood pressure-, lipid- or uric acid-lowering agents underwent an intravenous glucose tolerance test (IVGTT), with accompanying measurements of liver and renal function and haematology. Insulin sensitivity, SI , was measured by minimal model analysis of IVGTT glucose and insulin concentrations. Clustering of potential components of the metabolic syndrome was assessed by factor analysis. There were significant associations, independent of age, body mass index, alcohol intake, exercise and cigarette smoking, between SI and serum gamma glutamyl transferase (GGT), aspartate transaminase and alkaline phosphatase (ALP) activities, white blood cell count (WBC), erythrocyte sedimentation rate (ESR) and serum globulin concentration (all p < 0.01). On factor analysis, the principal factor (explaining 49.4% of variance) included in decreasing order of importance, body mass index, SI (negative), the HDL cholesterol/triglyceride concentration ratio (negative), fasting insulin, GGT activity, serum uric acid concentration, mean arterial pressure, ALP activity, blood haemoglobin concentration and fasting glucose. Although they are independent correlates of insulin sensitivity, indices of subclinical inflammation (WBC, ESR and globulin) may not cluster with other metabolic syndrome components. Raised GGT activity appears to be a prominent feature of the metabolic syndrome. HYPERINSULINAEMIA IS ASSOCIATED WITH ENDOTOXIN-MEDIATED INFLAMMATION IN OBESITY AND TYPE 2 DIABETES MELLITUS S.J. Creely1,2 , P.G. McTernan1 , F.F.M. M. Khanolkar3 , M. Evans3 , M. Kumar1,2
Fisher1 ,
1 Diabetes
and Metabolism Research Laboratories, Clinical Sciences Research Institute, Warwick Medical School, Coventry; 2 University Hospital of Coventry and Warwickshire, Coventry; 3 Heart Research Institute, Cardiff University, UK Obesity and type 2 diabetes (T2DM) are associated inflammation. Hyperinsulinaemia associates with bowel stasis, bacterial overgrowth and potentially increased gut permeability, suggesting gut as a source of endotoxin. We hypothesised that serum endotoxin is increased in association with hyperinsulinaemia and may cause the sub-clinical inflammation in T2DM. Further, the reduction of insulin achieved through the use of an insulin sensitizer leads to a reduction in serum endotoxin. Fasting serum samples were obtained from 51 T2DM patients with no complications or inflammatory conditions. 23 patients were on diet alone, 20 on oral agents, 7 on oral agents and insulin and 1 on insulin alone. Serum was also obtained from 33 age matched non-diabetic control subjects. A sub-group of previously untreated T2DM patients received rosiglitazone with fasting serum collected at baseline and 8 weeks. Endotoxin, insulin,
S4
Abstracts / Atherosclerosis 182 (2005) S1–S8
IL-6 and TNF-␣ were analysed using ELISA kits. Isolated adipocytes were cultured with or without 100 ng/mL LPS (14 h), IL-6 and TNF-␣ were measured in the conditioned media. Serum bacterial endotoxin levels were 43% higher in T2DM patients, and correlated with fasting insulin in the control subjects (r = 0·604, p = 0·0001). Treatment with rosiglitazone, reduced fasting serum insulin levels (↓89%, p = 0·018) and serum endotoxin (↓42%, p = 0·0229). Change in endotoxin correlated with change in fasting insulin. Endotoxin stimulated the secretion of pro-inflammatory cytokines from adipose tissue. Hyperinsulinaemia-associated inflammation in obesity and diabetes may be partially mediated by endotoxin. VASCULAR BENEFITS OF ROSIGLTAZONE AND METFORMIN COMPARED WITH SULPHONYLUREA AND METFORMIN COMBINATION THERAPY M. Khanolkar2 , A.W. Thomas1 , S.K. Jackson2 , A. Roberts2 , L. George2 , J. Geen3 , A. Rees2 , R.H.K. Morris1 , L.M. Evans2 1 School
of Applied Sciences, UWIC, Cardiff; 2 Department of Medicine, Cardiff University; 3 Clinical Biochemistry Department, Prince Charles Hospital, Merthyr, UK Hypetension is a major factor in the increased cardiovascular risk associated with insulin resistance (IR) and type 2 diabetes (T2D). Rosiglitazone is an insulin sensitizing oral hypoglycaemic agent with putative cardiovascular benefits. We conducted a randomised double blind study assessing the vascular and anti-hypertensive effects of rosiglitazone. 50 T2D patients on metformin (HbA1c > 6.5%) received rosiglitazone (R), 8 mg daily or gliclazide (G) 160 mg daily. 24 h ambulatory blood pressure, aortic compliance (augmentation index (Aix)), HbA1c, lipid profile, HOMA-IR, hs-CRP, oxidative stress (TBARS) and soluble markers of endothelial activation (e-selectin) were measured at baseline and at 3 months. Baseline parameters were similar in both groups, while Hba1c fell to a similar extent in both groups. HOMA-IR (1.12 ± 0.3 (R) versus 0.45 ± 0.1 (G)), hs-CRP (0.44 ± 0.4 (R) versus 0.13 ± 0.09 ng/l (G)), e-selectin (64.5 ± 21.1 (R) versus 22.9 ± 12.9 mol/l(G)) and Aix (19.3% (R) versus 3.1% (G)) and TBARS all fell to a greater extent in the rosiglitazone group (p < 0.05). There were also significant reductions in mean 24 h BP (131.3/78.9 ± 10.1/5.4 to 133.7/80.4 ± 8.8/5.9 (G) versus 133.2/79.4 ± 9.2/6.4 to 126/75.4 ± 8.6/4.5 (R), p < 0.05) and mean arterial pressure in the rosiglitazone group (88.3 ± 3.9 to 90.3 ± 7.8 (G) versus 88.9 ± 5.9 to 83.5 ± 4.5 (R)) compared to baseline and the gliclazide group. These BP changes correlated with the change in augmentation index and insulin resistance. Compared with glicalzide, rosiglitazone appears to exert greater vascular benefits including blood pressure lowering properties. Outcome studies are awaited to confirm the impact of these effects on vascular events.
CORONARY CALCIFICATION, HOMOCYSTEINE, c-REACTIVE PROTEIN AND THE METABOLIC SYNDROME IN TYPE 2 DIABETES: THE PREDICT STUDY I.F. Godsland1 , R.S. Elkeles1 , M.D. Feher3 , M.D. Flather2 , F. Nugara2 , M. Rubens2 , W. Richmond1 , M. Khan2 , J. Donovan2 , V. Anyaoku1 , for the PREDICT Study Group 1 Imperial College London and St Mary’s; 2 Royal Brompton,
London, UK; 3 Chelsea and Westminster Hospitals, London, UK The PREDICT Study aims to determine: (i) the association between cardiovascular risk factors and coronary calcification score (CACS) obtained by electron beam tomography and (ii) the predictive value of CACS for CHD events in type 2 diabetes. Having previously investigated relationships between CACS and conventional risk factors, we have now studied the novel risk factors, plasma high sensitivity C-reactive protein (CRP) and homocysteine, HOMA insulin resistance, serum apoproteins A1 and B concentrations, and metabolic syndrome (WHO and NCEP) in 573 subjects of the PREDICT (type 2) diabetic cohort. In univariate analyses significant correlates of CACS included age, waist–hip ratio, male gender, duration of diabetes (all p < 0.0001), homocysteine (p = 0.0007), systolic blood pressure (p = 0.002), use of antihypertensive drugs (p = 0.002) and CRP (negative, p = 0.006). Lipid- and insulin resistance-related variables, including the Metabolic Syndrome were unrelated to CACS. In a multivariate model that included all significant univariate correlates, CACS was independently associated with age (p < 0.0001), WHR (p < 0.02), male gender (p < 0.05) and duration of diabetes (p < 0.05). The negative association with CRP also remained, and was independent of statin use. The association with homocysteine, was reduced to borderline significance (p = 0.10). Age was, therefore, the principal factor affecting CACS in Type 2 diabetes, with further independent contributions from WHR, male gender and diabetes per se and a possible independent effect of homocysteine. Other cardiovascular risk factors appeared to have little positive effect. INCREASED RISK OF CARDIOVASCULAR DISEASE IN WOMEN WITH RHEUMATOID ARTHRITIS A.P. Yates1 , P.W. Pemberton1 , I. Laing2 , Y. Ahmad3 , D. Lokko4 , I.N. Bruce3 1 Clinical Research; 2 Clinical Biochemistry; 3 ARC Epidemi-
ology Unit; 4 Rheumatism Research Centre, Manchester Royal Infirmary, Oxford Road, Manchester M13 9WL, UK Background: Women with rheumatoid arthritis (RA) have increased morbidity and mortality due to coronary heart disease. Chronic systemic inflammation promotes atherosclerosis and arterial stiffness in patients, but other mechanisms may also be involved.
Abstracts / Atherosclerosis 182 (2005) S1–S8
Methods: Markers of lipid peroxidation, antioxidant status, inflammation and insulinaemia were measured in blood and urine from 46 RA patients and 48 age-matched controls. Plaque formation was measured using B-mode carotid Doppler scan. Current RA activity was assessed using disease activity score (DAS28). Results: Plaque was significantly increased (p = 0.042) in RA patients between 50–59 years old compared to agematched controls. Levels of 8-isoprostane, C-reactive protein (CRP), IL-6, insulin and adiponectin were all significantly increased (p = 0.004, p < 0.001, p < 0.001, p = 0.035, p = 0.012, respectively) in RA patients. Levels of selenium and LDL-cholesterol were both significantly decreased (p = 0.003, p = 0.025, respectively) in RA patients. 8isoprostane was positively associated with 10-year cardiac risk (r = 0.55, p < 0.001) in RA patients and more weakly with plaque (r = 0.26, p = 0.011) in all subjects. IL-6 was more strongly associated (r = 0.64, p < 0.001) with DAS28 than CRP (r = 0.32, p = 0.041), but neither marker correlated with plaque in RA patients. Insulin correlated positively with body mass index (r = 0.59, p < 0.001), systolic blood pressure (r = 0.37, p = 0.013) and glucose (r = 0.31, p = 0.034), and negatively with adiponectin (r = −0.43, p = 0.003). In the RA group, multivariate analysis showed that no single variable predicted plaque. Conclusions: The burden of atherosclerosis is increased in middle-aged women with RA. Oxidative stress and developing insulin resistance may work synergistically with inflammation to contribute to the atherogenic process. ACCURACY OF LIPID MEASUREMENTS IN UK LABORATORIES REVIEWED BY EXTERNAL QUALITY ASSESSMENT: DOES THIS MATCH THE EXPECTATIONS REQUIRED BY GUIDELINES FOR RISK CALCULATION AND GP PERFORMANCE TARGETS? R. Cramb1 , F. MacKenzie2 on behalf of the HEART UK Laboratory Sub-committee 1 Department
of Clinical Biochemistry, University Hospital Birmingham NHS Foundation Trust, Queen Elizabeth Hospital, Birmingham B15 2TH; 2 UK NEQAS Wolfson EQA Laboratory, P.O. Box 3909, Birmingham B15 2UE, UK The accurate measurement of lipids has taken a high priority with the need to assess primary risk for CHD and to meet targets in the new GP contracts. In the USA, since 1987, the National Cholesterol Education Program (NCEP) has set standards for lipid measurements. No such targets have been set in the UK. HEART UK with the UK National External Quality Assessment Scheme (UK NEQAS) for Lipid Investigations has circulated three single donation samples for lipid measurement each having values assigned by the CDC secondary reference laboratory in Glasgow. Our review shows that cholesterol measurements are accurate when triglyceride concentrations are low. However, increased triglycerides produce a negative bias in measured
S5
cholesterol of up to −20% dependent on the method used. HDL cholesterol methods showed a general positive bias that varied between 5 and 10%, with some methods showing a 30% bias. Triglyceride concentrations were negatively biased at low levels, but as their concentrations increased a method-dependent bias of anywhere between +12 and −15% was shown. These data support previous observations seen with pooled samples, distributed by UK NEQAS. Lipid results obtained from UK laboratories may adversely influence the assessment and treatment of patients using Framingham based risk scores and some patients in general practice may be denied access to treatment on the basis of current clinical targets. UK laboratory lipid measurements are not universally accurate and require re-examination. THE UTILITY OF DNA-BASED TESTING IN PATIENTS WITH FAMILIAL HYPERCHOLESTEROLAEMIA IN THE UK S.E. Humphries1 , R. Whittall1 , C. Hubbard1 , J.A. Cooper1 , A. Soutar2 , R. Naoumova2 , G.R. Thompson2 , M. Seed2 , P.N. Durrington4 , J.P. Miller5 , D.J.B. Betteridge1 , A. Neil6 1 RF&UCL; 2 Imperial
College; 3 University of Manchester;
5 Wythenshawe Hospital, Manchester; 6 University of Oxford,
UK Aims: To determine the impact of mutations in the genes for LDLR, APOB, PCSK9 in UK patients with definite familial hypercholesterolaemia (FH) in determining risk of coronary heart disease (CHD). Subjects: Four hundred and ten FH patients recruited from the UK Simon Broome Register, 158 with documented CHD. Methods: The LDLR was partially screened for mutations with SSCP (exons 3, 4, 6–10 and 14) and using a commercial kit for gross deletions and rearrangements. Common mutations in APOB (R3500Q) and PCSK9 (D374Y) were tested by specific assays. Results: Mutations were detected in 253 (61.9%) patients; 236 (57.7%) in LDLR; 10 (2.4%) in APOB and 7 (1.7%) in PCSK9. After adjusting for age, sex, smoking (ever versus never) and systolic blood pressure, the odds ratio of having CHD in those with an LDLR mutation was 1.84 (95% CI 1.10–3.06), for APOB 3.40 (0.71–16.36), and for PCSK9 19.96 (1.88–211.5), (p = 0.0003 overall). This high risk in LDLR and PCSK9-carrying subjects was partly explained by their poorer lipid profile than other groups (LDLR versus PCSK9 versus no-mutation, pre-treatment cholesterol 10.29 ± 1.85 mmol/l versus 13.12 ± 1.69 mmol/l versus 9.85 ± 1.90 mmol/l, (p = 0.0002). Conclusions: The higher CHD risk in patients carrying a detected LDLR mutation supports the usefulness of DNA testing in the management of FH patients. The relatively high prevalence of the D374Y mutation in PCSK9 and its high risk of CHD suggests that testing for this mutation will be clinically useful.
S6
Abstracts / Atherosclerosis 182 (2005) S1–S8
COMPARISON OF A NOVEL AUTOMATED ASSAY WITH TRADITIONAL TRIOLEIN ASSAY FOR DETERMINATION OF LIPOPROTEIN AND HEPATIC LIPASE ACTIVITIES A.S. Wierzbicki, S. Suryanarayanan, K. Lyttle, P.J. Lumb Department of Chemical Pathology, St. Thomas’ Hospital, London SE1 7EH, UK Previous methods to measure heparinised plasma total lipase (TL), lipoprotein lipase (LPL) and hepatic lipase (HL) activities have been time-consuming or imprecise. An automated multichannel analyser fluorescence-emission method using a commercial substrate was developed for assessment of lipase activities and compared with results obtained from a radioactive 14 C-triolein suspension assay. The automated assay was linear y = 0.199 + 12.42x; r2 = 0.985 with a detection limit of 0.50 pmol/ml/min. In males, the average and standard deviation for LPL and HL activity were 20.5 (7.2) pmol/ml/min and 7.6 (5.7) pmol/ml/min respectively. In females the ranges were 7.4(3.9) pmol/ml/min for LPL and 0.8(0.5) pmol/ml/min for HL. Assays in male patients (n = 15) with severe hypertriglyceridemia gave a median LPL of 3.1 (0.0–8.55) pmol/ml/min and HL 0.5 (0.1–2.5) pmol/ml/min while in women patients (n = 4) values were LPL 0.7 (0.0–1.1) pmol/ml/min and HL 1.0 (0.1–1.4) pmol/ml/min. All patients had significantly lower total and lipoprotein lipase activities. Hepatic lipase activities were non-significantly reduced compared to healthy controls. The new assay showed positive bias’ for TL and LPL and a profound negative bias for HL. The assay suffered interference from bilirubin, urea and creatinine and also from added plasma. The new automated method gives more precise results compared to the previous assay technique with considerable advantages in requiring less plasma, being more reproducible, non-radioactive and considerably faster and also in not giving negative results for LPL activity calculated by difference in high versus low salt conditions. DOES NIEMANN PICK C1-LIKE-1 PROTEIN REGULATES CHYLOMICRON ASSEMBLY THROUGH ITS EFFECT ON MTP? S. Lally, Tun T. Kyaw, D. Owens, G. Tomkin Department of Endocrinology, Trinity College Dublin, UK The Niemann Pick C1-like 1 (NP C1-L1) protein regulates cholesterol absorption but intestinal expression has not been described in humans. Aim: to investigate the relationship between intestinal mRNA levels of the NP C1-L1 and microsomal triglyceride transfer protein (MTP), factors influencing chylomicron composition. D2 gastroscopy biopsies were examined. Ethics committee approval and informed consent were obtained. Subjects (21) with normal duodenal biopsies were examined. The subjects were given an 1100 kcal fat meal and blood taken at 6 h. NP C1-L1 and MTP mRNA was measured using
real-time RT-PCR. Chylomicrons (density <0.006 g/l) were isolated and apo B48 and B100 determined. Results: Fasting serum cholesterol was 4.65 ± 1.1 mmol/l and serum triglyceride median 1.18 mmol/l. There was a significant positive correlation between NP C1-L1 mRNA and chylomicron apo B48 at 6h (r = 0.64, p < 0.05) and apo B100 (r = 0.75, p < 0.01). There was a positive correlation between NP C1-L1 mRNA and chylomicron cholesterol (r = 0.79, p < 0.01). There was a significant correlation between NP C1-L1 mRNA and microsomal triglyceride transfer protein (MTP) mRNA (r = 0.84, p < 0.01) but no correlation between NP C1-L1 mRNA and ABCG5 or G8 mRNA. Conclusion: This study suggests that NP C1-L1 plays a pivitol role in regulation of MTP and thus in the assembly of large triglyceride-rich lipoproteins. Since these particles are potentially atherogenic, inhibition of NP C1-L1 may have an important role to play in the reduction atherosclerosis. HUMAN HEPATITIS c VIRUS RNA CIRCULATES PREDOMINANTLY IN VLDL-LIKE LIPO-VIRALPARTICLES Soren U. Nielsen1 , Alastair D. Burt1,2 , Caroline Martin1 , Margaret F. Bassendine1 , Geoffrey L. Toms1 1 School of Clinical Medical Sciences, University of Newcastle, Newcastle upon Tyne, UK; 2 School of Clinical and Laboratory Sciences, University of Newcastle, Newcastle upon Tyne, UK
Introduction: Hepatitis C virus (HCV) RNA circulates in the blood of infected patients in heterogeneous particles associated with lipoprotein. The majority of HCV RNA is found in lipo-viro-particles (LVP) with density below 1.10 g/ml and these are the infectious particle of HCV. We have characterized these HCV-LVPs and sought to release virion particles from association with host lipoprotein by treatment with deoxycholic acid. Methods: Serum was fractionated in isotonic iodixanol density gradients and HCV RNA content of the fractions was assessed by quantitative real-time RT-PCR. Peak fractions containing HCV were further analysed by rate-zonal centrifugation to determine their sedimentation coefficients, and thereby, their size according to Stokes law. HCV RNA containing particles in gradient fractions were immunoprecipitated with polyclonal anti-apolipoprotein antibodies bound to protein G Sepharose. Results: Serum derived HCV RNA with a density below 1.10 g/ml could be precipitated with antibodies against apoB or apoE, suggesting an association of HCV RNA with host VLDL. Treatment with deoxycholic acid released HCV RNA containing putative virion particles with density 1.12 g/ml, a sedimentation coefficient of 215 S, a diameter of 57 nm and which remained precipitable with anti-apoB but not with anti-apoE. Conclusion: Blood derived HCV RNA was found to be predominantly associated with host lipoproteins containing
Abstracts / Atherosclerosis 182 (2005) S1–S8
apoB-100 and apoE. Thus the HCV-LVP appears to be a VLDL-like particle. HCV may utilise the lipid transport function of VLDL to promote the spread of infectious particles. RESPONSE TO ATORVASTATIN IN A CASE OF IDIOPATHIC HYPOALBUMINAEMIA COMPLICATED BY REMNANT HYPERLIPIDAEMIA C.S. Arun1 , H.K. Datta1,3 , R.D.G. Neely1,2 1 Lipid
and Metabolic Clinic; 2 Department of Clinical Biochemistry, Royal Victoria Infirmary; 3 School of Clinical and Laboratory Sciences, University of Newcastle, Newcastle upon Tyne, UK Idiopathic hypoalbuminaemia is a rare inherited disorder of primary albumin deficiency, usually associated with lipodstrophy, hyperlipoproteinaemia and compensatory increases in the plasma levels of other transport proteins. Although usually considered benign, the impact of this condition on the long-term risk of cardiovascular disease and other health outcomes is less certain. We report the case of an otherwise healthy, 27-year-old man with no family history of hyperlipidaemia who presented with subcutaneous lipomata, unexplained hypoalbuminaemia and severe mixed hyperlipidaemia (total cholesterol 15.4 mmol/L triglyceride 3.2 mmol/L). Following initial treatment with atorvastatin, lipid levels improved (total cholesterol 6.8 mmol/L, triglyceride 1.8 mmol/L) but elevated serum alphafetoprotein, liver enzymes and creatine kinase were noted. Despite the absence of oedema, serum albumin, measured using an immunonephelometric method, was 9 g/L. Albumin was undetectable in the urine (<5 mg/L). Following withdrawal of atorvastatin for 6 weeks, HDL-cholesterol and apolipoprotein A1 decreased, while apolipoprotein B, total cholesterol, LDL-, VLDLcholesterol and the VLDL-cholesterol/triglyceride ratio increased and beta-VLDL appeared in plasma. However, serum enzyme levels remained elevated. Apolipoprotein E was also increased and apoE genotying revealed homozygosity for ApoE2, consistent with remnant hyperlipidaemia. Remnant hyperlipidaemia has not previously been reported in association with idiopathic hypoalbuminaemia and may confer a greatly increased cardiovascular risk in this case, requiring treatment even in the absence of conventional cardiovascular risk factors.
S7
laemia (FH) and to compare the prevalence of clinical FH (elevated LDL-C) in relatives of probands with and without a detected mutation. Methods: Patients with a Simon Broome diagnosis of definite FH were recruited by a genetic nurse from the Lipid Clinic at the Royal Free Hospital. Patients were given information on the inheritance of FH, a pedigree was drawn, and a DNA test offered to try and identify the FH mutation, and where found, DNA testing offered to relatives. Results: Thirty three probands tested to date included 17 men and 16 women, ages ranging from 25 to 77 years (mean 54.2 years, median 56 years), with 13 (39%) with tendon xanthomas. Mutations were detected in 64% (21) of probands (4 APOB, and 17 LDLR). Fifteen different mutations were identified. The detailed pedigrees of 31 of the probands identified a total of 209 1st degree relatives, (3–15 per proband) of which 162 were living. Of these, 60.8% (127) were from probands with identified mutations, and 39.2% (82) from those with no identified mutation. Conclusion: The majority of FH patients have significant numbers of family members who could be identified by cascade testing. On average this should lead to contacting five 1st degree relatives per proband, of whom 50% should also have FH. These will become new probands for further cascading. The relative prevalence of clinical FH in mutation carriers and non-carriers will be compared. RELATIVE EFFICACY OF COMMERCIALLY AVAILABLE PLANT STEROL-CONTAINING FOODS ON PLASMA CHOLESTEROL AND HDL SUBCLASSES Nicola Harman, Kate Tabbernor, Bruce A. Griffin School of Biomedical and Molecular Sciences, University of Surrey, Guildford, Surrey GU2 7XH, UK
J.M. Farrer1 , D. Nair2 , G. Norbury3 , A. Taylor3 , S.E. Humphries1
Plant sterols have been shown to lower total and LDL cholesterol by 10–14% (∼0.45 mmol/l). Their effects on HDL cholesterol are neutral, and on HDL subclasses unknown. A study was designed to examine the relative efficacy of a range of sterol-containing food products (Flora Procactiv) on lowering cholesterol and to facilitate the development of a new technique for the separation of HDL subclasses. Three groups of normal, healthy subjects consumed the recommended daily intake of either spread (S, n = 6), milk (M, n = 7) or yogurt (Y, n = 7) for 2 weeks. Blood samples were taken at the beginning and end of the intervention and after a further 2 weeks for the analysis of plasma cholesterol (TC), triglycerides (TAG), HDL cholesterol and HDL subclasses. The latter were measured by iodixanol gradient ultracentrifugation.
1 London
%
DOES DNA INFORMATION COMPLEMENT LIPID MEASURES IN CASCADE TESTING IN FAMILIAL HYPERCHOLESTEROLAEMIA PATIENTS?
IDEAS Genetics Knowledge Park; 2 Royal Free Hospital; 3 Regional Molecular Genetics Laboratory, GOSH, London, UK Aim: To evaluate the utility of DNA testing in the diagnosis of patients and their relatives with familial hypercholestero-
TC
2 weeks +4.0Y , −18.7M* , −8.6S* 4 weeks −2.7Y , +6.9M , +7.2S
LDL-C
HDL-C
−0.5Y , −22.6M* , 11.5S +1.2Y , +10.0M , +11.0S
+6.3Y , −2.5M , −1.0S −3.0Y , −1.3M , −1.3S
Iodixanol centrifugation was shown to provide a more cost effective and rapid alternative to our existing techniques
S8
Abstracts / Atherosclerosis 182 (2005) S1–S8
for the separation of HDL2 and HDL3 subclasses. Total cholesterol was significantly decreased after 2 weeks on milk and spread (* p < 0.05) and returned to baseline levels 2 weeks after the intervention. In contrast, yogurt had no effect on plasma cholesterol. There was no overall effect on plasma TAG, HDL or HDL subclasses. In conclusion, while manufacturers recommend that the daily intake of plant sterols is achieved through a combination of all three products, this study demonstrates disparity in their individual cholesterol-lowering efficacy. CARDIOVASCULAR EFFECTS OF CONJUGATED LINOLEIC ACID AND FIBRATES: EVIDENCE FOR AN ANTIPLATELET EFFECT MEDIATED BY PLATELET PPAR␣ M. Khanolkar2 , A.W. Thomas1 , A. Roberts2 , L. George2 , J. Geen3 , A. Rees2 , R.H.K. Morris1 , L.M. Evans2 1 School
of Applied Sciences, UWIC, Cardiff; 2 Department of Medicine, Cardiff University; 3 Clinical Biochemistry Department, Prince Charles Hospital, Merthyr, UK Increased platelet activation is a recognised feature of insulin resistance (IR) and type 2 diabetes (T2D) and is an important factor in atherothrombosis. Recent studies have shown presence of functional PPAR␥ in human platelets capable of inhibiting platelet function. Conjugated linoleic acid (CLA), a ligand for both PPAR␣ and ␥ has been shown to inhibit platelet aggregation. We therefore hypothesised the presence of PPAR␣ in human platelets. Flow cytometry followed by western blotting confirmed the presence of PPAR␣ in human platelets, while Electrphoretic mobility shift assay (EMSA) confirmed the DNA binding ability of platelet PPAR␣. We also conducted in vitro studies on platelet rich plasma to study direct effects of Clofibrate (PPAR␣ agonist) and CLA at physiological concentrations on ADP induced platelet aggregation. A significant reduction platelet aggregation was observed with both compounds (Clofibrate:16.23 ± 4.26 ⇒ 6.14 ± 2.87 , p < 0.01; CLA:15.64 ± 3.97 ⇒ 3.11 ± 1.76 , p < 0.01). Pre-treating platelets with a PPAR␣-antagonist (MK-886) reduced the anti-platelet aggregatory effects of clofibrate and CLA, while pre-treatment with PPAR-␥ antagonist (GW-9662) did not alter the effects of CLA on platelet aggregation. This is the first study to demonstrate the presence of functionally active PPAR␣ with preserved DNA binding ability in human platelets and that PPAR␣ activation in platelets inhibits platelet activation. This study suggests that the potential cardiovascular benefits of both PPAR␣ and PPAR ␥ agonists may be partly mediated through modulating platelet function.
PULSE WAVE INTENSITY AND THE POTENTIAL IDENTIFICATION OF SUBCLINICAL CARDIOVASCULAR DISEASE IN TREATMENT NA¨ıVE TYPE 2 DIABETIC PATIENTS Aled Wyn Roberts1,2 , Frank Rakebrandt2 , Janice Swampillai2 , Laila Jones1 , Marc Evans1 , Alan Rees1 , Alan Fraser2 1 Department
of Diabetes and Endocrinology, University Hospital of Wales, Cardiff; 2 Wales Heart Research Institute, University Hospital of Wales, Cardiff, UK Background: Type 2 diabetes (T2D) is associated with increased risk of cardiovascular disease (CVD), which may be present prior to diagnosis of T2D. Diabetic cardiomyopathy in patients without coronary disease may be related to pathological changes in ventriculo-arterial coupling, with augmented central afterload caused by increased wave reflections from stiff conduit arteries. Pulse wave intensity has been developed as a non-invasive method of assessing the interaction of left ventricular contractility and relaxation with waves reflected from the periphery. We measured wave intensity in the common carotid artery (CCA) using prototype hardware (Aloka SSD-5500, 7.5 MHz lineararray probe) and analysed data using bespoke software (Matlab). Methods: We studied 18 subjects with T2D on diet control with no history of CHD. The right CCA was imaged, and arterial diameter measured by automated wall tracking was calibrated for pressure. Velocity was derived from Doppler colour flow. Wave intensity is the product of instantaneous rate of change of velocity and pressure. Local wave speed was calculated during early systole from a pressure-velocity loop, and separated forward and backwards travelling waves derived. Data from T2D subjects were compared with agematched controls, and statistical analyses performed using unpaired t-tests. Results: The ages of the T2D and control groups were comparable (means 67 versus 63 years). Systolic BP and MAP were higher in the T2D group. Epsilon and Beta, both measures of arterial stiffness, were significantly elevated in the T2D group (p = 0.02 and p < 0.01, respectively). The peak of the backward travelling compression wave, which augments central systolic pressure, was significantly higher in T2D subjects (p = 0.05). Conclusion: Analysis of our study group reveals reduced elasticity of the vasculature, in patients in the earliest stages of T2D. Non-invasive assessment of pulse wave intensity may provide evidence of subclinical vascular disease in high-risk patients, allowing the targeting of cardiovascular risk reduction therapy to those likely to most benefit.