UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Assessment of the appropriate treatment of coronary artery disease and dyslipidaemia in an angiographically… Francis, Michelle Catherine 2000

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2000-0639.pdf [ 6.88MB ]
Metadata
JSON: 831-1.0089732.json
JSON-LD: 831-1.0089732-ld.json
RDF/XML (Pretty): 831-1.0089732-rdf.xml
RDF/JSON: 831-1.0089732-rdf.json
Turtle: 831-1.0089732-turtle.txt
N-Triples: 831-1.0089732-rdf-ntriples.txt
Original Record: 831-1.0089732-source.json
Full Text
831-1.0089732-fulltext.txt
Citation
831-1.0089732.ris

Full Text

A S S E S S M E N T OF T H E A P P R O P R I A T E T R E A T M E N T OF C O R O N A R Y A R T E R Y DISEASE AND D Y S L I P I D A E M I A IN A N A N G I O G R A P H I C A L L Y DEFINED PATIENT P O P U L A T I O N by M I C H E L L E C A T H E R I N E F R A N C I S B.Sc.(Kin), Simon Fraser University, 1991 A THESIS S U B M I T T E D I N P A R T I A L F U L F I L M E N T OF . T H E R E Q U I R E M E N T F O R T H E D E G R E E OF M A S T E R OF S C I E N C E in T H E F A C U L T Y OF G R A D U A T E S T U D I E S (Department of Pathology and Laboratory Medicine) We accept this thesis as conforming to the required standard T H E U N I V E R S I T Y OF B R I T I S H C O L U M B I A August 2000 ©Michelle Catherine Francis, 2000 UBC Special Collections - Thesis Authorisation Form http://www.library.ubc.ca/spcoll/thesauth.html In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the Un i v e r s i t y of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y a v a i l a b l e for reference and study. I further agree that permission for extensive copying of th i s thesis for sch o l a r l y purposes may be granted by the head of my department or by his or her representatives. It i s understood that copying or pu b l i c a t i o n of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department The U n i v e r s i t y of B r i t i s h Columbia Vancouver, Canada 1 of 1 10/11/00 12:46 A M ABSTRACT Cardiovascular disease is the leading cause of death in British Columbia and Canada, with the majority of cases being the result of coronary artery disease ( C A D ) . C A D is a result of the interaction between genetic background and environmental influence. Several modifiable and non-modifiable risk factors contribute to the development and progression of atherosclerosis - the major cause of C A D . Numerous studies have shown that treatment of C A D risk factors, and dyslipidaemia in particular, reduces the incidence of C A D morbidity and mortality. Based on this evidence, national guidelines have been published for the treatment of dyslipidaemia and management of other risk factors of C A D . However, examination of treatment practices in Canada, the U S , and Europe has shown sub-optimal adherence to these guidelines. This thesis describes a questionnaire-based follow-up of consecutive patients referred for selective coronary angiography (SCA) between 1993 and 1995 at two tertiary Vancouver hospitals. L i p i d measurements, risk factor prevalence, and medication use at the time of S C A were available for all patients. The follow-up assessed changes in risk factors and medications, incidence of C A D morbidity, access to risk factor counseling, patient awareness of risk factors, and, in the case of deceased patients, cause of death. The 1988 Canadian Consensus Conference on Cholesterol ( C C C C ) and 1993 National Cholesterol Education Program (NCEP) guidelines were used to determine the appropriate treatment of dyslipidaemia in those who responded to the questionnaire. Respondents were more likely to have C A D , less likely to smoke, and more likely to live in rural areas than non-respondents. The prevalence of most risk factors was comparable to previous reports, however respondents were more obese and had a greater number of risk factors than the Canadian population. Between 1993 and 1997, 150 patients died, with cause of death available for 102. O f these, 70% died of cardiovascular causes. i i Awareness of risk factors was less than ideal among respondents. Only half were able to correctly judge their change in weight and exercise. Just over half knew their blood pressure, while only two-fifths could report their cholesterol level. Counseling for lifestyle risk factors was under-utilised in general, although appeared well targeted to patients requiring intervention. While the number of patients who would have required cholesterol monitoring, dietary therapy, or drug intervention differed between guidelines, the prevalence of appropriate dyslipidaemia treatment was equivalent regardless of the guideline employed. Three-quarters of respondents requiring cholesterol monitoring had their lipids checked during the follow-up period. Dietary therapy was appropriately administered in approximately 70% of patients requiring treatment. However, counseling by dieticians, which is explicitly required by C C C C guidelines, was seen in only 38%. L ip id lowering drugs ( L L D ) were not prescribed for over 40% of respondents requiring drug therapy. While the number of patients being treated appropriately for their dyslipidaemia was found to be slightly higher in the study population than in other studies published in the literature, the level of treatment was still sub-optimal, particularly with respect to utilisation of allied health professionals. With cardiovascular disease being the leading cause of death in British Columbia, more effort must be made to follow established guidelines using well-proven methods of treatment. T A B L E O F C O N T E N T S A B S T R A C T i i T A B L E O F C O N T E N T S iv A B B R E V I A T I O N S v i i i L I S T O F T A B L E S x L I S T OF F I G U R E S x i i A C K N O W L E D G E M E N T S x i i i 1 I N T R O D U C T I O N 1 1.1 C O R O N A R Y A R T E R Y D I S E A S E 1 1.1.1 Etiology 1 1.1.2 Incidence and Costs 4 1.1.3 Risk Factors 5 1.1.3.1 Major Risk Factors 6 1.1.3.2 Conditional Risk Factors 8 1.1.3.3 Predisposing Risk Factors 11 1.1.4 Treatment 12 1.1.4.1 Lifestyle Interventions 12 1.1.4.2 Drug Therapies for C A D Event Reduction 13 1.1.4.3 L i p i d Lowering 15 1.2 P U B L I S H E D G U I D E L I N E S F O R T R E A T M E N T O F D Y S L I P I D A E M I A 17 1.2.1 Canadian Consensus Conference on Cholesterol Guidelines 17 1.2.1.1 L ip id Risk Factors and Testing Priorities 17 1.2.1.2 Dietary Therapy 18 1.2.1.3 Drug Therapy 19 1.2.1.4 Other Risk Factors 19 1.2.1.5 Canadian Guidelines 2000 19 1.2.2 National Cholesterol Education Program Guidelines 22 1.3 P A T I E N T A W A R E N E S S OF H E A L T H 25 1.4 A P P R O P R I A T E T R E A T M E N T O F D Y S L I P I D A E M I A 27 1.5 R A T I O N A L E 30 1.6 H Y P O T H E S I S 31 1.7 SPECIFIC A I M S 31 iv 2 M E T H O D S 32 2.1 O R I G I N A L C O H O R T S E L E C T I O N 32 2.2 Q U E S T I O N N A I R E D E S I G N 33 2.2.1 Ethics Approval 34 2.3 M A I L - O U T 34 2.4 V I T A L STATISTICS I N F O R M A T I O N 35 2.5 M E D I C A L R E C O R D S 36 2.6 S T A T I S T I C A L A N A L Y S I S 36 3 R E S U L T S 37 3.1 R E S P O N S E R A T E 37 3.2 B A S E L I N E C H A R A C T E R I S T I C S 37 3.2.1 Demographics 37 3.2.2 L i p i d Profile 38 3.2.3 Risk Factor Prevalence 39 3.2.3.1 Subgroup analyses 40 3.2.4 Medication Usage 44 3.3 F O L L O W - U P C H A R A C T E R I S T I C S 45 3.3.1 Changes in Risk Factor Prevalence 45 3.3.2 Changes in Medication Usage 46 3.3.3 Morbidity 47 3.3.4 Mortality : 47 3.4 P R E D I C T O R S OF M O R B I D I T Y A N D M O R T A L I T Y 48 3.4.1 Morbidity 48 3.4.2 Mortality 48 3.5 P A T I E N T A W A R E N E S S OF H E A L T H 50 3.5.1 Changes in Health Measures 50 3.5.1.1 Weight 50 3.5.1.2 Exercise habit 52 3.5.1.3 Alcohol intake 52 3.5.2 Risk Markers 52 3.5.2.1 Hypertension 52 3.5.2.2 Cholesterol 53 3.5.3 Instillation of Awareness by Others 53 3.5.3.1 Risk factor counseling 53 3.5.3.2 Health professionals and laypersons 54 3.5.4 Comparison with Hospital Records 54 3.6 A P P R O P R I A T E T R E A T M E N T 55 3.6.1 C C C C Guidelines 55 3.6.1.1 L i p i d Testing 55 3.6.1.2 Dietary Therapy 55 3.6.1.3 Drug Therapy 56 3.6.1.4 Canadian Guidelines 2000 57 3.6.2 N C E P Guidelines 57 3.6.2.1 Annual cholesterol measurement 57 3.6.2.2 Dietary therapy 58 3.6.2.3 Drug therapy 58 3.7 P R E D I C T O R S OF A P P R O P R I A T E T R E A T M E N T 59 3.7.1 C C C C Guidelines 59 3.7.2 N C E P Guidelines 60 3.8 R E G I O N A L V A R I A T I O N S I N A P P R O P R I A T E T R E A T M E N T 60 4 DISCUSSION 62 4.1 L I M I T A T I O N S OF T H E S T U D Y 62 4.1.1 Methodology 62 4.1.2 Response Rate and Bias 63 4.2 P A T I E N T C H A R A C T E R I S T I C S 64 4.2.1 Demographics and Risk Factor Prevalence 64 4.2.2 Baseline Subgroup Analyses 68 4.2.2.1 A p o A I in L o w Risk Patients 68 4.2.2.2 Predictors of C A D in Younger versus Older Women 69 4.2.3 Morbidity 71 4.2.4 Mortality 72 4.3 A N A L Y S I S O F T R E A T M E N T 74 4.3.1 Awareness and Counseling of Risk Factors 74 4.3.2 Pharmacological Interventions 76 v i 4.3.3 Appropriate Treatment of Dyslipidaemia 78 4.3.3.1 Cholesterol Screening 79 4.3.3.2 Dietary Therapy 80 4.3.3.3 Drug Therapy 84 4.4 I M P A C T O N C L I N I C A L P R A C T I C E S 87 5 REFERENCES 91 b APPENDIX A 104" v i i ABBREVIATIONS 4S Scandinavian Simvastatin Survival Study A C E angiotensin converting enzyme A F C A P S / T e x C A P S A i r Force/Texas Coronary Atherosclerosis Prevention Study A H A American Heart Association apo apolipoprotein A S A acetylsalicylic acid B M I body mass index Ca-channel blocker calcium channel blocker C A B G coronary artery bypass graft C A D coronary artery disease C A R E Cholesterol and Recurrent Events C B S cystathionine (3-synthase C C C C Canadian Consensus Conference on Cholesterol C H F congestive heart failure C Q I N Clinical Quality Improvement Network C R P C-reactive protein C V D cardiovascular disease D B P diastolic blood pressure D V A - H I T Department of Veterans Affairs H D L Intervention Trial E D T A ethylenediaminetetraacetic acid F A T S Familial Atherosclerosis Treatment Study F E R H D L fractional esterification rate of cholesterol in H D L F P G fasting plasma glucose Hey homocyste ine v i i i H D L high density lipoprotein H D L - C high density lipoprotein cholesterol H E R S Heart and Estrogen/progestin Replacement Study H M G - C o A 3-hydroxy-3-methylglutaryl coenzyme A H R T hormone replacement therapy I C D International Classification of Diseases IL-1 interleukin-1 L D L low density lipoprotein L D L - C low density lipoprotein cholesterol LIPID Long-term Intervention with Pravastatin in Ischemic Disease L L D lipid lowering drugs Lp(a) lipoprotein(a) M I myocardial infarction M T H F R methylenetetrahydrofolate reductase N C E P National Cholesterol Education Program O x L D L oxidised L D L P D G F platelet-derived growth factor P T C A percutaneous transluminal coronary angioplasty SBP systolic blood pressure S C A selective coronary angiography S M C smooth muscle cell T C total cholesterol T G triglycerides V L D L very low density lipoprotein W O S C O P S West of Scotland Coronary Prevention Study ix LIST OF TABLES Table 1. A H A classification of major, conditional, and predisposing risk factors 5 Table 2. Risk assessment algorithm of the Canadian 2000 Guidelines 21 Table 3. Canadian 2000 Guidelines risk assignment for treatment decisions 22 Table 4. N C E P Guideline L D L - C thresholds for dietary and drug therapy 25 Table 5. Mean lipid levels for respondent and deceased patient groups, separated by L L D use at baseline 38 Table 6. Prevalence of baseline risk factors for non-respondent, respondent, and deceased patient groups 40 Table 7. Univariate analysis of lifestyle risk factors for C A D - and C A D + groups 41 Table 8. Univariate analysis of l ipid risk factors for C A D - and C A D + groups 42 Table 9. Multivariate logistic regression of C A D predictors 42 Table 10. Univariate analysis of C A D risk factors in women < 60 years 43 Table 11. Univariate analysis of C A D risk factors in women > 60 years 44 Table 12. Multivariate analysis of C A D predictors in women < 60 and > 60 44 Table 13. Proportion of non-respondents, respondents, and deceased patients taking medications at baseline 45 Table 14. Self-reported medication usage of respondents at follow-up, with comparison to baseline prevalence 46 Table 15. Self-reported revascularization procedures and M I of respondents at follow-up 47 Table 16. Variables predictive of all-cause mortality based on multivariate logistic regression 49 Table 17. Predictors of C V mortality based on multivariate logistic regression 49 Table 18. Pearson's x 2 associations between risk factor counseling and reported change in risk factor behaviour 54 Table 19. Proportion of respondents receiving appropriate dietary therapy with comparison between patients with high T C versus those with borderline high T C 55 Table 20. Proportion of respondents receiving appropriate drug therapy with comparison between patients with high T C versus those with borderline high T C 56 Table 21. Proportion of respondents receiving appropriate treatment according to C C C C guidelines 56 Table 22. Proportion of respondents receiving appropriate treatment according to Canadian 2000 guidelines 57 Table 23. Proportion of respondents receiving appropriate treatment according to N C E P guidelines 59 Table 24. Variables predictive of appropriate treatment according to C C C C Guidelines 59 Table 25. Variables predictive of appropriate treatment according to N C E P Guidelines 60 Table 26. Variables showing regional variation on univariate analysis 61 Table 27. Regional crossover effect of dietician counseling on self-reported low fat diet intake 61 Table 28. Comparison of the prevalence of l ipid risk factors found in C A D + male respondents with the D V A - H I T cohort 67 Table 29. Comparison of type of anti-hypertensive reported at follow-up with baseline use 77 x i LIST OF FIGURES Figure 1. Response to injury hypothesis of atherogenesis 3 Figure 2. Leading causes of death in British Columbia in 1998 4 Figure 3. N C E P Guideline treatment algorithm for primary prevention 24 Figure 4. N C E P Guideline treatment algorithm for secondary prevention 25 Figure 5. Post-MI discharge medications in Canada in 1998 27 Figure 6. Causes of death for 102 deceased cohort patients 48 Figure 7. Comparison of patient-reported weight change with actual weight change 51 xn A C K N O W L E D G E M E N T S I would like to express my gratitude to the people who contributed to the completion of this research. Foremost, I would like to thank my supervisor, Dr. Jiri Frohlich, for his invaluable guidance and mentorship throughout my academic and research studies, along with his encouragement of my future endeavours. I would also like to acknowledge the assistance of my supervisory committee members, in particular Dr. John H i l l for his input on manuscripts, Dr. John Spinelli for his statistical expertise and instruction, and Dr. Andy Ignaszewski for the opportunity to inform current clinical practices. In addition, I would like to extend my gratitude to my colleagues in the Atherosclerosis Specialty Lab and Healthy Heart Program for their helpful discussions and constructive feedback on my research directions. Special thanks are due to my husband, Andrew, who has supported me both emotionally and financially throughout my pursuits in research and medicine. Finally, I would like to acknowledge the funding support provided by the Medical Research Council o f Canada and Bristol-Myers Squibb for this study. x i i i 1 INTRODUCTION 1.1 CORONARY ARTERY DISEASE 1.1.1 Etiology The major pathological process underlying coronary artery disease ( C A D ) is atherosclerosis. The pathogenesis of the atherosclerotic process leading to C A D is a complex interplay of both genetic factors, which define the limit under which atherosclerosis develops, and environmental influences, which impact a person's risk within the l imi t 1 . These factors set in motion a process described by Russell Ross 2 as the "response-to-injury hypothesis". According to this theory, illustrated in Figure (1), the steps in atherosclerotic development begin with a mechanical or functional injury to the vascular endothelium, which leads to increased permeability to blood cells, lipoproteins, and certain hormones. Platelet aggregation occurs, with subsequent release of growth factors, such as platelet-derived growth factor (PDGF) , and chemoattractants. This stimulates smooth muscle cell ( S M C ) proliferation and their migration, along with macrophages, into the subintima region where the fatty streaks and plaques of atherosclerosis develop 3 . The SMC' s form a fibrous cap over a l ipid core. This core contains macrophages that have taken up oxidised low density lipoprotein ( L D L ) through an unregulated scavenger-receptor mediated process, at which point they are called foam cells 1 . Rupture of the fibrous cap may occur, leading to thrombosis and partial or complete occlusion of the lumen. The clinical manifestation of this process may be unstable angina or non-fatal or fatal myocardial infarction (MI). The presence of macrophages, together with T-lymphocytes, in the atherosclerotic plaque suggests an inflammatory reaction underlying the atherogenic process 2. Indeed, pathological 4 as well as c l in ica l 5 evidence has demonstrated a link between inflammation and C A D morbidity and mortality. The concentration of inflammatory markers such as C-reactive protein (CRP) 1 have been shown to increase in active coronary syndromes such as unstable angina, thus leading to the conclusion that inflammation plays a central role in plaque development and progression 4. In addition to plaque formation, vascular tone also modulates the pathology of atherosclerosis. In particular, endothelium-dependent vasodilation, a normal response to agonists such as acetylcholine, has been shown to be paradoxically altered to vasoconstriction in atherosclerotic coronary arteries6. This endothelial dysfunction, which may be measured using brachial ultrasound, is thought to result from a reduction in the bioavailability of the vasodilator nitric oxide 7 . 2 f a W c D E 3 C c o 03 O ' c S3 O CJ - 8 a © M g< ^ fe cn 8 © s * 1 © o « ^ 8 1 8 B 3 S U M _ » J3 < a o E c _ H 3 u o <u S o __ 05 « f i O 3 05 § i c o 2 CN CJ o CJ — O CJ —! c 03 S3 60 CJ 60 60 14 E/3 1.1.2 Incidence and Costs Coronary heart disease made up the greatest portion of all cardiovascular diseases ( C V D ) at 71%, and was the second leading cause o f death in British Columbia in 1998, accounting for 25% of all deaths. Overall, diseases of the circulatory system were responsible for the greatest proportion of deaths in British Columbia in 1998 - 36% in total 8 (Figure 2). Figure 2. Leading causes of death in British Columbia in 1998.8 The financial burden of C A D on society is felt through diagnostic procedures, medical interventions, lost work hours, and premature death. In 1993, C V D accounted for 15.3% o f the total cost of all illnesses in Canada, or $19.7 billion. C A D accounted for $7.8 bill ion. Indirect costs such as mortality, short, and long-term disability made up two thirds of the economic impact of C A D . Mortality alone was responsible for $4.6 bill ion (59% of total cost o f C A D ) . O f the direct costs, including hospital care, drugs, physician care expenditures, and research, hospital care made up 75.7%) of the total, or $1.6 b i l l ion 8 . 4 1.1.3 Risk Factors C A D is a multi-factorial disease linked to several modifiable and non-modifiable "risk factors" - a term coined by the long-running Framingham Heart Study 9 . These risk factors may be genetic, metabolic, or psychosocial in nature, and can be divided into three categories. Those risk factors that have a demonstrated independent relationship with C A D are "major" risk factors. Risk factors which have been shown to be associated with C A D , but the causal relationship has not been determined are known as "conditional". A n d finally, characteristics that worsen the impact of the major risk factors have been called "predisposing" risk factors 1 0 . Table (1) outlines the risk factors in each category. Table 1. A H A classification of major, conditional, and predisposing risk factors.1 0 Major Risk Factors Cigarette smoking Hypertension Obesity Physical inactivity Elevated serum total cholesterol and LDL-cholesterol L o w serum HDL-cholesterol Diabetes mellitus Advancing age Conditional Risk Factors Elevated serum triglycerides Small L D L particles Elevated serum homocysteine Elevated serum lipoprotein(a) Prothrombotic factors (eg. fibrinogen) Inflammatory markers (eg. C-reactive protein) Predisposing Risk Factors Abdominal obesity Family history of premature C A D Ethnic characteristics Psychosocial factors (eg. depression) 5 The Canada Heart Health surveys, completed between 1985 and 1990 found that 33% of women and 41% of men between the ages of 18 and 74 had two or more major risk factors for C A D 1 1 . 1.1.3.1 Major Risk Factors The important role of several key risk factors in the development and progression of C A D has been supported by numerous studies. Smoking tobacco has been linked to profound alterations in vasomotion and platelet adhesion. It has been shown to substantially increase fibrinogen and carboxyhaemoglobin levels, while decreasing high density lipoprotein-cholesterol ( H D L - C ) 1 2 . In Canada, it is responsible for more deaths due to cardiovascular disease than cancer. In 1996/97, 29% of individuals over the age of 15 smoked cigarettes, although the proportion of daily smokers decreased significantly beyond the age of 55 8 . The Framingham Study 9 brought to light the significant impact of hypertension on C A D incidence rates. Individuals with isolated systolic hypertension were more than twice as likely to suffer acute M I than those with a blood pressure <160 mmHg/<95 mmHg. High blood pressure increases overall cardiovascular risk by 2 to 3 t imes 1 3 . In Canada, 22% of individuals (26% of men and 18%> of women) had high blood pressure in 1985-90, and diagnosis of hypertension increases with advancing age 8. Hypertension has been described as the most useful single factor in characterising C A D r i s k 1 4 . Diabetes not only increases the risk of developing C A D , but also adversely affects the outcome, with mortality rates significantly higher for individuals with diabetes8. M e n with diabetes were 56% more likely to manifest signs of C A D within a 24 year follow-up period than men without diabetes, while women with diabetes were over twice as likely to develop C A D over the same time period than disease-free females. 9. In addition, Haffner et a l 1 5 showed that patients with diabetes and no previous evidence of C A D were at an equally high risk for C A D 6 morbidity and mortality as patients with previous M I , and thus required similarly aggressive treatment. Obesity is yet another modifiable lifestyle risk factor for C A D . The relationship between obesity and C A D risk is most likely mediated through associations with increased blood pressure, hypertriglyceridaemia, impaired glucose tolerance, and lower levels of H D L - C 1 6 . Obesity often results from a combination of high saturated fat intake and lack of physical activity. Physical inactivity is detrimental not only through its role in promoting obesity, but also as an independent risk factor for C A D . Habitual sedentarianism has been linked to an increased risk of M I 1 7 . Physical inactivity has been suggested to be as strong a risk factor for atherosclerosis as cigarette smoking, and has been linked to an increase in all-cause mortali ty 1 8 . A large number of both case-control 1 9 ' 2 0 and prospective epidemiological studies 2 1 have shown that plasma lipids and lipoproteins play an important role in C A D development. In particular, elevated levels of total cholesterol (TC) and LDL-cholesterol ( L D L - C ) , along with low levels of H D L - C represent major risk factors for the development of atherosclerosis 2 2, and thus form the basis of clinical guidelines aimed at reducing disease incidence and preventing cardiovascular morbidity and mortali ty 2 3 . According to a Canadian health survey between 1985 and 1990, 45% of men and 43% of women had a T C greater than the desirable level of 5.2 mmol/L, with 18% and 17%, respectively, in the highest risk group with a T C > 6.2 mmol /L 8 . Elevated T C has been correlated with C A D morbidity and mortality, but not all patients who suffer from C A D events have high T C levels. The Framingham Study provided evidence to support the importance of employing the ratio of T C to H D L - C as a better marker of r i s k 2 4 . The significance of depressed H D L - C levels even at moderate T C concentrations has important 7 implications for C A D prevention. A T C / H D L - C ratio greater than 5.0 for women and 5.5 for men was recommended as a prompt for l ipid lowering therapy. Non-modifiable risk factors for C A D include age and gender. A s all major forms of cardiovascular disease increase with advancing age, it is the dominant risk factor for heart disease 8. Males have a higher risk of developing C A D , particularly at a younger age, although the incidence of C A D becomes equal between men and women in the sixth and seventh decade 2 5 . 1.1.3.2 Conditional Risk Factors The role of triglyceride (TG) levels in the development of atherosclerosis has been controversial for a number of years. While several studies have shown an association between C A D and T G , adjustment for H D L - C tends to attenuate the relationship 2 6 . However, Stampfer et a l 2 7 showed T G levels to be independently predictive of M I in the Physicians' Health Study. The National Institutes of Health released a consensus statement acknowledging the variability of data and the lack of a causal link between elevated T G and disease 2 8 . T G levels have been shown to carry more weight in certain populations of patients, particularly w o m e n 2 6 and individuals with a high L D L - C / H D L - C ra t io 2 9 . The fractional esterification rate of cholesterol in H D L , or F E R H D L , has been shown to reflect both L D L 3 0 and H D L 3 1 particle size. It is a measure of the ratio of cholesteryl esters produced by endogenous lecithin cholesterol acyltransferase to the amount of cholesterol remaining after incubation of radiolabeled plasma. The greater the F E R H D L , the higher the preponderance of small H D L and small, dense L D L particles - both conditional risk factors for C A D . Homocysteine (Hey) is an intermediate in the metabolic processing of methionine. This pathway requires the enzymes methylenetetrahydrofolate reductase ( M T H F R ) and cystathionine 8 (3-synthase (CBS) along with cofactors folate, vitamin B6, and vitamin B12. Mutations in the M T H F R or C B S genes have been shown to lead to increased plasma levels of Hey, as have nutritional deficiencies in folate or the B vi tamins 3 2 . Hey levels over the norm of 10 umol/L have been associated with C A D , peripheral vascular disease, and cerebrovascular disease 3 3 . Proposed mechanisms by which elevated plasma Hey levels may lead to premature atherosclerosis are a) endothelial injury, particularly through the formation of reactive oxygen species, b) association with coagulation factors, and c) interaction with L D L particles leading to internalisation of L D L by macrophages and the oxidation of lipids by subsequently released H e y 3 2 . Lipoprotein(a) (Lp(a)) particles consist of an L D L particle with a disulfide linkage between apolipoprotein B-100 (apoB) and apo(a) 3 3 . Apo(a) has been demonstrated to have sequence homology to plasminogen, a precursor of the anticoagulant plasmin. A s such, the association between elevated Lp(a) concentration and C A D may be related to competitive inhibition of plasminogen binding at the site of intravascular injury, and thus the loss of beneficial anticoagulation. Conversely, apo(a) binds to fibrin, which is found at the site of injury. This attraction of Lp(a) may serve to bring L D L cholesterol to the injured site, which is then taken up by macrophages to form foam ce l l s 3 4 . Lp(a) levels are genetically determined, and may vary between individuals over 1000-fold 3 5 . However, levels greater than 0.20g/L have been associated with significantly increased risk of C A D 3 4 . There is no treatment known to reduce Lp(a) levels, and thus it has been suggested that reduction of other C A D risk factors take precedence in patients with elevated Lp(a) concentration 1 0. Fibrinogen influences platelet aggregation and is a substrate for fibrin formation 3 6 . A significant correlation between increased fibrinogen and C A D morbidity and mortality has been found in several studies 3 3 . Plasma viscosity is increased by fibrinogen, and this has been 9 associated with a negative prognosis from acute M I and unstable angina 3 6 . While no specific method to reduce fibrinogen has been discovered, it is known that smoking raises fibrinogen levels, but this can be reversed following smoking cessation 1 0 . The link between inflammation and atherosclerosis is believed to be through the macrophage-derived inflammatory mediator interleukin-1 (IL-1). The presence of IL-1 at the site of endothelial injury may generate procoagulation activity, thus transforming the typically nonthrombotic endothelium into a thrombogenic surface. C-reactive protein is an acute phase reactant induced by IL-1 , and has been used as a marker of inflammation 4. Elevations of C R P have been found in patients with unstable angina, and like fibrinogen, are believed to indicate a poor prognosis 3 6 . While not specifically stated in the American Heart Association ( A H A ) classification of conditional risk factors, apolipoproteins have also been shown to be independently associated with C A D , although the causative relationship is unclear. The combination of l ipid levels and lifestyle risk factors has only been able to explain half of the C A D incidence 3 7 . A s a result, researchers have examined the role of apolipoproteins - the protein moieties of lipoproteins - in C A D . A p o B , found on the surface of very low density lipoprotein ( V L D L ) and L D L particles, has been elevated in both men and women with C A D 3 8 ' 3 7 . Also , reduction of apo B through lipid lowering medications was independently correlated with angiographic regression in the Familial Atherosclerosis Treatment Study ( F A T S ) 3 9 . Small, dense L D L particles (pattern B) have been shown to be more atherogenic than large, buoyant L D L because they are more readily taken up by macrophages following oxidat ion 2 4 . Because apo B concentration is believed to be a reflection of the L D L particle number, and the ratio of L D L - a p o B / C has been shown to increase with the presence of the pattern B phenotype 4 0 , determination of apoB level may represent a more accurate measurement of atherogenicity than L D L cholesterol content. 10 Reduced plasma concentrations of apolipoprotein A I (apo AI) , the principal protein component of H D L molecules, have also been associated with C A D 4 1 ' 4 2 . Whether the predictive power of apo A I level on disease is greater than traditional l ipid measurements remains controversial. However, studies have shown that H D L particles containing only apo A I on the surface (LpAI) are selectively reduced in normolipaemic patients with C A D 4 3 . In addition, a preponderance of smaller H D L 3 particles, with fewer associated apo A I molecules, has been linked to C A D presence 4 4. 1.1.3.3 Predisposing Risk Factors Predisposing lifestyle risk factors that contribute to the incidence of C A D include consuming a high saturated fat diet, abdominal obesity, and excessive alcohol intake. Animal studies of dietary effects on cholesterol levels were an early indicator that consuming saturated fats and cholesterol lead to elevated levels of T C and L D L - C 4 5 . Epidemiological studies in humans have consistently demonstrated that C A D mortality is directly correlated with saturated fat in take 4 6 and that reducing dietary consumption of saturated fats leads to a decrease in blood cholesterol leve ls 1 6 . A diet high in saturated fats also contributes to the insulin resistance syndrome, a predisposing factor to both Type II diabetes and C A D 4 7 . The combination of a high fat diet and sedentarianism has lead to an increasing rate of obesity in North America. These metabolic effects are seen primarily with a pattern of fat distribution known as android obesity, which is characterised by central adiposity 4 8 . Abdominal obesity, defined as a waist circumference in men > 102 cm or > 88 cm in women, has been linked to disturbances in lipoprotein metabolism and to insulin resistance, and heightens the risk of C A D in obese individuals 4 9 . The association between alcohol intake and C A D is complex. Some studies have shown that moderate alcohol intake, on the order of one to two drinks per day, reduces C A D morbidity 11 and mor ta l i ty 5 0 ' 5 1 . Alcohol has been shown to raise H D L - C levels and lower L D L - C 5 2 by increasing the synthesis and/or the secretion of apolipoproteins, lipids, and H D L molecules 5 3 . However, it appears that excessive consumption of alcohol increases risk of C A D and all-cause mortal i ty 1 6 . Non-modifiable risk factors also play a role in predisposing individuals to atherogenesis. Having a family history of premature heart disease in a first-degree relative (parent, sibling, aunt/uncle) confers increased risk for developing C A D 5 4 . This association may be through either genetic or environmental effects, as lifestyle influences such as diet and level of physical activity may be imposed on all household members 8. 1.1.4 Treatment M u c h of the medical research over the past decades has focussed on identification and treatment of C A D risk factors. Many benefits have been gained from these efforts, with a death rate in 1997 from cardiovascular disease that was half the rate in 1969 8. It has been suggested that the decline in C A D mortality may be attributed to lifestyle changes including low fat diets, exercise, and smoking cessation, as well as improved medical and surgical care for pre-existing disease. 1.1.4.1 Lifestyle Interventions Meta-analysis of several research trials has demonstrated that exercise rehabilitation programs reduce C A D events, cardiac mortality, and overall mortal i ty 5 5 . Regular exercise lowers body weight and blood pressure, and can improve serum lipids and diabetes8. Equally important, studies have shown that exercise counseling improves perceived quality of life, reduces depression, and leads to increased psychosocial adjustment 5 6. Smoking cessation has been shown to reduce cardiovascular events in both primary and secondary prevention 1 2 . While mechanisms of this positive effect have not been studied, it has 12 been proposed that the cessation of smoking returns previously perturbed proatherogenic factors to normal levels. Several studies have shown that controlling hypertension with angiotensin converting enzyme ( A C E ) inhibitors, beta-blockers, or d iure t ics 5 7 ' 5 8 significantly reduces C A D mortality. Hypertension management also has an important role to play in the reduction of overall mortality and s t roke 5 9 ' 6 0 . Treatment of hypertension has also been demonstrated to be of particular benefit to reducing C A D risk in patients with diabetes 5 7 ' 5 8 . While no trials have shown definitive evidence that the appropriate control of blood sugar in patients with diabetes reduces C A D events, l ipid lowering in these patients is important in preventing C A D complications 1 2 . 1.1.4.2 Drug Therapies for C A D Event Reduction A variety of non-lipid lowering medications have been linked to beneficial outcomes in primary and secondary prevention of C A D morbidity and mortality. Meta-analysis of several aspirin ( A S A ) trials demonstrated an overall reduction in stroke and M I of 30%, and a decrease of 24%o in new vascular events with the administration of 300-325mg of A S A per day 6 1 . The Physician's Health S tudy 6 2 found a 44%> reduction in M I with a low dose of 325mg A S A every other day. There are several proposed mechanisms for the beneficial role of A S A in primary and secondary prevention of C A D . Most significant is believed to be the antiplatelet effects of aspirin - the irreversible inhibition of platelet-dependent cyclooxygenase 6 3 . In addition, A S A may impact on such contributory factors as haemostasis, atherogenesis, and f ibrinolysis 6 4 . A s inflammation emerges as an important factor in atherosclerosis, the ability of A S A to modify this process also serves as a possible mechanism of action 5 . The incidence of C A D in women approaches that in men after menopause 6 5 ' 6 6 . The proposed theories for this increase have centred around the protective effects of oestrogen on 13 heart disease. Prospective and clinical trials into the efficacy of hormone replacement therapy (HRT) for reducing C A D risk have given contradictory results. Many observational and prospective studies, including one by Nabulsi et a l 6 7 have shown a strong association between H R T use and a favourable cardiovascular risk profile. H R T users were shown to have higher H D L - C , lower L D L - C , and lower levels of coagulation and thrombogenic factors associated with atherosclerotic disease. However, data from the Heart and Estrogen/progestin Replacement Study (HERS) randomised clinical t r i a l 6 8 contradict these results. While l ipid changes observed in H E R S patients were comparable to other studies, a similar association with fewer C A D events was not seen. This secondary prevention trial using conjugated equine oestrogen and progestin showed an increase in C A D and venous thromboembolitic events in the first year, with a significant trend towards a subsequent reduction C A D risk over an additional three years. However, the overall reduction in cardiovascular risk with H R T never achieved significance. Due to the significance of oxidised L D L in the development of atherosclerosis, much attention has turned to the role that antioxidants may play in preventing C A D . Oxidised L D L has many negative effects on the atherogenic process. It acts as a chemoattractant for circulating monocytes and leads to cytotoxic damage of endothelial cells. This endothelial damage results in aggregation of platelets and subsequent growth factor release 2. In addition, oxidised L D L is more readily taken up by macrophages than native L D L , leading to foam cell formation 1. Antioxidants such as vitamins E and C, as well as beta-carotene, have been studied to assess their ability to reduce C A D morbidity and mortality. While observational studies show an independent relationship between antioxidant use and lower C A D event r i s k 6 9 , randomised clinical trials, including the H O P E study 7 0 have not demonstrated a significant benefit for antioxidants 7 1 ' 7 2 . 14 In addition to controlling hypertension, beta-blockers have been shown to reduce morbidity and mortality from C A D . The greatest benefit appears to be for patients who have survived the acute phase of a M I , with positive effects for lower risk individuals as w e l l 7 3 . 1.1.4.3 L i p i d Lowering L i p i d lowering may be accomplished by many different treatment modalities, from diet, to drug, to surgical means. Results from diet intervention trials vary, and suggest that dietary modification of serum lipids are less potent than l ipid lowering medications. The range of serum cholesterol lowering has been generally found on the order of 0% to 1 3 % 1 2 ' 7 4 . The addition of exercise to diet modification has been shown to reduce T C and L D L - C by 23%, and T G by 3 3 % 7 5 . The possibility has been raised that the beneficial effect of reductions in fat consumption may extend beyond simply lowering blood cholesterol levels. For example, one study by Lorgeril et a l 7 6 , where dietary fat was reduced to 30%> of total caloric intake and mirrored a Mediterranean diet, showed no mean change in cholesterol levels but demonstrated a 63% reduction in C A D events, along with significant reductions in cardiac and all-cause mortality. Medications commonly used for reducing total and L D L cholesterol include niacin, bile acid binding resins, fibrates, and 3-hydroxy-3-methylglutaryl-coenzyme A ( H M G - C o A ) reductase inhibitors, or "statins". Several large epidemiological studies have demonstrated a clear reduction in cardiovascular morbidity and mortality with drugs aimed at decreasing both T C and L D L - C , and benefits appear to be independent of initial T C leve ls 1 2 . In addition to studies with dichotomous outcome variables, angiographic regression studies have also demonstrated the positive effect to lowering cholesterol l e v e l s 3 9 ' 7 7 , albeit with lesion regression measures seemingly out of proportion to cardiac event reduction. A very small level of plaque regression was associated with large reductions in C A D event rates. 15 The Coronary Drug Project t r i a l 7 8 was the first to demonstrate a beneficial effect of niacin on overall mortality in patients with previous M I , following a 15 year follow-up. Niacin was also shown to significantly reduce the incidence of nonfatal M I compared to the placebo group. However, due to a number of dermatological and gastrointestinal side effects, there was a higher dropout rate for study patients taking niacin compared to placebo 7 9 . Bi le acid binding resins such as cholestyramine have been shown to lower cholesterol levels by preventing the reabsorption of cholesterol-derived bile acids in the small intestine. This leads to a subsequent upregulation of L D L receptors and increased uptake of L D L - C from the blood. The L ip id Research Clinics Coronary Primary Prevention Trial observed the effect of cholestyramine on lowering cholesterol in men and its role in primary prevention of C A D 8 0 . Mean reductions in L D L - C of 13% and in T C of 20% lead to a 30% reduction in C A D death. Fibrates have been shown not only to lower L D L - C , but also raise H D L - C 8 1 , and are the drug of choice in patients with hypertriglyceridaemia. The Helsinki Heart Study, which compared 600mg of gemfibrozil twice daily to placebo, showed a 34% reduction in cardiac endpoints with a 10% decrease in L D L - C and 10% increase in H D L - C 8 2 . Statin trials in particular have shown significant reductions in cardiovascular events and mortality with concurrent decreases in all-cause mortality as w e l l 8 3 . Primary prevention trials such as West of Scotland Coronary Prevention Study ( W O S C O P S ) 8 4 and A i r Force/Texas Coronary Atherosclerosis Prevention Study ( A F C A P S / T e x C A P S ) 8 5 have demonstrated reductions in risk of C A D events of 31-37% with a T C lowering of approximately 20% and L D L - C decrease of 25%. Preventing C A D morbidity and mortality with l ipid lowering in patients with established C A D has been shown in such studies as Scandinavian Simvastatin Survival Study (4S) 8 6 , Cholesterol and Recurrent Events ( C A R E ) 8 7 , and Long-term Intervention with Pravastatin in 16 Ischemic Disease ( L I P I D ) 8 8 . Reductions of 34% in major coronary events and 30%> reduction in all-cause mortality were found with a 25% decrease in total cholesterol and 35% decrease in LDL-cholesterol in the 4S study 8 6 . A s a result of the overwhelming evidence for the superior efficacy, potency, and low incidence of side effects, statins have largely supplanted the other major l ipid lowering medications as the drug of first choice 8 9 . While many of the l ipid lowering medications have been proven effective in monotherapy, there is increasing evidence that combination therapy is a safe and practical method of treating dyslipidaemia in patients who do not respond as well to single drugs. Combinations of statins and fibrates90, statins and n i ac in 9 1 , as well as concurrent use of three or more different l ipid lowering drugs have been shown to reduce T C , L D L - C , T G , and apoB while increasing H D L - C at a compound level greater than predicted for each individual medication 9 2 . 1.2 PUBLISHED GUIDELINES F O R T R E A T M E N T O F DYSLIPIDAEMIA 1.2.1 Canadian Consensus Conference on Cholesterol Guidelines The Canadian Consensus Conference on Cholesterol ( C C C C ) Guidel ines 9 3 were released in a final report in 1988 as a response to strong evidence that treatment of dyslipidaemia decreased cardiovascular morbidity and mortality. The recommendations of the panel extended to who should be screened for lipid levels, what risk factors were important to consider, at what point diet and drug therapy should be initiated, and how these therapies should be properly administered. 1.2.1.1 L i p i d Risk Factors and Testing Priorities Based on evidence from a wide range of scientific studies, the C C C C concluded that high T C and L D L - C , along with low H D L - C and an elevated T C / H D L - C ratio were strong independent risk factors for atherosclerosis. They also stipulated that T G be measured both as an important risk factor in certain groups, and to allow calculation of L D L - C . 17 According to the C C C C Guidelines, the following patients should have l ipid screening. Those with: • Cl inical evidence of C A D . • Family history of premature C A D or hyperlipidaemia in a parent, grandparent, or sibling. • Pre-existing conditions such as hypertension, diabetes, renal failure, and obesity, particularly abdominal obesity. 1.2.1.2 Dietary Therapy Recommendations for intensive dietary therapy were made for two different age groups based primarily on T C levels, with consideration given to other l ipid levels when T C was within a borderline elevated range. The guiding principles of diet modification include reducing total fat intake to no more than 30% of total calories, with saturated fats not exceeding 10%. Intensive dietary therapy, as set out by the C C C C , must include dietician counseling. The following l ipid levels necessitate diet intervention: 1. Adults > 30 years • T C > 6.2mmol/L • T C = 5.2-6.2mmol/L + • L D L - C > 3.4mmol/L or • H D L - C < 0.9mmol/L or • T G > 2.3mmol/L 2. Adults 18-29 years • T O 5.7mmol/L • T C = 4.6-5.7mmol/L + • L D L - C > 3.0mmol/L or 18 • H D L - C < 0.9mmol/L or • T G > 2.3mmol/L The desired treatment goal of the C C C C was to lower T C to 5.2mmol/L or less. 1.2.1.3 Drug Therapy C C C C Guidelines stipulated that drug therapy should be implemented only after six months of dietary therapy fails to reduce T C to the target level. Many potential lipid-lowering drugs ( L L D ) were recommended, including combination therapies. A l l drug therapy was to be accompanied by continuation of dietary therapy. Studies have since shown that combining medication with a low fat diet enhanced favourable lipid changes, and in particular resulted in a significant increase in H D L - C levels 9 4 . It should be noted that the C C C C Guidelines were released prior to the favourable conclusions of the large clinical trials of H M G - C o A reductase inh ib i to rs 8 4 ' 8 6 . A s such, drug therapy was viewed with more caution due to the failed reduction in all-cause mortality seen in fibrate, resin, and niacin t r a i l s 7 8 ' 8 0 . More recent revisions to the Canadian guidelines have designated specific cholesterol levels for administration of drug treatment that are similar to those in the National Cholesterol Education Program guidelines. 1.2.1.4 Other Risk Factors Based on epidemiological studies, the C C C C panel recommended counseling for several other C A D risk factors. They stipulated intervention on both an individual and population basis for cigarette smoking, hypertension, diabetes, obesity, and sedentary behaviour. 1.2.1.5 Canadian Guidelines 2000 In 2000, the Working Group on Hypercholesterolemia and Other Dyslipidemias published an updated series of recommendations for the treatment of dyslipidaemia 9 5 . In comparison to the 1988 C C C C Guidelines, the decision to treat is based on a multi-factorial risk 19 assessment, and target levels of L D L - C , T G , or the ratio of T C / H D L - C . A patient's risk is classified according to the algorithm shown in Table (2). Table 2. Risk assessment algorithm of the Canadian 2000 Guidelines.95 Risk points Risk Factor Men Women Age (yrs) 30-34 -1 -9 35-39 0 -4 40-44 1 0 45-49 2 3 50-54 3 6 55-59 4 7 60-64 5 8 65-69 6 8 70-74 7 8 TC (mmol/L) <4.14 -3 -2 4.15-5.17 0 0 5.18-6.21 1 1 6.22-7.24 2 2 >7.25 3 3 HDL-C (mmol/L) <0.90 2 5 0.91-1.16 1 2 1.17-1.29 0 1 1.30-1.55 0 0 > 1.56 -2 -3 SBP (mmHg) < 120 0 -3 120-129 0 0 130-139 1 1 140-159 2 2 > 160 3 3 Smoker N o 0 0 Yes 2 2 Add total risk points Calculate 10 year risk < 10% < 5 points < 9 points 10%-20% < 8 points < 14 points 20%-30% < 10 points < 17 points >30% > 10 points > 17 points Treatment with lifestyle modifications or L L D is then based on the levels of certain important l ipid parameters, given in Table (3). For patients at moderate or low risk, i f l ipid parameters are not lowered below treatment thresholds within a specific time frame, then L L D should be added to the regimen. Table 3. Canadian 2000 Guidelines risk assignment for treatment decisions.95 Level of Risk Treatment L D L - C mmol/L T C / H D L - C T G mmol/L Very High > 30% or C A D + or D M High 20%-30% Moderate 10%-20% Low < 10% L L D + diet L L D + diet diet L L D @ 3mos diet L L D @ 6mos >2.5 >3.0 >4.0 >5.0 >4 > 5 >6 >7 >2.0 >2.0 >2.0 >3.0 Significant changes from the C C C C Guidelines include the classification of patients with diabetes as very high risk for the development of C A D , and the early initiation of L L D in high risk individuals. In addition, risk classification is made more relevant by taking into account gender differences in many of the risk factors. 1.2.2 National Cholesterol Education Program Guidelines The American National Cholesterol Education Program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults released its second edition of recommendations for dyslipidaemia screening and treatment in J A M A in 1993 2 3 . These guidelines took a more specific and strict stance on treatment of l ipid risk factors for C A D . Algorithms for diagnosis of hyperlipidaemia utilised multiple l ipid measurements and risk factor assessment, and were different for primary and secondary prevention of C A D . 22 Classification of risk was based on T C , H D L - C , and number of other risk factors. Treatment decisions were then based on L D L - C levels. The algorithm for primary prevention is found in Figure (3). C A D risk factors considered the initial classification include: • M e n > 45 years of age. • Women > 55 years of age or with premature menopause and no H R T . • Family history of premature C A D in first degree relative. • Smoking • Hypertension • Diabetes • H D L - C < 0.9mmol/L —I Negative risk factor: H D L - C > 1.6mmol/L Nonfasting Total Cholesterol and HDL Cholesterol Assess Nonlipid Risk Factors TC < 5.2mmol/L TC 5.2-6.2mmol/L TC > 6.2mmol/L HDL-C > 0.9mmol/L HDL-C < 0.9mmol/L HDL > 0.9mmol/L < 2RF HDL < 0.9mmol/L >2RF Repeat TC & Diet & Exercise counseling HDL 5yrs Repeat TC & HDL 1 yr Fasting Lipid Analysis RF education RF education Fasting LDL Cholesterol LDL-C < 3.4mmol/L LDL-C 3.4-4.1 mmol/L <2RF LDL-C 3.4-4.1 mmol/L >2RF LDL-C > 4.1 mmol/L Repeat TC & HDL 5yrs RF education Diet & Exercise counseling Repeat TC & HDL 1 yr RF education Initiate appropriate therapy Figure 3. N C E P Guideline treatment algorithm for primary prevention.2 3 24 The secondary prevention treatment guidelines are outlined in Figure (4). In the case of patients with clinical evidence of C A D , T C and H D L - C levels are not used to classify risk. High risk is initially assumed and a much stricter L D L - C is used to determine appropriate treatment. Fasting LDL Cholesterol LDL-C < 2.6mmol/L LDL-C > 2.6mmol/L Diet & Exercise counseling Repeat TC & HDL 1 yr RF education Figure 4. NCEP Guideline treatment algorithm for secondary prevention.23 The appropriate therapy required by the N C E P Guidelines for primary and secondary prevention of C A D is listed in Table (4). Initiation levels for dietary and drug therapy are based on L D L - C levels. Table 4. NCEP Guideline LDL-C thresholds for dietary and drug therapy23 Risk Classification Dietary Therapy Drug Therapy N o C A D + < 2RF > 4.1 mmol/L > 4.9mmol/L N o C A D + > 2 R F > 3.4mmol/L >4.1mmol/L C A D >2.6mmol/L >3.4mmol/L 1.3 PATIENT AWARENESS OF HEALTH The prevention and management of heart disease requires not only the intervention of health professionals and administration of medical treatments, but also the involvement of patients in their own care. Patients who are aware of their health risks and understand the 25 behaviours that place them in danger of suffering from C A D are more likely to take an active role in modifying such factors as diet, physical activity, smoking habits. One important indication of patient health awareness is the accuracy with which individuals can report various physical measures. Self-reported anthropometric measurements such as weight and waist circumference have often been shown to deviate from real i ty 9 6 . It has been suggested that patient awareness of weight and other measures of obesity may reflect a cultural bias and can be differentiated by gender 9 7 . Some studies have demonstrated a tendency for obese individuals, particularly women, to underestimate their true w e i g h t 9 8 ' 9 9 , while others have concluded that self-reported weight is a reliable and accurate indicator of actual we igh t 1 0 0 . A long a similar vein, self-reporting of exercise and diet habits may be skewed towards perceived cultural ideals. A study by Lichtman et a l 1 0 1 showed that obese individuals were much more likely to under-report their caloric intake and overestimate their energy expenditure during physical activity. Adherence to an exercise program may be affected by such health beliefs as self-efficacy, perceived severity of disease, benefits of the exercise, and barriers that must be overcome 1 0 2 , as well as physical characteristics such as high body fat and weight, smoking, and C A D symptom l e v e l 1 0 3 . Knowledge of C A D risk factors is an important determinant of patient health awareness. While knowledge of cardiovascular risk factors has improved in recent yea r s 1 0 4 , there are indications that women 1 0 5 and less educated peop l e 1 0 6 may be uninformed regarding mortality and risk characteristics of C A D . One study examining the ability of participants drawn from active duty military personnel to identify important C A D risk factors showed that awareness ranged from 99% for being overweight, to 73% for elevated cholesterol levels, and 53.3% for diabetes 1 0 7 . 26 Awareness of personal risk for C A D is also vital in measuring health status. O f 1471 high risk individuals in a community-based cardiovascular risk reduction program, 73% knew their blood pressure, while only 15% knew their cholesterol level. However, an encouraging finding was that lack of risk status awareness did not correlate with self-initiated changes in lifestyle behaviours such as exercise and cholesterol reduct ion 1 0 8 . 1.4 APPROPRIATE TREATMENT OF DYSLIPIDAEMIA Except for age, dyslipidaemia is the most important predictor of C A D 3 3 . The overwhelming evidence that lowering cholesterol levels prevents both primary and secondary C A D events 8 3 , coupled with identification and treatment strategies for concurrent risk factors has led to the development of national guidelines for dyslipidaemia management. However, there is widespread evidence that treatment practices for C A D have been sub-optimal in Canada and around the world. Data from FastTrakll, a standardised quality assurance program implemented in over 100 hospitals in Canada, is shown in Figure (5). The gap between evidence-based recommendations and actual practice is seen in post-MI discharge medications 8. % of post-MI patients ASA Beta ACE blocker inhibitor Statin Discharge medications Figure 5. Post-MI discharge medications in Canada in 1998.8 27 A study by the Clinical Quality Improvement Network (CQIN) investigators 1 0 9 at four Canadian hospitals revealed drastic under-treatment of patients with a high one year risk for C A D events. Only 22% were prescribed diet therapy, and only 8% received lipid lowering medications. Moreover, only 28% had had a l ipid assessment performed in the five years prior to their index hospitalisation. A survey of Ontario physician practices with regard to C A D primary prevention showed that only one quarter started patients on drug therapy with a T C less that 6.22 mmol/L. This was by far less aggressive than seen in U S physicians. But despite having stronger stated attitudes towards dyslipidaemia treatment, along with stricter guidelines, American physicians fare no better in treating their patients appropriately. Several studies show rates of only 28% to 50% compliance with N C E P guidelines 110-11M12-113 0 f those patients prescribed the appropriate treatment, studies have shown that only one quarter to one third met the recommended L D L - C t a rge t 1 1 4 , 1 1 5 . Additional data from Britain demonstrated an 82% rate of appropriate treatment for hypertension according to guidelines, compared to only 17% for dysl ip idaemia 1 1 6 . Finally, data from the H E R S study revealed that only one-third of women with L D L - C levels over 4.1 mmol/L were receiving l ipid lowering medicat ions 1 1 7 . A n additional problem in the treatment of dyslipidaemia is patient non-compliance with taking cholesterol-lowering medications as prescribed. While L L D must be taken for at least two years to show a reduction in C A D mortality, a study of the Saskatchewan Prescription Drug P l a n 1 1 8 showed that only 25% of patients were still taking their medication after one year. Factors such as side effects, rapport with doctors, and cost all decrease the chance of patient adherence. Neither physicians nor patients have demonstrated adequate compliance to guidelines and recommendations for appropriate treatment. Public health information regarding the role of 28 cholesterol in heart disease has created confusion for both doctors and patients 1 1 9 , which in turn has contributed to sub-optimal management of dyslipidaemia in B C and Canada. 29 1.5 RATIONALE Death and disability from C A D is a major concern in Canada. Not only does atherosclerosis cost society a great deal financially, it also represents a significant burden on the emotional and psychological health of the people. A s such, the prevention and treatment of this disease should be paramount in national health practices. The evidence that treatment of dyslipidaemia and other risk factors for C A D reduces both morbidity and mortality is clear. Based on observational and prospective studies, basic science and clinical trials, guidelines for the diagnosis and management of dyslipidaemia and C A D risk have been drawn up and disseminated throughout North America. Despite all of this, numerous studies have indicated that appropriate treatment of patients with C A D or at risk of developing the disease is not given. Sub-optimal management of lipids using diet and medications is common. Interventions for other important risk factors such as smoking, exercise, diabetes, and hypertension are not being employed to the fullest potential. The aim of this study was to determine the level of appropriate treatment in a population of British Columbians who have angiographic documentation as to the presence and severity of C A D . Furthermore, information was gathered on the prevalence of adjuvant therapies, risk factor counseling, and anthropometric measures to assess what characteristics impact morbidity and mortality, as well to predict appropriate management according to guidelines. The design of the study also allowed assessment of patient awareness of health status, often a crucial element of successful outcomes. The knowledge gained from this research wi l l provide insight into the treatment practices of British Columbia physicians. It w i l l provide a basis for the education of patients and health professionals into better prevention strategies and management of cardiovascular health risks. 30 1.6 HYPOTHESIS The majority of patients seen for angiography at St. Paul's and Vancouver General hospitals between 1993 and 1994 who required dyslipidaemia treatment and risk factor modification for the purpose of primary or secondary C A D prevention w i l l not have received appropriate treatment according to published guidelines of the time. 1.7 SPECIFIC AIMS 1. To determine the appropriateness of treatment according to C C C C and N C E P guidelines in a population of angiographically defined British Columbians. 2. To establish the prevalence of C A D drug therapies such as A S A , anti-hypertensives, and H R T . 3. To ascertain the availability of risk factor counseling, including exercise, smoking cessation, and dietician interventions. 4. To assess patient awareness of health and health markers such as weight, change in exercise habits, and cholesterol levels. 5. To determine what l ipid and lifestyle risk factors independently predict C A D morbidity and mortality. 6. To determine what l ipid and lifestyle risk factors independently predict appropriate treatment. 31 2 METHODS 2.1 ORIGINAL COHORT SELECTION A total of 1109 consecutive individuals, referred to two Vancouver teaching hospitals for selective coronary angiography (SCA) , were recruited for the study between 1993 and 1995. A t the time of their S C A , fasting blood samples were collected in ethylenediaminetetraacetic acid ( E D T A ) , centrifuged, and the plasma aliquoted and stored at -70°C. On each sample, T C 1 2 0 , T G 1 2 1 , H D L - C 1 2 2 , and a p o B 1 2 3 were measured as previously described. FERHDL was determined using an isotopic assay method 1 2 4 . L D L - C was calculated using the Friedewald equat ion 1 2 5 for individuals with T G less than 4.5 mmol/L. Each angiogram was assessed semi-quantitatively by a cardiologist and scored as a) no evidence of narrowing in any vessel, or b) narrowing of either <50% or >50% in one, two, or three vessels. Patients in this study were considered to be angiographically positive for C A D whether they had less than or greater than 50% narrowing in at least one vessel. While this definition differs from that typically found in the literature, the reasoning behind it was that increasing evidence has shown that small plaques may contribute to cardiovascular morbidity and mortality more than large plaques 1 2 6 . Additionally, by including patients with any degree of narrowing in the C A D positive group, it was hoped that misclassification due to subjective differences in diagnosis around the 50%> mark could be avoided. A two-page questionnaire (Appendix 1) regarding clinical and lifestyle variables was administered to every patient by a nurse or attending cardiologist. Questions were asked regarding smoking, drinking, and exercise habits. Measurements of height, weight, sitting blood pressure, and waist circumference were obtained. Previous history of cardiovascular disease, diabetes, hypertension, and renal insufficiency in both the patient and other family members was determined. Current medications were recorded from the patient's chart. This information was 32 entered into a database and double-checked for accuracy. A l l participants in this study signed a consent form approved by both university and hospital ethical review boards. 2.2 QUESTIONNAIRE DESIGN The follow-up questionnaire (Appendix 2) was based closely on the original questionnaire given to patients at the time of angiography. Questions regarding smoking, exercise, and alcohol habits were taken word for word to ensure accurate comparison on these risk factor changes. Patients were asked to provide height and weight data, as well as an indication of whether their weight had changed in the intervening four years, in which direction, and by how much. Although waist circumference was measured at the original visit using the outdated World Health Organisation method of measuring maximum girth with the patient in the supine position, it was decided to have the patients repeat this form of measurement on follow-up. In this manner, determining the change in abdominal circumference would be more appropriate. In order to assess appropriateness of treatment as well as patient awareness and involvement in their health care, individuals were asked i f cholesterol had been measured in the past four years, the date of most recent measurement, and what the value was, i f known. A similar question was asked regarding blood pressure and previous diagnosis of hypertension. The prevalence of cardiovascular events and procedures was ascertained. Specifically, individuals were asked whether they had experienced percutaneous transluminal coronary angioplasty, coronary artery bypass graft, or myocardial infarction within the past four years since angiography. It was felt to be important that the prevalence of risk factor counseling and the utilisation of paramedical staff in treatment practices be assessed. Thus, patients were asked whether they had been counseled on various lifestyle risk factors including smoking, cholesterol, diet, and 33 exercise. Additionally, the person who served as counselor - doctor, nurse, dietician, exercise or physical therapist, or family and friends was requested. The last choice was intended to learn of societal and community involvement in patient health care. Medication information was requested. Patients were asked to supply type of medication, name, dose, and start date i f known. Having this information self-reported caused concern regarding incomplete information, but this in itself would provide data on patient understanding of their illness and treatment. Clinical and research scientists reviewed the questionnaire and provided feedback on its design, wording, and appropriateness. A group of volunteers filled out several questions on the questionnaire to assess readability and clarity, and modifications were made accordingly. 2.2.1 Ethics Approval A cover letter introducing the study and an informed consent form were drafted according to guidelines set out by St. Paul's Hospital and the University of British Columbia. These were sent, along with the finalised questionnaire and a research proposal, to the Ethics Committees from both institutions. Approval was granted by the University of British Columbia Behavioural Research Services review board and the St. Paul's Hospital Ethics Committee. Additional ethics approval was obtained from Vancouver General Hospital to allow access to health records for data acquisition and confirmation. 2.3 MAIL-OUT Record matching was attempted through the British Columbia Cardiac Registry to obtain vital status and most recent mailing addresses. Matches were made for 1027 individuals. Ninety-seven were known to be deceased as of December 31, 1997. The remaining individuals were mailed a questionnaire accompanied by a cover letter explaining the nature of the study, and two copies of an ethics board approved informed consent. Potential participants were to sign 34 both copies of the informed consent, and return one copy along with the completed questionnaire. A self-addressed, stamped envelope was provided for this purpose. Patients who had not responded within two weeks of mail-out were contacted by phone. A n y concerns they had about the questionnaire were addressed, and they were encouraged to complete and return the forms. Envelopes returned unopened and incorrect phone numbers were checked using hospital databases and Internet search engines to find a current address. A second mail-out was sent to non-responders in August 1999. Again, follow-up phone calls were made after two weeks to encourage participation. Patient returns were matched with the original information based on name, date of birth, and P H N number. They were assigned the same lab identification given to their original blood sample. Eight returns matched for name, but not for birth date or personal health number. These patients were considered lost to follow-up and the questionnaires were destroyed. A n Access 97 database was designed to capture the data from the returned questionnaires. Data was entered as questionnaires were returned and hard copies of the consent form and questionnaire were collated and filed. 2.4 V I T A L STATISTICS I N F O R M A T I O N Once the mail-out was complete, and all responders had returned their questionnaires, the British Columbia Vi ta l Statistics Agency was contacted. Patient identifying information was sent for all individuals, including those known to be deceased. Matches were made for 102 individuals for whom we obtained cause of death, indicated by International Classification of Diseases (ICD)-9 codes. There was some concern that 35 of the patients listed as deceased on the B C C R did not match with the Vi ta l Statistics database. However, the B C C R information was assumed to be correct, given the possibility that there may have been a problem with the 35 linkage to the Vi ta l Statistics database because P H N numbers were not available for those patients. 2.5 M E D I C A L R E C O R D S Permission was obtained from both St. Paul's Hospital and Vancouver General Hospital to review health records on the cohort. A random sample of records were requested and reviewed at the respective hospitals every week. In total, 129 patients with 223 discharges since 1994 were reviewed at St. Paul's Hospital and 95 patients with 165 discharges at Vancouver General Hospital. The information collected from the medical chart included history of diabetes, smoking, renal insufficiency, peripheral vascular disease, or stroke. Incidence of percutaneous transluminal coronary angioplasty (PTCA) , coronary artery bypass graft ( C A B G ) , M I , unstable angina, congestive heart failure, and other cardiovascular morbidities was recorded in order to confirm information given by patient report. Frequency of measurement and value was obtained for blood pressure and lab measurements, including T C , L D L - C , H D L - C , T G , and glucose where available. Evidence of counseling by dieticians or physiotherapists was also noted. 2.6 S T A T I S T I C A L A N A L Y S I S Statistical analysis was performed using SPSS 8.0 software. Differences between categorical variables, given as percentages, were determined using a Chi-square test. In the case of multiple comparisons between three groups - non-respondents, respondents, and deceased - a Bonferroni correction was applied. Continuous variables were assessed using Student's /-test and results are given as means ± standard error of the mean. Analysis of changes in medication usage was assessed using the McNemar test for change. Predictors of morbidity, mortality, and appropriate treatment were determined using stepwise Logistic Regression. Significance for all tests was defined as p < 0.05, and all analyses were two-tailed. 36 3 RESULTS 3.1 RESPONSE RATE A total of 1109 patients were recruited in the initial angiography study. Using original questionnaires and hospital databases, addresses were found for 1027 individuals. Linkage with the British Columbia Cardiac Registry database revealed that 97 of the 1027 patients were deceased as of December 31, 1997. Thus, 930 questionnaires were mailed out in two mailings with follow-up phone calls to encourage response. One hundred and thirty-eight questionnaires were returned unopened, and correct addresses could not be found using every means available. Ultimately, 314 completed questionnaires were returned. Either through returned letters, telephone calls, or subsequent Vi ta l Statistics record linkage, an additional 53 subjects were determined to have died since December 1997. However, five of the patients listed by Vi ta l Statistics as deceased had died after responding to the questionnaire. Thus in total, there were 314 respondents, 430 non-respondents, and 150 deceased patients from the original cohort, with 220 lost to follow-up. Response rate = respondents original cohort - lost to follow-up - (deceased - prior respondents) Response rate = 314 = 314 1109-220-(150-5) 744 Response rate = 42% 3.2 BASELINE CHARACTERISTICS 3.2.1 Demographics O f the 314 respondents, 237 were males (75%). Males made up 73% of the deceased group. This was not significantly different from the proportion of males among non-responders (72%). Eighty-one percent of follow-up patients had positive angiograms at their initial visit (CAD+), compared with 87% of deceased patients. There was a significantly smaller proportion of non-responders with C A D than among responders (75%; p<0.05). This may be explained by the possibility that C A D - individuals may not have seen the need to participate in the study. Patients with negative S C A were included, however, in order to determine risk factor prevalence and treatment in the context of primary prevention as well as secondary prevention. The average age of male respondents was 60.9±0.6 years, of females was 64.4±1.1 years (p<0.01). 3.2.2 Lipid Profile The mean l ipid levels at baseline for respondents and deceased patients are given in Table (5). There were no significant differences in the initial l ipid parameters comparing non-respondents to deceased or respondents. Means are separated for those taking L L D at baseline and those not on medications. Table 5. Mean lipid levels for respondent and deceased patient groups, separated by LLD use at baseline. Respondents Deceased LLD-(n=256) LLD+ (n-58) LLD-(n=131) LLD+(n=15) T C (mmol/L) 5.19±0.08 4.96±0.12 5.15±0.10 4.99±0.30 T G (mmol/L) 1.82±0.08 2.13±0.18 1.76±0.09 1.47±0.16 JTM r-r , ^ 3.63±0.07 3 .20±0.11* . 1 ( U „ n o . . ^ ^ L D L - C (mmol/L) , , 3.70+0.09 3.56+0.26 v ' (n=255) (n=57) - -Values given as mean + standard error of the mean. Comparison between groups for TG was based on logarithmically transformed values. * Significantly lower than LLD- respondents (p<0.005) 38 3.2.3 Risk Factor Prevalence Table (6) outlines the prevalence of risk factors at baseline for non-respondents, respondents, and deceased patients. It should be noted that five of the respondents subsequently died, and these patients have been included in both the respondent and deceased groups. Advanced age as a risk factor reflected men > 45 years or women > 55 years 2 3 . Hypertension was defined as a systolic blood pressure (SBP) > 140mmHg and a diastolic blood pressure (DBP) > 90mmHg, or taking anti-hypertensive medicat ions 1 2 7 . Overall obesity reflected a body mass index (BMI) greater than or equal to 30 kg/m 2 or an abdominal girth greater than 102cm in men or 88cm in women 1 0 . Elevated T C was defined as T C > 5.2 mmol/L, elevated L D L - C was > 3.4 mmol/L, and low H D L - C was < 0.9 m m o l / L 2 3 . Elevated T G reflected the C C C C guideline level of risk of > 2.3 m m o l / L 9 3 . FERHDL was used as a marker of small, dense L D L particles. Individuals with an FERHDL greater than the 95th percentile (> 17.97 for women, > 25.80 for men) were designated as having this conditional risk fac tor 1 2 8 . Abdominal obesity as a predisposing risk factor was defined as a waist circumference of greater than 102cm in men or greater than 88cm in w o m e n 1 2 9 . 39 Table 6. Prevalence of baseline risk factors for non-respondent, respondent, and deceased patient groups. Classification Risk Factor Non-respondents (n=430) Respondents (n=314) Deceased (n=150) Major M e n > 45 88%(n=310) 93% (n=237) 94% (n=109) Women > 55 78% (n=120) 88% (n=77) 83% (n=41) Current smokers 18%(n=425) 7% a (n=307) 21%(n=145) Former smokers 51% b (n=425) 64% (n=307) 57% (n=145) Hypertension 38% 37% 35% Diabetes 17% 16% 18% Obesity 51%(n=429) 49%(n=311) 48% (n=145) Physical inactivity 10% (n=420) 8% (n=305) 15%(n=142) Elevated T C 44% (n=423) 43% 40% (n=146) Elevated L D L - C 55%(n=418) 56%(n=312) 54% (n=146) L o w H D L - C 45% (n=423) 42%(n=313) 45% (n=146) Conditional Elevated T G 23% (n=423) 21% 19%(n=146) Elevated FERHDL 45% (n=423) 45%(n=313) 41%(n=146) Predisposing Abdominal obesity 48% (n=403) 44% (n=294) 50%(n=127) Family history 44% 48% 41% For variables with missing data, the total number of individuals is given in brackets and the prevalence is a percentage of this number. a Significantly lower than non-respondents and deceased (p<0.001) b Significantly lower than respondents (p<0.001) 3.2.3.1 Subgroup analyses (The first set of results makes up a portion of the paper: Francis M C , Frohlich JJ. Coronary artery disease in patients at low risk - apolipoprotein A I as an independent risk factor. Atherosclerosis. In press . ) 1 3 0 Two subgroup analyses were performed on the original cohort. The first was to determine the lipid and lifestyle risk factors that predicted C A D in those patients without any of the major risk factors. Individuals with T C < 5.2 mmol/L, H D L - C > 0.9 mmol/L, SBP < 140 mmHg and D B P < 90 mmHg, no diabetes, and no family history of premature C A D in first degree relatives were selected. Preliminary analysis indicated that apo A I was significantly 40 lower in patients with positive angiograms. Fifty-four patients met the selection criteria, 29 having positive evidence of C A D (CAD+) and 25 with no S C A evidence of disease ( C A D - ) . Chi-square analysis of gender in C A D + and C A D - patients revealed a significantly greater number of females in the control group. Therefore, all further univariate analyses of risk factors included an adjustment for gender. Because apoAI concentration is known to vary with age in females but not in ma le s 1 3 1 , and since we were adjusting for gender, we felt it important to adjust for age as well . Table (7) shows the results of logistic regression analysis for all lifestyle risk factors. The only variable showing a significant difference between patients with C A D and those without was gender, as expected. Table 7. Univariate analysis of lifestyle risk factors for C A D - and CAD+ groups. Risk Factor C A D - (n=25) CAD+ (n=29) Univariate p Adjusted p Gender <0.001 Male 44% 86% Female 56% 14% Age (yr) 58.9±2.22 63.8±2.49 ns Waist (cm) 76.1±7.3 82.9±5.6 ns ns B M I (kg/m 2) 25.2±0.9 25.9±0.9 ns ns Smoking ns ns Never 52% 21% Past 40% 62% Current 8% 14% Logistic regression of each l ipid variable with adjustment for age and gender is shown in Table (8). A lower apoAI concentration in C A D + individuals was the strongest factor associated with C A D (p < 0.001), while a lower L D L - C and higher FERHDL were also significant. Higher T G concentration in C A D + patients was of borderline significance. O f the parameter ratios examined, the ratio of apoB/apoAI showed a significant relationship with the presence of disease. 41 Table 8. Univariate analysis of lipid risk factors for C A D - and CAD+ groups. Risk Factor C A D - (n=25) CAD+ (n=29) Univariate p Adjusted p T C (mmol/L) 4.67±0.08 4.39±0.09 <0.05 ns T G (mmol/L) 1.30±0.11 1.63±0.15 ns <0.05 H D L - C (mmol/L) 1.21±0.05 1.10±0.03 <0.05 ns L D L - C (mmol/L) 3.48±0.09 3.07±0.12 <0.01 <0.05 Apo B (mg/dL) 83.8±3.2 83.4±2.9 ns ns Apo A I (mg/dL) 135.4±6.5 (n=22) 111.1+3.0 (n=28) <0.002 <0.001 FERHDL 15.56±1.24 19.08±1.10 <0.05 <0.01 T C / H D L - C 3.97±0.14 4.09±0.14 ns ns Apo B / A p o A I 0.66±0.05 (n=22) 0.77±0.04 (n=28) ns <0.05 Forward stepwise logistic regression was used to build a model of predictors for C A D in this cohort of patients free from major risk factors. Results are shown in Table (9). The only variables found to associate with disease presence were gender, age, and apoAI concentration. Risk factors believed to be potential confounders for the effect of apoAI were also tested using logistic regression. Table 9. Multivariate logistic regression of C A D predictors. Risk Factor O R p value Gender* 0.14 <0.02 Age 1.14 <0.01 Apo A I 0.94 <0.002 * Males coded as 0, females coded as 1. The association of reduced levels of apoAI with C A D presence remained significant after adjusting for FERHDL, H D L - C , T G , L D L - C , and use of medications such as beta-blockers or L L D (data not shown). Insulin/glucose ratio did not confound the effect of apoAI concentration, nor did waist circumference, B M I , or smoking (data not shown). 42 The second subgroup analysis was undertaken to test the hypothesis that different risk factors would be associated with C A D in women > 60 compared to women younger than 60. From the original cohort, all women who had been referred for their first angiography and had no history of taking L L D were selected. O f these 203 women, 76 were < 60 years of age, and 127 were > 60 (62%). Univariate analysis of C A D risk factors for younger women is presented in Table (10). Some l ipid parameters were significantly associated with C A D when analysed individually. However, when adjustment was made for the contribution of smoking, the effects were eliminated. Table 10. Univariate analysis of C A D risk factors in women < 60 years. Risk Factor C A D - (n=43) C A D + (n=33) p value Age (yrs) 50.4±1.1 52.2±1.4 ns Menopause 47% 64% ns Family history 47% 64% ns Hypertension 26% 21% ns Diabetes 5% 24% <0.01 Waist (cm) 80.4±4.4 82.5±5.4 ns B M I (kg/m 2) 27.2±0.8 27.2±1.1 ns Smoking (n=41) (n=32) <0.01 Never 54% 22% Former 34% 34% Current 12% 38% Univariate analysis of C A D predictors in women > 60 years showed that no lifestyle risk factors were associated with positive S C A , other than age. However, several l ipid risk factors were related to C A D in older women, even after adjustment for smoking. These results are shown in Table (11). 43 Table 11. Univariate analysis of C A D risk factors in women > 60 years. Risk Factor C A D - (n=62) C A D + (n=63) p value Age (yrs) 67.8±0.7 70.6±0.7 <0.01 T C (mmol/L) 5.20±0.12 5.56±0.11 <0.05 T G (mmol/L)* 1.39±0.08 1.87±0.12 <0.0001 L D L - C (mmol/L) 3.72±0.12 3.97±0.10 ns H D L - C (mmol/L) 1.17±0.04 1.11±0.04 ns T C / H D L - C 4.69±0.18 5.35±0.20 <0.05 Apo B (g /L) t 0.94±0.03 1.03±0.03 <0.05 FERHDL 17.27±0.87 19.03±0.89 ns * All analyses performed on logarithmically transformed values. t n=60 for CAD- group, n=62 for CAD+ group. Multivariate analysis of predictors (Table 12) revealed that diabetes and smoking were the sole variables associated with C A D presence in younger women. In older women, stepwise logistic regression of all variables showed that only advancing age and elevated triglycerides predicted disease. Table 12. Multivariate analysis of C A D predictors in women < 60 and > 60. W O M E N < 60 W O M E N > 60 Predictor O R p value Predictor O R p value Diabetes 26.7 <0.001 T G 15.7 <0.01 Smoking Age 1.1 <0.05 Former vs never 3.4 ns Current vs never 16.4 <0.001 3.2.4 Medication Usage Table (13) details the baseline medications of the cohort. H R T prevalence is given as a percentage of women, while cholesterol medication types are shown as a percentage of those taking L L D . 44 Table 13. Proportion of non-respondents, respondents, and deceased patients taking medications at baseline. Medication Non-Respondents (n=430) Respondents (n=314) Deceased (n=150) Aspir in 65% 71% 61% H R T 23% (n=120) 18%(n=77) 17% (n=41) Antioxidants 3% a 9% 5% Beta-blockers 47% 49% 47% A C E inhibitors 16% 18% 29% b Ca-channel blockers 37% 39% 36% L L D 17% 18% 11% Statins 73% (n=71) 82%(n=51) 85% (n=13) Fibrates 18% (n=71) 16%(n=51) 15%(n=13) Combination R x 4% (n=71) 2%(n=51) — a Significantly lower than respondents (p<0.002) b Significantly lower than non-respondents and respondents (p<0.005) 3.3 F O L L O W - U P C H A R A C T E R I S T I C S 3.3.1 Changes in Risk Factor Prevalence Fifty percent of respondents had hypertension at follow-up, either because they acknowledged taking anti-hypertensives, or reported that their S B P was > 140mmHg and their D B P was > 90mmHg. This is a significant increase over the number of respondents with hypertension at baseline (37%; p<0.001). There were significantly fewer obese individuals at follow-up (39% vs 49%; p<0.025). Obesity could not be assessed in one respondent due to missing information about waist circumference and weight on the returned questionnaire. It is important to note that both abdominal girth and weight were self-reported at follow-up, and thus there may be underestimation errors in the given values (see section 1.3). This may partially explain the significant decrease in the number of obese individuals at follow-up. There was no significant 45 difference in the number of respondents who reported being abdominally obese on the questionnaire compared to baseline prevalence. The only other lifestyle risk factor to change significantly over the intermediary four years was incidence of diabetes. Twenty-three percent (72) of respondents had diabetes at follow-up, compared to 16% (51) at baseline (p<0.05). 3.3.2 Changes in Medication Usage Self-reported medications at follow-up are given in Table (14), with baseline percentages provided for reference. The McNemar test for change was used to compare proportions to baseline values, and p-values can be found in the table. This test analyses the change in medication usage for those patients with information available at both baseline and follow-up. The number of available subjects for each drug is given in parentheses, and the prevalence is a percentage of this total. Table 14. Self-reported medication usage of respondents at follow-up, with comparison to baseline prevalence. Medication Baseline Follow-up p-value A S A (n=314) 71% 86% <0.001 H R T (n=77) 14% 17% ns Antioxidants (n=220) 9% 45% <0.001 Beta-blockers (n=189) 50% 58% <0.05 A C E inhibitors (n= 189) 23% 39% <0.001 Ca-channel blockers (n=189) 39% 46% ns L L D (n=314) 18% 55% <0.001 Statins (n=40) 83% 80% ns Fibrates (n=40) 15% 8% ns Combination rx (n=40) 3% 13% ns 46 3.3.3 Morbidity Table (15) details the proportion of follow-up patients reporting cardiovascular morbidity events. The first column contains events that occurred prior to angiography, while the second column gives the percentage of events within the intervening years before follow-up. Fifty-one percent of events at follow-up occurred in patients with no prior history of M I or revascularization procedure, while 49% occurred in patients who had already suffered a C A D event. Table 15. Self-reported revascularization procedures and MI of respondents at follow-up. Morbidity Baseline Follow-up P T C A 16% 36% C A B G 8% 24% M I 32% 14% 3.3.4 Mortality Five of the 150 deceased patients died after responding to the questionnaire. Data were obtained on cause of death from the B C Vi ta l Statistics Agency for 102 of the patients. Figure (6) breaks down the causes of death by category. Acute myocardial infarction (AMI) falls under the category of C A D , but has been separated to show that it makes up a substantial proportion of C A D deaths. Overall, the percentage of patients who died of cardiovascular disease in this cohort was 70%. 47 Other Cancer CAD Accidents Diabetes Stroke AMI Figure 6. Causes of death for 102 deceased cohort patients. 3.4 PREDICTORS O F MORBIDITY AND M O R T A L I T Y 3.4.1 Morbidity Stepwise logistic regression based on likelihood ratio was performed for 280 respondents with complete data, in order to establish those variables that predicted C A D morbidity during the follow-up period. The only significant predictor of P T C A , C A B G , and M I was a positive angiographic finding at baseline. Patients with positive angiograms had nearly a six-fold higher risk of requiring revascularization or suffering from a M I during the intervening four years (RR: 5.59, 95% CI: 2.71-11.56; p<0.0001). 3.4.2 Mortality Variables considered to have a possible association with mortality were tested using stepwise logistic regression for all responders, non-responders, and deceased patients (n=781). One hundred and fourteen patients were removed from the analysis because of missing data. Those individuals that were lost to follow-up were not included in the total because their vital 48 status could not be ascertained. Table (16) describes the baseline characteristics predictive of all-cause mortality. Table 16. Variables predictive of all-cause mortality based on multivariate logistic regression. Descriptor Level O R 9 5 % CI p-value Age 1.04 1.02-1.06 <0.0001 Exercise Occ (512) vs None (85) 0.56 0.31-0.999 <0.05 M o d (157) vs None 0.99 0.51-1.93 ns Heavy (27) vs None 0.48 0.13-1.84 ns Smoking Ever (559) vs Never (222) 1.77 1.10-2.86 O . 0 2 C A B G (69) 2.57 1.41-4.65 O . 0 0 5 L L D (132) 0.44 0.23-0.83 <0.GT T C / H D L - C 1.10 1.01-1.21 <0.05 Numbers in parenthesis represent the number of respondents in each category of risk factor Baseline predictors of cardiovascular death are given in Table (17). The total number of patients included in this analysis was 740 because cause of death could not be determined in an additional 41 individuals. Table 17. Predictors of C V mortality based on multivariate logistic regression. Descriptor Level O R 9 5 % CI p-value Age 1.06 1.03-1.10 O.0001 Exercise Occ (489) vs None (83) 0.41 0.19-0.86 0.05 M o d (143) vs None 0.82 0.35-1.89 ns Heavy (25) vs None 0.32 0.04-2.70 ns Smoking Ever (530) vs Never (210) 2.83 1.32-6.07 <0.005 T G L * 0.23 0.06-0.94 O . 0 5 T C / H D L - C 1.21 1.06-1.38 O . 0 1 * Analysis performed on logarithmically transformed values. 49 3.5 P A T I E N T A W A R E N E S S O F H E A L T H 3.5.1 Changes in Heal th Measures 3.5.1.1 Weight Self-reported weight change was separated into three categories: increased, decreased, and no change. The difference between patient-given weight at follow-up and measured baseline weight was calculated. To categorise actual weight change, the category of no change was defined as the follow-up weight being within ± 2 kg of baseline weight. This figure was chosen because it represented an approximate 5 lb change in weight. It was believed that this was a reasonable margin of error for self-assessment of weight change, and it has been used in other studies of self-reported weight accuracy 9 7 . A s such, a true increase in weight over the four years had to be an increase by more than 2 kg over baseline weight, and a decrease was a reduction by more than 2 kg. Self-reported weight and change was available for 300 patients. Figure (7) shows the breakdown of correct responses, as well as under- and over-estimation of true weight change. 50 Patient-stated weight change • No change • Decrease • Increase Increase Decrease N o change Recorded weight change Figure 7. Comparison of patient-reported weight change with actual weight change. Weight change as calculated by subtracting baseline measured weight from follow-up recorded weight is represented categorically along the X-axis . The tri-coloured bars indicate the proportion of respondents stating their weight increased (blue), decreased (maroon), or remained the same (yellow) in response to the question "Has your weight changed in the past four years?" In total, 53% of respondents correctly identified the change in their weight based on reported follow-up weight. There was no significant difference between men and women or C A D + and C A D - individuals. There was no difference in accuracy of reported weight change for patients in the three B M I classifications - healthy weight, overweight, and obese. However, significantly fewer respondents who were abdominally obese at follow-up were accurate in their assessment of weight change over four years (Non-obese: 58% correct; Obese: 44% correct; p<0.05). 51 3.5.1.2 Exercise habit Patients were asked in the original questionnaire to describe their exercise habits. A t follow-up, they were asked to state whether their exercise habits had changed in the interim four years, and their current level of participation in physical activity. O f those who said that their exercise levels had not changed, 54% correctly stated that their current exercise was identical to their level at baseline. A s for those who said their exercise habits had changed, only 42% reported an exercise level different from that given on the original questionnaire (p<0.05). 3.5.1.3 Alcohol intake Alcohol intake was assessed at both baseline and follow-up, using an identical question structure. Individuals were also asked whether their drinking habits had changed. Two hundred and sixty-five stated that there had been no change in their drinking pattern, with baseline and follow-up responses available for 250. For these patients, 70% accurately noted that their current alcohol intake was identical to that recorded on the original questionnaire. Only 54% of the 48 patients stating their drinking habit had changed correctly recorded so on the new questionnaire (p<0.05). 3.5.2 Risk Markers 3.5.2.1 Hypertension Patients were asked i f they knew their blood pressure, i f they had been diagnosed with hypertension, and i f they were taking anti-hypertensive medications. One hundred and seventy-three of the 314 respondents (55%) knew their blood pressure. O f these, 18 reported a SBP > 140 mmHg and a D B P > 90 mmHg. Sixteen (89%) reported having hypertension, and 13 (72%) stated they were on medications to lower their blood pressure. 52 Anti-hypertension medications were reportedly prescribed to 146 respondents. Ninety-six (66%) of these patients knew their blood pressure, and 71% had been diagnosed with hypertension, according to self-report. 3.5.2.2 Cholesterol Individuals were asked not only i f their cholesterol had been measured during the follow-up period, but also i f they knew their cholesterol level. O f all respondents, 238 (76%) had had their cholesterol measured within the four years since angiography. However, only 41% of the 314 respondents knew what their cholesterol level was. 3.5.3 Instillation of Awareness by Others 3.5.3.1 Risk factor counseling To assess the attention certain risk factors were receiving from health professionals and the general public, we inquired about counseling given to patients. In total, 224 (71%) of respondents reported being counseled regarding cholesterol. Fifty-one of the 58 (88%) patients on L L D at baseline reported receiving counseling, while 87% of those on L L D at follow-up said they were counseled about cholesterol. Blood pressure was reportedly discussed with 61% of all respondents, and with 93 of the 116 (80%) who had hypertension at baseline. Twenty-four percent said they were counseled about smoking; 87% of current smokers at follow-up reported being counseled, 8% of those who reported never smoking on the follow-up questionnaire, and 24% of patients who stated they had quit smoking. The importance of exercise in C A D risk reduction was said to have been addressed with 56% of respondents, diet and weight loss discussed with 62%, and 25% reported receiving counseling on alcohol intake. Table (18) gives the results of Pearson's % analyses of the associations between counseling of risk factors and patient-reported change in risk behaviour. 53 Table 18. Pearson's %2 associations between risk factor counseling and reported change in risk factor behaviour. Comparison p-value E t O H counseling vs E t O H change ns Exercise counseling vs Exercise change <0.01 Exercise instructor vs Exercise change <0.001 Diet counseling vs Diet change <0.02 Dietician counselor vs Diet change <0.001 3.5.3.2 Health professionals and laypersons The vast majority of counseling was performed by physicians - 84% of respondents reported being counseled by a doctor. Other health care professionals acted as a source of risk information much less frequently. Nurses reportedly counseled 14% of patients, dieticians 37%, exercise specialists 11%, and physio- or occupational therapists only 5%. Comparably, 15% of patients reported being counseled on cardiovascular risk factors by friends and family. 3.5.4 Comparison with Hospital Records Medical records were reviewed at Vancouver General Hospital and St. Paul's Hospital to verify patient-reported morbidity. One hundred and twenty-eight charts for respondents were screened for admissions for P T C A , C A B G , or M I . Fourteen (11%) patients were unaware that they had either had a revascularization procedure or suffered an acute M I . Two patients had been diagnosed with diabetes according to their hospital chart, yet did not report this fact on the follow-up questionnaire. In addition, three individuals stated that they did not suffer from diabetes at follow-up although they were noted to have diabetes on the original questionnaire. Only 22 of the 128 patients whose medical records were assessed had undergone lipid assessment in hospital (17%). In comparison to the T C levels at baseline, the concentration increased in 12 patients and decreased in 10. 54 3.6 A P P R O P R I A T E T R E A T M E N T 3.6.1 C C C C Guidelines 3.6.1.1 L i p i d Testing C C C C guidelines do not stipulate the frequency of l ipid testing that is appropriate, but they do specify which individuals should have their lipids assessed. Patients with previously diagnosed C A D , family history of premature C A D , hypertension, diabetes, renal insufficiency, or obesity should have regular cholesterol measurements. O f the responders who fell into these categories, which included all but four patients, 76% reported having their cholesterol measured in the intervening four years. 3.6.1.2 Dietary Therapy Using C C C C guideline criteria, 136 of the 314 respondents (43%>) required intensive dietary therapy for their dyslipidaemia. Table (19) details the number of patients who reported receiving the necessary treatment. Individuals have been separated into two groups - those with T C > 6.2 mmol/L, and those with TO=5.2-6.2 mmol/L who also met the other l ipid cut-offs for treatment. These groups are based on C C C C guidelines for treatment, which also separate patients based on the two categories of T C level. P-values for %2 tests comparing the proportions are also given. Table 19. Proportion of respondents receiving appropriate dietary therapy with comparison between patients with high T C versus those with borderline high T C . T C > 6.2 mmol/L (n=45) TC=5.2-6.2 mmol/L (n=91) p-value Diet counseling 62% 64% ns Dietician counselor 40% 35% ns L o w fat diet 86% (n=44) 66% (n=86) <0.025 55 3.6.1.3 Drug Therapy Lowering T C below 5.2 mmol/L was the stated goal of the 1988 C C C C guidelines. If this goal was not met after six months of dietary therapy, commencement of drug therapy was recommended. Forty-one percent of respondents were able to provide a recent cholesterol level. O f the 136 who should have received dietary therapy, 101 either had a current T C > 5.2 mmol/L, or their cholesterol level was unknown. Table (20) gives the proportions being managed appropriately with medications. Table 20. Proportion of respondents receiving appropriate drug therapy with comparison between patients with high T C versus those with borderline high T C . T C > 6.2 mmol/L TC=5.2-6.2 mmol/L p-value (n=35) (n=66) Cholesterol counseling 89% 70% <0.05 L i p i d lowering drug 74% 45% <0.GT Examining only those individuals known not to have met the T C target (n=32), only 53% were taking a L L D at follow-up. For those patients who reported meeting the target of T C < 5.2 mmol/L (n=35), 83% were prescribed a L L D . Table (21) outlines the total percentage of respondents receiving appropriate dyslipidaemia treatment according to C C C C guidelines. Table 21. Proportion of respondents receiving appropriate treatment according to C C C C guidelines. Appropriate treatment N Percentage Cholesterol check 311 76% Diet counseling 136 63% Dietician counselor 136 38% L o w fat diet 130 73% Cholesterol counseling 136 79% L L D 101 55% 56 3.6.1.4 Canadian Guidelines 2000 While appropriate treatment of patients assessed in 1993 and 1994 cannot be judged using guidelines published in 2000, a useful comparison may be made. Table (22) illustrates the percentage of patients who would be considered to have received proper treatment according to the 2000 Guidelines i f they were assessed today. Table 22. Proportion of respondents receiving appropriate treatment according to Canadian 2000 guidelines. Appropriate treatment N Percentage Cholesterol check 313 76% Diet counseling 274 64% L o w fat diet 268 72% Cholesterol counseling 274 75% L L D 274 59% 3.6.2 N C E P Guidelines 3.6.2.1 Annual cholesterol measurement According to N C E P guidelines, 285 of the respondents (91%) should have had their cholesterol monitored on a yearly basis. However, based on questionnaire responses, only 220 reported having their cholesterol checked at any time in the four years following their initial visit and S C A . Thus, 23% of these patients were not being followed appropriately regarding regular l ipid assessment. When the respondents are separated into those with angiographic evidence of C A D compared to those without, significantly more of the C A D + patients had their cholesterol checked within the follow-up period (79% vs 62%, p<0.025). O f note, because we asked simply i f cholesterol had been checked since S C A , we cannot ascertain i f l ipid analysis was, in fact, performed annually, as is stipulated by N C E P guidelines. 57 3.6.2.2 Dietary therapy In patients with pre-existing C A D , a fasting L D L - C measurement greater than or equal to 2.6 mmol/L is an indication for dietary therapy. Using these criteria, 211 of the C A D + respondents (67% of the total, 83%> of C A D + individuals) should have received dietary counselling and been placed on a low-fat diet. Sixty-six percent reported receiving counselling on their diet, while 71% indicated that they were on a low-fat/low-cholesterol diet. For the purposes of primary prevention, the number of risk factors in combination with a L D L - C cutoff determines the need for dietary therapy. Seventeen of the C A D - respondents met the criteria for dietary therapy. O f these, 59% had received dietary counselling, and 71%> of them were on a low-fat/low-cholesterol diet. In summary, 228 patients should have been placed on dietary therapy, based on N C E P guidelines for treatment. However, only 70%> of these individuals reported being on a low-fat/low cholesterol diet. Thus, 30% of respondents were not being treated appropriately according to accepted guidelines for C A D prevention. 3.6.2.3 Drug therapy A s for dietary therapy, initiation of drug therapy is based on presence of C A D , number of risk factors, and L D L - C levels. For those patients requiring secondary prevention, 57% (144) had L D L - C levels greater than 3.4 mmol/L, necessitating drug therapy. Sixty-three percent reported taking a lipid-lowering agent at follow-up. Only nine C A D - patients had L D L - C levels indicative of drug therapy. O f these, only 44% were prescribed a lipid-lowering drug. Thus of the 153 respondents (49%>) requiring medications to manage their dyslipidaemia according to guidelines, only 62% were receiving appropriate treatment. Table (23) shows the percentage of respondents receiving appropriate treatment according to N C E P guidelines. 58 Table 23. Proportion of respondents receiving appropriate treatment according to N C E P guidelines. Appropriate treatment N Percentage Cholesterol check 285 77% Diet counseling 227 65% L o w fat diet 227 70% Cholesterol counseling 153 75% L L D 153 62% 3.7 PREDICTORS O F A P P R O P R I A T E T R E A T M E N T 3.7.1 C C C C Guidelines Baseline risk factors thought to be associated with appropriate treatment were tested using stepwise logistic regression. Table (24) provides the results of multivariate analysis for four critical characteristics of appropriate dyslipidaemia management according to C C C C guidelines. Table 24. Variables predictive of appropriate treatment according to C C C C Guidelines. Guideline Predictor Odds Ratio 95% CI p-value Cholesterol checking Age 0.95 0.92-0.98 <0.002 (n=277) C A D + 3.28 1.61-6.68 <0.001 H D L - C 7.75 2.02-29.77 <0.002 Dietician counselor Diabetes 4.05 1.29-12.69 <0.01 (n=112) Rural vs urban* 0.39 0.17-0.89 <0.02 L o w fat diet Obese 0.24 0.08-0.66 <0.005 (n=108) P T C A 0.04 0.003-0.38 <0.001 T C 4.70 1.63-13.57 <0.001 Rural vs urban* 2.84 1.03-7.81 <0.05 Drug Therapy L L D unestim. unestim. <0.0005 (n=85) T C 3.23 1.20-8.66 <0.01 *Urban coded as 1, rural coded as 0. See section 3.8 for definition of rural and urban. 59 3.7.2 N C E P Guidelines Baseline characteristics predictive of appropriate treatment according to American guidelines are given in Table (25). Table 25. Variables predictive of appropriate treatment according to N C E P Guidelines. Guideline Predictor Odds ratio 95% CI p-value Cholesterol checking Age 0.95 0.92-0.99 <0.005 (n=257) C A D + 2.87 1.20-6.87 <0.02 H D L - C 7.37 1.77-30.70 <0.005 L o w fat diet Obese 0.49 0.25-0.98 <0.05 (n=200) Alcohol habit — ~ <0.05 occ vs none 0.51 0.19-1.36 ns mod vs none 0.25 0.08-0.78 <0.02 heavy vs none 0.11 0.01-0.83 <0.05 P T C A 0.39 0.16-0.95 <0.05 A S A 2.47 1.16-5.25 <0.02 Exercise ever vs never 5.20 1.90-14.27 <0.001 Drug Therapy P T C A 13.96 1.67-116.98 <0.001 (n=137) M I 0.21 0.07-0.65 <0.005 L L D 23.82 2.53-224.01 <0.0002 L D L - C 0.11 0.03-0.37 <0.0001 H D L - C unestim. unestim. <0.0001 T C / H D L - C 14.69 4.48-48.17 <0.0001 3.8 R E G I O N A L VARIATIONS IN A P P R O P R I A T E T R E A T M E N T Respondents were defined as either rural or urban based on mailing addresses. The British Columbia Government statistics web site (www.bcstats.gov.bc.ca) was used to obtain Health Area categories in order to classify the respondents. Urban dwellers included those l iving in Kelowna (health area 23), Kamloops (health area 24), the Lower Mainland (health areas 34-45), Prince George (health area 57), Victoria (health area 61), or Nanaimo (health area 68). Table (26) shows those variables that were significantly different between rural and urban 60 patients. Continuous variables have been analysed with a Student's /-test, categorical with a x test. Table 26. Variables showing regional variation on univariate analysis. Variable Rural (n=99) Urban (n=206) p-value Age 59.6±1.0 62.9±0.7 <0.01 B M I 29.0±0.5 27.4±0.3 <0.005 Baseline diabetes 5% 21% <0.001 Follow-up diabetes 12% 27% <0.005 Follow-up low fat diet 62% (n=97) 74% (n=201) <0.05 C C C C guidelines - low fat diet 26% 31% <0.05 C C C C guidelines - dietician counselor 21% 12% <0.05 Closer examination of the crossover effect seen with respect to patients on a low fat diet and counseled by a dietician in rural and urban areas suggests that dieticians have a stronger influence on patients living in urban centres, despite their less-frequent use. Table (27) shows the results of % analysis of low fat diet with adjustment for dietician counseling. Table 27. Regional crossover effect of dietician counseling on self-reported low fat diet intake. Rural Urban Low fat diet Low fat diet p-value N o dietician counselor 62% (n=63) 69% (n=126) ns Dietician counselor 62% (n=34) 83% (n=75) <0.02 There is also an association between counseling by a dietician and reported diet change for urban dwellers (p<0.001) but not for rural patients. 61 4 D I S C U S S I O N 4.1 L I M I T A T I O N S O F T H E S T U D Y 4.1.1 Methodology The diagnosis of C A D in this cohort was based on semi-quantitative selective coronary angiography (SCA) . While S C A has long been the gold standard for the assessment of coronary atherosclerosis presence and severity, there has been increasing evidence that its sensitivity is low. The diagnosis of percent stenosis relies on a comparison with adjacent segments of the artery. Due to the diffuse nature of atherosclerosis, there is a possibility of disease underestimation 1 3 2 . For example, i f a section of the lumen is reported to be 50% narrower than the surrounding vessel, but the entire vessel is already narrowed by 30%>, then the occlusion is in fact 80%). In addition, new theories of plaque expansion have suggested that the outer wall of the vessel may bulge as the plaque grows, thus leaving the lumenal space unchanged and C A D undetectable by S C A 1 3 3 . However, long-term studies of patients with chest pain and normal angiograms, have shown them to have a good prognosis, with no greater C A D morbidity than individuals not referred for S C A 1 3 4 . The nature of the follow-up study design has certain limitations. Whereas the baseline data on risk factors, morbidity, and medications was collected from hospital charts, the follow-up information was gathered solely by patient self-report. Thus the chance for inaccuracies is high. Patients may be unclear on what medications they are taking and their intended purpose. Numerous studies have demonstrated that people incorrectly report weight 9 7 , exercise, d i e t 1 3 5 , and other risk behaviours. Random medical record checks revealed that fourteen patients were unaware that they had suffered a heart attack or had a revascularization procedure since their original S C A . Therefore, the validity of the risk factor data acquired through the mail-out questionnaire must be viewed with caution. However, while self-reported data may have been 62 judged to be a limitation, it did provide important additional information about C A D treatment. The level of understanding and awareness of respondents could be assessed through comparison to "hard" baseline data. This was particularly true when the patients' perceptions of change were obtained along with a self-report of risk level, as was the case with exercise, weight, and alcohol intake - three variables particularly susceptible to misreport. Patient awareness of health and disease risk is invaluable in the prediction of the success of lifestyle modifications, medication compliance, and positive outcomes. 4.1.2 Response Rate and Bias Every effort was made to maximise response rate. The questionnaire was kept to one page, follow-up phone calls were made to non-responders to encourage response, and a second mail-out was made to those who could not be reached by phone or who stated that they would be wil l ing to complete a new questionnaire. However, despite these efforts, only 42% of those who received a questionnaire responded. While this response rate was less than ideal and endangered the analyses with potential biases, it fell within the range of response rates of 37% to 70% found in similarly designed postal s u r v e y s 1 3 6 ' 1 3 7 ' 1 3 8 . A s anticipated with a low response rate, some biases were found in the data collected. There were significant differences in key baseline characteristics that had to be taken into account when generalising the study findings to wider populations. Significantly more respondents had positive angiographic findings at their initial visit. One possible explanation of this bias may stem from the wording of the consent form. The purpose of the study was described as an assessment of treatment for C A D and dyslipidaemia in British Columbia. This may have misled potential participants with no evidence of C A D that their response was unnecessary, as was borne out by several of the follow-up phone calls. 63 Responders were biased towards non-smoking, which has been seen in other mail-out lifestyle and heart health s u r v e y s 1 3 9 ' 1 3 8 . Smokers may have been unwill ing to respond to a questionnaire assessing lifestyle modification and disease treatment since they themselves had been resistant or unable to stop smoking. It is also possible that individuals who had quit smoking were particularly interested in calling attention to their efforts to modify their risk behaviours. The smaller proportion of non-respondents taking antioxidants at baseline may reflect a lower interest in self-health. This would seem to fit with the higher frequency of current smokers among those who did not wish to participate in the follow-up. However, the absolute number of patients taking antioxidants was very small - 12 non-responders versus 28 responders - and so this difference was likely not clinically relevant. A further bias that appeared in the analysis of the data was the finding that patients living in rural British Columbia were more likely to respond to the questionnaire than urban dwellers (rural: 59%, urban: 48%; p < 0.02). One possible explanation is that individuals l iving outside of major centres may have been more accustomed to remote communication for health care, and were thus more comfortable returning a mail-out questionnaire regarding their health. This bias must be kept in mind in light of the finding that urbanites were less likely to receive appropriate dietician counseling while rural dwellers were less likely to report a low fat diet i f C C C C dietary therapy was indicated. These differences may have been more or less pronounced had there been a greater response rate. 4.2 P A T I E N T C H A R A C T E R I S T I C S 4.2.1 Demographics and R i s k Factor Prevalence Not unexpectedly, the vast majority of individuals in the cohort had positive angiograms and males outnumbered females on the order of three to one. The average age of males and 64 females was similar to other angiographic s tudies 1 4 0 and was in keeping with the observation that women with C A D are predominantly post-menopausal (84% of women). These individuals referred for S C A suffered from a large number of risk factors. Seventy-nine percent (77% female, 79% male) had two or more C A D risk factors, compared to 33% of women and 41% of men in the Canadian Heart Health Surveys 1 1 . Closer examination of the respondent group revealed other similarities and differences in risk factor prevalence compared with published cohorts. Baseline l ipid levels appeared relatively homogenous for the entire cohort, with the only difference being a significantly lower mean L D L - C concentration in respondents who were taking L L D at baseline. Mean T C levels were below 5.2 mmol/L and thus within a desirable range, as were T G (<2.3 mmol/L) and H D L - C (>0.9 mmol/L) levels. However, L D L - C was higher than desired (>3.4mmol/L) and the mean T C / H D L - C ratio was above the normal of five. This would seem reasonable given that this was a cohort of patients being referred for S C A , and these findings correspond to those from a similar published angiographic cohor t 1 4 0 . Comparison with the serum lipid levels of men from the Framingham Heart Study revealed that the men in this cohort had relatively similar L D L - C , H D L - C , and T C levels, along with mean ratio of T C / H D L - C 2 4 . The percentage of current smokers within the non-respondent and deceased groups appeared to be consistent with other study populat ions 3 9 ' 8 7 and with statistics for B C in 1996/1997 8. However, the respondent pool, with only 7% being current smokers, was far below the 25% found in the general population. These differences are difficult to interpret in light of the low response rate. The proportion of British Columbians with high blood pressure in 1996/1997 was 9% 8 . The study cohort was distinct from the general population in that it was made up of individuals 65 with the presence, signs, or potential for C A D . Dividing the cohort into C A D + and C A D -revealed that 33% of S C A negative individuals had hypertension, with 38% of C A D + having a diagnosis of high blood pressure. There was no significant difference between the proportion of C A D + with hypertension in this cohort compared to the LIPID study cohort 8 8 . However, the prevalence of hypertension in this cohort was significantly greater than that found in the A F C A P S / T e x C A P S study for C A D - patients (33% vs 22%; p<0.001). This variation is likely due to discrepancies in selection criteria. Among the noticeable differences, individuals in the A F C A P S / T e x C A P S study had higher l ipid levels at baseline, fewer individuals with a family history of premature C A D , and fewer patients taking anti-hypertensives 8 5. Diabetes was found less frequently among respondents than was been shown in other studies (17%o vs 24%; p<0.05) 1 4 1 . However, the number of C A D + patients with diabetes (18%) was identical to the proportion in a review of over 48,000 U S A patients 1 1 5 . A regional variation was found in the prevalence of diabetes. Patients l iving in urban centres were significantly more likely to have diabetes both at baseline and on follow-up than those l iving rurally. This was despite the fact that the mean B M I of rural patients was significantly higher than urban dwellers, although there was no difference in the number of obese individuals from each region. One speculation may be that patients with diabetes needed to live closer to health facilities for management of their disease. They would require access to dieticians, regular physician monitoring, and pharmacies for medication, which may be more accessible in a city centre. The incidence of obesity in the entire cohort was much greater than the 29%> found in the Canadian National Population Health Survey 8 . A n d yet, a B M I greater than 27 kg/m 2 was used as the guideline for obesity in the 1996/1997 survey - much less stringent than the criterion of > 30 kg/m used for this cohort. Presumably, the proportion of obese subjects would then be even greater in this cohort than the proportion seen in the Canadian population. This may be due to the older age of those referred for S C A , as weight increases with age, particularly in women 2 1 . Indeed, the Canadian National Population Survey was a measure of all individuals between the ages of 15 and 64, while the respondents in this study ranged in age from 37 to 87 at follow-up. Given the association of abdominal obesity with elevated T G , insulin resistance, and increased C A D r i s k 3 3 , the observation that almost half of all individuals in the original cohort had waist circumferences well above desirable measurements (< 88 cm for women, < 102 cm for men) is of great concern. The proportion of patients with T C levels greater than 5.2 mmol/L at baseline (43%) was comparable to that found in the 1985-1990 Canadian Heart Health Surveys 8 . Comparison of abnormal lipid levels in male C A D + patients was possible with a study of 8650 men recruited for the Department of Veterans Affairs H D L Intervention T r i a l 1 4 2 . Similarities and discrepancies are outlined in Table (28). Table 28. Comparison of the prevalence of lipid risk factors found in CAD+ male respondents with the DVA-HIT cohort. Lipid Risk Factor Respondents DVA-HIT Cohort p-value L D L - C > 3.4 mmol/L 53% 56% ns H D L - C < 0.9 mmol/L 45% 38% <0.05 T G > 2.3 mmol/L 24% 33% <0.01 The frequency of predisposing family history of C A D in this cohort was higher than proportions found in retrospective hospital studies of C C U patients (45% vs 36%; p<0.025) 1 1 3 and general admissions (45% vs 33%; p<0.01) 1 4 1 . The proportion of C A D + B C patients with family history was, however, significantly lower than those in the Heart Education and Research Trial (46% vs 74%; p . O . O O l ) 1 1 2 . 67 Very few risk factors changed in the intervening years between the original S C A and the follow-up questionnaire. Two that did - hypertension and diabetes - have been shown to be age-dependent1 ! ' 8 . Thus, the increase over four years was most likely explained simply by the aging of the population. Although the increases may also be a reflection of the high prevalence of obesity in the cohort, a predisposing factor for both conditions. The significant decrease in obesity at follow-up, combined with the observation that only half of the respondents were able to correctly gauge their weight change, may have been due to errors in self-report of weight and waist circumference. Indeed, many studies have shown that obese individuals in particular are prone to under-representation of their body weight 9 7 . 4.2.2 Baseline Subgroup Analyses The availability of risk factor information, l ipid profiles, and angiographic data on a large number of individuals allowed for the analysis of C A D predictors in specific sub-populations. The first study was an attempt to elucidate possible explanations for the occurrence of C A D in patients who were otherwise at very low risk of developing the disease due to their lack of major risk factors. The impetus for the second substudy was that C A D in women, as a whole, has been studied far less than in men, and the variation in incidence between younger and older women may be a result of a different risk profile between the age groups. 4.2.2.1 A p o A I in L o w Risk Patients (This discussion makes up a portion of the paper: Francis M C , Frohlich JJ. Coronary artery disease in patients at low risk - apolipoprotein A I as an independent risk factor. Atherosclerosis. In press . ) 1 3 0 This study examined the role of plasma apoAI concentration as an independent risk factor for C A D in patients with none of the major risk factors. Several previous studies have shown that apoAI is a stronger predictor of atherosclerosis than the traditional plasma l i p i d s 3 7 ' 1 4 3 ' 4 2 . 68 However, no prospective studies have supported a association between apoAI and C A D 1 4 4 ' 3 8 . Thus, current consensus is that apoAI is not a better predictor of atherosclerotic disease than H D L - C . However, this study was unique in that it looked at a select population of individuals in whom the strength of apoAI concentration versus H D L - C levels had not been examined. After univariate analysis several lipid parameters and ratios were predictive of the presence of C A D after adjustment for age and gender. Significantly lower levels of L D L -cholesterol were seen in C A D + patients, which would seem counterintuitive given the established link between high LDL-cholesterol and C A D . However, adjusting L D L - C for FERHDL, a marker for small lipoprotein particles, eliminated its association with C A D . This suggested that the relationship between L D L - C and presence of disease was more likely based on the preponderance of small, dense L D L particles. Multivariate logistic regression analysis revealed that, apart from gender and age, the sole predictive variable for C A D was apoAI concentration. For every increase of 1 mg/dL in apoAI, there was a concomitant 6% decrease in C A D risk. Thus, after taking into account all possible factors that may affect disease presence, it was a reduced level of serum apoAI that distinguished patients with disease from those with normal angiograms. The strong association of apoAI with C A D in low risk patients remained after adjustment for FERHDL, L D L - C , and other potential confounders such as H D L - C and obesity. These findings lend support to the suggestion by Kottke et a l 4 1 that low apoAI concentration may represent a threshold event in the development of C A D , particularly in patients with none of the other traditional risk factors. 4.2.2.2 Predictors of C A D in Younger versus Older Women Studies of C A D and its associated risk factors in women are less prevalent than research on men. The Framingham Heart Study was one of the first to demonstrate that different risk factors carry different weights in women versus men 9 . While the incidence of atherosclerosis is much lower in women 8 , there is a sharp increase in C A D after the age of 6 0 1 4 5 . It seemed reasonable to hypothesise that the change in C A D incidence may be due to a change in relevant risk factors in older women. Univariate analysis of women < 60 revealed strong associations between C A D and both diabetes and smoking. Both of these risk factors have previously been shown to have a greater impact on C A D in women 3 3 . While several l ipid parameters had a significant relationship with C A D when examined alone, in keeping with reports in the l i terature 1 4 6 , adjustment for smoking eliminated the relationships. Multivariate analysis of younger women showed that presence of diabetes and current smoking were the sole variables predictive of C A D . Both of these risk factors showed a strong association with C A D in a Scottish study of women under 6 0 1 4 6 . The risk profile of C A D women > 60 was very different from that of younger women. With regards to lifestyle risk factors, only age was significantly predictive of positive angiographic findings. This was expected given that C A D incidence increases with age in post-menopausal women 8 . Several l ipid parameters were related to C A D after univariate analysis. After adjustment for smoking, to allow comparison between younger and older women, T C (p<0.02), T G (p<0.001), and apoB (p<0.05) were significantly elevated in women with C A D . However, multivariate logistic regression of all possible predictors revealed that only age and T G level were associated with C A D in women > 60. T G levels have previously been linked to C A D in women more so than in m e n 2 9 . However, the association has tended to weaken or disappear upon multivariate analysis 3 3 . Also , none of these studies have looked at C A D risk factors solely in older women. The variation in risk factors predictive of C A D between younger and older women may be due to a survival effect. Young women who smoke or who have diabetes may be more likely to die before reaching 60. Those that survive into their 7th decade may be more susceptible to 70 elevated lipids, particularly T G . However, as this was a cross-sectional study, there is no way to ascertain that there was a longitudinal survival effect of smoking or diabetes on C A D in women. This analysis did suggest that risk factors associated with C A D in women < 60 years versus women over > 60 were different. 4.2.3 M o r b i d i t y Two-fifths of respondents who were referred for S C A had a prior history of either a M I or revascularization procedure. Over the intervening four years, there was an increase in the number of angioplasties and bypass grafts, with a concomitant decrease in the number of M i ' s . The annual rate of hospitalisation for M l in Canada for 1996/1997 was 8.9% 8. Averaged over four years, the rate of M I among respondents was 3.5%, just over a third of the national rate. Fifty-eight (57%) of those who had suffered a M I at baseline went on to have a revascularization procedure, while one-quarter had another heart attack. The hospitalisation rate for C A B G in Canada was 2% in 1996/1997 8. The average annual rate during the follow-up period for respondents was 6%, presumably much higher due to the high-risk nature of the patients, or perhaps the higher proportion of males in the cohort, who are much more likely to have surgical intervention than females 8. The prevalence of M I , P T C A , and C A B G as entered in the follow-up questionnaire must be viewed with care. Comparison to medical records revealed that out of 128 randomly reviewed charts, 14 patients (11%) failed to report that they had had a C A D event since the original S C A . Reasons for patient error are unclear. Those who were unaware of revascularization or M I were no older than the other patients were, and thus age-related factors such as dementia or memory loss were probably not at issue. Possible explanations may be that they were not informed by medical staff in the case of silent M I , or that they misunderstood the surgical procedure that they underwent. It is possible that a P T C A was performed during the 71 course of a planned angiography, and thus the patient may have been unaware or forgotten that the angioplasty was done. The fact that 11% of the respondents were unaware of having suffered a significant C A D event highlights a potential problem in patient communication and health consciousness, and presents an opportunity to enhance patient education. The finding that only angiographic status predicted future morbidity seemed appropriate. There were no differences in predictors of each morbidity event taken alone. Presence of C A D should have had the greatest association with C A D morbidity, overpowering any other associated factors. 4.2.4 Mor ta l i t y One hundred and fifty patients died between 1993 and 1999, indicating an overall mortality rate of 2.3% per year. With cause of death available for 102 individuals, death due to cardiovascular disease was established in 71 patients. Thus, the mortality rate from C V D in this cohort was estimated at 1.1% per year. The mortality rate from C V D in Canada was 0.5%» for 1997 8. Given that the patients in this cohort were selected for either presence of C A D or suspicion due to a number of risk factors, they were more prone to die from heart disease. Medical treatment of patients in the deceased groups differed significantly from respondents. A surprising finding was that deceased individuals were significantly more likely to have been prescribed A C E inhibitors at baseline. A C E inhibitors have been indicated for patients with renal insuff iciency 1 4 7 , and there were significantly more deceased patients with this condition. However, the numbers are too small to determine whether this would explain the difference. The H O P E t r i a l 1 4 8 has shown that the A C E inhibitor Ramipril had broad-ranging beneficial effects beyond merely hypertension control. A C E inhibitors are recommended for treatment of diabetes, as they reduce the incidence of diabetic neuropathy and other complications. Significantly more patients with diabetes among the deceased group were 72 receiving A C E inhibitors at baseline than others in the cohort. Also , A C E inhibitors were shown to be of great benefit to patients with congestive heart failure (CHF) . Although there was no data on the prevalence of C H F in the cohort, it may be reasonable to assume that a greater number of individuals who died may have suffered from the condition. Multivariate analysis revealed that the predictors of all-cause mortality and cardiovascular mortality were quite similar. With increasing age, the chance of death increased. Mortality was significantly affected by the lifestyle habits of exercise and smoking. All-cause mortality was 44% less likely for those who exercised occasionally while cardiovascular death was reduced by 59%. While there was no effect with increased exercise intensity, the number of individuals partaking in moderate and heavy exercise was much smaller. Current smoking or any history of smoking cigarettes increased the chance of overall mortality by 77% and cardiovascular mortality by almost three-fold. A n increase by one unit in the T C / H D L - C ratio increased the likelihood of C V D mortality by 21% and all-cause mortality by 10% There was some disagreement in predictors of total versus C V D mortality. For each 1 mmol/L increase in T G concentration, there was an associated 77% drop in C V D mortality, although T G did not predict all-cause mortality. This is an intriguing finding given that, first of all , the majority of studies in the literature have failed to show an independent relationship between C V D death and T G upon multivariate analysis 2 6 . Secondly, the mean T G concentration among those who died of cardiovascular causes was 1.49 mmol/L compared to 1.89 mmol/L for others in the cohort. T G was significantly lower for those dying of C V D (p<0.05; based on the logarithmic transformation due to skewed distribution). This finding was contrary to all other reports in the literature which, when an association between T G and C A D mortality has been seen, T G levels were elevated 2 6 . The reduction in T G at baseline, prior to death from C V D , may have been an indicator of an insidious process that resulted in the depletion of TG-containing 73 lipoproteins from the blood. It may have been the result of a low carbohydrate 1 4 9 or c a lo r i e 1 5 0 diet, which has been seen in many elderly individuals. A study by Wei et a l 1 5 1 demonstrated that low fasting plasma glucose (FPG) was independently associated with C V D mortality. They also showed that patients with low F P G had significantly lower T G levels. It may be possible that the lower T G concentration among patients in this cohort who died of cardiovascular causes may be a reflection of low F P G levels. It should be stressed that the absolute levels of T G were low among the cohort and were, for the most part, within C C C C guidelines. Overall mortality was further associated with C A B G and prescription of L L D at baseline. Both variables had the expected effect on mortality - patients who had undergone bypass had a two and a half fold higher risk of death, while L L D administration at baseline reduced the chance of death by 56%. Why a history of C A B G would be associated with overall mortality but not C V D mortality is uncertain. It may be due to the fact that the other predictors of C V D death conveyed higher risks that negated the effect of bypass surgery, or that the number of respondents who had undergone a C A B G prior to S C A was quite small. Use of L L D may only have been associated with overall death in part because of the small absolute numbers taking the medications. There may not have been enough patients on L L D to show a significant effect. 4.3 ANALYSIS O F T R E A T M E N T 4.3.1 Awareness and Counseling of Risk Factors A s smoking is a major risk factor for C A D , smoking cessation counseling should be a priority in the prevention of cardiovascular morbidity and mortality. Unfortunately, due to the significant response bias towards non-smokers returning the questionnaire, numbers for respondents were too small for any meaningful analysis. It was discovered that nearly three-quarters of patients who smoked at the time of their S C A were subsequently counseled about the 74 health risks of smoking. However, with so few respondents in this group (22), assessment of the effectiveness of the counseling on smoking cessation cannot be performed. Arguably, the first indicator of appropriate lifestyle management of C A D is the level of awareness the patient has regarding a particular risk factor. When asked about their blood pressure, just over half of the respondents were able to provide it. This is considerably lower than the 73% of respondents who knew their blood pressure in a community screening project of heart disease r i s k 1 0 8 . O f the 115 (37%) who stated that they had been diagnosed with hypertension, only 71% knew their blood pressure. This demonstrates less than optimal awareness of a particularly potent C A D risk factor among respondents. Additionally, only 71% of those who reported taking medications for high blood pressure stated that they had hypertension. If over one-quarter of the respondents taking anti-hypertensives did not understand that they suffer this medical condition, this may hamper medication compliance, which is particularly difficult in asymptomatic long-term conditions such as hypertension 1 5 2 . Counseling about the dangers of high blood pressure was relatively low for all respondents as a group (61%) but appeared to be well-targeted to those who had been diagnosed with hypertension at the original assessment (80%). Canadian guide l ines 1 2 7 for blood pressure management recommend weight loss, particularly in obese patients. Seventy percent of respondents who were obese and hypertensive (64) at baseline received counseling on diet and weight, with 64% receiving exercise counseling. Obesity was much more common among respondents than has been seen in the general population 8. Compounding this problem was the observation that only half of the patients were able to correctly gauge their weight change over four years. And , as expected based on previous reports 9 8 , abdominally obese individuals were significantly worse at judging their weight change 75 than non-obese individuals. Thus, patient awareness of this risk factor was disappointingly low among respondents. Treatment of obesity may be accomplished through several different approaches. Counseling patients to adopt a low fat diet, as a means of weight reduction, is important. Encouraging regular physical activity may also promote weight loss. Diet and weight loss were reportedly discussed with 68% of obese patients, and 60%> were counseled on increasing their exercise level. The prevalence of counseling in this high-risk population appeared to be sub-optimal. Treatment of risk factors such as obesity though counseling and education is widely believed to be an efficacious and cost-effective method of managing C A D 1 5 3 . While there was a significant correlation between risk factor counseling, particularly by experts in the field of diet and exercise, and changes in risk behaviour, the correlation was quantitatively small. A n d the effect on reducing obesity was moderate at best. Only 63% of obese respondents counseled on their diet and weight and 55%> of those who received exercise counseling reported a decrease in weight compared to baseline. Greater involvement of dieticians and exercise instructors in risk factor counseling may potentially improve the management of obesity. 4.3.2 Pharmacological Interventions A number of medications have been shown to reduce C A D morbidity and mortality as a peripheral benefit. Aspir in has been found to be particularly important in reducing events 6 2 and is recommended for patients without a risk for gastrointestinal bleeding. Aspir in use among respondents increased significantly over the four year follow-up period, and was comparable to 1998 rates of post-MI A S A prescription in Canada 8 . However, regular aspirin use among C A D + patients was significantly lower than rates found in both L I P I D 8 8 and C A R E 8 7 studies. Given 76 that meta-analyses of primary and secondary prevention trials have shown that risk reduction is greater in patients with established C A D 6 3 , targeting to C A D + individuals should be improved. Between 1993 and 1997, the use of antioxidants increased dramatically. Despite no conclusive evidence from randomised trials that antioxidants reduce the incidence of C A D and associated complicat ions 1 5 4 , many more respondents reported taking these supplements on the questionnaire. Self-medication with vitamins and other supplements has risen sharply over recent years, with Canadians spending almost four bill ion dollars per year on these medicine alternatives8. The desire to take personal control over health may explain the increase, as may the wish to use "natural" ways to treat medical problems. Prevalence of drug treatment for hypertension did not appear to change over the four years, with medications prescribed to 92% of hypertensives at baseline and 92% of self-reported follow-up hypertensives. The type of medication used to manage hypertension did appear to shift from the early-90's to the late-90's. The prevalence of drugs used specifically for blood pressure control at the time of S C A and as reported on the questionnaire is given in Table (29). Table 29. Comparison of type of anti-hypertensive reported at follow-up with baseline use. Medication Baseline Follow-up p-value Beta-blocker 53% 44% ns A C E inhibitor 30% 52% <0.001 Ca-channel blocker 50% 26% <0.001 It is apparent that prior to 1993, the majority of hypertensive patients were treated with either beta-blockers or Ca-channel blockers. However after 1994, a preference towards A C E inhibitors emerged to the deference of Ca-channel blockers. A C E inhibitors have been touted as the newest generation of drug for optimal management of hypertension 1 4 8 , and this may explain the increased use of this class of blood pressure medication. 77 A total of 111 respondents reported taking beta-blockers on the questionnaire. Over half of these said they were prescribed the drug for reasons other than hypertension. This was equivalent to 20% of all respondents, and represented a significant decrease compared to baseline adjuvant beta-blocker use (30%; p<0.025). One possible reason for the decline may have been the reduction in M I frequency in the four-year follow-up period compared to pre-S C A , given that post-MI patients have been shown to derive the greatest benefit from beta-blocker therapy 7 3 . There is also the possibility, because this data was self-reported, that respondents were not aware that they were taking beta-blockers, or that they were intended for C A D risk reduction rather than blood pressure control. O f note, 26 patients were not able to provide the type of heart drug they were taking, and so the use of beta-blockers for C A D prevention may have been greater at follow-up than reported. 4.3.3 Appropriate Treatment of Dyslipidaemia The relationship between dyslipidaemia and C A D has been well established in the scientific literature and widely disseminated in the popular media. Public education campaigns such as the American Heart Association's "Cholesterol L o w Down" program and the Canadian Heart and Stroke Foundation's "HeartSmart Women" have stressed the benefit o f not only maintaining a low cholesterol level but also the importance of simply knowing one's cholesterol level. However, only 41%> of respondents were able to give their T C level, despite over three-quarters of them reporting to have had their cholesterol measured within the last four years and 71%» stating that they were counseled about their cholesterol level. Reasons for this lack of personal awareness may be attributable to the uncertainty that many patients may have felt about cholesterol - particularly the confusion between "good" cholesterol versus "bad" cholesterol 1 1 9 . Two primary sets of guidelines for the treatment of dyslipidaemia were available to physicians in 1993. The C C C C guidelines were published in C M A J in 1988 9 3 , while the second 78 incarnation of the N C E P guidelines was released in J A M A in 1993 2 3 . The C C C C guidelines reflected the Canadian policy of dyslipidaemia treatment, and thus should have been most relevant to physicians treating patients in this cohort. However, the ubiquitous influence of American medicine on Canadian practices must also be considered. References to the N C E P cholesterol limits were found in several S P H and V G H hospital charts. Thus, analysis of appropriate treatment among respondents was performed using both sets of guidelines. The prevalence of appropriate dyslipidaemia management according to the 2000 Canadian guidelines has been included for comparison, but these guidelines cannot be used to judge the treatment practices of British Columbian physicians between 1993 and 1997. 4.3.3.1 Cholesterol Screening The C C C C guidelines 9 3 recommended lipid testing for all individuals with a high perceived risk of C A D or prior evidence of the disease. Based on these criteria, all but four of the respondents should have had lipids assessed. The N C E P guidelines 2 3 used a combination of T C and H D L - C level with the categorical number of risk factors. On this basis, only 91% of patients should have had annual screening of cholesterol levels. B y comparison, Canadian 2000 cri ter ia 9 5 for cholesterol screening were slightly altered from the 1988 C C C C guidelines. A n age limit was introduced so that all men over 40 and women over 50 were to be screened. Rather than considering a list o f individual risk factors for screening, major risk factors were grouped, with the presence of two or more necessitating l ipid assessment. Also , clinical evidence not only of C A D , but also peripheral vascular disease or carotid atherosclerosis was an indication for fasting l ipid screening. These criteria were met by all but one respondent. Despite the wide variation in guidelines for cholesterol screening, there was no difference in the number of respondents who were screened appropriately. Approximately three-quarters of those requiring l ipid assessment stated that their cholesterol had been measured within the previous four years. O f note, the N C E P guidelines for frequency of testing were much stricter than the C C C C . L i p i d profiles were required every year according to American recommendations. However, due to the wording of the questionnaire, the appropriate frequency of testing could not be ascertained. No program for repeated measures was given in the 1988 Canadian guidel ines 1 5 5 . Nor was there a stipulation for regular cholesterol screening in the recent 2000 Canadian guidelines - the frequency was left to clinical judgement 9 5 . The predictors of appropriate cholesterol screening were identical for both the C C C C and N C E P guidelines. The older the respondents were, the less likely they were to have had their cholesterol checked. Angiographic evidence of disease increased the chances of having cholesterol appropriately measured by approximately three-fold. Unexpectedly, for every 1 mmol/L increase in baseline H D L - C , there was a seven- to eight-fold greater likelihood that patients would have their lipids screened within the follow-up period. It is unclear why the presence of a negative risk factor for C A D would elicit greater vigilance on the part of physicians. Higher H D L - C , which would have been reflected in an elevated T C level, may have served as a risk marker to physicians who considered only T C in l ipid risk assessment. Conversely, patients with elevated H D L - C levels, which have been associated with such positive health behaviours as increased exercise 1 5 6 , may have taken a more active interest in their health, and may have requested closer monitoring of l ipid levels by their physicians. 4.3.3.2 Dietary Therapy A l l guidelines for the management of dyslipidaemia recommended diet as the cornerstone of therapy 1 5 5 . However, only the C C C C guidelines were explicit in the requirement for dietician intervention in the promotion of a suitable low fat diet. Policy makers were concerned about the accessibility of dieticians within the health management team 9 3 , and it would appear this was with good reason. O f those respondents requiring intensive dietary therapy according to C C C C 80 guidelines, only 38% reported being counseled by a dietician. This also varied regionally, with urban dwellers significantly less likely to have seen a dietician than their rural counterparts. Analysis of predictors o f dietician counseling revealed that patients with diabetes were four times more likely to have been counseled by a dietician as part of their recommended treatment. This was not unexpected given the necessity of dietician intervention for optimal nutritional management of adult-onset diabetes. In addition, respondents l iving in city centres were 61% less likely to be appropriately counseled by a dietician, in keeping with the previously described regional variation in dietician accessibility. Despite the sub-optimal use of allied health professionals, 63% of patients reported receiving dietary counseling and 73% stated that they were on a low fat diet. While encouraging, these findings should be viewed with caution. The type of diet was self-reported, and as previously shown in the li terature 1 0 1 , individuals, particularly obese individuals, tend to underestimate their dietary fat intake. Separation of respondents with high T C levels from those with borderline T C in combination with abnormalities in other l ipid measures revealed a disparity in the prevalence of patients on a low fat diet. While there was no difference in the frequency of dietary counseling between these two groups, those with borderline high T C were significantly less likely to report being on a low fat diet than those with T C above 6.2 mmol/L. A study involving focus group assessment of patient attitudes towards high cholesterol indicated that there was substantial confusion over "how high is too high" with regards to cholesterol, and a lack of confidence in the effect diet had on lowering cholesterol 1 1 9 . These concerns may explain the lower adherence to a low fat diet in respondents with borderline T C . The regional difference in dietician counseling was further confounded by a paradoxical variation in low fat diet prevalence among rural and urban respondents. Although patients l iving 81 in city centres were less likely to have seen a dietician, they were more likely to report that they were on a low fat diet. This may be an indication that dietician intervention was more effective for urbanites despite its limited use. Or perhaps patients l iving in a rural setting were successfully counseled to improve their dietary intake of fat, but were aware that the change was not enough to be considered a low fat diet. In effect, rural individuals may have started off with a poorer diet, and despite consulting a dietician, were not able to adequately reduce their fat intake. Unfortunately, there was no way to assess the actual dietary intake of respondents, and all that was available was the self-reported categorisation of their diet. N C E P guidelines have stricter criteria for dietary management of dyslipidaemia. A s such, more respondents would have required dietary therapy under the American guidelines (228 vs 135; p<0.001). The Canadian 2000 recommendations are even more rigorous than either of the older guidelines. Using the 2000 criteria for dietary therapy, almost 90% of respondents would have required implementation of a low fat diet to control their dyslipidaemia. In spite of these differences, the percentage of patients reporting they were on a low fat diet was nearly identical under all three guidelines, in the range of 70%>-73%. This figure appeared to fall within the range of literature studies on adherence to guidelines. McBr ide et a l 1 1 2 showed that only 55%) of patients requiring dietary therapy under N C E P guidelines received dietary counseling. A t the other end of the spectrum, 79% of patients requiring dietary intervention according to N C E P guidelines in a US-wide survey stated that they had been placed on a low fat diet by their phys i c i an 1 1 1 . While the percentage of respondents on a required low fat diet according to C C C C and N C E P guidelines was equal, the variables associated with appropriate dietary therapy were not all the same. Both obesity and history of P T C A prior to S C A were significantly associated with a reduced likelihood of being on a low fat diet. This was the expected effect of obesity on 82 appropriate diet; or rather it was logical that obese individuals would be found less likely to subscribe to a low fat diet. However, it seemed counterintuitive that patients undergoing revascularization would not make modifications to their diet in order to decrease their risk of C A D morbidity and mortality. Some studies have shown that post -PTCA patients are reluctant to change their lifestyle following the procedure, and many find the switch to a low fat diet drastic and an impediment to a satisfactory quality of l i f e 1 5 7 . Other predictors of appropriate dietary therapy according to C C C C guidelines included T C level and regionalisation. Every 1 mmol/L increase in T C was associated with a nearly five-fold higher likelihood of reporting a low fat diet. Rural dwellers were nearly three times less likely to be consuming a diet low in fat, once again reflecting the regional disparity in dietary fat intake. Lifestyle risk factors appeared to have stronger effect on NCEP-based dietary, therapy. For individuals who consumed moderate (1-2 drinks per day) and heavier (> 2 drinks per day) quantities of alcohol, the chance of eating low fat foods decreased by 75% to 90% respectively. Perhaps there was a perceived correlation between regular drinking and high fat intake among respondents requiring dyslipidaemia management. Patients who reported some level of physical activity were five times more likely to eat less fat. This may indicate that advice on weight loss and risk reduction emphasised not only healthier eating habits, but also the need for increased exercise. Or perhaps patients who made exercise a part of their lifestyle were more health-conscious, and thus more likely to reduce their dietary fat intake. The use of A S A by those requiring dietary therapy according to American recommendations was also associated with low fat diet, perhaps also indicative of greater personal interest in health. 83 4.3.3.3 Drug Therapy One important difference between the C C C C and N C E P guidelines, and a limitation to this study, was the target for cholesterol lowering. C C C C guidelines used a T C < 5.2 mmol/L as the desirable goal for dietary therapy. If this level was not reached after six months, then pharmacologic intervention was indicated 9 3 . N C E P guidelines, on the other hand, used L D L - C to determine the need for continued treatment. The L D L - C cut-off differed depending upon the presence of C A D and number of additional risk factors. I f L D L - C remained above the limit for drug therapy after an adequate trial of dietary therapy, then L L D s may need to be prescribed 2 3 . Self-reported T C levels were available for only 67 of the respondents who required dietary therapy under C C C C guidelines. The need for subsequent drug treatment was assumed for those whose follow-up T C did not meet the goal (n=32) and for those who did not provide a recent T C measurement (n=69). While almost 80% of those requiring drug therapy were counseled on their cholesterol levels, only 55% reported being prescribed a L L D at follow-up. There was also a significant difference in counseling and L L D prescription between respondents with baseline T C > 6.2 mmol/L and those with T C between 5.2 and 6.2 mmol/L. Three-quarters of those with highly elevated T C were appropriately receiving L L D , but only 45% of patients with borderline T C were on drug therapy. Literature reports have indicated that physicians tend to withhold medication for dyslipidaemia treatment until cholesterol levels are well above guideline cut-offs, and prefer to employ only dietary therapy for those with marginally elevated serum l i p i d s 1 5 8 . Doctors may have been concerned about drug compliance or medicalisation of patients in whom they perceive less risk of C A D morbidity and mortality. Also , for the sake of simplicity in the face of innumerable treatment guidelines, family physicians may have wished to use only T C as the criterion for drug therapy, rather than having to screen for other lipid parameters such as L D L - C and H D L - C 1 5 9 . 84 For those respondents requiring drug therapy according to C C C C guidelines who reported meeting T C targets at follow-up (n=35), over 80% were prescribed L L D s . It is unclear as to whether these individuals were more appropriately treated or whether this finding was merely an indication that L L D s were efficacious in lowering T C to the desired goal. A s with dietary therapy, the number of patients who should have been prescribed L L D s according to the 2000 Canadian guidelines was vastly greater than was borne out by the Canadian guidelines of the time (252 vs 101; p<0.001). The prevalence of respondents who were appropriately managed, however, was once again comparable between the two. Follow-up L D L - C levels were not available for respondents, and so target achievement could not be assessed according to N C E P guidelines. Therefore, all respondents who required dietary therapy were assumed to have required subsequent drug therapy i f their L D L - C levels were above recommended levels. This assumption was less than ideal, as some of these patients may have met L D L - C targets and not required L L D s . Hopefully, however, this was balanced by those whose L D L - C levels may have risen above drug target levels following S C A , but who were not included in the analysis due to the lack of recent l ipid levels. A greater number of respondents required medication management of their dyslipidaemia based on N C E P guidelines compared to C C C C guidelines (153 vs 101). Although it would appear that a greater proportion of respondents was treated appropriately with L L D under the American guidelines (62% vs 55%>), this difference did not achieve significance. There have been no studies looking directly at adherence to C C C C drug therapy guidelines. However, the C Q I N investigators did observe that only 8%> of high risk patients in an acute care setting received drug intervention for dysl ip idaemia 1 0 9 . While 55% is certainly a vast improvement over this finding, L L D treatment among respondents was less than optimal. Previous studies on adherence to N C E P guidelines have shown that appropriate drug therapy has 85 been administered to anywhere from 37% to 52% of pa t i en t s 1 6 0 ' 1 1 3 . The prevalence of appropriate drug therapy in this cohort of respondents appeared to be slightly better than the literature. However, 38% of patients remained inappropriately treated according to N C E P guidelines. This was sub-optimal and increasing efforts should be made to comply with evidence-based recommendations for dyslipidaemia management. Multivariate models of appropriate drug therapy predictors were somewhat unstable, primarily due to small numbers. However, differences in parameters associated with C C C C appropriate treatment versus N C E P appropriate treatment did emerge. Whether a respondent received L L D in accordance to Canadian guidelines was based solely on the prescription of L L D at baseline, and the initial T C level. While the effect of baseline l ipid medications could not be estimated using logistic regression, an increase in T C of 1 mmol/L improved the chances for proper L L D administration by over three-fold. The predictors of appropriate drug therapy based on N C E P guidelines were more varied and more variable. Having undergone P T C A prior to S C A increased the chances of receiving L L D by almost 14 times, while suffering a M I lowered the chances by almost 80%. That patients with a history of M I would be less likely to be treated appropriately with medications was distressing. Post-MI patients are at high risk for further C A D events, and the Number-Needed-to-Treat (NNT) has been shown to be lower than for patients with no history of atherosclerotic disease 1 6 1 . A possible explanation for this finding may be that angioplasties would have been performed at a tertiary care hospital under the care of cardiologists. On the other hand, diagnosis and treatment of M I may have occurred at a non-tertiary hospital with limited specialist resources. Prescription of L L D s following M I may have been less likely under these circumstances. In addition, studies have shown that l ipid levels decrease following an acute event, such as a M I . Although there was no record of the timing of M i ' s at baseline, i f an event had occurred within 3-6 months of the original S C A , the l ipid levels may have been falsely low; potentially below guideline thresholds for L L D treatment. The observation that respondents with higher L D L - C levels at baseline were less likely to be taking L L D at follow-up was equally disheartening. However, the large confidence intervals for many of the variables may denounce the model as a poor fit for the data. Other lipids associated with appropriate N C E P drug therapy were H D L - C concentration and T C / H D L - C ratio. The effect of H D L - C could not be estimated. However, an observed increase of one unit in T C / H D L - C ratio predicted a 14-fold increase in appropriate L L D use. Finally, baseline prevalence of L L D was highly associated with suitable medication therapy at follow-up. There are striking differences in the selection criteria for treatment between the three sets of guidelines. Direct comparisons of the ability of C C C C guidelines, N C E P guidelines, and a Framingham multivariate risk equation (a precursor of the Canadian 2000 guidelines) by Grover et a l 1 6 2 revealed that the N C E P guidelines were slightly better than C C C C guidelines in discriminating overall C H D risk, while the model based on the Framingham Heart Study showed the greatest ability to predict coronary deaths. However, regardless of number of respondents allocated to treatment groups, there was no difference in the prevalence of appropriate dyslipidaemia treatment using any of the guidelines. A n d thus it remains that the education of physicians on appropriate management of C A D risk factors along with the promotion of awareness and compliance among patients at risk is of utmost importance for the prevention of cardiovascular disease and death. 4.4 I M P A C T O N C L I N I C A L P R A C T I C E S The biases created by the low response rate weaken the applicability of the study findings to broader populations. Future studies using a mail-out questionnaire should make every attempt 87 to maximise the rate of response. Face-to-face and telephone interviews have been shown to have higher response rates, although they require considerably more time and money to carry o u t 1 6 3 . While obtaining self-reported data allowed for the assessment of patient awareness of C A D risk factors, the accuracy of information about medication usage, morbidity, and diet must be viewed with a degree of suspicion. Future researchers may wish to have the patients fill out the questionnaire with their physician or cardiologist. Money could be saved and data made more reliable by simply conducting a retrospective chart review. However, hospital charts often lack data about lifestyle risk factors, and patient awareness can not be assessed. This study was unique both in the population that was examined and the results that were obtained. Although there were limitations associated with a less-than-ideal response rate and self-reported data, the findings from this research can be used to inform C A D and dyslipidaemia treatment practices in British Columbia in the context of both primary and secondary prevention. Opportunities for lipid screening were missed in one-quarter of respondents. The use of dietary therapy, regardless of the guidelines used to judge the need for intervention, was appropriate in about 70% of respondents. While this was on par with literature reports, treatment of dyslipidaemia through diet must be improved in order to decrease health costs to society. Administration of L L D according to published recommendations was certainly sub-optimal. Not even two-thirds of patients who returned the questionnaire were receiving appropriate drug therapy. Although the Canadian 2000 guidelines could not be used to judge the treatment practices of physicians between 1993 and 1997, they most certainly provide a fresh opportunity to improve current practice. Using the criteria outlined for dietary and drug therapy, a significant number of British Columbians referred for S C A would require intervention to manage their dyslipidaemia. Physicians' compliance to guidelines may have been poor due to confusion or 88 misinformation regarding diagnostic and treatment recommendations. The simplified risk assessment algorithm of the recent Canadian guidelines may improve their ability to target patients for appropriate treatment. Along with guidelines for dyslipidaemia, the need to manage other C A D risk factors must be emphasised. This study revealed a disappointing lack of patient awareness regarding obesity, physical activity, high blood pressure, cholesterol, and C A D morbidity. Patient understanding of health behaviours and risk factors should be improved. In this way, increased compliance with medication and lifestyle modification, along with a reduction in morbidity and mortality may be achieved. The consistent use of "experts" to counsel patients on behaviour changes should be encouraged. Dieticians, exercise instructors, and smoking cessation nurses have an important role to play in educating patients about health risks, and providing them with new strategies to ensure that alteration of diet and exercise habits remains a lifelong priority. The treatment of C A D and dyslipidaemia in study respondents was not ideal. These results are from two university-affiliated tertiary care hospitals. Presumably, the appropriate management of patients may be worse at other hospitals without the support of highly trained specialists and access to the latest research. There have been changes in practice since this cohort was examined. St. Paul's Hospital now has a drug therapy discharge sheet (Appendix 3). Medications that have a significant impact on lowering C A D morbidity and mortality must be considered for all patients following M I or revascularization procedures, with reasons given i f the drug is not prescribed. C A D is the leading cause of death and disability in British Columbia and Canada. Many years of research have contributed to the understanding of risk factors and treatments for morbidity and mortality due to heart disease. Appropriate primary and secondary prevention of 89 C A D according to evidence-based guidelines is imperative in order to reduce health costs and improve the quality and length of life of British Columbians. 90 5 REFERENCES 1. Hegele R. The pathogenesis of atherosclerosis. C l i n Chimica Acta, 1996; 246: 21-38. 2. Ross R. The pathogenesis of atherosclerosis: a perspective for the 1990s. Nature, 1993; 362: 801-809. 3. Basha B and Sowers J. Atherosclerosis: A n update. A m Heart J, 1995; 131(6): 1192-1202. 4. Berk B , Weintraub W , and Alexander R. Elevation of C-reactive protein in "active" coronary artery disease. A m J Cardiol, 1990; 65: 168-172. 5. Ridker P, Cushman M , Stampfer M , Tracy R, and Hennekens C. Inflammation, aspirin, and the risk of cardiovascular disease in apparently healthy men. N Engl J Med, 1997; 336(14): 973-979. 6. Fuster V , Badimon L , Badimon J, and Chesebro J. The pathogenesis of coronary artery disease and the acute coronary syndromes. N Engl J Med, 1992; 326(5): 310-318. 7. Libby P, Aikawa M , Kinlay S, Selwyn A , and Ganz P. L i p i d lowering improves endothelial functions. Int J Cardiol, 2000; 74(Suppl 1): S3-S10. 8. Heart and Stroke Foundation of Canada. The changing face of heart disease and stroke in Canada 2000. 1999, Ottawa: Laboratory Centre for Disease Control, Health Canada, Statistics Canada, Canadian Institute for Health Information, Canadian Cardiovascular Society, Canadian Stroke Society, Heart and Stroke Foundation of Canada. 9. Kannel W . Contributions of the Framingham study to the conquest of coronary artery disease. A m J Cardiol, 1988; 62: 1109-1112. 10. Grundy S, Pasternak R, Greenland P, Smith Jr. S, and Fuster V . Assessment of cardiovascular risk by use of multiple-risk-factor assessment equations: A statement for healthcare professionals from the American Heart Association and the American College of Cardiology. Circulation, 1999; 100: 1481-1492. 11. Joffres M , Ghadirian P, Fodor J, Petrasovits A , Chockalingam A , and Hamet P. Awareness, treatment, and control of hypertension in Canada. A m J Hypertens, 1997; 10: 1097-1102. 12. Bairey-Merz C, Rozanski A , and Forrester J. The secondary prevention of coronary artery disease. A m J Med, 1997; 102(6): 572-581. 13. MacMahon S, Peto R, Cutler J, Collins R, Sorlie P, Neaton J, Abbott R, Godwin J, Dyer A , and Stamler J. Blood pressure, stroke, and coronary heart disease: Part 91 1, prolonged differences in blood pressure: prospective observational studies corrected for the regression dilution bias. Lancet, 1990; 335: 765-774. 14. Kannel W , McGee D , and Gordon T. A general cardiovascular risk profile: The Framingham Study. A m J Cardiol, 1976; 38: 46-51. 15. Haffner S, Lehto S, Ronnemaa T, Pyorala K , and Laakso M . Mortality from coronary heart disease in subjects with type 2 diabetes and in nondiabetic subjects with and without prior myocardial infarction. N Engl J Med , 1998; 339(4): 229-234. 16. Kannel W , Doyle J, Ostfeld A , Jenkins C, Kul ler L , Podell R, and Stamler J. Optimal resources for primary prevention of atherosclerotic diseases. Circulation, 1984; 70: 157A-205A. 17. Kannel W and Sorlie P. Some health benefits of physical activity: The Framingham Study. Arch Intern Med, 1979; 139: 857-861. 18. Paffenbarger R, Hyde R, Wing A , and Hsieh C. Physical activity, all-cause mortality, and longevity of college alumni. N Engl J Med , 1986; 314(10): 605-613. 19. Vega G and Sirtori C. Dyslipidemia and coronary heart disease. Curr Opin Lip ido l , 1996; 7: 179-182. 20. Corti M , Guralnik J, Salive M , Harris T, Ferrucci L , Glynn R, and Havl ik R. Clarifying the direct relation between total cholesterol levels and death from coronary heart disease in older persons. A n n Int M e d , 1997; 126(10): 753-760. 21. Dawber T, The Framingham Study: The Epidemiology of Atherosclerotic Disease. 1980, Cambridge: Harvard University Press. 22. Kannel W , Castelli W , and Gordon T. Cholesterol in the prediction of atherosclerotic disease: New perspectives based on the Framingham Study. A n n Int Med , 1979; 90: 85-91. 23. Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Summary of the second report on the National Cholesterol Education Program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel II). J A M A , 1993; 269(23): 3015-3023. 24. Kannel W . Range of serum cholesterol values in the population developing coronary artery disease. A m J Cardiol, 1995; 76: 69C-77C. 25. Chaimvimonvat V and Sternberg L . Coronary artery disease in women. Can Fam Physician, 1998; 44: 2709-2717. 26. Austin M . Plasma triglyceride and coronary heart disease. Arterioscl Thromb, 1991; 11(1): 2-14. 92 27. Stampfer M , Krauss R, Jing M , Blanche P, H o l l L , Sacks F , and Hennekens C. A prospective study of triglyceride level, low-density lipoprotein particle diameter, and risk of myocardial infarction. J A M A , 1996; 276(11): 882-888. 28. N I H Consensus Development Panel on Triglyceride, High-Density Lipoprotein, and Coronary Heart Disease. Triglyceride, high-density lipoprotein, and coronary heart disease. J A M A , 1993; 269(4): 505-510. 29. Assmann G and Schulte H . Relation of high-density lipoprotein cholesterol and triglycerides to incidence of atherosclerotic coronary artery disease (the P R O C A M experience). A m J Cardiol, 1992; 70: 733-737. 30. Ohta T, Saku K , Takata K , Nagata N , Maung K , and Matsuda I. Fractional esterification rate of cholesterol in high density lipoprotein ( H D L ) can predict the particle size of low density lipoprotein and H D L in patients with coronary heart disease. Atherosclerosis, 1997; 135: 205-212. 31. Dobiasova M , Stribrna J, Sparks D , Pritchard P, and Frohlich J. Cholesterol esterification rates in very low density lipoprotein-and low density lipoprotein-depleted plasma. Relation to high density lipoprotein subspecies, sex, hyperlipidemia, and coronary artery disease. Arterioscl Thromb, 1991; 11: 64-70. 32. Moghadasian M , McManus B , and Frohlich J. Homocysteine and coronary artery disease. Arch Intern Med, 1997; 157(10): 2299-2308. 33. Gensini G , Comeglio M , and Colella A . Classical risk factors and emerging elements in the risk profile for coronary artery disease. Eur Heart J, 1998; 19(Suppl A ) : A53-A61 . 34. Hegele R. Lipoprotein (a): A n emerging risk factor for atherosclerosis. Can J Cardiol, 1989; 5(5): 263-265. 35. Ridker P, Hennekens C, and Stampfer M . A prospective study of lipoprotein(a) and the risk of myocardial infarction. J A M A , 1993; 270(18): 2195-2199. 36. Toss H , Lindahl B , Siegbahn A , and Wallentin L . Prognostic influence of increased fibrinogen and C-reactive protein levels in unstable coronary artery disease. Circulation, 1997; 96(12): 4204-4210. 37. Kwiterovich P, Coresh J, Smith H , Bachorik P, Derby C , and Pearson T. Comparison of the plasma levels of apolipoproteins B and A - l , and other risk factors in men and women with premature coronary artery disease. A m J Cardiol, 1992; 69:1015-1021. 38. Lamarche B , Moorjani S, Lupien P, Cantin B , Bernard P, Dagenais G , and Despres J. Apolipoprotein A - I and B levels and the risk of ischemic heart disease during a five-year follow-up of men in the Quebec Cardiovascular Study. Circulation, 1996; 94: 273-278. 93 39. Brown G , Albers J, Fisher L , Schaefer S, L i n J, Kaplan C , Zhao X , Bisson B , Fitzpatrick V , and Dodge H . Regression of coronary artery disease as a result of intensive lipid-lowering therapy in men with high levels of apolipoprotein B . N Engl J Med , 1990; 323(19): 1289-1298. 40. Vega G and Grundy S. Comparison of apolipoprotein B to cholesterol in low density lipoproteins of patients with coronary heart disease. J L ip id Res, 1984; 25: 580-592. 41. Kottke B , Zinsmeister A , Holmes Jr. D R , Kneller R, Hallaway B , and Mao S. Apolipoproteins and coronary artery disease. Mayo C l i n Proc, 1986; 61: 313-320. 42. Maciejko J, Holmes D , Kottke B , Zinsmeister A , Dinh D , and Mao S. Apolipoprotein A - I as a marker of angiographically assessed coronary-artery disease. N Engl J Med , 1983; 309: 385-389. 43. Puchois P, Kandoussi A , Fievet P, Fourrier J, Bertrand M , Koren E , and Fruchart J. Apolipoprotein A - I containing lipoproteins in coronary artery disease. Atherosclerosis, 1987; 68: 35-40. 44. Warnick G . Measurement and clinical significance of high-density lipoprotein cholesterol subclasses, in Handbook of Lipoprotein Testing, N . Rifai , G . Warnick, and M . Dominiczak, Editors. 1997, A A C C Press: Washington. 251-266. 45. Taylor C, Manalo-Estrella P, and Cox G . Atherosclerosis in Rhesus monkeys. V . Marked diet-induced hypercholesteremia with xanthomatosis and severe atherosclerosis. Arch Path, 1963; 76: 239-249. 46. Connor W and Connor S. The key role of nutritional factors in the prevention of coronary heart disease. Prev Med , 1972; 1: 49-83. 47. Purnell J and Brunzell J. The central role of dietary fat, not carbohydrate, in the insulin resistance syndrome. Curr Opin Lip idol , 1997; 8: 17-22. 48. Despres J, Moorjani S, Lupien P, Tremblay A , Nadeau A , and Bouchard C. Regional distribution of body fat, plasma lipoproteins, and cardiovascular disease. Arteriosclerosis, 1990; 10: 497-511. 49. Pouliot M , Despres J, Lemieux S, Moorjani S, Bouchard C , Tremblay A , Nadeau A , and Lupien P. Waist circumference and abdominal sagittal diameter: Best simple anthropometric indexes of abdominal visceral adipose tissue accumulation and related cardiovascular risk in men and women. A m J Cardiol, 1994; 73: 460-468. 50. Hennekens C, Rosner B , and Cole D . Dai ly alcohol consumption and fatal coronary heart disease. A m J E p i d , 1978; 107(3): 196-200. 51. Wannamethee S and Shaper A . Lifelong teetotallers, ex-drinkers and drinkers: Mortality and the incidence of major coronary heart disease events in middle-aged British men. Int J Epidemiol, 1997; 26(3): 523-531. 94 52. Castelli W , Gordon T, Hjortland M , Kagan A , Doyle J, Hames C, Hulley S, and Zukel W . Alcohol and blood lipids: The Cooperative Lipoprotein Phenotyping Study. Lancet, 1977; 2: 153-155. 53. Savolainen M and Kesaniemi Y . Effects of alcohol on lipoproteins in relation to cornary heart disease. Curr Opin Lip idol , 1995; 6: 243-250. 54. Snowden C, McNamara P, Garrison R, Feinleib M , Kannel W , and Epstein F. Predicting coronary heart disease in siblings - A multivariate assessment: The Framingham Heart Study. A m J Epid, 1982; 115(2): 217-222. 55. O'Connor G , Buring J, Yusuf S, Goldhaber S, Olmstead E , Paffenbarger R, and Hennekens C. A n overview of randomized trials of rehabilitation with exercise after myocardial infarction. Circulation, 1989; 80: 234-244. 56. Dracup K , Moser D , Marsden C, Taylor S, and Guzy P. Effects of a multidimensional cardiopulmonary rehabilitation program on psychosocial function. A m J Cardiol, 1991; 68: 31-34. 57. Hansson L , Zanchetti A , Carruthers S, Dahlof B , Elmfeldt D , Julius S, Menard J, Rahn K , Wedel H , and Westerling S. Effects of intensive blood-pressure lowering and low-dose aspirin in patients with hypertension: principal results of the Hypertension Optimal Treatment (HOT) randomised trial. Lancet, 1998; 351: 1755-1762. 58. Curb J, Pressel S, Cutler J, Savage P, Applegate W , Black H , Camel G , Davis B , Frost P, Gonzalez N , Guthrie G , Oberman A , Rutan G , and Stamler J. Effect of diuretic-based antihypertensive treatment on cardiovascular disease risk in older diabetic patients with isolated systolic hypertension. J A M A , 1996; 276(23): 1886-1892. 59. Collins R, Peto R, MacMahon S, Hebert P, Fiebach N , Eberlein K , Godwin J, Qizilbash N , Taylor J, and Hennekens C. Blood pressure, stroke, and coronary heart disease: Part 2, short-term reductions in blood pressure: overview of randomised drug trials in their epidemiological context. Lancet, 1990; 335: 827-838. 60. Sutton-Tyrrell K , Alcorn H , Herzog H , Kelsey S, and Kul ler L . Morbidity, mortality, and antihypertensive treatment effects by extent of atherosclerosis in older adults with isolated systolic hypertension. Stroke, 1995; 26: 1319-1324. 61. Antiplatelet Trialists' Collaboration. Secondary prevention of vascular disease by prolonged antiplatelet treatment. B M J , 1988; 296: 320-331. 62. Steering Committee of the Physicians' Health Study Research Group. Final report on the aspirin component of the ongoing Physicians' Health Study. N Engl J Med , 1989; 321(3): 129-135. 95 63. Hennekens C, Buring J, Sandercock P, Collins R, and Peto R. Aspir in and other antiplatelet agents in the secondary and primary prevention of cardiovascular disease. Circulation, 1989; 80(4): 749-756. 64. Cook N , Chae C, Mueller F, Landis S, Saks A , and Hennekens C, Mis-medication and under-utilization of aspirin in the prevention and treatment of cardiovascular disease,. 1999, Medscape General Medicine. 65. Hjortland M , McNamara P, and Kannel W . Some atherogenic concomitants of menopause: The Framingham Study. A m J Epid, 1976; 103(3): 304-311. 66. Mosca L , Grundy S, Judelson D , K i n g K , Limarcher M , Oparil S, Pasternak R, Pearson T, Redberg R, Smith Jr. SC, Winston M , and Zinberg S. Guide to preventive cardiology for women. J A m C o l l Cardiol, 1999; 33: 1751-1755. 67. Nabulsi A , Folsom A , White A , Patsch W , Heiss G , W u K , and Szklo M . Association of hormone-replacement therapy with various cardiovascular risk factors in postmenopausal women. N Engl J Med , 1993; 328(15): 1069-1075. 68. Hulley S, Grady D , Bush T, Furberg C, Herrington D , Riggs B , and Vittinghoff E . Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. J A M A , 1998; 280(7): 605-613. 69. Hoffman R and Garewal H . Antioxidants and the prevention of coronary heart disease. Arch Intern Med, 1995; 155(3): 241-246. 70. The Heart Outcomes Prevention Evaluation Study Investigators. Vitamin E supplementation and cardiovascular events in high-risk patients. N Engl J Med , 2000; 342: 154-160. 71. Rapola J, Virtamo J, Haukka J, Heinonen O, Albanes D , Taylor P, and Huttunen J. Effect of vitamin E and beta carotene on the incidence of angina pectoris: A randomized, double-blind, controlled trial. J A M A , 1996; 275(9): 693-698. 72. Virtamo J, Rapola J, Ripatti S, Heinonen O, Taylor P, Albanes D , and Huttunen J. Effect of vitamin E and beta carotene on the incidence of primary nonfatal myocardial infarction and fatal coronary heart disease. A r c h Intern Med, 1998; 158(6): 668-675. 73. The Beta-Blocker Pooling Project Research Group. The Beta-Blocker Pooling Project (BBPP) : subgroup findings from randomized trials in post infarction patients. Eur Heart J, 1988; 9: 8-16. 74. Hjermann I, Holme I, and Leren P. Oslo Study Diet and Antismoking Trial. A m J Med , 1986; 80(Suppl 2A) : 7-11. 75. Barnard R. Effects of life-style modification on serum lipids. Arch Intern Med, 1991; 151: 1389-1394. 96 76. Longeril M D , Renaud S, Mamelle N , Salen P, Martin J, Monjaud I, Guidollet J, Touboul P, and Delaye J. Mediterranean alpha-linolenic acid-rich diet in secondary prevention of coronary heart disease. Lancet, 1994; 343: 1454-1459. 77. Jukema J, Bruschke A , Boven A v , Reiber J, Ba l E , Zwinderman A , Jansen H , Boerma G , van Rappard F M , and L ie K . Effects of l ipid lowering by pravastatin on progression and regression of coronary artery disease in symptomatic men with normal to moderately elevated serum cholesterol levels: The Regression Growth Evaluation Statin Study (REGRESS) . Circulation, 1995; 91: 2528-2540. 78. Canner P, Berge K , Wenger N , Stamler J, Friedman L , Prineas R, and Friedewald W . Fifteen year mortality in Coronary Drug Project patients: Long-term benefit with niacin. J A m C o l l Cardiol, 1986; 8(6): 1245-1255. 79. The Coronary Drug Project Research Group. Clofibrate and niacin in coronary heart disease. J A M A , 1975; 231(4): 360-381. 80. L i p i d Research Clinics Program. The L i p i d Research Clinics Coronary Primary Prevention Trial Results: Reduction in incidence of coronary heart disease. J A M A , 1984; 251(3): 351-364. 81. Stark R. Review of the major intervention trials of lowering coronary artery disease risk through cholesterol reduction. A m J Cardiol, 1996; 78(Suppl 6A): 13-19. 82. Frick M , Elo O, Haapa K , Heinonen O, Heinsalmi P, Helo P, Huttenen J, Kaitaniemi P, Koskinen P, Manninen V , Maenpaa H , Malkonen M , Manttari M , Norola S, Pasternack A , Pikkarainen J, Romo M , Sjoblom T, and Nikk i l a E . Helsinki Heart Study: Primary-prevention trial with gemfibrozil in middle-aged men with dyslipidemia: Safety of treatment, changes in risk factors, and incidence of coronary heart disease. N Engl J Med , 1987; 317: 1237-1245. 83. Pearson T. Primary and secondary prevention of coronary artery disease: Trials of l ipid lowering with statins. A m J Cardiol, 1998; 82(1 OA): 28S-30S. 84. Shepherd J, Cobbe S, Ford I, Isles C, Lorimer A , Macfarlane P, M c K i l l o p J, and Packard C. Prevention of coronary heart disease with pravastatin in men with hypercholesterolemia. N Engl J Med , 1995; 333(20): 1301-1307. 85. Downs J, Clearfield M , Wels S, Whitney E , Shapiro D , Beere P, Langendorfer A , Stein E , Kruyer W , and Gotto A . Primary prevention of acute coronary events with lovastatin in men and women with average cholesterol levels: Results of A F C A P S / T e x C A P S . J A M A , 1998; 279(20): 1615-1622. 86. Scandinavian Simvastatin Survival Study Group. Randomised trial o f cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet, 1994; 344: 1383-1389. 87. Sacks F, Pfeffer M , Moye L , Rouleau J, Rutherford J, Cole T, Brown L , Warnica J, Arnold J, Wun C, Davis B , and Braunwald E . The effect of pravastatin on 97 coronary events after myocardial infarction in patients with average cholesterol levels. N Engl J Med , 1996; 335(14): 1001-1009. 88. The Long-Term Intervention with Pravastatin in Ischaemic Disease (LIPID) Study Group. Prevention of cardiovascular events and death with pravastatin in patients with coronary heart disease and a broad range of initial cholesterol levels. N Engl J Med , 1998; 339(19): 1349-1357. 89. Wise G and Schultz T. Hyperlipidemia: When does treatment make a difference? Postgrad Med, 1996; 100(1): 138-149. 90. Wiklund O, Bergman M , Bondjers G , Linden T, Odman B , Saarinen I, Kron B , and Wright I. Pravastatin and gemfibrozil alone and in combination for the treatment of hypercholesterolemia. A m J Med , 1993; 94: 13-20. 91. Davignon J, Roederer G , Montigny M , Hayden M , Tan M , Connelly P, Hegele R, McPherson R, Lupien P, Gagne C, Little J, and Co l in P. Comparative efficacy and safety of pravastatin, nicotinic acid and the two combined in patients with hypercholesterolemia. A m J Cardiol, 1994; 73: 339-345. 92. Pasternak R, Brown L , Stone P, Silverman D , Gibson C, and Sacks F. Effect of combination therapy with lipid-reducing drugs in patients with coronary heart disease and "normal" cholesterol levels. A n n Int Med , 1996; 125(7): 529-540. 93. The Canadian Consensus Conference on the Prevention of Heart and Vascular Disease by Altering Serum Cholesterol and Lipoprotein Risk Factors. Canadian Consensus Conference on Cholesterol: Final Report. C M A J , 1988; 139(Suppl 11): 1-8. 94. Clifton P, Wight M , and Nestel P. Is fat restriction needed with H M G C o A reductase inhibitor treatment? Atherosclerosis, 1992; 93: 59-70. 95. Fodor J, Frohlich J, Genest J, and McPherson P. Recommendations for the management and treatment of dyslipidemia: Report of the Working Group on Hypercholesterolemia and Other Dyslipidemias. C M A J , 2000; 162(10): 1441-1447. 96. Palta M , Prineas R, Berman R, and Hannan P. Comparison of self-reported and measured height and weight. A m J Epid, 1982; 115(2): 223-230. 97. Pirie P, Jacobs D , Jeffery R, and Hannan P. Distortion in self-reported height and weight data. J A m Diet Assoc, 1981; 78: 601-606. 98. Rowland M . Self-reported weight and height. A m J C l i n Nutr, 1990; 52: 1125-1133. 99. DelPrete L , Caldwell M , English C, Banspach S, and Lefebvre C. Self-reported and measured weights and heights of participants in community-based weight loss programs. J A m Diet Assoc, 1992; 92: 1483-1486. 98 100. Stewart A . The reliability and validity of self-reported weight and height. J Chron Dis, 1982; 35: 295-309. 101. Lichtman S, Pisarska K , Berman E , Pestone M , Dowl ing H , Offenbacher E , Weisel H , Heshka S, Matthews D , and Heymsfield S. Discrepancy between self-reported and actual caloric intake and exercise in obese subjects. N Engl J Med , 1992; 327(27): 1893-1898. 102. Robertson D and Keller C. Relationships among health beliefs, self-efficacy, and exercise adherence in patients with coronary artery disease. Heart Lung, 1992; 21(1): 56-63. 103. Mirotznik J, Feldman L , and Stein R. The health belief model and adherence with a community center-based, supervised coronary heart disease exercise program. J Comm Health, 1995; 20(3): 233-247. 104. Frank E , Winkleby M , Fortmann S, and Farquhar J. Cardiovascular disease risk factors: Improvements in knowledge and behavior in the 1980s. A m J Public Health, 1993; 83(4): 590-593. 105. Wi lcox S and Stefanick M . Knowledge and perceived risk of major diseases in middle-aged and older women. Health Psyc, 1999; 18(4): 346-353. 106. Davis S, Winkelby M , and Farquhar J. Increasing disparity in knowledge of cardiovascular disease risk factors and risk-reduction strategies by socioeconomic status: Implications for policymakers. A m J Prev Med , 1995; 11(5): 318-323. 107. Fleetwood J and Packa D . Determinants of health-promoting behaviors in adults. J Cardiovasc Nurs, 1991; 5(2): 67-79. 108. Sutterer J, Carey M , Silver D , and Nash D . Risk factor knowledge, status, and change in a community screening project. J Comm Health, 1989; 14(3): 137-147. 109. The Clinical Quality Improvement Network (CQIN) Investigators. L o w incidence of assessment and modification of risk factors in acute care patients at high risk for cardiovascular events, particularly among females and the elderly. A m J Cardiol, 1995; 76: 570-573. 110. Cohen M , Byrne M , Levine B , Gutowski T, and Adelson R. L o w rate of treatment of hypercholesterolemia by cardiologists in patients with suspected and proven coronary artery disease. Circulation, 1991; 83: 1294-1304. 111. Giles W , Anda R, Jones D , Serdula M , Merritt R, and DeStefano F. Recent trends in the identification and treatment of high blood cholesterol by physicians. J A M A , 1993; 269(9): 1133-1138. 112. McBr ide P, Schrott H , Plane M , Underbakke G , and Brown R. Primary care practice adherence to National Cholesterol Education Program guidelines for patients with coronary heart disease. Arch Intern Med , 1998; 158(11): 1238-1244. 99 113. Frolkis J, Zyzanski S, Schwartz J, and Suhan P. Physician noncompliance with the 1993 National Cholesterol Education Program (NCEP-ATPI I ) guidelines. Circulation, 1998; 98: 851-855. 114. Marcelino J and Feingold K . Inadequate treatment with H M G - C o A reductase inhibitors by health care providers. A m J Med , 1996; 100: 605-610. 115. Sueta C, Chowdhury M , Boccuzzi S, Smith S, Alexander C, Londhe A , Lul la A , and Simpson R. Analysis of the degree of undertreatment of hyperlipidemia and congestive heart failure secondary to coronary artery disease. A m J Cardiol, 1999; 83(9): 1303-1307. 116. Campbell N , Thain J, Deans H , Ritchie L , and Rawles J. Secondary prevention in coronary heart disease: baseline survey of provision in general practice. B M J , 1998; 316: 1430-1434. 117. Schrott H , Bittner V , Vittinghoff E , Herrington D , and Hulley S. Adherence to National Cholesterol Education Program treatment goals in postmenopausal women with heart disease: The Heart and Estrogen/Progestin Replacement Study (HERS). J A M A , 1997; 277(16): 1281-1286. 118. Health Services Utilization and Research Commission. Adherence to cholesterol-lowering drugs in Saskatchewan. 1997, Health Services Utilization and Research Commission: Saskatoon, p. 1-8. 119. McKeehan C, Blue eggs and beans: Patient responses to high cholesterol. 1995, Health Services Utilization and Research Commission: Saskatoon. 120. Kattermann R, Jaworek D , Mol ler G , Assmann G , Bjorkhem I, Svensson L , Borner K , Boerma G , Leijnse B , and Desager J. Multicentre study of a new enzymtic method of cholesterol determination. J C l i n Chem C l i n Biochem, 1984; 22: 245-251. 121. Nagele U , Hagele E , Sauer G , Wiedemann E , Lehmann P, Wahlefeld A , and Gruber W . Reagent for the enzymatic determination of serum total triglycerides with improved lypolytic efficiency. J C l i n Chem C l i n Biochem, 1984; 22: 165-174. 122. Burstein M , Scholnick H , and Morfin R. Rapid method for the isolation of lipoproteins from human serum by precipitation with polyanions. J L ip id Res, 1970; 11: 583-595. 123. Sternberg J. Rate nephelometer for measuring specific proteins by immunoprecipitin reactions. C l i n Chem, 1977; 23: 1456-1464. 124. Dobiasova M and Frohlich J. Assays of lecithin cholesterol acyltransferease ( L C A T ) . in Methods in Molecular Biology, J. Ordovas, Editor. 1998, Humana Press Inc.: Totowa N J . 217-230. 100 125. Friedewald W , Levy R, and Frederickson D . Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. C l i n Chem, 1972; 18: 499-502. 126. Haft J, Haik B , Goldstein J, and Brodyn N . Development of significant coronary artery lesions in areas of minimal disease: A common mechanism for coronary disease progression. Chest, 1988; 94: 731-736. 127. Campbell N , Burgess E , Choi B , Taylor G , Wilson E , Cleroux J, Fodor G , Leiter L , and Spence D . Methods and an overview of the Canadian recommendations. C M A J , 1999; 160(9 Suppl): S1-S6. 128. Adler L , Mean F E R - H D L levels in men and women. 1997. 129. Lean M , Han T, and Morrison C. Waist circumference as a measure for indicating need for weight management. B M J , 1995; 311: 158-161. 130. Francis M and Frohlich J. Coronary artery disease in patients at low risk -apolipoprotein A I as an independent risk factor. Atherosclerosis, 2000; In press. 131. Schaefer E , Lamon-Fava S, Ordovas J, Cohn S, Schaefer M , Castelli W , and Wilson P. Factors associated with low and elevated plasma high density lipoprotein cholesterol and apolipoprotein A - I levels in the Framingham Offspring Study. J L i p i d Res, 1994; 35: 871-882. 132. Yamashita T, Colombo A , and Tobis J. Limitations of coronary angiography compared with intravascular ultrasound: Implications for coronary interventions. Prog Cardiovasc Dis, 1999; 42(2): 91-138. 133. Nakamura Y , Takemori H , Shiraishi K , Inoki I, Sakagami M , Shimakura A , Usuda K , Kubota K , Takata S, and Kobayashi K . Compensatory enlargement of angiographically normal coronary segments in patients with coronary artery disease: In vivo documentation using intravascular ultrasound. Angiology, 1996; 47(8): 775-781. 134. Isner J, Salem D , Jr. JB , and Levine H . Long-term clinical course of patients with normal coronary arteriography: Follow-up study of 121 patients with normal or nearly normal coronary arteriograms. A m Heart J, 1981; 102(4): 645-653. 135. Forbes G . Diet and exercise in obese subjects: Self-report versus controlled measurements. Nut rRev, 1993; 51(10): 296-300. 136. Choi B , Pak A , and Purdham J. Effects of mailing strategies on response rate, response time, and cost in a questionnaire study among nurses. Epidemiology, 1990; 1: 72-74. 137. Eaker S, Bergstrom R, Bergstrom A , Adami H , and Nyren O. Response rate to mailed epidemiologic questionnaires: A population-based randomized trial of variations in design and mailing routines. A m J Epid, 1998; 147(1): 74-82. 101 138. H i l l A , Roberts J, Ewings P, and Gunnell D . Non-response bias in a lifestyle survey. J Publ Health Med, 1997; 19(2): 203-207. 139. Smith C and Nutbeam D . Assessing non-response bias: a case study from the 1985 Welsh Heart Health Survey. Health Educ Res, 1990; 5(3): 381-386. 140. Romm P, Green C, Reagan K , and Rackley C. Relation of serum lipoprotein cholesterol levels to the presence and severity of angiographic coronary artery disease. A m J Cardiol, 1991; 67: 479-483. 141. Danias P, O'Mahony S, Radford M , Korman L , and Silverman D . Serum cholesterol levels are underevaluated and undertreated. A m J Cardiol, 1998; 81(11): 1353-1356. 142. Rubins H , Robins S, Collins D , Iranmanesh A , Wi l t T, Mann D , Mayo-Smith M , Faas F , Elam M , Rutan G , Anderson J, Kashyap M , and Schectman G . Distribution of lipids in 8,500 men with coronary artery disease. A m J Cardiol, 1995; 75:1196-1201. 143. Franzen J and Fex G . L o w serum apolipoprotein A - I in acute myocardial infarction survivors with normal H D L cholesterol. Atherosclerosis, 1986; 59: 37-42. 144. Sharrett A , Patsch W , Sorlie P, Heiss G , Bond M , and Davis C. Associations of lipoprotein cholesterols, apolipoproteins A - I and B , and triglycerides with carotid atherosclerosis and coronary heart disease: the Atherosclerosis Risk in Communities (ARIC) study. Arterioscl Thromb, 1994; 14: 1098-1104. 145. Broemeling L and Mielke C. Coronary risk assessment in women. Lancet, 1999; 354(9176): 426-427. 146. Tunstall-Pedoe H , Woodward M , Tavendale R, A 'Brook R, and McCluskey M . Comparison of the prediction by 27 different factors of coronary heart disease and death in men and women of the Scottish heart health study: cohort study. B M J , 1997; 315(7110): 722-729. 147. Lewis E , Hunsicker L , Bain R, and Rohde R. The effect of angiotensin-converting-enzyme inhibition on diabetic nephropathy. N Engl J Med , 1993; 329(20): 1456-1462. 148. The Heart Outcomes Prevention Evaluation Study Investigators. Effects of an angiotensin-converting-enzyme inhibitor, Ramipril , on cardiovascular events in high-risk patients. N Engl J Med , 2000; 342(3): 145-153. 149. Volek J, Gomez A , and Kraemer W . Fasting lipoprotein and postprandial triacylglycerol responses to a low-carbohydrate diet supplemented with n-3 fatty acids. J A m C o l l Nutr, 2000; 19(3): 383-391. 150. Mayes P. L i p i d transport and storage, in Harper's Biochemistry, R. Murray, et al, Editors. 1993, Appleton & Lange: Norwalk. 250-265. 102 151. Wei M , Gibbons L W , Mitchel l T L , Kampert JB , Stern M P , and Blair S N . L o w Fasting Plasma Glucose Level as a Predictor of Cardiovascular Disease and A l l -Cause Mortality. Circulation, 2000; 101(17): 2047-2052. 152. Leupker R. Current status of cholesterol treatment in the community: The Minnesota Heart Survey. A m J Med, 1997; 102(2S): 37-42. 153. Ebrahim S. Systematic review of randomised controlled trails of multiple risk factor interventions for preventing coronary heart disease. B M J , 1997; 314: 1666-1674. 154. Stocker R. Dietary and pharmacological antioxidants in atherosclerosis. Curr Opin Lip idol , 1999; 10(6): 589-597. 155. Toronto working Group on Cholesterol Policy. Asymptomatic hypercholesterolemia: a clinical policy review. J C l i n Epidemiol, 1990; 43(10): 1029-1121. 156. Kokkinos P and Fernhall B . Physical activity and high density lipoprotein cholesterol levels: What is the relationship? Sports M e d , 1999; 28(5): 307-314. 157. Gulanick M , Bl i ley A , Perino B , and Keough V . Recovery patterns and lifestyle changes after coronary angioplasty: The patient's perspective. Heart Lung, 1998; 27: 253-262. 158. Langner N , Hasselback P, Dunkley G , and Corber S. Attitudes and practices of primary care physicians in the management of elevated serum cholesterol levels. C M A J , 1989; 141: 33-38. 159. Tanenbaum T, Sampalis J, Battista R, Rosenberg E , and Joseph L . Early detection and treatment of hyperlipidemia: physician practices in Canada. C M A J , 1990; 143(9): 875-881. 160. Majumdar S, Gurwitz J, and Soumerai S. Undertreatment of hyperlipidemia in the secondary prevention of coronary artery disease. J Gen Intern Med, 1999; 14(12): 711-717. 161. Rembold C. Number-needed-to-treat analysis of the prevention of myocardial infarction and death by antidyslipidemic therapy. J Fam Pract, 1996; 42(6): 577-586. 162. Grover S, Coupal L , and H u X . Identifying adults at increased risk of coronary disease: how well do the current cholesterol guidelines work? J A M A , 1995; 274(10): 801-806. 163. Brambilla D and M c K i n l a y S. A comparison of responses to mailed questionnaires and telephone interviews in a mixed mode health survey. A m J Epid, 1987; 126(5): 962-971. 103 APPE>lblX A Study of Cholesterol Esterificatioo and Coronary Artery Disease Coordinated by : Drs J Frohlich and H Pritchard Department of Pathology and Laboratory Medicine University of British Columbia Funded by : Medical Research Council of Canada Addressograph Stamp • What is your name ? • Date of Birth (d/m/y) • Sex M Q F Q If female, have you passed menopause 7 If yes, do you take hormone replacement therapy 7 Y Q hD Y Q hD • What is your ethnic background? • Have any of your close relatives (parent, brother, sister, uncle, etc) had significant heart disease (Ml or angina) before the age of 60? Y D N O • If yes, how many relatives ? _ 1 Which category best describes your exercise habits 7 1 What category best describes your smoking habits 7 If you smoked in the past, how many years ago did you stop? • What category best describes your drinking habits? •Are you a vegetarian 7 YD HD D none Q occasional (slow walks several times a week) Q moderate (active, ie: jogging twice a week) Q heavy (daily vigorous exercise) I have never smoked • occasional (less than 10 cigarettes a week) • moderate (approximately one pack a day) D heavy (more than one pack a day) , and how much did you smoke (pks/day)? D never D occasional (less than 1 drink a day) D moderate (1 to 2 drinks a day) • heavy (more than 3 drinks a day) If yes, for how many years have you been a vegetarian 7 Patient ID# TC FC TG Comment Apo B APOAI APO All LpAl HDLTC HDL FC HDLTG Plasma LCAT HDLLCAT FER MER HDL HDL HDL HDL HDL 2b 3a 3b Atherosclerosis Specialty Laboratory Phone: [604]875-3087 Fax: [604]875-2787 104 ^Clinical Information • - :T to'be" completed by the physician/designate 9 Weight Height Waist circumference kg cm cm Blood Pressure (sitting R arm) Any significant weight gain/loss over last 3 months ( ± 3kg (71b)) ? Y d N • Documented history of disease A Personal history CAD If previous CAD present : • PVD ' • CVD • Diabetes Mellitus . • Renal Insufficiency • Hypertension B Family history • CAD • PVD • CVD • Diabetes Mellitus • Hypertension • • • Angiography • Angioplasty • Bypass surgery • MI Findings Positive Negative I—I Q Indicate all current medications and specify : • ASA Q ACE inhibitors Q Beta blockers Q Calcium channel blockers Q Antioxidants (vitamin E, carotene) l~j Lipid lowering drugs • Other Q Result of current angiography : • How many vessels affected ? V • What is the percentage obstruction ? None • l Q 2Q 3 Q <50% • >50% • JFB • January 19S4 • CAD26 P a t t e r n s o f T r e a t m e n t f o r C o r o n a r y A r t e r y D i s e a s e Coordinated by: Drs J Frohlich and H Pritchard, M Francis | Department of Pathology and Laboratory Medicine M I University of British Columbia ^ 1 Funded by: Medical Research Council of Canada and Bristol-Myers Squibb This form should take approximately 10-15 minutes to complete. Please try to answer all of the questions to the best of your ability.. Today's date: dd/mm/yy What is your Personal Health Number? . (On the front of your CareCard) Name: Birthdate: Fine Lot dd/mm/yy How tall are you? in/cm How much do you weigh? lbs/kg Has your weight changed in the past four years? D Increased O Decreased D Stayed the same If your weight has changed, by how much approximately? . lbs/kg "With a tape measure, lie on your back, relax, and measure around your waist at the level of your belly button. What is the measurement? in/cm SMOKING Have you ever regularly smoked tobacco? LJ Yes U No If yes, how many years? _1 If you smoked, which category best describes your O I don't smoke anymore present smoking habit? When did you quit?-O Occasional (less than 10 cigs/week) O Moderate (approx. 1 pack/day) O Heavy (more than 1 pack/day) EXERCISE _ Has there been any change in your exercise habits in the past 4 years? HI Yes LJ No Si Which category best describes your exercise habits? p Don't exercise O Occasional (slow walks many times/week) O Moderate (active, ie: jogging 2x/week) [ j Heavy (daily vigorous exercise) A L C O H O L Has there been any change in your drinking habits in the past 4 years? 0 Yes O No 31 Which category best describes your drinking habits? O Don't drink P7 Occasional (Less than one drink/day) O Moderate (1 to 2 drinks/day) " D Heavy (more than 3 drinks/day) NUTRITION Have you made any changes to your diet in the past 4 years? O Yes O No _l Which category best describes your eating habits? j_J Traditional North American diet O Low fat/cholesterol diet O Calorie-reduced diet O Vegetarian 0 Other (eg. Jenny Craig) — CHOLESTEROL Has your cholesterol been checked in the past four years? DYes D N o When was it last measured? What was the most recent measurement? D Don't know HYPERTENSION Do you have high blood pressure? DYes D N o When was it last measured? . • What was the measurement? D Don't know DIABETES Do you have diabetes? LJ Yes D NO Jfl If yes, do you take insulin? D Yes CI No DOCTOR/HOSPITAL VISITS In the past four years, have you been seen for heart trouble or blood vessel disease? D Yes D No In the past four years, have you had any of the following: Angioplasty DYes D N o Bypass surgery D Yes D No Heart attack D Yes D No What is your family doctor's name? When was the last time you visited a heart specialist? LIFESTYLE COUNSELLING Has anyone discussed any of the following issues with you? DYes D N o iS If yes, which ones were discussed? Sl Who discussed it/them with you? D Cholesterolievel D Doctor D Blood pressure D Nurse D Smoking ( c i i e c k ^ a p p l y ) D Dietician/nutritionist D Exercise D Exercise instructor D Alcohol D Physical/occupational therapist D Diet/Weight D Family/friend MEDICATIONS Which of the following types of medications do you take regularly? (check all that apply) D Cholesterol medication D Hormone Replacement Therapy D Blood pressure medication D Anti-depressants D Heart medication D Viumins D Diabetic medication D Natural medicines D A^irin/Blood thinners JJ None ^ Other If possible, please include a list of the medications you take including n a m * of drug, dose, and date started. Thank you for filling out this questionnaire. Your help with this research is appreciated. Healthy Heart Program Michelle Francis Phone: (604)631-5866 SL Paul's Hospital Fax: (604)631-5590 e-mail: michemac@unixg.ubcca \Sl2ry H tALTH CARE | \j St Paul's Hospital CARDIAC DISCHARGE ORDERS Discharge Prescription Order Form N a m e : Allergy: Q Computer-generated medication profile (prepared by a pharmacist & reviewed with patient) requested. (NOTE: Complete 24 hrs prior to discharge) DRUG DOSE/FREQUENCY QUANTITY/REFILL ACE INHIBITORS. • Not indicated • Contraindicated • Other: • Enalaprjl • Quinapril • Ramipril • Other • No need to fill at this time ANTICOAGULANTS/ANTIPLATELETS • Not indicated • • ASA Contraindicated • Other • Clopidogrel • No need to fill at this time • Warfarin • No need to fill at this time BETA-BLOCKERS • Not indicated • Contraindicated • Other: • Acebutolol • Atenolol • Metoprolol • Other • No need to fill at this time CALCIUM CHANNEL BLOCKERS • Not indicated • Contraindicated • Other: • Amlodipine • Felodipine • Diitiazem CD • Verapamil SR • Other • No need to fill at this time DIURETICS • Furosemide • Spironolactone • Hydrodorothiazide • Other • No need to fill at this time LIPID-LOWERING MEDICATIONS • Not indicated • Contraindicated • Other • Atrovastatin • Lovastatin • Pravastatin • Simvastatin • Other • No need to fill at this time • Digoxin • No need to fill at this time NITRATES • Nitroglycerin patch • Isosorbide dinitrate • Other • No need to fill at this time • Nitroglycerin spray OTHER MEDICATIONS: Discontinue all other medications if not listed on this sheet. Distribution: White copy: Give to patient on discharge. Canary Copy: Fax copy of this form to Dr. and place back on chart Pink copy: St. Paul's Hospital Pharmacy Form No. NF180T (09/99) Signature/MSC # Date 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0089732/manifest

Comment

Related Items