Comparison of an Olive oil enriched diet to a Low fat diet Intervention study

Abstract Epidemiological studies provide good evidence that a diet high in fat may be associated with low prevalence of coronary heart disease (CHD). Over 200 diet studies since the 1950s have unequivocally demonstrated that saturated fatty acid increases plasma cholesterol, while mono-unsaturated fatty acid (MUFA) and polyunsaturated fatty acid tend to lower plasma cholesterol. The OLIVE study is a randomised diet study of patients with CHD documented by coronary angiography. One hundred and eighty patients will be enrolled and undergo angiography at baseline and 2.5 years later. A low fat diet will be compared to a high fat Mediterranean-type diet rich in MUFA and enriched with olive oil. The low fat diet will have total fat providing less than 30% of energy. The Mediterranean diet will have 35% to 40% of energy as fat, with approximately 50% of fat being MUFA, which is typical of the Cretan diet in the 1960s. In addition patients will be randomised to ?-linolenic acid (1.8 g) or placebo capsules (olive oil, 2 g). This is the largest trial of its type and the only trial to directly compare a low fat diet to a Mediterranean-type diet using angiographic end-points, Uniquely, it will assess the impact of psychological factors on progression of CHD. [Aust J Nutr Diet 1998;55(4 Suppl)S24-S29].

Keywords: cardiovascular disease; diet; monounsaturated fats; low fat diet; Mediterranean diet; nutrition

Background
A key risk factor for the development of coronary heart disease (CHD) is an elevated serum cholesterol or, more specifically, elevated low density lipoprotein (LDL) cholesterol. At least 20 randomised trials have examined the effect of lowering plasma cholesterol on CHD using serial coronary angiography. Lowering plasma cholesterol 10% or more, by diet, drugs or ileal bypass surgery is associated with less progression and greater regression of CHD. Therapy, sufficient to inhibit progression, is associated with less angina, lower acute myocardial infarction (AMI) rates and less frequent revascularisation procedures ( 1). improvement may occur within a few months of initiation of therapy. Diet modification is in principle the cornerstone of lowering plasma cholesterol and most health authorities agree with the recent American recommendation of restricting total dietary fat intake to less than 30% of total energy (E) and saturated fatty acids (SFA) to less than 8 to 10% of energy ( 2).

Diet and coronary atherosclerosis progression
In six coronary angiographic studies, the effects of diet on CHD were assessed independently of the effects of drugs ( 3-8). These studies compared 'usual care' diets to low fat diets. The low fat diets usually required a total fat intake of less than 30% of energy. A strong correlation of saturated fatty acid (SFA) and mono-unsaturated fatty acid (MUFA) intake was found in these studies. These studies could be interpreted as showing that MUFA intake is associated with progression or new lesion formation in coronary arteries.

A stepwise logistic regression analysis of the Stanford Coronary Risk Intervention Project ( 9) demonstrated that the major dietary factor responsible for new lesion formation was calories consumed as fat (P=0.008). SFA (P = 0,03), MUFA (P = 0.012) and polyunsaturated fatty acid (PUFA) (P = 0.006) intakes all predicted new lesion formation. The Cholesterol Lowering Atherosclerosis Study (CLAS) ( 6) obtained similar results with total fat (OR=0.90, confidence interval [CI] 0.83-0.98) and PUFA (odds ratio [OR] = 0.78, CI 0.65-0.93) being the most significant dietary factors related to new lesion formation. There was a non-significant trend for MUFA (OR = 0.86, CI 0.73-1.03). In the St Thomas' Atherosclerosis Regression Study ( 8) total fat intake and LDL cholesterol levels were both independently correlated with progression. In contrast, MUFA was inversely correlated with progression (R = -0.33, P = 0.16). SFA intake correlated inversely with regression, while MUFA and PUFA had no correlation.

None of the angiographic studies were designed to differentiate between the types of fat and lesion change. Nonetheless, these studies suggest that different fats have different effects on progression and regression rates, and new lesion formation.

Epidemiology
The Mediterranean diet refers to the traditional food consumed in regions bordering the Mediterranean Sea including Greece, Southern Italy, Southern France, Turkey, Morocco and Spain.

The Mediterranean diet may be moderately low in total fat (25% E) as in parts of Italy, or high in fat (? 40% E) as in Crete ( 10). Intake of fat, irrespective of amount, is almost exclusively from olive oil. The seven countries study clearly demonstrated that a high MUFA intake is associated with a very low incidence of CHD. The death rate from CHD in the Finnish cohort of the study was more than 30 times that of Crete, though both had fat intake approximately 40% of energy ( 11). Recently published data from the multinational monitoring of trends and determinants of cardiovascular disease (MONICA) project ( 12) indirectly confirm the seven countries study findings in that low prevalence of CHD can occur in communities with high fat intake if it is mainly unsaturated.

There have been over 200 studies published since the first human diet-cholesterol studies in the 1950s.

Dietary studies have focused on manipulation of macronutrients and, in particular, fats and their effect on serum lipids and lipoproteins. SFA is the key fat that influences LDL levels: the major atherogenic lipoprotein. MUFA and PUFA have little direct effect on LDL levels. High intake of PUFA lowers high density lipoprotein (HDL) cholesterol. An olive oil enriched diet has been viewed as a diet rich in MUFA and specifically oleic acid. Olive oil contains more than 100 compounds, including ten frequently-occurring polyphenols ( 13). These 'minor' components may have significance in disease prevention. The polyphenolic fraction varies from 50 to 800 mg/kg of oil depending on factors such as the soil, ripeness of the fruit and the way the oil was produced and stored ( 14). These polyphenols have wide-ranging biological activities including inhibition of LDL oxidation, stimulation of anti-oxidant enzymes, and antiplatelet, antifungal and anticarcinogenic properties. Refining olive oil reduces the phenolic content ( 15). When a diet is enriched with olive oil, LDL is more difficult to oxidise. (Inhibition of oxidation is due to MUFA and polyphenolic content.) Oxidation of LDL is probably a key step in atherosclerosis initiation and progression.

A traditional Cretan diet includes a high intake and large variety of fruits, vegetables and herbs, which are associated with a high intake of polyphenols and fibres.

Animal studies
Animal studies have demonstrated that diets rich in MUFA and ?-linolenic acid (ALA) inhibit the development of atherosclerosis. The first study to demonstrate this was a five-year diet study in 21 African green monkeys ( 16). There were three experimental diets enriched with either SFA, MUFA or PUFA. The MUFA diet was associated with significantly less coronary atherosclerosis. The type of fat in the diet was a better predictor of atherosclerosis than plasma LDL levels. A similar result was seen in a hamster study which compared a PUFA to a MUFA enriched diet. The hamsters on the PUFA diet had 82% more fatty streaks in the aorta ( 17).

ALA is a polyunsaturated fat which has significant anti-inflammatory action. In New Zealand, rabbits on an atherogenic diet with flaxseed supplementation (high in ALA) had 46% less aortic atherosclerosis development. The anti-atherogenic effect of ALA supplementation was unrelated to changes in plasma cholesterol ( 18).

Diet intervention studies
A meta-analysis of the diet clinical end-point trials demonstrated both poor efficacy of low fat diets in lowering of serum cholesterol (on average by 3.2%), and a negligible effect in preventing coronary events such as AMI and sudden cardiac death ( 19).

In marked contrast, the Lyon Diet Study ( 20, 21) demonstrated surprisingly greater efficacy than the recent large trials using 3-hydroxy-3-methylglutaryl Co-enzymeA (HMG-CoA) reductase inhibitors commonly known as statins ( 22-24). The Lyon Diet Study involved 605 patients who were randomised to either 'usual care' dietary advice or a Mediterranean diet. The trial was scheduled to run for five years but was stopped at 27 months due to the clear benefit of the Mediterranean diet. Serum cholesterol levels were similar in both groups, but there was a 70% reduction (P<0.0001) in major coronary events and a 37% reduction (P < 0.005) in major and minor events. Analysis of the diet and serum fatty acids showed that benefit correlated with an increase in oleic acid and ?-linolenic acid in the plasma. It has been hypothesised that ?-linolenic acid was a key dietary factor in preventing the coronary events ( 21). The results are inconsistent with the current diet-heart paradigm and have not been compelling enough to change physician behaviour.

The current diet-heart paradigm states that diet therapy lowers CHD risk by lowering plasma cholesterol (mainly LDL). Unfortunately the recommended low fat diets are ineffective in lowering cholesterol ( 25). In large prevention trials the average reduction is 3.2% ( 16). A reduction of 5% in serum cholesterol would decrease mortality by 4% over five years or so. For a diet trial to have adequate power to detect a 4% difference in mortality requires more than 200 000 patients ( 26). To reliably detect a difference in coronary endpoints requires more than 50 000 patients.

Fortunately, changes in coronary arteries assessed by repeat angiography are good surrogate markers for CHD events during the peiod of study. A meta-analysis of trials that used arteriographic end-points demonstrates that lack of progression or regression of coronary lesions was associated with 47% less cardiovascular events during the period between repeat angiograms ( 27). Furthermore, lesion changes are powerful predictors of future CHD events. Progression or new lesion formation is associated with double risk of any CHD event and quadruple risk of sudden cardiac death over the subsequent three to five years ( 28-30). If this study shows a clear benefit of a Mediterranean diet, then a large intervention trial may be unnecessary. An angiographic trial also permits assessment of possible mechanisms of benefit.

Psychological factors
Despite optimal management of known classic risk factors, up to 40% of patients have a recurrent coronary event within six years of sustaining an AMI ( 22, 23). Psychological factors may play a significant role in the continuing high morbidity and mortality ( 31). Psychological factors such as hostility correlate with extent of CHD on angiography. No studies have assessed psychological factors as predictors of progression. This is another unique aspect of the OLIVE study.

Research plan
Patients

One hundred and eighty patients will be recruited into the study. Patients will be eligible if they fulfil the following criteria: written informed consent obtained; male and female, aged over 18 years of age; with CHD and undergoing coronary angiography (and not proceeding to coronary artery bypass grafting). Patients will be excluded from the study if they: have a life-threatening illness likely to lead to death before the end of the trial; are unlikely to be compliant with dietary recommendations for the period of the trial; are unable to control their dietary intake; or are advised not to participate by their medical practitioner.

Study design

The design is a randomised, 2 x 2 factorial prospective randomised open with blinded endpoints (PROBE) study of (a) two diets and (b) supplementation with ALA (1.8 g) or placebo (olive oil 2 g). Coronary angiography identifying a qualifying lesion will be performed within three months prior to randomisation. During the four-week run-in phase, patients will be assessed for eligibility. Medical history, medications and usual diet will be recorded and blood tests will be performed during this phase. At Week 0 patients who are eligible to enter the study will be randomised to one of the two diet regimens and independently one of the supplements. Patients will then receive dietary counselling and instruction, and have a physical examination. Patients will be monitored on a three-monthly basis for the duration of the study which will run for 2.5 years. Coronary angiography will be performed within three months prior to commencement and at the conclusion of the study.

Dietary intervention
At Week 0, participants will be randomised to either:

a low fat diet (<30% energy [E], low fat [LF] diet) supplemented with ?-linolenic acid (1.8 g);
a low fat diet (<30% E, LF diet) supplemented with placebo (olive oil 2 g);
a diet rich in mono-unsaturated fat (total fat 35-40% E, Mediterranean diet) supplemented with ?-linolenic acid (1.8 g). MUFA intake to be twice the SFA intake; or
a diet rich in mono-unsaturated fat (total fat 35-40% E, Mediterranean diet) supplemented with placebo (olive oil 2 g). MUFA intake to be twice the SFA intake.
At Week 0, the study nutritionists will provide individual dietary counselling and instruction to enable participants to follow the allocated diet. Nutritionists will provide follow-up dietary counselling and instruction at Weeks 1, 3 and 7 and then every three months for the first 12 months, then every six months for the duration of the study. Dietary instruction will be detailed and personalised to enable participants to adapt their usual diet to the allocated diet without significant change in energy consumption. Prior to each visit, participants will complete a three-day food record, and diet history will be recorded at each visit. Three random 24-hour dietary recalls will be performed each year. Three-day food records, dietary histories, and 24-hour recalls will be analysed using the FoodWorks dietary analysis software (Xyris Software, Brisbane, version 1.05, 1998). In addition, dietary adherence will be monitored biochemically by assessment of fatty acid content of plasma cholesteryl esters.

Blood tests
Prior to randomisation and at three months, the following fasting blood tests will be performed: electrolytes, liver function tests, total cholesterol, triglycerides, HDL, LDL, Lp(a), apoA1 and apoB, fatty acid content of plasma cholesteryl ester analysis and full blood count. Total cholesterol, triglycerides, HDL and LDL will be repeated and results averaged.

At three months and 12 months, the following fasting blood tests will be performed: electrolytes, liver function tests, total cholesterol, triglycerides, HDL and LDL.

At the conclusion of the study, the following fasting blood tests will be performed: electrolytes, liver function tests, total cholesterol, triglycerides, HDL, LDL, apoA1 and apoB and fatty acid content of plasma cholesteryl ester fraction. Total cholesterol, triglycerides, HDL and LDL will be repeated and results averaged. All biochemical measurements will be analysed at a pathology laboratory accredited by the Australian National Testing Authority.

Psychological factors
Standard self-report questionnaires will assess the best documented psychological risk factors for CHD: depression, Beck depression inventory (BD-II) ( 32-34); anxiety, state-trait anxiety inventory (STAI) ( 35, 36); anger, /state-trait anger expression inventory (STAXI) ( 37, 38); social support, social support questionnaire (SSQ) ( 39, 40); explanatory style, attributional style questionnaire (ASQ) ( 41, 42); and hopelessness ( 43).

Coronary angiography
Baseline coronary angiography will be performed using the standard Judkin's technique by one of the study cardiologists at the Wesley and Greenslopes Private Hospitals. Seven-French catheters and non-ionic contrast media will be used. Patients will be given two puffs of glyceryl trinitrate or one anginine at least two minutes prior to the study angiogram. All angiograms will be recorded on cine film at 12.5 frames per second and the basic sequence of views followed and modified for optimal projection

Left Right

i left anterior oblique i LAO 30° cranial 10°
(LAO) 45° cranial
20°

ii right anterior oblique ii RAO 30° caudal 10°
(RAO) 25° caudal
20°

iii anterior posterior
cranial 40°

iv lateral
Other views may be added to visualise optimally all portions of the coronary vessels. At the conclusion of the study, all patients will have a follow-up coronary angiogram. This follow-up measure will use the same protocol and camera angles as the baseline measure.

Clinical endpoints
Coronary angiography
The cine film obtained at angiography will be analysed off-line using a quantitative computerised angiography (QCA) utilising the Cardiovascular Measurement System (CMS-MEDIS) (Medical Imaging Systems, Nuenen, the Netherlands, version 3). Three proximal segments will be analysed in each of the right, circumflex, left main and left anterior descending arteries. Measurements of lesions will be: minimum obstruction diameter in millimetres and as percentage of stenosis, mean segment diameter (millimetres). An angiographic score will be calculated. The most important measurement is minimum obstruction diameter as this relates to coronary flow, and its change is most closely associated with changes in anginal symptoms.

Measurements will be made by an investigator blinded to the subjects' dietary allocation. On exit from the study identical views will be obtained. QCA measurements will be made at the same positions in the arteries for direct comparison over time. Regression and progression are defined as greater than 10% change in stenosis or greater than 0.4 mm change in luminal diameter ( 44). Follow-up angiography will be 2.5 years after the baseline angiogram. Subjects who require coronary angiography for clinical reasons before the end of the study will have these films analysed if greater than one year from baseline. If bypass surgery or angioplasty is performed during the trial period, then arteries affected by these procedures will be excluded from the analysis.

Statistical analysis
Randomisation
Patients will be randomised to either a low fat diet or Mediterranean-type diet in blocks of ten, stratified by baseline level of LDL. Patients will further be randomised to receive ALA or matched placebo, in a 2 x 2 factorial design. This second step in the randomisation procedure will be double-blind so that neither the investigator nor patient is aware which capsule is provided.

Sample size
Primary endpoints are absolute lumen change in coronary lesions and change of calculated percentage of stenosis as measured by coronary angiography. Precision measurements in our laboratory have demonstrated standard deviation of both mild and severe lesions of less than 3% (< 0.1 mm). The power calculation is based on sample size of 150, (180 recruited, with maximum 30 lost to follow-up). The study has 80% power to detect a change in percentage of stenosis of between 2% and 3%. Less than this would not be clinically relevant.

Statistical analysis
All angiograms will be scored by a clinician who is blinded to the patients' treatment allocation. Comparison of group differences in baseline demographic, clinical, risk factor and angiographic measurements will be done using a two sample t-test for continuous variables and a ?² test for categorical variables, The principal endpoint is the average rate of angiographic change. For each patient a percentage and absolute diameter (millimetres) change will be calculated, and this will be divided by the time between the baseline and follow-up angiograms. The between groups difference will be evaluated using a two sample t-test. In a secondary analysis, a multivariate regression analysis will be performed to predict the degree of angiographic change by baseline demographic, clinical, risk factor and baseline angiographic variables plus the group to which the patient was randomised.

Progress
Recruitment began in October 1997. By September 1998 69 eligible patients had been approached and 26 patients had been randomised. The most common reason for refusal to participate was lack of time or inconvenience (43 patients). The next most common (seven patients) was a preference for the Mediterranean diet. Baseline and three-month data on diet analyses (Figures 1 and 2) and serum lipids have been obtained on the first 20 patients. Analysis so far demonstrates good adherence to the allocated diet. In the low fat diet group, mean fat intake was 27% E at baseline, and at three months decreased to 20.3% E. In the Mediterranean diet group mean fat intake at baseline was similar at 26% E. In contrast, at three months the fat intake bad increased to 37% E, with 48.3% of fat being MUFA. Three month mean serum cholesterol in the low fat group was 4.57 mmol/L and in the Mediterranean diet group was 4.44 mmol/L. Diet modification has not had a significant effect on serum lipids at three months, which is not unexpected in patients on cholesterol-towering drugs. Consistent with current best practice, 19 out of 20 are on statins. It is expected that recruitment will continue until the end of 1999, with the final repeat angiogram occurring in mid-2002.

This research is supported by grants from Wesley Research Institute and International Olive Oil Council.

Acknowledgments: the OLIVE investigators
Cardiologists: Assoc Prof David Colquhoun, Dr David Cross, Dr Geoffrey Holt, Dr Mark Dooris, Dr Stephen Cox, Dr Narendra Kewal; Dietitians: Dr Shawn Somerset, Ms Pare Horsley, Ms Cindy Hamill, Mr Andrew McNeil: Pathologist: Dr Bruce Campbell; Endocrinologist: Assoc Prof Richard Jackson: Epidemiologist: Assoc Prof Paul Glasziou; Scientists: Dr Bill Chain, Dr Jeffrey Smith, Dr Justin Westhuyzen: Renal Physician: Dr David Saltissi; Psychologist: Ms Jeannet Weyers; Study Coordinators and Data Managers: Ms Debbie Richards; Ms Donna Griffiths, Ms Mary Anne Mackey; Consultants: Ms Rosemay Stanton, Prof Curt Ellison.

GRAPH: Figure 1. Dietary analysis of total fat intake at baseline and at three months for low fat (LF) and Mediterranean (Med) diets

GRAPH: Figure 2. Dietary analysis of type of fat intake at baseline and at three months for low fat (LF) and Mediterranean (Med) diets

References
(1.) Brown BG, Zhao XQ, Sacco DE, Albers JJ. Arteriographic view of treatment to achieve regression of coronary atherosclerosis and to prevent plaque disruption and clinical cardiovascular events. Br Heart J 1993;69(1 Suppl):S48-S53.

(2.) American Heart Association medical/scientific statement. Dietary guidelines for healthy American adults; a statement for health professionals from the nutrition committee, American Heart Association. Circulation 1996;94:1795-800.

(3.) Arntzenius AC, Kromhout D, Barth JD, Reiber JH, Bruschke AV, Buis B, et al. Diet, lipoproteins and the progression of coronary atherosclerosis: the Leiden Intervention Trial. N Engl J Med 1985;312:805-11.

(4.) Ornish D, Brown SE, Scherwitz LW, Billings JH, Armstrong WT, Ports TA, et al. Can lifestyle changes reverse coronary heart disease? The Lifestyle Heart Trial. Lancet 1990;336:129-33.

(5.) Schuler G, Hambrecht R, Schlierf G, Niebauer J, Hauer K, Neumann J, et al. Regular physical exercise and low fat diet. Effects on progression of coronary artery disease. Circulation 1992;86:1-11.

(6.) Blankenhorn DH, Johnson RL, Mack WJ, el Zein HA, Vailas LI. The influence of diet on the appearance of new lesions in human coronary arteries. JAMA 1990;263:1646-52.

(7.) Watts GF, Jackson P. Mandalia S, Brunt JN, Lewis ES, Coltart DJ, et al. Nutrient intake and progression of coronary artery disease. Am J Cardiol 1994;72:328-32.

(8.) Watts GF, Lewis B, Brunt J, Lewis, ES, Coltart DJ, Smith LD, et al. Effects on coronary artery disease of lipid lowering diets, or diet plus cholestyramine, in the St. Thomas' Atherosclerosis Regression Study (STARS). Lancet 1992;339:563-9.

(9.) Haskell WL, Alderman EL, Fair JM, Maron DJ, Mackey SF, Superko HR, et al. Effects of intensive multiple risk factor reduction on coronary atherosclerosis and clinical cardiac events in men and women with coronary artery disease. The Stanford Coronary Risk Intervention Project (SCRIP). Circulation 1994;89:975-90.

(10.) Ferro-Luzzi A, Sette S. The Mediterranean diet: an attempt to define its present and past composition. Eur J Clin Nutr 1989;43(2):13-29.

(11.) Keys A, Menotti A, Aravanis C, Blackburn H, Djordevic BS, Buzina R, et al. The seven countries study: 2289 deaths in 15 years. J Prev Med 1984;13:141-54.

(12.) Tunstall-Pedoe H, Kuulasmaa K, Amouyel P. Armeiler D, Rajakangas AM, Pajak A, et al. Myocardial infarction and coronary deaths in the World Health Organization MONICA Project: registration procedures, event rates, and case fatality rates in 38 populations from 21 countries in four continents. Circulation 1994;90(1):583-612.

(13.) Visioli F, Galli C. The effect of minor constituents of olive oil on cardiovascular disease: new findings. Nutr Rev 1998;56(5)142-7.

(14.) Olias JM, Perez AG. Rios JJ, Sanz LC. Aroma of virgin olive oil: biogenesis of the 'green' odor notes. J Agric Food Chem 1993:41:2368-73.

(15.) Colquhoun DM, Hicks BJ, Reed AW, Phenolic content of olive oil is reduced in extraction and refining: analysis of phenolic content of three grades of olive and ten seed oils. Asia Pac J Clin Nutr 1996;5:106-7.

(16.) Carr TP, Sawyers NK, Rudel LL. Dietary monounsaturated fat protects against coronary artery atherosclerosis in African green monkeys. Circ 1993;88(4):3030.

(17.) Nicolosi RJ, Courtemanche KV, Behr SR. Increased LDL oxidation susceptibility and enhanced aortic atherogenesis in hamsters fed high polyunsaturated vs. monounsaturated vegetable oils, Circ 1993;88(4):3032.

(18.) Prasad K. Dietary flax seed in prevention of hypercholesterolemic atherosclerosis. Atherosclerosis 1997;132:69-76.

(19.) Holme I. Relation of coronary heart disease incidence and total mortality to plasma cholesterol reduction in randomised trials: the use of meta-analysis. Br Heart J 1993:69(1 Suppl):42-7.

(20.) de Lorgeril M, Salon P, Martin JL, Mamelle N, Monjaud L Touboul P, et al. Mediterranean diet and cardioprotective nutrients. JACC 1996;28:1103-8.

(21.) de Lorgeril M, Renaud S, Mamelle N, Salen P, Martin JL, Monjaud I, et al. Mediterranean alpha-linolenic acid rich diet in secondary prevention of coronary heart disease. Lancet 1994;343:1454-9.

(22.) Scandinavian Simvastatin Survival Study Group, Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin survival study (4S). Lancet 1994:344:1383-9.

(23.) Sacks FM, Pfeffer MA, Moye LA, Rouleau JL, Rutherford JD, Cole TG, et al. The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. N Engl J Med 1996:335:1001-9.

(24.) Long-term intervention with Pravastin in ischaemic disease (LIPID) trial. New Eng J Med 1998;339:1349 57.

(25.) Roberts WC. The ineffectiveness of a commonly recommended lipid-lowering diet in significantly lowering the serum total and low-density lipoprotein cholesterol levels. Am J Cardiology 1994;73:623-4.

(26.) Collins R, Keech A, Peto R, Sleight P, Kjekshus J, Wilhelmsen L, el al. Cholesterol and total mortality: need for larger trials. BMJ 1992;304:1689.

(27.) Rousseau JE. Lipid lowering interventions in angiographic trials. Am J Cardiol 1995;76:86C-92C.

(28.) Buchwald H, Matts JP, Fitch LL, Campos CT, Sanmarck ME, Amplatz K, et al. Changes in sequential coronary arteriograms and subsequent coronary events. JAMA 1992;268:1429-33.

(29.) Blankenhorn DH, George Lyman and Duff Memorial Lecture: arterial imaging and atherosclerosis reversal. Arteriosel Thromb 1994;14:177-92.

(30.) Waters D, Craven TE, Lesperance J. Prognostic significance of progression of coronary atherosclerosis. Circulation 1993:87:1067-75,

(31.) Linden W, Stossel C, Maurice J. Psychosocial interventions for patients with coronary artery disease, Arch I Med. 1996;156:745-52.

(32.) Beck AT. Beck depression inventory manual. Australia: The Psychological Corporation, Harcourt Brace, 1996.

(33.) Tanaka-Matsumi J, Kameoka VA. Reliabilities and concurrent validities of popular social desirability. J Consult Clin Psychol 1986;54(3):328-33.

(34.) Beck AT, Steer RA, Garvin M. Psychometric properties of the Beck depression inventory: Twenty five-year evaluations. Clin Psychol Rev 1988;8:77-100.

(35.) Speilberger CD. State-trait anxiety inventory for adults. Palo Alto, California: Mind Garden. 1983.

(36.) Speilberger CD. Sydeman SJ. State-trait anxiety inventory and state-trait anger expression inventory. In: Matuish ME, editor. The use of psychological tests for treatment planning and outcome assessment. Hillsdale, NJ: LEA, 1994.

(37.) Speilberger CD. State-trait anger expression inventory (STAXI) professional manual. Florida: Psychological Assessment Resources Inc. 1996.

(38.) Speilberger CD, Sydeman SJ. State-trait anxiety inventory and state-trait anger expression inventory. In Matuish ME, editor. The use of psychological tests for treatment planning and outcome assessment. Hillsdale, NJ: LEA, 1994.

(39.) Samson IG, Sarason RB, Shearin EN. A brief measure of social support: the social support questionnaire. J Pers Social Psychol 1983;4:127-39.

(40.) McDowall I, Newall C. Measuring healthy: a guide to rating scales and questionnaires. New York: Oxford University Press, 1996.

(41.) Golin S, Sweeney PD, Schaeffer DE. The causality of causal attributions in depression: a cross-lagged panel correlational analysis. J Abnorm Psychol 1981;90:14-22.

(42.) Sweeney PD, Anderson K, Bailey S. Attributional styles in depression: a meta-analytic review. J Personal Soc Psychol 1986;50:974-91.

(43.) Everson SA, Goldberg DE, Kaplan GA, Cohen RD, Pukkala E, Tuomilchto J, el al. Hopelessness and risk of mortality and incidence of myocardial infarction and cancer. Psychosomatic Med 1996;58:113-21.

(44.) de Groot E, Jukema JW, van Boron AJ, Reiber JH, Zwinderman AH, Lie KI, et al. Effect of Pravastatin on progression and regression of coronary atherosclerosis and vessel wall changes in carotid and femoral arteries: a report from the Regression Growth Evaluation Statin Study (REGRESS). Am J Cardiol 1995;76:40C-6C.

~~~~~~~~

By David M. Colquhoun; Shawn Somerset; Paul Glasziou; Debbie Richards and Jeannet Weyers; For the OLIVE study investigators

D. Colquhoun, MBBS, FRACP, Director of Cardiovascular Research, Wesley and Greenslopes Private Hospitals, Brisbane, Queensland. Correspondence: D.M. Colquhoun, The Wesley Medical Centre, 40 Chasely Street, Auchenflower Qld 4066

S. Somerset, PhD, BSc(Hons), DipNutDiet, MDAA, Research Nutritionist/Dietitian, Wesley and Greenslopes Private Hospitals, Brisbane, Queensland

P. Glasziou, MBBS, PhD, FRAPHM, Associate Professor in Clinical Epidemiology, Wesley and Greenslopes Private Hospitals, Brisbane, Queensland; Department of Social and Preventive Medicine, University of Queensland

D. Richards, BSc, DipNatrDiet, GradDip Health Promotion, MDAA, Research Co-ordinator, Wesley and Greenslopes Private Hospitals, Brisbane, Queensland

J. Weyers, BPsych (Hons), Research Fellow, Wesley and Greenslopes Private Hospitals, Brisbane, Queensland

Share this with your friends