Intensive Lifestyle Changes for Reversal of Coronary Heart Disease Dean Ornish, MD; Larry W. Scherwitz, PhD; James H. Billings, PhD, MPH; K. Lance Gould, MD; Terri A. Merritt, MS; Stephen Sparler, MA; William T. Armstrong, MD; Thomas A. Ports, MD; Richard L. Kirkeeide, PhD; Charissa Hogeboom, PhD; Richard J. Brand, PhD Context.—The Lifestyle Heart Trial demonstrated that intensive lifestyle THE LIFESTYLE Heart Trial was the changes may lead to regression of coronary atherosclerosis after 1 year. first randomized clinical trial to investi- Objectives.—To determine the feasibility of patients to sustain intensive lifestyle gate whether ambulatory patients could changes for a total of 5 years and the effects of these lifestyle changes (without be motivated to make and sustain com- lipid-lowering drugs) on coronary heart disease. prehensive lifestyle changes and, if so, whether the progression of coronary Design.—Randomized controlled trial conducted from 1986 to 1992 using a atherosclerosis could be stopped or re- randomized invitational design. versed without using lipid-lowering Patients.—Forty-eight patients with moderate to severe coronary heart disease drugs as measured by computer-as- were randomized to an intensive lifestyle change group or to a usual-care control sisted quantitative coronary arteriogra- group, and 35 completed the 5-year follow-up quantitative coronary arteriography. phy. This study derived from earlier Setting.—Two tertiary care university medical centers. studies that used noninvasive mea- Intervention.—Intensive lifestyle changes (10% fat whole foods vegetarian diet, sures.1,2 aerobic exercise, stress management training, smoking cessation, group psycho- After 1 year, we found that experi- social support) for 5 years. mental group participants were able to Main Outcome Measures.—Adherence to intensive lifestyle changes, changes make and maintain intensive lifestyle changes and had a 37.2% reduction in in coronary artery percent diameter stenosis, and cardiac events. low-density lipoprotein (LDL) choles- Results.—Experimental group patients (20 [71%] of 28 patients completed terol levels and a 91% reduction in the 5-year follow-up) made and maintained comprehensive lifestyle changes for 5 frequency of anginal episodes.3 Average years, whereas control group patients (15 [75%] of 20 patients completed 5-year percent diameter stenosis regressed follow-up) made more moderate changes. In the experimental group, the average from 40.0% at baseline to 37.8% 1 year percent diameter stenosis at baseline decreased 1.75 absolute percentage points later, a change that was correlated with after 1 year (a 4.5% relative improvement) and by 3.1 absolute percentage points the degree of lifestyle change. In con- after 5 years (a 7.9% relative improvement). In contrast, the average percent diam- trast, patients in the usual-care control eter stenosis in the control group increased by 2.3 percentage points after 1 year group made more moderate changes in (a 5.4% relative worsening) and by 11.8 percentage points after 5 years (a 27.7% lifestyle, reduced LDL cholesterol lev- els by 6%, and had a 165% increase in the relative worsening) (P = .001 between groups. Twenty-five cardiac events occurred frequency of reported anginal episodes. in 28 experimental group patients vs 45 events in 20 control group patients during Average percent diameter stenosis pro- the 5-year follow-up (risk ratio for any event for the control group, 2.47 [95% con- gressed from 42.7% to 46.1%. fidence interval, 1.48-4.20]). Given these encouraging findings, Conclusions.—More regression of coronary atherosclerosis occurred after 5 we extended the study for an additional years than after 1 year in the experimental group. In contrast, in the control group, 4 years to determine (1) the feasibility coronary atherosclerosis continued to progress and more than twice as many car- of patients sustaining intensive changes diac events occurred. in diet and lifestyle for a much longer JAMA. 1998;280:2001-2007 time, and (2) the effects of these changes on risk factors, coronary atherosclero- sis, myocardial perfusion, and cardiac From the Department of Medicine (Dr Ornish), and Division of Cardiology, University of Texas Medical events after 4 additional years. the Division of Cardiology (Dr Armstrong), California School, Houston (Drs Gould and Kirkeeide); and the Pacific Medical Center, San Francisco; the Department Preventive Medicine Research Institue, Sausalito, Calif of Medicine (Dr Ornish), the Division of Cardiology, (Drs Ornish Scherwitz, and Billings, Mr Sparler, and METHODS Cardiac Catheterization Laboratory, Cardiovascular Ms Merritt). The design, recruitment, and study Research Institute (Dr Ports), and the Division of Reprints: Dean Ornish, MD, Preventive Medicine Re- 3-5 Biostatistics (Drs Brand and Hogeboom), School of search Institute, 900 Bridgeway, Suite 1, Sausalito, CA population were previously described. Medicine, University of California, San Francisco; the 94965 (e-mail: [email protected]). In brief, we recruited men and women JAMA, December 16, 1998—Vol 280, No. 23 Lifestyle Heart Trial—Ornish et al 2001 ©1998 American Medical Association. All rights reserved. Table 1.—Baseline Characteristics of Experimental and Control Groups* form after being fully informed of the Experimental Control P study requirements. Characteristic (n = 20) (n = 15) Value Patients completed a 3-day diet diary Men, No. 20 12 at baseline and after 1 and 5 years to .07 Women, No. 0 3 assess nutrient intake and dietary ad- 6 Age, mean (SD), y 57.4 (6.4) 61.8 (7.5) .08 herence. Methods of lipid assays were 3 Education, mean (SD), y 15.5 (2.7) 14.5 (3.4) .29 the same as previously reported. These Employed, No. 14 6 .10 3-day diet diaries were analyzed with a Body mass index, mean (SD), kg/m2 28.4 (4.1) 25.4 (3.5) .03 software package (CBORD Diet Ana- No. with history of myocardial infarction 12 5 .17 lyzer; CBORD Group Inc; Ithaca, NY) Average No. of lesions studied, mean (SD) 5.3 (2.7) 5.3 (3.2) .93 using the US Department of Agricul- No. with history of percutaneous transluminal 54..99 turedatabase.Also,patientswereasked coronary angioplasty to complete a questionnaire reporting No. with history of coronary artery bypass graft 1 0 ..99 the frequency and duration of exercise Reported angina, No. (%) 11 (55) 6 (40) .49 and of each stress management tech- *Values are statistics unless otherwise indicated. P values are 2-tailed. nique. Information from these sources wasquantifiedintocontinuousscoresus- with coronary atherosclerosis docu- to 5-year comparisons. ing an a priori determined formula. The mented by quantitative coronary arte- Four experimental and 4 control pa- adherence measure was a continuous riography. tients who had an angiogram at 1 year did score reflecting daily intake of choles- We identified 193 patients as poten- not have a third angiogram after 5 years. terol (in milligrams), fat (in grams), tially eligible for our study who agreed Three of these 4 patients in the experi- frequency and duration of exercise, fre- to undergo quantitative coronary angi- mental group refused a third angiogram quency and duration of stress manage- ography. Following angiography, 93 (patients only volunteered for a 1-year ment techniques, and smoking. A score patients remained eligible and were ran- study that was subsequently extended), of 1.0 equalled 100% adherence but domly assigned to experimental or con- and 1 died between years 1 and 4; of the 4 scores could be greater than 1.0 if par- trol groups using a randomized invita- control group patients who did not un- ticipants exceeded the recommended in- tional design to minimize crossover, dergo a third angiogram, 1 died, 2 under- tensive lifestyle changes. ethical concerns, nocebo effects, and went revascularization of the arterial The technicians responsible for per- dropout. Of these 93 patients who were lesions under study, and 1 developed forming all medical tests were blinded to eligible, 53 were randomly assigned to Parkinson disease and became too ill to patient group assignment. Also, different the experimental group and 40 to the be safely tested. Cine arteriograms made personnel implemented the lifestyle in- usual-care control group. Patients were in San Francisco, Calif, were sent to the tervention, conducted the tests, and com- thencontactedandinvitedtoparticipate University of Texas Medical School, putedstatisticalanalyses,althoughthedi- in the study; 28 (53%) and 20 (50%) Houston, for blinded quantitative analy- etitian was made aware of the nutrient agreed to participate in the experimen- ses as previously described in detail.4 analysis to monitor patients’ safety and tal and control groups, respectively. The All results, except lesion changes at 1 adherence. Quantitative coronary arte- primary reason for refusal in the experi- year (18 experimental and 15 control riograms were blindly analyzed without mental group was not wanting to un- subjects) and cardiac events after 5 knowledge of group assignment. dergo intensive lifestyle changes and/or years (all 28 experimental and 20 control not wanting a second coronary angio- subjects), are based on the total of 35 Program Intervention gram; control patients refused primarily patients (20 experimental and 15 control Experimental group patients were because they did not want to undergo a subjects) who had both baseline and 5- prescribed an intensive lifestyle pro- secondangiogram.Todetectpossiblese- year angiograms. From these 35 pa- gram that included