ORIGINAL INVESTIGATION Gemfibrozil in the Treatment of Dyslipidemia An 18-Year Mortality Follow-up of the Helsinki Heart Study Leena Tenkanen, PhD; Matti Ma¨ntta¨ri, MD; Petri T. Kovanen, MD; Hanna Virkkunen, MSc; Vesa Manninen, MD Background: The Helsinki Heart Study was a double- Cox proportional hazards models with and without blind, placebo-controlled primary prevention trial among covariates. 4081 dyslipidemic middle-aged men to test the efficacy of gemfibrozil in the prevention of coronary heart dis- Results: During the follow-up until 1995, subjects in the ease (CHD). After the 5-year trial, the participants were OG group had a 32% lower RR of CHD mortality (P=.03) notified of their treatment group and invited to con- compared with those in the OP group, and when followed tinue or start gemfibrozil therapy free of charge through up until 2000, the RR was 23% lower (P=.05). Overall, there 1995. Approximately two thirds of participants in both were no differences in all-cause or cancer mortality. How- groups chose gemfibrozil therapy. In this 18-year fol- ever, those in the OG group with both body mass index low-up through 2000, we compared the CHD, cancer, and triglyceride level in the highest tertiles had a 71% lower Ͻ and all-cause mortality among subjects in the original gem- RR of CHD mortality (P .001), a 33% lower RR of all- fibrozil (OG) group (n=2046) with those in the origi- cause mortality (P=.03), and a 36% lower RR of cancer mor- nal placebo (OP) group (n=2035). tality (P=.22) compared with those in the OP group. Conclusion: Long-term mortality follow-up showed that Methods: To provide an overview of the absolute risks patients with dyslipidemia benefited from beginning treat- in the 2 treatment groups as well as risk differences ment with gemfibrozil early, especially if their dyslipid- between them, we calculated crude mortality rates and emia entailed factors related to the metabolic syndrome. presented Kaplan-Meier plots of survival with log-rank tests. We also estimated the relative risks (RRs) using Arch Intern Med. 2006;166:743-748 HE HELSINKI HEART STUDY from the treatment was characterized by was a 5-year primary pre- a high baseline triglyceride (TG) level vention trial to test the hy- and an elevated ratio of LDL-C to pothesis that gemfibrozil HDL-C.5 Further analyses indicated that use reduces coronary heart gemfibrozil use reduced CHD risk espe- Tdisease (CHD) incidence by lowering low- cially in overweight subjects with addi- density lipoprotein cholesterol (LDL-C) tional risk factors known to contribute and increasing high-density lipoprotein or predispose to the metabolic syn- cholesterol (HDL-C) levels. During the drome.6 The effect of gemfibrozil therapy Author Affiliations: Helsinki original 5-year, double-blind phase of the in subjects with insulin resistance was Heart Study, Helsinki, Finland trial, a 34% reduction in cardiac end points, later confirmed in the Veterans Affairs (Drs Tenkanen, Mänttäri, and but no difference in all-cause mortal- High Density Lipoprotein Intervention Manninen and Ms Virkkunen); 1,2 7 School of Public Health, ity, was seen. When the 5-year trial Trial (VA-HIT). Tampere University, Tampere, ended, the participants in both treatment Finland (Dr Tenkanen and groups were offered gemfibrozil for the fol- Ms Virkkunen); Helsinki lowing 3.5 years. At the end of the 3.5- See also pages 715 and 737 University Jorvi Hospital, year open-label phase, the effect of long- Espoo, Finland (Dr Ma¨ntta¨ri); term vs short-term treatment with The present study is an 18-year fol- Helsinki University Central gemfibrozil was studied. No material dif- low-up of the original gemfibrozil (OG) and Hospital, Helsinki ferences in all-cause or coronary mortal- original placebo (OP) groups of the Hel- (Dr Manninen); and Wihuri ity were seen between the original treat- sinki Heart Study to examine the effects of Research Institute, Helsinki ment groups,3 whereas cancer mortality a 5-year earlier start of gemfibrozil treat- (Drs Ma¨ntta¨ri and Kovanen). was slightly (nonsignificantly) higher in ment on all-cause, CHD, and cancer mor- Group Information: 4 Drs Tenkanen, Mänttäri, and the original gemfibrozil group. tality. We also investigated whether the Manninen and Ms Virkkunen Subsequent subgroup analyses of the findings from the subgroup analyses of are members of the Helsinki double-blind trial data showed that the the double-blind trial phase would persist Heart Study Group. group that obtained the greatest benefit in the long-term analysis of mortality. (REPRINTED) ARCH INTERN MED/ VOL 166, APR 10, 2006 WWW.ARCHINTERNMED.COM 743 ©2006 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/01/2021 Free Gemfibrozil A OG Group Group Double Blind Open Label Treatment 2.20 OP Group Placebo Gemfibrozil (n = 1320) OP 2.00 (n = 2035) No Drug (n = 608) 1.80 1.60 Gemfibrozil Gemfibrozil (n = 1301) OG 1.40 (n = 2046) No Drug (n = 660) 1.20 TG Level, mmol/L 8.5-y Follow-up 1.00 13-y Follow-up 18-y Follow-up B Clinical Follow-up 1.40 Register-Based Follow-up 1.35 1.30 1981-1982 1986-1987 1990 1995 2000 1.25 Calendar Year 1.20 HDL-C Level, mmol/L 1.15 Figure 1. Schematic presentation of the different phases of the 18-year follow-up of the Helsinki Heart Study. OG indicates original gemfibrozil; OP, original placebo. C 5.20 METHODS 5.00 4.80 DIFFERENT PHASES 4.60 OF THE HELSINKI HEART STUDY 4.40 LDL-C Level, mmol/L 4.20 Double-blind Trial Baseline 1 2 3 4 5 6 7 8 The Helsinki Heart Study was a 5-year, double-blind clinical trial Years to test the hypothesis that lowering serum LDL-C and triglyc- Double Blind Open Label eride levels and elevating serum HDL-C levels with gemfibrozil 1981-1982 1986-1987 1990 reduces CHD risk in middle-aged dyslipidemic men. The de- Calendar Year sign, conduct, and main results of the study have been de- 1,2 scribed in detail. Briefly, the volunteers for the trial were se- Figure 2. Development of annual mean values of triglyceride (TG) (A), lected in 1980 through 1982 by 2 successive screenings from men high-density lipoprotein cholesterol (HDL-C) (B), and low-density lipoprotein aged 40 to 55 years, employed by 2 government agencies and 5 cholesterol (LDL-C) (C) in the original placebo (OP) and original gemfibrozil industrial companies. To be eligible for the second screening or (OG) groups during the double-blind and open-label trials. To convert to for the trial, the subjects’ non–HDL-C level had to be 200 mg/dL milligrams per deciliter, divide by 0.0113 for TG and by 0.0259 for HDL-C and LDL-C. (5.2 mmol/L) or higher, and for the trial they also had to be with- out evidence of CHD or any other major illness. Of 18 936 men tality by linkage to population-based registers, especially the at the first screening, 4081 met these criteria and were willing Cause-of-Death Register kept by Statistics Finland, Helsinki. In to participate in the trial. Of these, 2035 were randomized to re- Finland, there is a centralized nationwide population registra- ceive placebo and 2046 to receive gemfibrozil, 1200 mg/d, 2 cap- tion system based on personal identification numbers, kept since sules twice daily. Compliance with medication was studied by 1966. This system ensures that the registration of deaths is vir- 8 several methods (eg, capsule counting). tually complete. The Cause-of-Death Register was found to be very reliable in terms of CHD deaths compared with the World Open-Label Phase Health Organization Monitoring of Trends and Determinants in Cardiovascular Disease (WHO MONICA) project data9 and the At the end of the 5-year trial, all subjects (including dropouts) were Helsinki Heart Study data.10 informed about their lipid values during the trial and invited to par- ticipate in an open 3.5-year trial extension with 2 annual visits and Present Mortality Follow-up a choice to take gemfibrozil (Figure 1). About the same propor- tion of subjects in the 2 original treatment groups chose to take This study extends the follow-up from the beginning of the double- gemfibrozil: 66.3% in the OG group and 68.5% in the OP group. blind trial in 1980-1982 through 2000 (18-year follow-up) and Compliance monitoring showed that in both groups the partici- is fully register based. The following CHD codes were moni- 3 pant took, on average, 3.5 capsules of gemfibrozil per day. tored: International Classification of Diseases, Eighth Revision codes 410 to 414 through 1986; International Classification of Diseases, Register-Based Follow-up Phase Ninth Revision codes 4100 to 4199 from 1987 through 1995; and International Statistical Classification of Diseases, 10th Revision codes At the end of the open-label phase in June 1990, subjects taking I20 to I25 from 1996 through 2000. These codes imply the pres- gemfibrozil were offered a free drug supply through a central phar- ence of 1 or several of the following conditions: angina, myocar- macy through 1995, but no regular follow-up visits. The trial co- dial infarction, myocardial infarction with complications, other hort was then followed up for all-cause, CHD, and cancer mor- ischemic heart diseases, and chronic ischemic heart disease. (REPRINTED) ARCH INTERN MED/ VOL 166, APR 10, 2006 WWW.ARCHINTERNMED.COM 744 ©2006 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/01/2021 Table 1.