How To Lower Triglycerides And Increase Hdl

Epidemiological evidence has suggested that high-density lipoprotein cholesterol (HDL-C) levels are inversely correlated with coronary artery disease (CAD) and are independently predictive of cardiovascular morbidity and mortality.¹ ² ³ ⁴ In many prospective studies, triglyceride levels were predictive of CAD in univariate and multivariate analyses controlling for total cholesterol or low-density lipoprotein cholesterol (LDL-C) levels.⁵ ⁶ ⁷ Although in some analyses this association did not persist after adjustment for HDL-C,⁸ the predictive value of high triglycerides has been confirmed in a meta-analysis.⁹ Nevertheless, direct evidence for the clinical benefit of elevating HDL-C or reducing blood triglyceride levels is scarce because the efficacy of lipid-modifying drugs that lower triglyceride levels and raise HDL-C levels had not been directly assessed in large-scale clinical trials in CAD patients. In the Coronary Drug Project,¹⁰ nicotinic acid, which elevates HDL-C and lowers LDL-C, decreased coronary morbidity; however, HDL-C measurement was not part of the study protocol. An analysis of the joint effects of baseline triglyceride and lipoprotein cholesterol levels, conducted in the framework of the primary prevention Helsinki Heart Study, demonstrated a strong interdependence of LDL-C, HDL-C, and triglycerides as predictors of CAD risk and a beneficial effect of treatment with gemfibrozil.¹¹

The relation of serum total cholesterol and LDL-C with the development and progress of atherosclerosis and CAD has been demonstrated in numerous clinical and epidemiological studies. Moreover, the benefit of LDL-C reduction has now been strongly supported by the significant decrease in cardiovascular events, including cardiovascular mortality, achieved by reducing LDL-C with hepatic hydroxymethylglutaryl coenzyme A reductase inhibitors (statins).¹² ¹³ ¹⁴ ¹⁵ ¹⁶ However, in the Cholesterol and Recurrent Events¹⁴ study, this benefit was not observed in the subgroup of patients with baseline LDL-C concentrations <125 mg/dL. About 40% of patients with CAD have LDL-C levels <130 mg/dL, and most of these patients also have low levels of HDL-C, with or without increased triglyceride levels. In a previous publication from the large Bezafibrate Infarction Prevention (BIP) Registry population, we found that 25% of CAD patients had cholesterol levels <200 mg/dL (mean LDL-C, 117 mg/dL); among these patients, more than half had HDL-C levels <35 mg/dL. In addition, 17% of patients with CAD in the BIP Registry had combined low HDL-C (<35 mg/dL) and high triglyceride levels (≥200 mg/dL).¹⁷

The BIP study¹⁸ was designed and initiated in 1990. The primary question of the trial was whether bezafibrate, which raises HDL-C and reduces triglycerides, would reduce CAD mortality and nonfatal myocardial infarction (MI) in patients with established CAD, HDL-C <45 mg/dL, and moderately elevated cholesterol. Recent studies on the mode of action of fibrates indicate that some of these effects are mediated via the peroxisome proliferator-activated receptor pathway,¹⁹ which alters the transcription rate of genes encoding for proteins that control lipoprotein metabolism. The triglyceride-lowering effect is thus linked to an induction of lipoprotein lipase–mediated lipolysis and to lowered apoC-III production, and the HDL-increasing effect is due to an induction in the synthesis of apoAI and apoAII.²⁰ ²¹ ²²

Methods

Study Design and Patients

Between February 1990 and October 1992, 15 524 male and female patients with CAD aged 45 to 74 years were screened for inclusion in the BIP study in 18 of the 25 cardiology departments in Israel. After an initial 2 months on a lipid-lowering diet, 3122 patients who fulfilled the inclusion criteria were randomized to the study between May 1990 and January 1993. A total of 32 of the randomized patients, who were equally distributed between the placebo and bezafibrate treatment groups, never began study medication and were excluded from analysis. Inclusion criteria for men and women comprised the following: age of 45 to 74 years, history of MI ≥6 months but <5 years before enrollment into the study and/or stable angina pectoris confirmed by coronary angiography, and/or radionuclear studies or standard exercise tests. In addition, a lipid profile of serum total cholesterol between 180 to 250 mg/dL, LDL-C ≤180 mg/dL (≤160 mg/dL for patients <50 years), HDL-C ≤45 mg/dL, and triglycerides ≤300 mg/dL was required.

The main exclusion criteria were insulin-dependent diabetes mellitus, severe heart failure, unstable angina pectoris, hepatic or renal failure, known sensitivity to bezafibrate, or current use of lipid-modifying drugs.¹⁸

Patients were assigned consecutive randomization numbers within each recruiting center after giving written informed consent. They were allocated to receive either 400 mg of bezafibrate retard or placebo once a day, in addition to dietary advice. Patients were allowed to take prescribed medications for cardiac and other conditions except for lipid-lowering drugs. Lipid profiles, fibrinogen levels, and safety parameters were measured in the Central Laboratory at randomization, at 4 months, and annually thereafter until the end of the study. Additional details of the study design and the patients' baseline characteristics have been described elsewhere.¹⁸ ²³

Routine visits to the clinics were scheduled bimonthly for study medication distribution and compliance assessment by tablet count and every 4 months for clinical evaluation. Compliance was further assessed by annual measurements of alkaline phosphatase. During the 4-month visit, data on any adverse events (as defined in the study protocol), hospitalizations, and study outcomes were obtained. Study medication was withdrawn after the following: (1) a primary end point, (2) an adverse event deemed to be intolerable, (3) an increase in LDL-C to >210 mg/dL (or >190 mg/dL for patients aged <50 years) or triglycerides >500 mg/dL, or (4) safety variables exceeded predefined critical limits. All study participants, regardless of whether they continued to take the trial medication, were followed-up until the last patient had completed 5 years of follow-up.

In July 1994, after the publication of the Scandinavian Simvastatin Survival Study results,¹⁵ the International Review and Advisory Board approved the recommendation of the Steering Committee to add colestipol for patients on study medication if their LDL-C exceeded 180 mg/dL in 2 separate laboratory examinations after reinforcement of dietary advice. Colestipol was given concomitantly with the study medication to 165 patients (57 patients in the bezafibrate and 107 in the placebo group) during the study.

The trial was conducted independently of the sponsor (Boehringer Mannheim GmbH, which is now part of F. Hoffmann-La Roche, Ltd), and it was approved by the Helsinki Committees of each center and the central national Helsinki Committee.

Classification and Review of Study End Points

The primary end point of the study was fatal MI, nonfatal MI, or sudden death (occurring within 24 hours of onset of symptoms).¹⁸

Secondary end points, for patients free of primary end points, included hospitalization for unstable angina, percutaneous transluminal coronary angioplasty, and coronary artery bypass grafting. Stroke and death from any cause were also monitored. An independent Critical Event Committee, whose members were blinded to the treatment assignment, reviewed primary end points and all-cause mortality.

An independent International Review and Advisory Board regularly monitored the progress of the study and the incidence of adverse events. Two scheduled interim analyses were performed 4 and 5.5 years after the randomization of the first patient.

Laboratory Methods

Blood samples, which were collected in the 18 participating medical centers using standardized equipment and procedures, were transferred in cooled containers to the Central Laboratory at the Institute of Physiological Hygiene Laboratory at the E. Wolfson Medical Center, Holon. Blood samples were drawn after ≥12 hours of fasting to determine serum levels of cholesterol, HDL-C, triglycerides, and plasma fibrinogen. Laboratory measurements were performed using standard automated procedures with commercially available kits (Roche Diagnostics). HDL-C was measured by precipitation, and LDL was estimated using Friedewald et al's equation.²⁴ Fibrinogen was measured by an automated kinetic method. Accuracy and precision of lipid and lipoprotein determinations were under periodic surveillance by the Centers for Disease Control/National Heart, Lung, and Blood Institute's Lipids Standardization Program; other determinations, including the safety variables, were under surveillance by the Wellcome-Murex Diagnostic Clinical Chemistry Quality Assessment Program.

Statistical Analysis

The study design a priori assumed a cumulative event rate of 16% to 24% in the placebo arm of the study over 6 years and an expected reduction of the event rate of between 20% and 25%. Under these assumptions and using a 1-sided test, as originally planned, a sample size of between 2100 and 3300 would have provided a power of 80% to detect the expected reduction. However, during the course of the study, a decision was made to perform 2-sided rather than 1-sided statistical tests in light of the results of the Helsinki Heart Study II. Furthermore, the cumulative incidence of the primary end point under placebo turned out to be lower than expected. Under these circumstances, the randomization of 3000 patients provides a power between 62% and 85% to detect a 20% to 25% reduction in incidence rate with bezafibrate (α=0.05, 2-sided) when the cumulative incidence of the primary end point is 15%, as was observed in the placebo group.

Data were analyzed using SAS software.²⁵ All patients who took the study medication at least once (n=3090) were included in the intent-to-treat analysis. Baseline characteristics in the 2 study groups were compared using the χ² test for dichotomous parameters and the t test for continuous variables. Changes in laboratory parameters were calculated as the difference between the baseline value (measured before administration of the study medication) and the mean of the values measured in the annual laboratory examinations before the occurrence of a primary end point or during the entire follow-up period for patients free of a primary end point.

The cumulative probability of events was computed using the Kaplan-Meier life-table method. The curves of cumulative probability of event for patients in the placebo and the bezafibrate groups were compared using the log-rank test.

To determine which factors affected primary end points in subgroups of patients by high and low baseline triglyceride and HDL-C levels and by using the cut points recommended by the expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (National Cholesterol Education Program),²⁶ multivariate analyses were performed using Cox's proportional hazard stepwise regression modeling.

Results

Although 3122 patients were randomized to the study, 32 patients were excluded from analysis because they never started the study medication. Patients in the placebo (n=1542) and bezafibrate (n=1548) groups were well balanced in terms of clinical and laboratory baseline characteristics and concomitant medications (Table 1). The number of patients with prior MI (62% in the bezafibrate group and 61% in the placebo group) or angina (38% and 39% in the bezafibrate and placebo groups, respectively) as inclusion criteria was similar in both groups.

The study lasted for a mean of 6.2 years (range, 4.7 to 7.6 years). Vital status at the end of the study was ascertained for all patients except one. A total of 76% of the patients alive at the end of the study were on study medication (74% in the placebo group and 77% in the bezafibrate group). For 511 patients (17%), the study medication was withdrawn for reasons other than the occurrence of a primary end point or death. Of them, 373 patients (237 on placebo and 136 on bezafibrate) received open-label lipid-modifying treatment either before the occurrence of a primary end point or before the end of the study. Reasons for discontinuation of study medication were as follows: lipid levels exceeded predefined limits and necessitated treatment with a lipid-lowering drug (1.5% and 1.0% in the placebo and bezafibrate groups, respectively), adverse event (5.8% and 6.5%, respectively), patient's request to discontinue study medication (7.0% and 7.2%), and other miscellaneous reasons (2.6% and 1.5%).

Compliance, according to the tablet count, exceeded 90% in 74% of patients in both groups; it was between 75% and 90% for 17% of patients and <75% for the remaining 9% of patients. These data were confirmed in the bezafibrate group, in which alkaline phosphatase decreased by ≥10 U/L in 84% of patients, did not change in 10%, and increased by ≥10 U/L in 7% of patients.

Effect of Treatment on Lipid and Fibrinogen Levels

Average changes in lipid and fibrinogen levels are shown in Figure 1. The most marked changes were an increase of 18% in HDL-C and a reduction of 21% in triglycerides in the bezafibrate group. In the placebo group, values of total cholesterol and LDL-C remained stable for 3 years; thereafter, they declined (Figure 2).

Clinical Outcome

The effect of treatment on the primary end point (nonfatal and fatal MI and sudden death) is shown in Table 2. Among patients treated with bezafibrate, the crude rate of primary end points was 13.6% versus 15.0% in the placebo group (9.4% reduction; P=0.26). Figure 3 depicts Kaplan-Meier curves of the primary end point for the bezafibrate and placebo groups throughout the mean study period (6.2 years). The 2 curves started to separate after 2 years, but in the last 2 years of the study, a change in the slope of the placebo group becomes evident. The reduction in the cumulative probability of the primary end point at 6.2 years was 7.3% (P=0.24). Beyond 6.2 years, the number of patients at risk and the number of events were small and rather unstable (at 7 years, the reduction in the cumulative probability of the primary end point was 5.3%, but the standard error was 8.8% because only 10 events in the treatment group and 9 events in the placebo group had occurred).

Mortality rates were similar in both groups (Table 2). Among the 161 deaths in the bezafibrate group, 95 were due to cardiac causes, whereas in the placebo group, 88 of 152 deaths were attributed to cardiac causes (P=0.61). The distribution of all-cause and cardiac mortality was not different between the 2 study groups. Kaplan-Meier curves for all-cause mortality are shown in Figure 4. The incidence of secondary end points and stroke were comparable between the 2 groups (Table 2).

The study hypothesis was based on the effect of bezafibrate on baseline triglyceride and HDL-C levels. Therefore, we performed a post hoc analysis of the study primary end point by baseline HDL-C and triglyceride levels (Table 3). In patients with triglycerides <150 mg/dL, no clear benefit of bezafibrate treatment was observed. Among patients with baseline triglycerides ≥150 mg/dL, bezafibrate reduced the crude primary end point rate in direct relationship with the level of baseline triglycerides. Among patients with baseline triglycerides ≥200 mg/dL (225 patients in the placebo group and 234 in the bezafibrate group), bezafibrate reduced the cumulative probability of a primary end point by 39.5% (P=0.02), whereas among patients with triglycerides <200 mg/dL (1317 patients in the placebo group and 1314 in the bezafibrate group), the reduction in the cumulative probability of an end point was insignificant (Figure 5).

After adjustment for age, sex, prior MI, New York Heart Association class, angina class, and bezafibrate use, the relative risk for primary end points associated with bezafibrate treatment in the subgroup of patients with high baseline triglycerides (≥200 mg/dL) was 0.57 (95% confidence interval, 0.35 to 0.93). When the interaction between study treatment and different baseline triglyceride levels was further examined by low (<35 mg/dL) and high (≥35 mg/dL) HDL-C (Table 3), the effect of bezafibrate in patients with baseline triglycerides ≥200 mg/dL was of similar magnitude.

Safety

The overall incidence of any adverse event was 69% in both groups, and the frequency of each type of adverse event was similar in both groups. There were 85 cases (5.5%) of newly diagnosed fatal and nonfatal cancers in the bezafibrate group versus 91 cases (5.9%) in the placebo group, with no significant differences between the groups at any site. Seven patients in the placebo group and 5 patients in the bezafibrate group complained of muscular pains during follow-up. Creatine phosphokinase levels exceeding twice the upper normal limit (390 U/L for men and 260 U/L for women) were recorded in 5 patients (4 in the bezafibrate group). For the other safety laboratory parameters, small differences were observed between the study groups; these differences had no clinical significance.

Discussion

Although bezafibrate therapy led to a substantial increase in HDL-C and a reduction in triglycerides, the observed reduction in the primary end point was not as expected. In this respect, the time course of the Kaplan-Meier curves of the combined primary end point are intriguing. In the first 5 years of follow-up, the 2 curves diverged; at 5 years, they displayed a cumulative reduction in primary end points of 16.3% (P=0.09) between the bezafibrate and placebo groups. By the end of the study, the overall difference of the cumulative probability of primary end points was markedly reduced (Figure 3), reflecting an unexpected flattening of the placebo curve toward the end of the study. In previous lipid trials with statins and fibrates,¹⁴ ¹⁵ ²⁷ a continuous separation of the Kaplan-Meier curves was observed beginning at 12 to 24 months of follow-up and resulting in the desired effect of the active medication on the incidence of end points.

The convergence of the 2 curves of the primary end point could also be due to greater variation of the primary end point incidence rate toward the end of the study, which was caused by the relatively few events occurring in smaller groups of patients at risk in the last months of follow-up. The time course of the primary event curve of the placebo group was consistent with the decline in LDL-C levels in this group toward the end of the study (Figure 2). Of note, 373 patients, two-thirds of whom were randomized to placebo, received open-label lipid-modifying drugs before the occurrence of a primary end point or the end of the study. In addition, 164 patients (two thirds of whom were in the placebo group) with high LDL-C levels (>180 mg/dL) were given colestipol as adjuvant therapy to the study medication before the occurrence of a primary end point or the end of the study. It is to be expected that adding lipid-modifying therapy to patients in the placebo group would have a greater effect on outcome compared with patients in the bezafibrate group, where one effective therapy was added to or substituted for another.

A beneficial effect of fibric acid derivatives has been reported in the Helsinki Heart Study and in small angiographic secondary prevention trials in patients with isolated low HDL-C²⁸ and in young men after MI.²⁹ Recently, the results of the Veterans Affairs HDL Intervention Trial (VA-HIT), a secondary prevention study with the fibrate gemfibrozil, were published.²⁷ Comparison of mean lipid levels at baseline of patients recruited to VA-HIT and BIP shows that VA-HIT patients had lower HDL (32 mg/dL versus 34.6 mg/dL), lower LDL (111 mg/dL versus 148 mg/dL), and higher triglyceride levels (161 mg/dL versus 149 mg/dL) than BIP participants. After a mean follow-up of 5 years, a 22% reduction in the primary end point (defined as in our study) was observed in the VA-HIT study. By the end of 5 years of follow-up in the BIP study, the cumulative probability of primary end points was reduced by 16.3% with bezafibrate (P=0.09). It is noteworthy that although the placebo primary end point incidence rate in the VA-HIT study was 22%, this rate was 15% in the BIP study.

The low event rate in the BIP study may be partially explained by different patient characteristics and medical practice in the 2 study cohorts. The VA-HIT population was older and included more diabetic patients. Also, 2% of the placebo group and 1% of the gemfibrozil group were given open-label lipid medication in the VA-HIT study, whereas in the BIP study, 15% and 11% of patients in the placebo and bezafibrate groups, respectively, received such therapy. In addition, differences existed in the lipid responses of the 2 studies. In VA-HIT, HDL-C rose by 8%, in comparison to 18% in BIP, and triglycerides declined by 31%, as compared with 21% in BIP. We cannot rule out the possibility that reducing triglycerides may be more important than elevating HDL-C for secondary prevention in CAD patients with relatively low LDL-C levels and high triglycerides. Our data on patients with triglycerides >200 mg/dL may support this possibility.

The rates of adverse events, cancer, and mortality from any cause during follow-up were equally distributed between the 2 study groups. We can conclude from these data that long-term treatment with bezafibrate is safe.

In summary, bezafibrate was found to be safe and was effective in elevating HDL-C and lowering triglycerides. Although the overall effect of bezafibrate on the incidence of primary end points was moderate (P=0.24), the reduction in the primary end point was impressive in the subgroup of patients with high baseline triglycerides (≥200 mg/dL). The latter finding requires confirmation in a controlled, randomized trial designed to test this hypothesis because it was identified in post hoc analysis. Thus, bezafibrate may have a prominent role in the management of dyslipidemia and CAD when targeted to the subgroup of patients with high triglycerides.

Appendix A1

BIP Study Group: Participating Centers and Committee Membership

Participating Centers, Responsible Investigators and Physicians

Assaf Harofe Hospital, Zrifin: Zwi Schlesinger, MD; Zvi Vered, MD; Aharon Friedenson, MD. Barzilai Medical Center, Ashkelon: Leonardo Reisin, MD; Jamal Jafari, MD; Tatiana Flieb, MD. Beilinson Medical Center, Petach Tikva: Samuel Sclarovsky, MD; Yaakov Friedman, MD; Bruno Ostfeld, MD; Alejandro Solodky, MD. Bnei-Zion Hospital, Haifa: Edward Abinader, MD; Shmuel Rochfleish, MD. Carmel Hospital, Haifa: Abraham Palant, MD; Hanan Schneider, MD. Central Haemek Hospital, Afula: Tiberio Rosenfeld, MD; Suleiman Khalid MD. Edith Wolfson Medical Center, Holon: Yehezekiel Kishon, MD; Ron Narinsky, ³ MD; Rene Rotzak, MD; Anatoly Davidov, MD; Gregg Levine, MD. Hasharon Hospital, Petach Tikva: Izhar Zahavi, MD; Janash Vitrai, MD; Dror Diker, MD. Hillel-Yaffe Hospital, Hadera: Benyamin Pelled, MD; Joseph Pardu, MD; Jacob Galamidi, MD; Rasmi Majadla, MD. Ichilov Hospital, Sorasky Medical Center, Tel-Aviv: Shlomo Laniado, MD; Libi Sherf, MD; Shimon Braun, MD; Yemima Eschar, MD. Kaplan Hospital, Rehovot: Avraham Caspi, MD; Alexander Arditi, MD; Shulamit Botwin, MD; Lydia Arkavi, MD; Moshe Ziv, MD. Meir Hospital, Sapir Medical Center, Kfar-Saba: Daniel David, MD; Daniel Weisenberg, MD; Mira Kohanovski, MD; Simcha Meisel, MD. Naharia Hospital, Naharia: Nathan Rougin, MD; Malka Yahalom, MD; Alicia Glusman-Vazan, MD. Rambam Medical Center, Haifa: Walter Markiewitz, MD; Diav Motlak, MD; Jonathan Lessick, MD; Guenady Kagan, MD. Rivka Ziv Hospital, Tzfat: Alon Marmour, MD; Michael Flich, MD; Rachel Solomon, MD. Shaare Zedek Medical Center, Jerusalem: Dan Tzivoni, MD; Monty Zion, MD; Jonathan Balkin, MD. Sheba Medical Center, Tel Hashomer: Babeth Rabinowitz, MD; Eddy Barasch, MD; Zohar Brill, MD; Leon Aharon, MD; Aric Asman, MD. Soroka Medical Center, Be'er Sheba: Alexander Battler, MD; Moshe Gueron, MD; Natalio Cristal, MD; Noa Liel, MD; Bronislav Tsatskis, MD; Jacob Henkin, MD.

Review and Advisory Board

Gerd Assmann, MD; Peter Bauer, PhD; Shlomo Eisenberg, ³ MD; Lewis H. Kuller, MD; Baruch Modan, MD, Chairman; James Schoenberger, MD.

Principal Investigators

Daniel Brunner, MD; Jacob Agmon, MD; Elieser Kaplinsky, MD.

Steering Committee

Members of the Scientific Committee and Directors of Participating Centers.

Scientific Committee

Jacob Agmon, MD; Israel Bar Yehuda; Solomon Behar, MD; Daniel Brunner, MD, Chairman; Avraham Caspi, MD; Uri Goldbourt, PhD; Eran Graff, PhD; Elieser Kaplinsky, MD; Yehezekiel Kishon, MD; P. Dieter Lang, ¹ MD; Henrietta Reicher-Reiss, MD; Avraham Shotan, MD; Joachim Vollmar, ² MSc; Joshua Waysbort, MD.

Coordinating Center

Jacob Agmon, MD; Yisrael Bar-Yehuda; Solomon Behar, MD, Medical Director; Daniel Brunner, MD; Uri Goldbourt, PhD; Elieser Kaplinsky, MD; Henrietta Reicher-Reiss, MD; Avraham Shotan, MD.

Central Laboratory

Daniel Brunner, MD; Eran Graff, PhD, Director; Sara Schwartz, MSc; Joshua Waysbort, MD; Shoshana Schwartz, BSc; Tova Haimi, BSc; Rachel Lingel; Frima Nir; Ruth Sticlaru.

Critical Events and End Point Committee

Chaim Almog, MD; Alexander Battler, MD; Monty Zion, MD.

Stroke Monitoring

David Tanne, MD

Safety Evaluation

Siegfrid Hiemstra, MD; Eugene Heyman, PhD.**

Drug Supply Center

Klaus Kehne, Dr rer nat*

Epidemiology, Statistics, Computing, and Scientific Programming

Michal Benderly, MSc; Miriam Cohen; Mark Goldberg, BA; Uri Goldbourt, PhD; Lori Mandelzweig, MPH; Mitchell Snyder, PhD.

Data Entry

Dalia Ben-David; Yafit Makmal; Clara Shalom; Rachel Taub.

Study Monitors

Tzila Halevi, RN; Yemima Nahum, RN.

Secretarial Staff

Lynn Goodman; Leah Hirshkovitz; Tzila Pollak; Rachel Sinay.

Data Validation**

R. Flora, MD; Siegfrid Hiemstra, MD; Doris Kolb; Martin Scott, MSc; Joachim Vollmar, MSc.

Writing Group

Solomon Behar, MD; Daniel Brunner, MD; Elieser Kaplinsky, MD; Lori Mandelzweig, MPH; Michal Benderly, MSc.

1Boehringer Mannheim GmbH, Mannheim, Germany.

2PRA Europe, Mannheim, Germany.

3Deceased.

A complete list of BIP study participants is given in the Appendix.

**Figure 1.** Effect of treatment on lipids and fibrinogen. TC indicates total cholesterol. Values are in mg/dL.

**Figure 2.** Changes in lipid values throughout the study period in patients who completed 5 years of follow-up.

**Figure 3.** Kaplan-Meier curves for the cumulative probability of the primary end point.

**Figure 4.** Kaplan-Meier curves for the cumulative probability of all-cause mortality.

**Figure 5.** Kaplan-Meier curves for the primary end point in subgroups of patients with baseline triglycerides ≥200 mg/dL and <200 mg/dL.

**Table 1.** Baseline and Laboratory Characteristics of the 2 Study Groups
	Treatment Group
		Bezafibrate (n=1548)	Placebo (n=1542)
Men	1412 (91.2)	1413 (91.6)
Age, y	60.1 ±6.8	60.1±6.7
Systolic blood pressure, mm Hg	134±18	133±18
Diastolic blood pressure, mm Hg	81.2±9.0	80.8±9.0
Weight, kg	76.5±11.1	76.6±11.0
Height, cm	169.2±7.2	169.2 ±7.4
Body mass index, kg/m²	26.7±3.3	26.7 ±3.3
NYHA class ≥2	381 (24.9)	365 (24.1)
AP class ≥2	404 (26.1)	389 (25.3)
Prior myocardial infarction	1214 (78.6)	1194 (77.4)
Prior angina	875 (56.6)	891 (57.8)
Smoking	176 (11.4)	188 (12.1)
Diabetes	155 (10.0)	154 (10.0)
History of hypertension	482 (31.2)	518 (33.6)
Stroke	14 (0.9)	21 (1.4)
Transient ischemic attack	7 (0.5)	16 (1.0)
Peripheral vascular disease	51 (3.3)	56 (3.6)
COPD	36 (2.3)	52 (3.4)
Treatment at randomization
β-Blockers	580 (37.5)	609 (39.5)
Calcium antagonists	778 (50.3)	799 (51.8)
Anti-platelet aggregation	1095 (70.7)	1064 (69.0)
ACE inhibitors	185 (12.0)	197 (12.8)
Nitrates	793 (51.2)	781 (50.6)
Diuretics	210 (13.6)	224 (14.5)
Digitalis	61 (3.9)	47 (3.0)
Oral antidiabetic agents	78 (5.0)	78 (5.1)
Lipid values, mg/dL
Total cholesterol	212±17	213±18
HDL cholesterol	34.6±5.5	34.6±5.5
LDL cholesterol	148±17	149±17
Triglyceride	145±51	145±51
Fibrinogen	349±72	351±74

**Table 2.** Clinical Outcome
	Bezafibrate (n=1548)	Placebo (n=1542)	Reduction in Risk, %	P
Primary end point	211 (13.6)	232 (15.0)	−9.4	0.26
Non-fatal MI	150 (9.7)	172 (11.2)	−12.8	0.18
Fatal MI	18 (1.2)	17 (1.1)		0.87
Sudden death	43 (2.8)	43 (2.8)		0.98
Secondary end point	311 (20.1)	327 (21.2)	−4.9	0.44
UAP	76 (4.9)	82 (5.3)		0.61
CABG	144 (9.3)	157 (10.2)		0.41
PTCA	91 (5.9)	88 (5.7)		0.84
All end points combined	522 (33.7)	559 (36.3)	−6.6	0.14
Mortality	161 (10.4)	152 (9.9)		0.62
Cardiac	95 (6.1)	88 (5.7)		0.61
Noncardiac	66 (4.3)	64 (4.2)		0.87
Stroke	72 (4.6)	77 (5.0)		0.66
Ischemic stroke	59 (3.8)	69 (4.5)		0.36

**Table 3.** Cumulative Probability of Primary End Points at 6.2 Years of Follow-Up by Baseline Triglycerides and HDL-C Levels
	Bezafibrate, n (%)	Placebo, n (%)	Reduction, %	P
Triglycerides
<150 mg/dL	938 (12.6)	901 (13.7)	7.9	0.43
≥150 mg/dL	603 (16.3)	629 (17.1)	4.6	0.48
≥175 mg/dL	407 (15.9)	385 (20.3)	21.6	0.07
≥200 mg/dL	234 (12.0)	225 (19.7)	39.5	0.02
HDL-C <35 mg/dL and triglycerides
<150 mg/dL	378 (13.5)	382 (15.5)	12.4	0.46
≥150 mg/dL	420 (18.5)	436 (19.4)	4.5	0.56
≥175 mg/dL	294 (17.2)	286 (22.2)	22.6	0.09
≥200 mg/dL	184 (13.0)	162 (22.3)	41.8	0.02
HDL-C ≥35 and triglycerides
<150 mg/dL	560 (12.0)	518 (12.2)	1.6	0.77
≥150 mg/dL	183 (11.2)	193 (12.2)	8.5	0.59
≥175 mg/dL	113 (12.7)	99 (15.2)	16.8	0.45
≥200 mg/dL	50 (8.2)	63 (17.8)	35.9	0.33

Supported by a grant from Boehringer Mannheim GmbH, Mannheim, Germany, which is now part of F. Hoffmann-La Roche Ltd, Basel, Switzerland.

Footnotes

Correspondence to the BIP Study Group, Coordinating Center/S. Behar, MD, Neufeld Cardiac Research Institute, Sheba Medical Center, Tel-Hashomer 52621, Israel. E-mail [email protected]

References

1 Brunner D, Altman S, Loebl K, et al. Alpha-cholesterol percentages in coronary patients with and without increased total serum cholesterol levels and in healthy controls. J Atheroscler Res . 1962; 2:424–437.CrossrefMedlineGoogle Scholar
2 Franzen J, Johansson BW, Gustfson A. Reduced high density lipoproteins as a risk factor after acute myocardial infarction. Acta Med Scand . 1987; 221:357–362.CrossrefMedlineGoogle Scholar
3 Pekkanen J, Linn S, Heiss G, et al. Ten-year mortality from cardiovascular disease in relation to cholesterol level among men with and without preexisting cardiovascular disease. N Engl J Med . 1990; 322:1700–1707.CrossrefMedlineGoogle Scholar
4 Goldbourt U, Yaari S, Medalie JH. Isolated low HDL cholesterol as a risk factor for coronary heart disease mortality: a 21-year follow-up of 8000 men. Arterioscler Thromb Vasc Biol . 1997; 17:107–113.CrossrefMedlineGoogle Scholar
5 Brunner D, Altman S, Lobel K, et al. Serum cholesterol and triglycerides in patients suffering from ischemic disease and in healthy subjects. Atherosclerosis . 1977; 28:197–204.CrossrefMedlineGoogle Scholar
6 Austin MA. Plasma triglyceride and coronary heart disease. Arterioscler Thromb . 1991; 11:2–14.CrossrefMedlineGoogle Scholar
7 Assmann G, Schulte H, Cullen P. New and classical risk factors: the Munster Heart Study (PROCAM). Eur J Med Res . 1997; 2:237–242.MedlineGoogle Scholar
8 Austin MA. Plasma triglyceride as a risk factor for coronary heart disease: the epidemiologic evidence and beyond. Am J Epidemiol . 1989; 129:249–259.CrossrefMedlineGoogle Scholar
9 Hokanson JE, Austin MA. Plasma triglyceride level is a risk factor for cardiovascular disease independent of high-density lipoprotein cholesterol level: a meta-analysis of population-based prospective studies. J Cardiovasc Risk . 1996; 3:213–219.CrossrefMedlineGoogle Scholar
10 Berge KG, Canner PL. Coronary drug project: experience with niacin: Coronary Drug Project Research Group. Eur J Clin Pharmacol. 1991;40(suppl 1):S49–S51.Google Scholar
11 Manninen V, Tenkanen L, Koskinen P, et al. Joint effects of serum triglyceride and LDL cholesterol and HDL cholesterol concentrations on coronary heart disease risk in the Helsinki Heart Study: implications for treatment. Circulation . 1992; 85:37–45.CrossrefMedlineGoogle Scholar
12 Shepherd J, Cobbe SM, Ford I, et al, for the West of Scotland Coronary Prevention Study Group. Prevention of coronary heart disease with pravastatin in men with hypercholesterolemia. N Engl J Med . 1995; 333:1301–1307.CrossrefMedlineGoogle Scholar
13 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; 339:1349–1357.CrossrefMedlineGoogle Scholar
14 Sacks FM, Pfeffer MA, Moye LA, et al, for the Cholesterol and Recurrent Events Trial Investigators. The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. N Engl J Med . 1996; 335:1001–1009.CrossrefMedlineGoogle Scholar
15 Scandinavian Simvastatin Survival Study Group. Randomized trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet . 1994; 344:1383–1389.MedlineGoogle Scholar
16 Downs JR, Clearfield M, Weis S, et al. Primary prevention of acute coronary events with lovastatin in men and women with average cholesterol levels: results of AFCAPS/TexCAPS: Air Force/Texas Coronary Atherosclerosis Prevention Study. JAMA . 1998; 279:1615–1622.CrossrefMedlineGoogle Scholar
17 The Bezafibrate Infarction Prevention (BIP) study group. Lipids and lipoproteins in symptomatic coronary heart disease: distribution, intercorrelations, and significance for risk classification in 6700 men and 1500 women. Circulation . 1992; 86:839–848.CrossrefMedlineGoogle Scholar
18 Goldbourt U, Behar S, Reicher-Reiss H, et al, for the Bezafibrate Infarction Prevention Study Group. Rationale and design of a secondary prevention trial of increasing serum high-density lipoprotein cholesterol and reducing triglycerides in patients with clinical manifest atherosclerotic heart disease (the Bezafibrate Infarction Prevention Trial). Am J Cardiol . 1993; 71:909–915.CrossrefMedlineGoogle Scholar
19 Vu-Dac N, Schoonjans K, Laine B, et al. Negative regulation of the human apolipoprotein A-I promoter by fibrates can be attenuated by the interaction of the peroxisome proliferator-activated receptor with its response element. J Biol Chem . 1994; 269:31012–31018.CrossrefMedlineGoogle Scholar
20 Auwerx J, Schoonjans K, Fruchart JC, et al. Regulation of triglyceride metabolism by PPARs: fibrates and thiazolidinediones have distinct effects. Atheroscler Thromb . 1996; 3:81–89.CrossrefGoogle Scholar
21 Auwerx J, Schoonjans K, Fruchart JC, et al. Transcriptional control of triglyceride metabolism: fibrates and fatty acids change the expression of the LPL and apoC-III genes by activating the nuclear receptor PPAR. Atherosclerosis. 1995;124(suppl):S29–S37.Google Scholar
22 Staels B, Dallongeville J, Auwerx J, et al. Mechanism of action of fibrates on lipid and lipoprotein metabolism. Circulation . 1998; 98:2088–2093.CrossrefMedlineGoogle Scholar
23 Goldbourt U, Brunner D, Behar S, et al, for the BIP study group. Baseline characteristics of patients participating in the Bezafibrate Infarction Prevention (BIP) study. Eur Heart J. 1998;19(suppl H):H42–H47.Google Scholar
24 Friedewald WT, Levy RI, Frederickson DS. Estimation of the concentration of low density lipoprotein cholesterol in plasma, without use of preparative ultracentrifuge. Clin Chem . 1972; 18:499–502.CrossrefMedlineGoogle Scholar
25 SAS Institute Inc. SAS/STAT User's Guide. Version 6. 4th ed. Cary, NC: SAS Institute, Inc; 1987.Google Scholar
26 Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults, National Cholesterol Education Program. Second report of the Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel II). Circulation . 1994; 89:1329–1445.Google Scholar
27 Rubins H, Robins SJ, Collins D, et al, for the Veterans Affairs High-Density Lipoprotein Cholesterol Intervention Trial Study Group. Gemfibrozil for the secondary prevention of coronary heart disease in men with low levels of high-density lipoprotein cholesterol. N Engl J Med. . 1999; 341:410–418.CrossrefMedlineGoogle Scholar
28 Frick MH, Syvanne M, Nieminen MS, et al, for the Lopid Coronary Angiography Trial (LOCAT) study group. Prevention of the angiographic progression of coronary and vein-graft atherosclerosis by gemfibrozil after coronary bypass surgery in men with low levels of HDL cholesterol. Circulation. . 1997; 96:2137–2143.CrossrefMedlineGoogle Scholar
29 Ericsson CG, Hamsten A, Nilsson J, et al. Angiographic assessment of effects of bezafibrate on progression of coronary artery disease in young male post-infarction patients. Lancet. . 1996; 347:849–853.CrossrefMedlineGoogle Scholar