Optimization of beta-quantification methods for high-throughput applications.

BACKGROUND: Risk of cardiovascular disease is assessed, in part, by laboratory measurement of the concentrations of several lipoproteins. beta-Quantification is a method of lipoprotein measurement that uses ultracentrifugation to partially separate lipoprotein classes. Although beta-quantification is used largely in clinical and basic research, methods have not been described to allow the analysis of a large number of small-volume specimens with a short turnaround time. We report two variations of the traditional 5-mL method used by the Lipid Research Clinics Program that overcome these shortcomings. METHODS: Two lower-volume modifications of the traditional 5-mL beta-quantification method were developed. The methods used either 1 or 0.23 mL of specimen and required substantially less time for analysis (20 and 6 h, respectively) than the 5-mL method (2.5 days). The goal was to develop ultracentrifugation methods such that the concentration of cholesterol in the bottom fraction, from which LDL-cholesterol concentration is calculated, agreed with the 5-mL method. Fresh serum specimens (n = 45) were analyzed by the three methods to determine comparability of the methods based on the recovery of cholesterol in the bottom fraction after ultracentrifugation. To evaluate intrarun precision, replicate specimens (n = 17) were analyzed in a single run for each method. This experiment also evaluated how quickly the fractions would remix after separation by ultracentrifugation. For the 1-mL method, accuracy of the measurement of LDL- and HDL-cholesterol concentrations and the interrun precision were established by analysis of frozen serum specimens provided by the CDC, which established target values for the pools using reference methods. RESULTS: No clinically significant differences in cholesterol concentrations in the bottom fraction were observed for the 1- and 0.23-mL methods, which had mean biases of 0.8% and 1.5% relative to the 5-mL method, respectively. Intra- and interrun variability was acceptable for each method, e.g., <1.8% for cholesterol in the bottom fraction. Ultracentrifuged specimens were stable for at least 4 h with no evidence of contamination of cholesterol in the bottom fraction. For comparison specimens provided by the CDC, the 1-mL method met the accuracy and precision goals of the National Cholesterol Education Program for the measurement of HDL- and LDL-cholesterol concentrations (goals: total error <13% and <12%, respectively), with total errors of 6.45% and 5.43%, respectively. CONCLUSIONS: Both the 1- and 0.23-mL beta-quantification methods are suitable substitutes for the traditional 5-mL method for use in clinical and basic research for the determination of LDL-cholesterol concentration. Both methods provide much higher throughput and require substantially less specimen volume. The 0.23-mL method can be performed in 1 day, but it is slightly less precise than the 1-mL method. In our laboratory setting, as many as 80 specimens are routinely processed per day using the 1-mL method.

Standardization of lipoprotein reporting.

We wanted to ascertain whether the current format of lipid laboratory reports seemed adequate to promote identification and treatment of patients with dyslipidemia. In a random survey of lipid laboratory reports from 25 laboratories, we found great inconsistencies among reporting formats and contents. Fewer than half the laboratories correctly reported the ranges for cholesterol, only 4 correctly reported ranges for high-density lipoprotein cholesterol, only 2 correctly reported ranges for triglycerides, and none presented low-density lipoprotein cholesterol ranges in terms of risk factors for coronary heart disease. Reports typically were disjointed and difficult to read. The current practice of reporting results for lipid panels is confusing and does not follow the National Cholesterol Education Program (NCEP) guidelines. We recommend that reporting of results be standardized, and a "model" standardized report is presented herein, based on consensus from a team of experts. The standardized report uses current recommendations for ranges, follows the flowcharts of the NCEP guidelines, and takes the patient's clinical condition (the number of risk factors and the presence of coronary heart disease) into consideration. Standardizing lipid reports should decrease confusion and perhaps increase application of the guidelines and patient compliance with treatment.

VLDL, apolipoproteins B, CIII, and E, and risk of recurrent coronary events in the Cholesterol and Recurrent Events (CARE) trial.

BACKGROUND: Plasma triglyceride concentration has been an inconsistent independent risk factor for coronary heart disease, perhaps because of the metabolic heterogeneity among VLDL particles, the main carriers of triglycerides in plasma. METHODS AND RESULTS: We conducted a prospective, nested case-control study in the Cholesterol and Recurrent Events (CARE) trial, a randomized placebo-controlled trial of pravastatin in 4159 patients with myocardial infarction and average LDL concentrations at baseline (115 to 174 mg/dL, mean 139 mg/dL). Baseline concentrations of VLDL-apolipoprotein (apo) B (the VLDL particle concentration), VLDL lipids, and apoCIII and apoE in VLDL+LDL and in HDL were compared in patients who had either a myocardial infarction or coronary death (cases, n=418) with those in patients who did not have a cardiovascular event (control subjects, n=370) in 5 years of follow-up. VLDL-cholesterol, VLDL-triglyceride, VLDL-apoB, apoCIII and apoE in VLDL+LDL and apoE in HDL were all interrelated, and each was a univariate predictor of subsequent coronary events. The significant independent predictors were VLDL-apoB (relative risk [RR] 3.2 for highest to lowest quintiles, P:=0.04), apoCIII in VLDL+LDL (RR 2.3, P:=0.04), and apoE in HDL (RR 1.8, P:=0.02). Plasma triglycerides, a univariate predictor of coronary events (RR 1.6, P:=0.03), was not related to coronary events (RR 1.3, P:=0.6) when apoCIII in VLDL+LDL was included in the model, whereas apoCIII remained significant. Adjustment for LDL- and HDL-cholesterol did not affect these results. CONCLUSIONS: The plasma concentrations of VLDL particles and apoCIII in VLDL and LDL are more specific measures of coronary heart disease risk than plasma triglycerides perhaps because their known metabolic properties link them more closely to atherosclerosis.

Differences in the Abeta40/Abeta42 ratio associated with cerebrospinal fluid lipoproteins as a function of apolipoprotein E genotype.

The epsilon4 allele of apolipoprotein E (ApoE) is a risk factor for Alzheimer's disease (AD). ApoE, which is important for lipid metabolism, is also a major constituent of cerebrospinal fluid (CSF) lipoproteins (LPs). Although ApoE in the CSF is derived from the central nervous system, the relation between LP metabolism in plasma and CSF is not clear. Soluble amyloid-beta (Abeta) protein may normally be associated with CSF LPs. It is converted in AD to a fibrillar form in brain parenchyma. ApoE and CSF LPs may regulate this process. The purpose of this study was to characterize CSF LPs from healthy, cognitively normal, fasted, elderly individuals at different risk for AD based on ApoE genotype. Lipid composition of CSF LPs did not differ with ApoE genotype. Interestingly, plasma and CSF high-density lipoprotein (HDL) cholesterol and apolipoprotein AI (ApoAI) levels were correlated. Importantly, as assessed by size-exclusion chromatography, Abeta in CSF coeluted in fractions containing LPs and was influenced by ApoE genotype: E4-positive subjects displayed significant elevations in Abeta40/Abeta42 ratios. These results suggest that plasma ApoAI/HDL levels can influence CSF ApoAI/HDL levels and that interactions between Abeta and central nervous system LPs may reflect changes in brain Abeta metabolism before the onset of clinical disease.

Analytical and clinical performance of a detergent-based homogeneous LDL-cholesterol assay: a multicenter evaluation.

BACKGROUND: LDL-cholesterol (LDL-C) concentrations currently are determined in most clinical laboratories using the Friedewald calculation. This approach has several limitations and may not always meet the current total error recommendation in LDL-C measurement of

Selection, validation, standardization, and performance of a designated comparison method for HDL-cholesterol for use in the cholesterol reference method laboratory network.

BACKGROUND: Accurate and precise HDL-cholesterol (HDL-C) measurements are essential for effective application of National Cholesterol Education Program treatment guidelines. The Cholesterol Reference Method Laboratory Network (CRMLN) assists manufacturers of in vitro diagnostic products to establish traceability to the accuracy base. CRMLN sought to implement a designated comparison method (DCM) that overcomes the impracticalities of the expensive and labor-intensive reference method for HDL-C. METHODS: CRMLN evaluated candidate DCMs and selected one that uses 50-kDa dextran sulfate with magnesium ions as the precipitation reagent followed by measurement of cholesterol by the CDC reference method. After validating the method, we transferred it to all CRMLN laboratories and successfully standardized it using CDC frozen serum reference materials. CRMLN laboratories participate in monthly performance evaluations. RESULTS: CRMLN laboratories were able to meet a precision goal, as indicated by SD, of /=1.09 mmol/L (42 mg/dL) 95.6% of the time. CRMLN is working to further improve its performance by implementing a bias criterion of 0.03 mmol/L (1 mg/dL) for all HDL-C concentrations. CONCLUSIONS: CRMLN selected, validated, standardized, and implemented a DCM for HDL-C that is accurate, robust, transferable, and practical. The DCM is being used to assist manufacturers in calibrating their products so that ultimately, clinical laboratories using the products will more accurately measure HDL-C.

Known mutations of apoB account for only a small minority of hypobetalipoproteinemia.

Low LDL cholesterol and apoB levels in plasma cosegregate with mutations of apoB in some kindreds with familial hypobetalipoproteinemia. Approximately 35 apoB mutations, many specifying apoB truncations, have been described. Based on the centile nomenclature where the full-length nature apoB consisting of 4536 amino acids is designated as apoB-100, only those truncations of apoB >25% of normal length are detectable in plasma. Previously, we reported on five unrelated kindreds with familial hypobetalipoproteinemia in whom although no apoB truncations were detectable in plasma, low apoB levels were nevertheless linked to the apoB gene. In one of those kindreds, we reported a donor splice site mutation in intron 5 (specifying apoB- 4). We now describe a nonsense mutation in exon 10 (apoB-9) in two of the other unrelated families. Both the apoB-4 and apoB-9 mutations have been reported by others in unrelated families. Recurrent mutations of apoB-40 and apoB-55 also have been reported, suggesting that recurrent mutations of apoB may account for an appreciable proportion of familial hypobetalipoproteinemia kindreds. To test this hypothesis, we searched for four apoB mutations whose products are not detected in plasma including the apoB-4, apoB-9, and two other previously reported mutations in exons 21 and 25. We studied three groups with plasma cholesterols <130 mg/dl in whom no apoB truncations were detected in plasma: a) 28 FHBL probands from St. Louis, b) 151 individual St. Louisians, and c) 28 individual Sicilians. One subject from the 28 kindreds and two subjects among 151 hypobeta individuals from St. Louis harbored the exon 10 mutation. None of the other mutations were detected. Thus, among hypobeta lipoproteinemic subjects without any detectable apoB truncations in plasma, <5% had an apoB truncation-producing mutation. As only about 0.5% of hypobeta lipoproteinemic subjects have plasma-detectable apoB truncations, our data suggest that the known apoB truncations account for only a small proportion of hypocholesterolemia.

Reflex testing I: algorithm for lipid and lipoprotein measurement in coronary heart disease risk assessment.

We reviewed the current literature in order to construct a reflex testing algorithm that maximizes clinical utility and cost-effectiveness of lipid and lipoprotein testing. The algorithm was based on the 2nd Report of the National Cholesterol Education Program Adult Treatment Panel guidelines for use of total cholesterol (TC), triglycerides (TG), HDL-C, and LDL-C, and published reports describing the clinical use of apolipoprotein B and lipoprotein (a). The success of this algorithm was tested in a low-risk general and a high-risk hyperlipidemic patient population. Lipid data and non-lipid risk factors were obtained from a national database and from patients seen at two lipid clinics. A total of 16,968 individuals from the National Health and Nutrition Examination Survey III database comprised the low-risk group, and 239 patients examined in the Hartford Hospital and Washington University Lipid Clinics comprised the high-risk group. We found a solid scientific base to support the NCEP guidelines and reasonable support for limited testing of apoB and Lp(a). According to the algorithm, the direct LDL-C assay was deemed unnecessary in 98% and 91% of low- and high-risk subjects, respectively, if one assumes that the Friedewald equation is adequate with TG < or = 4.00 g/l. With a more conservative cutoff of TG < or = 2.50 g/l, the algorithm canceled 92% and 81% of direct LDL tests, respectively. The algorithm also limited TG to 20 and 64%, apoB to 6 and 20%, and Lp(a) to 15 and 56%, of low- and high-risk groups, respectively. Use of a comprehensive, reflex algorithm for coronary heart disease risk assessment will substantially reduce the utilization of laboratory services without diminishing the clinical value of these tests. The algorithm will prevent the overuse of certain expensive tests (direct LDL) while promoting the limited use of underutilized tests [apoB and Lp(a)].

A new liquid homogeneous assay for HDL cholesterol determination evaluated in seven laboratories in Europe and the United States.

We evaluated a new liquid homogeneous assay for the direct measurement of high density lipoprotein cholesterol (HDL-C Plus) in seven laboratories. The assay includes two reagents which can be readily used in most available clinical chemistry analyzers. The total CVs of the new method were below 4.6% and the bias in relation to the designated comparison method was below 3.9%. The total error ranged between 4 to 7%. HDL-C values determined by this method were in good agreement with those obtained by the old homogeneous assay using lyophilized reagents, and other homogeneous and precipitation assays (0.944 < r < 0.996). The assay was linear up to at least 3.89 mmol/l HDL-C. Hemoglobin did not interfere, whereas in icteric samples slight deviations were observed. Lipemia up to 11.3 to 22.6 mmol/l triglycerides did not interfere with this homogeneous HDL-C assay. In samples of patients with paraproteinemia, discrepant results were seen. This liquid homogeneous HDL-C assay was easy to handle and produced similar results in all laboratories participating in this study. This method will enable clinical laboratories to reliably measure HDL-C for risk assessment of coronary heart disease.

Cardiovascular events and their reduction with pravastatin in diabetic and glucose-intolerant myocardial infarction survivors with average cholesterol levels: subgroup analyses in the cholesterol and recurrent events (CARE) trial. The Care Investigators.

BACKGROUND: Although diabetes is a major risk factor for coronary heart disease (CHD), little information is available on the effects of lipid lowering in diabetic patients. We determined whether lipid-lowering treatment with pravastatin prevents recurrent cardiovascular events in diabetic patients with CHD and average cholesterol levels. METHODS AND RESULTS: The Cholesterol And Recurrent Events (CARE) trial, a 5-year trial that compared the effect of pravastatin and placebo, included 586 patients (14.1%) with clinical diagnoses of diabetes. The participants with diabetes were older, more obese, and more hypertensive. The mean baseline lipid concentrations in the group with diabetes--136 mg/dL LDL cholesterol, 38 mg/dL HDL cholesterol, and 164 mg/dL triglycerides--were similar to those in the nondiabetic group. LDL cholesterol reduction by pravastatin was similar (27% and 28%) in the diabetic and nondiabetic groups, respectively. In the placebo group, the diabetic patients suffered more recurrent coronary events (CHD death, nonfatal myocardial infarction [MI], CABG, and PTCA) than did the nondiabetic patients (37% versus 25%). Pravastatin treatment reduced the absolute risk of coronary events for the diabetic and nondiabetic patients by 8.1% and 5.2% and the relative risk by 25% (P=0.05) and 23% (P<0.001), respectively. Pravastatin reduced the relative risk for revascularization procedures by 32% (P=0.04) in the diabetic patients. In the 3553 patients who were not diagnosed as diabetic, 342 had impaired fasting glucose at entry defined by the American Diabetes Association as 110 to 125 mg/dL. These nondiabetic patients with impaired fasting glucose had a higher rate of recurrent coronary events than those with normal fasting glucose (eg, 13% versus 10% for nonfatal MI). Recurrence rates tended to be lower in the pravastatin compared with placebo group (eg, -50%, P=0.05 for nonfatal MI). CONCLUSIONS: Diabetic patients and nondiabetic patients with impaired fasting glucose are at high risk of recurrent coronary events that can be substantially reduced by pravastatin treatment.

Effect of short-term recombinant growth hormone administration on plasma lipoproteins in elderly adults.

To characterize the effects of recombinant human growth hormone (rhGH) on plasma lipids and lipoproteins, rhGH was administered daily at a dose of 40 micrograms.kg-1 (Genentech) for 14 days in 7 healthy elderly male (67.4 +/- 1.9 years, 75.8 +/- 2.6 kg) adults. Six other healthy males (63.9 +/- 0.7 years, 77.8 +/- 3.8 kg) served as concurrent controls. Total plasma cholesterol (TC), triglycerides (TG), very-low-density lipoprotein-cholesterol, low-density lipoprotein-cholesterol, high-density lipoprotein-cholesterol, very-low-density lipoprotein-TG (VLDL-TG) and apolipoprotein AI and apolipoprotein B were determined after an overnight fast before and after the 14-day period of rhGH administration. Subcutaneous rhGH administration was physiologically effective, as shown by a threefold increase in insulin-like growth factor-I (from 110.8 +/- 8.2 to 355.5 +/- 41.6 ng.ml-1; p < 0.05). Plasma fasting insulin also increased from 38.0 +/- 6.5 to 129.9 +/- 43.8 mumol.l-1 (p < 0.05) at the end of the 14 days of rhGH treatment. With respect to plasma lipid/lipoprotein changes, rhGH administration increased plasma TG levels (from 1.5 +/- 0.3 to 2.2 +/- 0.4 mmol.l-1; p < 0.05) and VLDL-TG (from 1.1 +/- 0.3 to 1.8 +/- 0.4 mmol.l-1; p < 0.05), but did not change TC (from 5.0 +/- 0.4 to 5.2 +/- 0.3 mmol.l-1) or any other lipid/lipoprotein variables measured. No significant lipid changes were noted in the control group over the 14-day period. These data suggest that short-term rhGH treatment significantly alters plasma variables of TG profile, perhaps by altering metabolic parameters (i.e. synthesis and/or clearance rates) of VLDL metabolism.

Assessment of interlaboratory performance in external proficiency testing programs with a direct HDL-cholesterol assay.

Direct assays for the determination of HDL-cholesterol (HDL-C) have recently become available. The methods are precise, require small sample volume, and appear to be less affected by increased triglycerides than traditional precipitation methods. In this study, we describe the inter- and intralaboratory variability of the Boehringer Mannheim Corporation direct HDL-C assay and its performance in external proficiency testing surveys. A comparison study among three laboratories, using different analyzers and 85 serum specimens, showed a correlation coefficient (r) of 0.99. The direct HDL-C assay also showed good agreement with the ultracentrifugation-dextran sulfate-Mg2+ method (r = 0.98) and the Cholesterol Reference Method Laboratory Network-Designated Comparison Method (a = 0.98x + 4.75 mg/L, r = 0.98). Total error at medical decision levels ranged from -0.8% to +11.1%. Furthermore, this assay performed adequately in the College of American Pathologists and the ALERT surveys as well as the CDC Lipid Standardization Program and met all performance criteria of regulatory agencies.

Relationship between plasma LDL concentrations during treatment with pravastatin and recurrent coronary events in the Cholesterol and Recurrent Events trial.

BACKGROUND: Although LDL lowering has been shown to reduce recurrent coronary events in patients with coronary heart disease, little direct information is available on the extent of LDL lowering required to achieve this outcome. METHODS AND RESULTS: The Cholesterol and Recurrent Events (CARE) trial compared pravastatin and placebo in patients who had experienced myocardial infarction (MI) who had average concentrations of total cholesterol <240 mg/dL (baseline mean, 209 mg/dL) and LDL cholesterol (LDL) 115 to 174 mg/dL (mean, 139 mg/dL). Pravastatin reduced coronary death or recurrent MI by 24%. In multivariate analysis, the LDL concentration achieved during follow-up was a significant, although nonlinear, predictor of the coronary event rate (P=.007), whereas the extent of LDL reduction was not significant, whether expressed as an absolute amount (P=.97) or a percentage (P=.76). The coronary event rate declined as LDL decreased during follow-up from 174 to approximately 125 mg/dL, but no further decline was seen in the LDL range from 125 to 71 mg/dL. In multivariate analysis, triglyceride but not HDL concentrations during follow-up were weakly but significantly associated with the coronary event rate. CONCLUSIONS: The LDL concentrations achieved during treatment with pravastatin or placebo were associated with reduction in coronary events down to an LDL concentration of approximately 125 mg/dL. LDL concentrations <125 mg/dL during treatment were not associated with further benefit. Absolute or percentage reduction in LDL had little relationship to coronary events.

Performance of predispensed reagent and noncentrifugation methods for high-density lipoprotein cholesterol measurement.

OBJECTIVE: To evaluate the performance of high-density lipoprotein cholesterol methods with predispensed reagent (SPINPRO, ISO Spin, Vitros, and One Shots) and noncentrifugation separation (Magnetic-HDL) compared with traditional high-density lipoprotein cholesterol methods (PBI Plus, Boehringer-Mannheim, and Abbott). DESIGN: Precision was evaluated by running two concentrations of frozen human sera and two concentrations of lyophilized quality control material according to National Committee for Clinical Laboratory Standards document EP5-T2. Accuracy was evaluated by comparing sample results achieved by each method with those by the Cholesterol Reference Method Laboratory Network Designated Comparison Method and the Lipid Research Clinics' heparin-Mn++ method. Sera from donors with high triglyceride levels were used to challenge the ability of each method to measure lipemic samples for each method. SETTING: Outpatient clinic and university medical center. PATIENTS: Forty-two ambulatory donors with high-density lipoprotein cholesterol levels ranging from 30 to 102 mg/dL, cholesterol levels ranging from 107 to 679 mg/dL, and triglyceride levels ranging from 20 to 2450 mg/dL. MAIN OUTCOME MEASURE: Precision, accuracy, linearity, and ability to measure lipemic samples. RESULTS: Imprecision varied from CV 1.4% to 8.6%. The mean absolute bias for each method versus the Designated Comparison Method ranged from 2.6% to 13.3%. Deming regression analysis of all methods versus the heparin-Mn++ method gave slopes from 0.88 to 1.08 and intercepts from -4.5 to 0.0 mg/dL. All methods met the current 22% total error for individual specimens recommended by the National Cholesterol Education Program. Not all methods could measure high-density lipoprotein cholesterol in the presence of high concentrations of triglycerides. CONCLUSIONS: Performance of all methods met the current recommendations of the National Cholesterol Education Program. Not all methods met the 1998 performance goals.

Point: status of lipid and lipoprotein standardization.

Cholesterol and triglyceride standardization procedures have been used extensively and continuously since the 1950s. Definitive and Reference Methods, as well as primary and secondary standards, have been developed and maintained as the basis for evaluating the accuracy of results by various methods in many laboratories. But, although standardization efforts for apolipoprotein A-I and B measurements have been reported in detail in the scientific literature, much less has been reported in the area of total and lipoprotein cholesterol and triglyceride standardization efforts. Standardized cholesterol and triglyceride concentrations, determined in multiple large epidemiological and clinical studies, have been instrumental to the National Cholesterol Education Program panels that have assessed the lipoprotein values associated with risk of coronary disease, and have determined the cutpoints that are now used extensively by physicians to guide diagnosis and treatment of individual patients.

Induction of human skeletal muscle lipoprotein lipase gene expression by short-term exercise is transient.

Exercise increases skeletal muscle lipoprotein lipase (LPL) expression, but the time course of this response is not known. In the present study, we examined the time course of the LPL response to both short-term and acute exercise and measured circulating levels of putative regulators of muscle LPL. Nine adults underwent short-term exercise training (60-90 min of stationary cycling at 55-70% of leg ergometer peak oxygen uptake on 5 consecutive days). Five vastus lateralis biopsies were performed: before training, 20 h after the fourth bout (immediately before the 5th bout), and 0.2, 4, and 8 h after the fifth bout. After four bouts of exercise in 4 days, there was no increase in LPL mass or LPL mRNA exactly 20 h after the fourth bout. However, when tissues were sampled closer to the exercise bout on the 5th day, transient increases were seen. On day 5, LPL mRNA increased by 127% (P < 0.05) at 4 h postexercise and was followed by an increase in LPL mass of 93% (P < 0.05) at 8 h postexercise. Serum triglycerides decreased by 23% (P < 0.05) during the protocol. Two nonexercising subjects showed no consistent change in LPL mRNA or mass. Acute exercise transiently increased levels of norepinephrine (9-fold) and epinephrine (5-fold) and reduced insulin levels. Acute exercise preceded by four daily bouts of exercise induces a transient rise in LPL mRNA followed by rise in LPL mass, suggesting that these responses are temporally related. This induction of LPL gene expression may result from dynamic changes in serum catecholamines, plasma insulin, or events intrinsic to muscle contraction itself.

The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. Cholesterol and Recurrent Events Trial investigators.

BACKGROUND: In patients with high cholesterol levels, lowering the cholesterol level reduces the risk of coronary events, but the effect of lowering cholesterol levels in the majority of patients with coronary disease, who have average levels, is less clear. METHODS: In a double-blind trial lasting five years we administered either 40 mg of pravastatin per day or placebo to 4159 patients (3583 men and 576 women) with myocardial infarction who had plasma total cholesterol levels below 240 mg per deciliter (mean, 209) and low-density lipoprotein (LDL) cholesterol levels of 115 to 174 mg per deciliter (mean, 139). The primary end point was a fatal coronary event or a nonfatal myocardial infarction. RESULTS: The frequency of the primary end point was 10.2 percent in the pravastatin group and 13.2 percent in the placebo group, an absolute difference of 3 percentage points and a 24 percent reduction in risk (95 percent confidence interval, 9 to 36 percent; P = 0.003). Coronary bypass surgery was needed in 7.5 percent of the patients in the pravastatin group and 10 percent of those in the placebo group, a 26 percent reduction (P=0.005), and coronary angioplasty was needed in 8.3 percent of the pravastatin group and 10.5 percent of the placebo group, a 23 percent reduction (P=0.01). The frequency of stroke was reduced by 31 percent (P=0.03). There were no significant differences in overall mortality or mortality from noncardiovascular causes. Pravastatin lowered the rate of coronary events more among women than among men. The reduction in coronary events was also greater in patients with higher pretreatment levels of LDL cholesterol. CONCLUSIONS: These results demonstrate that the benefit of cholesterol-lowering therapy extends to the majority of patients with coronary disease who have average cholesterol levels.