- Email: [email protected]
Effects of Atorvastatin Versus Other Statins on Fasting and Postprandial C-Reactive Protein and Lipoprotein-Associated Phospholipase A2 in Patients With Coronary Heart Disease Versus Control Subjects
Effects of Atorvastatin Versus Other Statins on Fasting and Postprandial C-Reactive Protein and LipoproteinAssociated Phospholipase A2 in Patients With Coronary Heart Disease Versus Control Subjects Ernst J. Schaefer, MD, Judith R. McNamara, MT, Bela F. Asztalos, PhD, Timothy Tayler, PhD, Jennifer A. Daly, MD, Joi L. Gleason, MS, RD, Leo J. Seman, MD, PhD, Andrea Ferrari, RN, and Joel J. Rubenstein, MD The effects of atorvastatin (40 mg/day) versus placebo on fasting and postprandial plasma levels of high-sensitivity C-reactive protein (hs-CRP) and lipoprotein-associated phospholipase A2 (Lp-PLA2) were examined over 36 weeks in 84 patients who had coronary heart disease and low-density lipoprotein cholesterol levels >130 mg/dl and compared directly with the effects of fluvastatin, lovastatin, pravastatin, and simvastatin. Results were also compared with those obtained in ageand gender-matched control subjects (n ⴝ 84). Feeding increased median hs-CRP levels by 2% in patients (p ⴝ NS) and 22% in controls (p <0.01) and increased mean Lp-PLA2 values by 9% in patients (p ⴝ NS) but decreased values by 21% in controls (p <0.0001). Patients had 51% higher median hs-CRP values and 29% higher mean Lp-PLA2 values than did controls (p <0.05 for hs-CRP and Lp-PLA2) in the fasting state; however, Lp-
PLA2 values were 62% higher (p <0.0001) in the fed state in patients compared with controls. Atorvastatin decreased median hs-CRP levels by 32% (p <0.01) and mean Lp-PLA2 values by 26% in patients (p <0.0001), with similar decreases in the fed state, and none of the other statins had any significant effect on these parameters. Change in Lp-PLA2 was significantly related to change in low-density lipoprotein cholesterol (p <0.01), with no significant relations with change in hs-CRP. Our data indicate greater differences in patients with coronary heart disease compared with controls in Lp-PLA2 in the fed state than in the fasting state and that atorvastatin is more effective than fluvastatin, lovastatin, pravastatin, or simvastatin for decreasing not only lowdensity lipoprotein cholesterol but also hs-CRP and Lp-PLA2. 䊚2005 by Excerpta Medica Inc. (Am J Cardiol 2005;95:1025–1032)
he focus of this study was the examination of the effects of atorvastatin versus those of other statins T (fluvastatin, lovastatin, pravastatin, and simvastatin)
of Lp-PLA2 and CRP are markers of inflammation and increased CHD risk. Lp-PLA2 is produced by macrophages, whereas CRP is primarily synthesized in the liver. We previously documented that patients who have CHD with low-density lipoprotein cholesterol levels ⬎130 mg/dl have high plasma levels of triglycerides and remnant lipoprotein cholesterol and triglyceride in the fasting and fed states compared with controls and that a fat challenge is not necessary to detect these abnormalities.1 In these studies, we also documented that fasting was not essential to detect abnormalities in direct low-density lipoprotein cholesterol, non– high-density lipoprotein cholesterol, and high-density lipoprotein cholesterol.1 In addition, we reported that atorvastatin therapy at 40 mg/day normalizes levels of all classes of triglyceride-rich lipoproteins and low-density lipoproteins (large, medium, and small low-density lipoproteins) in the fasting and fed states in patients who have CHD compared with control subjects2 and that atorvastatin is more effective than other tested statins in this regard.3
on high-sensitivity C-reactive protein (hs-CRP), lipoprotein-associated phospholipase A2 (Lp-PLA2), and plasma lipoprotein subspecies in the fasting and postprandial states (4 hours after a meal rich in saturated fat and cholesterol) in patients who had coronary heart disease (CHD) and a comparison of their parameters on and off therapy with those of age- and gendermatched normal control subjects. High plasma levels From the Cardiovascular Research and Lipid Metabolism Laboratories, Tufts University School of Medicine, Friedman School of Nutrition Science and Policy at Tufts University, and Jean Mayer USDA Human Nutrition Research Center on Aging at Tufts University, Boston, Massachusetts; and the Cardiology Division, Newton-Wellesley Hospital, Newton, Massachusetts. This study was supported by investigator-initiated research contracts from Parke-Davis/Pfizer, New York, New York; diaDexus, Inc., South San Francisco, California; and Otsuka America Pharmaceuticals, Inc., Rockville, Maryland. Manuscript received September 22, 2004; revised manuscript received and accepted January 3, 2005. Address for reprints: Ernst J. Schaefer, MD, Tufts University, 711 Washington Street, Boston, Massachusetts 02111. E-mail: ernst. [email protected]. ©2005 by Excerpta Medica Inc. All rights reserved. The American Journal of Cardiology Vol. 95 May 1, 2005
METHODS
Study subjects: The 84 patients in this study who had CHD were recruited from the clinic or by adver0002-9149/05/$–see front matter doi:10.1016/j.amjcard.2005.01.023
1025
tising and were required to have evidence of established CHD (coronary artery bypass grafting, angioplasty, documented myocardial infarction, significant coronary artery stenosis ⬎50% as assessed by angiography, or significantly decreased cardiac perfusion based on cardiac imaging with and without exercise). All subjects were studied after providing informed consent according to protocols approved by the human investigation review committee at the New England Medical Center (Boston, Massachusetts) and NewtonWellesley Hospital (Newton, Massachusetts). Patients were excluded if they had evidence of renal impairment, hypothyroidism, or liver dysfunction based on clinical chemistry testing or had previous adverse reactions to statins. In addition, patients were required to have a low-density lipoprotein cholesterol serum concentration ⬎130 mg/dl while off lipid-lowering medication for ⱖ6 weeks, including anion exchange resins, statins, fibric acid derivatives, fish oil, or niacin-containing products. Patients were maintained on other medications throughout the study, with no change, including calcium channel blockers,  blockers, and diuretics. Patients had previously received instruction on a National Cholesterol Education Program diet that contained ⬍30% of calories as fat, ⬍7% of calories as saturated fat, and ⬍200 mg/day of cholesterol. Patients had a lead-in baseline period of 4 weeks on this diet and did not use lipid-lowering medication. At the end of the 4-week lead-in period, patients were sampled after a 12-hour overnight fast and then randomized to receive atorvastatin or 1 of 4 other statins (fluvastatin, n ⫽ 26; lovastatin, n ⫽ 20; pravastatin, n ⫽ 17; or simvastatin, n ⫽ 16) for a protocol in which they were sampled in the fasting state after 4 weeks on 20 mg/day, after another 4 weeks on 40 mg/ day, after another 4 weeks on 80 mg/day or the maximal dose (40 mg for pravastatin and simvastatin), and then at 4 and 8 weeks during an 8-week placebo period. Thereafter, patients went on a second 12-week statin period in which they received 20 mg/day for 4 weeks, 40 mg/day for 4 weeks, and 80 mg/day or the maximal dose for 4 weeks with the sampling schedule. If they had not received atorvastatin in the first statin period, they received it during the second period. If they had received atorvastatin during the first statin period, they were randomized to 1 of the other 4 statins during the second statin period of 12 weeks. At the end of the 4-week lead-in period, at the end of the 40 mg/day dose of each statin, and at the end of 8-week period on placebo, most subjects (n ⫽ 61) were also sampled 4 hours after a fat-rich meal (obtained from a McDonald’s restaurant; 2 eggs, 2 sausages, and 2 muffins; 880 total calories, 500 calories (57%) from fat, 180 calories (20%) from saturated fat, and 510 mg of cholesterol). We also recruited an equal number of age- and gender-matched control subjects (n ⫽ 84) using the same exclusion criteria as for patients with CHD, except that they were also required to be clinically free of CHD by history, and there were no exclusion criteria concerning low-density lipoprotein choles1026 THE AMERICAN JOURNAL OF CARDIOLOGY姞
VOL. 95
terol. These subjects were studied only 1 time in the fasting state and 4 hours after the same fat-rich meal. Data on lipid results in the fasting and postprandial states at the end of the baseline period in patients who had CHD are not provided because they were not significantly different from values at the end of the 8-week placebo period. Lipoprotein analyses: Blood was drawn from each subject after a 12-hour overnight fast into tubes that contained ethylenediaminetetraacetic acid (final concentration 0.15%). Plasma was separated by centrifugation (2,500 rpm at 4°C for 20 minutes). Plasma concentrations of total cholesterol, triglycerides, and high-density lipoprotein cholesterol were measured fresh according to standard enzymatic methods and reagents obtained from Abbott Diagnostics (Irving, Texas) as described previously.4 Low-density lipoprotein cholesterol was measured directly with kits from Equal Diagnostics (Exton, Pennsylvania). Apolipoprotein-AI and apolipoprotein-B were measured by immunoturbidimetric assays from Wako Chemicals (Richmond, Virginia), and lipoprotein(a) was measured by mass with Macra kits5 (Wampole Laboratories, Cranbury, New Jersey) and by cholesterol content using kits from Genzyme Diagnostics (Cambridge, Massachusetts).6 Data on lipoprotein(a) mass are not reported because none of the statins induced statistically significant effects on this parameter compared with placebo or each other. Among-run precision studies indicated coefficients of variation of ⬍2% for total cholesterol, ⬍3% for triglycerides, ⬍4% for lowdensity lipoprotein cholesterol, ⬍5% for high-density lipoprotein cholesterol, apolipoprotein-AI, and lipoprotein(a), and ⬍10% for apolipoprotein-B and lipoprotein(a) cholesterol. Lipid assays are standardized through the Centers for Disease Control Lipid Standardization Program (Atlanta, Georgia). Remnant lipoprotein analyses: Isolation of remnant lipoprotein was based on removal of apolipoproteinAI– containing particles (high-density lipoprotein) and most apolipoprotein-B– containing particles (low-density lipoprotein, nascent very low-density lipoprotein, and nascent chylomicrons) using a previously described immunoseparation technique (JIMRO, Japan Immunoresearch Laboratories, Takasaki, Japan) that has been shown to leave particles characteristic of very low-density lipoprotein remnants and chylomicron remnants in the unbound fraction.7 Briefly, monoclonal antibodies to apolipoprotein-AI and specific monoclonal antibodies to apolipoprotein-B (JI-H), which do not recognize partially hydrolyzed, apolipoprotein-E– enriched lipoprotein remnants, were immobilized on agarose gel. Remnant lipoprotein cholesterol and remnant lipoprotein triglyceride concentrations were measured in plasma aliquots and stored at ⫺80°C until analysis. Thawed plasma was incubated with the gel for 2 hours, after which the gel, which contained the bound (nonremnant lipoprotein) lipoproteins, was precipitated with low-speed centrifugation (5 minutes at 135g). Remnant lipoprotein cholesterol and remnant lipoprotein triglyceride levels were then measured in the unbound supernates on an Abbott Spectrum CCx chemistry analyzer (Abbott DiMAY 1, 2005
TABLE 1 Fasting Concentrations for Controls and Patients on Placebo Measurement hs-CRP* (mg/L) Lp-PLA2 (ng/ml) Total cholesterol (mg/dl) Triglycerides* (mg/dl) LDL cholesterol (mg/dl) HDL cholesterol (mg/dl) Non–HDL cholesterol (mg/dl) Remnant lipoprotein cholesterol* (mg/dl) Remnant lipoprotein triglycerides* (mg/dl) Total cholesterol/HDL cholesterol ratio Nuclear magnetic resonance of small LDL (mg/dl) Nuclear magnetic resonance of LDL particle number Nuclear magnetic resonance of large HDL (mg/dl) hs-CRP ⬍3.0 mg/L hs-CRP ⬍2.0 mg/L LDL cholesterol ⬍100 mg/dl LDL cholesterol ⬍70 mg/dl Non–HDL cholesterol ⬍130 mg/dl Triglycerides ⬍150 mg/dl
Controls (n ⫽ 84)
Patients (n ⫽ 84)
3.3 ⫾ 6.2 230 ⫾ 81 179 ⫾ 37 92 ⫾ 50 98 ⫾ 39 52 ⫾ 14 126 ⫾ 37 6.3 ⫾ 2.0 18.5 ⫾ 14.6 3.6 ⫾ 1.2 32 ⫾ 28 1,266 ⫾ 334 30 ⫾ 13 76% 67% 66% 27% 66% 88%
3.6 ⫾ 5.5 260 ⫾ 91 277 ⫾ 68 204 ⫾ 85 184 ⫾ 56 37 ⫾ 10 240 ⫾ 69 11.3 ⫾ 11.2 42.4 ⫾ 25.8 7.9 ⫾ 2.9 96 ⫾ 66 2,455 ⫾ 711 14 ⫾ 9 70% 53% 0% 0% 0% 31%
Differences (%) 1 29 59 147 112 ⫺24 104 57 151 123 121 106 ⫺58
(⫺46 to 341)† ⫾ 69† ⫾ 45# (59 to 250) ⫾ 99# ⫾ 28# ⫾ 83# (0 to 22)# (20 to 360)# (61 to 208)# (⫺8 to 291)# ⫾ 98# (⫺76 to ⫺4)#
Normally distributed values are reported as mean ⫾ SD; other values are reported as medians (interquartile ranges). All p values were derived from 2-tailed, paired t-test analysis. *Values for hs-CRP, triglycerides, remnant lipoprotein cholesterol, and remnant lipoprotein triglycerides were log-transformed for statistical analysis.†p ⬍0.05; ‡p ⬍0.01; §p ⬍0.001;#p ⬍0.0001 for comparisons with placebo; the dataset comprised 12 women and 72 men; mean percentage ⫾ SD or median percent change from placebo. HDL ⫽ high-density lipoprotein; LDL ⫽ low-density lipoprotein.
TABLE 2 Fasting Concentrations on Placebo and Atorvastatin for All Subjects Measurement hs-CRP* (mg/L) Lp-PLA2 (ng/ml) Creatine phosphokinase* (U/L) Aspartate aminotransferase* (U/L) Arginine aminotransferase* (U/L) Total cholesterol (mg/dl) Triglycerides* (mg/dl) LDL cholesterol (mg/dl) HDL cholesterol (mg/dl) Non–HDL cholesterol (mg/dl) Apolipoprotein-AI (mg/dl) Apolipoprotein-B (mg/dl) Lipoprotein(a) cholesterol* (mg/dl) Lipoprotein (a) mass* (mg/dl) Remnant lipoprotein cholesterol* (mg/dl) Remnant lipoprotein triglycerides* (mg/dl) Total cholesterol/HDL cholesterol ratio Nuclear magnetic resonance of small LDL (mg/dl) Nuclear magnetic resonance LDL particle number Nuclear magnetic resonance of large HDL (mg/dl) hs-CRP ⬍3.0 mg/L hs-CRP ⬍2.0 mg/L LDL cholesterol ⬍100 mg/dl LDL cholesterol ⬍70 mg/dl Non–HDL cholesterol ⬍130 mg/dl Triglycerides ⬍150 mg/dl
Placebo (n ⫽ 84)
Atorvastatin (n ⫽ 84)
3.5 ⫾ 5.4 260 ⫾ 91 160 ⫾ 141 24 ⫾ 6 22 ⫾ 8 277 ⫾ 68 204 ⫾ 85 184 ⫾ 56 37 ⫾ 10 240 ⫾ 69 112 ⫾ 22 180 ⫾ 42 14.8 ⫾ 13.6 23.9 ⫾ 21.3 11.3 ⫾ 11.2 42.4 ⫾ 25.8 7.9 ⫾ 2.9 96 ⫾ 66 2,455 ⫾ 711 14 ⫾ 9 70% 53% 0% 0% 0% 31%
2.5 ⫾ 3.4 190 ⫾ 49 186 ⫾ 246 27 ⫾ 8 28 ⫾ 10 168 ⫾ 41 137 ⫾ 70 96 ⫾ 39 40 ⫾ 11 128 ⫾ 41 115 ⫾ 24 112 ⫾ 33 14.3 ⫾ 15.1 25.8 ⫾ 23.4 6.6 ⫾ 3.9 25.9 ⫾ 19.6 4.4 ⫾ 1.4 53 ⫾ 41 1,382 ⫾ 482 20 ⫾ 9 81% 65% 69% 24% 63% 67%
Differences (%) ⫺34 ⫺26 8 9 19 ⫺40 ⫺35 ⫺48 9 ⫺49 4 ⫺36 ⫺16 0 ⫺34 ⫺42 ⫺41 ⫺55 ⫺41 55
(⫺50 to ⫺5)‡ ⫾ 26# (⫺13 to 29) (0 to 25)# (8 to 41)# ⫾ 13# (⫺53 to ⫺12)# ⫾ 16# ⫾ 17† ⫾ 20%# ⫾ 18%† ⫾ 18# (⫺68 to 101)# (⫺7 to 21) (⫺57 to ⫺10)# (⫺63 to ⫺5)# (⫺52 to ⫺37)# (⫺73 to ⫺22)# ⫾ 21# (14 to 103)#
Normally distributed values are reported as mean ⫾ SD; other values are reported as medians (interquartile ranges). All p values were derived from 2-tailed, paired t-test analysis. *Values for hs-CRP, creatine phosphokinase, aspartate aminotransferase, arginine aminotransferase, triglycerides, lipoprotein(a) cholesterol, lipoprotein(a) mass, remnant lipoprotein cholesterol, and remnant lipoprotein triglycerides were log-transformed for statistical analysis. † p ⬍0.05; ‡p ⬍0.01; §p ⬍0.001; #p ⬍0.0001 for comparisons with placebo; the dataset comprised 12 women and 72 men; mean percentage ⫾ SD or median percent change from placebo. Abbreviations as in Table 1.
CORONARY ARTERY DISEASE/DIRECT COMPARISONS OF STATIN EFFECTS ON CRP AND LP-PLA2
1027
agnostics) that used 2-reagent enzymatic, colorimetric assays that contained a sensitive chromophore (Kyowa Medex, Tokyo, Japan). Precision studies yielded among-run remnant lipoprotein cholesterol imprecision for 2 levels of remnant lipoprotein control over 20 runs, i.e., 9.1% at 7 mg/dl and 7.3% at 24 mg/dl. Among-run remnant lipoprotein triglyceride imprecision levels for the same controls were 8.3% at 22 mg/dl and 5.0% at 109 mg/dl.7 Nuclear magnetic resonance analysis: Lipoprotein subclass distributions were determined by proton nuclear magnetic resonance spectroscopy.8 Subclasses were separated as chylomicrons (⬎200 nm); large (60 to 200 nm), medium (35 to 60 nm), and small (27 to 35 nm) very low-density lipoproteins; intermediatedensity lipoproteins (23 to 27 nm); large (21 to 23 nm), medium (20 to 21 nm), and small (18 to 20 nm) low-density lipoproteins; and large (9 to 13 nm), medium (8 to 9 nm), and small (7 to 8 nm) high-density lipoproteins. Very low-density lipoprotein subclass concentrations are expressed as milligrams per deciliter of triglycerides, and other lipoprotein subclass concentrations are expressed as milligrams per deciliter of cholesterol. Data on the chylomicron and intermediate-density lipoprotein fractions were omitted because the levels of lipids within the fractions were low, with no statistically significant differences across groups. Clinical chemistry analyses: Standardized serum measurements of liver transaminases and creatine phosphokinase were carried out at each 4-week time point in the study. Measurements of hs-CRP were performed on a Hitachi 911 (Roche Diagnostics, Indianapolis, Indiana) using the CRP-UL kit from Wako Chemicals. Within- and between-run coefficients of variation were ⬍5%. Measurements of LpPLA2 mass were performed by an enzyme-linked immunosorbent assay (diaDexus, Inc., South San Francisco, California) at the diaDexus facility in a blinded fashion. Within- and between-run coefficients of variation were ⬍7.5%. Statistical analysis: Statistical analyses were performed with patients who had CHD by comparing the effects of each statin at 40 mg/day with placebo, and its effects were compared with those observed on the same dose of atorvastatin in the same patients. The effects of statins at 40 mg/day in the fasting and fed states were compared with placebo and with values obtained on the same dose of atorvastatin. Values in control subjects who did not have CHD are provided for comparison. Student’s t test was used to compare mean levels of continuous measurements, and a chi-square statistic was calculated for categorical factors. For continuous measurements that were highly skewed, we performed comparisons using log-transformed values, although we report the untransformed means and SDs for each group. Analyses were also conducted separately by gender, but we report only pooled results because men and women had similar responses to statins. Similarly, we do not report baseline lipid data because these results were not statistically different from values obtained on placebo. For the postprandial studies, 1028 THE AMERICAN JOURNAL OF CARDIOLOGY姞
VOL. 95
statistical analyses were done to compare fasting values with postprandial values, values on medication with those on placebo, and values in patients who had CHD and used placebo and at 40 mg/day of statins with values on atorvastatin. We also carried out correlations for change in hs-CRP and Lp-PLA2 versus change in other parameters influenced by atorvastatin. One outlier that had a 8,531% increase in hs-CRP on atorvastatin was omitted from the correlation analysis.
RESULTS Baseline demographic information for subjects who had CHD (n ⫽ 84, 87% men and 13% women) and for age- and gender-matched controls (n ⫽ 84) were mean age (62 ⫾ 9 years), diabetes mellitus (0% of controls and 5% of patients), hypertension (31% of controls and 56% of patients), smoking (20% of controls and 18% of patients), and body mass index (26 ⫾ 4 kg/m2 for controls and 28 ⫾ 4 kg/m2 for patients). Mean values ⫾ SD for control subjects and subjects who had CHD and were using placebo with respect to fasting plasma levels of hs-CRP, Lp-PLA2, total cholesterol, triglycerides, lowdensity lipoprotein cholesterol, high-density lipoprotein cholesterol, non– high-density lipoprotein cholesterol, apolipoprotein-AI, apolipoprotein-B, and lipoprotein(a) or lipoprotein(a) cholesterol levels and to levels of remnant lipoprotein cholesterol and triglyceride, small lowdensity lipoprotein, low-density lipoprotein particle number, and large high-density lipoprotein are presented in Table 1. In addition, values for the ratio of total cholesterol to high-density lipoprotein cholesterol and percentage of patients who had hs-CRP levels ⬍3.0 and ⬍2.0 mg/L, low-density lipoprotein cholesterol levels ⬍100 and ⬍70 mg/dl, non– high-density lipoprotein cholesterol levels ⬍130 mg/dl, and triglyceride levels ⬍150 mg/dl are presented. Patients had significantly (p ⬍0.05) higher levels of hs-CRP, Lp-PLA2, total cholesterol, low-density lipoprotein cholesterol, triglycerides, remnant lipoprotein cholesterol, remnant lipoprotein triglycerides, total cholesterol/high-density lipoprotein cholesterol ratio, small low-density lipoprotein, and total low-density lipoprotein particle number and significantly (p ⬍0.05) lower levels of high-density lipoprotein cholesterol and large highdensity lipoprotein than did controls (Table 1). Median hs-CRP and mean Lp-PLA2 values were 51% and 29% higher in patients than in controls, respectively (p ⬍0.05). The effects of atorvastatin at 40 mg/day versus placebo on various parameters in 84 patients who had CHD are presented in Table 2. Median decreases in hs-CRP were 32% (p ⬍0.01) and mean decreases in Lp-PLA2 were 26% (p ⬍0.0001). There was a high degree of variability in the hs-CRP response to atorvastatin but somewhat less variability in the Lp-PLA2 response (Figure 1). For these parameters, there were increases in some patients (5 patients developed dramatic increases in hs-CRP and were excluded from the data shown in Figure 1 but not from the data analyses), but, in general, atorvastatin decreased these values substantially, with almost normal values for low-density lipoprotein cholesterol and triglycerides (Table 2). These increases may have been due to concurrent infections. Atorvastatin also MAY 1, 2005
which was moderately significant, and nonfasting hs-CRP level, which, although decreased by 29%, was not significant. Of note, for all these parameters, there were highly significant differences (p ⬍0.0001) between patients on placebo and controls except for Lp-PLA2. However, differences were very significant in the fed state because Lp-PLA2 was decreased by 21% in the fed state in controls but did not change significantly in patients. Moreover, although atorvastatin came close to normalizing Lp-PLA2, lowdensity lipoprotein cholesterol, and remnant lipoprotein cholesterol in patients versus controls, it did not do so for hs-CRP (especially in the fed state) or triglycerides. Table 4 presents a comparison of the effects of the different statins on fasting and postprandial levels of hsCRP and Lp-PLA2 and of low-density lipoprotein cholesterol. Atorvastatin was not only significantly more efficacious than other statins in decreasing low-density lipoprotein cholesterol but also the only statin that significantly decreased levels of hs-CRP and Lp-PLA2 in the fasting or fed state. Because of these effects, we reFIGURE 1. Variability (percent change from values on placebo) in response to atorvalated changes in fasting levels of hsstatin (40 mg/day) for hs-CRP, Lp-PLA2, creatine phosphokinase (CPK), and aspartate CRP and Lp-PLA2 induced by atoraminotransferase (AST) are shown. For hs-CRP, 5 extreme outliers have been omitted vastatin to other variables based on from the figure (ⴙ8,531%, ⴙ495%, ⴙ316%, ⴙ295%, and ⴙ242%); for CPK, 2 outlicorrelation coefficient analysis. For ers have been omitted from the figure (ⴙ592% and ⴙ499%). Lp-PLA2, the change was significantly correlated (p ⬍0.01) with changes in total cholesterol (r ⫽ resulted in modest median increases in creatine phos- 0.49), low-density lipoprotein cholesterol (r ⫽ 0.43), phokinase, aspartate aminotransferase, and alanine ami- and ratio of total cholesterol to high-density lipopronotransferase of 8%, 9%, and 19%, respectively, with tein cholesterol (r ⫽ 0.44). For hs-CRP, the change only the latter 2 changes being significant (p ⬍0.0001). was not significantly correlated with change in any The variability in response to atorvastatin by creatine parameter. phosphokinase and alanine aminotransferase is shown in Figure 1. As we previously reported, atorvastatin is very DISCUSSION We previously documented that fasting values of effective in decreasing levels of total cholesterol, lowdensity lipoprotein cholesterol, non– high-density li- low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, triglycerides, remnant lipoprotein poprotein cholesterol, triglycerides, and remnant licholesterol, remnant lipoprotein triglycerides, low-den- poprotein cholesterol are correlated with postprandial sity lipoprotein particle number, and small low-density values and that differences between patients who have lipoprotein and in increasing levels of large high-density CHD and controls with respect to these parameters lipoprotein as assessed by nuclear magnetic resonance can readily be detected in the fasting state.1 Therefore, a fat challenge is not necessary. Moreover, we found spectroscopy.2,3 Data that examined the effects of atorvastatin on that low-density lipoprotein cholesterol and high-denfasting and fed levels of hs-CRP, Lp-PLA2, low-density sity lipoprotein cholesterol can readily be measured in lipoprotein cholesterol, triglyceride, and remnant li- the fed state, provided that newer direct methods are poprotein cholesterol in 61 patients who had CHD are used and that these values are just slightly lower in the listed in Table 3. Decreases induced by atorvastatin for fed state than in the fasting state.1 In addition, we documented that atorvastatin is all these parameters were very similar in the fasting and fed states, and all decreases were highly significant very effective in normalizing many of the lipid abnor(p ⬍0.0001), except for fasting hs-CRP level (p ⬍0.05), malities found in patients who have CHD compared CORONARY ARTERY DISEASE/DIRECT COMPARISONS OF STATIN EFFECTS ON CRP AND LP-PLA2
1029
1030 THE AMERICAN JOURNAL OF CARDIOLOGY姞
TABLE 3 Fasting and Postprandial Concentrations on Placebo and 40 mg/day Atorvastatin Versus Matched Controls Patients With CHD Measurement
VOL. 95 MAY 1, 2005
Fasting hs-CRP* (mg/L) 4-h postprandial hs-CRP* (mg/L) Median postprandial change Fasting LP-PLA2 (ng/ml) 4-h postprandial Lp-PLA2 (ng/ml) Mean postprandial change Fasting Triglycerides* (mg/dl) 4-h Postprandial triglycerides* (mg/dl) Median postprandial change Fasting LDL cholesterol (mg/dl) 4-h postprandial LDL cholesterol (mg/dl) Mean postprandial change Fasting remnant lipoprotein cholesterol* (mg/dl) 4-h postprandial remnant lipoprotein cholesterol* (mg/dl) Median postprandial change
Controls (n ⫽ 61) 2.2 2.7 ⫹22% 230 177 ⫺21 95 151 ⫹53% 95 89 ⫺6 6.2 9.6
⫾ 4.5 ⫾ 5.7‡ (8 to 41) ⫾ 81 ⫾ 45储 ⫾ 13% ⫾ 52 ⫾ 84储 (34 to 83) ⫾ 34 ⫾ 32储 ⫾ 9% ⫾ 1.8 ⫾ 4.5储
⫹31% (13 to 50)
Placebo (n ⫽ 61) 2.6 3.1 ⫹2% 254 266 ⫹9 196 348 ⫹81% 164 155 ⫺5 11.8 19.5
⫾ 3.2 ⫾ 3.8 (⫺8 to 14) ⫾ 90 ⫾ 84 ⫾ 25% ⫾ 96 ⫾ 147储 (48 to 113) ⫾ 43 ⫾ 41储 ⫾ 8% ⫾ 13.9 ⫾ 15.1储
⫹42% (34 to 52)
Percent Differences Atorvastatin (n ⫽ 61) 2.3 3.0 ⫹2% 185 193 ⫹6 134 231 ⫹64% 91 84 ⫺7 6.5 11.6
⫾ 3.7 ⫾ 4.1 (⫺10 to 32) ⫾ 48 ⫾ 61 ⫾ 26% ⫾ 60 ⫾ 124储 (42 to 106) ⫾ 33 ⫾ 30储 ⫾ 10% ⫾ 4.0 ⫾ 7.1储
Controls/Placebo
Controls/Atorvastatin
Placebo/Atorvastatin
⫹68% (⫺39 to 504) ⫹14% (⫺36 to 527)
⫹20% (⫺61 to 358) ⫹53% (⫺34 to 352)
⫺34% (⫺54 to ⫺7)¶ ⫺29% (⫺51 to 5)
⫹23% (⫺20 to 44) ⫹62% (17 to 89)††
⫺10% (⫺36 to ⫺2)** ⫹17% (⫺17 to 37)
⫺23% (⫺36 to ⫺13)†† ⫺25% (⫺41 to ⫺13)††
⫹128% (48 to 207)†† ⫹141% (55 to 291)††
⫹48% (5 to 121)†† ⫹46% (6 to 162)††
⫺36% (⫺46 to ⫺14)†† ⫺39% (⫺50 to ⫺13)††
⫹94% (37 to 143)†† ⫹95% (31 to 134)††
⫹6% (⫺33 to 23) ⫹5% (⫺37 to 24)
⫺43% (⫺57 to ⫺36)†† ⫺44% (⫺59 to ⫺39)††
⫺4% (⫺36 to 30) ⫹10% (⫺25 to 100)
⫺40% (⫺50 to ⫺23)†† ⫺41% (⫺52 to ⫺24)††
¶
⫹55% (10 to 102)†† ⫹100% (30 to 171)††
⫹43% (29 to 54)
Normally distributed values are reported as mean ⫾ SD; other values are reported as medians (interquartile ranges). All p values were derived from 2-tailed, paired t-test analysis. *Values for hs-CRP, triglycerides, and remnant lipoprotein cholesterol were log-transformed for statistical analysis. † p ⬍0.05; ‡p ⬍0.01; §p ⬍0.001; 储p ⬍0.0001 for comparisons of fasting and postprandial values. ¶ p ⬍0.05; #p ⬍0.01; **p ⬍0.001; ††p ⬍0.0001 for paired t-test comparison among treatment groups; the dataset comprised 9 women and 52 men; controls were matched by age and gender to patients mean percentage ⫾ SD or median percent change from placebo. Abbreviation as in Table 1.
TABLE 4 Percentage Change in Fasting Lipoprotein-associated Phospholipase A2 Concentrations From Placebo After 40 mg/day Treament With Fluvastatin, Lovastatin, Pravastatin, or Simvastatin Versus Atorvastatin Parameter hs-CRP* (mg/L) LP-PLA2 (ng/ml) LDL cholesterol (mg/dl) Parameter hs-CRP* (mg/L) LP-PLA2 (ng/ml) LDL cholesterol (mg/dl) Parameter hs-CRP* (mg/L) LP-PLA2 (ng/ml) LDL cholesterol (mg/dl) Parameter hs-CRP* (mg/L) LP-PLA2 (ng/ml) LDL cholesterol (mg/dl)
Controls
Placebo
Fluvastatin
Atorvastatin
3.3 ⫾ 7.6 240 ⫾ 86 92 ⫾ 32
5.5 ⫾ 8.5 272 ⫾ 106 161 ⫾ 39¶¶
⫹2 (⫺27 to 37) ⫺1 ⫾ 19% ⫺16 ⫾ 16%储
⫺34 (⫺69 to 18)‡‡ ⫺24 ⫾ 17%§†† ⫺45 ⫾ 13%储††
Controls
Placebo
Lovastatin
Atorvastatin
2.9 ⫾ 5.0 232 ⫾ 78 100 ⫾ 31
3.0 ⫾ 0.3 233 ⫾ 69 157 ⫾ 28¶¶
Controls
Placebo
Pravastatin
4.6 ⫾ 6.2 246 ⫾ 91 110 ⫾ 43
3.1 ⫾ 4.0 264 ⫾ 80 193 ⫾ 53¶¶
⫹3 (⫺26 to 25) ⫺6 ⫾ 22% ⫺24 ⫾ 13%储
Controls
Placebo
Simvastatin
Atorvastatin
3.1 ⫾ 6.3 219 ⫾ 60 90 ⫾ 30
1.8 ⫾ 0.9 247 ⫾ 90 200 ⫾ 78¶¶
⫹26 (⫺26 to 48) ⫺10 ⫾ 26% ⫺37 ⫾ 16%储
⫺36 (⫺57 to ⫺20)‡ ⫺21 ⫾ 16%‡ ⫺45 ⫾ 15%储¶
1
⫹16 (⫺7 to 41) ⫺6 ⫾ 24% ⫺28 ⫾ 18%储
⫺32 (⫺54 to ⫺19)† ⫺16 ⫾ 26%‡¶ ⫺45 ⫾ 14%储†† Atorvastatin ⫺31 (⫺39 to ⫺7)§ ⫺26 ⫾ 17%储# ⫺45 ⫾ 14%储††
Normally distributed values are reported as mean ⫾ SDs; other values are reported as median and interquartile range. *Values for hs-CRP were log transformed before statistical analysis. All p values were derived from unpaired and paired 2-tailed t-test analyses. †p ⬍0.05; ‡p ⬍0.01; §p ⬍0.001; 储p ⬍0.0001 for paired comparisons of treatment with placebo; ¶p ⬍0.05; #p ⬍0.01; **p ⬍0.001; ††p ⬍0.0001 for paired comparisons between same doses of statins. ‡‡p ⬍0.05; §§p ⬍0.01; 储储p ⬍0.001; ¶¶p ⬍0.0001 for unpaired comparisons of controls with placebo; the data sets comprised 1 woman, 25 men (F), 2 women, 18 men (L), 6 women, 11 men (P), 2 women, 14 men (S); mean ⫾ SD% or median percent change from placebo. Abbreviation as in Table 1.
with controls, including increased low-density lipoprotein and non– high-density lipoprotein cholesterol, cholesterol in small dense low-density lipoprotein, total low-density lipoprotein particle number, triglycerides, triglycerides in different very low-density lipoprotein classes, postprandial triglycerides, and fasting and postprandial remnant lipoprotein cholesterol and triglycerides.2 Atorvastatin was found to decrease lipoprotein(a) cholesterol but not total lipoprotein(a) mass.2 Moreover, we noted that, although atorvastatin had a modest effect in increasing highdensity lipoprotein cholesterol, it significantly increased, but did not normalize, cholesterol in large high-density lipoprotein particles.2 In addition, we documented a wide range of patient variability in response to statins with regard to lipid modification.2,9 For this reason, statin comparisons should be done in the same subjects to remove genetic confounders in studying the efficacy of these statins. Other investigators have examined the effects of statins on lipoproteins, including remnants, in the fasting and postprandial states, but there have been no direct comparisons of all of these statins in the same subjects with respect to hs-CRP and Lp-PLA2 in the fasting and fed states. In addition, we documented that atorvastatin is more effective than other statins tested (fluvastatin, lovastatin, pravastatin, and simvastatin) in normalizing lipid abnormalities in patients who have CHD.3 The focus of this examination has been on hs-CRP and Lp-PLA2 as measured in the fasted and fed states in patients who have CHD and age- and gendermatched controls and examining the effects of atorvastatin and other statins on these parameters in patients
who have CHD. Other investigators have documented that hs-CRP and Lp-PLA2 are important inflammatory markers and markers of CHD risk.10 –13 Moreover, it has been documented that Lp-PLA2 is carried on lipoproteins, especially low-density lipoprotein.14,15 It has also been reported that statins can decrease the levels of these parameters.16 –18 A novel feature of our study is the finding that feeding a meal rich in saturated fat and cholesterol caused significant increases of 22% in hs-CRP and significant decreases of 21% in Lp-PLA2 in controls but not in patients. Therefore, differences in hs-CRP are somewhat less in patients than in controls in the fed state versus the fasting state (median difference 114% vs 168%), but differences were greater for LpPLA2 (mean difference 62% vs 23%). These findings need to be confirmed in larger studies. We have also documented that atorvastatin results in similar decreases in hs-CRP (34% and 29%) and Lp-PLA2 (23% and 25%) in patients who have CHD in the fasting and fed states. We have also documented that atorvastatin is much more effective than other statins tested (fluvastatin, lovastatin, pravastatin, and simvastatin) in decreasing not only lipid parameters in the fasting and fed states but also hs-CRP and Lp-PLA2. Other statins did not significantly decrease hs-CRP levels in our study in contrast to some other studies.19 These differences may be related to differences in sample size, duration of therapy, and baseline characteristics of the study subjects. A 17% decrease in Lp-PLA2 levels was reported with pravastatin in the West of Scotland Study, which was greater than what we observed.15 A recent intravascular ultrasound study that
CORONARY ARTERY DISEASE/DIRECT COMPARISONS OF STATIN EFFECTS ON CRP AND LP-PLA2
1031
involved ⬃500 evaluable patients who took atorvastatin (80 mg/day) for 18 months reported that hs-CRP decreased by 36% versus baseline, whereas pravastatin decreased this parameter by only 5%, which is consistent with our data.20 We tested for correlations of atorvastatin-induced changes in these parameters, with changes in lipid and other parameters. Changes in Lp-PLA2 were correlated with changes in total cholesterol, low-density lipoprotein cholesterol, and the total cholesterol/highdensity lipoprotein cholesterol ratio, consistent with Lp-PLA2 being carried on low-density lipoprotein, and these particles were substantially decreased by atorvastatin. For changes in hs-CRP, no significant correlations with atorvastatin-induced lipid changes were noted. In our data, hs-CRP increased with fat feeding as triglycerides and remnants increased in controls but not in patients. The mechanisms responsible for these interactions require further elucidation.21–25 The focus of current guidelines for lipid modification in patients who have CHD is for drastic decreases in low-density lipoprotein cholesterol.26 In our view, in the future, the emphasis may shift toward optimizing other lipoprotein particles, namely remnants, high-density lipoprotein, and lipoprotein(a), improving the ratio of total cholesterol to high-density lipoprotein cholesterol, and decreasing levels of inflammatory markers, such as hsCRP and LpPLA2, to decrease the risk for CHD. 1. Schaefer EJ, Audelin MC, McNamara JR, Shah PK, Tayler T, Daly JA, Augustin JL, Seman LJ, Rubenstein JJ. Comparison of fasting and postprandial plasma lipoproteins in subjects with and without coronary heart disease. Am J Cardiol 2001;88:1129 –1133. 2. Schaefer EJ, McNamara JR, Tayler T, Daly JA, Gleason JA, Seman LJ, Ferrari A, Rubenstein JJ. Effects of atorvastatin on fasting and postprandial lipoprotein subclasses in coronary disease patients versus control subjects. Am J Cardiol 2002;90:689 – 696. 3. Schaefer EJ, McNamara JR, Tayler T, Daly JA, Gleason JA, Seman LJ, Ferrari A, Rubenstein JJ. Direct comparisons of statin effects on fasting and postprandial lipoproteins in coronary heart disease patients versus control subjects. Am J Cardiol 2004;93:31–39. 4. McNamara JR, Schaefer EJ. Automated enzymatic standardized lipid analyses for plasma and lipoprotein fractions. Clin Chim Acta 1987;166:1– 8. 5. Jenner JL, Ordovas JM, Lamon-Fava S, Schaefer MM, Wilson PW, Castelli WP, Schaefer EJ. Effects of age, sex, and menopausal status on plasma lipoprotein(a) levels. The Framingham Offspring Study. Circulation 1993;87:1135–1141. 6. Seman LJ, Jenner JL, McNamara JR, Schaefer EJ. Quantitation of plasma lipoprotein(a) by cholesterol assay of lectin-bound lipoprotein(a). Clin Chem 1994;40:400 – 403. 7. McNamara JR, Shah PK, Nakajima K, Cupples LA, Wilson PWF, Ordovas JM, Schaefer EJ. Remnant lipoprotein cholesterol and triglyceride: reference ranges from the Framingham Heart Study. Clin Chem 1998;44:1224 –1232. 8. Otvos JD, Jeyarajah EJ, Bennett DW, Krauss RM. Development of a proton nuclear magnetic resonance spectroscopic method for determining plasma lipoprotein concentrations and subspecies distributions from a single, rapid measurement. Clin Chem 1992;38:1632–1638.
1032 THE AMERICAN JOURNAL OF CARDIOLOGY姞
VOL. 95
9. Pedro-Botet J, Schaefer EJ, Bakker-Arkema RG, Black DM, Stein FM, Corella D, Ordovas JM. Apolipoprotein E genotype affects plasma lipid response to atorvastatin in a gender-specific manner. Atherosclerosis 2001;158:183–194. 10. Ridker PM, Rifai N, Rose L, Buring JE, Cook NR. Comparison of C-reactive protein and low-density lipoprotein cholesterol levels in the prediction of first cardiovascular events. N Engl J Med 2002;347:1557–1565. 11. Sattar N, Gaw A, Scherbakova O, Ford I, O’Reilly DS, Haffner SM, Isles C, Macfarlane PW, Packard CJ, Cobbe SM, Shepherd J. Metabolic syndrome with and without C-reactive protein as a predictor of coronary heart disease and diabetes in the West of Scotland Coronary Prevention Study. Circulation 2003; 108:414 – 419. 12. Packard CJ, O’Reilly DS, Caslake MJ, McMahon AD, Ford I, Cooney J, Macphee CH, Suckling KE, Krishna M, Wilkinson FE, et al. Lipoproteinassociated phospholipase A2 as an independent predictor of coronary heart disease. West of Scotland Coronary Prevention Study Group. N Engl J Med 2000;343:1148 –1155. 13. Ballantyne CM, Hoogeveen RC, Bang H, Coresh J, Folsom AR, Heiss G, Sharrett AR. Lipoprotein-associated phospholipase A2, high-sensitivity C-reactive protein, and risk for incident coronary heart disease in middle-aged men and women in the Atherosclerosis Risk in Communities (ARIC) study. Circulation 2004;109:837– 842. 14. MacPhee CH, Moores KE, Boyd HF, Dhanak D, Ife RJ, Leach CA, Leake DS, Milliner KJ, Patterson RA, Suckling KE, et al. Lipoprotein-associated phospholipase A2, platelet-activating factor acetylhydrolase, generates two bioactive products during the oxidation of low-density lipoprotein: use of a novel inhibitor. Biochem J 1999;338:479 – 487. 15. Caslake MJ, Packard CJ. Lipoprotein-associated phospholipase A2 (plateletactivating factor acetylhydrolase) and cardiovascular disease. Curr Opin Lipidol 2003;14:347–352. 16. Horne BD, Muhlestein JB, Carlquist JF, Bair TL, Madsen TE, Hart NI, Anderson JL, Intermountain Heart Collaborative (IHC) Study Group. Statin therapy interacts with cytomegalovirus seropositivity and high C-reactive protein in reducing mortality among patients with angiographically significant coronary disease. Circulation 2003;107:258 –263. 17. Marz W, Winkler K, Nauck M, Bohm BO, Winkelmann BR. Effects of statins on C-reactive protein and interleukin-6 (the Ludwigshafen Risk and Cardiovascular Health study). Am J Cardiol 2003;92:305–308. 18. Wang TD, Chen WJ, Lin JW, Cheng CC, Chen MF, Lee YT. Efficacy of fenofibrate and simvastatin on endothelial function and inflammatory markers in patients with combined hyperlipidemia: relations with baseline lipid profiles. Atherosclerosis 2003;170:315–323. 19. Jialal I, Stein D, Balis D, Grundy SM, Adams-Huet B, Devaraj S. Effect of hydroxymethyl glutaryl coenzyme a reductase inhibitor therapy on high sensitive C-reactive protein levels. Circulation 2001;103:1933–1935. 20. Nissen SE, Tuzcu EM, Schoenhagan P, Brown BG, Ganz P, Vogel RA, Crowe T, Howard G, Cooper CJ, Brodie B, et al. Effect of intensive compared with moderate lipid-lowering therapy on progression of coronary atherosclerosis: a randomized controlled trial. JAMA 2004;291:1071–1080. 21. Koenig W, Lowel H, Baumert J, Meisinger C. C-reactive protein modulates risk prediction based on the Framingham score: implications for future risk assessment: results from a large cohort study in southern Germany. Circulation 2004;109:1349 –1353. 22. Saito M, Ishimitsu T, Minami J, Ono H, Ohrui M, Matsuoka H. Relations of plasma high-sensitivity C-reactive protein to traditional cardiovascular risk factors. Atherosclerosis 2003;167:73–79. 23. Luc G, Bard JM, Juhan-Vague I, Ferrieres J, Evans A, Amouyel P, Arveiler D, Fruchart JC, Ducimetiere P, PRIME Study Group. C-reactive protein, interleukin-6, and fibrinogen as predictors of coronary heart disease: the PRIME Study. Arterioscler Thromb Vasc Biol 2003;23:1255–1261. 24. van der Meer IM, de Maat MP, Kiliaan AJ, van der Kuip DA, Hofman A, Witteman JC. The value of C-reactive protein in cardiovascular risk prediction: the Rotterdam Study. Arch Intern Med 2003;163:1323–1328. 25. St-Pierre AC, Bergeron J, Pirro M, Cantin B, Dagenais GR, Despres JP, Lamarche B. Quebec Cardiovascular Study. Effect of plasma C-reactive protein levels in modulating the risk of coronary heart disease associated with small, dense, low-density lipoproteins in men (The Quebec Cardiovascular Study). Am J Cardiol 2003;91:555–558. 26. Grundy SM, Cleeman JI, Merz CN, Brewer HB Jr, Clark LT, Hunninghake DB, Pasternak RC, Smith SC Jr, Stone NJ. National Heart, Lung, and Blood Institute; American College of Cardiology Foundation; American Heart Association. Implications of recent clinical trials for the National Cholesterol Education Program Adult Treatment Panel III guidelines. Circulation 2004;110:227–239.
MAY 1, 2005
PLA2 values were 62% higher (p <0.0001) in the fed state in patients compared with controls. Atorvastatin decreased median hs-CRP levels by 32% (p <0.01) and mean Lp-PLA2 values by 26% in patients (p <0.0001), with similar decreases in the fed state, and none of the other statins had any significant effect on these parameters. Change in Lp-PLA2 was significantly related to change in low-density lipoprotein cholesterol (p <0.01), with no significant relations with change in hs-CRP. Our data indicate greater differences in patients with coronary heart disease compared with controls in Lp-PLA2 in the fed state than in the fasting state and that atorvastatin is more effective than fluvastatin, lovastatin, pravastatin, or simvastatin for decreasing not only lowdensity lipoprotein cholesterol but also hs-CRP and Lp-PLA2. 䊚2005 by Excerpta Medica Inc. (Am J Cardiol 2005;95:1025–1032)
he focus of this study was the examination of the effects of atorvastatin versus those of other statins T (fluvastatin, lovastatin, pravastatin, and simvastatin)
of Lp-PLA2 and CRP are markers of inflammation and increased CHD risk. Lp-PLA2 is produced by macrophages, whereas CRP is primarily synthesized in the liver. We previously documented that patients who have CHD with low-density lipoprotein cholesterol levels ⬎130 mg/dl have high plasma levels of triglycerides and remnant lipoprotein cholesterol and triglyceride in the fasting and fed states compared with controls and that a fat challenge is not necessary to detect these abnormalities.1 In these studies, we also documented that fasting was not essential to detect abnormalities in direct low-density lipoprotein cholesterol, non– high-density lipoprotein cholesterol, and high-density lipoprotein cholesterol.1 In addition, we reported that atorvastatin therapy at 40 mg/day normalizes levels of all classes of triglyceride-rich lipoproteins and low-density lipoproteins (large, medium, and small low-density lipoproteins) in the fasting and fed states in patients who have CHD compared with control subjects2 and that atorvastatin is more effective than other tested statins in this regard.3
on high-sensitivity C-reactive protein (hs-CRP), lipoprotein-associated phospholipase A2 (Lp-PLA2), and plasma lipoprotein subspecies in the fasting and postprandial states (4 hours after a meal rich in saturated fat and cholesterol) in patients who had coronary heart disease (CHD) and a comparison of their parameters on and off therapy with those of age- and gendermatched normal control subjects. High plasma levels From the Cardiovascular Research and Lipid Metabolism Laboratories, Tufts University School of Medicine, Friedman School of Nutrition Science and Policy at Tufts University, and Jean Mayer USDA Human Nutrition Research Center on Aging at Tufts University, Boston, Massachusetts; and the Cardiology Division, Newton-Wellesley Hospital, Newton, Massachusetts. This study was supported by investigator-initiated research contracts from Parke-Davis/Pfizer, New York, New York; diaDexus, Inc., South San Francisco, California; and Otsuka America Pharmaceuticals, Inc., Rockville, Maryland. Manuscript received September 22, 2004; revised manuscript received and accepted January 3, 2005. Address for reprints: Ernst J. Schaefer, MD, Tufts University, 711 Washington Street, Boston, Massachusetts 02111. E-mail: ernst. [email protected]. ©2005 by Excerpta Medica Inc. All rights reserved. The American Journal of Cardiology Vol. 95 May 1, 2005
METHODS
Study subjects: The 84 patients in this study who had CHD were recruited from the clinic or by adver0002-9149/05/$–see front matter doi:10.1016/j.amjcard.2005.01.023
1025
tising and were required to have evidence of established CHD (coronary artery bypass grafting, angioplasty, documented myocardial infarction, significant coronary artery stenosis ⬎50% as assessed by angiography, or significantly decreased cardiac perfusion based on cardiac imaging with and without exercise). All subjects were studied after providing informed consent according to protocols approved by the human investigation review committee at the New England Medical Center (Boston, Massachusetts) and NewtonWellesley Hospital (Newton, Massachusetts). Patients were excluded if they had evidence of renal impairment, hypothyroidism, or liver dysfunction based on clinical chemistry testing or had previous adverse reactions to statins. In addition, patients were required to have a low-density lipoprotein cholesterol serum concentration ⬎130 mg/dl while off lipid-lowering medication for ⱖ6 weeks, including anion exchange resins, statins, fibric acid derivatives, fish oil, or niacin-containing products. Patients were maintained on other medications throughout the study, with no change, including calcium channel blockers,  blockers, and diuretics. Patients had previously received instruction on a National Cholesterol Education Program diet that contained ⬍30% of calories as fat, ⬍7% of calories as saturated fat, and ⬍200 mg/day of cholesterol. Patients had a lead-in baseline period of 4 weeks on this diet and did not use lipid-lowering medication. At the end of the 4-week lead-in period, patients were sampled after a 12-hour overnight fast and then randomized to receive atorvastatin or 1 of 4 other statins (fluvastatin, n ⫽ 26; lovastatin, n ⫽ 20; pravastatin, n ⫽ 17; or simvastatin, n ⫽ 16) for a protocol in which they were sampled in the fasting state after 4 weeks on 20 mg/day, after another 4 weeks on 40 mg/ day, after another 4 weeks on 80 mg/day or the maximal dose (40 mg for pravastatin and simvastatin), and then at 4 and 8 weeks during an 8-week placebo period. Thereafter, patients went on a second 12-week statin period in which they received 20 mg/day for 4 weeks, 40 mg/day for 4 weeks, and 80 mg/day or the maximal dose for 4 weeks with the sampling schedule. If they had not received atorvastatin in the first statin period, they received it during the second period. If they had received atorvastatin during the first statin period, they were randomized to 1 of the other 4 statins during the second statin period of 12 weeks. At the end of the 4-week lead-in period, at the end of the 40 mg/day dose of each statin, and at the end of 8-week period on placebo, most subjects (n ⫽ 61) were also sampled 4 hours after a fat-rich meal (obtained from a McDonald’s restaurant; 2 eggs, 2 sausages, and 2 muffins; 880 total calories, 500 calories (57%) from fat, 180 calories (20%) from saturated fat, and 510 mg of cholesterol). We also recruited an equal number of age- and gender-matched control subjects (n ⫽ 84) using the same exclusion criteria as for patients with CHD, except that they were also required to be clinically free of CHD by history, and there were no exclusion criteria concerning low-density lipoprotein choles1026 THE AMERICAN JOURNAL OF CARDIOLOGY姞
VOL. 95
terol. These subjects were studied only 1 time in the fasting state and 4 hours after the same fat-rich meal. Data on lipid results in the fasting and postprandial states at the end of the baseline period in patients who had CHD are not provided because they were not significantly different from values at the end of the 8-week placebo period. Lipoprotein analyses: Blood was drawn from each subject after a 12-hour overnight fast into tubes that contained ethylenediaminetetraacetic acid (final concentration 0.15%). Plasma was separated by centrifugation (2,500 rpm at 4°C for 20 minutes). Plasma concentrations of total cholesterol, triglycerides, and high-density lipoprotein cholesterol were measured fresh according to standard enzymatic methods and reagents obtained from Abbott Diagnostics (Irving, Texas) as described previously.4 Low-density lipoprotein cholesterol was measured directly with kits from Equal Diagnostics (Exton, Pennsylvania). Apolipoprotein-AI and apolipoprotein-B were measured by immunoturbidimetric assays from Wako Chemicals (Richmond, Virginia), and lipoprotein(a) was measured by mass with Macra kits5 (Wampole Laboratories, Cranbury, New Jersey) and by cholesterol content using kits from Genzyme Diagnostics (Cambridge, Massachusetts).6 Data on lipoprotein(a) mass are not reported because none of the statins induced statistically significant effects on this parameter compared with placebo or each other. Among-run precision studies indicated coefficients of variation of ⬍2% for total cholesterol, ⬍3% for triglycerides, ⬍4% for lowdensity lipoprotein cholesterol, ⬍5% for high-density lipoprotein cholesterol, apolipoprotein-AI, and lipoprotein(a), and ⬍10% for apolipoprotein-B and lipoprotein(a) cholesterol. Lipid assays are standardized through the Centers for Disease Control Lipid Standardization Program (Atlanta, Georgia). Remnant lipoprotein analyses: Isolation of remnant lipoprotein was based on removal of apolipoproteinAI– containing particles (high-density lipoprotein) and most apolipoprotein-B– containing particles (low-density lipoprotein, nascent very low-density lipoprotein, and nascent chylomicrons) using a previously described immunoseparation technique (JIMRO, Japan Immunoresearch Laboratories, Takasaki, Japan) that has been shown to leave particles characteristic of very low-density lipoprotein remnants and chylomicron remnants in the unbound fraction.7 Briefly, monoclonal antibodies to apolipoprotein-AI and specific monoclonal antibodies to apolipoprotein-B (JI-H), which do not recognize partially hydrolyzed, apolipoprotein-E– enriched lipoprotein remnants, were immobilized on agarose gel. Remnant lipoprotein cholesterol and remnant lipoprotein triglyceride concentrations were measured in plasma aliquots and stored at ⫺80°C until analysis. Thawed plasma was incubated with the gel for 2 hours, after which the gel, which contained the bound (nonremnant lipoprotein) lipoproteins, was precipitated with low-speed centrifugation (5 minutes at 135g). Remnant lipoprotein cholesterol and remnant lipoprotein triglyceride levels were then measured in the unbound supernates on an Abbott Spectrum CCx chemistry analyzer (Abbott DiMAY 1, 2005
TABLE 1 Fasting Concentrations for Controls and Patients on Placebo Measurement hs-CRP* (mg/L) Lp-PLA2 (ng/ml) Total cholesterol (mg/dl) Triglycerides* (mg/dl) LDL cholesterol (mg/dl) HDL cholesterol (mg/dl) Non–HDL cholesterol (mg/dl) Remnant lipoprotein cholesterol* (mg/dl) Remnant lipoprotein triglycerides* (mg/dl) Total cholesterol/HDL cholesterol ratio Nuclear magnetic resonance of small LDL (mg/dl) Nuclear magnetic resonance of LDL particle number Nuclear magnetic resonance of large HDL (mg/dl) hs-CRP ⬍3.0 mg/L hs-CRP ⬍2.0 mg/L LDL cholesterol ⬍100 mg/dl LDL cholesterol ⬍70 mg/dl Non–HDL cholesterol ⬍130 mg/dl Triglycerides ⬍150 mg/dl
Controls (n ⫽ 84)
Patients (n ⫽ 84)
3.3 ⫾ 6.2 230 ⫾ 81 179 ⫾ 37 92 ⫾ 50 98 ⫾ 39 52 ⫾ 14 126 ⫾ 37 6.3 ⫾ 2.0 18.5 ⫾ 14.6 3.6 ⫾ 1.2 32 ⫾ 28 1,266 ⫾ 334 30 ⫾ 13 76% 67% 66% 27% 66% 88%
3.6 ⫾ 5.5 260 ⫾ 91 277 ⫾ 68 204 ⫾ 85 184 ⫾ 56 37 ⫾ 10 240 ⫾ 69 11.3 ⫾ 11.2 42.4 ⫾ 25.8 7.9 ⫾ 2.9 96 ⫾ 66 2,455 ⫾ 711 14 ⫾ 9 70% 53% 0% 0% 0% 31%
Differences (%) 1 29 59 147 112 ⫺24 104 57 151 123 121 106 ⫺58
(⫺46 to 341)† ⫾ 69† ⫾ 45# (59 to 250) ⫾ 99# ⫾ 28# ⫾ 83# (0 to 22)# (20 to 360)# (61 to 208)# (⫺8 to 291)# ⫾ 98# (⫺76 to ⫺4)#
Normally distributed values are reported as mean ⫾ SD; other values are reported as medians (interquartile ranges). All p values were derived from 2-tailed, paired t-test analysis. *Values for hs-CRP, triglycerides, remnant lipoprotein cholesterol, and remnant lipoprotein triglycerides were log-transformed for statistical analysis.†p ⬍0.05; ‡p ⬍0.01; §p ⬍0.001;#p ⬍0.0001 for comparisons with placebo; the dataset comprised 12 women and 72 men; mean percentage ⫾ SD or median percent change from placebo. HDL ⫽ high-density lipoprotein; LDL ⫽ low-density lipoprotein.
TABLE 2 Fasting Concentrations on Placebo and Atorvastatin for All Subjects Measurement hs-CRP* (mg/L) Lp-PLA2 (ng/ml) Creatine phosphokinase* (U/L) Aspartate aminotransferase* (U/L) Arginine aminotransferase* (U/L) Total cholesterol (mg/dl) Triglycerides* (mg/dl) LDL cholesterol (mg/dl) HDL cholesterol (mg/dl) Non–HDL cholesterol (mg/dl) Apolipoprotein-AI (mg/dl) Apolipoprotein-B (mg/dl) Lipoprotein(a) cholesterol* (mg/dl) Lipoprotein (a) mass* (mg/dl) Remnant lipoprotein cholesterol* (mg/dl) Remnant lipoprotein triglycerides* (mg/dl) Total cholesterol/HDL cholesterol ratio Nuclear magnetic resonance of small LDL (mg/dl) Nuclear magnetic resonance LDL particle number Nuclear magnetic resonance of large HDL (mg/dl) hs-CRP ⬍3.0 mg/L hs-CRP ⬍2.0 mg/L LDL cholesterol ⬍100 mg/dl LDL cholesterol ⬍70 mg/dl Non–HDL cholesterol ⬍130 mg/dl Triglycerides ⬍150 mg/dl
Placebo (n ⫽ 84)
Atorvastatin (n ⫽ 84)
3.5 ⫾ 5.4 260 ⫾ 91 160 ⫾ 141 24 ⫾ 6 22 ⫾ 8 277 ⫾ 68 204 ⫾ 85 184 ⫾ 56 37 ⫾ 10 240 ⫾ 69 112 ⫾ 22 180 ⫾ 42 14.8 ⫾ 13.6 23.9 ⫾ 21.3 11.3 ⫾ 11.2 42.4 ⫾ 25.8 7.9 ⫾ 2.9 96 ⫾ 66 2,455 ⫾ 711 14 ⫾ 9 70% 53% 0% 0% 0% 31%
2.5 ⫾ 3.4 190 ⫾ 49 186 ⫾ 246 27 ⫾ 8 28 ⫾ 10 168 ⫾ 41 137 ⫾ 70 96 ⫾ 39 40 ⫾ 11 128 ⫾ 41 115 ⫾ 24 112 ⫾ 33 14.3 ⫾ 15.1 25.8 ⫾ 23.4 6.6 ⫾ 3.9 25.9 ⫾ 19.6 4.4 ⫾ 1.4 53 ⫾ 41 1,382 ⫾ 482 20 ⫾ 9 81% 65% 69% 24% 63% 67%
Differences (%) ⫺34 ⫺26 8 9 19 ⫺40 ⫺35 ⫺48 9 ⫺49 4 ⫺36 ⫺16 0 ⫺34 ⫺42 ⫺41 ⫺55 ⫺41 55
(⫺50 to ⫺5)‡ ⫾ 26# (⫺13 to 29) (0 to 25)# (8 to 41)# ⫾ 13# (⫺53 to ⫺12)# ⫾ 16# ⫾ 17† ⫾ 20%# ⫾ 18%† ⫾ 18# (⫺68 to 101)# (⫺7 to 21) (⫺57 to ⫺10)# (⫺63 to ⫺5)# (⫺52 to ⫺37)# (⫺73 to ⫺22)# ⫾ 21# (14 to 103)#
Normally distributed values are reported as mean ⫾ SD; other values are reported as medians (interquartile ranges). All p values were derived from 2-tailed, paired t-test analysis. *Values for hs-CRP, creatine phosphokinase, aspartate aminotransferase, arginine aminotransferase, triglycerides, lipoprotein(a) cholesterol, lipoprotein(a) mass, remnant lipoprotein cholesterol, and remnant lipoprotein triglycerides were log-transformed for statistical analysis. † p ⬍0.05; ‡p ⬍0.01; §p ⬍0.001; #p ⬍0.0001 for comparisons with placebo; the dataset comprised 12 women and 72 men; mean percentage ⫾ SD or median percent change from placebo. Abbreviations as in Table 1.
CORONARY ARTERY DISEASE/DIRECT COMPARISONS OF STATIN EFFECTS ON CRP AND LP-PLA2
1027
agnostics) that used 2-reagent enzymatic, colorimetric assays that contained a sensitive chromophore (Kyowa Medex, Tokyo, Japan). Precision studies yielded among-run remnant lipoprotein cholesterol imprecision for 2 levels of remnant lipoprotein control over 20 runs, i.e., 9.1% at 7 mg/dl and 7.3% at 24 mg/dl. Among-run remnant lipoprotein triglyceride imprecision levels for the same controls were 8.3% at 22 mg/dl and 5.0% at 109 mg/dl.7 Nuclear magnetic resonance analysis: Lipoprotein subclass distributions were determined by proton nuclear magnetic resonance spectroscopy.8 Subclasses were separated as chylomicrons (⬎200 nm); large (60 to 200 nm), medium (35 to 60 nm), and small (27 to 35 nm) very low-density lipoproteins; intermediatedensity lipoproteins (23 to 27 nm); large (21 to 23 nm), medium (20 to 21 nm), and small (18 to 20 nm) low-density lipoproteins; and large (9 to 13 nm), medium (8 to 9 nm), and small (7 to 8 nm) high-density lipoproteins. Very low-density lipoprotein subclass concentrations are expressed as milligrams per deciliter of triglycerides, and other lipoprotein subclass concentrations are expressed as milligrams per deciliter of cholesterol. Data on the chylomicron and intermediate-density lipoprotein fractions were omitted because the levels of lipids within the fractions were low, with no statistically significant differences across groups. Clinical chemistry analyses: Standardized serum measurements of liver transaminases and creatine phosphokinase were carried out at each 4-week time point in the study. Measurements of hs-CRP were performed on a Hitachi 911 (Roche Diagnostics, Indianapolis, Indiana) using the CRP-UL kit from Wako Chemicals. Within- and between-run coefficients of variation were ⬍5%. Measurements of LpPLA2 mass were performed by an enzyme-linked immunosorbent assay (diaDexus, Inc., South San Francisco, California) at the diaDexus facility in a blinded fashion. Within- and between-run coefficients of variation were ⬍7.5%. Statistical analysis: Statistical analyses were performed with patients who had CHD by comparing the effects of each statin at 40 mg/day with placebo, and its effects were compared with those observed on the same dose of atorvastatin in the same patients. The effects of statins at 40 mg/day in the fasting and fed states were compared with placebo and with values obtained on the same dose of atorvastatin. Values in control subjects who did not have CHD are provided for comparison. Student’s t test was used to compare mean levels of continuous measurements, and a chi-square statistic was calculated for categorical factors. For continuous measurements that were highly skewed, we performed comparisons using log-transformed values, although we report the untransformed means and SDs for each group. Analyses were also conducted separately by gender, but we report only pooled results because men and women had similar responses to statins. Similarly, we do not report baseline lipid data because these results were not statistically different from values obtained on placebo. For the postprandial studies, 1028 THE AMERICAN JOURNAL OF CARDIOLOGY姞
VOL. 95
statistical analyses were done to compare fasting values with postprandial values, values on medication with those on placebo, and values in patients who had CHD and used placebo and at 40 mg/day of statins with values on atorvastatin. We also carried out correlations for change in hs-CRP and Lp-PLA2 versus change in other parameters influenced by atorvastatin. One outlier that had a 8,531% increase in hs-CRP on atorvastatin was omitted from the correlation analysis.
RESULTS Baseline demographic information for subjects who had CHD (n ⫽ 84, 87% men and 13% women) and for age- and gender-matched controls (n ⫽ 84) were mean age (62 ⫾ 9 years), diabetes mellitus (0% of controls and 5% of patients), hypertension (31% of controls and 56% of patients), smoking (20% of controls and 18% of patients), and body mass index (26 ⫾ 4 kg/m2 for controls and 28 ⫾ 4 kg/m2 for patients). Mean values ⫾ SD for control subjects and subjects who had CHD and were using placebo with respect to fasting plasma levels of hs-CRP, Lp-PLA2, total cholesterol, triglycerides, lowdensity lipoprotein cholesterol, high-density lipoprotein cholesterol, non– high-density lipoprotein cholesterol, apolipoprotein-AI, apolipoprotein-B, and lipoprotein(a) or lipoprotein(a) cholesterol levels and to levels of remnant lipoprotein cholesterol and triglyceride, small lowdensity lipoprotein, low-density lipoprotein particle number, and large high-density lipoprotein are presented in Table 1. In addition, values for the ratio of total cholesterol to high-density lipoprotein cholesterol and percentage of patients who had hs-CRP levels ⬍3.0 and ⬍2.0 mg/L, low-density lipoprotein cholesterol levels ⬍100 and ⬍70 mg/dl, non– high-density lipoprotein cholesterol levels ⬍130 mg/dl, and triglyceride levels ⬍150 mg/dl are presented. Patients had significantly (p ⬍0.05) higher levels of hs-CRP, Lp-PLA2, total cholesterol, low-density lipoprotein cholesterol, triglycerides, remnant lipoprotein cholesterol, remnant lipoprotein triglycerides, total cholesterol/high-density lipoprotein cholesterol ratio, small low-density lipoprotein, and total low-density lipoprotein particle number and significantly (p ⬍0.05) lower levels of high-density lipoprotein cholesterol and large highdensity lipoprotein than did controls (Table 1). Median hs-CRP and mean Lp-PLA2 values were 51% and 29% higher in patients than in controls, respectively (p ⬍0.05). The effects of atorvastatin at 40 mg/day versus placebo on various parameters in 84 patients who had CHD are presented in Table 2. Median decreases in hs-CRP were 32% (p ⬍0.01) and mean decreases in Lp-PLA2 were 26% (p ⬍0.0001). There was a high degree of variability in the hs-CRP response to atorvastatin but somewhat less variability in the Lp-PLA2 response (Figure 1). For these parameters, there were increases in some patients (5 patients developed dramatic increases in hs-CRP and were excluded from the data shown in Figure 1 but not from the data analyses), but, in general, atorvastatin decreased these values substantially, with almost normal values for low-density lipoprotein cholesterol and triglycerides (Table 2). These increases may have been due to concurrent infections. Atorvastatin also MAY 1, 2005
which was moderately significant, and nonfasting hs-CRP level, which, although decreased by 29%, was not significant. Of note, for all these parameters, there were highly significant differences (p ⬍0.0001) between patients on placebo and controls except for Lp-PLA2. However, differences were very significant in the fed state because Lp-PLA2 was decreased by 21% in the fed state in controls but did not change significantly in patients. Moreover, although atorvastatin came close to normalizing Lp-PLA2, lowdensity lipoprotein cholesterol, and remnant lipoprotein cholesterol in patients versus controls, it did not do so for hs-CRP (especially in the fed state) or triglycerides. Table 4 presents a comparison of the effects of the different statins on fasting and postprandial levels of hsCRP and Lp-PLA2 and of low-density lipoprotein cholesterol. Atorvastatin was not only significantly more efficacious than other statins in decreasing low-density lipoprotein cholesterol but also the only statin that significantly decreased levels of hs-CRP and Lp-PLA2 in the fasting or fed state. Because of these effects, we reFIGURE 1. Variability (percent change from values on placebo) in response to atorvalated changes in fasting levels of hsstatin (40 mg/day) for hs-CRP, Lp-PLA2, creatine phosphokinase (CPK), and aspartate CRP and Lp-PLA2 induced by atoraminotransferase (AST) are shown. For hs-CRP, 5 extreme outliers have been omitted vastatin to other variables based on from the figure (ⴙ8,531%, ⴙ495%, ⴙ316%, ⴙ295%, and ⴙ242%); for CPK, 2 outlicorrelation coefficient analysis. For ers have been omitted from the figure (ⴙ592% and ⴙ499%). Lp-PLA2, the change was significantly correlated (p ⬍0.01) with changes in total cholesterol (r ⫽ resulted in modest median increases in creatine phos- 0.49), low-density lipoprotein cholesterol (r ⫽ 0.43), phokinase, aspartate aminotransferase, and alanine ami- and ratio of total cholesterol to high-density lipopronotransferase of 8%, 9%, and 19%, respectively, with tein cholesterol (r ⫽ 0.44). For hs-CRP, the change only the latter 2 changes being significant (p ⬍0.0001). was not significantly correlated with change in any The variability in response to atorvastatin by creatine parameter. phosphokinase and alanine aminotransferase is shown in Figure 1. As we previously reported, atorvastatin is very DISCUSSION We previously documented that fasting values of effective in decreasing levels of total cholesterol, lowdensity lipoprotein cholesterol, non– high-density li- low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, triglycerides, remnant lipoprotein poprotein cholesterol, triglycerides, and remnant licholesterol, remnant lipoprotein triglycerides, low-den- poprotein cholesterol are correlated with postprandial sity lipoprotein particle number, and small low-density values and that differences between patients who have lipoprotein and in increasing levels of large high-density CHD and controls with respect to these parameters lipoprotein as assessed by nuclear magnetic resonance can readily be detected in the fasting state.1 Therefore, a fat challenge is not necessary. Moreover, we found spectroscopy.2,3 Data that examined the effects of atorvastatin on that low-density lipoprotein cholesterol and high-denfasting and fed levels of hs-CRP, Lp-PLA2, low-density sity lipoprotein cholesterol can readily be measured in lipoprotein cholesterol, triglyceride, and remnant li- the fed state, provided that newer direct methods are poprotein cholesterol in 61 patients who had CHD are used and that these values are just slightly lower in the listed in Table 3. Decreases induced by atorvastatin for fed state than in the fasting state.1 In addition, we documented that atorvastatin is all these parameters were very similar in the fasting and fed states, and all decreases were highly significant very effective in normalizing many of the lipid abnor(p ⬍0.0001), except for fasting hs-CRP level (p ⬍0.05), malities found in patients who have CHD compared CORONARY ARTERY DISEASE/DIRECT COMPARISONS OF STATIN EFFECTS ON CRP AND LP-PLA2
1029
1030 THE AMERICAN JOURNAL OF CARDIOLOGY姞
TABLE 3 Fasting and Postprandial Concentrations on Placebo and 40 mg/day Atorvastatin Versus Matched Controls Patients With CHD Measurement
VOL. 95 MAY 1, 2005
Fasting hs-CRP* (mg/L) 4-h postprandial hs-CRP* (mg/L) Median postprandial change Fasting LP-PLA2 (ng/ml) 4-h postprandial Lp-PLA2 (ng/ml) Mean postprandial change Fasting Triglycerides* (mg/dl) 4-h Postprandial triglycerides* (mg/dl) Median postprandial change Fasting LDL cholesterol (mg/dl) 4-h postprandial LDL cholesterol (mg/dl) Mean postprandial change Fasting remnant lipoprotein cholesterol* (mg/dl) 4-h postprandial remnant lipoprotein cholesterol* (mg/dl) Median postprandial change
Controls (n ⫽ 61) 2.2 2.7 ⫹22% 230 177 ⫺21 95 151 ⫹53% 95 89 ⫺6 6.2 9.6
⫾ 4.5 ⫾ 5.7‡ (8 to 41) ⫾ 81 ⫾ 45储 ⫾ 13% ⫾ 52 ⫾ 84储 (34 to 83) ⫾ 34 ⫾ 32储 ⫾ 9% ⫾ 1.8 ⫾ 4.5储
⫹31% (13 to 50)
Placebo (n ⫽ 61) 2.6 3.1 ⫹2% 254 266 ⫹9 196 348 ⫹81% 164 155 ⫺5 11.8 19.5
⫾ 3.2 ⫾ 3.8 (⫺8 to 14) ⫾ 90 ⫾ 84 ⫾ 25% ⫾ 96 ⫾ 147储 (48 to 113) ⫾ 43 ⫾ 41储 ⫾ 8% ⫾ 13.9 ⫾ 15.1储
⫹42% (34 to 52)
Percent Differences Atorvastatin (n ⫽ 61) 2.3 3.0 ⫹2% 185 193 ⫹6 134 231 ⫹64% 91 84 ⫺7 6.5 11.6
⫾ 3.7 ⫾ 4.1 (⫺10 to 32) ⫾ 48 ⫾ 61 ⫾ 26% ⫾ 60 ⫾ 124储 (42 to 106) ⫾ 33 ⫾ 30储 ⫾ 10% ⫾ 4.0 ⫾ 7.1储
Controls/Placebo
Controls/Atorvastatin
Placebo/Atorvastatin
⫹68% (⫺39 to 504) ⫹14% (⫺36 to 527)
⫹20% (⫺61 to 358) ⫹53% (⫺34 to 352)
⫺34% (⫺54 to ⫺7)¶ ⫺29% (⫺51 to 5)
⫹23% (⫺20 to 44) ⫹62% (17 to 89)††
⫺10% (⫺36 to ⫺2)** ⫹17% (⫺17 to 37)
⫺23% (⫺36 to ⫺13)†† ⫺25% (⫺41 to ⫺13)††
⫹128% (48 to 207)†† ⫹141% (55 to 291)††
⫹48% (5 to 121)†† ⫹46% (6 to 162)††
⫺36% (⫺46 to ⫺14)†† ⫺39% (⫺50 to ⫺13)††
⫹94% (37 to 143)†† ⫹95% (31 to 134)††
⫹6% (⫺33 to 23) ⫹5% (⫺37 to 24)
⫺43% (⫺57 to ⫺36)†† ⫺44% (⫺59 to ⫺39)††
⫺4% (⫺36 to 30) ⫹10% (⫺25 to 100)
⫺40% (⫺50 to ⫺23)†† ⫺41% (⫺52 to ⫺24)††
¶
⫹55% (10 to 102)†† ⫹100% (30 to 171)††
⫹43% (29 to 54)
Normally distributed values are reported as mean ⫾ SD; other values are reported as medians (interquartile ranges). All p values were derived from 2-tailed, paired t-test analysis. *Values for hs-CRP, triglycerides, and remnant lipoprotein cholesterol were log-transformed for statistical analysis. † p ⬍0.05; ‡p ⬍0.01; §p ⬍0.001; 储p ⬍0.0001 for comparisons of fasting and postprandial values. ¶ p ⬍0.05; #p ⬍0.01; **p ⬍0.001; ††p ⬍0.0001 for paired t-test comparison among treatment groups; the dataset comprised 9 women and 52 men; controls were matched by age and gender to patients mean percentage ⫾ SD or median percent change from placebo. Abbreviation as in Table 1.
TABLE 4 Percentage Change in Fasting Lipoprotein-associated Phospholipase A2 Concentrations From Placebo After 40 mg/day Treament With Fluvastatin, Lovastatin, Pravastatin, or Simvastatin Versus Atorvastatin Parameter hs-CRP* (mg/L) LP-PLA2 (ng/ml) LDL cholesterol (mg/dl) Parameter hs-CRP* (mg/L) LP-PLA2 (ng/ml) LDL cholesterol (mg/dl) Parameter hs-CRP* (mg/L) LP-PLA2 (ng/ml) LDL cholesterol (mg/dl) Parameter hs-CRP* (mg/L) LP-PLA2 (ng/ml) LDL cholesterol (mg/dl)
Controls
Placebo
Fluvastatin
Atorvastatin
3.3 ⫾ 7.6 240 ⫾ 86 92 ⫾ 32
5.5 ⫾ 8.5 272 ⫾ 106 161 ⫾ 39¶¶
⫹2 (⫺27 to 37) ⫺1 ⫾ 19% ⫺16 ⫾ 16%储
⫺34 (⫺69 to 18)‡‡ ⫺24 ⫾ 17%§†† ⫺45 ⫾ 13%储††
Controls
Placebo
Lovastatin
Atorvastatin
2.9 ⫾ 5.0 232 ⫾ 78 100 ⫾ 31
3.0 ⫾ 0.3 233 ⫾ 69 157 ⫾ 28¶¶
Controls
Placebo
Pravastatin
4.6 ⫾ 6.2 246 ⫾ 91 110 ⫾ 43
3.1 ⫾ 4.0 264 ⫾ 80 193 ⫾ 53¶¶
⫹3 (⫺26 to 25) ⫺6 ⫾ 22% ⫺24 ⫾ 13%储
Controls
Placebo
Simvastatin
Atorvastatin
3.1 ⫾ 6.3 219 ⫾ 60 90 ⫾ 30
1.8 ⫾ 0.9 247 ⫾ 90 200 ⫾ 78¶¶
⫹26 (⫺26 to 48) ⫺10 ⫾ 26% ⫺37 ⫾ 16%储
⫺36 (⫺57 to ⫺20)‡ ⫺21 ⫾ 16%‡ ⫺45 ⫾ 15%储¶
1
⫹16 (⫺7 to 41) ⫺6 ⫾ 24% ⫺28 ⫾ 18%储
⫺32 (⫺54 to ⫺19)† ⫺16 ⫾ 26%‡¶ ⫺45 ⫾ 14%储†† Atorvastatin ⫺31 (⫺39 to ⫺7)§ ⫺26 ⫾ 17%储# ⫺45 ⫾ 14%储††
Normally distributed values are reported as mean ⫾ SDs; other values are reported as median and interquartile range. *Values for hs-CRP were log transformed before statistical analysis. All p values were derived from unpaired and paired 2-tailed t-test analyses. †p ⬍0.05; ‡p ⬍0.01; §p ⬍0.001; 储p ⬍0.0001 for paired comparisons of treatment with placebo; ¶p ⬍0.05; #p ⬍0.01; **p ⬍0.001; ††p ⬍0.0001 for paired comparisons between same doses of statins. ‡‡p ⬍0.05; §§p ⬍0.01; 储储p ⬍0.001; ¶¶p ⬍0.0001 for unpaired comparisons of controls with placebo; the data sets comprised 1 woman, 25 men (F), 2 women, 18 men (L), 6 women, 11 men (P), 2 women, 14 men (S); mean ⫾ SD% or median percent change from placebo. Abbreviation as in Table 1.
with controls, including increased low-density lipoprotein and non– high-density lipoprotein cholesterol, cholesterol in small dense low-density lipoprotein, total low-density lipoprotein particle number, triglycerides, triglycerides in different very low-density lipoprotein classes, postprandial triglycerides, and fasting and postprandial remnant lipoprotein cholesterol and triglycerides.2 Atorvastatin was found to decrease lipoprotein(a) cholesterol but not total lipoprotein(a) mass.2 Moreover, we noted that, although atorvastatin had a modest effect in increasing highdensity lipoprotein cholesterol, it significantly increased, but did not normalize, cholesterol in large high-density lipoprotein particles.2 In addition, we documented a wide range of patient variability in response to statins with regard to lipid modification.2,9 For this reason, statin comparisons should be done in the same subjects to remove genetic confounders in studying the efficacy of these statins. Other investigators have examined the effects of statins on lipoproteins, including remnants, in the fasting and postprandial states, but there have been no direct comparisons of all of these statins in the same subjects with respect to hs-CRP and Lp-PLA2 in the fasting and fed states. In addition, we documented that atorvastatin is more effective than other statins tested (fluvastatin, lovastatin, pravastatin, and simvastatin) in normalizing lipid abnormalities in patients who have CHD.3 The focus of this examination has been on hs-CRP and Lp-PLA2 as measured in the fasted and fed states in patients who have CHD and age- and gendermatched controls and examining the effects of atorvastatin and other statins on these parameters in patients
who have CHD. Other investigators have documented that hs-CRP and Lp-PLA2 are important inflammatory markers and markers of CHD risk.10 –13 Moreover, it has been documented that Lp-PLA2 is carried on lipoproteins, especially low-density lipoprotein.14,15 It has also been reported that statins can decrease the levels of these parameters.16 –18 A novel feature of our study is the finding that feeding a meal rich in saturated fat and cholesterol caused significant increases of 22% in hs-CRP and significant decreases of 21% in Lp-PLA2 in controls but not in patients. Therefore, differences in hs-CRP are somewhat less in patients than in controls in the fed state versus the fasting state (median difference 114% vs 168%), but differences were greater for LpPLA2 (mean difference 62% vs 23%). These findings need to be confirmed in larger studies. We have also documented that atorvastatin results in similar decreases in hs-CRP (34% and 29%) and Lp-PLA2 (23% and 25%) in patients who have CHD in the fasting and fed states. We have also documented that atorvastatin is much more effective than other statins tested (fluvastatin, lovastatin, pravastatin, and simvastatin) in decreasing not only lipid parameters in the fasting and fed states but also hs-CRP and Lp-PLA2. Other statins did not significantly decrease hs-CRP levels in our study in contrast to some other studies.19 These differences may be related to differences in sample size, duration of therapy, and baseline characteristics of the study subjects. A 17% decrease in Lp-PLA2 levels was reported with pravastatin in the West of Scotland Study, which was greater than what we observed.15 A recent intravascular ultrasound study that
CORONARY ARTERY DISEASE/DIRECT COMPARISONS OF STATIN EFFECTS ON CRP AND LP-PLA2
1031
involved ⬃500 evaluable patients who took atorvastatin (80 mg/day) for 18 months reported that hs-CRP decreased by 36% versus baseline, whereas pravastatin decreased this parameter by only 5%, which is consistent with our data.20 We tested for correlations of atorvastatin-induced changes in these parameters, with changes in lipid and other parameters. Changes in Lp-PLA2 were correlated with changes in total cholesterol, low-density lipoprotein cholesterol, and the total cholesterol/highdensity lipoprotein cholesterol ratio, consistent with Lp-PLA2 being carried on low-density lipoprotein, and these particles were substantially decreased by atorvastatin. For changes in hs-CRP, no significant correlations with atorvastatin-induced lipid changes were noted. In our data, hs-CRP increased with fat feeding as triglycerides and remnants increased in controls but not in patients. The mechanisms responsible for these interactions require further elucidation.21–25 The focus of current guidelines for lipid modification in patients who have CHD is for drastic decreases in low-density lipoprotein cholesterol.26 In our view, in the future, the emphasis may shift toward optimizing other lipoprotein particles, namely remnants, high-density lipoprotein, and lipoprotein(a), improving the ratio of total cholesterol to high-density lipoprotein cholesterol, and decreasing levels of inflammatory markers, such as hsCRP and LpPLA2, to decrease the risk for CHD. 1. Schaefer EJ, Audelin MC, McNamara JR, Shah PK, Tayler T, Daly JA, Augustin JL, Seman LJ, Rubenstein JJ. Comparison of fasting and postprandial plasma lipoproteins in subjects with and without coronary heart disease. Am J Cardiol 2001;88:1129 –1133. 2. Schaefer EJ, McNamara JR, Tayler T, Daly JA, Gleason JA, Seman LJ, Ferrari A, Rubenstein JJ. Effects of atorvastatin on fasting and postprandial lipoprotein subclasses in coronary disease patients versus control subjects. Am J Cardiol 2002;90:689 – 696. 3. Schaefer EJ, McNamara JR, Tayler T, Daly JA, Gleason JA, Seman LJ, Ferrari A, Rubenstein JJ. Direct comparisons of statin effects on fasting and postprandial lipoproteins in coronary heart disease patients versus control subjects. Am J Cardiol 2004;93:31–39. 4. McNamara JR, Schaefer EJ. Automated enzymatic standardized lipid analyses for plasma and lipoprotein fractions. Clin Chim Acta 1987;166:1– 8. 5. Jenner JL, Ordovas JM, Lamon-Fava S, Schaefer MM, Wilson PW, Castelli WP, Schaefer EJ. Effects of age, sex, and menopausal status on plasma lipoprotein(a) levels. The Framingham Offspring Study. Circulation 1993;87:1135–1141. 6. Seman LJ, Jenner JL, McNamara JR, Schaefer EJ. Quantitation of plasma lipoprotein(a) by cholesterol assay of lectin-bound lipoprotein(a). Clin Chem 1994;40:400 – 403. 7. McNamara JR, Shah PK, Nakajima K, Cupples LA, Wilson PWF, Ordovas JM, Schaefer EJ. Remnant lipoprotein cholesterol and triglyceride: reference ranges from the Framingham Heart Study. Clin Chem 1998;44:1224 –1232. 8. Otvos JD, Jeyarajah EJ, Bennett DW, Krauss RM. Development of a proton nuclear magnetic resonance spectroscopic method for determining plasma lipoprotein concentrations and subspecies distributions from a single, rapid measurement. Clin Chem 1992;38:1632–1638.
1032 THE AMERICAN JOURNAL OF CARDIOLOGY姞
VOL. 95
9. Pedro-Botet J, Schaefer EJ, Bakker-Arkema RG, Black DM, Stein FM, Corella D, Ordovas JM. Apolipoprotein E genotype affects plasma lipid response to atorvastatin in a gender-specific manner. Atherosclerosis 2001;158:183–194. 10. Ridker PM, Rifai N, Rose L, Buring JE, Cook NR. Comparison of C-reactive protein and low-density lipoprotein cholesterol levels in the prediction of first cardiovascular events. N Engl J Med 2002;347:1557–1565. 11. Sattar N, Gaw A, Scherbakova O, Ford I, O’Reilly DS, Haffner SM, Isles C, Macfarlane PW, Packard CJ, Cobbe SM, Shepherd J. Metabolic syndrome with and without C-reactive protein as a predictor of coronary heart disease and diabetes in the West of Scotland Coronary Prevention Study. Circulation 2003; 108:414 – 419. 12. Packard CJ, O’Reilly DS, Caslake MJ, McMahon AD, Ford I, Cooney J, Macphee CH, Suckling KE, Krishna M, Wilkinson FE, et al. Lipoproteinassociated phospholipase A2 as an independent predictor of coronary heart disease. West of Scotland Coronary Prevention Study Group. N Engl J Med 2000;343:1148 –1155. 13. Ballantyne CM, Hoogeveen RC, Bang H, Coresh J, Folsom AR, Heiss G, Sharrett AR. Lipoprotein-associated phospholipase A2, high-sensitivity C-reactive protein, and risk for incident coronary heart disease in middle-aged men and women in the Atherosclerosis Risk in Communities (ARIC) study. Circulation 2004;109:837– 842. 14. MacPhee CH, Moores KE, Boyd HF, Dhanak D, Ife RJ, Leach CA, Leake DS, Milliner KJ, Patterson RA, Suckling KE, et al. Lipoprotein-associated phospholipase A2, platelet-activating factor acetylhydrolase, generates two bioactive products during the oxidation of low-density lipoprotein: use of a novel inhibitor. Biochem J 1999;338:479 – 487. 15. Caslake MJ, Packard CJ. Lipoprotein-associated phospholipase A2 (plateletactivating factor acetylhydrolase) and cardiovascular disease. Curr Opin Lipidol 2003;14:347–352. 16. Horne BD, Muhlestein JB, Carlquist JF, Bair TL, Madsen TE, Hart NI, Anderson JL, Intermountain Heart Collaborative (IHC) Study Group. Statin therapy interacts with cytomegalovirus seropositivity and high C-reactive protein in reducing mortality among patients with angiographically significant coronary disease. Circulation 2003;107:258 –263. 17. Marz W, Winkler K, Nauck M, Bohm BO, Winkelmann BR. Effects of statins on C-reactive protein and interleukin-6 (the Ludwigshafen Risk and Cardiovascular Health study). Am J Cardiol 2003;92:305–308. 18. Wang TD, Chen WJ, Lin JW, Cheng CC, Chen MF, Lee YT. Efficacy of fenofibrate and simvastatin on endothelial function and inflammatory markers in patients with combined hyperlipidemia: relations with baseline lipid profiles. Atherosclerosis 2003;170:315–323. 19. Jialal I, Stein D, Balis D, Grundy SM, Adams-Huet B, Devaraj S. Effect of hydroxymethyl glutaryl coenzyme a reductase inhibitor therapy on high sensitive C-reactive protein levels. Circulation 2001;103:1933–1935. 20. Nissen SE, Tuzcu EM, Schoenhagan P, Brown BG, Ganz P, Vogel RA, Crowe T, Howard G, Cooper CJ, Brodie B, et al. Effect of intensive compared with moderate lipid-lowering therapy on progression of coronary atherosclerosis: a randomized controlled trial. JAMA 2004;291:1071–1080. 21. Koenig W, Lowel H, Baumert J, Meisinger C. C-reactive protein modulates risk prediction based on the Framingham score: implications for future risk assessment: results from a large cohort study in southern Germany. Circulation 2004;109:1349 –1353. 22. Saito M, Ishimitsu T, Minami J, Ono H, Ohrui M, Matsuoka H. Relations of plasma high-sensitivity C-reactive protein to traditional cardiovascular risk factors. Atherosclerosis 2003;167:73–79. 23. Luc G, Bard JM, Juhan-Vague I, Ferrieres J, Evans A, Amouyel P, Arveiler D, Fruchart JC, Ducimetiere P, PRIME Study Group. C-reactive protein, interleukin-6, and fibrinogen as predictors of coronary heart disease: the PRIME Study. Arterioscler Thromb Vasc Biol 2003;23:1255–1261. 24. van der Meer IM, de Maat MP, Kiliaan AJ, van der Kuip DA, Hofman A, Witteman JC. The value of C-reactive protein in cardiovascular risk prediction: the Rotterdam Study. Arch Intern Med 2003;163:1323–1328. 25. St-Pierre AC, Bergeron J, Pirro M, Cantin B, Dagenais GR, Despres JP, Lamarche B. Quebec Cardiovascular Study. Effect of plasma C-reactive protein levels in modulating the risk of coronary heart disease associated with small, dense, low-density lipoproteins in men (The Quebec Cardiovascular Study). Am J Cardiol 2003;91:555–558. 26. Grundy SM, Cleeman JI, Merz CN, Brewer HB Jr, Clark LT, Hunninghake DB, Pasternak RC, Smith SC Jr, Stone NJ. National Heart, Lung, and Blood Institute; American College of Cardiology Foundation; American Heart Association. Implications of recent clinical trials for the National Cholesterol Education Program Adult Treatment Panel III guidelines. Circulation 2004;110:227–239.
MAY 1, 2005