Effect of statins on LDL particle size in patients with familial combined hyperlipidemia: a comparison between atorvastatin and pravastatin

Nutrition, Metabolism & Cardiovascular Diseases (2005) 15, 47e55

www.elsevier.com/locate/nmcd

Effect of statins on LDL particle size in patients with familial combined hyperlipidemia: a comparison between atorvastatin and pravastatin Cesare R. Sirtoria,*, Laura Calabresia, Livia Pisciottab, Luigi Cattinc, Paolo Pauciullod, Mario Montagnanie, Enzo Manzatof, Gabriele Bittolo Bong, Renato Fellinh a

Center E. Grossi Paoletti, Department of Pharmacological Sciences, University of Milano, Via Balzaretti 9, 20133 Milano, Italy b Department of Internal Medicine, University of Genova, Italy c Department of Clinical Sciences, University of Trieste, Italy d Department of Clinical and Experimental Medicine, University Federico II of Napoli, Italy e Policlinico le Scotte, Siena, Italy f Department of Medical Sciences, Policlinico, Padova, Italy g Division of Medicine, Ospedali Civili Riuniti, Venezia, Italy h Institute of Internal Medicine II, University of Ferrara, Italy Received 5 March 2004; received in revised form 30 July 2004; accepted 4 August 2004

KEYWORDS Familial combined hyperlipidemia; Small dense LDL; Statins

Summary Background and aim: Elevation of plasma cholesterol and/or triglycerides, and the prevalence of small dense low density lipoproteins (LDL) particles remarkably increase the risk in patients with familial combined hyperlipidemia (FCHL). There are, at present, inconsistent data on the effects of different treatments on size and density of LDL particles in FCHL patients. Methods and results: A multicenter, randomized, double-blind, double-dummy, parallel group study was designed to evaluate the effect of 3 months’ treatment with atorvastatin (10 mg/day) or pravastatin (20 mg/day) on the lipid/lipoprotein profile and LDL size in a total of 86 FCHL patients. Both statins significantly lowered plasma total and LDL cholesterol, with a significantly higher hypocholesterolemic effect observed with atorvastatin (ÿ26.8 G 11.1% and ÿ35.9 G 11.1%, respectively) compared to pravastatin (ÿ17.6 G 11.1% and ÿ24.5 G 10.2%). The percent

* Corresponding author. Tel.: C39 2 50318303; fax: C39 2 50319900. E-mail address: [email protected] (C.R. Sirtori). 0939-4753/$ - see front matter ª 2005 Published by Elsevier Ltd. doi:10.1016/j.numecd.2004.08.001

48

C.R. Sirtori et al. decrease in plasma triglycerides was highly variable, but more pronounced with atorvastatin (ÿ19.8 G 29.2%) than with pravastatin (ÿ5.3 G 48.6%). Opposite changes in LDL size were seen with the 2 treatments, with increased mean LDL particle diameter with atorvastatin, and decreased diameter with pravastatin, and significant between treatment difference in terms of percent modification vs baseline (C0.5 G 1.6% with atorvastatin vs ÿ0.3 G 1.8% with pravastatin). Conclusions: The present results support the evidence indicative of a greater hypocholesterolemic effect of atorvastatin compared to pravastatin, and in addition show a raising effect of atorvastatin on the size of LDL particles in FCHL patients. ª 2005 Published by Elsevier Ltd.

Introduction The effects of statins on low density lipoprotein (LDL) cholesterol levels are well established and likely to be responsible for the largest part of their coronary preventive activity [1]. However, in recent years, the definition of a clear heterogeneity of LDL particles in terms of density, diameter, and chemical composition [2] has suggested that quality rather than quantity of LDL may exert a direct influence on coronary risk [3]. Case control cohort studies have, in fact, demonstrated that the prevalence of small, dense LDL is an independent risk factor for coronary heart disease (CHD) [4,5]. Small LDL appear to have a delayed turnover, enhanced oxidizability, and to be less well recognized by the LDL receptor vs the case of normal size LDL [6,7]. The prevalence of small, dense LDL particles has been proposed as one of the biochemical markers of familial combined hyperlipidemia (FCHL), a lipid disorder originally described by Goldstein et al. [8] and present in 15e20% of myocardial infarction survivors. These individuals display, in the vast majority, the so called ‘‘B phenotype’’, characterized by LDL particles with ˚ (defined as mean peak diameter smaller than 255 A ‘‘small, dense particles’’) vs the case of ‘‘A phenotype’’ with mean peak diameters of the ˚ [9,10]. Small LDL particles greater than 255 A dense LDL of FCHL are characterized by apoB, triglyceride enrichment, and reduced cholesterol/ apoB ratio [10]. Among the mechanisms apparently linked to a reduced LDL particle size is an enhanced activity of phospholipase A2 (PLA2) [11]. The enzyme, whose activity is statistically associated with an enhanced cardiovascular risk [12], can modify LDL particles by reducing phospholipids and, as a consequence, increasing exposure of the glycosaminoglycan binding segments (particularly 3147e 3157 and 3359e3367) of apoB [11], thus leading to a higher retention in the arterial wall [13].

Modified LDL may also aggregate and fuse, enhancing accumulation of their lipids within the extracellular matrix [13]. There are, at present, inconsistent data on the effects of different treatments on size and density of LDL particles. While, in fact, fibrates tend to reduce LDL triglycerides and density and to increase LDL size in patients with hypertriglyceridemia [14], in the case of FCHL these changes are generally not statistically significant [15]. Furthermore, fibrates do not appear to affect density and composition of LDL in hypercholesteromic patients [16]. Studies on statins have provided inconstant findings. Whereas pravastatin showed apparently no effect on the LDL subclass pattern in patients with FCHL [17], this drug, combined with niacin seemed to be effective in reducing small, dense LDL in hypertriglyceridemic patients [18]. A recent study with fluvastatin in postmenopausal women supports the idea that the LDL subclasses with larger and lighter particles, as assessed by an ultracentrifugal technique, are increased following treatment [19]; in another report, this could not be confirmed by using a technique evaluating particle diameter [20]. Finally, a post-study evaluation of the CARE findings would indicate that, following pravastatin treatment, LDL size is reduced rather than increased [21]. While this finding, somewhat similar to what has been reported in hypercholesterolemic primates [22], might be a specific biochemical marker for patients with cholesterol elevations and large sized LDL, the results may be the cause for potential concern. It was, therefore, planned to carry out a multicenter, randomized, double-blind, double-dummy, parallel groups study evaluating percent changes of LDL-C induced by 3-month atorvastatin and pravastatin in FCHL patients. The major objective of this study was to evaluate possible modifications of LDL subclasses, investigating in particular the sensitivity of well characterized FCHL patients to

Statins and LDL particle size statin treatment. The results of the present study provide an indication that, in fact, atorvastatin treatment may be associated with possibly beneficial effects on the LDL particle profile.

Methods Patients and experimental design The study included 86 patients diagnosed as FCHL according to the following criteria: (1) a primary hyperlipidemia, defined by a plasma LDL-C above 160 mg/dl and/or triglyceride level above 200 mg/dl; (2) at least one first-degree relative with a hyperlipidemia phenotype different from the proband; and (3) a positive history of premature coronary artery disease in at least one blood-related subject or the proband. All patients had been on a standard low fat (30% of calories) diet for at least 1 month before randomization and during the entire study period. Patients were also encouraged to maintain a healthy life-style with further modifications of dietary, smoking and exercise habits, as considered appropriate. Lipid-lowering treatments were stopped at least 1 month before the start of the study. All participating subjects were fully informed of the modalities and end points of the study, which was approved by the Institutional Review Board, and signed an informed consent form. The study protocol was designed for a randomized, double-blind, double-dummy, parallel groups study, and conducted in 8 investigator sites. After a run-in period of 4 weeks, qualifying patients were randomly allocated to receive atorvastatin 10 mg/day (45 patients) or pravastatin 20 mg/day (41 patients) for 12 weeks. Plasma lipid levels, ECG and body mass index were monitored at baseline and after 6 and 12 weeks of treatment; apolipoprotein levels and LDL size were monitored only after 12 weeks of treatment. A standard battery of biochemical and hematological analyses was performed at baseline and after 6 and 12 weeks of treatment. Monitoring of drug intake was performed by pill counting at each visit.

Laboratory procedures Blood samples were collected after an overnight fast and serum was prepared by low speed centrifugation at 4 (C. Serum aliquots were stored at ÿ80 (C for LDL size evaluation. Plasma total cholesterol (TC) and triglyceride (TG) levels were determined with standard enzymatic techniques.

49 LDL-C was calculated by the Friedewald’s formula. Plasma HDL-cholesterol (HDL-C) levels were routinely measured after precipitation of the apo-B containing lipoproteins by dextran sulfateeMgCl2. Apolipoprotein A-I and B levels were determined by immunoturbidimetry, using commercially available polyclonal antibodies. LDL particle size distribution was analyzed by nondenaturing polyacrylamide gradient gel electrophoresis (GGE), using the Pharmacia Phast System (Pharmacia Biotech), as previously described [23]. Briefly, d ! 1.21 g/ml lipoprotein fraction was applied to precast 4e15% polyacrylamide gels and run at 400 V for 285 volthours. Gels were then stained with Coomassie Blue R250, and particle sizes calculated with Multi-Analyst software (BioRad Laboratories) using an LDL standard (25.7 nm), thyreoglobulin (17.0 nm), and apoferritin (12.2 nm) as calibrators [23].

Statistical analyses Determination of sample size The previously quoted study comparing atorvastatin 10 mg/day to pravastatin 20 mg/day in hypercholesterolemic patients showed a 12% (SD 13.15) between treatment difference in the LDL-C percent reduction after 16 weeks of treatment [24]. On this basis, 32 patients per treatment group were calculated to be required to detect a 12% between treatment difference with a SD of 13.5, a test power of 95% and aZ0.05. A total sample of 86 patients was considered adequate, in the hypothesis of 25% proportion of nonevaluable cases. Statistics Statistical analysis of lipid profile (TC, TG, HDL-C, LDL-C) was performed by means of ANOVA applied to ranks calculated from data collected at baseline and at the last available assessment during treatment, with sources of variation including subjects, treatments, and assessment visits, according to a split-plot model. ApoA-I and apoB levels and LDL particle size were determined at baseline and after 12 weeks. These parameters were analyzed by means of ANOVA applied to ranks calculated from data collected at these 2 assessment intervals, with sources of variation including subject, treatment and assessment visit, according to a split-plot model. In addition, for all efficacy parameters percent changes at last assessment vs baseline were calculated and derived ranks were submitted to ANOVA, with sources of variation including treatment. In addition to the planned

50

C.R. Sirtori et al.

analyses, ranks calculated from LDL particle size data were submitted to ANCOVA. Furthermore, in order to focus the analysis on the patient subset most likely sensitive to treatment effect on LDL particle size, a subgroup analysis was performed on this parameter, with patients classified on the basis of baseline LDL particle diameter (%25.5 nm or O25.5 nm). The analyses were carried out as described above for the total population, and included ANCOVA on ranks derived from absolute values, and ANOVA on ranks derived from percent changes at last assessment vs baseline, with sources of variation including treatment and LDL particle size class.

Results Lipid/apolipoprotein profile The characteristics and baseline lipid/apolipoprotein levels of the 2 groups of patients are summarized in Table 1. The baseline characteristics of lipid and lipoprotein parameters were well balanced between the 2 treatment groups, with the exception of triglyceride levels (Table 1). Indeed, TG values were highly variable (48e1048 mg/dl in the atorvastatin group, 92e850 mg/dl in the pravastatin group), with mean values being only slightly higher in patients randomized to atorvastatin vs those randomized to pravastatin, and with superimposable median values (225 and 211 mg/dl, respectively). After 6 weeks treatment, a decrease of TC, TG and LDL-C values was observed in both the atorvastatin and pravastatin groups (Table 2), being more pronounced with atorvastatin than with

Table 1

Characteristics of the patients at baseline Atorvastatin Pravastatin (n Z 45) (n Z 41)

Males (%) Females (%) Age (years) BMI (kg/m2) Total cholesterol (mg/dl) LDL-cholesterol (mg/dl) HDL-cholesterol (mg/dl) Triglycerides (mg/dl) Apolipoprotein A-I (mg/dl) Apolipoprotein B (mg/dl) ApoB/A-I ratio LDL size (nm)

44 56 54.8(11.8) 26.8(3.1) 305.9(54.1) 210.9(46.6) 45.6(13.2) 289.1(210.8) 166.4(28.4)

56 44 53.2(13.5) 26.9(3.8) 312.6(43.8) 225.0(43.9) 47.6(12.0) 237.3(138.2) 167.1(30.6)

153.2(33.5) 0.9(0.2) 25.7(0.5)

161.5(29.8) 1.0(0.2) 26.0(0.6)

Data are expressed as mean (SD).

pravastatin. In both groups mean HDL-C levels were almost unchanged (Table 2). After 12 weeks of treatment, the patients assessed in the atorvastatin and pravastatin groups showed slightly higher mean TC, TG and LDL-C values than those found at the 6-week interval (Table 2). Also at this assessment, mean TC and LDL-C values were significantly lower in the atorvastatin than in the pravastatin group. The 2 groups showed similar mean TG values as well as mean HDL-C levels (Table 2). At the last assessment, average TC and LDL-C percent modifications were ÿ26.8 G 11.1% and ÿ35.9 G 11.1%, respectively, with atorvastatin, vs ÿ17.6 G 11.1% and ÿ24.5 G 10.2% respectively, with pravastatin. The results of ANOVA showed a highly significant treatment effect (p ! 0.01 for TC, p ! 0.001 for LDL-C), time effect (p ! 0.001 for both parameters), as well as time by treatment interaction (p ! 0.01 for TC, p ! 0.001 for LDL-C), thus indicating a significantly more pronounced effect of atorvastatin vs pravastatin on both parameters. The percent decrease in plasma triglycerides was highly variable, but more pronounced with atorvastatin (ÿ19.8 G 29.2%) than with pravastatin (ÿ5.3 G 48.6%). A small, nonsignificant increase of HDL-C was observed with both atorvastatin (C6.7 G 19.63%) and pravastatin (C4.3 G 18.0%). As reported in Table 1, apoA-I, apoB, and the apoB/A-I ratio showed similar mean values at baseline in the 2 groups. After 12 weeks no important changes were observed in apoA-I levels, whereas mean apoB and apoB/A-I ratios were significantly lower in the atorvastatin compared to the pravastatin group (Table 2). Percent decreases of apoB and apoB/A-I ratio vs baseline were more pronounced with atorvastatin (ÿ21.7% and ÿ21.9%, respectively) than with pravastatin (ÿ15.9% and ÿ16.4%, respectively).

LDL size Mean LDL particle size at baseline was slightly but not significantly lower in the atorvastatin compared to the pravastatin group (25.7 G 0.5 nm and 26.0 G 0.6 nm, respectively). After 12 weeks of treatment, an average particle size increase was observed with atorvastatin (25.8 G 0.6 nm), while a decrease was observed after pravastatin (25.9 G 0.5 nm) (Fig. 1). A significant between treatment difference was observed in terms of percent modification vs baseline (C0.5 G 1.6% with atorvastatin vs ÿ0.3 G 1.8% with pravastatin, pZ0.04). The increase in LDL size was particularly evident in the 13 patients showing a clear net LDL

pattern B at baseline (LDL diameter % 25.5 nm). Six of these subjects (46%) were shifted to an LDL pattern A after treatment with atorvastatin, with LDL size increasing from 25.3 G 0.1 nm to 25.9 G 0.3 nm (Fig. 2).

255.3(33.7)* 167.8(30.2)* 48.3(13.4) 217.6(152.7) 172.5(29.2) 126.0(18.9)* 0.8(0.2)* 256.0(35.8)* 165.9(34.2)* 50.2(11.7) 211.2(133.6) e e e 312.6(43.8) 225.0(43.9) 47.6(12.0) 237.3(138.2) 167.1(30.6) 161.5(29.8) 1.0(0.2)

12 weeks Baseline

6 weeks

A vs P Pravastatin

Figure 1 Changes in individual LDL size after 12 weeks treatment with atorvastatin or pravastatin. Bars indicate mean values before and after treatment.

230.5(42.1)* 139.7(33.1)* 47.7(14.0) 216.6(126.7)** 169.1(31.9) 110.6(21.9)* 0.7(0.2)* 212.3(40.9)* 126.2(34.4)* 47.8(13.9) 183.9(95.2)* e e e

Joint or muscular pain at different sites was reported in 2 patients per treatment group; ECG modifications (QRS alterations) were reported as an adverse event in 1 patient in the pravastatin

Data are expressed as mean (SD). *p ! 0.001, **p ! 0.05 vs baseline.

305.9(54.1) 210.9(46.6) 45.6(13.2) 289.1(210.8) 166.4(28.4) 153.2(33.5) 0.9(0.2) Total cholesterol (mg/dl) LDL-cholesterol (mg/dl) HDL-cholesterol (mg/dl) Triglycerides (mg/dl) Apolipoprotein A-I (mg/dl) Apolipoprotein B (mg/dl) ApoB/A-I ratio

Baseline

6 weeks

12 weeks

Adverse events

Atorvastatin

Plasma lipid and lipoprotein levels at baseline, and after 6 and 12 weeks of treatment Table 2

51

!0.01 !0.01 ns ns ns 0.01 ns

Statins and LDL particle size

Figure 2 Gradient gel electrophoresis profiles of the LDL fraction before and after atorvastatin treatment in patients with starting LDL pattern B, and shifting to pattern A after treatment. LDL diameter increases from 25.3 to 25.9 nm.

52 group. Maximum severity of treatment associated with adverse events was judged to be mild in most cases with atorvastatin (11 of the 15 cases, 73.3%) and moderate in most cases with pravastatin (7 of the 9 cases, 77.8%), with a single severe event (vitreum detachment) being reported with atorvastatin. Relationship with the administered study medication was judged improbable in 4 of the 15 cases of the atorvastatin group (26.7%) and 1 of the 9 cases of the pravastatin group (11.1%), possible in 3 (20.0%) and 2 cases (22.2%) of the 2 groups, respectively. No action was taken in most cases of either treatment group. In addition, 1 patient withdrew consent while experiencing paresthesia and arthralgia, not requiring treatment discontinuation in the Investigator’s opinion. In no case code breaking was required. No increased frequency of laboratory abnormalities was apparent in the course of the study.

Discussion The present study, along with the confirmatory assessment of between treatment differences in LDL-C changes vs baseline, based on a statistical hypothesis of superiority of atorvastatin over pravastatin, was also designed to explore between treatment differences in the modification of LDL-C particle size pattern. In particular, a major objective was the direct evaluation of LDL size as determined by gradient gel electrophoresis. There is, at present, frequent confusion between data obtained by GGE, by sedimentation methods (i.e. ultracentrifugal separation), or other technologies that may, in fact, discriminate between particles of different sizes, but whose discriminative power is dependent on factors such as relative lipid, protein composition and others [25]. The overall results of the present study confirm the statistical hypothesis of a greater hypocholesterolemic effect of atorvastatin compared to pravastatin, with statistically and clinically significant more pronounced changes of circulating TC and LDL-C levels following treatment with atorvastatin. In addition, and differently from recent reports [26], both drugs appeared to affect HDL-C levels favorably, with an apparently slightly greater, non-statistically significant effect of atorvastatin. Drug dosages for the present study were suggested by the possibility that the 2 drugs might be associated to another agent, e.g. a fibrate, for FCHL patients who frequently display a significant hypertriglyceridemia. Thus, lower doses than generally used were selected for pravastatin,

C.R. Sirtori et al. whereas atorvastatin, also characterized by a higher interaction potential with other drugs vs pravastatin was used at a dose rated as somewhat more effective than that of pravastatin in hypercholesterolemic patients [24]. The 10 mg atorvastatin dose was, furthermore, recently suggested as most cost beneficial in coronary prevention [27], and it was selected in the combination therapy for high risk hypertensive patients in the ASCOT trial [28]. Evaluation of the possible effect of treatments on LDL particle size, a major objective of this study, showed altogether a satisfactory randomization. In fact the LDL sizes were, on average, only marginally smaller in the atorvastatin group vs pravastatin group at baseline. Opposite changes were seen in the 2 groups after 12 weeks of treatment, with increased mean LDL particle diameter with atorvastatin, and decreased diameter with pravastatin, and significant between treatment difference in terms of percent modification vs baseline. The atherogenic potential of small lipoproteins containing apolipoprotein B100 has been well determined in LDL receptor deficient animals [29]. These, while displaying similar cholesterolemia as mice with apo E deficiency, have a 3e4-fold higher atheroma formation. Possible correlations between LDL particle sizes and other metabolic determinants have been presented by a number of authors. A correlation with triglyceridemia has been suggested by some [4], indicating how the cardiovascular risk associated with reduced LDL size was cancelled after adjusting for the associated hypertriglyceridemia. Genetic linkage studies for the LDL subclass phenotype B included a variety of candidate genes [30] not all characterized by hypertriglyceridemia. Austin et al. [31] more recently, in a Japanese population, confirmed an enhanced risk associated with small LDL size (relative risk 1.28 with 95% CI, 1.01e1.063), the risk again being nonsignificant after adjusting for baseline risk factors. In the large Quebec Cardiovascular Study [5], instead, almost no impact of triglycerides on the relationship between small LDL particles and the risk of IHD was reported. These authors noted a 3.6-fold increase in the risk of IHD comparing individuals in the highest LDL size ˚ ) vs those with the lowest tertile (LDL O 265 A ˚ ). particle size distribution (%254 A The reported findings suggest that atorvastatin clearly leads to an improved LDL particle size profile, in a population characterized by LDL sizes generally below desirable values. They also indirectly confirm a potentially negative effect of pravastatin on this profile. Previous data by our

Statins and LDL particle size group with pravastatin in this type of patients only showed a slight reduction of LDL particle size [17]. In the report by Campos et al. [21] from the CARE study, the clear indication of a negative effect of the agent was attributed by the authors to a ‘‘normalization’’ of LDL sizes in hypercholesterolemic individuals, similar to the case of primates [22]. It should be noted that a prevalence of large LDL may also occur in coronary patients with perfectly normal plasma lipids (total cholesterol ! 200 mg/dl and triglycerides ! 250 mg/dl) [32], thus indicating that in this subpopulation of coronary patients (generally also displaying low HDL2 cholesterol levels) a paradoxical increase of LDL particle size may be found. According to the available evidence, while fibrates are generally associated with a reduction of LDL triglyceride and density accompanied by increased LDL diameter in patients with hypertriglyceridemia and FCHL [14], the effect of HMGCoA reductase inhibitors on LDL subclass pattern in patients with FCHL is unclear. While pravastatin did not reportedly affect LDL pattern in FCHL patients [21], in combination with niacin it was found to be effective in decreasing small, dense LDL in hypertriglyceridemic patients [18]. Fluvastatin was shown to induce a shift of LDL subclasses towards lighter particles [19], but no effect on the LDL size subclass pattern was reported [20]. The present results show that the greater hypocholesterolemic potency of atorvastatin, even when given at a low dose eventually allowing additional drug treatments, is associated with a small increase of LDL particle size, while opposite changes are apparent with the less potent pravastatin, and the effect of the 2 statins is significantly different. While essentially no study has shown a positive activity of pravastatin on LDL particle size, 4 studies with atorvastatin have potentially indicated increases of LDL size in patients not characterized by the FCHL phenotype. Hoogerbrugge and Jansen [33] showed an increased LDL size in male, not female patients with familial hypercholesterolemia treated with high doses of atorvastatin (40 and 80 mg/day). The LDL size, already large at baseline, ˚ to 269 G 6 A ˚ with 40 mg increased from 264 G 8 A ˚ with 80 mg atorvastatin. There and to 270 G 5 A appeared to be a small, statistically significant correlation between changes in TG levels and LDL size in male hypercholesterolemics. More recently Schaefer et al. [34] evaluated fasting and postprandial LDL subclasses (not sizes) in coronary patients on 40 mg/day of atorvastatin. The nuclear magnetic resonance (NMR) technology was used, thus providing little indication as to possible links with other, more frequently used techniques. The

53 results indicate that in these, nonphenotypically characterized patients, drug use was associated with a reduction of large, medium and small LDL, possibly somewhat more marked in the case of the small subfraction. Forster et al. [35] showed a reduction in the plasma concentration of small dense LDL in patients with combined hypelipidemia on 40 mg/day of atorvastatin. Finally, the lipoprotein profile appeared to be improved by high dose atorvastatin (80 mg/day) in patients with impaired fasting glucose/type II diabetes and combined dyslipidemia, particle sizes increasing from a mean ˚ after treatment [36]. Similar of 252.9 to 261.0 A findings in diabetic patients were very recently reported with a novel statin, pitavastatin, that also reduced remnant like particle cholesterol [37]. It should be noted that these last patients had relatively large LDL size at baseline and that, further, there appeared to be no correlation between changes in LDL size and triglyceridemia. The mechanism(s) whereby 1 statin may affect LDL particle sizes, vs a negative effect of another is, of course, difficult to hypothesize. The well documented promoting effect of PLA2 on small LDL formation [11,13] would suggest that an inhibitory activity of statins on PLA2 might beneficially affect LDL particle size. Limited information is available on the activity of different statins on serum PLA2, rated as an inflammatory marker. In a comparative study between simvastatin (40e80 mg) and atorvastatin (20e80 mg) in hypercholesterolemic patients, both drugs resulted in a significant reduction of PLA2, the reduction being around 13e15% vs baseline for both drugs [38]. It is possible, as suggested by Hoogerbrugge and Jansen [33], that higher doses of atorvastatin may lead to a larger effect in term of LDL size. ˚ increase in Presently, it is believed that a 10 A LDL size may be associated with a reduction of cardiovascular risk close to 10% [30,31]. The recent publication of the data from the ASCOT trial, suggesting a larger cardiovascular preventive effect of atorvastatin (10 mg/day) vs other statins [28], particularly pravastatin, would suggest an association between the LDL particle changes and coronary prevention. The present findings are, therefore, worthy of consideration, upon the selection of the most appropriate agent for the treatment of FCHL.

Acknowledgments This study was supported by a Grant from Pfizer, Italy.

54

References [1] LaRosa JC, He J, Vupputuri S. Effect of statins on risk of coronary disease: a meta-analysis of randomized controlled trials. JAMA 1999;282:2340e6. [2] Krauss RM, Burke D. Identification of multiple subclasses of plasma low density lipoproteins in normal humans. J Lipid Res 1982;23:97e104. [3] Austin MA, Breslow JL, Hennekens CH, Buring JE, Willett WC, Krauss RM. Low-density subclass patterns and risk of myocardial infarction. JAMA 1988;260:1917e21. [4] Stampfer MJ, Krauss RM, Ma J, Blanche PJ, Hou LG, Sacks FM, et al. A prospective study of triglyceride level, low-density lipoprotein particle diameter, and risk of myocardial infarction. JAMA 1996;276:882e8. [5] Lamarche B, Tchernof A, Moorjani S, Cantin B, Dagenais GR, Lupien PJ, et al. Small, dense low-density lipoprotein particles as a predictor of the risk of ischemic heart disease in men. Prospective results from the Quebec cardiovascular study. Circulation 1997;95:69e75. [6] Dejager S, Chapman MJ, Bruckert E. Dense low-density lipoprotein subspecies with diminished oxidative resistance predominate in combined hyperlipidemia. J Lipid Res 1993;34:295e308. [7] Campos H, Arnold KS, Balestra ME, Innerarity TL, Krauss RM. Differences in receptor binding of LDL subfractions. Arterioscler Thromb Vasc Biol 1996;16:794e801. [8] Goldstein JL, Schrott HG, Hazzard WR, Bierman EL, Motulsky AG. Genetic analysis of lipid levels in 176 families and delineation of a new inherited disorder, combined hyperlipidemia. J Clin Invest 1973;52:1544e68. [9] Austin MA, Krauss RM. Genetic control of low-density lipoprotein subclasses. Lancet 1986;2:585e92. [10] Kwiterovich Jr PO. HyperapoB: a pleiotropic phenotype characterized by dense low-density lipoproteins and associated with coronary artery disease. Clin Chem 1988; 34:71e7. [11] Hurt-Camejo E, Camejo G, Sartipy P. Phospholipase A2 and small dense low-density lipoprotein. Curr Opin Lipidol 2000;11:465e71. [12] Kugiyama K, Ota Y, Takazoe K, Moriyama Y, Kawano H, Miyao Y, et al. Circulating levels of secretory type II phospholipase A2 predict coronary events in patients with coronary artery disease. Circulation 1999;100:1280e4. ¨o [13] Hurt-Camejo E, Camejo G, Peilot H, O ¨rni K, Kovanen P. Phospholipase A2 in vascular disease. Circ Res 2001;89: 298e304. [14] Tsai MY, Yuan J, Hunnininghake DB. Effect of gemfibrozil on composition of lipoproteins and distribution of LDL subspecies. Atherosclerosis 1992;95:35e42. [15] Hokanson JE, Austin MA, Zambon A, Brunzell JD. Plasma triglyceride and LDL heterogeneity in familial combined hyperlipidemia. Arterioscler Thromb 1993;13:427e34. [16] Tilly-Kiesi M, Tikkanen MJ. Low density lipoprotein density and composition in hypercholesterolemic men treated with HMG-CoA reductase inhibitors and gemfibrozil. J Intern Med 1991;229:427e34. [17] Franceschini G, Cassinotti M, Vecchio G, Gianfranceschi G, Pazzucimi F, Murakami T, et al. Pravastatin effectively lowers LDL cholesterol in familial combined hyperlipidemia without changing LDL subclass pattern. Arterioscler Thromb 1994;14:1569e75. [18] O’Keefe JH, Garris WS, Nelson J, Windsor SL. Effects of pravastatin with niacin or magnesium on lipid levels and postprandial lipemia. Am J Cardiol 1995;76:480e4.

C.R. Sirtori et al. [19] Marz W, Scharnagl H, Abletshauser C, Hoffmann MM, Berg A, Keul J, et al. Fluvastatin lowers atherogenic dense low-density lipoproteins in postmenopausal women with the atherogenic lipoprotein phenotype. Circulation 2001; 103:1942e8. [20] Superko HR, DiRicco C, Krauss RM. Effects of fluvastatin on low-density lipoprotein peak particle diameter. Am J Cardiol 1997;80:78e81. [21] Campos H, Moye LA, Glasser SP, Stampfer MJ, Sacks FM. Low-density lipoprotein size, pravastatin treatment, and coronary events. JAMA 2001;286:1468e74. [22] Rudel LL, Parks JS, Hedrick CC, Thomas M, Williford K. Lipoprotein and cholesterol metabolism in diet-induced coronary artery atherosclerosis in primates. Role of cholesterol and fatty acids. Prog Lipid Res 1998;37: 353e70. [23] Calabresi L, Donati D, Pazzucconi F, Sirtori CR, Franceschini G. Omacor in familial combined hyperlipidemia: effects on lipids and low density lipoproteine subclasses. Atherosclerosis 2000;148:387e96. [24] Bertolini S, Bittolo Bon G, Cambell LM, Farnier M, Langan J, Mahla G, et al. Efficacy and safety of atorvastatin compared to pravastatin in patients with hypercholesterolemia. Atherosclerosis 1997;130:191e7. [25] Capell WH, Zambon A, Austin MA, Brunzell JD, Hokanson JE. Compositional differences of LDL particles in normal subjects with LDL subclass phenotype A and LDL subclass phenotype B. Arterioscler Thromb Vasc Biol 1996; 16:1040e6. [26] Kastelein JJP, Isaacsohn JL, Oser L, Hunninghake DB, Frohlich J, Davidson MH, et al. Comparison of effects of simvastatin versus atorvastatin on high-density lipoprotein cholesterol and apolipoprotein A-I levels. Am J Cardiol 2000;86:221e3. [27] Shepherd J. Resource management in prevention of coronary heart disease: optimizing prescription of lipidlowering drugs. Lancet 2002;359:2271e3. [28] Sever PS, Dahlof B, Poulter NR, Wedel H, Beevers G, Caulfield M, et al. Prevention of coronary and stroke events with atorvastatin in hypertensive patients who have average or lower-than-average cholesterol concentrations, in the Anglo-Scandinavian cardiac outcomes trial-lipid lowering arm (ASCOT-LLA): a multicentre randomised controlled trial. Lancet 2003;361:1149e58. [29] Ve ´niant MM, Sullivan MA, Kim SK, Ambroziak P, Chu A, Wilson MD, et al. Definiting the atherogenicity of large and small lipoproteins containing apolipoprotein B100. J Clin Invest 2000;106:1501e10. [30] Austin MA, Stephens K, Walden CE, Vijsman E. Linkage analysis of candidate genes and the small, dense low-density lipoprotein phenotype. Atherosclerosis 1999; 142:79e87. [31] Austin MA, Rodriguez BL, McKnight B, McNeely MJ, Edwards KL, Curb JD, et al. Low-density lipoprotein particle size, triglycerides, and high-density lipoprotein cholesterol as risk factors for coronary heart disease in older JapaneseeAmerican men. Am J Cardiol 2000;86: 412e6. [32] Campos H, Roederer GO, Lussier-Cacan S, Davignon J, Krauss RM. Predominance of large LDL and reduced HDL2 cholesterol in normolipidemic men with coronary artery disease. Arterioscler Thromb Vasc Biol 1995;15: 1043e8. [33] Hoogerbrugge N, Jansen H. Atorvastatin increases lowdensity lipoprotein in size and enhances high-density lipoprotein cholesterol concentration in male, but not in

Statins and LDL particle size female patients with familial hypercholesterolemia. Atherosclerosis 1999;146:167e74. [34] Schaefer EJ, McNamara JR, Tayler T, Daly JA, Gleason JA, Seman LJ, et al. Effects of atorvastatin on fasting and postprandial lipoprotein subclasses in coronary heart disease patients versus control subjects. Am J Cardiol 2002;90:689e96. [35] Forster LF, Stewart G, Bedford D, Stewart JP, Rogers E, Shepherd J, et al. Influence of atorvastatin and simvastatin on apolipoprotein B metabolism in moderate combined hyperlipidemic subjects with low VLDL and LDL fractional clearance rates. Atherosclerosis 2002;164: 129e45.

55 [36] Pontrelli L, Parris W, Adeli K, Cheung RC. Atorvastatin treatment beneficially alters the lipoprotein profile and increase low-density lipoprotein particle diameter in patients with combined dyslipidemia and impaired fasting glucose/type 2 diabetes. Metabolism 2002;51:334e42. [37] Sone H, Takahashi A, Shimano H, Ishibashi S, Yoshino G, Morisaki N, et al. HMG-CoA reductase inhibitor decreases small dense low-density lipoprotein and remnant-like particle cholesterol in patients with type-2 diabetes. Life Sci 2002;71:2403e12. [38] Wiklund O, Mattsson-Hulten L, Hurt-Camejo E, Oscarsson J. Effects of simvastatin and atorvastatin on inflammation markers in plasma. J Intern Med 2002;251:338e47.