Dr Klaus G. Parhofer MD, Medical Department II, Klinikum Grosshadern, Marchioninistr. 15, 81377 Munich, Germany (fax: +49 89 7095–8879; e-mail: parhofer@ med2.med.uni-muenchen.de).
Objectives. Atorvastatin is a new potent HMG-CoA reductase inhibitor. We evaluated whether patients with coronary heart disease and severe hypercholesterolaemia showing insufficient LDL (low-density lipoprotein) cholesterol reduction despite combined therapy with simvastatin and regular LDL apheresis will benefit from atorvastatin therapy.
Setting. Tertiary care centre, university hospital.
Methods. In 21 patients treated by LDL apheresis, concomitant simvastatin therapy (40 mg day−1) was replaced by atorvastatin (40 mg day−1) and increased to 60 and 80 mg day−1 (each for 3 months) if no side-effects were reported and NCEP treatment goals were not reached.
Results. In 20 of 21 patients (95%), atorvastatin resulted in significant reduction of LDL cholesterol compared with simvastatin (by 10%, additional 8% and additional 1%, with 40, 60 and 80 mg day−1, respectively). In four patients, NCEP treatment goals were reached (in three by atorvastatin alone, and in one by atorvastatin and apheresis). Patients with little reduction in LDL cholesterol to 40 mg day−1 atorvastatin benefited most by increasing the dose to 60 mg day−1 (additional 13% reduction), whilst those responding to atorvastatin 40 mg day−1 benefited less (additional 1.9% reduction). During atorvastatin therapy, significantly less plasma had to be treated during apheresis resulting in shorter apheresis time. Eight patients (38%) reported side-effects, resulting in discontinuation of atorvastatin in three (14%) and dose reduction in five patients (24%), whilst no elevation of biochemical markers was observed.
Conclusion. Concomitant atorvastatin therapy is superior to simvastatin therapy in patients with severe hypercholesterolaemia treated with regular LDL apheresis, but is associated with a high rate of subjective side-effects.
Atorvastatin is a new HMG-CoA reductase inhibitor which is approved for the treatment of hypercholesterolaemia and combined hyperlipidaemia [ 1]. Doses of atorvastatin ranging from 5 to 80 mg daily can reduce low-density lipoprotein (LDL) cholesterol by 17–61% [ 2–6] and serum triglycerides by 27–46% [ 3]. Even at the maximal approved dose (80 mg day−1), atorvastatin therapy is supposedly not associated with more side-effects than other HMG-CoA reductase inhibitors given at maximal doses [ 7–9].
Thus, atorvastatin seems to be more potent than other commonly used HMG-CoA reductase inhibitors, such as lovastatin [ 5, 8], simvastatin [ 7, 10], pravastatin [ 9] and fluvastatin [ 11]. Its greater efficacy may be related to the drug's longer half-life [ 1], which results in a prolonged inhibition of intracellular cholesterol synthesis [ 12]. Furthermore, animal studies indicate that atorvastatin, in contrast with other HMG-CoA reductase inhibitors, can significantly decrease the secretion rate of apolipoprotein B (apoB)-containing lipoproteins [ 13]. This result is confirmed by the observation that atorvastatin can be successfully used in patients with homozygous familial hypercholesterolaemia (FH) [ 14].
The National Cholesterol Education Program (NCEP) treatment goal is to reduce LDL cholesterol to less than 2.6 mmol L−1 in patients with coronary heart disease [ 15, 16]. If this goal cannot be achieved despite combined diet and drug therapy, patients are considered to be candidates for LDL apheresis, a time-consuming, expensive and invasive form of therapy. However, in many of these patients, LDL hypercholesterolaemia is so severe that the NCEP treatment goal cannot be achieved despite weekly LDL apheresis and treatment with a conventional HMG-CoA reductase inhibitor at the maximal approved dose. Because of its potency, atorvastatin may be useful in these patients, as it may reduce LDL cholesterol to such a degree that the NCEP treatment goal is reached with or without concomitant LDL apheresis therapy.
We evaluated the effect of atorvastatin (40–80 mg day−1) compared with simvastatin (40 mg day−1) on lipoprotein concentrations in a group of patients with severe hyperlipoproteinaemia and coronary heart disease treated with regular LDL apheresis. The aim of the study was to determine whether these patients with insufficient LDL cholesterol reduction, despite combined therapy with conventional HMG-CoA reductase inhibitors and apheresis, will benefit from atorvastatin therapy.
Patients and study design
We currently treat 32 patients with severe hyperlipoproteinaemia and documented coronary heart disease with regular LDL apheresis. All patients taking simvastatin (40 mg day−1) for a minimum of 6 months without occurrence of undesired reactions (n= 21) were included after giving informed written consent. In 17 of the 21 patients, the underlying hyperlipoproteinaemia was heterozygous familial hypercholesterolaemia (FH) and in four patients severe combined hyperlipidaemia. All patients adhered to a step I diet recommended by the NCEP [ 15]. LDL apheresis treatment was performed at weekly (n= 15) or bi-weekly intervals (n= 5); one subject was treated every three weeks. The methods used for LDL cholesterol elimination included immunoadsorption (Therasorb, Unterschleissheim, Germany, n= 12 patients), dextran sulphate adsorption (KANEKA, Osaka, Japan, n= 3), HELP apheresis (Braun, Melsungen, Germany, n= 2) and cascade filtration (Diamed, Cologne, Germany, n= 4).
Apheresis is standardized such that a post-apheresis LDL cholesterol of 50–60 mg dL−1 (1.3–1.6 mmol L−1) is reached. At an individually defined plasma volume (approximately 2 h after the beginning of apheresis) plasma cholesterol concentration is determined and the necessary plasma volume in order to reach that goal is estimated by comparison with prior aphereses in the same patient. In patients treated by apheresis techniques with plasma limit (HELP apheresis and cascade filtration), not more than the maximal approved plasma volume was processed.
Following 40 mg day−1 of simvastatin (treatment period 1), atorvastatin was given at 40 mg day−1 for a period of 3 months (treatment period 2). Atorvastatin was then increased to 60 mg day−1 (treatment period 3) in all patients who did not experience undesired reactions and who did not meet the NCEP treatment goals (average LDL cholesterol < 2.6 mmol L−1) during treatment period 2. Two months later, atorvastatin was given at the maximal approved dose (80 mg day−1) in all patients who did not meet the NCEP treatment goals with 60 mg day−1 of atorvastatin, but demonstrated any further decrease in LDL cholesterol with 60 mg day−1 compared with 40 mg day−1 of atorvastatin.
If undesired reactions were reported, the atorvastatin dose was decreased to the maximal tolerable dose. Patients who did not tolerate the initial dose (40 mg day−1) received simvastatin, if symptoms did not disappear within 2 weeks.
LDL cholesterol, HDL (high-density lipoprotein) cholesterol and plasma triglycerides were determined immediately before and after each apheresis treatment. For every patient, mean plasma lipid concentrations were calculated from a minimum of four representative apheresis treatments (range: four to eight aphereses) during each treatment period. The average LDL cholesterol was determined as the area under the curve as previously described [ 17]. Mean HDL cholesterol and mean plasma triglycerides were calculated from plasma samples drawn immediately before apheresis [ 17]. The plasma volume processed during LDL apheresis was documented for every patient. During simvastatin treatment, safety parameters (ALT, yGT, CK) were controlled monthly, whilst these parameters were determined bi-weekly during atorvastatin treatment. Patients were asked at every visit to report undesired reactions.
Blood was drawn in EDTA-containing tubes. Plasma was obtained by centrifugation (20 min, 1360 g). Cholesterol and triglyceride levels were measured enzymatically in plasma using an automated clinical chemistry analyser (Epos, Eppendorf, Germany). HDL cholesterol was determined after precipitation of apolipoprotein B (apoB)-containing lipoproteins by dextran sulphate and magnesium acetate. LDL cholesterol was determined by the formula of Friedewald et al. [ 18]. If plasma triglycerides were above 4.5 mmol L−1, very low-density lipoproteins (VLDLs) were isolated by preparative ultracentrifugation (18 h, d = 1.006 g mL−1, 270 000 g, 4 °C; Beckman Ti 50.4 rotor, Palo Alto, CA). In the supernatant, VLDL cholesterol and VLDL triglycerides were determined. In the infranatant (containing HDL and LDL) cholesterol was measured, then LDLs were precipitated by dextran sulphate and magnesium acetate. LDL cholesterol was calculated by subtraction of cholesterol in the supernatant (HDL cholesterol) from total cholesterol in the infranatant.
All results are expressed as means ± SD. Statistical evaluation was performed using Wilcoxon's matched-pairs signed rank test using the Statistical Package of Social Sciences software (SPSS/PC+, V3.1, SPSS Inc., Chicago, IL, USA). Differences between subgroups (male – female and responder – non-responder) were evaluated for statistical significance by the Mann–Whitney U-test using the same software.
In none of the 21 patients included in the study did atorvastatin have to be discontinued because of increases in biochemical markers. Furthermore, no significant differences were observed between treatment periods with regard to mean and maximal concentrations of yGT (17 U L−1, 4–71 U L−1), ALT (13 U L−1, 6–25 U L−1) and CK (40 U L−1, 9–98 U L−1). One patient showed a transient increase in alkaline phosphatase and yGT (maximal AP, 300 U L−1; maximal yGT, 90 U L−1) after switch to atorvastatin (40 mg day−1), which resolved 4 weeks later and did not recur during 60 mg day−1 of atorvastatin. However, eight of the 21 patients (38%) subjectively experienced side-effects such as myalgia (n= 5), depression (n= 2) and loss of appetite (n= 1). In three of these eight patients the side-effects were observed with 40 mg day−1 atorvastatin and the drug had to be discontinued. In the five other patients side-effects were only observed with higher atorvastatin doses (60 mg day−1, four patients; 80 mg day−1, one patient) and disappeared after dose reduction ( Fig. 1).
In 20 of 21 patients (95%) atorvastatin resulted in substantial reduction of LDL cholesterol compared with simvastatin ( Fig. 1 and 2). In four of these patients, atorvastatin decreased LDL cholesterol to such a degree that the NCEP treatment goal was reached. Furthermore, in three of these four patients, LDL apheresis treatment was discontinued, as LDL cholesterol levels were below 2.6 mmol L−1 during atorvastatin treatment alone. In one patient the treatment goal was reached at a dose of 60 mg day−1 atorvastatin, but he had to continue LDL apheresis to maintain this concentration.
Figure 2 shows the individual course of average LDL cholesterol in all patients (n= 21) during treatment periods 1 to 4. Patients were included in this analysis if undesired reactions were reported 4 or more weeks after change of medication (six patients). Patients who discontinued atorvastatin (n= 3) are shown by dashed lines. Almost all patients (n= 20) exhibited a further decrease in average LDL cholesterol during atorvastatin (40 mg day−1) compared with simvastatin (40 mg day−1) treatment. However, of the 15 patients eligible for an increase of the atorvastatin dose to 60 mg day−1, only nine patients demonstrated a further decrease in average LDL cholesterol. Of the remaining six patients, one discontinued 60 mg day−1 of atorvastatin within 1 week, whilst the other five did not exhibit any further reduction in LDL cholesterol and were thus not eligible for 80 mg day−1. Two of the nine patients showing LDL cholesterol reduction with 60 mg day−1 reported undesired reactions and one met the treatment goal. Therefore, six patients could be treated at the maximal dose. However, only one patient out of five tolerating this dose achieved a further reduction in LDL cholesterol at 80 mg day−1 compared with 60 mg day−1 atorvastatin.
Replacing simvastatin (40 mg day−1) with atorvastatin (40 mg day−1) decreased average LDL cholesterol by 10% (3.57 ± 0.54 vs. 3.21 ± 0.59 mmol L−1, P < 0.01) and triglycerides by 9% (1.72 ± 0.88 vs. 1.56 ± 0.78 mmol L−1, P < 0.01), whilst HDL cholesterol remained unchanged (1.22 ± 0.26 vs. 1.22 ± 0.26 mmol L−1, NS) ( Table 1). Furthermore, atorvastatin treatment resulted in a reduction of mean plasma volume treated during each apheresis (3260 ± 930 mL vs. 3070 ± 852 mL, P < 0.01).
Table 1. Mean plasma lipoprotein concentration and mean processed plasma volume with LDL apheresis during concomitant simvastatin and atorvastatin therapy
Increasing the dose to 60 mg day−1 in 14 patients resulted in a further reduction of LDL cholesterol by 8% (3.41 ± 0.52 vs. 3.13 ± 0.41 mmol L−1, P < 0.04), whereas plasma volume treated with apheresis remained unchanged (3150 ± 920 vs. 3200 ± 1100 mL, NS). The maximal approved dose of atorvastatin (80 mg day−1) induced no additional decrease in average LDL cholesterol (−0.9%, NS). However, less plasma volume had to be processed during apheresis in comparison to 60 mg day−1 of atorvastatin (2760 ± 450 vs. 3200 ± 880 mL, P < 0.03). Plasma triglycerides and HDL cholesterol remained unchanged with 60 or 80 mg day−1 in comparison with 40 mg day−1 of atorvastatin ( Table 1).
Patients were divided into responders (more than 10% decrease in LDL cholesterol during atorvastatin 40 mg day−1 compared with simvastatin 40 mg day−1, n= 8) and non-responders (less than 10% decrease in LDL cholesterol, n= 13). All but one non-responder had a significant decrease in LDL cholesterol when atorvastatin was increased to 60 mg day−1 (−13.3 ± 11%, P < 0.01), whilst responders to 40 mg day−1 did not benefit from an increase of atorvastatin to 60 mg day−1 (−1.9 ± 7%, NS).
Atorvastatin treatment decreased average LDL cholesterol similarly in men and women. However, in male patients the biggest benefit was achieved by changing therapy to 40 mg day−1 atorvastatin (−14%, P < 0.01), whilst increasing the dose to 60 mg day−1 resulted in only a small further reduction in LDL cholesterol (−4%, NS). In female patients, 40 mg day−1 atorvastatin resulted in only a moderate reduction (−6%); 60 mg day−1, however, resulted in a considerable further reduction in LDL cholesterol (−13%).
Exclusion of all patients treated by apheresis at bi-weekly or even longer intervals (n= 6), or of those subjects suffering from familiar combined hyperlipidaemia (n= 4), did not affect the results (data not shown).
We also analysed the results according to the apheresis technique used to evaluate whether the method of LDL elimination affects the response to atorvastatin ( Table 2). Patients treated by immunoadsorption (n= 12) showed a smaller decrease in LDL cholesterol (−7% vs −11%) and plasma triglycerides (−4% vs. −16%) with atorvastatin treatment than those treated by other techniques (dextran sulphate, HELP apheresis, cascade filtration, n= 9). Changing treatment from simvastatin to atorvastatin also resulted in a slight, but significant decrease in HDL cholesterol in the patients treated by immunoadsorption (−4%, P < 0.05), whereas it remained unchanged in the patients treated by non-immunoadsorption techniques ( Table 2).
Table 2. Plasma lipoproteins and plasma volume treated by immunoadsorption (n = 12) and non-immunoadsorption techniques (n = 9) with concomitant simvastatin (40 mg day−1) or atorvastatin (40 mg day−1) therapy
In patients with severe LDL hypercholesterolaemia and coronary heart disease treated by regular LDL apheresis, the use of atorvastatin at doses ranging from 40 to 80 mg day−1 decreases LDL cholesterol by 10–21% and triglycerides by 9–14% compared with 40 mg day−1 simvastatin. The NCEP treatment goal was met by four patients after change to atorvastatin, three of whom could discontinue LDL apheresis. With respect to biochemical markers the use of atorvastatin was safe; however, the percentage of patients reporting subjective side-effects was unexpectedly high (38%).
The recommended treatment goal for secondary prevention of coronary heart disease is a LDL concentration below 2.6 mmol L−1 [ 15, 16, 19]. In the patients included in this study LDL hypercholesterolaemia was so severe that this value was not achieved by regular LDL apheresis combined with simvastatin treatment. Changing the lipid-lowering medication to atorvastatin allowed apheresis to be discontinued in three patients because the NCEP treatment goal was reached. In a further patient, NCEP criteria were met by combined treatment with atorvastatin and LDL apheresis.
The dose-related effect of atorvastatin on LDL cholesterol in our study is comparable to findings of a placebo-controlled study evaluating the efficacy of atorvastatin given at doses between 2.5 and 80 mg day−1, which showed the biggest relative LDL reduction at lower doses of atorvastatin (20 mg day−1, 44% from baseline; 80 mg day−1, 61% from baseline) [ 4].
However, in our study, some subgroups, such as women and those showing little reduction with 40 mg day−1 of atorvastatin ('non-responders'), seem to benefit considerably from increasing the dose to 60 mg day−1. This may indicate the existence of a threshold to achieve maximal LDL cholesterol reduction with atorvastatin treatment. The male–female difference may be related to differences in body composition or in the severity and type of hyperlipoproteinaemia.
In previous studies [ 10, 11] comparing atorvastatin with simvastatin therapy, a greater benefit concerning LDL cholesterol and triglyceride reduction was usually observed with atorvastatin treatment than in the study reported here. In patients with moderate LDL hypercholesterolaemia, 40 mg day−1 of atorvastatin vs. 40 mg day−1 of simvastatin resulted in a further reduction of LDL cholesterol by 20% and of triglycerides by 55% [ 10]. In a recently published review it was concluded that 40 mg day−1 of atorvastatin can reduce LDL cholesterol by 18% compared with 40 mg day−1 of simvastatin [ 11].
The greater LDL cholesterol reduction by atorvastatin described in the studies above [ 10, 11], as compared with our own study, may be due to differences in study design and patient populations. Furthermore, the concomitant apheresis treatment may diminish differences between atorvastatin and simvastatin therapy concerning LDL cholesterol if average LDL cholesterol is used for comparison. Plasma volumes treated with each individual apheresis are regulated such that a post-apheresis LDL cholesterol of 1.3–1.6 mmol L−1 is reached, and the post-apheresis value is a major determinant of average LDL cholesterol. Therefore, more efficient concomitant drug therapy will result in less plasma volume processed during apheresis and therefore shorter apheresis times.
Data on the use of atorvastatin in patients with heterozygous FH treated with regular LDL apheresis are sparse. A study comparing the effect of atorvastatin at different doses to traditionally used HMG-CoA reductase inhibitors has not been conducted in this patient group. Preliminary results from a study comparing simvastatin with atorvastatin at identical doses (40 mg day−1) in patients treated by regular LDL apheresis show a decrease in LDL cholesterol by 9% and plasma triglycerides by 19% [ 20]. However, a direct comparison to our study is difficult as LDL cholesterol concentrations were not calculated from the area under the rebound curve, which estimates average LDL cholesterol concentration between LDL apheresis treatments more accurately. Furthermore, the plasma volume processed during LDL apheresis was not included in the analysis of the other study.
In a recently published study, the effect of atorvastatin (80 mg day−1) was evaluated in comparison to placebo in patients with homozygous FH treated with regular apheresis [ 14]. Atorvastatin resulted in a marked inhibition of cholesterol biosynthesis associated with a decrease of LDL cholesterol by 31%. However, this study was different, as patients with homozygous FH were studied and placebo was used for comparison. Our study was designed to answer the clinically important question as to whether the more common patient with heterozygous FH treated with apheresis and conventional HMG-CoA reductase inhibitor therapy will benefit from atorvastatin therapy.
In a subgroup analysis, we evaluated whether the response to atorvastatin treatment depends on the method of LDL elimination ( Table 2). The comparison of highly selective (immunoadsorption) vs. less selective techniques (non-immunoadsorption techniques) showed a similar further decrease in LDL cholesterol after the switch to atorvastatin (9% vs. 11%). However, plasma triglycerides were decreased in patients treated with non-immunoadsorption techniques and HDL cholesterol was reduced in patients treated by immunoadsorption. The decreased HDL cholesterol level in patients treated by immunoadsorption (−4%) seems paradoxical considering that this technique is the most selective method to eliminate LDL cholesterol [ 21]. This result is most likely related to differences in the patients' characteristics, since the group treated by non-immunoadsorption techniques had a somewhat less severe hyperlipoproteinaemia and included patients with familial combined hyperlipoproteinaemia.
Previous studies have shown that the administration of atorvastatin is safe [ 6, 10, 22–24]; the frequency of undesired reactions was not higher compared with other HMG-CoA reductase inhibitors [ 5, 7–9] given at equivalent doses. Our results confirm these studies, as we did not observe any significant differences in mean and maximal values of CK, yGT and ALT between simvastatin and atorvastatin.
However, a large proportion of our patients reported subjective side-effects such as myalgia (24%), depression (10%) and loss of appetite (5%). These symptoms resulted in dose reduction in five patients and drug discontinuation in three patients. This result is surprising since all patients had tolerated simvastatin without any complaints for at least 6 months. Since the plasma volume processed during LDL apheresis was less with atorvastatin than with simvastatin treatment, we cannot explain the occurrence of side-effects during atorvastatin treatment by altered LDL apheresis conditions. This high rate of side-effects may be related to patients' expectation that a possibly more effective drug will also be characterized by more frequent and more severe side-effects. Furthermore, we were reluctant to continue atorvastatin if possible side-effects were reported, since at the time of this study no data were available concerning atorvastatin–apheresis interactions. Therefore, our threshold in discontinuing or reducing the dose of atorvastatin was low.
In summary, we found that the use of atorvastatin instead of simvastatin can further reduce LDL cholesterol and plasma triglycerides in most patients with severe LDL hypercholesterolaemia treated by LDL apheresis. Atorvastatin therapy was associated with an unexpectedly high rate of subjective side-effects, whilst biochemical safety markers were not affected.
We conclude that concomitant atorvastatin therapy is superior to conventional HMG-CoA reductase inhibitor therapy in patients with severe hypercholesterolaemia and coronary heart disease treated with regular LDL apheresis.
We thank our patients for participating in the study, Ms K. Henze for technical assistance and the staff of the apheresis unit for organizational help. We appreciate the continuous support of our apheresis-related research activities by Therasorb (Unterschleissheim, Germany), Kaneka (Osaka, Japan), Braun (Melsungen, Germany), and Diamed (Cologne, Germany).