Author + information
- Received December 2, 2008
- Revision received March 23, 2009
- Accepted April 2, 2009
- Published online July 21, 2009.
- Takafumi Hiro, MD⁎,
- Takeshi Kimura, MD†,
- Takeshi Morimoto, MD‡,
- Katsumi Miyauchi, MD§,
- Yoshihisa Nakagawa, MD∥,
- Masakazu Yamagishi, MD¶,
- Yukio Ozaki, MD#,
- Kazuo Kimura, MD⁎⁎,
- Satoshi Saito, MD††,
- Tetsu Yamaguchi, MD‡‡,
- Hiroyuki Daida, MD§,
- Masunori Matsuzaki, MD⁎,⁎ (, )
- JAPAN-ACS Investigators
- ↵⁎Reprint requests and correspondence:
Dr. Masunori Matsuzaki, Division of Cardiology, Department of Medicine and Clinical Science, Yamaguchi University Graduate School of Medicine, Ube, Yamaguchi 755-8505, Japan
Objectives The objective of this study was to evaluate whether the regressive effects of aggressive lipid-lowering therapy with atorvastatin on coronary plaque volume (PV) in patients with acute coronary syndrome (ACS) are generalized for other statins in multicenter setting.
Background A previous single-center study reported beneficial regressive effects of atorvastatin in patients with ACS on PV of the nonculprit site by intravascular ultrasound (IVUS) evaluation. The effect of statins other than atorvastatin on PV has not been evaluated in the setting of ACS.
Methods The JAPAN-ACS (Japan Assessment of Pitavastatin and Atorvastatin in Acute Coronary Syndrome) study was a prospective, randomized, open-label, parallel group study with blind end point evaluation conducted at 33 centers in Japan. A total of 307 patients with ACS undergoing IVUS-guided percutaneous coronary intervention were randomized, and 252 patients had evaluable IVUS examinations at baseline and 8 to 12 months' follow-up. Patients were randomly assigned to receive either 4 mg/day of pitavastatin or 20 mg/day of atorvastatin. The primary end point was the percentage change in nonculprit coronary PV.
Results The mean percentage change in PV was −16.9 ± 13.9% and −18.1 ± 14.2% (p = 0.5) in the pitavastatin and atorvastatin groups, respectively, which was associated with negative vessel remodeling. The upper limit of 95% confidence interval of the mean difference in percentage change in PV between the 2 groups (1.11%, 95% confidence interval: −2.27 to 4.48) did not exceed the pre-defined noninferiority margin of 5%.
Conclusions The administration of pitavastatin or atorvastatin in patients with ACS equivalently resulted in significant regression of coronary PV (Japan Assessment of Pitavastatin and Atorvastatin in Acute Coronary Syndrome; NCT00242944).
Many large-scale pivotal clinical trials (1–3) have shown that 3-hydroxy-3-methyl-glutaryl coenzyme A reductase inhibitors (statins) reduce both atherogenic lipoproteins as well as cardiovascular morbidity and mortality. In addition, several previous multicenter studies in which the authors used intravascular ultrasound (IVUS) imaging revealed that statins attenuate the progression of atherosclerosis or even diminish plaque volume (4,5). An IVUS study in patients with acute coronary syndrome (ACS) demonstrated that statin therapy with 20 mg/day of atorvastatin could reduce nonculprit coronary plaque volume (6). However, this study was a relatively small trial conducted at a single center. This observation, if confirmed in a larger multicenter study, could address one of the mechanisms of improvement of clinical outcome provided by administration of statins in patients with ACS (7–10).
The effect of statins other than atorvastatin on plaque volume (PV) has not been evaluated in the setting of ACS. Pitavastatin is a statin that is commonly used in Japan, South Korea, and Thailand. It has been demonstrated that its ability to lower levels of low-density lipoprotein cholesterol (LDL-C) is comparable with that of atorvastatin (11). Therefore, a multicenter study using a central IVUS core laboratory was designed to assess the effect of pitavastatin on coronary PV compared with that of atorvastatin in patients with ACS.
Study design and ethical considerations
The JAPAN-ACS (Japan Assessment of Pitavastatin and Atorvastatin in Acute Coronary Syndrome) study was a prospective, randomized, open-label, parallel group study with blind end-point evaluation at 33 centers to examine the effect of 8 to 12 months' treatment with pitavastatin versus atrovastatin in coronary plaque regression in nonpercutaneous coronary intervention (PCI) sites of the culprit vessel in patients with ACS. A documentation of the present study design was published before the dataset was locked (12). This study was conducted according to the Declaration of Helsinki and with the approval of the institutional review boards of all 33 participating institutions. Written informed consent to participate was obtained from all of the patients enrolled.
Patient enrollment and randomization
Patients with ACS who satisfied all criteria for inclusion were selected after having a successful PCI under IVUS guidance. We defined ACS as unstable angina pectoris, non–ST-segment elevation myocardial infarction (MI) or ST-segment elevation MI. These diagnoses were made if patients met at least 2 of the following 3 conditions: 1) ischemic ECG changes; 2) the increase (≥2 times) in serum creatine kinase or creatine kinase, myocardial band, or a positive troponin T result; and 3) the presence of symptoms suggestive of ACS. A standard antiplatelet therapy and other medications for ACS were provided.
The patients were randomized within 72 h after PCI to receive either pitavastatin (4 mg) or atorvastatin (20 mg) daily. The dose of 20 mg/day of atorvastatin was selected because when such doses were used in the ESTABLISH (Demonstration of the Beneficial Effect on Atherosclerotic Lesions by Serial Volumetric Intravascular Ultrasound Analysis During Half a Year After Coronary Event) study they significantly reduced coronary PV in patients with ACS (6) and because this dose was the most intensive permitted one to reduce LDL-C in Japan at the beginning of this trial. In addition, the pitavastatin dosage of 4 mg/day was selected because it causes a similar LDL-C–lowering effect to 20 mg/day of atorvastatin (11). We did not include a control group of patients not receiving statin treatment because of ethical reasons. The randomization was stratified by the presence of diabetes mellitus, sex, and total cholesterol level by use of a web response system, which generated association sequence. The IVUS examination was performed at baseline and repeated after 8 to 12 months' administration of the allocated drugs.
Blood examinations for lipid levels and inflammatory markers were performed at baseline and follow-up at 8 to 12 months. Lipid profiles and other biomarkers were measured at SRL Co., Ltd., Tokyo, Japan, and pentraxin3 at Perseus Proteomics Inc., Tokyo, Japan. Safety was evaluated by regular medical examination and laboratory tests at 1, 3, and 8 to 12 months after enrollment. The independent event assessment committee evaluated major adverse cardiac events and any other adverse events.
Examination with IVUS
After IVUS-guided PCI of the culprit lesion of ACS, IVUS examination was performed in both the longest and least angulated culprit vessel segment meeting inclusion criteria. After 200 μg of intracoronary nitroglycerin was administered, a 40-MHz, 2.6-F (0.87-mm) IVUS catheter (Atlantis SR Pro2, Boston Scientific, Natick, Massachusetts) was advanced into the culprit vessel, and the transducer was positioned as far distally as could be safely reached. This procedure was designed to select the longest-possible vessel segment for analysis. A motorized pullback device withdrew the transducer at a speed of 0.5 mm/s. The consoles used were ClearView or Galaxy 2 systems (Boston Scientific). The same imaging system with the same type of IVUS catheter was used for both the baseline and the follow-up examinations When the angular span of the acoustic shadow of calcification or attenuation by some noncalcified tissues was >90°, the case was excluded. After an 8- to 12-month treatment period, IVUS examinations were performed under the conditions identical to the baseline.
Core laboratory analysis of IVUS
Two independent experienced investigators who were unaware of the patient group allocation performed the quantitative IVUS analysis at the central core laboratory. Baseline and follow-up IVUS images were reviewed together on a display, and target segments were selected. The target segment was determined at a non-PCI site (>5 mm proximal or distal to the PCI site) of culprit vessel with a reproducible index, usually a branch site, on the PCI vessel. Spotty calcification, side vein, and distances from side branch, orifice, left anterior descending-left circumflex branch bifurcation, and stent edge also were referred. Subsequently, every 6th image (0.1 mm apart) was manually traced on a commercially available IVUS measurement software (echoPlaque2, INDEC systems Inc., Santa Clara, California). Moreover, this software automatically interpolated the tracing of 5 cross sections in between the 2 manually traced images. Therefore, the volume was calculated from each of 0.017-mm spaced segments. The final cross section for measurement was obtained at a proximal fiduciary site.
The IVUS measurements were performed according to the standards of the American College of Cardiology and the European Society of Cardiology (13). These measurements are present in the standard manner, for which accuracy and reproducibility have been well established (14). The primary end point was the percent change in coronary PV during the observation period:Coronary PV was calculated as the sum of the differences between the external elastic membrane (EEM) and lumen area across all evaluated frames as: PV = Σ(EEMCSA− LUMENCSA), where EEMCSA= external elastic membrane cross-sectional area and LUMENCSA= luminal cross-sectional area.
Major secondary end points include nominal change in percent PV (%PV) and nominal change in normalized plaque volume (NPV) (follow-up minus baseline, respectively):where LMED= median value of observed length in all subjects and LMEASURED= observed length of each plaque.
A detailed structure of statistical analyses in the current study was described elsewhere (12). In brief, this study aims to evaluate whether the effect of pitavastatin on coronary PV would not be inferior to that of atorvastatin and vice versa. Two-sided noninferiority was evaluated by analysis of variance with adjustment for sex, the presence of diabetes mellitus, and total cholesterol level on admission as previously described (12). We decided noninferiority margin as follows: 1) in the ESTABLISH study, the % change in PV of atorvastatin was 13.1 ± 12.8%; 2) standard deviation (±12.8) multiplied by 0.36 (15) yielded 4.608, rounded to 5; and 3) the noninferiority margin was decided as 5%. We calculated 150 subjects in each group with an alpha level of 5%, a power of 80%, and a dropout rate of 30%.
We used full analysis set (FAS) of data for primary analyses. Data of patients were included in FAS if patients had ACS and measurable IVUS both at the enrollment and at follow-up. We prepared per-protocol analysis set of data if enrolled patients completely met the inclusion and exclusion criteria and followed the protocol as it was. If patients received the study drug at least once, they were included in the safety analysis set of data.
After the descriptive statistics, comparisons of continuous variables between the 2 groups were performed by the 2-sample ttest or Wilcoxon rank sum test, and those between the baseline and the follow-up by 1-sample ttests or Wilcoxon sign rank test according to their distributions. Comparisons of categorical values between the 2 groups were performed by chi-square tests and Fisher exact tests. We used general linear models to assess relationships between the percent change in coronary PV and several factors, including serum lipid profile at 8 to 12 months, or to assess interobserver and intraobserver variabilities for measuring plaque area. The numbers of adverse events were assessed to determine safety profiles. The significance level was 5% 2-sided (2.5% 1-sided), and all statistical analyses were performed by the use of the SAS system version 9.1 (SAS Institute, Cary, North Carolina).
The grouping of patients in the present study is shown in Figure 1.Between November 1, 2005, and October 31, 2006, 307 patients were enrolled at 33 centers in Japan, and 153 patients were randomly assigned to receive pitavastatin and 154 to atorvastatin. The IVUS images qualified for evaluation both at baseline and at follow-up were obtained in 125 patients (82%) in the pitavastatin group and in 127 patients (82%) in the atorvastatin group. The median follow-up time with intraquartile range in the pitavastatin group was 9.3 (range 8.5 to 10.3) months and 9.6 (range 8.6 to 10.5) months in the atorvastatin group, respectively.
There was no significant difference in baseline demographics and characteristics between the 2 groups (Table 1).Eighty-two percent of patients were men, and 29% of total patients had diabetes. Sixty-four percent of patients had ST-segment elevation MI, and drug-eluting stents were used in 32% and bare-metal stents in 66%. Plaques proximal to the PCI sites were analyzed in 70% of patients.
We found that LDL-C decreased from 130.9 ± 33.3 mg/dl (3.39 ± 0.86 mmol/l) at baseline to 81.1 ± 23.4 mg/dl (2.10 ± 0.61 mmol/l) at 8 to 12 months' follow-up (p < 0.001) in the pitavastatin group and from 133.8 ± 31.4 mg/dl (3.47 ± 0.81 mmol/l) to 84.1 ± 27.4 mg/dl (2.18 ± 0.71 mmol/l; p < 0.001) in the atorvastatin group (Table 2).We found that high-density lipoprotein cholesterol (HDL-C) as well as triglycerides showed comparable increase between the 2 groups. The inflammatory markers, high-sensitivity C-reactive protein, pentraxin3, and white blood cell counts were markedly increased at baseline and were not different between the 2 groups in terms of percent change.
Efficacy analysis with IVUS
We randomly selected 93 IVUS cross-sectional images from 31 patients to assess the intraobserver and interobserver variabilities for measuring plaque area by 2 independent technicians. The correlation coefficient and mean difference ± SD were 0.99 and 0.02 ± 0.24 mm2(of the absolute mean value, 6.97 ± 4.33 mm2, of the samples) for intraobserver variability and 0.98 and 0.13 ± 0.32 mm2for interobserver variability.
As a primary end point, the percent change in coronary PV showed a significant regression for both groups (−16.9 ± 13.9% in the pitavastatin group, −18.1 ± 14.2% in the atorvastatin group, and −17.5 ± 14.0% for total patients) (Table 3).Noninferiority of pitavastatin to atorvastatin and also atorvastatin to pitavastatin in terms of percent change in PV was proved (Fig. 2).The mean difference of drug effects on percent change in PV (μp – μa), adjusted for sex, the presence of diabetes mellitus, and total cholesterol level, was 1.11% (95% confidence interval [CI]: −2.27% to 4.48%). The upper limit of 95% CI of this difference did not exceed the pre-defined noninferiority margin of 5%. The direction of difference in per-protocol analysis set setting, mean value of 1.36% (95% CI: −2.15% to 4.88%), was consistent with FAS setting. Secondary efficacy end points such as %PV and normalized PV were significantly reduced in both groups (Table 3).
These benefits were associated with significant negative vessel remodeling in both groups (113.0 ± 59.3 mm3to 105.4 ± 55.0 mm3), which consequently provided slight but significant lumen enlargement (56.1 ± 59.3 mm3to 57.8 ± 30.5 mm3). Reduction in EEM volume correlated with the decreased PV (r = 0.7), but there was no correlation between change in lumen volume and change in PV (Fig. 3).Figure 4showed representative examples of IVUS in a single patient with ACS at the baseline and the follow-up period in pitavastatin group.
Plaque regression and biomarkers
Because there was no significant difference in percent change in PV between the 2 groups, the correlation between LDL-C level and percent change in PV was evaluated in the whole FAS patients. There were no significant correlations between LDL-C level at follow-up or at baseline and percent change in PV. Percent change in LDL-C level during the study period also did not significant correlate with percent change in PV (Fig. 5).In addition, there were no significant correlations between high-sensitivity C-reactive protein level at follow-up or at baseline and percent change in PV.
There were no significant differences in the prevalence of these major adverse cardiac events and adverse events between the pitavastatin group and the atorvastatin group (Table 4).The study drugs were discontinued because of either adverse reactions or abnormality of laboratory value only in 2.7% and 4.7% of the pitavastatin group and the atorvastatin group, respectively.
Our study demonstrated that aggressive lipid-lowering therapy with either pitavastatin 4 mg/day or atorvastatin 20 mg/day achieved significant regression of the coronary PV with negative vessel remodeling in patients with ACS based on a randomized, large-scale, multicenter, central IVUS core laboratory evaluation study. Therefore, the results provided support to the hypothesis that administration of statins after the onset of ACS has the potential to reverse the process of atherosclerosis, thereby improving clinical outcome (7–10). Moreover, the results showed that pitavastatin as well as atorvastatin provided a comparable benefit to reduce PV in such patients. This observation also generalized the effect of statins other than atorvastatin on PV in the setting of ACS.
The degree of percent change in PV was −17.5% for total patients in this study. This beneficial regressive effect was similar to that reported by the ESTABLISH single-center study (−13.1%) (6), even more than that of the REVERSAL trial (−0.4%, median in the atorvastatin group) that used a similar primary end point. One of the potential reasons for this might be the difference in clinical presentation (ACS vs. stable coronary artery disease). Evidence has accumulated that shows patients with ACS have many greater-risk nonculprit plaques (16–19). The plaques in the most-diseased 10-mm segments showed more regression than whole coronary artery in the REVERSAL trials (4), and the %PV at the baseline in this study was relatively large compared with those in both trials (JAPAN-ACS, ∼50%; REVERSAL, ∼40%). Furthermore, there was relatively greater proportion of the patients who were administered statins de novo after the entry of this trial. It is also possible there are genetic, racial, or ethnic differences in terms of response to statins.
The correlation between the reduction of LDL-C and the regression of PV was not significant in the present study as compared with previous placebo-controlled studies (Fig. 5) (6,20). One of the reasons might be that this study did not have a placebo arm of patients not receiving lipid-lowering therapy, which was not included for ethical reasons. Regression in PV was observed in a broad spectrum of patients regardless of the baseline LDL-C level. Pleiotropic effects unrelated to LDL-C reduction might be one of the mechanisms of plaque regression. Other pharmacologic and lifestyle interventions applied after the onset of ACS might contribute to the modification of the plaque. In addition, PV regression observed in our study was associated with negative vessel remodeling, which might suggest that nonculprit plaques in patients with ACS were stabilized by statins (21).
The observation of a single plaque in the culprit vessel may not represent the pan-coronary nature of a plaque. Meanwhile, it has been documented that the ACS may represent the pan-coronary process of vulnerable plaque development, suggesting that a single plaque can reflect the general feature of whole coronary artery (19). Another criticism may be that arteries undergoing mechanical interventions were included, which could have affected atheroma measurements. However, IVUS examination for nonculprit vessel in emergent patients with ACS was not possible because of ethical reasons. IVUS might not be appropriate to identify thrombosis. It has been reported that thrombosis can be identified by IVUS with a sensitivity of <50% (22). However, fresh thrombus, which is frequently seen in ACS, can be detected with a true-positive rate of 80% (23). Therefore, meticulous care was taken to exclude thrombus in the present study as strictly as possible with criteria that thrombus in an IVUS image is usually mobile and relatively low echoic, with a uniform texture having some scintillations, some microchannels, and a soft wavy surface.
Intensive statin therapy with 4 mg/day of pitavastatin or 20 mg/day of atorvastatin in patients with ACS resulted in significant regression of atheroma burden with negative vessel remodeling in a large-scale, multicenter trial using a central IVUS core laboratory in which the noninferiority of pitavastatin to atorvastatin was proved.
The authors acknowledge the contributions made by Izumi Miki and Saeko Minematsu for data management and by Hiroko Kanou, Natsuko Yamamoto, Tatsuhiro Fujimura, and Genta Hashimoto for IVUS core laboratory managements and IVUS planimetry.
For a list of JAPAN-ACS Investigators, please see the online version of this article.
The Japan Heart Foundation funded this study with an unrestricted grant from Kowa Pharmaceutical. Kowa Pharmaceutical participated in the preparation of the study design. However, investigators or the independent Clinical Research Coordinator (see Acknowledgments) made the final decision on the study design and database maintenance, wrote the manuscript, and decided to submit the article. An independent statistician (see Online Appendix) analyzed the data. Dr. Hiro has received honoraria for lectures from Kowa Pharmaceutical, Pfizer, and Astellas Pharma. Dr. Kimura is an advisory member of Kowa Pharmaceutical and Pfizer, and has received honoraria for lectures from Kowa Pharmaceutical, Pfizer, and Astellas Pharma. Dr. Morimoto has received honoraria for lectures from Kowa Pharmaceutical and Pfizer. Dr. Miyauchi has received honoraria for lectures from Kowa Pharmaceutical, Pfizer, and Astellas Pharma. Dr. Nakagawa has received honoraria for lectures from Kowa Pharmaceutical, Pfizer, and Astellas Pharma. Dr. Yamagishi has received honoraria for lectures from Kowa Pharmaceutical, Pfizer, and Astellas Pharma and has received a research grant from Kowa Pharmaceutical and Astellas Pharma. Dr. Ozaki is an advisory member of Kowa Pharmaceutical and has received honoraria for lectures from Pfizer and Kowa Pharmaceutical. Dr. Kimura is an advisory member of Kowa Pharmaceutical and has received honoraria for the lectures from Kowa Pharmaceutical and Astellas Pharma. Dr. Saito has received honoraria for lectures from Kowa Pharmaceutical. Dr. Daida is an advisory member of Kowa Pharmaceutical and has received honoraria for lectures and research grants from Kowa Pharmaceutical, Pfizer, and Astellas Pharma. Dr. Matsuzaki is an advisory member of Kowa Pharmaceutical and Pfizer and has received honoraria for lectures and research grants from Kowa Pharmaceutical, Pfizer, and Astellas Pharma.
- Abbreviations and Acronyms
- acute coronary syndrome
- C-reactive protein
- external elastic membrane
- full analysis set
- high-density lipoprotein cholesterol
- intravascular ultrasound
- low-density lipoprotein cholesterol
- percutaneous coronary intervention
- plaque volume
- Received December 2, 2008.
- Revision received March 23, 2009.
- Accepted April 2, 2009.
- American College of Cardiology Foundation
- Okazaki S.,
- Yokoyama T.,
- Miyauchi K.,
- et al.
- Mintz G.S.,
- Nissen S.E.,
- Anderson W.D.,
- et al.
- Wellek S.
- Falk E.,
- Shah P.K.,
- Fuster V.
- Asakura M.,
- Ueda Y.,
- Yamaguchi O.,
- et al.
- Schoenhagen P.,
- Ziada K.M.,
- Kapadia S.R.,
- Crowe T.D.,
- Nissen S.E.,
- Tuzcu E.M.