Impact of omega-3 polyunsaturated fatty acids on coronary plaque instability: An integrated backscatter intravascular ultrasound study

Atherosclerosis 218 (2011) 110–116

Contents lists available at ScienceDirect

Atherosclerosis journal homepage: www.elsevier.com/locate/atherosclerosis

Impact of omega-3 polyunsaturated fatty acids on coronary plaque instability: An integrated backscatter intravascular ultrasound study Tetsuya Amano a,∗ , Tatsuaki Matsubara b , Tadayuki Uetani a , Masataka Kato a , Bunichi Kato a , Tomohiro Yoshida a , Ken Harada a , Soichiro Kumagai a , Ayako Kunimura a , Yusaku Shinbo a , Katsuhide Kitagawa a , Hideki Ishii c , Toyoaki Murohara c a

Department of Cardiology, Chubu-Rosai Hospital, Nagoya, Japan Department of Internal Medicine, School of Dentistry, Aichi-Gakuin University, Nagoya, Japan c Department of Cardiology, Nagoya University Graduate School of Medicine, Nagoya, Japan b

a r t i c l e

i n f o

Article history: Received 4 March 2011 Received in revised form 8 May 2011 Accepted 24 May 2011 Available online 1 June 2011 Keywords: Omega-3 polyunsaturated fatty acids Integrated backscatter intravascular ultrasound Coronary plaque instability Fish oil Cardiovascular disease

a b s t r a c t Objective: To assess the impact of omega-3 polyunsaturated fatty acids (␻3 PUFAs) on coronary plaque instability. Methods: Serum content of eicosapentaenoic acid (EPA), docosapentaenoic acid (DPA), and docosahexaenoic acid (DHA) was measured in 336 of 368 consecutive patients suspected of having coronary artery disease who underwent coronary angiography. Conventional and integrated backscatter intravascular ultrasound (IB-IVUS) parameters were analyzed in 116 patients with 128 coronary plaques, using a 43MHz (motorized pullback 0.5 mm/s) intravascular catheter (View It, Terumo Co., Japan). Lipid-rich plaques were classified into two categories according to their components. Results: Patients with acute coronary syndrome had significantly lower levels of ␻3 PUFAs (especially of EPA and DPA) than those without it. IB-IVUS analyses showed that ␻3 PUFAs correlated inversely with % lipid volume and positively with % fibrous volume. Patients with low EPA levels, low DPA levels, and low DHA levels had a significantly higher % lipid volume (p = 0.048, p = 0.008, and p = 0.036, respectively) and a significantly lower % fibrous volume (p = 0.035, p = 0.008, and p = 0.034, respectively) than those with high levels of these fatty acids. Even after adjustment for confounders, the presence of both low EPA and low DPA levels proved to be an independent predictor for lipid-rich plaques in any of the two categories. Conclusions: A lower serum content of ␻3 PUFAs (especially of EPA and DPA) was significantly associated with lipid-rich plaques, suggesting the contribution to the incidence of acute coronary syndrome. © 2011 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Omega-3 polyunsaturated fatty acids (␻3 PUFAs) therapy continues to show great promise in primary, and particularly in secondary prevention of cardiovascular diseases. Most of the evidence for benefits of ␻3 PUFAs has been obtained from research on fish oils that typically contain eicosapentaenoic acid (EPA), docosapentaenoic acid (DPA), and docosahexaenoic acid (DHA), the long-chain fatty acids in this family [1–4]. Several biological effects of ␻3 PUFA have accounted for the potential benefits of ␻3 PUFA in atherosclerosis [5,6], heart failure [7,8], and arrhythmias [9,10]. However, data as to the direct correlation of ␻3 PUFA with coronary plaque composition are quite limited. Integrated

backscatter intravascular ultrasound (IB-IVUS) has recently been developed allowing analysis of the tissue components of coronary plaques in vivo [11–14]. Apart from its diagnostic utility, IB-IVUS has also proved useful to assess the prognosis of patients with coronary atherosclerosis, and the risk of experiencing a coronary event [15,16]. In the present study, we hypothesized that ␻3 PUFAs would be associated with the presence of acute coronary syndrome (ACS), and assessed the impact of ␻3 PUFAs on coronary plaque instability using IB-IVUS.

2. Methods 2.1. Patients and study design

∗ Corresponding author at: Department of Cardiology, Chubu-Rosai Hospital, Kohmei 1-10-6, Minato-ku, Nagoya 455-8530, Japan. Tel.: +81 052 652 5511; fax: +81 052 653 3533. E-mail address: [email protected] (T. Amano). 0021-9150/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.atherosclerosis.2011.05.030

This study was an observational investigation of 368 consecutive patients suspected of having coronary artery disease who underwent coronary angiography (CAG) between July 2009 and February 2010. In Analysis 1, 368 patients were allocated to the

T. Amano et al. / Atherosclerosis 218 (2011) 110–116

presence or absence of ACS. Thirty-two patients (2 ACS and 30 non-ACS patients) were excluded because serum biomarkers were not available (on-line supplementary Figure). ACS included acute myocardial infarction (AMI) and unstable angina pectoris (UAP). Diagnosis of AMI was based on elevation of at least one biomarker (creatine kinase {CK}, CK-MB or troponin T), characteristic electrocardiogram changes and a history of prolonged acute chest pain. UAP was defined as either angina with a progressive crescendo pattern or angina at rest. In Analysis 2, conventional and IB-IVUS parameters were measured in 158 patients who underwent percutaneous coronary intervention (PCI). After exclusion of 5 patients because of inestimable IVUS images and of 37 patients with ACS, data from 116 patients with 128 plaques were evaluated (on-line supplementary Figure). 2.2. Data collection and laboratory examinations The composition of serum fatty acids was determined by capillary gas chromatography. Total lipids were extracted by Folch’s procedure and then fatty acids were methylated with boron trifluoride and methanol; methylated fatty acids were analyzed using a Shimadzu GC-17A gas chromatograph (Shimadzu Corporation, Kyoto, Japan) and an Omegawax 250 capillary column (0.25 mm internal diameter × 30 m, Supelco, Inc., Bellefonte, PA, USA). Blood samples were collected from 336 patients before CAG or PCI. Hypertensive patients were those with documented blood pressure of >130/85 mm Hg on 2 or more occasions, or who were already on antihypertensive therapy. Diabetes mellitus was defined if the patient was taking any antihyperglycemic medication or had previously been diagnosed with diabetes mellitus. Smoking status was defined if at baseline the patient was a smoker or had quit less than a year before the study. We obtained informed consent from all subjects, and the investigation was approved by the ethics committee of Chubu Rosai Hospital.

111

frames) within the first 35 mm of the left anterior descending and left circumflex arteries, and in the worst plaque that could be observed by IVUS of the right coronary artery. In the conventional IVUS analysis, cross-sectional images were quantified for lumen cross-sectional area (LCSA), external elastic membrane (EEM), cross-sectional area (CSA), and plaque (P) + media (M) crosssectional area (P + M CSA = EEM CSA–LCSA) using software built into the IVUS system. The remodeling index was defined as the ratio of EEM CSA at the measured lesion (minimum luminal site) to the reference EEM CSA (average of the proximal and distal reference segments). The remodeling index was calculated in the segment with the minimal luminal area. The percentage fibrous area (fibrous area/plaque area, % FA), the percentage lipid area (lipid area/plaque area, % LA) and the percentage calcified area (calcified area/plaque area, % CA) were automatically calculated using the IB-IVUS system. Three-dimensional (3D) analysis of conventional and IB-IVUS images was performed to obtain the fibrous volume, lipid volume and calcified volume (separate sums of fibrous, lipid and calcified areas in the segment with the minimum luminal area at 1 mm axial intervals for 5 mm proximally and for 5 mm distally). A total of eleven IB-IVUS images were captured at intervals of 1 mm for a 10 mm length within each plaque. Then, percentage values were calculated (fibrous volume/plaque volume, %FV; lipid volume/plaque volume, %LV; and calcified volume/plaque volume, %CV). Lipid-rich plaques (LRP) were arbitrarily classified into two categories [Category 1; both an increase in %LV (>63%) and a decrease in %FV (<36%)], or [Category 2; both an increase in %LV (>75%) and a decrease in %FV (<25%)], which were 75th percentile of %LV and 25th percentile of %FV (Category 1), or 90th percentile of %LV and 10th percentile of %FV (Category 2) in the present study population. Variability in %LV and %FV determined by two physicians was also considered from thirty randomly selected records. 2.5. Statistical analyses

2.3. Coronary angiography and intravascular ultrasound procedure Before performing CAG and PCI, patients were administered an intracoronary 0.5 mg dose of isosorbide dinitrate in order to prevent coronary spasm. IVUS catheters (View It, Terumo Co., Japan) were inserted in 187 coronary vessels of 158 patients as far distally as possible before the PCI procedure. Continuous ultrasound images were acquired during withdrawal of the catheter through a segment of the artery at a constant rate of 0.5 mm/s. A personal computer (Windows XP Professional, CPU: 3.4 GHz) equipped with commercially available custom software (VISIWAVE IB, Terumo Co.) was connected to the IVUS imaging system (VISIWAVE, Terumo Co.) to obtain radio frequencies, and signal trigger outputs. Ultrasound backscattered signals were acquired using a 43 MHz (motorized pullback 0.5 mm/s) mechanically rotating IVUS catheter, digitized and subjected to spectral analysis. IB values for each tissue component were calculated as average power levels using a fast Fourier transform, measured in decibels (dB), of the frequency component of backscattered signals from a small volume of tissue [17]. The segmentation of each tissue component was entirely automated [12], and the excellent correlation of IB-IVUS and histology has been reported in the validation studies [13,17]. Quantitative coronary angiography (QCA) analysis was conducted, and the reference diameter and percentage diameter stenosis were measured with a validated automated edge-detection program (CMS-MEDIS Medical Imaging System, Leiden, The Netherlands).

Continuous and categorical variables were assessed as mean ± SD values and proportions, respectively. Continuous variables with a skewed distribution were listed as median (range). Univariate analysis was applied to compare various clinical and laboratory parameters between patients with and without ACS (Analysis 1), and to compare QCA and IVUS parameters between patients with low or high serum levels of ␻3 PUFAs (Analysis 2), by unpaired Student t-test for continuous normally distributed variables and by Mann–Whitney U-tests for non-normally distributed variables. Chi-square and Fisher exact tests were applied when appropriate for categorical variables. Three groupings of ␻3 PUFAs tertiles as to the relationship to ACS were evaluated using analysis of variance (ANOVA), and Bonferroni test was performed for multiple comparisons. Simple linear regression analysis was performed with %LV and %FV as the dependent variables. Logistic regression analysis was applied to study the association of ␻3 PUFAs with ACS (Analysis 1) and for best predictors of LRPs (Analysis 2), after adjusting for confounding and various risk factors. The Bland Altman test was applied to assess concordance of measurements. All variables with a p value <0.10 on univariate analysis were considered in the multivariate model. A p value <0.05 was considered as statistically significant. 3. Results 3.1. Association of ␻3 PUFAs with ACS

2.4. Measurement of conventional and IB-IVUS parameters Conventional and IB-IVUS parameters were measured in the worst plaque (a plaque burden ≥40% in at least 3 consecutive

Baseline characteristics of patients with and without ACS are shown in Table 1. Of the 368 patients subjected to CAG, 32 were excluded because data on serum biomarkers were not available;

112

T. Amano et al. / Atherosclerosis 218 (2011) 110–116

Table 1 Baseline characteristics of patients in Analysis 1 and Analysis 2. Analysis 1

Age, years Male gender Hypertension Diabetes mellitus Current smoker Old myocardial infarction Multiple vessel disease Previous PCI Previous CABG Triglycerides, mg/dL HDL-cholesterol, mg/dL LDL-cholesterol, mg/dL HOMA-IR EPA, wt% DPA, wt% DHA, wt% AA, wt% C-reactive protein, mg/dL Statins

Analysis 2

p value

ACS (n = 39)

Non-ACS (n = 297)

Non-ACS (n = 116)

65 ± 11 28 (72) 26 (67) 16 (41) 18 (46) 5 (13) 10 (26) 4 (10) 1 (3) 132 [32–515] 44 ± 11 124 ± 38 2.1 ± 1.3 1.3 [0.40–8.8] 0.53 [0.35–2.0] 4.1 [1.4–6.0] 5.7 ± 1.5 2.0 [0.1–53.7] 22 (43)

69 ± 10 206 (69) 218 (73) 152 (51) 68 (23) 69 (23) 120 (40) 66 (22) 17 (6) 112 [47–434] 49 ± 12 114 ± 33 1.7 ± 1.1 2.0 [0.45–9.6] 0.65 [0.32–1.6] 4.4 [1.6–8.3] 5.1 ± 1.4 1.0 [0.1–112.3] 174 (55)

70 ± 10 82 (71) 94 (81) 66 (57) 20 (17) 14 (12) 50 (43) 27 (23) 7 (6) 119 [44–434] 48 ± 15 109 ± 33 1.7 ± 1.1 1.9 [0.54–9.6] 0.64 [0.36–1.6] 4.3 [1.8–8.3] 5.2 ± 1.4 1.5 [0.1–112.3] 81 (70)

0.018 0.90 0.49 0.31 0.003 0.20 0.12 0.13 0.66 0.95 0.018 0.11 0.12 <0.001 <0.001 0.005 0.013 0.002 0.26

(%)

pANOVA<0.001

(A) p < 0.001

25

p = 0.011 20

15

10

5 0 High (>2.48)

Intermediate (1.48-2.48)

Low (<1.48)

The Presence of Acute Coronary Syndrome

The Presence of Acute Coronary Syndrome

Values are the mean ± SD or number of patients or median [range]. ACS, acute coronary syndrome; PCI, percutaneous coronary intervention; CABG, coronary artery bypass graft; HDL, high-density lipoprotein; LDL, low-density lipoprotein; HOMA-IR, homeostasis model assessment of insulin resistance; EPA, eicosapentaenoic acid; DPA, docosapentaenoic acid; DHA, docosahexaenoic acid; AA, arachidonic acid.

(%)

p < 0.001

(B)

p < 0.001

25 20

15

10

5 0 High (>0.71)

The Presence of Acute Coronary Syndrome

EPA Tertile (%)

pANOVA<0.001

Intermediate (0.57-0.71)

Low (<0.57)

DPA Tertile pANOVA=ns

(C)

25 20

15

10

5

0 High (>4.79)

Intermediate (3.80-4.79)

Low (<3.80)

DHA Tertile Fig. 1. Presence of acute coronary syndrome in patients distributed by their serum levels of omega-3 polyunsaturated fatty acids. ACS was significantly associated with the lower tertile of EPA in A (p < 0.001) and of DPA in B (p < 0.001). EPA, eicosapentaenoic acid; DPA, docosapentaenoic acid; DHA, docosahexaenoic acid.

T. Amano et al. / Atherosclerosis 218 (2011) 110–116

113

Table 2 Results of quantitative coronary angiography and intravascular ultrasound examination. EPA, wt%

QCA analysis Reference diameter, mm Diameter stenosis, % Conventional IVUS analysis Remodeling index Vessel volume, mm3 Lumen volume, mm3 Plaque volume, mm3 % Plaque volume, % IB IVUS analysis Lipid volume, mm3 Fibrous volume, mm3 Calcified volume, mm3 Lipid volume, % Fibrous volume, % Calcified volume, %

p value High ≥1.9 (n = 58)

Low <1.9 (n = 58) 2.8 ± 0.5 41 ± 11

DPA, wt% Low <0.64 (n = 58)

2.8 ± 0.6 40 ± 11

0.76 0.39

2.8 ± 0.6 42 ± 11

p value High ≥0.64 (n = 58)

DHA, wt%

2.8 ± 0.5 39 ± 12

0.63 0.056

0.97 138.8 53.2 85.6 61.9

± ± ± ± ±

0.23 45.8 26.2 33.0 11.2

0.98 132.8 55.2 77.6 57.9

± ± ± ± ±

0.19 40.6 22.8 30.7 12.1

0.58 0.46 0.66 0.18 0.064

0.99 137.2 52.4 84.8 61.6

± ± ± ± ±

0.20 50.0 25.6 35.6 11.4

0.98 134.4 56.0 78.4 58.2

± ± ± ± ±

0.22 35.5 23.5 27.8 12.1

0.36 0.72 0.42 0.28 0.11

48.8 35.4 1.3 55.5 42.9 1.6

± ± ± ± ± ±

27.6 13.3 1.5 14.6 13.8 1.6

41.1 35.5 1.0 50.0 48.6 1.4

± ± ± ± ± ±

24.7 14.1 0.9 15.1 14.7 1.1

0.11 0.98 0.16 0.048 0.035 0.59

49.8 33.9 1.1 56.4 42.2 1.4

± ± ± ± ± ±

30.2 13.2 1.2 14.9 14.2 1.4

40.1 37.0 1.2 49.1 49.3 1.6

± ± ± ± ± ±

21.1 14.0 1.3 14.4 14.0 1.4

0.047 0.22 0.65 0.008 0.008 0.50

p value High ≥4.3 (n = 58)

Low <4.3 (n = 58) 2.8 ± 0.6 41 ± 11

2.7 ± 0.5 40 ± 12

0.47 0.72

0.98 141.1 54.8 86.3 61.5

± ± ± ± ±

0.20 45.2 27.0 33.0 12.2

0.98 130.5 53.6 76.9 58.3

± ± ± ± ±

0.18 40.8 21.9 30.5 11.3

0.53 0.19 0.81 0.11 0.16

49.6 35.4 1.2 55.7 42.9 1.4

± ± ± ± ± ±

28.4 13.8 1.4 15.3 14.6 1.4

40.3 35.5 1.1 49.8 48.6 1.5

± ± ± ± ± ±

23.6 13.6 1.1 14.3 13.9 1.4

0.059 0.97 0.77 0.036 0.034 0.63

Values are the mean ± SD. EPA, eicosapentaenoic acid; DPA, docosapentaenoic acid; DHA, docosahexaenoic acid; AA, arachidonic acid; QCA, quantitative coronary angiography; IVUS, intravascular ultrasound; IB, integrated backscatter.

thus, 336 were included in Analysis 1. As compared to patients without ACS, those with ACS had significantly lower serum EPA levels (median, 1.3 [range 0.40–8.8], wt% vs. 2.0 [0.45–9.6], wt%, p < 0.001), lower serum DPA levels (0.53 [0.35–2.0], vs. 0.65 [0.32–1.6], p < 0.001), lower serum DHA levels (4.1 [1.4–6.0], vs. 4.4 [1.6–8.3], p = 0.005), and lower serum high-density lipoprotein (HDL)-cholesterol levels. ACS patients had greater serum levels of arachidonic acid and C-reactive protein (CRP) compared to those without ACS. Smoking was more frequent in patients with ACS as compared to those without it. Fig. 1 shows the rate of ACS in each tertile of EPA (A), DPA (B), and DHA (C). ACS was significantly associated with lower levels of EPA (p < 0.001) and DPA (p < 0.001), while DHA did not have a significant relationship. On multivariate logistic regression analysis after adjusting for confounding factors (age, gender, smoking, CRP, and HDL-cholesterol), low EPA level (odds ratio 2.10, 95% confidence interval 1.29–3.42, p = 0.003) or low DPA level (odds ratio 2.15, 95% confidence interval 1.33–3.49, p = 0.0019) was independently associated with the presence of ACS, while low DHA level (odds ratio 1.23, 95% confidence interval 0.77–1.97, p = 0.38) did not have a significant relationship. 3.2. Reproducibility of IB-IVUS parameters The correlation of %LV and %FV measured by 2 physicians who conducted IB-IVUS measurements independently was r2 = 0.84 (p < 0.001) and r2 = 0.83 (p < 0.001), respectively. In order to validate the %LV measurements obtained by both physicians, we compared %LV measurements of 30 randomly selected lesions. In BlandAltman plots, the mean differences in %LV were 0.90 ± 9.8%. The limits of agreement for plaque volume were between −12.0% and 13.8%.

low DPA levels, and low DHA levels had a significantly higher %LV (55.5 ± 14.6% vs. 50.0 ± 15.1%, p = 0.048; 56.4 ± 14.9% vs. 49.1 ± 14.4%, p = 0.008; and 55.7 ± 15.3% vs. 49.8 ± 14.3%, p = 0.036) and a significantly lower %FV (42.9 ± 13.8% vs. 48.6 ± 14.7%, p = 0.035; 42.2 ± 14.2% vs. 49.3 ± 14.0%, p = 0.008; and 42.9 ± 14.6% vs. 48.6 ± 13.9%, p = 0.034) as compared to their counterparts with higher median values of these ␻3 PUFA s (Table 2). Table 3 shows the results of multivariate logistic regression analysis for prediction of LRPs. On simple logistic regression analysis, only low DPA content and low EPA content were significantly associated with LRPs in any of the two categories (on-line supplementary Table 2). When EPA and DPA were independently included in the multivariate model to assess multi-co-linearity between EPA and DPA (Model 1), neither of these two factors was associated with LRPs. However, in Model 2 even after adjustment for confounding factors (age, gender, and BMI) and low DHA content, the presence of both low EPA content and low DPA content increased the predictive value for LRPs (Category 1; odds ratio 4.28, 95% confidence interval 1.37–13.3, p = 0.012, and Category 2; odds ratio 7.39, 95% confidence interval 1.25–43.6, p = 0.027, respectively). When statin use was entered into the model as a dependent variable (Model 3), the presence of both low EPA content and low DPA content were again significantly and independently associated with LRPs (Category 1; odds ratio 3.94, 95% confidence interval 1.28–12.1, p = 0.017, and Category 2; odds ratio 7.05, 95% confidence interval 1.18–42.1, p = 0.032, respectively) (Table 3). Representative conventional and IB-IVUS images of LRPs or nonLRPs are shown in Fig. 2. The percentage of lipid and fibrous volumes were 63.3% and 35.8% in panel A, and 37.9% and 60.4% in panel B, respectively. 4. Discussion

3.3. Association of ␻3 PUFAs with tissue characteristics of coronary plaques On simple linear regression analysis, serum EPA levels (logtransformed), serum DPA levels (log-transformed), and serum DHA levels (log-transformed) were inversely correlated with %LV (r = −0.265, p = 0.004; r = −0.278, p = 0.003; and r = −0.223, p = 0.043) and positively correlated with %FV (r = 0.268, p = 0.004; r = 0.281, p = 0.002; and r = 0.219, p = 0.018) (on-line supplementary Table 1). Patients were then divided into two groups according to the median value of ␻3 PUFA. Those with low EPA levels,

Results from previous interventional and observational studies have provided evidence of the protective effects of ␻3 PUFA against cardiovascular disease [1–4]. In the present study, our patients with ACS had lower ␻3 PUFA levels than those without ACS. In addition, the presence of ACS was significantly associated with low levels of ␻3 PUFA (especially of EPA and DPA). ␻3 PUFAs have several protective mechanisms against adverse outcomes, such as reduced platelet aggregation [18,19], vasodilation [20,21], antiproliferation [22], and antiarrhythmic effects [9,10]. Regarding the association of ␻3 PUFA with atherosclerosis,

114

T. Amano et al. / Atherosclerosis 218 (2011) 110–116

Fig. 2. Representative images of conventional intravascular ultrasound and integrated backscatter color-coded maps of coronary artery plaques in patients with lipid-rich plaques and those with non lipid-rich plaques. The lipid and fibrous volumes were 63.3% and 35.8% in panel A, and 37.9% and 60.4% in panel B, respectively. Blue, lipid pool; green yellow, fibrous lesion; red, calcified lesion.

several studies have shown serum levels to correlate inversely with subclinical facets of atherosclerosis such as carotid intimamedia thickening [8,23]. Few studies, however, have examined the relevance of ␻3 PUFA serum levels regarding coronary plaques, especially for the plaque tissue components. In the present study, IB-IVUS analyses showed that ␻3 PUFAs correlated inversely with %LV and positively with %FV. Furthermore, the presence of both low EPA content and low DPA content proved to be an independent

predictor of LRPs even after adjustment for confounding factors. Measurement of lipid contents by IB-IVUS has proven useful in the diagnosis [11–14] and in the assessment of the prognosis of coronary atherosclerosis, as a risk for experiencing a coronary event [15,16]. However, its invasive nature does not allow IB-IVUS to assess the entire coronary tree, especially in the general population. Therefore, the relationship of ␻3 PUFAs with coronary tissue components measured by IB-IVUS may allow in the future a

T. Amano et al. / Atherosclerosis 218 (2011) 110–116

115

Table 3 Multiple logistic regression analysis for prediction of lipid-rich plaque. Multiple logistic regression Model 1 OR (95% CI)

Model 2 p value

Category 1; LRP (% lipid volume > 75th percentile and % fibrous volume < 25th percentile) 2.21 (0.68–7.22) 0.19 Low EPA < median (vs. ≥median) 2.52 (0.80–7.91) 0.11 Low DPA < median (vs. ≥median) Low DPA and low EPA – – Low DHA < median (vs. ≥median) 1.00 (0.30–3.28) 0.99 – – High HOMA-IR > median (vs. ≤median) – – Low HDL-cholesterol < 40 mg/dL Category 2; LRP (% lipid volume > 90th percentile and % fibrous volume < 10th percentile) 2.45 (0.40–15.0) 0.33 Low EPA < median (vs. ≥median) Low DPA < median (vs. ≥median) 2.60 (0.46–14.5) 0.28 Low DPA and low EPA – – Low DHA < median (vs. ≥median) 2.63 (0.41–16.8) 0.31 – – High HOMA-IR > median (vs. ≤median) – – Low HDL-cholesterol < 40 mg/dL

Model 3

OR (95% CI)

p value

OR (95% CI)

p value

– – 4.28 (1.37–13.3) 0.012 1.04 (0.33–3.33) 0.94 – –

– – 0.012 0.94 – –

– – 3.94 (1.28–12.1) 1.14 (0.36–3.60) – –

– – 0.017 0.82 – –

– – 7.39 (1.25–43.6) 0.027 2.12 (0.33–13.6) 0.43 – –

– – 0.027 0.43 – –

– – 7.05 (1.18–42.1) 2.10 (0.33–13.5) – –

– – 0.032 0.44 – –

After adjustment for confounding variables (age, gender, and body-mass index). Each listed variable was adjusted for all other variables in each model. OR, odds ratio; CI, confidential interval; EPA, eicosapentaenoic acid; DPA, docosapentaenoic acid; DHA, docosahexaenoic acid; HOMA-IR, homeostasis model assessment of insulin resistance; HDL, high-density lipoprotein.

clearer insight on risk stratification and a better evaluation of the therapeutic strategies. Recently, Cawood et al. examined ␻3 PUFA incorporation into subclinical atherosclerotic plaques and their association with plaque inflammation and stability, concluding that atherosclerotic plaques readily incorporated EPA although EPA and DHA were simultaneously administered, and a higher EPA content in the plaques was associated with a reduced number of foam cells and T cells, less inflammation and increased stability [24]. These findings showed that EPA was mainly incorporated in atherosclerotic plaques and could have beneficial effects on its stability, suggesting a different mechanism of EPA and DHA in terms of their protective effects against the development of arteriosclerosis. In the present study, coronary plaques in patients with low EPA and low DPA levels had a significantly higher %LV and a significantly lower %FV as compared to their high counterparts, with higher levels of EPA and DPA; however, EPA and DPA were not independently associated with LRPs. Although the underlying mechanism is not clear in our study, some plausible mechanisms can be suggested. DPA is an elongated metabolite of EPA and is an intermediary product between EPA and DHA. The serum DPA levels in patients after oral administration of pure EPA were elevated and correlated with serum EPA levels [25]. Actually, in the present study, serum EPA levels were significantly correlated with serum DPA levels (r = 0.662, p < 0.001). Although the literature on DPA is limited, the available data suggest that DPA is retro-converted back to EPA, however, it does not appear to be readily metabolized to DHA [26]. In addition, a previous study has addressed overlapping biological effects of EPA and DPA on endothelial cell migration and proliferation, which are important processes in the control of the wound-healing response of blood vessels. These data suggest that the stimulatory effect of EPA on endothelial cell migration occurs via DPA, and that EPA and DPA act together as an anti-atherogenic factor [27]. In the JELIS (Japan EPA Lipid Intervention Study) trial, 18,645 patients (14,981 in primary prevention and 3664 in secondary prevention) with hypercholesterolemia were randomized to receive statin alone or statin plus highly purified EPA (18.00 mg/day). At the end of 5-years, those randomized to EPA showed a 19% lower occurrence of major cardiovascular events [3]. On the other hand, Kromhout et al. have recently shown that low doses of a mixture of EPA and DHA or plant-derived ␣-linoleic acid did not reduce the occurrence of cardiovascular events in people that had suffered myocardial infarction [28]. These controversial results pose questions and challenges for the future use of these lipids in

cardiovascular therapy. Considering the results of present study, the beneficial effects of ␻3 PUFAs on atherosclerotic change might not be the same for EPA, DPA and DHA; thus, further confirmation of the therapeutic target and clinical implications is needed.

5. Limitations This study has several limitations. First, it was performed at a single-center and the study population was relatively small. Second, thrombus formation was not color coded and analyzed, thereby hindering rigorous calculation of plaque components in the target lesion, especially in patients with ACS. Third, we did not analyze the entire artery segment by IB-IVUS. However, it has been reported that vulnerable plaques clustered in the proximal coronary arteries; 90% of those are present within the first 30–33 mm of the left anterior descending and left circumflex arteries, and within the first 74 mm of the right coronary artery [29]. In the present study, IB-IVUS parameters were measured in the worst plaque within 35 mm from the ostium of the left anterior descending and left circumflex arteries, and in the worst plaque that could be observed in the right coronary artery by IVUS. Therefore, the LRPs analyzed here could portend possible plaque instability. Finally, the observational nature of the study does not allow us to make definitive conclusions regarding the beneficial effects of ␻3 PUFAs intake on coronary plaque stability. Our findings need further confirmation of any validated and clinical implications.

6. Conclusions ACS was significantly associated with lower levels of ␻3 PUFAs (especially of EPA and DPA). The presence of both low EPA content and low DPA content proved to be an independent predictor of LRPs measured by IB-IVUS. These findings might help explain the protective mechanisms by which ␻3 PUFAs, mainly of EPA and DPA, contribute to decrease the risk of atherosclerotic change.

Acknowledgments We thank Masaaki Soma, PhD for excellent assistance with the manuscript. There is neither a conflict of interest nor financial support in connection with the present study.

116

T. Amano et al. / Atherosclerosis 218 (2011) 110–116

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.atherosclerosis.2011.05.030.

[15]

[16]

References [17] [1] Kromhout D, Bosschieter EB, de Lezenne Coulander C. The inverse relation between fish consumption and 20-year mortality from coronary heart disease. N Engl J Med 1985;312:1205–9. [2] Albert CM, Hennekens CH, O’Donnell CJ, et al. Fish consumption and risk of sudden cardiac death. JAMA 1998;279:23–8. [3] Yokoyama M, Origasa H, Matsuzaki M, et al. Effects of eicosapentaenoic acid on major coronary events in hypercholesterolaemic patients (JELIS): a randomised open-label, blinded endpoint analysis. Lancet 2007;369:1090–8. [4] Rissanen T, Voutilainen S, Nyyssonen K, Lakka TA, Salonen JT. Fish oil-derived fatty acids, docosahexaenoic acid and docosapentaenoic acid, and the risk of acute coronary events: the Kuopio ischaemic heart disease risk factor study. Circulation 2000;102:2677–9. [5] Sekikawa A, Ueshima H, Kadowaki T, et al. Less subclinical atherosclerosis in Japanese men in Japan than in White men in the United States in the post-World War II birth cohort. Am J Epidemiol 2007;165:617–24. [6] Sekikawa A, Curb JD, Ueshima H, et al. Marine-derived n-3 fatty acids and atherosclerosis in Japanese, Japanese-American, and white men: a crosssectional study. J Am Coll Cardiol 2008;52:417–24. [7] GISSI-HF Investigators, et al. Effect of n-3 polyunsaturated fatty acids in patients with chronic heart failure (the GISSI-HF trial): a randomised, double-blind, placebo-controlled trial. Lancet 2008;372:1223–30. [8] Yamagishi K, Iso H, Date C, et al. Fish, omega-3 polyunsaturated fatty acids, and mortality from cardiovascular diseases in a nationwide community-based cohort of Japanese men and women the JACC (Japan Collaborative Cohort Study for Evaluation of Cancer Risk) Study. J Am Coll Cardiol 2008;52:988–96. [9] Anand RG, Alkadri M, Lavie CJ, Milani RV. The role of fish oil in arrhythmia prevention. J Cardiopulm Rehabil Prev 2008;28:92–8. [10] Leaf A, Albert CM, Josephson M, et al. Prevention of fatal arrhythmias in highrisk subjects by fish oil n-3 fatty acid intake. Circulation 2005;112:2762–8. [11] Kawasaki M, Takatsu H, Noda T, et al. In vivo quantitative tissue characterization of human coronary arterial plaques by use of integrated backscatter intravascular ultrasound and comparison with angioscopic findings. Circulation 2002;105:2487–92. [12] Okubo M, Kawasaki M, Ishihara Y, et al. Tissue characterization of coronary plaques: comparison of integrated backscatter intravascular ultrasound with virtual histology intravascular ultrasound. Circ J 2008;72:1631–9. [13] Okubo M, Kawasaki M, Ishihara Y, et al. Development of integrated backscatter intravascular ultrasound for tissue characterization of coronary plaques. Ultrasound Med Biol 2008;34:655–63. [14] Amano T, Matsubara T, Uetani T, et al. Impact of metabolic syndrome on tissue characteristics of angiographically mild to moderate coronary lesions:

[18]

[19] [20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

integrated backscatter intravascular ultrasound study. J Am Coll Cardiol 2007;49:1149–56. Sano K, Kawasaki M, Ishihara Y, et al. Assessment of vulnerable plaques causing acute coronary syndrome using integrated backscatter intravascular ultrasound. J Am Coll Cardiol 2006;47:734–41. Amano T, Matsubara T, Uetani T, et al. Lipid-rich plaques predict nontarget-lesion ischemic events in patients undergoing percutaneous coronary intervention. Circ J 2010;75:157–66. Kawasaki M, Hattori A, Ishihara Y, et al. Tissue characterization of coronary plaques and assessment of thickness of fibrous cap using integrated backscatter intravascular ultrasound: comparison with histology and optical coherence tomography. Circ J 2010;74:2641–8. Hirai A, Terano T, Hamazaki T, et al. The effects of the oral administration of fish oil concentrate on the release and the metabolism of [14C] arachidonic acid and [14C] eicosapentaenoic acid by human platelets. Thrombus Res 1982;28:285–98. Tamura Y, Hirai A, Terano T, et al. Clinical and epidemiological studies of eicosapentaenoic acid (EPA) in Japan. Prog Lipid Res 1986;25:461–6. Hamazaki T, Hirai A, Terano T, et al. Effects of orally administered ethyl ester of eicosapentaenoic acid (EPA; C20:5, omega-3) on PGI2-like substance production by rat aorta. Prostaglandins 1982;23:557–67. Okuda Y, Kawashima K, Sawada T, et al. Eicosapentaenoic acid enhances nitric oxide production by cultured human endothelial cells. Biochem Biophys Res Commun 1997;232:487–91. Terano T, Shiina T, Tamura Y. Eicosapentaenoic acid suppressed the proliferation of vascular smooth muscle cells through modulation of various steps of growth signals. Lipids 1996;31:301–4. Thies F, Garry JM, Yaqoob P, et al. Association of n-3 polyunsaturated fatty acids with stability of atherosclerotic plaques: a randomised controlled trial. Lancet 2003;361:477–85. Cawood AL, Ding R, Napper FL, et al. Eicosapentaenoic acid (EPA) from highly concentrated n-3 fatty acid ethyl esters is incorporated into advanced atherosclerotic plaques and higher plaque EPA is associated with decreased plaque inflammation and increased stability. Atherosclerosis 2010;212: 252–9. Okuda Y, Mizutani M, Ogawa M, et al. Long-term effects of eicosapentaenoic acid on diabetic peripheral neuropathy and serum lipids in patients with type II diabetes mellitus. J Diabetes Complications 1996;10:280–7. Kaur G, Cameron-Smith D, Garg M, Sinclair AJ. Docosapentaenoic acid (22:5n-3): a review of its biological effects. Prog Lipid Res 2011;50: 28–34. Kanayasu-Toyoda T, Morita I, Murota S. Docosapentaenoic acid (22:5, n-3) is a potent stimulator of endothelial cell migration on pretreatment in vitro. Prostaglandins Leukot Essent Fatty Acids 1996;54:319–25. Kromhout D, Giltay EJ, Geleijnse JM, Alpha Omega Trial Group. n-3 fatty acids cardiovascular events after myocardial infarction. N Engl J Med 2010;363:2015–26. Cheruvu PK, Finn AV, Gardner C, et al. Frequency and distribution of thin-cap fibroatheroma and ruptured plaques in human coronary arteries: a pathologic study. J Am Coll Cardiol 2007;50:940–9.