Vascular composition data supporting the role of N-3 polyunsaturated fatty acids in the prevention of cardiovascular disease events

N-3 polyunsaturated fatty acids (PUFAs) are thought to have protective effects against cardiovascular disease. Here, we report the relationship between serum PUFA concentrations and plaque composition, as evaluated by virtual histology-intravascular ultrasound (VH-IVUS). Consecutive patients (n=61) who underwent percutaneous coronary intervention (PCI) were pre-operatively examined using VH-IVUS to assess the composition of culprit plaques. Gray-scale IVUS and VH-IVUS data of fibrous, fibro-fatty, necrotic core, and dense calcium regions of plaques were estimated at the minimal luminal area sites of culprit lesions. Serum levels of high-sensitivity C-reactive protein (hsCRP) and PUFAs, including eicosapentaenoic acid (EPA), docosahexaenoic acid (DHA), and arachidonic acid (AA), were compared between patients with (ACS, n=27) and without acute coronary syndrome (non-ACS, n=34) before PCI. Multiple logistic regression analysis of the data showed that EPA/AA under the median was more highly associated with ACS than hsCRP over the median. In addition, EPA/AA was negatively correlated with the percentage of fibrous plaque regions and EPA/AA and DHA/AA were positively correlated with the percentage of dense calcium regions in plaques. Furthermore, the correlation index of EPA/AA was the most highly (R=0.513) correlated with the percentage of dense calcium regions in plaques.

The data provides information on the effect of PUFAs on the coronary vascular composition of cardiovascular disease patients.
This data is valuable for further clinical examination of PUFAs and experimental vascular models. The analysis of plaque composition based on the gray-scale IVUS and VH-IVUS data is expected to be useful for assessing how PUFAs or other substances impact coronary vascular composition.
The data may be potentially valuable to other researchers examining the relationship between PUFAs and vascular calcification.

Data
The serum levels of eicosapentaenoic acid (EPA), EPA/arachidonic acid (AA), docosahexaenoic acid (DHA), and DHA/AA in acute coronary syndrome (ACS) patients were lower than those in stable coronary disease (SCD) patients (Table 1) (Fig. 1). In addition, the vessel and plaque areas in ACS patients were larger than those in SCD patients (Fig. 2); however, the percentage of fibrous, fibrofatty, necrotic core, and dense calcium regions at sites of minimal luminal area within culprit lesions were comparable between the two patient groups ( Table 2, Fig. 3), although the percentage of dense calcium regions in plaques tended to be higher in SCD patients than ACS patients ( Table 2, Fig. 3). Univariate and multiple logistic regression analyses revealed that low DHA, low EPA, low DHA/AA, and low EPA/AA were independent factors for the risk of ACS (Tables 3 and 4). The percentage of dense calcium regions in plaques positively correlated with EPA/AA (Table 5, Fig. 5), EPA (Table 5, Fig. 6), and DHA (Table 5, Fig. 7), and EPA/AA was also inversely correlated with %plaque burden (ratio of plaque area to vessel area) ( Table 5, Fig. 4).

Study population
This study was approved by the Fukushima Red Cross Hospital Ethics Committee (April 24, 2012), and written informed consent was obtained from all study subjects before enrollment. The study Table 1 Comparisons of patient characteristics and laboratory data between ACS and SCD patients.

IVUS image acquisition and analysis
Intravascular ultrasound was used to examine culprit lesions. A phased-array, 20-MHz, 3.2-F IVUS catheter (Eagle Eye, Volcano Corp., Rancho Cordova, CA) was placed into the distal coronary artery and pulled back to the aorto-ostial junction using a motorized catheter pull-back system set at 0.5 mm/s (Eagle Eye, Volcano Corp.). The gray-scale IVUS and captured radiofrequency (VH-IVUS) data were analyzed using echoPlaque 4.0 software (INDEC Systems, Inc., City, State). Corresponding images of IVUS examinations were identified at culprit lesions between segments. The gray-scale IVUS and VH-IVUS images were analyzed at sites of minimal luminal area within culprit lesions, as previously described [4]. Gray-scale IVUS analysis was performed according to the American College of Cardiology Clinical Expert Consensus Document on Standards for Acquisition, Measurement and Reporting of Intravascular Ultrasound Studies [5]. In the conventional gray-scale IVUS analysis, crosssectional images were quantified for luminal diameter, vessel diameter, intima diameter, luminal area, vessel area and plaque area. Plaque burden was calculated as the ratio of plaque area to vessel area. The remodeling index was calculated as the ratio of vessel area at the site of the measured lesion (sites of minimal luminal area) to the reference vessel area (average of the proximal and distal reference segments). The four VH-IVUS plaque components were color-coded as follows: dark green (fibrous), light green (fibro fatty), red (necrotic core), and white (dense calcium), and are reported as the area or percentage of plaque area. Representative images of lesions from ACS (low EPA/AA ratio; 0.29) and SCD patients (high EPA/AA ratio; 0.82) are shown in Fig. 1. The intraobserver (r¼ 0.98, 0.98 and 0.99) and interobserver (r ¼0.96, 0.97, and 0.98) variability for the measurements were determined to be acceptable (Table 2).

Laboratory analyses
Blood examinations for lipid levels were performed within 2 days prior to PCI. Fasting blood samples were collected and after centrifugation and prompt freezing at À 80°C, serum samples were shipped to SRL, Inc. (Tokyo, Japan) for the measurement of dihomo-gamma-linolenic acid (DGLA), EPA, DHA, and AA levels using a gas chromatography method [6]. Plasma concentrations of hsCRP were quantified using the nephelometry method [7].      Table 5.

Statistical analysis
Data are presented as the mean 7standard deviation. Categorical data are presented as number (n) and percentage (%). Categorical and continuous variables were compared between the ACS and SCD groups using the chi-square test and Student's unpaired t-test, respectively (Tables 1 and 2, Figs. 2 and 3). Risk factors of ACS were identified using univariable and multivariable logistic regression analysis. All continuous data were categorized because linearity on the logit-scale could not be achieved with continuous covariables. The factors showing P o0.2 in the univariable test were entered into a multivariable logistic analysis. Significant prognostic factors were selected with a forward selection strategy using the likelihood ratio statistic (Table 3). In addition, multiple logistic analyses were performed using two models to avoid multicollinearity (Table 4). All statistical assessments were two-sided and evaluated at the 0.05 level of significance. Correlations of the EPA, EPA/AA, DHA, and DHA/AA data between the gray-scale and VH-IVUS data are shown in Table 5 and Figs. 4-7. All statistical analyses were performed using SPSS 19.0 for Windows (SPSS Inc., Chicago, IL).

Limitations
Two limitations of this data warrant mention. First, the data was obtained at a single center from a relatively small number of patients. Data from a larger number of patients and multiple centers are     Table 5. Fig. 6. Correlation between EPA and %DC. EPA was significantly correlated with %DC (R¼ 0.334, P ¼0.01). This figure is based on the data presented in Table 5.
needed to verify the correlations detected here. Second, the relationship between plaque calcification and PUFAs could not be determined from the data. Further experimental examination will be needed to determine the factors underlying this relationship.  Table 5.