Data on gender and subgroup specific analyses of omega-3 fatty acids in the Ludwigshafen Risk and Cardiovascular Health Study

This paper contains additional data related to the research article “Omega-3 fatty acids and mortality in patients referred for coronary angiography – The Ludwigshafen Risk and Cardiovascular Health Study” (Kleber et al., in press) [1]. The data shows characteristics of the Ludwigshafen Risk and Cardiovascular Health (LURIC) study according to tertiles of omega-3 fatty acids as well as stratified by gender. The association of proportions of omega-3 fatty acids measured in erythrocyte membranes with different causes of death is investigated with a special focus on modeling the association of EPA with mortality in a nonlinear way. Further, the association of omega-3 fatty acids with all-cause mortality adjusted for high-sensitive C-reactive protein as a marker of systemic inflammation is examined as well as the association of EPA with cause-specific death.

lity adjusted for high-sensitive C-reactive protein as a marker of systemic inflammation is examined as well as the association of EPA with cause-specific death.
& Fasting blood samples were obtained by venipuncture at study entry. Fatty acid methyl esters were generated from erythrocytes that had been stored at À 80°C by acid transesterification. Experimental features Erythrocyte fatty acid composition was analyzed according to the HS-Omega-3 Index technology [2]. Fatty acid methyl esters were analyzed by gas chromatography using a GC2010 gas chromatograph (Shimadzu, Duisburg, Germany) equipped with a 100-m SP2560 column (Supelco, Bellefonte, PA) and using hydrogen as carrier gas. Fatty acids were identified by comparison with a standard mixture of fatty acids characteristic of erythrocytes. Results are presented as a percentage of total identified fatty acids after response factor correction. Data source location Ludwigshafen Heart Center in South-West Germany Data accessibility Data is within this article.

Value of the data
Gender-stratified analysis allows the identification of gender-specific effects of omega-3 fatty acids. Examination of nonlinear effects on mortality risk is important to define safe reference values. Investigation of cause-specific mortality might provide hints to possible pathways affected by omega-3 fatty acids.

Data
The data presented in this paper includes Figures and Tables that show the results of gender and subgroup stratified distribution of omega-3 fatty acids proportions in the LURIC study as well as multivariate adjusted analyses of their association with all-cause and cause-specific mortality that extend the results reported in [1] (Figs. 1-7 and Tables 1-6).

Subjects
The LURIC study consists of 3316 Caucasians with an indication for coronary angiography that were between 1997 and 2000 at the Ludwigshafen Heart Center in South-West Germany. A detailed description of the study can be found in Winkelmann et al. [3]. The study was approved by the 'Landesärztekammer' Ethics Committee of the Rheinland-Pfalz state in Germany. Informed written consent was obtained from all participants.

Laboratory procedures
The fatty acid composition of erythrocyte membranes was analyzed using the HS-Omega-3 Index s methodology as described in [1,2]. Results are given as a percentage of total identified fatty acids after response factor correction.

Statistical analyses
We present the mean and the standard deviation of continuous data when normally distributed and the median and 25th and 75th percentile for variables with a skewed distribution. Categorical data are presented as percentages. ANOVA was used to compare continuous variables between groups (variables with a skewed distribution were log-transformed before entering analysis) and the chisquare test was used for categorical variables. The association of omega-3 fatty acid levels in tertiles     or as Z-transformed values with mortality was assessed by Cox proportional hazard regression with the same adjustments as in [1]. Examination of scaled Schoenfeld residuals provided no evidence for a violation of the proportional hazard assumption. All tests were two-sided and a p value o0.05 was considered statistically significant. SPSS v22.0 (IBM, Ehningen, Germany) and R v3.2.3 (http://www.rproject.org) were used for all analyses. The R-package 'rms' was used for the generation of hazard ratio plots.  In a Cox regression model including age, sex, BMI, LDL-C, HDL-C, TG, smoking, alcohol intake, diabetes mellitus, hypertension, physical exercise, lipid lowering therapy and EPA, the EPA proportion was modeled as restricted cubic spline with three knots and plotted against the log hazard. EPA: eicosapentaenoic acid.