Abstract 855: Long-term Intra-individual Variability of Triglycerides Predicts the Risk of CHD Irrespective of Triglyceride Levels
There is strong evidence that triglyceride levels (TG) are highly variable between and within individuals, even if standardization of blood sampling procedures and laboratory techniques are used. Few studies have examined the role of intra-individual variability per se as a CHD risk factor. Positive associations have been reported between the variability of total and LDL cholesterol and CVD events. While TG is the most variable lipid, our knowledge of the clinical consequences of this variability is scarce. We examined the effect of long-term intra-individual variability of fasting TG levels on the risk of clinical CHD, irrespective of the subject’s TG level. Since we did not have prior knowledge about the nature of the relationship between TG variability and CHD, we investigated this relationship using generalized additive models (GAM) that relaxes the assumption of linearity and allows more flexibility in exploring the relation between TG variability (measured by intrapersonal standard deviation of log TG) and CHD. Study subjects (n=1355, 51% Hispanic) from the San Luis Valley Health Studies, ages 20 –74 at baseline, were followed for up to 16 years (median=10 ys). Most subjects (60%) had 3 TG measurements (range: 2– 6), with a mean (SD) time interval between measurements = 5 years (2.25). There were 137 incident CHD events. In univariate analysis, TG variability was significantly related to CHD (P=0.03) and this relationship was non-linear. Subjects with moderate TG variability (SD log TG 0.2– 0.8) had higher CHD risk compared to those with more extreme variability. In multivariate analysis, the same non-linear pattern persisted, and the nonlinear term for TG variability remained significant (p=0.002) after adjusting for mean TG (p=0.02), other lipids, lipid lowering therapy, HDL-C variability, and other CHD risk factors. In conclusion, the biologic variability of TG significantly predicts CHD risk in a non-linear way, independent of TG level and other CHD risk factors. Further studies are needed to replicate our findings, and to explain the mechanisms that relate TG variability to CHD risk. In addition, repeated TG measurements are required to assign CHD risk accurately, not only based on TG levels, but also based on TG variability.