There has been controversial evidence regarding the relationship between isomers of circulating trans-fatty acids (TFAs) and mortality. This study aimed to ascertain the relationships between plasma TFAs and overall or cause-specific mortality of the general population in two independent subsets from the US National Health and Nutrition Examination Survey (1999-2000 and 2009-2010 cycles). Plasma TFA isomers (C16:1n-7t, C18:1n-7t, C18:1n-9t and C18:2n-6,9t) in 3439 adults free of cancer or severe cardiovascular disease were analyzed by gas chromatography/mass spectrometry. Overall, 259 died among 1376 individuals over a median follow-up of 15.6 years in the 1999-2000 cycle, and 105 died in the latter subset of 2063 subjects during a median of 5.9 years. Cox proportional hazards regression was conducted to estimate the hazard ratios of mortality. The main isomer of industrially derived TFAs,elaidic acid (C18:1n-9t) was considerably associated with long-term total mortality in the 1999-2000 cycle after adjusting for confounders, with a 54% increase in the top tertile compared with the bottom one. However, the association disappeared with halving C18:1n-9t by 2009-2010. In contrast, neither of the ruminant-derived TFAs (C16:1n-7t and C18:1n-7t) suggested any inverse correlations with all-cause death, mortality due to heart disease, cancer or other causes. The major isomer of industrial TFAs, the higher circulating C18:1n-9t might be associated with increased long-term mortality. The associations with death risk turned slight with the reduction of TFAs consumption by half. However, dietary guidelines should rigorously identify the healthy effect of animal TFAs consumption.