The associations of different lipid measures and related indices with incident hypertension during a median follow-up of 12·89 years were examined. Fasting levels of total cholesterol (TC), TAG, HDL-cholesterol and LDL-cholesterol, and related indices (TC/HDL-cholesterol and TAG/HDL-cholesterol) were determined in 7335 Iranian adults (men = 3270) free of hypertension, aged 39·0 (sd 13·2) years. Multivariate Cox proportional hazard regression was applied and lipid parameters were considered either as categorical or continuous variables. During follow-up, 2413 (men = 1126) participants experienced hypertension. Using the first quartile as reference, significant trends were found between quartiles of TAG, HDL-cholesterol, TC/HDL-cholesterol and TAG/HDL-cholesterol in multivariate models; moreover considering these measures as continuous variables, a 1 sd increase in each of these parameters was significantly associated with the risk of incident hypertension; the corresponding hazard ratios and CI were 1·06 (95 % CI 1·02, 1·10), 0·94 (95 % CI 0·89, 0·98), 1·04 (95 % CI 1·01, 1·09) and 1·04 (95 % CI 1·01, 1·07), respectively. The associations between lipid measures and incident hypertension did not change after excluding lipid lowering drug users and those with type 2 diabetes mellitus and these associations were independent of the categories of baseline blood pressure (P for interaction > 0·08). These findings indicate that TAG, TAG/HDL-cholesterol and TC/HDL-cholesterol were independently associated with higher risk, while HDL-cholesterol was associated with lower risk of incident hypertension.