In order to assess the role of HDL on longevity, we studied HDL subfraction distribution in centenarian women compared with a group of weight- and gender-matched healthy normolipidemic controls. We did not find any significant difference in the mean plasma lipid, apolipoprotein, and Lp(a) levels. On the contrary, in spite of similar HDL-cholesterol concentrations (1.32 ± 0.41 mmol/l in centenarians vs. 1.32 ± 0.25 mmol/l in controls, p = not significant), HDL2b and HDL3a levels were, respectively, significantly increased and significantly reduced in centenarians in comparison with controls (HDL2b 32.4 ± 9.2% in centenarians vs. 23.4 ± 7.7% in controls, p < 0.002, and HDL3a 26.3 ± 9.8% in centenarians vs. 34.1 ± 7.3% in controls, p < 0.01). Moreover, HDL2b levels were significantly raised and HDL3a levels were significantly reduced in centenarians in comparison with both ‘middle-aged’ and ‘elderly’ subjects, whereas no difference for any HDL subfraction was found between the two groups of controls of different ages. Age was significantly correlated with HDL2b and HDL3a (respectively, +0.452, p < 0.001, and –0.370, p < 0.01) in all subjects, but not with all the other lipid, lipoprotein and apolipoprotein parameters, but we observed a large overlapping of individual values of HDL2b between centenarians and controls. Since HDL2b levels were found to be inversely correlated with coronary heart disease risk, we could speculate that, in some cases, this may probably favor a healthy ageing, but long-term longitudinal studies are necessary to define the relative importance of HDL subfractions distribution as a marker of longevity. Probably other factors or clinical characteristics play a major role in the ageing process.