Background: Subclinical cobalamin deficiency is common in the elderly, but the sensitivity and specificity of serum total cobalamin for this diagnosis is poor. Serum holotranscobalamin (holoTC), a measure of biologically available cobalamin, is considered a better marker for early cobalamin depletion than total cobalamin. However, in elderly populations, healthrelated reference intervals for holoTC and correlations to renal function are not entirely clear. Methods: HoloTC was determined with an automated microparticle enzyme immunoassay (AxSYM Õ ) in 790 elderly non-vitamin-supplemented Swedish men, median age 75.3 years. Renal function was assessed with creatinine, cystatin C and estimated glomerular filtration rate (eGFR calculated from creatinine). Results: Median holoTC was 51.8 pmol/L, the health-related reference interval 19.6-132.3 pmol/L. There was no significant difference in mean holoTC in probands with normal compared to high creatinine (P ¼ 0.80) and cystatin C (P ¼ 0.82). No significant differences between the quartiles of creatinine or cystatin C in mean of log holoTC were seen. HoloTC correlated strongly with total cobalamin (r ¼ 0.69, P < 0.001), weaker with eGFR creatinine (r ¼ À0.09, P < 0.05) and creatinine (r ¼ 0.09, P < 0.05), the latter correlation was only seen in subjects with creatinine <100 mmol/L. HoloTC correlated negatively with plasma total homocysteine (r ¼ À0.24, P < 0.001), but not with cystatin C and age. Conclusions: Serum holoTC in healthy elderly men shows the same distribution as earlier described for a younger reference population. In this group of elderly subjects, holoTC did not correlate to reduced renal function. Thus, holoTC appears to be a promising tool for evaluating cobalamin status also in elderly populations.