In the past decade, moderately elevated homocysteine concentration has achieved wide-spread recognition as an independent risk factor for vascular diseases, such as stroke and peripheral vascular disease, as well as for an impaired nutritional status. In general, EDTA plasma is used for the determination of homocysteine. However, from the pre-analytical point of view it is important, that, when plasma is not separated from blood cells within 30 minutes, homocysteine levels increase in samples significantly by about 10% per hour. This 10% increase is very important, because the normal range is between 5 and 15 micromol/l and moderately elevated homocysteine concentrations above 15 micromol/l may signify an increased risk of vascular disease. These preliminary cut-off points show that there is only a small difference between normal and moderately elevated homocysteine concentrations. Most blood samples are obtained outside the hospital, and in these cases homocysteine concentrations will be falsely elevated, if no precautions are taken, such as immediate centrifugation and separation of plasma and cells. This aspect is critical both for clinical studies and in patient care outside the hospital. But even in the hospital it is difficult to separate plasma and cells within 30 minutes. In the past, different approaches were adopted to solve this problem. Potential stabilisers were sodium fluoride (4 g/l) and 3-deazaadenosine (100 micromol/l). Sodium fluoride initially increased the homocysteine concentration, which dropped below the initial values after 72 h. On the other hand, 3-deazaadenosine stabilised homocysteine concentrations for 24 h, but increased it within 72 h by roughly 10%. However, this stabiliser is restricted to HPLC technology but does not work reliably with immunoassays. Lysis of blood stabilised homocysteine, but homocysteine concentrations were systematically lower requiring totally new reference ranges. In addition, acidic citrate (0.5 mol/l) was evaluated, which seems to stabilise plasma homocysteine concentrations at ambient temperatures for several hours. However, small but systematic deviations at baseline are observed. This stabilisation procedure does not interfere with immunoassays. Because immunoassays will be the future method of choice for robust and easy to perform homocysteine measurements, because they easily allow the analyses of high sample numbers, homocysteine stabilisation in whole blood is still an important matter. It must be solved before homocysteine determinations are introduced as a general screening for vascular risk factors in non-specialist laboratories.