Excessive albumin losses during HC (haemocatharsis) are considered a potential cause of hypoalbuminemia—a key risk factor for mortality. This review on total albumin losses considers albumin “leaking” into the dialysate and losses due to protein/membrane interactions (i.e. adsorption, “secondary membrane formation” and denaturation). The former are fairly easy to determine, usually varying at the level of ~ 2 g to ~ 7 g albumin loss per session. Such values, commonly accepted as representative of the total albumin losses, are often quoted as limits/standards of permissible albumin loss per session. On albumin mass lost due to adsorption/deposition, which is the result of complicated interactions and rather difficult to determine, scant in vivo data exist and there is great uncertainty and confusion regarding their magnitude; this is possibly responsible for neglecting their contribution to the total losses at present. Yet, many relevant in vitro studies suggest that losses of albumin due to protein/membrane interactions are likely comparable to (or even greater than) those due to leaking, particularly in the currently favoured high-convection HDF (haemodiafiltration) treatment. Therefore, it is emphasised that top research priority should be given to resolve these issues, primarily by developing appropriate/facile in vivo test-methods and related analytical techniques.