Urea Kt/V, calculated according to the variable volume single pool urea kinetic model (UKM), has been accepted as the yardstick reflecting the adequacy of haemodialysis therapy. However, the classical algorithm of UKM requires great care in dialyser urea clearance (K) measurement in order to avoid major inaccuracies in estimating the urea distribution volume (V). Thus, we suggest a modified algorithm of UKM which avoids the measurement of K. It assumes an arbitrary V value and then calculates kinetically K as a function of the assumed V value. The rationale of the modified algorithm can be derived from the knowledge that the classical algorithm imposes a proportionality ration between K and V: given a particular set of data, a change in the attributed value of K leads to a proportional change in the calculated V value, so that the ratio K/V remains nearly constant. Aims of the study were (1) to validate the modified algorithm by comparing the resulting Kt/V and normalised protein catabolic rate (NPCR) values with the homologous ones obtained using the classical algorithm in a group of 33 patients on thrice-weekly haemodialysis; plasma water urea concentrations were used with the classical algorithm (CApw) and the modified algorithm (MApw); and (2) to verify the possibility of using plasma urea concentrations with the modified algorithm (MAp) instead of the more rigorous plasma water concentrations. NPCR (g/kg per day) was 1.33 +/- 0.05 in CApw, 1.29 +/- 0.05 in MApw and 1.28 +/- 0.04 in MAp. Kt/V was 1.27 +/- 0.03 in CApw, 1.25 +/- 0.03 in MApw and 1.26 +/- 0.03 in MAp.(ABSTRACT TRUNCATED AT 250 WORDS)