It has been widely accepted that there is little use for saline treatment in the syndrome of inappropriate secretion of ADH (SIADH). However, having observed that most SIADH patients increased their plasma sodium (PNa) after 2 l isotonic saline over 24 h, we investigated whether urine osmolality or the sum of urinary sodium and potassium (UNa + K) predicted this response, in 17 consecutive patients with chronic SIADH. The initial measure of urinary sodium plus potassium (UNa + K t0) was weakly correlated to the change in PNa (DPNa) after infusion (r = -0.51; p < 0.05), while initial urine osmolality (UOSM t0) was a much better predictor (y = -0.024x + 12.90; r = -0.81; p < 0.001). The lack of predictive value for UNa + K t0 was probably because urine electrolyte concentrations were not maximal for the corresponding initial UOSM. This reflects differences in salt intake between the patients. The theoretical maximal value for UNa + K t0 (th max UNa + K t0) for a given USOM t0, was as good a predictor as UOSM t0 (th max UNa + K vs. DPNa: r = -0.81; p < 0.001). A theoretical model describing the effect of 2 l isotonic saline infusion on DPNa as a function of UNa + K, produced values comparable to those observed in our patients. Only 6/17 patients, those with UOSM > 530 mOsm/kg, had their hyponatraemia aggravated by 2 l isotonic saline. Many SIADH patients have lower UOSM; in most such patients, 2 l of isotonic saline will improve PNa.