Background: The syndrome of inappropriate antidiuresis (SIAD) is the most frequent cause of hyponatremia. Its diagnosis requires a decreased serum osmolality, urinary osmolality >200 mosm/kg, clinical euvolemia, and urinary sodium (UNa) >30 mmol/l. The natriuretic effect of diuretics impairs the diagnostic accuracy of UNa and thus, limits the diagnostic and therapeutic decision in hyponatremic patients substantially. We therefore examined the accuracy to predict SIAD of alternative markers, thought not to interact with diuretics.
Methods: In a prospective study 86 consecutive hyponatremic patients (serum sodium <130 mmol/l) were classified according to their history, clinical evaluation, and saline response (isotonic saline) into a SIAD and a non-SIAD group. In both groups the following biochemical parameters were tested in a diuretic collective (DC) and in a non-diuretic collective (NDC): UNa, serum uric acid concentration, fractional excretion of sodium, urea, and uric acid (FEUA). The parameters accuracy to predict SIAD was assessed with a receiver-operating-charcateristic curve analysis.
Results: 31 patients were diagnosed SIAD (36%), 53 patients were classified non-SIAD (64%), the mean age was 66±15 years, 57 patients were on diuretics (68%). In NDC, UNa was the most accurate test parameter to predict SIAD (AUC 0.96), however, showed a significant loss of accuracy in DC (AUC 0.80; P<0.01 vs NDC). In DC, FEUA had a significant better performance, compared with all of the alternative parameters (AUC 0.96; all P<0.01) and showed a positive predictive value of 100% choosing a cutoff value of 12%.
Conclusion: In patients not taking diuretics, UNa is a sufficient test parameter to predict SIAD. In patients on diuretics, FEUA is the most accurate predictor of SIAD and shows a discriminative quality, similar to UNa, in patients not taking diuretics.