Several nonconsecutive 24-h urinary collections are considered the gold standard for estimating dietary salt intake. As those samples are logistically demanding, we aimed to describe the variability of 24-h sodium urinary excretion over consecutive days and report its adequacy with sodium intake.
We enrolled 16 healthy male volunteers in a prospective controlled study. All participants randomly received a low salt diet (LSD) (3 g/day of NaCl), a normal salt diet (NSD) (6 g/day of NaCl), and a high salt diet (HSD) (15 g/day of NaCl) for 7 days in a crossover design without wash-out period.
On day 6, median sodium urinary excretion was 258 (216-338), 10 (8-18), and 87 (69-121) mmol/day for HSD, LSD, and NSD, respectively (P < .001). When considering days 4-6, sodium urinary excretion was in steady state as models with and without interaction term "diet type X sample day" were not significantly different (P = .163). On day 6, area under the curve (AUC) of receiver operating characteristic for urinary sodium excretion to detect HSD was 1.0 (1.0-1.0) and a cut-point of 175 mmol/day was 100% sensitive and specific to detect HSD. On day 6, receiver operating characteristic AUC to detect LSD was 0.993 (0.978-1.0) and a cut-point of 53 mmol/day was 96.4% sensitive and 100% specific to detect LSD.
A steady state of sodium balance, where sodium intake is proportional to its excretion, is reached within a few days under a constant diet in the real-life setting. Categorization of salt consumption into low (3 g/day), normal (6 g/day), or high (15 g/day) based on a single 24-h urine collection is nearly perfect. Based on these results, repeated nonconsecutive urine collection might prove unnecessary to estimate sodium intake in daily clinical practice provided that diet is rather constant over time.