Clinical significance of urinary sodium measurement for predicting pulmonary decongestion: a prospective observational study

Diuretic therapy based on the urinary sodium concentration (UNa) is recommended for patients with acute decompensated heart failure (HF). This prospective study evaluated the clinical impact of the UNa on pulmonary decongestion. A total of 450 patients diagnosed with acute decompensated HF who required intravenous furosemide were enrolled. Diuretic therapy was guided by an algorithm based on the UNa measured 1 h after intravenous furosemide administration. Radiographic signs of congestion were compared between patients with a UNa of < 70 mEq/L (n = 42) and patients with a UNa of ≥ 70 mEq/L (n = 408) using the congestion score index (CSI) measured on days 1 and 3. The change in CSI was smaller in the UNa < 70 mEq/L group than in the UNa ≥ 70 mEq/L group. And the median improvement in CSI in the UNa < 70 mEq/L group was significantly smaller than that in the UNa ≥ 70 mEq/L group. Multivariable logistic regression analysis demonstrated that the serum sodium and C-reactive protein levels and preadmission beta-blocker use were significantly associated with a UNa of < 70 mEq/L. An analysis stratified by the median value (20%) of CSI improvement demonstrated that patients with a < 20% improvement in CSI had higher rates of UNa < 70 mEq/L. In the multivariable analysis, in which the UNa was included as a continuous variable, UNa was significantly associated with poor CSI improvement (P = 0.047). In a receiver operating characteristic analysis, the optimal cutoff value for distinguishing a 20% improvement in CSI was 103 mEq/L. A UNa measurement 1 h after intravenous furosemide administration may be useful in predicting pulmonary decongestion in patients with acute decompensated HF, however, the optimal cutoff value requires further validation.