Combination of Urinary Sodium/Creatinine Ratio and Plasma Brain Natriuretic Peptide Level Predicts Successful Tolvaptan Therapy in Patients With Heart Failure and Volume Overload.

Abstract

To evaluate the short-term clinical and hemodynamic effects of tolvaptan therapy and to identify predictors of the therapeutic outcomes, we retrospectively recruited 60 consecutive hospitalized heart failure (HF) patients (70 ± 11 years) with volume overload. The subjects were divided into two groups on the basis of the changes in HF symptom scores and hemodynamic status assessed by right heart catheterization after tolvaptan therapy (median: 7 days). The majority of patients were successfully treated (group 1). However, 22% of patients (group 2) were unsuccessfully treated, in whom 1) the HF symptom score worsened or 2) there was a stationary HF symptom score ≥ 6 points, and mean PCWP > 18 mmHg and mean RAP > 10 mmHg, after tolvaptan therapy. HF symptom scores, hemodynamic parameters, and plasma brain natriuretic peptide (BNP) level improved in group 1, but all of these parameters remained unchanged in group 2. Lower urine sodium/creatinine ratio (UNa/UCr) and higher BNP level at baseline were independently associated with unsuccessful tolvaptan therapy, and UNa/UCr best predicts unsuccessful tolvaptan therapy with a cut-off value of 46.5 mEq/g·Cr (AUC 0.847, 95% CI: 0.718-0.976, sensitivity 77%, specificity 81%, P < 0.01). Double-positive results of UNa/UCr < 46.5 mEq/g·Cr and plasma BNP level > 778 pg/mL predicted unsuccessful tolvaptan therapy with high diagnostic accuracy (sensitivity 54%, specificity 100%, positive predictive value 100%, negative predictive value 89%, and accuracy 90%). In summary, short-term tolvaptan therapy ameliorated HF symptoms and provided hemodynamic improvement in the majority of patients, and UNa/UCr and BNP level strongly predicted the therapeutic outcomes.