Clinical implications with tolvaptan on monitored bioimpedance-defined fluid status in patients with cirrhotic ascites: an observational study

Abstract

Background: Prognostic value or clinical implications of fluid status monitoring in liver cirrhosis are not fully elucidated. Tolvaptan, an orally available, selective vasopressin V2-receptor antagonist approved for hyponatremia in the United States and European Union. It is also used for cirrhotic ascites at a relatively low dose (3.75 mg to 7.5 mg) in Japan, exerts its diuretic function by excreting electrolyte-free water. We hypothesized that bioimpedance-defined dynamic changes in fluid status allow prediction of response of V2 antagonism and survival in cirrhotic patients.

Methods: In this prospective observational study, 30 patients with decompensated liver cirrhosis who were unresponsive to conventional diuretics were enrolled. Detailed serial changes of body composition that were assessed by using non-invasive bioimpedance analysis (BIA) devices, along with biochemical studies, were monitored at 5 time points.

Results: Sixteen patients were classified as short-term responders (53%). Rapid and early decrease of BIA-defined intracellular water, as soon as 6 h after the first dose (ΔICW_BIA%-6 h), significantly discriminated responders from non-responders (AUC = 0.97, P < 0.0001). ΔICW_BIA%-6 h was highly correlated with the change of BIA-derived phase angle of trunk, e.g. reduced body reactance operated at 50 kHz after 24 h of the first dose of tolvaptan. Lower baseline blood urea nitrogen and lower serum aldosterone were predictive of a rapid and early decrease of ICW_BIA. A rapid and early decrease of ICW_BIA in response to tolvaptan was also predictive of a better transplant-free survival.

Conclusions: BIA-defined water compartment monitoring may help predict short-term efficacy and survival in decompensated cirrhotic patients treated with tolvaptan.

Keywords: Ascites; Impedance; Liver cirrhosis; Vasopressin antagonism.

Fig. 1

Serial monitoring of body weight, changes in BW, extracellular water to total body water ratio, and serum sodium concentration after add-on tolvaptan at low doses. Body weight (BW, in kilograms, a, changes in body weight (ΔBW, in kilograms, b, the extracellular water/total body water (ECW/TBW) ratio (using the bioimpedance analysis [BIA] method, c, and serum sodium concentration (in mmol/L, d were monitored as illustrated in Figure S2, at 0, 6, 24, 72, and 168 h (missing data: 6 for BIA at 168 h) after add-on tolvaptan at low doses was administered. Patients with 1.5 kg or more body weight reduction (horizontal light gray line in b) on day eight were defined as responders (open circles); all other patients were defined as non-responders (closed triangles). Medians with interquartile ranges are shown in bars. Abbreviations: ECW, extracellular water; TBW, total body water. *, P < 0.05

Fig. 2

Serial monitoring of bioimpedance-defined intracellular water, extracellular water, and the percent of change in both parameters by bioimpedance analysis after add-on tolvaptan at low doses. Bioimpedance-defined intracellular body water standardized by body surface area (BSA) (ICW_BIA-c, L/m², a, extracellular body water standardized by BSA (ECW_BIA-c, L/m², b, and percent of change in both compartments compared to day 0 (panel c) were monitored as illustrated in Figure S2, at 0, 6, 24, 72, and 168 (missing data: 6 for BIA at 168 h) hours after add-on tolvaptan at low doses was administered. Patients with 1.5 kg or more body weight reduction on day eight were defined as responders (open circles for ICW_BIA; open squares for ECW_BIA; with thin lines), and others as non-responders (closed upward triangles for ICW_BIA; closed downward triangles for ECW_BIA; with dashed lines). Medians with interquartile ranges are shown in bars. In panel c, P-values compared ICWs with respect to response are shown, whereas P-values referring to ECW_BIA, comparisons are shown in brackets. Abbreviations: ICW, intracellular water; ECW, extracellular water; BIA; bioimpedance analysis; BSA, body surface area; NS, not significant. *, P < 0.05; **, P < 0.01; ***, P < 0.0001

Fig. 3

Correlation analysis of the change of the phase angle highly and the percent change of bioimpedance-defined intracellular water at 6 h after add-on tolvaptan. Patients with 1.5 kg or more body weight reduction on day eight were defined as responders (open circles), and others as non-responders (closed triangles). Phase angle (PA) was measured at 50 Hz. ICW, intracellular water; BIA, bioimpedance analysis; R, correlation coefficient. **, P < 0.001

Fig. 4

Receiver operating characteristic curves for predicting response to add-on tolvaptan. Comparison of the prognostic value of various parameters for the discrimination of responders and non-responders to add-on tolvaptan treatment is shown. The receiver operating characteristic (ROC) curves for the percent change in bioimpedance-defined intracellular water at 6 h after add-on tolvaptan (ΔICW_BIA%-6 h), and other parameters are shown. Detailed statistics, including statistical comparison of the ROC curves by the DeLong method, are presented in Table S2. Abbreviations: MELD, model for end-stage liver disease; CPT, Child-Pugh-Turcotte; T-Bil, total bilirubin; AST, aspartate aminotransferase; ALT, alanine aminotransferase; AVP, arginine vasopressin

Fig. 5

Serial monitoring of urine osmolality, serum sodium and blood urea nitrogen and the percent of change in both parameters in patients whose intracellular water decreased or increased after add-on tolvaptan. Patients whose bioimpedance-defined intracellular water (ICW_BIA) decreased (open circles with lines) or increased (closed circles with dashed lines) after add-on tolvaptan were compared. Urine osmolality (mOsm/kg. H₂O; a), serum sodium (mmol/L; b, left) and blood urea nitrogen (BUN, mg/dL; b, right) were monitored as illustrated in Figure S2, at 0, 6, 24, 72, and 168 h. The combined percent changes in both serum sodium (Na) and urea nitrogen (UN) were compared in ICW_BIA decreased/increased cases (c, left) and responders/non-responders (c, right). Spearman correlation analyses of the combined percent changes in both serum sodium (Na) and urea nitrogen (UN) and that of ICW_BIA and bioimpedance-defined extracellular water (ECW_BIA) were shown in d. Medians with interquartile ranges are shown in bars. *, P < 0.05; **, P < 0.01

Fig. 6

Kaplan–Meier analysis for long-term survival. In the 30 cases studied, transplant-free survival from the administration of add-on tolvaptan by Kaplan–Meier analysis is compared. Patients are stratified according to (a) bioimpedance-defined intracellular water (ICW_BIA) decreases or increases after add-on tolvaptan; eight vs ten events; b response to tolvaptan; eight vs ten events; c Model for End-stage Liver Disease (MELD) score at 18; fifteen vs three events; d Child-Pugh-Turcotte (CPT) grade B or C; eight vs ten events; e MELDNa score at 20; nine vs nine events. P-values and hazard ratio from log-rank tests are shown. *P < 0.05