Impact of Ultrafiltration on Serum Sodium Homeostasis and its Clinical Implication in Patients With Acute Heart Failure, Congestion, and Worsening Renal Function

Abstract

Background: The relationship between changes in serum sodium with clinical events in acute heart failure patients using different decongestive events has not been investigated. This study aimed to describe changes in serum sodium levels during decongestion therapy in patients receiving stepped pharmacological therapy versus ultrafiltration.

Methods and results: We studied 188 patients who were enrolled in the CARRESS-HF trial (Cardiorenal Rescue Study in Acute Decompensated Heart Failure). Treatment-induced hyponatremia was defined as admission normonatremia (≥135 mEq/L) with a subsequent decrease (<135 mEq/L) during hospitalization. Patients treated with ultrafiltration had significantly lower sodium levels than those with conventional treatment at days 1, 4, and 7 (all P<0.01), whereas those at day 30 were similar between the groups. Changes in sodium levels in patients with ultrafiltration were negatively correlated to those in serum creatinine and plasma renin activity. The incidence of treatment-induced hyponatremia was significantly higher in the ultrafiltration group than those receiving conventional treatment (P=0.002). Although patients with discharge hyponatremia had a higher risk for composite end point of all-cause death, rehospitalization, or unscheduled hospital visit in comparison to those without (adjusted hazard ratio, 2.01; 95% confidence interval, 1.09-3.70; P=0.025), the risk was comparable between patients with treatment-induced hyponatremia and those who did not experience any hyponatremia (adjusted hazard ratio, 0.99; 95% confidence interval, 0.50-1.96; P=0.99).

Conclusions: Fluid removal by ultrafiltration was associated with a decrease in serum sodium levels compared with diuretic treatment but returned to baseline levels at day 30. Discharge hyponatremia but not treatment-induced hyponatremia was associated with worse clinical outcomes.

Clinical trial registration: URL: http://www.clinicaltrials.gov. Unique identifier: NCT00608491.

Keywords: creatinine; diuretics; heart failure; hyponatremia; ultrafiltration.

Figure 1

Study patient flow. Treatment-induced hyponatremia (TIH) was defined as admission normonatremia with a subsequent decrease during hospitalization. Decompensation hyponatremia was defined as admission hyponatremia that disappeared with decongestive treatment at discharge.

Figure 2

Comparison of changes in serum sodium levels between patients on pharmacological therapy versus ultrafiltration. * Denotes P<0.05 and † denotes P<0.01.

Figure 3

Comparison of changes in serum sodium, creatinine, and plasma renin activity in patients receiving ultrafiltration (A) and those with pharmacological therapy (B). PRA = plasma renin activity.

Figure 3

Comparison of changes in serum sodium, creatinine, and plasma renin activity in patients receiving ultrafiltration (A) and those with pharmacological therapy (B). PRA = plasma renin activity.

Figure 4

The Comparison of Kaplan-Meier survival curves for freedom from the primary outcome measures between patients with and without discharge hyponatremia (A), those with and without admission hyponatremia (B), those with and without treatment induced hyponatremia (C), among patients stratified by patterns of sodium levels during decongestion therapy (D), and those with persistent hyponatremia versus treatment induced hyponatremia (E).

Figure 4

The Comparison of Kaplan-Meier survival curves for freedom from the primary outcome measures between patients with and without discharge hyponatremia (A), those with and without admission hyponatremia (B), those with and without treatment induced hyponatremia (C), among patients stratified by patterns of sodium levels during decongestion therapy (D), and those with persistent hyponatremia versus treatment induced hyponatremia (E).

Figure 4

The Comparison of Kaplan-Meier survival curves for freedom from the primary outcome measures between patients with and without discharge hyponatremia (A), those with and without admission hyponatremia (B), those with and without treatment induced hyponatremia (C), among patients stratified by patterns of sodium levels during decongestion therapy (D), and those with persistent hyponatremia versus treatment induced hyponatremia (E).

Figure 4

The Comparison of Kaplan-Meier survival curves for freedom from the primary outcome measures between patients with and without discharge hyponatremia (A), those with and without admission hyponatremia (B), those with and without treatment induced hyponatremia (C), among patients stratified by patterns of sodium levels during decongestion therapy (D), and those with persistent hyponatremia versus treatment induced hyponatremia (E).

Figure 4

The Comparison of Kaplan-Meier survival curves for freedom from the primary outcome measures between patients with and without discharge hyponatremia (A), those with and without admission hyponatremia (B), those with and without treatment induced hyponatremia (C), among patients stratified by patterns of sodium levels during decongestion therapy (D), and those with persistent hyponatremia versus treatment induced hyponatremia (E).