Daily Low-Volume Paracentesis and Clinical Complications in Patients With Refractory Ascites

Abstract

Importance: The potential association of low-volume paracentesis of less than 5 L with complications in patients with ascites remains unclear, and individuals with cirrhosis and refractory ascites (RA) treated with devices like Alfapump or tunneled-intraperitoneal catheters perform daily low-volume drainage without albumin substitution. Studies indicate marked differences regarding the daily drainage volume between patients; however, it is currently unknown if this alters the clinical course.

Objective: To determine whether the incidence of complications, such as hyponatremia or acute kidney injury (AKI), is associated with the daily drainage volume in patients with devices.

Design, setting, and participants: This retrospective cohort study of patients with liver cirrhosis, RA, and a contraindication for a transjugular intrahepatic portosystemic shunt who received either device implantation or standard of care (SOC; ie, repeated large-volume paracentesis with albumin infusion), and were hospitalized between 2012 and 2020 were included. Data were analyzed from April to October 2022.

Interventions: Daily ascites volume removed.

Main outcomes and measures: The primary end points were 90-day incidence of hyponatremia and AKI. Propensity score matching was performed to match and compare patients with devices and higher or lower drainage volumes to those who received SOC.

Results: Overall, 250 patients with RA receiving either device implantation (179 [72%] patients; 125 [70%] male; 54 [30%] female; mean [SD] age, 59 [11] years) or SOC (71 [28%] patients; 41 [67%] male; 20 [33%] female; mean [SD] age, 54 [8]) were included in this study. A cutoff of 1.5 L/d or more was identified to estimate hyponatremia and AKI in the included patients with devices. Drainage of 1.5 L/d or more was associated with hyponatremia and AKI, even after adjusting for various confounders (hazard ratio [HR], 2.17 [95% CI, 1.24-3.78]; P = .006; HR, 1.43 [95% CI, 1.01-2.16]; P = .04, respectively). Moreover, patients with taps of 1.5 L/d or more and less than 1.5 L/d were matched with patients receiving SOC. Those with taps of 1.5 L/d or more had a higher risk of hyponatremia and AKI compared with those receiving SOC (HR, 1.67 [95% CI, 1.06-2.68]; P = .02 and HR, 1.51 [95% CI, 1.04-2.18]; P = .03), while patients with drainage of less than 1.5 L/d did not show an increased rate of complications compared with those receiving SOC.

Conclusions and relevance: In this cohort study, clinical complications in patients with RA performing low-volume drainage without albumin infusion were associated with the daily volume drained. Based on this analysis, physicians should be cautious in patients performing drainage of 1.5 L/d or more without albumin infusion.

Figure 1.. Ninety-Day Incidence of Hyponatremia, Acute Kidney Injury (AKI), and Severe AKI in Patients With Daily Paracentesis of 1.5 L/d or More and Less Than 1.5 L/d

Death and liver transplantation were considered as competing events. Patients chronically dependent on hemodialysis at baseline were excluded from analysis regarding AKI or severe AKI. Overall, 13 patients adapted their daily drainage volume after a mean of 35 days in the follow-up (n = 9 in the 1.5 L/d or more group and n = 4 in the less than 1.5 L/d group). Schoenfeld residual plots indicated that the proportional hazards assumption was met in the respective analyses.

Figure 2.. Ninety-Day Incidence of Hyponatremia, Acute Kidney Injury (AKI), and Severe AKI in Patients With Daily Paracentesis of 1.5 L/d or More and Standard of Care (SOC)

Death and LTx were considered as competing events. Patients chronically dependent on hemodialysis at baseline were excluded from analysis regarding AKI/severe AKI. Schoenfeld residual plots indicated that the proportional hazards assumption was met in the respective analyses.

Figure 3.. Ninety-Day Incidence of Hyponatremia, Acute Kidney Injury (AKI), and Severe AKI in Patients With Daily Paracentesis Less Than 1.5 L/d

A, Hyponatremia; B, AKI; and C, severe AKI in patients with daily paracentesis of less than 1.5 L/d and SOC. Death and liver transplantation were considered as competing events. Patients chronically dependent on hemodialysis at baseline were excluded from analysis regarding AKI or severe AKI. Schoenfeld residual plots indicated that the proportional hazards assumption was met in the respective analyses. SOC indicates standard of care.