Home Management of Refractory Ascites in Decompensated Cirrhosis With Long‐Term Abdominal Drains, a Pilot Study

1. Introduction

The prevalence of decompensated liver cirrhosis (DC) is increasing worldwide [1]. Ascites‐related readmissions are the predominant cause of hospitalizations in DC [1, 2]. Refractory ascites (RA), characterized by diuretic non‐responsiveness or intolerance, is encountered in 10% of patients with cirrhosis and is associated with reduced survival without liver transplantation (LT) or trans‐jugular intrahepatic portosystemic shunts (TIPSS) [3]. Large volume paracentesis (LVP) and albumin infusions are the only therapeutic options in those with RA who cannot undergo LT or TIPSS. LVPs are often required weekly or fortnightly, placing significant burden on hospital systems and often leading to unplanned hospitalizations. These recurrent admissions, with adverse impacts on patients' quality of life (QoL) and health expenditure, are potentially avoidable if ascites drainage can be regularly performed in patients' homes. Since ascites drainage in patients unsuitable for LT is a palliative procedure, it is best approached along principles of palliative care. Drainage using indwelling catheters is a well‐accepted model of care in patients with malignant ascites and hydrothorax [4]. There is limited evidence supporting this procedure in cirrhotic patients with RA [5]. The aim of this study was therefore to explore the feasibility, effectiveness, safety, and acceptability of home drainage of ascites with long‐term abdominal drains (LTAD) in an Australian health care setting as a management pathway for RA. The complete study protocol is included as Supporting Information, Section 1.

2. Methods

2.1. Study Design and Location

This was a feasibility pilot study conducted at a tertiary hepatology unit from September 2022 to September 2024, with a 6‐month follow‐up.

2.2. Participants

Patients older than 18 years with DC and RA, ineligible to undergo LT or TIPSS, were included in the study. Patients with alcoholic liver disease were considered for LTAD only after allowing sufficient time for recompensation.

RA was defined as ascites:

• –Unresponsive to sodium restriction, and high‐dose diuretic treatment (spironolactone 400 mg/day and/or furosemide 160 mg/day) with or without intolerance to diuretics.
• –Recurred rapidly after LVP (≥ 1 LVP/month), no clinical response to weekly albumin infusions.

Patients with loculated ascites, active infection, and no access to community nursing support were excluded.

3. LTAD Insertion and Drainage

After obtaining informed consent, Rocket LTAD catheters (Rocket Medical, Watford, UK) were inserted by interventional radiologists. Participants underwent complete drainage with albumin replacement. They were discharged the next day with sufficient drainage kits for 4 weeks of drainage. The local community nurses' pathway was utilized for ongoing LTAD drains. Nurses were provided with instructions and a referral form regarding the frequency and amount of drainage to be done for each participant. Participants underwent drainage two to three times per week at home as guided by their abdominal discomfort. During each visit, one to two liters of ascites was drained, as per the previously published experience [5]. Antibiotic prophylaxis with norfloxacin or equivalent was given throughout the duration of LTAD being in situ. No albumin replacement was given. Whenever bacterial peritonitis (BP) was suspected, ascitic fluid was sampled via LTAD and from the abdominal wall. Emergency contact numbers for reporting any adverse events were provided.

4. Results

4.1. Feasibility

Fifty‐one patients with DC and RA were screened. Thirty‐seven of these patients were excluded for the following reasons: deceased prior to consent (8), unable to access community nursing (3), underwent LT (7), underwent or being worked up for TIPSS (6) and improvement in ascites control (13). Fourteen participants were eligible and consented for LTAD. Two consenting participants died prior to drain insertion. Twelve participants had drains inserted and were included in the analysis.

4.2. Baseline Demographics and Clinical Characteristics

Baseline demographics are shown in Table 1. There was no significant difference between baseline or end of study creatinine and Model for End‐stage Liver Disease (MELD) (p = 0.2 and p = 0.3, respectively, by Wilcoxon Signed Rank Test). Three patients were on diuretics during the study period for peripheral edema. Median follow‐up duration post drain insertion was 103 days.

TABLE 1.

Demographics of study participants (N = 12).

Patient characteristics	N = 12
Age in years	71 (12)
Male, n(%)	8 (67)
Etiology
Alcohol	4
MAFLD	6
Non‐cirrhotic portal hypertension	2
Child Pugh Score	9 (1)
Baseline creatinine	153 (91)
End of study creatinine	175 (67)
MELD at insertion	16 (9)
End of study MELD	19 (5)
Median duration of LTAD (days)	103 (87)
Comorbidities
Cardiac failure	3
Renal dysfunction	6
Ongoing alcohol use	1
Hyponatremia compromising diuretic therapy	2

Note: All continuous variables expressed in median (interquartile ratio).

4.3. Effectiveness

The number of LVPs before and after LTAD is listed in Table S1. There were no hospitalizations required for LVP after LTAD insertion.

Details of LTAD drains are listed in Table S2. The number of LTADs ranged from 5 to 87 during a maximum of 6 months follow‐up. Frequency of home drains was two to three times per week in most cases.

4.4. Safety

Six patients died during the study period. None of the deaths were related to LTAD. Details of adverse events are listed in Table S3. There were two instances of BP, including one episode that occurred within 48 h of drain insertion due to missed antibiotic prophylaxis. The other case was related to multiple procedures for repositioning and reinsertion of a blocked drain. In both instances, there was resolution of infection with intravenous antibiotics without removal of LTAD. There was one instance of abdominal discomfort that required a day stay in hospital for manipulation of the drain. There was one case of severe discomfort and leakage around LTAD that led to study withdrawal. Minor adverse events included one cellulitis at the LTAD site, four instances of self‐limiting abdominal discomfort, and one incidentally detected BP at drain insertion that was managed with intravenous antibiotics.

4.5. Acceptability

Five patients and four carers completed the satisfaction survey as shown in Table S4. Other participants were unable to complete the survey within the specified timeframe due to the worsening of their illness. All responding participants preferred LTAD to LVPs and reported high levels of satisfaction with LTAD. Three of four participants who completed 6 months of study opted for the continuation of LTAD. The fourth patient had improved liver function with alcohol abstinence and no further accumulation of ascites.

5. Discussion

Management of RA in DC patients without definitive options should be in line with the principles of palliative care by prioritizing symptomatic management and preserving QoL. This pilot study explored the option of LTAD in an Australian health care setting and found it to be a safe, effective, and acceptable alternative to LVPs.

Only 12/51 (24%) patients with RA had an LT or TIPSS option for management, highlighting the need for evidence‐based palliative care for the majority of patients with DC and RA [6]. Eight of the 51 patients had far too advanced liver disease to be considered for LTAD. This demonstrates the need to consider LTAD early during the course of decompensation when definitive treatment options no longer exist, but patients remain well enough to tolerate palliative drains. With long‐term albumin infusions emerging as an option to prevent and reduce ascites accumulation, the appropriate timing and patient population for LTAD need to be defined. We recommend that LTAD should be discussed early during the course of RA in DC, when TIPSS or LT is not an option, and when the frequency of LVPs is unchanged with albumin infusions.

Patients with DC are at high risk of infections which can be life threatening [7]. The fear of infections precluded prior trials on LTAD. With encouraging results of lower frequency of peritonitis with LTAD compared to LVPs (6% vs. 11%), from the REDUCE trial in the UK, we designed this study under antibiotic prophylaxis [5]. It was encouraging that BP was encountered in only two participants, in the setting of a prescription error leading to absence of antibiotic prophylaxis and LTAD blockage requiring re‐intervention. In both instances, BP was managed without the need to remove the LTAD. A meta‐analysis including 18 studies revealed a low incidence of infections and non‐infectious complications with LTAD [8]. This is confirmed by a large retrospective cohort study that reported no increased incidence of peritonitis with LTAD, but an increased incidence of acute kidney injury [9]. In our study, there was no significant change in serum creatinine post‐LTAD insertion, despite lack of support with albumin infusions. This is likely explained by frequent small volume ascitic fluid drainage not resulting in post‐paracentesis circulatory dysfunction. We did not encounter any new onset hyponatraemia.

Local complications including cellulitis and leakage were mild and non‐serious, similar to the REDUCE study. Only one patient had persistent pain and leakage that required removal of LTAD. This patient had multiple prior abdominal surgeries and anxiety, which likely affected LTAD tolerability. Interestingly, this patient subsequently had LTAD insertion post study, which was well tolerated. We recommend that patients be counseled about the possible need for hospitalizations for LTAD adjustments, antibiotics, and minor abdominal discomfort before the procedure is offered.

Our study had excellent participant and carer acceptability; they reported less stress due to the avoided hospital visits.

Notwithstanding the small numbers in this study, its strength lies in its prospective design with a well‐planned care model involving both community nursing support and supervision by hospital staff. More objective assessment of patients with the EQ‐5D‐5L questionnaire would have been helpful in studying the impact of LTAD on QoL. However, feedback and interviews supported the patients' and carers' positive experience.

In summary, we have established the feasibility of LTAD in an Australian health care setting with its excellent community nursing network. Its effectiveness was established with patients not requiring any further hospitalizations for ascitic drainage. Most importantly, BP was infrequent. LTAD was highly acceptable to participants and carers who preferred LTAD to regular LVPs. We recommend large prospective studies to evaluate the safety and cost‐effectiveness of LTADs in patients with RA.

Ethics Statement

This study was approved by the Southern Adelaide Clinical Human Research Ethics Committee (reference 2022/HRE00044) on 17 June 2022.

Consent

All patients provided written informed consent.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1. Supplementary Material.

Ramachandran J., Bragg K., Narayana S., et al., “Home Management of Refractory Ascites in Decompensated Cirrhosis With Long‐Term Abdominal Drains, a Pilot Study,” JGH Open 9, no. 10 (2025): e70228, 10.1002/jgh3.70228.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.