Midodrine versus placebo in patients with acute-on-chronic liver failure with ascites—A randomized trial (MIDAS trial)

Abstract

Background:

Ascites is common in patients with acute-on-chronic liver failure (ACLF) and can complicate the course of ACLF due to circulatory dysfunction and hemodynamic compromise. Oral midodrine has been variably reported to improve hemodynamics and reduce complications of ascites in patients with decompensated cirrhosis, but its role in patients with high MELD ACLF remains unknown, which we aimed to assess.

Methods:

In this single-center, double-blind, randomized trial, patients with ACLF were assigned 1:1 to receive midodrine (5–7.5 mg tid) or a placebo for 30 days, alongside standard care. Transplant-free survival (TFS) at 1 month was the primary outcome, while secondary outcomes included TFS at 3 months, the proportion of patients achieving ascites control, changes in mean arterial pressure (MAP) and plasma renin activity at 1 month, the incidence of cirrhosis complications, and diuretic-related complications.

Results:

One hundred thirty patients with ACLF were enrolled. Midodrine use was not associated with improvement in TFS at 1 (HR, 0.99 [95% CI: 0.37–2.65]) and 3 months (HR, 0.93 [95% CI: 0.48–1.8]). Ascites control at day 30 (placebo: 9.2% [95% CI: 3.5–19] vs. midodrine: 20% [95% CI: 12.3–33.5]; p=0.08) and 90 (placebo: 44.6% [95% CI: 32.2–57.5] vs. midodrine: 53.8% [95% CI: 41–66.3]; p=0.29) was comparable. Midodrine increased the MAP (1.9 vs. 6.9 mm Hg; p=0.003) with an insignificant reduction in plasma renin activity levels (delta change: 0.22±1.6 vs. −1.6±2.2; p=0.09). Cirrhosis-related and diuretic-related complications were similar in both groups.

Conclusions:

Midodrine increases MAP in patients with ACLF and high MELD scores without TFS benefit and does not reduce the rate of ascites-related complications.

INTRODUCTION

Ascites is a significant complication in the natural history of liver disease and serves as a marker of poor prognosis. Midodrine is a vasoconstrictor (alpha-1 agonist) that has been evaluated in the management of complications of ascites, including refractory ascites, hepatorenal syndrome-acute kidney injury (AKI), hyponatremia, and for the prevention of paracentesis-induced circulatory dysfunction.^1–5 A randomized study reported a survival benefit with midodrine in patients with cirrhosis and refractory ascites within a small cohort. ¹ While the combination of midodrine, octreotide and albumin is suboptimal in the management of hepatorenal syndrome-AKI, it is advocated to be used in situations where terlipressin or norepinephrine is contraindicated.^3,6 Midodrine has also been shown to be a cost-effective measure with similar efficacy to albumin in preventing paracentesis-induced circulatory dysfunction, both in patients with cirrhosis and acute-on-chronic liver failure (ACLF).^7–9 Similarly, we recently observed that in patients with cirrhosis and refractory ascites undergoing large-volume paracentesis, the addition of midodrine to albumin reduced both the frequency and volume of ascites removed, and related complications. ¹⁰ Midodrine aids in controlling ascites by increasing mean arterial pressure (MAP), which subsequently leads to increased urine volume and sodium excretion.^1,10 However, the midodrine and albumin for cirrhotic patients in the waiting list for liver transplantation trial, which evaluated the combination of midodrine and albumin in patients with cirrhosis and ascites awaiting liver transplantation, reported no survival benefit or improvement in MAP. ¹¹ It is worth noting that in this trial, over 50% of patients underwent liver transplantation within a median period of 42 days (range: 18–95 d). ¹¹ Additionally, the study included patients with a MELD score of 16–17. We previously reported, in a retrospective pilot study, that patients with ACLF receiving midodrine might achieve better ascites control with fewer diuretic-related complications, though without a survival benefit. ¹² The limitations of our prior study prompted us to conduct a randomized controlled trial on repurposing the role of midodrine in patients with ACLF. ¹³ Consequently, we aimed to evaluate the benefit of midodrine in patients with ACLF with high MELD scores.

METHODS

Study design

This double-blind, placebo-controlled, randomized trial (MIDAS-midodrine for ACLF with ascites) was conducted at AIG Hospitals, Hyderabad. Patients were randomly allocated to receive either oral midodrine or a matching placebo for 30 days. Recruitment began on January 16, 2023, and the trial concluded on July 14, 2025. The study received approval from the Institutional Ethics Committee (Asian Institiute of Gastroenterology /IEC-BH&R 32/07.2022-02) and was registered with the Clinical Trial Registry-India (CTRI/2023/01/048926). Written informed consent was obtained from all participants, and the study adhered to the modified ethical principles of the Declaration of Helsinki and Istanbul.

Participants selection criteria

Consecutive patients with ACLF as identified by APASL criteria, aged between 18 and 75 years, presenting to the outpatient department, were included. We excluded those without clinical ascites, those who received terlipressin, had active untreated sepsis, or shock requiring vasopressor support (within the previous 7 d), those with HCC or PVT, Budd-Chiari syndrome, pregnant women, patients with hypertension, recent midodrine exposure (within the previous 7 d), and those unwilling to participate.

Study interventions

Patients were randomly assigned to receive midodrine 5 mg 3 times daily, with the dose increased to 7.5 mg 3 times daily if the MAP did not rise by at least 5 mm Hg by day 15. The placebo group received matching placebos 3 times daily, and the number of tablets was adjusted based on the MAP on day 15. Treatment was administered for 30 days. All patients were scheduled for compulsory follow-up visits on days 15, 30, and 90, and were asked to seek medical attention for any clinical issues in the interim. The lower midodrine dose used in the current study was intended to reflect clinical practice, in which midodrine is typically initiated at 15 mg/d and titrated based on MAP response, and to avoid the pill burden for these patients. In cases of drug intolerance or grade 2 or higher hypertension, therapy was permanently discontinued. For patients who developed stage 1 hypertension, the dose of the drug was reduced to half. All participants with ACLF and ascites received standard care according to European Association for the Study of the Liver (EASL) guidelines, which included a salt restriction of 2 g/d and diuretics as tolerated. ¹⁴ Weekly albumin infusions (20–25 g/wk for 4–6 wk) were recommended for patients with ascites ¹² ; however, due to limited funding for albumin, its use depended on patient willingness. Plasma renin activity (PRA) was measured on fresh morning samples using the ELISA method.

Definitions

ACLF was defined according to APASL criteria (type A) as patients with jaundice (serum bilirubin >5 mg/dL) and coagulopathy (international normalized ratio >1.5) who developed ascites within 4 weeks, with or without HE. ¹⁵ Ascites control was defined as a 1-grade improvement on ultrasonography (USG) and/or ascites clearance, coupled with a 10% reduction in body weight and abdominal circumference (measured at the umbilicus). ¹² Patients with symmetrical abdominal distension on clinical examination, along with USG evidence of fluid in multiple quadrants, were considered grade 2, and patients with marked abdominal distension and tense abdomen with or without respiratory distress, and USG showing diffuse anechoic fluid in all quadrants with floating bowel loops were considered grade 3. ¹⁶ Diuretic-related complications were defined according to EASL criteria. ¹⁴ Diuretic-induced HE refers to the development of encephalopathy without any other precipitating factor. An increase in serum creatinine by >100%, reaching a value above 2 mg/dL in patients with ascites responding to treatment, was considered diuretic-induced renal injury. AKI was staged according to the Kidney Disease: Improving Global Outcomes (KDIGO) guidelines. Hyponatremia was defined as a serum sodium level below 130. A decline in serum sodium of >10 mmol/L to <125 mmol/L was regarded as diuretic-induced hyponatremia. A serum potassium level <3 mmol/L or >6 mmol/L, despite appropriate measures, was considered diuretic-induced hypo- or hyperkalemia. Hypertension stage 1 was defined as systolic blood pressure of 120–129 and/or diastolic blood pressure of 80–89, and stage 2 as systolic blood pressure≥140 and/or diastolic blood pressure ≥90. ¹⁷ A MAP at baseline of <80 mm Hg was considered low, and those with a MAP of 80 mm Hg or higher were considered high. ¹⁸

Study outcomes

The primary outcome was transplant-free survival (TFS) at 1 month. Secondary outcomes included TFS at 3 months, ascites control, number of paracenteses, changes in MAP at 1 month, changes in PRA at day 30, cirrhosis-related complications, incidence of diuretic-related complications (acute kidney injury [AKI], dyselectrolytemia, and encephalopathy), incidence of infections, frequency and duration of hospitalizations, and drug-related adverse events. Following the completion of the study, a subgroup analysis was performed on patients who received albumin. Post hoc analysis was also performed to assess the impact of ACLF grade, MAP, and the presence or absence of AAH on outcomes.

Sample size and randomization

Previously, we noted that the TFS at day 30 is 70% in patients with ACLF. ¹⁹ Considering this as a superiority trial with at least 20% improvement in survival with the addition of midodrine, ¹ and an alpha error of 5% and power of 80%, we needed to enroll 62 patients in each group. Accounting for a 5% attrition rate, we needed to enroll 65 patients in each group. A total of 130 patients were planned to be recruited for the study. For this single-center study, a randomization list was prepared using an online computer-generated sequence with block sizes of 10, and patients were randomly assigned in a 1:1 ratio to receive midodrine or placebo. The investigator and the patients were blinded to the allocated treatment. Midodrine 2.5 mg tablets and a matching placebo were prepared by Cipla India Pvt Ltd and provided in a bottle containing 200 tablets. Each individual was given one bottle (containing 200 tablets) on day 1 by the clinical trial coordinator, who ensured drug adherence by counting the tablets remaining in the bottle at each subsequent visit and questioning the patient about consumption.

Statistical analysis

This was an intention-to-treat analysis and included all patients who received at least one dose of the drug or placebo. SPSS version 31 (IBM Corp) was used for data analysis. Baseline characteristics were compared between the 2 groups using Student t test for continuous variables and were expressed as mean (SD) or median (range) as appropriate. All the p values are reported as 2-sided. Similarly, the 2-sided Fisher exact test was used to compare categorical variables between the groups, expressed as proportions (n, %). For post hoc subgroup analysis of patients who received albumin, the ANOVA test was used for continuous variables and the χ² test for categorical variables among the 4 groups (ie, placebo only vs. albumin+placebo vs. midodrine alone vs. midodrine+albumin). Patients who were lost to follow-up were considered as events in outcomes. Kaplan-Meier analysis was performed to assess the TFS, and comparisons between groups were made using the Log-rank test. The primary outcomes are presented as proportions with 95% CIs. A p value <0.05 was considered statistically significant.

Funding

Cipla India Pvt Ltd provided the midodrine 2.5 mg tablets and matching placebos as an academic gesture. However, Cipla India Pvt Ltd had no role in the preparation of the protocol, the conduct of the trial, the interpretation of the results or the preparation of the manuscript.

RESULTS

One hundred thirty patients were randomly assigned equally to 2 groups after excluding 128 patients (grade I or no ascites: 38; sepsis/septic shock: 35; recent terlipressin therapy: 21; hypertension: 16; denial of consent: 9; HCC/Budd-Chiari syndrome: 6; recent midodrine exposure: 3). Two patients in each group were lost to follow due to logistical reasons (Figure 1). The baseline characteristics are presented in Table 1, where the groups were comparable. The mean MELD score was 24.1±3.8 in the placebo arm and 23.3±3.8 in the midodrine arm (p=0.82).

FIGURE 1.

Consort chart showing the trial enrollment details. Abbreviations: BCS, Budd-Chiari syndrome; ITT, intention to treat.

TABLE 1.

Baseline characteristics among the included patients

Variables	Placebo arm	Midodrine arm	p
Age (y)	42.6±9.9	42.9±9.5	0.87
BMI (kg/m2)	26±4.9	25.2±4.5	0.34
Abdominal girth (cm)	107.8±16.5	104.5±15.9	0.25
Men, n (%)	58 (89.2)	61 (93.8)	0.53
Women, n (%)	7 (10.8)	4 (6.2)
Diabetes mellitus, n (%)	10 (15.4)	11 (16.9)	1
Hypothyroidism, n (%)	5 (7.7)	4 (6.2)	1
Etiology, n (%)
Alcohol	45 (69.2)	50 (76.9)	0.63
Viral	16 (24.6)	11 (16.9)
DILI	1 (1.5)	3 (4.6)
AIH	3 (4.6)	2 (3.1)
Concomitant AAH, n (%)	34 (52.3)	31 (47.7)	0.72
Ascites grade, n (%)
2	39 (60)	40 (61.5)	1
3	26 (40)	25 (38.5)
Pedal edema, n (%)	27 (41.5)	28 (43.1)	1
Prior HE, n (%)	11 (16.9)	9 (13.8)	0.8
Carvedilol 3.125 mg/d, n (%)	7 (10.8)	5 (7.7)
Carvedilol 6.25 mg/d, n (%)	1 (1.5)	4 (6.2)	0.29
Rifaximin, n (%)	14 (21.5)	12 (18.5)	0.66
Blood pressure (mm Hg), n (%)
SBP	110.7±8.4	109.4±6.7	0.31
DBP	67.6±5.5	67.3±4.7	0.74
MAP	82±5.2	81.3±4.2	0.44
Hemoglobin (g/dL)	10.1±1.8	10.2±1.7	0.76
TLC (x 109/L)	9895.1±5485.2	10935.2±7002.7	0.34
Platelets (x 109/L)	158.1±66.1	155.3±86.1	0.83
Total bilirubin (mg/dL)	10 (5.2–35.9)	11.8 (5.2–29.9)	0.27
Direct bilirubin (mg/dL)	5.4 (2.3–23.2)	5.9 (1.5–17.6)	0.35
AST(U/L)	120 (47–526)	103 (25–598)	0.35
ALT (U/L)	45 (14–388)	51.5 (15–755)	0.85
Albumin (g/dL)	2.7±0.4	2.8±0.5	0.62
Sodium (meq/dL)	132.1±4	131.7±4.5	0.58
Creatinine (mg/dl)	0.82±0.2	0.88±0.3	0.22
INR	2.2±0.6	2.1±0.5	0.28
Lactate (mmol/L)	0.91±0.3	0.84±0.4	0.33
CTP score	10.9±1.4	10.9±1.5	0.9
CTP class, n (%)
B	11 (16.9)	10 (15.4)	1
C	54 (83.1)	55 (84.6)
MELD	23 (18–37)	23 (18–36)	0.82
CLIF-C ACLF	38.8±6.8	38.9±7.3	0.95
AARC	7.2±1.5	7.3±1.4	0.77
AARC grade, n (%)			0.42
1	42 (64.6)	36 (55.4)
2	19 (29.2)	26 (40)
3	4 (6.2)	3 (4.6)

Note: AARC APASL ACLF Research Consortium.

Abbreviations: AAH, Alcohol-associated hepatitis; AIH, autoimmune hepatitis; BMI, body mass index; CLIF-C, Chronic Liver Failure Consortium; CTP, Child Turcotte Pugh; DBP, diastolic blood pressure; INR, international normalized ratio; MAP, mean arterial pressure; SBP, systolic blood pressure; TLC, total leukocyte counts.

Treatment during the 90-day period

The mean duration of drug therapy was 28.6± 4.1 days in the placebo group and 28.2±5.4 days in the midodrine group (p=0.64). The median dose of midodrine was 15 mg/d (range: 7.5–22.5). Eight patients in the placebo arm and 9 in the midodrine arm received concomitant carvedilol. The proportion of patients receiving diuretics was 86.2% (56/65) in the placebo arm compared to 93.8% (61/65) (p=0.24). The mean daily dose of torsemide was 28.3±13.5 mg (n=15) in placebo arm and 31.4±12.2 mg (n=25) in midodrine arm (p=0.46); dose of furosemide was 21.5±5.3 mg (n=40) in placebo versus 22.5±6.7 mg (n=32) in midodrine (p=0.48); and the dose of spironolactone was 60.9±20.8 mg (n=55) versus 65.8±24.7 mg (n=60) in midodrine arm (p=0.25). The cumulative dose of loop diuretics was higher in the midodrine arm (40.1±26.2 vs. 31.5±21.7 mg/d in the placebo arm; p = 0.06). The duration of diuretic therapy was 53.8±31.8 days compared to 52±29.6 in the midodrine arm (p=0.76). Sixty-one percent (40/65) in the placebo arm and 60% (39/65) in the midodrine arm received albumin therapy during the study period, and the dose of albumin was similar between the 2 groups (95.4±26.8 vs. 94.6±22.5 g in the midodrine arm; p=0.89). The serum albumin levels were comparable between the 2 groups at day 30 (placebo: 3±0.45 vs. 2.9±0.5 in midodrine; p=0.4) and 90 (placebo: 3.2±0.46 vs. 3.1±0.41 in midodrine arm; p=0.11) (Supplemental Figure S1, http://links.lww.com/HC9/C349).

TFS at 1 and 3 months

TFS at 1 month was 87.7% (57/65; 95% CI: 77.1–94.5; Log-rank p=0.99) in both groups. TFS at 3 months was 72.3% (47/65; 95% CI: 59.8–82.7) in the placebo arm and 73.8% (48/65; 95% CI: 61.4–84; Log-rank p = 0.83). (Figure 2) Only 2 patients in the placebo arm and 3 in the midodrine arm underwent liver transplantation within 90 days. The causes of mortality are listed in Supplemental Table S1, http://links.lww.com/HC9/C349. The most common cause of mortality was infections (38.8% [7/18] vs. 23.5% [4/17]; p=0.33) and liver failure with multiorgan failure (33.3% [4/18] vs. 35.2% [5/17]; p=0.9) in both groups.

FIGURE 2.

Transplant-free survival among the 2 groups.

Ascites control and the number of paracenteses

At day 30, 9.2% (6/65; 95% CI: 3.5–19) of patients in the placebo arm compared to 20% (13/65; 95% CI: 12.3–33.5) in the midodrine arm (OR, 2.45 [95% CI: 0.87–6.93]; p=0.08) had ascites control. Similarly, at day 90, 44.6% (29/65; 95% CI: 32.2–57.5) in the placebo group, compared to 53.8% (35/65; 95% CI: 41–66.3) in the midodrine arm, had ascites control (OR, 1.44 [95% CI: 0.72–2.88; p=0.29]). The proportion of patients undergoing paracentesis was 33.8% (22/65) in the placebo arm compared to 18.5% (12/65) in the midodrine arm (p=0.07). Twelve percent (8/65) in the placebo arm required 2 paracenteses compared to 1.5% (1/65) in the midodrine arm (p=0.01). The amount of fluid removed was similar between the groups (placebo arm: 3763.6±1304.7 vs. 3458.3±1346.6 mL; p=0.52).

Change in MAP at 1 month and effect on PRA

The baseline MAP in the placebo arm (82±5.2 mm Hg) and midodrine arm (81.3±4.2 mm Hg) was comparable (p=0.44). At day 30, the MAP in the placebo arm was 83.9±8 mm Hg compared to 88.2±8.5 in the midodrine arm (p=0.007). At day 90, MAP was comparable between the 2 groups (Table 2). The mean baseline PRA was 19.7±0.58 in the placebo arm compared to 20.1±0.89 in the midodrine arm (p=0.32). At day 30, PRA levels remained the same in the placebo arm (20±1.8; delta change: 0.22±1.6) while the PRA levels were reduced in the midodrine arm (18.5±2.2; delta change: −1.6±2.2) (p=0.09) (Figure 3).

TABLE 2.

Blood pressure at baseline, day 30 and day 90

	Placebo	Midodrine	p
Blood pressure (baseline)
SBP	110.7±8.4	109.4±6.7	0.31
DBP	67.6±5.5	67.3±4.7	0.74
MAP	82±5.2	81.3±4.2	0.44
Blood pressure (on day 30)
SBP	111.6±12.7	117.4±11.4	0.01
DBP	70.1±8.6	73.5±8.1	0.03
MAP	83.9±8	88.2±8.5	0.003
Blood pressure (on day 90)
SBP	116±6.7 (n=47)	118.4±7.6 (n=48)	0.11
DBP	73.5±5.4 (n=47)	75.2±5.6 (n=48)	0.13
MAP	87.7±5 (n=47)	89.6±5.5 (n=48)	0.08

Note: All values in mmHg.

Abbreviations: DBP, diastolic blood pressure; MAP, mean arterial pressure; SBP, systolic blood pressure.

FIGURE 3.

Delta change in plasma renin activity at day 30 from baseline.

Cirrhosis complications, adverse events, and diuretic-related complications

Seventy-four percent (48/65) in the placebo arm and 81.5% (53/65) in the midodrine arm developed any adverse events/complications (p=0.4). Similar proportion of patients developed AKI (27.7% vs. 23.1%; p=0.68), hyponatremia (32.3% vs. 30.8%; p=1), HE (23.1% vs. 13.8%; p=0.25) and infections (26.2% vs. 23.1%; p=0.83) (Table 3 and Supplemental Table S2, http://links.lww.com/HC9/C349). Twenty-nine percent (19/65) in the placebo arm and 24.6% (17/65) in the midodrine arm developed diuretic-induced complications (p=0.55). The maximum serum creatinine (placebo arm: 2.5 [1.6–3.5] vs. 2.23 [1.7–3.7] mg/dL; p=0.88) and the lowest sodium (placebo: 127 [115–129) meq/dL vs. midodrine: 127 [109–129] meq/dL; p=0.83) were similar between the 2 groups. The stage of AKI according to KDIGO was also between the 2 groups (AKI 1, 2, 3: placebo: 16.7%, 33.3%, 50% vs. 20%, 20%, 60%; p=0.69). Spontaneous bacterial peritonitis, pneumonia, and urinary tract infection were common sites of infection in both groups. MELD score at day 30 (placebo: 22.8±5.9 vs. midodrine: 22.1±4.9; p=0.46) and day 90 (placebo: 18.5±5.1 vs. midodrine: 18.8±5.2; p=0.8) were comparable.

TABLE 3.

Cirrhosis-related and diuretic-related complications were noted among the 2 groups

	Placebo (n=65)	Midodrine (n=65)	p
HE	15 (23.1)	9 (13.8)	0.25
Infections	17 (26.2)	15 (23.1)	0.83
SBP	4	5
Pneumonia	4	3
Pneumonia and SBP	0	1
UTI	3	3
SSTI	3	2
Occult	2	1
Spontaneous bacteremia	1	0
AKI	18 (27.7)	15 (23.1)	0.68
KDIGO stage 1	3 (16.7)	3 (20)	0.69
Stage 2	6 (33.3)	3 (20)
Stage 3	9 (50)	9 (60)
Hyponatremia (<130)	21 (32.3)	20 (30.8)	1
125–129	15 (23.1)	14 (21.5)
<125	6 (9.2)	6 (9.2)
Hypokalemia (<3.5 meq/dL)	13 (20)	7 (10.8)	0.22
Potassium (<3 meq/dL)	8 (12.3)	1 (1.5)	0.03
Potassium (≥6 meq/dL)	2 (3.1)	2 (3.1)	1
Diuretic-induced complications	Placebo (n=65)	Midodrine (n=65)
Total	19 (29.2)	17 (26.1)
AKI	3	6
Hyponatremia	3	4
Cramps	2	2
Hypokalemia	1	0
Hyperkalemia	0	1
HE	0	1
Hypokalemia+HE	5	1
AKI+hyponatremia	1	0
AKI+hypokalemia	1	0
Hyponatremia+HE	0	1
Hyponatremia+hyperkalemia	0	1
Hyponatremia+hyperkalemia+cramps	2	0
AKI+hyponatremia+hypokalemia	1	0

Abbreviations: AKI, acute kidney injury; KDIGO, Kidney Disease: Improving Global Outcomes; SBP, spontaneous bacterial peritonitis; SSTI, skin and soft tissue infection; UTI, urinary tract infection.

Hospitalizations

Twenty-six patients (40%) in the placebo arm and 25 (38.5%) in the midodrine arm required hospitalizations (p=0.2). The median duration of hospitalization was 4 (1–20) days in the placebo arm and 6 (1-23) days in the midodrine arm (p=0.1).

Drug-related adverse events

Twenty-two percent (14/65) in the placebo arm and 30.8% (20/65) in the midodrine arm developed adverse events due to the drug (p=0.31). Five patients in the placebo arm discontinued drugs (abdominal pain-2, itching-2, and itching and rash over lower limbs-1). Four in the midodrine arm discontinued the drug (abdominal pain-1; fatigue and body pains-1; stage 2 hypertension-1; urinary retention-1). The remaining 16 patients (who had stage 1 hypertension) in the midodrine arm tolerated lower doses of midodrine. The most common adverse event was a rise in blood pressure in both groups (7 in the placebo arm and 17 in the midodrine arm) (Supplemental Table S3, http://links.lww.com/HC9/C349).

Subgroup analysis of patients who received albumin

Twenty-five patients received only a placebo, 40 received only albumin, 26 received midodrine alone, and 39 received a combination of midodrine and albumin (Supplemental Table S4, http://links.lww.com/HC9/C349). Survival was similar across the 4 groups, regardless of albumin addition (Supplemental Figure S2, http://links.lww.com/HC9/C349). The control of ascites was noted in 4% (1/25) in the placebo-only arm, 12.5% (5/40) in the albumin-only arm, 11.5% (3/26) in the midodrine-only arm and 25.6% (10/39) in the midodrine and albumin arm at day 30 (p=0.09). Similarly, the control of ascites was noted in 40% (10/25) in the placebo-only arm, 47.5% (19/40) in the albumin-only arm, 46.5% (12/26) in the midodrine-only arm and 59% (23/39) in the midodrine and albumin arm at day 90 (p=0.49). Similar complications were noted among the four groups, except the proportion of patients undergoing ascitic tapping, which was lower in the albumin group (placebo: 60% vs. midodrine+albumin:15.4% vs. albumin alone: 17.5% vs. midodrine alone: 23.1%; p<0.001) (Supplemental Table S5, http://links.lww.com/HC9/C349).

Impact of AAH on outcomes

Fifty-two percent (34/65) in the placebo arm and 47.7% (31/65) in the midodrine arm had AAH. TFS was 73.5% (25/34) in the AAH group compared to 71% (22/31) in the non-AAH group receiving placebo (p=1). Similarly, TFS was 71% (22/31) in the AAH group and 76.5% (26/34) in the non-AAH group in those receiving midodrine (p=0.77). Ascites control at day 90 was 47.1% (16/34) in the AAH group compared to 41.9% (13/31) in the non-AAH group in the placebo arm (p=0.8). Ascites control in those receiving midodrine was also similar in AAH (45.2%; 14/31) and non-AAH groups (61.8%; 21/34; p=0.21).

Impact of ACLF grade and MAP on outcomes

The proportion of patients with ACLF grade 1, 2, and 3 was similar among the 2 groups. There was no difference in ascites control and TFS at days 30 and 90 in patients receiving placebo and midodrine (Supplemental Table S6, http://links.lww.com/HC9/C349). The proportion of patients with low MAP was similar between the 2 groups (24.6% [n=16/65] in the placebo group vs 38.5% [25/65] in the placebo group; p=0.13). In patients with low MAP, midodrine improved ascites control at day 30 (0% [0/16] vs. 28% [7/25]; p=0.03) but had no effect on ascites control at day 90 (31.3% [5/16] vs. 52% [13/25]; p=0.21). TFS at days 30 and 90 was similar between the 2 groups (Supplemental Table S7, http://links.lww.com/HC9/C349).

DISCUSSION

In this double-blind, placebo-controlled randomized trial, we noted that in patients with high MELD, ACLF, and ascites who met the APASL criteria for type A ACLF: (1) midodrine did not provide any survival benefit, (2) midodrine did not lead to ascites control, (3) midodrine increased the MAP without any significant change in PRA levels, and (4) midodrine was well tolerated but did not reduce the complications of diuretics and incidence of infections or hospitalizations.

ACLF is a syndrome driven by impaired inflammatory response to a trigger, usually alcohol or viruses, and is associated with high short-term mortality. ²⁰ Improving survival in these patients is critical, and several measures have been assessed. These measures usually focus on preventing infections, a major reason for mortality. Etiology-agnostic measures evaluated in ACLF include granulocyte colony-stimulating factor,^21,22 prophylactic antibiotics, ¹⁹ omega-3 fatty acids, ²³ nonselective beta-blockers, ²⁴ fecal microbiota transplantation, ²⁵ and plasma exchange. ²⁶ Ascites is common in ACLF, and hemodynamic compromise increases the risk of renal failure and mortality. ²⁷ Therefore, we aimed to assess the role of midodrine in improving survival by controlling ascites and preventing hemodynamic alterations.^1,10 Midodrine improved MAP but did not provide any survival benefit, reduce the incidence of ascites complications, or improve ascites control. Midodrine is known to increase the MAP and has been reported to reduce the portal pressure, although the mechanism remains unclear. ²⁸ Previously, a 5 mm Hg increase in MAP was reported to be associated with improved renal recovery in patients with ACLF and AKI. ²⁹ Midodrine increased the MAP by 7 mm Hg, which was probably insufficient to improve the long-term outcomes of ACLF. Possible explanations for the lack of benefit in the current study include lower midodrine doses and a relatively sicker patient population. The lower dose used in the current study was intended to reflect clinical practice and avoid a higher pill burden. We noted higher survival in both the midodrine and placebo arms compared to those included in the sample size calculation. ¹⁹ This may be due to inclusion of relatively stable low MELD ACLF and stringent exclusion criteria in current study compared to our previous study on ACLF and antibiotics. ¹⁹ Further, post hoc power based on the observed TFS was low (<10%), indicating that larger sample sizes would be needed to confirm smaller effects. Lastly, the lack of survival benefit in the current study may also stem from the inflammatory milieu of ACLF overriding the hemodynamic benefits of the midodrine, as infections and organ failures dominated mortality, mirroring real-world data on ACLF. ³⁰ We also did not measure portal pressure or other inflammatory biomarkers (cytokines or von Willebrand factor), which could have provided additional insights into the mortality and hemodynamics.

There are a few differences between the MACTH trial and the MIDAS trial, although both highlight the lack of survival benefit with midodrine. ¹¹ MACTH included patients with decompensated cirrhosis (awaiting liver transplant) who had a MELD score of 16–17, compared to a mean of 23 in the current study, who met the ACLF type A criteria in the current study. The median dose and duration of midodrine were 23 mg/d for 63 days in the MACTH trial, compared to 15 mg/d for 28 days in the current study. Although the combination of midodrine and albumin attenuated the severity of AKI and hyponatremia in the MACTH trial, such effects were not observed in our study. In the current study, approximately 95 g of albumin was administered over 6 weeks, whereas in the MACTH trial, approximately 360 g was administered over 9 weeks (40 g every 2 wk). The use of albumin for long-term and hospitalized patients is quite common in clinical practice.^31,32 Contrasting reports suggest long-term albumin therapy in patients with cirrhosis to be beneficial due to its immunomodulatory effects and volume expansion mechanism.^33–36 Long-term albumin therapy has been reported to be associated with a reduction in the incidence of infections, HE, AKI, and ascites-related complications.^34,36,37 Although albumin addition did not provide a survival benefit in ACLF, it did reduce complications. In our cohort, a higher proportion of patients in the albumin group received diuretics and tolerated higher diuretic doses. Despite this, the incidence of AKI was lower in the albumin group, and the proportion of patients requiring paracentesis was lower as well. While there have been no studies on albumin dosing in ACLF as of now, the dose chosen in our study was arbitrary and not protocolized, which is a major limitation. However, we believe this study opens new avenues for assessing the role of albumin in ACLF.

The study had several limitations, including single-center recruitment, modest power for survival benefit and exploratory analysis, interobserver variability on USG to identify ascites control, the absence of objective evaluation of dietary salt restriction, and variable albumin use. Midodrine was used for 30 days; it is unknown whether a longer duration (60–90 days) or higher doses would have provided additional benefit in this population. Lastly, the included patients were outpatients with ACLF type A (APASL) and lower MELD scores, and the results may not be generalizable to high MELD ACLF type B (EASL-CLIF and NACSELD) patients, who are typically in the inpatient cohort.

In conclusion, oral midodrine, although safe in patients with ACLF, cannot be routinely recommended for all outpatients based on the findings of this study. Future multicenter trials should standardize albumin use and include biomarkers to better define the role of midodrine in patients with high MELD scores and ACLF.

Supplementary Material

hc9-10-e0966-s001.docx ^{(7.6MB, docx)}