1
Management of AscitesManagement of Ascites
I have no disclosures to make relative to my presentation. I have no disclosures to make relative to my presentation.
Guadalupe García-Tsao, MDProfessor of Medicine
Yale University
Chief, Digestive Diseases SectionVA-CT Healthcare System
Guadalupe García-Tsao, MDProfessor of Medicine
Yale University
Chief, Digestive Diseases SectionVA-CT Healthcare System
Cirrhosis Heart failureHeart failure
Peritoneal tuberculosisPeritoneal tuberculosis
Cirrhosis is the Most Common Cause of AscitesCirrhosis is the Most Common Cause of Ascites
Others Pancreatic Budd-Chiari syndrome Nephrogenic ascites
Others Pancreatic Budd-Chiari syndrome Nephrogenic ascites
Peritoneal malignancyPeritoneal malignancy
CIRRHOSIS IS THE MOST COMMON CAUSE OF ASCITES
Source of the main 3 causes of ascitesSource of the main 3 causes of ascites
Entity Source Pathophysiology
Cirrhosis Hepatic sinusoid
Fibrosis and nodules causing sinusoidal and post-sinusoidal obstruction
Heart failure Hepatic sinusoid
Congestion of liver due to right heart failure (post-hepatic block)
Peritoneal malignancy/TB
Peritoneum Inflammation or infiltrationof the peritoneum
Patients with cirrhotic ascites have an HVPG of at
least 12 mmHg (nl 3-5)Morali et a. J Hepatol 2002
2
(out of the sinusoid)(out of the sinusoid) (into the sinusoid)(into the sinusoid)
SINhydSINhyd PERoncPERonc++ SINoncSINonc PERhydPERhyd== ++
SINhydSINhyd ==
HVPGHVPG ==
SINoncSINonc PERoncPERonc--
SerumalbuminSerum
albuminAscitesalbuminAscitesalbumin
-- = SAAG= SAAG
Rationale Behind the Serum-Ascites Albumin Gradient (SAAG)
Rationale Behind the Serum-Ascites Albumin Gradient (SAAG)
Hoefs J, J Lab Clin Med 1983; 102:260Hoefs J, J Lab Clin Med 1983; 102:260
The Serum-Ascites Albumin Gradient (SAAG) Correlates With Sinusoidal Pressure
The Serum-Ascites Albumin Gradient (SAAG) Correlates With Sinusoidal Pressure
SAAG (g/dL)SAAG (g/dL)
HVPG (mmHg)HVPG (mmHg)
r = 0.73r = 0.73
3030
2020
1010
00
00 1.01.0 2.02.0 3.03.0
1111
1.11.1
Cirrhotic ascitesCirrhotic ascites Cardiac ascitesCardiac ascites
Peritoneal malignancyPeritoneal
malignancy
1.11.1
4.04.0
3.03.0
2.02.0
1.01.0
00
Serum –ascites albumin
gradient (g/dL)
Serum –ascites albumin
gradient (g/dL)
7.07.0
Serum-Ascites Albumin Gradient and Ascites Protein Levels in the Most Common Causes of Ascites
Serum-Ascites Albumin Gradient and Ascites Protein Levels in the Most Common Causes of Ascites
SERUM-ASCITES ALBUMIN GRADIENT (SAAG) AND ASCITES PROTEIN LEVELS IN THE MOST COMMON CAUSES OF ASCITES
SAAG is an indicator of sinusoidal pressure. If >1.1 ascites is coming
from the sinusoid
SAAG is an indicator of sinusoidal pressure. If >1.1 ascites is coming
from the sinusoid
3
THE PERMEABILITY OF THE HEPATIC SINUSOID VARIES IN HEALTH AND DISEASE
In cirrhosis, the hepatic sinusoid
is less leaky
In cirrhosis, the hepatic sinusoid
is less leaky
The Permeability of the Hepatic Sinusoid Varies in Health and Disease
The Permeability of the Hepatic Sinusoid Varies in Health and Disease
HepatocytesHepatocytes
The normal sinusoid is “leaky”
The normal sinusoid is “leaky”
SinusoidSinusoid
SinusoidSinusoid
fibrous tissue deposition “capillarization” of sinusoidfibrous tissue deposition
“capillarization” of sinusoid
no basement membrane
no basement membrane
Cirrhotic ascitesCirrhotic ascites Cardiac ascitesCardiac ascites
Peritoneal malignancyPeritoneal
malignancy
1.11.1
4.04.0
3.03.0
2.02.0
1.01.0
00
Serum –ascites albumin
gradient (g/dL)
Serum –ascites albumin
gradient (g/dL)
Serum-Ascites Albumin Gradient and Ascites Protein Levels in the Most Common Causes of Ascites
Serum-Ascites Albumin Gradient and Ascites Protein Levels in the Most Common Causes of Ascites
Runyon, Ann Intern Med 1992; 117:215Runyon, Ann Intern Med 1992; 117:215
SERUM-ASCITES ALBUMIN GRADIENT (SAAG) AND ASCITES PROTEIN LEVELS IN THE MOST COMMON CAUSES OF ASCITES
SAAG is an indicator of sinusoidal pressure. If >1.1 ascites is coming
from the sinusoid
SAAG is an indicator of sinusoidal pressure. If >1.1 ascites is coming
from the sinusoid
Ascitic fluid total protein
(g/dL)
Ascitic fluid total protein
(g/dL)
7.07.0
5.05.0
3.03.0
2.02.0
00
2.52.5
(75)(75)
Ascites protein is an indicator of leakiness of sinusoid, >2.5 the sinusoid is leaky (i.e. normal)
Ascites protein is an indicator of leakiness of sinusoid, >2.5 the sinusoid is leaky (i.e. normal)
CONDITIONCONDITION SAAGSAAGASCITESPROTEINASCITESPROTEIN
Cirrhosis
Peritoneal malignancy
Heart failure
Cirrhosis
Peritoneal malignancy
Heart failure
high low
low high
high high
high low
low high
high high
CutoffCutoff 1.1 g/dL1.1 g/dL 2.5 g/dL2.5 g/dL
SAAG and ascites total protein can establish the differential among the main causes of ascites
SAAG and ascites total protein can establish the differential among the main causes of ascites
4
Serum BNP has a higher diagnostic accuracy for cardiac ascites than SAAG/ascites protein
Serum BNP has a higher diagnostic accuracy for cardiac ascites than SAAG/ascites protein
Farias et al. Hepatology 2014; 59:1043-51
Test LR (+)(rules in)
LR(-) (rules out)
SAAG >1.1; prot >2.5 9.63
Serum BNP >364 pg/mL 168.09
SAAG <1.1; prot < 2.5 1.272
Serum BNP < 182 pg/mL 0.000
Patients with new onset ascites
364
182
Rules in
Rules out
Compensatedcirrhosis
Compensatedcirrhosis
Decompensatedcirrhosis
Decompensatedcirrhosis
Natural History of Chronic Liver DiseaseNatural History of Chronic Liver Disease
Ascites VH Encephalopathy
Ascites VH Encephalopathy
Chronic liver
disease
Chronic liver
disease
VH= variceal hemorrhage
DeathDeath
In a cohort of patients with compensated cirrhosis, ascites was the most common decompensating event
0 24 48 72 96 120 144 168 192 216 240
0.00
0.25
0.50
0.75
1.00
AscitesAscites
HEHE
JaundiceJaundice
VHVH
DecompensationDecompensation
months
D’Amico G. Gastroenterology 2001; 120: A2D’Amico G. Gastroenterology 2001; 120: A2
5
Intrahepatic resistance
Portal (sinusoidal) hypertension
Cirrhosis
Ascites
Splanchnic / systemic vasodilatationSplanchnic / systemic vasodilatation
Effective arterial blood volume
Sodium retention
Activation of neurohumoral systems
DiureticsDiuretics
Spironolactone is More Effective Than Furosemide in Uncomplicated AscitesSpironolactone is More Effective Than Furosemide in Uncomplicated Ascites
Response No response Total
Spironolactone 18 1 19(150-300 mg/d)
Response No response Total
Spironolactone 18 1 19(150-300 mg/d)
Perez-Ayuso et al. Gastroenterology 1983; 84:961 Perez-Ayuso et al. Gastroenterology 1983; 84:961
SPIRONOLACTONE IS MORE EFFECTIVE THAN FUROSEMIDE IN CIRRHOTIC PATIENTS WITH ASCITESSPIRONOLACTONE IS MORE EFFECTIVE THAN FUROSEMIDE IN CIRRHOTIC PATIENTS WITH ASCITES
Furosemide 11 10 21(80-160 mg/d)Furosemide 11 10 21(80-160 mg/d)
Treatment of ascitesTreatment of ascites
Not an emergency, treat ascites in a stepwise unhurried manner
Other complications (GI bleed AKI, infection) are absent or have resolved
If patient uncomfortable large volume paracentesis
Treatment aimed at achieving a negative sodium balance
Not an emergency, treat ascites in a stepwise unhurried manner
Other complications (GI bleed AKI, infection) are absent or have resolved
If patient uncomfortable large volume paracentesis
Treatment aimed at achieving a negative sodium balance
6
Less frequent dose reductions are needed when spironolactone is started alone
Spironolactone + Furosemide
(n=50)
Spironolactone + Furosemide
(n=50)
Spironolactone alone*(n=50)
Spironolactone alone*(n=50)
Response Rate 94% 98%
Time to Response 12.8 days 12.3 days
Dose reductionneeded 34% 68%
Response Rate 94% 98%
Time to Response 12.8 days 12.3 days
Dose reductionneeded 34% 68%p=0.002p=0.002
Santos et al., J Hepatol 2003; 39:187Santos et al., J Hepatol 2003; 39:187 * Followed by furosemide if necessary* Followed by furosemide if necessary
In addition to spironolactone-based diuretics….
In addition to spironolactone-based diuretics….
Salt restriction (2g/day = ~90mEq/day) Do not compromise nutritional status
Avoid non-steroidal anti-inflammatory drugs
No water restriction unless serum Na <130 mEq/L
Low threshold to perform a diagnostic paracentesis to investigate SBP
Salt restriction (2g/day = ~90mEq/day) Do not compromise nutritional status
Avoid non-steroidal anti-inflammatory drugs
No water restriction unless serum Na <130 mEq/L
Low threshold to perform a diagnostic paracentesis to investigate SBP
Follow weight and labs (BUN, creatinine, lytes)
Weight loss goals 2-3 lb a week; no more than 1 lb / day
If no weight loss Make sure patient is not on NSAIDs Check urine Na. If any of the following, patient
is eating too much salt: > 50 mEq/L or greater than daily Na intake Spot UNa >UK (correlates with a 24-hour sodium
excretion >78 mEq/L)
Follow weight and labs (BUN, creatinine, lytes)
Weight loss goals 2-3 lb a week; no more than 1 lb / day
If no weight loss Make sure patient is not on NSAIDs Check urine Na. If any of the following, patient
is eating too much salt: > 50 mEq/L or greater than daily Na intake Spot UNa >UK (correlates with a 24-hour sodium
excretion >78 mEq/L)
Management of AscitesManagement of Ascites
7
Hepatic HydrothoraxHepatic Hydrothorax
Occurs in ~6% of patients with cirrhosis Occurs in ~6% of patients with cirrhosis Krok KL, Cardenas A. Semin Respir Crit Care Med 2012; 33: 3-10.
Due to trans-diaphragmatic movement of fluid from the peritoneum to the pleural space through diaphragmatic defects
Due to trans-diaphragmatic movement of fluid from the peritoneum to the pleural space through diaphragmatic defects
Management same as for cirrhotic ascites Management same as for cirrhotic ascites
Intrahepatic resistance
Portal (sinusoidal) hypertension
Cirrhosis
Ascites
Splanchnic / systemic vasodilatationSplanchnic / systemic vasodilatation
Effective arterial blood volume
Sodium retention
Refractory Ascites
Activation of neurohumoral systems
Large volume-paracentesis (LVP):• Local therapy• Recurrence of ascites is the rule• May be associated with post-paracentesis circulatory dysfunction
8
LVP Without Albumin Leads to Increases in Renin, Renal Failure and HyponatremiaLVP Without Albumin Leads to Increases in Renin, Renal Failure and Hyponatremia
BeforeBefore BeforeBeforeAfterAfter AfterAfter
AlbuminAlbumin No albuminNo albumin
1212
88
44
00
Plasma renin activity (ng/mL/h)Plasma renin activity (ng/mL/h) Renal failure / HyponatremiaRenal failure / Hyponatremia
AlbuminAlbumin No albuminNo albumin
2020
1515
1010
00
55
p<0.1p<0.1
p<0.1p<0.1
nsns
Post-paracentesis circulatory
dysfunction (PCD)
Post-paracentesis circulatory
dysfunction (PCD)
%%
Gines et al., Gastroenterology 1988; 94:1493Gines et al., Gastroenterology 1988; 94:1493
LVP WITHOUT ALBUMIN LEADS TO INCREASES INCIDENCE OF POST-PARACENTESIS CIRCULATORY DYSFUNCTION (PCD)
Consequences of post-paracentesis circulatory dysfunction (PCD)
Consequences of post-paracentesis circulatory dysfunction (PCD)
Shorter time to ascites recurrence
Higher incidence of hyponatremia and renal dysfunction
Higher mortality
Shorter time to ascites recurrence
Higher incidence of hyponatremia and renal dysfunction
Higher mortality
Gines et al., Gastroenterology 1996; 111:1002; Ruiz del Arbol et al., Gastroenterology 1997; 113:579 Gines et al., Gastroenterology 1996; 111:1002; Ruiz del Arbol et al., Gastroenterology 1997; 113:579
CONSEQUENCES OF POST-PARACENTESIS CIRCULATORY DYSFUNCTION (PCD)
Post-paracentesis circulatory dysfunction (PCD) is lowest in patients receiving albumin after LVPPost-paracentesis circulatory dysfunction (PCD) is lowest in patients receiving albumin after LVP
Development of PCD
Development of PCD
%%
Ascites removedAscites removedOverallOverall <5-6 L<5-6 L >5-6 L>5-6 L
7070
6060
5050
4040
3030
2020
1010
00
No expanderSalineSynthetic expanderAlbumin*
No expanderSalineSynthetic expanderAlbumin*
Gines et al., Gastroenterology 1988; 94:1493; Gines et al., Gastroenterology 1996; 111:1002;Sola-Vera et al., Hepatology 2003; 37:1147
Gines et al., Gastroenterology 1988; 94:1493; Gines et al., Gastroenterology 1996; 111:1002;Sola-Vera et al., Hepatology 2003; 37:1147
*6-8 g per liter of ascites removed*6-8 g per liter of ascites removed
9
Intrahepatic resistance
Portal (sinusoidal) hypertension
Cirrhosis
Ascites
Splanchnic / systemic vasodilatationSplanchnic / systemic vasodilatation
Effective arterial blood volume
Renal vasoconstriction
Water retention
Sodium retention
Refractory Ascites
Hypo-natremia
Hepatorenal syndrome
Activation of neurohumoral systems
ALBUMINALBUMIN
LVPLVP
Other volume expanders?Vasoconstrictors?
Favors albumin
Favors control
Compared to alternative treatment, albumin reduces the rate of PCD
Compared to alternative treatment, albumin reduces the rate of PCD
Bernardi et al. Hepatology 2012;55:1172.Bernardi et al. Hepatology 2012;55:1172.
PCD : 18% (2/11) with Albumin25% (2/8) with Octreotide/Midodrine
(p=0.574)
Recurrence of ascites is no different in patients treated with LVP + albumin vs. octreotide/midodrine
Recurrence of ascites is no different in patients treated with LVP + albumin vs. octreotide/midodrine
Bari et al. Accepted Clin Gastroenterol Hepatol.Bari et al. Accepted Clin Gastroenterol Hepatol.
10
Transjugular Intrahepatic Portosystemic ShuntTransjugular Intrahepatic Portosystemic Shunt
Hepatic veinHepatic vein
Portal veinPortal veinSplenic veinSplenic vein
Superior mesenteric veinSuperior mesenteric vein
TIPSTIPS
THE TRANSJUGULAR INTRAHEPATIC PORTOSYSTEMIC SHUNT
Recurrence of ascitesRecurrence of ascites
Lebrec (1996) 20
EncephalopathyEncephalopathy
DeathDeath
Death (excluding Lebrec)Death (excluding Lebrec)
Better TIPSBetter TIPS Better LVPBetter LVP
.01.01
0.14 (0.08-0.26)0.14 (0.08-0.26)
2.34 (1.41-3.87)2.34 (1.41-3.87)
0.90 (0.44-1.81) Heterogeneity 2 p=0.050.90 (0.44-1.81) Heterogeneity 2 p=0.05
.1.1 .5.5 11 55 1010
0.74 (0.40-1.37)0.74 (0.40-1.37)
In refractory ascites, TIPS is more effective than LVP in preventing ascites recurrence In refractory ascites, TIPS is more effective than LVP in preventing ascites recurrence
D’Amico et al. Gastroenterology 2005; 129:1282D’Amico et al. Gastroenterology 2005; 129:1282
Odds ratioOdds ratio
Salerno et al. Gastroenterology 2007;133:825–834Salerno et al. Gastroenterology 2007;133:825–834
In a meta-analysis of individual patient data, survival was better with TIPS than LVP
In a meta-analysis of individual patient data, survival was better with TIPS than LVP
SurvivalSurvival EncephalopathyEncephalopathy
P=0.005
p=0.36
Greater survival benefit in patients treated with TIPS who had a MELD
score <15
Greater survival benefit in patients treated with TIPS who had a MELD
score <15
TIPS= transjugular intrahepatic portosystemic shunt
LVP= large-volume paracentesis
*individual data meta-analysis
11
Refractory hepatic hydrothoraxRefractory hepatic hydrothorax
A trial of in-hospital diuretic therapy should be attempted
Serial thoracenteses – may be required too frequently
Chest tube or indwelling catheter should not be placed ( infection, AKI)
TIPS may need to be considered earlier Clinical response (67%) and survival are
also associated with pre-TIPS MELD <15
A trial of in-hospital diuretic therapy should be attempted
Serial thoracenteses – may be required too frequently
Chest tube or indwelling catheter should not be placed ( infection, AKI)
TIPS may need to be considered earlier Clinical response (67%) and survival are
also associated with pre-TIPS MELD <15
Dhanasekaran et al. Am J GE 2010.
Peritoneo-Venous Shunt (PVS) is Useful in the Treatment of Refractory Ascites
Use of jugular vein will hinder TIPS placement
Use of jugular vein will hinder TIPS placement
Intraabdominal adhesions may complicate liver
transplant surgery
Intraabdominal adhesions may complicate liver
transplant surgery
One-way valve
One-way valve
Indicated in malignant ascites or patients who are not transplant or TIPS
candidates
ALFA pump transfersascites into the bladder
Pilot safety study of Automated Low-Flow pump for refractory Ascites (ALFA) (n=40)Pilot safety study of Automated Low-Flow pump for refractory Ascites (ALFA) (n=40)
Placed under general anesthesia
6-month followup
LVP 3.4 0.2 per month
Infections antibiotic prophylaxis (76%42%)
Catheter dislodgement/problems (10/40=25%)
Surgical complications (5/40)
Progressive decrease in serum albumin
13 early termination, 8 died, 2 txp
Placed under general anesthesia
6-month followup
LVP 3.4 0.2 per month
Infections antibiotic prophylaxis (76%42%)
Catheter dislodgement/problems (10/40=25%)
Surgical complications (5/40)
Progressive decrease in serum albumin
13 early termination, 8 died, 2 txp
Bellot et al. J Hepatol 2013;58:922-7
12
Hernandez-Gea et al. Am J Gastroenterol 2012; 107:418-27Hernandez-Gea et al. Am J Gastroenterol 2012; 107:418-27
In patients with large varices that have not bled, a decrease in HVPG >10% leads to less ascites, RA and HRS
Ascites
Hepatorenalsyndrome
Refractory ascites
Cirrhotic ascitesCirrhotic ascites
The most common decompensating event in cirrhosis
It is not an emergency unless complicated by infection or hepatorenal syndrome
Ideal treatment strategies should be based on its pathophysiology Increase sodium excretion
Decrease sinusoidal pressure
Remove fluid while replenishing intravascular volume
The most common decompensating event in cirrhosis
It is not an emergency unless complicated by infection or hepatorenal syndrome
Ideal treatment strategies should be based on its pathophysiology Increase sodium excretion
Decrease sinusoidal pressure
Remove fluid while replenishing intravascular volume