Page 1 of 33
TECHNICAL APPENDIX:SUMMARY OF LITERATURE REVIEW
CHAPTER 1— ASCITES
Ascites is the most common complication of cirrhosis. Approximately 50% of patients with “compensated” cirrhosis, i.e., without having developed one of these complications, develop ascites during 10 years of observation. Once patients develop ascites, the probability of survival at 5 years is close to 50%.10 The first section will discuss the indicators applicable to initial evaluation and management of patients with ascites.
1A.DIAGNOSIS AND INITIAL EVALUATION
Wong et al recently conducted a comprehensive review on the utility, safety, and accuracy in cirrhotic patients with ascites.11 This review provided data for several of the QIs that are pertinent to the diagnosis and evaluation of this subset of patients with cirrhosis.
Abdmonial paracentesis at the time of ascites diagnosis.
Corresponding guideline
- Abdominal paracentesis should be performed and ascitic fluid should be obtained from all patients with clinically apparent new ascites.
- The initial laboratory investigation of ascitic fluid should include an ascitic fluid cell count and differential, total protein, and serum-ascites albumin gradient (SAAG).
Abdominal paracentesis with appropriate ascitic fluid analysis is the most rapid method of diagnosing the cause of ascites. Fluid due to portal hypertension (cause of ascites in cirrhosis) can be readily differentiated from fluid due to other causes. Wong et al found 4 prospective trials that tested the accuracy of theSAAG in diagnosing portal hypertension.11 In all these trials, invasive measures of portal pressure wasused as the reference standard. The diagnostic accuracy of the tests aresummarized as likelihood ratio or LRs, the likelihood that a given test result would occur in a patient with the target disorder as opposed to a patient without. Pooling the1029 samples from the 4 studies, an abnormal SAAG of 1.1 g/dL or greater increased the likelihood of portal hypertension (LR, 4.6; 95% confidence interval [CI],1.6-12.9). A normal SAAG < 1.1 g/dL was found to be very useful in decreasing the likelihood of portal hypertension diagnosis (summary LR, 0.06; 95% CI, 0.02-0.20).
Although several prospective and retrospective studies of diagnostic and therapeutic paracentesis report on significant bleeding and death, the systematic review found these studies to be too heterogeneous to allow meaningful metaanalysis.11 The overall rate of significant bleeding ranged from0%to 2.7% and of death from 0% to 17%. Variability in bleeding rates was likely influenced by preexisting renal dysfunction in one study. The heterogeneity in the risk of death was primarily due to a retrospective study of therapeutic paracentesis in hospitalized patients at a tertiary care hospital. The review found that, in general, paracentesis is a safe procedure with minimal risks. We did not find any data that prospectively examined the impact of timely paracentesis on clinical outcomes (e.g., initiation of treatment, length of hospital stay, etc.) in patients with new onset cirrhosis.
Grade II-3
Class I, Level C
Paracentesis at the time of hospital admission
Corresponding guideline
- Abdominal paracentesis should be performed and ascitic fluid should be obtained from all inpatients with clinically apparent ascites.
Infection occurs in as many as 27% of cirrhosis patients admitted to the hospital for evaluation and management of symptoms associated with ascites. In the recent systematic review, Wong et al found 14 prospective studies that evaluated the diagnostic accuracy of the ascitic fluid cell count for diagnosing spontaneous bacterial peritonitis (SBP).11The reference standard for diagnosing spontaneous bacterial peritonitis was a culture with positive results. The review found that the most accurate parameters to diagnose SBP were an ascitic white blood count (WBC) count of greater than 1000 cells/μL (summary LR, 9.1; 95% CI, 5.5- 15.1) and an ascitic PMN count of greater than 500 cells/μL (summary LR, 10.6; 95% CI, 6.1-18.3). While the literature supports a threshold of greater than 500 cells/μL as most accurate, most experts initiate antibiotic therapy when the PMN count is greater than 250 cells/μL, given still the greatly increased likelihood of SBP at this lower PMN count threshold (summary LR, 6.4; 95% CI, 4.6-8.8). Conversely, an ascitic PMN count of 250 cells/μL or less significantly lowers the likelihood of SBP (summary LR, 0.20;95% CI, 0.11-0.37).
In a recent report, 12% of emergency department patients with ascites were found to have SBP on routine paracentesis.12 Although this estimate is lower than what has been reported in the older literature, clinical characteristics and physician assessment were insufficient in the diagnosis or exclusion of spontaneous bacterial peritonitis in this study.12
Given the potential fatality of SBP,13, 14 and the effectiveness of timely treatment, guidelines recommend abdominal paracentesis in all hospitalized patients with ascites to rule out underlying infection.
Grade II-3
Class I, Level C
Replacement of fresh frozen plasma or platelets with paracentesis
Corresponding guideline
- Prophylactic use of fresh frozen plasma or platelets before paracentesis is not recommended.
Wong et al found 2 prospective studies that systematically obtained prothrombin times and platelet counts prior to paracentesis in all patients.11 The first study included both diagnostic and therapeutic procedures in the emergency department under ultrasound guidance. No bleeding complications occurred in any of the 96 diagnostic procedures (95% CI, 0%-3.8%), even though 6 paracenteses were carried out with a platelet count of less than 50 000/mm3and 27 paracenteses were performed with an international normalized ratio (INR) of more than 1.5. Only 2 episodes of minor bleeding occurred among the 314 therapeutic procedures (0.0064%; 95% CI, 0.0008%-0.023%). Both occurred with an INR of 2.5 to 2.9 and platelet count of 50 000 to 99 000/mm3. Neither resulted in hospitalization or transfusion.15 The second study included 1100 large-volume paracenteses performed by trained endoscopy assistants in an outpatient setting. There was no significant bleeding in any patient (95% CI, 0%-0.33%) despite 598 procedures that were performed with a platelet count of less than 50 000/mm3and 292 with an INR of greater than 2.0.16Therefore, preprocedure coagulation studies or replacements are likely unnecessary prior to paracentesis.
Grade II-3
Class III, Level B
1B.TREATMENT OF ASCITES
Treatment of ascites improves patient’s symptoms and health-related quality of life. The cornerstone of ascites treatment is a combination of diuretics and sodium (or salt) restriction.
Sodium restriction and diuretic therapy
Corresponding guideline
- First-line treatment of patients with cirrhosis and ascites consists of sodium restriction and diuretics (oral spironolactone and furosemide).
Diuretics are effective in reducing the volume of ascites in a significant number of patients with normal renal function. In a large multi-center randomized trial a combination of oral spironolactone and furosemide, beginning with 100 mg and 40 mg respectively, was effective in more than 90% of patients with severe ascites.17 In a recent randomized trial, the response rate, the rapidity of ascites mobilization, and the incidence of complications induced by diuretic therapy were similar in patients with moderate ascites managed with dual therapy compared to those managed with spironolactone alone (starting dose of spironolactone,100 mg).18 On the contrary, furosemide alone is less effective and resulted in more hypokalemia than spironolactone alone or in combination with furosemide 19, 20
Gatta et al examined the role of salt restriction alone in the treatment of patients with ascites and found that salt restriction alone induces spontaneous diuresis in a very small proportion (7 of 51, 13.7%) of patients.21 We found 3 controlled trials that evaluated the efficacy of salt restriction in combination with diuretics in the treatment of cirrhotic ascites. All studies allowed patients to be treated with a combination of spironolactone and loop diuretics at standard dosages. In the study by Descos et al, no difference was found in the frequency of total or partial regression of ascites after a month of treatment in patients with salt restriction of 500 gm/dayversus those without salt restriction.22 Similarly, Runyon et al found no difference in the daily weight loss or total dose of diuretics in patients who received low sodium diet (10 mmol/day) compared to those on an unrestricted diet. Gauthier et al reported that although ascites disappeared significantly faster with a salt restricted diet (21 mmol/day), there was no difference in the resolution of ascites at 3 month follow up.23
In summary, these data show that spironolactone alone or in combination with furosemide is effective in reducing the fluid overload in patients with ascites. The combination treatment may expedite diuresis and therefore may be preferable in patients with severe ascites. The usefulness of salt restriction is unclear given the evidence from controlled trials.
Diuretics
Grade I
Class I, level A
Fluid restriction in patients with hyponatremia
Corresponding guideline
- Fluid restriction is not necessary unless serum sodium is less than 120-125 mmol/L.
There are no data to support any particular threshold for initiating water restriction in patients with hyponatremia and cirrhosis. A serum sodium less than 120-125 mmol/L has been suggested as a reasonable threshold by the AASLD guidelines.14 We did not find direct data comparing fluid restriction versus no restriction in patients with hyponatremia and cirrhosis. However, fluid restriction to 1 or 1.5 liters per day was used as a control strategy in 3 recently published randomized controlled trials that evaluated the efficacy of selective antagonists of vasopressin (vaptans) in improving hyponatremia.24-26 Reversal of hyponatremia (defined as serum sodium of >135 mmol/day) was seen in 0% to 26% of patients managed with water restriction alone.24, 25 Of note, there was no significant change in the mean serum sodium levels between enrollment and end of study in all trials. The mean serum sodium at entry was more than 125 mmol/L and results were not stratified based on the degree of baseline hyponatremia. Given these data, it is unclear whether and to what extent fluid restriction results in improvement of hyponatremia in patients with cirrhosis.
Grade III
Class IIb, Level B
Counseling regarding alcohol use in alcohol related liver disease
Corresponding guideline
- Patients with ascites who are thought to have an alcohol component to their liver injury should abstain from alcohol consumption.
Abstinence from alcohol can result in dramatic improvement in the reversible component of alcoholic liver disease. Patients with Child Pugh C cirrhosis due to alcohol and who stop drinking have an approximately 75% 3-year survival, but all those continue to drink die in 3 years.27 Ascites may resolve or become more responsive to medical therapy with abstinence and time. Therefore, providing counseling or therapy to patients with cirrhosis who are currently drinking alcohol should be a recommended component of care in these patients.
Grade II-2
Class I, level C
Colloid replacement with large volume paracentesis
Corresponding guideline
- For large-volume paracentesis (5 liters), an albumin infusion of 8-10 g/L of fluid removed can be considered
Ten to 15% of patients with cirrhotic ascites are resistant to standard medical therapy or develop clinically significant complications of diuretics (i.e., have refractory ascites).10These patients are commonly treated with serial large volume paracentesis. Wong et al found 9 randomized studies that evaluated the use of plasma expanders with large volume paracentesis.11 In all 9 studies, there was no significant difference between therapeutic paracentesis with and without volume expansion with albumin, nor with non-albumin plasma expanders compared with albumin for either hyponatremia or renal impairment. Similarly, there was no significant between group difference in enceophalopathy in 7 studies, with data from 666 procedures, that reported encephalopathy as an outcome. In addition, albumin compared with no plasma expansion or other plasma expanders did not confer any statistically significant difference in the risk of death. Therefore, it is unclear if colloid replacement improves meaningful outcomes in patients with cirrhosis undergoing large volume paracentesis.
Grade II-2
Class II-b, level A
Consideration for transjugular intrahepatic portosystemic shunt in refractory ascites
Corresponding guideline
- Transjugular intrahepatic portosystemic shunt should be considered in appropriately selected patients who meet criteria similar to those of published randomized trials.
Several randomized controlled trials have compared transjugular intrahepatic portosystemic shunt (TIPS) with large volume paracentesis in cirrhotic patients with refractory ascites.28-31All four trials conclude that TIPS reduces the recurrence rates of ascites. Moreover, a recent meta-analysis of individual patient data showed that TIPS significantly improved transplant free survival of cirrhotic patients with refractory ascites compared to large volume paracentesis.32Specifically, the actuarial probability of transplant free survival at 6, 12, 24, and 36 months was 75%, 63%, 9%, and 38% for patients allocated to TIPS and 65%, 52%, 35%, and 29% for patients allocated to paracentesis, respectively.32 These data support the use of TIPS in patients with refractory ascites who meet the criteria similar to those of the above published trials. However, the performance of TIPS is highly dependent on local expertise and this option may not be readily available to all patients with cirrhosis.
Criteria used in randomized trials of TIPS include less than grade 2 hepatic encephalopathy; no portal vein thrombosis; bilirubin <5mg/dl, creatinine <3 mg/dl; absence of liver cancer or hepatic hydrothorax; prothrombin time >40%, platelet count >40,000/mm3; no bacterial infection, cardiac failure, parenchymal renal disease, or alcoholic hepatitis; and Child Pugh score <11.
Grade I
Class I, level B
1C.SPONTANEOUS BACTERIAL PERITONITIS
Treatment of spontaneous bacterial peritonitis (SBP)
Corresponding guideline
- Patients with ascitic fluid PMN counts greater than or equal to 250 cells/mm3 should receive empiric antibiotics therapy e.g., intravenous cefotaxime 2 g every 8 hours.
The prognosis of cirrhotic patients with SBP has improved greatly during the last two decades because of early recognition and treatment with effective and safe antibiotics. Cefotaxime (third-generation cephalosporin) is currently the treatment of choice for SBP, because it is highly effective against most causative organisms and lacks nephrotoxic effect. Moreover, treatment with cefotaxime is associated with an extremely low incidence of superinfections. Using cefotaxime, infection resolution is obtained in 80%–90% and the hospital survival rate has increased to 70%–80% compared with the 0%–40% hospital survival rate in studies published before the introduction of this antibiotic.
Cefotaxime or similar third generation cephalopsporin has been shown to be superior to ampicillin and tobramycin.33 Treatment for 5 days is as efficacious as 10 days treatment in carefully characterized patients with SBP.34 Oral ofloxacin is as effective as intravenous cefotaxime in uncomplicated SBP.35
Grade 1
Class 1, level A
Antibiotics prophylaxis for SBP
Corresponding guidelines
- Short-term inpatient norfloxacin (or bactrim) should be given to prevent bacterial infections in patients with cirrhosis and gastrointestinal hemorrhage; a quinolone antibiotic can be given intravenously while the patient is actively bleeding.
- Patients who have survived an episode of SBP should receive long-term prophylaxis with daily norfloxacin (or trimethoprim/sulfamethoxazole)
- In patients with cirrhosis and ascites but no gastrointestinal bleeding, either short-term (inpatient-only) or long-term outpatient use of daily norfloxacin (or trimethoprim /sulfamethoxazole) can be justified when theascitic fluid total protein is < 1g/dl or serum bilirubin is >2.5mg/dl
We found a meta-analysis of 8 trials that evaluated the efficacy of antibiotic prophylaxis compared with placebo or no antibiotic prophylaxis in 864 cirrhotic patients with gastrointestinal bleeding. Antibiotic prophylaxis for cirrhotic inpatients with gastrointestinal bleeding wasfound to be efficacious in reducing the number of deaths (17% versus 21.5% in antibiotic versus placebo group; relative risk, RR=0.73, 95% CI, 0.55,0.95) as well as the number of bacterial infections (16% versus 42%; RR= 0.4, 95%CI=0.32, 0.51).36 The randomized trials did not stratify patients based on the source of gastrointestinal bleeding. Therefore, it is unclear if the type of gastrointestinal bleeding matter. Similarly, the trials did not present data stratified by severity of liver disease. However, meta-analysis of trials where at least 30% of the patients presented with severe liver disease (Child-Pugh C) showed a tendency towards reduction in mortality and a significant reduction in the incidence of bacterial infections in patients receiving antibiotics. Given these data, all patients with cirrhosis—regardless of the stage of liver disease—presenting with gastrointestinal bleeding of any source are likely to benefit from antibiotic prophylaxis.
Antibiotics were initiated just before or just after the endoscopic procedure. All patients underwent endoscopy within the first 24 hours of presentation in these trials. Therefore, it is prudent to say that antibiotics should be started within the first 24 hours of presentation.
Grade 1
Class I, level A
We found one placebo controlled randomized trial evaluating the efficacy of antibiotic prophylaxis in cirrhotic patients that had a past episode of SBP. Antibiotic prophylaxis with norfloxacin 400 mg/day was found to be efficacious in reducing the number of bacterial infections at the mean follow up of 6.5 months (12% in norfloxacin versus 35% in placebo; RR=0.48, 95% CI0.26, 0.88).37 The overall probability of SBP recurrence at 1 yr of follow-up was 20% in the norfloxacin group and 68% in the placebo group. There was also a trend towards improved patient survival at the end of 1-year (RR=0.70, 95% CI= 0.30, 1.66). The randomized trials did not stratify patients based on the severity of cirrhosis. Regardless, this randomized study supports the use of antibiotic prophylaxis in patients with documented episode of SBP.
Grade 1
Class I, level B
We found a meta-analysis of 4 trials that compared the efficacy of antibiotic prophylaxis compared with placebo or no antibiotic prophylaxis in 143 cirrhotic patients with low protein ascitic fluid.38 The meta-analysis found that antibiotic prophylaxis for 1 to 4 months was effective in reducing the number of deaths (RR=0.61, 95% CI 0.35, 1.09) and bacterial infections (RR=0.0.29, 95% CI 0.17, 0.52). Three of the 4 trials used quinolones (2 used norfloxacin and 1 used ciprofloxacin) whereas one trial tested the use of trimethoprim/sulfamethoxazole in prophylaxis of SBP.
A recent randomized controlled trial compared long-term antibiotic prophylaxis with daily norfloxacin (400 mg) versusplacebo in patients with low protein ascites and advanced liver failure (defined as Child Pugh score of 9 or more with serum bilirubin 3 mg/dl) or impaired renal function (serum creatinine level 1.2 mg/dL, blood urea nitrogen 25 mg/dL, or serum sodium level 130 mEq/L).39Antibiotic prophylaxis improved patient survival at both 3 (94% versus 62% ) and 12 months (60% versus48%). Sixty percent of patients in the control group developed SBP during 12 months follow-up. This risk was reduced to 7% in those receiving norfloxacin. Similarly, there was a significant reduction in the proportion of patients developing hepato-renal syndrome in the treatment versus placebo groups (28% vs 41%). These data support the use of norfloxacin prophylaxis in patients with low protein ascites and advanced liver or renal dysfunction.
Novella et al compared long-term prophylaxis with norfloxacin versus short-term prophylaxis during hospital admissions only.39, 40, 40 After an average follow-up of 43 weeks, they found only one episode of SBP in patients receiving long-term prophylaxis compared to 9 patients (16.9%) in the short-term prophylaxis group. There was no difference in the rates of bacterial infections, hospitalizations, or mortality between the 2 study groups. This study included cirrhotic patients with ascites and low ascitic fluid total protein and/or high serum bilirubin levels, but did not stratify results based on these inclusion criteria.