Diagnosis and Management of Cirrhosis and Its Complications A Review

Abstract

IMPORTANCE

Cirrhosis affects approximately 2.2 million adults in the US. From 2010 to 2021, the annual age-adjusted mortality of cirrhosis increased from 14.9 per 100 000 to 21.9 per 100 000 people.

OBSERVATIONS

The most common causes of cirrhosis in the US, which can overlap, include alcohol use disorder (approximately 45% of all cases of cirrhosis), nonalcoholic fatty liver disease (26%), and hepatitis C (41%). Patients with cirrhosis experience symptoms including muscle cramps (approximately 64% prevalence), pruritus (39%), poor-quality sleep (63%), and sexual dysfunction (53%). Cirrhosis can be diagnosed by liver biopsy but may also be diagnosed noninvasively. Elastography, a noninvasive assessment of liver stiffness measured in kilopascals, can typically confirm cirrhosis at levels of 15 kPa or greater. Approximately 40% of people with cirrhosis are diagnosed when they present with complications such as hepatic encephalopathy or ascites. The median survival time following onset of hepatic encephalopathy and ascites is 0.92 and 1.1 years, respectively. Among people with ascites, the annual incidence of spontaneous bacterial peritonitis is 11% and of hepatorenal syndrome is 8%; the latter is associated with a median survival of less than 2 weeks. Approximately 1% to 4% of patients with cirrhosis develop hepatocellular carcinoma each year, which is associated with a 5-year survival of approximately 20%. In a 3-year randomized clinical trial of 201 patients with portal hypertension, nonselective β-blockers (carvedilol or propranolol) reduced the risk of decompensation or death compared with placebo (16% vs 27%). Compared with sequential initiation, combination aldosterone antagonist and loop diuretics were more likely to resolve ascites (76% vs 56%) with lower rates of hyperkalemia (4% vs 18%). In meta-analyses of randomized trials, lactulose was associated with reduced mortality relative to placebo (8.5% vs 14%) in randomized trials involving 705 patients and reduced risk of recurrent overt hepatic encephalopathy (25.5% vs 46.8%) in randomized trials involving 1415 patients. In a randomized clinical trial of 300 patients, terlipressin improved the rate of reversal of hepatorenal syndrome from 39% to 18%. Trials addressing symptoms of cirrhosis have demonstrated efficacy for hydroxyzine in improving sleep dysfunction, pickle brine and taurine for reducing muscle cramps, and tadalafil for improving sexual dysfunction in men.

CONCLUSIONS AND RELEVANCE

Approximately 2.2 million US adults have cirrhosis. Many symptoms, such as muscle cramps, poor-quality sleep, pruritus, and sexual dysfunction, are common and treatable. First-line therapies include carvedilol or propranolol to prevent variceal bleeding, lactulose for hepatic encephalopathy, combination aldosterone antagonists and loop diuretics for ascites, and terlipressin for hepatorenal syndrome.

Cirrhosis affects approximately 2.2 million adults in the US¹ and is associated with mortality rates of 21.9 per 100 000 people.^2,3 Cirrhosis is defined as the fibrotic replacement of liver tissue that can result from any chronic liver disease. Most prevalent cases of cirrhosis are caused by alcohol use disorder (approximately 45% of all cirrhosis cases), hepatitis C (41%), and nonalcoholic fatty liver disease (26%), with many patients having overlapping causes.⁴ However, hepatitis C is now curable with direct- acting antivirals and most newly diagnosed cirrhosis is due to nonalcoholic fatty liver disease (NAFLD) (accounting for 61.8% of incident cases) and alcohol use disorder (accounting for 20.0%).⁵

Outcomes for patients with cirrhosis can be improved with evidence-based therapies directed toward both the etiology of cirrhosis^6–12 and its complications.^13–21 Recent innovations include noninvasive risk stratification of cirrhosis^22,23 as well as interventions that improve survival by preventing or reducing the complications of cirrhosis.²⁴ Such complications include variceal hemorrhage, ascites, and hepatic encephalopathy. People with cirrhosis have reduced quality of life.²⁵ Poor quality of life is associated with many common symptoms²⁶ such as muscle cramps,^27,28 poor-quality sleep,²⁹ pruritus,^30,31 and sexual dysfunction,^32,33 which can be improved with therapy.

This review summarizes the current evidence regarding the diagnosis and management of cirrhosis and its complications (Box).

Box. Common Questions on the Management and Complications of Cirrhosis.

Does my patient need a liver biopsy to diagnose cirrhosis?

No, Cirrhosis can be accurately diagnosed using sequential noninvasive testing such as the fibrosis-4 index followed by a liver stiffness measurement obtained by elastography (eg, vibration-controlled transient elastography or magnetic resonance elastography). Because liver stiffness measurements are also prognostic, they can be used, for example, to determine which patients need endoscopy to screen for esophageal varices.

What is the most common cause of cirrhosis?

Most prevalent cases of cirrhosis in the US are caused by alcohol use disorder, nonalcoholic fatty liver disease (NAFLD), and hepatitis C infection. Most incident cases of cirrhosis are caused by NAFLD; however, there is also an increase in alcohol-related cirrhosis, particularly among young people.

What can be done to improve survival among patients with compensated cirrhosis?

Survival for patients with cirrhosis is improved with control of their underlying chronic liver disease (eg, alcohol use disorder, viral hepatitis, NAFLD). Beyond that, screening for liver cancer with biannual ultrasound and α-fetoprotein is associated with higher rates of curative treatment when cancer is detected. When patients develop portal hypertension, nonselective β-blockers (particularly carvedilol or propranolol) are associated with lower rates of decompensation or death.

Methods

We searched PubMed (January 1, 2000, to March 10, 2023) for systematic reviews, meta-analyses, randomized clinical trials (RCTs), and relevant guidelines. We prioritized recent RCTs of higher quality based on rigor of study design, sample size, and length of follow-up. Of 8887 articles retrieved, 115 were included, consisting of 9 practice guidelines, 3 consensus statements, 25 RCTs, 17 meta-analyses, 7 systematic reviews (without meta-analysis), and 54 observational cohort studies.

Discussion

Epidemiology

The causes of cirrhosis vary by context and many overlap. In a study of 68 673 patients from a national sample of patients in the Veterans Administration (2020–2021), the causes of cirrhosis were hepatitis C (24.0%), alcohol related (27.9%), hepatitis C and alcohol related (17.4%), NAFLD related (25.9%), and due to other conditions (3.7%).⁴ Patients diagnosed with cirrhosis typically have a mean age of 59.5 to 62.4 years.^34,35 Patients with NAFLD cirrhosis often present at a mean age of 67 years.³⁴ However, cirrhosis is now more common among younger patients. The incidence of cirrhosis by age 35 years was 46.9 per 100 000 people among those born during or after 1980 compared with 32.6 per 100 000 born between 1945 and 1960.³⁶ A total of 54% to 60% of cirrhosis cases occur among men.^35–37 Between 2010 and 2021, age-adjusted mortality from cirrhosis increased from 14.9 to 21.9 per100 000 people.^2,3 Cirrhosis mortality increased from 1.1 to 3.3 per 100 000 people aged 25 to 34 years from 2010 to 2020³ due to increases in alcohol-related liver disease.² The epidemiology of cirrhosis and its complications are described further in Table 1.^{24,34,36,38–56}

Table 1.

Epidemiology of Cirrhosis and Its Complications in the US

Condition	Prevalence/incidence	Mortality	Hospitalization rates
Cirrhosis	Prevalence, 840–1058/100 00034,36	56 989 in 2020a	56.4–125.7 of every 100 000 hospitalizations involve cirrhosis38,39 25.8% 30-d readmission rate40
Hepatic encephalopathy (HE)	40% 5-y risk overall among patients with cirrhosis 10% 1-y risk if Child Ab with portal hypertension 25% 1-y risk if Child Bb with portal hypertension41 25%–80% Develop covert HEc depending on Child class42,43a	Median survival, 0.95 y and 2.5 y for those aged ≥65 y or <65 y, respectively44	HE presents in 10.1% of cirrhosis hospitalizations45
Ascites	40% Lifetime risk among patients with cirrhosis 5.1% at 1 y46 20% at 3 y24	Median survival, 1.08 y46	Ascites present in 47.7% of cirrhosis hospitalizations45
Variceal bleeding	Varices form at a rate of 9%/y in patients with cirrhosis47 Variceal bleeding occurs at a rate of 1.2%–4% within 3 y among patients with cirrhosis48	17.7% 6-wk mortality after variceal bleeding49	Variceal bleeding occurs in up to 70.6 of every 100 000 hospitalizations in the US50 and 10.5% of cirrhosis hospitalizations45
Hepatocellular carcinoma	Annual incidence in viral hepatitis B- and hepatitis C-related cirrhosis was approximately 3.2%–4.8%; and 1.4%–2.1% in NAFLD51	<20% 5-y survival, dependent on stage of detection; 11 000 annual deaths nationally52	NA
Spontaneous bacterial peritonitis	11% Annual risk in patients with ascites53	Median survival after diagnosis of spontaneous bacterial peritonitis, 11 wk54	Spontaneous bacterial peritonitis occurs in 3.1% of cirrhosis hospitalizations55
Hepatorenal syndrome	8% Annual risk in patients with ascites56	Median survival is 2 wk after diagnosis of hepatorenal syndrome56	Hepatorenal syndrome occurs in 1.2% of cirrhosis hospitalizations45

Abbreviations: NA. not available: IMAFLD, nonalcoholic fatty liver disease.

^aThis estimate is derived from the primary cause of death listed on death certificates compiled in the CDC WONDER database.

^bChild A and B refer to the Child classification: class A refers to patients without ascites and normal albumin, bilirubin, and international normalized ratio: class B refers to patients with ascites and/or mild abnormalities in laboratory indices.

^cCovert HE is an early stage of HE characterized by deficits in executive function, poor sleep quality, and poor balance.

Pathophysiology of Cirrhosis

Chronic inflammatory liver injury causes activation of hepatic myofibroblasts and macrophages, which increase collagen accumulation (fibrosis) in the extracellular matrix. This process disrupts the connection between hepatocytes and sinusoids where blood flows, leads to formation of nodules of fibrosis, and impedes portal inflow resulting in portal venous hypertension. Chronic liver injury results in increased vasoconstrictor signaling (such as endothelin-1) and decreased production of vasodilators (such as nitric oxide), further restricting sinusoidal flow. Inflammatory injury from alcohol or steatosis also increases vascular resistance. In both NAFLD and alcohol-related liver disease, heritable factors in lipid metabolism have been associated with progression of liver disease.⁵⁷ In addition, chronic liver injury causes hepatocyte loss and reduces the liver’s capacity for metabolic activity including protein synthesis, detoxification, nutrient storage, and bilirubin clearance. Proteins synthesized by the liver include albumin, hormones (eg, thrombopoietin [responsible for platelet production]), and hemostatic factors (procoagulants and anticoagulants).⁵⁸ The multisystem impact of these processes is depicted in Figure 1.

Figure 1. Impact of Portal Hypertension and Hepatic Insufficiency on Cirrhosis Pathophysiology.

Cirrhosis leads to intrahepatic resistance, which causes portal hypertension and, at later stages, hepatic insufficiency, which disrupts the liver’s normal metabolic functions. Together these features cause gut-barrier disruption and portosystemic shunting, resulting in the multisystem complications of cirrhosis, eg, hepatic encephalopathy, sarcopenia, ascites, and kidney injury.

Over time, patients may progress from compensated cirrhosis without clinical manifestations to decompensated cirrhosis with variceal hemorrhage, ascites, or hepatic encephalopathy. Portal hypertension, defined as a pressure gradient between the hepatic and portal vein of 10 mm Hg or greater, promotes development of varices (collateral vessels that shunt portal blood to systemic veins and often result in dilated esophagogastric channels prone to hemorrhage). Disrupted portal flow causes decreased cardiac return and decreased central blood volume, leading to increased plasma renin activity, increased renal-tubular affinity for sodium, peripheral volume expansion, and kidney vasoconstriction, predisposing patients to ascites, hyponatremia, and kidney injury, particularly in the setting of volume depletion, infection, or hemorrhage. Increasing portal pressure induces ascites from hepatic sinusoids. Increased sinusoidal pressure causes increased lymph production, which extravasates into the peritoneum when lymphatic drainage capacity is exceeded. Gut-derived toxins, such as ammonia and bacterial products that induce systemic inflammation, cause hepatic encephalopathy. Hepatic encephalopathy can develop at low ammonia levels in the context of infection.⁵⁹ While the mechanisms are incompletely characterized, the presence of hepatic fibrosis and hepatic injury from inflammation contribute to the genetic and epigenetic aberrations that lead to the development of hepatocellular carcinoma.

Diagnosing Cirrhosis

Medical history and physical examination can identify patients with or at risk for cirrhosis. Patients with cirrhosis frequently experience muscle cramps (64% prevalence) and pruritus (39%),²⁶ poor-quality sleep (63%),⁶⁰ and sexual dysfunction (53%).⁶¹ Risk factors, such as diabetes or alcohol use, and symptoms, such as muscle cramps, pruritus, sleep disorder, and sexual dysfunction, are neither sensitive nor specific for the diagnosis of cirrhosis.⁶² Most physical examination findings are not sensitive for cirrhosis but some offer specificity greater than 90%: these include Terry nails (white discoloration, absent lunula, dark pink at tip), gynecomastia, caput medusa, facial telangiectasia, palmar erythema, decreased body hair, testicular atrophy, and jaundice.⁶²

Screening for cirrhosis in the general population is not currently recommended.⁶³ However, patients with established chronic liver disease with abnormal liver enzymes, hepatic steatosis on imaging, or viral hepatitis should be evaluated for cirrhosis. Liver biopsy is considered the criterion standard to diagnose cirrhosis, although it is being increasingly replaced by noninvasive methods for fibrosis assessment. Biopsy is reserved for patients with noninvasive testing that is inconclusive or technically inadequate or when the underlying chronic liver disease is unclear.

Serologic measures and imaging-based indices are used to diagnose cirrhosis. Compared with biopsy, these measures are less expensive, safer, and simpler to follow longitudinally (Figure 2). The most common serologic tests capture indirect signs of liver fibrosis and dysfunction (eg, thrombocytopenia, reflecting reduced platelet production and splenic sequestration and a higher ratio of aspartate aminotransferase to alanine aminotransferase).⁶² The fibrosis-4 index (FIB-4; age, alanine aminotransferase, aspartate aminotransferase, platelet count) is a widely accepted risk-stratification tool that, for people with either NAFLD or alcohol-related liver disease, classifies scores as low (<1.30), intermediate (1.30–2.67), and high (>2.67). Age increases the FIB-4; for patients older than 65 years, the lower risk threshold is 2.0 or less (while high-risk thresholds remain the same).⁶⁴ Cutoffs such as less than 1.45 and greater than 3.25 have been developed for hepatitis C. FI B-4 has high negative predictive value (96%) but low positive predictive value (63%) for cirrhosis.⁶⁵ Risk stratification using FIB-4, for which values less than 1.3 offer a negative likelihood ratio of 0.4 for advanced fibrosis,⁶⁶ is recommended by societal guidelines in patients with known NAFLD (or risk factors such as diabetes or obesity).⁶³

Figure 2. Noninvasive Diagnosis and Risk Stratification of Patients at Risk for Cirrhosis.

The presence of cirrhosis is associated with an increased risk of complications such as liver cancer and decompensation including ascites, hepatic encephalopathy, and variceal hemorrhage. It is important to evaluate for the presence of cirrhosis in people with risk factors or any diagnosed chronic liver disease. While physical examination findings may be suggestive, it is recommended to stratify risk for all using the FIB-4 followed by elastography for at-risk patients. After identifying patients with cirrhosis, optimal care may involve referral to a hepatologist, liver cancer screening, and consideration of endoscopy for varices screening and/or initiation of nonselective β-blockers. BMI indicates body mass index.

^a Terry nails identified by white discoloration, absent lunula, and dark pink tips.

^b Biopsy is of highest value when the diagnosis of the underlying liver disease is unclear or noninvasive tests yield indeterminate or discordant results. The role of biopsy is also based on patient preference and clinical context.

Additional testing is needed in the setting of an elevated FIB-4 score (eg, ≥1.3 for patients with NAFLD, ≥2.0 for patients >65 years old).⁶³ Sequential testing of patients with liver disease risk factors using FIB-4 followed by elastography can provide posterior probabilities of cirrhosis of 89% or greater.⁶⁶ Elastography provides a liver stiffness measurement (LSM; measured in kilopascals [kPa]) that correlates with the abundance of fibrosis.⁶⁶ LSM of 15 kPa or greater by vibration-controlled transient elastography (VCTE), an ultrasound-based method that uses a handheld probe for point-of-care assessments, identified cirrhosis with 95.5% specificity (62% positive predictive value) in a cohort of 5648 patients with both VCTE and liver biopsy. Conversely, LSM of 10 kPa or greater had a sensitivity of 74.9% (88% negative predictive value).^22,67 Magnetic resonance elastography has fewer technical failures than VCTE in patients with high (>40) body mass index (calculated as weight in kilograms divided by height in meters squared). However, cost and access limit widespread use of magnetic resonance elastography. Liver inflammation (ie, alanine aminotransferase >120 IU/L⁶⁸) and central venous congestion from heart failure can also increase liver stiffness, generating false-positives from elastography.

Diagnosis of Portal Hypertension

Portal pressures can be estimated using a transjugular catheter to determine the hepatic venous pressure gradient, a measure of the pressure gradient across the liver. Clinically significant portal hypertension (CSPH) is defined as a gradient of 10 mm Hg or greater (normal <5 mm Hg).²² In a study of 213 patients with pressure gradients less than 10 mm Hg, approximately 90% remained decompensation free for at least 4 years.⁶⁹ Pressure measurements, though safe, are costly; can only be obtained in specialized units; and have high (26%) within-individual variance.⁷⁰ The optimal noninvasive alternative for identifying patients with CSPH involves a combination of liver stiffness from VCTE and platelet counts.⁷¹ Thrombocytopenia (ie, platelet count <110 × l0⁹/L) in patients with liver disease is both highly suggestive of cirrhosis⁶² and associated with a patient’s risk of ascites and variceal bleeding.⁷² Among 518 persons with cirrhosis from Europe and Canada, a nomogram based on LSM and platelet counts was developed to predict CSPH.⁷¹ For example, patients with an LSM of 25 kPa or greater and any platelet count have a prevalence of CSPH of at least 66%, increasing to 90% or more for patients with platelet counts of 110 × 10⁹/L or less.⁷¹

All patients with varices have CSPH. Because portal pressures are not routinely measured, it is recommended to screen for varices in patients with cirrhosis every year if decompensated or every 2 to 3 years if compensated (2 years if the patient is actively drinking alcohol or chronic liver disease is uncontrolled, eg, untreated hepatitis B or C or autoimmune hepatitis).²³ However, guidelines also suggest that noninvasive tests can rule out CSPH^22,23 and therefore obviate the need for endoscopy. Among 7387 patients pooled from 26 studies, LSM less than 20 kPa and platelet count greater than 150 × 10⁹/L provided a negative likelihood ratio of 0.09 for high-risk (large and/or thin-walled) varices.⁷³ According to these data, only 2.2% of high-risk varices were missed if endoscopy was not performed.⁷³

Diagnosing Complications of Cirrhosis

The diagnosis of ascites can be made using abdominal ultrasonography or cross-sectional imaging. Flank dullness, shifting dullness, and fluid wave elicited by physical examination offer 94%, 83%, and 50% sensitivity and 29%, 56%, and 82% specificity, respectively.⁷⁴ Spontaneous bacterial peritonitis is diagnosed after paracentesis with ascites concentrations of neutrophil count greater than 250/μL.⁷⁵ Up to one-third of patients with spontaneous bacterial peritonitis do not have fever or pain. Therefore, diagnostic paracentesis is recommended for all hospitalized patients with cirrhosis and ascites.^53,76 Hepatorenal syndrome is defined as kidney injury in the presence of large-volume ascites if there is a 50%, or 0.3 mg/dL, or greater increase in serum creatinine within 7 days from the last measure that does not respond to 2 days of intravenous fluids to establish normal intravascular volume.⁵³

Hepatic encephalopathy is a clinical diagnosis. It presents as a spectrum on the West Haven Criteria scale (0 to 4 scale, where 0 indicates no deficits and 4 indicates coma). Overt hepatic encephalopathy (grades ≥2) presents with asterixis, disorientation, lethargy, and coma. Covert hepatic encephalopathy (grades ≤1) may present as deficits in executive function, sleep disorder, vegetative behavior, and gait disturbance. The criterion-standard diagnostic for covert hepatic encephalopathy is greater than or equal to 4 SDs below healthy control performance on the 5-test paper-pencil battery called the Psychometric Hepatic Encephalopathy Score. This battery can be replaced by some bedside measures⁴² including the Animal Naming Test (in a prospective cohort of 327 patients, <15 and <10 animals per minute offered sensitivities for diagnosing hepatic encephalopathy of 70% and 15%, respectively, and specificities of 63% and 92%, respectively) or the EncephalApp Stroop Test (>198 seconds on a computerized version of the Stroop test of attention offered 80% sensitivity and 61% specificity in a prospective cohort of 277 patients).⁴² Covert hepatic encephalopathy can also present as recent falls (40% within the prior year for those diagnosed with covert hepatic encephalopathy vs 12.9% for those without)⁷⁷ and poor-quality sleep (mean of 10.3 vs 7.6 on the Pittsburgh Sleep Quality Index where >5 reflects poor sleep).⁷⁸ An algorithm based on age, sex, and self-reported loss of balance, irritability/impatience, anorexia, and disinterest in physical activity can identify covert hepatic encephalopathy with a sensitivity of 80% and specificity of 79%.⁴²

Screening for Hepatocellular Carcinoma

Although randomized trials of screening for hepatocellular carcinoma (HCC) are lacking, screening with biannual abdominal ultrasound and serum α-fetoprotein is recommended by the American Association for the Study of Liver Diseases to improve early HCC detection in patients with cirrhosis, regardless of etiology.^76,79 In a meta-analysis of 32 observational studies that included 13 367 patients, screening for HCC was associated with early-stage detection (58.8% vs 27.0%) and increased rates of curative therapies (58.2% vs 34.0%) in comparison with no screening.⁸⁰ Longer screening intervals (ie, annual screening) have not been prospectively compared with semiannual screening.

Patients with cirrhosis and a greater than 1.0-cm mass on screening ultrasound or with a rising or elevated α-fetoprotein level (cutoff > 20 ng/mL) should undergo further diagnostic workup to evaluate for HCC. Though biopsy is diagnostic, multiphasic contrast-enhanced cross-sectional imaging can be used to make the diagnosis.⁷⁹ A solid lesion exhibiting specific features (eg, arterial-phase hyperenhancement and portal venous phase washout) in a patient with cirrhosis can be diagnosed as HCC.⁷⁹

Prognosis of Cirrhosis

Although survival varies with age at diagnosis and extrahepatic comorbidities,⁸¹ patients with compensated cirrhosis have a median survival of 12 years according to a pooled analysis of 806 prospectively followed up patients.⁸² Survival is reduced after any decompensation (Figure 3). Patients with compensated cirrhosis and small varices have a 6% 1-year risk of bleeding, while patients with large varices and decompensated cirrhosis have a 42% to 76% 1-year risk of bleeding.⁸³ In-hospital mortality after variceal hemorrhage is approximately 14.5% overall and as low as 0% for patients with previously compensated cirrhosis.⁸⁴ Ascites in the setting of cirrhosis was associated with a median survival of 1.1 years in a cohort of 13 265 patients enrolled in Medicare.⁴⁶ Median survival time following incident overt hepatic encephalopathy was 0.92 years in a study of 49164 patients with cirrhosis enrolled in Medicare.^37,44 Compared with patients with cirrhosis without any hepatic encephalopathy, covert hepatic encephalopathy was also associated with worse outcomes. Such outcomes included a higher 1-year risk of car crashes (17% of 97 patients with covert hepatic encephalopathy vs 3% of 70 without)⁸⁵ and, in a cohort of 170 patients with cirrhosis (56% with covert hepatic encephalopathy), higher rates of hospitalization (47% vs 15%) and death (18% vs 3%).⁴³

Figure 3. Natural History of Cirrhosis, Its Complications, and Modifiable Factors.

The consequences of cirrhosis are depicted on a timeline from the development of compensated cirrhosis to death or transplant. Biomarkers associated with lower risk of decompensation and death are shown in green; those of indeterminate significance in yellow, and biomarkers and events associated with a higher risk of decompensation and death in red; disease-modifying measures and interventions are shown in blue. The liver stiffness measures are shown in kilopascals and are derived from vibration-controlled transient elastography.

^a Ten to 19 kPa denotes intermediate risk, 15 kPa is the threshold for assuming cirrhosis, 20 kPA is high risk, and 25 kPa is highest risk and assumes clinical portal hypertension.

Prognostic Systems

Factors associated with reduced survival include lower serum levels of albumin, higher international normalized ratios (INRs), and elevated bilirubin levels. These are 3 components of the Child-Turcotte-Pugh (CTP) score, which also includes ascites and hepatic encephalopathy. Bilirubin and INR are included in the Model for End-stage Liver Disease-Sodium (MELD-Na) score along with creatinine and sodium levels. The CTP ranges from 5 (75% 5-year survival) to 15 (20% 5-year survival if >12)⁸⁶; the MELD-Na ranges from 6 (1.9% 90-day mortality) to 40 (71.3% 90-day mortality).⁸⁷ MELD is best suited to short-term prognostication for patients with decompensated cirrhosis and it is used to prioritize organ allocation on the transplant waitlist; the CTP is used for long-term prognostics and complements MELD by describing the patient’s compensation status. When patients with cirrhosis require hospitalization, patients with acute-on-chronic liver failure have an increased rate of near-term mortality.⁸⁸ Organ failures include severe hepatic encephalopathy (disorientation and/or coma), shock, requirement for mechanical ventilation, and kidney failure requiring dialysis. Thirty-day survival is 95% for patients with decompensated cirrhosis and no organ failure.⁸⁸ Survival is reduced for patients with organ failures and infections. For patients with 2 organ failures, survival with or without infection is 62% or 84%, respectively; for those with 4 organ failures, survival is 0% or 24%.⁸⁸

Treatment

Symptoms of Cirrhosis

Cirrhosis is associated with multiple common physical and psycho-logical symptoms that can be improved with treatment (Table 2).^{13–21,24,27–33,53,89–94} In an RCT of 80 patients, compared with tap water, 1 sip of pickle brine at cramp onset significantly reduced cramp severity at 28-day follow-up (2.3- vs 0.4-point reduction on a 10-point visual analog scale).²⁷ In a 2-week randomized, double-blind, crossover trial of 30 patients, compared with placebo, 1000 mg of taurine twice daily significantly reduced leg cramping (7 fewer cramps compared with placebo).²⁸ Cholestyramine (4–16 g daily) is considered first-line therapy for pruritus given its safety profile but randomized trials in cirrhosis are lacking.³⁰ In a 4-week RCT of 16 patients with cirrhosis, naltrexone significantly improved pruritus compared with placebo with a mean (SD) 54% (10%) reduction in pruritus severity compared with an 8% (10%) increase as measured by a 100-mm visual analog scale.³¹ In a 10-day RCT of 35 patients with sleep disorder, compared with placebo, hydroxyzine, 25 mg, nightly was associated with a significant 40% improvement from baseline on a 10-point visual analog scale of sleep quality (vs 0% for placebo).²⁹ One patient in the hydroxyzine group developed overt hepatic encephalopathy (disorientation). Alcohol cessation may improve sexual function, with 25% of 60 men abstinent for 6 months or longer achieving self-reported normal sexual function.³² In a 12-week, randomized trial of 140 men, compared with placebo, tadalafil, 10 mg, improved erectile function based on patient report (63% vs 30%).³³

Table 2.

Treatment of cirrhosis

	Symptoms and diagnosis	First-line therapies	Effectiveness	Adverse effects
Symptoms of cirrhosis
Muscle cramps	Patient reported	Acute therapy: sips of pickle juice at cramp onset Preventive therapy: taurine, 500–1000 mg, twice daily	Pickle juice: reduction of 2.3 points on VAS for cramp severity compared with 0.4 for a tap-water control among 80 patients in a randomized trial27 Taurine: 7 fewer cramps compared with placebo over 2 wk28 in a randomized, double-blind crossover trial of 30 patients	Pickle juice: none observed in trial27 Taurine: none observed in trial28
Pruritus	Patient reported	Moisturizers Cholestyramine, 4–16 g daily Naltrexone, 50 mg daily	Cholestyramine: efficacy data from trials enrolling people with cirrhosis is lacking but considered first-line therapy given safety30 Naltrexone: placebo-controlled RCT of 16 patients followed up for 4 wk; naltrexone significantly improved pruritus as measured by a 100-mm VAS of pruritus severity with a mean (SD) 54% (10%) reduction from baseline compared with an 8% (10%) with placebo31	Cholestyramine: none observed in trials30 Naltrexone: 50% experienced 3 d of malaise with nausea (vs 0% with placebo); 62.5% experienced mild abdominal cramps (vs 12.5% with placebo)31
Sleep disorder	Patient reported	Sleep hygiene (improving environment, relaxation, limiting caffeine) Hydroxyzine, 25 mg, nightly	Hydroxyzine: in a 10-d randomized, double-blind, placebo-controlled trial of 35 patients with covert HE, 40% with improved self-reported sleep and 70% with improved sleep efficiency by actigraphy (noninvasive method of monitoring rest/activity cycles) compared with 0% and 16% worsening among those receiving placebo, respectively29	Hydroxyzine: risk of increased confusion after 10 d (6% vs 0% with placebo)29
Sexual dysfunction	Patient reported	Cessation of smoking and alcohol use, counseling, PDE-5i (ie, tadalafil, 10 mg)	Cessation of alcohol use associated with 25% rate of achieving self-reported normal sexual function among 60 men abstinent from alcohol for >6 mo32 Tadalafil: in a 12-wk, randomized, placebo-controlled trial of 140 men, 63% had improved erectile function compared with 30% receiving placebo33	None observed in trial
Complications of cirrhosis
Hepatic encephalopathy (HE)	Covert HE: deficits in executive function, sleep disorder, vegetative behavior, and gait disturbance Overt HE: asterixis, disorientation, lethargy, and coma	Lactulose, 10–20 g, 2–3 times daily with goal of 2–3 soft bowel movements Rifaximin, 550 mg, twice daily Nutrition, 1.25 g of protein/kilogram of actual body weight/d and nighttime snack (>250 kcal)	In meta-analyses of randomized trials, lactulose was associated with reduced mortality relative to placebo (8.5% vs 14%) in randomized trials involving 705 patients and reduced risk of recurrent overt HE (25.5% vs 46.8%) in randomized trials involving 1415 patients14 In a 6-mo placebo-controlled, double-blind RCT, rifaximin reduced hospitalization for HE to 13.6% vs 22.6% with placebo among 299 patients with prior HE with lactulose13 In a 6-mo RCT of 120 patients with covert HE, the 60 patients receiving protein supplements to achieve 1–1.5 g/kg daily protein intake had lower rates of overt HE relative to placebo (10% vs 21.7%)15	Lactulose: relative to placebo, bloating (46% vs 15%), diarrhea (29% vs 37% but higher if excess intake), nausea (15% vs 2%)14 Rifaximin: high cost, no other adverse events seen in trial13 High-protein diet: none15
Ascites	Physical examination (distended abdomen, shifting dullness) and free fluid on abdominal imaging	Aldosterone antagonists (spironolactone, 50–400 mg) Loop diuretics (furosemide, 40–160 mg)	Combined aldosterone antagonists and loop diuretics resolved ascites in 76% of patients compared with 56% who received aldosterone antagonists alone in a randomized trial of 100 patients16 Paracentesis relieves abdominal symptoms but ascites recurs	Combination aldosterone antagonists and loop diuretics: hyperkalemia (4%), kidney injury (12%) Aldosterone antagonists alone: hyperkalemia (18%), kidney injury (16%) Not reported in trials of patients with ascites but observed in other trials of spironolactone: gynecomastia (rate unknown)53
Varices and variceal bleeding	Endoscopic evaluation	Prevention of bleeding: carvedilol, 12.5 mg, daily (can be used for primary or secondary prophylaxis); propranolol, 40 mg, twice a day increased up to 160 mg twice a day with goal heart rate of 55 beats/min Active bleeding:  Intravenous octreotide/terlipressin ceftriaxone (1 g/d for 5 d) Band ligation or sclerotherapy of varices	Primary prophylaxis with carvedilol or propranolol in a double-blind, placebo-controlled trial of 201 patients with CSPH, carvedilol or propranolol reduced risk of decompensation or death (16% vs 27%) by 3 y24 Treatment of hemorrhage: vasoactive medications (in addition to endoscopic therapy) Octreotide: associated with hemostasis at 5 d in 77% compared with 58% with placebo in a meta-analysis of randomized trials17 Terlipressin: associated with hemostasis at 5 d in 65% compared with 46% with placebo in a meta-analysis of randomized trials89 Prophylactic antibiotics (in addition to endoscopic therapy): associated with reduced mortality to 18.5% vs 22.2% with placebo in a meta-analysis of randomized trials18 Secondary prophylaxis with preemptive TIPS: in a randomized trial of 63 patients with acute variceal bleeding, preemptive TIPS within 72 h improved 1-y survival to 86% vs 61% for the control group19	Carvedilol or propranolol: weakness (23% vs 17%)24 Octreotide: hyperglycemia (23% vs 13% with placebo)17 Terlipressin: no difference compared with placebo in meta-analysis of trials, can be associated with digital ischemia and abdominal cramping90 TIPS: none observed in trial19
Spontaneous bacterial peritonitis (SBP)	Ascites concentrations of neutrophils >250/μL can present without fever or pain in up to one-third of cases	Third-generation cephalosporin for 5 d Albumin, 1.5g/kg, on day 1 and 1 g/kg on day 2 Meropenem and daptomycin for nosocomial SBP	Albumin: compared with antibiotics alone, reduces 3-mo mortality in RCT of 126 patients from 41% to 22%; kidney injury occurred in 33% vs 6%20 If nosocomial SBP: in a randomized trial of 32 patients, compared with ceftazidime, meropenem and daptomycin increased SBP resolution (87% vs 25%) but not 90-d survival91	Not reported in trial. However, albumin infusions can increase risk of pulmonary edema (4% vs 1% from placebo in a trial of albumin for hospitalized patients with cirrhosis)92
Hepatorenal syndrome (HRS)	50% or ≥0.3 mg/dL increase in serum creatinine within 7 d from the last measure and does not respond to 2 d of volume expansion	Terlipressin, 0.85 mg, IV every 6 h Norepinephrine, 0.5–3 mg/h, IV	In placebo-controlled (1:2 ratio), randomized trial, terlipressin improved kidney recovery (39% vs 18%) in 300 patients with HRS21 In a randomized trial of 46 patients comparing terlipressin vs norepinephrine, the rate of achieving creatinine <1.5 mg/dL was 39.1% vs 43.4%93	Terlipressin: death due to respiratory failure (11% vs 2% from placebo)21 Norepinephrine: none reported in trial93; however, 25% experienced tachyarrhythmia in a cohort study, 10% requiring treatment discontinuation94

Abbreviations: CSPH, clinically significant portal hypertension: IV. intravenous: PDE-5i, phosphodiesterase-5 inhibitor; QOL, Quality of Life; RCT, randomized clinical trial; TIPS, transjugular intrahepatic portosystemic shunt; VAS, visual analog scale.

SI conversion factor: to convert creatinine to μmol/L, multiply by 88.4.

Underlying Etiology of Cirrhosis

Patients with cirrhosis of any cause may benefit from evaluation for liver transplant when they have developed a decompensation or HCC.⁷⁶ However, control of the underlying etiology improves the prognosis of cirrhosis by slowing its progression and may reverse fibrosis. For example, after 12 months of follow-up, 43% of 37 patients with cirrhosis and hepatitis C cured with direct-acting antivirals experienced regression to a lower fibrosis stage.⁹⁵ Among 96 patients with hepatitis B cirrhosis treated with tenofovir and followed up for 240 weeks, 28% no longer had cirrhosis on biopsy.⁹⁶ Alcohol use can worsen the prognosis of any chronic liver disease; alcohol use disorder should be identified and treated.⁹⁷ In an observational longitudinal study of 33 682 patients with cirrhosis and alcohol use disorder from the Veterans Administration, behavioral or pharmacotherapy (eg, naltrexone) for alcohol use disorder was associated with significantly reduced 180-day mortality (2.6% vs 3.9%) and cirrhosis decompensation (6.5% vs 11.6%).⁹⁸ Etiology-specific therapies and their effects on cirrhosis outcomes are provided in Table 3.^{6–12,99–104}

Table 3.

Therapies for Cirrhosis and Its Complications, According to Etiology^a

	Risk factors	First-line therapies	Effectiveness	Adverse effects
Alcohol-related liver disease	Alcohol use disorder, obesity	Abstinence from alcohol. Counselling and medications such as naltrexone	In a meta-analysis of observational studies, abstinence was associated with reduced risk of mortality after 1.5 y (HR, 0.51 [95% CI, 0.33–0.81])12	NA
Nonalcoholic fatty liver disease (NAFLD)	Insulin resistance, obesity, metabolic syndrome	Weight loss. Nutritionist referral, medical weight loss therapies, and bariatric surgery in highly selected patients (ie, without portal hypertension)	No randomized trials in patients with cirrhosis Among 1158 patients with NAFLD and fibrosis, bariatric surgery was associated with 2.3% 10-y risk of cirrhosis or decompensation compared with 9.6% for 508 matched patients who did not undergo surgery6	Rates of adverse events are unknown. Inadvertent protein restriction while dieting could worsen sarcopenia and increase the risk of hepatic encephalopathy. Bariatric surgery is associated with a 4.7% risk of decompensation in patients with compensated cirrhosis99
Hepatitis C	Blood transfusions prior to 1990, shared needles or drug implements, rarely sexual transmission	Direct-acting antivirals (eg, sofosbuvir/velpatisvir or glecaprevir/pibrentasvir)	Observational data sources: SVR, compared with not achieving SVR, is associated with lower 10-y all-cause mortality (8.9% vs 26.0%)7 SVR, compared with not achieving SVR, is associated with a lower risk of HCC (3.3 vs 13.2/1000 person-years)100	Headache: 25%
Hepatitis B	Vertical transmission, sexual transmission, shared needles, or drug implements	Virological control with antiviral therapy (eg, tenofovir alafenamide, tenofovir disoproxil fumarate, entecavir)	Randomized (2:1), placebo-controlled trial of lamivudine in 651 patients with advanced fibrosis or cirrhosis: compared with placebo, treatment lowered risk of HCC (3.9% vs 7.4%)8 Observational study of tenofovir disoproxil fumarate: compared with no treatment (291 patients), antiviral therapy (797 patients) was associated with a lower risk of HCC (9.8% vs 14.9%), decompensation (1.1% vs 13.1%), and death (1.1% and 13.1%) in people with cirrhosis from Hong Kong, South Korea, and California101	Lamivudine associated with cough relative to placebo (14% vs 7%)8 Tenofovir disoproxil fumarate was associated with decreased bone density (1.72% in hip, 2.29% in spine) after 48 wk of therapy102 Among patients with hepatitis B and HIV coinfection, 0% receiving tenofovir alafenamide discontinued therapy for kidney complications within 144 wk compared with 3.6% receiving tenofovir disoproxil fumarate103
Hemochromatosis	Autosomal recessive inheritance; 2 copies of the C282Y variant of the HFE gene	Iron depletion (phlebotomy) with a goal of 50–100 μg/L of ferritin	In a nationwide, observational study of Danish patients, 10-y overall survival was 70% among 66 patients receiving phlebotomy and 20% of 62 untreated patients9	Venipuncture site bleeding and infection, anemia, and syncope (rates unknown)
Primary biliary cholangitis (PBC)	More common among women and first-degree relatives of patients with PBC	Ursodeoxycholic acid (UDCA) at 13–15 mg/kg/d	In a pooled analysis of 548 patients in 3 randomized trials, UDCA improved transplant-free 4-y survival compared with placebo (17% vs 24%)10	Not observed in trials10 Rare adverse events may include weight gain (<5 lbs),104 loose stool, and hair thinning
Primary sclerosing cholangitis (PSC)	Ulcerative colitis	No proven therapy	NA	NA
Autoimmune hepatitis	Unknown	Combination prednisone (20–40 mg/d) and azathioprine (50–150 mg/d)	In a meta-analysis of randomized trials of 3–4 y duration, the rates of remission and overall mortality associated with combination therapy (received by 44 patients) was 43% and 0% while placebo (received by 33 patients) was associated with 39% mortality and 0% remission11	Not reported in meta-analysis Prednisone is associated with adverse effects such as weight gain, rash, cataracts, infection risk, and osteoporosis Azathioprine is associated with leukopenia, pancreatitis, nonmelanoma skin cancer, lymphoma, and increased infection risk

Abbreviations: HCC, hepatocellular carcinoma: HR, hazard ratio; NA, not applicable; SVR, sustained virologie response (cure of hepatitis C).

^aLiver transplant is the definitive therapy for decompensated cirrhosis of any cause: the underlying chronic liver disease may recur and require treatment.

Complications of Cirrhosis

Varices and Portal Hypertension

Nonselective β-blockers (eg, carvedilol or propranolol) reduce portal pressure by reducing splanchnic blood flow. Because of its α-blocking effects, carvedilol also reduces intrahepatic resistance.²² β-Blockers are standard of care for people with large varices or prior bleeding.^58,76 If large varices are encountered on endoscopy, carvedilol (optimally dosed at 12.5 mg daily) is preferred to other β-blockers (grade B evidence, strong recommendation) according to the Baveno VII consensus statement.²² In an RCT of 152 patients, compared with band ligation every 2 weeks until variceal eradication, participants randomized to carvedilol (without banding) had lower rates (10% vs 23%) of variceal bleeding after 20 months of follow-up.¹⁰⁵ Esophageal ulcers caused by the band ligation resulted in bleeding in about 8% of patients in the band ligation group.¹⁰⁵ Patients with portal hypertension alone also benefited from β-blockers. In a 3-year, placebo-controlled, RCT of 201 patients with CSPH, propranolol (or carvedilol for those who did not respond to propranolol) reduced the risk of decompensation or death (16% vs 27%).^24,48 Variceal bleeding should be treated with band ligation during timely endoscopy (<24 hours after presentation),¹⁰⁶ vasoactive medications (compared with placebo, octreotide was associated with higher rates of hemostasis at 5 days [77% vs 58%] in a meta-analysis of randomized trials),¹⁷ and prophylactic antibiotics (associated with reduced short-term mortality to 18.5% vs 22.2% with placebo in a meta-analysis of randomized trials¹⁸). In a randomized trial of 63 patients with acute variceal bleeding who achieved initial hemostasis, transjugular intrahepatic portosystemic shunt (TIPS, a stent placed in a tract created to connect branches of the hepatic and portal veins) performed within 72 hours (compared with no TIPS placement) improved 1-year survival (61% vs 86%).¹⁹

Ascites

In an RCT of 100 patients that compared sequential therapy with aldosterone antagonists followed by the addition of loop diuretics to a guideline-recommended combination of both diuretics, ascites resolved at a higher rate with combination therapy (76% vs 56%) and was associated with lower rates of hyperkalemia (4% vs 18%).^16,53 Sodium restriction (<2 g/d) is recommended because greater Intake may be associated with worse ascites.⁵³ However, sodium restriction must be carefully monitored, ideally under the care of a nutritionist. Sodium restriction may not improve the effectiveness of diuretics (as seen in a clinical trial of 115 hospitalized patients randomized to daily sodium intake of 2760 mg or 920 mg resulting in rates of refractory ascites of 5.7% vs 4.8%, respectively).¹⁰⁷ Further, many patients who successfully restrict sodium do not meet daily calorie and protein goals.⁵³

Paracentesis is associated with temporary relief for patients with symptomatic ascites. Multiple paracenteses, despite attempts to optimize diuretic dosage, should prompt referral for TIPS. In a meta-analysis of 305 patients in randomized trials, compared with treatment without TIPS, TIPS was associated with reduced risk of recurrent ascites (42% vs 89%) and reduced 2-year mortality (51% vs 65%), but more hepatic encephalopathy episodes per year (mean [SD1],1.1 [1.9] vs 0.63 [1.2]).¹⁰⁸ While the risk of TIPS rises with MELD score and age, rather than using an absolute MELD cutoff for TIPS candidacy, expert consensus recommends a multidisciplinary approach and shared decision-making (level of evidence, 2a).¹⁰⁹

Hyponatremia

Hyponatremia (<135 meq/L) is common among patients with decompensated cirrhosis, affecting up to 31% of a national sample of 13 940 patients undergoing liver transplant evaluation.⁸⁷ Though evidence supporting its management is limited, hyponatremia is treated by addressing volume depletion (if present) and optimizing diuretic doses; fluid restriction is reserved for patients with levels less than 125 mmol/L despite optimization.⁵³

Hypoalbuminemia

Because hypoalbuminemia (<3.5 g/dL) is common among patients with ascites, an open-label RCT enrolling 431 patients with ascites refractory to diuretics and hypoalbuminemia investigated the role of weekly infusions of 40 g of 25% human albumin solution compared with no infusions¹¹⁰; 18-month survival was higher in the albumin group (77%) compared with the no infusion group (66%).¹¹⁰ However, albumin infusion is not yet recommended for clinical care.⁵³ In an open-label randomized trial enrolling 777 hospitalized patients with cirrhosis and albumin level less than 3.0 g/dL, albumin infusions (mean dose, 200 g) targeted to increase the albumin level to more than 3.0 g/dL did not improve the rate of a composite outcome (new infection, kidney dysfunction, or death) for up to 14days (29.7% vs 30.2% for the standard care group). The rate of pulmonary edema (4% vs 1%) was higher for the albumin group.⁹²

Spontaneous Bacterial Peritonitis

Spontaneous bacterial peritonitis should be treated with guideline-recommended third-generation cephalosporins such as ceftriaxone, 2 g, daily and intravenous albumin.⁵³ In an RCT of 126 patients, compared with antibiotics alone, 25% albumin (1.5 g/kg on day 1,1 g/kg on day 2) reduced mortality from 41% to 22% at 3-month follow-up.²⁰ After a first episode of spontaneous bacterial peritonitis, patients should receive secondary prophylaxis with suppressive oral antibiotics (eg, trimethoprim/sulfamethoxazole or ciprofloxacin).⁵³ Primary prophylaxis may not be effective given the prevalence of resistant organisms in the community. Antibiotic use is associated with adverse events (eg, trimethoprim-related hyperkalemia, antibiotic-related diarrhea, or Clostridioides difficile infection).⁵³

Hepatorenal Syndrome

Optimal therapy for hepatorenal syndrome includes carefully monitored volume expansion with intravenous albumin and vasoconstrictor therapy to increase mean arterial pressure and kidney perfusion. In an RCT of 300 patients, compared with placebo, terlipressin improved kidney function (creatinine ≤1.5 mg/dL; 39% vs 18%) but was associated with an increased risk of death due to respiratory failure (11% vs 2% with placebo).²¹ In a meta-analysis, norepinephrine, 0.5 to 3 mg/h, was noninferior to terlipressin with a 50% pooled rate of kidney injury reversal.¹¹¹

Coagulopathy

Despite low platelet counts or prolonged INR, bleeding after low-risk procedures (eg, paracentesis, endoscopy) is rare. For example, the rate of major bleeding was 0.2% in a pooled analysis of 2113 patients with INR greater than 1.5 and/or platelet count less than 50 × 10⁹/L undergoing paracentesis.¹¹² In guidelines from the American Association for the Study of Liver Diseases and the Society of Interventional Radiology, neither prophylactic plasma, platelet transfusions, nor vitamin K supplementation is recommended.^58,112

Hepatic Encephalopathy

Patients who present with clinically manifest hepatic encephalopathy (grade ≥2) should be evaluated for infection, gastrointestinal bleeding, dehydration, and receipt of psychoactive medications.⁷⁶ Patients should receive intravenous hydration and lactulose, starting at a dose of 60 cc, followed by 20 cc of lactulose every 1 to 2 hours until a bowel movement occurs, followed by maintenance with sufficient lactulose to achieve 2 to 3 soft bowel movements per day.⁴³ After the acute episode resolves, secondary prophylaxis also includes rifaximin, 550 mg, twice a day.¹³ In a meta-analysis of 705 patients with hepatic encephalopathy, compared with placebo, lactulose was associated with reduced mortality (8.5% vs 14%).¹⁴ Compared with placebo, lactulose was associated with reduced recurrent overt hepatic encephalopathy (25.5% vs 46.8%) in a meta-analysis of RCTs enrolling 1415 patients with hepatic encephalopathy.¹⁴ In a meta-analysis of 3 clinical trials of 126 participants receiving lactulose, lactulose was associated with an improvement of 6.92 (95% CI, 6.66–7.18) in the Sickness Impact Profile (ranges from 0 [best] to 68 [worst]; clinically important differences are >4).¹¹³ In a 6-month placebo-controlled, double-blind RCT, rifaximin reduced hospitalization for hepatic encephalopathy from 22.6% to 13.6% among 299 patients with prior hepatic encephalopathy taking lactulose.¹³ Guidelines recommend that all patients should receive education about nutrition, including consumption of 1 g of protein per kilogram of actual body weight, 30 to 40 kCal/kg, and a nighttime snack, as overnight fasting exacerbates catabolism (Table 2).^114,115

Limitations

This review has several limitations. First, this was not a systematic review. Second, some relevant research may have been missed. Third, the quality of the included evidence was not formally assessed.

Conclusions

Approximately 2.2 million US adults have cirrhosis. Many symptoms such as muscle cramps, poor sleep, pruritus, and sexual dysfunction are common and treatable. First-line therapies include carvedilol or propranolol to prevent variceal bleeding, lactulose for hepatic encephalopathy, combination aldosterone antagonists and loop diuretics for ascites, and terlipressin for hepatorenal syndrome.