Portal hypertension in children

Watch the interview with the author

Watch a video presentation of this article

Answer questions and earn CME

Abbreviations

EGD

esophagogastroduodenoscopy

EST

endoscopic sclerotherapy

EVL

endoscopic variceal ligation

TIPS

transjugular intrahepatic portosystemic shunt

Many liver and vascular diseases cause portal hypertension in children (Table 1), which may give rise to severe and life‐threatening complications, including hemorrhaging from esophageal varices, ascites, hepatopulmonary syndrome, portopulmonary hypertension, and hepatic encephalopathy. The effective prevention and management of portal hypertension and its complications, therefore, are important goals for improving outcomes for affected children.

Table 1.

Causes of Portal Hypertension in Children

Location of Lesion	Diagnostic Group	Examples
Hepatic	Cirrhosis resulting from cholestatic disease	Biliary atresia, progressive familial intrahepatic cholestasis, primary sclerosing cholangitis, cystic fibrosis liver disease, intestinal failure–associated liver disease
	Cirrhosis resulting from hepatocellular disease	Autoimmune hepatitis, chronic viral hepatitis B and C, alpha‐1‐antitrypsin deficiency, nonalcoholic fatty liver disease
	Other fibrotic liver disease	Congenital hepatic fibrosis/Caroli disease
Prehepatic	Portal vein occlusion	Portal vein thrombosis, tumor (e.g., infiltration by hepatoblastoma, hepatocellular carcinoma, or compression by large focal nodular hyperplasia)
	Nodular regenerative hyperplasia	Drug therapy (6‐thioguanine and azathioprine), Turner syndrome
	Portal venopathy or portal sclerosis	Schistosomiasis, idiopathic, human immunodeficiency virus infection, cystic fibrosis liver disease
Posthepatic	Hepatic vein obstruction	Budd‐Chiari syndrome, inferior vena cava obstruction, congestive heart failure
		Veno‐occlusive disease (sinusoidal obstruction syndrome)

Epidemiology and Natural History of Varices in Children

Up to 70% of children with biliary atresia or portal vein thrombosis have esophageal varices.1, 2 Variceal hemorrhaging occurs in 17% to 29% of children with biliary atresia over a 5‐ to 10‐year period3 and in up to 50% of children with portal vein thrombosis by 16 years of age.2 Predictors of variceal bleeding in children include the variceal size assessed by esophagogastroduodenoscopy (EGD) as well as red marks on varices and the presence of gastric varices.1, 2 The mortality rate from variceal bleeding in children is poorly characterized but ranges from 2.5% to 20% in children with biliary atresia and from 0% to 2% in children with portal vein thrombosis.2, 3, 4

Diagnosis of Portal Hypertension and Assessment of Its Severity

Portal hypertension usually is diagnosed in children with liver disease or portal vein thrombosis when splenomegaly, evidence of portal‐systemic collaterals, or both are found during a physical examination or on abdominal ultrasound scans. Other associated complications may occur: thrombocytopenia and leukopenia due to hypersplenism, hematemesis or melena due to variceal bleeding, cyanosis or low oxygen saturation on pulse oximetry due to hepatopulmonary syndrome, and abdominal distension due to ascites. A minority of children will have reversal of blood flow in the portal vein shown by Doppler ultrasonography; the severity of portal hypertension is thought to be worse in such children. Although hepatic venous pressure gradient measurements are feasible and safe in children, including infants, the role of this measurement in the clinical care of children with portal hypertension has not yet been defined.5

Diagnosis of Varices

No evidence‐based recommendation can be made to screen routinely for varices in children with cirrhosis or other risk factors because there is currently no evidence that any therapy is safe and effective for the prevention of the first variceal bleed in children. When surveyed, however, many parents of children with portal hypertension indicate that they would accept endoscopy for their children to provide information about the risk of bleeding, even if no therapy is available to reduce that risk.6 Expert opinion on the management of portal hypertension has recently suggested that children beyond puberty may be managed according to adult guidelines.7

The currently accepted best available test for the diagnosis of varices is EGD. However, EGD has important limitations, including a lack of validated grading systems for variceal size and appearance, poor interobserver variation (κ statistics of 0.36‐0.6), and the requirement for significant sedation or general anesthesia when it is performed in children.

Therefore, there has been considerable effort to find a noninvasive test for esophageal varices (Table 2). Future studies will need to determine the real‐world effectiveness of simple noninvasive tests, for which patient access and uptake may be much better in comparison with EGD.

Table 2.

Noninvasive Tests for the Diagnosis of Varices

Tests Studied in Adults	Summary of Pediatric Studies
Computerized tomography scan
Magnetic resonance imaging scan	Sensitivity of 75% and specificity of 87% in 34 children with biliary atresia
	Better performance in children older than 10 years
Ultrasound scan
Transient elastography	Sensitivity of 97%, specificity of 80%, and area under the receiver operator characteristic (AUROC) curve of 0.88 in 44 children with biliary atresia
Platelet count, spleen size, and platelet count/spleen size ratio	Sensitivity of 97.7% and specificity of 26.8% for splenomegaly in 111 children with portal hypertension of various causes
	Sensitivity of 81%, specificity of 70%, and AUROC curve of 0.79 for platelet counts and sensitivity of 81%, specificity of 73%, and AUROC curve of 0.80 for a clinical prediction rulea in 108 children with portal hypertension of various causes
Capsule endoscopy
Liver fibrosis blood tests (e.g., FibroTest, aminotransferase‐to‐platelet ratio index, Lok's model, and Forns' index)

^aThe clinical prediction rule included the platelet count, spleen size z score for age, and albumin.

Primary Prophylaxis for Variceal Bleeding

Primary prophylaxis for variceal hemorrhaging in adults is the established standard of care, but the widespread use of primary prophylaxis in children is controversial because of the lack of controlled pediatric data. Current practices among pediatric hepatologists vary significantly.6 The published pediatric experience does not support evidence‐based recommendations and is summarized in Table 3. Particular concerns are related to the fear that infants receiving beta‐blockers may experience worse outcomes as a result of hypovolemic shock because of their reliance on a tachycardic response to shock and their poor ability to increase stroke volume. Endoscopic variceal ligation (EVL) as primary prophylaxis in children is well tolerated, with a low subsequent bleeding rate and no reports of major complications. Endoscopic sclerotherapy (EST) has been piloted for primary prophylaxis in infants, for whom currently available equipment for EVL is too large. However, further safety data are required before its widespread use can be recommended.14

Table 3.

Studies of Primary Prophylaxis of Variceal Bleeding in Children

Study	Treatment	Design	n	Follow‐Up	Bleeding (%)
Shashidhar (1999)8	Beta‐blocker	Case series	17	3 years	35
Ozsoylu (2000)9	Beta‐blocker	Case series	45	5 years	16
Erkan (2003)10	Beta‐blocker	Case series	10	5.2 years	10
Paquet (1985)11	Sclerotherapy	Case series	2	10 years	0
Howard (1988)12	Sclerotherapy	Case series	17	2.5 years	0
Maksoud (1991)13	Sclerotherapy	Case series	26	2.4 years	42
Goncalves (2000)4	Sclerotherapy	Randomized controlled trial	100	4.5 years	6% EST versus 42% control
Duché (2008)14	Sclerotherapy	Case series	13	8 months	8
Lampela (2012)15	Sclerotherapy	Case series	16	1.7 years	25
Cano (1995)16	EVL	Case series	4	Not given	0
Sasaki (1998)17	EVL	Case series	9	23 months	10
Celinska‐Cedro (2003)18	EVL	Case series	37	16 months	0

For children with portal vein thrombosis, portal hypertension may be treated by the creation of a surgical shunt between the mesenteric vein and the left portal vein at a point just beneath the Rex fossa with an autologous vein graft, which is usually taken from the internal jugular (a meso‐left portal vein bypass or Rex bypass). This procedure restores portal venous flow through the liver. The balance between risks and benefits of this procedure for primary prophylaxis has yet to be defined.

Management of Acute Variceal Bleeding

Unfortunately, minimal data are available from studies of children with acute variceal hemorrhage, and no significant new studies in this area have been published in the last 5 years. Table 4 details several important components of the management of acute variceal hemorrhage.

Table 4.

Management of Acute Variceal Bleeding

	Adult Recommendationsa	Summary of Pediatric Studies and Expert Opinionb
Red cell transfusion	Fluid resuscitation aims to preserve tissue perfusion. Provide red cell transfusion to a target hemoglobin concentration of 70‐80 g/L while accounting for other factors, including comorbidities, hemodynamic stability, and ongoing bleeding.	There are no pediatric data in studies of children with variceal hemorrhage. Studies of other pediatric populations suggest that a target of 70‐80 g/L is appropriate.
Vasoactive drugs	Initiate somatostatin, octreotide, vapreotide, or terlipressin as soon as variceal bleeding is suspected and continue for 3‐5 days.	Case series suggest that the adult approach is safe and likely to be effective for children. No pediatric randomized controlled trials are available. Expert opinion supports the application of the adult recommendations to children.
Antibiotics	Short‐term antibiotics in all patients, started immediately on admission (e.g., oral norfloxacin, intravenous ciprofloxacin, or intravenous ceftriaxone).	There are no pediatric data on the incidence of infection in children with variceal hemorrhage. Expert opinion recommends maintaining a high index of suspicion and commencing antibiotics early if any signs of sepsis are present.
Endoscopy	EGD should be performed within 12 hours for diagnosis and treatment using EVL (or EST if EVL is technically difficult).	Case series support the expert opinion that EGD should be performed as soon as possible (within 24 hours) and once the patient is hemodynamically stable.
Management of rebleeding, TIPS	Consider early placement of TIPS (within 72 hours) if there is a high risk of treatment failure (e.g., Child‐Pugh class C). Repeat attempt at endoscopic therapy if rebleeding occurs within 5 days. TIPS is indicated in patients with bleeding that cannot be controlled or recurs despite drugs and EVL.	Minimal pediatric data are available. Expert opinion agrees with adult recommendations under the assumption that an attempt at repeat endoscopic therapy has been undertaken.
Balloon tamponade	Balloon tamponade should be used as a temporizing measure (for a maximum of 24 hours) in patients with uncontrollable bleeding for whom a more definitive therapy is planned (e.g., TIPS or endoscopic therapy).	Use only very rarely in patients with massive bleeding as a temporary bridge (maximum 24 hours) to a more definitive therapy, preferably in an intensive care unit with experienced health care professionals.

^aDerived from guidelines issued by the American Association for the Study of Liver Disease and the Baveno V Consensus Workshop.

^bThe expert opinions are taken from Shneider et al.7

Secondary Prophylaxis of Variceal Bleeding

Children who have experienced a variceal bleed should be offered secondary prophylactic treatment to reduce the otherwise high risk of recurrent bleeding. Multiple published case series and 1 randomized controlled clinical trial describe secondary prophylaxis in children with varices (Table 5). The frequency of rebleeding seems to be greater with EST than with EVL. There are currently inadequate data concerning the efficacy and safety of nonselective beta‐blockers to recommend their routine use for secondary prophylaxis in children.

Table 5.

Secondary Prophylaxis of Variceal Bleeding in Children

Study Design	Pediatric Studies (n)	Subjects (n)	Treatment Method With Number of Studies	Duration of Follow‐up	Frequency of Rebleeding
Retrospective case series	19	Median: 36 (range = 4–197)	Beta‐blocker: 2	3 and 5 years	25% and 53%
			EVL: 3	1–12 years	18% on average
			EST: 12	1–10 years	27% on average
			EVL + EST: 1	2 years	11% with EST alone and 3% combined
			EST + beta‐blocker: 1	2 years	25% with EST alone and 16% combined
Prospective case series	2	50, 28	EST: 1	19 months	26%
			EVL: 1	21 months	14%
Randomized controlled trial	1	49	EST versus EVL	2 years	25% with EST and 4% with EVL

Children with portal vein thrombosis who have experienced a variceal bleed should be considered for Rex bypass surgery or, when the Rex bypass is not feasible, a distal splenorenal shunt.

If bleeding recurs despite endoscopic therapy in a child with cirrhosis or an intrahepatic cause for portal hypertension, then treatment options include the continuation of previous secondary prophylactic measures (e.g., EVL), the placement of a transjugular intrahepatic portosystemic shunt (TIPS), portosystemic shunt surgery, and liver transplantation. The choice of treatment is dictated by the individual patient's circumstances, including the presence of indications for liver transplantation, the estimated longevity of native liver function, and the available expertise for TIPS placement or surgery.

Summary

Recommended approaches to the management of portal hypertension and esophageal varices in children continue to rely on expert opinion, low‐quality pediatric studies, and the extrapolation of the results of studies in adult populations. There is a pressing need for high‐quality pediatric studies, and expert panels recently have itemized the priorities for future pediatric research in this area.7, 19 Multicenter and international collaborations will be required to provide the evidence needed to advance the clinical care that is provided to affected children.