Prophylactic endoscopic sclerotherapy of large esophagogastric varices in infants with biliary atresia

CASE STUDIES

Prophylactic endoscopic sclerotherapy of large esophagogastric varices in infants with biliary atresia ´, MD, Dalila Habe `s, MD, Philippe Roulleau, MD, Vincent Haas, MD, Mathieu Duche Emmanuel Jacquemin, MD, PhD, Olivier Bernard, MD, PhD Biceˆtre, France

Background: Esophageal varices–related GI bleeding occurs frequently and early in life in children with biliary atresia and it may be life threatening. Objective: We report the results of prophylactic sclerotherapy in 13 infants with biliary atresia and large varices. Patients: Mean age was 13 months, mean weight was 8.2 kg, mean total serum bilirubin was 258 mmol/L, and mean prothrombin time was 78%. Esophageal varices were grade III (11 patients) or II (2 patients), with red signs in all infants and gastric varices in 12. None had GI bleeding. Intervention: Sclerotherapy was performed with the patient under continuous intravenous octreotide therapy in 7 infants. Results: In 8 children a complete or almost complete eradication of varices was obtained; none of these children bled later, 4 underwent liver transplantation, 3 are alive without liver transplantation, and 1 died of sepsis after 9 months awaiting liver transplantation. In 4 children a partial eradication was obtained and liver transplantation was performed. None of these children bled. One other child bled to death after 2 sessions of sclerotherapy. Limitations: Four ulcers and 2 stenoses occurred in 6 children with no octreotide versus no ulcer and 1 stenosis in 7 children receiving octreotide. Conclusion: These results (1) indicate that primary prevention of GI bleeding by sclerotherapy of esophageal varices is technically feasible and fairly effective in infants with biliary atresia and large varices, even in those with end-stage liver disease, (2) suggest that decreasing the risk of bleeding may allow liver transplantation under better conditions, and (3) further suggest that octreotide associated with sclerotherapy lowers the rate of complications.

Biliary atresia, affecting 1 in 20,000 newborn infants, is an important cause of cirrhosis in children and the main indication for pediatric liver transplantation.1-3 Its current management includes, sequentially, the operation described by Kasai (hepatic portoenterostomy or hepatic portocholecystostomy), followed by liver transplantation, either because of failure of the Kasai operation or because of late complications of cirrhosis.4 The development of portal hypertension may be extremely rapid in these children, and GI bleeding has been reported in children below 2 years of age.5,6 In these

Copyright ª 2008 by the American Society for Gastrointestinal Endoscopy 0016-5107/$32.00 doi:10.1016/j.gie.2007.11.005

infants, who often are in poor condition because of jaundice, malnutrition, or ascites, GI bleeding from esophagogastric varices can have devastating consequences, including acceleration of liver failure, ascites, bacterial peritonitis, kidney failure, acute ischemic liver necrosis, and death.7,8 The risk of GI bleeding can be estimated from the size of the esophageal varices, the appearance of the esophageal mucosa, and the presence of gastric varices: the presence of large and tense (grade II or III) esophageal varices, red mucosal signs, and gastric varices is associated in adults and in children with a high risk of spontaneous GI bleeding.9,10 Here we report a prospective trial of systematic detection of severe varices and primary prophylaxis of GI bleeding by endoscopic sclerotherapy in infants who were followed up after Kasai operation for biliary atresia in an attempt to prevent the complications of GI bleeding.

732 GASTROINTESTINAL ENDOSCOPY Volume 67, No. 4 : 2008

www.giejournal.org

Abbreviation: GOV, gastroesophageal varices.

´ et al Duche

Prophylactic sclerotherapy of varices in infants

TABLE 1. Clinical and biochemical features at the time of endoscopy of 13 infants with biliary atresia and major endoscopic signs of portal hypertension

Ascites

Weight (kg)

Serum bilirubin (mmol/L)

Serum albumin (g/L)

Prothrombin time (%)

20

0

11.1

33

37

95

2

19

0

11.5

367

31

64

3

19

0

9

37

38

80

4

17

Mild

8.2

54

39

82

5

10

Mild

5.5

389

32

91

6

11

Major

9.6

579

36

83

7

17

0

8.5

14

40

100

8

11

0

7.3

308

37

94

9

16

0

8.6

89

37

75

10

6

Mild

5.5

292

21

53

11

10

0

6.7

370

25

78

12

12

Mild

9.3

469

34

72

13

5

Mild

6.5

232

35

58

Patient no.

Age (mo)

1

PATIENTS AND METHODS

The initial upper GI endoscopic evaluation was carried out with the patient under general anesthesia or intrarec-

tal midazolam sedation (0.2-0.3 mg/kg), and the severity of endoscopic findings was graded as follows: (1) esophageal varices: grade I when varices were flattened by insufflation, grade II when varices were not flattened by insufflation but separated by healthy mucosa, and grade III when varices were not flattened by insufflation and confluent; (2) mucosal red signs: they were estimated as mild or severe according to previous descriptions11,12; (3) gastric varices were recorded as described by Sarin and Kumar13: type I gastroesophageal varices (GOV) extending from the esophagus across the cardia; type II GOV not extending across the cardia; presence of red signs on the GOV mucosa was recorded as well as the presence of portal hypertensive gastropathy.13-15 Endoscopic sclerotherapy was carried out by the same endoscopist (M.D.) using an Olympus fiberoptic XQ 20 endoscope (Olympus, Tokyo, Japan) with the patient under general anesthesia with intubation; 1.2% aetoxysclerol was injected at a mean volume of 9 mL per session and a mean volume of 2.5 mL per injection site. Systemic antibiotic treatment with amoxicillin-clavulanic acid was given before the procedure. Endoscopy sessions were repeated with the objective of reducing as much as possible the risk of bleeding, adjusting the number of sessions to the evolution of the endoscopic signs. Two to 5 weekly sessions were followed by endoscopy/sclerotherapy sessions every 1 to 3 months, then every 6 months for 1 year, and then every year for the children who did not require liver transplantation. Proton pump inhibitor was given for at least 2 months. Bleeding in the course or after completion of the sclerotherapy process were recorded as

www.giejournal.org

Volume 67, No. 4 : 2008 GASTROINTESTINAL ENDOSCOPY 733

From 1989 to 2005, 303 infants operated on for biliary atresia and with clinical and ultrasonographic signs of portal hypertension10 had an upper GI endoscopy at younger than age 20 months. Among these, 13, who had not bled before, had large, tense esophagogastric varices and underwent primary prophylaxis of bleeding by endoscopic sclerotherapy.

Patients There were 9 girls and 4 boys: 10 had undergone hepatic portoenterostomy and 3 had undergone hepatic portocholecystostomy at a mean age of 65 days (range 20-150 days) (Table 1). At the time of endoscopy showing large and tense varices, the mean patient age was 13 months (range 5-20 months), and their mean weight was 8.2 kg (range 5.5-11.5 kg). All but one were jaundiced, with total serum bilirubin concentrations ranging from 33 to 579 mmol/L (normal !17 mmol/L), and 8 patients had a serum bilirubin level greater than 200 mmol/L. Ascites was clinically present in 6 patients. Severe malnutrition required continuous nocturnal nasogastric feeding in 7 patients and total parenteral nutrition in 1 patient. The mean prothrombin time was 78% (range 53%-100%) (normal O70%), and the mean total serum albumin concentration was 34 g/L (range 21-40 g/L) (normal O35 g/L).

Methods

´ et al Duche

Prophylactic sclerotherapy of varices in infants

TABLE 2. Results of prophylactic endoscopic sclerotherapy in 13 infants with biliary atresia and large esophageal varices Endoscopy before sclerotherapy Esophageal varices Patient (no./grade/ no. red signs)

Gastric varices (grade/ red signs)

Endoscopy after sclerotherapy

Sclerotherapy

No. of Octreotide sessions therapy Complications

Esophageal varices (no./grade/ red signs)

Gastric varices (grade/ red signs)

Follow-up*

Time (d)

GI bleeding

Outcome

1

3/mild

None/ none

3

No

Ulcerationy

1/I/none

None/ none

274

None

Death, fulminant sepsis

2

3/III/mild

GOV1/ present

2

No

Ulcerationy

1/I/none

None/ none

257

None

Liver transplantation

3

3/III/major

GOV1/ present

4

No

Stenosisy/GI bleeding

None/n.a./ none

None/ none

4

1/II/major

GOV1/ present

2

No

GI bleeding

1/II/mild

GOV1/ none

18

Yes

Death, refractory bleeding

5

3/III/ mild

GOV1/ none

4

No

GI bleeding/ ulceration

2/II/mild

GOV1/ none

72

None

Liver transplantation

6

3/III/major

GOV1/ none

5

No

GI bleeding/ ulceration

2/ II/mild

GOV1/ none

198

None

Liver transplantation

7

3/III/ major GOV1/ none

3

Yes

None

1/I/none

None/ none

785

None

Alive, bilirubin 9 mmol/L

8

2/II/major

GOV1/ present

6

Yes

None

1/I/none

None/ none

869

None

Liver transplantation

9

1/III/major

GOV1/ present

6

Yes

None

None/n.a./ none

None/ none

355

None

Alive, bilirubin 17 mmol/L

10

3/III/major

GOV11/ present

5

Yes

Stenosisy

None/n.a./ present

None/ none

94

None

Liver transplantation

11

1/III/major

GOV1/ present

6

Yes

None

1/II/0

None/ none

180

None

Liver transplantation

12

3/III/major

GOV1/ present

4

Yes

None

2/III/mild

GOV1/ present

63

None

Liver transplantation

13

1/III/ major GOV1/ none

8

Yes

None

None/n.a./ none

GOV1/ present

261

None

Liver transplantation

1800 None

Alive, bilirubin 35 mmol/L

n.a., Not applicable. *After last sclerotherapy session. yDetected during endoscopy; there were no clinical symptoms.

well as the complications of the technique and the endoscopic features at the end of the sclerotherapy. From 2000 on, octreotide was given intravenously during sclerotherapy at a continuous dose of 2 mg/kg per hour, starting 1 hour before the first session, progressively tapered down to 1 mg/kg per hour until the next session and stopped when the GI bleeding risk was considered significantly reduced, as shown by a decrease of the grade of the varices and the disappearance of the mucosal red signs and the gastric varices.

The end points of the study were the time of death or liver transplantation or the last visit. This study was carried out in accordance with the Helsinki Declaration as revised in 1989, and informed consent was obtained from the parents of all children.

734 GASTROINTESTINAL ENDOSCOPY Volume 67, No. 4 : 2008

www.giejournal.org

RESULTS All infants had endoscopic signs of impending GI bleeding (Table 2). Eleven infants had grade III esophageal

´ et al Duche

Prophylactic sclerotherapy of varices in infants

varices and the other 2 had grade II esophageal varices; all had red signs on the esophageal mucosa and 12 had GOV type 1, including 8 with red signs. Two infants, awaiting liver transplantation at ages 6 and 11 months, respectively, had the most impressive endoscopic signs with pseudotumoral esophageal varices completely obstructing the lumen and were treated in the intensive care unit under general anesthesia and mechanical ventilation. The mean number of endoscopic sclerotherapy sessions was 4.2 (range 2 to 8). In a first group of 8 infants (patients 1-3, 7-10, and 13 in Table 2), a complete, or almost complete, eradication of esophagogastric varices was obtained; none of them bled again, except for one minor bleeding in the course of the procedure not requiring blood transfusion. Three of these children are alive without the need for liver transplantation 1, 2, and 5 years after completion of sclerotherapy; their total serum bilirubin concentration is 9 to 35 mmol/L and no relapse of varices was observed. Four other infants underwent liver transplantation 3 to 28 months after completion of sclerotherapy, the latter requiring 2 additional sclerotherapy sessions for relapse of varices at 1 year. One child died with fulminant sepsis while awaiting transplantation abroad, 9 months after eradication of varices. All but one of the infants in whom a complete or almost complete eradication was achieved had grade III esophageal varices, including the 2 with the most severe signs as described above. Besides 2 cases of mild stenosis not requiring any treatment, no complication related to the endoscopic procedure was observed. In a second group of 4 infants (patients 5, 6, 11, and 12 in Table 2), a decrease in the grade of esophagogastric varices and the disappearance of red mucosal signs resulting in a reduction of risk of bleeding were observed, and liver transplantation was performed 2 to 6 months after completion of sclerotherapy, respectively. None of these infants bled while awaiting liver transplantation, except for 1 small episode not requiring blood transfusion in 2 patients each in the course of sclerotherapy. All had grade III esophageal varices, with red signs. Patient 4, early in the experience, died because of massive refractory GI bleeding 10 days after the second session. During the period of follow-up, a mean increase in weight (excluding ascites) of þ1 standard deviation [SD] (range 0.5-2) was observed in 7 children, whereas 5 others had stable growth and 1 lost 1 S. Prothrombin time improved (þ13, to 38%) in 5 children and remained stable in the other 8. The total serum bilirubin concentration improved or remained normal in the 3 children who did not require liver transplantation. No portal vein thrombosis was observed; the mean diameter of the portal vein did not change: it was 4 mm (range 2-6 mm) before sclerotherapy and 4.5 mm (range 2-8 mm) at the end of follow-up. Continuous octreotide infusion was given to the last 7 infants in the study. No side effects were recorded. As

shown in Table 2, no GI bleeding and no ulcers were observed in this group of patients, as opposed to 4 episodes of GI bleeding and 4 ulcers during the course of the procedure in the 6 earlier infants not treated with octreotide. The subjective impression of the endoscopist was that, in children receiving octreotide therapy, there was a lower bleeding tendency from the varix while removing the injection needle, a lesser need for multiple injections on the same site, and the overall feeling of an easier procedure. Endoscopic signs of portal hypertensive gastropathy were present in 10 children before starting sclerotherapy; gastropathy was considered mild (ie, described as red snake skin) in 5 infants and severe (ie, red spots or hemorrhagic gastritis) in the other 5. After completion of sclerotherapy, signs of gastropathy appeared or increased in 2 infants, decreased in 5, and remained stable in 4; in the remaining 2 children follow-up was too short to provide any information in that respect. There was an equal distribution of the evolution of the signs of gastropathy in children who received octreotide and in those who did not. Portal hypertensive gastropathy was not responsible for clinically apparent GI bleeding during the period of study.

www.giejournal.org

Volume 67, No. 4 : 2008 GASTROINTESTINAL ENDOSCOPY 735

DISCUSSION Primary prophylaxis for variceal bleeding is currently recommended in adult patients with cirrhosis and medium-sized or large varices because the risk of bleeding is considered high in these patients and carries a risk of death.16 Both endoscopic banding and b-blockade are used for that purpose with similar results. In children, it is commonly held that treatment of portal hypertension should await a first episode of GI bleeding, and unlike what is advocated in adults, primary prophylaxis is a matter of controversy.17 However, two previous reports support the feasibility and efficacy of a prophylactic treatment. In a randomized study of 100 children with a variety of liver diseases, endoscopic sclerotherapy reduced the risk of bleeding over a 4-year period; most children in this report were older (mean age 4.3 years), the severity of the endoscopic findings were not described, and the liver condition was good (Child A) in more than 80% of cases.18 Another study reports on the efficacy of prophylactic ligation of esophageal varices in 37 children with various causes of portal hypertension and severe endoscopic signs who did not bleed after a mean follow-up of 16 months after completion of treatment; all children in this study were older than 4 years.19 The current report extends these results to a homogeneous population of infants with biliary atresia who in the majority of instances had end-stage liver disease. There is, indeed, a justification to consider prophylactic treatment in infants with biliary atresia and portal hypertension for the following reasons: (1) cirrhosis is almost

´ et al Duche

Prophylactic sclerotherapy of varices in infants

invariably present in these children regardless of the result of the Kasai operation20; (2) even in children who are free of jaundice after the Kasai operation, 20% to 50% display varices on endoscopy21; (3) children with biliary atresia are prone to bleeding at an early age5,6,10; (4) in case of failure of the Kasai operation, liver transplantation has to be performed in the first 3 years of life in the majority of patients, including a fair proportion before age 1 year22; (5) the results of liver transplantation for biliary atresia in children before age 1 year are worse than in older children23; (6) in children operated on for biliary atresia who remain jaundiced, GI bleeding carries a significant risk of death or of requiring urgent transplantation,8 and preventing GI bleeding in infants with biliary atresia who display endoscopic signs of impending bleeding would therefore provide the child with a better chance of growth, allow the wait for a suitable liver, and lower the need for a transplantation in an emergency, a factor also associated with a poorer outcome4; and (7) in addition, it has been shown that children without a failed Kasai operation have extended survival if their GI bleeding is managed aggressively and effectively.8 Therefore, an effective and relatively safe means of preventing GI bleeding in infants with biliary atresia and portal hypertension appears warranted. There are no convincing studies on the efficacy of b-blocking agents in children with portal hypertension, especially in infants.19,24 Endoscopic band ligation, the preferred method in adults and older children,16,17 is not commonly performed in infants because of the size of most available equipment17; indeed, most studies reporting the efficacy of variceal ligation in children deal with children aged 3 years or more.19,25 Endoscopic variceal sclerotherapy remains, therefore, a valid option in infants; the apparent high risk of complications of the technique, an issue often raised to favor band ligation instead of sclerotherapy in children,19,24 was not the case in this series. Subcutaneous octreotide was introduced in 1997 as an adjuvant treatment to sclerotherapy in adults with cirrhosis who had bled once; it was shown to reduce the risk of rebleeding during endoscopic treatment.26,27 Although the numbers are small and the study was not controlled, the results reported here suggest that continuous octreotide therapy could prove useful and safe in infants undergoing endoscopic treatment of esophageal varices, both in terms of risk of rebleeding and in terms of risk of ulceration. One cannot exclude, however, that the lower rate of complications observed under octreotide therapy could be related, at least in part, to the experience of the endoscopist. Although these results, as a whole, suggest that primary prophylaxis sclerotherapy should be considered in infants with biliary atresia and severe endoscopic signs of portal hypertension, they cannot guarantee success in all instances, as shown by the unfortunate bleeding to death of one child. Possibly subcutaneous octreotide should be

given for several months after the end of sclerotherapy in infants awaiting liver transplantation, as suggested for adults.26 In conclusion, the results reported here indicate that in infants with biliary atresia, with major endoscopic signs of portal hypertension and a severe liver condition in the majority of cases, endoscopic sclerotherapy can be carried out successfully and relatively safely and can result in a disappearance or reduction in the size of the esophageal varices, a decrease or disappearance of red signs and of gastric varices, and a reduction in the risk of bleeding in the short term, therefore eliminating one of the factors aggravating the liver condition during the waiting period for liver transplantation. In addition, the results suggest that the use of octreotide may be beneficial in limiting the risk of bleeding during the procedure.

736 GASTROINTESTINAL ENDOSCOPY Volume 67, No. 4 : 2008

www.giejournal.org

DISCLOSURE The authors report that there are no disclosures ´ fe ´ rrelevant to this publication. Supported by Centre de re ´ sie des voies biliaires (France). ence national de l’atre REFERENCES 1. Sokol RJ, Mack C, Narkewicz MR, et al. Pathogenesis and outcome of biliary atresia: current concepts. J Pediatr Gastroenterol Nutr 2003;37: 4-21. 2. Tiao G, Ryckman FC. Pediatric liver transplantation. Clin Liver Dis 2006;10:169-97. 3. Chardot C, Carton M, Spire-Bendelac N, et al. Epidemiology of biliary atresia in France: a national study 1986-96. J Hepatol 1999;31:1006-13. 4. Otte JB, de Ville de Goyet J, Reding R, et al. Sequential treatment of biliary atresia with Kasai portoenterostomy and liver transplantation: a review. Hepatology 1994;20:41-8S. 5. Lilly JR, Stellin G. Variceal hemorrhage in biliary atresia. J Pediatr Surg 1984;19:476-9. 6. Stringer MD, Howard ER, Mowat AP. Endoscopic sclerotherapy in the management of esophageal varices in 61 children with biliary atresia. J Pediatr Surg 1989;24:438-42. 7. Gartner JC Jr, Jaffe R, Malatack JJ, et al. Hepatic infarction and acute liver failure in children with extrahepatic biliary atresia and cirrhosis. J Pediatr Surg 1987;22:360-2. 8. Miga D, Sokol RJ, Mackenzie T, et al. Survival after first esophageal variceal hemorrhage in patients with biliary atresia. J Pediatr 2001;139:291-6. 9. North Italian Endoscopic Club for the Study and Treatment of Esophageal Varices. Prediction of the first variceal hemorrhage in patients with cirrhosis of the liver and esophageal varices: a prospective multicenter study. N Engl J Med 1988;319:983-9. 10. Bernard O, Alvarez F, Brunelle F, et al. Portal hypertension in children. Clin Gastroenterol 1985;14:33-55. 11. Beppu K, Inokuchi K, Koyanagi N, et al. Prediction of variceal hemorrhage by esophageal endoscopy. Gastrointest Endosc 1981;27:213-8. 12. Cales P, Zabotto B, Meskens C, et al. Gastroesophageal endoscopic features in cirrhosis: observer variability, interassociations, and relationship to hepatic dysfunction. Gastroenterology 1990;98:156-62. 13. Sarin SK, Kumar A. Gastric varices: profile, classification, and management. Am J Gastroenterol 1989;84:1244-9. 14. Papazian A, Braillon A, Dupas JL, et al. Portal hypertensive gastric mucosa: an endoscopic study. Gut 1986;27:1199-203.

´ et al Duche

Prophylactic sclerotherapy of varices in infants 24. Shashidhar H, Langhans N, Grand RJ. Propranolol in prevention of portal hypertensive hemorrhage in children: a pilot study. J Pediatr Gastroenterol Nutr 1999;29:12-7. 25. Sasaki T, Hasegawa T, Nakajima K, et al. Endoscopic variceal ligation in the management of gastroesophageal varices in postoperative biliary atresia. J Pediatr Surg 1998;33:1628-32. 26. Jenkins SA, Baxter JN, Critchley M, et al. Randomised trial of octreotide for long term management of cirrhosis after variceal haemorrhage. BMJ 1997;315:1338-41. 27. Avgerinos A, Nevens F, Raptis S, et al. Early administration of somatostatin and efficacy of sclerotherapy in acute oesophageal variceal bleeds: the European Acute Bleeding Oesophageal Variceal Episodes (ABOVE) randomised trial. Lancet 1997;350:1495-9.

15. McCormack TT, Sims J, Eyre-Brook I, et al. Gastric lesions in portal hypertension: inflammatory gastritis or congestive gastropathy? Gut 1985;26:1226-32. 16. Boyer TD. Primary prophylaxis for variceal bleeding: are we there yet? Gastroenterology 2005;128:1120-2. 17. McKiernan PJ. Treatment of variceal bleeding. Gastrointest Endosc Clin North Am 2001;11:789-812. 18. Goncalves ME, Cardoso SR, Maksoud JG. Prophylactic sclerotherapy in children with esophageal varices: long-term results of a controlled prospective randomized trial. J Pediatr Surg 2000;35:401-5. 19. Celinska-Cedro D, Teisseyre M, Woynarowski M, et al. Endoscopic ligation of esophageal varices for prophylaxis of first bleeding in children and adolescents with portal hypertension: preliminary results of a prospective study. J Pediatr Surg 2003;38:1008-11. 20. Gautier M, Valayer J, Odievre M, et al. Histological liver evaluation 5 years after surgery for extrahepatic biliary atresia: a study of 20 cases. J Pediatr Surg 1984;19:263-8. 21. Kasai M, Ohi R, Ohuchi N. Reversibility and aggravation of portal hypertension in biliary atresia patients. In: Kasai M, editor. Biliary atresia and related disorders. Amsterdam: Excerpta Medica; 1983. p. 291-6. 22. Fouquet V, Alves A, Branchereau S, et al. Long-term outcome of pediatric liver transplantation for biliary atresia: a 10-year follow-up in a single center. Liver Transpl 2005;11:152-60. 23. Goss JA, Shackleton CR, McDiarmid SV, et al. Long-term results of pediatric liver transplantation: an analysis of 569 transplants. Ann Surg 1998;228:411-20.

Reprint requests: Olivier Bernard, MD, PhD, Division of Pediatric Hepatology, Department of Pediatrics, Hopital Biceˆtre, 78 rue du Ge´ne´ral Leclerc, Le Kremlin Biceˆtre Cedex 94275, France.

www.giejournal.org

Volume 67, No. 4 : 2008 GASTROINTESTINAL ENDOSCOPY 737

Received June 4, 2007. Accepted November 5, 2007. Current affiliations: Division of Radiologie (M.D.), Division of He´patologie (M.D., D.H., E.J., O.B.), Division of Re´animation (V.H.), Division of Pe´diatriques, and Anesthe´sie-Re´animation (P. R.), Hopital Bicetre, Le Kremlin Biceˆtre, France.