ORIGINAL ARTICLE

Revisiting Extra Hepatic Portal Vein Obstruction in Childrenfrom the North Indian Gangetic Plain

Sundeep Goyal & V. K. Dixit & A. K. Jain & O. P. Mishra &

S. K. Jena & Jayant Ghosh

Received: 4 April 2013 /Accepted: 20 September 2013 /Published online: 10 October 2013# Dr. K C Chaudhuri Foundation 2013

AbstractObjectives To study the demographic features, etiology, clini-cal, radiology and laboratory findings in children with Extrahepatic portal vein obstruction (EHPVO) from the North IndianGangetic Plain.Methods A prospective analysis of 53 patients of EHPVO(<14 y of age) was done. Data for clinical presentation,laboratory workup, endoscopic procedures, growth and devel-opment, management and outcome were analyzed.Results A total of 53 patients (32 male, 21 female) with meanage of 8.66 ± 3.32 y at presentation were included. Growthretardation was present in 30 (56.60 %) patients at time ofdiagnosis. The most common presenting symptoms wereupper gastrointestinal bleeding (86.80 %) and an incidentalfinding of splenomegaly (13.2 %). Hematological parametersof hypersplenism were present in 9 (17 %) patients. Protein C(PC), Protein S (PS), antithrombin III (AT) deficiency werefound in 8 (20 %), 4 (10 %) and 6 (15 %) of the patientsrespectively in 40 patients tested. On first endoscopy, esoph-ageal varices were present in all patients. Endoscopic sclero-therapy (EST)/band ligation (EBL) was performed in 46bleeder patients and success rate was 83.3 % for sclerotherapyand 90 % for band ligation.Conclusions The etiology of EHPVO in the majority of pa-tients remains elusive. It results in impaired somatic growth.Sclerotherapy and endoscopic banding are effective meansfor treatment for bleeding varices. It is still not clear whether

deficiency of anticoagulant proteins is a primary event orsecondary to disease process.

Keywords Extra hepatic portal venous obstruction .

Esophageal varices . Thrombophilia . Protein C . Protein S .

Antithrombin III

Introduction

Extra hepatic portal vein obstruction (EHPVO) is defined asobstruction of the extra hepatic portal vein with or withoutinvolvement of the intra hepatic portal veins or splenic orsuperior mesenteric veins [1, 2]. EHPVO is the commonestcause of portal hypertension (PHT) in children [3, 4]. Inadults, hypercoagulable states play an important role in thepathogenesis of EHPVO in a significant proportion of casesand are due to deficiencies of natural anticoagulants likeprotein C (PC), protein S (PS), anti thrombin III (AT) or excessproduction of procoagulants due to factor V leiden or pro-thrombin gene mutation [2]. Local factors like inflammation,trauma and malignancy are also linked to portal vein throm-bosis (PVT) in adults. The etiology of extra hepatic portal veinobstruction is highly diverse in children. The three broadcategories are; infections (omphalitis, neonatal umbilical sep-sis, umbilical vein cannulation, sepsis, surgery etc.), congen-ital anomaly (obstruction along the line of left and rightvitelline veins) and prothrombic state. However most of thecases remain idiopathic [1, 5, 6]. Deficiency of natural anti-coagulant proteins is common in EHPVO and is probably asecondary phenomenon in most cases, occurring as a part ofglobal disturbance of coagulation variables [7–9].

Variceal bleeding is the commonest complication leadingto diagnosis of EHPVO in children; however, management ofEHPVO extends to broad spectrum of complications likegrowth failure, hypersplenism and ectopic varices.

S. Goyal (*) :V. K. Dixit :A. K. Jain : S. K. Jena : J. GhoshDepartment of Gastroenterology, Institute of Medical Sciences,Banaras Hindu University, Varanasi 221005, Indiae-mail: drsundeepgoyal@gmail.com

O. P. MishraDepartment of Pediatrics, Institute of Medical Sciences,Banaras Hindu University, Varanasi, India

Indian J Pediatr (May 2014) 81(5):429–433DOI 10.1007/s12098-013-1257-7

The aims of this study were to assess demographic features,etiology, clinical and laboratory findings, with special refer-ence to thrombophilic factors like protein C, protein S andantithrombin III deficiency in children with EHPVO in north-eastern part of India.

Material and Methods

The study involved 53 patients who consecutivelypresented to the Gastroenterology unit, at Banaras HinduUniversity hospital, in Varanasi, India from June 2008through June 2012. The diagnosis of EHPVO was basedon the history of upper gastrointestinal bleeding,presence of splenomegaly, endoscopic demonstrationof esophageal varices, and the documentation ofcavernomatous transformation of the portal vein on ultra-sonography. The exclusion of associated liver disease wascarried out through clinical, laboratorial and radiologicalevaluation. Various variables like age at the diagnosis, birthconditions (weight and complications in the perinatal peri-od, such as infections, use of umbilical catheter andomphalitis history), and form of disease presentation wererecorded. The anthropometric data including weight wasmeasured in light clothing and without shoes to the nearesthalf-kilogram on a calibrated digital scale. Height wasmeasured in standing position and recumbent length wasmeasured in younger (<2 y) children to the nearest half-centimeter. Stunted growth was defined as height forage<−2 Z score and wasting as weight for height<−2Z score. All patients with upper gastrointestinal bleeding(UGIB), after the emergency approach, underwent asecondary endoscopic prophylaxis—sclerotherapy (EST)or endoscopic band ligation (EBL). Three percent phe-nol was used as sclerosant. Patients underwent endosco-py every 3–4 wk until the variceal eradication.

Protein C and protein S were studied with Zumutest proteinC & S kit (Hyphen Bio Med, France), a two-site ELISAmethod for the measurement of human protein C and pro-tein S in citrated plasma. Antithrombin III was estimated byDiffuplate (Bio Cientifica SA, Argentina), a radial immu-nodiffusion (RID) for the determination of antithrombin IIIin serum.

The institution’s ethical committee approved the studyprotocol and written informed consent was obtained fromthe patients or their parent/guardian as appropriate.

The statistical analysis was done by SPSS (version 20.0).The descriptive statistics was calculated for continuous andqualitative data. Mean±SDwas calculated for continuous dataand frequency, and percentage were calculated for qualitativedata.

Results

Fifty three consecutive patients with portal vein thrombosis(PVT), without associated liver disease were evaluated. Thirtytwo (60.38 %) patients were males and 21 (39.62 %) werefemales. The age at the time of diagnosis ranged from 2 to14 y, with mean age of 8.66 ±3.32 y at diagnosis. Thedemographic and clinical characteristics are summarizedin Table 1.

Growth retardation was present in 30 (56.60 %) children.Wasting was found in 10 (18.86 %) cases, whereas stuntingwas found in 15 (28.3 %) of cases. Both stunting and wastingwere found in 5 (9.44 %) cases.

History of home delivery was found in 33 (62.3 %) caseswhereas 20 (37.7 %) cases had delivery in hospital. Previous

Table 1 Demographic data, clinical and laboratory characteristics ofpatients

Variable N (%)

Age (mean ± SD) y 8.66 ±3.32

Male 32 (60.38)

Female 21 (39.62)

History of delivery

Home delivery 33 (62.3)

Hospital delivery 20 (37.7)

Hypercoagulable state (N =40)

Protein C deficiency 8 (20)

Protein S deficiency 4 (10)

Anti-thrombin III deficiency 6 (15)

MTHFR/JAK2 mutation (N =5) 0

Clinical presentation

Upper GI bleeding 46 (86.8)

Splenomegaly 7 (13.2)

Physical and laboratory findings

Splenomegaly 49 (92.5)

Hepatomegaly 7 (13.2)

Ascites 4 (7.5)

Elevated liver enzymes 5 (9.4)

Hypersplenism 9 (17)

Growth retardation 30 (56.60)

Wasting 10 (18.86)

Stunting 15 (28.3)

Stunting and wasting 5 (9.44)

Initial endoscopy findings

Esophageal varices 53 (100)

Small caliber 25 (47.16)

Large caliber 28 (52.83)

Gastric varices (GOV1) 18 (33.96)

PHG 10 (18.86)

430 Indian J Pediatr (May 2014) 81(5):429–433

records were available in 40 patients and were not suggestiveof umbilical catheterization, neonatal sepsis, abdominal infec-tion, cardiovascular malformations or abdominal surgery inany of the patients. Regarding the clinical course, upper GIbleeding (hematemesis and/or melena) was the most common(86.8 %) clinical presentation, followed by incidental findingof splenomegaly in 7 (13.2 %) children with EHPVO. Outof 46 bleeders 40 patients had past history of self-limitingbleeding episodes. Mean numbers of bleeding episodesbefore presentation were 1.9. At the time of diagnosis, 49(92.5 %) patients were found to have splenomegaly. Hepa-tomegaly was present in 7 (13.2 %) cases; however, none ofthe patients with hepatomegaly showed abnormal liverfunction tests. Ascites was present in 4 (7.5 %) cases butwithout chronic liver disease. Five (9.4 %) patients hadtransiently elevated aspartate aminotransferase and/or ala-nine aminotransferase (<two times of upper limit of nor-mal) at some time during follow up. Nine (18.87 %) pa-tients had hypoalbuminemia on diagnosis, which improvedon follow up examination. Nine (17 %) patients had signsof hypersplenism including leucopenia (TLC <4000/mm3),thrombocytopenia (PLT <150,000/mm3).

The initial endoscopy finding showed presence of esopha-geal varices in all (100 %) patients and 46 (86.8 %) hadbleeding at the time of diagnosis. Small esophageal varices(<5 mm) were seen in 25 (47.16 %) patients while largevarices present in rest 28 (52.83 %) children. Gastric varices[all are gastroesophageal varices-1 (GOV1)] were present in18 (33.96 %) patients. Portal hypertensive gastropathy (allmild) was present in 10 (18.86 %) patients. During followup, two more patients had episode of hematemesis. Patientswith GI bleed underwent secondary prophylaxis. Out of 48bleeders, 36 (75 %) patients were treated with sclerotherapyand 12 (25 %) patients with band ligation. Variceal oblitera-tion was achieved successfully in 41 (85.42 %) cases; in 30(83.3 %) cases with 2–10 sessions (mean 5.43±2.01) in ESTgroup and in 11 (90 %) children with 2–4 sessions (mean 2.56±088) in EBL group. Esophageal stricture formed in five(13.9 %) patients who underwent sclerotherapy and it wastreated successfully with endoscopic dilatations. Followingvariceal eradication, mild Portal hypertensive gastropathy(PHG) progressed to severe PHG in six patients and fourpatients with mild PHG had failed endotherapy and werereferred for shunt surgery. Of the total of eight patients whohad failed endotherapy, two of them underwent side-to-sidelienorenal shunting and it was successful. One patient had co-existing thrombosis of both splenic and superior mesentericvein. Following endoscopic obliteration of esophageal varicesin 41 patients, 12 of them also had GOV1, which showed,increased size on follow-up and one patient developed GOV2.None of them bled from gastric varices. Rest six patients withGOV1 were in failed endo-therapy group.

Forty patients were investigated for PC, PS and AT defi-ciency. Methyl-tetra-hydro-folate reductase (MTHFR ) muta-tion and Janus kinase 2 (JAK2 ) V617F mutation wereassessed in five patients. Low PC and PS activity were foundin 8 (20%) and 4 (10%) cases respectively. The AT deficiencywas present in 6 (15 %) patients (Table 1).MTHFR mutationand JAK2 V617F mutationwere not detected in any of the fivecases studied.

Discussion

The present study summarizes a tertiary level referral centerexperience of EHPVO in children up to 14 y of age. In adults,a significant proportion of cases are due to procoagulant statebut it is not true for children. The etiology of EHPVO inpediatric age group is multifactorial, with association ofomphalitis, umbilical vein catherization, abdominal trauma,neonatal sepsis, dehydration, congenital agenesis or atresia ofportal vein. However no definite cause can be elucidated inmajority of cases [6, 9–11]. We found low PC activity only in20 % and low PS activity in 10 % of cases and none of thepatients tested showedMTHFR/JAK2 mutation. However, wedid not obtain blood samples from parents or sibling of indexcases, to determine whether the deficiency was hereditary. Arecent meta-analysis and older studies also substantiate theview that inherited AT, PC, and PS deficiencies are rare inEHPVO [12]. The mechanism for lower blood levels of AT,PC and PS remains unclear but may result from a combinationof reduced hepatic blood flow and portosystemic shuntingitself [7–9]. A larger study with more patients using geneticmarkers, especially prothrombin gene mutation along withJAK2 and MTHFRase and testing of parents and/or siblingmight provide more insight.

As observed in the present study, PVT in children does notshow gender predominance and its most common initial man-ifestations are upper gastrointestinal bleeding and spleno-megaly, which are in agreement with results reported in theliterature [3, 9–14]. Splenomegaly related to portal hyper-tension, even when not detected as the first manifestation, ispresent in most of the patients. In children it is reported in63 % to 92 % of the patients [9–11]. The results of this studycorroborate the others.

Laboratory results mostly showed normal values of ami-notransferases, albumin and coagulation tests in the indexstudy. Transient elevation of aminotransferases was found in9.4 % of cases. However, none of the patients with derangedliver enzymes had ascites. Ascites was present in a total of7.5 % cases and 18.87 % patients had hypoalbuminemia ondiagnosis and all of them presented with upper gastro intesti-nal bleeding (UGIB). Albumin may have decreased transient-ly and is associated with ascites, after episodes of UGIB [15,

Indian J Pediatr (May 2014) 81(5):429–433 431

16]. A proportion of patients with long-term portal hyperten-sion secondary to PVT may show ascites, low albumin,prolonged prothrombin time and deranged aminotransferases[15, 16]. This hepatic dysfunction may be attributed toprolonged reduction of the portal blood flow and/or develop-ment of portal biliopathy, which would indicate that PVT hasa progressive character [15, 17, 18]. Portal biliopathy, whichhas a progressive and silent course, usually manifests mainlyin the adulthood age though biliary abnormalities are seen inalmost all patients on MRCP/ERCP [18, 19]. All of thepresent patients were asymptomatic for biliopathy. Symptomssuch as abdominal pain, jaundice and fever developed in noneduring study period.

Hypersplenism leading to thrombocytopenia was detectedin 17 % of the index patients. Hypersplenism, manifesting asthrombocytopenia associated with splenomegaly, is a frequentmanifestation in pediatrics cases requiring hematology con-sultation [20, 21]. Growth retardation was present in almosthalf of the children (56.60 %) in present study. Literaturesupports the current observation that EHPVO occurring inthe pre-pubertal period could result in growth retardation inaround 50% of the children [1]. Few suggested hypothesis forreduced growth are: decreased portal blood supply to the liverand deprivation of hepatotrophic hormones; poor insulin de-livery to the liver and threreby decreased production of insulinlike growth factor-1 (IGF-1) and insulin like growth factorbinding protein-3 (IGFBP-3); poor substrate utilization or/andmalabsorption due to portal hypertensive enteropathy[22–25]. Improvement of growth after restoration of hepaticblood flow with mesenteric-left-portal bypass or Rex shunthas been documented [26, 27].

At the time of diagnosis, even the patients who did notbleed, had esophageal varices at the first endoscopic exami-nation in the index study. Esophageal varices were present inall cases and 90.57 % of them bled. Variceal obliteration wasachieved in 83.3 % cases in EST group and in 90 % ofchildren in EBL group. Esophageal stricture formed in13.9 % patients treated with sclerotherapy. There is enoughevidence for sclerotherapy and banding as effective secondaryprophylactic therapies for bleeding esophageal varices, butcomplications like ulcer (8 % to 30 %) and stricture (6 % to20 %) are associated with EST [3, 5, 6, 9, 11, 28, 29]. On theother hand, EVL has the advantages of rapid eradication ofvarices requiring fewer sessions and fewer associated com-plications. Endoscopic band ligation followed by sclero-therapy has been another approach tested to prevent recur-rence of varices [30, 31]. Shunt surgery is reserved for patientswho fail endoscopic therapy, have significant growth retarda-tion in prepubertal age, symptomatic portal biliopathy andsymptomatic hypersplenism. It can also be offered to patientswho demand a ‘one-time’ treatment [1]. Two of the indexpatients underwent shunt surgery for failed endotherapy and itwas successful.

Conclusions

EHPVO is the commonest cause of portal hypertension andvariceal bleeding in children. The etiological factors are stillmostly unknown. Though mortality related to variceal bleed-ing is low, but morbidity due to massive splenomegaly withhypersplenism, growth failure is significant. Endoscopy is aneffective means to control acute variceal bleed as well as forlong-term eradication of varices. There is growing interest inshunt surgery as an alternative method and also for refractoryand complicated cases.

Conflict of Interest None.

Role of Funding Source None.

References

1. Sarin SK, Sallanoo JD, Chawla YK, Amarapurkar D, Hamid S,Hashizume M, et al; Members of the APASL Working Party onPortal Hypertension. Consensus on extra-hepatic portal vein obstruc-tion. Liver Int. 2006;26:512–9.

2. PrimignaniM. Portal vein thrombosis, revisited. Dig Liver Dis. 2010;42:163–70.

3. Poddar U, Thapa BR, Rao KL, Singh K. Etiological spectrum ofesphageal varices due to portal hypertension in Indian children: Is itdifferent from the West? J Gastroenterol Hepatol. 2008;23:1354–7.

4. Poddar U, Thapa BR, Puri P, Girish CS, Vaiphei K, Vasishta RK,et al. Non-cirrhotic portal fibrosis in children. Indian J Gastroenterol.2000;19:12–3.

5. Alvarez F, Bernard O, Brunelle F, Hadchouel P, Odièvre M, AlagilleD. Portal obstruction in children. I. Clinical investigation and hem-orrhage risk. J Pediatr. 1983;103:696–702.

6. Poddar U, Thapa BR, Singh K. Endoscopic sclerotherapy in children:Experience with 257 cases of extrahepatic portal venous obstruction.Gastrointest Endosc. 2003;57:683–6.

7. Yachha SK, Aggarwal R, Sharma BC, Misra RN, Aggarwal A, NaikSR. Functional protein C and anti–cardiolipin antibody in childrenwith portal vein thrombosis. Indian J Gastroenterol. 2001;20:47–9.

8. Fisher NC, Wilde JT, Roper J, Elias E. Deficiency of natural antico-agulant proteins C, S, and antithrombin in portal vein thrombosis: Asecondary phenomenon? Gut. 2000;46:534–9.

9. Abd El-hamid N, Taylor RM, Marinello D, Mufti GJ, Patel R, Mieli-Vergani G, et al. Aetiology and management of extrahepatic portalvein obstruction in children: King’s College Hospital experience. JPediatr Gastroenterol Nutr. 2008;47:630–4.

10. Gürakan F, Eren M, Koçak N, Yüce A, Ozen H, Temizel IN, et al.Extrahepatic portal vein thrombosis in children: Etiology and long-term follow-up. J Clin Gastroenterol. 2004;38:368–72.

11. Weiss B, Shteyer E, Vivante A, Berkowitz D, Reif S, Weizman Z,et al. Etiology and long-term outcome of extrahepatic portal veinobstruction in children. World J Gastroenterol. 2010;16:4968–72.

12. Qi X, De Stefano V, Wang J, Bai M, Yang Z, Han G, et al. Prevalenceof inherited antithrombin, protein C, and protein S deficiencies inportal vein system thrombosis and Budd-Chiari syndrome: A system-atic review and meta-analysis of observational studies. J GastroenterolHepatol. 2013;28:432–42. doi:10.1111/jgh.12085.

432 Indian J Pediatr (May 2014) 81(5):429–433

13. Arora NK, Lodha R, Gulati S, Gupta AK, Mathur P, Joshi MS, et al.Portal hypertension in north Indian children. Indian J Pediatr. 1998;65:585–91.

14. Mittal SK, Kalra KK, Aggarwal V. Diagnostic upper GI endoscopyfor hemetemesis in children: Experience from a pediatric gastroen-terology centre in north India. Indian J Pediatr. 1994;61:651–4.

15. Rangari M, Gupta R, Jain M, Malhotra V, Sarin SK. Hepatic dys-function in patients with extrahepatic portal venous obstruction. LiverInt. 2003;23:434–9.

16. Webb LJ, Sherlock S. The aetiology, presentation and natural historyof extra-hepatic portal venous obstruction. Q J Med. 1979;48:627–39.

17. Bilodeau M, Aubry MC, Houle R, Burnes PN, Ethier C. Evaluationof hepatocyte injury following partial ligation of the left portal vein. JHepatol. 1999;30:29–37.

18. Chandra R, Kapoor D, Tharakan A, Chaudhary A, Sarin SK. Portalbiliopathy. J Gastroenterol Hepatol. 2001;16:1086–92.

19. Sarin SK, Agarwal SR. Extrahepatic portal vein obstruction. SeminLiver Dis. 2002;22:43–58.

20. Bayraktar Y, Harmanci O. Etiology and consequences of thrombosisin abdominal vessels. World J Gastroenterol. 2006;12:1165–74.

21. Wang JT, Zhao HY, Liu YL. Portal vein thrombosis. HepatobiliaryPancreat Dis Int. 2005;4:515–8.

22. Sarin SK, Bansal A, Sasan S, Nigam A. Portal-vein obstruction inchildren leads to growth retardation. Hepatology. 1992;15:229–33.

23. Mehrotra RN, Bhatia V, Dabadghao P, Yachha SK. Extra hepaticportal vein obstruction in children: Anthropometry, growth hormoneand insulin like growth factor I. J Pediatr Gastroenterol Nutr. 1997;25:520–3.

24. Menon P, Rao KL, Bhattacharya A, Thapa BR, Chowdhary SK,Mahajan JK, et al. Extrahepatic portal hypertension: Quality oflife and somatic growth after surgery. Eur J Pediatr Surg. 2005;15:82–7.

25. Nihal N, Bapat MR, Rathi P, Shah NS, Karvat A, Abraham P, et al.Relation of insulin-like growth factor-1 and insulin-like growth factorbinding protein-3 levels to growth retardation in extrahepatic portalvein obstruction. Hepatol Int. 2009;3:305–9.

26. Lautz TB, Sundaram SS, Whitington PF, Keys L, Superina RA.Growth impairment in children with extrahepatic portal vein obstruc-tion is improved bymesenterico-left portal vein bypass. J Pediatr Surg.2009;44:2067–70.

27. Stringer MD. Improved body mass index after mesenterico-portalbypass. Pediatr Surg Int. 2007;23:539–43.

28. Dilawari JB, Chawla YK, Ramesh GN, Mitra SK, Walia BN. Endo-scopic sclerotherapy in children. J Gastroenterol Hepatol. 1989;4:155–60.

29. Yachha SK, Sharma BC, Kumar M, Khanduri A. Endoscopic sclero-therapy for esophageal varices in children with extrahepatic portalvein obstruction: A follow-up study. J Pediatr Gastroenterol Nutr.1997;24:49–52.

30. Poddar U, Thapa BR, Singh K. Band ligation plus sclerotherapyversus sclerotherapy alone in children with extrahepatic portal veinobstruction. J Clin Gastroenterol. 2005;39:626–9.

31. Poddar U, Bhatnagar S, Yachha SK. Endoscopic band ligationfollowed by sclerotherapy: Is it superior to sclerotherapy in childrenwith extrahepatic portal venous obstruction? J Gastroenterol Hepatol.2011;26:255–9.

Indian J Pediatr (May 2014) 81(5):429–433 433