Congenital portosystemic shunts (CPSSs) are rare vascular malformations that allow intestinal blood to bypass the liver and are classically divided into intra- and extrahepatic shunts (Fig. 1, Table 1) (1,2). An extrahepatic shunt type I, which is also called “Abernethy malformation,” is associated with absence of the portal vein, whereas an extrahepatic shunt type II is characterized by preserved portal venous flow to the liver (1). Intrahepatic shunts can occur in one or in both lobes and consist of one or multiple portosystemic connections (2). A patent ductus venosus is commonly referred to as intrahepatic shunt type 5, despite its course in the ligamentum venosum, because it originates from the left portal vein. Portal hypertension is not usually a feature of CPSS, in contrast to secondary portosystemic shunts found in the setting of liver cirrhosis or portal vein occlusion. Recently, CPSSs have been reported more frequently because of improved imaging capabilities (3,4) and the development of innovative therapeutic solutions (5–7).
On the contrary, the literature on CPSS is widely scattered among numerous clinical case reports and case series, and a focused systematic review of phenotypic characteristics, treatment options, and outcomes has hitherto not been available. We present here a comprehensive review of the published literature of this condition, to which we also add the findings from patients with CPSS who have attended our institution. We aim to provide guidance for the identification and care of affected patients.
A PubMed database search was performed using the search terms “congenital portosystemic shunt,” “congenital absence of portal vein,” “Abernethy malformation,” and “patent ductus venosus” and limited to the English language. From the resulting 292 articles, the title, abstract, and if unclear, the manuscript, were reviewed. Articles were excluded if they provided insufficient data or the type of shunt was not specified, or they discussed patients with clinical problems outside our target group (eg, arteriovenous or surgical shunts, or shunts caused by cirrhosis or portal vein thrombosis). Reference lists of the relevant articles identified were searched to identify further relevant articles. In total, patient data from 195 articles were extracted and analyzed (the complete list of articles can be reviewed at http://links.lww.com/MPG/A199). The minimal information required to include a patient in the analysis was sex, age at diagnosis, and type of shunt. If available, results of investigations (eg, laboratory data), treatments, and follow-up data were retrieved as well. In cases in which different articles discussed the same patient, the extractable data were combined.
Additionally, we identified consecutive patients with CPSS who were referred and seen in the pediatric liver clinic at The Hospital for Sick Children between 2003 and 2010 and retrospectively reviewed their medical charts with research ethics board approval.
Publications identified by our literature review were mostly small case series that illustrated 1 association, complication, or intervention. The largest cohort was a single-center case series, which included 22 patients identified during a 22-year time period (8). We found no randomized controlled trials of therapeutic interventions.
The results of our analysis of the published cases are summarized in Tables 2 and 3. In total we evaluated 316 published cases of CPSS. Of the 185 with an extrahepatic shunt, 103 (56%) had a type I and 82 (44%) a type II shunt. An intrahepatic shunt was found in 131 patients. There was no significant difference in the proportion of male and female patients affected by extrahepatic shunts; however, the majority of intrahepatic shunts were found in male patients. The age at diagnosis ranged from prenatal to 84 years; 66% of the patients were diagnosed before 12 years of age and 24% in adulthood. There was no significant age difference at time of diagnosis between patients with an extrahepatic or intrahepatic shunt. All of the patients underwent at least ≥1 imaging modalities (ultrasound, computerized tomography, or magnetic resonance imaging [MRI]), which confirmed the diagnosis of CPSS. For 157 (49%) patients, comments on liver enzymes were found. Exact values were available for 104 (66% of these patients), and for the others statements included “normal,” “slightly elevated,” “mildly elevated,” and “elevated.” Sixty-nine (43%) patients had normal transaminases; however, it may be more likely that pathological values are reported. These values would mean an underestimation of patients with normal liver enzymes. The mean for all reported ALT levels was 47 IU/L with a maximum of 160 and the mean for AST 60 IU/L, with a maximum of 280. Hyperbilirubinemia, hyperammonemia, and elevated bile acids (mean 89 μmol/L, range 25–285, documented in 47 cases) were common among patients with CPSS, reflecting the result of intestinal blood bypassing the liver.
More than half of the patients presented with symptoms that led to the diagnosis of CPSS, including neurological involvement (hyperammonemia, altered consciousness, and developmental delay in 23%) and pulmonary involvement (dyspnea and hypoxia in 14%). In 40 neonates, hypergalactosemia (as part of the investigation for galactosemia in some countries) or conjugated hyperbilirubinemia led to investigations that identified the diagnosis.
Complications involved mainly the central nervous system, the liver, and the cardiorespiratory system (Table 2). Liver malignancy developed in 12 patients (4 boys, 8 girls) with an extrahepatic shunt (8–18). Hepatoblastoma (4 cases) was diagnosed in early childhood and 3 patients with hepatocellular carcinoma were younger than 12 years at diagnosis. Treatment consisted of resection or partial hepatectomy in 7 patients and liver transplantation in 4, without evidence of tumor recurrence during the reported follow-up between 6 months and 4 years.
Complications of CPSS directly caused 8 of 22 reported deaths (respiratory failure, malignant liver tumor). Brain abscess has been described to complicate CPSS because of shunting of contaminated blood from the intestine through CPSS and intrapulmonary shunts to the brain (19,20).
Interventional or surgical treatment including liver transplantation was considered in 142 (45%), and follow-up data were available for 161 (51%) patients (Table 3). Nearly all of the patients (96%) with available follow-up data showed improvement of symptoms after interventional shunt closure; however, 2 patients died because of complications of liver transplantation and 5 after shunt closure. Spontaneous shunt closure occurred in 11 patients; 10 had an intrahepatic shunt.
Shunt size was evaluated in 71 patients, either by measurement of the shunt diameter on imaging (median 9.6 mm, range 3.5–25.0 mm) or functionally by per-rectal scintigraphy. The latter test calculates the ratio of uptake of a rectally administered tracer in the lungs compared with the liver and thus quantifies the amount of blood traversing the portosystemic shunt (21). The average shunt ratio was 54% (range 15%–95%). Extrahepatic shunts were larger than intrahepatic shunts (median diameter 12.7 ± 5 vs 8.3 ± 4.8 mm, median shunt ratio of 60% ± 21% vs 50% ± 24%, respectively). In 4 patients undergoing both measurements, there was no linear correlation between diameter and shunt fraction, indicating that caution is required when deducing functional characteristics of a CPSS from the anatomical appearance.
A mean portal venous pressure of 19 mmHg (maximum 32 mmHg) was measured after occlusion or banding of the shunt in 29 patients. In 6 patients who underwent nonocclusive shunt reduction procedures, portal pressure increased during a total occlusion test to a mean pressure of 38.9 mmHg. In 14 patients with measurements before and after intervention, average portal pressure increase was 7.3 ± 4 mmHg.
Histopathological examination of 87 patients’ liver biopsies was, of course, not standardized. In many, but not all, cases, absence or paucity of portal vein branches in the portal tracts was reported. Bile ductules appeared normally developed. Cirrhosis was not described.
Details of our own patients are summarized in Table 4, including the clinical presentation, type of shunt, associated abnormalities, complications, and therapy offered. In all but 2 of our 12 patients, we found associated abnormalities, including congenital cardiac disease (n = 4), Down syndrome (n = 2), cutaneous hemangiomas (n = 2), polysplenia (n = 1), and a solitary kidney (n = 1). Complications affecting the central nervous systems (hyperammonemia, MRI changes of the globi pallidi, developmental/behavioral abnormality) were found in 6 patients, followed by pulmonary complications in 2 patients and a benign liver mass in 1 patient. The intrahepatic shunts in 4 patients closed spontaneously before 1 year of age. In 1 patient with hepatopulmonary syndrome (HPS), closure of the shunt with an Amplatzer allowed cessation of supplemental oxygen therapy. One patient's multiple associated conditions preclude shunt closure and 5 patients are well without specific treatment.
This case series and systematic review of the literature reveal that CPSS is a disorder presenting in both children and adults, often with complex associations and complications, and for which interventions to achieve shunt closure are indicated in symptomatic patients. CPSS manifests primarily with neurodevelopmental, hepatic, and pulmonary involvement. Laboratory abnormalities (including elevated ammonia, galactose, conjugated bilirubin, bile acids, transaminases) should raise suspicion of CPSS and prompt liver ultrasound imaging and further investigation to screen for other organ involvement (Fig. 2).
CPSS should be sought as a cause of unexplained mental retardation, behavioral issues, and learning difficulties, especially in cases with additional features associated with CPSS such as cardiac defects, abdominal syndromes, or cutaneous hemangiomas. Blood ammonia may be used as first-line screening for CPSS in these cases; however, its level does not reliably correlate with the degree of encephalopathy and may be normal in some patients. The spectrum of encountered neurodevelopmental abnormalities includes subtle cognitive deficits detected only by neuropsychological tests (22), unexplained mental retardation (23), and postprandial unconsciousness (24). It may be difficult to determine the contribution of CPSS to neurological abnormality in patients with comorbidities such as trisomy 21. Further research is required to determine the diagnostic use of increased signal intensity in the basal ganglia on T1-weighted MRI images because of augmented deposition of manganese, which has escaped hepatic clearance (23,25,26).
Hepatic complications of CPSS include the development of benign liver tumors in 25% of reported cases, most commonly in patients with extrahepatic CPSS. It is postulated that abnormal hepatic blood circulation promotes tumor growth (27). The presence of malignant liver tumors in 4% of reported cases suggests a causal connection in these patients in whom no other risk factors for liver malignancy were identified. Acute or chronic inflammation does not appear as a predisposing or contributing factor because the liver enzymes were normal or only mildly elevated and there were no signs of advanced fibrosis or cirrhosis detected in liver biopsies. α-Fetoprotein was elevated in most cases with hepatocellular carcinoma or hepatoblastoma (9–12), but not in all (13), and therefore its role as a screening test for carcinoma is as controversial here as in other settings (28). Overall, regular screening should be undertaken to identify liver tumors and to monitor for any change in size or appearance with imaging modalities such as ultrasound scan in combination with regular determination of α-fetoprotein (Fig. 2).
The development of HPS (defined by hypoxemia and intrapulmonary vascular dilatations) and pulmonary hypertension (PH) may be explained by exposure of the pulmonary vascular bed to intestinal vasoactive mediators that have bypassed the liver (29–31). Although their pathogenesis is incompletely defined, recent evidence suggests a pivotal role for intravascular accumulation of CD68-positive macrophages (32). Lung biopsies of patients with CPSS and PH showed an obliteration of pulmonary arteries with microthrombotic lesions and intima fibrosis (33). It is not yet possible to determine the extent to which CPSS may exacerbate PH caused by congenital cardiac disease. Dyspnea during exercise is an indication to further investigate for HPS and PH in patients with CPSS, even though physical findings may be absent (Fig. 2).
Spontaneous closure of CPSS is mainly described in patients with intrahepatic shunts and occurs mostly within the first year of life. Therapeutic procedures for intrahepatic CPSS should therefore be postponed in asymptomatic infants younger than 1 year. Because of the high incidence and severity of later complications, shunt closure procedures should be considered for symptomatic patients and those with shunts that persist beyond the first year of life. The location of the shunt (intrahepatic or extrahepatic) should not otherwise influence the decision to treat because neurological and pulmonary complications affect both groups to the same extent (Fig. 2). The authors of the diverse manuscripts have presented objective (although uncontrolled) evidence of postinterventional improvement of encephalopathy (24), normalization of ammonia level (23,33,34), resolution of brain MRI changes (25), improvement in SpO2 (19,31), resolution of pulmonary disease (7,35), and regression of liver masses (9).
The choice of a surgical or an interventional radiology shunt closure technique depends on local expertise, the anatomy of the shunt, and the condition of the patient. Some investigators suggest a 2-stage approach if transient balloon occlusion of the shunt causes significant elevation of the portal pressure >22 or 32 mmHg (8,34). The 2-stage approach involves an initial reduction in shunt diameter using either surgical banding or placement of a reduction stent by an interventional radiologist (35). It is suggested that this approach allows adaptation of the rudimentary intrahepatic portal venous system, even when the main portal vein appears to be absent, before later full shunt closure (7). The published experience with this approach to therapy suggests the need to reconsider previous treatment paradigms that recommended liver transplantation if the portal vein was severely hypoplastic or absent. Unfortunately, adequately powered and controlled studies of different therapeutic approaches are unlikely to be feasible because of the rarity of this abnormality.
The main limitation of these data is the potential for referral and publication bias, leading to more extraordinary or severe cases and positive results from therapeutic interventions being more likely to be reported. Thus, the incidence and severity of the associated morbidity of CPSS and the success of therapy may be overestimated. Additionally, this literature review and our case series rely on retrospective data not collected for research purposes.
In summary, we present a comprehensive summary of the published data and our own clinic experience regarding CPSS. Because CPSS manifests with complications affecting multiple systems, we recommend that CPSS be sought in patients with unexplained PH, hypoxemia, HPS, liver tumors, and mental retardation or behavior changes. Patients diagnosed as having CPSS should be screened regularly for these associated morbidities. Treatment should be considered on a case-by-case basis for all patients after the first year of life or earlier in symptomatic cases. Closure of CPSS by interventional radiology techniques or surgery is emerging as the preferred first-line treatment in appropriate cases.
1. Morgan G, Superina R. Congenital absence of the portal vein: two cases and a proposed classification system for portosystemic vascular anomalies. J Pediatr Surg 1994; 29:1239–1241.
2. Park JH, Cha SH, Han JK, et al. Intrahepatic portosystemic venous shunt. AJR Am J Roentgenol 1990; 155:527–528.
3. Goo HW. Extrahepatic portosystemic shunt in congenital absence of the portal vein depicted by time-resolved contrast-enhanced MR angiography. Pediatr Radiol 2007; 37:706–709.
4. Collard B, Maleux G, Heye S, et al. Value of carbon dioxide wedged venography and transvenous liver biopsy in the definitive diagnosis of Abernethy malformation. Abdom Imaging 2006; 31:315–319.
5. Soejima Y, Taguchi T, Ogita K, et al. Auxiliary partial orthotopic living donor liver transplantation for a child with congenital absence of the portal vein. Liver Transpl 2006; 12:845–849.
6. Gillespie MJ, Golden A, Sivarajan VB, et al. Transcatheter closure of patent ductus venosus with the Amplatzer vascular plug in twin brothers. Pediatr Cardiol 2006; 27:142–145.
7. Kuo MD, Miller FJ, Lavine JE, et al. Exploiting phenotypic plasticity for the treatment of hepatopulmonary shunting in abernethy malformation. J Vasc Interv Radiol 2010; 21:917–922.
8. Franchi-Abella S, Branchereau S, Lambert V, et al. Complications of congenital portosystemic shunts in children: therapeutic options and outcomes. J Pediatr Gastroenterol Nutr 2010; 51:322–330.
9. Witters P, Maleux G, George C, et al. Congenital veno-venous malformations of the liver: widely variable clinical presentations. J Gastroenterol Hepatol 2008; 23:e390–e394.
10. Murray CP, Yoo SJ, Babyn PS. Congenital extrahepatic portosystemic shunts. Pediatr Radiol 2003; 33:614–620.
11. Barton JW III, Keller MS. Liver transplantation for hepatoblastoma in a child with congenital absence of the portal vein. Pediatr Radiol 1989; 20:113–114.
12. Pichon N, Maisonnette F, Pichon-Lefievre F, et al. Hepatocarcinoma with congenital agenesis of the portal vein. Jpn J Clin Oncol 2003; 33:314–316.
13. Joyce AD, Howard ER. Rare congenital anomaly of the portal vein. Br J Surg 1988; 75:1038–1039.
14. Konstas AA, Digumarthy SR, Avery LL, et al. Congenital portosystemic shunts: imaging findings and clinical presentations in 11 patients. Eur J Radiol 2011; 80:175–181.
15. Morotti RA, Killackey M, Shneider BL, et al. carcinoma and congenital absence of the portal vein in a child receiving growth hormone therapy for turner syndrome. Semin Liver Dis 2007; 27:427–431.
16. Marois D, van Heerden JA, Carpenter HA, et al. Congenital absence of the portal vein. Mayo Clin Proc 1979; 54:55–59.
17. Kawano S, Hasegawa S, Urushihara N, et al. Hepatoblastoma with congenital absence of the portal vein: a case report. Eur J Pediatr Surg 2007; 17:292–294.
18. Lundstedt C, Lindell G, Tranberg KG, et al. Congenital absence of the intrahepatic portion of the portal vein in an adult male resected for hepatocellular carcinoma. Eur Radiol 2001; 11:2228–2231.
19. Ohnishi Y, Ueda M, Doi H, et al. Successful liver transplantation for congenital absence of the portal vein complicated by intrapulmonary shunt and brain abscess. J Pediatr Surg 2005; 40:e1–e3.
20. Alvarez AE, Ribeiro AF, Hessel G, et al. Abernethy malformation: one of the etiologies of hepatopulmonary syndrome. Pediatr Pulmonol 2002; 34:391–394.
21. Kira T, Ikeda S, Sera Y, et al. Evaluation of persistence of ductus venosus with tc-99m DTPA galactosyl human serum albumin liver scintigraphy and I-123 iodoamphetamine per-rectal portal scintigraphy. Ann Nucl Med 2000; 14:213–216.
22. Ortiz M, Cordoba J, Alonso J, et al. Oral glutamine challenge and magnetic resonance spectroscopy in three patients with congenital portosystemic shunts. J Hepatol 2004; 40:552–557.
23. Ferrero GB, Porta F, Biamino E, et al. Remittent hyperammonemia in congenital portosystemic shunt. Eur J Pediatr 2010; 169:369–372.
24. Uchino T, Endo F, Ikeda S, et al. Three brothers with progressive hepatic dysfunction and severe hepatic steatosis due to a patent ductus venosus. Gastroenterology 1996; 110:1964–1968.
25. Cho YK, Chang NK, Ma JS. Successful transcatheter closure of a large patent ductus venosus with the Amplatzer vascular plug II. Pediatr Cardiol 2009; 30:540–542.
26. Usuki N, Miyamoto T. A case of congenital absence of the intrahepatic portal vein diagnosed by MR angiography. J Comput Assist Tomogr 1998; 22:728–729.
27. Kondo F. Benign nodular hepatocellular lesions caused by abnormal hepatic circulation: etiological analysis and introduction of a new concept. J Gastroenterol Hepatol 2001; 16:1319–1328.
28. Di Bisceglie AM. Issues in screening and surveillance for hepatocellular carcinoma. Gastroenterology 2004; 127:S104–S107.
29. Hoeper MM, Krowka MJ, Strassburg CP. Portopulmonary hypertension and hepatopulmonary syndrome. Lancet 2004; 363:1461–1468.
30. Rodriguez-Roisin R, Krowka MJ. Hepatopulmonary syndrome--a liver-induced lung vascular disorder. N Engl J Med 2008; 358:2378–2387.
31. Yagi H, Takada Y, Fujimoto Y, et al. Successful surgical ligation under intraoperative portal vein pressure monitoring of a large portosystemic shunt presenting as an intrapulmonary shunt: report of a case. Surg Today 2004; 34:1049–1052.
32. Thenappan T, Goel A, Marsboom G, et al. A central role for CD68(+) macrophages in hepatopulmonary syndrome. Reversal by macrophage depletion. Am J Respirat Crit Care Med 2011; 103:1080–1091.
33. Ohno T, Muneuchi J, Ihara K, et al. Pulmonary hypertension in patients with congenital portosystemic venous shunt: a previously unrecognized association. Pediatrics 2008; 121:e892–e899.
34. Kamimatsuse A, Onitake Y, Kamei N, et al. Surgical intervention for patent ductus venosus. Pediatr Surg Int 2010; 26:1025–1030.
35. Marx M, Huber WD, Crone J, et al. Interventional stent implantation in a child with patent ductus venosus and pulmonary hypertension. Eur J Pediatr 2001; 160:501–504.