Answers are grounded in this paper and cite their sources.
Rarely, an intrahepatic portosystemic venous shunt (PSVS) develops as a collateral circulation secondary to portal hypertension (1,2). When a patient does not show any clinical evidence of cirrhosis or portal hypertension, however, the intrahepatic PSVS may be congenital in origin and represent persistence of the prenatal omphalomesenteric venous system (3,4). PSVSs can be found with symptoms of portosystemic encephalopathy (3,5-7) or malformations including congenital heart disease and minor anomalies (8-10), although some cases have recently been detected incidentally by abdominal sonography (11). The patency of the ductus venosus beyond birth seems to be the most likely explanation for the communication between the umbilical portion of the left portal vein and the inferior vena cava (IVC). To our knowledge, only six patients with a patent ductus venosus, (PDV) acting as a PSVS have been previously described (5-7,9-11). We report a young girl with an almost complete PSVS through a PDV complicated by tumor-like lesions of the liver diagnosed as focal nodular hyperplasia (FNH). CASE REPORT The patient, a girl, was born after the mother's first full-term pregnancy without any complications. No other member of the family suffered from congenital anomalies. The patient was breast-fed from her first day. No abnormal findings, including hypergalactosemia, were detected by neonatal metabolic screening test at 4 days of age. The patient was in good health until 2 weeks before admission. At age 3 years and 6 months, she was referred to our hospital for acute leukemia. On admission, physical examination revealed a solid tumor 2 cm in diameter in the right orbit, moderate scoliosis, and mild splenomegaly. Developmental and neurological examination showed no abnormalities. The histological findings of the bone marrow and the solid tumor in the right orbit were compatible with acute myeloblastic leukemia (AML; FAB M2). Blood chemistry showed several abnormal findings; total protein (5.9 g/dl), albumin (2.5 g/dl), triglyceride (86 mg/dl), and cholesterol (71 mg/dl) were decreased, while total bile acid (76 μM; normal, <10) was considerably elevated. Postprandial ammonia was also elevated, to 184 μg/dl, despite a normal level before meals. Plasma aminogram showed a remarkably low molar ratio of branched-chain (BC) to aromatic (A) amino acids (BC:A, 0.96; normal, 2.36-3.86). The remaining results, including alanine aminotransferase, aspartate aminotransferase, total bilirubin, direct bilirubin, creatinine, urea nitrogen, glucose, and α-fetoprotein, were normal. A markedly elevated indocyanine green retention rate of 52% at 15 min (normal, <10%) and a delayed decreasing pattern of galactose levels in the oral galactose tolerance test indicated the presence of a portosystemic venous shunt. Electroencephalogram and electrocardiogram were normal. Abdominal ultrasonography demonstrated a slightly atrophic right lobe of the liver with a small hypoechogenic round mass, 1 cm in diameter. In addition, the umbilical portion of the left portal vein (1.5 cm in diameter and 2.5 cm in length) was dilated. Color Doppler sonography revealed a circular pattern of blood flow within this dilated vessel, which drained from the left portal vein (Fig. 1A). In the sagittal plane, the blood flow was observed to enter the IVC directly through a small channel (0.5 cm in diameter), which was interpreted as the ductus venosus (Fig. 1B). Venules of the intrahepatic portal venous system other than the left portal vein and its umbilical portion were not visible, and no vessels showing the blood flow pattern of the portal vein could be detected in the liver by pulsed-wave Doppler sonography. The splenic and superior mesenteric veins were seen to drain normally into the main portal vein by sonography. In the venous phase of the superior mesenteric arteriogram (Fig. 2), the dilated umbilical portion of the left portal vein and the IVC were seen, although the right portal vein and its peripheral branches could not be detected. The patency of the ductus venosus was confirmed by placing a catheter into the umbilical portion of the left portal vein from the IVC by an approach from the femoral vein. Almost the entire splanchnic venous return from the splenic and the superior mesenteric veins seemed to bypass the liver through the PDV into the IVC. Angiography via the celiac artery revealed the hypertrophic intrahepatic branches of the hepatic artery, which were assumed to compensate for the lack of blood supply by the portal vein, while the hepatic nodule was not detected. To evaluate the extent of portosystemic shunting, portal scintigraphy with transrectal N-isopropyl-p-[123I] iodoamphetamine IMP) was carried out (12). A lung image was obtained instead of a liver image until 60 min after IMP administration, suggesting that nearly all splanchnic venous blood bypassed the liver. Needle biopsy of the liver revealed slightly expanded hepatocytes with clear cytoplasm, with no evidence of cirrhosis or infiltration by leukemic cells. The portal vessels in the portal tracts were almost normal in appearance. The patient underwent standard induction therapy for remission of AML with etopside (VP-16), methotrexate, cytosine arabinoside (ara-C), and hydrocortisone. Although the treatment was successful in clearing the myeloblasts from the bone marrow and reducing the size of the chloroma, the patient remained severely aplastic for 10 weeks and died of Candida pneumonia. On autopsy, the liver was found to contain several round, well-demarcated nodules, 0.3-1.2 cm in diameter, most of which were near the surface of the liver (Fig. 3). Microscopically, these nodules were composed of mature hepatocytes without lobular architecture or portal triads. Fibrous septa, developed poorly within the nodules, showed areas of proliferating bile ductules and dilated vascular channels, although the central scar was absent. These histological findings were compatible with focal nodular hyperplasia. The ductus venosus, 4 mm in diameter and 6 mm in length, was found to connect the dilated umbilical portion of the portal vein and the IVC. DISCUSSION The ductus venosus acts as a bypass of the liver, enabling about half of the blood to pass from the placenta to the IVC during fetal life. It usually closes within 2 weeks after birth (13) following the interruption of umbilical blood flow and the decrease in pressure in the portal sinus, and it becomes a fibrotic cord in the neonatal period (14). Once obliterated, the ductus venous seldom reopens because of portal hypertension (6). In our case, there was no evidence of portal hypertension and the PDV was considered to be congenital in origin. To our knowledge, PDV has been reported in only six patients, including five congenital cases (5-7,9-11). All congenital cases, except for one neonate, were complicated by an absence or hypoplasia of the intrahepatic portal venous system. In our case as well, the intrahepatic portal venous system, except for the left portal vein and its umbilical portion, could not be visualized by ultrasonography or angiography, although the portal vessels were detected microscopically. One can postulate that the ductus venosus may remain patient as a bypass tract of the portal vein because of elevated vascular resistance due to poor development of the intrahepatic portal venous system (9). Alternatively, it is possible that the presence of the shunt may eliminate the normal pressure gradient between the portal vein and the IVC, thus preventing the development of the intrahepatic portal vein (15). Surgical ligation of the ductus venosus is not considered to be advisable when a patient with congenital PDV also has a defect of the normally functional intrahepatic portal vein, as in our case, since this may result in severe intestinal congestion (6). Focal nodular hyperplasia (FNH), a rare benign lesion of the liver in children (16), is a tumor-like lesion occurring in an almost normal liver, usually solitary but sometimes multiple (17-19). FNH consists of a well-circumscribed lesion of hyperplastic liver parenchyma, often with a central stellate scar. The nodule is thought to be supplied exclusively by arterial blood (18,19). In contrast to our case, patients with FNH occasionally present with either abdominal pain or mass (20). Although several theories on the pathogenesis of FNH have been proposed, linking it with inflammation, hamartoma, or oral contraceptives (21), many researchers believe that the lesions arise as a hyperplastic response of the liver parenchyma to differential blood flow caused by preexisting arterial malformation; wellperfused areas become hyperplastic and ischemic areas become atrophic (18,19). The nonuniformity of the blood supply has also been hypothesized to be related to the development of other kinds of tumor-like lesions including nodular regenerative hyperplasia or partial nodular hyperplasia (22). As a cause of this nonuniformity, thrombosis (15), vascular anomalies (23,24), and blockage of the vessel by deformed red blood cells (25) have been suggested. Apart from our case, the combination of tumor-like hepatic lesions and congenital portosystemic shunts associated with abnormal development of the portal vein has been reported in four patients. FNH was found in one female child and two young women with congenital absence of the portal vein and PSVSs (26-29). Wanless et al. (15) reported a 29-year-old man with hypoplasia of the portal vein and PDV who presented with partial nodular transformation of the liver. In these patients and in our case, the pathogenetic mechanism for the hepatic nodules is not fully understood. Further research may elucidate the relation between the abnormal patterns of hepatic blood supply caused by an absence or hypoplasia of the portal vein and the development of these hepatic nodules. A large systemic return of splanchnic venous flow may result in portosystemic encephalopathy, even when not complicated by severe hepatic dysfunction due to cirrhosis (3,5-7). Our case manifested no symptoms of encephalopathy despite an almost complete portosystemic venous shunt. According to previous studies, encephalopathy has not been a significant clinical feature of children with congenital portosystemic shunts, except for a 3.5-year-old girl who was reported to suffer from somnolence, headache, and vomiting triggered by protein-rich food (7). Generally, elderly patients with portosystemic shunts are likely to present clinical manifestations thought to be associated with intolerance of the aging brain to toxic metabolites (4,7). It is possible that portosystemic shunts in patients who remain asymptomatic, especially children, may be overlooked. Several abnormalities in laboratory tests that could be clues to finding a PSVS were observed in our case. These findings, including an increased level of postprandial ammonia, bile acids, and aromatic amino acids and a delayed decreasing pattern in the galactose tolerance test, appeared to represent an influx of splanchnic blood directly into the systemic circulation without removal of these substances by the liver (5,7,11,30). Recent studies have described two patients with PSVSs who had newborn hypergalactosemia detected on routine screening (7,30). The liver in our case was assumed to be supplied mainly by systemic blood via the compensatively dilated hepatic artery, instead of splanchnic blood, which might influence the various functions of the liver to some degree. For example, the fall in the concentration of proteins and lipids seen in our patient may reflect an impairment of hepatic synthetic functions. Furthermore, it is possible that the intractable suppression of the bone marrow after a standard course of chemotherapy for AML in our case could be related to a dysfunction of the hepatic drug-metabolizing system, although direct proof, such an increased serum concentration of the drugs, was not available. We have reported a 3.5-year-old girl with PDV acting as a portosystemic shunt and associated with FNH of the liver. Ultrasonography in combination with color Doppler imaging, a noninvasive procedure, was efficient in detecting the presence of PDV. We believe that more frequent application of this technique will identify PDV in more cases, thereby contributing to the understanding of the pathophysiology of PDV and related diseases. Acknowledgment: The authors thank Dr. Y. Itai and Dr. T. Ahmed for their helpful suggestions and assistance in preparing this manuscript.FIG. 1. A:: Transverse view of the liver by color Doppler sonography reveals a circular flow pattern in the remarkably dilated umbilical portion (small arrows) of the left portal vein (large arrow). B: Sagittal scan shows that this large vessel directly joins the IVC through a small channel (curved arrow), the ductus venosus.FIG. 2.: Superior mesenteric arteriogram, venous phase. The main portal vein (arrowheads), continued by the superior mesenteric vein (arrows), drains into the dilated umbilical portion (solid arrows) of the left portal vein, which in turn enters the IVC through the ductus venosus. No architecture of the intrahepatic portal venous system, except for the left portal vein and its umbilical portion, is visible.FIG. 3.: Section of the liver at autopsy shows several round, well-demarcated nodules beneath the surface of the liver (arrows).
No takes yet. Share an insight, caveat, or question.
Matsubara et al. (1996) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: