bs_bs_banner
Pediatrics International (2014) 56, e102–e105
doi: 10.1111/ped.12456
Patient Report
Hepatic focal nodular hyperplasia with congenital portosystemic shunt Yuki Cho,1 Taro Shimono,2 Hiroyasu Morikawa,3 Haruo Shintaku1 and Daisuke Tokuhara1 Departments of 1Pediatrics, 2Radiology, and 3Hepatology, Osaka City University Graduate School of Medicine, Osaka, Japan Abstract
Hepatic focal nodular hyperplasia (FNH) is a rare benign tumor in children. Vascular anomalies have been identified as pathological features of FNH, but the etiology remains unclear. We describe a rare case including the time course of formation of hepatic FNH in response to congenital portosystemic shunt (PSS). A 4-month-old girl was identified on newborn mass screening to have hypergalactosemia, but no inherited deficiencies in galactose-metabolizing enzymes were found. Ultrasonography and per-rectal portal scintigraphy showed intrahepatic PSS of the right lobe as a cause of the hypergalactosemia. At age 12 months, the patient had elevated hepatic enzymes and small hypoechoic hepatic lesions around the shunt. On abdominal contrast-enhanced ultrasonography spoke-wheel sign and central stellate scar were seen, which are typical features of hepatic FNH without biopsy. Congenital intrahepatic PSS should be evaluated on abdominal contrast-enhanced ultrasonography and observed over time because of its potential to develop into hepatic FNH.
Key words contrast-enhanced ultrasonography, galactosemia, hepatic focal nodular hyperplasia, portosystemic shunt.
Hepatic focal nodular hyperplasia (FNH) is a rare benign and usually asymptomatic tumor in children.1,2 Arterial and venous malformations have been confirmed in this tumor on histology, thus indicating the role of pre-existing arterial or venous malformation, or both, in its formation.3–5 There is as yet, however, no direct information on the development of arterial or venous anomalies in this condition. Here, we report the time course of development of hepatic FNH originating from a congenital venous malformation in the form of portosystemic shunt (PSS).
Case report The patient was an 18-month-old girl born at a weight of 2355 g and a gestational age of 38 weeks after a normal pregnancy. At age 4 months, the baby was referred to hospital because of hypergalactosemia (blood galactose, 6.3 mg/dL, Paigen method) after routine newborn screening, which had been delayed because of child neglect. Physical examination showed no essential abnormalities and no hepatomegaly. On laboratory examination no inherited deficiencies were seen in galactosemetabolizing enzymes: serum aspartate aminotransferase (AST), 37 IU/L (normal, 25–68 IU/L); alanine aminotransferase (ALT), 24 IU/L (normal, 13–56 IU/L); total bile acid (TBA), 16.7 μmol/L (normal, 7.0–14.8 μmol/L); γ-glutamyl-transpeptidase (GGTP), 91 IU/L (normal, 15–150 IU/L); and ammonia, 79 μg/dL (normal, 0–70 μg/dL). Amino acid pattern was normal. The baby was suspected of having congenital PSS, which is a Correspondence: Daisuke Tokuhara, MD PhD, Department of Pediatrics, Osaka City University Graduate School of Medicine, 1-4-3 Asahimachi, Abenoku, Osaka 545-8585, Japan. Email: m1155519 @med.osaka-cu.ac.jp Received 27 November 2013; revised 2 June 2014; accepted 3 July 2014.
© 2014 Japan Pediatric Society
major cause of hypergalactosemia in the absence of inherited galactose-metabolizing enzyme deficiencies. At patient age 6 months, color Doppler ultrasonography showed intrahepatic shunt flow between a right branch of the portal vein and a right hepatic vein. There were no other abnormal signals in the right lobe on B-mode imaging (Fig. 1a). In order to evaluate the shunt severity in addition to the presence of PSS, per-rectal portal scintigraphy using 99mTc-pertechnetate was performed, and the amount of radionuclide shunting the liver and reaching the systemic blood after injection into the rectum was measured as described previously.6 On per-rectal portal scintigraphy shunt index was elevated, at 21.4% (cut-off >10%), thus supporting the presence of intrahepatic PSS. At patient age 12 months, laboratory examination showed mild elevation of liver enzymes (AST, 72 IU/L; ALT, 58 IU/L), total bilirubin, 0.2 mg/dL, TBA, 33.3 μmol/L; GGTP, 36 IU/L; α-fetoprotein, 7.4 ng/mL; and ammonia, 69 μg/dL. The patient had no fever and no hepatomegaly. Abdominal B-mode and color Doppler ultrasonography done at the same time showed several small hypoechoic hepatic lesions but there were no further specific findings (Fig. 1b). On the basis of blood examination and ultrasound, the patient was suspected of having developed an unknown hepatic lesion in addition to PSS. To evaluate the hepatic lesions, contrast-enhanced computed tomography (CT) and magnetic resonance imaging (MRI) were done at 14 and 15 months of age, respectively. A hypervascular 4 cm mass associated with a spoke-wheel-like arterial flow and central stellate scar, which are typical imaging features of hepatic FNH, were seen (Fig. 2). At age 18 months, the patient had no symptoms and had normal development. On laboratory tests AST had decreased to 37 IU/L and ALT to 26 IU/L. On B-mode ultrasonography at age 18 months, it was possible to visualize the
FNH with congenital portosystemic shunt
a
b
c
Fig. 1 Time course of development of hepatic focal nodular hyperplasia on ultrasonography. (a) No parenchymal lesions were found at age 6 months. (b) Several small hypoechoic lesions (white circle) were found at age 12 months. (c) A hypoechoic central scar (white circle) was evident at age 18 months.
central stellate scar, which was more evident than at age 12 months (Fig. 1b,c), but it was difficult to visualize the tumor’s contours. On contrast-enhanced ultrasound (CEUS), however, the hypervascular tumor with its central stellate scar was clearly visualized on arterial-phase imaging (Fig. 3a, Video S1), as was as the spoke-wheel arterial flow (Fig. 3b, Video S2); these findings further characterized the hepatic FNH.
e103
Discussion Despite the accumulation of FNH studies, the etiology of this condition remains uncertain and is therefore of great interest. One possible reason for the obscure nature of the etiology of FNH is that the tumor is asymptomatic and is usually found incidentally,4,7 thus making it difficult to observe the origin or process of FNH formation. In the present patient, we identified the congenital intrahepatic PSS first and could therefore record the process of hepatic FNH formation in response to the presence of the PSS. These findings should help further the understanding of the role of congenital venous anomalies in the formation of hepatic FNH in children. Histologically, hepatic FNH is characterized by a central or peripheral fibrous stellate area with radiating septa containing malformed vascular structures and cholangiolar proliferation.3,5 Recently accumulated studies now allow us to diagnose hepatic FNH, without the need for histology, through the use of imaging modalities such as CEUS, contrast-enhanced CT, and MRI, on the basis of the characteristic images of the spoke-wheel sign and central scar.8–10 In the present patient, conventional B-mode ultrasonography was insufficient to detect FNH, but the spoke-wheel sign and central scar were clearly identifiable on CEUS, which was therefore useful for effective and non-invasive (i.e. without biopsy) evaluation. In the present patient, we considered that CEUS provided some advantages compared to CT and MRI for evaluation because CEUS carries no radiation exposure, does not require the use of a large amount of contrast medium (unlike contrast-enhanced CT), and can provide real-time dynamic images of tumors. Several studies have suggested the pathology of hepatic FNH formation. Wanless et al. found that FNH lesions were supplied by an anomalous artery draining directly into the sinusoids of the nodule.5 Their pathologic findings led to speculation that FNH is a hyperplastic response of the hepatic parenchyma to a preexisting arterial spider-like malformation. In one case of FNH occurring after chemotherapy, histopathology showed hepatic thrombosis and necrosis, suggesting that thrombosis results in hepatic ischemia or necrosis, which subsequently results in nodule formation following hepatic arterial recanalization and tissue reperfusion.11 These pieces of evidence support the important role of blood flow anomalies in the pathomechanics of FNH, but until now there has been no direct evidence that blood flow anomalies develop into FNH. In this regard, we have provided direct evidence of the process of FNH formation around a congenital PSS. We hypothesize that portal–hepatic anomalous venous flow in the present patient stimulated the surrounding parenchyma and caused hyperplastic response. At age 12 months, the patient was found to have a small hypoechoic hepatic lesions and elevated liver enzymes. Hepatic FNH is generally asymptomatic, without liver enzyme elevation. The elevated liver enzymes in the present patient may have reflected hepatocellular injury following a reconstructive or hyperplastic response of the parenchyma; such elevations can therefore act as a biochemical markers of the need for contrastenhanced ultrasound or CT, or both, in patients with congenital © 2014 Japan Pediatric Society
e104
Y Cho et al.
a
b
c
Fig. 2 (a,b) Contrast-enhanced computed tomography (CT) at age 14 months. (a) Arterial phase CT shows early homogeneous enhancement, low-attenuation central scar (arrowhead), and spoke-wheel pattern of arteries (arrows). (b) Portal venous phase CT shows a shunt flow (white arrow) between a posterior segmental branch of the right portal vein (arrowhead) and a right hepatic vein (arrow). (c) T2-weighted magnetic resonance imaging (T2WI) in the axial plane at age 15 months. The lesion remains isointense but the scar has hyperintensity (arrow).
PSS. In support of the association between hepatic FNH and PSS, Lalonde et al. reported a case of intrahepatic PSS associated with FNH in an adolescent.4 Both PSS and FNH in their patient were detected incidentally on elevation of liver enzymes; it was therefore not assumed that the PSS was the origin of the FNH. The present case is, therefore, to our knowledge, the first to explain the etiological role of congenital PSS in FNH formation. Congenital PSS, which is a major cause of hypergalactosemia without inherited enzyme deficiency at newborn mass screening, can cause hepatic encephalopathy, manganese deposition in the brain, and pulmonary hypertension.12 Careful long-term follow up is therefore necessary. The present case provides novel evidence of the potential association between congenital PSS and FNH. Congenital PSS is generally monitored on color Doppler ultrasonography and blood examination (TBA and ammonia), but use of these follow-up methods could underestimate the presence
a
b
Fig. 3 Contrast-enhanced ultrasound at age 18 months. (a) Tumor shows dense parenchymal staining (arrowheads) associated with lack of a blood signal within the hypoechoic nodule (arrows) in the early vascular phase (
Pediatrics International (2014) 56, e102–e105
doi: 10.1111/ped.12456
Patient Report
Hepatic focal nodular hyperplasia with congenital portosystemic shunt Yuki Cho,1 Taro Shimono,2 Hiroyasu Morikawa,3 Haruo Shintaku1 and Daisuke Tokuhara1 Departments of 1Pediatrics, 2Radiology, and 3Hepatology, Osaka City University Graduate School of Medicine, Osaka, Japan Abstract
Hepatic focal nodular hyperplasia (FNH) is a rare benign tumor in children. Vascular anomalies have been identified as pathological features of FNH, but the etiology remains unclear. We describe a rare case including the time course of formation of hepatic FNH in response to congenital portosystemic shunt (PSS). A 4-month-old girl was identified on newborn mass screening to have hypergalactosemia, but no inherited deficiencies in galactose-metabolizing enzymes were found. Ultrasonography and per-rectal portal scintigraphy showed intrahepatic PSS of the right lobe as a cause of the hypergalactosemia. At age 12 months, the patient had elevated hepatic enzymes and small hypoechoic hepatic lesions around the shunt. On abdominal contrast-enhanced ultrasonography spoke-wheel sign and central stellate scar were seen, which are typical features of hepatic FNH without biopsy. Congenital intrahepatic PSS should be evaluated on abdominal contrast-enhanced ultrasonography and observed over time because of its potential to develop into hepatic FNH.
Key words contrast-enhanced ultrasonography, galactosemia, hepatic focal nodular hyperplasia, portosystemic shunt.
Hepatic focal nodular hyperplasia (FNH) is a rare benign and usually asymptomatic tumor in children.1,2 Arterial and venous malformations have been confirmed in this tumor on histology, thus indicating the role of pre-existing arterial or venous malformation, or both, in its formation.3–5 There is as yet, however, no direct information on the development of arterial or venous anomalies in this condition. Here, we report the time course of development of hepatic FNH originating from a congenital venous malformation in the form of portosystemic shunt (PSS).
Case report The patient was an 18-month-old girl born at a weight of 2355 g and a gestational age of 38 weeks after a normal pregnancy. At age 4 months, the baby was referred to hospital because of hypergalactosemia (blood galactose, 6.3 mg/dL, Paigen method) after routine newborn screening, which had been delayed because of child neglect. Physical examination showed no essential abnormalities and no hepatomegaly. On laboratory examination no inherited deficiencies were seen in galactosemetabolizing enzymes: serum aspartate aminotransferase (AST), 37 IU/L (normal, 25–68 IU/L); alanine aminotransferase (ALT), 24 IU/L (normal, 13–56 IU/L); total bile acid (TBA), 16.7 μmol/L (normal, 7.0–14.8 μmol/L); γ-glutamyl-transpeptidase (GGTP), 91 IU/L (normal, 15–150 IU/L); and ammonia, 79 μg/dL (normal, 0–70 μg/dL). Amino acid pattern was normal. The baby was suspected of having congenital PSS, which is a Correspondence: Daisuke Tokuhara, MD PhD, Department of Pediatrics, Osaka City University Graduate School of Medicine, 1-4-3 Asahimachi, Abenoku, Osaka 545-8585, Japan. Email: m1155519 @med.osaka-cu.ac.jp Received 27 November 2013; revised 2 June 2014; accepted 3 July 2014.
© 2014 Japan Pediatric Society
major cause of hypergalactosemia in the absence of inherited galactose-metabolizing enzyme deficiencies. At patient age 6 months, color Doppler ultrasonography showed intrahepatic shunt flow between a right branch of the portal vein and a right hepatic vein. There were no other abnormal signals in the right lobe on B-mode imaging (Fig. 1a). In order to evaluate the shunt severity in addition to the presence of PSS, per-rectal portal scintigraphy using 99mTc-pertechnetate was performed, and the amount of radionuclide shunting the liver and reaching the systemic blood after injection into the rectum was measured as described previously.6 On per-rectal portal scintigraphy shunt index was elevated, at 21.4% (cut-off >10%), thus supporting the presence of intrahepatic PSS. At patient age 12 months, laboratory examination showed mild elevation of liver enzymes (AST, 72 IU/L; ALT, 58 IU/L), total bilirubin, 0.2 mg/dL, TBA, 33.3 μmol/L; GGTP, 36 IU/L; α-fetoprotein, 7.4 ng/mL; and ammonia, 69 μg/dL. The patient had no fever and no hepatomegaly. Abdominal B-mode and color Doppler ultrasonography done at the same time showed several small hypoechoic hepatic lesions but there were no further specific findings (Fig. 1b). On the basis of blood examination and ultrasound, the patient was suspected of having developed an unknown hepatic lesion in addition to PSS. To evaluate the hepatic lesions, contrast-enhanced computed tomography (CT) and magnetic resonance imaging (MRI) were done at 14 and 15 months of age, respectively. A hypervascular 4 cm mass associated with a spoke-wheel-like arterial flow and central stellate scar, which are typical imaging features of hepatic FNH, were seen (Fig. 2). At age 18 months, the patient had no symptoms and had normal development. On laboratory tests AST had decreased to 37 IU/L and ALT to 26 IU/L. On B-mode ultrasonography at age 18 months, it was possible to visualize the
FNH with congenital portosystemic shunt
a
b
c
Fig. 1 Time course of development of hepatic focal nodular hyperplasia on ultrasonography. (a) No parenchymal lesions were found at age 6 months. (b) Several small hypoechoic lesions (white circle) were found at age 12 months. (c) A hypoechoic central scar (white circle) was evident at age 18 months.
central stellate scar, which was more evident than at age 12 months (Fig. 1b,c), but it was difficult to visualize the tumor’s contours. On contrast-enhanced ultrasound (CEUS), however, the hypervascular tumor with its central stellate scar was clearly visualized on arterial-phase imaging (Fig. 3a, Video S1), as was as the spoke-wheel arterial flow (Fig. 3b, Video S2); these findings further characterized the hepatic FNH.
e103
Discussion Despite the accumulation of FNH studies, the etiology of this condition remains uncertain and is therefore of great interest. One possible reason for the obscure nature of the etiology of FNH is that the tumor is asymptomatic and is usually found incidentally,4,7 thus making it difficult to observe the origin or process of FNH formation. In the present patient, we identified the congenital intrahepatic PSS first and could therefore record the process of hepatic FNH formation in response to the presence of the PSS. These findings should help further the understanding of the role of congenital venous anomalies in the formation of hepatic FNH in children. Histologically, hepatic FNH is characterized by a central or peripheral fibrous stellate area with radiating septa containing malformed vascular structures and cholangiolar proliferation.3,5 Recently accumulated studies now allow us to diagnose hepatic FNH, without the need for histology, through the use of imaging modalities such as CEUS, contrast-enhanced CT, and MRI, on the basis of the characteristic images of the spoke-wheel sign and central scar.8–10 In the present patient, conventional B-mode ultrasonography was insufficient to detect FNH, but the spoke-wheel sign and central scar were clearly identifiable on CEUS, which was therefore useful for effective and non-invasive (i.e. without biopsy) evaluation. In the present patient, we considered that CEUS provided some advantages compared to CT and MRI for evaluation because CEUS carries no radiation exposure, does not require the use of a large amount of contrast medium (unlike contrast-enhanced CT), and can provide real-time dynamic images of tumors. Several studies have suggested the pathology of hepatic FNH formation. Wanless et al. found that FNH lesions were supplied by an anomalous artery draining directly into the sinusoids of the nodule.5 Their pathologic findings led to speculation that FNH is a hyperplastic response of the hepatic parenchyma to a preexisting arterial spider-like malformation. In one case of FNH occurring after chemotherapy, histopathology showed hepatic thrombosis and necrosis, suggesting that thrombosis results in hepatic ischemia or necrosis, which subsequently results in nodule formation following hepatic arterial recanalization and tissue reperfusion.11 These pieces of evidence support the important role of blood flow anomalies in the pathomechanics of FNH, but until now there has been no direct evidence that blood flow anomalies develop into FNH. In this regard, we have provided direct evidence of the process of FNH formation around a congenital PSS. We hypothesize that portal–hepatic anomalous venous flow in the present patient stimulated the surrounding parenchyma and caused hyperplastic response. At age 12 months, the patient was found to have a small hypoechoic hepatic lesions and elevated liver enzymes. Hepatic FNH is generally asymptomatic, without liver enzyme elevation. The elevated liver enzymes in the present patient may have reflected hepatocellular injury following a reconstructive or hyperplastic response of the parenchyma; such elevations can therefore act as a biochemical markers of the need for contrastenhanced ultrasound or CT, or both, in patients with congenital © 2014 Japan Pediatric Society
e104
Y Cho et al.
a
b
c
Fig. 2 (a,b) Contrast-enhanced computed tomography (CT) at age 14 months. (a) Arterial phase CT shows early homogeneous enhancement, low-attenuation central scar (arrowhead), and spoke-wheel pattern of arteries (arrows). (b) Portal venous phase CT shows a shunt flow (white arrow) between a posterior segmental branch of the right portal vein (arrowhead) and a right hepatic vein (arrow). (c) T2-weighted magnetic resonance imaging (T2WI) in the axial plane at age 15 months. The lesion remains isointense but the scar has hyperintensity (arrow).
PSS. In support of the association between hepatic FNH and PSS, Lalonde et al. reported a case of intrahepatic PSS associated with FNH in an adolescent.4 Both PSS and FNH in their patient were detected incidentally on elevation of liver enzymes; it was therefore not assumed that the PSS was the origin of the FNH. The present case is, therefore, to our knowledge, the first to explain the etiological role of congenital PSS in FNH formation. Congenital PSS, which is a major cause of hypergalactosemia without inherited enzyme deficiency at newborn mass screening, can cause hepatic encephalopathy, manganese deposition in the brain, and pulmonary hypertension.12 Careful long-term follow up is therefore necessary. The present case provides novel evidence of the potential association between congenital PSS and FNH. Congenital PSS is generally monitored on color Doppler ultrasonography and blood examination (TBA and ammonia), but use of these follow-up methods could underestimate the presence
a
b
Fig. 3 Contrast-enhanced ultrasound at age 18 months. (a) Tumor shows dense parenchymal staining (arrowheads) associated with lack of a blood signal within the hypoechoic nodule (arrows) in the early vascular phase (
