Hepatic dysfunction after the Fontan surgical palliation runs an indolent course. Moreover, there is no standard method of evaluating hepatic dysfunction. Magnetic resonance elastography has emerged as an advanced screening tool for preclinical detection of hepatic fibrosis and cirrhosis. We describe the case of a patient who had undergone Fontan palliation, and then developed liver nodules and elevated tumor markers 18 years later. Her case illustrates the challenges in diagnostic management of hepatic dysfunction and the potential role of magnetic resonance elastography in monitoring these patients.

Hepatic dysfunction after the Fontan operation runs an indolent course. Extracardiac sequelae arising from long-standing supraphysiologic right-sided heart pressure or flow-related abnormalities are frequently encountered and contribute substantially to morbidity and death in these patients. Patients who have had the Fontan operation are at increased risk of developing liver disease. Eventually these patients can develop cirrhosis and its related sequelae, including hepatocellular carcinoma. The length of time for the evolution of these hepatic changes is unclear, and the severity of liver dysfunction is often underestimated on the basis of serum biochemical testing. The conventional standard for the diagnosis and staging of liver fibrosis is percutaneous biopsy. Liver biopsy is invasive and expensive, has poor patient acceptance, is prone to interobserver variability and sampling errors, and has a complication rate of 3%, with a mortality rate of 0.03%.1 

Liver dysfunction often precedes laboratory or ultrasonographic detection of pathologic conditions. Early cirrhosis can be missed, and rare cases of hepatocellular carcinoma have been reported.2  Moreover, there is no standard means of evaluating patients for hepatic dysfunction after the Fontan procedure. Magnetic resonance elastography (MRE) has emerged as an advanced screening tool for preclinical detection of hepatic fibrosis and cirrhosis. Multiple studies have shown a strong correlation between MRE-measured hepatic stiffness and the stage of fibrosis at histology. The emerging literature indicates that MRE can serve as a safe, less expensive, and more accurate alternative to invasive liver biopsy.3 

The MRE is performed with use of a mechanical acoustic driver and can be performed in conjunction with cardiac magnetic resonance imaging (MRI). Mean liver stiffness is calculated on automatically generated stiffness maps, in units known as kilopascals (kPa). We describe the case of a patient in whom liver nodules suspect for malignancy developed 18 years after the Fontan operation. She also had elevated tumor markers. This report describes the challenges in the diagnosis of hepatic dysfunction and the role of MRE in patients after a Fontan operation.

Case Report

A 21-year-old nursing student who had undergone a fenestrated extracardiac Fontan procedure at 3 years of age presented for her annual cardiac follow-up examination. She was born with pulmonary atresia with intact ventricular septum and a hypoplastic monopartite right ventricle without evidence of right-ventricular-dependent coronary circulation. Her previous surgical procedures had included a right modified Blalock-Taussig shunt (at 5 days of age), a right bidirectional cavopulmonary anastomosis, and a Fontan fenestration purse-string closure.

At initial presentation, a screening abdominal ultrasonogram revealed a 1-cm liver nodule. Her alpha-fetoprotein level was mildly elevated to 9.7 ng/mL (normal, <6 ng/mL). Magnetic resonance imaging of the liver revealed fibrotic and cirrhotic changes, with 3 nodules that were suspect for malignant neoplasms. A transthoracic echocardiogram revealed a patent Fontan connection, borderline left ventricular (LV) enlargement, normal LV systolic function (calculated ejection fraction, 0.60), and trivial mitral regurg ta i tion. She was subsequently referred to our institution for study of the liver nodules.

Laboratory testing revealed normal values for aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase, total bilirubin, direct bilirubin, total protein, albumin, gamma-glutamyl transpeptidase, prothrombin time, hemoglobin, and platelets. The patient's serum was negative for hepatitis B core antibody, positive for hepatitis B surface antibody, and negative for hepatitis C surface antibody.

Cardiac catheterization revealed patent Fontan and cavopulmonary connections (Fig. 1). The mean pressure in the Fontan conduit and branch pulmonary arteries was 13 mmHg. The LV end-diastolic pressure was 11 mmHg. There was mild proximal left pulmonary artery hypoplasia (diameter, 10 mm) and a trivial residual shunt at the fenestration.

Fig. 1.

A) Angiogram shows widely patent lateral-tunnel Fontan and Glenn connection, and a relatively small left pulmonary artery without distinct obstruction. B) Cardiac catheterization-derived mean pressure was 13 mmHg, and pulmonary arteriolar resistance was normal.

L = vessel width

Fig. 1.

A) Angiogram shows widely patent lateral-tunnel Fontan and Glenn connection, and a relatively small left pulmonary artery without distinct obstruction. B) Cardiac catheterization-derived mean pressure was 13 mmHg, and pulmonary arteriolar resistance was normal.

L = vessel width

The MRI revealed 3 hypervascular lesions in the liver, the largest of which was 2.1 × 1.6 cm. The arterial phase showed nodular enhancement, and the venous-hepatocytic phase showed uptake consistent with focal nodular hyperplasia. There was moderately elevated mean liver stiffness of 4.8 kPa, consistent with stage 4 fibrosis (normal level, <2.5 kPa) (Fig. 2). However, in the region of the liver nodules, the stiffness was normal. Normal stiffness is observed in benign nodules, but in malignant nodules stiffness is markedly elevated.3 

Fig. 2.

A) Magnetic resonance elastogram (MRE) displays elevated mean liver stiffness of 4.8 kPa. In the region of the liver nodule (dashed circle), the stiffness was normal (<2.5 kPa). B) Noncontrast magnetic resonance image (axial view) reveals a 2.1 × 1.6-cm hypervascular liver nodule (arrow). C) Arterial contrast phase shows hyperenhancement consistent with focal nodular hyperplasia (arrow). D) Venous-hepatocyte contrast phase shows uptake that is also consistent with focal nodular hyperplasia (arrow). In hepatocellular carcinoma, contrast uptake is usually absent. Our “normal,” as determined by MRE, is a value that has arisen from our clinical experience among more than 3,500 patients with liver dysfunction primarily related to infectious or inflammatory hepatitis. Normal stiffness is observed in benign nodules, but elevated stiffness is common in malignant nodules.

Fig. 2.

A) Magnetic resonance elastogram (MRE) displays elevated mean liver stiffness of 4.8 kPa. In the region of the liver nodule (dashed circle), the stiffness was normal (<2.5 kPa). B) Noncontrast magnetic resonance image (axial view) reveals a 2.1 × 1.6-cm hypervascular liver nodule (arrow). C) Arterial contrast phase shows hyperenhancement consistent with focal nodular hyperplasia (arrow). D) Venous-hepatocyte contrast phase shows uptake that is also consistent with focal nodular hyperplasia (arrow). In hepatocellular carcinoma, contrast uptake is usually absent. Our “normal,” as determined by MRE, is a value that has arisen from our clinical experience among more than 3,500 patients with liver dysfunction primarily related to infectious or inflammatory hepatitis. Normal stiffness is observed in benign nodules, but elevated stiffness is common in malignant nodules.

Subsequently, a percutaneous liver biopsy was performed. Samples were stained with hematoxylin & eosin and Masson's trichrome. Examination under light microscopy at 40× and 100× magnification revealed centrilobular sinusoidal dilation and regions of patchy, dense, centrilobular perisinusoidal fibrosis. These findings were consistent with hepatic fibrosis and venous outflow impairment (Fig. 3). No evidence of malignancy was identified in this liver biopsy specimen.

Fig. 3.

A) Hematoxylin & eosin staining reveals centrilobular sinusoidal dilation (orig. ×100). B) Masson's trichrome staining shows patchy, dense, centrilobular perisinusoidal fibrosis (orig. ×100). Taken together, these findings are consistent with hepatic fibrosis and venous outflow impairment. No evidence of malignancy was identified in this liver biopsy specimen.

Fig. 3.

A) Hematoxylin & eosin staining reveals centrilobular sinusoidal dilation (orig. ×100). B) Masson's trichrome staining shows patchy, dense, centrilobular perisinusoidal fibrosis (orig. ×100). Taken together, these findings are consistent with hepatic fibrosis and venous outflow impairment. No evidence of malignancy was identified in this liver biopsy specimen.

Discussion

As we indicated at the outset, hepatic dysfunction after the Fontan procedure is both slow to develop and hard to detect. Changes in liver parenchyma precede laboratory or ultrasonographic detection. It is important to document evidence of liver dysfunction to guide future surgical planning, monitor complications, and decide the timing for heart transplantation. There is no standard diagnostic means of detecting hepatic dysfunction in patients after a Fontan operation. Liver biopsy has been the primary screening tool.4  Unfortunately, this test has substantial disadvantages, including invasive risk, surgical complications, expense, poor acceptance by patients, interobserver variability among pathologists, and sampling errors.

Magnetic resonance elastography has emerged as a screening tool for preclinical detection of hepatic fibrosis and cirrhosis. It enables targeted biopsy if the findings raise concern (especially if pre-malignant lesions are identified). The results of a recent pilot study exemplified the feasibility of using MRE to screen post-Fontan patients: among 17 patients, the median liver stiffness was 5.1 kPa, which indicated severe fibrosis.5  However, liver biopsy specimens and cardiac catheterization data were not analyzed; therefore, the relationships of histopathology and hemodynamics to liver stiffness were unknown.

Conclusion. As a supplement to these findings of Serai and colleagues,5  our own results show that MRE-derived liver stiffness can corroborate cardiac catheterization-derived pressure measurements and biopsy results. Magnetic resonance elastography can prove itself useful in establishing correlates of detection and staging of liver fibrosis in patients who have undergone the Fontan operation. Moreover, MRE could be helpful in guiding recommendations for liver biopsy. However, the pathologic implication of the MRE has not yet been validated and is limited by the potential presence of passive congestion, which is not easily diagnosed in the absence of cardiac catheterization. A worthwhile study might prospectively evaluate Fontan patients by MRE imaging, liver biopsy for pathologic validation, and cardiac catheterization for the detection of obstruction in the Fontan circuit (which might exhibit abnormally elevated upstream hepatic pressure). Such information could be used to produce an estimated timeframe for the development of liver f ibrosis, and to tailor a liver-surveillance regimen accordingly. Ultimately, more study is needed to determine the prospective role of MRE in the detection of hepatic fibrosis and cirrhosis after the Fontan operation.

References

References
1.
Bulut
OP
,
Romero
R
,
Mahle
WT
,
McConnell
M
,
Braithwaite
K
,
Shehata
BM
,
et al
.
Magnetic resonance imaging identifies unsuspected liver abnormalities in patients after the Fontan procedure
.
J Pediatr
2013
;
163
(
1
):
201
6
.
2.
Asrani
SK
,
Warnes
CA
,
Kamath
PS.
Hepatocellular carcinoma after the Fontan procedure
.
N Engl J Med
2013
;
368
(
18
):
1756
7
.
3.
Venkatesh
SK
,
Yin
M
,
Ehman
RL.
Magnetic resonance elastography of liver: technique, analysis, and clinical applications
.
J Magn Reson Imaging
2013
;
37
(
3
):
544
55
.
4.
Rychik
J
,
Veldtman
G
,
Rand
E
,
Russo
P
,
Rome
JJ
,
Krok
K
,
et al
.
The precarious state of the liver after a Fontan operation: summary of a multidisciplinary symposium
.
Pediatr Cardiol
2012
;
33
(
7
):
1001
12
.
5.
Serai
SD
,
Wallihan
DB
,
Venkatesh
SK
,
Ehman
RL
,
Campbell
KM
,
Sticka
J
,
et al
.
Magnetic resonance elastography of the liver in patients status-post Fontan procedure: feasibility and preliminary results
.
Congenit Heart Dis
2014
;
9
(
1
):
7
14
.

Author notes

From: Division of Pediatric Cardiology (Drs. Cetta and Poterucha, and Ms Novak) and Department of Radiology (Dr. Venkatesh), Mayo Clinic College of Medicine, Rochester, Minnesota 55905

Presented at the 14th Annual International Symposium on Congenital Heart Disease; St. Petersburg, Florida, 15–18 February 2014.

Mayo Clinic holds patents and has a financial interest through royalties related to magnetic resonance elastography.