Jaundice in Polycystic Liver Disease: More Than Meets the Eye

Watch a video presentation of this article

Answer questions and earn CME

Abbreviations

ALT

alanine transaminase

AST

aspartate transaminase

CA

cancer antigen

CT

computed tomography

ERCP

endoscopic retrograde cholangiopancreatography

MELD

Model for End‐Stage Liver Disease

MRCP

magnetic resonance

cholangiopancreatography

MRI

magnetic resonance imaging

PKD

polycystic kidney disease

PLD

polycystic liver disease

Key Points

• Polycystic liver disease (PLD) rarely results in jaundice, and this finding is likely to indicate that a process other than PLD may be present.

• Surgical intervention is reserved for patients with PLD with dominant cysts that correlate with symptoms.

• Liver transplantation should be considered in PLD in symptomatic patients who meet criteria for Model for End‐Stage Liver Disease (MELD) exception status.

Case Report

The patient is a 69‐year‐old man with a history of PLD and polycystic kidney disease (PKD) who presented with progressive jaundice. He was diagnosed with autosomal dominant PKD many years prior, on the basis of family history of a sister with PKD and a great grandparent with PLD/PKD. He had been asymptomatic and had normal liver enzymes and normal kidney function until he presented with jaundice and was found to have total bilirubin concentration of 11 mg/dL, alkaline phosphatase of 387 U/L, aspartate transaminase (AST) of 178 U/L, and alanine transaminase (ALT) of 190 U/L. He experienced lower extremity edema a month into his presentation but had no known history of ascites, hepatic encephalopathy, or gastrointestinal bleeding. The only medication he had taken in the preceding 5 months was lisinopril/hydrochlorothiazide, although this was stopped 2 months prior to presentation. He denied use of herbal or alternative supplements. He had no history of significant alcohol use. Serological evaluation was negative for viral hepatitis, autoimmune hepatitis, hemochromatosis, alpha‐1‐antitrypsin deficiency, and primary biliary cholangitis.

Which of the following is the next best step in management?

1. Liver biopsy

2. CT abdomen

3. Endoscopic retrograde cholangiopancreatography (ERCP)

When assessing jaundice, an important first step is abdominal imaging to evaluate the biliary tree (Fig. 1). Abdominal computed tomography (CT) showed numerous hepatic cysts throughout the entire liver measuring up to 11 cm (Fig. 2), with no obvious biliary tract abnormalities. Because of a high suspicion for biliary tract abnormalities and concern that the CT scan was not able to adequately visualize the biliary tree, magnetic resonance imaging with cholangiopancreatography (MRI/MRCP) was performed. MRI/MRCP showed similar findings as the CT scan, and the biliary tree was not easily evaluable due to the high cyst burden. Despite MRI/MRCP findings, high degree of concern for biliary obstruction existed, given no history of alcohol or medication use and negative serological evaluation. The patient underwent ERCP with sphincterotomy. The major finding was an attenuated biliary tree, although it was unclear whether this was related to compression of the biliary tree from cysts. No biliary stent was placed because of difficulty in cannulating the common bile duct.

FIG 1.

Evaluation of cholestasis.

FIG 2.

CT scan showing multiple hepatic cysts.

Despite sphincterotomy, the bilirubin worsened up to a peak of 33 mg/dL over the course of the month. AST was in the range of 140 to 150 U/L, ALT 150 to 180 U/L, and alkaline phosphatase ~300 U/L. He was seen in the transplant clinic for worsened jaundice and ongoing diagnostic uncertainty, and was ultimately admitted to the Liver Transplant Unit.

A high level of suspicion for unrecognized liver disease still existed. In addition, the possibility of lisinopril‐induced drug‐induced liver injury was reconsidered but ultimately dismissed because of cholestatic liver injury not being consistent with the pattern of injury typically seen with this drug. Liver biopsy was considered to evaluate for intrinsic liver disease but was not performed because of a relative lack of liver tissue spared from cysts. Thus, a repeat ERCP was performed.

The repeat ERCP showed signs of extrinsic compression of the biliary tree at the level of the common hepatic duct, with upstream dilation of the left and right intrahepatic ducts (Fig. 3); brushings of the intrahepatic biliary tree showed high‐grade dysplasia. Cancer antigen (CA) 19‐9 returned elevated at 2297 U/mL (upper limit of normal is 38), and carcinoembryonic antigen was minimally elevated at 5.1 µg/mL (<5.1 µg/mL is normal). The constellation of biliary strictures with abnormal cytology in combination with an elevation in CA 19‐9 was highly concerning for cholangiocarcinoma.

FIG 3.

ERCP showing compression of the biliary tree at the level of the common hepatic duct, with upstream dilation of the left and right intrahepatic ducts.

The patient ultimately experienced development of acute kidney injury requiring dialysis, hepatic encephalopathy, and coagulopathy. Liver transplantation was not offered because of concern for advanced intrahepatic cholangiocarcinoma portending a high risk for posttransplant recurrence. The patient was transitioned to hospice care and died at home.

Discussion

PLD is a rare disease most commonly seen in association with PKD but can be isolated without concomitant PKD. The natural history of PLD with regard to mortality is driven largely by kidney dysfunction in the case of PKD, caused by progression to chronic kidney disease or end‐stage renal disease. Not surprisingly, isolated PLD tends to have a somewhat more benign prognosis.

Patients with severe PLD, as defined by height‐adjusted liver volume >1800 mL/m, are more likely to be female, have lower platelet count and albumin, higher alkaline phosphatase, poorer kidney function, and worse quality of life. ¹ Although complications of portal hypertension, such as ascites or varices, can arise from PLD with high cyst burden (Mayo class C or D; Table 1), it is rare to see true signs of hepatic synthetic dysfunction with jaundice or coagulopathy unless another cause of chronic liver disease is present. Fibrosis may result from hepatic venous outflow obstruction, but development of cirrhosis is rare, although the outflow obstruction likely plays a role in development of ascites and varices in PLD. ²

TABLE 1.

Mayo classification of polycystic liver disease

Types	A	B	C	D
Symptoms	None or mild	Moderate to severe	Moderate to severe	Moderate to severe
Cyst findings	Focal	Focal	Diffuse	Diffuse
Spared remnant volume	≥3	≥2	≥1	<1
Portal or hepatic vein occlusion	No	No	No	Yes

Rarely, patients with PLD, such as the one described in this case, may develop complications of biliary tract disease with jaundice. Patients with PKD have been observed to have a higher prevalence of common bile duct dilation than controls, although the mechanism and clinical significance of this is unclear. ³ It has been postulated that biliary obstruction may be related to cholangiocyte cilia dysfunction causing impairments in bile flow, ⁴ although external compression from cysts may play a role. Although case reports have demonstrated improvement in jaundice following percutaneous or surgical decompression of cysts, it may be challenging to determine which cyst may be the culprit in the case of diffuse disease.

A recently published case series from the Mayo clinic described 29 patients with definite or suspected bacterial cholangitis in the setting of PLD with or without associated PKD. ⁵ Predictors of cholangitis included prior ERCP with sphincterotomy, cholelithiasis, or cholecystectomy prior to the first episode of cholangitis, duodenal diverticulum, or prior kidney transplantation. Jaundice was likely rare even in these patients, because elevations of total bilirubin >3 likely occurred in only about 16% of patients (mean bilirubin 1.6, standard deviation 1.4). Thus, it is very important to evaluate for other causes of liver disease when patients with PLD present with jaundice, as was done in this case.

CA 19‐9 can be elevated in the setting of malignancy, such as cholangiocarcinoma, pancreatic cancer, gallbladder cancer, adenocarcinoma of the ampulla of Vater, or ovarian cancer, as well as the benign conditions cholangitis or cirrhosis. CA 19‐9 can be challenging to interpret in the setting of biliary obstruction, although levels >300 U/mL are more specific for cholangiocarcinoma, with even higher levels typically more suggestive of cholangiocarcinoma rather than benign causes of biliary obstruction. Cholangiocarcinoma has been reported rarely in association with PLD ⁶ , ⁷ ; because of this rarity, the pathogenesis of cholangiocarcinoma in PLD is not well understood, although it is likely due to longstanding biliary stasis. In our patient’s case, we have limited data to draw on to determine whether this was his risk factor. The available laboratory data from 1 year prior to his presentation showed completely normal liver tests, making significant cholestasis unlikely. Because Caroli disease and syndrome have known associations with cholangiocarcinoma, this diagnosis was considered, but his imaging studies, including cholangiography, were not consistent with this condition.

The management of PLD is generally driven by the patient’s clinical symptoms, with the predominant chronic symptoms being pain, fullness, or early satiety resulting from massive hepatomegaly with large hepatic cysts, which can have a substantial impact on quality of life. More severe, acute pain can occur when cysts hemorrhage (unclear incidence) or become infected (risk of 1% in all patients with liver cysts). ⁸ Cyst aspiration can be performed easily but is unlikely to provide relief for more than days to weeks because of rapid cyst recurrence. When combined with sclerotherapy, the effects of cyst aspiration may be more durable, although due to variability in disease severity at presentation and variability in follow‐up, the overall response rate and duration are not clear. ⁹ Surgical intervention, such as cyst fenestration (also known as unroofing) or hepatic segmental resection of dominant cysts, is associated with higher degrees of effectiveness (86%‐92%), with resection carrying higher rates of complications than fenestration (51% versus 23%). ⁸ Liver transplantation is reserved for severely symptomatic patients with a large cyst burden that is not amenable to surgical intervention. Current criteria for MELD exception points in the US include Mayo class C or D cyst burden associated with severe symptoms and at least one of the following: (1) hepatic decompensation, (2) concurrent hemodialysis, or (3) estimated glomerular filtration rate less than 20 mL/min. The only available effective pharmacotherapy is currently lanreotide, a long‐acting somatostatin analogue, which has been demonstrated to reduce the rate of liver growth in several small clinical trials lasting 6 months, ¹⁰ as well as a more recent larger study that treated patients for 2.3 years. ¹¹

This case illustrates an exceedingly rare complication of a rare disease. Because jaundice is rare in PLD, its presence should prompt evaluation for biliary obstruction and other causes of liver disease.