Download - Nodular regenerative hyperplasia mimicking cirrhosisoftheliver · Nodular regenerative hyperplasia ofthe liver usuallypresentswithsigns ofportalhyperten-sion with little evidence

Transcript
Page 1: Nodular regenerative hyperplasia mimicking cirrhosisoftheliver · Nodular regenerative hyperplasia ofthe liver usuallypresentswithsigns ofportalhyperten-sion with little evidence

Gut, 1990,31,725-727

Nodular regenerative hyperplasia mimickingcirrhosis of the liver

J A McDonald, D M Painter, N D Gallagher, G W McCaughan

AbstractNodular regenerative hyperplasia of the liverusually presents with signs of portal hyperten-sion with little evidence of obvious liverdisease. We report a 47 year old man whopresented with clinical signs of decompen-sated cirrhosis, recurrent encephalopathy, andtense ascites but at liver transplant was foundto have nodular regenerative hyperplasiaassociated with a portal vein thrombosis.

Nodular regenerative hyperplasia of the liver isan unusual condition of unknown aetiology. Itis characterised histologically by multipleparenchymal nodules throughout the liver, madeup of hepatocytes that are hyperplastic in someareas and atrophic in others. Central veins areusually absent or atretic and there is little or nofibrosis or inflammation, which distinguishesthis disorder from cirrhosis.

Patients usually present with oesophagealvariceal haemorrhage associated with a few or nosigns of liver disease.2-7 We describe a patient,referred for a liver transplant, who presentedwith recurrent bacterial peritonitis, recurrentepisodes of hepatic encephalopathy, and signs ofend stage liver disease. The preoperative diagno-sis was decompensated cirrhosis but at operationthe resected liver showed nodular regenerativehyperplasia without evidence of cirrhosis.

The A W MorrowGastroenterology andLiver Centre andDepartment ofAnatomical Pathology,Royal Prince AlfredHospital, MissendenRoad, Camperdown,Australia.J A McDonaldD M PainterN D GallagherGW McCaughanCorrespondence to:Dr G W McCaughan, A WMorrow Gastroenterology andLiver Centre, Royal PrinceAlfred Hospital, MissendenRoad, Camperdown, NSW2050, Australia.

Accepted for publication4 September 1989

PatientA 47 year old man in end stage liver failure wasreferred for liver transplant assessment. He haddeveloped alopecia totalis at age 7 years and acutenephrotic syndrome at 17 years but had made a

complete recovery. At 37 years (in 1978) hedeveloped submassive hepatic necrosis, theaetiology of which was unknown. He gave no

history of drug or alcohol abuse. The autoanti-body screen, hepatitis A and B antibodies,cytomegalovirus, Epstein-Barr, and toxoplasmo-sis antibodies were all negative. The copper,iron, and al antitrypsin studies were normal. Onadmission to hospital at this time he had hepaticencephalopathy and bacterial peritonitis. Herequired peritoneal dialysis for one week andsuffered ototoxicity and renal toxicity secondaryto oral neomycin treatment. Four months laterhe was discharged in good health with someresidual splenomegaly but no other stigmata ofliver disease, and he returned to a normalworking life.

In 1986, when he was 45 years old, hedeveloped lethargy and ascites. Hypothyroidismwas diagnosed and treated but with no sympto-matic improvement. He was admitted to hospitalon five occasions (March, June, August,

September, and October) in 1987 with episodesof hepatic encephalopathy confirmed by electro-encephalography. There were no precipitatingaetiologies except for a single episode of spon-taneous bacterial peritonitis. He was transferredfor liver transplantation assessment duringNovember 1987.On admission to hospital he was grossly

oedematous with ascites, tender gynaecomastia,testicular atrophy, and splenomegaly 5 cm belowthe costal margin. He was not jaundiced and didnot have spider naevi. A full blood count showedpancytopenia with a haemoglobin of 7-4 g/dl,platelet count of 55 x 109/1, and a white cell countof 2 6x 109/1. The prothrombin index was 85%and the APTT was 34 seconds. The patient'selectrolyte studies were within the normal range,the serum creatinine concentration was 177iimol/l (range 55-120 [tmol/1) and creatinineclearance was 42 ml/min, with no proteinuria.The bilirubin concentration was 24 ,umol/l (range0-18), albumin 26 g/l (range 40-52), alkalinephosphatase 80 IU/l (< 160), aspartate transami-nase 28 IU/l (range 5-55), and gammaglutamyltransferase 27 IU/l (range 0-55). The prealbuminvalue was within normal range at 170 mg/ml(150-450). The patient was euthyroid. His hepa-titis A and B serology was negative, antinuclearfactor, antimitochondrial antibody, and smoothmuscle antibody were negative, his ca antitryp-sin value was normal, as were iron and copperstudies.An abdominal computed tomogram showed

noticeable ascites, a very small nodular liver, andan appreciably enlarged spleen. There wasevidence of large collateral venous channelswithin the abdomen. The coeliac and superiormesenteric angiograms showed a tortuoussplenic vein, with no filling of the portal vein andretrograde flow down the superior mesentericvein communicating with the inferior venacava. Hepatic wedge venography showed anatretic portal vein but during this investigationthe portal flow was towards the liver. Thepatient's hospital stay was marked by recurrentbouts of hepatic encephalopathy, often associ-ated with episodes of spontaneous bacterialperitonitis. Citrobacter freundii, Enterobactercloacae, and Streptococcus faecalis were grown onseparate occasions from the ascitic fluid.The patient underwent orthotopic liver trans-

plantation on 16 January 1988. At the time oftransplantation he was in grade IV hepatic coma.At surgery the portal vein was found to have anirregular lumen, indicating previous thrombosisand recanalisation. Macroscopically the liver hadan irregular nodular appearance, with nodulesranging from 1 to 0-1 cm in diameter (Fig 1). Theliver weighed 973 g. Under light microscopy theliver looked diffusely nodular. The overall archi-

725

on June 19, 2020 by guest. Protected by copyright.

http://gut.bmj.com

/G

ut: first published as 10.1136/gut.31.6.725 on 1 June 1990. Dow

nloaded from

Page 2: Nodular regenerative hyperplasia mimicking cirrhosisoftheliver · Nodular regenerative hyperplasia ofthe liver usuallypresentswithsigns ofportalhyperten-sion with little evidence

McDonald, Painter, Gallagher, McCaughan

.. --

*:

Figure 1: The macroscopic liver.

aw j 've *s ;

Figure 2: Reversed lobulation. Note 'central' portal tract, circumferentially aligned thin livercell cords, and sinusoidal dilatation.

Figure 3: Portal tract devoid ofportal vein radicles.

tecture was maintained but the lobular architec-ture was disturbed, with reversed labulation insome areas. Hepatocytes were hypertrophic insome areas and atrophic in others. The atrophichepatocytes were often arranged circumferenti-ally around nodules and accompanied bysinusoidal dilatation (Fig 2). No inflammatoryinfiltrate could be seen. Portal vein radicles wereabsent from many portal tracts (Fig 3) andhilar branches of the portal vein showed old(recanalised) thrombosis. There was no fibrosis.Overall, the appearances were characteristic ofnodular regenerative hyperplasia. The patientregained consciousness and had good graftfunction. His recovery was complicated,however, by recurrent episodes of sepsisdue to opportunistic organisms. He died fourmonths after the transplant with herpes zosterencephalitis.

DiscussionObvious signs of liver disease such as ascites andhepatic encephalopathy are an uncommonpresentation in patients with nodular regenera-tive hyperplasia. A small proportion of patients,however, go on to develop hepatic decompensa-tion after long term follow up,2A particularlyafter portosystemic shunt surgery.8 The usualpresentation is with variceal bleeding but withrelatively normal liver synthetic function, thusmimicking extrahepatic portal hypertension.27Splenomegaly with only a slightly raised serumalkaline phosphatase value are the usual find-ings.

This patient had histological evidence ofnodular regenerative hyperplasia but the clinicalfindings were those of decompensated cirrhosis.The severity of the clinical picture may, how-ever, have been contributed to by the presence ofan old portal vein thrombosis. In addition,hepatofugal flow was found during coeliac andsuperior mesenteric angiography. This may alsohave contributed to uncontrollable hepaticencephalopathy. Rector9 reported hepatofugalportal flow in five patients with cirrhosis, all withhepatic encephalopathy and ascites and nohistory of gastrointestinal bleeding. Similar find-ings were also reported in a patient with non-cirrhotic portal hypertension presenting withchronic encephalopathy.'° Despite this, the find-ing of encephalopathy in non-cirrhotic portalhypertension is very uncommon, although itshould be recognised that it may be precipitatedby portosystemic shunt surgery.8The pathology findings in this patient suggest

the diagnosis of nodular regenerative hyper-plasia. The nodules present were small anddiffuse in nature; there was no portal fibrosis.These conclusions are supported by Dr KamalIshak (personal communication). The associatedportal vein thrombosis, however, raises thepossibility of partial nodular transformation ofthe liver. Portal vein thrombosis is seen infre-quently in nodular regenerative hyperplasia butis common in partial nodular hyperplasia.'°"Partial nodular hyperplasia, however, ischaracterised by irregular sized nodules that areparticularly prominent in the perihilar region,

726

on June 19, 2020 by guest. Protected by copyright.

http://gut.bmj.com

/G

ut: first published as 10.1136/gut.31.6.725 on 1 June 1990. Dow

nloaded from

Page 3: Nodular regenerative hyperplasia mimicking cirrhosisoftheliver · Nodular regenerative hyperplasia ofthe liver usuallypresentswithsigns ofportalhyperten-sion with little evidence

Nodular regenerative hyperplasia mimicking cirrhosis ofthe liver 727

associated with atrophy of the remaining liver. "These features were not present in this patient.Wanless et al'2 have suggested that nodularregenerative hyperplasia is secondary to smallportal vein tributary thrombosis, with atrophy ofnearby hepatocytes and regeneration of otherhepatocytes. The difficulty in finding smallportal vein tributaries in the portal tracts in ourpatient would be consistent with this hypothesis.Many authors have reported this liver problem inassociation with longstanding congestive cardiacfailure,' myeloproliferative disorders,4 'I

Raynaud's syndrome,5 CRST syndrome (cal-cinosis cutis, Raynaud's phenomenon, sclero-dactyly, telangiectasia),3 Walderstrom'smacroglobulinaemia,'3 bacterial endocarditis,"miliary tuberculosis,6 and Felty's syndrome."However, it may occur de novo.27The connection between the previous sub-

massive hepatic necrosis and nodular regenera-tive hyperplasia is unclear. Although it is wellestablished that inactive cirrhosis may be the endresult of submassive necrosis of the liver,'6 thispatient raised the possibility that abnormalregeneration of the liver may occasionally lead tonodular regenerative hyperplasia. Alternatively,the finding of an old portal vein thrombosis withloss of small intrahepatic venous channels raisesthe possibility that this disorder may be theresult of portal venous obliteration after peri-toneal dialysis 10 years previously, rather thansubmassive massive necrosis.

In conclusion, this report shows that nodularregenerative hyperplasia can present with endstage liver failure, and is a reminder that aclinical diagnosis without histological investiga-tions can be misleading.

We thank Dr K G Ishmak for histological review and Ms WendyMacpherson and Maria Davidson for typing the manuscript. Alsomany thanks to Dr Ian Mackay for referring the patient.

1 Steiner PE. Nodular regenerative hyperplasia of the liver.AmJ7 Pathol 1959; 35: 943-53.

2 Stromeyer FW, Ishak KG. nodular transformation (nodular,regenerative' hyperplasia) of the liver. Human Pathol 1981;12: 60-71.

3 Lurie B, Novis B, Bank S, Silber W, Botha JBC, Marks IN.CRST syndrome and nodular transformation of the liver: acase report. Gastroenterology 1973; 64: 457-61.

4 Dudley FP, Cebon J, Ireton HJC, Yeomans ND. A 70 year oldman with portal hypertension. Aust NZ 7 Med 1985; 15:461-7.

5 Mones JM, Saldana MJ, Albores-Saavedra J. Nodularregenerative hyperplasia of the liver. Arch Pathol Lab Med1984; 108: 741-3.

6 Rougier P, Degott C, Rueff B, Benhamou JP. Nodularregenerative hyperplasia of the liver. Gastroenterology 1978;75: 169-72.

7 Guarda LA, Hales MR. Nodular regenerative hyperplasia ofthe liver: report of two cases and review of the literature.JClinGastroenterol 1981;3: 157-64.

8 Kingham JGC, Levison DA, Stansfield AG, Dawson AM.Non-cirrhotic idiopathic portal hypertension: a long termfollow up study. QJ Med 1981; 199: 259-68.

9 Rector WG, Hoefs JC, Hossack HF, Everson GT. Hepato-fugal portal flow in cirrhosis: observations of hepatichemodynamics and the nature of the arterioportal communi-cations. Hepatology 1988; 8: 16-20.

10 Takashi M, Igarashi M, Hino S, Goto N, Okuda K. Chronicportal-systemic encephalopathy with normal portal veinpressure possibly due to noncirrhotic portal fibrosis. Dig DisSci 1984; 29: 669-73.

11 Sherlock S, Feldman CA, Moran B, Scheuer PJ. Partialnodular transformation of the liver with portal hypertension.AmJ7 Med 1966; 40: 195-203.

12 Wanless IR, Godwin TA, Allen F, Feder A. Nodular regenera-tive hyperplasia of the liver in hematologic disorders: apossible response to obliterative portal venopathy. Medicine1980; 80: 367-79.

13 Wanless IR, Solt LC, Kortan P, Dech JHN, Gardiner GW,Prokipohuk EJ. Nodular regenerative hyperplasia of theliver associated with macroglobulinemia. Am J Med 1981;70: 1203-9.

14 Knowles DM, Kaye GI, Godman GC. Nodular regenerativehyperplasia of the liver. Gastroenterology 1975; 69: 746-51.

15 Blendis LM, Parkinson MC, Shilkin KM, Williams R.Nodular regenerative hyperplasia of the liver in Felty'ssyndrome. QJ Med 1974; 169: 25-32.

16 Ware AJ, Eigenbrodt EH, Combes B. Prognostic significanceof submassive hepatic necrosis in acute hepatitis. Gastro-enterology 1975; 68: 519-24.

on June 19, 2020 by guest. Protected by copyright.

http://gut.bmj.com

/G

ut: first published as 10.1136/gut.31.6.725 on 1 June 1990. Dow

nloaded from