HISTOLOGICAL PATTERN OF LIVER DISEASES IN JOS

UNIVERSITY TEACHING HOSPITAL, JOS:

A TEN YEAR RETROSPECTIVE STUDY (2000-2009)

BY

VHRITERHIRE AKPOBOME RAYMOND (MBBS)

DISSERTATION SUBMITTED IN PARTIAL FULFILMENT OF THE

REQUIREMENTS FOR THE AWARD OF THE FELLOWSHIP OF THE

NATIONAL POSTGRADUATE MEDICAL COLLEGE OF NIGERIA IN

PATHOLOGY

MAY, 2013

ii

iii

iv

v

CONTENTS

Title page i

Supervisors ii

Certification iii

Declaration iv

Table of Contents v

Dedication vi

Acknowledgement vii

Abstract viii

Introduction 1

Aims and Objectives 4

Literature Review 5

Methodology 17

Results 19

Tables and Figures 27

Discussion 49

References 56

Appendix A: WHO histological classification of tumours of the liver and

intra-hepatic bile ducts 63

Appendix B: Knodell scoring system 66

Appendix C: Ethical clearance 67

vi

DEDICATION

To all those who expend their days seeking cures to diseases and end to all causes

of human suffering; truly, they are angels on earth, whose beautiful minds bear

immortal seeds of boundless hope.

To Benita, my little lovely daughter whose curiosity never fails to amuse me.

vii

ACKNOWLEDGEMENT

I wish to appreciate the dedicated tutorship by Professor B M Mandong and the

other consultants in the department.

Professor Mandong, Professor Abdulkareem and Dr Echejoh are the giants upon

whose shoulders I stood to see; and whose previous research work guided me

through the initially shadowy path of this project.

I also wish to thank the other resident doctors in the department and friends who

provided helpful suggestions and assistance during the course of this work.

The other staff in the laboratory have been very helpful and supportive most

especially Pastor Steve, Mrs Rifkat Bot, Kizito, Eno, Ju and Madame Mary. Mrs Bot

was ever available to skilfully section and stain the tissue specimens.

viii

ABSTRACT

Background: Liver disease is among the common causes of hospital visitation and

admission in most of tropical Africa. Among the liver diseases, viral hepatitis and

other stigmata of chronic liver disease are the commonest reasons for admission. In

most medical wards, patients with liver cirrhosis and hepatocellular carcinoma

constitute a significant proportion of admissions.

Objective of this study: This study was undertaken to document the histological

appearances of diseases in liver biopsies in Jos.

Methods: All documented cases of liver biopsies over a period of ten years were

reviewed.

Results: There were a total of 374 histologically diagnosed liver biopsy specimens

during the ten-year period of review (2000-2009). Liver diseases were found to be

more common in males than in females with male female ratio of 1.9:1. Most cases

occurred within the 31-40 years age bracket. Chronic viral hepatitis was the most

common diagnosis (61%) followed by hepatocellular carcinoma (20.3%) and

cirrhosis (7.8%). Chronic viral hepatitis and cirrhosis occurred most frequently in the

fourth decade while hepatocellular carcinoma peaked a decade later. More than half

of the cases of chronic hepatitis (51.1%) had Knodell scores between the range of

5-8 points while in cirrhosis more than half (51.7%) scored 13-18 points. Analysis of

variance (ANOVA) of these scores in both conditions failed to demonstrate any

statistically significant difference between males and females (P-value = 0.63).

There was poor correlation between age and Knodell scores (Pearson’s r = 0.0067).

ix

The patterns of hepatocellular carcinoma observed included trabecular (71.1%),

acinar/pseudoglandular (13.2%), clear cell (6.6%) and fibrolamellar (1.3%).

Conclusion: This study has demonstrated that there is a high frequency of

occurrence of histologically diagnosed chronic hepatitis and hepatocellular carcinoma

in Jos.

Keywords: liver, hepatitis, Knodell score, hepatocellular carcinoma.

1

CHAPTER ONE

INTRODUCTION

The liver weighing between 1.4 - 1.6 Kg is the second largest organ in the body. It is

roughly triangular and located mainly in the right upper quadrant of the abdomen

inferior to the diaphragm. As seen under the light microscope, cords of polyhedrally

shaped hepatocytes are organised into hexagonal lobules with a central vein and

portal tracts at the corners. The vascular sinusoids separate these cords, lined by

fenestrated flattened endothelial cells. Kupffer cells are attached to the luminal

border of the endothelial cells.1

Histological examination following a liver biopsy is the traditional gold standard for

the evaluation of liver diseases. Liver biopsy was first described by Ehrlich in 1883

but it gained wider acceptance after Menghini again described the procedure in

1958.2

The last 50 years has witnessed a dramatic change in the indications for performing

a liver biopsy because of better understanding of liver diseases, newer disease

entities and the availability of newer and advanced radiological, immunological,

biochemical and genetic markers.3

The common indications for requesting a liver biopsy include abnormal liver function

tests, fever of unknown origin, evaluation of chronic liver disease for diagnosis,

grading, staging and assessment of therapy, evaluation of cholestatic disorders,

tumours, unexplained hepatomegaly, evaluation of a potential living donor for liver

transplantation, drug reactions and non-alcoholic steatohepatitis (NASH).2

2

The approaches for obtaining liver tissue are percutaneous, transjugular,

laparoscopic and intra-operative. The percutaneous liver biopsy technique is the

most commonly used. It may be a blind procedure or guided by ultrasound or

computed tomography scan. Liver tissue may also be obtained during an autopsy.4

The needles for percutaneous liver biopsy are grouped into three categories: suction

needles (Menghini, Klatskin, Jamshidi), cutting needles (Vim-Silverman, Tru-cut),

and spring-loaded cutting needles that have a built-in triggering mechanism.2

Disposable Tru-cut needles are now used in many centres. The adequacy of the

specimen size is of concern to pathologists. A biopsy length of not less than 1.0 cm

to 1.5 cm is generally considered adequate for the evaluation of parenchymal liver

diseases. Some authors recommend larger biopsy sizes up to 2.5 cm in length.2 The

tissue is fixed in 10% formalin, processed and stained using standard methods.

There are concerns about cost, post-biopsy complications and inter-observer

variation in the interpretation of liver biopsy specimens, but it has remained the

procedure of choice for diagnosis and monitoring of patients with liver disorders.5,6

The wide range of morphological changes which occur following injury to

hepatocytes in disease conditions include degeneration, intracellular accumulation,

necrosis, apoptosis, inflammation, regeneration, and fibrosis.1

A systematic approach to the evaluation of liver biopsy is essential in order to ensure

that important diagnostic findings are not overlooked. The slide is first of all scanned

under low magnification to assess the entire architecture. In the portal tracts, the

portal veins, hepatic arteries, bile ducts and lymphatics are examined. The limiting

plate, periportal hepatocytes, sinusoids, pericentral hepatocytes and terminal hepatic

3

venules are next observed for changes. Focal lesions are studied carefully. Findings

are correlated with clinical information.

At least six stains including haematoxylin and eosin (H&E), and histochemical stains

are employed. These include periodic acid Schiff (PAS), Perls’ Prussian blue,

trichrome (or Van Gieson), reticulin and Victoria blue stains. Fibrosis, even in the

early stages is highlighted by reticulin silver impregnation. The trichrome stain

highlights Mallory hyaline and megamitochondria in alcoholic liver disease. Victoria

blue, Orcein and Gomori aldehyde fuchsin stains detect hepatitis B surface antigen

and elastic fibres which are abundant in cirrhosis.4

Other techniques include immunohistochemistry, molecular studies (in situ

hybridization, polymerase chain reaction), electron microscopy and metal assays

(Wilson disease, genetic haemochromatosis). Non-invasive techniques of estimation

of the stage and grade of liver disease include the use of fibroscan, aspartate

aminotransferase to platelet ratio index (APRI).7,8 Serology and imaging studies are

also used in the investigation of liver diseases.4

Although there is an increasing use of these non-invasive methods, proper

diagnostic evaluation still requires histological examination of liver biopsy specimens.

This study examined the pattern of liver diseases on biopsy specimens submitted at

the histopathology laboratory over a period of ten years. This study will contribute to

the existing pool of information on the morphological pattern of histologically

diagnosed liver diseases in our environment.

4

CHAPTER TWO

AIMS AND OBJECTIVES

Aim

The aim of this study is to describe the histological pattern of liver diseases

diagnosed in Jos University Teaching Hospital, Jos, during the period of study.

Objectives

1. To determine the specific histological diagnosis of all the cases;

2. To determine the histological activity index in cases of hepatitis and cirrhosis

using the Knodell scoring system;

3. To determine the sex and age distribution of the various histological

diagnoses;

4. To compare results with observations in similar research done in other

centres.

5

CHAPTER THREE

LITERATURE REVIEW

1. CHRONIC VIRAL HEPATITIS

Chronic hepatitis is a symptomatic, biochemical or serological evidence of continuing

or relapsing hepatic disease for more than six months with histologically documented

inflammation and necrosis.1 Aetiological agents of chronic hepatitis include hepatitis

B, C, and D, drugs, autoimmune and genetic disorders. It is a precursor lesion for

cirrhosis and hepatocellular carcinoma. The clinical features are extremely variable

and are not predictive of outcome.1

Chronic hepatitis formed 31.7% of liver biopsies reported in Jos .9 Previously,

Osuafor et al had found chronic active hepatitis to constitute only 6.5% of a series of

154 liver biopsies histologically analysed at Enugu.10 In a similar study in Kano,

chronic hepatitis was the commonest histological diagnosis (40.5%). Minimal necro-

inflammation without fibrosis was present in 24.4% of these cases. Minimal necro-

inflammation with fibrosis occurred in 3 (8.1%) but severe necro-inflammation with

or without fibrosis and periportal inflammation were seen in only 1 (2.7%) of the

patients.11

Lesi and colleagues in Lagos found chronic hepatitis to be histologically confirmed in

39 (53%) out of a total of 97 patients with clinical and sonographic evidence of

probable chronic liver disease.12

Ndububa et al studied seventy chronic hepatitis patients at Ibadan and observed 57

(75.7%) to be symptomatic while 24.3% were asymptomatic. The asymptomatic

6

cases were found during tests before blood transfusion or routine medical tests.

They were all positive for hepatitis B surface antigen (HBsAg) and negative for anti-

hepatitis C virus (HCV) antibody. Only one of the six symptomatic patients tested

positive for anti-HCV. All of the asymptomatic cases studied had a Knodell

histological grade of ≤ 8 and none with fibrosis. On the other hand, 56.6% of the 53

symptomatic patients with a histological grade of ≥ 9 had stage 3 or 4 fibrosis.

Severe necro-inflammation (HAI score ≥ 14) was found in 11 (20.8%) of the

symptomatic patients.13 This shows that the presence of symptoms may represent a

more severe disease.

The high prevalence of HBV infection in Nigeria was demonstrated by Olubuyide et

al in another study conducted at Ibadan in which 59.3% of hepatocellular carcinoma

patients were HbsAg positive. 14 An investigation into the association between

chronic HBV infection and hepatocellular carcinoma in Maiduguri revealed 86.8% of

114 HCC patients to be positive for HBsAg. 15 This was a significantly higher figure

than was found at Ibadan.

There is a low prevalence of hepatitis C virus infection in most of Africa. The

literature, however, displays an interesting pattern. Nwokedi et al observed HCV

seroprevalence rate of 6.2% in 1007 patients tested at Aminu Kano University

Teaching Hospital, Kano, over a period of three years.16 A slightly higher figure of

9.3% was observed in cirrhotic patients studied in The Gambia.17

However, Egypt has been reported to have the highest country wide prevalence of

HCV infection where eight to ten million of the 68 million population is estimated to

be infected18. Strickland et al studied the correlates of HCV in Egypt in a case control

7

study of 237 patients with liver disease. Anti-HCV antibody positive patients formed

58.2%. This high figure was linked to previous injection treatment for

schistosomiasis in 65.9% of the patients.18

The histological features of mild chronic hepatitis consist of inflammation limited to

the portal tracts and liver architecture remains well preserved. In HCV infection

there are lymphoid aggregates and focal steatosis. As the disease progresses,

interface hepatitis and bridging necrosis increase leading to fibrosis and invariably

cirrhosis.

The definitive diagnosis of chronic hepatitis, monitoring of response to therapy and

the progress of the disease rely heavily on liver biopsy.

2. GRANULOMATOUS HEPATITIS

Granulomas are aggregates of macrophages, often admixed with other inflammatory

cells, which usually result from chronic antigen presentation. Hepatic granulomas are

reportedly present in 2 - 10% of all liver biopsy specimens.19 Tuberculosis,

sarcoidosis, primary biliary cirrhosis, drugs and several infectious agents have been

implicated as causative agents. Primary biliary cirrhosis is the most common cause in

the western world.20

Granulomatous hepatitis was reported as the most common diagnosis among

patients who died of HIV/AIDS in Jos, accounting for 34% of the cases. Fourteen per

cent were non-specific aggregates of epithelioid cells.21 Four cases of granulomas

were also reported in another study of 227 liver biopsies in Jos. Two of these were

8

tuberculoid with caseous necrosis and one was a case of schistosomiasis. The

remaining one was a non-specific aggregate of epithelioid cells.9 Osuafor et al

documented granulomatous hepatitis comprising 1.3% of 154 liver biopsy cases at

Enugu.10

Muthuphei analysed 72 liver biopsies from children in Ga-Rankuwa Hospital, South

Africa and found schistosomiasis in 7 (9.7%) and tuberculosis in 2 (2.8%).22

Schistosomiasis is the leading cause of hepatic granulomas in Saudi Arabia

accounting for 54% of cases in one study.20 Amarapurkar et al studied the livers of

60 HIV positive autopsies in Mumbai, India. On histological examination of the livers,

tubercular granulomas were seen in 19 (31.6%) cases of disseminated tuberculosis.

Granulomas were typical caseating epithelioid cell type in 14 (73.6%) and in 5 cases,

granulomas were not typical. Acid fast bacilli were demonstrated in 4 (6.6%) cases,

all of which showed presence of granulomas.23

3. ALCOHOLIC LIVER DISEASE

The clinical spectrum of alcoholic liver disease consists of fatty change, hepatitis and

cirrhosis. Alcohol is the leading cause of liver cirrhosis in the developed countries

while infection with hepatitis B and or C is mainly responsible for progressive liver

disease in most of Africa.24

Ndububa et al investigated the contribution of alcohol to chronic liver disease at Ile-

Ife and reported that alcoholic liver disease made up 6 (14.4%) of the 145 patients

studied. Three of these alcoholic liver disease patients had cirrhosis while the other

three had hepatitis. Histologically, these patients’ liver biopsies showed the

9

characteristic features of centrilobular polymorphonuclear infiltrates, centrilobular

hepatocyte swelling, ballooning degeneration, macrovesicular steatosis and Mallory

bodies (hyaline bodies).25

4. NON ALCOHOLIC FATTY LIVER DISEASE (NAFLD)

Non-alcoholic fatty liver disease (NAFLD) is an increasingly recognised condition

which may progress into end stage liver disease. The pathological picture resembles

that of alcohol induced injury but it occurs in patients who do not abuse alcohol. It is

a spectrum of liver damage ranging from steatosis to steatohepatitis, advanced

fibrosis and cirrhosis. Non-alcoholic fatty liver disease is frequently found in patients

undergoing liver biopsy for an unexplained abnormal liver function studies. It has

been associated with obesity, type 2 diabetes, hyperlipidaemia, hyperinsulinemia and

insulin resistance. Some children with type 1 diabetes mellitus have been found with

NAFLD.26

Ndububa et al in Ile-Ife, reported 2 (1.4%) histologically diagnosed cases of non-

alcoholic fatty liver disease in their study.25 The liver biopsy features include

steatosis, mixed inflammatory cell infiltrate, hepatocyte ballooning and necrosis,

glycogen nuclei, Mallory’s hyaline, and fibrosis.26

5. LIVER CIRRHOSIS

Cirrhosis is a diffuse bridging fibrosis with on-going necroinflammation and attempt

at regeneration with nodule formation.1 Cirrhosis is often asymptomatic or

associated with only mild clinical symptoms. It is the final common pathway for most

10

liver diseases. Although cirrhotic persons can lead relatively normal lives for several

years, they are at a high risk for liver decompensation and progression to

hepatocellular carcinoma.1

It is the eighth leading cause of death in adults worldwide, responsible for 382, 000

deaths in the year 2002 according to the World Health organisation 2003 world

health report.27 Abdulkareem et al reported 17.7% cirrhosis in 345 liver biopsy

specimens examined in Lagos.28 Similar studies in Jos and Ife showed cirrhosis

accounting for 28% and 41.38% respectively25,29.

Similarly, Adeniji et al studied 251 liver biopsy specimens at Ilorin and found that

cirrhosis constituted 35 (13.9%) of the total. Forty per cent were macronodular

cirrhosis while 11.4% were micronodular.30 This figure is lower than the reports from

Lagos and Jos.

In terms of ranking of diagnostic entities, the findings from Lagos and Jos are similar

because cirrhosis ranked third after hepatocellular carcinoma and chronic hepatitis in

both locations.

Liver cirrhosis which has progressed to hepatocellular carcinoma is depicted by the

presence of the two conditions on biopsies from the same patient. Echejoh et al

found cirrhosis co-existing with hepatocellular carcinoma in 27.4% of 53 cases of

cirrhosis studied. The other 72.6% consisted of cirrhosis alone.29

Other disease conditions may also be seen on biopsy from cirrhotic patients. Falaiye

et al reported a case of a 45 years old farm worker who presented at Lagos

University Teaching Hospital with hepatomegaly. Investigation showed amoebic liver

11

abscess. Furthermore, biopsy revealed cirrhosis with patchy destruction of the liver

parenchyma separated by bands of fibrosis.31

6. BENIGN NEOPLASMS

Benign epithelial tumours of the liver include liver cell adenoma, bile duct adenoma,

bile duct cystadenoma and biliary papillomatosis. The non-epithelial tumours include

haemangioma and angiomyolipoma. Tumour–like lesions which occur in the liver are

cysts, focal nodular hyperplasia, nodular regenerative hyperplasia and peliosis.

Benign tumours of the liver are rare and not commonly found compared with the

malignant ones. Only one (0.6%) each of bile duct adenoma,

haemangioendothelioma and cavernous haemangioma were reported in a study in

Jos.9

6.1. HEPATIC ADENOMA

Hepatic adenoma is histologically composed of benign appearing hepatocytes

arranged in plates of one or two cells in thickness and they are most frequently

found in premenopausal women who have used oral contraceptives.1 Hepatic

adenomas have rarely been seen in men who have been on steroids or diagnosed

with glycogen storage diseases type I or III. Three cases of hepatic adenoma

occurring in males were reported from New Zealand.32 Nine cases of hepatic

adenoma were reported in Korea between 1987 and 2001. Interestingly, six of these

patients were males and three, females.33

12

7. MALIGNANT NEOPLASMS

Liver cancers occur worldwide and contribute immensely to morbidity and mortality.

Malignant tumours of the liver may be primary or metastatic.

Mandong et al reported cancer of the liver as the fourth most common of all

malignant tumours in Jos between 1985 and 1994 but it dropped to become the fifth

in the next study period of 1995 – 2002. The most common cancers were those of

the breast followed by cervix and prostate.34 From Sokoto, Malami and colleagues

reported a low figure of 9 (2.2%) liver cancers in males and none in females out of a

total of 298 cancers diagnosed from 1999 to 2004.35

7.1. HEPATOCELLULAR CARCINOMA (HCC)

Hepatocellular carcinoma which is the most common form of primary liver cell

carcinoma represents the fifth most common cancer in the world and the third most

frequent cause of mortality among cancer patients.36 According to global cancer

statistics, it was responsible for more than 500,000 deaths and more than 600,000

new cases worldwide. This tumour has the highest occurrence in South East Asia

and sub-Saharan Africa (120/100,000) and lowest in the USA (1.8/100, 000).36,37,38

One of the earliest papers on hepatocellular carcinoma in Nigeria emerged from

Ibadan in 1967 and was published in the British Journal of Cancer. It was a report

by Williams and colleagues, of four cases of childhood hepatocellular carcinoma seen

at the University College Hospital (UCH).39

A study of 227 liver biopsy specimens in Jos within a ten-year period (1995-2004)

showed a frequency of 71 (31.3%) hepatocellular carcinoma. There were 25

13

(40.4%) females and 46 (27.9%) males. The highest numbers of patients (29.6%)

were between the age ranges of 41-50 years.9 Similarly, primary liver cell carcinoma

was seen in 27% of liver biopsies examined in Kano.11

A prospective study of 100 histologically confirmed cases of HCC in Maiduguri, north

eastern Nigeria showed the highest prevalence among those aged 40-58 years and a

male to female ratio of 4:1.40 Osuafor et al found hepatocellular carcinoma to

constitute 38 (24.6%) of 154 liver biopsies. This was the most common histologically

diagnosed liver pathology at Enugu.10

Malignant tumours of the liver accounted for 105 (14.7%) of 713 malignant tumours

of gastroenterological organs studied in Lagos by Abdulkareem et al. Hepatocellular

carcinoma constituted 75% of these 105 cases.41 In a study aimed at evaluating the

temporal and biological trends in liver cancers in south western Nigeria, Otegbayo et

al analysed 1234 cases of histologically confirmed liver cancers seen at Ibadan for a

period of over two decades. Hepatocellular carcinoma accounted for 92% of these

cases demonstrating that it is the predominant liver cancer in the south west. The

peak age incidence occurred in the fifth decade and the male /female ratio was

2.5:1 42 similar to the findings in Jos.43

Adeniji et al noted hepatocellular carcinoma to be the most common malignant

tumour of the liver at Ilorin because it constituted 63.7% of all malignant tumours of

the liver in a series of 251 biopsy specimens.44 A similar study done by Seleye-

Fubara and colleagues in Port Harcourt showed HCC making up 3.6% of 2105

malignancies. The highest frequency occurred in the age range 40-49 years and a

male to female ratio of 2:1 similar to previous observations. The trabecular pattern

14

was the most common histological type (49.3%) followed by the

pseudoglandular/acinar pattern (28%).45

Four major aetiological factors associated with HCC have been established. These

are chronic viral infection (HBV, HCV), chronic alcoholism, non-alcoholic

steatohepatitis (NASH) and food contaminants (primarily aflatoxins).1 Benvegnu et al

investigated the pattern of hepatocellular carcinoma development in hepatitis B virus

and hepatitis C virus related cirrhosis in Italy. Four hundred and one patients with

cirrhosis were followed up for 14 – 189 months. Eight (16%) out of 50 HBsAg

positive and 53 (18.7%) out of 284 anti-HCV positive patients developed HCC during

follow-up. On the basis of ultrasound or computed tomography scans, two

morphologic patterns were observed in this study. HCC developed as a nodular type

in 81.8% while an infiltrating mass was found in 18.2% of the patients. Older age,

longer duration and more advanced stage of cirrhosis were all significantly

associated with increased risk of developing nodular but not infiltrating HCC. On the

other hand, male sex, HBsAg positivity and dual HBsAg and anti-HCV positivity were

significant risk factors for development of infiltrating but not nodular HCC.46

The National Cancer Institute Pathology Registry in Egypt shows that HCC make up

11.75% of the malignancies of all digestive organs and 1.68% of total malignancies.

The report of El Zayadi’s study of 2005 shows a rising incidence of HCC in Egypt

which has been explained by the increasing prevalence of the risk factors such as

the emergence of HCV, HBV, improvements in screening programs and diagnostic

tools. The increased survival of cirrhotic patients giving time for progression to HCC

was also noted.47

15

The carcinogenic effect of aflatoxin on hepatocytes following chronic exposure is well

documented by the International Agency for Research on Cancer (IARC).48 Kuniholm

et al reported that moderate and high lifetime groundnut consumption in The

Gambia was associated with 2-fold and 3-fold increase in the risk of cirrhosis and by

extension, an increased risk of acquiring hepatocellular carcinoma.17 In Nigeria,

33% of maize samples from different agro-ecological zones were observed to be

contaminated with aflatoxin49.

Aflatoxin B1 is metabolised by cytochrome p450 enzymes to its reactive form, AFB1-

5-9-oxide. It binds covalently with DNA at the N7 position of guanine causing G:C

>> T:A transversion in codon 249ser of p53 gene. This results in the amino acid

substitution of arginine with serine. This p53 mutation is more frequent in countries

where there is also increased exposure to HBV suggesting a synergism between

AFB1 and HBV in the aetiology of HCC.50

Igetei et al in a pioneering study of p53 codon 249 mutations in HCC patients in

south western Nigeria used cell-free DNA. Evaluation of the 79 plasma samples from

the HCC patients through restriction fragment length polymorphism revealed

mutation in 7.6% of them. This figure is lower than what is reported from Asia and

other sub-Saharan countries. The authors attributed this to the staple food of this

region that is based mainly on tubers. Perhaps a higher figure will be obtained from

the northern part of the country where the diet consists mainly of groundnuts and

grains. This study also revealed a younger age (34.5±11.64 years) of the patients

with p53 mutations.51 This is similar to observations in Asia and The Gambia

suggesting synergism with HBV.52,53

16

7.2. HEPATOBLASTOMA

Hepatoblastoma accounts for 0.2-5.8% of malignancies in childhood but 25 -45% of

all primary liver cell tumours, and forms about 50% of those that are malignant.

Eighty three to 92% occur before the age of five and 66% are seen within the first

two years of life.54 Otegbayo et al42 reported a 0.24% frequency of hepatoblastoma

at Ibadan, while 0.9% was recorded in Jos.9

7.3. CHOLANGIOCARCINOMA

Cholangiocarcinoma is a much less common cancer arising from the common bile

duct cells.

Cholangiocarcinoma formed 1.9% of 154 liver biopsies at Enugu. In that same

study, hepatocellular carcinoma constituted 24.6%.10 At Ibadan, cholangiocarcinoma

constituted 15 (1.2%) compared with 1136 (92%) hepatocellular carcinoma out of

1234 liver cancers.42 Other rare cancers reported in the Ibadan study include

lymphoma (0.16%) and hepatoblastoma (0.24%).42

7.4. METASTATIC TUMOURS

Three (4.8%) females and 6 (3.6%) males out of 227 liver biopsies in Jos were

found to have metastatic adenocarcinomas. One metastatic nephroblastoma was

seen in this study.9 Metastatic carcinoma accounted for 12 (54.5%) males and 10

(45.5%) females of 1,234 liver cancer cases studied at Ibadan.42

17

CHAPTER FOUR

METHODOLOGY

This study is retrospective and hospital based. In this study, the archival records and

materials of liver biopsy specimens accessioned in the Histopathology laboratory of

Jos University Teaching Hospital were analysed. The hospital receives close to 2,500

specimens annually.

Materials and Methods

The materials for this study consisted of the archival records of liver biopsy

specimens submitted at the Department of Histopathology, Jos University Teaching

Hospital, from January 2000 to December 2009.

The age, sex, diagnoses and other relevant information were obtained from the

records. Serological test results for hepatitis B surface antigen were obtained from

the clinical records. In diagnosed cases where additional information was needed,

like in the cases of chronic hepatitis and cirrhosis without staging and grading, the

tissue blocks were retrieved, re-sectioned and stained with haematoxylin and eosin.

Special stains such as Masson’s trichrome, Gordon and Sweet’s silver impregnation

and Perls’ Prussian blue were also employed.

World Health Organisation histological classification of tumours of the liver and

intra-hepatic bile ducts (2000) was used in this study50 (Appendix A). The

histological activity in hepatitis and cirrhosis was assessed with the Knodell scoring

system (Appendix B).

18

Statistical Analysis

Descriptive data obtained in this research were expressed as number, proportion,

and mean ± SD (standard deviation). The results of analysis of the different

diagnostic entities were presented on age and sex distribution tables. One way

analysis of variance (ANOVA) was computed to ascertain if there was any statistically

significant gender differences in the Knodell scores in chronic hepatitis and cirrhosis

based on assumptions of normality of distribution and parametric statistics.

Pearson’s correlation analysis was performed to determine the relationship between

age and Knodell scores obtained in chronic hepatitis. All statistical computations

were performed using Microsoft Excel (2007 version).

Inclusion Criteria

All available records of specimens submitted and reported at the department of

histopathology, Jos University Teaching Hospital, from January 2000 to December

2009 inclusive were used.

Exclusion Criteria

The entries of liver biopsy specimens not showing the complete records were

excluded in this analysis. Liver fine needle aspiration biopsy and cytology records

were also not used because they could not be scored or graded.

Ethical Clearance

The Ethical Committee of the Jos University Teaching Hospital granted ethical

clearance for this study.

19

CHAPTER FIVE

RESULTS

Annual Distribution of Liver Biopsy histological Diagnoses

There were three hundred and ninety one liver biopsy specimens accessioned in Jos

University Teaching Hospital from January 2000 to December 2009. Seventeen of

them were assessed to be inadequate and were not included in this analysis. These

included specimens which were necrotic, not properly fixed or processed, or with

less than five portal regions. Therefore, three hundred and seventy-four specimens

were analysed in this study.

As shown in figure 1, there was an increasing trend from 16 specimens in 2000 to

56 in 2005. A sudden decline occurred in 2006 before rising to a plateau between

2007 and 2008 with 78 and 79 specimens respectively. A sudden decrease was

again noted in 2009.

Liver diseases were generally observed to be more common in males than females.

There were a total of 245 (65.5%) males and 129 (34.5%) females with a male to

female ratio of 1.9:1.

The biopsy specimens were from patients with ages ranging from 4 months to 81

years. Most of the cases occurred within the 31 – 40 years age bracket. Seventy-

seven males and 47 females seen in this age group gave a total of 124 which

accounted for 33.2% of the total number of liver specimens. This was closely

followed by the preceding decade (21-30 years age group) which had 104 (27.8%).

Liver diseases occurred more in males (73) than in females (31) within this age

20

group. Next, the 41-50 years age group had 47 males and 20 females which

altogether formed 17.9%. Eight (2.1%) children between 0 – 10 years old had liver

diseases. However, at the other extreme of age, only a male older than 80 years

was found (table 1). Analysis of the various histologically diagnosed liver diseases

demonstrated that chronic hepatitis, hepatocellular carcinoma and cirrhosis occurred

frequently (table 2).

The most frequently diagnosed liver disease was chronic viral hepatitis with 228

(61%) out of the total of 374 liver disease specimens. Hepatocellular carcinoma

occurred about three times less frequently than chronic viral hepatitis. There were

76 (20.3%) cases of this malignant tumour.

Cirrhosis was significantly noted to be much less frequent with 29 (7.8%) cases seen

during this ten year period.

Other less frequent observations include liver with normal histology (4%) and

steatosis (1.9%). Figure 3 shows a case of hepatic steatosis. There were four cases

(1.1%) each of hepatoblastoma and metastatic adenocarcinoma to the liver.

Cholangiocarcinoma, non-alcoholic fatty liver disease, polycystic liver disease and

schistosomiasis were each diagnosed twice (0.5%). Conditions with only a single

case (0.3%) reported include capillary haemangioma, cavernous haemangioma,

focal nodular hyperplasia, haemangioendothelioma and tuberculosis (table 2).

21

Chronic hepatitis

Chronic hepatitis occurred most frequently in the fourth decade with 87 (38.2%) out

of 228 biopsies from patients within this age group. Chronic hepatitis occurred

predominantly in males with a male: female ratio of 1.9:1. The peak occurrence was

in the 31-40 years age group (table 3).

Hepatitis B virus (HBV) infection was responsible for 219 (96.1%) cases of chronic

hepatitis. All the hepatitis B virus cases were also positive for hepatitis B surface

antigen on serological testing. This information was obtained from the clinical

records. Figure 4 shows a case of chronic hepatitis due to hepatitis B virus infection.

There were five cases of hepatitis C viral infection while co-infection with hepatitis B

virus was present in four (1.8%).

Distribution of Knodell scores in chronic hepatitis.

The histological activity index in chronic hepatitis and cirrhosis was assessed using

the Knodell scoring system. The age and sex distribution of the Knodell scores

obtained in the slides diagnosed with hepatitis B virus infection is presented on

Table 4. The scores were arbitrarily grouped into 1-4, 5-8, 9-12, 13-18 and 19-22

points based on previous studies 13,55. The highest frequency (51.1%) was obtained

in those cases with scores between the ranges of 5-8 points. This was followed by

Knodell scores of 1-4 points, constituting 29.2%. Only two (0.9%) out of the 219

histologically diagnosed cases of chronic hepatitis B virus infection were scored 13-

18 points.

22

Cirrhosis

There were 29 cases of liver cirrhosis and their peak incidence paralleled that of

chronic hepatitis. Fourteen (48.3%) of them occurred within the 31-40 age bracket

and this was followed by the third decade (20.7%) similar to what was observed in

chronic hepatitis (Table 5). Figures 5, 6, and 7 show cases of liver cirrhosis.

Distribution of Knodell scores in cirrhosis

Slightly more than half (51.7%) of the cirrhosis cases had Knodell scores between

13 to 18 points. Thirty-four point five per cent were scored 9-13 points while 13.8%

were scored 5-8 points. In this study, there was no case of cirrhosis with a Knodell

score beyond 18 points (Table 6). Higher scores were obtained in the cases of

cirrhosis with a mean of 13.7±3.2 when compared with non-cirrhotic cases which

had a mean of 6.6±2.6.

Analysis of effect of gender on Knodell Scores in chronic hepatitis and

cirrhosis

The Knodell scores in males with chronic hepatitis B virus infection averaged 6.6 ±

2.5. There was a slightly higher mean of 6.8 ± 2.6 in females. Similarly, males had

lower mean scores (13.0 ± 3.4 S.D.) than females (15.3 ± 2.3) in cirrhosis (Table

7).

23

One-way analysis of variance (ANOVA) was performed to determine if there was a

statistically significant effect of gender on necro-inflammation and fibrosis, based on

the Knodell scores obtained, in chronic hepatitis and cirrhosis.

In chronic hepatitis B infection, the calculated variance ratio, F (0.23) was less than

the critical F value (6.75). In addition, the p-value computed, 0.63 is more than the

significance level for alpha (α), 0.01. Hence, the null hypothesis (Ho: µ male Knodell

scores = µ female Knodell scores) was not rejected. Therefore, there was no

statistically significant difference between the male and female scores in chronic

hepatitis B viral infection (Table 8).

Similarly, in cirrhosis, the calculated variance ratio, F (0.99), was less than the

critical F value (7.68) and the p-value, 0.033, is more than the alpha level of

significance (α) of 0.01. Hence, the null hypothesis (Ho: µ male Knodell scores = µ

female Knodell scores) was not rejected. Therefore, there was no statistically

significant difference between the male and female scores in cirrhosis (Table 8).

Correlation analysis of Knodell scores and age in chronic hepatitis

Pearson’s correlation analysis of age and all the Knodell scores in chronic hepatitis B

viral infection returned the value, r = 0.0067. With a degree of freedom (df) of 217,

there was a two tailed P-value of 0.9218 and one-tailed value of 0.4609. Pearson’s r

value is much less than both the two-tailed and one-tailed P-values. Therefore, there

was a weak positive linear correlation between the age of the patient and the degree

24

of liver damage. The scatter chart below illustrates this weak relationship between

the Knodell scores and the age of the patients (Figure 2).

Benign and malignant tumours

The benign tumours seen in this study included a case each of capillary

haemangioma, cavernous haemangioma, focal nodular hyperplasia; and two cases

of polycystic liver disease. They all make up 1.3% of the total biopsy specimens.

Eighty-eight malignant tumours were present, constituting 22.5% of the 391 total

number of biopsy specimens. Hepatocellular carcinoma was the most common and

made up 86.4% of the malignant tumours. This was followed by metastatic cancer

(5.7%), hepatoblastoma (4.5%), cholangiocarcinoma (2.3%) and

haemangioendothelioma (1.1%) (Table 9).

Hepatocellular carcinoma

There were 76 cases of hepatocellular carcinoma, 50 males and 26 females giving a

male to female ratio of 1.9:1 similar with the observation in chronic hepatitis B virus

infection. These have an average age of 46.2 ±13.9 years. The peak frequency

(30.3%) occurred in the 41-50 year age group with 17 males and 6 females (Table

10). Figures 8, 9 and 10 show cases of hepatocellular carcinoma.

25

Morphological patterns of hepatocellular carcinoma

The distribution of the various histological patterns of hepatocellular carcinoma

showed a preponderance of the trabecular pattern, which constituted 71.1% of the

total 76 cases of hepatocellular carcinoma studied. The proportion of males (4.7%)

with this pattern is more than twice that in females (22.4%). In addition, most of

them are males within the fifth decade. The next common pattern is the

acinar/pseudoglandular occurring in 13.2% of the cases. There was combined

occurrence of trabecular and acinar pattern in 7.9% of the cases. Other variants

include the clear cell (6.6%), and fibrolamellar (1.3%). There was only one case of

fibrolamellar pattern occurring in a 43 years old male. Overall, most of the

hepatocellular carcinoma cases were trabecular patterns in males within the fifth and

sixth decade (table 10).

Co-existence of hepatocellular carcinoma with other disease conditions

Hepatocellular carcinoma may co-exist with other conditions. In this study, 18

(23.7%) were cases of this cancer arising on a background of cirrhosis. However, a

larger proportion (75%) existed in isolation without observable combination with any

other disease condition. The males within this group alone made up half of the total

cases of this cancer (Table 11).

Cholangiocarcinoma

There were two cases of cholangiocarcinoma in 15 and 64 years old males, both of

which made up 0.5% of the total liver biopsy specimens.

26

Hepatoblastoma

The four cases of hepatoblastoma observed occurred in children between the ages

of 7 and 9 years with a mean of 8.3 ± 0.96 years. There was a nine years old

female and the remaining were males.

Metastases

Five cases of metastases were seen, four males and one female all constituting

1.1% of the total biopsies. The age range was 32-81 years. They were all cases of

metastatic adenocarcinoma. Figure 11 shows a case of metastatic adenocarcinoma.

27

Table 1: Age and sex distribution of liver biopsies

Age (Years) Male Female Total

Percentage

(%)

0-10 5 3 8 2.1

11-20 9 8 17 4.5

21-30 73 31 104 27.8

31-40 77 47 124 33.2

41-50 47 20 67 17.9

51-60 19 14 33 8.8

61-70 10 5 15 4.0

71-80 4 1 5 1.3

81-90 1 0 1 0.3

Total

245

(65.5%)

129

(34.5%) 374 100.0

28

Table 2: Sex distribution of liver diseases diagnosed by histology

Diagnosis Male Female Total Percentage (%)

Normal Liver 9 6 15 4.0

Inflammatory Diseases

Chronic Viral Hepatitis 149 79 228 61.0

Tuberculosis 1 0 1 0.3

Schistosomiasis 1 1 2 0.5

Metabolic Diseases

Steatosis 5 2 7 1.9

Other Conditions

Liver Cirrhosis 20 9 29 7.8

Benign Tumours and Tumour-like Conditions

Polycystic Liver Disease 0 2 2 0.5

Focal Nodular Hyperplasia 0 1 1 0.3

Capillary Haemangioma 0 1 1 0.3

Cavernous Haemangioma 1 0 1 0.3

Primary Malignant Tumours

Hepatocellular Carcinoma 50 26 76 20.3

Cholangiocarcinoma 2 0 2 0.5

Hepatoblastoma 3 1 4 1.1

Haemangioendothelioma 1 0 1 0.3

Metastatic Tumours

Adenocarcinoma 3 1 4 1.1

Total (Percentage) 245 (65.5%) 129 (34.5%) 374 100.0

29

Table 3: Age and sex distribution of chronic hepatitis

Age (Years) Male Female Total Percentage (%)

0-10 1 0 1 0.4

11-20 8 6 14 6.1

21-30 51 22 73 32.0

31-40 56 31 87 38.2

41-50 23 10 33 14.5

51-60 6 7 13 5.7

61-70 4 3 7 3.1

Total 149 79 228 100.0

30

Table 4: Age and sex distribution of Knodell scores in chronic hepatitis B virus

(HBV) infection

Age (Years)

Knodell Scores

Total Percentage (%) 1-4 5-8 9-12 13-18

M F M F M F M F

0-10 0 0 0 0 1 0 0 0 1 0.5

11-20 3 1 1 4 3 1 1 0 14 6.4

21-30 16 5 28 12 6 5 0 0 72 32.9

31-40 19 8 24 15 11 6 1 0 84 38.4

41-50 4 4 12 2 4 3 0 0 29 13.2

51-60 1 3 3 4 1 0 0 0 12 5.5

61-70 0 0 3 3 0 0 0 0 6 2.7

71-80 0 0 1 0 0 0 0 0 1 0.5

Total 43 21 72 40 26 15 2 0 219 100.0

M: male, F: female

31

Table 5: Age and sex distribution of liver cirrhosis

Age (Years) Male Female Total Percentage

11-20 0 1 1 3.4

21-30 5 1 6 20.7

31-40 11 3 14 48.3

41-50 3 2 5 17.2

51-60 0 2 2 6.9

61-70 1 0 1 3.4

TOTAL 20 (69%) 9 (31%) 29 100.0

32

Table 6: Age and sex distribution of Knodell scores in cirrhosis

Knodell scores

1-4 5-8 9-12 13-18

Age (Years) M F M F M F M F Total Per cent (%)

0-10 0 0 0 0 0 0 0 0 0 0.0

11-20 0 0 0 0 0 1 0 0 1 3.4

21-30 0 0 2 0 1 0 2 1 6 20.7

31-40 0 0 0 1 6 1 4 2 14 48.3

41-50 0 0 0 1 1 0 1 2 5 17.2

51-60 0 0 0 0 0 0 0 2 2 6.9

61-70 0 0 0 0 0 0 1 0 1 3.4

71-80 0 0 0 0 0 0 0 0 0 0.0

Total 0 0 2 2 8 2 8 7 29 100.0

M: male, F: female

33

Table 7: Knodell scores mean and standard deviation in chronic hepatitis B viral

infection and cirrhosis.

Sex

Hepatitis B Virus Infection Cirrhosis

Mean Score S.D. Mean Score S.D.

Male 6.6 2.5 13.0 3.4

Female 6.8 2.6 15.3 2.3

34

Table 8: Analysis of variance (ANOVA) tables for male and female Knodell scores

from chronic hepatitis and cirrhosis

Chronic hepatitis B infection

ANOVA

Source of

Variation SS Df MS F P-value F crit

Between Groups 1.52 1 1.52 0.23 0.63 6.75

Within Groups 1417.48 217 6.53

Total 1419.00 218

Cirrhosis

ANOVA

Source of

Variation SS Df MS F P-value F crit

Between Groups 108.91 1.00 108.91 0.99 0.33 7.68

Within Groups 2985.09 27.00 110.56

Total 3094 28

SS: sum of squares; df: degree of freedom; MS: mean sum of squares; F: F value;

P-value: calculated probability value; F crit: tabulated critical F value.

35

Table 9: Frequency of distribution of liver cancer types by sex

Diagnosis

Sex

Total Percentage (%) Males Females

Hepatocellular Carcinoma 50 26 76 86.4

Cholangiocarcinoma 2 0 2 2.3

Hepatoblastoma 3 1 4 4.5

Haemangioendothelioma 1 0 1 1.1

Metastases 4 1 5 5.7

Total 60 28 88 100.0

36

Table 10: Age and sex distribution of the patterns of hepatocellular carcinoma

Trabecular Acinar Trabecular and

Acinar

Clear Cell Fibrolamellar Total

Age M F M F M F M F M F Sum %

<10 0 1 0 0 0 0 0 0 0 0 1 1.3

11-20 0 0 0 0 0 0 0 0 0 0 0 0.0

21-30 5 2 2 0 0 0 0 2 0 0 11 14.5

31-40 4 5 2 1 0 1 0 1 0 0 14 18.4

41-50 12 4 2 1 1 1 1 0 1 0 23 30.3

51-60 10 3 1 0 1 2 0 0 0 0 17 22.4

61-70 3 2 0 0 0 0 1 0 0 0 6 7.9

71-80 3 0 1 0 0 0 0 0 0 0 4 5.3

Total 37 17 8 2 2 4 2 3 1 0 76 100

37

Table 11: Distribution of hepatocellular carcinoma (HCC) co-existing with cirrhosis

Diagnosis Male % Female % Total Cases Percentage

HCC alone 39 51.3 19 25 58 76.3

HCC and Cirrhosis 11 14.5 7 9.2 18 23.7

Total 50 65.8 26 34.2 76 100

38

Figure 1: Histogram showing the annual distribution of the number of liver biopsy

specimens in Jos University Teaching Hospital from January 2000 to December 2009.

The percentages indicated represent the proportion of the total number of liver

specimens (374) analysed in this study.

16(4.3%) 11(2.9%)

23(6.1%) 19(5.1%)

32(8.6%)

56(15%)

31(8.3%)

78(20.9%) 79(21.1%)

29(7.8%)

0

10

20

30

40

50

60

70

80

90

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

Num

ber

of

bio

psi

es

Year

39

Figure 2: Scatter chart shows weak linear correlation between Knodell scores and

age of the patient.

0

2

4

6

8

10

12

14

16

0 10 20 30 40 50 60 70 80

Knodel

sco

res

Age (Years)

40

Figure 3: Photomicrograph of fatty liver in a 42 years old male. Many of the

hepatocytes have cytoplasmic microvesicles. These have coalesced in some of the

cells to form large fat vacuoles appearing as round clear spaces pushing the nuclei

peripherally (H & E, X 20).

41

Figure 4: Photomicrograph of hepatitis B virus infection exhibiting portal

inflammation, piecemeal necrosis, ballooning of hepatocytes with inflammatory cells

extending into the parenchyma (H & E, X10).

Inset: Some of the hepatocytes have pale, eosinophilic relatively finely granular

cytoplasm (ground glass appearance) and eosinophilic granules in the nuclei (sanded

nuclei). Lymphocytes are seen on the right infiltrating the liver parenchyma from a

portal region (H & E, X 40).

42

Figure 5: Photomicrograph of liver cirrhosis showing thick fibrous bands

surrounding nodules of regenerated hepatocytes (H & E, X 4).

43

Figure 6: Photomicrograph of liver cirrhosis. Abnormal deposition of reticulin fibres

and irregularity of hepatocyte plates is demonstrated in this reticulin silver

impregnation stain (Gordon and Sweet, X 10).

44

Figure 7: Photomicrograph of liver cirrhosis. This is a trichrome stain showing an

increased deposits of collagen within the fibrous bands separating lobules of

hepatocytes (Masson's trichrome, X 10)

45

Figure 8: Photomicrograph of hepatocellular carcinoma arising on a background of

cirrhosis. A vascularised fibrous band is present in the middle of the field. The

neoplastic cells are in sheets and irregular cords which are several cells thick. The

cells have increased nucleo-cytoplasmic ratio and striking pleomorphism. Most of the

cells have vesicular nuclei with prominent nucleoli (H & E, X40)

46

Figure 9: Photomicrograph of hepatocellular carcinoma, trabecular pattern. The

malignant cells are arranged in thick irregular trabecula (H & E, X 20).

47

Figure 10: Photomicrograph of hepatocellular carcinoma, acinar/pseudoglandular

pattern. These malignant cells are forming gland-like structures (H & E, X 40).

48

Figure 11: Photomicrograph showing metastatic adenocarcinoma in a 65 years

male (H & E, X 40).

49

CHAPTER SIX

DISCUSSION

The liver biopsy specimens used in this study cut across the different age groups, in

both sexes. The ratio of adults to children is 49:1. This is higher than the finding in

previous liver biopsy reports both in Jos, Enugu and Lagos10, 28, 29.

The incidence of liver diseases in males has been noted to be higher than in

females. In this study, the male to female ratio is 1.9:1. This ratio compares

favourably with the observations of Ndububa et al and Abdulkareem et al who both

reported 1.8:1 and it is significantly less than 3:1 reported by Osuafor and

colleagues10, 13, 28.

The most common diagnoses were chronic hepatitis, followed by hepatocellular

carcinoma and cirrhosis. A similar finding was reported in Kano11. Meanwhile, in the

east and south west, hepatocellular carcinoma consistently occurred most frequently

in different studies10,28,30.

The peak incidence of chronic hepatitis occurred in the 31-40 years age group in Jos.

This observation is consistent with reports by other workers9, 28. Most of the patients

are males and had Knodell scores ranging from five to eight points signifying mild

necro-inflammatory activity (Table 5). The older patients did not demonstrate a

greater degree of liver damage and Pearson correlation analysis returned a weak

linear correlation between Knodell scores and age. Perhaps, a further study with a

larger sample size, correlating the scores and duration of illness will give a significant

statistical result and computing with the duration of infection may predictably yield

50

the strongest positive correlation. Poynard et al found acceleration of fibrosis with

increasing age in a series of 4,852 patients with chronic liver diseases56. Older age at

diagnosis appears to be an important determinant of progression to cirrhosis and

hepatocellular carcinoma. This may be caused by the aging of the immune system

which can no longer contain the disease or because of the longer duration of

infection57.

Fibrosis appears to progress more slowly in females than in males with chronic

hepatitis B infection, suggesting that oestrogens have a protective effect on

fibrogenesis56. Rigamonti and colleagues demonstrated the effect of gender on

necro-inflammation. They conducted a univariate and multivariate analysis of the

prognostic variables in chronic hepatitis C and found that gender modulates the

progression of chronic hepatitis C only in younger patients. Women ≤ 50 years

showed lower necro-inflammatory and fibrosis scores than men of comparable age

while men and women >50 years did not exhibit any difference in the severity of the

disease58. In contrast to the observations of these researchers, the analysis of

variance (ANOVA) of the scores obtained from males and females with chronic

hepatitis B, in this study, failed to demonstrate a significant statistical difference

suggesting an insignificant influence of gender on histological activity index. There

has actually been contrasting data on the influence of gender on the risk of

progression of chronic hepatitis57.

The peak incidence of cirrhosis occurred in the fourth decade. This is lower than the

fifth decade reported in Jos and Lagos, and sixth decade observed at Illorin28, 30.

51

The occurrence of hepatocellular carcinoma found in this study (19.4%) is less than

previous observations in Jos (31.3%), Kano (27%), Lagos (33%), and Enugu

(24.6%)10,11,28,43.

The peak occurrence of hepatocellular carcinoma was in the 41-50 years age group,

a decade later than chronic hepatitis and cirrhosis. This concurs with findings in

Lagos, Ibadan, Enugu, Jos and Maiduguri 15, 42, 43, 60, 61.

The predominance of males with hepatocellular carcinoma is confirmed in this study.

This tumour was more common in males (50, 65.8%) than females (26, 34.2%)

giving a male female ratio of 1.9:1. This figure, however, is less than the

observations in Lagos (2.6:1), Maiduguri (2.5:1), Enugu (2.3:1) and Port Harcourt

(2:1)15, 45, 60, 61.

The aetiological factors implicated in hepatocellular carcinoma include viral hepatitis,

chronic alcoholism and aflatoxin exposure, with cirrhosis as the final common

pathway for most of the liver diseases. The significantly low frequency of cirrhosis

(7.8%) compared with chronic viral hepatitis (61%) and hepatocellular carcinoma

(20.3%) portends certain implications. Perhaps, most cases of viral hepatitis in this

region remained asymptomatic until progression to hepatocellular carcinoma. This

observation probably also demonstrates the important role played by other factors

such as aflatoxin in the aetiopathogenesis of hepatocellular carcinoma in our

environment where grains are abundantly consumed and there are no adequate

storage facilities. Synergism between aflatoxin and hepatitis B virus infection in

rapidity of progression to cancer has been previously reported52. In the case control

study conducted by Igetei and colleagues in the south west, all the hepatocellular

52

carcinoma patients who were positive for p53 codon 249 mutation were also positive

for hepatitis B surface antigen (HBsAg) but slightly more than half of the those

negative for this mutation were HBsAg positive suggesting synergism51. Certainly,

more comprehensive information on the contribution of aflatoxin B1 toxicity to liver

disease in this environment will be made available if a study similar to theirs is

performed in other regions of the country.

Cirrhosis was found in this study, to be most common a decade before

hepatocellular carcinoma peak occurrence. Hepatocellular carcinoma occurred in

23.7% of the cases, on a background of cirrhosis. The inference is that the process

leading to the cancer began as an inflammatory condition or otherwise, which

caused diffuse fibrosis with regeneration, a fertile ground for dysplasia and

eventually malignant transformation.

The architectural patterns of hepatocellular carcinoma observed in this study are the

trabecular, pseudoglandular/acinar, acinar and trabecular combined, clear cell and

fibrolamellar. The trabecular pattern, in which the tumour cells grow in cords

separated by sinusoidal blood spaces, was the most common, making up to 71.1%

of the hepatocellular carcinoma cases. Seleye-Fubara and colleagues have reported

the predominance of this histological pattern in Port Harcourt45. They reported

49.3% of 75 hepatocellular carcinoma cases to be of the trabecular pattern, which is

a much lower figure than observed in this study. Similar to the findings in Port

Harcourt, the next common variant is the acinar/pseudoglandular pattern (13.2%).

However, over 50% of the hepatocellular carcinoma cases in Lagos were reported to

be pseudoglandular in pattern60.

53

Other histological patterns of hepatocellular carcinoma not found in this study

include the compact, scirrhous and undifferentiated.

Immunostains distinguish hepatocellular carcinoma from other focal lesions or

malignancies especially cholangiocarcinoma and metastatic adenocarcinoma.

Hepatocellular carcinoma gives a positive result with alpha-fetoprotein, pCEA, Hep-

Par1, glypican 3 and cytokeratin 8-18; and gives negative results with cytokeratin 7,

19 and 20. By contrast, cholangiocarcinoma is positive to cytokeratin 7, 19 and 20,

while metastatic adenocarcinoma is positive only to cytokeratin 20 and negative to

the other immunohistochemical markers61.

The frequency of occurrence of hepatoblastoma (4.5%) is similar to previous reports

in Jos and higher than 2.4% reported at Ibadan by Otegbayo et al and 2% reported

at Lagos by Abdulkareem et al9, 42, 60.

The metastatic tumours observed in this study were all adenocarcinomas and

represented only 1.1% of all the liver biopsies and 5.7% of the malignant

neoplasms. This figure is less than 10% recorded in Lagos and 22% at Ibadan but

higher than 2.6% previously reported in Jos9, 42, 60.

Conclusion

This analysis of 374 histologically diagnosed liver biopsy specimens in Jos

demonstrated that there is high frequency of occurrence of chronic hepatitis and

hepatocellular carcinoma in the north central region of Nigeria. This study showed

an insignificant modulating effect of gender on necro-inflammatory activity and

54

fibrosis progression in chronic hepatitis. However, a weak positive correlation was

observed between disease progression and age. Hepatocellular carcinoma occurred

three times less than chronic hepatitis and the trabecular pattern was found to be

the most common. Despite the increasing popularity of non-invasive modalities of

liver evaluation, liver biopsy will remain, in a long time to come, the most effective

means of evaluation of liver disease especially in a resource-depleted setting.

Recommendations

I wish to make the following recommendations based on the findings of this study

and general observations during the course of this research.

1. There should be an increase in public awareness campaigns to enlighten the

public on the risks of transmission of hepatitis B virus. Health care workers

should adhere strictly to blood handling guidelines.

2. There should be a two pronged approach to the long term prevention of

hepatocellular carcinoma. There should be education of farmers on proper

preservation of food products and implementation of government policy on

provision of storage facilities. This will complement efforts being made

towards vaccination against hepatitis B virus infection.

3. There should be improvement in the synergism between surgeons,

physicians, radiologists and pathologists in the management of patients with

liver diseases. I believe this will increase the diagnostic yield of liver biopsies

and minimise instances of specimen inadequacy and incompletely or

incorrectly filled laboratory request forms.

55

4. Provision of facilities for immunohistochemistry in the histopathology

laboratory of the Jos University Teaching Hospital will greatly enhance

diagnostic accuracy of liver specimens.

5. A future study involving a morphometric correlation of necro-inflammation

and fibrosis with non-invasive markers (liver enzymes, telomere length, etc.,)

will aid in developing a model for accurate prognostication of disease

progression in patients from whom a liver biopsy cannot be obtained.

56

REFERENCES

1. Crawford JM, Liu C. Liver and biliary tract. In: Kumar V, Abbas AK, Fausto N, Aster

JC. Robbin and Cotran Pathologic basis of disease. 8th ed. Philadelphia: Elsevier;

2010.p. 833-890.

2. Sheela H, Seela W, Caldwell C, Boyer JL, Jain D. Liver biopsy: evolving role in the

new millennium. J Clin Gastroenterol 2005; 39:603-610.

3. Siegel CA, Silas AM, Suriawinata AA, Van-Leeuwen DJ. Liver biopsy 2005: when and

how? Cleveland Clin J Med 2005; 72(3):199-224.

4. Geller SA, Pitman MB, Nichols WS. Cellular molecular diagnostic techniques. In: Burt

AD, Portmann BC, Ferrell LD. MacSween’s Pathology of the liver. 5th ed. Philadelphia

(PA), USA: Elsevier; 2007. p. 120-128.

5. Friedman LS. Controversies in liver biopsy: who, where, when, how, why? Current

Gastroenterol Rep 2009; 6:30-36.

6. Piccino T, Sagnelli E, Pasquale G, Giusti G. Complications following percutaneous

liver biopsy. A multicentre retrospective study on 68, 276 biopsies. J Hepatol 1986;

2:165-173.

7. Streinu-Cercel A, Preoţescu L, Streinu-Cercel A, Călin R, Budulac A, Streinu-Cercel O.

Correlations between non-invasive explorations and liver biopsy in the determination

of liver fibrosis in chronic viral hepatitis C. Therap Pharmacol Clin Toxicol

2009;13(3):281-286.

8. Loaeza-del-Castillo A, Paz-Pineda F, Oviedo-Cárdenas E, Sánchez-Ávila F, Vargas-

Vorácková F. AST to platelet ratio index (APRI) for the non-invasive evaluation of

liver fibrosis. Ann Hepatol 2008; 7(4):350-357.

57

9. Echejoh GO, Tanko N M, Manasseh A N, Mandong BM, Ogala-Echejoh SE, Ladep GN

et al. Clinico-pathologic correlation of liver biopsy in Jos, central Nigeria. J Chinese

Clin Med 2007; 2(10):557-562.

10. Osuafor TO, Ikerionwu SE, Ukabam SO. Liver biopsy: experience at Enugu, eastern

Nigeria. Scand J Gastroenterol 1986; 124 Suppl:107S-112S.

11. Samaila AA, Mohammed AZ, Borodo MM, Tijjani BM. Histopathological findings in

liver biopsies and clinical correlation at Kano, Nigeria. Sahel Med J 2008; 11(1):20-

23.

12. Lesi OA, Kehinde MO, Anomneze EE, Wali SS. Hepatitis C infection and risk of chronic

liver disease in Lagos. Nig Q J Hosp Med 2002; 12(1-4):1-5.

13. Ndububa DA, Ojo OS, Adetiloye VA, Durosinmi MA, Olasode BJ, Famurewa OC et al.

Chronic hepatitis in Nigerian patients: a study of 70 biopsy proven cases. West Afr J

Med 2005; 24(2):107-111.

14. Olubuyide IO, Aliyu B, Olaleye OA, Ola SO, Olawuyi F, Malabu UH, et al. Hepatitis B

and C virus and hepatocellular carcinoma. Trans R Soc Trop Med Hyg 1997;91:38-41

15. Ajayi BB, Nggada HA, Moses AE. Hepatocellular carcinoma among patients diagnosed

with and without hepatitis B surface antigenaemia in a Nigerian tertiary hospital. Afr

J Microbiol Res 2007 Dec.1:121-124.

16. Nwokedi EE, Ilyasu Z, Emokpae MA, Dutse AI, Taura AA. Hepatitis C virus infection

among teaching hospital patients in Kano, Nigeria: a retrospective study. Ann Afr

Med 2006; 5(4):185-187.

58

17. Kuniholm MH, Lesi OA, Mendy M, Akano A, Sam O, Hall AJ, et al. Aflatoxin exposure

and viral hepatitis in the etiology of liver cirrhosis in the Gambia, West Africa. Env

Health Perpect 2008; 116(11):1553-57.

18. Strickland GT, Elhefni H, Salman T, Waked I, Abdel-Hamid M, Mikhail NN, et al. Role

of hepatitis C infection in chronic liver disease in Egypt. Am J Trop Med Hyg 2002;

67(4):436-42.

19. Lamps L W. Hepatic granulomas, with an emphasis on infectious causes. Adv Anat

Pathol 2008; 15:309–318.

20. Wainwright H. Hepatic granulomas. Eur J Gastroenterol Hepatol 2007;19:93–95

21. Echejoh GO, Mandong BM, Tanko MN, Manasseh AN, Okeke EN, Agaba EI. Hepatic

histopathological findings in HIV patients at post-mortem in Jos university teaching

hospital, Nigeria. Trop Doct 2006; 36: 228-231.

22. Muthuphei M N. Childhood liver diseases in Ga-Rankuwa hospital, South Africa. East

Afr Med J 2000; 77(9):508-509.

23. Amarapurkar A, Sangle NA. Histological spectrum of liver in HIV - autopsy study. Ann

Hepatol 2005; 4(1):47-51.

24. Mitra AK. Hepatitis C-related hepatocellular carcinoma: Prevalence around the world,

factors interacting, and role of genotypes. Epidemiol Rev 1999; 21(2):180-187.

25. Ndububa DA, Ojo OS, Adetiloye VA, Aladegbaiye AO, Adebayo RA, Adekunle O. The

contribution of alcohol to chronic liver disease in patients from south west Nigeria.

Nig J Clin Pract 2010; 13(4):360-364.

26. Angulo P. Non-alcoholic fatty liver disease. New Engl J Med 2002; 346(16):1221-

1231.

59

27. World Health Organisation. The world health report 2003. Geneva. World Health

Organisation; 2003.p.17. Available from: www.who.int\whr\2003\en\index.html

[accessed on 2rd March, 2011].

28. Abdulkareem FB, Banjo AA, Elesha SO, Daramola AO. Histopathological study of liver

diseases at Lagos University Teaching Hospital, Nigeria (1989-2000). Nig Postgrad

Med J 2006; 13(1):41-6.

29. Echejoh GO, Tanko MN, Manasseh AN, Silas AO, Ogala-Echejoh S, Mandong BM.

Liver cirrhosis in Jos, north central Nigeria. Jos J Med 2008; 3(1):26-29.

30. Adeniji KA, Anjorin AS. Histopathological assessment of the pattern liver cirrhosis in a

tropical population. Afr J Med Med Sci 2002; 31:367-369.

31. Falaiye JM, Okeke GCE, Fregene AO. Amoebic abscess in the cirrhotic liver. Gut

1980; 21:161-163.

32. Ronald M, Woofield J, McCall J, Koea J. Hepatic adenomas in male patients. HPB

2004; 6(1):25-27. DOI: 10.1080/13651 8203 10020846

33. Kim DH, Kim SU, Nam DH, Choi YJ, Park SM, Lee CK, Kim DY. A case of

hepatocellular carcinoma within hepatocellular adenoma in a non-cirrhotic male. Kor

J Int Med 2009; 24(2):147-152. DOI: 10.3904/kjim.2009.24.2.147

34. Mandong BM, Madaki AKJ, Manasseh AN. Malignant diseases in Jos: a follow up. Ann

Afr Med 2003; 2(2):49-53.

35. Malami SA, Pindiga UH, Abimiku BA, Mungadi IA, Abdullahi AD, Dauda A et al. A

descriptive retrospective study of the pattern of malignant diseases in Sokoto, north

western Nigeria (1999-2004). J Med Sci 2007; 7(6):1033-1038.

60

36. Bosch FX, Ribes J, Cleries R, Diaz M. Epidemiology of hepatocellular carcinoma. Clin

Liver Dis 2005; 9:191-211.

37. Parkin DM, Bray F, Ferley J, Pisani P. Global cancer statistics 2000. CA Cancer J Clin

2005; 55:74-108.

38. Stefaniuk P, Cianciara J, Wiercinska-Drapalo A. Present and future possibilities for

early diagnosis of hepatocellular carcinoma. World J Gastroenterol 2010 28;

16(4):418-424.

39. Williams AO, Edington GM, Obakponovwe PC. Hepatocellular carcinoma in infancy

and childhood. Br J Cancer 1967; 21(3):474-482.

40. Mustapha SK, Bolori MT, Ajayi NA, Ngadda HA, Pindiga UH, Gashau W et al.

hepatocellular carcinoma in north-eastern Nigeria: a prospective clinical study of 100

cases. Internet J Gastroenterol 2007;6(1)

41. Abdulkareem FB, Faduyile FA, Daramola A O, Rotimi O, Banjo AAF, Elesha SO, et al.

Malignant Gastrointestinal Tumours in South Western Nigeria: A Histopathologic

Analysis of 713 Cases. West Afr J Med 2009; 28(3):173-176.

42. Otegbayo JA, Oluwasola OA, Akere A, Ogunbiyi JO. Temporal and biological trends in

liver cancers at a university hospital in southwest Nigeria. Trop Doct 2006; 36:28-30.

43. Echejoh GO, Tanko MN, Manasseh AN, Ogala-Echejoh S, Ugoya SO, Mandong BM.

Hepatocellular carcinoma in Jos, Nigeria. Nig J Med 2008; 17(2):210-213.

44. Adeniji KA, Anjorin AS. The pattern of malignant tumours of the liver in a tertiary

institution in Nigeria. Afr J Med Med Sci 2004; 33:27-30.

45. Seleye-Fubara D, Jebbin NJ. Hepatocellular carcinoma in Port Harcourt, Nigeria:

Clinico-pathologic study of 75 cases. Ann Afr Med 2007; 6(2):54-57.

61

46. Benvegnu L, Alberti A. Pattern of hepatocellular carcinoma development in hepatitis

B virus and hepatitis C virus related cirrhosis. Antivir Res 2001; 52:199-207.

47. Wagida AA, Khaled HM, Amra HA, El-Nezami H, Loffredo C. Changing pattern of

hepatocellular carcinoma and its risk factors in Egypt: possibilities for prevention.

Mutat Res 2008; 659:176-184.

48. International Agency for Research on Cancer. Aflatoxins. IARC monographs

evaluating carcinogenic risks in humans. 1993; 56:245-395.

http://monographs.iarc.fr/ENG/Monographs/vol82/mono82.pdf [accessed 2nd March,

2011].

49. Udoh JM, Cardwel KF, Ikotun T. Storage structures and aflatoxin content of maize in

five agro-ecological zones of Nigeria. J Stored Products Res 2000; 36:187-201.

50. Hirohashi S, Ishak KG, Kojiro M, Wanless IR, Theis ND et al. Tumours of the liver and

intrahepatic bile ducts. In: Hamilton SR, Aaltonen LA [editors]. Pathology and

genetics of tumours of the digestive system. Lyon (France): IARC press; 2000.p.159-

172.

51. Igetei R, Otegbayo JA, Ndububa DA, Lesi OA, Anumudu CI, Hainaut P, Gormally E.

Detection of p53 codon 249 mutation in Nigerian patients with hepatocellular

carcinoma using novel evaluation of cell-free DNA. Ann Hepatol 2008; 7(4):339-344.

52. Kew MC. Synergistic interaction between aflatoxin B1 and hepatitis B virus in

hepatocellular carcinogenesis. Liver Int 2003; 23:405-409.

53. Turner PC, Mendy M, Whittle H, Fortuin M, Hall AJ, Wild CP. Hepatitis B infection and

aflatoxin biomarker levels in Gambian children. Trop Med Int Health 2000; 5:837-

841.

62

54. Anthony PP. Tumours of the hepatobiliary system. In: Fletcher CDM [editor].

Diagnostic histopathology of tumours. Vol 1. 2nd ed. London: Churchill Livingstone;

2000. p. 411-450.

55. Datta GS. Histopathological scoring of chronic viral hepatitis. Hep B Annual 2004;

1:92-112.

56. Poynard T, Mathurin P, Lai C, Guyader D, Poupon R, Tainturier M et al. A comparison

of fibrosis progression in chronic liver disease. J Hepatol 2003; 38:257-265.

57. Fattovich G, Zagni I, Scattolini C. Natural history of hepatitis B and prognostic factors

of disease progression. In: Marcellin P, editor. Management of patients with viral

hepatitis. Paris: Association pour l'amélioration de la prise en charge des malades

atteints d'hépatite virale (APMAHV); 2004.p.203-220.

58. Rigamonti C, Andorno S, Maduli E, Capelli F, Boldorini R, Sartori M. Gender and liver

fibrosis. Aliment Pharmacol Ther 2005; 21:1445-1451.

59. Kim SU, Han KH, Ahn SH. Non-invasive assessment of liver fibrosis: the gap between

ideal and real. J Gastroenterol Hepatol 2011; 26:937-942.

60. Abdulkareem FB, Banjo AF, Elesha SO. Hepatic neoplasms in Lagos: a ten-year

review of surgical biopsies (1989-1998). Nig Postgrad Med J 199; 6(3):126-129.

61. Nwokediuko SC, Ijoma UN, Obienu O. Liver cancer in Enugu, south east Nigeria.

Insight bioinformatics 2011; 1(1):1-5.

62. Yeh MM, Swanson PE. Hepatobiliary system. In: Gattuso P, Reddy VB, David O, Spitz

D, Haber MH. Differential diagnosis in surgical pathology. 2nd ed. Philadelphia.

Saunders-Elsevier; 2010.p.428-430.

63

APPENDIX A

WHO HISTOLOGICAL CLASSIFICATION OF TUMOURS OF THE LIVER AND

INTRAHEPATIC BILE DUCTS (2000)

1. Epithelial tumours

1.1. Benign:

1.1.1. Hepatocellular adenoma

1.1.2. Focal nodular hyperplasia

1.1.3. Intrahepatic bile duct adenoma

1.1.4. Intrahepatic bile duct cystadenoma

1.1.5. Biliary papillomatosis

1.2. Malignant:

1.2.1. Hepatocellular carcinoma (liver cell carcinoma)

1.2.2. Intrahepatic cholangiocarcinoma (peripheral bile duct adenoma)

1.2.3. Bile duct cystadenocarcinoma

1.2.4. Combined hepatocellular and cholangiocarcinoma

1.2.5. Hepatoblastoma

1.2.6. Undifferentiated carcinoma

2. Non-epithelial tumours

2.1. Benign:

2.1.1. Angiomyolipoma

2.1.2. Lymphangioma and lymphangiomatosis

2.1.3. Haemangioma

2.1.4. Infantile haemangioendothelioma

64

2.2. Malignant

2.2.1. Epithelioid haemangioendothelioma

2.2.2. Angiosarcoma

2.2.3. Embryonal sarcoma (undifferentiated sarcoma)

2.2.4. Rhabdomyosarcoma

2.2.5. Others

3. Miscellaneous tumours

3.1. Solitary fibrous tumour

3.2. Teratoma

3.3. Yolk sac tumour (endodermal sinus tumour)

3.4. Carcinosarcoma

3.5. Kaposi sarcoma

3.6. Rhabdoid tumour

3.7. Others

4. Haematopoietic and lymphoid tumours

5. Secondary tumours

6. Epithelial abnormalities

6.1. Liver cell dysplasia (liver cell change)

6.1.1. Large cell type (large cell change)

6.1.2. Small cell type (small cell change)

6.2. Dysplastic nodules (adenomatous hyperplasia)

6.2.1. Low grade

6.2.2. High grade (atypical adenomatous hyperplasia)

6.3. Bile duct abnormalities

65

6.3.1. Hyperplasia

6.3.2. Dysplasia (bile duct epithelium and peribiliary glands)

6.3.3. Intraepithelial carcinoma (carcinoma in situ)

7. Miscellaneous lesions

7.1. Mesenchymal hamartoma

7.2. Nodular transformation (nodular regenerative hyperplasia)

7.3. Inflammatory pseudotumour

66

APPENDIX B

KNODELL’S SCORING SYSTEM

Score

Periportal ± bridging necrosis

None 0

Mild piece meal necrosis (PMN) 1

Moderate PMN (<1/2 circumference) 3

Marked PMN (>1/2 circumference) 4

Moderate PMN + bridging necrosis 5

Marked PMN + bridging necrosis 6

Multilobular necrosis 10

Intralobular degeneration and focal

necrosis

None 0

Mild (<1/3 of lobules or nodules) 1

Moderate (1/3-2/3 of lobules or nodules) 3

Marked (>2/3 of lobules or nodules) 4

Portal inflammation

None 0

Mild (<1/3 of portal tract) 1

Moderate (1/3-2/3 of portal tract) 3

Marked (2/3 of portal tracts) 4

Fibrosis

None 0

Fibrous portal expansion 1

Bridging fibrosis (P-P) or (P-C) 3

Cirrhosis 4

67

68