SINONASAL TUMORS IN ADULT PATIENTS:

CLINICOPATHOLOGICAL PERSPECTIVE FROM CHRIS HANI

BARAGWANATH ACADEMIC HOSPITAL

Dr Lungile Precious Setoaba

0108047F

MBBCH (Wits), FCORL (SA)

A research report submitted to the Faculty of Health Sciences, University of the

Witwatersrand, Johannesburg, in partial fulfilment of the requirements for the degree of

Master of Medicine in Otorhinolaryngology

February 2021

Declaration

I, Dr Lungile Precious Setoaba, declare that this research report is my own work. It is

submitted for the degree of MMed (Otorhinolaryngology) at the University of the

Witwatersrand, Johannesburg. It has not been submitted before for any degree or

examination at this or any other University.

Dr Lungile Precious Setoaba

On this 30th day of January 2021

lungileprecioussetoaba
Stamp

iii

Dedication

I would like to dedicate this paper to my late mother, Thembekile Agnes Setoaba, who

passed away in 2002. She has always been my pillar of strength. I will forever hold her

shining smile in my heart.

Publications and presentations

This paper has not been published nor presented at a congress.

Acknowledgements

I wish to express my gratitude to all the following:

1. A special thank you to Professor Velaphi in Paediatrics who introduced me to my

excellent and dedicated supervisor Dr S. Pather. Thank you for your kindness and

professionalism Dr Pather.

2. Staff at the NHLS Laboratories for their assistance.

3. Dr Masege the head of ENT Department and the co-supervisor for this

dissertation.

4. Miss Promise Gumbo who assisted wholeheartedly with statistics.

ABBREVIATIONS AND DEFINITIONS OF TERMS

CHBH - Chris Hani Baragwanath Hospital

DLBCL - Diffuse large B-cell lymphoma

DWI - Diffusion Weighted Imaging

EBV- Epstein Barr Virus

EBV ISH - Epstein Barr Virus In Situ Hybridization

ENT- Ear Nose and Throat

HIV - Human Immunodeficiency Virus

NHLS - National Health Laboratory Service

NPC - Nasopharyngeal Carcinoma

ORL - Otorhinolaryngology

SNOMED - Systematized Nomenclature of Medicine

SCC - Squamous Cell Carcinoma

SNUC - Sinonasal Undifferentiated Carcinoma

IMRT- Intensity Modulated Radiation Therapy

WHO - World Health Organization

vii

Contents

Title page i

Declaration ii

Dedication iii

Publications and presentations iv

Acknowledgements

Abbreviations and definition of terms

Contents vii

List of figures ix

List of tables ix

Abstract xi

1 Introduction

1.1 Background 1

1.2

1.3

Anatomy

Epidemiology and aetiology

1.4 Classification of sinonasal tumors 6

1.5

1.6

1.7

1.8

Imaging of sinonasal tumors

Tumor staging

Presentation and management

Rationale

1.9 Aims and objectives 15

viii

2 Methodology 16

2.1 Study population 16

2.2 Sampling strategy 17

2.3 Data collection 17

2.4 Statistical analysis 17

2.5 Descriptive statistics 18

3.1

Results

Results in tables and figures

4 Discussion 48

5 Current application 55

6 Limitations of the current study 55

7 Conclusion 56

Appendix A: Data collection sheet

Appendix B: Ethics clearance certificate

Appendix C: Turn it in receipt

References 60

List of figures

Figure 1 Gender 23

Figure 2 Age 23

Figure 3 Tumor types 28

Figure 4 Concomitant pathology 29

Figure 5 HIV status 31

List of Tables

Table 1 Normality of sample distribution 19

Table 2 Symptoms 25

Table 3 Tumor site 26

Table 4 Tumor and tumor-like subtypes 30

Table 5 CD4 count 32

Table 6 Tumor types by gender 33

Table 7 Chi-square tests on benign and malignant tumors by gender 33

Table 8 Tumor types by age 34

Table 9 Mann-Whitney test statistics 34

Table 10 HIV results by tumour types 35

Table 11 Chi-square tests on HIV results by benign and malignant

tumours

Table 12 HIV results by malignant tumor subtypes 37

Table 13 Chi-square tests on HIV results by malignant tumor subtypes 38

Table 14 Omnibus tests of model coefficiencients 39

Table 15 Logistics regression model summary 40

Table 16 Variables in the equation 40

Table 17

Table 18

Table 19

Table 20

Table 21

Table 22

Table 23

Table 24

Table 25

Table 26

Table 27

Logistic regression model summary

Omnibus tests of model coefficients

Logistic regression model summary

Lymphoma cases associated with EBV in the study

Chi –square tests on EBV and malignant tumors

Omnibus tests of model coefficients

EBV by tumor types

Chi-square tests on EBV and malignant tumors

Omnibus tests of model of coefficients

Logistic regression model summary

Variables in the equation

Abstract

Sinonasal tumors are rare tumors of the nasal cavity and paranasal sinuses.

Demographics of adult patients with sinonasal tumours who presented at the

Otorhinolaryngology (ORL) department at Chris Hani Baragwanath Academic Hospital

between July 2013 and July 2016 were reviewed. The spectrum of these sinonasal

tumours and the influence of concomitant pathology such as Epstein Barr Virus (EBV)

were evaluated. Presenting symptoms of benign and malignant tumors were compared.

This record review study evaluated a spectrum of histopathological types of sinonasal

tumors by using SNOMED codes to access the histopathology reports at National Health

Laboratory Service (NHLS) at CHBAH. Patient demographics, presenting signs and

symptoms and associated concomitant pathology were reviewed. Lymphoma was the

most prevalent tumor at 45% and was followed by invasive poorly differentiated

squamous cell carcinoma at 19%. The average age range of patients was 30-39 years.

The nasal cavity was found to be the most common site for these tumours in 66% of the

cases, followed by the maxillary sinus at 25%. Human Immunodeficiency Virus

infection was documented in 45% of the cases and these were mostly associated with

malignant tumours. There was a significant association between malignant tumor

subtype, particularly lymphoma and EBV. EBV was present in 26% of the cases that

were assessed for the virus. Nasal obstruction was the most common presenting

symptom. Lymphoma is the most common sinonasal tumour in adult patients at the

CHBAH Otolaryngology department. HIV and EBV were found to be important viruses

that influenced their development. Squamous cell carcinoma was found to be less

common compared to what has been described in the literature, and, perharps HIV has

contributed to this shift in the trend.

1. INTRODUCTION

1.1 Background

The paranasal sinuses and the nasal cavity in the human body occupy a relatively limited

anatomical space, yet these sites may harbor complex and diverse neoplasms. These

neoplasms may develop form a variety of structures that are indigenous to these regions.

While some of the tumors may be similar to those occurring elsewhere in the human

body such as squamous cell carcinoma and adenocarcinoma, a few of the tumors such as

the olfactory neuroblastoma are unique to this region. (1) It is essential to appreciate the

complex anatomical structure displayed by the paranasal sinuses to appreciate the extent

of the tumors outlined in this study.

1.2 Anatomy

1.2.1 The nasal cavity

The nasal cavity is formed by the bony and cartilaginous skeleton. It extends anteriorly

from the nasal vestibule to terminate posteriorly at the choanae. It is separated in the

midline by the nasal septum. The bony framework is formed by the nasal bones, the

frontal process of the maxilla and the nasal part of the frontal bone. The cartilaginous

frame work is formed by the septum, lateral nasal cartilage, major and minor alar

cartilages. The nasal septum is formed by the septal cartilage anteriorly, the vomer

posterior inferiorly, and the perpendicular plate of the ethmoid bone posteriorly. The

floor is formed by the palate which is formed by the palatine processes of the maxillae

and the horizontal portions of the palatine bones. The roof is formed by the very thin

cribriform plate. The lateral walls are formed by the superior, middle and inferior

turbinates. The turbinates are attached anteriorly and have a free edge posteriorly. The

ethmoid and the maxillary ostia are also found on the lateral wall and open into the nasal

cavity. The nasal fossa is related laterally to the superior turbinate and medially to the

nasal septum. The nasal fossa houses the olfactory recess which appears as yellowish

mucosal epithelium. The mucosa houses olfactory bipolar olfactory epithelium which

crosses through the cribriform plate and terminate at the olfactory bulb. The nasal cavities

and paranasal sinuses are lined by pseudostratified columnar ciliated epithelium with

goblet cells. The blood supply to the nasal septum is from the external and internal

carotid artery. The septum is supplied by the branch of the internal carotid artery which

is the ophthalmic artery that gives off the anterior and posterior ethmoidal arteries which

supply the septum. Posteriorly the septum receives blood supply from the sphenopalatine

and greater palatine arteries which are branches of the maxillary artery, a branch of the

external carotid artery. The facial artery gives off the septal branch which also supplies

the septum. The venous drainage is via the sphenopalatine, facial and ophthalmic veins.

The lymphatic drainage is via the submandibular and upper deep cervical nodes. The

blood supply to the lateral wall of the nasal cavity is via the anterior, posterior ethmoidal

artery, the sphenopalatine artery, the lateral nasal branch which is the branch of facial

artery, and the greater palatine artery. (52,54)

1.2.2 The posterior nasal space (PNS)

This area is also referred to as the chaonae. It is formed by the two oval openings

between the nasal cavities and the nasopharynx. The rigid openings are completely

surrounded by bone. Inferiorly, it is formed by the posterior border of the horizontal

plate of the pterygoid process. Medially it is formed by the posterior surface of the

vomer. The roof of the choanae is formed anteriorly by the ala of the vomer and the

vaginal process of the medial plate of the pterygoid process. Posteriorly it is formed by

the body of sphenoid bone.(52)

1.2.3 The paranasal sinuses

The paranasal sinuses develop as outgrowths from the nasal cavities and erode into the

surrounding bones. Sinuses are absent at birth. They are lined by respiratory mucosa,

which is ciliated and secretes mucous. The paranasal sinuses are formed by the bilateral

paired frontal and maxillary sinuses, the ethmoidal sinuses which are divided into

anterior, middle and posterior air cells and the sphenoid sinus. All paranasal sinuses open

into the nasal cavities. Paranasal sinuses are innervated by the branches of the trigeminal

nerve.

1.2.4 The ethmoid sinus complex

The ethmoidal sinuses are thin walled cavities in the ethmoidal labyrinth. They are

formed by 3 large and 18 small sinuses. The ethmoidal sinuses are divided into anterior,

middle, and posterior air cells according to the location of their ostia. The posterior air

cells are larger and fewer than the anterior air cells. The anterior group which is referred

to as the infundibular group is made up of about 11 air cells open at the infundibulum.

The middle group which is referred to as bullar sinuses and is usually made up of 3 cells,

opens on or above the ethmoid bulla. The posterior group which is usually made up of

between one and seven cells opens into the superior meatus. This air cell complex lies

between the orbit and upper nasal fossa. The cribriform plate connects the left and right

group of ethmoidal complex. The cribriform plate is an essential landmark with regards

to sinonasal tumors. Cribriform plate erosion signifies erosion of the skull base and

extension of the tumor to the intracranial cavity. The medial wall of the ethmoids is

formed by the lamina from which the middle and superior turbinates are attached. The

lateral ethmoid wall is formed by a thin lamina papyracea that separates ethmoid cells

from the orbit. The roof of the ethmoids is formed by the fovea ethmoidalis which is the

medial extension of the orbital plate of the frontal bone. (1) The blood supply to the

ethmoid sinuses is via the sphenopalatine artery and the anterior and posterior ethmoidal

arteries. The nerve supply is via the orbital branches pterygopalatine, the anterior and

posterior ethmoidal nerves. The lymphatic drainage is via the submandibular nodes and

the retropharyngeal group of lymph nodes. (52)

1.2.5 The frontal sinus

The frontal sinuses are paired sinuses located between the anterior and posterior cranial

tables. They are prominent in males and rarely symmetrical. The floor of the frontal

sinus forms the roof of the orbit. It is bounded posteriorly by the anterior cranial fossa.

The frontal sinus drains into the frontal recess. The frontal recess is occupied by the cells

that determine the drainage pathway of the sinus. The frontal sinus ostium is the

narrowest region between the frontal sinus and the frontal recess. The anterior part is

formed by the frontal sinus beak and the skull base posteriorly. The frontal sinus is

sometimes associated with Kuhn (frontal) cells namely:

I – Type I frontal cell (one air cell superior to agger nasi)
II – Type II frontal cell (multiple air cells superior to agger nasi and inferior to orbital
roof)
III – Type III frontal cell (reaches the frontal sinus and is continuous with agger nasi cell)
IV – Type IV frontal cell is located within the frontal sinus

The Agger nasi cell is the anterior most air cell found anterolateral and inferior to the

frontal recess. (54) The arterial supply to the frontal sinus is via the supraorbital and

anterior ethmoidal arteries. (53) The nerve supply is from the supraorbital nerve. The

lymphatic drainage to these sinuses is via the submandibular nodes. (52)

1.2.6 The sphenoid sinus

The sphenoid sinus is located at approximately the center of the skull above the
nasopharynx. It is paired and lies within the body of sphenoid. Its posterior wall is
formed by the clivus. It relates laterally to the cavernous sinus, the internal carotid artery
and cranial nerves II–VI, and it is intimately related to the optic canal. The optic nerve
and internal carotid artery may run directly beneath the mucosa of the lateral wall of the
sphenoid sinus, without a bony covering. The sphenoid sinus is bordered superiorly by
the sella turcica, the pituitary and by the anterior and middle cranial fossae. Inferiorly it
is related to the nasal cavities. It opens at the sphenoethmoidal recess. The sphenoid sinus
has 3 types of cells which signify the extent of pneumatization:

I. Conchal type - It is commonly found in children. The area of inferior to the sella
turcica is a solid block of bone that has no pneumatization.

II. Presellar type – With the type, pneumatization does not extend beyond the
coronal plane defined by the anterior sellar wall.

III. Sellar type - pneumatization extends into the body of the sphenoid and extends to
the clivus

The blood supply to the sphenoid sinus is via the posterior ethmoidal artery. (52,55) The
nerves supply the posterior ethmoidal nerves and orbital branches. The lymphatic
drainage is via the retropharyngeal group of nodes.

1.2.7 The maxillary sinus

The maxillary sinuses are paired sinuses that lie within the body of the maxilla. Each
sinus is pyramidal in shape consisting of the base, apex, roof and the floor. The
superolateral surface (roof) is related above to the orbit. The anterolateral surface is
related below to the roots of the upper molar and premolar teeth and in front of the face.
The posterior wall is related posteriorly to the infratemporal fossa. The maxillary sinus
opens into the hiatus semilunaris. It borders the nasal cavity laterally. Behind the
maxillary sinus is the pterygopalatine fossa, which is traversed by the maxillary artery
along with branches of the trigeminal nerve and autonomic nervous system. The
variation of the maxillary sinus is the presence of the Haller cell, commonly known as the
infraorbital cell. They are the extension of the anterior ethmoidal cells along the floor of
the orbit. (56) The blood supply to the maxillary sinus is via the facial artery, the greater
palatine artery, and the infraorbital vessels. The nerve supply is via the infraorbital
nerve, and superior alveolar nerves. The lymphatic drainage is via the submandibular
lymph nodes. (52)

1.3 Epidemiology and etiology

Sinonasal tumors occur in the nasal cavity and paranasal sinuses. In the literature, it is

noted that 3% of sinonasal tumors are malignant. (1,2) With regards to the anatomical

site, 60% of sinonasal tumors originate in the maxillary sinus, 20-30% in the nasal cavity,

10-15% in the ethmoid sinus and 1% in the sphenoid and/or frontal sinuses. (3) The most

common (70-80%) malignant sinonasal histological type is the squamous cell carcinoma,

followed by adenocarcinoma and adenoid cystic carcinoma (10% each). (4) These tumors

are seen more commonly in males than females and are frequently seen in the fourth and

eighth decades of life. (5)

The annual incidence of cancer of the nasal cavity and paranasal sinuses is reported to be

low in most populations (1 in 100 000 per annum). (6) However, higher rates are

recorded in Japan and certain parts of China and India. Squamous cell carcinoma is the

most common. Over time, a stable incident rate and a slight decline has been reported in

recent decades.

Benign sinonasal tumors have been associated with exposure to certain viruses such as

the Human Papillomavirus (HPV) 6 and 11. HIV infection often presents as an

association in certain patients with these tumors, but its role has not been entirely

determined. Exposure to allergens, air pollution and industrial carcinogens influences the

development of benign tumors. (6,7,8) Tobacco, alcohol and industrial exposure to heavy

metal particles (such as nickel and chromium), particularly for workers in the leather,

textile, furniture and wood industries, are considered carcinogenic and are associated

with various types of malignant sinonasal tumors. Adenocarcinoma is known to be

associated with exposure to wood dust and leather tanning. (8)

1.4 Classification of sinonasal tumors

The World Health Organization (WHO) has a broad classification for sinonasal tumors.

This classification divides sinonasal tumors according to the tissue of origin and whether

they are benign or malignant. The histological types include epithelial tumors,

hematolymphoid tumors, skin and muscle tumors, bone and cartilage tumors,

neuroectodermal tumors, germ cell tumors and metastatic tumors. (1)

1.4.1 Benign neoplasms

Epithelial tumors

The benign epithelial tumors consist of papillomas, salivary gland adenomas and mixed

tumors known as pleomorphic adenomas. Papillomas in the sinonasal tract may have an

inverted growth pattern. These are termed inverted papilloma and categorized as

sinonasal Schneiderian papilloma. These neoplasms are known to have high recurrence

rates and can be locally aggressive. HPV has been implicated in the development of this

tumor. Treatment is complete surgical resection. Incomplete resection results in tumour

reoccurrence. (1,36)

Soft tissue tumors

Soft tissue tumours are very rare tumours of the sinonasal tract and consist of nerve

sheath tumours, hemangioma, angiofibroma, meningioma, myxoma, and leiomyoma.

Neurofibroma is a peripheral nerve tumour and it is very rare in the sinonasal tract. It is

associated mainly with neurofibromatosis. Associated with this tumour are

schwannomas which arise from the nerve sheath and are associated with

Neurofibromatosis 2 (NF2). Malignant transformation of these tumours is very rare and

often associated with neurofibroma. (9) Juvenile angiofibroma is a vascular tumour that

originates from the sphenopalatine foramen and grows towards the nasal cavity, or may

extend into the nasopharynx or extend laterally into the pterygopalatine foramen. It is

more common in males and if found in females genetic mosaicism should be considered.

(10) It commonly presents in a young male with nasal obstruction and epistaxis.

Preoperative embolization prior to surgical excision is performed to minimize the risk of

intra-operative bleeding.

Bone and cartilage tumors

Bone and cartilage tumours may affect the sinonasal tract and present with disfigurement

of the face and compression of the cranial nerves and other vital structures. They consist

of osteoma, fibrous dysplasia, and ossifying fibroma. Fibrous dysplasia is the most

common of the tumours and commonly involves the craniofacial skeleton. Osteoma is

most commonly found in the frontal and ethmoid sinuses. They are often found

incidentally and are asymptomatic. These may present as multiple tumours in multiple

sites in Gardner syndrome. (11) Ossifying fibroma consists of mattered bone originating

from mesenchyme of peridontal ligament and commonly occurs in the mandible and less

commonly in the sinus walls and the maxilla. They are seen mostly in young females and

can be locally aggressive. (12)

1.4.2 Malignant neoplasms

Squamous cell carcinoma is the most common epithelial malignancy followed by

adenocarcinoma. These are both commonly found in males. Occupational exposure to

irritants such as wood dust, chrome and nickel are considered significant risk factors for

the development of these tumors. (5,6,8,13) Sinonasal undifferentiated carcinoma

(SNUC) is an aggressive neoplasm without evidence of squamous or glandular

differentiation. It is often considered a neuroendocrine tumor due to the similarities it

shares with those tumors. It is commonly seen in males and often presents with nodal

disease and distant metastasis. This tumor has a poor prognosis with a 5-year survival

rate of less than 20%. (14)

Adenoid cystic carcinoma is the most common salivary gland tumor and is found

predominantly in men. It has the potential for perineural spread and intracranial extension

affecting most commonly the maxillary division of the Trigerminal nerve. (15)

Non-Hodgkin lymphoma is the most common hematolymphoid neoplasm of the

sinonasal tract. (16) Sinonasal lymphoma tends to occur more commonly in the nasal

cavity compared to the paranasal sinuses. Other types of hematolymphoid neoplasms

include plasmacytoma, myeloid, histiocytic sarcomas and Langerhans cell histiocytosis.

Neuroectodermal malignancies

Esthesioneuroblastoma is a neuroendocrine tumor arising from the olfactory nerve. It

commonly occurs in males and has no known associated risk factors to date. It is a highly

vascular tumor and epistaxis is often a commonly presenting symptom. Intracranial and

intra-orbital extension is often seen at presentation. Nodal disease occurs in 15-20% of

the cases and it is often associated with recurrence. (17)

Sinonasal melanoma is a very rare mucosal melanoma. (22) It is a very aggressive

sinonasal tumor and caries a very poor prognosis. The lesion tends to be vascular and

epistaxis is a common presentation.

Malignant tumors of soft tissue, bone and cartilage

Mesnchymal tumors originate from muscle, bone and cartilage. Osteosarcoma and

chondrosarcoma occur commonly in adults, while rhabdomyosarcoma is seen mostly in

children. These tumors tend to be aggressive with a 5-year survival rate of 47%. (23)

1.5 Imaging of sinonasal tumors

Imaging is key to the evaluation and diagnosis of the sinonasal tumors. It is helpful in

assessing the precise location of the lesion and assists with identifying involved local and

distant structures. Imaging is also essential in planning surgical and oncological

treatment. The computed tomography scan (CT scan) provides detailed bone anatomy

and involvement of surrounding structures, the vascularity of the lesion and texture. In

sinonasal pathology, essential anatomical structures such as the anterior and middle

cranial fossa, orbit, pterygopalatine fossa, palate, or infratemporal fossa (masticator and

parapharyngeal space) are of importance to evaluate as their involvement influences

surgical planning.

A CT scan is the most commonly used imaging modality because of its wider

availability, easy access, lower cost, and potential to offer greater anatomic detail. In

comparison to MRI, CT is particularly effective in delineating calcifications and

evaluating the pattern of bone invasion. In addition, certain lesions on imaging are

typical of a specific diagnosis, though histological diagnosis is still required. Bone

changes such as bone erosion and destruction can give information with regards to tumor

aggression. High grade malignancies such as lymphoma and the squamous cell

carcinoma show extensive bony destruction, whereas small round cell tumors show

permeative invasion and lack of bone destruction. Benign lesions and low grade

malignancies may cause bone expansion due to their slow and expansive growth.

Calcifications are observed in some sinonasal disorders, such as adenocarcinoma,

olfactory neuroblastoma, inverted papilloma, fibrous dysplasia, osteoma, osteosarcoma,

cartilaginous tumor, fungal sinusitis, and dentigerous tumor. Characteristic patterns of

bone invasion help predict the tumor histology. Contrast-enhanced CT is invaluable for

the identification of the feeding artery and for the diagnosis of highly vascular tumors.

The magnetic resonance imaging (MRI) provides significant information with regards to

the soft tissue involvement and neural structures. Malignant tumors usually exhibit

nonspecific hyper intensity on T2-weighted images (T2WI) and hypo- to isointensity on

T1-weighted images (T1WI). On T2WI, mucinous or cartilaginous tumors show marked

hyperintensity, hypercellular tumors show slight hyperintensity and tumors with fibrosis,

calcification, or flow void show hypointensity. On T1WI, hyperintensity within a tumor

is indicative of the presence of methemoglobin, melanin, lipid, protein, and mineral

elements. Diffusion-weighted image (DWI) with measurement of apparent diffusion

coefficient (ADC) captures the degree of Brownian movement of the water molecules in

tissues, which serves as a useful imaging biomarker. Low-ADC lesions with strong

diffusion restriction indicate hypercellularity, abscess, or hemorrhage, whereas high-

ADC lesions indicate hypocellularity, mucus, cartilage, or fluid. Therefore, DWI with

ADC measurement is can be useful to differentiate between benign and malignant

tumors. (18, 19, 20, 21)

Tumor location plays a significant role in arriving at a differential diagnosis. Tumors

involving the region of the cribriform plate and upper nasal cavity suggest diagnoses such

as olfactory neuroblastoma or meningioma. Inverted Schneiderian papilloma occurs

predominantly along the lateral wall of cavity and the medial wall of the maxillary

sinus.(36) In the lower maxilla, odontogenic lesions should be considered. Such lesions

arise in the bone of the alveolar process and during their growth elevate the floor of the

maxillary sinus. Fibroosseous lesions typically arise from bone and follow the contour of

the bone. On imaging it is seen as a radiodense lesion. Correlation of imaging studies

with histologic appearance is crucial in the evaluation of bony lesions.

1.6 Tumor staging

Staging of the sinonasal tumors is complex. The tumors are staged according to the

tumors that start in the nasal cavity or ethmoids, and maxillary sinus. Tumor stage is not

characterized by size, but the extent of sites and local structures involved. The late stage

of the tumor is characterized by invasion of adjacent structures such as the orbit, base of

skull, cranial nerves, brain and facial skin. Below is the detailed staging for sinonasal

tumors:

TX- Primary Tumour cannot be assessed

T0 -No evidence of primary Tumour

Tis =Carcinoma in situ

Maxillary Sinus

T1 Tumor limited to the maxillary sinus mucosa with no erosion or destruction of

bone.

T2 Tumor causing bone erosion or destruction, including extension into the hard

palate and/or middle nasal meatus, except extension to the posterior wall of the

maxillary sinus and pterygoid plates.

T3 Tumor invades any of the following: bone of the posterior wall of the

maxillary sinus,subcutaneous tissues, or medial wall of the orbit, pterygoid fossa,

or ethmoid sinuses.

T4a Moderately advanced local disease

Tumour invades anterior orbital contents, skin of cheek, pterygoid plates,

infratemporal fossa, cribriform plate, sphenoid or frontal sinuses.

T4b Moderately advanced local disease

Tumor invades any of the following: orbital apex, dura, brain, middle cranial

fossa, cranial nerves other than maxillary division of trigeminal nerve. (V2),

nasopharynx, or clivus.

Nasal cavity and ethmoid sinus

T1 Tumor restricted to any one subsite, with or without bony invasion

T2 Tumor invades two subsites in a single region or extending to involve

an adjacent region within the nasoethmoidal complex, with or without bony

invasion

T3 Tumor extends to invade the medial wall or oor of the orbit, maxillary sinus,

palate, or cribriform plate

T4a Moderately advanced local disease. Tumor invades any of the following:

anterior orbital contents, skin of nose or cheek, minimal extension to anterior

cranial fossa, pterygoid plates, sphenoid or frontal sinuses

T4b Moderately advanced local disease

Tumor invades any of the following: orbital apex, dura, brain, middle cranial

fossa, cranial nerves other than maxillary division of trigeminal nerve (V2),

nasopharynx, or clivus.

Regional lymph nodes

NX Regional lymph nodes cannot be assessed.

N0 regional nodes metastasis.

N1 Metastasis in a single ipsilateral lymph node, 3 cm or less in greatest

dimension.

N2 Metastasis in a single ipsilateral lymph node, more than 3 cm but not more

than 6 cm in its greatest dimension; or in multiple ipsilateral lymph nodes, no

more that 6 cm in its greatest dimension; or in bilateral or contralateral lymph

nodes, none greater than 6 cm in their greatest dimensions.

N2a Metastasis in a single ipsilateral lymph node, more than 3 cm but not more

than 6 cm in greatest dimension.

N2b Metastasis in multiple ipsilateral lymph nodes, none more that 6 cm in

greatest dimension.

N2c Metastasis in bilateral or contralateral lymph nodes, none more than 6 cm in

greatest dimension .

N3 Metastasis in a lymph node more than 6 cm in greatest dimension.

Stage grouping

Stage 0 Tis N0 M0

Stage I T1 N0 M0

Stage II T2 N0 M0

Stage III T3 N0 M0

T1 N1 M0

T2 N1 M0

T3 N1 M0

Stage IVA T4a N0 M0

T4a N1 M0

T1 N2 M0

T2 N2 M0

T3 N2 M0

T4a N2 M0

Stage IVB Any T N3 M

T4b Any N M0

Stage IVC Any T Any N M1

Clinical stage grouping by T and N status

Sinonasal tumor staging adapted from: Deschler DG, Moore MG, Smith RV, eds.
Quick Reference Guide to TNM Staging of Head And Neck Cancer and Neck Dissection
Classification, 4th ed. Alexandria, VA: American Academy of Otolaryngology-Head and
Neck Surgery Foundation, 2014.

1.7 Presentation and management

Malignant neoplasms of this region may lead to morbidity and disfigurement. Patients

with sinonasal tumors present to the ENT surgeon with nonspecific complaints such as

epistaxis, nasal obstruction, headache, nasal congestion, nasal discharge, swelling, and

facial pain or numbness. (25) Loss of vision and diplopia is most often as a result of the

tumor compressing or invading the orbital nerve or direct involvement of occulomotor

nerve. Epiphoria is due to the obstruction or infiltration of the lacrimal duct. Trismus

heralds an advanced tumor invading the muscles of mastication and at times, the invasion

of the pterygoid plates. A neck mass is also a sign of advanced disease in the setting of

carcinoma due to the likelihood of nodal metastases. Hearing loss may result from

nasopharyngeal extension of the tumor obstructing the Eustachian tube with resultant

middle ear effusion. What is noted further is that 9-12% of patients are frequently

N T1 T2 T3 T4a T4b

N0 I II III IVa IVb

N1 III III III IVa IVb

N2 IVa IVa IVa IVa IVb

N3 IVb IVb IVb IVb IVb

asymptomatic and this further contributes to a delay in diagnosis and therefore an

advanced stage at presentation. (25,26)

Diagnosis begins with a thorough clinical history and physical examination. Computed

tomography/magnetic resonance imaging (CT/MRI) scans are done to stage the tumor

locally and to evaluate for the presence of metastasis. Biopsy of the tumor is necessary to

make a final diagnosis. Tumor proximity to vital structures such as the brain, optic

nerves, and internal carotid artery pose significant challenges for their treatment and may

be the source of the significant morbidity of the patients. Almost all the benign tumors

have a tendency to recur with locally destructive capability and have a 5–15% likelihood

of progressing to malignancy. Surgery (open or endoscopic) is the mainstay of treatment

with or without radiation therapy and/or chemotherapy. (26) Sinonasal tumors carry a

poor prognosis, despite an early diagnosis, radical surgical resection and strict follow-up.

1.8 Rationale

A number of patients present to the ENT division at CHBAH with sinonasal tumors of

varying subtypes. A retrospective assessment of the clinical and histopathological

spectrum of sinonasal tumors allowed for comparison with documented published

literature and brought to our attention the burden of sinonasal disease and the prominent

histological subtypes presenting to the unit. This is important, as there had been a

notable difference in the most common histological type presenting in the CBHAH ENT

unit in comparison to what has been documented in the literature. In addition, the role of

the human immunodeficiency virus (HIV) infection in sinonasal tumors will be

evaluated.

1.9 Aims and objectives

1.9.1 Aim of the research

To document the clinical and pathological spectrum of benign and malignant sinonasal

tumors in an HIV-seroprevalent South African setting

1.9.2 Objectives

• To describe the demographic details of adult patients (age and gender) and the

topographic site of biopsy sampled at Chris Hani Baragwanath Academic

Hospital (CHBAH) over a three year time frame, July 2013 to July 2016.

• To describe the pathological spectrum of sinonasal tumour subtypes occurring in

HIV positive and negative patients.

• To describe the presence of concomitant pathology such as cytomegalovirus

infection, granulomatous inflammation or parasites.

• To compare the presenting symptoms of benign and malignant tumours.

2. Methodology

A record review study was conducted to document the spectrum of benign and malignant

sinonasal tumors in patients who presented at the ENT clinical unit at CHBAH between

July 2013 and July 2016. Ethics approval of this study was obtained from the Human

Research Ethics Committee, Medical, at the University of the Witwatersrand (clearance

certificate number M170668).

2.1 Study population

This study included 53 adult male and female patients, above the age of 18 years, who

presented with a sinonasal mass at CHBAH, ORL department between July 2013 and

July 2016. These patients were clinically assessed between the time frame of July 2013

and July 2016. Biopsies were performed with the intent of establishing a

histopathological diagnosis at the National Health Laboratory Service (NHLS). HIV

positive and negative patients were included in this study.

2.2 Sampling strategy

The histopathological reports of all patients included in this study were retrieved from the

NHLS following a SNOMED (Systematized Nomenclature of Medicine) search of the

laboratory database. The following codes were used to retrieve cases (T-code represents

the anatomical topography and M-code represents neoplastic subtype): T-2200 (nasal

sinus structure), T-21002/3 (nasal region), T-22100 (maxillary sinus), T-22200 (frontal

sinus), T- 22300 (ethmoid sinus), T22400 (sphenoid sinus), M-80000 (benign neoplasm),

M-80001 (neoplasm: uncertain whether benign or malignant) and M-0003 (neoplasm

malignant). The histopathological reports were retrieved. The patient’s demographic

information, the clinical features and the pathological details were documented from the

reports. The HIV status and CD4 count were documented from the NHLS database.

Concomitant pathology was also documented.

2.3 Data collection

Data was collected using the SNOMED electronic records from the NHLS Corporate

Data Warehouse and patient records were retrieved using episode numbers. A data

collection sheet (Appendix A) was utilized to detail age, gender, HIV status, CD4 count,

nasal symptoms, tumor site/topographic region, gross appearance and histopathological

findings. There was no mention of other concomitant symptoms either than nasal

obstruction and therefore, nasal obstruction was selected as the commonest symptom.

Concomitant pathology was also documented.

2.4 Statistical analysis

The data was analysed using IBM SPSS Statistics. Descriptive statistics were run to

produce and present basic features of the study sample in the form of frequencies and

percentages as well as measures of central tendency. Normality of the sample data was

tested in order to determine if parametric or non-parametric methods would be more

suitable for the main analysis. Among the inferential methods used were the Pearson’s

Chi square, the Mann-Whitney test and binomial logistic regression. All the multivariate

and statistical significance testing was done at a 95% confidence interval.

2.5 Descriptive statistics

2.5.1 Normality tests

The data was first tested for normality to determine if parametric or non-parametric

methods would be more appropriate for the subsequent inferential analysis. Skewness is

a measure of symmetry in a distribution and a value between -0.5 and 0.5 is regarded as

indicative of approximately symmetric distribution. Table 1 shows that skewness for

gender (0.79), age (0.74) and tumor type (1.64) were all above the recommended

threshold for parametric testing. Tests suitable for analysis of data that is not sufficiently

normally distributed were therefore utilised.

Table 1: Normality of sample distribution

Statistic Std. Error
Gender Mean 1.68 0.06

Median 2.00
Variance 0.22
Std. Deviation 0.47
Skewness 0.79 0.33

Age Mean 44.64 1.85
Median 42.00
Variance 182.35
Std. Deviation 13.50
Skewness 0.74 0.33

Tumour
type

Mean 1.19 0.05
Median 1.00
Variance 0.16
Std. Deviation 0.39
Skewness 1.64 0.33

HIV Mean 1.58 0.08
Median 2.00
Variance 0.32
Std. Deviation 0.57
Skewness 0.30 0.33

3. Results

The collected data comprised a total of 53 cases, of these, 36 (68%) were male and 17

(32%) were female patients (Figure 1). The largest concentrations were in the 30-39 age

range (29%) and 40-49 age range (30%), with about 6 in 10 (59%) of the patients falling

in the 30-49 range. (Figure2) The nasal cavity was specified as the site of the tumor in

two thirds (66%) of the cases. The second most common site was the maxillary sinus at

25% (Table 3). There was a total of 43 (81%) malignant tumor and 10 (19%) benign

tumor cases in the study. (Figure 3) Lymphoma was the most prevalent tumor subtype at

45% and was followed by invasive poorly differentiated squamous cell carcinoma (SCC)

at 19% (Table 4). The most common benign tumor was inverted papilloma and tumor-

like inflammatory polyps. The presenting concomitant pathology is described in Figure 4.

Two thirds (66%) of the cases did not have any accompanying pathology while EBV was

demonstrated in 26% of cases tested for EBV and choanal atresia in 2% of the cases.

HIV testing was non-reactive in 27 of the 53 cases (51%) and reactive in 24 cases (45%).

The HIV status of two of the patients in this sample was not specified. (Figure 4) The

CD4 count was recorded in 18 of the 24 cases that tested positive for HIV. The recorded

CD4 count ranged widely between a minimum of 5 cells/uL and a maximum of 787

cells/uL, with a mean of 336 cells/uL as shown in Table 5. The relationship between

benign and malignant tumors and HIV was examined, first, using the Chi-square test and

then using the logistic regression method. The percentage of cases testing positive for

HIV was higher amongst patients with malignant tumors at 56% compared to only 10%

amongst patients with benign tumors (Table 10). The differences in HIV results between

malignant and benign tumor cases were statistically significant (Chi-square=6.857,

p=0.01). Therefore, cases with malignant tumors tended to be significantly more

associated with HIV reactive results compared to those with benign tumors (Table11).

The results showed a significant association between malignant tumors and HIV positive

results, the analysis further examined if the HIV association varied between the different

diagnosed malignant tumor subtypes (viz. lymphoma, SCC and other less prevalent

subtypes). As shown in Table 12, the percentage of cases testing positive for HIV among

patients with lymphoma was 83% compared to 40% amongst patients with SCC, while

none of the 8 patients diagnosed with other tumor subtypes tested positive for HIV. The

differences in HIV results between patients with lymphoma, SCC and other malignant

tumor subtypes were statistically significant (Chi-square=17.838, p=0.00).

The percentage of HIV reactive results was significantly higher in lymphoma cases

compared to cases with SCC as well as those with other types of malignant tumors.

(Table 13) The logistic regression model explained 53% of the variance in HIV test

results based on malignant tumor subtypes as reflected by the Nagelkerke R² value =

0.53. A malignant tumor subtype was thus a significant predictor of HIV test results in

patients (Wald chi-square 11.703, p=0.00) with a strong association between lymphoma

and reactive HIV test results.

EBV was tested in certain types of lymphomas. Table 20 shows the types of lymphomas

associated with EBV, those that were EBV negative , and those that were not specifically

tested for EBV. As shown in Table 23, 33% of malignant tumor cases had EBV. Table

24 shows that the EBV prevalence was associated with patients with malignant tumors

(Chi-square = 4.425, p=0.04). The results showed a significant association between

malignant tumors and EBV. The analysis extended to examine the association between

EBV and specific malignant tumor subtypes namely, lymphoma and SCC. The

differences in EBV between patients with Lymphoma and SCC and were statistically

significant (Chi-square=7.528, p=0.02). The percentage of cases with EBV was

significantly higher in lymphoma cases. Binomial logistic regression was then performed

to determine if EBV could be predicted from the presenting malignant tumor subtypes.

In this instance malignant tumor subtypes were analysed as the predictor variables of

EBV as shown in the omnibus tests table, the logistic regression model was statistically

significant (Chi-square=7.163, p=0.01). A malignant tumor subtype was thus a

significant predictor of EBV in patients (Wald chi-square 4.975, p=0.03), with a strong

association between lymphoma and positive EBV status.

The presenting nasal and other symptoms were not described in 24 of the 53 cases

(45%). Where the symptoms were described, nasal obstruction was the most common at

34%, a figure about twice that of nasal mass, the next most common symptom. At least

12 of the cases presented with more than one nasal symptom. (Table 2) Proptosis was

present in 9% of the patients. Cranial nerve fallout, sinusitis and facial numbness were

equally distributed in 4% of the patients. The less commonly presenting symptoms

included epistaxis and cervical lymphadenopathy, facial asymmetry, hearing loss, nasal

bridge deformity, otitis media with effusion and skin involvement

3.1 Results in tables and figures

The collected data had a total of 53 cases of which 36 (68%) were male and 17 (32%)

female patients

Figure 1: Gender

Figure 2: Age

68%

32%

Gender (n=53)

Male Female

13%

29% 30%

13%
15%

20-29 30-39 40-49 50-59 60 and older

Age (n=53)

Figure 2 shows the age brackets and the largest concentrations were in the 30-39 age

range (29%) and 40-49 age range (30%), with about 6 in 10 (59%) of the patients

therefore falling in the 30-49 range

Presenting symptoms

The presenting nasal and other symptoms were not described in 24 of the 53 cases (45%).

Where the symptoms were described, nasal obstruction was the most common at 34%, a

figure about twice that of nasal mass, the next common symptom. At least 12 of the cases

presented more than one nasal symptom, hence the count exceeding 100% in the results

table.

Table 2: Percentage of presnting symptoms

Symptoms Number of patients Percentage

Not described 24 45%

Nasal obstruction 18 34%

Nasal mass 8 15%

Visual symptoms/Proptosis 5 9%

Associated tonsillar mass/hypertrophy 2 4%

Cranial nerve fallout 2 4%

Facial numbness 2 4%

Sinusitis 2 4%

Cervical lymphadenopathy 1 2%

Epistaxis 1 2%

Facial asymmetry 1 2%

Hearing loss 1 2%

Nasal bridge deformity 1 2%

Otitis media with effusion 1 2%

Premaxillary mass extension 1 2%

Semter’s triad 1 2%

Tumor sites

The nasal cavity was given as the site of the tumour in two thirds (66%) of the cases. The

second most common site was the maxillary sinus at 25% as shown in the table below.

Notably, the tumour covered more than one specific site in many of the patients, hence

the counts that in excess of 100% .

Table 3: Percentage of tumor sites

Tumour site Number of patients Percentage

Nasal cavity 35 66%

Maxillary sinus 13 25%

Posterior nasal cavity 7 13%

Ethmoids 6 11%

Posterior nasal space 6 11%

Nasopharynx 3 6%

Orbit 3 6%

Frontal sinus 2 4%

Nasal mass 2 4%

Base of tongue 1 2%

Bilateral nasal cavity 1 2%

Cribriform 1 2%

Lateral pharyngeal wall 1 2%

Middle meatus 1 2%

Nasal skin 1 2%

Oral cavity 1 2%

Palate 1 2%

Parapharyngeal space 1 2%

Posterior nasal mass 1 2%

Skull base 1 2%

Tumor types

There was a total of 43 (81%) malignant tumor and 10 (19%) benign tumor cases in the

study.

Figure 3

81%

19%

Tumor types (n=53)

Malignant Benign

Concomitant pathology

The presenting concomitant pathology is described in Figure 4. Two thirds (66%) of the

cases did not have any accompanying pathology while EBV was diagnosed in 26% of the

cases tested for EBV and choanal atresia in 2% of the cases.

Figure 4

Tumor and tumor-like subtypes

Lymphoma was the most prevalent tumour subtype at 45% and was followed by Invasive

poorly differentiated squamous cell carcinoma (SCC) at 19%. (Table 4)

26%

66%

Choanal atresia

Bacterial colonies

EBV

None

Concomitant pathology (n=53)

Table 4

Tumor and tumor-like subtypes

Number of

patients Percentage

Lymphoma 24 45%

SCC 10 19%

Inflammatory polyp 3 6%

Inverted papilloma 3 6%

Adenoid cystic carcinoma 2 4%

Lymphoid hyperplasia 2 4%

Melanoma 2 4%

Vascular tumours 2 4%

Alveolar Rhadomyosarcoma 1 2%

Chondrosarcoma 1 2%

Nasopharyngeal carcinoma 1 2%

Sinonasal Adenocarcinoma 1 2%

Spindle cell carcinoma 1 2%

HIV status

HIV testing was non-reactive in 27 of the 53 cases (51%) and reactive in the other 24
cases (45%). The HIV status of two of the patients in the sample was not specified.

Figure 5

51% 45%

4%
HIV (n=53)

Non reactive Reactive Unspecified

CD4 count (cells/uL)

The CD4 count was recorded in only 18 of the 24 cases that tested positive to HIV. The

recorded CD4 count ranged widely between a minimum of 5 to a maximum of

787cells/uL, with a mean of 336 as shown below.

Table 5

Statistic Std. Error
CD4 count Mean 335.72 62.01

95% Confidence Interval for
Mean

Lower
Bound 204.89

Upper
Bound 466.56

5% Trimmed Mean 329.02
Median 240.00
Variance 69220.45
Std. Deviation 263.10
Minimum 5
Maximum 787
Range 782
Interquartile Range 490.25
Skewness 0.64 0.54
Kurtosis -1.18 1.04

Tumor types by gender

The Pearson’s Chi square test was run to compare the prevalence of tumor types by

gender and 65% of female patients had malignant tumors compared to 89% among male

patients as shown below

Table 6

Type of tumor Gender

Female (n=17) Male (n=36)

Malignant tumors 65% 89%

Benign tumors 35% 11%

Chi-square tests on benign and malignant tumors by gender

The results in Table 7 show that prevalence in benign and malignant tumors differed

significantly by gender (Chi Square=4.411, p=0.04), with the proportion of malignant

tumour cases markedly higher amongst males compared to females.

Table 7

Value

Asymptotic
Sig.
(2-sided)

Pearson Chi-Square 4.411 1 0.04*

Likelihood Ratio 4.146 1 0.04

N of Valid Cases 53

p<0.05* = statistically significant

Tumor types by age

Table 8 shows how benign and malignant tumours were distributed across the different

age groups. The Mann-Whitney test was used to compare the differences in tumour types

by patient age and the test statistics in Table 9 show that differences in tumour types by

age were not statistically significant (U=32.00, p=0.39).

Table 8

Type of tumor Age

20-29 30-39 40-49 50-59 >60

n=7 n=15 n=16 n=7 n=8

Malignant 57% 80% 94% 71% 88%

Benign 43% 20% 6% 29% 13%

Table 9: Mann-Whitney test statistics

Type of

tumour

Mann-Whitney U 32.00

Wilcoxon W 123.00

Z -0.87

Asymp. Sig. (2-tailed) 0.39

p<0.05* = statistically significant

HIV results by tumor types

The Pearson’s Chi square test was run to compare differences in HIV status between

patients with benign and malignant tumors. The two cases whose classification in terms

of HIV was unknown were excluded from this analysis, leaving 51 cases. As shown in

Table 10, the percentage of cases testing positive to HIV was higher amongst patients

with malignant tumours at 56% compared to only 10% amongst patients with benign

tumors.

Table 10

Tumour types N HIV

Reactive Non-reactive

Malignant 41 56% 44%

Benign 10 10% 90%

Total 51 47% 53%

Chi-square tests on HIV results by benign and malignant tumors

Table 11 shows that these differences in HIV results between malignant and benign

tumor cases were statistically significant (Chi-square=6.857, p=0.01). Therefore, cases

with malignant tumors tended to have significantly more HIV reactive results compared

to those with benign tumors.

Table 11

Value

Asymptotic
Sig.
(2-sided)

Pearson Chi-Square 6.857 1 0.01*

Likelihood Ratio 7.796 1 0.01

N of Valid Cases 51

p<0.05* = statistically

HIV results by malignant tumor subtypes

With the results showing a significant association between malignant tumors and HIV

positive results, the analysis extended to examining if the HIV association varied between

the different diagnosed malignant tumor subtypes (viz. Lymphoma, SCC and other less

prevalent subtypes). As shown in the table below, the percentage of cases testing positive

to HIV among patients with Lymphoma was 83% compared to 40% amongst patients

with SCC, while none of the 8 patients diagnosed with other tumour subtypes tested

positive to HIV.

Table 12

Malignant tumour

subtypes

N HIV

Reactive Non-reactive

Lymphoma 23 83% 17%

SCC 10 40% 60%

Other 8 0% 100%

Total 41 56% 44%

Chi-square tests on HIV results by malignant tumor subtypes

Table 13 shows that the above differences in HIV results between patients with

Lymphoma, SCC and other malignant tumour subtypes were statistically significant (Chi-

square=17.838, p=0.00). The percentage of HIV reactive results was significantly higher

in Lymphoma cases compared to cases with SCC as well as those with other types of

malignant tumours.

Table 13

Value

Asymptotic
Sig.

(2-sided)
Pearson Chi-Square 17.838 2 0.00*

Likelihood Ratio 21.513 2 0.00

N of Valid Cases 41

p<0.05* = statistically significant

Predicting HIV status from malignant tumor subtypes

Binomial logistic regression was then performed to determine if HIV test results could be

predicted from the presenting malignant tumor subtypes. In this analysis malignant

tumor subtypes were allocated as the predictor variables and HIV results as the outcome

variable. As shown in the omnibus tests table, the logistic regression model was

statistically significant (Chi-square=20.543, p=0.00).

Table 14: Omnibus tests of model coefficients

Chi-square df Sig.

Step 1 Step 20.543 1 0.00

Block 20.543 1 0.00

Model 20.543 1 0.00*

p<0.05* = statistically significant

Table 15: Logistic regression model summary

Step -2 Log Likelihood Cox & Snell R Square

Nagelkerke R

Square

1 35.684 0.394 0.53

Malignant tumor subtype was thus a significant predictor of HIV test results in patients

(Wald chi-square 11.703, p=0.00), with a strong association between Lymphoma and

reactive HIV test results.

Table 16: Variables in the Equation

B S.E. Wald df Sig. Exp(B)

Malignant tumor

subtype
2.376 0.695 11.703 1 0.00* 10.764

Constant -4.044 1.132 12.755 1 0.00 0.018

p<0.05* = statistically significant

The logistic regression model explained 53% of the variance in HIV test results based on

malignant tumour subtypes as reflected by the Nagelkerke R² value = 0.53 in Table 17

Table 17: Logistic regression model summary

Step -2 Log Likelihood Cox & Snell R Square

Nagelkerke R

Square

1 35.684 0.394 0.53

Malignant tumor subtype was thus a significant predictor of HIV test results in patients

(Wald chi-square 11.703, p=0.00), with a strong association between Lymphoma and

reactive HIV test results.

Table 18: Omnibus tests of model coefficients

Chi-square df Sig.

Step 1 Step 7.796 1 0.01

Block 7.796 1 0.01

Model 7.796 1 0.01*

p<0.05* = statistically significant

The logistic regression model explained 19% of the variance in HIV test results as

reflected by the Nagelkerke R² value = 0.19 in Table 19.

Table 19: Logistic regression model summary

Step -2 Log Likelihood Cox & Snell R Square

Nagelkerke R

Square

1 62.73 0.14 0.19

Table 20: Lymphoma cases associated with EBV in the study

Different types of lymphoma were encountered in the study. EBV ISH was performed in

certain lymphomas.

Lymphoma

subtype

Number of

lymphoma

cases

EBV ISH +ve EBV ISH -ve EBV ISH not

performed

Diffuse Large

B-cell

Lymphoma

9 4 1 4

Peripheral T-

cell

Lymphoma

1 - -

Plasmablastic

Lymphoma

11 5 2 4

Burkitt

Lymphoma

Total: 22 9 2 11

Table 21 shows that EBV prevalence was significantly high in patients with lymphoma

and SCC (Chi-square = 11.518, p=0.00).

Table 21: Chi-square tests on EBV and malignant tumors

Value

Asymptotic
Sig.
(2-sided)

Pearson Chi-Square 11.518 2 0.00*

Likelihood Ratio 11.809 2 0.00

N of Valid Cases 53

p<0.05* = statistically significant

Table 22: Omnibus tests of model coefficients

Chi-square df Sig.

Step 1 Step 0.996 1 0.32

Block 0.996 1 0.32

Model 0.996 1 0.32

p<0.05* = statistically significant

As shown in Table 23, 33% of malignant tumors were associated with EBV

Table 23: EBV by tumor types

Tumor types N EBV

Positive None

Malignant 43 33% 67%

Benign 10 - -

Total 53 33% 67%

Table 24 shows that EBV is prevalent in patients with malignant tumors (Chi-square =

4.425, p=0.04).

Table 24: Chi-square tests on EBV and malignant tumors

Value

Asymptotic
Sig.

(2-sided)
Pearson Chi-Square 4.425 1 0.04*

Likelihood Ratio 6.933 1 0.01

N of Valid Cases 53

p<0.05* = statistically significant

EBV according to malignant tumor subtypes

The analysis examined the association between EBV and specific malignant tumor

subtypes (viz. Lymphoma, and SCC ).

Predicting EBV from malignant tumor subtypes

Binomial logistic regression was then performed to determine if EBV could be predicted

from the presenting malignant tumor subtypes. In this instance malignant tumor subtypes

were analysed as the predictor variables and EBV and as shown in the omnibus tests

table, the logistic regression model was statistically significant (Chi-square=7.163,

p=0.01).

Table 25: Omnibus tests of model coefficients

Chi-square df Sig.

Step 1 Step 7.163 1 0.01

Block 7.163 1 0.01

Model 7.163 1 0.01*

p<0.05* = statistically significant

The logistic regression model explained 21% of the variance in EBV across the

malignant tumour subtypes as reflected by the Nagelkerke R² value = 0.21 in Table 26.

Table 26: Logistic regression model summary

Step -2 Log Likelihood Cox & Snell R Square

Nagelkerke R

Square

1 47.104 0.153 0.21

Malignant tumor subtype was thus a significant predictor of EBV in patients (Wald chi-

square 4.975, p=0.03), with a strong association between Lymphoma and positive EBV.

Table 27: Variables in the Equation

B S.E. Wald df Sig. Exp(B)

Malignant tumour

subtype
-1.346 .604 4.975 1 0.03* 0.260

Constant 1.268 .870 2.123 1 0.15 3.553

p<0.05* = statistically significant

4. Discussion

In this study, 53 histopathological reports of sinonasal biopsies of patients presenting at

the CHBAH ORL department between July 2013 and July 2016 were analysed. The

study aimed to evaluate the demographic details of these patients, the spectrum of

sinonasal tumors, concomitant pathology and the presenting symptoms of benign and

malignant tumors. 68% of the cases were male and 32% were female of age range 30-49

years. Malignant tumors were found to be the highest in males (89%) compared to

females (65%) (p=0.04). The most common site of presentation was the nasal cavity

followed by the maxillary sinus. The most common tumors were malignant tumors

(81%). Lymphoma was found to be the most prevalent tumor subtype (45%), which is

contrary to what has been previously described in the literature. The squamous cell

carcinoma (SCC) is known in the literature to be the most common sinonasal tumor,

followed by adenoid cystic carcinoma. (4) In this study, however, lymphoma was found

to be the most common sinonasal tumor, followed by the SCC.

The SCC is classified as keratinizing and non-keratinizing. It is then further classified

into SCC variants such as papillary, verrucous, spindle, basaloid, to mention but a few.

The subtypes encountered in this study were basaloid and spindle cell variants.

Treatment for sinonasal SCC varies depending on the stage, patient physical well-being

and comorbidities, tumor type and staging of the tumor. Basic principles are surgical

resection followed by radiotherapy. Recently, endoscopic surgical approaches have

superseded the conventional open approaches for tumor excision. Directed radiotherapy,

such as intensity modulated radiation therapy (IMRT) and gamma knife radiotherapy

have improved treatment outcomes and decreased morbidity, especially for those patients

with tumors adjacent to vital structures such as the orbit, skull base, and brain. For

advanced stage tumors, especially those that are inoperable, targeted treatments have not

been practiced. Palliative chemotherapy is reserved for advanced irresectable tumors.

(40)

Adenoid cystic carcinoma is the most common salivary gland tumor of the sinonasal tract

and has been described in the literature as the third most common tumor. The current

study demonstrated similar findings. It is a slow growing tumor with a tendency for

recurrence, local and regional spread with a predilection for perineural spread and distant

metastasis. Surgery and postoperative radiation/photon therapy is the mainstay treatment,

however, it is associated with high recurrence rate. (41) Sinonasal melanoma is

described in the literature as a rare mucosal melanoma accounting for less than 1% of the

malignant sinonasal tumors. It is seen in adult patients with the average age of 64. (22)

In this study, this tumor was found in a young 27-year-old female. Its presentation

demonstrated malignant aggression with skull base erosion. Vascular tumors in this

study constituted 4% of the sinonasal tumors. These were the lobular capillary

haemangioma and cavenous hemangioma. The cavenous hemangioma was associated

with a unilateral choanal atresia on a CT scan. Interestingly, this was a finding in a 49

year old patient.

The alveolar rhabdomyosarcoma is a rare mesenchymal tumor involving skeletal

muscles, often seen in children. In this study it was found in a 53 year old male patient.

Overall, this tumor constituted 2% of the cases in this review. Chondrosarcoma is an

uncommon neoplasm that accounts for between 1% and 4% of all primary bone

neoplasms. The most common origin of this tumor is the sacrum, however, 28% to 36%

are seen in the clival region of the skull. It is very rare in the sinonasal tract. (27) Our

study included a case of low grade chondrosarcoma. Nasopharyngeal carcinoma was

found in one patient and was associated with EBV. Genetic, environmental and dietary

factors together with Epstein Barr Virus (EBV) infection are implicated as the causative

factors of NPC. (28)

Sinonasal adenocarcinoma (non-intestinal type) in this study constitutes 2% of the tumors

encountered. These tumors are divided into two types: intestinal and non-intestinal

sinonasal adenocarcinoma. These constitute 10-20% of malignant tumors of the

sinonasal tract. Histologically, these resemble adenocarcinoma or adenoma of the

intestines, or exceptionally normal small intestinal mucosa. Spindle cell carcinoma

constitutes about 2% of the sinonasal tumors reviewed in this study. This tumor is a

variant of squamous cell carcinoma and exhibits spindled and pleomorphic tumor cells.

This tumor has commonly been seen in the larynx especially the glottis, then the oral

cavity, it is however very rare in the sinonasal tract with only a few cases having been

reported. (29)

Benign tumors-like lesions such as the inflammatory polyps and lymphoid hyperplasia

constitute 6% of all tumors in this review. These patients presented with mild symptoms

of nasal obstruction, Samter’s triad and otitis media with effusion. The malignant tumors

had some of the severe symptoms and signs. Nasal obstruction, nasal mass and visual

disturbance/proptosis constitute the most common clinical presentations. Other less

common associated clinical symptoms were cranial nerve fallout, facial numbness,

sinusitis, epistaxis and skin involvement. These severe features were mostly associated

with malignant tumors.

The role of concomitant infection was evaluated and HIV was found to be of importance.

In this study, 45% of the patients were HIV reactive. The percentage of cases testing

positive for HIV was higher amongst patients with malignant tumors at 56% compared to

only 10% amongst patients with benign tumours. This means that a higher percentage of

malignant tumors were associated with HIV (p=0.01). This study further demonstrated

that HIV was highly associated with Lymphoma (83%) and SCC (40%) (p=0.00). EBV

was associated with 33% of the malignant tumors (p=0.04). EBV was associated with

50% of the lymphoma cases compared to SCC (10%) (p=0.02). EBV is a gamma-

herpesvirus first discovered by Epstein and Barr over half a century ago with the aid of

electron microscopy. It is known to infect 90–95% of people worldwide. The virus enters

the human body at a young age and remains persistent for years. Transmission occurs

primarily through saliva. Infection occurs first within oropharyngeal epithelial cells

where the IgA:EBV complex is formed at the IgA receptor and enters the cell. (41) The

nearby circulating B cells that come close to the vicinity of the infected epithelial cells

become infected through viral attachment to CD21 and then enter into the nucleus. (42)

Once it has gained nuclear entry, the EBV viral genome becomes circular and starts to

express EBV nuclear antigen leader protein (EBNA-LP) and EBNA-2. Within 24–48

hours additional EBNAs and latent membrane proteins (LMPs) are produced, leading to

cell growth, transformation and suppression of apoptosis through increase in BCL2

expression. Full expression of all EBV proteins transforms the naive B cells into

activated lymphoblasts. EBV-infected lymphoblasts continue to differentiate within the

germinal center into memory B cells, the reservoir of EBV infection. For the duration of

latent infection, EBV expresses nuclear antigens, membrane proteins, small non-coding

RNA and cell transcripts that contribute to genomic maintenance and evasion of host

immunosurveillance. EBV lymphomagenesis is dependent on expression of viral gene

products that may inhibit apoptosis and promote proliferation via MYC activation or

through inhibition of tumor suppressor genes.(43)

EBV is not only associated with malignancy, but there are certain benign syndromes

which are linked to this virus. Infectious mononucleosis is one such syndrome, which

affects adolescents and adults. Symptoms manifest as fever, lymph node enlargement,

and inflammation of the pharynx. Enlargement of the liver and spleen, and, associated

petechial lesions of the palate are present in most of the patients. Other associated less

common complications include blood dyscrasias, inflammation of the myocardium, liver

dysfunction, ulcers of the genitalia, splenic rupture and neurological complications such

as Guillain-Barre, encephalitis and meningitis. Chronic EBV infection is another

syndrome associated with EBV. It is characterized by severe illness of more than six

months that starts as a primary EBV infection with abnormal EBV antibody titers,

presence of organ disease evidenced by a confirming histologic report, such as hepatitis,

pneumonitis, bone marrow hypoplasia or uveitis. This is accompanied by demonstration

of EBV antigens or DNA tissue. (48) X-linked lymphoproliferative disease is also a

syndrome associated with EBV. It is an inherited disease of males, who are unable to

control infection with EBV. The gene found on the X chromosome known as SAP

(signaling lymphocyte activation molecule associated protein), encodes a protein located

on the surface of Tcells that interacts with proteins on the surface of Tcells and natural

killer cells. When this gene is absent in these patients results in impairment of normal

interaction of T and B cells which results in unregulated growth of EBV infected cells.

(48)

EBV was the first virus associated with the development of malignant lymphoma, and

has been reported to play a role in the development of a variety of HIV-related

lymphomas. According to a different series, EBV has been identified in the HIV positive

setting in 30–90% of DLBCL, between30–40% in Burkit Lymphoma, 70–80% in

Plasmablastic Lymphoma, and almost 100% Hodgkin Lymphoma. The lymphoma

subtypes observed in the HIV positive setting can be grouped in 3 different categories:

1. Immunocompetent patients with lymphoma

a. Burkitt lymphoma

b. Diffuse large B-cell lymphoma

c. Classical Hodgkin lymphoma

d. MALT lymphoma

e. Peripheral T-cell lymphoma

2. Lymphoma associated with HIV

a. Primary effusion lymphoma

b. Plasmablastic lymphoma

c. HHV8 positive diffuse large B-cell lymphoma, not

otherwise specified

3. Lymphoma associated with other immunodeficiency

syndromes :

- Polymorphic B cell lymphoma(PTLD-like)

Plasmablastic lymphoma was the commonest lymphoma encountered in our study at

CHBAH. It was first described in 1997 as a HIV related lymphoma affecting young male

patients involving the jaw and oral cavity. Subsequent studies have also shown its

association with other forms of immunosuppression such as organ transplant

immunosuppression. It is known to be an aggressive form of non-Hodgkins lymphoma

with diffuse proloferation of malignant clones resmbling immunoblasts, but have plasma

cell immunophenotype with associated absence of B-cell antigens. HIV associated

plasmablastic lymphoma is an AIDS defining illness. (44) EBV is positive in 70-80% of

the cases and our study depicted a similar finding.

Diffuse large B-cell lymphoma (DLBCL) is a subtype of high grade B cell non Hodgkins

lymphoma which was the second most commonly encountered lymphoma in our study.

DLBCL has various subtypes such as the central nervous system lymphoma affecting the

brain, and primary mediastinal B cell lymphoma occurring in the chest typically in young

patients. Most cases of DLBCL do not fall in any category and are termed not otherwise

specified (DLBCL-NOS). Diagnosis requires tissue biopsy and imaging for staging.

Bone marrow may also be included and, a lumbar puncture is often performed to

determine the presence of cancer cells in the brain and spine. DLBCL is treated with a

combination of chemotherapy and the monoclonal antibody rituximab. The treatment

regimen is referred to as R-CHOP ( rituximab, cyclophosphamide, doxorubicin,

vincristine, and prednisone). (45)

Burkitt lymphoma was found in our study. It is a high grade malignant lymphoma of

small, noncleaved B cells. In central Africa, Burkitt's lymphoma is associated with

Plasmodium falciparum malaria. These tumors usually present in the jaw, and, over 90%

of the cases are associated with EBV. Infection with malaria is associated with

diminished ability of the T-cell to control of proliferating EBV infected B-cells and,

enhance their proliferation. The cells contain a chromosomal translocation involving

chromosomes 8 and 14, 22, or 2. The translocations result in the positioning of the c-myc

oncogene (chromosome 8) near the immunoglobulin heavy chain (chromosome 14) or

light-chain (chromosome 2 or 22) constant region, leading to abnormal regulation of the

c-myc gene. This results in increased tumorigenicity. High titers of antibody to EBV

structural proteins are associated with high risk for Burkitt's lymphoma (46)

Lymphoblastic lymphoma, which was seen in this review, is a rare lymphoma and

almost 100% of the cases were found to be associated with EBV infection. It is a type of

non-Hodgkin lymphoma associated with immunosuppression due to exposure to

pesticides or radiation. It arises from immature T cells in the majority of cases and

immature B cells in the remainder of cases. It can affect individuals who are HIV

negative who have immunosuppression secondary to organ transplants. (30,31) The

associated concomitant pathology namely the bacterial colony was insignificant,

however, it may have contributed to sinusitis as the presenting symptom in some cases.

EBV is implicated in the pathogenesis of other tumors such as the lymphomatoid

granulomatosis, central nervous system lymphomas, nasal T-cell/ natural killer cell

lymphomas, smooth cell tumors in immunocompromised patients and gastric carcinomas.

EBV and HIV

A vast association between HIV and EBV was noted in this review, and it is the author’s

opinion that these viruses seem to support each other in some way that is not currently

understood. T cells from patients with AIDS suppress EBV-infected B cells less

effectively than do cells from normal controls and results proliferation of EBV infected

cells. EBV is found in substantial amounts in oropharyngeal secretions of patients with

HIV, and this is also associated with high antibody titers. EBV associated non-

Hodgkins’s lymphoma in HIV patients is associated with a reduction in EBV-specific

cytotoxic T cells and an elevated EBV viral load (49,51)

EBV diagnostics

The diagnostic strategies used to detect EBV differ between immunocompromised and

immunocompetent persons as a result of distinct therapeutic intervention required.

Serology offers a dependable criteria for interpretation of results despite a high degree of

variability with EBV. EBV genome encodes a number of different unique genes. The

important ones are the viral capsid antigens (VCA), the early antigens (Eas), and the

EBNAs: EBNA-1 and EBNA-2. There are only three essential serological markers for

detection of EBV namely VCA immunoglobulin G (IgG), VCA IgM, and EBNA-1 IgG.

Interpretation of serological findings should allow EBV infection stage specific

diagnoses. This means that diagnosis should correlate with the clinical picture. In

immunocompetent persons, three diagnosis are relevant: primary or acute infection

(mononucleosis), a past infection that excludes monocucleosis, and the absence of EBV

specific antibodies. In the case of positive results for VCA IgG and EBNA-1 IgG and in

the absence of VCA IgM, a past infection is considered. If the results for VCA IgG,

VCA IgM and EBNA-1 are negative, the patient is EBV susceptible. If tests for VCA

IgM and VCA IgG are positive and those for EBNA-1 are negative then acute or primary

infection can be considered to be present. EBER in situ hybridization (EBRER-ISH) has

been promoted as the most successful test in picking up and localizing latent EBV in

tissue samples (50). This test was used in our study to detect presence of EBV in the

selected submitted specimens. It was an essential diagnostic test especially in tumors that

were borderline such as the plasma cell neoplasm which overlaps with plasmablastic

lymphoma, where a considered differential was multiple myeloma, plasmacytoma or

plasmablastic lymphoma. The detection of EBV using EBER-ISH , tumor morphology

and high proliferation index of the tumor, in context of HIV infection, leads to the

confirmation of plasmablastic lymphoma.

5. Current application

This review has revealed how HIV has influenced the sinonasal disease spectrum and

completely changed what has been described in the literature for many years, this being

the fact that SCC is the most common malignant sinonasal tumor and, osteoma the most

common benign tumor of the sinonasal tract. Researchers and clinicians can now

approach sinonasal pathology with knowledge in mind that the most common pathology

is a malignat tumor and this should allude the treating clinician of the need to test for

HIV as the strong association between malignant tumors and HIV has been demonstrated.

The clinician should also treat this pathology with some urgency because of the nature of

progression of the disease and the tendency to spread to essential adjacent structures such

as the orbit, skull base and cranial nerves.

6. Limitations of the current study

As predicted, there were limitations with regards to certain patients whose HIV status

was unknown and records were not found. Another limitation was the fact that this was a

retrospective study, the researchers were unable to assess the outcomes in the patients

selected for the study. Presenting symptoms were not specified in the request forms and

patients were not interviewed with regards to these. Due to financial factors, EBV status

was not routinely assessed in all tumors.

7. Conclusion

Sinonasal tumors in adult patients at the CHBAH ORL Department are a very diverse

spectrum and demonstrate very interesting characteristics. Male patients younger than 50

years of age tend to have more aggressive malignant tumors and these tumors are noted

to be lymphoma and SCC. These malignant tumors showed a strong association with

HIV and EBV. The most common site for the presentation of these tumours was the

nasal cavity and the maxillary sinus respectively. As a result, the most common

presenting symptoms were a nasal mass and nasal obstruction. The sinonasal tumors

known to affect older patients, such as melanoma were seen to affect even younger

patients in this study.

Expanding on this research, it would be interesting to evaluate if ARV treatment

influences the disease outcome in terms of tumor aggression as compared to patients who

are not initiated on HAART. It is absolutely essential for primary health care clinicians

to refer patients with nasal obstuction promptly to prevent delay in diagnosis of such

aggressive nasal tumors as seen in this study.

For the practicing ORL specialist, it is worth remembering that even the rarest tumors

occuring in other sites can be found in the sinonasal tract. Indeed, a varying degree

disease spectum in CHBH of sinonasal tumors was demonstrated in this study.

Appendix A : Data collection sheet

Hospital No:

Study No:

Male ☐

Female ☐

Age

HIV :

Reactive ☐

Non reactive ☐

CD4 Count:

Date of CD4 count:

Nasal symptoms

Tumor site

Gross appearance

Date of biopsy:

Histological findings:

Tumor subtype:

Concomitant pathology:

Appendix B: Ethics clearance certificate

Appendix C: Turn it in receipt

References

1. Barnes L, Eveson JW, Reichart P, Sidransky D, editors. World Health Organization

classification of tumours. Pathology and genetics of head and neck tumours. Lyon:

IARC Press; 2005. p. 15-17.

2. Zimmer LA, Carrau RL. Neoplasms of the nose and paranasal sinuses. In: Bailey BJ,

Johnson JT, Newlands SD, editors. Head and neck surgery – Otolaryngology. 4th ed.

Philadelphia: Lippincott Williams & Wilkins; 2006. p. 1481-1500.

3. Bridger GP, Mendelsohn MS, Baldwin M, Smee R. Paranasal sinus cancer. Aust N Z

J Surg. 1991;61(4):290-294.

4. Weymuller EA, Gal TJ. Neoplasms of the nasal cavity. In: Cummings CW, Flint PW,

Harker LA, editors. Otolaryngology – Head and neck surgery. 4th ed. Place of

Publication: Mosby; 2005.

5. Robin PE, Powell DJ, Stansbie JM. Carcinoma of the nasal cavity and paranasal

sinuses: Incidence and presentation of different histological types. Clin Otolaryngol

Allied Sci. 1979;4(6):431-456.

6. Caplan LS, Hall HI, Levine RS, Zhu K. Preventable risk factors for nasal cancer. Ann

Epidemiol. 2000;10(3):186-191.

7. Al-Mujaini A, Wali U, Alkhabori M. Functional endoscopic sinus surgery:

Indications and complications in the ophthalmic field. Oman Med J. 2009;24(2):70-

80.

8. d’Errico A, Pasian S, Baratti A, Zanelli R, Alfonzo S, Gilardi L, et al. A case-control

study on occupational risk factors for sino-nasal cancer. Occup Environ Med.

2009;66(7):448-455.

9. Cakmak O, Yavuz H, Yucel T. Nasal and paranasal sinus schwannomas. Eur Arch

Otorhinolaryngoly. 2003;260(4):195-197.

10. Lloyd C, McHugh K. The role of radiology in head and neck tumours in children.

Cancer Imaging. 2010;10(1):49-61.

11. Erdogan N, Demir U, Songu M, Ozenler NK, Uluç E, Dirim B. A prospective study

of paranasal sinus osteomas in 1,889 cases: Changing patterns of localization.

Laryngoscope. 2009;119(12):2355-2359.

12. Alawi F. Benign fibro-osseous diseases of the maxillofacial bones. A review and

differential diagnosis. Am J Clin Pathol. 2002;118 Suppl:S50-70.

13. Llorente JL, López F, Suárez C, Hermsen MA. Sinonasal carcinoma: Clinical,

pathological, genetic and therapeutic advances. Nat Rev Clin Oncol. 2014;11(8):460-

472.

14. Frierson HF Jr, Mills SE, Fechner RE, Taxy JB, Levine PA. Sinonasal

undifferentiated carcinoma: An aggressive neoplasm derived from Schneiderian

epithelium and distinct from olfactory neuroblastoma. Am J Surg Pathol.

1986;10(11):771-779.

15. Husain Q, Kanumuri VV, Svider PF, Radvansky BM, Boghani Z, Liu JK, et al.

Sinonasal adenoid cystic carcinoma: Systematic review of survival and treatment

strategies. Otolaryngol Head Neck Surg. 2013;148(1):29-39.

16. Shohat I, Berkowicz M, Dori S, Horowitz Z, Wolf M, Taicher S, et al. Primary non-

Hodgkin’s lymphoma of the sinonasal tract. Oral Surg Oral Med Oral Pathol Oral

Radiol Endod. 2004;97(3):328-331.

17. Bak M, Wein RO. Esthesioneuroblastoma: A contemporary review of diagnosis and

management. Hematol Oncol Clin North Am. 2012;26(6):1185-1207.

18. Razek AA, Sieza S, Maha B. Assessment of nasal and paranasal sinus masses by

diffusion-weighted MR imaging. J. Neuroradiol. 2009;36(4):206-211.

19. Sasaki M, Eida S, Sumi M, Nakamura T. Apparent diffusion coefficient mapping for

sinonasal diseases: Differentiation of benign and malignant lesions. AJNR Am J

Neuroradiol. 2011;32(6):1100-1106.

20. Wang X, Zhang Z, Chen Q, Li J, Xian J. Effectiveness of 3 T PROPELLER DUO

diffusion-weighted MRI in differentiating sinonasal lymphomas and carcinomas. Clin

Radiol. 2014;69(11):1149-1156.

21. Eggesbø HB. Imaging of sinonasal tumours. Cancer Imaging. 2012;136-152.

22. Clifton N, Harrison L, Bradley PJ, Jones NS. Malignant melanoma of nasal cavity

and paranasal sinuses: Report of 24 patients and literature review. J Laryngol Otol.

2011;125(5):479-485.

23. Wu AW, Suh JD, Metson R, Wang MB. Prognostic factors in sinonasal sarcomas:

Analysis of the surveillance, epidemiology and end result database. Laryngoscope.

2012;122(10):2137-2142.

24. Lathi A, Syed MM, Kalakoti P, Qutub D, Kishve SP. Clinico-pathological profile of

sinonasal masses: A study from a tertiary care hospital of India. Acta

Otorhinolaryngol Ital. 2011;31(6):372-377.

25. Jackson RT, Fitz-Hugh GS, Constable WC. Malignant neoplasms of the nasal cavities

and paranasal sinuses: (A retrospective study). Laryngoscope. 1977;87(5Pt 1):726-

736.

26. Shirazi N, Bist SS, Selvi TN, Harsh M. Spectrum of sinonasal tumours: A 10-year

experience at a tertiary care hospital in North India. Oman Med J. 2015;30(6):435-

440.

27. Heffelfinger MJ, Dahlin DC, MacCarty CS, Beabout JW. Chordomas and

cartilaginous tumors at the skull base. Cancer. 1973;32(2):410-420.

28. Van Hasselt CA, Gibb AG, editors. Nasopharyngeal carcinoma. Hong Kong: Chinese

University Press; 1991.

29. Thompson LD, Wieneke JA, Miettinen M, Heffner DK. Spindle cell (sarcomatoid)

carcinomas of the larynx: A clinicopathologic study of 187 cases. AM J Surg Pathol.

2002;26(2):153-170.

30. Au WY, Pang A, Choy C, Chim CS, Kwong YL. Quantification of circulating

Epstein-Barr virus (EBV) DNA in the diagnosis and monitoring of natural killer cell

and EBV-positive lymphomas in immunocompetent patients. Blood.

2004;104(1):243-249.

31. Borenstein J, Pezzella F, Gatter KC. Plasmablastic lymphomas may occur as post-

transplant lymphoproliferative disorders. Histopathology. 2007;51(6):774-777.

32. Hopkin N, McNicoll W, Dalley VM, Shaw HJ. Cancer of the paranasal sinuses and

nasal cavities. Part I. Clinical features. J Laryngol Otol. 1984;98(6):585-595.

33. McNicoll W, Hopkin N, Dalley VM, Shaw HJ. Cancer of the paranasal sinuses and

nasal cavities. Part II. Results of treatment. J Laryngol Otol. 1984;98(7):707-718.

34. Lund VJ. Diagnosis and treatment of nasal polyps. BMJ. 1995;311(7017):1411-1414.

35. Mansell NJ, Bates GJ. The inverted Schneiderian papilloma: A review and literature

report of 43 new cases. Rhinology. 2000;38(3):97-101.

36. Svane-Knudsen V, Jørgensen KE, Hansen O, Lindgren A, Marker P. Cancer of the

nasal cavity and paranasal sinuses: A series of 115 patients. Rhinology.

1998;36(1):12-14.

37. Deschler DG, Moore MG, Smith RV, editors. Quick reference guide to TNM staging

of head and neck cancer and neck dissection classification. 4th ed. Alexandria, VA:

American Academy of Otolaryngology – Head and Neck Surgery Foundation; 2014.

38. Drake RL, Vogl W, Mitchell AWM, Gray H. Gray’s anatomy for students.

Philadelphia: Churchill Livingstone Elsevier; 2010. p. 1013-1029.

39. Lewis JS Jr. Sinonasal squamous cell carcinoma: A review with emphasis on

emerging histologic subtypes and the role of human papillomavirus. Head Neck

Pathol. 2016;10(1):60-67.

40. Naficy S, Disher MJ, Esclamado RM. Adenoid cystic carcinoma of the paranasal

sinuses. Am J Rhinol. 1999;13(4):311-314.

41. Sixbey JW, Nedrud JG, Raab-Traub N, Hanes RA, Pagano JS. Epstein-Barr virus

replication in oropharyngeal epithelial cells. N Engl J Med. 1984;310(19):1225-1230.

42. Nemerow GR, Wolfert R, McNaughton ME, Cooper NR. Identification and

characterization of the Epstein-Barr virus receptor on human B lymphocytes and its

relationship to the C3d complement receptor (CR2). J Virol. 1985;55(2):347-351.

43. Hemann MT, Bric A, Teruya-Feldstein J, Herbst A, Nilsson JA, Cordon-Cardo C, et

al. Evasion of the p53 tumour surveillance network by tumour-derived MYC mutants.

Nature. 2005;436(7052):807-811.

44. Carbone A, Gloghini A. Plasmablastic lymphoma: One or more entities? Am J

Hematol. 2008;83(10):763-764.

45. Sehn LH, Donaldson J, Chhanabhai M, Fitzgerald C, Gill K, Klasa R, et al.

Introduction of combined CHOP plus rituximab therapy dramatically improved

outcome of diffuse large B-cell lymphoma in British Colombia. J Clin Oncol.

2005;23(22):5027-5032.

46. Inghirami G, Grignani F, Sternas L, Lombardi L, Knowles DM, Dalla-Favera R.

Down-regulation of LFA-1 adhesion receptors by C-myc oncogene in human B

lymphoblastoid cells. Science. 1990;250(4981):682-686.

47. Straus SE. The chronic mononucleosis syndrome. J Infect Dis. 1988;157(3):405-412.

48. Sayos J, Wu C, Morra M, Wang N, Zhang X, Allen D, et al. The X-linked

lymphoproliferative-disease gene product SAP regulates signals induced through the

co-receptor SLAM. Nature. 1998;395(6701):462-469.

49. Jenson H, McIntosh K, Pitt J, Husak S, Tan M, Bryson Y, et al. Natural history of

primary Epstein-Barr virus infection in children of mothers infected with human

immunodeficiency virus type 1. J Infect Dis. 1999;179(6):1395-1404.

50. Gärtner BC, Hess RD, Bandt D, Kruse A, Rethwilm A, Roemer K, et al. Evaluation

of four commercially available Epstein-Barr virus enzyme immunoassays with an

immunofluorescence assay as the reference method. Clin Diagn Lab Immunol.

2003;10(1):78-82.

51. Linke-Serinsöz E, Fend F, Quintanilla-Martinez L. Human immunodeficiency virus

(HIV) and Epstein-Barr virus (EBV) related lymphomas, pathology view point.

Semin Diagn Pathol. 2017;34(4):352-363.

52. Drake RL, Wayne AV, Mitchell AWM. Gray’s Anatomy for students. Philadelphia:

Churchill Livingstone Elsevier; 2010. P 1013-1029.

53. Wormald PJ, Wemer H, Claudio C, et al. The international frontal sinus anatomy

classification and classification of the extent of endoscopic frontal sinus surgery. Int

Forum Allergy Rhinol. 2016 Jul;6(7):677-96

54. Stammberger HR, Kennedy DW (1995) Paranasal sinuses: anatomic terminology and

nomenclature. The Anatomic Terminology Group. Ann Otol Rhinol Laryngol Suppl

Oct;167:7–16

55. Polavaram R, Devaiah AK, Sakai O.(2004)Anatomic variants and pearls functional

endoscopic sinus surgery. Otolaryngol Clin North Am 2004;37(2):221–42.

56. Kantarci M, Karasen RM. Remarkable anatomic variations in paranasal sinus region

and their clinical importance. Eur J Radiol 2004;50(3):296–302.