EXPOSURE TO LOUD NOISE AND RISK OF ACOUSTIC NEUROMA

DISSERTATION

Presented in Partial Fulfillment of the Requirements for

the Degree Doctor of Philosophy in the Graduate

School of The Ohio State University

By

Colin G. Edwards, M.S.

* * * * *

The Ohio State University

2007

Dissertation Committee: Approved by

Professor Randall A. Harris, Co-Adviser

Professor Judith A. Schwartzbaum, Co-Adviser

Professor John M. Crawford Co-Advisers Graduate Program in Public Health

Copyright by Colin G. Edwards

2007

ii

ABSTRACT This dissertation presents three related investigations that evaluate the previously

reported association between loud noise exposure and the risk of acoustic neuromas, as

well as the proposed biological basis for the association. The goal of the first

investigation was to examine further the role of loud noise in acoustic neuroma etiology.

In a population-based case-control study conducted from 1999-2002 in Sweden, reports

of occupational and nonoccupational loud noise exposure of 146 acoustic neuroma cases

and 564 controls were compared. Individuals reporting loud noise exposure from any

source were found to be at increased risk for acoustic neuromas. The findings of an

increased risk of acoustic neuromas with loud noise exposure support previous research.

The goal of the second investigation was to further examine the association

between noise exposure and acoustic neuroma using an objective measure of exposure. A

total of 793 acoustic neuroma cases were identified between 1987 and 1999 from the

Swedish Cancer Registry, to which 101,756 controls were frequency matched.

Occupational information, available for 599 of the cases and 73,432 of the controls, was

obtained from censuses and linked to a job exposure matrix. Of three studies of noise

exposure and acoustic neuroma risk to date, this is the first to use a job exposure matrix

iii

to assess exposure. Contrary to previous study results the findings do not demonstrate an

increased acoustic neuroma risk related to occupational noise exposure.

A mechanism of acoustic neuroma tumorigenesis during the cellular repair

process following acoustic trauma has been proposed, whereby cellular division results in

DNA replication errors which may in turn lead to chromosomal changes essential for

neoplastic transformation. In the third investigation, an extensive literature search was

conducted to evaluate the biological plausibility of this hypothesis. The tumor typically

involves the vestibular rather than the acoustic division of the eighth nerve, however

intralabyrinthine and cochlear nerve schwannomas have been reported. Additionally,

evidence of vestibular damage in rodents has been demonstrated following acoustic

trauma. The proposed hypothesis is therefore plausible, however further research is

needed to elucidate the precise biological basis for the association between loud noise

and acoustic neuromas.

iv

Dedicated to my wife and children

v

ACKNOWLEDGMENTS

I wish to thank Professor Judith Schwartzbaum for her support, enthusiasm, and

knowledge of epidemiologic methods, all of which were invaluable during my entire

doctoral training. I also thank her for providing me with the opportunity to collaborate

with researchers at the Karolinska Institute during my doctoral dissertation research.

Additional thanks go to Professor Randall Harris for his advice in the field of cancer

epidemiology and for his encouragement at the time of my doctoral candidacy exam and

during the completion of my dissertation. Many thanks are also due to Professor John

Crawford for his guidance, encouragement, and friendship.

In addition, I extend gratitude to Professors Maria Feychting and Anders Ahlbom

at the Institute of Environmental Medicine at the Karolinska Institute in Stockholm, for

twice inviting me to the Karolinska Institute to obtain and begin the analysis of my two

dissertation research data sets. I also thank Drs. Stefan Lönn and Ulla Forssén, former

doctoral students at the Karolinska Institute, for their guidance and advice. The advice

and support of Professor Thomas Prior is also appreciated. Thanks also go to Professor

Clive Edwards for his insightful comments on the final draft of my dissertation and to

Ms. Lynn Higginbotham for her administrative support.

Lastly, but by no means least, my immeasurable thanks to my wife Anne for her

unwavering support, understanding, and patience during the almost eight years of my

vi

part-time doctoral studies. Without her support for my trips to Sweden and my late nights

at the library, my doctoral studies and this dissertation research would not have been

possible.

vii

VITA

1987 – 1992 B.S. Microbiology, The Ohio State University

1993 – 1995 M.S. Preventive Medicine,

The Ohio State University

1995 – 2005 Research Associate, Comprehensive Cancer Center, The Ohio State University

2005 – present Senior Research Specialist, Comprehensive

Cancer Center, The Ohio State University

PUBLICATIONS 1. Edwards CG, Schwartzbaum JA, Lönn S, Ahlbom A and Feychting M: Exposure to Loud Noise and Risk of Acoustic Neuroma. American Journal of Epidemiology 163(4):327-333, 2006. 2. Marcucci G, Mrózek K, Ruppert AS, Maharry K, Kolitz JE, Moore JO, Mayer RJ, Pettenati MJ, Powell BL, Edwards CG, Sterling LJ, Vardiman JW, Schiffer CA, Carroll AJ, Larson RA, Bloomfield CD. Prognostic factors and outcome of core binding factor acute myeloid leukemia patients with t(8;21) differ from those of patients with inv(16): a Cancer and Leukemia Group B study. Journal of Clinical Oncology 23(24):5705-17, 2005. 3. Marcucci G, Baldus CD, Ruppert AS, Radmacher MD, Mrózek K, Whitman SP, Kolitz JE, Edwards CG, Vardiman JW, Powell BL, Baer MR, Moore JO, Perrotti D, Caligiuri MA, Carroll AJ, Larson RA, de la Chapelle A and Bloomfield CD. Overexpression of the ETS Transcription Factor ERG Predicts a Worse Outcome in Acute Myeloid Leukemia with Normal Karyotype: A Cancer and Leukemia Group B Study, Journal of Clinical Oncology 23(36):9234-42, 2005.

viii

4. Bloomfield CD, Ruppert AS, Mrózek K, Kolitz JE, Moore JO, Mayer RJ, Edwards CG, Sterling LJ, Vardiman JW, Carroll AJ, Pettenati MJ, Stamberg J, Byrd JC, Marcucci G, Larson RA. Core binding factor acute myeloid leukemia. Cancer and Leukemia Group B (CALGB) Study 8461. Annals of Hematology 83(Suppl 1): S84-5, 2004. 5. Marcucci G, Mrózek K, Ruppert AS, Archer KJ, Pettenati MJ, Heerema NA, Carroll AJ, Koduru PRK, Kolitz JE, Sterling LJ, Edwards CG, Anastasi J, Larson RA, Bloomfield CD: Abnormal Cytogenetics at Date Of Morphologic Complete Remission Predicts Short Overall and Disease-Free Survival and Higher Relapse Rate in Adult Acute Myeloid Leukemia: Results From Cancer and Leukemia Group B (CALGB) 8461. Journal of Clinical Oncology 22(12):2410-18, 2004. 6. Byrd, JC, Rupert A, Mrózek, K, Carroll AJ, Edwards CG, Arthur DC, Pettenati MJ, Stamberg J, Koduru PRK, Moore JO, Mayer RJ, Davey FR, Larson RA and Bloomfield CD: Repetitive Cycles of High Dose Cytarabine Benefit Patients with Acute Myeloid Leukemia and inv(16)(p13q22) or t(16;16)(p13;q22): Results from Cancer and Leukemia Group B (CALGB 8461). Journal of Clinical Oncology 22(6):1087-94, 2004. 7. Byrd JC, Mrózek K, Dodge RK, Carroll AJ, Edwards CG, Arthur DC, Pettenati MJ, Patil SR, Rao KW, Watson MS, Moore JO, Stone RM, Mayer RJ, Davey FR, Schiffer CA, Larson RA and Bloomfield CD: Pre-Treatment Cytogenetic Abnormalities are Predictive of Induction Success, Cumulative Incidence of Relapse and Overall Survival in Adult Patients with de novo Acute Myeloid Leukemia: Results from Cancer and Leukemia Group B (CALGB 8461). Blood 100(13): 4325-4336, 2002. 8. Mrózek K, Prior TW, Edwards CG, Marcucci G, Carroll AJ, Snyder PJ, Koduru PRK, Theil KS, Pettenati MJ, Archer KJ, Caligiuri MA, Vardiman JW, Kolitz J, Larson RA and Bloomfield CD: A Comparison of Cytogenetic and Molecular Genetic Detection of t(8;21) and inv(16) in Adults with de novo Acute Myeloid Leukemia: A Cancer and Leukemia Group B Study. Journal of Clinical Oncology 19(9):2482-2492, 2001. 9. Byrd JC, Dodge R, Carroll A, Baer M, Edwards CG, Stamberg J, Qumsiyeh M, Moore JO, Mayer RJ, Davey F, Schiffer CA, Bloomfield CD: Patients with t(8;21)(q22;q22) and Acute Myeloid Leukemia Have Superior Failure-Free and Overall Survival When Repetitive Cycles of High-Dose Cytarabine are Administered. Journal of Clinical Oncology 17(12): 3767-3775, 1999.

ix

FIELDS OF STUDY Major field of study: Public health Area of specialization: Cancer epidemiology Minor field of study: Molecular pathology

x

TABLE OF CONTENTS

Page Abstract ii Dedication iv Acknowledgments v Vita vii List of Tables xiii Chapters: 1. Introduction 1 1.1 Need for the Present Studies and Overview of Acoustic

Neuromas 1 1.2 Sporadic Acoustic Neuromas and Bilateral Neurofibromatosis 4 1.3 Histopathology 7 1.4 Natural History 9 1.5 Clinical Presentation 13 1.6 Diagnostic Modalities 16 1.7 Management 19 1.8 Incidence, Age and Sex Distribution 22 1.9 Risk Factors 24 1.10 References 31 2. Exposure to Loud Noise and Risk of Acoustic Neuroma 37 2.1 Abstract 32 2.2 Introduction 33 2.3 Materials and Methods 34

2.3.1 Study Design and Population 34 2.3.2 Acoustic Neuroma Case Ascertainment 34

xi

2.3.3 Controls 35 2.3.4 Data Collection and Loud Noise Exposure Assessment 36 2.3.5 Statistical Analysis 37

2.4 Results 38 2.5 Discussion 47

2.5.1 Loud Noise Exposure and Risk of Acoustic Neuroma 47 2.5.2 Other Risk Factors 48 2.5.3 Diagnostic Delay 49 2.5.4 Acoustic Trauma and Tumorigenesis 50 2.5.5 Recall and Selection Bias 51 2.5.6 Conclusion 52

2.6 Acknowledgments 52 2.7 References 53

3. Exposure to Occupational Noise and Risk of Acoustic Neuroma 56 3.1 Abstract 56

3.2 Introduction 57 3.3 Materials and Methods 58 3.3.1 Study Design and Population 58 3.3.2 Acoustic Neuroma Case Ascertainment 58 3.3.3 Controls 59 3.3.4 Census Data 59 3.3.5 Job Exposure Matrix 59 3.3.6 Individual Noise Exposure Assessment 60 3.3.7 Statistical Analysis 62 3.4 Results 62 3.5 Discussion 69 3.5.1 Occupational Noise Exposure and Risk of Acoustic

Neuroma 69 3.5.2 Comparison of Results to Prior Study of Self-

Reported Noise Exposure and Potential for Recall Bias 71

3.5.3 Diagnostic Delay and the Healthy Worker Survivor Effect 71

3.5.4 Misclassification of Exposure and Confounding 72 3.5.5 Conclusion 74 3.6 Acknowledgments 74 3.7 References 75 4. Acoustic Neuroma: Histopathology, Natural History, and the Role of

Acoustic Trauma in Tumorigenesis 78

4.1 Abstract 78 4.2 Epidemiology 79

xii

4.3 Genetics 81 4.4 Histopathology 83

4.5 Pathology of the Vestibulocochlear Nerve 85 4.6 Nerve Origin of Acoustic Neuromas 86 4.7 Natural History 87 4.8 Growth Rate 89 4.9 Noise-Induced Mechanical Damage of the Cochlea 94 4.10 Acoustic Trauma in the Pathogenesis of Acoustic Neuroma 95 4.11 Conclusion 97 4.12 References 99

5. Summary and Conclusions 106 5.1 Summary of Key Study Findings 106 5.2 Summary of the Role of Acoustic Trauma in Acoustic

Neuroma Tumorigenesis 107 5.3 Study Limitations and Recommendations for Future Research 108 5.4 References 110 List of References 111

xiii

LIST OF TABLES Table Page 2.1 Selected characteristics of the 710 study participants, Sweden, 1999-

2002 42 2.2 Odds ratios for acoustic neuroma according to loud noise exposure,

Sweden, 1999-2002 43 2.3 Odds ratios for acoustic neuroma according to duration of loud noise

exposure, Sweden, 1999-2002 44 2.4 Odds ratios for acoustic neuroma according to loud noise exposure

type, Sweden, 1999-2002 45 2.5 Odds ratios for acoustic neuroma according to loud noise exposure

and latency period, Sweden, 1999-2002 46 3.1 Selected characteristics of all study participants and of the study

participants with information on exposure, Sweden, 1987-1999 65 3.2 Most frequent occupations with high (≥ 85 dB) loud noise exposure

among cases and controls for men and women, Sweden, 1987-1999 66 3.3 Odds ratios and 95% confidence intervals for acoustic neuroma in

relation to loud noise exposure level according to latency period duration, Sweden, 1987-1999 67

3.4 Odds ratios and 95% confidence intervals for exposure to loud noise

at any time (ever) for one, two, three, or four consecutive censuses closest in time prior to reference year, Sweden, 1987-1999 68

1

CHAPTER 1

INTRODUCTION 1.1 Need for the Present Studies and Overview of Acoustic Neuromas

The incidence of acoustic neuromas has been increasing and this observation

provided motivation for these studies of loud noise exposure as a potential risk factor for

this tumor (3, 54, 57). In addition, previous studies of risk factors for acoustic neuromas

have generally grouped the tumor under the broad heading of “brain tumors” and few

studies have examined acoustic neuromas alone. A study of loud noise exposure as a

potential risk factor for acoustic neuromas was therefore warranted. An association

between occupational loud noise exposure and acoustic neuromas has been demonstrated

in a study of men performed in 1989, however no subsequent studies have attempted to

confirm the association (71). The previous study proposed a hypothesis that mechanical

trauma may contribute to acoustic neuroma tumorigenesis (71). Further examination of

the plausibility of this hypothesis was also of interest as part of this dissertation research.

Additionally, loud noise exposure remains a significant public health problem, especially

with the increase in the use of in-the-ear technology listening devices such as MP3

players, further supporting the need for studies of loud noise exposure and risk of

acoustic neuroma.

2

Brain tumor registry data were not readily available in the United States and

collaboration with researchers at the Institute of Environmental Medicine at the

Karolinska Institute in Stockholm, Sweden provided the opportunity to analyze Swedish

brain tumor data. Swedish Cancer Registry data have high coverage and accuracy, and in

addition data were readily available from the continuously updated Swedish Population

Registry. A large number of cases of acoustic neuroma were identified for the studies

reported in Chapters 2 and 3, and controls were readily available for matching to the

cases. Finally, the previous study of loud noise exposure and risk of acoustic neuroma did

not report findings for nonoccupational exposure and included only men (71). The

Swedish acoustic neuroma data provided both occupational and nonoccupational loud

noise exposure information for both males and females, providing additional motivation

for further study of this potential risk factor.

Acoustic neuroma, also known as vestibular schwannoma, acoustic schwannoma,

or benign neurilemoma, is a benign tumor originating in the region of Scarpa’s ganglion

at the junction of peripheral and central myelin of the of the vestibular division of the

eighth cranial nerve (1-4). The seventh and eighth cranial nerves are housed in the

confined bony structure of the internal auditory canal (2). The slow-growing acoustic

neuroma tumor usually results from the abnormal proliferation of Schwann cells covering

the superior vestibular division or the inferior vestibular or cochlear division of the nerve

as it passes through the internal auditory canal (1, 5). The tumors can also originate

outside the canal (6). Historically, the tumor has been termed “acoustic neuroma”

although the name “vestibular schwannoma” in fact describes the tumor more accurately.

In order to maintain consistency with the majority of the published literature the term

3

“acoustic neuroma” is used throughout this dissertation document. Schwann cells are

named after the German physiologist and histologist Theodor Schwann, 1810-1882, who

was the first to describe the nerve sheath cells (7). These cells are the glial or

“supporting” cells of the peripheral nervous system (8). Acoustic neuromas comprise

approximately 5-10% of all intracranial tumors and account for 71-90% of all tumors

involving the cerebellopontine angle (1, 9). The remainder of cerebellopontine angle

tumors are meningiomas and cholesteatomas (1).

1.2 Sporadic Acoustic Neuromas and Bilateral Neurofibromatosis

Acoustic neuromas present clinically as two distinct types, the bilateral hereditary

type and the unilateral sporadic type (2, 10). The majority of acoustic neuromas are

unilateral, however in approximately 4-5% of cases the tumors are bilateral (9, 11). In the

bilateral expression of the disease, the germ cell experiences a genetic defect, whereas in

the unilateral expression the defect occurs in the somatic cell (10). The hallmark of the

genetic syndrome neurofibromatosis type 2 (NF2) or “central neurofibromatosis” is the

development of bilateral acoustic neuromas (5, 10). In contrast, individuals with

neurofibromatosis type 1 (NF1), also known as “von Recklinghausen’s

neurofibromatosis” or “peripheral neurofibromatosis” rarely develop acoustic neuromas

(4, 10).

Neurofibromatosis types 1 and 2 are characterized by the presence of

neurofibromas and peripheral and central nervous system tumors (4). The incidence of

NF1 is approximately 40 per 100,000 and that for NF2 is 1 per 33,000-40,000 (2, 4). A

deletion of a tumor suppressor gene on the long arm of chromosome 22 has been

4

established as the cause of NF2 and the majority of these patients develop bilateral

acoustic neuromas about 20 years earlier in life than those with unilateral acoustic

neuromas (2, 3, 12). The disorder is inherited in an autosomal dominant manner with

95% penetrance (5, 10). However, in approximately 50% of patients with NF2 no family

history is obtained and these cases are likely the result of new germline mutations (13).

NF2 is also associated with an increased incidence of other intracranial tumors, such as

gliomas and meningiomas, neurofibromas, and facial nerve neuromas (2, 5, 10). The gene

for NF1 has been found on chromosome 17 and these patients rarely develop acoustic

neuromas (4, 10). NF1 is characterized by multiple cutaneous neurofibromas and café-au-

lait spots (13).

Both the sporadic acoustic neuromas and the bilateral hereditary acoustic

neuromas exhibit remarkable genomic stability and are thought to be a result of the

functional loss of the tumor suppressor gene on chromosome 22 (14, 15). The NF2 gene

was mapped to 22q12.2 in 1986 and identical candidate NF2 genes were isolated in 1993

(13, 14). The gene encodes a protein of 595 amino acids named merlin (for moesin-ezrin-

radixin-like protein), also known as schwannomin, and mutation in the NF2 gene causes

loss of expression of this functional tumor suppressor protein (13, 16). It is thought that

inactivating mutations of the NF2 gene may result in a truncated protein product (17).

Classical merlin knockout mice have been shown to develop a variety of tumors,

including highly malignant tumors, but they do not develop schwannomas (18).

The primary NF2 constitutional mutations in NF2 patients are point mutations,

whereas the main causal event in sporadic acoustic neuromas are small deletions (19). It

is likely that the NF2 protein is involved in intracellular signaling, modulation of cell

5

motility, and suppression of mitogenesis in schwannoma cells (13, 20). The mechanism

of tumorigenesis acts in accordance with Knudson’s “two hit” mutation model, identical

to the retinoblastoma model of carcinogenesis involving the 13q chromosomal region

(15, 21). A more complete understanding of other genes whose expression are

deregulated during acoustic neuroma tumorigenesis will provide better understanding of

why NF2 mutation leads to tumor formation (16). The present research on loud noise

exposure as a risk factor for acoustic neuromas includes only patients with unilateral,

sporadic acoustic neuromas.

1.3 Histopathology

The appearance of many pathologic markers in acoustic neuromas has been

described in the literature. These include Luse bodies, fibrous collagen bundles, myelin

sheath irregularities, and subepithelial nerve fiber losses, all observed in the vestibule

(22). The tumors have a remarkably diffuse yellow appearance, a firm consistency, and

when compared to other benign intracranial tumors acoustic neuromas have the least

proliferative status (7, 9, 23). It is the distal part of the eighth nerve, with Schwann cells

enclosing the axons, where an overproliferation of Schwann cells leads to the formation

of an acoustic neuroma (7). Acoustic neuromas can reach sizes of up to several

centimeters in diameter and thus most of the cells comprising a tumor are not adjacent to

the axon (24). As a result, acoustic neuroma cells can survive and proliferate in the

absence of axon-derived growth factors, in a similar manner to mature, denervated

Schwann cells (24). Acoustic neuromas derive their arterial blood supply primarily from

the branches of the basilar arteries, as well as from branches of the vertebral arteries (9).

6

As the tumor grows it follows the direction of least resistance, often medially into the

cerebellopontine angle, at which stage it may be of considerable size (10). As a result, an

acoustic neuroma is often comprised of two parts, the stalk within the internal auditory

canal and the main portion occupying the cerebellopontine angle (10).

Schwann cells are multivalent neuro-ectodermal cells which are considered

homologous to oligodendroglia of the central nervous system, both of which form and

maintain the myelin sheath (25, 26). The cell has a large round or elongated nucleus with

a prominent nucleolus and finely distributed chromatin (25). Surrounded by basement

membrane, the Schwann cell has a histological appearance marked by two prominent

tissue types (25). Type A Antoni cells which are composed of a closely packed, cellular,

fibrillary structure, with small spindle-shaped densely staining nuclei, and type B Antoni

cells which are comprised of a less cellular, loosely arranged, reticular structure (7, 9,

23). The type B Antoni cells may contain mucineous and microcystic changes (27).

Approximately 15 percent of tumors will contain cysts, characteristic of cystic acoustic

neuromas (27). The type A Antoni cells are intermingled with the type B cells, however

the two cells types are generally rather well demarcated (9). The transition from type A

tissue to type B tissue, or vice versa, can be abrupt or continuous (9).

The vestibular and cochlear nerves transmit sensory input regarding balance and

hearing, respectively (27). The vestibulocochlear nerve measures approximately 20 mm

in length from the nerve root entry zone of the brain stem to the labyrinth end organ (10).

The vestibular nerve has an embryologically disordered arrangement of sheath cells that

predisposes it to develop schwannomas (28). The cochlear fibers are smaller, greater in

number, and darker staining than the vestibular fibers (29). However, the plane between

7

the cochlear and vestibular fibers is often not clear (29, 30). The fibers of both nerves

originate in separate end organs and have separate central connections, however they do

travel as one nerve through the posterior cranial fossa and internal auditory canal (31).

The capsule typically divides the tumor from the nerves (29).

Finally, inflammation is a recurrent hallmark of acoustic neuromas on histological

sections, especially in Antoni type B areas, and inflammatory components have been

shown to influence benign tumor size and growth (32). In one study the degree of

inflammation increased with time, which is consistent with the concept that inflammation

is a degenerative change (32).

1.4 Natural History

The natural history of untreated acoustic neuromas is unpredictable (2, 78). In a

study of in vivo and in vitro growth models, Charabi states that any attempt to elucidate

the natural history of acoustic neuromas must have as its basis a clear understanding of

the environment in which the disease develops (7). This includes an understanding of the

normal anatomy and histology of the vestibulocochlear nerve, growth factors, histological

changes, and factors that determine rates and patterns of tumor growth (7). It has been

well documented in the literature that some tumors grow, some regress, some do not

grow at all, and some display a variable pattern of growth (33, 34).

Serial radiographic studies on patients who are not surgical candidates or who

refuse surgery have provided some insight into the natural history of tumor expansion

(10, 35). However, factors such as loss of inhibitor regulation, local growth factors, and

other cell cycle regulators that may help predict tumor growth have not yet been

8

identified (2, 10). Several growth factors, including nerve growth factor, glial growth

factor, platelet-derived growth factor, basic fibroblast growth factor, transforming growth

factor-β, epidermal growth factor, and vascular endothelial growth factor have been

proposed as potentially playing a role in the pathogenesis of acoustic neuromas, but all

require further evaluation (7, 8, 16, 21, 24, 36). In addition, no consistent correlation has

been found between growth rate of acoustic neuromas and patient age or tumor size (10,

37). Studies using monoclonal antibody assays have, however, shown that clinical growth

rates will correlate with the fraction of cells found in the proliferative phase of the cell

cycle (10). Studies using the monoclonal antibody Ki-67, proliferating cell nuclear

antigen (PCNA), and DNA flow cytometric analysis have been used to examine the

growth rate and proliferative potential of acoustic neuromas (7, 38).

Some studies have shown that tumor growth rate at the start of follow-up is a

good predictor of future tumor growth rate (2, 35). Many studies have shown that the rate

of tumor growth in a given patient is consistent over the course of the disease and is

established during the first one to three years of observation (1, 2, 35). However, serial

imaging follow-up should not be terminated based solely on tumor quiescence, as some

acoustic neuroma patients do experience a delayed onset of tumor growth (39).

It has been shown that tumors larger than 20 mm are statistically more likely to

grow (2). Another study found a statistically significant inverse relationship between

tumor size and patient age (40). Other studies have reported a tendency for higher growth

rate in larger tumors and in younger patients (41, 42). Nevertheless, it is not possible to

demonstrate any pathoanatomic feature of acoustic neuroma that correlates with the

clinical course of the tumor (43, 44).

9

Studies have examined the proportion of acoustic neuroma patients who

experienced tumor growth. Growth varied widely from 15-90% of the cases followed (34,

41, 45). In studies examining the growth rate of acoustic neuromas, the mean annual

tumor growth rates ranged between 0.7 and 4.8 mm per year (33, 35, 46). Recent studies

have shown that slow to medium growth of acoustic neuromas is in the range of 0.2 to

2.3 mm/year (1, 2, 10). A growth rate of >2 mm per year is seen in only 22-29% of

untreated acoustic neuromas (2, 35). In comparison, a fast tumor growth rate pattern is

considered to be approximately 10 mm per year (10). Methods of calculating tumor

growth include tumor diameter measurement and tumor volume calculation (33). The

results of two studies of tumor growth based on tumor diameter did not differ from the

results based on tumor volume (47, 48). However, some studies have demonstrated more

accurate tumor growth measurements using tumor volume calculations (42, 44).

It is important to note that potentially less than 1 percent of acoustic neuromas

demonstrate enough growth to become clinically active, which is indicative of a very

slow or arrested growth (33, 34). The emergence of gadolinium-enhanced magnetic

resonance imaging (MRI) has allowed the detection of smaller and often asymptomatic

tumors as small as 2 mm in diameter (35). As a result, the likelihood of finding an

acoustic neuroma that would have never become clinically significant has increased

enormously (33).

Tumors that exhibit enough growth to become clinically active encroach on the

vestibulocochlear nerve and are likely to cause unilateral high-frequency sensorineural

hearing loss, tinnitus, disequilibrium, or vertigo and as a result are at increased risk of

being diagnosed. Features that can potentially distinguish tumors with fast tumor growth

10

from those with slow or arrested growth include the influence of growth factors such as

the neuregulin-1 (NRG-1) and/or neuregulin-2 (NRG-2) proteins, as well as angiogenic

factors (7, 8, 16, 20, 21, 24, 36). Also, tumor vascularity, the presence of inflammation in

the Antoni type B cellular areas, a variation in basal apoptosis rates of acoustic neuroma

schwannoma cells, and a possible hormonal influence may also be factors influencing

tumor growth rate (1, 32, 49, 73). Additional features of fast-growing tumors include the

presence of Antoni B tissues containing extratumoral or intratumoral cystic components,

the over-expression or amplification of G1 regulators of the cell cycle such as cyclin D1

and cyclin D3, and the expression of erbB2 and erbB3 membrane tyrosine kinases (16,

20, 24, 74). When examining risk factors that may increase the likelihood of tumor

diagnosis, the only endogenous or exogenous factors associated with the features of fast-

growing tumors are patient sex and repeated environmental insults to the

vestibulocochlear nerve that may lead to chronic inflammation. Proinflammatory

cytokines such as IL-6, IL-1β, and TNF-α may initiate an inflammatory response after

loud noise exposure and thus may be involved in cochlear damage (75).

1.5 Clinical Presentation

Symptoms of acoustic neuroma are highly variable and include unilateral high-

frequency sensorineural hearing loss, tinnitus, dysequilibrium, pressure in the ear, otalgia,

and occasionally vertigo, which result from pressure exerted by the tumor upon the

cochlear and vestibular portions of the eighth cranial nerve (2, 6, 12). Approximately 10

percent of cases of unilateral progressive hearing loss are caused by acoustic neuromas

(1). Hearing loss is the most common finding in patients with acoustic neuromas, with

11

more than 95 percent of patients experiencing this symptom over the course of their

disease (2, 5). The mechanism of hearing loss is cochlear nerve compression (2). This

symptom may be of several years’ duration prior to diagnosis (37, 49). Alternative

explanations for hearing loss and tinnitus are made in the majority of acoustic neuroma

cases, resulting in diagnostic delay due to the patient or due to the physician (50).

Diagnostic delay is the period between the appearance of the first symptom and the time

that first medical attention is sought. The average time from onset of symptoms to

clinical diagnosis has been shown to range from approximately 4 to 7.3 years (12, 27,

37).

Nevertheless, as many as 5 to 12 percent of patients with newly diagnosed

acoustic neuromas have normal hearing, in part as a result of the detection of smaller

tumors by means of MRI (2, 49). In most cases, the onset of hearing loss is gradual, but

in 5 to 15 percent of cases it may be sudden if compression of the internal auditory artery

occurs (1, 2, 5). Another consequence of cochlear nerve dysfunction is tinnitus, usually

presenting with concomitant hearing loss (2, 5). Dysequilibrium and vertigo are

vestibular symptoms present in approximately 50 percent of patients (2). Ironically,

vestibular symptoms tend to be late in the course of the disease and mild in nature,

despite the predominant origin of acoustic neuromas on the vestibular division of the

eighth cranial nerve (1).

Symptoms related to any intracranial mass lesion may also be present. These

include sensory changes on the face or tongue, decreased corneal reflex, direction-

changing nystagmus, ipsilateral numbness of the face, facial nerve motor dysfunction, a

slurring of speech, ataxia, gait disturbance, an incoordination of one or both upper

12

extremities, numbness or tingling of the malar eminence, and occasionally long tract

signs (1, 2, 6). Facial nerve dysfunction generally occurs late in the course of the disease

and is rarely associated with small tumors (2). Decreased or absent corneal reflex is a

consequence of trigeminal nerve dysfunction, although this deficit is rarely noticed by

patients (2). Trigeminal nerve dysfunction is, however, responsible for the more common

complaint of numbness or tingling of the malar eminence (2). Brainstem compressive

symptoms include ipsilateral upper or lower extremity dysfunction and cerebellar

symptoms include ataxia and gait disturbance (2). These symptoms occur very late in the

disease and are observed with very large tumors (2).

Raised intracranial pressure, headaches, nausea, vomiting, obtundation,

papilloedema, and dullness of mental faculties are gradual and persistent symptoms

characteristic of the later stage of the disease (2, 6). These symptoms appear gradually

and are persistent (2). The appearance of headache as a symptom of an acoustic neuroma

is typically a sequela of hydrocephalus (2).

In summary, the symptoms of acoustic neuromas are highly dependent on tumor

size (1). Most patients with small tumors present with unilateral hearing loss, tinnitus,

and symptoms of vestibular nerve compression (2). Patients with larger tumors present

with symptoms of trigeminal nerve dysfunction, facial nerve dysfunction, and symptoms

related to increased intracranial pressure (2). Finally, continued growth of the tumor often

results in symptoms related to brainstem and cerebellar compression (2).

13

1.6 Diagnostic Modalities

Only 5 percent of patients presenting with symptoms suspicious for acoustic

neuroma will in fact have the tumor (1). The primary diagnostic modalities for acoustic

neuromas are audiometric assessment and imaging assessment (2, 5). Audiometry is a

sensitive, nonspecific method of screening for acoustic neuromas and should include

puretone air and bone conduction thresholds, speech reception thresholds, and speech

discrimination scores (1, 51). Asymmetric high-frequency sensorineural hearing loss is

demonstrated in the majority of patients with acoustic neuromas who undergo

audiograms (1, 2). High frequency hearing loss is attributable to the compression by the

tumor of those cochlear nerve fibers arranged in the outer part of the nerve (1). Loss of

speech discrimination is also associated with acoustic neuromas (1, 2). However, it is

important to note that a normal audiogram does not rule out acoustic neuroma, as 12

percent of patients with acoustic neuromas may have normal audiograms (2, 12).

On physical exam spontaneous nystagmus may be evidence of vestibular

dysfunction (2). When present, the nystagmus is usually horizontal (2). Vestibular

testing, such as electronystagmography is a quantitative methods of assessing

spontaneous and induced nystagmus using periorbital electrodes to detect the electrical

field shifts caused by the moving eye (1). However, it has shown to be inconsistent in its

ability to diagnose acoustic neuromas and it is not cost effective (2). The sensitivity of

electronystagmography is 78-98%, whereas the specificity is very poor (1). The

diagnostic value of vestibular testing is thought to be inferior to that of audiometry, as the

14

vestibular system has a good compensatory ability in the presence of an acoustic neuroma

tumor (52). Consequently, the role of vestibular tests in the diagnostic work-up for

acoustic neuromas continues to diminish (2, 49).

The auditory brainstem response which records the passage of electrical events in

the eighth nerve and brainstem following a sound stimulus, is the most reliable and

reproducible audiometric testing available for detecting acoustic neuromas (1, 4, 5). An

advantage of auditory brainstem response testing over electronystagmography is its lower

cost (52). This test replaced older traditional audiological tests such as speech

audiometry, stapedial reflex audiometry, loudness balance, and Carhart’s tone decay test

(5). Auditory brainstem response has high sensitivity of up to 93-100%, but its specificity

is somewhat lower (1, 2, 5). The sensitivity of auditory brainstem response makes it very

practical for ruling out acoustic tumors when a negative result is found (1). However, its

ability to detect small acoustic neuromas is limited and auditory evoked responses may

not be possible if severe hearing loss is present (2, 52).

The diagnosis of acoustic neuromas has been revolutionized by the introduction

of MRI with gadolinium enhancement, which remains the most sensitive diagnostic

method for detecting small acoustic neuromas (2, 5). It is particularly useful in its ability

to detect intracanalicular tumors (49). In addition, MR imaging exhibits exceptional soft

tissue visualization, but it does not show bony details (2). The gadolinium enhancement

allows a small tumor surrounded by bone to be seen as a bright signal bordered by a

black background (2). This technique is able to detect very small tumors, with a very low

false-positive rate, and is considered the gold standard for the detection of acoustic

15

tumors (2, 5). In addition, both sides of the head can be imaged at the same time (5). The

disincentive to the use of MRI as a screening tool for acoustic neuromas is its higher cost

(52).

If MRI facilities are not available, computed axial tomography (CT) scanning is

the next best approach, although its role is limited in detecting small acoustic neuromas

due to insufficient soft tissue differentiation and resolution (2, 5). CT scanning will

consistently miss tumors smaller than 15 mm, limiting its value as an effective screening

tool for acoustic neuromas (1).

In general, single tests are diagnostically inefficient for the diagnosis of acoustic

neuromas because of the unacceptable level of false-negative results, false-positive

results, or both that occur (53). The decision as to which test to use is dependent on level

of suspicion, as well as clinical judgment (52). Specific factors involved in the decision

are family history, asymmetry of auditory symptoms, brainstem evaluation findings, and

symptoms consistent with NF2 (52).

1.7 Management

The optimal treatment for a patient with an acoustic neuroma is total excision of

the tumor in a single surgical procedure with minimal morbidity and mortality, combined

with preservation of neurologic function (52). If a patient is not a candidate for total

surgical excision then other management options include observation, subtotal surgical

excision, and various types of radiation therapy, including stereotactic radiotherapy (52).

The decision as to whether and when a patient should be treated remains a complex issue.

Younger patients with evidence of tumor growth or with progressive neurologic

16

symptoms are clearly candidates for surgical intervention (52). Conversely, elderly

patients without evidence of tumor growth or severe neurologic deficits may be best

served by long-term observation (52). In addition, patients with very small, asymptomatic

acoustic neuromas identified by MRI with contrast enhancement may also be good

candidates for observation (12, 52).

Observation, also known as radiologic surveillance, is a mode of conservative

management involving the serial assessment of tumor growth by radiologic imaging, in

conjunction with periodic clinical assessment of symptom progression (39). The major

assumption with radiologic surveillance is that the majority of acoustic neuromas will not

grow and that immediate surgical or radiological intervention poses a greater risk of

complications to the patient than the potential outcome of delayed intervention (39). This

management option is chosen by some patients due to advanced age, poor medical

condition, or fear of surgery (49).

Specific preoperative findings have been identified to help predict postoperative

outcome (52). For example, a patient with mild hearing loss and normal auditory

brainstem response test who undergoes surgical excision of a tumor is more likely to

experience hearing preservation than a patient with a tumor larger than 20 mm or with a

tumor that fills the fundus of the internal auditory canal (52). The three major surgical

approaches for total removal of an acoustic neuroma tumor are each characterized by

advantages and disadvantages (49, 52). The middle fossa approach is suitable for small

tumors limited to the internal auditory canal and provides increased probability for

hearing preservation, whereas it is not suitable for larger tumors (12, 52). The

translabyrinthine approach does not facilitate hearing preservation, however it does allow

17

facial nerve preservation during the removal of small and large tumors (12, 52). Finally,

the suboccipital or retrosigmoid approach is suitable for small or large tumors and allows

identification of the brain stem, cranial nerves, and cerebral vasculature (12, 52). This

approach facilitates hearing preservation in patients with small tumors (52).

Total tumor removal is accomplished in the majority of cases and genuine tumor

recurrence is rare (5). The results of surgical excision of an acoustic neuroma are

predominantly dependent on the tumor size and the skill and experience of the surgical

team (5). The surgical management of acoustic neuromas is also enhanced by

intraoperative real-time neurologic monitoring (5, 52). This includes the preservation of

facial nerve function, as well as the use of intraoperative auditory brainstem response

monitoring (12, 52). Hearing preservation has been reported to be more successful in

tumors smaller than 20 mm in diameter (49).

Patients unable or unwilling to undergo surgical excision of a tumor may be

candidates for radiotherapy (12, 52). Options for radiotherapy include conventional

photon beam therapy, particle beam therapy, or stereotactic radiotherapy (52). The latter

treatment uses multisource cobalt 60 units for single-dose, external gamma ray therapy

and retardation of tumor growth has been observed in the majority of patients treated with

this type of radiotherapy (12, 52). In 45 percent of patients treated with stereotactic

radiotherapy tumor shrinkage is achieved (12, 52). An important component of

management is patient follow-up whereby repeated neurologic examination, audiologic

assessment, and radiographic imaging are performed every 3 to 6 months initially to

every 1 to 2 years, depending on the patient’s clinical course (35, 52).

18

Finally, chemotherapy has been reported in the treatment of patients with bilateral

acoustic neuromas (49). One report of a chemotherapeutic treatment regimen included the

use of cyclophosphamide, doxorubicin, and dacarbazine (49). However, surgical and

radiotherapeutic intervention remain the primary treatment modalities in use today.

1.8 Incidence, Age and Sex Distribution

The true incidence of acoustic neuromas has been difficult to estimate (2, 4). The

incidence has been reported to range from 1 to 20 per million per year (3, 4, 54). In

studies of temporal bone materials, undiagnosed acoustic neuromas were found in 0.57-

2.7% of the bones studied (33, 34). A study of acoustic neuromas in a review of 24,000

brain MRIs reported a prevalence of 0.07% (2). When comparing the clinical incidence

of acoustic neuromas to the prevalence of occult acoustic neuromas ascertained from

histopathological studies of temporal bones, it can be concluded that the vast majority of

tumors that exist are never clinically manifested (33, 53).

Acoustic neuromas are most commonly diagnosed between the ages of 30 and 68

(1, 35). Reported cases of acoustic neuroma occurring in childhood are rare and in such

patients other evidence of NF2 should be investigated (1, 9). In the study of 146 cases of

acoustic neuroma presented in Chapter 2 the median age of the cases was 52 years (55).

In the second study of 793 cases of acoustic neuroma presented in Chapter 3 the median

age of the cases was 54 years.

The sex ratio (females/males) for acoustic neuromas has been reported to be >1

(3, 5, 12, 56). However, data from the Central Brain Tumor Registry of the United States

(CBTRUS) do not support a female/male difference (57). Nevertheless, the studies

19

presented in Chapters 2 and 3 analyze Swedish data and the sex ratio (females/males) of

acoustic neuromas in the Nordic counties has been shown to be >1. This sex ratio may

indicate that hormones play a role in the etiology of acoustic neuromas (3, 58). Little is

reported in the literature regarding acoustic neuromas and ethnic distribution. The tumor

has been reported to be higher in whites than in non-whites (57) as well as being

uncommon in individuals of African ancestry (5).

According to some studies the incidence of acoustic neuromas has been

increasing (3, 54, 57). This may be due to several factors. Steady improvements in

diagnosis, such as the introduction of auditory brainstem response, CT, and MRI could

explain part of the increased incidence (3, 57). Increased awareness among physicians

and patients of the symptoms of acoustic neuromas may have caused an increase in the

reporting of the tumor (3, 54). Changes in classification or coding may also explain some

of the trend, whereby registries may have misclassified nonvestibular schwannomas as

acoustic neuromas (3, 57). Or, there may be a true increase in the incidence of these

tumors (57). The trend may also lend support to the emerging hypotheses regarding an

environmental cause of these tumors. Established and hypothesized risk factors have been

reported extensively in the literature and include ionizing radiation exposure, cellular

telephone use, specific occupations, a possible hormonal cause, and loud noise exposure.

1.9 Risk Factors

The causes of acoustic neuromas are largely unknown. Relatively few studies

have addressed this specific tumor type, as acoustic neuromas are seldom analyzed

separately in risk factor analyses and are more often grouped under the general heading

20

of brain tumors (59). The current epidemiologic literature on brain tumors is largely

concerned with risk of glioma or risk of brain tumors in general (60). Ionizing radiation

exposure is the only well-established exogenous risk factor for acoustic neuroma and has

been confirmed in studies of radiation treatments and dental X-rays (61, 62). Individuals

who underwent radiation treatment of tinea capitis during childhood were found to

develop an excess of benign and malignant brain tumors of various histological types,

including acoustic neuromas (55, 62). In addition, a study of atomic bomb survivors

found that the intracranial tumor subtype most strongly related to ionizing radiation

exposure was acoustic neuroma (61, 62). A statistically significant dose-related excess of

nervous system tumors, including schwannomas was observed in the cohort of atomic

bomb survivors (61). The excess relative risk per sievert (Sv) of absorbed dose for

schwannomas was 4.5 (95 percent confidence interval: 1.9, 9.2) and the dose-response

relationship was linear (61). Exposure to even moderate doses (i.e. <1 Sv) of radiation

was associated with an elevated incidence of nervous system tumors, including

schwannomas (61).

The association between the non-ionizing radiation from cellular telephones and

the risk of acoustic neuroma has been examined in several studies. However, the

evidence of an association so far is limited (63-70). This tumor is of particular interest in

relation to cellular telephones because the radiofrequency radiation emitted during

transmission is absorbed within a small area of the head close to the handset, which

includes the vestibular division of the eighth cranial nerve where the majority of acoustic

neuromas develop (64, 66). There is general agreement that the heating of brain tissue by

radiofrequency radiation from cellular telephones is negligible and that and that any

21

carcinogenic effect would have to be mediated through a nonthermal mechanism (63, 69).

In contrast to ionizing radiation, radiofrequency fields do not have enough energy to

break chemical bonds or to cause DNA damage (66).

Of ten published studies to date examining cellular telephones and risk of acoustic

neuroma, three showed any significantly increased risks (66, 67). A study performed by

Hardell et al. showed an increased risk with use of analogue phones, with an odds ratio of

4.4 (95 percent confidence interval: 2.1, 9.2) for both ipsilateral and contralateral use (66,

69). This increased risk was observed in short-term users, however, the study has been

heavily criticized for methodological and analytical limitations (64, 66). The second

study performed by Lönn et al. reported a relative risk associated with cellular telephone

use of at least 10 years’ duration to be 1.9 (95 percent confidence interval: 0.9, 4.1).

When the analysis was restricted to tumors on the same side of the head as the telephone

was normally used, the relative risk increased to 3.9 (95 percent confidence interval: 1.6,

9.5) (64, 66). In the study no increased risk was observed for short-term cellular

telephone use and a short latency period, regardless of tumor or cellular telephone

laterality (64). Finally, the third pooled study of two case-control studies conducted by

Hardell et al. demonstrated significantly increased risk of acoustic neuromas for analogue

cellular telephones, digital cellular telephones, and cordless telephones (70). The highest

risk was found for analogue cellular telephone use with a latency period of >15 years

with an odds ratio of 3.8 (95 percent confidence interval: 1.4, 10) (70).

The seven published studies that showed no increased risk were small and had

few long-term users (64, 66, 67). Results of these studies pertain primarily to analogue

telephones which became widely used in the mid-1990s. It is important to note that the

22

effects of digital telephones are difficult to separate from that of analogue telephones

because almost all analogue users are also users of digital telephones (64). One study that

found an increased risk of acoustic neuromas with more than 10 years of analogue

telephone use did not, however, find an increased risk with digital telephone use (64).

Studies of ionizing radiation have shown that the induction period of radiation-induced

solid tumors is likely to be at least 10 years, therefore, a carcinogenic effect following a

long induction period would remain undetected at the current time (64).

Elevated risk of acoustic neuromas has been associated with specific occupations

such as truck drivers, gas station attendants, sales representatives, and teachers (60, 71).

However, it remains unclear what specific environmental exposure within those

occupations is responsible for the increased risk.

As mentioned previously, the sex ratio (females/males) for acoustic neuroma is >1

which suggests a possible hormonal cause (3, 58). In a study of brain tumors and

menopausal status, menopausal women had a greatly reduced risk of developing

meningiomas (odds ratio = 0.58, 95 percent confidence interval: 0.18, 1.90), but a greater

risk of developing gliomas or acoustic neuromas (relative risk = 1.77, 95 percent

confidence interval: 0.67, 4.68) (58). The reduction in risk for meningiomas may be

attributable to the decreased estrogen levels during menopause, which is compatible with

the previous finding that certain brain tumors have been found to increase in size during

pregnancy, when estrogen levels are elevated (10, 58, 72). The study does not provide a

hypothesis for the increased risk of acoustic neuromas during menopause, although the

23

authors do suggest that since males are at lower risk for developing acoustic neuromas

than females, it is possible that the relative elevation of androgen levels after menopause

is providing a protective effect (58).

Exposure to loud noise has been suggested as a potential risk factor for acoustic

neuromas in two studies (55, 71). The first study by Preston-Matin et al. (1989) included

86 men in Los Angeles County aged 25-69 years, diagnosed with acoustic neuroma

between 1978 and 1985. The cases were pair-matched to neighborhood controls. Study

participants reported past loud noise exposures, and in addition self-reported occupational

histories were reviewed by an occupational hygienist to determine whether or not

significant noise exposure occurred. Loud noise was classified as either impact noise or

continuous noise. In the study the odds ratio for ever having a job involving exposure to

extremely loud noise was 2.2 (95 percent confidence interval: 1.12, 4.67). The study also

found a dose-response effect for years of employment in an occupation with loud noise

exposure (P for trend = 0.02). More cases reported exposure to extremely loud noise on

jobs held 10 or more years before the year of diagnosis with an odds ratio of 3.0

(P=0.004). The OR for exposure for 20 or more years during the period 10 or more years

before diagnosis was 13.2 (95 percent CI: 2.01, 86.98). Two significant weaknesses of

this study were that it included only men and it was not possible to report findings for

nonoccupational loud noise exposure, as too few study participants reported such

exposure. Recall bias is a potential problem in this study, as the study participants

responded to a questionnaire inquiring about loud noise and other environmental

exposures. The study did propose a hypothesis that mechanical trauma may contribute to

acoustic neuroma tumorigenesis (71).

24

A similar hypothesis was proposed in a subsequent study by Edwards et al. (2006)

examining loud noise exposure in 146 acoustic neuroma cases diagnosed between 1999

and 2002 in Sweden. Given the findings of the earlier study of loud noise and acoustic

neuromas, the goal of the second study was to further examine the role of loud noise in

acoustic neuroma etiology. The population-based case-control study of both sexes

compared reports on type and duration of occupational and non-occupational loud noise

exposure. Controls were selected randomly from the study base and were frequency

matched on age, sex and residential area. Those reporting occupational loud noise

exposure were at increased risk for acoustic neuromas with an odds ratio of 1.43 (95

percent confidence interval: 0.96, 2.13). Exposure to nonoccupational loud noise also

increased the risk for acoustic neuromas (odds ratio = 1.38, 95 percent confidence

interval: 0.80, 2.36). Individuals reporting loud noise exposure from any source were

found to be at increased risk for acoustic neuromas with an odds ratio of 1.55 (95%

confidence interval: 1.04, 2.30). Also of interest were those study participants exposed to

loud noise with hearing protection. The odds ratio for those individuals was 0.92 (95

percent confidence interval: 0.51, 1.64) (55).

In the study by Edwards et al. the risk of acoustic neuromas was also examined

according to duration of loud noise exposure. A dose-response effect was evident with

increasing years of loud noise exposure (P for trend = 0.0056). In an analysis of loud

noise exposure type, exposure to loud noise from machines, power tools and/or

construction increased the risk for acoustic neuromas (odds ratio = 1.79, 95% confidence

interval: 1.11, 2.89), as did exposure to loud music (odds ratio = 2.25, 95% confidence

interval: 1.20, 4.23). Finally, the significance of latency period and the risk of acoustic

25

neuromas was evaluated. The odds ratio for a latency period of ≥13 years since first loud

noise exposure from any source was 2.12 (95% confidence interval: 1.40, 3.20). The

findings of an increased risk of acoustic neuromas with loud noise exposure in the second

study supported the previous research. However, further research was recommended in

order to validate self-reports of loud noise exposure (55).

A third study of objectively measured occupational noise and the risk of acoustic

neuromas was conducted in a large cancer registry based case-control study. The study

was conducted in order to facilitate further examination of the role of loud noise in

acoustic neuroma etiology. A total of 793 acoustic neuroma cases aged 21-84 years were

identified between 1987 and 1999 from the Swedish Cancer Registry. The 101,756

controls randomly selected from the study base were frequency matched to cases on age,

sex, and calendar year of diagnosis. Occupational information, available for 599 of the

cases and 73,432 of the controls, was obtained from censuses and linked to a job

exposure matrix based on actual noise measurements. Contrary to previous study results,

all risk estimates were close to unity, regardless of noise exposure level or parameter. The

overall odds ratio for exposure to ≥85 decibels of noise was 0.89 (95 percent confidence

interval: 0.64, 1.23). Risk estimates for occupational noise exposure with a 5 year, 10

year, and 15 year latency period were all close to unity and all of the CIs for the three

latency periods for both low and high noise included the null. Odds ratios for exposure to

occupational noise according to one, two, three, or four consecutive censuses closest in

time prior to reference year were all slightly above unity but all of the corresponding CIs

included the null. The overall results of the study did not support the hypothesis that

26

occupational noise exposure is a risk factor for acoustic neuromas and the effect of non-

differential misclassification of exposure must be considered as a potential cause of the

negative findings.

2.10 References

1. Hart RG, Davenport J. Diagnosis of acoustic neuroma. Neurosurgery 1981;9:450-463.

2. Ho SY, Kveton JF. Acoustic neuroma assessment and management. Otolaryngol

Clin North Am 2002;35:393-404. 3. Howitz MF, Johansen C, Tos M, Charabi S, Olsen JH. Incidence of Vestibular

Schwannoma in Denmark, 1977-1995. Am J Otol 2000;21:690-694. 4. Tos M, Charabi S, Thomsen J. Clincial experience with vestibular schwannomas:

epidemiology, symptomatology, diagnosis, and surgical results. Eur Arch Otorhinolaryngol 1998;255:1-6.

5. Chandler CL, Ramsden RT. Acoustic schwannoma. Br J Hosp Med 1993;49:336-

343. 6. Miller MH, Doyle TJ, Geier SR. Acoustic neuroma in a population of noise

exposed workers. Laryngoscope 1981;91:363-371. 7. Charabi S. Acoustic neuroma/vestibular schwannoma in vivo and in vitro growth

models. Acta Otolaryngol 1997;Suppl 530:6-27. 8. Cardillo MR, Filipo R, Monini S, Aliotta N, Barbara M. Transforming growth

factor-β1 expression in human acoustic neuroma. Am J Otol 1999;20:65-68. 9. Nager GT. Acoustic Neurinomas. Acta Otolaryngol 1985;99:245-261. 10. Lanser MJ, Sussman SA, Frazer K. Epidemiology, pathogenesis, and genetics of

acoustic tumors. Otolaryngol Clin North Am 1992;25:499-520. 11. National Institutes of Health consensus development conference statement on

acoustic neuroma, December 11-13, 1991. Arch Neurol, 1994:201-207. 12. Spoelhof GD. When to suspect acoustic neuroma. Am Fam Physician

1995;52:1768-1774.

27

13. Moffat DA, Irving RM. The molecular genetics of vestibular schwannoma. J Laryngol Otol 1995;109:381-384.

14. Irving RM, Moffat DA, Hardy DG, Barton DE, Xuereb JH, Maher ER. Molecular

genetic analysis of the mechanism of tumorigenesis in acoustic neuroma. Arch Otolaryngol Head Neck Surg 1993;119:1222-1228.

15. Seizinger BR, Martuza RL, Gusella JF. Loss of genes on chromosome 22 in

tumorigenesis of human acoustic neuroma. Nature (London) 1986;322:644-647. 16. Welling DB, Lasak JM, Akhmametyeva E, Ghaheri B, Chang L-S. cDNA

microarray analysis of vestibular schwannomas. Otol Neurotol 2002;23:736-748. 17. Mawrin C, Kirches E, Boltze C, Dietzmann K. PTEN is not altered in sporadic

vestibular schwannomas. Histopathology 2002;40:526-530. 18. Hanemann CO, Bartelt-Kirbach B, Diebold R, Kämpchen K, Langmesser S,

Utermark T. Differential gene expression between human schwannoma and control Schwann cells. Neuropathol Appl Neurobiol 2006;32:605-614.

19. Warren C, James LA, Ramsden RT, et al. Identification of recurrent regions of

chromosome loss and gain in vestibular schwannomas using comparative genomic hybridisation. J Med Genet 2003;40:802-806.

20. Stonecypher MS, Chaudhury AR, Byer SJ, Carroll SL. Neuregulin growth factors

and their ErbB receptors form a potential signaling network for schwannoma tumorigenesis. J Neuropathol Exp Neurol 2006;65:162-175.

21. Irving RM, Moffat DA, Hardy DG, et al. A molecular, clinical, and

immunohistochemical study of vestibular schwannoma. Otolaryngol Head Neck Surg 1997;116:426-430.

22. Sans A, Bartolami S, Fraysse B. Histopathology of the peripheral vestibular

system in small vestibular schwannomas. Am J Otol 1996;17:326-24. 23. Kleihues P, Cavenee WK. Pathology and genetics of tumours of the nervous

system. Lyon: International Agency for Research on Cancer, 1997. 24. Hansen MR, Roehm PC, Chatterjee P, Green SH. Constitutive neuregulin-1/ErbB

signaling contributes to human vestibular schwannoma proliferation. Glia 2006;53:593-600.

25. Manolidis L, Baloyannis SJ. Pathology of acoustic nerve schwannoma in vivo and

in vitro. Acta Otolaryngol 1985;99:280-284.

28

26. Tallan EM, Harner SG, Beatty CW. Does the distribution of Schwann cells correlate with the observed occurrence of acoustic neuromas? Am J Otol 1993;14:131-134.

27. Spickler EM, Govila L. The vestibulocochlear nerve. Semin Ultrasound CT MR

2002;23:218-237. 28. Jorgensen MB. Intracochlear neurinoma. Acta Otolaryngol 1962;54:227-232. 29. Sterkers JM, Perre J, Viala P, Foncin J-F. The origin of acoustic neuromas. Acta

Otolaryngol 1987;103:427-431. 30. Forton G, Moeneclaey L, Declau F, Marquet J. The involvement of the cochlear

nerve in neurinomas of the eighth cranial nerve. Arch Otorhinolaryngol 1989;246:156-160.

31. Schefter RP, Harner SG. Histologic study of the vestibulocochlear nerve. Ann

Otol Rhinol Laryngol 1986;95:146-50. 32. Labit-Bouvier C, Crebassa B, Bouvier C, Andrac-Meyer L, Magnan J, Charpin C.

Clinicopathologic growth factors in vestibular schwannomas: a morphological and immunohistochemical study of 69 tumours. Acta Otolaryngol 2000;120:950-954.

33. Rosenberg SI. Natural history of acoustic neuromas. Laryngoscope 2000;110:497-

508. 34. Yoshimoto Y. Systematic review of the natural history of vestibular schwannoma.

J Neurosurg 2005;103:59-63. 35. Strasnick B, Glasscock ME, Haynes D, McMenomey SO, Minor LB. The natural

history of untreated acoustic neuromas. Laryngoscope 1994;104:1115-1119. 36. Sturgis EM, Woll SS, Aydin F, Marrogi AJ, Amedee RG. Epidermal growth

factor receptor expression by acoustic neuromas. Laryngoscope 1996;106:457-462.

37. Thomsen J, Tos M. Acoustic neuromas. Diagnostic delay, growth rate and

possible non-surgical treatment. Acta Otolaryngol 1988;Suppl 452:26-33. 38. Kesterson L, Shelton C, Dressler L, Berliner KI. Clinical behavior of acoustic

tumors. Arch Otolaryngol Head Neck Surg 1993;119:269-271. 39. Selesnick SH, Johnson G. Radiologic surveillance of acoustic neuromas. Am J

Otol 1998;19:846-849.

29

40. Nutik SL, Babb MJ. Determinants of tumor size and growth in vestibular schwannomas. J Neurosurg 2001;94:922-926.

41. Fucci MJ, Buchman CA, Brackmann DE, Berliner KI. Acoustic tumor growth:

implications for treatment choices. Am J Otol 1999;20:495-499. 42. Niemczyk K, Vaneecloo FM, Lemaitre L, et al. The growth of acoustic neuromas

in volumetric radiologic assessment. Am J Otol 1999;20:244-248. 43. Charabi S, Tos M, Thomsen J, Charabi B, Mantoni M. Vestibular schwannoma

growth: the continuing controversy. Laryngoscope 2000;110:1720-1725. 44. Herwadker A, Vokurka EA, Evans DGR, Ramsden RT, Jackson A. Size and

growth rate of sporadic vestibular schwannoma: predictive value of information available at presentation. Otol Neurotol 2005;26:86-92.

45. Stangerup S-E, Caye-Thomasen P, Tos M, Thomsen J. The natural history of

vestibular schwannoma. Otol Neurotol 2006;27:547-552. 46. Mirz F, Jørgensen B, Fiirgaard B, Lundorf E, Pedersen CB. Investigations into the

natural history of vestibular schwannomas. Clin Otolaryngol 1999;24:13-18. 47. Charabi S, Thomsen J, Mantoni M, et al. Acoustic neuroma (vestibular

schwannoma): growth and surgical and nonsurgical consequences of the wait-and-see policy. Otolaryngol Head Neck Surg 1995;113:5-14.

48. Laasonen EM, Troupp H. Volume growth rate of acoustic neurinomas.

Neuroradiology 1986;28:203-207. 49. Brackmann DE, Kwartler JA. A review of acoustic tumors: 1983-1988. Am J Otol

1990;11:216-232. 50. van Leeuwen JPPM, Harhangi BS, Thewissen NPMW, Thijssen HOM, Cremers

CWRJ. Delays in the diagnosis of acoustic neuromas. Am J Otol 1996;17:321-325.

51. Eldridge R, Parry D. Summary: vestibular schwannoma (acoustic neuroma)

consensus development conference. Neurosurgery 1992;30:962-964. 52. Acoustic neuroma. National Institutes of Health consensus statement online 1991

Dec. 11-13, 1991:1-24. 53. Thomsen J, Tos M. Acoustic neuroma: clinical aspects, audiovestibular

assessment, diagnostic delay, and growth rate. Am J Otol 1990;11:12-19.

30

54. Tos M, Stangerup SE, Cayé-Thomasen P, Tos T, Thomsen J. What is the real incidence of vestibular schwannoma? Arch Otolaryngol Head Neck Surg 2004;130:216-220.

55. Edwards CG, Schwartzbaum JA, Lönn S, Ahlbom A, Feychting M. Exposure to

loud noise and risk of acoustic neuroma. Am J Epidemiol 2006;163:327-333. 56. Inskip PD, Tarone RE, Hatch EE, et al. Sociodemographic indicators and risk of

brain tumours. Int J Epidemiol 2003;32:225-233. 57. Propp JM, McCarthy BJ, Davis FG, Preston-Martin S. Descriptive epidemiology

of vestibular schwannomas. Neuro-oncol 2006;8:1-11. 58. Schlehofer B, Blettner M, Wahrendorf J. Association between brain tumors and

menopausal status. J Natl Cancer Inst 1992;84:1346-1349. 59. Forssén UM, Lönn S, Ahlbom A, Savitz DA, Feychting M. Occupational

magnetic field exposure and the risk of acoustic neuroma. Am J Ind Med 2006;49:112-118.

60. Rajaraman P, De Roos AJ, Stewart PA, et al. Occupation and risk of meningioma

and acoustic neuroma in the United States. Am J Ind Med 2004;45:395-407. 61. Preston DL, Ron E, Yonehara S, et al. Tumors of the nervous system and pituitary

gland associated with atomic bomb radiation exposure. J Natl Cancer Inst 2002;94:1555-1563.

62. Ron E, Modan B, Boice JD, et al. Tumors of the brain and nervous system after

radiotherapy in childhood. N Engl J Med 1988;319:1033-1039. 63. Inskip PD, Tarone RE, Hatch EE, et al. Cellular-telephone use and brain tumors.

N Engl J Med 2001;344:79-86. 64. Lönn S, Ahlbom A, Hall P, Feychting M. Mobile phone use and the risk of

acoustic neuroma. Epidemiology 2004;15:653-659. 65. Muscat JE, Malkin MG, Shore RE, et al. Handheld cellular telephones and risk of

acoustic neuroma. Neurology 2002;58:1304-1306. 66. Schoemaker MJ, Swerdlow AJ, Ahlbom A, et al. Mobile phone use and risk of

acoustic neuroma: results of the Interphone case-control study in five North European countries. Br J Cancer 2005;93:842-848.

31

67. Hardell LO, Carlberg M, Söderqvist F, Mild KH, Morgan LL. Long-term use of cellular phones and brain tumors - increased risk associated with use for >10 years. Occup Environ Med 2007. Apr 4; [Epub ahead of print].

68. Lönn S, Ahlbom A, Hall P, Feychting M. Long-term mobile phone use and brain

tumor risk. Am J Epidemiol 2005;161:526-535. 69. Hardell L, Mild KH, Carlberg M. Further aspects on cellular and cordless

telephones and brain tumors. Int J Oncol 2003;22:399-407. 70. Hardell L, Carlberg M, Mild KH. Pooled analysis of two case-control studies on

the use of cellular and cordless telephones and the risk of benign brain tumours diagnosed during 1997-2003. Int J Oncol 2006;28:509-518.

71. Preston-Martin S, Thomas DC, Wright WE, Henderson BE. Noise trauma in the

aetiology of acoustic neuromas in men in Los Angeles County, 1978-1985. Br J Cancer 1989;59:783-786.

72. Klinken L, Thomsen J, Rasmussen BB, Wiet RJ, Tos M. Estrogen and

progesterone receptors in acoustic neuromas. Arch Otolaryngol Head Neck Surg 1990;116:202-204.

73. Utermark T, Kaempchen K, Antoniadis G, Hanemann CO. Reduced Apoptosis

Rates in Human Schwannomas. Brain Pathol 2005;15:17-22. 74. Neff BA, Oberstien E, Lorenz M, Chaudhury AR, Welling DB, Chang L-S.

Cyclin D1 and D3 expression in vestibular schwannomas. Laryngoscope 2006;116:423-426.

75. Fujioka M, Kanzaki S, Okano HJ, Masuda M, Ogawa K, Okano H.

Proinflammatory cytokines expression in noise-induced damaged cochlea. J Neurosci Res 2006;83:575-583.

32

CHAPTER 2

EXPOSURE TO LOUD NOISE AND RISK OF ACOUSTIC NEUROMA1

2.1 Abstract

Exposure to occupational loud noise has been previously identified as a possible

risk factor for acoustic neuroma in only one relatively small (n=86 cases) case-control

study of men. The goal of the present study was to further examine the role of loud noise

in acoustic neuroma etiology. In our population-based case-control study of both sexes

conducted from 1999-2002 in Sweden, I compared reports on type and duration of

occupational and non-occupational loud noise exposure of 146 acoustic neuroma cases

and 564 controls. Controls were randomly selected from the study base and were

frequency matched on age, sex and residential area. I found that individuals reporting

loud noise exposure from any source were at increased risk for acoustic neuroma (odds

ratio (OR) = 1.55, 95% confidence interval (CI): 1.04, 2.30). Exposure to loud noise from

machines, power tools and/or construction increased the risk for acoustic neuroma

1Edwards CG, Schwartzbaum JA, Lönn S, Ahlbom A, Feychting M. Exposure to loud

noise and risk of acoustic neuroma. Am J Epidemiol 2006;163(4):327-333 by permission

of Oxford University Press.

33

(OR = 1.79, 95% CI: 1.11, 2.89), as did exposure to loud music (OR = 2.25, 95% CI:

1.20, 4.23). The OR for a latency period of ≥13 years since first loud noise exposure from

any source was 2.12 (95% CI: 1.40, 3.20). The findings of an increased risk of acoustic

neuroma with loud noise exposure support previous research.

2.2 Introduction

Acoustic neuroma, also referred to as vestibular schwannoma, is a benign tumor

of the vestibular division of the eighth cranial nerve (1, 2). This tumor results in hearing

loss, tinnitus and dysequilibrium (3, 4). The tumor constitutes from 6-10 percent of all

intracranial tumors, with an incidence of 1-20 per million per year (5-7). The sex ratio

(females/males) for acoustic neuroma has been reported to be >1 (1, 6, 8, 9) and the

tumor occurs mainly in individuals aged 50 years or older (6). Acoustic neuromas present

clinically as two distinct types, the bilateral hereditary type and the unilateral sporadic

type (4, 8). In the present study we examine unilateral acoustic neuroma which comprises

90-95 percent of all acoustic neuromas (6, 10).

We report the results from the Swedish portion of the INTERPHONE study, an

international collaborative case-control study of brain tumors, acoustic neuroma, and

parotid gland tumors in relation to mobile phone use and other potential risk factors (11).

In a recent publication from this study, an association between acoustic neuroma and

mobile phone use was reported that is awaiting confirmation from other studies (12). In

the present study we focus instead on another potential acoustic neuroma risk factor, loud

noise exposure. The only available study to date examining loud noise exposure and

34

acoustic neuroma risk was limited by the relatively small number of acoustic neuroma

cases available for inclusion in the analysis, as well as the restriction of the study

population to men (13).

2.3 Materials and Methods

2.3.1 Study design and population A population based case-control study was

conducted including all individuals aged 20 to 69 years who resided in three geographical

regions covered by the regional cancer registries in Stockholm, Göteborg, and Lund,

Sweden, which comprised a population of approximately 3.1 million people. Data were

collected during the period 1999-2002. Study approval was obtained from the

institutional Ethics Committee and oral informed consent was obtained from all study

participants.

2.3.2 Acoustic neuroma case ascertainment Eligible cases were all

patients diagnosed with acoustic neuroma (ICD-10 C72.4 and ICD-O-2 9560.0) during

the period from 1st of September 1999 to 31st of August 2002 in the areas covered by the

Lund and Göteborg Cancer Registries and from 1st of January 2000 to 31st of August

2002 in the Stockholm Cancer Registry area. Continuous identification of cases

throughout the study period was achieved through collaboration with neurosurgery,

oncology, neurology, and otorhinolaryngology clinics at hospitals within the

geographical regions covered by the study. Any patients missed during weekly visits to

these clinics were subsequently identified during quarterly regional Cancer Registry

searches. The first medical examination resulting in diagnosis of acoustic neuroma was

used as the date of diagnosis and also as the reference date for exposure calculations.

35

Medical records for all cases were examined to confirm the diagnosis and to determine

the side of the head on which the tumor was located. Histopathological reports were used

to verify diagnosis in 58 cases (40%) and the remaining cases were diagnosed by

computerized axial tomography or magnetic resonance imaging. Histological

classification of the cases as acoustic neuroma was performed by pathologists at the

hospitals in the three study regions where the cases underwent surgery or biopsy. We

identified a total of 160 eligible acoustic neuroma case patients, of whom 146 (91

percent) were interviewed. Those cases who did not participate included 11 (7 percent)

who would not consent to the study and three (2 percent) who could not be contacted.

2.3.3 Controls Controls were randomly selected approximately every 2

months from the continuously updated Swedish population registry. The number of

controls required for each case was stipulated by the common protocol of the

INTERPHONE study (one per brain tumor case, two per acoustic neuroma case, and

three per parotid gland tumor case). Of the 838 controls identified for inclusion in the

study, 564 (67 percent) were interviewed. Those controls who did not participate

included 127 (15 percent) who would not consent to the study and 147 (18 percent) who

could not be reached by phone in order to make an appointment for an interview. There

were no major differences between the respondents and the non-respondents with regard

to age, sex and area of residence. For controls, the reference date was defined as the date

of identification of the control, adjusted for the average time difference between the date

of diagnosis and the date of identification of the cases within the same matching stratum.

This assured a comparable length of follow-up for cases and controls.

36

2.3.4 Data collection and loud noise exposure assessment Identification

of cases and controls occurred prospectively from September 2000 through August 2002,

and retrospectively in the 12 months prior to September 2000 in the Lund and Göteborg

study regions and in the 8 months prior September 2000 in the Stockholm study region.

The study participant contact and interview procedures were similar for cases and

controls. Interviews were conducted by study nurses or the study neuro-psychologist

using laptop computers. Study participants who would not participate in a personal

interview were offered a telephone interview. Those who refused participation in any

kind of interview were offered the option of completing a short written questionnaire

instead. The written questionnaire was focused on mobile phone use and did not include

questions about loud noise. Therefore, participation rates are lower in the analyses

reported here compared to the analysis of mobile phone use described earlier (12). A

proxy respondent was used in two cases where the case had died before the first contact

by study personnel. Eligibility criteria for both cases and controls required that they were

not completely deaf prior to the reference date. One control was excluded for this reason.

The cases and controls also had to possess the intellectual and language skills necessary

to complete the interview. Three cases and 18 controls were excluded because they did

not speak Swedish; four controls were excluded because of insufficient intellectual skills.

The study participants were asked if they were exposed to occupational loud noise

and also if they were exposed to regular non-occupational loud noise. Exposure to loud

noise was defined as a level of 85 dB or more. This cut-off was explained to the study

participants by means of a diagram depicting a decibel scale and associated loud noise

exposures. If they were exposed to either occupational loud noise, regular

37

non-occupational loud noise, or both, the study participants were then asked to specify

the activities in which they were exposed to loud noise and the year in which the

exposure started and the year in which it stopped. They were also asked if there were any

years during this period in which they were unexposed.

Total years of loud noise exposure were categorized into less than 5 years, 5-14

years and 15 years or more (with cut points at approximately the 25th and 75th percentile

for controls). In the analysis of type of loud noise exposure, the following categories were

created: (1) exposure to machines, power tools and/or construction; (2) exposure to

motors, including airplanes; (3) exposure to loud music, including employment in the

music industry; and (4) exposure to screaming children, sports events and/or restaurants

or bars. The remainder was classified as “other” types of loud noise exposure. Ninety-

seven percent of the loud noise exposure responses could be categorized into one of the

four loud noise exposure types. Data regarding the use of hearing protection were also

collected. This would allow us to determine if there was any difference between study

participants who were unexposed and participants who used hearing protection. Finally,

complete job history was collected for the 20 years preceding diagnosis, but not in

sufficient detail to be used as a loud noise exposure validation tool.

2.3.5 Statistical analysis Unconditional logistic regression models adjusted

for age (5 year categories), sex and local cancer registry region were used to estimate

odds ratios (ORs) and their respective 95 percent confidence intervals (CIs) (PROC

LOGISTIC in SAS, version 8; SAS Institute, Inc., Cary, North Carolina) (14). The OR

was used as an estimate of the relative risk in the analysis of the interview data. In the

analysis of loud noise exposure type (Table 2.4), we also adjusted for highest level of

38

education which we used as a proxy for socio-economic status. In the evaluation of

potential confounding variables, the data presented in Table 2.2 were stratified on

ionizing radiation exposure due to medical treatment, as well as on mobile phone use.

The odds ratios did not differ within the different strata, discounting these variables as

potential confounders or effect modifiers. In addition, radiation exposure and mobile

phone use were added to the logistic regression model. However, the changes in the ORs

were negligible and therefore these variables were not included in the final model. Tests

for trend were calculated using the Cochran-Armitage test for trend. All tests of statistical

significance performed were two-sided.

2.4 Results

Basic demographic characteristics of cases and controls are presented in Table

2.1. The median age of the cases and controls was 52 years and 54 years, respectively.

Tumor location was more common on the right side (59 percent) than on the left side (41

percent).

Study participants were initially classified as unexposed to loud noise or exposed

to occupational and/or regular non-occupational loud noise, with either hearing protection

or without hearing protection. Ten percent of the cases and 11 percent of the controls who

were exposed to occupational loud noise, regular non-occupational loud noise, or both,

reported using hearing protection most of the time.

The ORs for acoustic neuroma by loud noise exposure are shown in Table 2.2.

The OR for occupational loud noise exposure (OR = 1.43, 95 percent CI: 0.96, 2.13) was

similar to the OR for regular non-occupational loud noise exposure (OR = 1.38, 95

39

percent CI: 0.80, 2.36). When study participants exposed to occupational loud noise or

regular non-occupational loud noise (or both) were combined, the OR increased to 1.55

(95 percent CI: 1.04, 2.30). The OR for the group exposed to loud noise with hearing

protection was 0.92, 95 percent CI: 0.51, 1.64. In all subsequent analyses the subjects

using hearing protection were categorized as unexposed to loud noise. In an additional

analysis, those study participants using hearing protection were excluded. There was a

slight rise in the odds ratios for the three loud noise exposure categories listed in Table

2.2. Finally, the overall analysis presented in Table 2.2 was stratified on hearing loss. The

patterns within the hearing loss and the no hearing loss groups were found to be

comparable.

Table 2.3 displays results according to duration of loud noise exposure,

combining occupational and regular non-occupational loud noise. For males and females

combined, all of the exposure duration groups had elevated ORs, with the highest OR at

1.64 (95 percent CI: 0.91, 2.91) for the group with 5-14 years of loud noise exposure. A

dose-response effect is evident with increasing years of loud noise exposure (P for trend

= 0.0056). When the loud noise exposure duration data were analyzed separately for

males and females, the highest OR for males was 2.12 (95 percent CI: 0.99, 4.57) for 5-

14 years of loud noise exposure and the highest OR for females was 3.34 (95 percent CI:

1.32, 8.43) for ≥15 years of loud noise exposure. An additional analysis of loud noise

exposure duration was performed, excluding the five years prior to diagnosis. However,

the pattern in the additional analysis was similar to that of the original analysis. Finally,

the duration of exposure analysis was repeated, dropping hearing protection from the

40

comparison group. There were negligible changes in the odds ratios for the analysis of

the male study participants, the female study participants and the combined group.

Table 2.4 presents type of loud noise exposure grouped into four distinct

categories. A significantly greater proportion of men than women were exposed to

machines, power tools and construction, as well as to motors, including airplanes. A

greater proportion of men were exposed to loud music, including employment in the

music industry and a greater proportion of women were exposed to screaming children,

sports events and/or restaurants/bars, although the latter differences were not significant.

All of the categories had elevated ORs. The two categories with the highest ORs were

exposure to machines, power tools and/or construction, with an OR of 1.79 (95 percent

CI: 1.11, 2.89) and exposure to music, including employment in the music industry with

an OR of 2.25 (95 percent CI: 1.20, 4.23). Subgroup logistic regression analyses were

performed for males and females separately. The resulting ORs were in the same

direction for both subgroups, although the numbers of study participants in each

subgroup were too small for meaningful analysis and are therefore not presented.

To evaluate the significance of latency period and the risk of acoustic neuroma we

analyzed the data using three latency periods. The data were analyzed for those study

participants with <13 years, 13-26 years, and ≥27 years since first regular loud noise

exposure (with cut points at approximately the 25th and 75th percentile for controls). The

data are presented in Table 2.5 for occupational and/or regular non-occupational loud

noise exposure only, as the numbers of study participants in the analyses of other

categories of loud noise exposure were too small for meaningful analysis. The OR

increased with increasing latency period (P for trend = 0.0029). The OR for ≥13 years

41

since first regular loud noise exposure was 2.12 (95% CI: 1.40, 3.20) (data not shown).

An additional analysis of latency period was performed, excluding the five years prior to

diagnosis. However, the pattern in the additional analysis was similar to that of the

original analysis.

42

Table 2.1. Selected characteristics of the 710 study participants, Sweden, 1999-2002.

Cases (n=146) Controls (n=564) No. % No. % Sex Male 77 53 274 49 Female 69 47 290 51 Age at reference date (years) 20-29 8 5 43 8 30-39 18 12 74 13 40-49 27 19 103 18 50-59 57 39 188 33 60-69 36 25 156 28 Highest Education Compulsory school 29 20 118 21 Vocational/secondary school 36 25 151 27 Upper secondary school 26 18 115 20 University 53 36 177 31 Unknown 2 1 3 1

43

1 OR, odds ratio, from unconditional logistic regression analysis, adjusted for age, sex and region 2 CI, confidence interval 3 Reference category includes individuals never exposed to loud noise 4 Exposure to occupational loud noise, or non-occupational loud noise, or both occupational and non-occupational loud noise Table 2.2. Odds ratios for acoustic neuroma according to loud noise exposure, Sweden, 1999-2002.

Cases Controls Exposure to loud noise No. % No. % OR1 95% CI2 Never3 72 49 340 60 1.00 With hearing protection 15 10 61 11 0.92 0.51, 1.64 Occupational 64 44 195 35 1.43 0.96, 2.13 Non-occupational 27 18 77 14 1.38 0.80, 2.36 Occupational and/or non-occupational4 74 51 223 40 1.55 1.04, 2.30

43

44

1 OR, odds ratio, from unconditional logistic regression analysis, adjusted for age, sex and region 2 CI, confidence interval 3 4 male controls and 6 female controls did not specify duration of loud noise exposure 4 Cochran-Armitage test for trend Table 2.3. Odds ratios for acoustic neuroma according to duration of loud noise exposure, Sweden, 1999-2002.

Males and females Cases Controls Years of exposure to loud noise No. % No.3 % OR1 95% CI2 P for trend4 Never or used hearing protection 87 60 401 71 1.00 < 5 years 14 10 38 7 1.51 0.77, 2.95 5-14 years 20 14 51 9 1.64 0.91, 2.91 ≥ 15 years 25 17 64 11 1.56 0.91, 2.66 0.0056 Males Years of exposure to loud noise Never or used hearing protection 40 52 177 65 1.00 < 5 years 8 10 18 7 1.71 0.67, 4.38 5-14 years 14 18 27 10 2.12 0.99, 4.57 ≥ 15 years 15 19 48 18 1.18 0.60, 2.32 0.11 Females Years of exposure to loud noise Never or used hearing protection 47 68 224 77 1.00 < 5 years 6 9 20 7 1.24 0.44, 3.52 5-14 years 6 9 24 8 1.01 0.36, 2.81 ≥ 15 years 10 14 16 6 3.34 1.32, 8.43 0.024

44

45

1 OR, odds ratio, from unconditional logistic regression analysis, adjusted for age, sex, region and education 2 CI, confidence interval Table 2.4. Odds ratios for acoustic neuroma according to loud noise exposure type, Sweden, 1999-2002.

Cases Controls Exposure to occupational and regular non-occupational loud noise No. % No. % OR1 95% CI2 Never or used hearing protection 87 60 401 71 1.00 Machines, power tools and/or construction 35 24 90 16

1.79 1.11, 2.89 Motors, including airplanes 12 8 34 6

1.30 0.63, 2.67 Music, including employment in the music industry 21 14 48 9

2.25 1.20, 4.23 Screaming children, sports events and/or restaurants/bars 7 5 23 4

1.40 0.56, 3.49

45

46

1 OR, odds ratio, from unconditional logistic regression analysis, adjusted for age, sex and region 2 CI, confidence interval 3 includes exposure to occupational and/or regular non-occupational loud noise 4 10 controls did not specify year in which loud noise began 5 Cochran-Armitage test for trend

Table 2.5. Odds ratios for acoustic neuroma according to loud noise exposure and latency period, Sweden, 1999-2002.

Cases Controls4 Time since first regular loud noise exposure (years)3 No. % No. % OR1 95% CI2 P for trend5 Never or used hearing protection 87 60 401 71 1.0 <13 6 4 36 6 0.68 0.26, 1.77 13-26 30 21 78 14 1.74 1.06, 2.84 ≥27 23 16 39 7 2.15 1.19, 3.86 0.0029

46

47

2.5 Discussion

2.5.1 Loud noise exposure and risk of acoustic neuroma Exposure to

any loud noise, to occupational loud noise and to regular non-occupational loud noise

were all associated with an increased risk of acoustic neuroma. Each of our three

categories of loud noise exposure duration were associated with an increased risk of

acoustic neuroma, with the highest risk of acoustic neuroma for study participants

exposed to loud noise for a duration of 5-14 years. The two types of loud noise exposure

with the highest risk of acoustic neuroma were exposure to loud noise from machines,

power tools and/or construction and exposure to loud noise from music, including

employment in the music industry.

The results of the present study are in agreement with the only other known study

examining loud noise exposure and the risk of acoustic neuroma (13). The previous study

reported results in 86 males, a slightly larger number than the 77 males in the present

study. Not only was our study able to replicate the previous study’s findings in males, but

perhaps more importantly an elevated risk of acoustic neuroma was also found in

females. In the previous study, the OR for ever having a job involving exposure to

extremely loud noise was 2.2 (95 percent CI: 1.12, 4.67). The study also found a dose-

response effect for years of employment in an occupation with loud noise exposure (P for

trend = 0.02). The OR for exposure for 20 or more years during the period 10 or more

years before diagnosis was 13.2 (95 percent CI: 2.01, 86.98). The previous study included

only men and was not able to report findings for non-occupational loud noise exposure,

as too few study participants reported such exposure.

48

In our study, elevated risk of acoustic neuroma was found with all loud noise

exposure duration categories in males, which is consistent with the previous study (13)

and elevated risk of acoustic neuroma was found with all of the loud noise exposure

duration categories in females. No comparison can be made with the previous study for

the latter group, as they did not collect data on occupational loud noise exposure among

females. One recent study of occupation and risk of meningioma and acoustic neuroma

found that those occupations associated with an increased risk of acoustic neuroma were

not occupations with which one would expect an unusually high exposure to loud noise

(60). However, the study was only able to estimate loud noise exposure indirectly, as no

data on loud noise exposure were collected. Finally, in the present study tumor location

was found to be more common on the right side than on the left side. Other data also

indicate a similar uneven distribution with regard to tumor laterality (16).

2.5.2 Other risk factors In a previous report from the Swedish

INTERPHONE study of mobile phone use and the risk of acoustic neuroma, the relative

risk associated with mobile phone use of at least 10 years’ duration was shown to be 1.9

(95 percent CI: 0.9, 4.1). When the analysis was restricted to tumors on the same side of

the head as the phone was normally used, the relative risk increased to 3.9 (95 percent CI:

1.6, 9.5). The study had a larger number of exposed acoustic neuroma cases than any of

six previous studies examining mobile phone use and thus was better powered to study

the effects of long-term mobile phone use (12). The findings of this recent study will

require confirmation in other studies. Mobile phone use was evaluated in the present

study and was not found to be a confounding variable.

49

Ionizing radiation exposure is the only well-established exogenous risk factor for

acoustic neuroma. It has been found to increase acoustic neuroma risk among individuals

who underwent radiation treatment of tinea capitis during childhood and who developed

an excess of benign and malignant brain tumors of various histological types, including

acoustic neuroma, and among survivors of the atomic bombings in Japan (17, 18).

Ionizing radiation exposure due to medical treatment was evaluated in the present study

and was not found to be a confounding variable.

It has been suggested that female hormones may also increase acoustic neuroma

risk, although the evidence for this association is suggestive rather than definitive (6, 19).

One line of evidence suggesting that female hormones play a role in the etiology of

acoustic neuroma is that the incidence is higher in women than in men (1, 6, 8, 9). A

population-based case-control study of brain tumors, including acoustic neuroma, and

menopausal status, found a tendency towards an increased risk of acoustic neuroma for

menopausal women (19). It is suggested that this is perhaps due to the cessation of

estrogen production after menopause or even perhaps due to a protective effect of the

relative elevation of androgen levels after menopause.

2.5.3 Diagnostic delay The majority of acoustic neuroma tumors grow

slowly (20, 21). In our study it is likely that many of the cases had the tumor for several

years before a clinical diagnosis was made. Diagnostic delay is the period between the

appearance of the first symptom and the time that first medical attention is sought.

According to another study, the delay from the first symptom until diagnosis averaged

more than 5 years (22). In these patients the diagnostic delay ranged from 2 to 30 years.

Hence, it is very difficult to predict the growth rate and as a result the latency period of

50

acoustic neuroma. In our analysis, however, we did observe an increased risk of acoustic

neuroma as the latency period was increased (P for trend = 0.0029). An increased risk of

acoustic neuroma was found only with a latency of at least 13 years between exposure

and diagnosis. This is consistent with the hypothesis that for slow-growing tumors such

as acoustic neuroma, one would expect a higher risk with a longer latency period and

lower risk with a shorter latency period.

2.5.4 Acoustic trauma and tumorigenesis The results of the present

study support the hypothesis that acoustic trauma due to loud noise exposure contributes

to tumorigenesis. Damage to the structures of the ear caused by intense sound exposure

appears to be caused by similar mechanisms in all mammals (23). It has been reported

previously that damage to cochlear hair cells produced by acoustic trauma stimulates

mitotic replication of normally postmitotic cells in the chicken and quail (24, 25). The

supporting cells or perhaps unidentified stem cells that replicate as a result of the acoustic

trauma, do not divide in the absence of trauma (24). Experimental studies in rodents have

demonstrated mechanical damage to the organ of Corti and surrounding tissue, including

the eighth cranial nerve, as a result of intense impulse noise (26, 27). Oxidative DNA

damage in the cochlea following intense noise exposure was observed in one

experimental study of rodents (28). If cancer risk is proportional to the number of

proliferating cells, as has been previously postulated (29), then it is plausible that a

benign tumor such as acoustic neuroma may arise as a result of cochlear hair cell trauma.

During the cellular repair process, cellular division results in DNA replication errors

which may in turn lead to chromosomal changes essential for neoplastic transformation

(13).

51

2.5.5 Recall and selection bias The tendency for patients with a tumor to

focus on the reasons that they may have developed the disease is a potential source of

recall bias in our study. As our study is a case-control interview study and as many of the

cases were exposed to loud noise for 5 or more years’ duration, we cannot exclude the

possibility that recall bias occurred. In their search for exposure to a putative risk, the

cases may have focused on occupational exposures, including exposure to loud noise. In

addition, 91 percent of the cases in the study reported unilateral hearing loss at the time

of the interview, which is the primary symptom seen in patients with acoustic neuroma.

This may have made the cases more aware of past loud noise exposures prior to their

diagnosis than the controls, of which only 29 percent reported hearing loss at the time of

the interview.

In the present study there was higher participation rate in the cases than in the

controls, which could have caused selection bias. If willingness to participate was higher

among those controls exposed to loud noise, then the risk of acoustic neuroma would

have been underestimated. However, such selection bias is unlikely, as the primary aim

of the INTERPHONE study was the evaluation of mobile phone use as a possible risk

factor for acoustic neuroma and other brain tumors, not the evaluation of loud noise

exposure. In addition, cases with loud noise exposure and subsequent hearing impairment

may have sought medical attention for their hearing impairment and as a result may have

been diagnosed with acoustic neuroma. This may have resulted in earlier detection of

acoustic neuroma. Such a scenario would be possible in the music industry for example,

in which hearing impairment is likely to be more prevalent than in other occupations.

52

This possible detection bias may have caused an overestimation of the true effect of loud

noise exposure and is an issue that warrants further investigation in future studies.

The standardized face-to-face interviews used in this study decreased the

likelihood of recall bias and provided more reliable answers to detailed questions than

self-administered questionnaires (30). Although the interviewers were not blinded as to

case and control status, the potential for interviewer bias was minimized by the use of a

standard set of questions read verbatim from the laptop computer by trained study

personnel. Also, it has been reported that patients with acoustic neuroma do not generally

have memory deficits and as a result this should not have affected the quality of our data

(22).

2.5.6 Conclusion We conclude that our data support the hypothesis that loud

noise exposure is a risk factor for acoustic neuroma. Further research is needed to

validate self-reports of loud noise exposure and evaluate the effect of potential detection

bias.

2.6 Acknowledgments

I acknowledge funding from the European Union Fifth Framework Program,

“Quality of Life and Management of living Resources” (contract QLK4-CT-1999-

01563), the Swedish Research Council, and the International Union against Cancer

(UICC). The UICC received funds for this purpose from the Mobile Manufacturers’

Forum and GSM Association. Provision of funds to the INTERPHONE study

investigators via the UICC was governed by agreements that guaranteed

53

INTERPHONE’s complete scientific independence. These agreements are publicly

available at http://www.iarc.fr/ENG/Units/RCAd.html.

I wish to thank the regional Cancer Registries and the area hospital clinics for

their collaboration. I also thank the research nurses for their skillful work.

2.7 References

1. Chandler CL, Ramsden RT. Acoustic schwannoma. Br J Hosp Med 1993;49:336-343.

2. Tos M, Charabi S, Thomsen J. Clincial experience with vestibular schwannomas:

epidemiology, symptomatology, diagnosis, and surgical results. 1998;255:1-6. 3. Miller MH, Doyle TJ, Geier SR. Acoustic neuroma in a population of noise

exposed workers. Laryngoscope 1981;91:363-371. 4. Ho SY, Kveton JF. Acoustic neuroma assessment and management. Otolaryngol

Clin North Am 2002;35:393-404. 5. Tos M, Stangerup SE, Cayé-Thomasen P, et al. What is the real incidence of

vestibular schwannoma? Arch Otolaryngol Head Neck Surg 2004;130:216-220. 6. Howitz MF, Johansen C, Tos M, et al. Incidence of Vestibular Schwannoma in

Denmark, 1977-1995. Am J Otol 2000;21:690-694. 7. Tos M, Thomsen J. Epidemiology of acoustic neuromas. J Laryngol Otol

1984;98:685-692. 8. Spoelhof GD. When to suspect acoustic neuroma. Am Fam Physician

1995;52:1768-1774. 9. Inskip PD, Tarone RE, Hatch EE, et al. Sociodemographic indicators and risk of

brain tumours. Int J Epidemiol 2003;32:225-233. 10. Lanser MJ, Sussman SA, Frazer K. Epidemiology, pathogenesis, and genetics of

acoustic tumors. Otolaryngol Clin North Am 1992;25:499-520. 11. Cardis E, Kilkenny M. International case-control study of adult brain, head and

neck tumors: results of the feasibility study. Radiat Prot Dosim 1999;83:179-183.

54

12. Lönn S, Ahlbom A, Hall P, et al. Mobile phone use and the risk of acoustic neuroma. Epidemiology 2004;15:653-659.

13. Preston-Martin S, Thomas DC, Wright WE, et al. Noise trauma in the aetiology of

acoustic neuromas in men in Los Angeles County, 1978-1985. Br J Cancer 1989;59:783-786.

14. Breslow NE, Day NE. Statistical methods in cancer research. The analysis of

case-control studies. Lyon: International Agency for Research on Cancer, 1980. 15. Rajaraman P, De Roos AJ, Stewart PA, et al. Occupation and risk of meningioma

and acoustic neuroma in the United States. Am J Ind Med 2004;45:395-407. 16. Inskip PD, Tarone RE, Hatch EE, et al. Laterality of brain tumors.

Neuroepidemiol 2003;22:130-138. 17. Preston DL, Ron E, Yonehara S, et al. Tumors of the nervous system and pituitary

gland associated with atomic bomb radiation exposure. J Natl Cancer Inst 2002;94:1555-1563.

18. Ron E, Modan B, Boice JD, et al. Tumors of the brain and nervous system after

radiotherapy in childhood. N Engl J Med 1988;319:1033-1039. 19. Schlehofer B, Blettner M, Wahrendorf J. Association between brain tumors and

menopausal status. J Natl Cancer Inst 1992;84:1346-1349. 20. Strasnick B, Glasscock ME, Haynes D, et al. The natural history of untreated

acoustic neuromas. Laryngoscope 1994;104:1115-1119. 21. Valvassori GE, Guzman M. Growth rate of acoustic neuromas. Am J Otol

1989;10:174-176. 22. Thomsen J, Tos M. Acoustic neuroma: clinical aspects, audiovestibular

assessment, diagnostic delay, and growth rate. Am J Otol 1990;11:12-19. 23. Consensus conference: noise and hearing loss. Consensus Development

Conference on Noise and Hearing Loss: JAMA, 1990:3185-90. 24. Corwin JT, Cotanche DA. Regeneration of sensory hair cells after acoustic

trauma. Science 1988;240:1772-1774. 25. Ryals BM, Rubel EW. Hair regeneration after acoustic trauma in adult Coturnix

quail. Science 1988;240:1774-1776.

55

26. Hammernik RP, Turrentine G, Wright CG. Surface morphology of the inner sulcus and related epithelial cells of the cochlea following acoustic trauma. Hearing Res 1984;16:143-160.

27. Chan E, Suneson A, Ulfendahl M. Acoustic trauma causes reversible stiffness

changes in auditory sensory cells. Neuroscience 1998;83:961-968. 28. Van Campen LE, Murphy WJ, Franks JR, et al. Oxidative DNA damage is

associated with intense noise exposure in the rat. Hearing Res 2002;164:29-38. 29. Albanes D, Winick M. Are cell number and cell proliferation risk factors for

cancer? J Natl Cancer Inst 1988;80:772. 30. O'Toole BI, Battistutta D, Long A, et al. A comparison of costs and data quality

of three health survey methods: mail, telephone and personal home interview. Am J Epidemiol 1986;124:317-328.

56

CHAPTER 3

OCCUPATIONAL NOISE EXPOSURE AND RISK OF ACOUSTIC NEUROMA1

3.1 Abstract

Two prior epidemiologic studies of occupational noise exposure based on self-

report have suggested an association with acoustic neuroma. The goal of the present

study was to further examine the association between noise exposure and acoustic

neuroma using an objective measure of exposure in the form of a job exposure matrix. A

total of 793 acoustic neuroma cases aged 21-84 years were identified between 1987 and

1999 from the Swedish Cancer Registry. The 101,756 controls randomly selected from

the study base were frequency matched to cases on age, sex, and calendar year of

diagnosis. Occupational information, available for 599 of the cases and 73,432 of the

controls, was obtained from censuses and linked to a job exposure matrix based on actual

noise measurements. All risk estimates were close to unity, regardless of noise exposure

1Edwards CG, Schwartzbaum JA, Nise G, Forssén UM, Ahlbom A, Lönn S, Feychting

M. Occupational Noise Exposure and Risk of Acoustic Neuroma. Accepted for

publication in American Journal of Epidemiology on June 28, 2007 by permission of

Oxford University Press.

57

level or parameter. The overall odds ratio for exposure to ≥85 decibels of noise was 0.89

(95 % confidence interval: 0.64, 1.23). Contrary to previous study results, the findings do

not demonstrate an increased acoustic neuroma risk related to occupational noise

exposure even allowing for a long latency period. The effect of non-differential

misclassification of exposure must be considered as a potential cause of the negative

findings.

3.2 Introduction

Acoustic neuroma, also referred to as vestibular schwannoma, constitutes from 6-

10 percent of all intracranial tumors, with an incidence of 1-20 per million per year (1-3).

The sex ratio (females/males) has been reported to be >1 (1, 4-6) and the tumor occurs

mainly in individuals aged 50 years or older (1). Although benign, the tumor can cause

significant morbidity due to its location on the vestibular division of the eighth cranial

nerve in the internal auditory canal (4, 7, 8). In the present study we examine unilateral

sporadic acoustic neuroma which comprises 90-95 percent of all acoustic neuromas (1,

9).

The only well-established exogenous risk factor for acoustic neuromas is ionizing

radiation. A link between acoustic neuroma and mobile phone use has been suggested,

although the increased risk, if real, seems to be associated with longer duration of use,

generally 10 or more years of use prior to diagnosis (10, 11). In the present study we

focus on occupational noise exposure as a potential risk factor for acoustic neuroma. Two

previous studies have examined occupational noise and its relationship to acoustic

neuroma with consistent results (12, 13). In the first study elevated risks for acoustic

58

neuroma were found for occupational noise exposure based on self-reported occupational

histories reviewed by an occupational hygienist (13). In the second study elevated

acoustic neuroma risks were detected for self-reported regular exposure to occupational

and nonoccupational noise (12). In both studies a dose-response effect was evident with

increasing years of noise exposure (12, 13). Limitations of the prior studies include their

use of self-reported exposure, as both studies analyzed participant interview data.

The purpose of the present study was to investigate exposure to objectively

measured occupational noise and acoustic neuroma risk in a large register based case-

control study. This would facilitate further examination of the role of noise in acoustic

neuroma etiology.

3.3 Materials and Methods

3.3.1 Study Design and Population A register-based, case-control study

was conducted in which the source population included all residents of Sweden between

1987 and 1999 who were gainfully employed according to any census performed between

1975 and 1990. The study was approved by the Ethics Committee at the Karolinska

Institutet.

3.3.2 Acoustic Neuroma Case Ascertainment Eligible cases included all

patients diagnosed with acoustic neuroma (ICD9 code 1920 and histopathological code

451, classified according to WHO/HS/CANC/24.1 histology code) between 1987 and

1999. We identified 793 cases of acoustic neuroma reported to the national Swedish

Cancer Registry who met these criteria. In Sweden physicians and pathologists must

59

notify the Cancer Registry of every case of acoustic neuroma. In our study, reference year

was defined as the year of acoustic neuroma diagnosis.

3.3.3 Controls Controls were selected randomly from the continuously

updated Swedish Population Registry from among individuals never diagnosed with

acoustic neuroma or other intracranial tumors, pancreatic cancer, or hematological

malignancies (controls were selected simultaneously for studies of the latter two tumor

types, as part of another larger case-control study). On December 31 of each year of the

study, controls were frequency matched on age and sex to cases of acoustic neuroma,

pancreatic cancer, and hematological malignancies diagnosed during that year. The year a

control was selected was used as the reference year for the control. Controls could only

be selected once and cases could not be selected as controls. Registry information

necessary for the analysis was readily available, as all study participants could be linked

to other Swedish registers by means of the national registration number unique to each

individual in Sweden (14). In our study the entire set of controls (n=101,756) was used.

3.3.4 Census Data The study participants’ occupations were obtained

from censuses performed by Statistics Sweden in 1975, 1980, 1985, and 1990. Census

data, collected in September of each census year, provided the occupational codes and

socioeconomic statuses for the study participants. Occupations were categorized using a

three-digit occupational coding system.

3.3.5 Job Exposure Matrix The job exposure matrix (JEM) used in this

study was a cross-classification between numerous occupations and actual noise

measurements taken during different time periods. The noise exposure information used

for construction of the JEM derives from measurements performed from the late 1960s

60

and onward all over Sweden at occupational medicine clinics and occupational health

services units. The exposure estimate for a 5 year period was based on all available

measurements for that period. The estimated decibel (dB) level for each occupation

during each of the 5 year time periods was coded as either <75 dB, 75-84 dB, or ≥85 dB

based on the consensus of three occupational hygienists. This JEM covers 320 different

occupations, with up to four different measurements made per occupation.

3.3.6 Individual Noise Exposure Assessment Instead of depending on self-

reported noise exposure as was done in prior studies, occupational exposure in the present

study was based on knowledge of job titles (15, 16). Information regarding noise

exposure for each study participant was assessed by linking the newly created JEM to

each subject’s occupational code in each of four censuses. Exposure could be determined

by the occupation at each census and by the noise measurement for each occupation

during the pericensal time period. For each occupation, noise measurements were made

for the 5 year periods preceding and following each of the 4 census years. Consequently,

individual exposure data were not available for 1995 to 1999 (the last follow-up period),

as 1990 was the last census from which occupational codes were available. In addition,

only exposure occurring prior to the reference date was considered in the analysis.

Unfortunately, data on noise-induced hearing loss among study participants were not

available for use as an indicator of noise exposure.

Although occupation was reported only every 5 years, if the noise measurement

for an occupation was ≥75 dB for the 5 years before the census, then a study participant

reporting that occupation could be considered exposed at each of the 5 years prior to the

census. Equivalently, if the noise measurement for the 5 year period was after the census,

61

the study participant could be assigned the noise measurement for that occupation for the

5 years including and following the census. However, exposure status for an occupation

could not be based on an average of both the years preceding the census and following it,

as combining exposure statuses for the years preceding and following a census might lead

to further exposure misclassification.

All analyses were all performed initially in two ways, first assigning exposure

before the census and then assigning exposure for the period including and following the

census. Differences in risk estimates and accompanying 95 percent CIs between the two

times of exposure assignment were negligible. Therefore, we arbitrarily selected the dB

measurement categories for the 5 years preceding each census to assess exposure in the

final model.

In the present analysis, ever being exposed to low noise versus never being

exposed was defined as ever holding an occupation with exposure between 75 and 84 dB

based on any census, compared to never having an occupation with noise exposure. Ever

being exposed to high noise versus never being exposed was similarly defined using

exposure of ≥85 dB.

To evaluate the impact of different latency periods we examined the effect of time

since first noise exposure on acoustic neuroma risk. Given the length of the observation

period (1975-1990), the semidecadal censuses, and based on estimates of the latency

period for acoustic neuroma reported in prior studies, latency periods of 5, 10, and 15

years were chosen (12, 17, 18). A 20 year latency period was not used, as there were too

few observations for meaningful analysis. In the analysis, risk estimates for low noise

exposure as well as for high noise exposure were obtained for each of the three latency

62

periods, as well as for no latency period. The reference category for each comparison in

this analysis included those whose first exposure was within the latency period as well as

those never exposed. The latency period analysis assumes implicitly that exposure during

the latency period has no effect on the developing tumor. Finally, the availability of

occupational data from multiple consecutive censuses allowed for the assessment of

exposure duration, in which we investigated the effect of low noise exposure and high

noise exposure at one, two, three, and four consecutive censuses prior in time to the

reference year.

3.3.7 Statistical Analysis Unconditional logistic regression models

adjusted for age, sex, and socioeconomic status were used to estimate odds ratios (OR)

and their respective 95 percent confidence intervals (CI) with SAS, version 9.1, statistical

software (SAS Institute, Inc., Cary, North Carolina) (19). The OR was used as an

estimate of relative risk in the analysis of the JEM data. Age was evaluated for inclusion

in the model as both a categorical and a continuous variable. As there was a negligible

difference between these two representations of age and in order to reduce the number of

variables, age was included as a continuous variable in the final model. Socioeconomic

status was categorized into eight groups as shown in Table 3.1. All statistical significance

tests were two-sided.

3.4 Results

Basic demographic characteristics of cases and controls are presented in Table

3.1. The total number of cases and controls in the study was 793 and 101,756,

respectively. The ages of the cases ranged from 21-84 years. Differences among the

63

demographic variable distributions between cases and controls can be seen for age and

sex because controls were matched on these variables also to types of tumors other than

acoustic neuroma. Of all study participants, less than one percent were missing

information on occupation and approximately two percent were missing information on

socioeconomic status. The JEM enabled 72 percent of the study participants to be

assigned an exposure level, allowing 599 cases and 73,432 controls to be included in the

final analysis. There were no meaningful differences in the distributions of sex, age, or

socioeconomic status between all study participants and those with information on

exposure.

Table 3.2 shows the most frequent occupations with high (≥ 85 dB) noise

exposure according to case status and sex. Men were significantly more often employed

in occupations with high noise ≥85 dB (14 percent of men and 3 percent of women), as

well as in occupations with low (75-84 dB) noise (63 percent of men and 43 percent of

women).

In Table 3.3 the OR for ever being exposed to low noise and the OR for ever

being exposed to high noise were both slightly below unity and the CIs included the null.

Odds ratios are presented for acoustic neuroma according to time since first noise

exposure or latency period. All of the CIs for the three latency periods for both low and

high noise included the null.

Odds ratios for acoustic neuroma are presented in Table 3.4 for exposure to

occupational noise according to one, two, three, or four consecutive censuses closest in

time prior to reference year. The ORs for the four low noise exposure durations were

slightly above unity but all of the corresponding CIs included the null. For the high noise

64

exposure category the ORs were generally at or close to unity and again the CIs all

included the null. As part of the analysis Tables 3.3 and 3.4 were combined (i.e., duration

of exposure stratified on latency period), however the numbers of observations were too

small for meaningful analysis.

65

Table 3.1. Selected characteristics of all study participants and of the study participants with information on exposure, Sweden, 1987-1999.

Cases Controls

Eligible With exposure

information Eligible With exposure

information (n=793) (n=599; 76%) (n=101,756) (n=73,432; 72%) No. % No. % No. % No. % Sex Male 391 49 332 55 58,950 58 47,212 64 Female 402 51 267 45 42,806 42 26,220 36 Number with information on occupation in at least one census 791 99.7 599 100 100,887 99.1 73,432 100 Age at reference date (years) <20 0 0 0 0 703 1 161 0.2 20-39 131 17 85 14 7,419 7 5,144 7 40-59 373 47 306 51 27,099 27 23,032 31 60-79 282 36 205 34 58,545 58 40,527 55 ≥80 7 1 3 1 7,990 8 4,568 6 Socio-economic status based on occupation Unskilled employees in goods and service production 223 28 180 30 24,059 24 19,363 26 Skilled employees in goods and service production 103 13 96 16 11,685 12 10,962 15 Assistant non-manual employees 121 15 73 12 12,785 13 7,230 10 Intermediate non-manual employees 138 17 101 17 12,441 12 8,139 11 Upper-level executives 77 10 50 8 7,910 8 5,299 7 Self-employed professionals 38 5 31 5 5,330 5 4,101 6 Agricultural 23 3 23 4 3,772 4 3,760 5 Miscellaneous and unclassified 66 8 45 8 21,624 22 13,683 19

65

66

Table 3.2. Most frequent occupations with high (≥ 85 dB) noise exposure among cases and controls for men and women, Sweden, 1987-1999.

Cases* Controls Males • Construction carpenters and joiners (n=24) • Construction carpenters and joiners (n=3,778) • Bench carpenters and cabinet makers (n=24) • Forest workers and log-drivers (n=2,554) • Forest workers and log-drivers (n=14) • Bench carpenters and cabinet makers (n=2,308) • Sheet metal workers (n=1,744) • Plumbers and pipe fitters (n=1,503) Females • Other production related work (n=4) • Spinners, weavers, knitters and dyers (n=426) • Sheet metal workers (n=3) • Plastic products workers (n=300) • Butchers and meat preparers (n=2) • Bench carpenters and cabinet makers (n=296) • Spinners, weavers, knitters and dyers (n=2) • Other production related work (n=178) • Canning workers (n=168) * high noise exposure occupations held by more than one case

66

67

Table 3.3. Odds ratios and 95% confidence intervals for acoustic neuroma in relation to noise exposure level according to latency period duration, Sweden, 1987-1999.

<75 dB‡ ≥75-84 dB ≥85 dB Cases Controls Cases Controls OR* 95% CI† Cases Controls OR* 95% CI† No latency period 178 16,761 354 46,783 0.96 0.77, 1.20 67 9,888 0.89 0.64, 1.23 ≥5 year latency period 180 17,083 352 46,515 0.99 0.80, 1.23 67 9,834 0.93 0.67, 1.28 ≥10 year latency period 212 20,789 327 43,378 1.09 0.89, 1.32 60 9,265 0.99 0.72, 1.36 ≥15 year latency period 283 26,929 262 38,363 1.00 0.83, 1.22 54 8,140 1.04 0.76, 1.42 * OR, odds ratio, from unconditional logistic regression analysis, adjusted for age, sex and socioeconomic status † CI, confidence interval ‡ reference category includes study participants with first exposure during the latency period and those never exposed

67

68

Table 3.4. Odds ratios and 95% confidence intervals for exposure to noise at any time (ever) for one, two, three, or four consecutive censuses closest in time prior to reference year, Sweden, 1987-1999.

≥75-84 dB‡ ≥85 dB‡ Cases Controls OR* 95% CI† Cases Controls OR* 95% CI† Exposed in: One census 250 23,740 1.13 0.86, 1.49 25 2,578 0.94 0.58, 1.52 Two censuses 183 16,711 1.18 0.88, 1.58 15 1,496 1.02 0.56, 1.83 Three censuses 131 11,955 1.21 0.88, 1.66 10 1,026 1.00 0.88, 1.66 Four censuses 100 10,006 1.15 0.81, 1.63 9 796 1.17 0.81, 1.63 * OR, odds ratio, from unconditional logistic regression analysis, adjusted for age, sex and socioeconomic status † CI, confidence interval ‡ reference category includes 178 cases and 16,761 controls never exposed to occupational noise (i.e., exposure <75 dB)

68

69

3.5 Discussion

3.5.1 Occupational Noise Exposure and Risk of Acoustic Neuroma The

present study is the third to examine the association between noise exposure and acoustic

neuroma. No evidence was found of an increased acoustic neuroma risk among study

participants working in occupations with noise exposure, regardless of exposure level,

exposure duration, or latency period. This is in contrast to previous studies examining the

role of noise exposure as a possible risk factor for acoustic neuroma that demonstrated an

elevated risk of the tumor with exposure to either occupational or nonoccupational noise,

or both (12, 13).

In the first study to examine noise in the etiology of acoustic neuromas by

Preston-Martin et al. controls were matched to cases aged 25-69 years diagnosed with

acoustic neuroma (13). Self-reported occupational histories were reviewed by an

occupational hygienist to determine if significant noise exposure occurred. The study

found an OR of 2.2 (95 percent CI: 1.12, 4.67) for ever having a job involving exposure

to extremely loud noise. The authors also found a dose-response effect for years of

employment in an occupation with noise exposure (P for trend = 0.02) (13).

In the second study, occupational and nonoccupational noise exposure were

evaluated in a population based case-control study of 146 men and women aged 20-69

years with acoustic neuroma (12). Type and duration of noise exposure were ascertained

through self-report with noise defined as ≥85 dB. Exposure to regular noise from

machines, power tools and/or construction was found to increase acoustic neuroma risk

(OR = 1.79, 95 percent CI: 1.11, 2.89), as did regular exposure to loud music (OR = 2.25,

95 percent CI: 1.20, 4.23). When latency period was examined, an increased acoustic

70

neuroma risk was found only with a latency of at least 13 years between first regular

noise exposure and the time of diagnosis, with an OR of 2.12 (95 percent CI: 1.40, 3.20)

(12).

The first study by Preston-Martin et al. contained a relatively small number of

cases (n=86) and was restricted to men (13). In the second study of both men and women,

data on 146 cases of acoustic neuroma were analyzed, although the number of men in the

study was in fact smaller than the first study (12). It is important to note that while the

total number of cases in the second study (n=146) was considerably less than the present

study (n=599), the statistical power of the second study was in fact higher due to the

larger number of cases (n=74 or 51 percent) exposed to ≥85 dB than in the present study

(n=67 or 11 percent). The negative findings of the present study may therefore be due to

lack of statistical power.

The strengths of our present study include the high quality of data available for

analysis. Data obtained from the Swedish Cancer Registry have high coverage and

accuracy. In excess of 98 percent of histologically confirmed cancers in Sweden are

reported to the Cancer Registry (14). Of all cancer cases reported to the registry,

including both malignant and benign tumors, close to 80 percent are histologically

confirmed (14). Of the acoustic neuromas reported to the Cancer Registry, 99 percent are

histologically confirmed (Swedish Cancer Registry). Although the quality of the registry

is high it is important to note that acoustic neuroma is a benign tumor and may therefore

be diagnosed many years after the tumor has developed or it may never be diagnosed at

all.

71

The census registries cover the whole population and therefore the information on

occupation collected was remarkably complete. The risk of selection bias was minimized

through the random selection of controls from the population and the potential for recall

bias was eliminated through the use of objectively collected occupational data obtained

from censuses. In addition, the JEM included a large number of different occupations and

was based on actual measurements of noise.

3.5.2 Comparison of Results to Prior Study of Self-Reported Noise Exposure and

Potential for Recall Bias The present study was conducted, in part, in an

effort to replicate the results of the second study of self-reported noise exposure and

acoustic neuroma, through the use of a JEM and census data collected independently of

disease (12). Accordingly, the first explanation for the negative findings of the present

study was the effect of recall bias in the second study. The tendency for patients with a

tumor to focus on the reasons that they may have developed the disease was a potential

source of recall bias. At the time of the interview, 91 percent of the cases in the second

study reported unilateral hearing loss. This may have made the cases more aware of past

noise exposures prior to their diagnosis than the controls, of which only 29 percent

reported hearing loss (12).

3.5.3 Diagnostic Delay and the Healthy Worker Survivor Effect Diagnostic

delay is the period between the appearance of the first symptom and the time that first

medical attention is sought. As the majority of acoustic neuroma tumors grow slowly it is

likely that many of the cases had the tumor for several years before a clinical diagnosis

was made (18, 20). According to one study, the delay from the first symptom of acoustic

neuroma until diagnosis averaged more than 5 years, with a range of 2 to 30 years (17).

72

In addition, study participants working in high noise exposure occupations may have

developed hearing loss or tinnitus and consequently may have left their occupation or

transferred to an occupation with lower noise exposure. This potential source of bias is

termed the healthy worker survivor effect (21). Such a situation could potentially explain

the observed odds ratios below unity when no, or a very short latency period was used in

our analyses (22). Additionally, the slow growth of acoustic neuromas may have resulted

in the inclusion in the exposure assessment of a time period during which the tumor was

already present, in contradiction to study methodology whereby etiologically relevant

exposures are only those prior to disease onset. However, the latency calculation was

assigned to address the issue of time between disease onset and diagnosis and

furthermore this method of estimating exposure also eliminates bias that results when

workers leave their occupations due to noise exposure or symptoms of a tumor (the

healthy worker survivor effect).

3.5.4 Misclassification of Exposure and Confounding Another concern is

non-differential misclassification of exposure, whereby exposure is misclassified

similarly among study participants with and without disease (23, 24). JEMs are often

suspected of producing greater non-differential misclassification than do questionnaires

(16). When individuals who perform different tasks in different work environments are

grouped together under the same occupational title, and are classified as exposed or

unexposed depending on whether the probability of exposure exceeds a given threshold

(i.e., either 74 or 84 dB), misclassification will inevitably occur (16, 25, 26). If present,

misclassification may lead to an underestimation of the effect estimate with associated

loss of statistical power (15, 23, 26). Also, in our study nonoccupational noise exposure is

73

not considered. If there exists a true association between noise and acoustic neuroma,

then nonoccupational noise contributes to the risk burden and to exposure

misclassification.

Nevertheless, misclassification may in fact be less than anticipated when defining

exposure as occupations with noise ≥ 85 dB. Occupations with this noise exposure level

are relatively uncommon with only 13 percent of controls classified as such,

corresponding to a high specificity. Therefore, even in the unlikely event that a majority

of study participants who are classified as exposed are misclassified, most participants

classified as unexposed are most likely correctly classified.

It is possible that imprecision of the JEM may be a result of the noise

measurements being taken later than the time of the actual exposures, as the noise

exposures within each occupation may have changed over time (27). In addition, census

data on occupation were obtained every 5 years and reflect the occupational status only at

one point in time, and therefore may not accurately reflect the occupational exposure

between the censuses or if more than one job was held at the same time (26, 28, 29).

There were also a limited number of measurements taken for each occupation. An

additional limitation of the study includes the potential for reporting of acoustic neuromas

to the Cancer Registry as unspecified tumors. However, such misclassification of cases

should be unrelated to exposure.

Finally, as the exposure assessment in the present study was conducted using a

JEM, data on the use of hearing protection were not available. In the second study of

occupational and nonoccupational noise and acoustic neuroma risk, the OR for study

participants exposed to noise with hearing protection was close to the null and therefore

74

these individuals were categorized as unexposed (12). The observed ORs in the present

study that were close to or below the null in the high noise exposure analysis may be

attributable, in part, to the use of hearing protection. At an occupational noise exposure

level of ≥85 dB and during the years of our study the use of hearing protection would

have been commonplace. This, and the aforementioned issues, may contribute to the

imprecision inherent in assigning exposure according to occupational categories in a

study such as ours, that may ultimately attenuate a true effect.

3.5.5 Conclusion In summary, the overall results of the study do not

support the hypothesis that occupational noise exposure is a risk factor for acoustic

neuroma. In the present study we used an objective measure of noise exposure in the

form of a JEM. Yet, because we had no direct measure of each individual's exposure to

noise, but rather used occupational categories to estimate this exposure, the effect of non-

differential misclassification of exposure must be considered as a potential cause of the

negative findings.

3.6 Acknowledgments

The study was funded by the Swedish Council for Working Life and Social Research. I

thank Helena Pettersson for preparation of the data and Kati Maharry, M.A.S. for

assistance with the statistical analysis.

75

3.7 References

1. Howitz MF, Johansen C, Tos M, et al. Incidence of Vestibular Schwannoma in Denmark, 1977-1995. Am J Otol 2000;21:690-694.

2. Tos M, Stangerup SE, Cayé-Thomasen P, et al. What is the real incidence of

vestibular schwannoma? Arch Otolaryngol Head Neck Surg 2004;130:216-220. 3. Tos M, Thomsen J. Epidemiology of acoustic neuromas. J Laryngol Otol

1984;98:685-692. 4. Chandler CL, Ramsden RT. Acoustic schwannoma. Br J Hosp Med 1993;49:336-

343. 5. Spoelhof GD. When to suspect acoustic neuroma. Am Fam Physician

1995;52:1768-1774. 6. Inskip PD, Tarone RE, Hatch EE, et al. Sociodemographic indicators and risk of

brain tumours. Int J Epidemiol 2003;32:225-233. 7. Ho SY, Kveton JF. Acoustic neuroma assessment and management. Otolaryngol

Clin North Am 2002;35:393-404. 8. Tos M, Charabi S, Thomsen J. Clincial experience with vestibular schwannomas:

epidemiology, symptomatology, diagnosis, and surgical results. Eur Arch Otorhinolaryngol 1998;255:1-6.

9. Lanser MJ, Sussman SA, Frazer K. Epidemiology, pathogenesis, and genetics of

acoustic tumors. Otolaryngol Clin North Am 1992;25:499-520. 10. Lönn S, Ahlbom A, Hall P, et al. Mobile phone use and the risk of acoustic

neuroma. Epidemiology 2004;15:653-659. 11. Schoemaker MJ, Swerdlow AJ, Ahlbom A, et al. Mobile phone use and risk of

acoustic neuroma: results of the Interphone case-control study in five North European countries. Br J Cancer 2005;93:842-848.

12. Edwards C, Schwartzbaum J, Lönn S, et al. Exposure to loud noise and risk of

acoustic neuroma. Am J Epidemiol 2006;163:327-333. 13. Preston-Martin S, Thomas DC, Wright WE, et al. Noise trauma in the aetiology of

acoustic neuromas in men in Los Angeles County, 1978-1985. Br J Cancer 1989;59:783-786.

76

14. Mattsson B, Wallgren A. Completeness of the Swedish Cancer Register. Acta Radiologica - Oncology 1984;23:305-313.

15. Bouyer J, Hémon D. Comparison of three methods of estimating odds ratios from

a job exposure matrix in occupational case-control studies. Am J Epidemiol 1993;137:472-480.

16. Bouyer J, Hémon D. Retrospective evaluation of occupational exposures in

population-based case-control studies: general overview with special attention to job exposure matrices. Int J Epidemiol 1993;22:S57-64.

17. Thomsen J, Tos M. Acoustic neuroma: clinical aspects, audiovestibular

assessment, diagnostic delay, and growth rate. Am J Otol 1990;11:12-19. 18. Valvassori GE, Guzman M. Growth rate of acoustic neuromas. Am J Otol

1989;10:174-176. 19. Breslow NE, Day NE. Statistical methods in cancer research. The analysis of

case-control studies. Lyon: International Agency for Research on Cancer, 1980. 20. Strasnick B, Glasscock ME, Haynes D, et al. The natural history of untreated

acoustic neuromas. Laryngoscope 1994;104:1115-1119. 21. Arrighi HM, Hertz-Picciotto I. The evolving concept of the healthy worker

survivor effect. Epidemiology 1994;5:189-196. 22. Richardson D, Wing S, Steenland K, et al. Time-related aspects of the healthy

worker survivor effect. Ann Epidemiol 2004;14:633-639. 23. Brenner H, Savitz DA, Jöckel K-H, et al. Effects of Nondifferential exposure

misclassification in ecologic studies. Am J Epidemiol 1992;135:85-95. 24. Kauppinen T. Assessment of exposure in occupational epidemiology. Scand J

Work Environ Health 1994;20 special issue:19-29. 25. Guénel P, Nicolau J, Imbernon E, et al. Exposure to 50-Hz electric field and

incidence of leukemia, brain tumors, and other cancers among French electric utility workers. Am J Epidemiol 1996;144:1107-1121.

26. Goldberg M, Kromhout H, Guénel P, et al. Job exposure matrices in industry. Int

J Epidemiol 1993;22:S10-15. 27. Forssén UM, Lönn S, Ahlbom A, et al. Occupational magnetic field exposure and

the risk of acoustic neuroma. Am J Ind Med 2006;49:112-118.

77

28. Axelson O. Negative and non-positive epidemiological studies. Int J Occ Med Environ Health 2004;17:115-121.

29. Teschke K, Olshan AF, Daniels JL, et al. Occupational exposure assessment in

case-control studies: opportunities for improvement. Occup Environ Med 2005;59:575-594.

78

CHAPTER 4

ACOUSTIC NEUROMA: HISTOPATHOLOGY, NATURAL HISTORY, AND THE ROLE OF ACOUSTIC TRAUMA IN TUMORIGENESIS

4.1 Abstract

Loud noise exposure has been shown to increase the risk of acoustic neuromas in

two prior studies, however a recent study of occupational loud noise and acoustic

neuromas finds no such association. A mechanism of acoustic neuroma tumorigenesis

during the cellular repair process following acoustic trauma has been proposed, whereby

cellular division results in DNA replication errors which may in turn lead to

chromosomal changes essential for neoplastic transformation. Furthermore, it is known

that deletions on chromosome 22 are the main causal event in acoustic neuromas.

However, the question remains as to whether this proposed mechanism is biologically

plausible. The author conducted an extensive search of the published literature with a

focus on the epidemiology, histopathology, genetics, nerve origin, natural history, and

growth rate of acoustic neuromas, as well as vestibulocochlear nerve pathology and

noise-induced cochlear and vestibular damage in acoustic neuroma pathogenesis, in order

to evaluate the biological plausibility of the proposed hypothesis that loud noise exposure

is a risk factor for acoustic neuromas. The tumor typically involves the vestibular rather

than the acoustic division of the eighth cranial nerve, however intralabyrinthine

79

schwannomas and cochlear nerve schwannomas have been reported in the literature.

Oxidative DNA damage in the cochlea and mechanical damage to the organ of Corti have

been demonstrated in rodents, as well as lesions in the cochlear epithelium of chickens

following intense noise exposure. Additionally, evidence of vestibular damage in rodents

has been demonstrated following acoustic trauma. Based on an extensive review of the

literature the proposed biological mechanism explaining the association between loud

noise exposure and acoustic neuromas is plausible, however further research is needed to

first confirm that such an association exists and second to further elucidate the precise

biological basis for the association.

4.2 Epidemiology

Acoustic neuroma, also referred to as vestibular schwannoma, has an incidence of

1-20 per million per year (1,2) and occurs most often in patients between the ages of 30

and 68 years (3, 4). The sex ratio (females/males) for acoustic neuromas has been

reported to be >1 (4-7). Acoustic neuromas account for 5-10 percent of all intracranial

tumors and comprise 71-90 percent of tumors of the cerebellopontine angle (3, 8). The

only well-established exogenous risk factor for acoustic neuromas is ionizing radiation. It

has been found to increase acoustic neuroma risk among individuals who underwent

radiation treatment of tinea capitis during childhood and who developed an excess of

benign and malignant brain tumors of various histological types, including acoustic

neuromas, and among survivors of the atomic bombings in Japan (9, 10).

A possible association between sex hormones and the growth of acoustic

neuromas has been proposed by researchers, based on both epidemiological and clinical

80

features of the tumor (7). However, evidence demonstrating that hormones influence

acoustic neuroma growth is inconclusive (11). One epidemiological line of evidence

suggesting that female hormones play a role in the etiology of acoustic neuromas is that

the incidence is higher in women than in men (5, 12). The clinical observation that

acoustic neuromas seldom become symptomatic before the onset of puberty supports the

concept that hormones influence tumor growth (13). In addition, increased growth of

acoustic neuromas due to hormonal changes during pregnancy has been described (13,

14). Sporadic reports of estrogen receptors in acoustic neuromas can be found in the

literature (15). However, several studies have demonstrated a lack of estrogen and

progesterone receptors in acoustic neuromas (7, 14, 16, 17).

Another possible risk factor for acoustic neuroma that has been examined at

length in the literature over the past several years is cellular telephone use. A link

between acoustic neuroma and the use of cellular telephones has been suggested,

although the increased risk, if real, seems to be associated with a longer duration of use,

generally 10 or more years of use prior to diagnosis (18, 19). A lesser studied potential

risk factor for acoustic neuromas is exposure to loud noise. Two previous studies have

examined occupational loud noise and its relationship to acoustic neuromas with

comparable results (20, 21).

In the first study, elevated risks for acoustic neuromas were found for

occupational loud noise exposure based on self-reported exposures, as well as self-

reported occupational histories that were reviewed by an occupational hygienist (21). In

the second study elevated risks for acoustic neuromas were found for self-reported

regular exposure to both occupational and nonoccupational loud noise (20). The type of

81

loud noise with the highest risk estimate was regular exposure to loud music with an odds

ratio of 2.25 (95 percent confidence interval: 1.20, 4.23) (20). This is particularly

significant, as the use of in-the-ear technology listening devices such as MP3 players has

increased considerably over the past several years. Therefore, not only is there a risk of

noise-induced hearing loss with loud noise exposure, but this research suggests that there

exists an increased risk of acoustic neuromas. Furthermore, in both studies a dose-

response effect was evident with increasing years of loud noise exposure (20, 21).

4.3 Genetics

The functional loss of a tumor suppressor gene located on the long arm of

chromosome 22 is thought to be the cause of all cases of sporadic and NF2-associated

acoustic neuromas (11, 19). Mutation in the NF2 gene causes loss of expression of the

functional tumor suppressor protein named merlin (for moesin-ezrin-radixin-like protein),

also known as schwannomin (30, 31, 39). A member of the protein 4.1 superfamily,

merlin presents a high degree of amino-acid sequence homology with the ERM (ezrin,

radixin, moesin) proteins that link the cytoskeleton to the cell/plasma membrane (30, 31).

Merlin, a protein of 595 amino acids, plays an important role in intracellular signaling,

modulation of cell motility, and suppression of mitogenesis in schwannoma cells (30, 37,

39). Therefore, it is thought that a loss of merlin expression in Schwann cells leads to

aberrant cell growth and formation of an acoustic neuroma (37).

Chromosome 22 is the second smallest autosome, constituting only two percent of

the total human haploid genome (13). However, multiple studies have suggested that the

long arm of chromosome 22 contains a relatively high number of functional genes and is

82

involved in an unexpectedly high number of inherited and acquired cancers in humans

(13, 33). In patients with unilateral sporadic acoustic neuromas, the disease is the result

of inactivation of both alleles of the NF2 gene by acquired somatic mutations (11, 40).

Deletions on chromosome 22 are the main causal event in unilateral acoustic neuromas,

but point mutations and occasional missense mutations have also been reported (31, 38).

Complete loss of one copy of chromosome 22 is often seen in acoustic neuromas (38,

41). In addition, molecular genetic studies of acoustic neuromas have demonstrated loss

of heterozygosity on chromosome 22 (38, 40).

It is important to note that the genomic stability of acoustic neuromas is

significant, with chromosomal losses being highly specific for chromosome 22 (33, 42).

In a study of 43 acoustic neuroma patient samples the authors found extensive deletions

involving most or all of the long arm of chromosome 22 in most tumors with

chromosome 22 loss (40). An investigation of the tumor suppressor genes VHL, APC,

WT2, and NF1 in these 43 samples suggested that they are not important in the

pathogenesis of acoustic neuromas (40). In addition, a number of studies of the p53 tumor

suppressor gene have demonstrated that it too does not play a central role in acoustic

neuroma tumorigenesis (13, 40). Finally, an analysis of the role of the PTEN tumor

suppressor gene in the tumorigenesis of sporadic acoustic neuromas found that PTEN was

not altered in the 30 tumors studied (43).

Cytogenetic studies of acoustic neuroma have suggested that chromosomes 11,

13, and 19 could be involved in acoustic neuroma tumorigenesis, in addition to

chromosome 22 (38). Also, a study of loss of heterozygosity in 76 acoustic neuroma

tumors showed 10 percent of tumors with gain of 9q34, a gain frequently seen in a

83

variety of solid tumors, including colorectal cancer, prostate cancer, parathyroid

adenomas, lung adenocarcinomas, non-Hodgkin’s lymphomas, and adrenal cortical

tumors (38). However, other than chromosome 22, no consistent regions of interest have

been identified through the analysis of a large number of tumors (38).

The concept of the multistep pathogenesis model whereby tumors result from

mutations in several tumor suppressor genes and proto-oncogenes holds true for benign

tumors, as well as for malignant tumors (40). Therefore, mutation in the NF2 gene may

be only the initial step in a series of events that result in the development of an acoustic

neuroma (13). The identification of the other interacting genes whose expression are

deregulated during tumorigenesis will lead to a more comprehensive understanding of

why NF2 mutation leads to the formation of this tumor (31).

4.4 Histopathology

Acoustic neuromas are histologically benign neoplasms of the

neurolemmal sheath of the eighth cranial nerve (5, 22, 23). It is important to note that

there are two distinct clinical presentations for acoustic neuromas. Approximately 95

percent of diagnosed acoustic neuromas are sporadic, nonfamilial, and unilateral (1, 24).

The remaining four to five percent of patients exhibit bilateral acoustic neuromas

associated with the inherited syndrome named neurofibromatosis type 2 (NF2) (1, 8).

The distal part of the eighth nerve is comprised, in part, by Schwann cells, that

sheath the axons (8, 25). These multivalent neuroectodermal cells are considered

homologous to oligodendroglia of the central nervous system, both of which form and

maintain the myelin sheath (26, 27). It is this portion of the nerve where an

84

overproliferation of Schwann cells leads to the formation of an acoustic neuroma (25).

The tumors are rounded and encapsulated, usually appearing as a single mass, with a

capsule dividing the tumor from the nerves (11, 28). Electron microscopic examination of

the neoplastic Schwann cell reveals a large round or elongated nucleus with a distinctive

nucleolus and finely dispersed chromatin (27). In the extracellular space can be found

numerous collagen units with abnormally long spacing, called Luse bodies (27). The

tumors have a remarkably diffuse yellow appearance and compared to other benign

intracranial tumors acoustic neuromas have the least proliferative status (25, 29).

Additionally, decreased basal apoptosis rates of acoustic neuroma schwannoma cells in

vitro and in vivo have been observed compared to normal Schwann cells (30).

The histological appearance of acoustic neuromas is marked by two prominent

types of tissue. Type A Antoni cells which are composed of a closely packed, cellular,

fibrillary structure, with small spindle-shaped densely staining nuclei, and type B Antoni

cells which are comprised of a less cellular, loosely arranged, reticular structure (8, 25,

29). The same tumor may contain both histological types (12). The loosely arranged

Antoni B tissues may contain extratumoral or intratumoral cysts characteristic of cystic

acoustic neuromas (31).

Acoustic neuromas derive their main arterial blood supply from the basilar

arteries, as well as from the vertebral arteries (8). A relationship between sex and tumor

vascularity has been shown, with fewer blood vessels in tumors of males and twice the

frequency of heavy vascularization in tumors of females (15). In addition, it has been

suggested that angiogenic factors may play a role in tumor growth (1, 32). Additional

factors implicated in the pathogenesis of sporadic and NF2-associated acoustic neuromas

85

are nerve growth factor, glial growth factor, platelet-derived growth factor, basic

fibroblast growth factor, transforming growth factor-β, epidermal growth factor, and

vascular endothelial growth factor, however their roles in the pathogenic process require

further evaluation (25, 31, 33-36). Additionally, neuregulin-1 (NRG-1) and/or

neuregulin-2 (NRG-2) proteins have been suggested as possible growth factors involved

in the promotion of acoustic neuroma tumorigenesis (34, 37). It has also been

demonstrated that the erbB2 and erbB3 membrane tyrosine kinases to which NRG-1 and

NRG-2 bind, are expressed by neoplastic Schwann cells in the majority of acoustic

neuromas (34, 37).

Inflammation is a frequent feature found in histological sections of acoustic

neuromas and inflammatory tumor components can influence tumor size and growth (32).

In a clinicopathologic study of growth factors in 69 acoustic neuroma tumor samples, the

authors of one study found a statistically significant correlation between symptom

duration of longer than one year and degree of inflammation, which is consistent with the

concept that inflammation is a degenerative change (32).

4.5 Pathology of the Vestibulocochlear Nerve

The vestibulocochlear or eighth nerve transmits sensory information essential for

balance and hearing and measures approximately 17-18 mm in men and 16-17 mm in

women (24, 25). The two branches of the nerve are the vestibular and cochlear that

course through the cerebellopontine angle as two distinct nerves (25, 44). The fibers of

86

both nerves originate in separate end organs and have separate central connections,

however they do travel as one nerve through the posterior cranial fossa and internal

auditory canal (45).

The cochlear nerve transmits sensory input regarding hearing and the fibers of the

nerve originate from the hair cells of the organ of Corti located in the cochlea (44). The

vestibular nerve transmits sensory input regarding balance and is usually larger than the

cochlear nerve (44, 45). The vestibular branch has two divisions in the internal auditory

canal, the superior and the inferior (25, 46). Cadaveric investigations have demonstrated

that the superior vestibular nerve is typically larger in size than the inferior vestibular

nerve (44). Nerve fibers of the superior and inferior divisions form a single trunk and

merge with the cochlear nerve in the internal auditory canal (45). Although the cochlear

fibers are smaller, greater in number, and darker staining than the vestibular fibers, the

cleavage plane between the fibers of the vestibular and cochlear nerve is not always clear

(45, 47, 48).

4.6 Nerve Origin of Acoustic Neuromas

The name “acoustic neuroma” is somewhat of a misnomer as it implies

involvement of the acoustic nerve, when in fact the tumor typically involves the

vestibular rather than the acoustic division of the eighth cranial nerve (1, 46, 49). The

embryologically disordered arrangement of sheath cells in the vestibular nerve

predisposes it to develop an acoustic neuroma (50, 51). The tumors are also known as

vestibular schwannomas, as they represent a neoplasia of the Schwann cells on the

87

vestibular nerve (8, 25). As a rule, pure vestibular nerve schwannomas do not infiltrate

the cochlear nerve (52). Only in von Recklinghausen’s disease is infiltration of the

cochlear nerve found (23, 52).

The exact nerve origin of acoustic neuromas has been the subject of a long

standing controversy (53). Although it is now widely accepted that acoustic neuromas

originate from the vestibular nerve rather than the acoustic nerve, the precise origin of the

tumor on the vestibular nerve is the subject of continued debate. It has been reported that

the tumors usually arise on the superior division of the nerve, with fewer tumors arising

on the inferior division (12, 51, 54). In contrast, other reports suggest that the majority of

tumors arise from the inferior vestibular nerve (46, 48, 55, 56). More recent reports,

however, suggest that the tumors arise with equal frequency from the superior and

inferior divisions of the nerve (44, 57).

Acoustic neuromas may also originate from the facial nerve (26). In addition,

several cases of intralabyrinthine acoustic neuromas have been reported in the literature

(49). These are tumors of the vestibular or cochlear nerves that involve the cochlea, the

semicircular canals, the vestibule, or a combination of these inner ear structures (59).

Finally, cochlear nerve schwannomas have been reported extensively in the literature

(59). Although clearly less prevalent than vestibular nerve schwannomas, cochlear nerve

schwannomas are not exceptional (50, 55, 60, 61).

4.7 Natural History

The number of reports regarding the growth of acoustic neuromas is

numerous, yet the natural history of the tumor remains somewhat controversial (2, 62).

88

Of all the questions regarding the natural history of acoustic neuromas in the literature,

the most important involve to the relationship between tumor growth rate, tumor size, and

patient age (63). Growth studies cannot replicate the complete natural history of acoustic

neuromas, yet they can uncover details about tumor growth, growth rates, and growth

patterns during a defined observation period (25). In order to discern the natural history

of acoustic neuromas there must first be an understanding of the environment in which

the disease develops, including the normal anatomy and histology of the vestibular nerve,

as well as the growth factors involved in tumor development (25).

Histopathological studies of temporal bones have shown the prevalence of

acoustic neuromas to be from 0.57 to 2.7% (62, 63). When comparing the clinical

incidence of acoustic neuromas, to the prevalence of occult acoustic neuromas from

histopathological studies of temporal bones, it can be concluded that the vast majority of

tumors that exist are never clinically manifested (62, 64). This is due to the very slow or

arrested growth of acoustic neuromas, with potentially less than one percent of tumors

acquiring enough growth to become clinically active (62, 63). However, the rapid

development of more advanced imaging studies has led to the identification of tumors

that would have previously remained clinically undetectable (62). In a study of 164

patients the average diameter of the tumors gradually decreased from 33 mm in 1980 to

22 mm in 1992 (65). During the whole study period CT scanning was available, however

during the latter years of the study magnetic resonance imaging (MRI) became the gold

standard for imaging acoustic neuromas, thus allowing the detection of tumors with a

diameter of only a few millimeters (65).

89

4.8 Growth Rate

The majority of clinical studies of acoustic neuroma growth rate have used

serial imaging, either by means of MRI or computed axial tomography (CT) imaging (66-

68). Radiologic surveillance is a mode of conservative management involving the serial

assessment of tumor growth by radiologic imaging, in conjunction with periodic clinical

assessment of symptom progression (69). Serial radiologic imaging follow up intervals

for patients who have had surgery, irradiation, or who are undergoing observation may

range from every three to six months initially to every one to two years (1, 11, 66).

However, the intervals between follow up are highly dependent on the patient’s clinical

course (1). It has been suggested that follow up by MRI can be terminated after five

years, as tumor growth occurs only within the first four years after diagnosis (2). Other

studies suggest that when tumor growth occurs it does so within the first year of follow

up (67, 70). Nevertheless, it is important to note that serial imaging follow-up should not

be terminated based solely on tumor quiescence, as some acoustic neuroma patients

experience a delayed onset of tumor growth (69).

In addition to classification of acoustic neuromas by size and location,

classification of the growth rate of a tumor is one of the most useful tools for clinical

management (11). However, the growth of acoustic neuromas, typically measured in mm

per year, is largely unpredictable (11, 62). Some tumors may remain unchanged in size

for many years, some may regress, some may increase in size at a rate of up to 20 mm in

diameter per year, and others may display a variable pattern of growth (2, 11, 62). In a

large series of patients reviewed by Charabi et al., five tumor growth patterns were

90

observed: continuous growth, no measurable growth, no measurable growth followed by

continuous growth, negative growth, and various tumor growth patterns (25, 71).

The percentage of tumors that grow varies widely and has been reported to be

from 15 to 90 percent (2, 63, 72). One factor responsible for the wide variation in growth

rates is the length of the observation period in each study (2, 25). Ascertainment of

growth rate may also depend on the criteria used for the determination of growth (in mm)

(2). Comparison of serial imaging studies is often hindered by the limited numbers of

patients included in each study, the inconsistent inclusion of cystic and NF2 tumors, and

the variation of the imaging modalities used (25, 71). It is also important to note that in

acoustic neuroma growth studies the patients who are assigned to observation only are

selected, with patients receiving immediate surgical or radiation treatment excluded from

such observation groups (73). Yet another problem limiting the evaluation of growth

studies in patients undergoing observation is the absence of a well-defined “endpoint” for

the observation period (25, 71). Many factors can initiate the transition from a period of

observation to surgical intervention, including tumor growth change, histopathologic

change, clinical symptom change, or a change in patient consent for treatment (71).

Monitoring of tumor growth is best achieved with gadolinium-enhanced MRI

(11). Mean annual tumor growth rate can vary from 0.7 mm to 4.8 mm per year (22, 62,

66). Although widely variable, the criterion for determination of tumor growth is often

reported as a change of the largest tumor diameter of more than two mm (2, 66, 69). The

largest anteroposterior and mediolateral dimensions are often used (68, 69, 74). In order

to assess tumor growth most accurately, it has been suggested that the size of the tumor

within the posterior fossa and the degree of penetration into the intracanalicular space

91

should be evaluated (11, 69). Some groups promote tumor volume (in cubic cm) as the

ideal tumor measurement, whereas others find no difference in tumor growth analysis

when using either tumor volume or tumor diameter (24, 62, 75, 76). One further method

of growth rate analysis is tumor volume doubling time (VDT) which represents the

amount of time (in years) it would take for a tumor volume to double if growth were

linear (6, 24). Rapidly growing tumors have a VDT of less than 12 months, moderately

growing tumors have a VDT of 12-36 months, and slow-growing tumors have a VDT of

more than 36 months (6). This is in contrast to volume growth rate for malignant brain

tumors which have a VDT of as short as 20 days (75).

Numerous studies examining various aspects of the growth rate of acoustic

neuromas can be found in the literature. Several studies report no relationship between

tumor size and duration of symptoms (16, 62, 64, 74). Other studies report no correlation

between tumor growth and patient age or follow up length (13, 67, 77). Other studies

have reported no significant correlation between patient age and tumor size (24, 74). In

another study no significant differences in tumor size between males and females were

found (44). As a result, it is not possible to demonstrate any pathoanatomic feature of

acoustic neuroma that correlates with the clinical course of the tumor (24, 71, 78).

In contrast, other studies have found various correlations between tumor growth

rate, tumor size, patient age, duration of symptoms, and patient sex. For example, one

study did find a statistically significant inverse relationship between tumor size and

patient age (17). Other studies have reported a tendency for higher growth rates in larger

tumors and in younger patients (6, 25, 72, 76). Yet another study demonstrated that

patients with shorter duration of symptoms had tumors that statistically grew faster (16,

92

25, 62). In a study examining possible gender differences in the growth rates of acoustic

neuromas a tendency for tumors to grow more rapidly in females than males was

observed (6). Another study of 433 patients with unilateral acoustic neuromas found that

tumor size was larger in the female than in the male group (17). However, even with such

findings no single predictive factor for tumor growth has yet been identified for

widespread clinical use (6, 22, 68, 78).

As a result, when attempting to predict the growth rate and choosing a therapeutic

option for an acoustic neuroma several factors should be taken into account, including

patient age and sex, tumor pathology, hearing level, and tumor size at diagnosis (6, 71).

Insight into the growth rate of a tumor must be ascertained, in part, by serial imaging

studies and the choice of imaging interval cannot be guided by baseline data (24). Finally,

it is important to note that duration of symptoms is uncertain subjective data provided by

the patient and the reproducibility of such data is questionable (25, 32).

Diagnostic delay is the period between the appearance of the first symptom and

the time that first medical attention is sought. Thomsen and Tos have suggested that a

period of one year from the appearance of the first symptom until diagnosis is a situation

in which there is no diagnostic delay (64, 78). Any longer period than this and diagnostic

delay is considered present. The typical early symptom of an acoustic neuroma is gradual

hearing loss which may not interfere enough with the patient’s daily living to justify

medical attention (62). As a result, the failure to seek medical attention when such

symptoms first appear, combined with the failure of primary care physicians or

neurotologists to correctly diagnose the cause of the symptoms, often leads to lengthy

diagnostic delays. The average time from onset of symptoms to clinical diagnosis ranges

93

from approximately 4 to 7.3 years (5, 44, 78). In one study of 233 patients the average

diagnostic delay was 7.1 years, with a range of 2 to 30 years (64).

Earlier studies of Schwann cell-related markers were limited to the

immunohistochemical recognition of S-100 protein (25). Schwannoma cells

characteristically express S-100 protein strongly throughout the tumor (14, 29). More

recently, immunohistochemical investigation using the monoclonal antibody Ki-67 has

been used to study the growth rate of acoustic neuromas (67, 79). During cell

proliferation the nuclear antibody Ki-67 is found on all human cells, whereas it is absent

in cells in the G0 or noncycling phase (62). In one study using the Ki-67 technique two

growth rates were demonstrated in acoustic neuromas (25). In addition, the cellular

proliferative fraction in acoustic neuromas was found to be the least among a group of

benign and malignant brain tumors (25). Another study demonstrated a significant

relationship between Ki-67 counts and duration of symptoms (25, 62). Tumors with a

high proliferative status had a short duration of symptoms, whereas tumors with low

proliferative status had a long duration of symptoms (25). In yet another study the

macroscopic growth of acoustic neuromas in athymic nude mice was correlated with a

high rate of cellular proliferation as expressed by Ki-67 (25).

The proliferating cell nuclear antigen (PCNA) is an auxiliary protein for DNA

polymerase. (25). It accumulates in the nuclei of dividing cells predominantly in the S

phase of the cell cycle (25). Therefore, PCNA has been used as an indicator of

proliferative activity and growth rate of a variety of cancers, including acoustic neuromas

(25, 33). The application of PCNA to determine growth rate in two studies revealed

proliferation of between 0.00 and 8.87 percent and a statistically significant difference

94

between the proliferation in tumors smaller than 3 cm and those larger than 3cm in one

study, and an increased rate of PCNA-positive cells in hyperdiploid tumors compared to

diploid tumors in a second study (25). Another study of 22 cases of acoustic neuroma

demonstrated a relationship between tumor size and the PCNA labeling index (33). The

PCNA index provides an estimate of the percentage of tumor cells studied that are in the

cell cycle (33). Studies have also demonstrated that tumor growth is not uniformly

distributed within a tumor and that proliferation appears to be more active near the

surface or in the capsule compared to the center of the tumor (25, 79).

DNA flow cytometric studies have been utilized as an estimate of the proliferative

activity of acoustic neuromas as determined by the fraction of tumor cells in the DNA

synthetic or S phase of the cell cycle (7, 38). In three studies no correlation was found

between the percentage of cells in the S phase of the cell cycle and tumor size, patient

age, or duration of symptoms (25). Studies have demonstrated a wide range of S phase

fractions suggesting a large variation in the growth potential of the acoustic neuromas

examined (80). Finally, studies have suggested that the G1 regulators of the cell cycle,

cyclin D1 and cyclin D3, play a significant role in the regulation of Schwann cell

proliferation and differentiation (81). The cyclin D1 gene is known to be rearranged,

over-expressed, and/or amplified in a number of human malignancies and cyclin D3 is

known to be over-expressed and/or amplified in some tumors (81).

4.9 Noise-induced mechanical damage of the cochlea

Oxidative DNA damage has been demonstrated in the cochlea of rodents

following intense noise exposure (82, 83). Yamashita et al. (2004) demonstrated that

95

mechanical trauma occurring during loud noise exposure is accompanied by oxidative

stress-induced formation of reactive oxygen species (ROS). These reactive species

directly cause cell damage by destroying DNA, as well as cell membranes. They also act

as signaling molecules for the upregulation of genes involved in apoptosis. Cochlear

damage following intense noise exposure continues to spread for a period of days along

the cochlear turn, even after the noise exposure is terminated. Free radical reactions are

fast and transitory, typically terminating within milliseconds to seconds. Therefore, either

newly formed ROS continue to appear even after the noise exposure has ceased, or other

delayed biochemical events occur that cause cell damage (83). In addition, it has been

suggested by Fujioka et al. that noise-induced cochlear damage may involve an early-

phase inflammatory response. A key regulator in inflammation is the fibroblast that

produces several cytokines. Proinflammatory cytokines are released in a number of

organs following tissue damage. Several reports suggest that proinflammatory cytokines

such as IL-6, IL-1β, and TNF-α may initiate an inflammatory response after loud noise

exposure and thus may be involved in cochlear damage (84).

4.10 Acoustic Trauma in the Pathogenesis of Acoustic Neuroma

A study of intense noise exposure in chickens found that lesions were produced in

the cochlear epithelium (85). Furthermore, it has been reported previously that damage to

cochlear hair cells produced by acoustic trauma stimulates mitotic replication of normally

postmitotic cells in the chicken and quail (85, 86). The supporting cells or perhaps

unidentified stem cells that have been found to replicate as a result of the acoustic trauma,

do not divide in the absence of trauma (85). As a result of this discovery in the avian

96

cochlea the possibility of self-repair of hair cells following acoustic trauma in

mammalian ears should not be ruled out (85).

Experimental studies in rodents have demonstrated mechanical damage to the

organ of Corti and surrounding tissue as a result of intense impulse noise (87-89).

Damage to the structures of the ear caused by intense sound exposure appears to be

caused by similar mechanisms in all mammals (90). In addition, myelin-related Schwann

cell proliferation during Wallerian degeneration has been shown in tissue culture in

response to direct injury to the cells (91). In these tissue models, damaged Schwann cells

were replaced and new basal lamina and myelin were produced (91).

The majority of published studies have focused on cochlear damage following

loud noise exposure, with few studies addressing vestibular damage (92). In a study by

Watanabe et al. (2004), an immunohistochemical examination of the vestibules of guinea

pigs was performed after acoustic stimulation. In the study inducible nitric oxide synthase

(iNOS) was detected in the vestibules of the noise exposed group, but not in the

vestibules of the control group. iNOS generates large amounts of nitric oxide, a free

radical that reacts with superoxides, resulting in damage to surrounding tissues.

Pathological processes such as inflammation are thought to promote the expression of

iNOS. The findings of this research indicate that the expression of iNOS participates in

the pathogenesis of vestibular damage resulting from acoustic trauma (92). Additional

evidence of vestibular damage following acoustic trauma was found in an

electronystagmographic study of 326 men suffering from acoustic trauma, in which

evidence of vestibular injury was found in combination with cochlear damage (93). Also,

inflammation has been shown to be a recurrent hallmark of acoustic neuromas on

97

histological sections, especially in Antoni type B areas, and inflammatory components

have been shown to influence benign tumor growth (32).

If cancer risk is proportional to the number of proliferating cells, as has been

previously postulated (94), then it is plausible that a benign tumor such as AN may arise

as a result of cochlear or vestibular acoustic trauma. During the cellular repair process,

cellular division results in DNA replication errors which may in turn lead to

chromosomal changes essential for neoplastic transformation (20, 21). This hypothesis is

consistent with the common mechanism of tumorigenesis that exists for all acoustic

neuromas. Selective losses of genes on chromosome 22 result in loss of expression of the

NF2 gene product and functional tumor suppressor protein merlin, in combination with

the deregulation of other interacting genes, ultimately resulting in tumor formation (42,

43). In the case of sporadic unilateral acoustic neuromas, loss of genes on chromosome

22 can be viewed as analogous to that of genes on chromosome 11 in Wilms’ tumor and

chromosome 13 in retinoblastoma (42).

4.11 Conclusion

In conclusion, although the majority of acoustic neuromas are found on the

vestibular nerve, intralabyrinthine schwannomas (including cochlear schwannomas), and

cochlear nerve schwannomas have been reported. Two published manuscripts have

suggested a hypothesis for a mechanism of acoustic neuroma tumorigenesis following

cochlear or vestibular acoustic trauma (20, 21). The same hypothesis can be applied to

acoustic neuromas located on the vestibular nerve, as studies have demonstrated

vestibular damage, in addition to cochlear damage, following acoustic trauma (92, 93).

98

To summarize, in NF2-associated acoustic neuromas a patient inherits a

chromosome 22 that has a deleted or mutated NF2 locus and then a random mutation of

the remaining NF2 locus removes the inhibition provided by the NF2 gene product

merlin (42, 43). In patients with unilateral sporadic acoustic neuromas, the disease is the

result of inactivation of both alleles of the NF2 gene by acquired somatic mutations (11,

40). Thus, the mechanism of tumorigenesis acts in accordance with Knudson’s “two hit”

mutation model, identical to the retinoblastoma model of carcinogenesis involving

chromosome 13 (15, 21).

Also, with only a small fraction of all acoustic neuromas demonstrating enough

growth to become clinically active, the identification of such tumors is important.

Features that can potentially distinguish tumors with fast tumor growth from those with

slow or arrested growth include the influence of growth factors, tumor vascularity,

inflammation, basal apoptosis rates, hormones, cystic components, cyclins, and

membrane tyrosine kinases (3, 25, 30-37, 58, 81). When examining risk factors that may

increase the likelihood of tumor diagnosis, the only endogenous or exogenous factors

associated with the features of fast-growing tumors are patient sex and repeated

environmental damage to the vestibulocochlear nerve that may lead to chronic

inflammation, as has been shown with loud noise exposure. Finally, further studies are

needed to confirm that there exists an association between loud noise exposure and

acoustic neuroma, as well as to elucidate the precise biological basis for the proposed

mechanism of tumorigenesis.

99

4.12 References

1. National Institutes of Health consensus development conference statement on acoustic neuroma, December 11-13, 1991. Arch Neurol, 1994:201-207.

2. Stangerup S-E, Caye-Thomasen P, Tos M, Thomsen J. The natural history of

vestibular schwannoma. Otol Neurotol 2006;27:547-552. 3. Hart RG, Davenport J. Diagnosis of acoustic neuroma. Neurosurgery 1981;9:450-

463. 4. Gaughan RK, Harner SG. Acoustic neuroma and pregnancy. Am J Otol

1993;14:88-91. 5. Spoelhof GD. When to suspect acoustic neuroma. Am Fam Physician

1995;52:1768-1774. 6. Ogawa K, Kanzaki J, Ogawa S, Yamamoto M, Ikeda S, Shiobara R. The growth

rate of acoustic neuromas. Acta Otolaryngol 1991;487:157-163. 7. Curley JWA, Ramsden RT, Howell A, Healy K, Lye RH. Oestrogen and

progesterone receptors in acoustic neuroma. J Laryngol Otol 1990;104:865-867. 8. Nager GT. Acoustic Neurinomas. Acta Otolaryngol 1985;99:245-261. 9. Preston DL, Ron E, Yonehara S, et al. Tumors of the nervous system and pituitary

gland associated with atomic bomb radiation exposure. J Natl Cancer Inst 2002;94:1555-1563.

10. Ron E, Modan B, Boice JD, et al. Tumors of the brain and nervous system after

radiotherapy in childhood. N Engl J Med 1988;319:1033-1039. 11. Acoustic neuroma. National Institutes of Health consensus statement online 1991

Dec. 11-13, 1991:1-24. 12. Chandler CL, Ramsden RT. Acoustic schwannoma. Br J Hosp Med 1993;49:336-

343. 13. Lanser MJ, Sussman SA, Frazer K. Epidemiology, pathogenesis, and genetics of

acoustic tumors. Otolaryngol Clin North Am 1992;25:499-520. 14. Klinken L, Thomsen J, Rasmussen BB, Wiet RJ, Tos M. Estrogen and

progesterone receptors in acoustic neuromas. Arch Otolaryngol Head Neck Surg 1990;116:202-204.

100

15. Kasantikul V, Brown WJ. Estrogen receptors in acoustic neurilemmomas. Surg Neurol 1981;15:105-109.

16. Charabi S, Thomsen J, Mantoni M, et al. Acoustic neuroma (vestibular

schwannoma): growth and surgical and nonsurgical consequences of the wait-and-see policy. Otolaryngol Head Neck Surg 1995;113:5-14.

17. Nutik SL, Babb MJ. Determinants of tumor size and growth in vestibular

schwannomas. J Neurosurg 2001;94:922-926. 18. Lönn S, Ahlbom A, Hall P, Feychting M. Mobile phone use and the risk of

acoustic neuroma. Epidemiology 2004;15:653-659. 19. Schoemaker MJ, Swerdlow AJ, Ahlbom A, et al. Mobile phone use and risk of

acoustic neuroma: results of the Interphone case-control study in five North European countries. Br J Cancer 2005;93:842-848.

20. Edwards CG, Schwartzbaum JA, Lönn S, Ahlbom A, Feychting M. Exposure to

loud noise and risk of acoustic neuroma. Am J Epidemiol 2006;163:327-333. 21. Preston-Martin S, Thomas DC, Wright WE, Henderson BE. Noise trauma in the

aetiology of acoustic neuromas in men in Los Angeles County, 1978-1985. Br J Cancer 1989;59:783-786.

22. Mirz F, Jørgensen B, Fiirgaard B, Lundorf E, Pedersen CB. Investigations into the

natural history of vestibular schwannomas. Clin Otolaryngol 1999;24:13-18. 23. Karlan MS, Basek M, Potter GB. Intracochlear neurilemoma. Arch Otolaryngol

1972;96:573-575. 24. Herwadker A, Vokurka EA, Evans DGR, Ramsden RT, Jackson A. Size and

growth rate of sporadic vestibular schwannoma: predictive value of information available at presentation. Otol Neurotol 2005;26:86-92.

25. Charabi S. Acoustic neuroma/vestibular schwannoma in vivo and in vitro growth

models. Acta Otolaryngol 1997;Suppl 530:6-27. 26. Tallan EM, Harner SG, Beatty CW. Does the distribution of Schwann cells

correlate with the observed occurrence of acoustic neuromas? Am J Otol 1993;14:131-134.

27. Manolidis L, Baloyannis SJ. Pathology of acoustic nerve schwannoma in vivo and

in vitro. Acta Otolaryngol 1985;99:280-284.

101

28. Sterkers JM, Perre J, Viala P, Foncin J-F. The origin of acoustic neuromas. Acta Otolaryngol 1987;103:427-431.

29. Kleihues P, Cavenee WK. Pathology and genetics of tumours of the nervous

system. Lyon: International Agency for Research on Cancer, 1997. 30. Utermark T, Kaempchen K, Antoniadis G, Hanemann CO. Reduced Apoptosis

Rates in Human Schwannomas. Brain Pathol 2005;15:17-22. 31. Welling DB, Lasak JM, Akhmametyeva E, Ghaheri B, Chang L-S. cDNA

microarray analysis of vestibular schwannomas. Otol Neurotol 2002;23:736-748. 32. Labit-Bouvier C, Crebassa B, Bouvier C, Andrac-Meyer L, Magnan J, Charpin C.

Clinicopathologic growth factors in vestibular schwannomas: a morphological and immunohistochemical study of 69 tumours. Acta Otolaryngol 2000;120:950-954.

33. Irving RM, Moffat DA, Hardy DG, et al. A molecular, clinical, and

immunohistochemical study of vestibular schwannoma. Otolaryngol Head Neck Surg 1997;116:426-430.

34. Hansen MR, Roehm PC, Chatterjee P, Green SH. Constitutive neuregulin-1/ErbB

signaling contributes to human vestibular schwannoma proliferation. Glia 2006;53:593-600.

35. Sturgis EM, Woll SS, Aydin F, Marrogi AJ, Amedee RG. Epidermal growth

factor receptor expression by acoustic neuromas. Laryngoscope 1996;106:457-462.

36. Cardillo MR, Filipo R, Monini S, Aliotta N, Barbara M. Transforming growth

factor-β1 expression in human acoustic neuroma. Am J Otol 1999;20:65-68. 37. Stonecypher MS, Chaudhury AR, Byer SJ, Carroll SL. Neuregulin growth factors

and their ErbB receptors form a potential signaling network for schwannoma tumorigenesis. J Neuropathol Exp Neurol 2006;65:162-175.

38. Warren C, James LA, Ramsden RT, et al. Identification of recurrent regions of

chromosome loss and gain in vestibular schwannomas using comparative genomic hybridisation. J Med Genet 2003;40:802-806.

39. Moffat DA, Irving RM. The molecular genetics of vestibular schwannoma. J

Laryngol Otol 1995;109:381-384.

102

40. Irving RM, Moffat DA, Hardy DG, Barton DE, Xuereb JH, Maher ER. Molecular genetic analysis of the mechanism of tumorigenesis in acoustic neuroma. Arch Otolaryngol Head Neck Surg 1993;119:1222-1228.

41. Bijlsma EK, Brouwer-Mladin R, Bosch DA, Westerveld A, Hulsebos TJM.

Molecular characterization of chromosome 22 deletions in schwannomas. Genes Chromosom Cancer 1992;5:201-205.

42. Seizinger BR, Martuza RL, Gusella JF. Loss of genes on chromosome 22 in

tumorigenesis of human acoustic neuroma. Nature (London) 1986;322:644-647. 43. Mawrin C, Kirches E, Boltze C, Dietzmann K. PTEN is not altered in sporadic

vestibular schwannomas. Histopathology 2002;40:526-530. 44. Spickler EM, Govila L. The vestibulocochlear nerve. Semin Ultrasound CT MR

2002;23:218-237. 45. Schefter RP, Harner SG. Histologic study of the vestibulocochlear nerve. Ann

Otol Rhinol Laryngol 1986;95:146-50. 46. Komatsuzaki A, Tsunoda A. Nerve origin of the acoustic neuroma. J Laryngol

Otol 2001;115:376-379. 47. Silverstein H. Cochlear and vestibular gross and histologic anatomy (as seen from

postauricular approach). Otolaryngol Head Neck Surg 1984;92:207-11. 48. Forton G, Moeneclaey L, Declau F, Marquet J. The involvement of the cochlear

nerve in neurinomas of the eighth cranial nerve. Arch Otorhinolaryngol 1989;246:156-160.

49. Eldridge R, Parry D. Summary: vestibular schwannoma (acoustic neuroma)

consensus development conference. Neurosurgery 1992;30:962-964. 50. Ayani N, Shea JJ. Asymptomatic schwannoma of the cochlear nerve. Arch

Otolaryngol 1982;108:501-503. 51. Krueger WWO, Kemper JV. Cochlear schwannomas. Skull Base Surg

2000;10:87-88. 52. Perre J, Viala P, Foncin J-F. Involvement of cochlear nerve in acoustic tumours.

Acta Otolaryngol 1990;110:245-252. 53. Pineda A, Feder BH. Acoustic Neuroma: A Misnomer. Am Surg 1967;33:40-43.

103

54. Nordström C-H. Re: "Exposure to loud noise and risk of acoustic neuroma". Am J Epidemiol 2006;164:706.

55. Barbieri M, Bruzzo M, Mora R, Meller R, Chays A, Magnan J. Cochlear

schwannomas. Skull Base 2001;11:241-244. 56. Gersdorff MCH, Decat M, Duprez T, Deggouj N. Intracochlear schwannoma. Eur

Arch Otorhinolaryngol 1996;253:374-376. 57. Khurana VG, Link MJ, Driscoll CLW, Beatty CW. Evolution of a cochlear

schwannoma on clinical and neuroimaging studies. Case report. J Neurosurg 2003;99:779-782.

58. Brackmann DE, Kwartler JA. A review of acoustic tumors: 1983-1988. Am J Otol

1990;11:216-232. 59. Doyle KJ, Brackmann DE. Intralabyrinthine schwannomas. Otolaryngol Head

Neck Surg 1994;110:517-523. 60. Leonetti JP. Cochlear neuromas. Am J Otol 1998;19:499-502. 61. Jorgensen MB. Intracochlear neurinoma. Acta Otolaryngol 1962;54:227-232. 62. Rosenberg SI. Natural history of acoustic neuromas. Laryngoscope 2000;110:497-

508. 63. Yoshimoto Y. Systematic review of the natural history of vestibular schwannoma.

J Neurosurg 2005;103:59-63. 64. Thomsen J, Tos M. Acoustic neuroma: clinical aspects, audiovestibular

assessment, diagnostic delay, and growth rate. Am J Otol 1990;11:12-19. 65. van Leeuwen JPPM, Harhangi BS, Thewissen NPMW, Thijssen HOM, Cremers

CWRJ. Delays in the diagnosis of acoustic neuromas. Am J Otol 1996;17:321-325.

66. Strasnick B, Glasscock ME, Haynes D, McMenomey SO, Minor LB. The natural

history of untreated acoustic neuromas. Laryngoscope 1994;104:1115-1119. 67. Valvassori G, Shannon M. Natural history of acoustic neuromas. Skull Base Surg

1991;1:165-167. 68. Charabi S, Thomsen J, Tos M, Charabi B, Mantoni M, Børgesen SE. Acoustic

neuroma/vestibular schwannoma growth: past, present and future. Acta Otolaryngol 1998;118:327-332.

104

69. Selesnick SH, Johnson G. Radiologic surveillance of acoustic neuromas. Am J

Otol 1998;19:846-849. 70. Valvassori GE, Guzman M. Growth rate of acoustic neuromas. Am J Otol

1989;10:174-176. 71. Charabi S, Tos M, Thomsen J, Charabi B, Mantoni M. Vestibular schwannoma

growth: the continuing controversy. Laryngoscope 2000;110:1720-1725. 72. Fucci MJ, Buchman CA, Brackmann DE, Berliner KI. Acoustic tumor growth:

implications for treatment choices. Am J Otol 1999;20:495-499. 73. Charabi S, Tos M, Thomsen J, Charabi B, Mantoni M. Vestibular schwannoma

growth--long-term results. Acta Otolaryngol 2000;Suppl 543:7-10. 74. van Leeuwen JPPM, Cremers CWRJ, Thewissen NPMW, Harhangi BS, Meijer E.

Acoustic neuroma: correlation among tumor size, symptoms, and patient age. Laryngoscope 1995;105:701-707.

75. Laasonen EM, Troupp H. Volume growth rate of acoustic neurinomas.

Neuroradiology 1986;28:203-207. 76. Niemczyk K, Vaneecloo FM, Lemaitre L, et al. The growth of acoustic neuromas

in volumetric radiologic assessment. Am J Otol 1999;20:244-248. 77. Kawamoto Y, Uozumi T, Kiya K, et al. Clinicopathologic growth factors of

acoustic neuromas. Surg Neurol 1995;43:546-52. 78. Thomsen J, Tos M. Acoustic neuromas. Diagnostic delay, growth rate and

possible non-surgical treatment. Acta Otolaryngol 1988;Suppl 452:26-33. 79. Charabi S, Engel P, Charabi B, et al. Growth of vestibular schwannomas: in situ

model employing the monoclonal antibody Ki-67 and DNA flow cytometry. Am J Otol 1996;17:301-306.

80. Kesterson L, Shelton C, Dressler L, Berliner KI. Clinical behavior of acoustic

tumors. Arch Otolaryngol Head Neck Surg 1993;119:269-271. 81. Neff BA, Oberstien E, Lorenz M, Chaudhury AR, Welling DB, Chang L-S.

Cyclin D1 and D3 expression in vestibular schwannomas. Laryngoscope 2006;116:423-426.

105

82. Van Campen LE, Murphy WJ, Franks JR, Mathias PI, Toraason MA. Oxidative DNA damage is associated with intense noise exposure in the rat. Hearing Res 2002;164:29-38.

83. Yamashita D, Jiang H-Y, Schacht J, Miller JM. Delayed production of free

radicals following noise exposure. Brain Res 2004;1019:201-209. 84. Fujioka M, Kanzaki S, Okano HJ, Masuda M, Ogawa K, Okano H.

Proinflammatory cytokines expression in noise-induced damaged cochlea. J Neurosci Res 2006;83:575-583.

85. Corwin JT, Cotanche DA. Regeneration of sensory hair cells after acoustic

trauma. Science 1988;240:1772-1774. 86. Ryals BM, Rubel EW. Hair regeneration after acoustic trauma in adult Coturnix

quail. Science 1988;240:1774-1776. 87. Hammernik RP, Turrentine G, Wright CG. Surface morphology of the inner

sulcus and related epithelial cells of the cochlea following acoustic trauma. Hearing Res 1984;16:143-160.

88. Chan E, Suneson A, Ulfendahl M. Acoustic trauma causes reversible stiffness

changes in auditory sensory cells. Neuroscience 1998;83:961-968. 89. Hammernik RP, Turrentine G, Roberto M, Salvi R, Henderson D. Anatomical

correlates of impulse noise-induced mechanical damage in the cochlea. Hearing Res 1984;13:229-247.

90. Consensus conference. Noise and hearing loss. JAMA 1990;263:3185-90. 91. Salzer J, Bunge R. Studies of Schwann cell proliferation. I. An analysis in tissue

culture of proliferation during development, Wallerian degeneration, and direct injury. J Cell Biol 1980;84:739-752.

92. Watanabe K-I, Inai S, Hess A, Michel O, Yagi T. Acoustic stimulation promotes

the expression of inducible nitric oxide synthase in the vestibule of guinea pigs. Acta Otolaryngol 2004;553:54-57.

93. Man A, Segal S, Naggan L. Vestibular involvement in acoustic trauma (An

electronystagmographic study). J Laryngol Otol 1980;94:1395-1400. 94. Albanes D, Winick M. Are cell number and cell proliferation risk factors for

cancer? J Natl Cancer Inst 1988;80:772.

106

CHAPTER 5

SUMMARY AND CONCLUSIONS

5.1 Summary of Key Study Findings

A study of loud noise and acoustic neuromas performed in 1989 using self-

reported loud noise exposures and occupational histories reviewed by an occupational

hygienist found an increased risk of acoustic neuromas (1). In an effort to further evaluate

the association between loud noise exposure and acoustic neuromas, a larger study was

performed using self-reported exposure to occupational and nonoccupational loud noise

(2). The second study, the largest study of loud noise and acoustic neuroma risk to date,

also found an increased risk of acoustic neuromas (2). Furthermore, in both studies a

dose-response effect was evident with increasing years of loud noise exposure (1, 2).

In order to validate the self-reports of loud noise exposure used in the second

study, a more objective study of occupational noise exposure and risk of acoustic

neuromas was conducted using a job exposure matrix. The third study, however, did not

demonstrate an increased acoustic neuroma risk related to occupational noise exposure,

even allowing for a long latency period. In summary, two studies using self-reported loud

noise exposure have demonstrated an association between loud noise exposure and

acoustic neuromas, however, one study using a more objective measure of noise exposure

found no such association (1, 2). As a result of these findings, it was suggested that

107

additional studies are needed to confirm that there exists an association between loud

noise exposure and acoustic neuromas. Additionally, a recommendation that further

research is needed to evaluate the effect of potential detection bias was made.

5.2 Summary of the Role of Acoustic Trauma in Acoustic Neuroma Tumorigenesis

As a result of the inconsistent study conclusions, an investigation into the biology

of acoustic neuromas was conducted. The histopathology, natural history, and role of

acoustic trauma in acoustic neuroma tumorigenesis were reviewed in the medical

literature. The hypothesis proposed in the first study and in the study described in

Chapter 2 suggested that following cochlear or vestibular acoustic trauma and during the

cellular repair process, cellular division results in DNA replication errors which may in

turn lead to chromosomal changes essential for neoplastic transformation (1, 2). This

hypothesis is consistent with the common mechanism of tumorigenesis that exists for all

acoustic neuromas whereby selective losses of genes on chromosome 22 result in loss of

expression of the NF2 gene product and functional tumor suppressor protein merlin,

ultimately resulting in tumor formation (3, 4).

However, this proposed mechanism of tumorigenesis has been challenged, as the

primary location of acoustic neuromas is on the vestibular division of the eighth cranial

nerve, not on the acoustic division (5, 6). In a published criticism of the study described

in Chapter 2, the author states that “There is no obvious relation between the stimulus of

sound and the development of a tumor in fairly distant cells surrounding a nerve that has

nothing to do with hearing.” (6). Although it has been shown in the literature that the

majority of acoustic neuromas are found on the vestibular nerve, intralabyrinthine

108

schwannomas, including cochlear schwannomas, and cochlear nerve schwannomas have

been reported (7, 8). It is therefore plausible that acoustic trauma to the labyrinth or

acoustic nerve could contribute to the development of an intralabyrinthine or cochlear

nerve schwannoma. Additionally, in support of the plausibility of this hypothesis, it is

important to note that with the majority of acoustic neuromas located on the vestibular

nerve, studies have nevertheless demonstrated vestibular damage in addition to cochlear

damage, following acoustic trauma (9). Thus, in addition to confirming the association

between loud noise exposure and risk of acoustic neuromas, additional studies are needed

to elucidate the precise biological basis for the proposed mechanism of tumorigenesis.

5.3 Study Limitations and Recommendations for Future Research

The analysis of the first acoustic neuroma data set presented in Chapter 2

reinforced the significance of one of Hill’s criteria for causation, namely biological

plausibility. When conducting epidemiologic research it is important that careful

consideration be given to the biological basis for any proposed causal hypothesis. A

statistically verified association, as was found in the study, must be accompanied by a

sound, biologically plausible hypothesis for the mechanism of tumorigenesis.

In addition, caution must be taken when reporting the findings of epidemiologic

research, as published research is often disseminated rapidly by the media and can have

widespread public health and legal implications. Although consistency of association

increases the likelihood that an association is causal, a demonstrated association merely

supports a proposed hypothesis, it does not prove it. The findings of the study reported in

Chapter 2 were consistent with the findings of the first study evaluating the association

109

between loud noise exposure and acoustic neuromas. This did not, however, prove that

loud noise causes acoustic neuromas, only that further research examining the association

was perhaps warranted.

In the analysis and discussion of the two studies presented, it was necessary to

carefully address the potential weaknesses of the exposure assessment methodologies

used, as well as the effect of potential biases, including confounding. Examples of

potential epidemiologic problems encountered in this research include recall bias,

detection bias, selection bias, non-differential exposure misclassification, and

confounding from the healthy worker survivor effect. Epidemiologic problems such as

non-differential exposure misclassification can cause an underestimation of the effect

estimates, or even a reverse association, as can result from the healthy worker survivor

effect for example (10).

Finally, in the study of occupational and nonoccupational loud noise exposure and

acoustic neuroma risk presented in Chapter 2, the highest risk estimates were found for

regular exposure to loud music, including employment in the music industry (2).

Research is needed on the health effects of exposure to in-the-ear technology devices,

such as MP3 players. Not only is there a risk of noise-induced hearing loss with loud

noise exposure, but this research suggests that there exists an increased risk of acoustic

neuromas. However, with the introduction of MP3 players in the late 1990s, it will be

undoubtedly be many years before a clear picture of the health effects of these popular

devices will emerge.

110

5.4 References

1. Preston-Martin S, Thomas DC, Wright WE, Henderson BE. Noise trauma in the aetiology of acoustic neuromas in men in Los Angeles County, 1978-1985. Br J Cancer 1989;59:783-786.

2. Edwards CG, Schwartzbaum JA, Lönn S, Ahlbom A, Feychting M. Exposure to

loud noise and risk of acoustic neuroma. Am J Epidemiol 2006;163:327-333. 3. Mawrin C, Kirches E, Boltze C, Dietzmann K. PTEN is not altered in sporadic

vestibular schwannomas. Histopathology 2002;40:526-530. 4. Seizinger BR, Rouleau G, Ozelius LJ, et al. Common Pathogenetic mechanism

for three tumor types in bilateral acoustic neurofibromatosis. Science 1987;236:317-319.

5. Komatsuzaki A, Tsunoda A. Nerve origin of the acoustic neuroma. J Laryngol

Otol 2001;115:376-379. 6. Nordström C-H. Re: "Exposure to loud noise and risk of acoustic neuroma". Am J

Epidemiol 2006;164:706. 7. Barbieri M, Bruzzo M, Mora R, Meller R, Chays A, Magnan J. Cochlear

schwannomas. Skull Base 2001;11:241-244. 8. Doyle KJ, Brackmann DE. Intralabyrinthine schwannomas. Otolaryngol Head

Neck Surg 1994;110:517-523. 9. Watanabe K-I, Inai S, Hess A, Michel O, Yagi T. Acoustic stimulation promotes

the expression of inducible nitric oxide synthase in the vestibule of guinea pigs. Acta Otolaryngol 2004;553:54-57.

10. Richardson D, Wing S, Steenland K, McKelvey W. Time-related aspects of the

healthy worker survivor effect. Ann Epidemiol 2004;14:633-639.

111

LIST OF REFERENCES

Adelman S. Identification of acoustic neuroma in noise-exposed workers. Scand Audiol 1983;12:247-250. Albanes D, Winick M. Are cell number and cell proliferation risk factors for cancer? J Natl Cancer Inst 1988;80:772. Arrighi HM, Hertz-Picciotto I. The evolving concept of the healthy worker survivor effect. Epidemiology 1994;5:189-196. Axelson O. Negative and non-positive epidemiological studies. Int J Occ Med Environ Health 2004;17:115-121. Ayani N, Shea JJ. Asymptomatic schwannoma of the cochlear nerve. Arch Otolaryngol 1982;108:501-503. Barbieri M, Bruzzo M, Mora R, Meller R, Chays A, Magnan J. Cochlear schwannomas. Skull Base 2001;11:241-244. Bijlsma EK, Brouwer-Mladin R, Bosch DA, Westerveld A, Hulsebos TJM. Molecular characterization of chromosome 22 deletions in schwannomas. Genes Chromosom Cancer 1992;5:201-205. Bouyer J, Dardenne J, Hémon D. Performance of odds ratios obtained with a job-exposure matrix and individual exposure assessment with special reference to misclassification errors. Scand J Work Environ Health 1995;21:265-271.

112

Bouyer J, Hémon D. Comparison of three methods of estimating odds ratios from a job exposure matrix in occupational case-control studies. Am J Epidemiol 1993;137:472-480. Bouyer J, Hémon D. Retrospective evaluation of occupational exposures in population-based case-control studies: general overview with special attention to job exposure matrices. Int J Epidemiol 1993;22:S57-64. Bouyer J, Hémon D. Studying the performance of a job exposure matrix. Int J Epidemiol 1993;22:S65-71. Brackmann DE, Kwartler JA. A review of acoustic tumors: 1983-1988. Am J Otol 1990;11:216-232. Brenner H, Savitz DA, Jöckel K-H, Greenland S. Effects of Nondifferential exposure misclassification in ecologic studies. Am J Epidemiol 1992;135:85-95. Breslow NE, Day NE. Statistical methods in cancer research. The analysis of case-control studies. Lyon: International Agency for Research on Cancer, 1980. Brookhouser PE. Prevention of noise-induced hearing loss. Prev Med 1994;23:665-669. Cardillo MR, Filipo R, Monini S, Aliotta N, Barbara M. Transforming growth factor-β1 expression in human acoustic neuroma. Am J Otol 1999;20:65-68. Cardis E, Kilkenny M. International case-control study of adult brain, head and neck tumors: results of the feasibility study. Radiat Prot Dosim 1999;83:179-183. Chan E, Suneson A, Ulfendahl M. Acoustic trauma causes reversible stiffness changes in auditory sensory cells. Neuroscience 1998;83:961-968. Chandler CL, Ramsden RT. Acoustic schwannoma. Br J Hosp Med 1993;49:336-343. Charabi S. Acoustic neuroma/vestibular schwannoma in vivo and in vitro growth models. Acta Otolaryngol 1997;Suppl 530:6-27.

113

Charabi S, Engel P, Charabi B, Jacobsen GK, Overgaard J, Thomsen J, Tos M. Growth of vestibular schwannomas: in situ model employing the monoclonal antibody Ki-67 and DNA flow cytometry. Am J Otol 1996;17:301-306. Charabi S, Thomsen J, Mantoni M, Charabi B, Jørgensen B, Børgesen SE, Gyldensted C, Tos M. Acoustic neuroma (vestibular schwannoma): growth and surgical and nonsurgical consequences of the wait-and-see policy. Otolaryngol Head Neck Surg 1995;113:5-14. Charabi S, Thomsen J, Tos M, Charabi B, Mantoni M, Børgesen SE. Acoustic neuroma/vestibular schwannoma growth: past, present and future. Acta Otolaryngol 1998;118:327-332. Charabi S, Tos M, Thomsen J, Charabi B, Mantoni M. Vestibular schwannoma growth: the continuing controversy. Laryngoscope 2000;110:1720-1725. Charabi S, Tos M, Thomsen J, Charabi B, Mantoni M. Vestibular schwannoma growth--long-term results. Acta Otolaryngol 2000;Suppl 543:7-10. Checkoway H, Pearce N, Crawford-Brown DJ. Research methods in occupational epidemiology. 1989. Christensen HC, Schüz J, Kosteljanetz M, Poulsen HS, Thomsen J, Johansen C. Cellular telephone use and risk of acoustic neuroma. Am J Epidemiol 2004;159:277-283. Cocco P, Dosemeci M, Heineman EF. Brain cancer and occupational exposure to lead. J Occup Environ Med 1998;40:937-942. Corwin JT, Cotanche DA. Regeneration of sensory hair cells after acoustic trauma. Science 1988;240:1772-1774. Curley JWA, Ramsden RT, Howell A, Healy K, Lye RH. Oestrogen and progesterone receptors in acoustic neuroma. J Laryngol Otol 1990;104:865-867. Davis FG, McCarthy B, Jukich P. The descriptive epidemiology of brain tumors. Neuroimaging Clin N Am 1999;9:581-594.

114

Doyle KJ, Brackmann DE. Intralabyrinthine schwannomas. Otolaryngol Head Neck Surg 1994;110:517-523. Edwards CG, Schwartzbaum JA, Lönn S, Ahlbom A, Feychting M. Exposure to loud noise and risk of acoustic neuroma. Am J Epidemiol 2006;163:327-333. Edwards CG, Schwartzbaum JA, Lönn S, Ahlbom A, Feychting M. Re: "Exposure to loud noise and risk of acoustic neuroma". Am J Epidemiol 2006;164:706-707. Eldridge R, Parry D. Summary: vestibular schwannoma (acoustic neuroma) consensus development conference. Neurosurgery 1992;30:962-964. Ferner RE. Neurofibromatosis 1 and neurofibromatosis 2: a twenty first century perspective. Lancet Neurol 2007;6:340-351. Feychting M. Health effects of static magnetic fields-a review of the epidemiological evidence. Prog Biophys Mol Biol 2005;87:241-246. Forssén UM, Lönn S, Ahlbom A, Savitz DA, Feychting M. Occupational magnetic field exposure and the risk of acoustic neuroma. Am J Ind Med 2006;49:112-118. Forssén UM, Rutqvist L, Ahlbom A, Feychting M. Occupational magnetic fields and female breast cancer: a case-control study using Swedish population registers and new exposure data. Am J Epidemiol 2005;161:250-259. Forton G, Moeneclaey L, Declau F, Marquet J. The involvement of the cochlear nerve in neurinomas of the eighth cranial nerve. Arch Otorhinolaryngol 1989;246:156-160. Fucci MJ, Buchman CA, Brackmann DE, Berliner KI. Acoustic tumor growth: implications for treatment choices. Am J Otol 1999;20:495-499. Fujioka M, Kanzaki S, Okano HJ, Masuda M, Ogawa K, Okano H. Proinflammatory cytokines expression in noise-induced damaged cochlea. J Neurosci Res 2006;83:575-583.

115

Gaughan RK, Harner SG. Acoustic neuroma and pregnancy. Am J Otol 1993;14:88-91. Gersdorff MCH, Decat M, Duprez T, Deggouj N. Intracochlear schwannoma. Eur Arch Otorhinolaryngol 1996;253:374-376. Glasscock ME, Hart MJ, Vrabec JT. Management of bilateral acoustic neuroma. Otolaryngol Clin North Am 1992;25:449-469. Goldberg M, Imbernon E. The use of job exposure matrices for cancer epidemiology research and surveillance. Arch Public Health 2002;60:173-185. Goldberg M, Kromhout H, Guénel P, Fletcher AC, Gérin M, Glass DC, Heederick D, Kauppinen T, Ponti A. Job exposure matrices in industry. Int J Epidemiol 1993;22:S10-15. Gosztonyi RE. The effectiveness of hearing protective devices. J Occup Med 1975;17:569-580. Grayson J. Radiation exposure, socioeconomic status, and brain tumor risk in the US Air Force: a nested case-control study. Am J Epidemiol 1996;143:480-486. Gridley G, Nyren O, Dosemeci M, Moradi T, Adami H-O, Carroll L, Hoar Zahm S. Is there a healthy worker effect for cancer incidence among women in Sweden? Am J Ind Med 1999;36:193-199. Guénel P, Nicolau J, Imbernon E, Chevalier A, Goldberg M. Exposure to 50-Hz electric field and incidence of leukemia, brain tumors, and other cancers among French electric utility workers. Am J Epidemiol 1996;144:1107-1121. Hammernik RP, Turrentine G, Roberto M, Salvi R, Henderson D. Anatomical correlates of impulse noise-induced mechanical damage in the cochlea. Hearing Res 1984;13:229-247.

116

Hammernik RP, Turrentine G, Wright CG. Surface morphology of the inner sulcus and related epithelial cells of the cochlea following acoustic trauma. Hearing Res 1984;16:143-160. Hanemann CO, Bartelt-Kirbach B, Diebold R, Kämpchen K, Langmesser S, Utermark T. Differential gene expression between human schwannoma and control Schwann cells. Neuropathol Appl Neurobiol 2006;32:605-614. Hansen MR, Roehm PC, Chatterjee P, Green SH. Constitutive neuregulin-1/ErbB signaling contributes to human vestibular schwannoma proliferation. Glia 2006;53:593-600. Hardell L, Mild KH, Carlberg M. Further aspects on cellular and cordless telephones and brain tumors. Int J Oncol 2003;22:399-407. Hardell L, Carlberg M, Mild KH. Pooled analysis of two case-control studies on the use of cellular and cordless telephones and the risk of benign brain tumours diagnosed during 1997-2003. Int J Oncol 2006;28:509-518. Hardell LO, Carlberg M, Söderqvist F, Mild KH, Morgan LL. Long-term use of cellular phones and brain tumors - increased risk associated with use for >10 years. Occup Environ Med 2007. Apr 4; [Epub ahead of print]. Hart RG, Davenport J. Diagnosis of acoustic neuroma. Neurosurgery 1981;9:450-463. Hernán MA, Hernández-Diaz S, Robins JM. A structural approach to selection bias. Epidemiology 2004;15:615-625. Herwadker A, Vokurka EA, Evans DGR, Ramsden RT, Jackson A. Size and growth rate of sporadic vestibular schwannoma: predictive value of information available at presentation. Otol Neurotol 2005;26:86-92. Hilding DA, House WF. "Acoustic neuroma": comparison of traumatic and neoplastic. J Ultrastruct Res 1965;12:611-623.

117

Ho SY, Kveton JF. Acoustic neuroma assessment and management. Otolaryngol Clin North Am 2002;35:393-404. Howitz MF, Johansen C, Tos M, Charabi S, Olsen JH. Incidence of Vestibular Schwannoma in Denmark, 1977-1995. Am J Otol 2000;21:690-694. Inskip P, Tarone RE, Brenner A, Fine HA, Black PM, Shapiro WR, Selker RG, Linet MS. Handedness and risk of brain tumors in adults. Cancer Epidemiol Biomarkers Prev 2003;12:223-225. Inskip PD, Tarone RE, Hatch EE, Wilcosky TC, Fine HA, Black PM, Loeffler JS, Shapiro WR, Selker RG, Linet MS. Sociodemographic indicators and risk of brain tumours. Int J Epidemiol 2003;32:225-233. Inskip PD, Tarone RE, Hatch EE, Wilcosky TC, Selker RG, Fine HA, Black PM, Loeffler JS, Shapiro WR, Linet MS. Laterality of brain tumors. Neuroepidemiol 2003;22:130-138. Inskip PD, Tarone RE, Hatch EE, Wilcosky TC, Shapiro WR, Selker RG, Fine HA, Black PM, Loeffler JS, Linet MS. Cellular-telephone use and brain tumors. N Engl J Med 2001;344:79-86. Irving RM, Moffat DA, Hardy DG, Barton DE, Xuereb JH, Holland FJ, Maher ER. A molecular, clinical, and immunohistochemical study of vestibular schwannoma. Otolaryngol Head Neck Surg 1997;116:426-430. Irving RM, Moffat DA, Hardy DG, Barton DE, Xuereb JH, Maher ER. Molecular genetic analysis of the mechanism of tumorigenesis in acoustic neuroma. Arch Otolaryngol Head Neck Surg 1993;119:1222-1228. Jorgensen MB. Intracochlear neurinoma. Acta Otolaryngol 1962;54:227-232. Karlan MS, Basek M, Potter GB. Intracochlear neurilemoma. Arch Otolaryngol 1972;96:573-575.

118

Kasantikul V, Brown WJ. Estrogen receptors in acoustic neurilemmomas. Surg Neurol 1981;15:105-109. Kauppinen T. Assessment of exposure in occupational epidemiology. Scand J Work Environ Health 1994;20 special issue:19-29. Kawamoto Y, Uozumi T, Kiya K, Fujioka Y, Kurisu K, Kawamoto K, Sugiyama K. Clinicopathologic growth factors of acoustic neuromas. Surg Neurol 1995;43:546-52. Kesterson L, Shelton C, Dressler L, Berliner KI. Clinical behavior of acoustic tumors. Arch Otolaryngol Head Neck Surg 1993;119:269-271. Khurana VG, Link MJ, Driscoll CLW, Beatty CW. Evolution of a cochlear schwannoma on clinical and neuroimaging studies. Case report. J Neurosurg 2003;99:779-782. Kleihues P, Cavenee WK. Pathology and genetics of tumours of the nervous system. Lyon: International Agency for Research on Cancer, 1997. Klinken L, Thomsen J, Rasmussen BB, Wiet RJ, Tos M. Estrogen and progesterone receptors in acoustic neuromas. Arch Otolaryngol Head Neck Surg 1990;116:202-204. Komatsuzaki A, Tsunoda A. Nerve origin of the acoustic neuroma. J Laryngol Otol 2001;115:376-379. Kraus H, Richrath W, Dullweber L, Gerbig D, Prieshof R, Tillmann J. Noise and the RNA synthesis of the cochlea. Acta Otolaryngol 1975;80:230-237. Krueger WWO, Kemper JV. Cochlear schwannomas. Skull Base Surg 2000;10:87-88. Laasonen EM, Troupp H. Volume growth rate of acoustic neurinomas. Neuroradiology 1986;28:203-207.

119

Labit-Bouvier C, Crebassa B, Bouvier C, Andrac-Meyer L, Magnan J, Charpin C. Clinicopathologic growth factors in vestibular schwannomas: a morphological and immunohistochemical study of 69 tumours. Acta Otolaryngol 2000;120:950-954. Lanser MJ, Sussman SA, Frazer K. Epidemiology, pathogenesis, and genetics of acoustic tumors. Otolaryngol Clin North Am 1992;25:499-520. Leonetti JP. Cochlear neuromas. Am J Otol 1998;19:499-502. Li C-Y, Sung F-C. A review of the healthy worker effect in occupational epidemiology. Occup Med 1999;49:225-229. Lönn S, Ahlbom A, Hall P, Feychting M. Mobile phone use and the risk of acoustic neuroma. Epidemiology 2004;15:653-659. Lönn S, Klaeboe L, Hall P, Mathiesen T, Auvinen A, Christensen HC, Johansen C, Salminen T, Tynes T, Feychting M. Incidence trends of adult primary intracerebral tumors in four nordic countries. Int J Cancer 2004;108:450-455. Man A, Segal S, Naggan L. Vestibular involvement in acoustic trauma (An electronystagmographic study). J Laryngol Otol 1980;94:1395-1400. Manolidis L, Baloyannis SJ. Pathology of acoustic nerve schwannoma in vivo and in vitro. Acta Otolaryngol 1985;99:280-284. Mattsson B, Wallgren A. Completeness of the Swedish Cancer Register. Acta Radiologica - Oncology 1984;23:305-313. Mawrin C, Kirches E, Boltze C, Dietzmann K. PTEN is not altered in sporadic vestibular schwannomas. Histopathology 2002;40:526-530. Miller MH, Doyle TJ, Geier SR. Acoustic neuroma in a population of noise exposed workers. Laryngoscope 1981;91:363-371.

120

Mirz F, Jørgensen B, Fiirgaard B, Lundorf E, Pedersen CB. Investigations into the natural history of vestibular schwannomas. Clin Otolaryngol 1999;24:13-18. Moffat DA, Irving RM. The molecular genetics of vestibular schwannoma. J Laryngol Otol 1995;109:381-384. Monoh K, Ishikawa K, Yasui N, Mineura K, Andoh H, Togawa K. p53 tumor suppressor gene in acoustic neuroma. Acta Otolaryngol 1998;Suppl 537. Muscat JE, Malkin MG, Shore RE, Thompson S, Neugut A I, Stellman SD, Bruce J. Handheld cellular telephones and risk of acoustic neuroma. Neurology 2002;58:1304-1306. Nager GT. Acoustic Neurinomas. Acta Otolaryngol 1985;99:245-261. National Institutes of Health consensus development conference statement on acoustic neuroma, December 11-13, 1991. Arch Neurol, 1994:201-207. National Institutes of Health consensus statement online 1991 Dec. Acoustic neuroma. 11-13, 1991:1-24. Navas-Acién A, Pollán M, Gustavsson P, Plato N. Occupation, exposure to chemicals and risk of gliomas and meningiomas in Sweden. Am J Ind Med 2002;42:214-227. Neff BA, Oberstien E, Lorenz M, Chaudhury AR, Welling DB, Chang L-S. Cyclin D1 and D3 expression in vestibular schwannomas. Laryngoscope 2006;116:423-426. Neff BA, Welling DB, Akhmametyeva E, Chang L-S. The molecular biology of vestibular schwannomas: dissecting the pathogenic process at the molecular level. Otol Neurotol 2006;27:197-208. Niemczyk K, Vaneecloo FM, Lemaitre L, Lejeune JP, Skarzynski H, Dubrulle F, Vincent C. The growth of acoustic neuromas in volumetric radiologic assessment. Am J Otol 1999;20:244-248.

121

Noise and hearing loss. Consensus conference. JAMA 1990;263:3185-90. Nordström C-H. Re: "Exposure to loud noise and risk of acoustic neuroma". Am J Epidemiol 2006;164:706. Nutik SL, Babb MJ. Determinants of tumor size and growth in vestibular schwannomas. J Neurosurg 2001;94:922-926. Ogawa K, Kanzaki J, Ogawa S, Yamamoto M, Ikeda S, Shiobara R. The growth rate of acoustic neuromas. Acta Otolaryngol 1991;487:157-163. O'Toole BI, Battistutta D, Long A, Crouch K. A comparison of costs and data quality of three health survey methods: mail, telephone and personal home interview. Am J Epidemiol 1986;124:317-328. Perneczky A. Blood supply of acoustic neurinomas. Acta Neurochir 1980;52:209-218. Perre J, Viala P, Foncin J-F. Involvement of cochlear nerve in acoustic tumours. Acta Otolaryngol 1990;110:245-252. Pfeiffer SE, Sundarraj N, Dawson G, Kornblith PL. Human acoustic neuromas: nervous system specific biochemical parameters. Acta Neuropathol 1979;47:27-31. Pineda A, Feder BH. Acoustic Neuroma: A Misnomer. Am Surg 1967;33:40-43. Plato N, Steineck G. Methodology and utility of a job-exposure matrix. Am J Ind Med 1993;23:491-502. Preston DL, Ron E, Yonehara S, Kobuke T, Fujii H, Kishikawa M, Tokunaga M, Tokuoka S, Mabuchi K. Tumors of the nervous system and pituitary gland associated with atomic bomb radiation exposure. J Natl Cancer Inst 2002;94:1555-1563.

122

Preston-Martin S, Thomas DC, Wright WE, Henderson BE. Noise trauma in the aetiology of acoustic neuromas in men in Los Angeles County, 1978-1985. Br J Cancer 1989;59:783-786. Propp JM, McCarthy BJ, Davis FG, Preston-Martin S. Descriptive epidemiology of vestibular schwannomas. Neuro-oncol 2006;8:1-11. Pyykkö I, Toppila E, Starck J, Juhola M, Auramo Y. Database for a hearing conservation program. Scand Audiol 2000;29:52-58. Rajaraman P, De Roos AJ, Stewart PA, Linet MS, Fine HA, Shapiro WR, Selker RG, Black PM, Inskip PD. Occupation and risk of meningioma and acoustic neuroma in the United States. Am J Ind Med 2004;45:395-407. Rasmussen N, Tribukait B, Thomsen J, Holm L-E, Tos M. Implications of DNA Characterization of Human Acoustic Neuromas. Acta Otolaryngol 1984;406:278-281. Richardson D, Wing S, Steenland K, McKelvey W. Time-related aspects of the healthy worker survivor effect. Ann Epidemiol 2004;14:633-639. Ron E, Modan B, Boice JD, Alfandary E, Stovall M, Chetrit A, Katz L. Tumors of the brain and nervous system after radiotherapy in childhood. N Engl J Med 1988;319:1033-1039. Rosenberg SI. Natural history of acoustic neuromas. Laryngoscope 2000;110:497-508. Rossi G, Robecchi MG, Penna M. Effects of acoustic trauma on Corti's ganglion. Acta Otolaryngol 1976;81:270-277. Ryals BM, Rubel EW. Hair regeneration after acoustic trauma in adult Coturnix quail. Science 1988;240:1774-1776. Salzer J, Bunge R. Studies of Schwann cell proliferation. I. An analysis in tissue culture of proliferation during development, Wallerian degeneration, and direct injury. J Cell Biol 1980;84:739-752.

123

Sans A, Bartolami S, Fraysse B. Histopathology of the peripheral vestibular system in small vestibular schwannomas. Am J Otol 1996;17:326-24. Savitz D, Cai J, van Wijngaarden E, Loomis D, Mihlan G, Dufort V, Kleckner RC, Nylander-French L, Kromhout H, Zhou H. Case-cohort analysis of brain cancer and leukemia in electric utility workers using a refined magnetic field job-exposure matrix. Am J Ind Med 2000;38:417-425. Schefter RP, Harner SG. Histologic study of the vestibulocochlear nerve. Ann Otol Rhinol Laryngol 1986;95:146-50. Schlehofer B, Blettner M, Wahrendorf J. Association between brain tumors and menopausal status. J Natl Cancer Inst 1992;84:1346-1349. Schoemaker MJ, Swerdlow AJ, Ahlbom A, Auvinen A, Blaasaas KG, Cardis E, Christensen HC, Feychting M, Hepworth SJ, Johansen C, Klaeboe L, Lönn S, McKinney PA, Muir K, Raitanen J, Salminen T, Thomsen J, Tynes T. Mobile phone use and risk of acoustic neuroma: results of the Interphone case-control study in five North European countries. Br J Cancer 2005;93:842-848. Seizinger BR, Martuza RL, Gusella JF. Loss of genes on chromosome 22 in tumorigenesis of human acoustic neuroma. Nature 1986;322:644-647. Seizinger BR, Rouleau G, Ozelius LJ, Lane AH, St George-Hyslop P, Huson S, Gusella JF, Martuza RL. Common Pathogenetic mechanism for three tumor types in bilateral acoustic neurofibromatosis. Science 1987;236:317-319. Selesnick SH, Johnson G. Radiologic surveillance of acoustic neuromas. Am J Otol 1998;19:846-849. Sheppard L. Insights on bias and information in group-level studies. Biostatistics 2003;4:265-278. Silverstein H. Cochlear and vestibular gross and histologic anatomy (as seen from postauricular approach). Otolaryngol Head Neck Surg 1984;92:207-11.

124

Spickler EM, Govila L. The vestibulocochlear nerve. Semin Ultrasound CT MR 2002;23:218-237. Spoelhof GD. When to suspect acoustic neuroma. Am Fam Physician 1995;52:1768-1774. Stangerup S-E, Caye-Thomasen P, Tos M, Thomsen J. The natural history of vestibular schwannoma. Otol Neurotol 2006;27:547-552. Sterkers JM, Perre J, Viala P, Foncin J-F. The origin of acoustic neuromas. Acta Otolaryngol 1987;103:427-431. Stonecypher MS, Chaudhury AR, Byer SJ, Carroll SL. Neuregulin growth factors and their ErbB receptors form a potential signaling network for schwannoma tumorigenesis. J Neuropathol Exp Neurol 2006;65:162-175. Strasnick B, Glasscock ME, Haynes D, McMenomey SO, Minor LB. The natural history of untreated acoustic neuromas. Laryngoscope 1994;104:1115-1119. Sturgis EM, Woll SS, Aydin F, Marrogi AJ, Amedee RG. Epidermal growth factor receptor expression by acoustic neuromas. Laryngoscope 1996;106:457-462. Svensson E, Morata T, Nylén P, Krieg E, Johnson A-C. Beliefs and attitudes among Swedish workers regarding the risk of hearing loss. Int J Audiol 2004;43:585-593. Tallan EM, Harner SG, Beatty CW. Does the distribution of Schwann cells correlate with the observed occurrence of acoustic neuromas? Am J Otol 1993;14:131-134. Teschke K, Olshan AF, Daniels JL, De Roos AJ, Parks CG, Schulz M, Vaughan TL. Occupational exposure assessment in case-control studies: opportunities for improvement. Occup Environ Med 2005;59:575-594. Thomsen J, Tos M. Acoustic neuroma: clinical aspects, audiovestibular assessment, diagnostic delay, and growth rate. Am J Otol 1990;11:12-19.

125

Thomsen J, Tos M. Acoustic neuromas. Diagnostic delay, growth rate and possible non-surgical treatment. Acta Otolaryngol 1988;Suppl 452:26-33. 't Mannetje A, Kromhout H. The use of occupation and industry classifications in general population studies. Int J Epidemiol 2003;32:419-428. Tos M, Charabi S, Thomsen J. Clincial experience with vestibular schwannomas: epidemiology, symptomatology, diagnosis, and surgical results. Eur Arch Otorhinolaryngol 1998;255:1-6. Tos M, Stangerup SE, Cayé-Thomasen P, Tos T, Thomsen J. What is the real incidence of vestibular schwannoma? Arch Otolaryngol Head Neck Surg 2004;130:216-220. Tos M, Thomsen J. Epidemiology of acoustic neuromas. J Laryngol Otol 1984;98:685-692. Tos M, Thomsen J, Charabi S. Incidence of acoustic neuromas. Ear Nose Throat J 1992;71:391-393. Utermark T, Kaempchen K, Antoniadis G, Hanemann CO. Reduced Apoptosis Rates in Human Schwannomas. Brain Pathol 2005;15:17-22. Valvassori G, Palacios E. Growth rate of acoustic neuromas. Ear Nose Throat J 1998;77:373. Valvassori G, Shannon M. Natural history of acoustic neuromas. Skull Base Surg 1991;1:165-167. Valvassori GE, Guzman M. Growth rate of acoustic neuromas. Am J Otol 1989;10:174-176. Van Campen LE, Murphy WJ, Franks JR, Mathias PI, Toraason MA. Oxidative DNA damage is associated with intense noise exposure in the rat. Hearing Res 2002;164:29-38.

126

van Leeuwen JPPM, Cremers CWRJ, Thewissen NPMW, Harhangi BS, Meijer E. Acoustic neuroma: correlation among tumor size, symptoms, and patient age. Laryngoscope 1995;105:701-707. van Leeuwen JPPM, Harhangi BS, Thewissen NPMW, Thijssen HOM, Cremers CWRJ. Delays in the diagnosis of acoustic neuromas. Am J Otol 1996;17:321-325. Warren C, James LA, Ramsden RT, Wallace A, Baser ME, Varley JM, Evans DG. Identification of recurrent regions of chromosome loss and gain in vestibular schwannomas using comparative genomic hybridisation. J Med Genet 2003;40:802-806. Watanabe K-I, Inai S, Hess A, Michel O, Yagi T. Acoustic stimulation promotes the expression of inducible nitric oxide synthase in the vestibule of guinea pigs. Acta Otolaryngol 2004;553:54-57. Welling DB, Lasak JM, Akhmametyeva E, Ghaheri B, Chang L-S. cDNA microarray analysis of vestibular schwannomas. Otol Neurotol 2002;23:736-748. Wesseling C, Pukkala E, Neuvonen K, Kauppinen T, Boffetta P, Partanen T. Cancer of the brain and nervous system and occupational exposures in Finnish women. J Occup Environ Med 2002;44:663-668. Yamashita D, Jiang H-Y, Schacht J, Miller JM. Delayed production of free radicals following noise exposure. Brain Res 2004;1019:201-209. Yoshimoto Y. Systematic review of the natural history of vestibular schwannoma. J Neurosurg 2005;103:59-63.