s1042368008000144.pdf

Neurosurg Clin N Am 19 (2008) 207–216

Conservative Management of Acoustic Neuromas

Julian M. Nedzelski, MD, FRCS(C) * ,

David A. Schessel, PhD, MD, Andrew Pﬂeiderer, MB, FRCS(E),

Edward E. Kassel, MD, FRCP(C), David W. Rowed, MD, FRCS(C)

Sunnybrook Health Science Centre, and the University of Toronto, Toronto, Ontario, Canada

Improved diagnostic screening, most notably

auditory evoked brain stem response testing and

more sophisticated and generally available imag-

ing as well as a better informed population, have

resulted in increasing numbers of smaller and

less symptomatic acoustic neuromas being diag-

nosed [1] .

Advances in surgical technique, anesthesia,

and perioperative care have dramatically reduced

both operative mortality and morbidity. Although

the mortality associated with acoustic neuroma

excision was on the order of 80% in the early

1900s, this ﬁgure has been reduced to approxi-

mately 1% in most current series. Lower cranial

nerve palsy and signiﬁcant residual trigeminal

hypesthesia is now a rare and generally isolated

event [2,3] .

As a result, several authors have suggested that

all tumors, with few exceptions, should be

removed following diagnosis [4,5] . Others have

proposed that a more expectant attitude is an ac-

ceptable alternative to immediate surgical excision

[6–9] .

A conservative (non–tumor excision) strategy

is based on the premise that surgical removal may

pose a greater risk. It assumes that given the

nature of acoustic neuroma clinical behavior, ie,

pattern of growth, these tumors will not result in

either mortality or signiﬁcant morbidity within

the expected life span of a given individual.

The ability to predict a tumor’s potential for

growth would clearly be of beneﬁt in the assess-

ment of such a treatment strategy. A number of

studies have been undertaken to answer this

question. Studies of individuals with acoustic

neuromas have been carried out to deﬁne their

rate and pattern of growth. Overall tumor growth

is generally considered slow [10] . Annual rates of

growth of 0.2 cm per year or less have been noted

in the majority of cases followed [11,12] . Impor-

tantly 40% or more of tumors studied had no

growth, and in several cases, demonstrable tumor

shrinkage occurred [13–15] . Very importantly,

however, several studies have unequivocally iden-

tiﬁed a percentage of patients in whom tumor

growth is relatively rapid, i.e., exceeding 0.2 cm

per year [16,17] .

A number of patients have been successfully

followed for up to 10 years without appreciable

change in symptoms [7] . Such studies conﬁrm the

feasibility of adopting a conservative course of

management in selected patients. If a strategy

could be developed that reliably predicts the

future growth of acoustic neuromas, a more ratio-

nal approach in the management of these patients

would evolve. A conservative policy that ulti-

mately results in a large tumor, causing signiﬁcant

disability in an older patient who has been fol-

lowed for years, is undesirable.

Most clinical studies have used serial imaging

as a method of determining tumor growth.

Annual tumor growth rate has in many instances

become the predominant factor in considering

future tumor enlargement [6] . No statistically sig-

niﬁcant correlation between tumor size and pa-

tient age at the time of presentation has been

found [7,15] .

This article originally appeared in the Otolaryngologic

clinics of NA: Vol 25, Issue 3, June 1992; p. 691–706.

* Corresponding author. Department of Otolaryngol-

ogy, SunnybrookHealth Science Center, 2075 BayviewAv-

enue, Suite A-217, Toronto, Ontario, Canada M4N-3M5.

doi:10.1016/j.nec.2008.02.013

neurosurgery.theclinics.com

208

NEDZELSKI et al

Other parameters have been used to predict

tumor growth. Attempts have been made to

correlate the percentage of tumor cells undergoing

mitosis to the clinical course of aﬀected individ-

uals. The use of 5-bromo-deoxyuridine followed

by immunolocalization has supported the clinical

impression of the overall slow growth of acoustic

tumors [18] . Flow cytometric study has deﬁned

a variable mitotic rate, suggesting a variable

growth pattern [19] . Immunohistochemical study,

using the monoclonal antibody KI-67, has re-

vealed a high mitotic rate associated with a faster

growing tumor [20] . In the clinical studies pub-

lished to date, however, no correlation was found

between presenting tumor size and the rate of cell

proliferation. In one study, a number of tumors

studied with ﬂow cytometry were found to have

proliferative potential equivalent to some malig-

nant tumors. The usefulness of such a comparison

is unclear, given the inability to correlate this lab-

oratory ﬁnding to actual tumor behavior. Clearly

factors other than actual cellular turnover rates

inﬂuence changing tumor size. Hemorrhage, cystic

degeneration, and scarring also play a role.

If a nontreatment management strategy is to

be a serious alternative in the management of

patients with acoustic neuromas, a deﬁnition of

selection criteria, guidelines as to follow-up, and

clinical assessment and imaging parameters are

crucial. In an attempt to answer some of these

concerns, a prospective study was initiated at

Sunnybrook Health Science Centre, University

of Toronto, in 1978, to follow a select group of

mostly older patients with unilateral, previously

untreated acoustic neuromas.

Table 1

Rationale for conservative management (n ¼ 50)

Rationale

Age R 65 years 60–64 50–59

Age

Age and medical

problems

Medical problems

Refusal

or better hearing ear d

reasons for selecting a conservative course of

management. Note that of the patients under

age 65, extenuating circumstances and refusal

were responsible for the decision. Of particular in-

terest are two individuals, one age 50 with an

acoustic tumor in the only-hearing ear (the other

ear having been deafened following mastoid sur-

gery), and one age 60, who had severe Meniere’s

disease in the opposite ear.

In patients selected or who themselves have

opted for conservative management of their tumor,

a follow-up policy has been adopted that includes

regular full neurotologic examination together

with serial CT scanning at 6-month intervals. In

those patients followed for a number of years in

whom there has been no clinical or radiologic

evidence of tumor growth, the follow-up interval

has been extended to 1 year. Conventional con-

trast-enhanced CT examination of the posterior

cranial fossa using either 5-mm or 1.5-mm slices

through the internal auditory canal has been the

most frequently used method to monitor tumor

growth during the follow-up period. In two in-

dividuals with lesions conﬁned to the internal

auditory canal, either air/CT meatography or

more recently magnetic resonance imaging (MRI)

has been used to monitor tumor growth. MRI was

used to monitor one such patient in this series.

All scans were assessed, and all measurements

related to tumor size were carried out by one

neuroradiologist (EEK). The radiologist was un-

aware of the patients’ clinical status and future

management plans. All measurements were made

more than once to validate their accuracy. In this

study, the reliability coecient obtained was

0.998, for simple replication. A value of 1 repre-

sents 100% repeatability/accuracy.

Tumor size would be best expressed by an

exact measurement of tumor volume [17] . Because

this has not been possible, a conservative method

has been adopted whereby the mean of the

maximum anteroposterior and mediolateral

Study group

Four hundred seventy-four patients with

acoustic neuromas were seen at Sunnybrook

Health Science Centre, between the years 1976

and 1991. A conservative approach (non–tumor

excision) was adopted in 56 individuals with

unilateral acoustic neuromas. All patients with

recurrent or persistent tumors following previous

surgery as well as those with bilateral tumors have

been excluded. Of the 56 patients in whom

conservative management was embarked on, six

were excluded for lack of good quality computed

tomography (CT) images.

Our study group therefore consisted of 50

patients, 34 of whom were female and 16 male,

with an age range of 50 to 83 years (mean ¼ 68.1).

Thirty-nine were 65 or over. Table 1 details the

Tumor in only

CONSERVATIVE MANAGEMENT OF ACOUSTIC NEUROMAS

209

dimensions of the cerebellopontine angle mass is

used to represent tumor size. As described, inter-

nal auditory canal tumor content was not in-

cluded in the measurement. The method and

rationale for use of this technique are described

in detail in a previous publication [7] . All patients

included in this study had at least two CT scans

(range 2 to 18), of sucient quality to allow accu-

rate measurements to be made. The period of

follow-up in each case includes only the interval

from the ﬁrst to the last scan from which accurate

measurement was possible. Clinical follow-up out-

side these limits during which scans were judged

to be of poor quality has been excluded. Although

this has inevitably led to a shortened follow-up

interval (mean follow-up 41.7 months; range 7

to 152 months), a more accurate assessment of

tumor growth has been ensured. For instance,

the individual followed for the longest time in

our study had a CT scan over 2 years earlier,

which because of its poor quality rendered it inad-

missible. This reduced the follow-up period in her

instance from 181 to 152 months. Twenty-eight

patients have been followed for 3 or more years,

with 10 for more than 5 years.

Data Analysis

Annual tumor growth rate was calculated in

two ways ( Fig. 1 ). Method 1 uses the initial and

ﬁnal tumor size divided by the number of years

of follow-up. Thus it does not take variability of

growth within the total follow-up time into

account. Method 2 accounts for growth variabil-

ity. It is the mean of the individual annual growth

rates obtained between each interval of assess-

ment during the follow-up period.

Tumor size was measured to 0.001 cm using

a micrometer. This value was rounded oﬀ to the

nearest 0.01 cm (0.1 mm). Although the repro-

ducibility of these measurements was demon-

strated, for reasons of clinical signiﬁcance,

a tumor was judged to have changed in size only

if the total growth was 0.1 cm or more. The

calculated growth rate, however, was the total

growth divided by the number of years of follow-

up and as such could be less than 0.1 cm per year.

In an eﬀort to determine whether tumor

growth is constant or signiﬁcantly variable in the

same individual, a measure of change in growth

rate between consecutive follow-up intervals was

Fig. 1. Calculation of tumor growth rate. (Method 1) Growth rate is calculated as the diﬀerence between the ﬁrst and

last measurement and expressed in cm/yr; (Method 2) growth rate is calculated as the mean of individual measurements

taken between the ﬁrst and last study.

210

NEDZELSKI et al

also calculated (growth variability). This was

determined in 30 of the 50 patients. Small growth

variability values represent constant growth,

whereas large changes indicate signiﬁcant vari-

ability, whether it be an increase or decrease in

size ( Fig. 2 ). Multiple regression analysis was also

used to determine whether age or initial tumor

size was related to subsequent growth.

nine patients, the tumor was noted to decrease

in size (range of –0.51 to –0.01 cm per year).

The CT images in Fig. 4 illustrate a reduction in

tumor size from 2.2 cm to 1.05 cm, a 1.15 cm de-

crease over a 5.5 year follow-up period. Seventeen

patients had no measurable change in tumor size.

The interval CT scans of one such individual,

followed for nearly 13 years (presently age 85),

is shown in Fig. 5 . Measurable tumor growth

occurred in 24 patients (range 0.01 to 0.98 cm

per year). The others are as listed in Table 2 .

The maximum growth measured was 1.15 cm

over a 14-month interval in a 66-year-old woman.

Tumor size increased from 1.45 cm to 2.6 cm

( Fig. 6 ). Although less dramatic, in terms of abso-

lute size, another patient, age 65, realized a three-

fold increase over a 9-month period (0.38 to 1.0 cm)

( Fig. 7 ). In the patients followed for more than

3 years and more than 5 years, the growth rates

were þ 0.04 cm per year and –0.02 cm per year,

respectively.

Results

Growth rate

Using method 1, the mean annual growth rate

for the study group was determined to be 0.11 cm

per year, whereas method 2, which took into

account variations in yearly growth, was 0.11 cm

per year. All growth rates subsequently referred to

are those obtained by method 1, which is easier to

calculate. These rates range between an actual

reduction in tumor size of 0.51 cm per year to an

increase of 0.98 cm per year. Fig. 3 illustrates the

distribution of growth rates of tumors for all

patients. Note that the majority of patients dem-

onstrate little, if any, growth. These ﬁndings are

summarized in Table 2 . Fully 78% of individuals

had a growth rate of less than 0.2 cm per year. In

Growth variability

The mean variation in growth rate for the

30 tumors studied is 0.13. The distribution,

illustrated in Fig. 8 , indicates that the majority

Fig. 2. Examples of growth rate patterns illustrating signiﬁcant diﬀerences in the variability of growth. (Left) Constant

rate of growth in a fast and slow (bottom line) growing tumor; (Right) a highly variable rate of growth in a single tumor.

CONSERVATIVE MANAGEMENT OF ACOUSTIC NEUROMAS

211

Fig. 3. Distribution of annual growth rates of study group (N ¼ 50). Note that 0 growth includes rate changes between

0.099 and þ 0.099 cm/y. (See Table 2 for summary.)

of tumors demonstrate little or no variation in

their growth rate, that is to say, a constant pattern

of growth occurs regardless of the absolute rate.

48.2 months. Thirty-four are known to be alive and

well, of whom 29 remain currently under review.

The other ﬁve have been contacted by phone but

decline to attend further clinical andCT evaluation.

Apart froma progression of their unilateral hearing

loss, all patients remain asymptomatic.

One patient has died during the follow-up

period. Death was from causes unrelated to the

acoustic tumor.

Other factors aﬀecting growth

Multiple regression analysis showed that there

was no relationship between either the age of the

patient or the initial tumor size with the annual

growth rate. Of interest, a signiﬁcant correlation

was noted between patient age and tumor size.

That is, the older the patient, the larger the tumor

at presentation.

Surgical intervention (n ¼ 2). Two patients have

had a ventriculoperitoneal shunt inserted because

of the development of associated signiﬁcant hy-

drocephalus (mean tumor size 2.6 cm). In one

patient, this was performed shortly after presen-

tation. The shunt required revision 6 years later

because of recurrent ataxia associated with in-

creasing hydrocephalus. Clinical symptoms with

associated hydrocephalus necessitated a shunt in

the second patient 2 years after diagnosis, in spite

of no obvious tumor enlargement.

Outcome of follow-up

Patients with tumor growth less than 0.2 cm per

year (n ¼ 39)

Table 3 lists the outcome of the 39 patients in

whom tumor growth has been less than 0.2 cm

per year.

No surgical intervention (n ¼ 37). Of the 39

patients, 37 have not required any formof operative

intervention. The mean follow-up for this group is

Patients with growth rate greater than 0.2 cm per

year (n ¼ 11)

The outcome of patients with tumor growth

exceeding 0.2 cm per year is listed in Table 4 .

Table 2

Annual growth rate of tumors (n ¼ 50)

Measurable reduction with R 0.1 cm net

change

% 0.1 cm/y

4 (8%)

No surgical intervention (n ¼ 2). Despite a docu-

mented tumor growth of more than 0.2 cm per

year, no surgery has been carried out to date on

two patients. One has refused surgery (age 68,

tumor presently 2.7 cm, growth rate 0.26 cm per

year); the other underwent stereotactic irradiation.

O 0.1 cm/y

5 (10%)

No growth

17 (34%)

Growth with R 0.1 cm net change

% 0.1 cm/y

7 (14%)

O 0.1 cm/y

6 (12%)

Surgical intervention (n ¼ 9). Nine patients have

undergone planned translabyrinthine removal of

R 0.2 cm/y

11 (22%)

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika: