Consevative Rx of An

7/27/2019 Consevative Rx of An

1/6


2/6

importance of an accurate assessment of the growth ofthe tumor has led some authors to calculate volumerather than to observe the diameter of the neuroma. Somehave calculated the volume of the tumor using amathematical formula (8), and others have calculatedvolume using specific software (9). Even though volu-metric calculation of tumor growth is theoretically more

accurate, the choice of the most efficient method re-mains debatable.In this study, the authors present the selection criteria

for conservative management of AN, the follow-up, andthe outcome in these patients. They also describe theresults of the radiologic assessment and the follow-up ofthese patients.

MATERIAL AND METHODS

Study groupBetween January 1989 and December 1998, a population of

282 patients with posterior fossa tumor, including AN, werefollowed up or treated in our department. Among them, 179

patients (64%) were operated on, 6 patients (2%) underwentradiation therapy, and 97 (34%) patients were initially ob-served. Before patients were included in the latter group, all theadvantages and disadvantages of all the existing methods forthe treatment of AN were fully explained to them.

The group whose tumors were managed conservatively hada mean age of 63 years (range 2989) and included 34 men and63 women. There were 4 patients aged 70 years. One patient had bilateral ANcaused neurofibromatosis type 2. The length of follow-upranged from 4 months to 10 years (mean 31 months). The initialpresenting symptoms are given in Table 1. Hearing loss was themost common initial symptom, followed by vertigo and tinni-tus. In 9% of the patients, sudden deafness inaugurated thesymptoms. Twenty-eight patients had moderate deafness (meanpure-tone average [PTA] 030 dB), 48 patients had moderate tosevere deafness (PTA 3070 dB), 12 patients had severe deaf-ness (PTA 7090 dB), and 9 patients had total deafness (PTA>90 dB).

The reasons for choosing initial conservative treatment arelisted in Table 2 and are detailed here. In 28 patients (29%),general anesthesia was contraindicated because of general dis-ease; 24 patients (25%) refused surgery, and 18 (19%) elderlypatients (>70 years) were treated conservatively because oftheir age. In 9 patients, AN was asymptomatic; in 5 patients, wefound a local contraindication to surgery; in 3 patients, thetumor was found in the only hearing ear. Other reasons arelisted in Table 2 for the remaining 10 patients.

The clinical stage of AN was determined according to Koos

classification ((10)) for the 97 patients (98 tumors): stage I(intracanalicular tumor with a longitudinal diameter of 110mm): 30 AN; stage II (intracanalicular and intracisternal tumorwith a longitudinal diameter 20 mm): 47 AN; stage III (in-trameatal and intracisternal tumor with a longitudinal diameter30 mm): 13 AN; and stage IV (intrameatal and intracisternaltumor with a longitudinal diameter 30 mm): 8 AN.

The mean size of the AN in the population was 12 mm; itwas 10.37 mm in patients aged 60 years. Stage I and II AN in patients aged >60

years occurred in 62% of the population.The conservative management consisted in clinical follow-up every 6 months with pure-tone and speech audiometry, andMRI at 1 year. Follow-up MRI was obtained every 1 or 2 years.

AN growth rate analysisAmong the untreated patients, 87 underwent at least 2 neu-

roradiologic examinations of the cerebellopontine angle. If apatient underwent radiation therapy or surgery, the latest radio-logic examination taken into account was the one performedbefore the procedure. The radiographic study performed for theinitial and follow-up examinations varied with the evolution ofimaging techniques for the diagnosis of AN. Among these 87patients, size measurement of the tumor was based primarily onMRI and secondarily on computed tomography (CT). The im-aging measurements were made by the same observer (C.C.).

In our center, MRI is done with a Siemens (Paris, France)Magnetom Vision 1.5 Tesla. Sequences were acquired accord-ing to the following protocols: axial T1-weighted images, axialT1-weighted gadolinium-enhanced images with saturation offat (Fat-Sat), axial T2-weighted images, three-dimensionalFourier-transform constructive interference in the steady state(CISS 3D), and flash 3D for the large tumors. A correlationbetween two MRI was performed in 84%, between an initialCT and a MRI in 14%, and between two CT in 2%. The meaninterval between the initial and control radiologic examinationswas 1.3 years.

Volume measurementTumor size was calculated by estimating the mean of the

greatest anteroposterior and mediolateral tumor extent as de-

scribed by Nedzelski et al. ((7)) and therefore was not estimatedonly by the larger diameter of the tumor, including its in-trameatal root (Figs. 1 and 2):

Tumor size (AP + ML)/2.

The growth per annum was calculated by dividing the changein tumor size (mm) by the follow-up period (year). Tumorgrowth pattern was determined for all patients for whom threeneuroradiologic exams were performed.

Of these 87 patients, we also studied the volume of the tumorof 6 patients, for whom the mean interval between the first and

TABLE 1. Main initial symptom (N = 97)

Symptom No. (%)

Progressive hearing loss 49 (51)Vertigo 19 (20)Tinnitus 16 (16)Sudden deafness 9 (9)Unknown 2 (2)Trigeminal neuralgia 1 (1)Retroauricular pain 1 (1)

TABLE 2. Reasons for selecting initialconservative management

Reason No. of patients (N 97)

Contraindication to general anesthesia 28Patient preference 24Age 18Asymptomatic tumor 9

Local contraindication to surgery 5Tumor in the only hearing ear 3Other reasons 10

858 Y. J. SHIN ET AL.

The American Journal of Otology, Vol. 21, No. 6, 2000


3/6

second examinations was 13 months (range 323) and betweenthe second and third examinations 18 months (range 427).

Tumor volume was measured by determining the area of thetumor with a computer program. In this method, the outline ofthe tumor was drawn on each slice (Fig. 1). We averaged eachslice and obtained an average area measurement. The averagedslice area was multiplied by slice interval to obtain a slicevolume, and slice volumes were added to obtain a tumor vol-ume (Fig. 2). The growth per annum was calculated by dividingthe change in tumor volume (mm3) by the follow-up period(year).

We compared the results of this volume measurement madein our center with a method based on mathematical formulas asreported by Charabi et al. (8).

Statistical analysisStatistical analysis was performed classic bivariate tests: un-

paired and paired Students t tests, one-way analysis of vari-

ance (ANOVA), and Pearsons correlation coefficient. We alsoused Students test, the Mann-Whitney U test, and Spearmansrank correlations for the other parameters. The level of signifi-cance was 5%.

RESULTS

Clinical outcomeOf the 97 patients studied, 6 required surgery and 6

required stereotactic radiotherapy. Surgical interventionwas required for above-average growth of the tumor.Among the irradiated group, one of the patients who

benefited from gamma-knife surgery was later operatedon because his tumor grew rapidly (growth 5 mm).Evolution was stable for three patients after a follow-upperiod ranging from 3 to 6 years. For two other patients,a slight increase (


4/6

VolumeIn six patients, we compared the volume calculation

from mathematical formulas based on anteroposteriorand mediolateral measures as reported by Charabi et al.(8), with the volume measurement based on a computerprogram. These two results were not similar, especiallyfor large AN. A paired comparison of the means using arepeated measures ANOVA adjusted on patients madeevidence of higher values given by the volume calculated

with mathematical formulas (p 0.028) with no signifi-cant interaction. This difference between the two serieswas confirmed by a nonparametric Wilcoxons ttest (p 60 years. For Charabi etal. (8), Bederson et al. (12), and Ogawa et al. (21), thesefigures are higher in a younger population. Our studycould not support these findings. It would be interestingto compare this growth rate in a homogeneous popula-tion (stages III) of young patients with a homogeneouspopulation (stages III) of elderly patients.

Difficulties of the conservative managementThe first major difficulty is the nonobservance of this

management. In our series, 19 patients (20%) did notreturn to their clinical/radiologic evaluation appoint-ments. Although this nonobservance has not been re-ported by other authors, we think it represents a majordifficulty of this policy. All the parameters for long-termfollow-up are gathered in the organization of our center:a specific register is filled, and data are gathered in asoftware program. This part of the follow-up is per-formed by one of the authors (I.G.). There are twoweekly specific consultations led by the senior author tomonitor these patients. If a patient does not come to the

appointment, we write and call the patient for anotherappointment. Some reasons for this nonobservance maybe the following: this is an elderly population; despite adetailed explanation, the absence of classic treatment(surgery, radiotherapy) may lead patients to obtain a sec-ond opinion at another center that applies a differentpolicy; or patients with few symptoms may tend to waituntil their hearing worsens. This is a major concern,because we have demonstrated that 16% of the patientswith a grade I or II neuroma lost their candidacy duringthe follow-up.

Another difficulty is to find the appropriate and accu-rate means for evaluating tumor growth. Since the reportby Wazen et al. (11), two-dimensional measurement on

C T o r M RI h as b ee n the f ir st metho d u se d ( 47,11,12,16). More recently, volume measurement of thetumor was proposed as an alternative to the two-dimensional measurement. Interest in volume measure-ment stems from the fact that the relationship betweendiameter and volume in AN is not equivalent to the re-lationship of a spherical tumor with the same diameters.In tumors with large diameters, small increases in diam-eter correspond to much larger increases in tumor vol-ume than would have occurred with similar increases indiameter of small tumors. However, the choice of themost efficient method for volume calculation is stillquestionable. We analyzed and compared volume calcu-lation with mathematical formulas and volume measure-

ment from the area of the tumor on all slices in sixpatients. This comparison was not performed for thewhole population because this technique was introducedfor the last patients included in the study period andrequired at least three examinations. A paired compari-son of the means, by repeated measures ANOVA ad-justed on patients, gave evidence of higher values givenby the volume calculated with mathematical formulas (p

0.028). This means that we can consider that thevolume calculation method presents a trend toward gen-

erating higher values, regardless of the patient. The dif-ference was more evident for large tumors, as seen inpatients D. and V. in Table 5. This finding, althoughmade in a small population, stresses the importance ofusing a single and unique method of volume measure-ment for the follow-up of one patient. Moreover, bothmethods are time-consuming for the neuroradiologist to

calculate and therefore have to be evaluated in terms ofcost efficiency. Computerized assessment of tumor vol-ume based on semiautomated volume calculation withMRI could improve this limitation. Finally, calculationof the systemic error and accuracy should be perfectlydelineated, as Long et al. (18) and Niemcyzyk et al. (20)did. Those authors reported 25% and 15% margin oferror, which are still relatively high figures for theseslow-growing tumors. As Sarazin et al. (9) have said, thehigher the tumoral volume, the lower the margin of error.These latter authors also confirmed a 15% margin oferror. Niemczyck et al. (20) state that methods usingadditional computer software to three-dimensionalanalysis should be more accurate. Such estimations,

though they may be true theoretically, are so far prelimi-nary, and more studies involving larger populations andcost-efficiency data are necessary to validate computer-assisted methods.

Finally, in our practice and at the current time, theauthors think that three-dimensional volume measure-ment from the shape of the tumor is the most efficientand reliable method for assessing tumor growth. Al-though still hampered by a relatively high margin oferror, the methods used seem to be more convincing (9).

The third difficulty is to apply a policy based on theevolution of clinical signs and radiologic findings. In ourseries, surgery or radiotherapy performed because of tu-mor growth was required in 12 patients (12%). This fig-

ure is lower than in other reports because the decision tomodify our attitude is not only dependent on small ra-diologic variations of volume of the AN. Bederson et al.(12) reported that 19% of their patients required surgicalintervention an average of 14 months after the patientinitially sought treatment. Strasnick et al. (14) andCharabi et al. (8) reported indications for surgery or ra-diotherapy in 24% and 34%, respectively, of their pa-tients.

Our decision to modify the management is not basedonly on the overall growth of the tumor, becausewe think that the anatomical relations of the AN withthe neighboring structures, its anterior extension, andthe initial growth rate of the tumor have to be taken

into account. Finally, decision-making factors alsoinclude modification of symptoms and the patientslife expectancy.

CONCLUSION

Conservative management of AN is accompanied bythese difficulties: long-term follow-up of patients, unpre-dictability of the tumor growth pattern, and correlationsbetween clinical and radiologic findings. A reliable and

861CONSERVATIVE MANAGEMENT OF ACOUSTIC NEUROMAS



6/6

reproducible radiologic method is of great importance inevaluating tumor size and growth. Further studies areneeded to establish the Accuracy and reliability of radio-logic evaluation of tumor size.

Acknowledgment: The authors thank Professor A. Bonafe,Dr. M. V. Cadene, and Ms. N. Iked for their contributions tothis study and Mr. M. Ardente and Mr. M. Bouamra for manu-

script preparation.

REFERENCES

1. Deguine O, Maillard A, Bonafe A, el Adouli H, Tremoulet M,Fraysse B. Pre-operative and per-operative factors conditioninglong-term facial nerve function in vestibular schwannoma surgerythrough translabyrinthine approach. J Laryngol Otol 1998;112:4415.

2. Sekhar L, Gormley W, Wright D. The best treatment of vestibularschwannoma (acoustic neuroma): microsurgery or radiosurgery?

Am J Otol 1996;17:67682.3. Shin YJ, Fraysse B, Sterkers O, Bouccara D, Rey A, Lazorthes Y.

Hearing restoration in posterior fossa tumor. Am J Otol 1998;19:64953.

4. Silverstein H, McDaniel A, Norrell H, Wazen J. Conservative

management of acoustic neuroma in the elderly patient. Laryngo-scope 1985;95(7 Pt 1):76670.

5. Wiet RJ, Young NM, Monsell EM, OConnor CA, Kazan R. Ageconsiderations in acoustic neuroma surgery: the horns of a di-lemma. Am J Otol 1989;10:17780.

6. Gardner G, Moretz WH, Robertson JH, et al. Non-surgical man-agement of small and intracanalicular acoustic tumors. Otolaryn-gol Head Neck Surg 1986;94:32833.

7. Nedzelski JM, Canter RJ, Kassel EE, Rowed DW, Tator CH. Is notreatment good treatment in the management of acoustic neuromasin the elderly? Laryngoscope 1986;96:8259.

8. Charabi S, Thomsen J, Mantoni M, et al. Acoustic neuroma (ves-tibular schwannoma): growth and surgical and nonsurgical conse-quences of the wait-and-see policy. Otolaryngol Head Neck Surg1995;113:514.

9. Sarazin L, Jolivet O, Doyon D. Computerized evaluation of tumorvolume with MRI: applications to the surveillance of acoustic neu-rinoma. J Radiol 1993;74:45560.

10. Koos WT, Spetzler RT, Pendl G, Perneczy A, Lang L, eds. Coloratlas of microsurgery. Stuttgart: Georg Thieme Verlag, 1985:188.

11. Wazen J, Silverstein H, Norrell H, Besse B. Preoperative andpostoperative growth rates in acoustic neuromas documented withCT scanning. Otolaryngol Head Neck Surg 1985;93:1515.

12. Bederson JB, von Ammon K, Wichmann WW, Yasargil MG. Con-

servative treatment of patients with acoustic tumors. Neurosurgery1991;28:64650; discussion 6501.

13. Nedzelski JM, Schessel DA, Pfleidera A, et al. Conservative man-agement of acoustic neuroma. Otolaryngol Clin North Am 1991;25:691704.

14. Strasnick B, Glasscock ME III, Haynes D, McMenomey SO, Mi-nor LB. The natural history of untreated acoustic neuromas. La-ryngoscope 1994;104:11159.

15. Sterkers O, Badr Eldine M, Martin N, Viala P, Sterkers J, eds. Slowversus rapid growing acoustic neuroma. Amsterdam: Kugler,1992:1457.

16. Luetje CM, Whittaker CK, Davidson KC, Vergara GG. Spontane-ous acoustic tumor involution: a case report. Otolaryngol Head

Neck Surg 1988;98:957.

17. Rosenberg SI, Silverstein H, Gordon MA, Flanzer JM, WillcoxTO, Silverstein J. A comparison of growth rates of acoustic neu-

romas: nonsurgical patients versus subtotal resection. OtolaryngolHead Neck Surg 1993;109(3 Pt 1):4827.

18. Long SA, Arriaga M, Nelson RA. Acoustic neuroma volume:MRI-based calculations and clinical implications. Laryngoscope1993;103:10936.

19. Fraysse B, Delisle MB, Cognard C, et al. In: Acoustic schwanno-mas: correlations between magnetic resonance imaging and histo-pathologic appearance. 2nd International Conference on AcousticTumors, Rome, 1999.

20. Niemczyk K, Vaneecloo FM, Lemaitre L, et al. The growth ofacoustic neuromas in volumetric radiologic assessment. Am J Otol1999;20:2448.

21. Ogawa K, Kanzaki J, Ogawa S, Yamamoto M, Ikeda S, ShiobaraR. The growth rate of acoustic neuroma. Acta Otolaryngol 19911991;487(suppl):15763.

862 Y. J. SHIN ET AL.


Documents

Consevative Rx of An