Methylprednisolone, valacyclovir, or the combination for vestibular neuritis

The new england journal of medicine or the Combination for Vestibular Neuritis Michael Strupp, M.D., Vera Carina Zingler, M.D., Viktor Arbusow, M.D., Daniel Niklas, Klaus Peter Maag, M.D., Ph.D., Marianne Dieterich, M.D., Sandra Bense, M.D., Diethilde Theil, D.V.M., Klaus Jahn, M.D., b a c k g r o u n d
Vestibular neuritis is the second most common cause of peripheral vestibular vertigo.
(M.S., V.C.Z., V.A., D.N., D.T., K.J., T.B.) Its assumed cause is a reactivation of herpes simplex virus type 1 infection. Therefore, and Epidemiology and Biometrics (K.P.M.), corticosteroids, antiviral agents, or a combination of the two might improve the out- University of Munich, Munich; and the De-partment of Neurology, University of Mainz, come in patients with vestibular neuritis.
Mainz (M.D., S.B.) — both in Germany.
Address reprint requests to Dr. Strupp at the Department of Neurology, Universityof Munich, Klinikum Grosshadern, Mar- We performed a prospective, randomized, double-blind, two-by-two factorial trial in chioninistr. 15, 81377 Munich, Germany, or which patients with acute vestibular neuritis were randomly assigned to treatment with at [email protected].
placebo, methylprednisolone, valacyclovir, or methylprednisolone plus valacyclovir.
Vestibular function was determined by caloric irrigation, with the use of the vestibular Copyright 2004 Massachusetts Medical Society. paresis formula (to measure the extent of unilateral caloric paresis) within 3 days afterthe onset of symptoms and 12 months afterward.
Of a total of 141 patients who underwent randomization, 38 received placebo, 35methylprednisolone, 33 valacyclovir, and 35 methylprednisolone plus valacyclovir.
At the onset of symptoms there was no difference among the groups in the severity ofvestibular paresis. The mean (±SD) improvement in peripheral vestibular function atthe 12-month follow-up was 39.6±28.1 percentage points in the placebo group,62.4±16.9 percentage points in the methylprednisolone group, 36.0±26.7 percentagepoints in the valacyclovir group, and 59.2±24.1 percentage points in the methylpred-nisolone-plus-valacyclovir group. Analysis of variance showed a significant effect ofmethylprednisolone (P<0.001) but not of valacyclovir (P=0.43). The combination ofmethylprednisolone and valacyclovir was not superior to corticosteroid monotherapy.
c o n c l u s i o n s
Methylprednisolone significantly improves the recovery of peripheral vestibular func-tion in patients with vestibular neuritis, whereas valacyclovir does not.
Downloaded from nejm.org at BOSTON UNIVERSITY on June 7, 2012. For personal use only. No other uses without permission. Copyright 2004 Massachusetts Medical Society. All rights reserved. m e t h y l p r e d n i s o l o n e , v a l a c y c l o v i r , o r t h e c o m b i n a t i o n f o r v e s t i b u l a r n e u r i t i s these treatments in patients with vestibular neu- most common cause of peripheral vestib- ritis, in whom we assessed vestibular function at v ular vertigo (the first being benign parox- baseline and the change after 12 months.
ysmal positional vertigo). It accounts for 7 percentof the patients who present at outpatient clinics specializing in the treatment of dizziness1 and has
an incidence of about 3.5 per 100,000 population.2 patients
The key signs and symptoms of vestibular neuritis Patients 18 through 80 years of age were recruited
are the acute onset of sustained rotatory vertigo, from emergency departments in two hospital cen-
postural imbalance with Romberg’s sign (i.e., falls, ters specializing in the diagnosis and treatment of
with the eyes closed, toward the affected ear), hori- vertigo, at the University of Munich and the Univer-
zontal spontaneous nystagmus (toward the unaf- sity of Mainz, between January 1, 1998, and June
fected ear) with a rotational component, and nau- 30, 2002. All patients underwent complete neuro-
sea. Caloric testing (irrigation of the ear with warm logic, neuro-ophthalmologic, and neuro-otologic
or cold water) invariably shows ipsilateral hypo- examination as well as electronystagmography
responsiveness or nonresponsiveness.
(including caloric irrigation), neuro-orthoptic ex- In the past, either an inflammation of the ves- amination (which provides detailed measurement tibular nerve3-5 or labyrinthine ischemia6 was pro- of eye movements), cranial magnetic resonanceposed as a cause of vestibular neuritis. Currently, imaging, laboratory testing, and measurement ofa viral cause is favored. The evidence, however, re- blood pressure and heart rate. The study was ap-mains circumstantial.1,7,8 Postmortem studies have proved by the local ethics committee, and writtenshown atrophy of the vestibular nerve and the ves- informed consent was obtained from all patients.
tibular sensory epithelium that is similar to the his- As in a previous study,20 the diagnosis of vestib- topathological findings in known viral disorders, ular neuritis was based on four criteria. There wassuch as herpes zoster oticus.9 Herpes simplex virus a history of the acute or subacute (i.e., within min-type 1 (HSV-1) DNA has been detected on autopsy utes to hours) onset of severe, prolonged rotatorywith the use of the polymerase chain reaction in vertigo, nausea, and postural imbalance. On clini-about two of three human vestibular ganglia.10,11 cal examination, there was a horizontal spontane-This indicates that the vestibular ganglia are latently ous nystagmus with a rotational component towardinfected by HSV-1, as are other cranial-nerve gan- the unaffected ear (fast phase) without evidence ofglia.12-14 A similar cause is also assumed for Bell’s a central vestibular lesion, and the head-thrust testpalsy and is strongly supported by the detection of (performed by turning the head of the patient rap-HSV-1 DNA in the endoneurial fluid of affected idly to the right and left to provoke compensatorypersons.15 eye movements) showed an ipsilateral deficit of the Recovery after vestibular neuritis is usually in- horizontal semicircular canal.17 Caloric irrigation complete.1,7 In a study of 60 patients, horizontal showed hyporesponsiveness or lack of respon-semicircular canal paresis was found in about 90 siveness of the horizontal canal of the affected ear.
percent one month after the onset of symptoms and (The maximal slow-phase velocity during caloric ir-in 80 percent after six months; the caloric responses rigation with water at 30°C and 44°C should be lessnormalized in only 42 percent.16 On the basis of the than three degrees per second on the affected side,incidence of this condition,2 a substantial and per- and the asymmetry between the two sides shouldmanent unilateral dynamic deficit of the vestibulo- be more than 25 percent as measured with the useocular reflex, which cannot be compensated for by of Jongkees’s formula for vestibular paresis.21,22)other mechanisms,17,18 develops in approximately Finally, there was a perceived displacement of ver-4000 people per year in the United States alone. This ticality and the eyes rotated toward the affected eardeficit leads to impaired vision and postural imbal- without showing vertical divergence of one eyeance during walking and especially during head above the other.23,24movement toward the affected ear.19 Patients were excluded if they had a history of Despite the assumed viral cause of vestibular vestibular dysfunction before the acute onset of neuritis, the effects of corticosteroids, antiviral symptoms or had symptoms that began more thanagents, or the two in combination are uncertain.1,8 three days before recruitment; if they had addition-We conducted a prospective, randomized trial of al cochlear symptoms, such as tinnitus or acute Downloaded from nejm.org at BOSTON UNIVERSITY on June 7, 2012. For personal use only. No other uses without permission. Copyright 2004 Massachusetts Medical Society. All rights reserved. The new england journal of medicine hearing loss before, during, or after the onset of patients also received antiemetic agents (50 tovertigo; if they had central ocular motor dysfunction 150 mg of dimenhydrinate a day) for a maximum ofor central vestibular dysfunction; if they had other three days.
signs or symptoms of brain-stem or cerebellar dis- All patients were admitted to the hospital for at orders, abnormal findings on magnetic resonance least one day and up to seven days (they were dis-imaging of the brain stem or cerebellum in diffu- charged when they were able to walk unassistedsion-weighted images or of hyperintense lesions in with their eyes closed). During the hospital stay,T -weighted images in combination with contrast compliance with the assigned regimen was checked enhancement in T -weighted images, a history of by the physicians and nurses by counting the cap- psychiatric disorders, glaucoma, ongoing infection, sules. On discharge from the hospital, all patientssevere diabetes mellitus (a fasting blood glucose were provided with the study drugs for the follow-level >180 mg per deciliter [10.0 mmol per liter] on ing days (through day 22) in standardized pack-admission, despite treatment), or severe hyperten- ages of the daily regimen with written instructionssion (blood pressure on admission >180 mm Hg for taking the drugs. Compliance was checked insystolic or >110 mm Hg diastolic); or if there were an interview within one week after treatment wascontraindications to the use of corticosteroids, such completed.
as peptic ulcer disease or known osteoporosis (on During hospitalization, the patients’ blood pres- the basis of bone-density testing or a history of frac- sure was measured three times per day and theture), or to valacyclovir, such as dysfunction of the blood glucose levels at least once per day (four timesliver (i.e., known cirrhosis of the liver or alanine per day for patients with known diabetes mellitus).
aminotransferase levels two times the upper limit After discharge, patients with known hypertensionof the normal range or higher) or dysfunction of were instructed to measure their blood pressurethe kidneys (i.e., creatinine level >2.6 mg per deci- at least three times per day, and those with knownliter [230 µmol per liter] in women and >3.5 mg per diabetes to measure their blood glucose level fourdeciliter [310 µmol per liter] in men), malignant dis- times per day. Medication was to be adjusted by theease, or heart failure.
patient’s physician. All patients received writteninformation about possible adverse effects of meth- r a n d o m i z a t i o n a n d t r e a t m e n t
ylprednisolone and valacyclovir, as well as a stan- Patients were randomly assigned (by means of com- dardized protocol with open questions about ad-puter-generated block randomization) to one of verse effects that might have occurred before thefour treatment groups: the placebo group, the meth- patients were included in the study. They were in-ylprednisolone group, the valacyclovir group, and structed to inform the investigators about any ad-the methylprednisolone-plus-valacyclovir group. verse effects as soon as possible, by telephone, fax,Methylprednisolone (or a matching placebo) was or e-mail. Adverse effects of the medication wereadministered daily as a single morning dose of assessed three to four weeks after treatment was100 mg on days 1 through 3, 80 mg on days 4 started; at that time, patients were asked whetherthrough 6, 60 mg on days 7 through 9, 40 mg on they had had any adverse effects, although they weredays 10 through 12, 20 mg on days 13 through 15, not asked about specific effects.
10 mg on days 16 through 18, and 10 mg on days Treatment was stopped if patients did not want 20 and 22. Valacyclovir, an l-valyl ester of acyclovir to continue or if they did not comply with the regi-
(or matching placebo), was given as two 500-mg men (i.e., did not take the study drug at least twice),
capsules three times daily for seven days. Valacyclo- if adverse effects developed during treatment, or if
vir was used in this study, because the serum con- signs or symptoms (such as tinnitus or hearing loss)
centrations that result from its use are similar to developed during the course of the disease that were
those resulting from intravenous acyclovir25 and not compatible with vestibular neuritis. Patients
because it is given at less frequent intervals than who did not return for the 12-month follow-up
oral acyclovir. The study drugs were first adminis- examination were excluded from the final analysis.
tered to all subjects on the day of admission, which
was within three days after the onset of symptoms. efficacy analysis
Patients also received 150 mg of pirenzepine (a As a measure of unilateral vestibular loss, the mean
muscarinic M1–receptor antagonist) once a day to peak slow-phase velocity during caloric irrigation
reduce the secretion of gastric acid. If necessary, with water at 30°C and 44°C was measured and
Downloaded from nejm.org at BOSTON UNIVERSITY on June 7, 2012. For personal use only. No other uses without permission. Copyright 2004 Massachusetts Medical Society. All rights reserved. m e t h y l p r e d n i s o l o n e , v a l a c y c l o v i r , o r t h e c o m b i n a t i o n f o r v e s t i b u l a r n e u r i t i s automatically analyzed with the use of IGOR Pro software (version 3.13, WaveMetrics) on the first orsecond day of hospitalization and at the 12-month Of 157 patients who underwent screening, 141 metfollow-up. Because the nystagmus induced by ca- the criteria for inclusion and were willing to partic-loric irrigation may vary considerably among sub- ipate. Of those 141 patients, 38 were randomly as-jects but only to a small extent in a healthy person, signed to the placebo group, 35 to the methyl-Jongkees’s vestibular paresis formula21,22,26 was prednisolone group, 33 to the valacyclovir group,used as the primary outcome variable in the efficacy and 35 to the methylprednisolone-plus-valacycloviranalysis. The extent of unilateral caloric paresis, group. Eight patients in the placebo group, six inexpressed as a percentage, was calculated with the the methylprednisolone group, six in the valacyclo-use of the following formula: {[(R30°+R44°)¡ vir group, and seven in the methylprednisolone- (L30° + L44°)] ÷ (R30° + R44° + L30° + L44°)} ¬ 100, plus-valacyclovir group were excluded (because the where, for example, R30° is the mean peak slow- patient did not want to continue treatment, was notphase velocity during caloric irrigation of the right compliant, had severe adverse effects and treatmentlabyrinth with water at 30°C (R denotes right, and was stopped, or was lost to follow-up) (Table 1).
L left, and 30° or 44° indicates the water tempera- Thirty patients in the placebo group, 29 in theture). With the use of this formula, a direct compar- methylprednisolone group, 27 in the valacyclovirison can be performed between the function of the group, and 28 in the methylprednisolone-plus-horizontal semicircular canals of the right and left valacyclovir group completed the study at 12labyrinths. The formula is highly reliable in detect- months, for a total of 114 patients. The groups diding unilateral peripheral vestibular loss.22 A 12- not differ with regard to mean age, sex ratio, andmonth follow-up was used, because there have been time from the onset of symptoms to the start ofreports of delayed spontaneous recovery of vestib- treatment (Table 1).
ular function.16,27 Calculations performed with the use of Jong- kees’s formula at the initial examination showed no s t a t i s t i c a l a n a l y s i s
significant differences in the extent of the periph- The sample size was calculated with the use of eral vestibular deficits among the groups at base-SampleStat software (SPSS) and was based on a line (Fig. 1 and Table 2). The mean extent of vestib-mean (±SD) difference between groups (calculated ular paresis was 78.9±24.0 percent in the placebowith Jongkees’s formula) of 25±26 percent. The group, 78.7±15.8 percent in the methylpredniso-calculation yielded a sample size of 30 patients in lone group, 78.4±20.0 percent in the valacyclovireach treatment group, assuming a t-test for two in- group, and 78.6±21.1 percent in the methylpred-dependent groups, with a two-sided alpha level of nisolone-plus-valacyclovir group. At the 12-month0.01 and a statistical power of 85 percent.
follow-up, the improvement in vestibular paresis Data are presented as means ±SD. A two-by-two was 39.6±28.1 percentage points among patients factorial analysis of variance (in which the factors in the placebo group, 62.4±16.9 percentage pointswere methylprednisolone and valacyclovir), used to in the methylprednisolone group, 36.0±26.7 per-compare the percentage of vestibular paresis mea- centage points in the valacyclovir group, and 59.2±sured at the initial examination of the patient and 24.1 percentage points in the methylprednisolone-the percentage measured at follow-up, was per- plus-valacyclovir group (Fig. 1 and Table 2). Analysisformed with the use of Statistica 6 software (Stat- of variance showed a significant effect of methyl-Soft). All reported P values are two-sided.
prednisolone (P<0.001), but not of valacyclovir An interim analysis was performed (in 2001) (P=0.43). Furthermore, there was no interaction be- after one year of follow-up of a total of 50 patients. tween methylprednisolone and valacyclovir (P=There was no significant difference between groups, 0.92), indicating that the addition of valacyclovirand the study was continued.
did not affect the efficacy of methylprednisolone.
Hoechst Pharma, Germany, supplied the study A combined analysis of the two groups that re- drugs and placebo but was not involved in the de- ceived methylprednisolone showed a change in thesign of the study, the data collection and analysis, percentage of vestibular paresis of 60.9±20.6 per-the preparation of the manuscript, or the decision cent (95 percent confidence interval, 55.4 to 66.3to publish the findings.
percent), as compared with 37.9±27.2 percentage Downloaded from nejm.org at BOSTON UNIVERSITY on June 7, 2012. For personal use only. No other uses without permission. Copyright 2004 Massachusetts Medical Society. All rights reserved. The new england journal of medicine Table 1. Baseline Characteristics of the 141 Patients.*
Methylprednisolone Valacyclovir
Methylprednisolone-
plus-Valacyclovir Group
Characteristic
Time from onset of symptoms to treatment * Plus–minus values are means ±SD.
points (95 percent confidence interval, 30.7 to 45.1 67-year-old man) 10 days after he started therapypercent) in the two groups who did not receive (despite the administration of pirenzepine to himmethylprednisolone. The pooled effect of valacy- and all other subjects). Methylprednisolone wasclovir (change, 47.8±27.8 percentage points; 95 stopped, and the bleeding was halted with a localpercent confidence interval, 40.3 to 55.3 percent) injection of epinephrine. Three patients reportedwas not significantly different from the change in dyspepsia and five reported mood swings, but allthe percentage of vestibular paresis without vala- these patients continued treatment. The adversecyclovir (50.8±25.8 percentage points; 95 percent effects resolved after the patients completed treat-confidence interval, 44.1 to 57.5 percent).
ment with corticosteroids. In two patients who had The treatment groups differed significantly in the normal fasting blood glucose levels on admission, number of patients who had a complete or almost hyperglycemia developed (fasting blood glucosecomplete recovery of peripheral vestibular function >180 mg per deciliter [10.0 mmol per liter]) during(defined as a difference of less than 25 percent treatment. Both patients started long-term treat-between the affected and unaffected labyrinths, as ment with oral antidiabetic agents, and the bloodcalculated with the use of Jongkees’s formula). glucose level normalized. Patients in the placeboThe number of patients who had complete or par- and valacyclovir groups reported no other adversetial recovery was 8 of 30 in the placebo group, 22 of effects that affected treatment.
29 in the methylprednisolone group, 10 of 27 in thevalacyclovir group, and 22 of 28 in the methylpred- nisolone-plus-valacyclovir group (placebo vs. meth-ylprednisolone, P<0.001; placebo vs. methylpred- Treatment with methylprednisolone alone signifi-nisolone plus valacyclovir, P<0.001).
cantly improved the long-term outcome of periph- In the methylprednisolone group, a gastric ulcer eral vestibular function among patients with vestib- with minor bleeding developed in one patient (a ular neuritis, whereas treatment with the antiviral Downloaded from nejm.org at BOSTON UNIVERSITY on June 7, 2012. For personal use only. No other uses without permission. Copyright 2004 Massachusetts Medical Society. All rights reserved. m e t h y l p r e d n i s o l o n e , v a l a c y c l o v i r , o r t h e c o m b i n a t i o n f o r v e s t i b u l a r n e u r i t i s Methylprednisolone
Vestibular Paresis (%)
Vestibular Paresis (%)
Follow-up
Follow-up
Follow-up
Follow-up
Valacyclovir
Methylprednisolone plus Valacyclovir
Vestibular Paresis (%)
Vestibular Paresis (%)
Follow-up
Follow-up
Follow-up
Follow-up
Figure 1. Unilateral Vestibular Loss within Three Days after the Onset of Symptoms and after 12 Months.
Vestibular function was measured for each patient in the four study groups with the use of caloric irrigation and Jongkees’s formula for ves-tibular paresis for a direct comparison of the function of the right and left labyrinths. Clinically relevant vestibular paresis was defined as an asymmetry greater than 25 percent between the right-sided and left-sided responses.26 In the box plot for each treatment group, the solid square indicates the mean, the horizontal lines the 25th, 50th, and 75th percentiles, the error bars above and below the boxes the SDs, and the crosses the 1st and 99th percentiles. Analysis of variance for the comparison of methylprednisolone and methylprednisolone plus valacy-clovir with placebo or valacyclovir alone showed significantly more improvement with methylprednisolone. The combination of methylpred-nisolone and valacyclovir was not superior to corticosteroid monotherapy.
agent valacyclovir did not improve the outcome. (placebo) group. Thirteen of the 16 patients whoThe combination of these drugs was no more ef- had been treated with corticosteroids had remis-fective than methylprednisolone alone.
sion of their symptoms within six hours of start- Previous data have supported the hypothesis that ing treatment. In another study27 that was neither corticosteroids have a beneficial effect on the course prospective nor placebo-controlled, 34 patients re-of acute peripheral vestibular vertigo. One double- ceived corticosteroid therapy for vestibular neuritisblind, prospective, placebo-controlled, crossover and 77 received no treatment. The recovery rate instudy28 included 20 patients who had the opportu- that study, as measured with the use of Jongkees’snity to switch medication within 24 hours of start- formula over a mean follow-up period of sevening treatment; in the final analysis, 16 patients had months, was twice as high among the patients whoreceived corticosteroids (beginning with a dose of received corticosteroids as among those who did32 mg per day) for eight days, and 4 patients had re- not, although corticosteroids had no significant ef-ceived placebo. At follow-up at four weeks, electro- fect on the symptoms.
nystagmography showed that values returned to For Bell’s palsy, which probably has the same normal in all 16 patients who had received cortico- pathogenesis as vestibular neuritis,15,29 one trialsteroids but in only 2 of the 4 patients in the control showed that the combination of acyclovir and cor- Downloaded from nejm.org at BOSTON UNIVERSITY on June 7, 2012. For personal use only. No other uses without permission. Copyright 2004 Massachusetts Medical Society. All rights reserved. The new england journal of medicine Table 2. Extent of Vestibular Paresis at Baseline and at 12 Months.*
Placebo Group
Methylprednisolone
Valacyclovir Group
Methylprednisolone-plus-
Extent of Vestibular Paresis
Group (N=29)
Valacyclovir Group (N=28)
* Percentages were calculated with the use of Jongkees’s formula for vestibular paresis (which indicates the extent of uni- lateral caloric paresis). Percentages below 25 percent are considered to be normal; higher percentages indicate a greater unilateral peripheral vestibular deficit. At baseline, the mean (±SD) percentage of vestibular paresis and the range among patients excluded were 72.1±35.0 percent (range, 38 to 100 percent) in the placebo group (n=8); 80±23.8 percent (range, 42 to 100 percent) in the methylprednisolone group (n=6); 75.3±22.2 percent (range, 41 to 100 percent) in the valacyclovir group (n=6); and 86.2±38.7 percent (range, 31 to 100 percent) in the methylprednisolone-plus-valacyclovir group (n=7).
† P values are for pairwise comparisons with the placebo group.
‡ P values by analysis of variance showed a significant effect of methylprednisolone but not of valacyclovir. There was no interaction between methylprednisolone and valacyclovir, indicating that the addition of valacyclovir did not affect the ef-ficacy of methylprednisolone. ticosteroids significantly improved the outcome as vestibular compensation.34 Data on symptoms andcompared with corticosteroids alone.30 However, on postural imbalance would not allow differentia-meta-analyses of studies of treatment for Bell’s tion between an improvement in peripheral vestib-palsy29,31 have shown contradictory results with re- ular function and an improvement in central ves-gard to the reported trials, and the authors conclud- tibular compensation, and therefore we did noted that corticosteroids are probably effective and collect these data. The percentage of improvementthat acyclovir (combined with prednisolone) is pos- in vestibular paresis cannot be directly translatedsibly effective in improving facial function.29 into clinical terms; nonetheless, methylpredniso- In our study, the antiviral drug did not improve lone therapy significantly increased the extent of the outcome in patients with vestibular neuritis, de- recovery, and the likelihood of complete recovery,spite the assumed viral cause. Replication of HSV-1 of peripheral vestibular function. We did not mea-in the vestibular ganglia may conceivably have al- sure vestibular function during the period betweenready occurred by the time the antiviral drug was the start of treatment and the 12-month assess-initiated — that is, within three days after the onset ment. Thus, we cannot estimate the effects of theof symptoms. The findings in two studies of the different regimens on the times to improvement.
treatment of herpes simplex encephalitis may pro- Furthermore, data on the potential adverse effectsvide some support for this hypothesis. In both stud- of methylprednisolone and valacyclovir therapyies, the most relevant prognostic factor was early were not systematically collected. Finally, we do notacyclovir therapy — within two days after admission have follow-up data on patients who did not take atto the hospital.32,33 Furthermore, there is good evi- least two doses of the assigned study drug or indence that the major damage in vestibular neuritis whom adverse effects developed that necessitatedis caused by the swelling and mechanical compres- stopping treatment. However, such patients madesion of the vestibular nerve within the temporal up only a small proportion of the total number ofbone, which is also assumed in Bell’s palsy.29 The patients, and at baseline they appeared similar toantiinflammatory effect, which results in reduced patients with complete follow-up. Our results showswelling, may explain why treatment with cortico- that methylprednisolone alone significantly im-steroids results in improvement in both disorders.
proved the extent of recovery of peripheral vestibu- Our study has several limitations. We did not as- lar function in patients with vestibular neuritis.
sess the duration and severity of symptoms (vertigo We are indebted to Judy Benson for editorial assistance with the manuscript, to Ruth Sandmann-Strupp, M.D., for helpful discus- and imbalance). In studies in animals, however, cor- sions, and to Stefan Glasauer, Ph.D., for his contribution to the ticosteroids have been shown to improve central statistical analysis.
Downloaded from nejm.org at BOSTON UNIVERSITY on June 7, 2012. For personal use only. No other uses without permission. Copyright 2004 Massachusetts Medical Society. All rights reserved. m e t h y l p r e d n i s o l o n e , v a l a c y c l o v i r , o r t h e c o m b i n a t i o n f o r v e s t i b u l a r n e u r i t i s r e f e r e n c e s
25. Hardman JG, Limbird LE, Molinoff PB,
dromes. 2nd ed. London: Springer, 1999.
sponse. Am J Pathol 2003;163:2179-84.
15. Murakami S, Mizobuchi M, Nakashiro
& Gilman’s the pharmacological basis of ma T. Vestibular neuronitis: epidemiologi- Y, Doi T, Hato N, Yanagihara N. Bell palsy cal survey by questionnaire in Japan. Acta and herpes simplex virus: identification of viral DNA in endoneurial fluid and muscle.
26. Honrubia V. Quantitative vestibular
function tests and the clinical examination.
Labyrinth- und Hörnerverkrankungen.
16. Okinaka Y, Sekitani T, Okazaki H, Miura
In: Herdman SJ, ed. Vestibular rehabilita- M, Tahara T. Progress of caloric response of tion. Philadelphia: F.A. Davis, 1994:113-64.
vestibular neuronitis. Acta Otolaryngol Sup- 27. Ohbayashi S, Oda M, Yamamoto M, et
mus due to certain positions of the head.
al. Recovery of the vestibular function after Acta Otolaryngol (Stockh) 1924;6:106-37.
17. Halmagyi GM, Curthoys IS. A clinical
vestibular neuronitis. Acta Otolaryngol Sup- sign of canal paresis. Arch Neurol 1988;45: 28. Ariyasu L, Byl FM, Sprague MS, Adour
common disorders of the vestibular system.
18. Curthoys IS, Halmagyi GM. Vestibular
KK. The beneficial effect of methylpredniso- compensation: a review of the oculomotor, lone in acute vestibular vertigo. Arch Oto- neural, and clinical consequences of unilat- laryngol Head Neck Surg 1990;116:700-3.
vertigo due to unilateral sudden partial loss eral vestibular loss. J Vestib Res 1995;5:67- 29. Grogan PM, Gronseth GS. Practice pa-
rameter: steroids, acyclovir, and surgery for 19. Borello-France DF, Whitney SL, Herd-
Bell’s palsy (an evidence-based review): re- man SJ. Assessment of vestibular hypofunc- port of the Quality Standards Subcommittee laryngol Head Neck Surg 1995;112:162-72.
tion. In: Herdman SJ, ed. Vestibular rehabil- of the American Academy of Neurology.
itation. Philadelphia: F.A. Davis, 1994:247- 30. Adour KK, Ruboyianes JM, von Doer-
20. Strupp M, Arbusow V, Maag KP, Gall C,
sten PG, et al. Bell’s palsy treatment with Brandt T. Vestibular exercises improve cen- tral vestibulospinal compensation after ves- prednisone alone: a double-blind, random- 10. Furuta Y, Takasu T, Fukuda S, Inuyama
tibular neuritis. Neurology 1998;51:838-44.
21. Jongkees LB, Maas J, Philipszoon A.
Clinical nystagmography: a detailed study 31. Sipe J, Dunn L. Aciclovir for Bell’s palsy
ganglia. Acta Otolaryngol Suppl 1993;503: of electro-nystagmography in 341 patients (idiopathic facial paralysis). Cochrane Data- with vertigo. Pract Otorhinolaryngol (Basel) 11. Arbusow V, Schulz P, Strupp M, et al.
32. McGrath N, Anderson NE, Croxson
Distribution of herpes simplex virus type 1 22. Fife TD, Tusa RJ, Furman JM, et al. As-
MC, Powell KF. Herpes simplex encephalitis in human geniculate and vestibular ganglia: sessment: vestibular testing techniques in treated with acyclovir: diagnosis and long implications for vestibular neuritis. Ann adults and children: report of the Therapeu- term outcome. J Neurol Neurosurg Psychia- 12. Theil D, Arbusow V, Derfuss T, et al.
mittee of the American Academy of Neurol- 33. Raschilas F, Wolff M, Delatour F, et al.
Prevalence of HSV-1 LAT in human trigemi- Outcome of and prognostic factors for her- nal, geniculate, and vestibular ganglia and 23. Böhmer A, Rickenmann J. The subjec-
pes simplex encephalitis in adult patients: its implication for cranial nerve syndromes.
tive visual vertical as a clinical parameter of results of a multicenter study. Clin Infect Dis vestibular function in peripheral vestibular 13. Nahmias AJ, Roizman B. Infection with
diseases. J Vestib Res 1995;5:35-45.
34. Cameron SA, Dutia MB. Lesion-induced
24. Curthoys IS, Dai MJ, Halmagyi GM. Hu-
plasticity in rat vestibular nucleus neurones man ocular torsional position before and af- dependent on glucocorticoid receptor acti- 14. Theil D, Derfuss T, Paripovic I, et al. La-
ter unilateral vestibular neurectomy. Exp tent herpesvirus infection in human trigem- Copyright 2004 Massachusetts Medical Society. Downloaded from nejm.org at BOSTON UNIVERSITY on June 7, 2012. For personal use only. No other uses without permission. Copyright 2004 Massachusetts Medical Society. All rights reserved.

Source: http://www.ed.bmc.org/library/core-curriculum/NEJMvestibularneuritis.pdf