The Laryngoscope C 2014 The American Laryngological, V
Rhinological and Otological Society, Inc.
The Clinical Significance of Findings Obtained on 3D-FLAIR MR Imaging in Patients With Ramsay-Hunt Syndrome Mi S. Chung, MD; Jeong H. Lee, MD, PhD; Dae Y. Kim, MD; Young-Min Lim, MD, PhD; Joong H. Ahn, MD, PhD; Yu S. Sung, PhD; Young J. Choi, MD; Ra G. Yoon, MD; Jung H. Baek, MD Objectives/Hypothesis: To investigate the clinical significance of three-dimensional-fluid-attenuated inversion recovery (3D-FLAIR) magnetic resonance imaging (MRI) findings in patients with Ramsay-Hunt syndrome (RHS). Study Design: Case series. Methods: We enrolled 28 consecutive patients of RHS with temporal bone MRI. Initial clinical findings and outcome were assessed by House-Brackmann (HB) scales, electroneuronography (ENoG), and pure tone audiometry (PTA). Two radiologists evaluated the presence of abnormalities on pre-/postcontrast 3D-FLAIR for the cranial nerve (CN)-VII, CN-VIII, inner ear (IE), and the posterior fossa by consensus. The relative signal intensity and enhancement degree (rED) of the structures were measured using ImageJ (http://rsbweb.nih.gov/ij/). Statistical test correlated the clinical symptoms and the outcome with the analysis results of 3D-FLAIR images. Results: 3D-FLAIR demonstrated enhancement of CN-VII in all patients. Precontrast hyperintensity and enhancement were seen in eight and 16 patients with IE, and in four and six with CN-VIII, respectively. Precontrast hyperintensity of IE or CN-VIII was significantly associated with the presence of vertigo (P value < 0.05). Precontrast hyperintensity of IE or CN-VIII significantly correlated with clinical symptoms assessed by HB, ENoG, and PTA (P value < 0.05, respectively). rED of the vestibule moderately correlated with initial HB scale (r 5 0.391, P 5 0.039). There was no correlation between any of the 3DFLAIR findings and the follow-up HB. Conclusions: RHS shows frequent abnormalities of IE or CN-VIII, as well as CN-VII on pre-/postcontrast 3D-FLAIR images. Precontrast hyperintensity of IE/CN-VIII on 3D-FLAIR is significantly correlated with the severity of facial palsy, the presence of vertigo, and the degree of hearing impairment but not with clinical outcome. Key Words: Ramsay-Hunt Syndrome, MRI, 3D-FLAIR, facial palsy. Level of Evidence: 4. Laryngoscope, 125:950–955, 2015
INTRODUCTION Ramsay-Hunt syndrome (RHS) is diagnosed by detecting erythematous vesicular rashes in the external auditory canal or in the ear, as well as otalgia and ipsilateral peripheral facial palsy. Characteristic peripheral facial palsy is caused by inflammation of cranial nerve (CN)-VII, accounting for the reactivation of varicellazoster virus in the geniculate ganglion. Compared with patients with idiopathic facial palsy (Bell’s palsy), RHS has had poorer recovery of facial palsy despite the various treatment strategies.1 Furthermore, concurrent CNVIII and multiple cranial nerve involvement, comprising
From the Department of Radiology and Research Institute of Radiology (S.C., J.H.L., D.Y.K., Y.S.S., Y.J.C., R.G.Y., J.H.B.); Department of Neurology (Y-M.L.); and the Department of Otolaryngology (J.H.A.), University of Ulsan College of Medicine, Asan Medical Center, Seoul, Republic of Korea. Editor’s Note: This Manuscript was accepted for publication September 23, 2014. The authors have no funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Jeong Hyun Lee, MD, PhD, Department of Radiology and Research Institute of Radiology, University of Ulsan College of Medicine, Asan Medical Center, 88, Olympic-ro 43-gil, Songpa-gu, Seoul 138–736, Korea. E-mail: [email protected] DOI: 10.1002/lary.24973
Laryngoscope 125: April 2015
950
48% and 1.8% 3.2 % of the RHS patients, respectively, was associated with the severe initial presentation and the poorer prognosis of facial palsy.2–4 Owing to the gadolinium accumulation in the inflamed tissue with breakdown of the blood–nerve barrier, contrast-enhanced T1-weighted magnetic resonance (MR) imaging (CE-T1WI) can visualize the enhancement of CN-VII in almost all patients with RHS.5–7 However, abnormality of the inner ear (IE) on CE-T1WI has been rarely demonstrated, despite the relatively frequent ear symptoms including hearing loss or vertigo.8–10 Therefore, the role of CE-T1WI has been questioned for the management of RHS, especially in the assessment of the symptom severity or prediction of its outcome.5,11,12 Three-dimensional-fluid-attenuated inversion recovery (3D-FLAIR) MR imaging has recently been used to reveal subtle changes in the cranial nerve or IE due to its sensitivity for detecting high protein concentration and abnormal contrast enhancement, compared to conventional T1WI. However, the imaging findings of 3DFLAIR MR imaging of RHS have rarely been reported,13,14 whereas the diagnostic advantages of 3DFLAIR MR in patients with idiopathic facial palsy have recently been published.15 Therefore, the purpose of our
Chung et al.: 3D-FLAIR MR Findings in Ramsay-Hunt Syndrome Patients
TABLE I. Summary of Patient Demographics and Clinical Data. Characteristics
Results
Male:Female
16:12
Age (years) Mean 6 SD
57 6 18 years
Range Underlying disease Diabetes Hypertension Miscellaneous* Duration from onset to MR imaging Mean 6 SD (days) Inner ear symptoms Hearing difficulty Vertigo
19–78 years 4 2 4 7.6 6 6.8 5 6
Tinnitus Clinical follow-up period
4
Mean 6 SD (months)
4.7 6 6.4
*Included hepatocellular carcinoma (n 5 1), thyroid cancer (n 5 1), lung cancer (n51), and valvular heart disease (n 5 1), respectively. MR 5 magnetic resonance; SD 5 standard deviation.
study is to evaluate the imaging findings of RHS on pre-/postcontrast 3D-FLAIR MRI and to correlate the 3D-FLAIR MR findings with the symptom severity and clinical outcome in patients with RHS.
defined as the attainment of HB scale I or II, whereas others were defined as a poor outcome. Our institutional review board approved this study, and informed consent was waived due to the retrospective nature of the study.
MR Examination All MR examinations, including coronal T1- and T2weighted, turbo spin-echo images, and axial 3D T2 and FLAIR images, were performed on a 3T MR unit (Achieva; Philips Medical System, Best, The Netherlands) with an 8-channel head coil. We used a variable, refocusing flip-angle technique to obtain isotrophic 3D T2-weighted and FLAIR MR images. For enhancement, Gd-DOTA (Dotarem; Guerbet, Paris, France) at 0.2 mmol/kg of body weight were used, and axial 3D-FLAIR images were acquired with a time delay of 7 minutes. Axial pre-/postcontrast 3D-FLAIR MR imaging were performed with the following parameters: time to repetition (TR) 5 8000 ms; effective time to echo (TE) 5 280 ms; inversion time (TI) 5 2400 ms; fat saturation 5 spectral attenuated inversion recovery; flip angle 5 90 ; number of signal averages (NSA) 5 1; echo train length (ETL) 5 80; number of encoding steps 5 300; FOV 5 180 3 180 mm; reconstruction matrix 5 512 3 512; slice thickness 5 1.2 mm; spacing between slices 5 0.6 mm; and sensitivity encoding (SENSE) factor 5 2. Axial 3D T2 imaging was performed using the following parameters: TR 5 2000 msec; effective TE 5 250 msec; flip angle 5 90 ; NSA 5 2; ETL 5 70; number of encoding steps 5 300; field of view (FOV) 5 180 3 180 mm; reconstruction matrix 5 512 3 512; slice thickness 5 1.2 mm; spacing between slices 5 0.6 mm; and SENSE factor 5 2.
Analysis of MR Imaging MATERIALS AND METHODS Patients Data Between January 2009 and June 2012, 31 consecutive patients were diagnosed as RHS with facial palsy and otoscopic findings of vesicles around the ipsilateral ear or the external auditory canal. They underwent pretreatment temporal bone MRI using a 3T MR unit at our institution. Among these patients, 28 (male:female 5 16:12) were enrolled in our study, after excluding three patients with insufficient clinical followup information. The electronic medical records were reviewed in order to collect demographic patient information, including age, sex, the presence of underlying diseases, and the duration from the symptom onset to MR imaging and treatment. The presence of ear symptom such as hearing difficulty or vertigo was also evaluated in all patients. Initial symptom severity was assessed using the initial House-Brackmann (HB) scale (HBini), electroneuronography (ENoGini), and pure tone audiometry (PTAini). HBini was scored according to the degrees of facial palsy at the time of the first physical examination.16 ENoGini was measured at least 4 days following the onset of facial palsy because Wallerian degeneration of the peripheral facial nerve takes approximately 72 hours to become evident after the onset of the symptom.17 For describing ENoGini result, the percentage of amplitude loss was calculated as 100 3 (1 – [amplitude on affected side/amplitude on healthy side]). PTAini was performed in 14 patients to measure the degree of hearing impairment. The clinical outcome regarding the degree of facial nerve impairment was evaluated using the final HB scale (HBfu) during the clinical follow-up. Based on HBfu, a good outcome was
Laryngoscope 125: April 2015
Coronal T1- and T2-weighted TSE and axial 3D-FLAIR images were used to identify any coexisting diseases in the temporal bone and the brain stem. MR imaging analysis was divided into two different sessions, that is, visual assessment and quantitative measurement of the signal intensity or enhancement degree using ImageJ, a public domain, Java-based image processing program developed at the National Institutes of Health (NIH, Bethesda, MD; http://rsbweb.nih.gov/ij/). For visual assessment, two radiologists, with 17 and 3 years of clinical experience interpreting temporal bone MRI, reviewed pre- and postcontrast 3D-FLAIR MR images of all patients in consensus manner blind to the clinical data. We evaluated the presence of abnormalities in the CN-VII to CN-XII, IE, the brainstem, and the cerebellum. CN-VII was subdivided into three parts including the anterior genu, the canalicular segment, and others (the labyrinthine segment, the tympanic segment, and mastoid segments). We searched for the presence of precontrast hyperintensity or contrast enhancement of the structures on pre- and postcontrast 3DFLAIR MR images. The precontrast hyperintensity of the cochlea and vestibule on 3D-FLAIR images was defined as when the signal was more hyperintense than that of the cerebellar hemisphere. Thickening of a nerve segment was also considered as a precontrast abnormality for the cranial nerves with reference to the contralateral unaffected side. Quantitative analysis was performed by one radiologist with 3 years of clinical experience. We reviewed the axial 3D-FLAIR images to search for any abnormality of the pons where a reference region-of-interest (ROI) was to be placed. After importing the DICOM files of the axial 3D T2 and pre- and postcontrast- enhanced 3D-FLAIR MR images into ImageJ, the operator placed ROIs on the affected cochlea, the vestibule, and the corresponding pons. ROIs were directly placed on the entire cochlea or vestibule on each slice
Chung et al.: 3D-FLAIR MR Findings in Ramsay-Hunt Syndrome Patients
951
TABLE II. Inner Ear Symptoms, Results of Physical Examination, and Precontrast 3D-FLAIR MR Images of Total Study Group. Initial Symptoms
Patient No.
Sex
Age
Site
HBini
HBfu
ENoGini
1
M
34
Right
2 3
F M
69 78
Right Right
V H
4
F
43
Right
T
5 6
M F
33 40
Left Left
7
F
68
Left
4
3
59
4 4
2 4
59 53.3
47
2
1
45.2
8
2 4
1 1
49 49
32
4
3
63
8 9
F M
65 48
48
Left Left
4 3
1 1
49.2 62
32
10
M
11 12
F M
23
Left
4
2
88
62
28 30
Right Left
2 4
2 3
70 81
13
10
M
19
Left
3
2
23
14 15
F F
74 66
Right Right
H/V
4 3
2 2
57 55
67 13
16
M
42
Right
T/V
2
2
25
20
17 18
F M
61 76
Right Right
T
4 4
1 4
33.3 54
19
M
32
Left
4
4
88
20 21
M M
63 37
Left Right
4 3
3 1
52.8 58
22
F
36
Right
3
1
51.2
23 24
F M
47 55
Left Right
4 4
4 4
51.7 66
25
M
56
Right
4
1
55.6
26 27
M M
71 71
Left Right
5 4
2 1
66.9 26
55 63
28
F
58
Left
3
3
52.6
10
H/V
H/T/V V H
PTAini
Hyperintensity on Pre-3D-FLAIR
CO CO/VE/CN-VIII CO/VE
CN-VIII
CO
CO/VE/CN-VIII
CO
11 20
CO/VE CO/VE CO/VE/CN-VIII
CN-VIII 5 vestibulocochlear nerve; CO 5 cochlea; ENoGini 5 initial results of electroneuronography; F 5 female; H 5 hearing difficulty; HBini 5 initial HouseBrackmann scales; HBfu 5 follow-up House-Brackmann scales; M 5 male; Pre-3D-FLAIR 5 precontrast 3D-FLAIR; PTAini 5 initial results of pure tone audiometry; VE 5 vestibule.
of the 3D-FLAIR images using the Wand tool of ImageJ, which allowed reliable definition of the boundary of the cochlea or the vestibule with increased signal intensity. When the signal intensity was insufficient to define the boundary, the ROI placements were checked by reference to axial 3D T2 MR images obtained at the same level. Circular ROIs of similar size (42–46 mm2) were carefully placed at the center of the pons on the same level as analyzed IE. We obtained the mean value of the signal intensities within the ROIs of the cochlea, the vestibule, and the pons, respectively, and calculated the relative signal intensities (rSI) of the cochlea and the vestibule to the pons. We also calculated the relative enhancement degree (rED) of the cochlea and the vestibule using rSIs on pre- and postcontrast 3D-FLAIR MR images using the following equation: [rSI on postcontrast images 2 rSI on precontrast images]/rSI on precontrast images 3 100.
Statistical Analysis The Mann-Whitney test was performed to investigate the correlation between the abnormal findings on 3D-FLAIR MRI and the initial symptom severity. The Chi-square test or Fisher’s extract test was used to evaluate the association between the abnormal findings seen on 3D-FLAIR MRI and the inner
Laryngoscope 125: April 2015
952
ear symptom or the clinical outcome. Spearman’s correlation coefficient was used to investigate the correlation between the initial symptom severity/clinical outcome and rSI/rED. Statistical analysis was performed using SPSS 19.0. for Windows (SPSS Icnc, Chicago, IL). For all statistical analyses, a P-value < .05 was considered to indicate a significant difference.
RESULTS Table I and Table II show the summary of the demographic data, inner ear symptoms, results of physical examination, and precontrast 3D-FLAIR MR images of the patients in this study. An equal number of ears (left:right 5 14:14) were included from 28 patients, and there was no bilateral involvement. Nine patients had underlying disease including malignant tumor (n 5 3), diabetes (n 5 3), hypertension (n 5 1), both diabetes and hypertension (n 5 1), and valvular heart disease (n 5 1). The mean clinical follow-up period of patients with complete recovery was 2.06 6 1.47 months (range: 1–4 months) and of those without recovery was 4.73 6 6.37 months (range: 1–28 months). The mean values of HBini
Chung et al.: 3D-FLAIR MR Findings in Ramsay-Hunt Syndrome Patients
Fig. 1. MR images of a 47-year-old female with right facial palsy (HBini scale II) and tinnitus for 3 days. Precontrast 3D-FLAIR image (A) shows precontrast hyperintensity (long arrow) of the canalicular segment of the right CN-VII and the adjacent meatal area. Postcontrast 3D-FLAIR image (B) demonstrates enhancement of the canalicular (long arrow) and labyrinthine (short arrow) segments (short arrow) and the anterior genu (arrowhead) of the right CN-VII. The cochlea and the vestibule showed no abnormality on pre-/postcontrast 3D-FLAIR images. ENoGini measured 2 weeks later showed a 45% neural response of the ipsilateral CN-VII. The patient’s facial palsy was fully recovered to HBfu scale I after treatment with acyclovir during follow-up. CN-VII5 facial nerve; ENoGini 5 initial results of electroneuronography; HBfu 5 follow-up House-Brackmann scales; HBini 5 initial House-Brackmann scales; Pre-3D-FLAIR 5 precontrast 3D-FLAIR.
and HBfu in all patients were 3.5 6 0.79 (range, 2– 5) and 2.2 6 1.12 (range, 1– 4), respectively. Initial ENoG was 55.1 6 16.07% (range, 23–88%). Visual assessment of 3D-FLAIR MR images demonstrated precontrast hyperintensity in 18 patients (64%) and contrast enhancement in all patients for CN-VII (Fig. 1). Precontrast hyperintensity and contrast enhancement were frequently depicted in the cochlea, the vestibule, and the CN-VIII (Fig. 2). There was only one case of a brainstem abnormality that showed hyperintensity of the ipsilateral facial nucleus on precontrast 3D-FLAIR images. The detailed results are summarized in Table III. Statistical test revealed that HBini was significantly associated with precontrast hyperintensities of the cochlea, vestibule, and CN-VIII (Table IV). ENoGini was also significantly correlated with precontrast hyperintensity of the cochlea. Hearing impairment measured by PTAini in 14 patients had a significant correlation with the
presence of precontrast hyperintensity of the cochlea or CN-VIII. There was no correlation between HBini or ENoGini, and the precontrast hyperintensity of CN-VII or the brainstem. There was no statistical significance between HBini or ENoGini, and the presence of enhancement of the CN-VII, CN-VIII, or the inner ear. The clinical outcome of facial palsy assessed by HBfu had no correlation with any abnormality noted on 3D-FLAIR MR images by visual assessment. Among the subjective inner ear symptoms, vertigo was the only one that showed significant correlation with abnormalities on 3D-FLAIR MR images, including precontrast hyperintensity of the cochlea, the vestibule, and CN-VIII, as well as contrast enhancement of the cochlea and CN-VIII (P value < 0.05). Quantitative analysis demonstrated no significant correlation between the initial symptom severity/clinical outcome and rSI or rED of the IE (P value > 0.05), except rED of the vestibule showing a moderate
Fig. 2. MR images of a 76-year-old female with right facial palsy (HBini scale IV) with right hearing difficulty and vertigo for 3 days. PTAini showed moderate hearing loss (47dB) on the right side. Precontrast 3D-FLAIR image (A) shows hyperintense change and thickening of the canalicular segment (arrow) and the anterior genu (arrowhead) of CN-VII. Another precontrast 3D-FLAIR image at the lower level also demonstrates hyperintense change of the cochlea (double arrows), the vestibule (short arrow), and CN-VIII (long arrow). Postcontrast 3D-FLAIR image at the same level (C) clearly depicts strong enhancement of the cochlea (double arrows), the vestibule (short arrow), and CN-VIII (long arrow). Note that the distal meatal area is also strongly enhanced by the contrast agent. ENoGini for the damaged facial nerve presented a 59% neural response. After treatment with acyclovir, the patient’s facial palsy completely recovered to HBfu scale I during follow-up. CN-VII5 facial nerve; CN-VIII 5 vestibulocochlear nerve; ENoGini 5 initial results of electroneuronography; HBini 5 initial HouseBrackmann scales; HBfu 5 follow-up House-Brackmann scales; MR 5 magnetic resonance; Pre-3D-FLAIR 5 precontrast 3D-FLAIR; PTAini 5 initial results of pure tone audiometry.
Laryngoscope 125: April 2015
Chung et al.: 3D-FLAIR MR Findings in Ramsay-Hunt Syndrome Patients
953
TABLE III. Frequency of Abnormal Findings on Precontrast and Postcontrast 3D-FLAIR MR Images According to the Anatomic Locations by Visual Assessment.
CN-VII
Precontrast 3D-FLAIR
Postcontrast 3D-FLAIR
18 (64)
28 (100)
Anterior genu
8 (29)
12 (43)
Canalicular segment Others*
16 (57) 14 (50)
27 (96) 9 (32)
8 (29)
16 (57)
8 (29) 6 (21)
16 (57) 5 (18)
CN-VIII
4 (14)
6 (21)
Brain stem
1 (4)
†
1 (4)†
IE Cochlea Vestibule
Numbers in parentheses mean the percentages among all patients. *Labyrinthine, tympanic, and mastoid segments. † The lesion involved the ipsilateral facial nucleus. CN-VII5 facial nerve; CN-VIII 5 vestibulocochlear nerve; IE 5 inner ear.
correlation with HBini (r 5 0.391; P 5 0.039). HBfu indicating the outcome of facial palsy also had no correlation with any abnormality seen on 3D-FLAIR MR images and with quantitative analysis.
DISCUSSION In this study, we evaluated the imaging findings of 3D-FLAIR MRI of RHS in 28 consecutive patients presenting with acute facial palsy, as well as the relationships between the MR imaging findings and the clinical symptom severity or the outcome of the facial paralysis. The main results of our study can be summarized as follows: 1) 3D-FLAIR MRI demonstrated frequent involvement of the cochlea and the vestibule by RHS with incidences up to 29% of precontrast hyperintensity and 57% of contrast enhancement, in addition to contrast enhancement of CN-VII in all patients. 2) The precontrast hyperintensities of IE and CN-VIII by visual assessment were significantly correlated with the clinical symptoms and signs, including the presence of vertigo or hearing impairment measured by PTAini and the initial severity of facial palsy scaled by HBini or ENoGini. 3) rED of the vestibule showed a significant correlation with the initial severity of the facial palsy (HBini). Although enhancement of CN VII on CE-T1WI is very well-known in RHS patients,5–7,18 the role of MRI has been questioned because most studies failed to reveal any correlation between the enhancement pattern and the clinical symptoms and/or outcome of facial palsy.5,6,9,12,19 In this study, we demonstrated that abnormal findings of IE and CN-VIII, in addition to CNVII, are frequently found in RHS patients using 3DFLAIR, and they are significantly associated not only with audiovestibular symptoms but with the severity of the facial palsy. A recent study reported that there is a significant difference on 3D-FLAIR MR imaging between those patients with and those without audiovestibular disturbance among 15 patients with facial nerve paralysis.14 Laryngoscope 125: April 2015
954
They suggested that 3D-FLAIR could be a useful tool for detecting the abnormality of IE in patients with facial nerve paralysis and audiovestibular symptoms.14 Sugiura et al. also correlated the enhancement of IE on 3D-FLAIR with hearing deterioration in a patient with RHS in a case report.13 All of these results, including ours, suggest that the 3D-FLAIR sequence could be more efficient than T1WI for evaluating RHS patients because 3D-FLAIR can demonstrate the abnormality of CN-VIII and IE correlated with clinical symptoms because of its higher sensitivity for detecting hemorrhage or proteineous fluid caused by inflammation14,20 and lower concentration of gadolinium contrast agent.21,22 There have been no previous reports correlating the abnormality of CN-VIII and/or IE on MR images with the severity of facial palsy. This might resulted from the rare demonstration of their abnormality on MR using CE-T1WIs and the fact that audiovestibular evaluation is not routinely performed in RHS patients. Our results suggest that increased involvement of the structures (not only CN-VII but also CN-VIII and IE) on 3D-FLAIR might be related to the increased viral burden, and subsequently to more severe symptom presentation. Therefore, detecting the additional involvement of IE or CNVIII using 3D-FLAIR could be important in order to perform a thorough evaluation regarding the extent of initial viral involvement. However, in our study, recovery of facial palsy was not correlated with abnormalities of CN-VIII and/or IE on 3D-FLAIR. Owing to the all of CN-VII enhanced on postcontrast 3D-FLAIR images, the presence of enhancement in CN-VII on visual analysis could not reflect the patient’s symptoms. Thus, we did not attempt to correlate the degree or pattern of CN-VII enhancement with the prognosis of facial palsy. There has been a published report suggesting that quantitative ROI analysis of CE-T1WI data is helpful for predicting the outcome of acute facial nerve palsy in patients with Bell’s palsy.23 However, we thought it impractical to measure all of the enhancing segments of CN-VII in order to assess the prognosis. In addition, it is difficult to separate the enhancing CN-VII adjacent to also enhancing CN-VIII; and it is difficult to confidently divide the enhancing and nonenhancing area of CN-VII. Even more importantly, most of the studies that attempted to reveal the relationship between the severity of facial palsy and CN-VII abnormality in RHS TABLE IV. P-values Representing Correlations Between The Initial Severity of Facial Palsy (Hbini; Enogini) or Hearing Impairment (PTAini) and the Presence of Precontrast Hyperintensity on 3D-FLAIR MR Images by Visual Assessment. HBini
ENoGini
PTAini
Cochlea Vestibule
0.009 0.021
0.009 0.450
0.045 0.137
CN-VIII
0.043
0.404
0.047
CN-VIII 5 vestibulocochlear nerve; ENoGini 5 initial results of electroneuronography; HBini 5 initial House-Brackmann scales; PTAini 5 initial results of pure tone audiometry.
Chung et al.: 3D-FLAIR MR Findings in Ramsay-Hunt Syndrome Patients
patients failed to demonstrate any significance using T1WIs. There are several limitations to our study. Our study results are not free from the intrinsic limitations of all retrospective studies. Second, the possibility of measurement error should be considered in our study resulting from our attempt to measure the signal intensity of a small structure such as IE. To make our analysis reliable and reproducible, we used the wand tool of the ImageJ program, which used the seeded regiongrowing method with a constant threshold for the ROIs in order to determine the boundary of the IE and which has been used for quantitative measurement of brain parenchymal lesions24,25 and for signal intensity measurement of the whole volume of the ROI. Finally, the small number of enrolled patients in this study will need validation of our findings in further studies.
CONCLUSION In conclusion, 3D-FLAIR MRI is useful for detecting abnormalities of CN-VII and IE/CN-VIII in patients with RHS. Precontrast hyperintensity and enhancement of IE/CN-VIII on 3D-FLAIR MRI are significantly correlated with the clinical symptoms and signs, including the presence of vertigo or hearing impairment and the severity of facial palsy, although not with the clinical outcome. Therefore, detecting additional involvement of IE or CN-VIII using 3D-FLAIR might be helpful in order to perform a thorough evaluation regarding the extent of disease involvement.
Acknowledgment We acknowledge the technical assistance provided by Biomedical Imaging Infrastructure, Department of Radiology and Research Institute of Radiology in Asan Medical Center, Seoul, Korea.
BIBLIOGRAPHY 1. Ryu EW, Lee HY, Lee SY, Park MS, Yeo SG. Clinical manifestations and prognosis of patients with Ramsay Hunt syndrome. Am J Otolaryngol 2012;33:313–318. 2. Kim YH, Chang MY, Jung HH, et al. Prognosis of Ramsay Hunt syndrome presenting as cranial polyneuropathy. Laryngoscope 2010; 120:2270– 2276. 3. Wayman DM, Pham HN, Byl FM, Adour KK. Audiological manifestations of Ramsay Hunt syndrome. J Laryngol Otol 1990;104:104–108. 4. Shim HJ, Jung H, Park DC, Lee JH, Yeo SG. Ramsay Hunt syndrome with multicranial nerve involvement. Acta Otolaryngol 2011;131: 210–215.
Laryngoscope 125: April 2015
5. Tada Y, Aoyagi M, Tojima H, et al. Gd-DTPA enhanced MRI in Ramsay Hunt syndrome. Acta Otolaryngol Suppl 1994;511:170–174. 6. Yanagida M, Ushiro K, Yamashita T, Kumazawa T, Katoh T. Enhanced MRI in patients with Ramsay-Hunt’s syndrome. Acta Otolaryngol Suppl 1993;500:58–61. 7. Kuo MJ, Drago PC, Proops DW, Chavda SV. Early diagnosis and treatment of Ramsay Hunt syndrome: the role of magnetic resonance imaging. J Laryngol Otol 1995;109:777–780. 8. Downie AC, Howlett DC, Koefman RJ, Banerjee AK, Tonge KA. Case report: prolonged contrast enhancement of the inner ear on magnetic resonance imaging in Ramsay Hunt syndrome. Br J Radiol 1994;67: 819–821. 9. Rovira Canellas A, Sanchez Torres C, Grive Isern E, Capellades Font J, Castella Fierro E, Gili Planas J. Ramsay-Hunt syndrome and highresolution 3DFT MRI. J Comput Assist Tomogr 1993;17:495–497. 10. Zammit-Maempel I, Campbell RS. Prolonged contrast enhancement of the inner ear on MRI in Ramsay Hunt syndrome. Br J Radiol 1995;68:334– 335. 11. Brandle P, Satoretti-Schefer S, Bohmer A, Wichmann W, Fisch U. Correlation of MRI, clinical, and electroneuronographic findings in acute facial nerve palsy. Am J Otol 1996;17:154–161. 12. Jonsson L, Tien R, Engstrom M, Thuomas KA. Gd-DPTA enhanced MRI in Bell’s palsy and herpes zoster oticus: an overview and implications for future studies. Acta Otolaryngol 1995;115:577–584. 13. Sugiura M, Naganawa S, Nakata S, Kojima S, Nakashima T. 3D-FLAIR MRI findings in a patient with Ramsay Hunt syndrome. Acta Otolaryngol 2007;127:547–549. 14. Nakata S, Mizuno T, Naganawa S, et al. 3D-FLAIR MRI in facial nerve paralysis with and without audio-vestibular disorder. Acta Otolaryngol 2010;130:632–636. 15. Lim HK, Lee JH, Hyun D, et al. MR diagnosis of facial neuritis: diagnostic performance of contrast-enhanced 3D-FLAIR technique compared with contrast-enhanced 3D-T1-fast-field echo with fat suppression. AJNR Am J Neuroradiol 2012;33:779–783. 16. Vrabec JT, Backous DD, Facial Nerve Disorders Committee, et al. Facial Nerve Grading System 2.0. Otolaryngol Head Neck Surg 2009;140:445– 450. 17. Grosheva M, Wittekindt C, Guntinas-Lichius O. Prognostic value of electroneurography and electromyography in facial palsy. Laryngoscope 2008;118:394–397. 18. Kim J, Chung SM, Moon IS, Lee HK, Lee WS. Correlation between enhanced MRI and surgical findings in herpes zoster oticus. Acta Otolaryngol 2009;129:900–905. 19. Sweeney CJ, Gilden DH. Ramsay Hunt syndrome. J Neurol Neurosurg Psychiatry 2001;71:149–154. 20. Sugiura M, Naganawa S, Sato E, Nakashima T. Visualization of a high protein concentration in the cochlea of a patient with a large endolymphatic duct and sac, using three-dimensional fluid-attenuated inversion recovery magnetic resonance imaging. J Laryngol Otol 2006;120:1084– 1086. 21. Fukuoka H, Hirai T, Okuda T, et al. Comparison of the added value of contrast-enhanced 3D fluid-attenuated inversion recovery and magnetization-prepared rapid acquisition of gradient echo sequences in relation to conventional postcontrast T1-weighted images for the evaluation of leptomeningeal diseases at 3T. AJNR Am J Neuroradiol 2010;31: 868–873. 22. Mathews VP, Caldemeyer KS, Lowe MJ, Greenspan SL, Weber DM, Ulmer JL. Brain: gadolinium-enhanced fast fluid-attenuated inversion-recovery MR imaging. Radiology 1999;211:257–263. 23. Kress B, Griesbeck F, Stippich C, Bahren W, Sartor K. Bell palsy: quantitative analysis of MR imaging data as a method of predicting outcome. Radiology 2004;230:504–509. 24. Hanna-Pladdy B, Heilman KM, Foundas AL. Cortical and subcortical contributions to ideomotor apraxia: analysis of task demands and error types. Brain 2001;124:2513–2527. 25. Patel B, Gimi B, Vachha B, Agadi S, Koral K. Optic nerve and chiasm enlargement in a case of infantile Krabbe disease: quantitative comparison with 26 age-matched controls. Pediatr Radiol 2008; 38:697–699.
Chung et al.: 3D-FLAIR MR Findings in Ramsay-Hunt Syndrome Patients
955
Rhinological and Otological Society, Inc.
The Clinical Significance of Findings Obtained on 3D-FLAIR MR Imaging in Patients With Ramsay-Hunt Syndrome Mi S. Chung, MD; Jeong H. Lee, MD, PhD; Dae Y. Kim, MD; Young-Min Lim, MD, PhD; Joong H. Ahn, MD, PhD; Yu S. Sung, PhD; Young J. Choi, MD; Ra G. Yoon, MD; Jung H. Baek, MD Objectives/Hypothesis: To investigate the clinical significance of three-dimensional-fluid-attenuated inversion recovery (3D-FLAIR) magnetic resonance imaging (MRI) findings in patients with Ramsay-Hunt syndrome (RHS). Study Design: Case series. Methods: We enrolled 28 consecutive patients of RHS with temporal bone MRI. Initial clinical findings and outcome were assessed by House-Brackmann (HB) scales, electroneuronography (ENoG), and pure tone audiometry (PTA). Two radiologists evaluated the presence of abnormalities on pre-/postcontrast 3D-FLAIR for the cranial nerve (CN)-VII, CN-VIII, inner ear (IE), and the posterior fossa by consensus. The relative signal intensity and enhancement degree (rED) of the structures were measured using ImageJ (http://rsbweb.nih.gov/ij/). Statistical test correlated the clinical symptoms and the outcome with the analysis results of 3D-FLAIR images. Results: 3D-FLAIR demonstrated enhancement of CN-VII in all patients. Precontrast hyperintensity and enhancement were seen in eight and 16 patients with IE, and in four and six with CN-VIII, respectively. Precontrast hyperintensity of IE or CN-VIII was significantly associated with the presence of vertigo (P value < 0.05). Precontrast hyperintensity of IE or CN-VIII significantly correlated with clinical symptoms assessed by HB, ENoG, and PTA (P value < 0.05, respectively). rED of the vestibule moderately correlated with initial HB scale (r 5 0.391, P 5 0.039). There was no correlation between any of the 3DFLAIR findings and the follow-up HB. Conclusions: RHS shows frequent abnormalities of IE or CN-VIII, as well as CN-VII on pre-/postcontrast 3D-FLAIR images. Precontrast hyperintensity of IE/CN-VIII on 3D-FLAIR is significantly correlated with the severity of facial palsy, the presence of vertigo, and the degree of hearing impairment but not with clinical outcome. Key Words: Ramsay-Hunt Syndrome, MRI, 3D-FLAIR, facial palsy. Level of Evidence: 4. Laryngoscope, 125:950–955, 2015
INTRODUCTION Ramsay-Hunt syndrome (RHS) is diagnosed by detecting erythematous vesicular rashes in the external auditory canal or in the ear, as well as otalgia and ipsilateral peripheral facial palsy. Characteristic peripheral facial palsy is caused by inflammation of cranial nerve (CN)-VII, accounting for the reactivation of varicellazoster virus in the geniculate ganglion. Compared with patients with idiopathic facial palsy (Bell’s palsy), RHS has had poorer recovery of facial palsy despite the various treatment strategies.1 Furthermore, concurrent CNVIII and multiple cranial nerve involvement, comprising
From the Department of Radiology and Research Institute of Radiology (S.C., J.H.L., D.Y.K., Y.S.S., Y.J.C., R.G.Y., J.H.B.); Department of Neurology (Y-M.L.); and the Department of Otolaryngology (J.H.A.), University of Ulsan College of Medicine, Asan Medical Center, Seoul, Republic of Korea. Editor’s Note: This Manuscript was accepted for publication September 23, 2014. The authors have no funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Jeong Hyun Lee, MD, PhD, Department of Radiology and Research Institute of Radiology, University of Ulsan College of Medicine, Asan Medical Center, 88, Olympic-ro 43-gil, Songpa-gu, Seoul 138–736, Korea. E-mail: [email protected] DOI: 10.1002/lary.24973
Laryngoscope 125: April 2015
950
48% and 1.8% 3.2 % of the RHS patients, respectively, was associated with the severe initial presentation and the poorer prognosis of facial palsy.2–4 Owing to the gadolinium accumulation in the inflamed tissue with breakdown of the blood–nerve barrier, contrast-enhanced T1-weighted magnetic resonance (MR) imaging (CE-T1WI) can visualize the enhancement of CN-VII in almost all patients with RHS.5–7 However, abnormality of the inner ear (IE) on CE-T1WI has been rarely demonstrated, despite the relatively frequent ear symptoms including hearing loss or vertigo.8–10 Therefore, the role of CE-T1WI has been questioned for the management of RHS, especially in the assessment of the symptom severity or prediction of its outcome.5,11,12 Three-dimensional-fluid-attenuated inversion recovery (3D-FLAIR) MR imaging has recently been used to reveal subtle changes in the cranial nerve or IE due to its sensitivity for detecting high protein concentration and abnormal contrast enhancement, compared to conventional T1WI. However, the imaging findings of 3DFLAIR MR imaging of RHS have rarely been reported,13,14 whereas the diagnostic advantages of 3DFLAIR MR in patients with idiopathic facial palsy have recently been published.15 Therefore, the purpose of our
Chung et al.: 3D-FLAIR MR Findings in Ramsay-Hunt Syndrome Patients
TABLE I. Summary of Patient Demographics and Clinical Data. Characteristics
Results
Male:Female
16:12
Age (years) Mean 6 SD
57 6 18 years
Range Underlying disease Diabetes Hypertension Miscellaneous* Duration from onset to MR imaging Mean 6 SD (days) Inner ear symptoms Hearing difficulty Vertigo
19–78 years 4 2 4 7.6 6 6.8 5 6
Tinnitus Clinical follow-up period
4
Mean 6 SD (months)
4.7 6 6.4
*Included hepatocellular carcinoma (n 5 1), thyroid cancer (n 5 1), lung cancer (n51), and valvular heart disease (n 5 1), respectively. MR 5 magnetic resonance; SD 5 standard deviation.
study is to evaluate the imaging findings of RHS on pre-/postcontrast 3D-FLAIR MRI and to correlate the 3D-FLAIR MR findings with the symptom severity and clinical outcome in patients with RHS.
defined as the attainment of HB scale I or II, whereas others were defined as a poor outcome. Our institutional review board approved this study, and informed consent was waived due to the retrospective nature of the study.
MR Examination All MR examinations, including coronal T1- and T2weighted, turbo spin-echo images, and axial 3D T2 and FLAIR images, were performed on a 3T MR unit (Achieva; Philips Medical System, Best, The Netherlands) with an 8-channel head coil. We used a variable, refocusing flip-angle technique to obtain isotrophic 3D T2-weighted and FLAIR MR images. For enhancement, Gd-DOTA (Dotarem; Guerbet, Paris, France) at 0.2 mmol/kg of body weight were used, and axial 3D-FLAIR images were acquired with a time delay of 7 minutes. Axial pre-/postcontrast 3D-FLAIR MR imaging were performed with the following parameters: time to repetition (TR) 5 8000 ms; effective time to echo (TE) 5 280 ms; inversion time (TI) 5 2400 ms; fat saturation 5 spectral attenuated inversion recovery; flip angle 5 90 ; number of signal averages (NSA) 5 1; echo train length (ETL) 5 80; number of encoding steps 5 300; FOV 5 180 3 180 mm; reconstruction matrix 5 512 3 512; slice thickness 5 1.2 mm; spacing between slices 5 0.6 mm; and sensitivity encoding (SENSE) factor 5 2. Axial 3D T2 imaging was performed using the following parameters: TR 5 2000 msec; effective TE 5 250 msec; flip angle 5 90 ; NSA 5 2; ETL 5 70; number of encoding steps 5 300; field of view (FOV) 5 180 3 180 mm; reconstruction matrix 5 512 3 512; slice thickness 5 1.2 mm; spacing between slices 5 0.6 mm; and SENSE factor 5 2.
Analysis of MR Imaging MATERIALS AND METHODS Patients Data Between January 2009 and June 2012, 31 consecutive patients were diagnosed as RHS with facial palsy and otoscopic findings of vesicles around the ipsilateral ear or the external auditory canal. They underwent pretreatment temporal bone MRI using a 3T MR unit at our institution. Among these patients, 28 (male:female 5 16:12) were enrolled in our study, after excluding three patients with insufficient clinical followup information. The electronic medical records were reviewed in order to collect demographic patient information, including age, sex, the presence of underlying diseases, and the duration from the symptom onset to MR imaging and treatment. The presence of ear symptom such as hearing difficulty or vertigo was also evaluated in all patients. Initial symptom severity was assessed using the initial House-Brackmann (HB) scale (HBini), electroneuronography (ENoGini), and pure tone audiometry (PTAini). HBini was scored according to the degrees of facial palsy at the time of the first physical examination.16 ENoGini was measured at least 4 days following the onset of facial palsy because Wallerian degeneration of the peripheral facial nerve takes approximately 72 hours to become evident after the onset of the symptom.17 For describing ENoGini result, the percentage of amplitude loss was calculated as 100 3 (1 – [amplitude on affected side/amplitude on healthy side]). PTAini was performed in 14 patients to measure the degree of hearing impairment. The clinical outcome regarding the degree of facial nerve impairment was evaluated using the final HB scale (HBfu) during the clinical follow-up. Based on HBfu, a good outcome was
Laryngoscope 125: April 2015
Coronal T1- and T2-weighted TSE and axial 3D-FLAIR images were used to identify any coexisting diseases in the temporal bone and the brain stem. MR imaging analysis was divided into two different sessions, that is, visual assessment and quantitative measurement of the signal intensity or enhancement degree using ImageJ, a public domain, Java-based image processing program developed at the National Institutes of Health (NIH, Bethesda, MD; http://rsbweb.nih.gov/ij/). For visual assessment, two radiologists, with 17 and 3 years of clinical experience interpreting temporal bone MRI, reviewed pre- and postcontrast 3D-FLAIR MR images of all patients in consensus manner blind to the clinical data. We evaluated the presence of abnormalities in the CN-VII to CN-XII, IE, the brainstem, and the cerebellum. CN-VII was subdivided into three parts including the anterior genu, the canalicular segment, and others (the labyrinthine segment, the tympanic segment, and mastoid segments). We searched for the presence of precontrast hyperintensity or contrast enhancement of the structures on pre- and postcontrast 3DFLAIR MR images. The precontrast hyperintensity of the cochlea and vestibule on 3D-FLAIR images was defined as when the signal was more hyperintense than that of the cerebellar hemisphere. Thickening of a nerve segment was also considered as a precontrast abnormality for the cranial nerves with reference to the contralateral unaffected side. Quantitative analysis was performed by one radiologist with 3 years of clinical experience. We reviewed the axial 3D-FLAIR images to search for any abnormality of the pons where a reference region-of-interest (ROI) was to be placed. After importing the DICOM files of the axial 3D T2 and pre- and postcontrast- enhanced 3D-FLAIR MR images into ImageJ, the operator placed ROIs on the affected cochlea, the vestibule, and the corresponding pons. ROIs were directly placed on the entire cochlea or vestibule on each slice
Chung et al.: 3D-FLAIR MR Findings in Ramsay-Hunt Syndrome Patients
951
TABLE II. Inner Ear Symptoms, Results of Physical Examination, and Precontrast 3D-FLAIR MR Images of Total Study Group. Initial Symptoms
Patient No.
Sex
Age
Site
HBini
HBfu
ENoGini
1
M
34
Right
2 3
F M
69 78
Right Right
V H
4
F
43
Right
T
5 6
M F
33 40
Left Left
7
F
68
Left
4
3
59
4 4
2 4
59 53.3
47
2
1
45.2
8
2 4
1 1
49 49
32
4
3
63
8 9
F M
65 48
48
Left Left
4 3
1 1
49.2 62
32
10
M
11 12
F M
23
Left
4
2
88
62
28 30
Right Left
2 4
2 3
70 81
13
10
M
19
Left
3
2
23
14 15
F F
74 66
Right Right
H/V
4 3
2 2
57 55
67 13
16
M
42
Right
T/V
2
2
25
20
17 18
F M
61 76
Right Right
T
4 4
1 4
33.3 54
19
M
32
Left
4
4
88
20 21
M M
63 37
Left Right
4 3
3 1
52.8 58
22
F
36
Right
3
1
51.2
23 24
F M
47 55
Left Right
4 4
4 4
51.7 66
25
M
56
Right
4
1
55.6
26 27
M M
71 71
Left Right
5 4
2 1
66.9 26
55 63
28
F
58
Left
3
3
52.6
10
H/V
H/T/V V H
PTAini
Hyperintensity on Pre-3D-FLAIR
CO CO/VE/CN-VIII CO/VE
CN-VIII
CO
CO/VE/CN-VIII
CO
11 20
CO/VE CO/VE CO/VE/CN-VIII
CN-VIII 5 vestibulocochlear nerve; CO 5 cochlea; ENoGini 5 initial results of electroneuronography; F 5 female; H 5 hearing difficulty; HBini 5 initial HouseBrackmann scales; HBfu 5 follow-up House-Brackmann scales; M 5 male; Pre-3D-FLAIR 5 precontrast 3D-FLAIR; PTAini 5 initial results of pure tone audiometry; VE 5 vestibule.
of the 3D-FLAIR images using the Wand tool of ImageJ, which allowed reliable definition of the boundary of the cochlea or the vestibule with increased signal intensity. When the signal intensity was insufficient to define the boundary, the ROI placements were checked by reference to axial 3D T2 MR images obtained at the same level. Circular ROIs of similar size (42–46 mm2) were carefully placed at the center of the pons on the same level as analyzed IE. We obtained the mean value of the signal intensities within the ROIs of the cochlea, the vestibule, and the pons, respectively, and calculated the relative signal intensities (rSI) of the cochlea and the vestibule to the pons. We also calculated the relative enhancement degree (rED) of the cochlea and the vestibule using rSIs on pre- and postcontrast 3D-FLAIR MR images using the following equation: [rSI on postcontrast images 2 rSI on precontrast images]/rSI on precontrast images 3 100.
Statistical Analysis The Mann-Whitney test was performed to investigate the correlation between the abnormal findings on 3D-FLAIR MRI and the initial symptom severity. The Chi-square test or Fisher’s extract test was used to evaluate the association between the abnormal findings seen on 3D-FLAIR MRI and the inner
Laryngoscope 125: April 2015
952
ear symptom or the clinical outcome. Spearman’s correlation coefficient was used to investigate the correlation between the initial symptom severity/clinical outcome and rSI/rED. Statistical analysis was performed using SPSS 19.0. for Windows (SPSS Icnc, Chicago, IL). For all statistical analyses, a P-value < .05 was considered to indicate a significant difference.
RESULTS Table I and Table II show the summary of the demographic data, inner ear symptoms, results of physical examination, and precontrast 3D-FLAIR MR images of the patients in this study. An equal number of ears (left:right 5 14:14) were included from 28 patients, and there was no bilateral involvement. Nine patients had underlying disease including malignant tumor (n 5 3), diabetes (n 5 3), hypertension (n 5 1), both diabetes and hypertension (n 5 1), and valvular heart disease (n 5 1). The mean clinical follow-up period of patients with complete recovery was 2.06 6 1.47 months (range: 1–4 months) and of those without recovery was 4.73 6 6.37 months (range: 1–28 months). The mean values of HBini
Chung et al.: 3D-FLAIR MR Findings in Ramsay-Hunt Syndrome Patients
Fig. 1. MR images of a 47-year-old female with right facial palsy (HBini scale II) and tinnitus for 3 days. Precontrast 3D-FLAIR image (A) shows precontrast hyperintensity (long arrow) of the canalicular segment of the right CN-VII and the adjacent meatal area. Postcontrast 3D-FLAIR image (B) demonstrates enhancement of the canalicular (long arrow) and labyrinthine (short arrow) segments (short arrow) and the anterior genu (arrowhead) of the right CN-VII. The cochlea and the vestibule showed no abnormality on pre-/postcontrast 3D-FLAIR images. ENoGini measured 2 weeks later showed a 45% neural response of the ipsilateral CN-VII. The patient’s facial palsy was fully recovered to HBfu scale I after treatment with acyclovir during follow-up. CN-VII5 facial nerve; ENoGini 5 initial results of electroneuronography; HBfu 5 follow-up House-Brackmann scales; HBini 5 initial House-Brackmann scales; Pre-3D-FLAIR 5 precontrast 3D-FLAIR.
and HBfu in all patients were 3.5 6 0.79 (range, 2– 5) and 2.2 6 1.12 (range, 1– 4), respectively. Initial ENoG was 55.1 6 16.07% (range, 23–88%). Visual assessment of 3D-FLAIR MR images demonstrated precontrast hyperintensity in 18 patients (64%) and contrast enhancement in all patients for CN-VII (Fig. 1). Precontrast hyperintensity and contrast enhancement were frequently depicted in the cochlea, the vestibule, and the CN-VIII (Fig. 2). There was only one case of a brainstem abnormality that showed hyperintensity of the ipsilateral facial nucleus on precontrast 3D-FLAIR images. The detailed results are summarized in Table III. Statistical test revealed that HBini was significantly associated with precontrast hyperintensities of the cochlea, vestibule, and CN-VIII (Table IV). ENoGini was also significantly correlated with precontrast hyperintensity of the cochlea. Hearing impairment measured by PTAini in 14 patients had a significant correlation with the
presence of precontrast hyperintensity of the cochlea or CN-VIII. There was no correlation between HBini or ENoGini, and the precontrast hyperintensity of CN-VII or the brainstem. There was no statistical significance between HBini or ENoGini, and the presence of enhancement of the CN-VII, CN-VIII, or the inner ear. The clinical outcome of facial palsy assessed by HBfu had no correlation with any abnormality noted on 3D-FLAIR MR images by visual assessment. Among the subjective inner ear symptoms, vertigo was the only one that showed significant correlation with abnormalities on 3D-FLAIR MR images, including precontrast hyperintensity of the cochlea, the vestibule, and CN-VIII, as well as contrast enhancement of the cochlea and CN-VIII (P value < 0.05). Quantitative analysis demonstrated no significant correlation between the initial symptom severity/clinical outcome and rSI or rED of the IE (P value > 0.05), except rED of the vestibule showing a moderate
Fig. 2. MR images of a 76-year-old female with right facial palsy (HBini scale IV) with right hearing difficulty and vertigo for 3 days. PTAini showed moderate hearing loss (47dB) on the right side. Precontrast 3D-FLAIR image (A) shows hyperintense change and thickening of the canalicular segment (arrow) and the anterior genu (arrowhead) of CN-VII. Another precontrast 3D-FLAIR image at the lower level also demonstrates hyperintense change of the cochlea (double arrows), the vestibule (short arrow), and CN-VIII (long arrow). Postcontrast 3D-FLAIR image at the same level (C) clearly depicts strong enhancement of the cochlea (double arrows), the vestibule (short arrow), and CN-VIII (long arrow). Note that the distal meatal area is also strongly enhanced by the contrast agent. ENoGini for the damaged facial nerve presented a 59% neural response. After treatment with acyclovir, the patient’s facial palsy completely recovered to HBfu scale I during follow-up. CN-VII5 facial nerve; CN-VIII 5 vestibulocochlear nerve; ENoGini 5 initial results of electroneuronography; HBini 5 initial HouseBrackmann scales; HBfu 5 follow-up House-Brackmann scales; MR 5 magnetic resonance; Pre-3D-FLAIR 5 precontrast 3D-FLAIR; PTAini 5 initial results of pure tone audiometry.
Laryngoscope 125: April 2015
Chung et al.: 3D-FLAIR MR Findings in Ramsay-Hunt Syndrome Patients
953
TABLE III. Frequency of Abnormal Findings on Precontrast and Postcontrast 3D-FLAIR MR Images According to the Anatomic Locations by Visual Assessment.
CN-VII
Precontrast 3D-FLAIR
Postcontrast 3D-FLAIR
18 (64)
28 (100)
Anterior genu
8 (29)
12 (43)
Canalicular segment Others*
16 (57) 14 (50)
27 (96) 9 (32)
8 (29)
16 (57)
8 (29) 6 (21)
16 (57) 5 (18)
CN-VIII
4 (14)
6 (21)
Brain stem
1 (4)
†
1 (4)†
IE Cochlea Vestibule
Numbers in parentheses mean the percentages among all patients. *Labyrinthine, tympanic, and mastoid segments. † The lesion involved the ipsilateral facial nucleus. CN-VII5 facial nerve; CN-VIII 5 vestibulocochlear nerve; IE 5 inner ear.
correlation with HBini (r 5 0.391; P 5 0.039). HBfu indicating the outcome of facial palsy also had no correlation with any abnormality seen on 3D-FLAIR MR images and with quantitative analysis.
DISCUSSION In this study, we evaluated the imaging findings of 3D-FLAIR MRI of RHS in 28 consecutive patients presenting with acute facial palsy, as well as the relationships between the MR imaging findings and the clinical symptom severity or the outcome of the facial paralysis. The main results of our study can be summarized as follows: 1) 3D-FLAIR MRI demonstrated frequent involvement of the cochlea and the vestibule by RHS with incidences up to 29% of precontrast hyperintensity and 57% of contrast enhancement, in addition to contrast enhancement of CN-VII in all patients. 2) The precontrast hyperintensities of IE and CN-VIII by visual assessment were significantly correlated with the clinical symptoms and signs, including the presence of vertigo or hearing impairment measured by PTAini and the initial severity of facial palsy scaled by HBini or ENoGini. 3) rED of the vestibule showed a significant correlation with the initial severity of the facial palsy (HBini). Although enhancement of CN VII on CE-T1WI is very well-known in RHS patients,5–7,18 the role of MRI has been questioned because most studies failed to reveal any correlation between the enhancement pattern and the clinical symptoms and/or outcome of facial palsy.5,6,9,12,19 In this study, we demonstrated that abnormal findings of IE and CN-VIII, in addition to CNVII, are frequently found in RHS patients using 3DFLAIR, and they are significantly associated not only with audiovestibular symptoms but with the severity of the facial palsy. A recent study reported that there is a significant difference on 3D-FLAIR MR imaging between those patients with and those without audiovestibular disturbance among 15 patients with facial nerve paralysis.14 Laryngoscope 125: April 2015
954
They suggested that 3D-FLAIR could be a useful tool for detecting the abnormality of IE in patients with facial nerve paralysis and audiovestibular symptoms.14 Sugiura et al. also correlated the enhancement of IE on 3D-FLAIR with hearing deterioration in a patient with RHS in a case report.13 All of these results, including ours, suggest that the 3D-FLAIR sequence could be more efficient than T1WI for evaluating RHS patients because 3D-FLAIR can demonstrate the abnormality of CN-VIII and IE correlated with clinical symptoms because of its higher sensitivity for detecting hemorrhage or proteineous fluid caused by inflammation14,20 and lower concentration of gadolinium contrast agent.21,22 There have been no previous reports correlating the abnormality of CN-VIII and/or IE on MR images with the severity of facial palsy. This might resulted from the rare demonstration of their abnormality on MR using CE-T1WIs and the fact that audiovestibular evaluation is not routinely performed in RHS patients. Our results suggest that increased involvement of the structures (not only CN-VII but also CN-VIII and IE) on 3D-FLAIR might be related to the increased viral burden, and subsequently to more severe symptom presentation. Therefore, detecting the additional involvement of IE or CNVIII using 3D-FLAIR could be important in order to perform a thorough evaluation regarding the extent of initial viral involvement. However, in our study, recovery of facial palsy was not correlated with abnormalities of CN-VIII and/or IE on 3D-FLAIR. Owing to the all of CN-VII enhanced on postcontrast 3D-FLAIR images, the presence of enhancement in CN-VII on visual analysis could not reflect the patient’s symptoms. Thus, we did not attempt to correlate the degree or pattern of CN-VII enhancement with the prognosis of facial palsy. There has been a published report suggesting that quantitative ROI analysis of CE-T1WI data is helpful for predicting the outcome of acute facial nerve palsy in patients with Bell’s palsy.23 However, we thought it impractical to measure all of the enhancing segments of CN-VII in order to assess the prognosis. In addition, it is difficult to separate the enhancing CN-VII adjacent to also enhancing CN-VIII; and it is difficult to confidently divide the enhancing and nonenhancing area of CN-VII. Even more importantly, most of the studies that attempted to reveal the relationship between the severity of facial palsy and CN-VII abnormality in RHS TABLE IV. P-values Representing Correlations Between The Initial Severity of Facial Palsy (Hbini; Enogini) or Hearing Impairment (PTAini) and the Presence of Precontrast Hyperintensity on 3D-FLAIR MR Images by Visual Assessment. HBini
ENoGini
PTAini
Cochlea Vestibule
0.009 0.021
0.009 0.450
0.045 0.137
CN-VIII
0.043
0.404
0.047
CN-VIII 5 vestibulocochlear nerve; ENoGini 5 initial results of electroneuronography; HBini 5 initial House-Brackmann scales; PTAini 5 initial results of pure tone audiometry.
Chung et al.: 3D-FLAIR MR Findings in Ramsay-Hunt Syndrome Patients
patients failed to demonstrate any significance using T1WIs. There are several limitations to our study. Our study results are not free from the intrinsic limitations of all retrospective studies. Second, the possibility of measurement error should be considered in our study resulting from our attempt to measure the signal intensity of a small structure such as IE. To make our analysis reliable and reproducible, we used the wand tool of the ImageJ program, which used the seeded regiongrowing method with a constant threshold for the ROIs in order to determine the boundary of the IE and which has been used for quantitative measurement of brain parenchymal lesions24,25 and for signal intensity measurement of the whole volume of the ROI. Finally, the small number of enrolled patients in this study will need validation of our findings in further studies.
CONCLUSION In conclusion, 3D-FLAIR MRI is useful for detecting abnormalities of CN-VII and IE/CN-VIII in patients with RHS. Precontrast hyperintensity and enhancement of IE/CN-VIII on 3D-FLAIR MRI are significantly correlated with the clinical symptoms and signs, including the presence of vertigo or hearing impairment and the severity of facial palsy, although not with the clinical outcome. Therefore, detecting additional involvement of IE or CN-VIII using 3D-FLAIR might be helpful in order to perform a thorough evaluation regarding the extent of disease involvement.
Acknowledgment We acknowledge the technical assistance provided by Biomedical Imaging Infrastructure, Department of Radiology and Research Institute of Radiology in Asan Medical Center, Seoul, Korea.
BIBLIOGRAPHY 1. Ryu EW, Lee HY, Lee SY, Park MS, Yeo SG. Clinical manifestations and prognosis of patients with Ramsay Hunt syndrome. Am J Otolaryngol 2012;33:313–318. 2. Kim YH, Chang MY, Jung HH, et al. Prognosis of Ramsay Hunt syndrome presenting as cranial polyneuropathy. Laryngoscope 2010; 120:2270– 2276. 3. Wayman DM, Pham HN, Byl FM, Adour KK. Audiological manifestations of Ramsay Hunt syndrome. J Laryngol Otol 1990;104:104–108. 4. Shim HJ, Jung H, Park DC, Lee JH, Yeo SG. Ramsay Hunt syndrome with multicranial nerve involvement. Acta Otolaryngol 2011;131: 210–215.
Laryngoscope 125: April 2015
5. Tada Y, Aoyagi M, Tojima H, et al. Gd-DTPA enhanced MRI in Ramsay Hunt syndrome. Acta Otolaryngol Suppl 1994;511:170–174. 6. Yanagida M, Ushiro K, Yamashita T, Kumazawa T, Katoh T. Enhanced MRI in patients with Ramsay-Hunt’s syndrome. Acta Otolaryngol Suppl 1993;500:58–61. 7. Kuo MJ, Drago PC, Proops DW, Chavda SV. Early diagnosis and treatment of Ramsay Hunt syndrome: the role of magnetic resonance imaging. J Laryngol Otol 1995;109:777–780. 8. Downie AC, Howlett DC, Koefman RJ, Banerjee AK, Tonge KA. Case report: prolonged contrast enhancement of the inner ear on magnetic resonance imaging in Ramsay Hunt syndrome. Br J Radiol 1994;67: 819–821. 9. Rovira Canellas A, Sanchez Torres C, Grive Isern E, Capellades Font J, Castella Fierro E, Gili Planas J. Ramsay-Hunt syndrome and highresolution 3DFT MRI. J Comput Assist Tomogr 1993;17:495–497. 10. Zammit-Maempel I, Campbell RS. Prolonged contrast enhancement of the inner ear on MRI in Ramsay Hunt syndrome. Br J Radiol 1995;68:334– 335. 11. Brandle P, Satoretti-Schefer S, Bohmer A, Wichmann W, Fisch U. Correlation of MRI, clinical, and electroneuronographic findings in acute facial nerve palsy. Am J Otol 1996;17:154–161. 12. Jonsson L, Tien R, Engstrom M, Thuomas KA. Gd-DPTA enhanced MRI in Bell’s palsy and herpes zoster oticus: an overview and implications for future studies. Acta Otolaryngol 1995;115:577–584. 13. Sugiura M, Naganawa S, Nakata S, Kojima S, Nakashima T. 3D-FLAIR MRI findings in a patient with Ramsay Hunt syndrome. Acta Otolaryngol 2007;127:547–549. 14. Nakata S, Mizuno T, Naganawa S, et al. 3D-FLAIR MRI in facial nerve paralysis with and without audio-vestibular disorder. Acta Otolaryngol 2010;130:632–636. 15. Lim HK, Lee JH, Hyun D, et al. MR diagnosis of facial neuritis: diagnostic performance of contrast-enhanced 3D-FLAIR technique compared with contrast-enhanced 3D-T1-fast-field echo with fat suppression. AJNR Am J Neuroradiol 2012;33:779–783. 16. Vrabec JT, Backous DD, Facial Nerve Disorders Committee, et al. Facial Nerve Grading System 2.0. Otolaryngol Head Neck Surg 2009;140:445– 450. 17. Grosheva M, Wittekindt C, Guntinas-Lichius O. Prognostic value of electroneurography and electromyography in facial palsy. Laryngoscope 2008;118:394–397. 18. Kim J, Chung SM, Moon IS, Lee HK, Lee WS. Correlation between enhanced MRI and surgical findings in herpes zoster oticus. Acta Otolaryngol 2009;129:900–905. 19. Sweeney CJ, Gilden DH. Ramsay Hunt syndrome. J Neurol Neurosurg Psychiatry 2001;71:149–154. 20. Sugiura M, Naganawa S, Sato E, Nakashima T. Visualization of a high protein concentration in the cochlea of a patient with a large endolymphatic duct and sac, using three-dimensional fluid-attenuated inversion recovery magnetic resonance imaging. J Laryngol Otol 2006;120:1084– 1086. 21. Fukuoka H, Hirai T, Okuda T, et al. Comparison of the added value of contrast-enhanced 3D fluid-attenuated inversion recovery and magnetization-prepared rapid acquisition of gradient echo sequences in relation to conventional postcontrast T1-weighted images for the evaluation of leptomeningeal diseases at 3T. AJNR Am J Neuroradiol 2010;31: 868–873. 22. Mathews VP, Caldemeyer KS, Lowe MJ, Greenspan SL, Weber DM, Ulmer JL. Brain: gadolinium-enhanced fast fluid-attenuated inversion-recovery MR imaging. Radiology 1999;211:257–263. 23. Kress B, Griesbeck F, Stippich C, Bahren W, Sartor K. Bell palsy: quantitative analysis of MR imaging data as a method of predicting outcome. Radiology 2004;230:504–509. 24. Hanna-Pladdy B, Heilman KM, Foundas AL. Cortical and subcortical contributions to ideomotor apraxia: analysis of task demands and error types. Brain 2001;124:2513–2527. 25. Patel B, Gimi B, Vachha B, Agadi S, Koral K. Optic nerve and chiasm enlargement in a case of infantile Krabbe disease: quantitative comparison with 26 age-matched controls. Pediatr Radiol 2008; 38:697–699.
Chung et al.: 3D-FLAIR MR Findings in Ramsay-Hunt Syndrome Patients
955
