Predictors of round window membrane visibility in pediatric cochlear implant surgery using temporal bone HRCT: A retrospective study

Accepted Manuscript Predictors of round window membrane visibility in pediatric cochlear implant surgery using temporal bone HRCT: a retrospective study Jianqing Chen, M.D., Yingwei Wu, M.D., Jun Shi, M.D., Huan Jia, M.D., Ph.D., Zhaoyan Wang, M.D., Ph.D., Zhihua Zhang, M.D., Ph.D., Hao Wu, M.D., Ph.D PII:

S0165-5876(19)30138-7

DOI:

https://doi.org/10.1016/j.ijporl.2019.03.017

Reference:

PEDOT 9431

To appear in:

International Journal of Pediatric Otorhinolaryngology

Received Date: 10 December 2018 Revised Date:

11 February 2019

Accepted Date: 14 March 2019

Please cite this article as: J. Chen, Y. Wu, J. Shi, H. Jia, Z. Wang, Z. Zhang, H. Wu, Predictors of round window membrane visibility in pediatric cochlear implant surgery using temporal bone HRCT: a retrospective study, International Journal of Pediatric Otorhinolaryngology, https://doi.org/10.1016/ j.ijporl.2019.03.017. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT 1

Predictors of round window membrane visibility in pediatric

2

cochlear

3

retrospective study

4

Jianqing Chen1,3,4, M.D., Yingwei Wu2, M.D., Jun Shi1,3,4, M.D., Huan Jia1,3,4, M.D.,

5

Ph.D., Zhaoyan Wang1,3,4, M.D., Ph.D., Zhihua Zhang1,3,4, M.D., Ph.D., Hao Wu1,3,4,

6

M.D., Ph.D.

7

1. Department of Otolaryngology, Head & Neck Surgery, Shanghai Ninth People’s

8

Hospital, affiliated to Shanghai Jiaotong University School of Medicine, Shanghai,

9

China

surgery

using

temporal

bone

HRCT:

a

TE D

M AN U

SC

RI PT

implant

2. Department of Radiology, Shanghai Ninth People’s Hospital, affiliated to Shanghai

11

Jiaotong University School of Medicine, Shanghai, China

12

3. Ear Institute, Shanghai Jiaotong University School of Medicine, Shanghai, China

13

4. Shanghai Key Laboratory of Translational Medicine on Ear and Nose Diseases

14

(14DZ2260300), Shanghai, China

AC C

EP

10

ACCEPTED MANUSCRIPT Correspondance: Hao Wu, M.D., Ph.D., 639 Zhizaoju Road, Shanghai 200011, China

16

([email protected]); Zhihua Zhang, M.D., Ph.D., 639 Zhizaoju Road, Shanghai

17

200011, China ([email protected])

18

Conflicts of Interest and Source of Funding: The authors declare that they have no

19

conflict of interest. This work was supported by grants from the National Natural

20

Science Foundation of China [grant number 81470681] and Shanghai Talent

21

Development Foundation [grant number 2017120]

M AN U

SC

RI PT

15

AC C

EP

TE D

22

ACCEPTED MANUSCRIPT 1

Predictors of round window membrane visibility in pediatric

2

cochlear

3

retrospective study

4

Abstract

5

Objective: To predict round window membrane (RWM) visibility and electrode

6

insertion sites using high-resolution computed tomography (HRCT) in pediatric

7

cochlear implant surgery.

8

Materials and methods: Sixty-two ears of 36 infants less than 1 year old were

9

included in our study. Intraoperative RWM visibility was classified into three types

10

corresponding to three different surgical approaches. Radiologic parameters were

11

measured on preoperative axial temporal HRCT images and correlated with RWM

12

visibility and surgical approaches.

13

Results: A significant correlation was found between the degree of RWM visibility

14

and the following two parameters: 1) a line (lw) was drawn from the posterior margin

15

of the RWM to the intersection point of the posterior wall of the external auditory

16

canal(EAC) and mastoid cortex. Another line (lf) was drawn between the posterior

17

margin of the RWM and the lateral margin of the FN. The angle between lw and lf was

18

measured as angle A, P<0.01, R2= -0.809; 2) a line (lm) was drawn from the anterior

19

to posterior margin of the RWM, and the angle between lm and lf was measured as

20

angle B, P<0.01, R2= -0.850. A nonsignificant correlation was found between the

21

degree of RWM visibility and the facial recess width, p>0.05, R2= -0.00015.

22

Conclusion: RWM visibility showed a high correlation with the two angular

23

measurements (angle A and angle B) and was associated with electrode insertion sites.

24

In children less than one year old, surgeons can depend on those two parameters in

25

predicting the RWM visibility.

26

Key Words: Cochlear implant, HRCT, round window membrane, surgical visibility.

using

temporal

bone

HRCT:

a

RI PT

surgery

AC C

EP

TE D

M AN U

SC

implant

27 1

ACCEPTED MANUSCRIPT INTRODUCTION

29

Cochlear implantation (CI) has become a widely accepted treatment option for

30

patients with severe to profound sensorineural hearing loss. It is mainly performed via

31

a posterior tympanotomy approach. The electrode could be inserted through a

32

cochleostomy or round window approach. In recent years, the round window (RW)

33

approach has attracted increasing attention. The RW approach for CI has many

34

advantages, such as less risk of intracochlear trauma, because it does not require

35

directly drilling into the cochlea, which reduces the possibility of bone dust entering

36

the scala tympani[1-3]. However, the RW approach is not always feasible for every

37

patient. In cases in which the round window membrane (RWM) cannot be exposed

38

properly despite maximal effort, other approaches such as the extended RW approach

39

or cochleostomy approach should be considered. Therefore, a presurgical method of

40

predicting whether the RW approach is possible would be of great value.

SC

M AN U

TE D

EP

41

RI PT

28

High-resolution computed tomography (HRCT) of temporal bone provides

43

radiographic information about inner ear structures, including the relative locations of

44

the facial nerve (FN), chorda tympani, external auditory canal (EAC), and round

45

window membrane (RWM). Several reports have attempted to predict the RWM

46

visibility using different radiological measurements[4-8]; however, these studies

47

focused mostly on adult patients. Few studies have examined children and especially

48

infants less than 1 year old. The purpose of our study is to determine an applicable

AC C

42

2

ACCEPTED MANUSCRIPT 49

method using HRCT imaging to preoperatively predict RWM visibility and its

50

implications in the success of the RW approach in infants less than 1 year old during

51

CI surgery.

RI PT

52 PATIENTS AND METHODS

54

Ethical approval for this study was given by the local ethics committee. Thirty-six

55

patients (62 ears) underwent CI from December 2016 to December 2017, including 13

56

(36.1%) girls and 23 (63.9%) boys. Their ages ranged from 7 to 12 months (mean:

57

10.15 ± 3.32 months). The exclusion criteria were inner ear malformations such as an

58

incomplete partition, a common cavity and cochlear hypoplasia.

59

HRCT images of the temporal bone (0.625 mm slice thickness) were obtained. Slices

60

with maximum visibility of the round window niche (RWN) and RWM were selected

61

to measure the following parameters: (1) a line (lw) was drawn from the posterior

62

margin of the RWM to the intersection point at the posterior wall of the EAC and

63

mastoid cortex. Another line (lf) was drawn between the posterior margin of the

64

RWM and the lateral margin of the FN. The angle between lw and lf was measured as

65

angle A (Fig. 1); (2) a line (lm) was drawn from the anterior to the posterior margin of

66

the RWM, and the angle between lm and lf was measured as angle B (Fig. 1); (3) The

67

FR width was measured perpendicularly from the EAC line to the anterolateral part of

68

the FN. Radiological measurements were performed independently by an ENT

69

surgeon and a radiologist. The mean values of the two examiners were analyzed.

AC C

EP

TE D

M AN U

SC

53

3

ACCEPTED MANUSCRIPT The patient was in a supine position with the head turned to the opposite side. A linear

71

skin incision was made approximately 2.5-3 cm behind the ear. A standard

72

mastoidectomy and posterior tympanostomy were performed. After the facial recess

73

(FR) was fully opened, and the positions of the patient’s head and the microscope

74

were adjusted, the RWM was observed through the FR. Before identifying the RW, a

75

posterior tympanostomy was performed using a 2 mm diamond burr, and the EAC

76

was drilled as thinly as possible to obtain the largest view of the RWM. A senior ENT

77

surgeon blinded to the HRCT images reviewed the surgical videos. RWM visibility

78

was then graded and classified as follows[4]: Type I: the RW was entirely exposed,

79

Type II: the RW was partially exposed, Type III: the RW could not be identified (Fig.

80

2). The surgical approaches concerning electrode insertion sites were classified into

81

three groups: RW approach, extended RW approach, cochleostomy (Fig. 3).

82

TE D

M AN U

SC

RI PT

70

Statistical analysis

84

All results were statistically analyzed using SPSS version 17.0 (Chicago, IL, U.S.A.).

85

A p value < 0.05 was considered statistically significant. Differences in radiographic

86

measurements for the different types of intraoperative RW visibility were analyzed

87

using the Kruskal-Wallis test, and Dunn’s multiple-comparisons procedure was used

88

as a post hoc test if to the data did not exhibit a normal distribution. Differences in

89

radiographic measurements among different types of intraoperative RWM visibility

90

were analyzed using one-way ANOVA, and the Bonferroni test was used as a post hoc

AC C

EP

83

4

ACCEPTED MANUSCRIPT 91

test for data with a normal distribution. Spearman’s correlation (R2) was used to

92

evaluate correlations between different variables.

93 Fig. 1 Angular parameters of A and B.

95

l: A line drawn along the posterior wall of the EAC. lf: The line from the posterior

96

margin of the RWM to the lateral margin of the FN. lw: The line from the posterior

97

margin of the RWM to the intersection of the posterior wall of the EAC and the

98

mastoid cortex. A: Angle between lf and lw. lm: A line drawn from the posterior to

99

anterior margin of the RWM. B: Angle between lf and lm. P: Posterior margin of the

SC

M AN U

100

RI PT

94

RWM.

101

Fig. 2 Intraoperative RWM visibility.

103

2a: type I RWM (complete exposure), 2b: type II RWM (partial exposure), 2c: type

104

III RWM (unexposed).

EP AC C

105

TE D

102

106

Fig. 3 Surgical approaches.

107

3a: Round window approach, 3b: Extended round window approach, 3c:

108

Cochleostomy. The arrow shows the vertical segment of the facial nerve. The red part

109

represents the drilling area.

110 111

5

ACCEPTED MANUSCRIPT RESULTS

113

The RWM visibility of all patients was as follows: type I in 38 ears (61.3%), type II in

114

17 ears (27.4%) and type III in 7 ears (11.3%).

115

The average angle A (between lw and lf) was 12.4±1.8°. The average angle B

116

(between lm and lf) was 32.2±5.2°. The mean measurement of the FR width was

117

3.48±0.54 mm(Table 1). A significant correlation was found between the types of

118

RWM visibility and the following two parameters: 1) angle A (between lw and lf),

119

P<0.01, Fig. 4; 2) angle B (between lm and lf), P<0.01, Fig. 5; However, a

120

nonsignificant correlation was found between the degree of RWM visibility and the

121

FR width, P=0.8944.

M AN U

SC

RI PT

112

123

TE D

122

Fig. 4. Comparison of angle A among the 3 types RWM visibility.

125

One-way ANOVA of the 3 groups showed significant differences among angle A for

126

the three different types of RWM visibility. A Bonferroni test showed significant

127

differences between Type I and Type II, Type I and Type III, and Type II and Type III

128

RWM visibility. **: P<0.01. R2= -0.809

AC C

129

EP

124

130

Fig. 5. Comparison of angle B among the 3 types of RWM visibility.

131

The Kruskal-Wallis test of the 3 groups showed significant differences among angle B

132

in the three different groups of RWM visibility. Dunn’s multiple-comparisons test

6

ACCEPTED MANUSCRIPT 133

showed significant differences between Type I and Type II and between Type I and

134

Type III RWM visibility. **: P<0.01. R2= -0.850.

135 DISCUSSION

137

Currently, the RW approach is commonly used in hearing preservation studies[9-13].

138

With this technique, the amount of bone drilling is less than that of the cochleostomy,

139

thereby reducing acoustic trauma and the possibility of bone fragment entry or

140

bleeding into the scala tympani[14]. As the RW is an important anatomical landmark

141

in CI, surgeons routinely review HRCT images to identify any anatomical deformities

142

or insertion difficulties. When the RW is not adequately exposed and the RW

143

approach is not accessible, alternative surgical steps should be considered, such as an

144

extended RW approach or cochleostomy, for electrode insertion. If the RWM

145

visibility could be predicted before surgery, surgeons could be well prepared, the risk

146

of facial nerve damage would be decreased, and hearing preservation benefits would

147

be increased.

148

As some differences in the temporal bone structure exist between adults and

149

children[15], the preoperative HRCT evaluation used in adults may not be applicable

150

in children. Currently, it is widely accepted that the size of the mastoid cavity and FR

151

and the morphology of the RWN are important factors in CI surgery[8]. In children

152

younger than 2 years old, the tympanic annulus and mastoid pneumatization are

153

underdeveloped[16-17]. Lloyd found a statistically significant reduction in basal turn

AC C

EP

TE D

M AN U

SC

RI PT

136

7

ACCEPTED MANUSCRIPT angulation relative to the midsagittal plane with increasing age[18]; this angulation is

155

important because a surgeon depends heavily on anatomic relationships to estimate

156

the anatomy hidden by bone. A reliable method is needed to predict RWM visibility

157

in children and especially in infants less than 1 year old.

158

In the RW approach, RWM visibility is important. Leong et al. developed the St

159

Thomas’ Hospital (STH) classification to evaluate the accessibility of RWM electrode

160

insertion, which was shown to be a sensible method for decision-making regarding

161

surgical approaches[19]. The RWM has a conical shape, with a postero-superior part

162

and an antero-inferior part. We made two angular measurements to predict RWM

163

exposure. is the first involves the angle between the line from the posterior margin of

164

the RWM to the FN lateral margin and the line from the posterior margin of the RWM

165

to the exterior surface of the posterior wall of the EAC (angle A), which represents

166

the antero-inferior part of the RWM. The other measurement consisted of the angle

167

between the line from the posterior margin of the RWM to the lateral margin of the

168

FN and the line from the anterior to the posterior margin of the RWM (angle B),

169

which represents the postero-superior part of the RWM. The statistical analysis of the

170

types of RWM visibility in the current study revealed a significant difference in these

171

two angles among the three different types RWM visibility (Type I, II and III) (p

172

<0.01). The postero-superior part of the RWM, which leads to the scala tympani, is

173

covered by bony overhangs called the RWN. Drilling of the RWN can lead to the

174

acoustic trauma of the intracochlear structures. Shapira et al. evaluated the

AC C

EP

TE D

M AN U

SC

RI PT

154

8

ACCEPTED MANUSCRIPT relationship between the orientation of the RWM and insertion trauma in 15 temporal

176

bones and demonstrated that approximately 10° changes in the orientation to the

177

posterior aspect could increase the risk of traumatic insertion through the RW

178

approach[20]. Mirsalehi et al. found that incomplete visibility of the RWM (less than

179

100%) may increase the risk of residual hearing loss[21]. Our results showed that

180

incomplete RWM visibility could be predicted before surgery through HRCT to get

181

surgeons better prepared for this condition.

182

Initially, most surgeons thought that a larger FR could provide better RW visibility.

183

However, we observed that the FR width did not differ significantly among the three

184

groups. Kashio et al. reported a similar finding, that intraoperative RW niche visibility

185

was not associated with FR width[4]. Bielamowicz found no statistically significant

186

differences in the surgical dimensions between adults and children[22]. Therefore, the

187

FR width is not a determinant for RW exposure for CI surgery in children.

188

There were some limitations of our study. As the measurement lines were drawn

189

manually, deviations could be significant when performing the measurements and

190

predicting RWM visibility. A study including a larger number of cases is required to

191

confirm these findings. It should also be noted that the measurements were obtained

192

from axial CT images, but the RWM is a three-dimensional structure; our angular

193

measurements in the axial plane cannot fully represent the intra-operative surgical

194

view of the RWM. Further studies including 3D reconstruction of the RW may better

195

address this problem.

AC C

EP

TE D

M AN U

SC

RI PT

175

9

ACCEPTED MANUSCRIPT 196 CONCLUSION

198

Preoperative temporal HRCT measurement is a useful method to evaluate the

199

potential challenge of exposing the RWM during cochlear implant surgery. The

200

values of the two angles described could be used to predict RWM visibility, which

201

might be a potential tool for preoperative planning in pediatric CI surgery.

SC

RI PT

197

202 References

204

[1] Pau HW, Just T, Bornitz M, Lasurashvilli N, Zahnert T. Noise exposure of the

205

inner ear during drilling a cochleostomy for cochlear implantation. The Laryngoscope

206

2007;117:535–40.

207

[2] James AL, Papsin BC. Cochlear implant surgery at 12 months of age or younger.

208

Laryngoscope 2010;114:2191–5.

209

[3] Briggs RJS, Tykocinski M, Stidham K, Roberson JB. Cochleostomy site:

210

implications for electrode placement and hearing preservation. Acta Otolaryngol

211

(Stockh) 2005;125:870–6.

212

[4] Kashio A, Sakamoto T, Karino S, Kakigi A, Iwasaki S, Yamasoba T. Predicting

213

round window niche visibility via the facial recess using high-resolution computed

214

tomography. Otol Neurotol 2015;36:e18-23.

215

[5] Fouad YA, Elaassar AS, El-Anwar MW, Sabir E, Abdelhamid A, Ghonimy M.

216

Role of Multislice CT Imaging in Predicting the Visibility of the Round Window in

AC C

EP

TE D

M AN U

203

10

ACCEPTED MANUSCRIPT Pediatric Cochlear Implantation. Otol Neurotol 2017;38:1097–103.

218

[6] Xie L-H, Tang J, Miao W-J, Tang X-L, Li H, Tang A-Z. Preoperative evaluation

219

of cochlear implantation through the round window membrane in the facial recess

220

using high-resolution computed tomography. Surg Radiol Anat 2018; 40(6):705-711.

221

[7] Park E, Amoodi H, Kuthubutheen J, Chen JM, Nedzelski JM, Lin VY. Predictors

222

of round window accessibility for adult cochlear implantation based on pre-operative

223

CT scan: a prospective observational study. J Otolaryngol Head Neck Surg

224

2015;44:20.

225

[8] Vaid S, Vaid N. Imaging for cochlear implantation: structuring a clinically

226

relevant report. Clin Radiol 2014;69:e9.

227

[9] Bruce IA, Todt I. Hearing Preservation Cochlear Implant Surgery. Adv

228

Otorhinolaryngol 2018;81:66–73.

229

[10] Jiam NT, Limb CJ. The impact of round window vs cochleostomy surgical

230

approaches on interscalar excursions in the cochlea: Preliminary results from a

231

flat-panel computed tomography study. World J Otorhinolaryngol - Head Neck Surg

232

2016;2:142–7.

233

[11] Richard C, Fayad JN, Doherty J, Linthicum FH Jr. Round Window Versus

234

Cochleostomy Technique in Cochlear Implantation. Otol Neurotol 2012;33:1181–7.

235

[12] Nordfalk KF, Rasmussen K, Bunne M, Jablonski GE. Deep round window

236

insertion versus standard approach in cochlear implant surgery. Eur Arch

237

Otorhinolaryngol 2016;273:1–8.

AC C

EP

TE D

M AN U

SC

RI PT

217

11

ACCEPTED MANUSCRIPT [13] Roland PS, Wright CG, Isaacson B. Cochlear implant electrode insertion: the

239

round window revisited. Laryngoscope 2007;117:1397–1402.

240

[14] Adunka O, Unkelbach MH, Mack M, Hambek M, Gstoettner W, Kiefer J.

241

Cochlear implantation via the round window membrane minimizes trauma to cochlear

242

structures:

243

2004;124:807–12.

244

[15] McRackan TR, Reda FA, Rivas A, Noble JH, Dietrich MS, Dawant BM, et al.

245

Comparison of cochlear implant relevant anatomy in children versus adults. Otol

246

Neurotol 2012;33:328–34.

247

[16] Woolley AL, Oser AB, Lusk RP, Bahadori RS. Preoperative temporal bone

248

computed tomography scan and its use in evaluating the pediatric cochlear implant

249

candidate. Laryngoscope 1997;107:1100–1106.

250

[17] Dahm MC, Shepherd RK, Clark GM. The postnatal growth of the temporal bone

251

and its implications for cochlear implantation in children. Acta Oto-Laryngol Suppl

252

1993;505:1–39.

253

[18] Lloyd SK, Kasbekar AV, Kenway B, Prevost T, Hockman M, Beale T, et al.

254

Developmental

255

implantation. Otol Neurotol 2010;31:902.

256

[19] Leong AC, Jiang D, Agger A, Fitzgerald-O’Connor A. Evaluation of round

257

window accessibility to cochlear implant insertion. Eur Arch Oto-Rhino-Laryngol

258

2013;270:1237–42.

histologically

controlled

insertion

study.

Acta

Otolaryngol

AC C

EP

TE D

M AN U

SC

a

RI PT

238

changes

in

cochlear

orientation–implications

12

for

cochlear

ACCEPTED MANUSCRIPT [20] Shapira Y, Eshraghi AA, Balkany TJ. The perceived angle of the round window

260

affects electrode insertion trauma in round window insertion – an anatomical study.

261

Acta Otolaryngol (Stockh) 2011;131:284–9.

262

[21] Mirsalehi M, Mohebbi S, Ghajarzadeh M, Lenarz T, Majdani O. Impact of the

263

round window membrane accessibility on hearing preservation in adult cochlear

264

implantation. Eur Arch Oto-Rhino-Laryngol 2017;274:3049.

265

[22] Bielamowicz SA, Coker NJ, Jenkins HA, Igarashi M. Surgical dimensions of the

266

facial recess in adults and children. Arch Otolaryngol Head Neck Surg 1988;114:534–

267

7.

M AN U

SC

RI PT

259

AC C

EP

TE D

268

13

ACCEPTED MANUSCRIPT Table 1. Preoperative measurements and surgical approaches for the three types of

RW visibility Intraoperative view

Preoperative measurements

Type I

Mean degree of

Width of

of angle A°

angle B°

FR (mm)

13.8±0.9

35.2±1.1

3.50±0.6

(complete Type II

Type III

8.4±0.5

(unexposed) P<0.01

31.5±1.1

3.53±0.5

M AN U

11.6±0.6

(partial exposure)

27.2±0.9

3.26±0.3

P<0.01

P>0.05

EP

TE D

RWM: round window membrane, FR: facial recess

AC C

approaches Round

SC

Types of RWM

RI PT

Surgical Mean degree

window Extended round window

Cochleostomy

ED

M AN

D

M A

D

M AN

ED M AN

ED M AN