人工耳蜗电听觉特性与语后聋植入者言语识别的相关性研究
|
摘要
人工耳蜗是目前唯一能使全聋病人恢复听觉言语交流能力的医学装置，近年来受到国内的普遍重视和应用。但患者植入人工耳蜗后听觉言语康复效果呈现出巨大差异，使耳科医生和听力学家面临一个挑战：如何解释和预测人工耳蜗存在的大量个体差异，哪些因素起着至关重要的作用?由于人工耳蜗的电极是通过电刺激听神经而使病人恢复音感的，因而术前使用蜗外骨岬电极或术中、术后使用植入电极本身测量到的电生理特性，有望帮助我们预测或解释植入后的效果。
目前我国植入人工耳蜗的病例数量增长很快，为系统进行人工耳蜗效果评价及相关因素分析提供了很好的素材，但国内缺乏临床实用的、符合国际规范的标准化汉语普通话测试材料，使得听觉言语康复效果的判定没有依据，成为制约此项研究的瓶颈。
本研究利用我院有限的成人语后聋人工耳蜗植入病例，拟初步分析人工耳蜗植入后的电听觉特性(阈值及输入-输出关系、兴奋空间分布、不应期恢复函数等)与汉语声调辨别、单音节识别、语句识别的关系。研究共分三个部分：
第一部分 汉语测听材料的编制与数字化编辑
我们首先将国家标准GB／T15508-1995中颁布的语言清晰度测试单音节KXY表、多音节KXC表、KXJ句表，经数字化录音、编辑后刻录成CD，作为成人语后聋植入者术后言语识别能力的测试材料。为便于今后的临床使用，我们基于音位平衡原则，又编制了25个字为一张、共9张表的简化版KXY单音节表。针对汉语作为有声调语言的特殊性，我们基于毗邻激发模型(NAM，neighborhood activation model)编制了双音节词汇声调毗邻词表。
博士学位论文
人工耳蜗电听觉特性与语后聋植入者言语识别的相关性研究前言
第二部分残余听神经电听觉特性的方法学研究
第二部分应用神经反应遥测困盯)技术记录了90名植入了Nucleus24人
工耳蜗患者的听神经电诱发复合动作电位(E CAP)，其中16名患者在术后1、
2、3个月时对所有22个电极进行了系统的ECAP记录。
本文采用优化的刺激与记录参数来观察ECAP波形，但仍有8例患者未
引出ECAP波形，阳性率为91 .1%。ECAP的幅值增长函数呈现侧向拉伸的/
丁/，引用不同的数值点进行AGF(幅值增长函数)拟合会得到不同的ECAP阂
值及幅值增长函数的斜率。与耳蜗底回相比，耳蜗顶端ECAP的闽值较低、
幅值增长函数斜率较大。16名患者ECAP在术后3个月内变异性很小。
结论:测算ECAP幅值增长函数的斜率及闽值时，应尽可能了解ECAP
幅值增长函数的全貌;患者残存听神经的电听觉特性在术后很稳定，只须测
试一次即可;ECAP闭值和幅值增长函数斜率在一定程度上是残余听神经元
在耳蜗空间分布数量上的反映。
第三部分人工耳蜗电听觉特性与言语识别率的关系
对人工耳蜗电听觉特性的描述，可分成ECAP阂值及幅值增长函数的斜
率、电极的兴奋区间、ECAP不应期恢复时间常数三个方面。
1.对9名语后聋成人(含3名16岁以上的青少年)进行汉语声调辨别、单
音节字(包括75个字的KXY表和25个字的汉语LNT字表)及语句识别测试，
其声调及单音节字的识别率与ECAP闭值及幅值增长函数的斜率有一定的相
关性，但只有声调与ECAP斜率的相关系数具有统计学意义。语句识别率与
ECAP的闽值及斜率无关，可能与语后聋患者具有较强的语言认知能力有关。
语后聋患者对LNT难、易字表的得分无显著性差异，表明所使用的汉语单音
节LNT字表可能不足以体现言语认知能力对言语识别的贡献。
2.应用NRT中扣减算法提供的前掩蔽方案，将探测脉冲与掩蔽脉冲施
加在不同的电极上，可测算出患者每一导电极的兴奋区域。本文测定了20
博士学位论文
人工耳蜗电听觉特性与语后聋植入者言语识别的相关性研究前言
名患者(含4名语后聋患者)的空间分布特征，发现每一导电极的兴奋区域都
交互重叠，存在明显的极间干扰。病史长的患者的电极兴奋区域稍窄、蜗底
电极的兴奋区域稍窄，提示兴奋区域的宽窄可间接反映残存听神经的数量。4
名患者的言语识别率与兴奋区域的宽窄似乎无关，提示耳蜗的时间编码机制
可能对言语识别的意义更大，因为极间干扰的明显存在会使部位编码无法实
施。
3.逐渐增加NRT技术中掩蔽脉冲与探测脉冲的时间间隔，将会有越来
越多的听神经纤维从不应期中恢复，扣减算法的应用前提消失，ECAP幅值
逐渐降低，二者的函数关系成为不应期恢复函数，反映了残存听神经对刺激
速率的最高响应能力。本文测试了38例患者的ECAP不应期恢复函数，呈指
数函数A=c+Ke一叭特征(:为时间常数)，说明残存的听神经纤维具有不同的
时间特性。结果显示，后天聋、病史短的患者，其不应期恢复函数的时间常
数长，提示该时间常数可间接反映残存听神经的数量。9名语后聋患者的言
语识别率与该时间常数具有一定的相关性，但未见统计学意义，这可能与患
者未个性化地设置言语编码策略的刺激速率有关。
总结
ECAP本质上是残存听神经纤维同步放电的空间累积，其幅值增长函数的
斜率是听神经元数量的直接反映，而用NRT技术获得的兴奋区域和不应期恢
复函数由于受到电极间干扰和各神经纤维不同时间特性的影响，只能间接地
反映听神经的状
Cochlear Implant (CI) is the unique medical and bioengineering equipment for the totally deaf persons to rehabilitate their aural/oral communicative abilities. Chinese ENT surgeons and audiologists have recognized its wider perceived applications and achievements throughout the world. But the benefits from a cochlear implant vary enormously between individuals. The challenge for the clinicians is to explain why there is such a wide variability of outcome and to develop ways to more accurately predict potential benefit preoperatively. A cochlear implant is an electrical device that takes over the function of a damaged cochlea. The auditory physiological properties evoked by electrical stimulation via a needle electrode placed onto the promontory or the electrodes implanted in the inner ear are believed to be an indicator of the speech perception of CI users.
The steady increase of cochlea implant cases in China in the past few years provides an opportunity in collecting robust outcome measures in a short period. It would be advantageous for clinical investigation about the key factors in making precise predictions of post-implantation benefits. Unfortunately, there is very little in the way of standardized test material to be used for speakers of Mandarin, the clinical research concerning accurate comparisons of performance and comparisons across languages becomes more difficult.
The electrically evoked compound action potential (ECAP) is the
synchronized response of peripheral auditory nerve to electric current pulses delivered by intra-cochlear electrodes. The study investigated its characteristics in the sensitivity, growth, and temporal, spatial properties of all the adult post-lingual deafened CI recipients who implanted in our hospital. We also measured their speech recognition scores on tests of Mandarin tones, monosyllabic, and sentences. The relationship between electrical auditory properties and speech perceptual performance was analyzed in order to determine the key factors of electrical auditory characteristics that would influence CI outcomes. The study consists of three parts:
Part One:
Compilation of Mandarin Test Material and Digital Edit of KXY
The KXY monosyllabic word lists, KXC multisyllabic word lists, KXJ sentence lists which were published as appendixes in the Chinese National Standard GB/T15508-1995: Acoustics?Speech articulation testing method were selected as the Mandarin test materials for evaluation of speech perceptual abilities of CI recipients in our study. The recording, edit and normalization of test items were digitally processed and all test materials were burned into CD finally.
On account of facility in clinical practice, we simplified the 75-items KXY monosyllabic lists into 25-items lists based on Phonemic Balance consideration. We also compiled Mandarin spondee tonal lexical neighborhood test lists based on the assumption of the Neighborhood Activation Model (NAM).
Part Two:
Protocol of ECAP recording
This part outlines a series of experiments using Neural Response Telemetry (NRT) to measure the electrically evoked compound action potential (ECAP). 90 Nucleus 24 cochlear implant users participated in the study. ECAP thresholds and growth function slopes of 16 CI recipients were measured for all electrodes at post-operative 1,2, 3-months mapping sessions,
Results: 1). We manipulate the optimized stimulating and recording parameters to elicit the ECAP, 91.1% of all 90 recipients showed good ECAP morphology in total 22 electrodes; 2). The shape of ECAP amplitude growth function looks like /. When a linear fit was applied to the amplitude growth function data to determine threshold and slope, the saturated and sub-threshold responses had to be excluded; 3). The slope of ECAP growth function was higher in the apical region of the cochlea, while ECAP threshold was lower in the same region; 4). There was considerably less change of ECAP over the subsequent post-operative time intervals.
Conclusion: 1). ECAP threshold should be determined by regression of
目前我国植入人工耳蜗的病例数量增长很快，为系统进行人工耳蜗效果评价及相关因素分析提供了很好的素材，但国内缺乏临床实用的、符合国际规范的标准化汉语普通话测试材料，使得听觉言语康复效果的判定没有依据，成为制约此项研究的瓶颈。
本研究利用我院有限的成人语后聋人工耳蜗植入病例，拟初步分析人工耳蜗植入后的电听觉特性(阈值及输入-输出关系、兴奋空间分布、不应期恢复函数等)与汉语声调辨别、单音节识别、语句识别的关系。研究共分三个部分：
第一部分 汉语测听材料的编制与数字化编辑
我们首先将国家标准GB／T15508-1995中颁布的语言清晰度测试单音节KXY表、多音节KXC表、KXJ句表，经数字化录音、编辑后刻录成CD，作为成人语后聋植入者术后言语识别能力的测试材料。为便于今后的临床使用，我们基于音位平衡原则，又编制了25个字为一张、共9张表的简化版KXY单音节表。针对汉语作为有声调语言的特殊性，我们基于毗邻激发模型(NAM，neighborhood activation model)编制了双音节词汇声调毗邻词表。
博士学位论文
人工耳蜗电听觉特性与语后聋植入者言语识别的相关性研究前言
第二部分残余听神经电听觉特性的方法学研究
第二部分应用神经反应遥测困盯)技术记录了90名植入了Nucleus24人
工耳蜗患者的听神经电诱发复合动作电位(E CAP)，其中16名患者在术后1、
2、3个月时对所有22个电极进行了系统的ECAP记录。
本文采用优化的刺激与记录参数来观察ECAP波形，但仍有8例患者未
引出ECAP波形，阳性率为91 .1%。ECAP的幅值增长函数呈现侧向拉伸的/
丁/，引用不同的数值点进行AGF(幅值增长函数)拟合会得到不同的ECAP阂
值及幅值增长函数的斜率。与耳蜗底回相比，耳蜗顶端ECAP的闽值较低、
幅值增长函数斜率较大。16名患者ECAP在术后3个月内变异性很小。
结论:测算ECAP幅值增长函数的斜率及闽值时，应尽可能了解ECAP
幅值增长函数的全貌;患者残存听神经的电听觉特性在术后很稳定，只须测
试一次即可;ECAP闭值和幅值增长函数斜率在一定程度上是残余听神经元
在耳蜗空间分布数量上的反映。
第三部分人工耳蜗电听觉特性与言语识别率的关系
对人工耳蜗电听觉特性的描述，可分成ECAP阂值及幅值增长函数的斜
率、电极的兴奋区间、ECAP不应期恢复时间常数三个方面。
1.对9名语后聋成人(含3名16岁以上的青少年)进行汉语声调辨别、单
音节字(包括75个字的KXY表和25个字的汉语LNT字表)及语句识别测试，
其声调及单音节字的识别率与ECAP闭值及幅值增长函数的斜率有一定的相
关性，但只有声调与ECAP斜率的相关系数具有统计学意义。语句识别率与
ECAP的闽值及斜率无关，可能与语后聋患者具有较强的语言认知能力有关。
语后聋患者对LNT难、易字表的得分无显著性差异，表明所使用的汉语单音
节LNT字表可能不足以体现言语认知能力对言语识别的贡献。
2.应用NRT中扣减算法提供的前掩蔽方案，将探测脉冲与掩蔽脉冲施
加在不同的电极上，可测算出患者每一导电极的兴奋区域。本文测定了20
博士学位论文
人工耳蜗电听觉特性与语后聋植入者言语识别的相关性研究前言
名患者(含4名语后聋患者)的空间分布特征，发现每一导电极的兴奋区域都
交互重叠，存在明显的极间干扰。病史长的患者的电极兴奋区域稍窄、蜗底
电极的兴奋区域稍窄，提示兴奋区域的宽窄可间接反映残存听神经的数量。4
名患者的言语识别率与兴奋区域的宽窄似乎无关，提示耳蜗的时间编码机制
可能对言语识别的意义更大，因为极间干扰的明显存在会使部位编码无法实
施。
3.逐渐增加NRT技术中掩蔽脉冲与探测脉冲的时间间隔，将会有越来
越多的听神经纤维从不应期中恢复，扣减算法的应用前提消失，ECAP幅值
逐渐降低，二者的函数关系成为不应期恢复函数，反映了残存听神经对刺激
速率的最高响应能力。本文测试了38例患者的ECAP不应期恢复函数，呈指
数函数A=c+Ke一叭特征(:为时间常数)，说明残存的听神经纤维具有不同的
时间特性。结果显示，后天聋、病史短的患者，其不应期恢复函数的时间常
数长，提示该时间常数可间接反映残存听神经的数量。9名语后聋患者的言
语识别率与该时间常数具有一定的相关性，但未见统计学意义，这可能与患
者未个性化地设置言语编码策略的刺激速率有关。
总结
ECAP本质上是残存听神经纤维同步放电的空间累积，其幅值增长函数的
斜率是听神经元数量的直接反映，而用NRT技术获得的兴奋区域和不应期恢
复函数由于受到电极间干扰和各神经纤维不同时间特性的影响，只能间接地
反映听神经的状
Cochlear Implant (CI) is the unique medical and bioengineering equipment for the totally deaf persons to rehabilitate their aural/oral communicative abilities. Chinese ENT surgeons and audiologists have recognized its wider perceived applications and achievements throughout the world. But the benefits from a cochlear implant vary enormously between individuals. The challenge for the clinicians is to explain why there is such a wide variability of outcome and to develop ways to more accurately predict potential benefit preoperatively. A cochlear implant is an electrical device that takes over the function of a damaged cochlea. The auditory physiological properties evoked by electrical stimulation via a needle electrode placed onto the promontory or the electrodes implanted in the inner ear are believed to be an indicator of the speech perception of CI users.
The steady increase of cochlea implant cases in China in the past few years provides an opportunity in collecting robust outcome measures in a short period. It would be advantageous for clinical investigation about the key factors in making precise predictions of post-implantation benefits. Unfortunately, there is very little in the way of standardized test material to be used for speakers of Mandarin, the clinical research concerning accurate comparisons of performance and comparisons across languages becomes more difficult.
The electrically evoked compound action potential (ECAP) is the
synchronized response of peripheral auditory nerve to electric current pulses delivered by intra-cochlear electrodes. The study investigated its characteristics in the sensitivity, growth, and temporal, spatial properties of all the adult post-lingual deafened CI recipients who implanted in our hospital. We also measured their speech recognition scores on tests of Mandarin tones, monosyllabic, and sentences. The relationship between electrical auditory properties and speech perceptual performance was analyzed in order to determine the key factors of electrical auditory characteristics that would influence CI outcomes. The study consists of three parts:
Part One:
Compilation of Mandarin Test Material and Digital Edit of KXY
The KXY monosyllabic word lists, KXC multisyllabic word lists, KXJ sentence lists which were published as appendixes in the Chinese National Standard GB/T15508-1995: Acoustics?Speech articulation testing method were selected as the Mandarin test materials for evaluation of speech perceptual abilities of CI recipients in our study. The recording, edit and normalization of test items were digitally processed and all test materials were burned into CD finally.
On account of facility in clinical practice, we simplified the 75-items KXY monosyllabic lists into 25-items lists based on Phonemic Balance consideration. We also compiled Mandarin spondee tonal lexical neighborhood test lists based on the assumption of the Neighborhood Activation Model (NAM).
Part Two:
Protocol of ECAP recording
This part outlines a series of experiments using Neural Response Telemetry (NRT) to measure the electrically evoked compound action potential (ECAP). 90 Nucleus 24 cochlear implant users participated in the study. ECAP thresholds and growth function slopes of 16 CI recipients were measured for all electrodes at post-operative 1,2, 3-months mapping sessions,
Results: 1). We manipulate the optimized stimulating and recording parameters to elicit the ECAP, 91.1% of all 90 recipients showed good ECAP morphology in total 22 electrodes; 2). The shape of ECAP amplitude growth function looks like /. When a linear fit was applied to the amplitude growth function data to determine threshold and slope, the saturated and sub-threshold responses had to be excluded; 3). The slope of ECAP growth function was higher in the apical region of the cochlea, while ECAP threshold was lower in the same region; 4). There was considerably less change of ECAP over the subsequent post-operative time intervals.
Conclusion: 1). ECAP threshold should be determined by regression of
引文
1．岳月 寂静之声：人工耳蜗的发展．科学 1999，51：40-43．
2．曹克利．人工耳蜗的应用与进展．世界医疗器械．2000，6：50-52．
3. Kirk KR. Challenges in the clinical investigation of cochlear implant outcomes. In Niparko JK, ed. Cochlear Implants: Principles & Practices. Philadelphia: Lippincott Williams & Wilkins, 2000: 225-259.
4．许时昂，韩德民．回顾．见韩德民主编<人工耳蜗>本嗣裎郎霭嫔纾?003，1-6．
5．Ng M，Niparko JK，Nager GT．重度和极重度感音神经性聋的内耳病理学．见(人工耳蜗植入：原理与实践)．Niparko JK et al．著，王直中、曹克利等译．人民卫生出版社，2003年，北京．
6．曹克利．人工耳蜗的进展与应用．中华医学会耳鼻咽喉科高级医师研修班汇编，2001．66-72．
7. Nadol JB. Patterns of neural degeneration in the human cochlea and auditory nerve: implications for cochlear implantation. Otolaryngol. Head Neck Surg. 1997; 117: 220-228.
8．梁之安．听觉感受和辨别的神经机制，上海，上海科技教育出版社，1999．
9. Kiang NYS. Discharge Patterns of Single Fibers in the Cat's Auditory Nerve. MIT Press, Cambridge, Mass. 1965.
10．刘爱国，王正敏．听神经时间编码特性．中国眼耳鼻喉科杂志，2002，2：183-186．
11．刘爱国，王正敏．电刺激下听神经兴奋性研究．国外医学耳鼻喉科分册，2003，27：216-220．
12．祝小莉、曹克利 电刺激诱发电位在人工耳蜗植入术中的作用．听力学及言语疾病杂志，2000，8：247-248．
13. Shallop JK, Beiter AL, Goin DW, et al. Electrically evoked auditory brainstem responses(EABR) and middle latency responses(EMLR) obtained from patients with the Nucleus multichannel cochlear implant. Ear Hear 1990, 11: 5-15.
14. Starr A, Brackmann DE. Brain stem potentials evoked by electrical stimulation of the cochea in human subjects. Ann. Otol. Laryngol. Rhinol. 1979, 88, 59-67.
15. Gibson W, Sanli H. The role of round window electrophysiological techniques in the selection of children for cochlear implants. Advances in Oto-Rhino-Laryngology, 1999, 57, 148-151
16. Abbas PJ, Brown CJ, Shallop JK et al. Summary of results using the Nucleus CI24M implant to record the electrically evoked compound action potential. Ear and Hearing, 1999, 20: 45-59.
17. Nilsson M, McCaw V, Soli SD. User Manual: A minimal speech test battery for adult cochlear implant patients. House Ear Institute, Los Angeles, CA. 1996.
18．张家骤．国家标准GB／T 15508-1995《声学 语言清晰度测试方法》1995．
2．曹克利．人工耳蜗的应用与进展．世界医疗器械．2000，6：50-52．
3. Kirk KR. Challenges in the clinical investigation of cochlear implant outcomes. In Niparko JK, ed. Cochlear Implants: Principles & Practices. Philadelphia: Lippincott Williams & Wilkins, 2000: 225-259.
4．许时昂，韩德民．回顾．见韩德民主编<人工耳蜗>本嗣裎郎霭嫔纾?003，1-6．
5．Ng M，Niparko JK，Nager GT．重度和极重度感音神经性聋的内耳病理学．见(人工耳蜗植入：原理与实践)．Niparko JK et al．著，王直中、曹克利等译．人民卫生出版社，2003年，北京．
6．曹克利．人工耳蜗的进展与应用．中华医学会耳鼻咽喉科高级医师研修班汇编，2001．66-72．
7. Nadol JB. Patterns of neural degeneration in the human cochlea and auditory nerve: implications for cochlear implantation. Otolaryngol. Head Neck Surg. 1997; 117: 220-228.
8．梁之安．听觉感受和辨别的神经机制，上海，上海科技教育出版社，1999．
9. Kiang NYS. Discharge Patterns of Single Fibers in the Cat's Auditory Nerve. MIT Press, Cambridge, Mass. 1965.
10．刘爱国，王正敏．听神经时间编码特性．中国眼耳鼻喉科杂志，2002，2：183-186．
11．刘爱国，王正敏．电刺激下听神经兴奋性研究．国外医学耳鼻喉科分册，2003，27：216-220．
12．祝小莉、曹克利 电刺激诱发电位在人工耳蜗植入术中的作用．听力学及言语疾病杂志，2000，8：247-248．
13. Shallop JK, Beiter AL, Goin DW, et al. Electrically evoked auditory brainstem responses(EABR) and middle latency responses(EMLR) obtained from patients with the Nucleus multichannel cochlear implant. Ear Hear 1990, 11: 5-15.
14. Starr A, Brackmann DE. Brain stem potentials evoked by electrical stimulation of the cochea in human subjects. Ann. Otol. Laryngol. Rhinol. 1979, 88, 59-67.
15. Gibson W, Sanli H. The role of round window electrophysiological techniques in the selection of children for cochlear implants. Advances in Oto-Rhino-Laryngology, 1999, 57, 148-151
16. Abbas PJ, Brown CJ, Shallop JK et al. Summary of results using the Nucleus CI24M implant to record the electrically evoked compound action potential. Ear and Hearing, 1999, 20: 45-59.
17. Nilsson M, McCaw V, Soli SD. User Manual: A minimal speech test battery for adult cochlear implant patients. House Ear Institute, Los Angeles, CA. 1996.
18．张家骤．国家标准GB／T 15508-1995《声学 语言清晰度测试方法》1995．