癫痫样放电是否需要治疗？特邀综述《癫痫杂志》

StarCare

转发群友分享！

文献网址：https://pmc.ncbi.nlm.nih.gov/articles/PMC6294573/



癫痫样放电是否需要治疗？特邀综述《癫痫杂志》

美国卫生与公众服务部公共访问
作者手稿
《癫痫》杂志。2018年12月14日可在公共医学中心查阅；2015年10月正式发表，56（10）：1492-1504，doi:10.1111/epi.13108

作者：伊万·桑切斯·费尔南德斯（医学博士）1,2、托比亚斯·洛登肯珀（医学博士）2、阿里斯泰亚·S·加兰诺普洛（医学博士、哲学博士）3、所罗门·L·莫舍（医学博士）3,4
1 美国马萨诸塞州波士顿市波士顿儿童医院神经科癫痫与临床神经生理科、哈佛医学院
2 西班牙巴塞罗那市巴塞罗那大学圣胡安·德迪乌医院儿童神经科
3 美国纽约州布朗克斯市蒙特菲奥雷医疗中心/爱因斯坦癫痫管理中心，阿尔伯特·爱因斯坦医学院索尔·R·科里神经科、多米尼克·P·普尔普拉神经科学系、发育医学实验室
4 美国纽约州布朗克斯市阿尔伯特·爱因斯坦医学院蒙特菲奥雷医疗中心儿科

通信作者：托比亚斯·洛登肯珀（医学博士），哈佛医学院神经科副教授，波士顿儿童医院癫痫与临床神经生理科，地址：美国马萨诸塞州波士顿市朗伍德大道300号费根9号楼，邮编02115；电话：617-355-2443；传真：617-730-0463；邮箱：tobias.loddenkemper@childrens.harvard.edu

利益冲突披露：所有作者均无相关利益冲突需披露。
我们确认已阅读本刊关于伦理出版相关问题的立场声明，并声明本研究报告符合相关准则。

摘要

目的：评估尖波、棘慢波等非癫痫发作模式下出现的癫痫样放电（EDs）对认知功能的影响，并探讨考虑对癫痫样放电进行治疗的适用情况。
方法：文献综述。
结果：癫痫样放电可短暂干扰人类认知功能，长期频繁的癫痫样放电可能损害人类远期认知功能。但由于潜在病因、癫痫发作、药物作用等多种混杂因素，癫痫样放电对远期认知结局的影响尚无统一结论。现有研究的局限性包括缺乏标准化的癫痫样放电定量方法，以及尚未形成公认的自动化尖波定量手段。目前尚无确凿证据支持或反对对癫痫样放电进行治疗，因此提出一种基于非循证医学的实用处理方案：无症状人群的癫痫样放电无需治疗，因治疗风险可能大于其不确定的获益；当患者出现无法用潜在病因、共病或癫痫发作严重程度解释的认知功能障碍/倒退或神经系统症状时，可考虑对癫痫样放电进行试验性治疗；对于合并癫痫或癫痫样放电且存在认知/神经功能障碍的患者，为控制潜在的癫痫综合征，需进行针对性治疗。
意义：癫痫样放电可在短期内导致人类认知或神经功能障碍，其对远期认知结局的影响证据相互矛盾，目前尚无证据支持或反对对无症状性癫痫样放电进行治疗。

关键词：抗癫痫药物；认知；癫痫样放电；记忆；睡眠

一、引言

约1%~5%的人群脑电图（EEG）中可检测到癫痫样放电，癫痫患者的癫痫样放电检出率显著高于健康对照人群。癫痫样放电包括孤立或短程发放的尖波、多棘波、锐波或棘慢波复合波，无明显临床相关症状。癫痫样放电可急性干扰人类的认知或神经功能，其持续时间和分布区域与功能障碍类型相关，但慢性癫痫样放电对神经元环路及远期认知结局的确切影响尚未明确，抑制癫痫样放电的治疗获益与风险也缺乏清晰结论。

本研究旨在探讨癫痫样放电对认知功能的影响（聚焦于非典型癫痫发作模式/状态下出现的癫痫样放电），并为癫痫样放电的治疗时机提供建议。研究仅针对非急性门诊场景，排除意识障碍状态，且以临床数据为核心（近期已有相关综述探讨动物实验研究结果）。

二、癫痫样放电的定量分析与认知功能

（一）癫痫样放电的客观定量尝试

多数癫痫样放电定量方法为睡眠期癫痫性电持续状态（ESES）专门设计，该脑电图表现为睡眠期几乎持续的癫痫样放电。

1. 棘慢波指数（SWI）与棘慢波百分比（SWP）：二者用于评估脑电图中癫痫样放电占总监测时长的比例。最初针对ESES的研究中，棘慢波指数被定义为慢波睡眠期癫痫样放电的占比，但因方法学细节缺失、计算解读标准不一，其可重复性受限。有研究提出将棘慢波指数定义为至少出现1次癫痫样放电的1秒时间窗占总时长的百分比，该方法后被采纳并重新命名为棘慢波百分比。棘慢波百分比的优势为可重复性强、资源消耗低，但当癫痫样放电频率超过1次/秒时，无法区分脑电图的严重程度。

2. 尖波频率（SF）：指单位时间内癫痫样放电的总数，通常以100秒为统计单位。若未实现自动化，该方法较棘慢波百分比更耗时，且易出现抽样误差，但对高频尖波的癫痫样放电严重程度具有更好的区分度。

3. 计算机化定量方法：自动化方法可实现癫痫样放电的快速、可重复定量，但仅能识别符合预设参数的癫痫样放电（或其组成部分），需根据不同患者调整参数、筛查排除伪差，且无法识别所有形态的癫痫样放电，因此尚未在临床实践中广泛应用。

（二）癫痫样放电影响定量分析的挑战

癫痫样放电评估中的诸多细节问题，使其对临床结局的影响难以直接定量分析。

1. 分布与偏侧性：临床功能障碍与癫痫样放电的定位相关，位于功能区、多灶性或广泛性的癫痫样放电，其危害性高于非认知相关脑区的癫痫样放电。

2. 波幅差异：高波幅与低波幅癫痫样放电对认知功能的潜在危害是否存在差异，目前尚未明确。

3. 形态学特征：癫痫样放电包括尖波、多棘波、锐波的不同组合形式，可伴或不伴后续慢波，有研究认为后续慢波也可干扰皮质功能。

4. 睡眠-觉醒周期：癫痫样放电的严重程度并非始终独立于睡眠-觉醒阶段，不同阶段的癫痫样放电对癫痫性脑病严重程度的影响尚不明确。由于多数认知测试依赖患者的主动反应（需在觉醒状态下完成），因此觉醒状态下的认知功能评估更易实施；而睡眠相关的认知缺陷，若无睡眠期主动测试，更易被认为与信息储存和神经可塑性相关。

（三）癫痫样放电与认知/功能结局的相关性

癫痫样放电的短期（急性）影响评估仅局限于可通过快速反应时评估的认知功能，且认知评估本身可能改变癫痫样放电的表现，进一步增加结果解读的复杂性。癫痫样放电对远期认知功能的影响评估则受限于其认知效应的情境性和时间依赖性，且由于脑电图和特异性认知测试通常在症状出现后进行，症状发生前的认知功能和脑电图基线状态难以准确评估，症状（认知、癫痫发作）与癫痫样放电的时间演变规律也常不明确。

三、癫痫样放电对认知功能的影响

（一）无癫痫病史的无症状人群

无癫痫病史的无症状人群中，儿童癫痫样放电的检出率为0~6%，成人为0~7%。健康人群中偶然发现的癫痫样放电，其临床意义尚未明确：后续发展为癫痫的概率较低（儿童约6%，成人约2%），且有研究提示此类人群可能存在轻微认知和行为异常，但该结论仍需前瞻性、严格对照的大样本研究进一步验证。

（二）合并非癫痫性神经系统疾病的癫痫样放电患者

无癫痫发作史的住院和门诊患者，癫痫样放电检出率（2%~14%）高于社区人群研究结果，其认知层面的意义尚不明确，可能为脑功能异常的生物标志物。

1. 注意缺陷多动障碍（ADHD）：一项纳入48例平均年龄9.4岁（6.7~14.9岁）注意缺陷多动障碍儿童的研究中，16例伴中央颞区尖波的患儿、16例无中央颞区尖波的患儿与16例健康对照儿童相比，中央颞区尖波在冲动性增加、反应抑制和干扰控制能力下降的注意缺陷多动障碍儿童中更常见。此外，睡眠多导监测显示注意缺陷多动障碍儿童的癫痫样放电检出率更高：42例接受监测的注意缺陷多动障碍儿童中，53.1%存在癫痫样放电，其中3例监测期间出现癫痫发作，其余合并语言障碍、运动性失用等其他神经系统共病。另一项纳入23例4~17岁史密斯-莱姆利-奥皮茨综合征患儿的系列研究中，51%的患儿脑电图存在异常（多为癫痫样放电），且同一患儿某次脑电图中癫痫样放电的出现，可预测其注意缺陷多动障碍症状严重程度平均增加27%。目前尚无证据证实，排除其他潜在病因后，癫痫样放电是导致注意缺陷多动障碍症状的直接原因。

2. 语言障碍：睡眠多导监测或标准脑电图研究证实，语言障碍儿童的癫痫样放电发生率更高。52例4~11岁语言发育障碍患儿中，50%检测到癫痫样放电（84%位于左侧），显著高于年龄匹配对照组的10%。少数研究显示，语言发育障碍的类型与癫痫样放电的发生率和严重程度相关：睡眠多导监测中，表达性发育性语言障碍儿童的癫痫样放电检出率（37.5%）高于对照组（5.1%），其中包括非快速眼动睡眠期的广泛性癫痫样放电，9例患儿中有2例在觉醒或快速眼动睡眠期也检测到癫痫样放电；56%的表达性发育性语言障碍儿童存在频繁癫痫样放电（睡眠期棘慢波指数2.5%~66%）。目前无证据表明此类癫痫样放电是导致语言障碍的直接原因，其更可能是潜在病因的表现；值得注意的是，上述研究中2例合并癫痫样放电的对照儿童，睡眠期棘慢波指数达21.6%，但无语言功能障碍。

3. 孤独症谱系障碍（ASD）与智力障碍（ID）：孤独症谱系障碍和智力障碍人群的癫痫及癫痫样放电发生率均显著高于健康对照人群，但其癫痫样放电是加重认知损害的原因，还是仅为脑功能异常的表现，目前尚未明确。总体而言，20%~60%的孤独症谱系障碍儿童存在癫痫样异常，排除癫痫患者后，仍有8%~20%的孤独症谱系障碍儿童检测到癫痫样放电。与其他孤独症谱系障碍亚型相比，阿斯伯格综合征患者的癫痫样放电和临床癫痫发作发生率更低；而攻击行为史与癫痫样放电相关，与临床癫痫发作无明显关联。

即使无癫痫病史，孤独症倒退的儿童也更易出现癫痫样放电，但二者的时间关联难以明确。与单纯语言倒退（朗道-克莱夫纳综合征，LKS）相比，孤独症倒退与癫痫发作（8% vs 33%）或癫痫样放电（28% vs 56%）的关联性更低，发病年龄更小，且更常合并发育迟缓。因此，相较于单纯语言倒退，癫痫样放电在孤独症倒退的发病机制中可能作用更小，或作为替代标志物的价值更弱。现有实验证据也支持另一种可能性：此类年龄特异性综合征潜在病因的发育影响，可能受年龄、性别或脑区特异性因素的进一步调节。

（三）癫痫患者

多项研究探讨了局灶性或广泛性癫痫样放电在特定综合征中的短期和远期影响：癫痫样放电可影响中枢信息处理、刺激反应时和认知应答。另有研究对比癫痫样放电与癫痫发作对学习和认知的影响，发现无癫痫发作时，癫痫样放电对注意力和信息处理速度无影响或仅存在轻微独立影响，高频癫痫样放电时该影响更明显。

1. 癫痫样放电的短期影响

（1）对神经心理测试的影响：癫痫样放电引发的短期认知损害并非一定由全面注意力障碍导致，更可能反映受累脑区特定功能的受损，且认知缺陷与癫痫样放电的出现时间相关。癫痫样放电需出现在“正确的脑区”和“正确的时间”，才会导致特定的认知功能障碍。
（2）对日常活动的影响：20例儿童的学业任务表现评估显示，癫痫样放电频率越高，测试表现越差，尤以算术学科为著。6例合并癫痫样放电的受试者在高速公路以90km/h匀速行驶420km的驾驶表现监测显示，癫痫样放电出现期间，3例受试者的车道保持能力受损，1例存在受损趋势；该损害即使在癫痫发作多年缓解的患者中仍可出现，且标准实验室认知测试（如科尔西积木测试、短时言语记忆测试）结果，无法预测受试者的驾驶测试表现。提示癫痫样放电可损害对注意力要求较高的任务表现（即使在癫痫无发作患者中），但该损害无法通过标准神经心理测试预测。
（3）癫痫样放电与短暂性认知损害（TCI）、非惊厥性小发作的关系：若癫痫样放电导致可测量的认知改变，则其并非亚临床表现，而是临床性短暂性认知损害或发作期表现（轻微癫痫发作）。多项视觉、听觉反应测试发现，癫痫样放电出现的脑电时段与反应延迟存在时间关联，有研究提出癫痫样放电的影响始于尖波出现前，止于后续慢波结束。另一项纳入188例6~18岁儿童的研究中，在2小时认知测试同时进行脑电图监测，将研究人群分为：测试期间出现非惊厥性小发作的癫痫患儿（表现为凝视或轻微动作，持续数秒）、测试期间无非惊厥性小发作的癫痫患儿、无癫痫的健康对照。结果显示，非惊厥性小发作可显著损害认知功能（发作时长多短于19秒，以局灶性发作为主，失神发作和肌阵挛发作为次）；无癫痫发作的癫痫样放电也可损害认知功能，但影响更轻微。

癫痫样放电是否为非惊厥性癫痫发作谱系中的轻度表现？或其是否为易诱发认知功能障碍和/或癫痫发作的潜在病因/状态的早期标志物？明确上述问题，有助于优化癫痫样放电患者的管理方案。
（4）结果解读的挑战：短期认知测试的研究对象存在固有偏倚，因研究通常优先选择觉醒清醒状态下癫痫样放电频繁的受试者，因此研究结果可能无法适用于癫痫样放电频率较低或仅睡眠期出现癫痫样放电的人群。此外，外界刺激和任务注意力水平，也可能改变癫痫样放电的频率和特征。

2. 癫痫样放电的长期影响

（1）儿童局灶性癫痫综合征：儿童中央颞区棘波癫痫（CECTS）、帕纳约托普洛斯综合征、晚发性儿童枕叶癫痫（加斯陶型）具有共同特征：癫痫发作相对轻微、发作频率低，癫痫样放电频繁且睡眠期易化，即使癫痫发作缓解后，癫痫样放电仍可持续数年。儿童中央颞区棘波癫痫研究显示，睡眠期癫痫样放电越频繁，神经心理发育异常（尤以言语领域为著）越明显；另有研究发现，29%的患儿存在学业和家庭功能受损，且不良预后的预测因素与癫痫样放电相关，与癫痫发作无关，但也有研究未发现尖波频率与认知表现存在相关性。一项纳入26例合并学习困难的局灶性癫痫样放电患儿（19例为儿童中央颞区棘波癫痫、2例为帕纳约托普洛斯综合征、1例为局灶性癫痫发作、4例无癫痫发作史）的研究显示，基线中枢信息处理速度与癫痫样放电频率呈负相关（未排除无癫痫发作史患儿）；随访发现，中枢信息处理速度下降与癫痫样放电频率增加、癫痫发作持续存在均相关，但目前无法明确癫痫发作与癫痫样放电对中枢信息处理的独立影响。

局灶性癫痫综合征患儿中，左侧中央颞区癫痫样放电者在复杂语言任务中表现更差，枕叶尖波者在同步信息处理中表现更差，尤以视觉转换任务为著。综上，基于儿童局灶性癫痫合并癫痫样放电的小样本系列研究，提示癫痫样放电可能导致远期认知结局变差，且该损害与癫痫样放电的定位相关。
（2）全面性癫痫：儿童失神癫痫患者的棘慢波放电可在短期内干扰注意力、意识和信息处理，且有证据提示其存在远期认知缺陷。一项对比16例儿童失神癫痫、14例1型糖尿病和15例健康儿童的研究显示，儿童失神癫痫患儿在智力、记忆、学业成就、精细运动速度或信息处理速度方面与其他两组无显著差异，但在问题解决、词语流畅性、复杂运动控制、注意力/行为抑制和社会心理功能方面表现更差；且其远期社会心理结局，较青少年类风湿关节炎等非癫痫性慢性疾病患者更差。青少年肌阵挛癫痫也可能存在远期轻微的额叶加工功能障碍，推测癫痫样放电可能通过渐进性干扰丘脑-额叶神经元环路，导致该环路介导的功能缺陷，但功能缺陷与癫痫样放电也可能均为遗传决定的丘脑-额叶神经元环路异常的生物标志物。
（3）合并语言/认知倒退的癫痫性脑病：睡眠期癫痫性电持续状态的特征为非快速眼动睡眠期几乎持续的尖波发放，常合并不同程度的认知倒退，其持续时间历来被认为是认知预后不良的标志物。一项纳入209例患者的文献综述显示，睡眠期癫痫性电持续状态持续超过2年，与认知预后不良相关；30例对治疗有反应的睡眠期癫痫性电持续状态患者研究显示，其持续时间与随访时认知功能恢复至基线水平的程度相关：持续时间短于13个月的患儿中，50%认知功能恢复至基线，而持续时间超过18个月的患儿无一恢复。癫痫手术前后认知功能的客观评估显示，睡眠期癫痫性电持续状态的缓解可改善认知功能，但由于研究样本量较小，且存在潜在病因、癫痫发作负荷、抗癫痫药物治疗等混杂因素，不能将认知改善完全归因于癫痫样放电的减少。另有一项纳入7例患者的研究，未发现睡眠期癫痫性电持续状态的持续时间与认知结局存在关联。
（4）共同病理生理机制：部分基因突变可使同一患者出现不同程度的孤独症谱系障碍、癫痫样放电、癫痫和智力障碍，提示其存在共同的发病机制，可能由异常的突触可塑性和兴奋/抑制失衡介导。这为此类疾病的治疗提供了新方向，但也提示癫痫样放电可能仅为潜在病理过程的表现，而非认知倒退的原因。

多种潜在机制可导致以显著癫痫样放电为特征的睡眠期癫痫性电持续状态，GRIN2A基因突变和早期丘脑病变均与远期癫痫样放电显著和睡眠期癫痫性电持续状态的发生相关，进一步证实了不同病因可通过共同通路导致相似临床表型的观点。

3. 癫痫样放电认知影响的评估方法

长期暴露于癫痫样放电是否损害认知功能、治疗癫痫样放电是否可逆转该潜在损害，目前仍无明确答案。癫痫样放电对短期认知功能的影响，通常通过同一受试者在癫痫样放电出现和消失两种状态下的对比评估，实验均在觉醒状态下数小时内完成。睡眠期癫痫性电持续状态或儿童中央颞区棘波癫痫患者适用于该评估方法：可在基线时进行正式神经心理评估，通过抗癫痫药物减少其癫痫样放电后，再次进行干预后神经心理评估。

但治疗前后的对比分析，仅能校正癫痫发作负荷等有限混杂因素，无法校正性别、遗传背景等个体固有因素。此类研究对短期结局评估有一定价值，但远期结局的解读受限于难以区分：治疗相关的认知改善，与自然发育、教育体系或生活环境变化带来的认知能力提升。若初步研究结果良好，可进一步开展大样本研究，纳入同质性较高的人群（如同一综合征或相同遗传背景），在考虑多种已知混杂因素的前提下，对比接受/未接受治疗的癫痫样放电患者的发育轨迹；且两类研究均需对癫痫样放电的负荷、类型和定位进行细致评估。

四、是否需要治疗？

在获得更充分的循证医学证据前，临床需做出治疗决策。对癫痫样放电治疗的犹豫，源于目前尚无确凿证据证实其与远期认知损害存在因果关系。本研究针对不同临床场景，提出一种基于非循证医学的癫痫样放电实用管理方案。

（一）无症状人群

无症状无癫痫病史人群中，癫痫样放电的认知影响尚不明确，且后续发展为癫痫的风险较低（可能与遗传特征相关）。因此，无症状无癫痫病史人群中偶然发现的癫痫样放电无需治疗，因治疗获益尚不明确。

（二）无癫痫发作但合并癫痫样放电及认知功能障碍/倒退的患者

有报道显示，抗癫痫药物可改善此类患者的部分认知功能，但大样本研究结果不一。一项双盲单交叉试验中，8例6~12岁合并学习和行为问题的癫痫样放电患儿，随机接受丙戊酸钠或安慰剂治疗，结果显示丙戊酸钠治疗组患儿注意力分散增加、反应时延长、记忆评分降低，且无临床改善。一项前瞻性开放标签研究中，6例7~15岁合并局灶性癫痫样放电的学习困难患儿，接受左乙拉西坦治疗，随访10周发现，4例患儿的广泛记忆学习评估量表评分改善，但韦氏个体成就测试评分无变化。综上，尽管部分患者接受抗癫痫药物治疗后认知功能有所改善，但这可能反映疾病的自然波动，目前尚无足够证据推荐对无癫痫发作但合并认知功能障碍/倒退的癫痫样放电患者使用抗癫痫药物，且抗癫痫药物可能加重认知功能损害。

（三）癫痫性脑病合并癫痫样放电及认知功能障碍/倒退的患者

对于合并倒退的癫痫性脑病患者，促肾上腺皮质激素、大剂量糖皮质激素或免疫治疗被报道可有效改善脑电图表现并治疗认知倒退；大剂量苯二氮䓬类药物减少睡眠期癫痫性电持续状态患者的癫痫样放电后，部分患者的认知功能得到改善。32例睡眠期癫痫性电持续状态患儿研究显示，丙戊酸钠或乙琥胺减少癫痫样放电后，患者认知功能改善；对睡眠期癫痫性电持续状态患者的致痫灶行癫痫手术，至少半数患者的认知损害停止进展甚至改善。但由于研究数据收集缺乏对照，且疾病本身存在自然波动，同时受其他因素潜在影响，结果解读需谨慎。综上，对于癫痫性脑病合并可能与癫痫样放电相关的认知功能障碍/倒退的患者，可考虑试验性治疗。

（四）癫痫发作控制良好但合并癫痫样放电及认知功能障碍/倒退的患者

一项双盲安慰剂对照单交叉研究中，61例7~17岁癫痫发作控制良好（局灶性或全面性癫痫发作无/偶发）且合并行为和/或认知问题的患儿，随机接受拉莫三嗪或安慰剂治疗，结果显示拉莫三嗪治疗后癫痫样放电减少的患者，其行为表现（通过父母和教师康纳斯评定量表评估）得到改善，且对癫痫发作频率无显著影响。因此，对此类患者（尤其认知功能障碍呈进行性加重者），可考虑抗癫痫药物试验性治疗。

（五）合并癫痫样放电、认知功能障碍/倒退且癫痫发作持续存在的患者

对于合并认知功能障碍/倒退且癫痫发作持续存在的患者，需进行治疗以控制癫痫发作；目前尚无足够证据支持，将癫痫样放电的消失作为现有治疗的更佳终点。

（六）癫痫样放电的治疗方案

评估抗癫痫药物对癫痫样放电疗效的研究较少，且结果解读需考虑癫痫样放电的自然波动。一项大样本儿科研究对比213对抗癫痫药物使用前后的脑电图，癫痫样放电抑制率（治疗后脑电图中所有类型癫痫样放电完全消失）为：苯巴比妥22%、卡马西平33%、丙戊酸钠46%，且该数值在局灶性/广泛性癫痫样放电、脑电图监测间隔时间、新生儿/非新生儿脑电图中均保持稳定。一项双盲单交叉试验中，12例4~21岁难治性全面性癫痫患者，添加拉莫三嗪或安慰剂治疗，结果显示拉莫三嗪可显著减少癫痫样放电的持续时间和密度。一项双盲安慰剂对照随机交叉试验中，小剂量肌内注射氯硝西泮可显著减少儿童的癫痫样放电。

对于睡眠期癫痫性电持续状态患者，大剂量苯二氮䓬类药物可减少部分患者的癫痫样放电；一项纳入18例5~10岁睡眠期癫痫性电持续状态患儿的安慰剂对照双盲交叉试验中，左乙拉西坦可中度减少夜间癫痫样放电。44例接受长期氢化可的松治疗的睡眠期癫痫性电持续状态患者中，21例（48%）脑电图恢复正常，但其中14例复发。

综上，丙戊酸钠、拉莫三嗪、左乙拉西坦、苯二氮䓬类药物和糖皮质激素等药物，可能对癫痫样放电有抑制作用，但目前无法明确该作用是否独立于其抗癫痫发作效应（即癫痫发作对癫痫样放电的影响）。此外，部分抗癫痫药物可能加重癫痫样放电或某些负面行为（如多动）。

五、结论

癫痫样放电可通过脑区特异性方式，短期内损害人类认知功能；但其对远期认知结局的影响，证据相互矛盾。目前尚无足够证据支持单独针对癫痫样放电进行治疗，但当怀疑癫痫样放电导致认知功能障碍和行为问题时，可考虑试验性治疗。

致谢

伊万·桑切斯·费尔南德斯获阿方索·马丁·埃斯库德罗基金会和人疱疹病毒6型基金会的癫痫性脑病研究基金资助。
托比亚斯·洛登肯珀任职于长期（癫痫和重症监护室）监测实验室认证委员会、美国临床神经生理学会、美国临床神经生理委员会，担任《癫痫发作》杂志副主编，在波士顿儿童医院开展视频脑电图长程监测、常规脑电图及其他电生理检查并收取相关费用，获美国患者中心结局研究协会、支付方-提供者质量倡议项目支持，以及美国癫痫学会、美国癫痫基金会、癫痫治疗项目、儿科癫痫研究基金会、Cure基金会、丹尼迪德基金会的研究资助，同时获卫材公司、灵北公司、厄普舍史密斯公司的研究者发起研究资助。
阿里斯泰亚·S·加兰诺普洛获美国国防部、Cure基金会、美国国家神经疾病和中风研究所、优时比公司、赫弗家族基金会、巴里西格尔家族基金会、阿贝戈德斯坦/约书亚卢里基金会、劳里马什/丹莱维茨家族基金会的研究资助；获摩根&克莱普尔出版社、爱思唯尔出版社、约翰利比欧洲出版社的著作版税，以及美国国防部的评审酬金；无与本研究相关的利益冲突。
所罗门·L·莫舍担任神经外科和神经科查尔斯弗罗斯特讲席教授，获美国国家卫生研究院、Cure基金会、美国国防部、优时比公司、赫弗家族基金会、巴里西格尔家族基金会、阿贝戈德斯坦/约书亚卢里基金会、劳里马什/丹莱维茨家族基金会的研究基金资助；获爱思唯尔出版社的《神经疾病生物学》杂志副主编年度薪酬，以及两部合编著作的版税；获灵北公司、优时比公司的咨询费；拥有缺血性卒中多打击模型的专利（美国专利号：US7863499）；无与本研究相关的利益冲突。

参考文献

1. 卡斯泰莱因-诺斯特·特伦尼特DG, 费尔梅伦R. 亚临床癫痫样放电对复杂任务和认知功能的影响：对机组人员和空中交通管制员的意义[J]. 癫痫与行为, 2005,6:31-34.

2. 大久保洋, 松浦正, 浅井哲等. 健康儿童的癫痫样脑电图放电：患病率、情绪和行为相关性及遗传影响[J]. 癫痫杂志, 1994,35:832-841.

3. 阿尔茨JH, 宾尼CD, 斯密特AM等. 局灶性和广泛性癫痫样脑电图活动引发的选择性认知损害[J]. 大脑, 1984,107(1):293-308.

4. 卡斯泰莱因-诺斯特·特伦尼特DG, 里默斯马JB, 宾尼CD等. 亚临床癫痫样脑电图放电对驾驶行为的影响[J]. 脑电图与临床神经生理学, 1987,67:167-170.

5. 克林JK, 斯科特RC, 霍姆斯GL等. 海马发作间期癫痫样活动对人类认知的干扰[J]. 神经学, 2013,81:18-24.

6. 奥尔登坎普AP, 阿伦兹J. 癫痫样脑电图放电对认知功能的影响：“短暂性认知损害”的概念是否仍成立？[J]. 癫痫与行为, 2004,5(增刊1):S25-S34.

7. 霍姆斯GL, 伦克-桑蒂尼PP. 发作间期癫痫样异常在认知损害中的作用[J]. 癫痫与行为, 2006,8:504-515.

8. 利希特EA, 雅各布森RH, 藤川DG. 亚临床癫痫样放电导致的慢性额叶功能损害[J]. 癫痫与行为, 2002,3:96-100.

9. 查普曼KE等. 控制癫痫活动可改善预后[J]. 癫痫杂志, 2015.

10. 加兰诺普洛AS, 莫舍SL. 婴儿痉挛症和早期癫痫性脑病的发病机制及新的候选治疗方案：来自临床前研究的观点[J]. 神经疾病生物学, 2015,79:135-149.

11. 加兰诺普洛AS, 博伊科A, 拉多F等. 睡眠期癫痫性电持续状态相关的神经精神异常谱[J]. 脑发育, 2000,22:279-295.

12. 帕特里G, 利亚古比S, 塔西纳里CA. 儿童睡眠诱发的亚临床“癫痫性电持续状态”：6例临床和脑电图研究[J]. 神经学文献, 1971,24:242-252.

13. 桑切斯·费尔南德斯I, 查普曼KE, 彼得斯JM等. 巴别塔：北美地区睡眠期癫痫性电持续状态和睡眠期持续性棘慢波的概念与术语调查[J]. 癫痫杂志, 2013,54:741-750.

14. 埃比A, 波兹南斯基N, 费尔赫伦D等. 左乙拉西坦治疗睡眠期持续性棘慢波癫痫综合征的疗效：12例经验[J]. 癫痫杂志, 2005,46:1937-1942.

15. 桑切斯·费尔南德斯I, 彼得斯JM, 哈吉洛祖S等. 睡眠期持续性棘慢波的临床分期和脑电图演变[J]. 癫痫杂志, 2012,53:1185-1195.

16. 查瓦库拉V, 桑切斯·费尔南德斯I, 彼得斯JM等. 尖波的自动化定量分析[J]. 癫痫与行为, 2013,26:143-152.

17. 休蒙DA, 欧文RJ. 局灶性尖波诱发的脑功能障碍与后续慢波相关[J]. 神经病学年鉴, 1988,23:131-137.

18. 迪克尔曼S, 博恩J. 睡眠的记忆功能[J]. 自然综述神经科学, 2010,11:114-126.

19. 卡普德维拉O, 达亚特E, 海尔兰迪什-戈扎尔L等. 健康儿童睡眠期癫痫样活动的患病率[J]. 睡眠医学, 2008,9:303-309.

20. 苏EL. 无癫痫发作史者的发作间期癫痫样放电：其意义何在？[J]. 临床神经生理学杂志, 2010,27:229-238.

21. 山姆MC, 苏EL. 社区中无癫痫患者的癫痫样放电意义[J]. 癫痫杂志, 2001,42:1273-1278.

22. 霍尔特曼M, 马泰A, 赫尔曼U等. 中央颞区尖波增加注意缺陷多动障碍患儿的冲动性：一项神经心理学初步研究[J]. 脑发育, 2006,28:633-640.

23. 西尔维斯特里R, 加利亚诺A, 卡拉雷塞T等. 视频多导睡眠监测记录的注意缺陷多动障碍患儿发作期和发作间期脑电图异常[J]. 癫痫研究, 2007,75:130-137.

24. 施赖伯JM, 兰纳姆DC, 特雷舍尔WH等. 脑电图放电的变化可预测史密斯-莱姆利-奥皮茨综合征患者的注意缺陷多动障碍严重程度[J]. 神经学, 2014,83:151-159.

25. 皮卡德A, 谢利奥特-埃罗F, 布斯克拉维M等. 发育性语言障碍的睡眠脑电图[J]. 发育医学与儿童神经学, 1998,40:595-599.

26. 迪韦勒鲁瓦-奥梅C, 吉莱P, 比亚尔C等. 发育性语言障碍儿童的单侧半球功能研究[J]. 神经心理学, 1995,33:823-834.

27. 塔奇曼R. 癫痫中的孤独症与社会认知：对综合癫痫护理的启示[J]. 当代神经学观点, 2013,26:214-218.

28. 马利根CK, 特劳纳DA. 孤独症谱系障碍中癫痫样异常的发生率及行为相关性[J]. 孤独症与发育障碍杂志, 2014,44:452-458.

29. 巴拉班-吉尔K, 塔奇曼R. 癫痫和癫痫样脑电图：与孤独症及语言障碍的关联[J]. 精神发育迟滞与发育障碍研究综述, 2000,6:300-308.

30. 麦克维卡KA, 巴拉班-吉尔K, 拉平I等. 语言倒退儿童的癫痫样脑电图异常[J]. 神经学, 2005,65:129-131.

31. 加兰诺普洛AS, 莫舍SL. 癫痫假说：基于动物模型的发育相关论证[J]. 癫痫杂志, 2009,50(增刊7):37-42.

32. 埃布斯SC, 艾吉DM, 登博尔JT等. 良性局灶性尖波频率的变化与中枢信息处理速度的变化相关：一项为期2年的前瞻性随访研究[J]. 癫痫与行为, 2015,43C:8-15.

33. 宾尼CD, 卡斯泰莱因-诺斯特·特伦尼特DG, 斯密特AM等. 癫痫样脑电图放电与认知的相互作用[J]. 癫痫研究, 1987,1:239-245.

34. 卡斯泰莱因-诺斯特·特伦尼特DG, 巴克DJ, 宾尼CD等. 亚临床癫痫样脑电图放电的心理效应（一）：学业能力[J]. 癫痫研究, 1988,2:111-116.

35. 奥尔登坎普AP, 拜特勒J, 阿伦兹J等. 亚临床癫痫样脑电图放电对认知激活的急性影响[J]. 功能性神经学, 2005,20:23-28.

36. 尼科莱J, 埃布斯S, 比曼斯DP等. 发作间期癫痫样脑电图放电和非惊厥性小发作对认知的影响[J]. 癫痫杂志, 2012,53:1051-1059.

37. 宾尼CD. 癫痫样放电期间的认知损害：治疗脑电图是否合理？[J]. 柳叶刀神经病学, 2003,2:725-730.

38. 帕纳约托普洛斯CP, 迈克尔M, 桑德斯S等. 儿童良性局灶性癫痫：已确立和新认识的综合征评估[J]. 大脑, 2008,131:2264-2286.

39. 菲利皮尼M, 博尼A, 詹诺塔M等. 儿童中央颞区棘波癫痫谱系的神经心理发育：癫痫样活动的长期影响[J]. 癫痫与行为, 2013,28:504-511.

40. 马萨R, 圣马丁A, 卡坎久R等. 特发性中央颞区癫痫复杂演变的脑电图预测指标[J]. 神经学, 2001,57:1071-1079.

41. 金村浩, 佐野文, 青柳健等. 连续脑电图变化能否预测中央颞区癫痫的非典型临床特征？[J]. 发育医学与儿童神经学, 2012,54:912-917.

42. 沃尔夫M, 魏斯科普夫N, 塞拉E等. 儿童良性部分性癫痫：选择性认知缺陷与脑电图/脑磁图定位的局灶性尖波相关[J]. 癫痫杂志, 2005,46:1661-1667.

43. 伯曼R, 根岸M, 维斯特尔M等. 典型儿童失神癫痫的同步脑电图、功能磁共振成像和行为学研究[J]. 癫痫杂志, 2010,51:2011-2022.

44. 科南特LL, 威尔丰A, 因格莱斯C等. 儿童失神癫痫的执行功能及相关功能障碍[J]. 癫痫与行为, 2010,18:414-423.

45. 威尔雷尔EC, 坎菲尔德CS, 坎菲尔德PR等. 典型儿童失神癫痫的远期社会心理结局：有时是披着羊皮的狼[J]. 儿科学与青少年医学文献, 1997,151:152-158.

46. 皮亚齐尼A, 特纳K, 维尼奥利A等. 青少年肌阵挛癫痫的额叶认知功能障碍[J]. 癫痫杂志, 2008,49:657-662.

47. 万德施奈德B, 科普UA, 克利格尔M等. 青少年肌阵挛癫痫患者及其健康同胞的前瞻性记忆[J]. 神经学, 2010,75:2161-2167.

48. 桑切斯·费尔南德斯I, 洛登肯珀T, 彼得斯JM等. 睡眠期癫痫性电持续状态：临床表现和病理生理学[J]. 儿科神经学, 2012,47:390-410.

49. 鲁塞勒C, 勒沃尔M. 睡眠期持续性棘慢波与认知功能的关系//博阿莫瓦诺A, 比罗M, 德翁纳L等. 睡眠期持续性棘慢波[M]. 伦敦：约翰利比出版社, 1995:123-133.

50. 克拉默U, 萨吉L, 戈德堡-斯特恩H等. 儿童睡眠期癫痫性电持续状态的临床谱和药物治疗[J]. 癫痫杂志, 2009,50:1517-1524.

51. 卡拉巴约RH, 邦焦尔尼L, 塞尔索西莫R等. 伴分流脑积水的儿童睡眠期持续性棘慢波癫痫性脑病：9例研究[J]. 癫痫杂志, 2008,49:1520-1527.

52. 洛登肯珀T, 科斯莫G, 科塔加尔P等. 儿童睡眠期癫痫性电持续状态的癫痫手术治疗[J]. 神经外科杂志, 2009,64:328-337.

53. 佩尔托拉ME, 柳科宁E, 格兰斯特伦ML等. 手术对睡眠期癫痫性电持续状态脑病的疗效[J]. 癫痫杂志, 2011,52:602-609.

54. 斯科尔特斯FB, 亨德里克斯MP, 雷尼尔WO. 认知损害与睡眠期持续性棘慢波[J]. 癫痫与行为, 2005,6:167-173.

55. 莱斯卡G, 鲁道夫G, 布鲁诺N等. GRIN2A基因突变与获得性癫痫性失语及相关儿童局灶性癫痫和伴言语语言功能障碍的脑病[J]. 自然遗传学, 2013,45:1061-1066.

56. 莱斯卡G, 鲁道夫G, 拉巴尔梅A等. 朗道-克莱夫纳型和睡眠期持续性棘慢波型癫痫性脑病：基因组分析揭示与孤独症的关联[J]. 癫痫杂志, 2012,53:1526-1538.

57. 梅福德HC, 巴特肖ML, 霍夫曼EP. 基因组学、智力障碍与孤独症[J]. 新英格兰医学杂志, 2012,366:733-743.

58. 诺瓦里诺G, 埃尔-菲沙维P, 开塞利H等. BCKD激酶突变导致一种潜在可治的伴癫痫孤独症[J]. 科学, 2012,338:394-397.

59. 萨欣M. 结节性硬化症和孤独症的靶向治疗试验：不再是梦想[J]. 当代神经生物学观点, 2012,22:895-901.

60. 塔奇曼R, 赫茨D, 马穆纳斯LA. 美国国家神经疾病和中风研究所癫痫与孤独症谱系障碍研讨会报告[J]. 神经学, 2013,81:1630-1636.

61. 莱姆克JR, 拉尔D, 赖因塔勒EM等. GRIN2A基因突变导致伴中央颞区尖波的特发性局灶性癫痫[J]. 自然遗传学, 2013,45:1067-1072.

62. 古泽塔F, 巴塔利亚D, 韦雷迪切C等. 早期丘脑损伤与癫痫及睡眠期持续性棘慢波相关[J]. 癫痫杂志, 2005,46:889-900.

63. 克斯伯根KJ, 德弗里斯LS, 莱滕FS等. 新生儿丘脑出血与睡眠期癫痫性电持续状态高度相关[J]. 癫痫杂志, 2013,54:733-740.

64. 桑切斯·费尔南德斯I, 武冈M, 塔斯E等. 睡眠易化性癫痫样活动患者的早期丘脑病变[J]. 神经学, 2012,78:1721-1727.

65. 戈登K, 鲍登H, 坎菲尔德P等. 丙戊酸治疗无临床发作但脑电图呈重度癫痫样改变的学习障碍[J]. 儿童神经学杂志, 1996,11:41-43.

66. 罗嫩GM, 理查兹JE, 坎宁安C等. 丙戊酸钠能否改善有癫痫样发放但无临床发作儿童的学习能力？[J]. 发育医学与儿童神经学, 2000,42:751-755.

67. 明茨M, 莱戈夫D, 斯科尔奈恩奇J等. 未被充分认识的癫痫谱系：左乙拉西坦对亚临床尖波产生相关神经心理功能的影响[J]. 儿童神经学杂志, 2009,24:807-815.

68. 奥尔廷斯基P, 米多尔KJ. 抗癫痫药物的认知副作用[J]. 癫痫与行为, 2004,5(增刊1):S60-S65.

69. 德内格里M, 巴列托MG, 巴塔利亚FM等. 地西泮直肠推注试验后短疗程地西泮治疗癫痫性电持续状态[J]. 脑发育, 1995,17:330-333.

70. 桑切斯·费尔南德斯I, 哈吉洛祖S, 埃克西奥卢Y等. 大剂量地西泮对睡眠期癫痫性电持续状态中睡眠易化性尖波的短期疗效[J]. 儿科神经学, 2012,46:312-318.

71. 桑切斯·费尔南德斯I, 彼得斯JM, 安S等. 大剂量地西泮治疗睡眠期持续性棘慢波的远期疗效[J]. 儿科神经学, 2013,49:163-170.

72. 柳科宁E, 坎托拉-索尔萨E, 佩陶R等. 32例睡眠期癫痫性电持续状态脑病患儿的远期结局[J]. 癫痫杂志, 2010,51:2023-2032.

73. 普雷斯勒RM, 鲁宾逊RO, 威尔逊GA等. 治疗发作间期癫痫样放电可改善伴行为问题癫痫患儿的行为表现[J]. 儿科学杂志, 2005,146:112-117.

74. 利本森MH, 哈尔达尔A, 平托AL. 发作间期癫痫样活动儿童的尖波计数稳定性[J]. 癫痫发作, 2014.

75. 利本森MH, 卡拉瓦莱B. 不同抗癫痫药物对儿童发作间期癫痫样活动的抑制作用是否存在差异？[J]. 儿科神经学, 2001,24:214-218.

76. 埃里克松AS, 克努特松E, 内加德A. 拉莫三嗪对难治性癫痫年轻患者癫痫样放电的影响[J]. 癫痫杂志, 2001,42:230-236.

77. 达林M, 克努特松E, 阿马克P等. 小剂量氯硝西泮减少癫痫患儿癫痫样活动的随机双盲研究[J]. 癫痫杂志, 2000,41:308-315.

78. 拉尔松PG, 巴克KA, 比约内斯H等. 左乙拉西坦对睡眠期局灶性夜间癫痫样活动的影响：一项安慰剂对照双盲交叉研究[J]. 癫痫与行为, 2012,24:44-48.

79. 布扎图M, 布尔特奥C, 阿尔图扎拉C等. 糖皮质激素治疗睡眠期持续性棘慢波癫痫综合征[J]. 癫痫杂志, 2009,50(增刊7):68-72.

80. 卡瓦祖蒂GB, 卡佩拉L, 纳林A. 正常儿童癫痫样脑电图模式的纵向研究[J]. 癫痫杂志, 1980,21:43-55.

81. 博鲁西亚克P, 齐尔鲍尔M, 延克AC. 健康儿童癫痫样放电的患病率：数字脑电图前瞻性研究新数据[J]. 癫痫杂志, 2010,51:1185-1188.

82. 霍尔特曼M, 贝克尔K, 肯特纳-菲古拉B等. 注意缺陷多动障碍患儿的中央颞区尖波发生率升高[J]. 癫痫杂志, 2003,44:1241-1244.

83. 西贝尔林克BM, 巴克DJ, 宾尼CD等. 亚临床癫痫样脑电图放电对儿童的心理效应（二）：一般智力测试[J]. 癫痫研究, 1988,2:117-121.

84. 丰塞卡LC, 特德鲁斯GM, 帕切科EM. 儿童良性中央颞区棘波癫痫的癫痫样脑电图放电：反应性与短暂性认知损害[J]. 癫痫与行为, 2007,11:65-70.


Should Interictal Epileptiform Discharges Be Treated? A Critical Review

US Department of Health and Human Services Public Access
Author Manuscript
Epilepsia. Available online 14 December 2018; Published online 2015 October, 56(10):1492-1504, doi:10.1111/epi.13108

Authors

Ivan Sánchez Fernández, MD1,2, Tobias Loddenkemper, MD2, Aristea S. Galanopoulou, MD, PhD3, Solomon L. Moshe, MD3,4
1 Division of Epilepsy and Clinical Neurophysiology, Department of Neurology, Boston Children’s Hospital, Harvard Medical School, Boston, Massachusetts, USA
2 Department of Pediatric Neurology, Sant Joan de Déu Hospital, University of Barcelona, Barcelona, Spain
3 Saul R. Korey Department of Neurology, Dominick P. Purpura Department of Neuroscience, Developmental Medicine Laboratory, Montefiore Medical Center/Einstein Epilepsy Management Center, Albert Einstein College of Medicine, Bronx, New York, USA
4 Montefiore Medical Center, Albert Einstein College of Medicine, Bronx, New York, USA

Corresponding Author

Tobias Loddenkemper, MD, Associate Professor of Neurology, Harvard Medical School, Division of Epilepsy and Clinical Neurophysiology, Boston Children’s Hospital, 300 Longwood Avenue, Fenwood 9, Boston, Massachusetts 02115, USA; Tel: 617-355-2443; Fax: 617-730-0463; E-mail: tobias.loddenkemper@childrens.harvard.edu

Conflict of Interest Disclosure

All authors have no conflicts of interest to disclose.
We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

Abstract

Objective: To assess the impact of interictal epileptiform discharges (EDs) such as spikes and spike-wave complexes on cognitive function and to explore the indications for considering treatment of EDs.
Methods: Literature review.
Results: EDs can transiently disrupt cognitive function in humans, and frequent chronic EDs may impair long-term cognitive outcomes. However, due to multiple confounding factors such as underlying etiology, seizures, and medication effects, there is no consensus on the impact of EDs on long-term cognitive outcomes. Limitations of existing studies include the lack of standardized quantitative methods for EDs and the absence of a widely accepted automated spike quantification approach. Currently, there is no conclusive evidence to support or refute the treatment of EDs. Therefore, a practical, non-evidence-based management algorithm is proposed: EDs in asymptomatic individuals do not require treatment, as the risks of treatment may outweigh the uncertain benefits; trial treatment of EDs may be considered in patients with cognitive impairment/regression or neurological symptoms that cannot be explained by the underlying etiology, comorbidities, or seizure severity; for patients with epilepsy or EDs who have cognitive/neurological dysfunction, targeted treatment is required to control the underlying epilepsy syndrome.
Significance: EDs can cause short-term cognitive or neurological dysfunction in humans, and the evidence regarding their impact on long-term cognitive outcomes is conflicting. Currently, there is no evidence to support or refute the treatment of asymptomatic EDs.

Key words: Antiepileptic drugs; Cognition; Interictal epileptiform discharges; Memory; Sleep

1. Introduction

Interictal epileptiform discharges (EDs) can be detected in the electroencephalogram (EEG) of approximately 1% to 5% of the general population, with a significantly higher detection rate in patients with epilepsy compared to healthy controls. EDs include isolated or short bursts of spikes, polyspikes, sharp waves, or spike-wave complexes without obvious clinical symptoms. EDs can acutely disrupt cognitive or neurological function in humans, and the duration and distribution of EDs are associated with the type of functional impairment. However, the exact impact of chronic EDs on neuronal circuits and long-term cognitive outcomes remains unclear, and the benefits and risks of treatment aimed at suppressing EDs are also not well defined.

This review aims to explore the impact of EDs on cognitive function (focusing on EDs occurring in non-ictal patterns/states) and to provide recommendations for the timing of treatment of EDs. The review is limited to non-acute outpatient settings, excludes states of impaired consciousness, and focuses on clinical data (animal experimental studies have been reviewed recently).

2. Quantitative Analysis of EDs and Cognitive Function

2.1 Attempts at Objective Quantification of EDs

Most quantitative methods for EDs were specifically designed for electrical status epilepticus during sleep (ESES), an EEG pattern characterized by almost continuous EDs during sleep.

1. Spike-Wave Index (SWI) and Spike-Wave Percentage (SWP): Both are used to assess the proportion of EDs in the total monitoring time. In initial studies on ESES, the SWI was defined as the proportion of EDs during slow-wave sleep, but its reproducibility is limited due to missing methodological details and inconsistent standards for calculation and interpretation. Some studies have proposed defining the SWI as the percentage of 1-second time windows with at least one ED relative to the total duration, which was later adopted and renamed the SWP. The advantage of the SWP is its high reproducibility and low resource consumption, but it cannot distinguish the severity of EEG when the frequency of EDs exceeds 1 per second.

2. Spike Frequency (SF): Refers to the total number of EDs per unit time, usually calculated per 100 seconds. Without automation, this method is more time-consuming than the SWP and is prone to sampling errors, but it has better discrimination for the severity of EDs with high spike frequencies.

3. Computerized Quantitative Methods: Automated methods can achieve rapid and reproducible quantification of EDs, but they can only identify EDs (or their components) that meet preset parameters. Parameters need to be adjusted for different patients, artifacts need to be screened and excluded, and not all morphological types of EDs can be identified. Therefore, they have not been widely used in clinical practice.

2.2 Challenges in Quantitative Analysis of ED Impact

Numerous detailed issues in the assessment of EDs make it difficult to directly quantify their impact on clinical outcomes.

1. Distribution and Lateralization: Clinical dysfunction is related to the localization of EDs. EDs located in functional areas, multifocal, or generalized are more harmful than those in non-cognitive related brain regions.

2. Amplitude Differences: It is currently unclear whether high-amplitude and low-amplitude EDs have different potential harms to cognitive function.

3. Morphological Characteristics: EDs include different combinations of spikes, polyspikes, and sharp waves, which may be accompanied by or without subsequent slow waves. Some studies suggest that subsequent slow waves can also interfere with cortical function.

4. Sleep-Wake Cycle: The severity of EDs is not always independent of the sleep-wake stage, and the impact of EDs in different stages on the severity of epileptic encephalopathy remains unclear. Since most cognitive tests rely on the patient's active response (which needs to be completed in the awake state), the assessment of cognitive function in the awake state is easier to implement; cognitive deficits related to sleep, without active testing during sleep, are more likely to be considered related to information storage and neural plasticity.

2.3 Correlation Between EDs and Cognitive/Functional Outcomes

Assessments of the short-term (acute) effects of EDs are limited to cognitive functions that can be evaluated by rapid reaction time, and the cognitive assessment itself may alter the manifestation of EDs, further increasing the complexity of result interpretation. Assessments of the long-term effects of EDs on cognitive function are limited by the situational and time-dependent nature of their cognitive effects. Moreover, since EEG and specific cognitive tests are usually performed after the onset of symptoms, it is difficult to accurately assess the baseline cognitive function and EEG status before the onset of symptoms, and the temporal evolution of symptoms (cognitive, seizures) and EDs is often unclear.

3. Impact of EDs on Cognitive Function

3.1 Asymptomatic Individuals Without a History of Epilepsy

The detection rate of EDs in asymptomatic children without a history of epilepsy is 0~6%, and 0~7% in adults. The clinical significance of incidentally detected EDs in healthy individuals remains unclear: the probability of subsequent development of epilepsy is low (approximately 6% in children and 2% in adults), and some studies suggest that such individuals may have mild cognitive and behavioral abnormalities, but this conclusion still needs to be further verified by prospective, well-controlled large-sample studies.

3.2 Patients with EDs and Non-Epileptic Neurological Disorders

The detection rate of EDs in inpatients and outpatients without a history of seizures (2%~14%) is higher than that in community-based population studies, and their cognitive significance is unclear, which may be a biomarker of brain function abnormalities.

1. Attention-Deficit/Hyperactivity Disorder (ADHD): In a study involving 48 children with ADHD with a mean age of 9.4 years (6.7~14.9 years), 16 children with centrotemporal spikes, 16 children without centrotemporal spikes, and 16 healthy control children were compared. Centrotemporal spikes were more common in children with ADHD with increased impulsivity, decreased response inhibition, and impaired interference control. In addition, polysomnography showed a higher detection rate of EDs in children with ADHD: among 42 children with ADHD who underwent monitoring, 53.1% had EDs, of whom 3 had seizures during monitoring, and the rest had other neurological comorbidities such as language disorders and motor apraxia. Another series of studies involving 23 children with Smith-Lemli-Opitz syndrome aged 4~17 years found that 51% of the children had abnormal EEG (mostly EDs), and the presence of EDs in a single EEG of the same child could predict an average increase of 27% in the severity of ADHD symptoms. Currently, there is no evidence to confirm that EDs are the direct cause of ADHD symptoms after excluding other potential etiologies.

2. Language Disorders: Polysomnography or standard EEG studies have confirmed that the incidence of EDs is higher in children with language disorders. Among 52 children with language developmental disorders aged 4~11 years, 50% had EDs detected (84% located in the left side), which was significantly higher than 10% in age-matched controls. A few studies have shown that the type of language developmental disorder is related to the incidence and severity of EDs: in polysomnography, the detection rate of EDs in children with expressive developmental language disorder (37.5%) is higher than that in controls (5.1%), including generalized EDs during non-rapid eye movement (NREM) sleep. Among 9 children, 2 also had EDs detected during wakefulness or rapid eye movement (REM) sleep; 56% of children with expressive developmental language disorder had frequent EDs (spike-wave index during sleep 2.5%~66%). There is currently no evidence that such EDs are the direct cause of language disorders; more likely, they are a manifestation of the underlying etiology. It is worth noting that in the above study, 2 control children with EDs had a spike-wave index of 21.6% during sleep but no language dysfunction.

3. Autism Spectrum Disorder (ASD) and Intellectual Disability (ID): The incidence of epilepsy and EDs in individuals with ASD and ID is significantly higher than that in healthy controls, but it remains unclear whether EDs are the cause of aggravated cognitive impairment or just a manifestation of brain function abnormalities. Overall, 20%~60% of children with ASD have epileptiform abnormalities, and after excluding patients with epilepsy, 8%~20% of children with ASD still have EDs detected. Compared with other ASD subtypes, patients with Asperger syndrome have a lower incidence of EDs and clinical seizures; a history of aggressive behavior is associated with EDs but not with clinical seizures.

Even without a history of epilepsy, children with autistic regression are more likely to have EDs, but the temporal correlation between the two is difficult to determine. Compared with isolated language regression (Landau-Kleffner syndrome, LKS), autistic regression has a lower correlation with seizures (8% vs 33%) or EDs (28% vs 56%), a younger age of onset, and is more often complicated with developmental delay. Therefore, compared with isolated language regression, EDs may play a smaller role in the pathogenesis of autistic regression or have weaker value as a surrogate marker. Existing experimental evidence also supports another possibility: the developmental impact of the underlying etiology of such age-specific syndromes may be further regulated by age, gender, or brain region-specific factors.

3.3 Patients with Epilepsy

Multiple studies have explored the short-term and long-term effects of focal or generalized EDs in specific syndromes: EDs can affect central information processing, stimulus reaction time, and cognitive responses. Other studies have compared the effects of EDs and seizures on learning and cognition, and found that in the absence of seizures, EDs have no or only a slight independent effect on attention and information processing speed, and this effect is more obvious when EDs are frequent.

3.3.1 Short-Term Effects of EDs

(1) Impact on Neuropsychological Tests: The short-term cognitive impairment caused by EDs is not necessarily caused by a general attention disorder, but more likely reflects the impairment of specific functions in the affected brain regions, and cognitive deficits are related to the timing of EDs. EDs need to occur at the "right brain region" and "right time" to cause specific cognitive dysfunction.
(2) Impact on Daily Activities: An assessment of academic task performance in 20 children showed that the higher the frequency of EDs, the worse the test performance, especially in arithmetic. Monitoring of driving performance in 6 subjects with EDs who drove 420 km at a constant speed of 90 km/h on the highway showed that during the occurrence of EDs, 3 subjects had impaired lane-keeping ability, and 1 had a tendency of impairment; this impairment can still occur in patients who have been seizure-free for many years, and the results of standard laboratory cognitive tests (such as the Corsi Block Tapping Test and short-term verbal memory test) cannot predict the driving test performance of the subjects. This suggests that EDs can impair the performance of tasks with high attention requirements (even in seizure-free patients), but this impairment cannot be predicted by standard neuropsychological tests.
(3) Relationship Between EDs, Transient Cognitive Impairment (TCI), and Nonconvulsive Seizures: If EDs cause measurable cognitive changes, they are not subclinical manifestations but clinical transient cognitive impairment or ictal manifestations (minor seizures). Multiple visual and auditory response tests have found a temporal correlation between the EEG time period of EDs and reaction delay. Some studies have proposed that the impact of EDs starts before the appearance of spikes and ends at the end of subsequent slow waves. Another study involving 188 children aged 6~18 years conducted EEG monitoring during 2-hour cognitive tests, and divided the study population into: epileptic children with nonconvulsive seizures during the test (manifested as staring or slight movements, lasting several seconds), epileptic children without nonconvulsive seizures during the test, and healthy controls without epilepsy. The results showed that nonconvulsive seizures can significantly impair cognitive function (most seizures last less than 19 seconds, mainly focal seizures, followed by absence seizures and myoclonic seizures); EDs without seizures can also impair cognitive function, but the impact is milder.

Are EDs a mild manifestation in the spectrum of nonconvulsive seizures? Or are they early markers of underlying etiologies/states that are prone to induce cognitive dysfunction and/or seizures? Clarifying these issues will help optimize the management plan for patients with EDs.
(4) Challenges in Result Interpretation: The study population of short-term cognitive tests has inherent biases, because studies usually prefer subjects with frequent EDs in the awake state. Therefore, the study results may not be applicable to populations with low frequency of EDs or EDs only occurring during sleep. In addition, external stimuli and the level of task attention may also change the frequency and characteristics of EDs.

3.3.2 Long-Term Effects of EDs

(1) Pediatric Focal Epilepsy Syndromes: Benign childhood epilepsy with centrotemporal spikes (BCECTS), Panayiotopoulos syndrome, and late-onset childhood occipital epilepsy (Gastaut type) share common characteristics: relatively mild seizures, low seizure frequency, frequent EDs that are facilitated during sleep, and EDs can persist for several years even after seizure remission. Studies on BCECTS have shown that the more frequent EDs during sleep, the more obvious neuropsychological developmental abnormalities (especially in the language domain); other studies have found that 29% of children have impaired academic and family functions, and the predictors of poor prognosis are related to EDs but not to seizures. However, some studies have not found a correlation between spike frequency and cognitive performance. A study involving 26 children with focal EDs and learning difficulties (19 with BCECTS, 2 with Panayiotopoulos syndrome, 1 with focal seizures, and 4 without a history of seizures) showed a negative correlation between baseline central information processing speed and the frequency of EDs (excluding children without a history of seizures); follow-up found that the decline in central information processing speed was related to the increase in the frequency of EDs and the persistence of seizures, but it is currently impossible to clarify the independent impact of seizures and EDs on central information processing.

Among children with focal epilepsy syndromes, those with left centrotemporal EDs perform worse in complex language tasks, and those with occipital spikes perform worse in synchronous information processing, especially in visual switching tasks. In summary, small-sample series studies based on pediatric focal epilepsy with EDs suggest that EDs may lead to poor long-term cognitive outcomes, and this damage is related to the localization of EDs.
(2) Generalized Epilepsy: Spike-wave discharges in patients with childhood absence epilepsy (CAE) can transiently interfere with attention, consciousness, and information processing, and there is evidence of long-term cognitive deficits. A study comparing 16 children with CAE, 14 children with type 1 diabetes mellitus, and 15 healthy children showed that children with CAE had no significant differences in intelligence, memory, academic achievement, fine motor speed, or information processing speed compared with the other two groups, but performed worse in problem-solving, verbal fluency, complex motor control, attention/behavioral inhibition, and psychosocial function; their long-term psychosocial outcomes are worse than those of patients with non-epileptic chronic diseases such as juvenile rheumatoid arthritis. Juvenile myoclonic epilepsy (JME) may also have long-term mild frontal lobe processing dysfunction. It is speculated that EDs may cause functional deficits mediated by this circuit through progressive interference with the thalamic-frontal neuronal circuit, but functional deficits and EDs may also be biomarkers of abnormal thalamic-frontal neuronal circuits determined by genetics.
(3) Epileptic Encephalopathies with Language/Cognitive Regression: Electrical status epilepticus during sleep (ESES) is characterized by almost continuous spike discharges during NREM sleep, often complicated with varying degrees of cognitive regression. Its duration has historically been considered a marker of poor cognitive prognosis. A literature review involving 209 patients showed that ESES lasting more than 2 years is associated with poor cognitive prognosis; a study of 30 ESES patients who responded to treatment showed that the duration of ESES is related to the degree of cognitive function recovery to baseline during follow-up: among children with a duration of less than 13 months, 50% recovered to baseline cognitive function, while none of the children with a duration of more than 18 months recovered. Objective assessment of cognitive function before and after epilepsy surgery showed that the remission of ESES can improve cognitive function, but due to the small sample size of the study and confounding factors such as underlying etiology, seizure burden, and antiepileptic drug treatment, the cognitive improvement cannot be completely attributed to the reduction of EDs. Another study involving 7 patients did not find a correlation between the duration of ESES and cognitive outcomes.
(4) Common Pathophysiological Mechanisms: Some gene mutations can cause different degrees of ASD, EDs, epilepsy, and ID in the same patient, suggesting a common pathogenesis, which may be mediated by abnormal synaptic plasticity and excitatory/inhibitory imbalance. This provides a new direction for the treatment of such diseases, but also suggests that EDs may only be a manifestation of the underlying pathological process, rather than the cause of cognitive regression.

A variety of potential mechanisms can lead to ESES characterized by significant EDs. GRIN2A gene mutations and early thalamic lesions are both associated with significant long-term EDs and the occurrence of ESES, further confirming the view that different etiologies can lead to similar clinical phenotypes through common pathways.

3.3.3 Methods for Assessing the Cognitive Impact of EDs

Whether long-term exposure to EDs impairs cognitive function and whether treatment of EDs can reverse this potential damage remains unclear. The short-term cognitive impact of EDs is usually assessed by comparing the same subject in two states: the presence and absence of EDs, and the experiments are completed within several hours in the awake state. This assessment method is applicable to patients with ESES or BCECTS: a formal neuropsychological assessment can be performed at baseline, and a post-intervention neuropsychological assessment can be performed after reducing EDs with antiepileptic drugs.

However, the pre- and post-treatment comparative analysis can only correct limited confounding factors such as seizure burden, and cannot correct individual inherent factors such as gender and genetic background. Such studies have certain value for the assessment of short-term outcomes, but the interpretation of long-term outcomes is limited by the difficulty in distinguishing: cognitive improvement related to treatment from cognitive ability improvement brought about by natural development, education system, or changes in living environment. If the preliminary study results are good, large-sample studies can be further carried out, including populations with high homogeneity (such as the same syndrome or the same genetic background), and comparing the developmental trajectories of patients with EDs who received/ did not receive treatment under the premise of considering a variety of known confounding factors; both types of studies need to conduct detailed assessments of the burden, type, and localization of EDs.

4. Should EDs Be Treated?

In the absence of more sufficient evidence-based medical evidence, clinical treatment decisions need to be made. The hesitation in the treatment of EDs stems from the current lack of conclusive evidence to confirm a causal relationship between EDs and long-term cognitive impairment. Based on different clinical scenarios, this review proposes a practical, non-evidence-based management algorithm for EDs.

4.1 Asymptomatic Individuals

In asymptomatic individuals without a history of epilepsy, the cognitive impact of EDs is unclear, and the risk of subsequent development of epilepsy is low (may be related to genetic characteristics). Therefore, incidentally detected EDs in asymptomatic individuals without a history of epilepsy do not require treatment, as the benefits of treatment are unclear.

4.2 Patients Without Seizures but with EDs and Cognitive Impairment/Regression

There are reports that antiepileptic drugs can improve some cognitive functions in such patients, but the results of large-sample studies are inconsistent. In a double-blind, single crossover trial, 8 children aged 6~12 years with EDs and learning and behavioral problems were randomly treated with valproate or placebo. The results showed that children in the valproate treatment group had increased distractibility, prolonged reaction time, decreased memory scores, and no clinical improvement. In a prospective open-label study, 6 children aged 7~15 years with learning difficulties and focal EDs were treated with levetiracetam. A 10-week follow-up found that 4 children had improved scores on the Wide Range Assessment of Memory and Learning, but no changes in scores on the Wechsler Individual Achievement Test. In summary, although some patients have improved cognitive function after treatment with antiepileptic drugs, this may reflect the natural fluctuation of the disease. Currently, there is insufficient evidence to recommend the use of antiepileptic drugs for patients without seizures but with EDs and cognitive impairment/regression, and antiepileptic drugs may aggravate cognitive impairment.

4.3 Patients with Epileptic Encephalopathy, EDs, and Cognitive Impairment/Regression

For patients with epileptic encephalopathy complicated with regression, adrenocorticotropic hormone (ACTH), high-dose glucocorticoids, or immunotherapy have been reported to effectively improve EEG manifestations and treat cognitive regression; after reducing EDs in ESES patients with high-dose benzodiazepines, some patients' cognitive function has been improved. A study of 32 children with ESES showed that after reducing EDs with valproate or ethosuximide, patients' cognitive function improved; epilepsy surgery targeting the epileptogenic focus in ESES patients stopped the progression of cognitive impairment in at least half of the patients and even improved it. However, due to the lack of controls in the collection of study data, the natural fluctuation of the disease itself, and the potential impact of other factors, the results need to be interpreted with caution. In summary, for patients with epileptic encephalopathy complicated with cognitive impairment/regression that may be related to EDs, trial treatment may be considered.

4.4 Patients with Well-Controlled Seizures but with EDs and Cognitive Impairment/Regression

In a double-blind, placebo-controlled, single crossover study, 61 children aged 7~17 years with well-controlled seizures (no/occasional focal or generalized seizures) and behavioral and/or cognitive problems were randomly treated with lamotrigine or placebo. The results showed that patients with reduced EDs after lamotrigine treatment had improved behavioral performance (assessed by the Conners' Rating Scales by parents and teachers) without a significant impact on seizure frequency. Therefore, trial treatment with antiepileptic drugs may be considered for such patients (especially those with progressive cognitive impairment).

4.5 Patients with EDs, Cognitive Impairment/Regression, and Persistent Seizures

For patients with cognitive impairment/regression and persistent seizures, treatment is required to control seizures; currently, there is insufficient evidence to support that the disappearance of EDs is a better endpoint of existing treatments.

4.6 Treatment Options for EDs

There are few studies evaluating the efficacy of antiepileptic drugs on EDs, and the interpretation of results needs to consider the natural fluctuation of EDs. A large-sample pediatric study compared EEG before and after the use of 213 antiepileptic drugs. The suppression rate of EDs (complete disappearance of all types of EDs in EEG after treatment) was: phenobarbital 22%, carbamazepine 33%, valproate 46%, and this value remained stable in focal/generalized EDs, EEG monitoring intervals, and neonatal/non-neonatal EEG. In a double-blind, single crossover trial, 12 patients with refractory generalized epilepsy aged 4~21 years were treated with add-on lamotrigine or placebo. The results showed that lamotrigine can significantly reduce the duration and density of EDs. In a double-blind, placebo-controlled, randomized crossover trial, low-dose intramuscular clonazepam can significantly reduce EDs in children.

For patients with ESES, high-dose benzodiazepines can reduce EDs in some patients; in a placebo-controlled, double-blind crossover trial involving 18 children with ESES aged 5~10 years, levetiracetam can moderately reduce nocturnal EDs. Among 44 patients with ESES who received long-term hydrocortisone treatment, 21 (48%) had normalized EEG, but 14 of them relapsed.

In summary, drugs such as valproate, lamotrigine, levetiracetam, benzodiazepines, and glucocorticoids may have an inhibitory effect on EDs, but it is currently impossible to clarify whether this effect is independent of their antiseizure effect (i.e., the impact of seizures on EDs). In addition, some antiepileptic drugs may aggravate EDs or certain negative behaviors (such as hyperactivity).

5. Conclusion

EDs can acutely impair human cognitive function in a brain region-specific manner; however, the evidence regarding their impact on long-term cognitive outcomes is conflicting. Currently, there is insufficient evidence to support the treatment of EDs alone, but trial treatment may be considered when EDs are suspected to cause cognitive dysfunction and behavioral problems.

Acknowledgments

Ivan Sánchez Fernández was supported by research grants from the Alfonso Martín Escudero Foundation and the HHV-6 Foundation for epileptic encephalopathy research.
Tobias Loddenkemper is a member of the Board of Directors of the Long-Term (Epilepsy and Intensive Care Unit) Monitoring Laboratory Accreditation Committee, the American Clinical Neurophysiology Society, and the American Board of Clinical Neurophysiology. He serves as an Associate Editor of the journal Seizure. He conducts long-term video-EEG monitoring, routine EEG, and other electrophysiological examinations at Boston Children’s Hospital and receives related fees. He is supported by the Patient-Centered Outcomes Research Institute, the Payer-Provider Quality Initiative, and research grants from the American Epilepsy Society, the Epilepsy Foundation of America, the Epilepsy Therapy Project, the Pediatric Epilepsy Research Foundation, the Cure Foundation, and the Danny Did Foundation. He also receives investigator-initiated research grants from Eisai Inc., Lundbeck Inc., and Upsher-Smith Laboratories.
Aristea S. Galanopoulou is supported by research grants from the US Department of Defense, the Cure Foundation, the National Institute of Neurological Disorders and Stroke, UCB Pharma, the Heffer Family Foundation, the Barry Siegel Family Foundation, the Abe Goldstein/Joshua Lurie Foundation, and the Laurie Marsh/Dan Levitz Family Foundation; she receives book royalties from Morgan & Claypool Publishers, Elsevier, and John Libbey Eurotext, and review fees from the US Department of Defense; she has no conflicts of interest related to this study.
Solomon L. Moshe holds the Charles Frost Chair in Neurosurgery and Neurology. He is supported by research grants from the National Institutes of Health, the Cure Foundation, the US Department of Defense, UCB Pharma, the Heffer Family Foundation, the Barry Siegel Family Foundation, the Abe Goldstein/Joshua Lurie Foundation, and the Laurie Marsh/Dan Levitz Family Foundation; he receives annual compensation as an Associate Editor of the journal Biology of the Nervous System published by Elsevier and royalties from two co-edited books; he receives consulting fees from Lundbeck Inc. and UCB Pharma; he holds a patent for a multiple-hit model of ischemic stroke (US Patent No.: US7863499); he has no conflicts of interest related to this study.

References

1. Castelijn-Nostren TRENITT DG, Vermeulen R. The impact of subclinical epileptiform discharges on complex tasks and cognitive function: implications for aircrew and air traffic controllers[J]. Epilepsy & Behavior, 2005,6:31-34.

2. Okubo H, Matsuura M, Asai T, et al. Epileptiform EEG discharges in healthy children: prevalence, emotional and behavioral correlates, and genetic influences[J]. Epilepsia, 1994,35:832-841.

3. Alts JH, Binney CD, Smit AM, et al. Selective cognitive impairment induced by focal and generalized epileptiform EEG activity[J]. Brain, 1984,107(1):293-308.

4. Castelijn-Nostren TRENITT DG, Riemersma JB, Binney CD, et al. The effect of subclinical epileptiform EEG discharges on driving performance[J]. Electroencephalography and Clinical Neurophysiology, 1987,67:167-170.

5. Kring JK, Scott RC, Holmes GL, et al. Interictal epileptiform activity in the hippocampus interferes with human cognition[J]. Neurology, 2013,81:18-24.

6. Aldenkamp AP, Arends J. The impact of epileptiform EEG discharges on cognitive function: is the concept of 'transient cognitive impairment' still valid?[J]. Epilepsy & Behavior, 2004,5(Suppl 1):S25-S34.

7. Holmes GL, Lenck-Santini PP. The role of interictal epileptiform abnormalities in cognitive impairment[J]. Epilepsy & Behavior, 2006,8:504-515.

8. Licht EA, Jacobson RH, Fujikawa DG. Chronic frontal lobe dysfunction caused by subclinical epileptiform discharges[J]. Epilepsy & Behavior, 2002,3:96-100.

9. Chapman KE, et al. Controlling epileptic activity improves outcome[J]. Epilepsia, 2015.

10. Galanopoulou AS, Moshe SL. Pathogenesis and novel candidate therapeutic approaches for infantile spasms and early epileptic encephalopathies: insights from preclinical studies[J]. Biology of the Nervous System, 2015,79:135-149.

11. Galanopoulou AS, Boyko A, Rado F, et al. The spectrum of neuropsychiatric abnormalities associated with electrical status epilepticus during sleep[J]. Brain & Development, 2000,22:279-295.

12. Patry G, Liagouris S, Taschner CA. Sleep-induced subclinical "electrical status epilepticus" in children: a clinical and electroencephalographic study of 6 cases[J]. Archives of Neurology, 1971,24:242-252.

13. Sánchez Fernández I, Chapman KE, Peters JM, et al. The Tower of Babel: a survey of concepts and terminology for electrical status epilepticus during sleep and continuous spike-and-wave during sleep in North America[J]. Epilepsia, 2013,54:741-750.

14. Eby A, Poznanski N, Verhelen D, et al. Efficacy of levetiracetam in the treatment of continuous spike-and-wave during sleep syndrome: experience in 12 patients[J]. Epilepsia, 2005,46:1937-1942.

15. Sánchez Fernández I, Peters JM, Hajiouz S, et al. Clinical staging and electroencephalographic evolution of continuous spike-and-wave during sleep[J]. Epilepsia, 2012,53:1185-1195.

16. Chavakula V, Sánchez Fernández I, Peters JM, et al. Automated quantitative analysis of spikes[J]. Epilepsy & Behavior, 2013,26:143-152.

17. Shuman DA, Erwin RJ. Focal spike-induced brain dysfunction associated with the subsequent slow wave[J]. Annals of Neurology, 1988,23:131-137.

18. Diekelmann S, Born J. The memory function of sleep[J]. Nature Reviews Neuroscience, 2010,11:114-126.

19. Capdevila O, Dayyat E, Hellandish-Gozal L, et al. Prevalence of epileptiform activity during sleep in healthy children[J]. Sleep Medicine, 2008,9:303-309.

20. Su EL. Interictal epileptiform discharges in individuals without a history of seizures: what is their significance?[J]. Journal of Clinical Neurophysiology, 2010,27:229-238.

21. Sam MC, Su EL. The significance of epileptiform discharges in non-epileptic patients in the community[J]. Epilepsia, 2001,42:1273-1278.

22. Holtmann M, Mattheus A, Herrmann U, et al. Centrotemporal spikes increase impulsivity in children with attention-deficit/hyperactivity disorder: a preliminary neuropsychological study[J]. Brain & Development, 2006,28:633-640.

23. Silvestri R, Galliano A, Carrese T, et al. Ictal and interictal EEG abnormalities in children with attention-deficit/hyperactivity disorder recorded by video-polysomnography[J]. Epilepsy Research, 2007,75:130-137.

24. Schreiber JM, Lanham DC, Treiber WH, et al. Changes in EEG discharges predict ADHD severity in Smith-Lemli-Opitz syndrome[J]. Neurology, 2014,83:151-159.

25. Picard A, Cheliout-Heraut F, Bouscary M, et al. Sleep EEG in developmental language disorders[J]. Developmental Medicine & Child Neurology, 1998,40:595-599.

26. Duvelois-Oume C, Guillet P, Biar C, et al. Unilateral hemisphere function in children with developmental language disorders[J]. Neuropsychologia, 1995,33:823-834.

27. Tuchman R. Autism and social cognition in epilepsy: implications for comprehensive epilepsy care[J]. Current Opinion in Neurology, 2013,26:214-218.

28. Mulligan CK, Trauner DA. Prevalence and behavioral correlates of epileptiform abnormalities in autism spectrum disorders[J]. Journal of Autism and Developmental Disorders, 2014,44:452-458.

29. Baraban-Gil K, Tuchman R. Epilepsy and epileptiform EEGs: associations with autism and language disorders[J]. Mental Retardation and Developmental Disabilities Research Reviews, 2000,6:300-308.

30. McVicar KA, Baraban-Gil K, Laxer I, et al. Epileptiform EEG abnormalities in children with language regression[J]. Neurology, 2005,65:129-131.

31. Galanopoulou AS, Moshe SL. The epilepsy hypothesis: a developmentally relevant argument based on animal models[J]. Epilepsia, 2009,50(Suppl 7):37-42.

32. Ebus SC, Eggermont DM, den Boer JT, et al. Changes in benign focal spike frequency correlate with changes in central information processing speed: a 2-year prospective follow-up study[J]. Epilepsy & Behavior, 2015,43C:8-15.

33. Binney CD, Castelijn-Nostren TRENITT DG, Smit AM, et al. Interaction between epileptiform EEG discharges and cognition[J]. Epilepsy Research, 1987,1:239-245.

34. Castelijn-Nostren TRENITT DG, Bakker DJ, Binney CD, et al. Psychological effects of subclinical epileptiform EEG discharges in children (I): academic abilities[J]. Epilepsy Research, 1988,2:111-116.

35. Aldenkamp AP, Butler J, Arends J, et al. Acute effects of subclinical epileptiform EEG discharges on cognitive activation[J]. Functional Neurology, 2005,20:23-28.

36. Nicolai J, Ebus S, Beemans DP, et al. The impact of interictal epileptiform EEG discharges and nonconvulsive seizures on cognition[J]. Epilepsia, 2012,53:1051-1059.

37. Binney CD. Cognitive impairment during epileptiform discharges: is treating the EEG justified?[J]. The Lancet Neurology, 2003,2:725-730.

38. Panayiotopoulos CP, Michael M, Sanders S, et al. Benign focal epilepsies of childhood: an appraisal of established and newly recognized syndromes[J]. Brain, 2008,131:2264-2286.

39. Filippini M, Boni A, Giannotta M, et al. Neuropsychological development in the spectrum of benign childhood epilepsy with centrotemporal spikes: long-term effects of epileptiform activity[J]. Epilepsy & Behavior, 2013,28:504-511.

40. Massa R, San Martin A, Cacanju R, et al. Electroencephalographic predictors of complex evolution in idiopathic centrotemporal epilepsy[J]. Neurology, 2001,57:1071-1079.

41. Kanemura H, Sano F, Aoyagi K, et al. Can serial EEG changes predict atypical clinical features in centrotemporal epilepsy?[J]. Developmental Medicine & Child Neurology, 2012,54:912-917.

42. Wolff M, Weiskopf N, Serra E, et al. Benign partial epilepsy of childhood: selective cognitive deficits associated with focal spikes localized by EEG/magnetoencephalography[J]. Epilepsia, 2005,46:1661-1667.

43. Berman R, Negishi M, Westerveld M, et al. Simultaneous EEG, fMRI, and behavioral study of typical childhood absence epilepsy[J]. Epilepsia, 2010,51:2011-2022.

44. Conant LL, Wilfong A, Inglis C, et al. Executive function and associated dysfunction in typical childhood absence epilepsy[J]. Epilepsy & Behavior, 2010,18:414-423.

45. Wirrell EC, Canfield CS, Canfield PR, et al. Long-term psychosocial outcome of typical childhood absence epilepsy: sometimes a wolf in sheep's clothing[J]. Archives of Pediatrics & Adolescent Medicine, 1997,151:152-158.

46. Piazzi A, Turner K, Vignoli A, et al. Frontal cognitive dysfunction in juvenile myoclonic epilepsy[J]. Epilepsia, 2008,49:657-662.

47. Wanderscheid B, Kopp UA, Kliegel M, et al. Prospective memory in patients with juvenile myoclonic epilepsy and their healthy siblings[J]. Neurology, 2010,75:2161-2167.

48. Sánchez Fernández I, Loddenkemper T, Peters JM, et al. Electrical status epilepticus during sleep: clinical presentation and pathophysiology[J]. Pediatric Neurology, 2012,47:390-410.

49. Roussel C, Levavasseur M. Relationship between continuous spike-and-wave during sleep and cognitive function//Beaussart A, Biro M, Deonna L, et al. Continuous Spike-and-Wave During Sleep[M]. London: John Libbey & Co, 1995:123-133.

50. Kramer U, Sagi L, Goldberg-Stern H, et al. Clinical spectrum and medical treatment of electrical status epilepticus during sleep in children[J]. Epilepsia, 2009,50:1517-1524.

51. Calabayo RH, Bongiorni L, Celso Simão R, et al. Electrical status epilepticus during sleep epileptic encephalopathy in children with shunted hydrocephalus: a study of 9 cases[J]. Epilepsia, 2008,49:1520-1527.

52. Loddenkemper T, Cosmo G, Kotagal P, et al. Epilepsy surgery for electrical status epilepticus during sleep in children[J]. Journal of Neurosurgery, 2009,64:328-337.

53. Pertola ME, Luukinen E, Granström ML, et al. Efficacy of surgery for electrical status epilepticus during sleep encephalopathy[J]. Epilepsia, 2011,52:602-609.

54. Scholtens FB, Hendriks MP, Reijniers WO. Cognitive impairment and continuous spike-and-wave during sleep[J]. Epilepsy & Behavior, 2005,6:167-173.

55. Lesca G, Rudolf G, Bruno N, et al. GRIN2A mutations cause acquired epileptic aphasia and related childhood focal epilepsies and encephalopathies with speech and language dysfunction[J]. Nature Genetics, 2013,45:1061-1066.

56. Lesca G, Rudolf G, Labalme A, et al. Landau-Kleffner syndrome and electrical status epilepticus during sleep epileptic encephalopathies: genomic analysis reveals links to autism[J]. Epilepsia, 2012,53:1526-1538.

57. Mefford HC, Batshaw ML, Hoffman EP. Genomics, intellectual disability, and autism[J]. The New England Journal of Medicine, 2012,366:733-743.

58. Novarino G, El-Fishawy P, Kacar H, et al. BCKD kinase mutations cause a potentially treatable form of autism with epilepsy[J]. Science, 2012,338:394-397.

59. Sahin M. Targeted treatment trials for tuberous sclerosis complex and autism: no longer a dream[J]. Current Opinion in Neurobiology, 2012,22:895-901.

60. Tuchman R, Hertz D, Mamounas LA. Report of the NINDS workshop on epilepsy and autism spectrum disorders[J]. Neurology, 2013,81:1630-1636.

61. Lemke JR, Lal D, Reintaler EM, et al. GRIN2A mutations cause idiopathic focal epilepsies with centrotemporal spikes[J]. Nature Genetics, 2013,45:1067-1072.

62. Guzzetta F, Battaglia D, Verrecchia C, et al. Early thalamic injury is associated with epilepsy and continuous spike-and-wave during sleep[J]. Epilepsia, 2005,46:889-900.

63. Kersbergen KJ, de Vries LS, Leijten FS, et al. Neonatal thalamic hemorrhage is highly associated with electrical status epilepticus during sleep[J]. Epilepsia, 2013,54:733-740.

64. Sánchez Fernández I, Takeoka M, Tas E, et al. Early thalamic lesions in patients with sleep-facilitated epileptiform activity[J]. Neurology, 2012,78:1721-1727.

65. Gordon K, Bowden H, Canfield P, et al. Valproic acid treatment of learning-disabled children with severe epileptiform EEG changes but no clinical seizures[J]. Journal of Child Neurology, 1996,11:41-43.

66. Ronen GM, Richards JE, Cunningham C, et al. Does sodium valproate improve learning ability in children with epileptiform discharges but no clinical seizures?[J]. Developmental Medicine & Child Neurology, 2000,42:751-755.

67. Mintz M, Legoff D, Skolnick-Chin J, et al. An underrecognized epilepsy spectrum: the effect of levetiracetam on neuropsychological function associated with subclinical spike production[J]. Journal of Child Neurology, 2009,24:807-815.

68. Altintas P, Midor KJ. Cognitive side effects of antiepileptic drugs[J]. Epilepsy & Behavior, 2004,5(Suppl 1):S60-S65.

69. Denegri M, Valetto MG, Battaglia FM, et al. Short course diazepam after rectal diazepam trial in electrical status epilepticus[J]. Brain & Development, 1995,17:330-333.

70. Sánchez Fernández I, Hajiouz S, Eksioglu Y, et al. Short-term efficacy of high-dose diazepam on sleep-facilitated spikes in electrical status epilepticus during sleep[J]. Pediatric Neurology, 2012,46:312-318.

71. Sánchez Fernández I, Peters JM, Ahn S, et al. Long-term efficacy of high-dose diazepam for continuous spike-and-wave during sleep[J]. Pediatric Neurology, 2013,49:163-170.

72. Luukinen E, Kantola-Sorsa E, Peltola R, et al. Long-term outcome of 32 children with electrical status epilepticus during sleep encephalopathy[J]. Epilepsia, 2010,51:2023-2032.

73. Pressler RM, Robinson RO, Wilson GA, et al. Treating interictal epileptiform discharges improves behavior in epileptic children with behavioral problems[J]. The Journal of Pediatrics, 2005,146:112-117.

74. Libenson MH, Haldar A, Pinto AL. Stability of spike counts in children with interictal epileptiform activity[J]. Seizure, 2014.

75. Libenson MH, Caravale B. Do different antiepileptic drugs differ in their ability to suppress interictal epileptiform activity in children?[J]. Pediatric Neurology, 2001,24:214-218.

76. Eriksson AS, Knutsson E, Negård A. Effect of lamotrigine on epileptiform discharges in young patients with refractory epilepsy[J]. Epilepsia, 2001,42:230-236.

77. Dahlin M, Knutsson E, Amark P, et al. A randomized double-blind study of low-dose clonazepam in reducing epileptiform activity in children with epilepsy[J]. Epilepsia, 2000,41:308-315.

78. Larsson PG, Bakke KA, Bjørnes H, et al. Effect of levetiracetam on focal nocturnal epileptiform activity during sleep: a placebo-controlled double-blind crossover study[J]. Epilepsy & Behavior, 2012,24:44-48.

79. Buzatu M, Bulteau C, Artus C, et al. Corticosteroid treatment for continuous spike-and-wave during sleep syndrome[J]. Epilepsia, 2009,50(Suppl 7):68-72.

80. Cavazzuti GB, Cappella L, Nalin A. Longitudinal study of epileptiform EEG patterns in normal children[J]. Epilepsia, 1980,21:43-55.

81. Borusiak P, Zielbauer M, Janke AC. Prevalence of epileptiform discharges in healthy children: new data from a prospective digital EEG study[J]. Epilepsia, 2010,51:1185-1188.

82. Holtmann M, Becker K, Kentner-Figuera B, et al. Increased incidence of centrotemporal spikes in children with attention-deficit/hyperactivity disorder[J]. Epilepsia, 2003,44:1241-1244.

83. Siebelink BM, Bakker DJ, Binney CD, et al. Psychological effects of subclinical epileptiform EEG discharges in children (II): general intelligence tests[J]. Epilepsy Research, 1988,2:117-121.

84. Fonseca LC, Tedrus GM, Pacheco EM. Epileptiform EEG discharges in benign childhood epilepsy with centrotemporal spikes: reactivity and transient cognitive impairment[J]. Epilepsy & Behavior, 2007,11:65-70.