Abstract：
Abnormal patterns identified on electroencephalogram (EEG) are one of the primary diagnostic tests for epilepsy. However, epidemiological studies have established that both benign and epileptiform abnormalities (EAs) occur on the EEG of nonepileptic, seizure-free people as well. The reported rates of EAs in nonepileptic, seizure-free populations vary, and the true prevalence is unknown. The primary objective of this systematic review and meta-analysis was to estimate the overall prevalence of EAs in the EEG of people without a history of seizures. Secondary aims were to characterize (1) the cortical localization of focal abnormalities, (2) the proportion of findings that occurred during standard EEG stimulation protocols, and (3) the persistence and implications of abnormalities at follow-up. A comprehensive electronic search of six bibliographic databases was completed: Embase, MEDLINE, PsycInfo, Cumulative Index of Nursing and Allied Health Literature, Cochrane Central Register for Controlled Trials, and Web of Science. No search date restrictions were applied. Overall effect size was calculated using a generalized linear mixed-effects model. Fifty-three studies, totaling 73 990 individuals, met our inclusion criteria. The overall point prevalence of EAs was 1.74% (95% confidence interval [CI] = 1.13-2.67). Due to the risk of bias in the literature, especially from participant selection, we believe this to be an overestimate of the true prevalence. Prevalence of EAs was greater in children (2.45%, 95% CI = 1.41-4.21) and the elderly (5.96%, 95% CI = 1.39-22.13) compared with adults (.93%, 95% CI = .48-1.80). Reports of developing epilepsy after an EA-positive EEG were rare. The likelihood of subsequent positive findings on follow-up EEG may be as high as 50%. Our study has limitations in that males were overrepresented in the study samples, there is substantial heterogeneity among studies, and many studies provided insufficient detail about their exclusion criteria. Nonetheless, our estimates provide benchmark data for future studies examining EAs in clinical populations, particularly behavioral and psychiatric populations.
摘要：
在脑电图（EEG）上识别出的异常模式是癫痫的主要诊断测试之一。然而,流行病学研究已经确定，良性和癫痫样异常（EA），发生在非癫痫的脑电图中，无癫痫的人也是如此。报告的非癫痫患者的EA率，无癫痫发作的人群各不相同，和真正的流行是未知的。这项系统评价和荟萃分析的主要目的是评估没有癫痫发作史的人的EEG中EA的总体患病率。次要目的是表征i）局灶性异常的皮层定位；ii）在标准EEG刺激方案中发生的发现比例；iii）随访中异常的持久性和含义。完成了对六个书目数据库的全面电子搜索：EMBASE，MEDLINE,PsycINFO,护理和相关健康文献的累积指数，Cochrane中央控制试验登记册,和WebofScience。未应用搜索日期限制。使用广义线性混合效应模型计算总效应大小。53项研究,共有73,990人,符合我们的纳入标准。EA的总体点患病率为1.74％（95％CI：1.13-2.67）。由于文献中存在偏见的风险，特别是从参与者的选择，我们认为这是对真正流行率的高估。儿童中EA的患病率更高(2.45%，1.41-4.21)和老年人(5.96%，1.39-22.13）与成年人（0.93%，0.48-1.80）。EA阳性EEG后发生癫痫的报道很少。后续脑电图结果为阳性的可能性可能高达50%。我们的研究有局限性，因为研究样本中男性比例过高，研究之间存在很大的异质性，许多研究提供的关于其排除标准的细节不足.尽管如此，我们的估计为未来在临床人群中检查EA的研究提供了基准数据，特别是行为和精神人群。
