Background

Transient loss of consciousness (TLOC) may be mistaken for other disorders like epilepsy. Our objectives were to identify symptoms that could help differentiate epilepsy from syncope among children with TLOC and to validate previously suggested criteria.

Methods

We retrospectively reviewed the charts of patients aged 18 years or younger who presented with TLOC attacks from January 2008 to December 2018 at King Saud University Medical City, Riyadh, Saudi Arabia. Symptoms from which epilepsy and syncope could be predicted with high accuracy were included in the previously suggested criteria. The discriminative abilities of current and previous criteria were examined in receiver-operating characteristic analyses.

Results

Data from 46 patients, 32 with confirmed epilepsy and 14 with syncope, were included in this analysis. The mean age was 12.1 years (S.D., 4.3 years), and 60.9% of the patients were girls. According to our proposed criteria, the sensitivity, specificity, and accuracy of symptoms in predicting epilepsy were 68.8%, 85.7%, and 73.9%, respectively, and the area under the curve was 0.814 (confidence interval 0.686 to 0.941, P = 0.001). According to previously suggested criteria, the sensitivity, specificity, and accuracy of symptoms in predicting epilepsy were 63.2%, 62.5%, and 63.0%, respectively, and the area under the curve was 0.730 (confidence interval 0.541 to 0.92, P = 0.063).

Conclusions

A number of self-reported/observed symptoms can be used to distinguish epilepsy from syncope with high discriminative ability. The current findings still need to be validated in larger, preferably multiple populations before they can be safely relied upon.

中文翻译：

---

儿童晕厥发作：与心脏或癫痫有关吗？

背景

短暂性意识丧失 (TLOC) 可能会被误认为是癫痫等其他疾病。我们的目标是确定有助于区分 TLOC 儿童癫痫和晕厥的症状，并验证先前建议的标准。

方法

我们回顾性审查了 2008 年 1 月至 2018 年 12 月在沙特阿拉伯利雅得的沙特国王大学医学城发生 TLOC 攻击的 18 岁或以下患者的图表。可以高精度预测癫痫和晕厥的症状包含在先前建议的标准中。在接收者操作特征分析中检查了当前和以前标准的辨别能力。

结果

该分析包括来自 46 名患者的数据，其中 32 名确诊为癫痫，14 名患有晕厥。平均年龄为 12.1 岁（SD，4.3 岁），60.9% 的患者为女孩。根据我们提出的标准，症状预测癫痫的敏感性、特异性和准确性分别为 68.8%、85.7% 和 73.9%，曲线下面积为 0.814（置信区间 0.686 至 0.941，P  = 0.001） . 根据之前提出的标准，症状预测癫痫的敏感性、特异性和准确性分别为63.2%、62.5%和63.0%，曲线下面积为0.730（置信区间0.541～0.92，P  =0.063） .

结论

许多自我报告/观察到的症状可用于区分癫痫和晕厥，并具有高辨别力。目前的发现仍然需要在更大的人群中进行验证，最好是在多个人群中进行验证，然后才能安全地依赖它们。