Department of Psychiatry, Faculty of Medicine, The Chinese University of Hong Kong, Shatin, N.T., Hong Kong, China.
Department of Psychiatry, Faculty of Medicine, The Chinese University of Hong Kong, Shatin, N.T., Hong Kong, China.
Sleep Med. 2019 Apr;56:128-134. doi: 10.1016/j.sleep.2019.01.018. Epub 2019 Jan 28.
To investigate whether baseline electromyography (EMG) activity during rapid eye movement (REM) sleep predicts the development of neurodegenerative diseases over time in patients with idiopathic REM sleep behavior disorder (iRBD).
A total of 216 patients with polysomnography-confirmed iRBD were recruited from September 1997 to December 2016 with a mean follow-up duration of 5.0 ± 3.7 years (median: 4.0, range: 0.5-19.0). Neurodegenerative diseases were ascertained according to standard diagnostic criteria during follow-up. Time-dependent receiver operating characteristic (ROC) curves were employed to evaluate the dynamic predictive performance of EMG activity over time. Both tonic and phasic EMG activity were dichotomized into 'mild' and 'severe' categories by the ROC curves estimated optimal cut-offs.
A total of 58 patients (26.9%) developed neurodegenerative diseases. The predictive performance of tonic EMG activity was stable (area under the curve of approximately 0.68) over time, while the performance of phasic EMG activity was significantly superior to chance only after five years of follow-up. The optimal cut-off for prediction at five years was 15.4% (sensitivity, 0.69; specificity, 0.57) and 7.8% (sensitivity, 0.79; specificity, 0.47) for tonic and phasic EMG activity, respectively. Cox proportional hazards regression analyses further revealed that severe tonic (adjusted HR: 2.76, 95% CI: 1.35-5.62) and phasic EMG activity (adjusted HR: 3.10, 95% CI: 1.10-8.71) were associated with early development of Parkinson's disease (PD) and dementia with Lewy bodies (DLB), respectively.
Tonic but not phasic EMG activity may serve as a stable biomarkers for predicting the progression of neurodegeneration in iRBD.
研究特发性 REM 睡眠行为障碍（iRBD）患者的 REM 睡眠期间基线肌电图（EMG）活动是否随时间推移预测神经退行性疾病的发展。
共招募了 216 名经多导睡眠图（PSG）证实的 iRBD 患者，从 1997 年 9 月至 2016 年 12 月入组，平均随访时间为 5.0±3.7 年（中位数：4.0，范围：0.5-19.0）。在随访期间根据标准诊断标准确定神经退行性疾病。采用时间依赖性接受者操作特征（ROC）曲线评估 EMG 活动随时间的动态预测性能。通过 ROC 曲线估计的最佳截断值，将肌紧张和阵发性 EMG 活动均分为“轻度”和“重度”两类。
共有 58 例患者（26.9%）发生神经退行性疾病。肌紧张 EMG 活动的预测性能随时间保持稳定（曲线下面积约为 0.68），而肌阵挛 EMG 活动的性能仅在随访 5 年后才明显优于机会水平。预测 5 年的最佳截断值分别为 15.4%（敏感性，0.69；特异性，0.57）和 7.8%（敏感性，0.79；特异性，0.47），用于肌紧张和肌阵挛 EMG 活动。Cox 比例风险回归分析进一步表明，严重的肌紧张（调整后的 HR：2.76，95%CI：1.35-5.62）和肌阵挛 EMG 活动（调整后的 HR：3.10，95%CI：1.10-8.71）分别与帕金森病（PD）和路易体痴呆（DLB）的早期发病相关。
肌紧张但不是肌阵挛 EMG 活动可能作为 iRBD 神经退行性变进展的稳定生物标志物。
