Tamborska Arina, Bashford James, Wickham Aidan, Iniesta Raquel, Masood Urooba, Cabassi Cristina, Planinc Domen, Hodson-Tole Emma, Drakakis Emmanuel, Boutelle Martyn, Mills Kerry, Shaw Chris
Department of Basic and Clinical Neuroscience, Maurice Wohl Clinical Neuroscience Institute, King's College London, London, UK.
Department of Bioengineering, Imperial College London, London, UK.
Brain Commun. 2020 Sep 7;2(2):fcaa141. doi: 10.1093/braincomms/fcaa141. eCollection 2020.
Delayed diagnosis of amyotrophic lateral sclerosis prevents early entry into clinical trials at a time when neuroprotective therapies would be most effective. Fasciculations are an early hallmark of amyotrophic lateral sclerosis, preceding muscle weakness and atrophy. To assess the potential diagnostic utility of fasciculations measured by high-density surface electromyography, we carried out 30-min biceps brachii recordings in 39 patients with amyotrophic lateral sclerosis, 7 patients with benign fasciculation syndrome, 1 patient with multifocal motor neuropathy and 17 healthy individuals. We employed the surface potential quantification engine to compute fasciculation frequency, fasciculation amplitude and inter-fasciculation interval. Inter-group comparison was assessed by Welch's analysis of variance. Logistic regression, receiver operating characteristic curves and decision trees discerned the diagnostic performance of these measures. Fasciculation frequency, median fasciculation amplitude and proportion of inter-fasciculation intervals <100 ms showed significant differences between the groups. In the best-fit regression model, increasing fasciculation frequency and median fasciculation amplitude were independently associated with the diagnosis of amyotrophic lateral sclerosis. Fasciculation frequency was the single best measure predictive of the disease, with an area under the curve of 0.89 (95% confidence interval 0.81-0.98). The cut-off of more than 14 fasciculation potentials per minute achieved 80% sensitivity (95% confidence interval 63-90%) and 96% specificity (95% confidence interval 78-100%). In conclusion, non-invasive measurement of fasciculation frequency at a single time-point reliably distinguished amyotrophic lateral sclerosis from its mimicking conditions and healthy individuals, warranting further research into its diagnostic applications.
肌萎缩侧索硬化症的延迟诊断会阻碍患者在神经保护疗法最有效的时候尽早进入临床试验。肌束震颤是肌萎缩侧索硬化症的早期特征，早于肌肉无力和萎缩。为了评估通过高密度表面肌电图测量的肌束震颤的潜在诊断效用，我们对39例肌萎缩侧索硬化症患者、7例良性肌束震颤综合征患者、1例多灶性运动神经病患者和17名健康个体进行了30分钟的肱二头肌记录。我们使用表面电位量化引擎来计算肌束震颤频率、肌束震颤幅度和肌束震颤间期。通过韦尔奇方差分析进行组间比较。逻辑回归、受试者工作特征曲线和决策树辨别了这些测量方法的诊断性能。肌束震颤频率、肌束震颤幅度中位数和肌束震颤间期<100毫秒的比例在各组之间存在显著差异。在最佳拟合回归模型中，肌束震颤频率增加和肌束震颤幅度中位数增加与肌萎缩侧索硬化症的诊断独立相关。肌束震颤频率是预测该疾病的单一最佳指标，曲线下面积为0.89（95%置信区间0.81 - 0.98）。每分钟超过14个肌束震颤电位的临界值达到了80%的灵敏度（95%置信区间63 - 90%）和96%的特异性（95%置信区间78 - 100%）。总之，在单个时间点对肌束震颤频率进行非侵入性测量能够可靠地将肌萎缩侧索硬化症与其模仿病症及健康个体区分开来，有必要对其诊断应用进行进一步研究。
