Assessment of EEG Responses to Spinal Cord Stimulation Waveforms in Chronic Pain Patients

慢性疼痛是一个严重的问题，据估计影响了大约5000万成年美国人。它占医疗保健支出，残疾和生产力损失的6000多亿美元，仍然是患者寻求医疗保健的最常见原因之一。脊髓刺激（SCS）是美国食品药品监督管理局（FDA）批准的神经调节治疗，用于缓解慢性顽固性疼痛。虽然SCS可有效用于许多患有难治性慢性疼痛病症的患者，但相当一部分患者的疼痛抑制效果不佳或不充分。考虑到每年有超过30，000人因慢性疼痛接受SCS植入，并且由于缺乏关于最佳刺激参数的共识而导致疼痛缓解不满意，因此每位患者的年成本超过24，000美元，因此提高SCS在慢性难治性疼痛中的临床疗效至关重要。除了传统的刺激（强直），使用新的刺激波形，如突发和高频刺激（HFS）是一种方法。这些波形不会引起感觉异常（刺痛或麻木的感觉），并且它们的效果通常延迟24- 48小时，不像强直刺激，其中效果是立即的。然而，这些波形的脊髓上效应仍然不是很清楚。也不知道为什么有些患者对一种波形的反应优于另一种波形。另一个局限性是SCS疼痛缓解的评估几乎完全取决于患者的自我报告，缺乏SCS植入物结果的客观测量可能导致永久性植入失败的数量增加。在这方面，调查神经模式的SCS波形和评估SCS引起的疼痛缓解清醒和麻醉慢性疼痛患者使用脑电图（EEG）似乎是必不可少的。我们的初步数据表明，α峰频率和α和θ波段功率比的变化可能与SCS诱导的疼痛缓解相关。这些模式的空间分布与所使用的波形密切相关。我们假设术中EEG表现出与术后EEG相似的空间-频谱模式。我们预测，疼痛缓解将最好地与增加α波段功率在额叶和躯体感觉区的爆发刺激和躯体感觉皮质区HFS。在这项提议中，我们的目标是确定α波段特征与清醒患者SCS诱导的疼痛缓解之间的关系（目标1），确定“睡眠”和清醒状态EEG之间的空间-频谱差异，并表征疼痛缓解的术中神经相关性（目标2），并检查当疼痛缓解最小的患者被重新编程以体验疼痛缓解时α带特征的变化（目标3）。具体来说，在目标3中，我们将根据这些神经特征开发一个数学模型，并将患者临时编程为EEG引导设置。我们将研究评估的神经模式对选择最佳波形的预测能力。该项目的直接目标是开发量化的神经信号缓解疼痛。这些客观标记可以形成器械选择的客观指标，这可能会减少与植入物失败相关的医疗保健支出，并有助于指导SCS参数以更好地缓解疼痛，特别是在疼痛缓解程度最低的患者中。总体而言，这些客观的神经生理学指标可用于更个性化的SCS治疗，具有更高的疗效，本研究可有助于神经技术和神经工程方法的发展。

Chronic pain is a significant problem and it has been estimated to affect about 50 million adult Americans. It accounts for more than 600 billion dollars in health care expenditures, disability and loss of productivity and remains one of the most common reasons that patients seek healthcare. Spinal cord stimulation (SCS) is a Food and Drug Administration (FDA)-approved neuromodulation treatment to relieve chronic refractory pain. While SCS can be used effectively in many patients with refractory chronic pain conditions, a significant portion of the patients receive suboptimal or inadequate pain suppression. Considering that over 30,000 individuals receive SCS implantation on an annual basis for chronic pain and an annual cost of over $24,000 per patient resulting from lack of satisfactory pain relief due to lack of consensus regarding optimal stimulation parameters, it is crucial to improve the clinical efficacy of SCS in chronic refractory pain. Beside the traditional stimulation (tonic), using novel stimulation waveforms such as burst and high frequency stimulation (HFS) is one approach. These waveforms do not induce paresthesia (sensation of tingling or numbness) and their effects are often delayed 24- 48 hours unlike tonic stimulation, where the effects are immediate. However, the supraspinal effects of these waveforms are still not very clear. It is also unknown why some patients respond well to one waveform over another. Another limitation is that assessment of the pain relief by SCS depends almost exclusively on patient self-reports and lack of objective measurements of SCS implants results might lead to increased number of failed permanent placements. In this regard, investigating neural patterns of SCS waveforms and assess SCS-induced pain relief by using electroencephalography (EEG) in awake and anesthetized chronic pain patients appears essential. Our preliminary data suggests that changes in alpha peak frequency and ratio of alpha and theta band power might be correlated to SCS-induced pain relief. The spatial distribution of these patterns is closely related to which waveform is used. We hypothesize that intra-operative EEG demonstrates similar spatio-spectral patterns to post-operative EEG. We predict that pain relief will correlate best with increased alpha band power in frontal and somatosensory regions in burst stimulation and somatosensory cortical regions in HFS. In this proposal, our objectives are to determine the relationship between alpha band features and SCS-induced pain relief in awake patients (Aim 1), determine the spatio-spectral differences between “sleep” and awake states EEG and characterize the intra-operative neural correlates of pain relief (Aim 2), and examine the changes in alpha band features when patients with minimal pain relief are reprogrammed to experience pain relief (Aim 3). Specifically, in Aim 3, we will develop a mathematical model based on these neural features and patients will be temporarily programmed to this EEG-guided setting. We will investigate the predictive power of the assessed neural patterns on selection of the optimal waveform. The immediate goal of this project is to develop quantified, neural signatures of pain relief. These objective markers could form an objective metric for device selection, which might reduce the healthcare expenditures associated with failed implants and help in guiding SCS parameters for better pain relief, especially in patients with minimal pain relief. Overall, these objective neurophysiological measures can be used for more personalized SCS therapy with higher efficacy, and, this study can contribute to development of neurotechnologies and neuroengineering methodologies.