Design and evaluation of an autonomic nerve monitoring system based on skin sympathetic nerve activity

Highlights

• CVDs can show symptomatic cardiac autonomic function changes.

• System to record electrocardiogram (ECG) and skin sympathetic nerve activity (SKNA).

• Combining a commercial analog front end (AFE) with a first-stage amplifier to reduce the system noise floor.

• Noninvasive and real-time assessment of sympathetic nerve activity during anesthesia.

Abstract

The autonomic nervous system is closely related to cardiovascular diseases (CVDs). Simultaneous non-invasive recording of skin sympathetic nerve activity (SKNA) and electrocardiogram (ECG) is a new method for autonomic nervous system real-time assessment. This study presented a portable monitoring system based on SKNA. A system-level modification by combining a commercial analog front end (AFE) chip with a low-noise first-stage amplifier was implemented to reduce the system noise floor without a high-cost customized chip. An adaptive power-line-interference (PLI) filter and outliers clipping were developed to reject the PLI and motion artifacts in the signal. The laboratory experiment and clinical experiment were conducted to verify the performance and effectiveness of the proposed system. The laboratory results show that the proposed AFE has a much lower noise floor with 0.1 ${\mu V}_{\mathit{rms}}$ than the reference systems (PowerLab system). Moreover, the correlation coefficient of the envelope of SKNA (eSKNA) is 0.83, and the correlation coefficient of the heart rate variability (HRV) index is 0.99, suggesting a good performance in signal recording. The clinical results show that the proposed system can reflect sympathetic nerve activity significantly before and after anesthesia injection (p < 0.01), indicating its better feasibility in this application scenario than HRV-based devices (p = 0.40). The system achieves a comparable performance to the reference system and satisfactory performance for the autonomic nervous system assessment in clinical application.

Introduction

Cardiovascular diseases (CVDs) are the leading cause of death, accounting for 32.1% of the death cases globally [1], [2]. Some CVDs can show symptomatic cardiac autonomic function changes, manifested as a disorder or structural damage of sympathetic and vagal innervation [3]. These symptoms can develop into cardiac arrhythmia and other diseases such as gastrointestinal neurosis and hyperventilation syndrome [3]. For example, the dual activation of sympathetic and parasympathetic systems is the most common trigger of atrial fibrillation [4]. In patients with ischemic heart diseases, elevated sympathetic activity arises ventricular arrhythmias [5]. Monitoring the sympathetic nervous system can reflect one’s autonomic nerve function [3]. As one of the two branches of the autonomic nervous system, the sympathetic nervous system functions as an accelerator and ensures the human body’s physiological needs in a tense state, taking part in various physiological processes [6]. Hence, sympathetic nerve monitoring is of interest to patients and care providers for disease diagnosis and risk stratification. If proper autonomic nervous system activity evaluations are available, therapies such as neuromodulation could be provided to improve the outcomes of the treatment for some patients with CVDs [5].

Microneurography [7] and heart rate variability (HRV) analysis from electrocardiogram (ECG) are two [8] conventional methods of estimating sympathetic nerve activity. Microneurography can record sympathetic nerve activity by inserting a fine tungsten needle electrode into the nerves [9]. This method is invasive and thus not the most feasible and convenient solution for human subject research that targets daily monitoring applications. HRV requires proper sinus node function [10], and it is not practicable for patients with atrial fibrillation because their rhythm is not sinus rhythm [11]. In addition, HRV cannot reflect the dynamic changes of sympathetic nerve activity because of their indirect parameters from segments that last 5 min. Photoplethysmogram (PPG) and galvanic response (GSR) have the potential to evaluate the autonomic nervous system [13]. Still, the relationship between these physiological signals and the autonomic nervous system needs to be further verified. Therefore, it is challenging to perform effective and dynamic sympathetic nerve monitoring and evaluation using conventional methods. A research team recently proposed a new method for sympathetic nerve assessment [12] by collecting simultaneously non-invasive recordings of ECG and skin sympathetic nerve activity (SKNA) [11], [12], [13], [14]. The sympathetic nerve activity can be quantified by analyzing different frequency bands of the recorded signals (ECG: 0.5–150 Hz, SKNA: 500–1000 Hz). Fig. 1 shows a typical example of a sympathetic burst during the Valsalva Maneuver (VM). The VM increases blood pressure, heart rate, and sympathetic nerve activity level. As shown in Fig. 1 (a), the parasympathetic nerves originate from the brain stem (green line) while the sympathetic nerve arises from the spinal cord (red line). Especially, the stellate ganglion sends post-ganglionic sympathetic fibers both to the heart and skin. Electrodes can measure the subcutaneous sympathetic nerve activity to reflect the cardiac sympathetic nerve activity. The burst of SKNA signals reflects the activation of sympathetic activity, as shown in the red dotted box of Fig. 1 (b) [11]. It is a synthetic behavior of compound action potentials generated by a series of neurologically activated motor units. It has been shown that SKNA bursts represent critical indications of the starting and termination of cardiac arrhythmias [3].

PowerLab systems were used as the reference equipment for SKNA measurement in literature [12], [18]. However, the instrument is bulky and not dedicated to autonomic nerve assessment. Thus, they are cumbersome for physicians and patients to operate and unsuitable for portable daily monitoring. The lack of a dedicated monitoring system limits the practical application and evaluation of such a new modality despite its promising application potentials in clinical assessments. Therefore, a low-cost, small-profile, and portable autonomic nerve assessment system is needed.

Our previous work [17] presented a prototype design of a portable autonomic nerve monitoring system. To the best of our knowledge, the prototype in [17] is the first dedicated device specially designed for autonomic nerve monitoring. Our pilot study indicates that the system can be used for the simultaneous non-invasive recording of ECG and SKNA. However, the previous pilot study needs further research in several aspects [19]. Firstly, our pilot results indicate that the proposed system’s noise performance requires further improvement to ensure satisfactory SKNA signal quality [20]. Moreover, the previous analysis only evaluates the calculation of SKNA, lacking a comprehensive comparison and evaluation with HRV-based metrics. Lastly, lab and clinical experiments have not thoroughly validated the system’s noise floor and clinical effectiveness.

In this manuscript, we extensively addressed these aspects point-by-point. Firstly, we designed a low-noise analog-front-end (AFE) circuit. The analytical and experimental results reveal a much lower noise floor than the reference systems and the prototype in our previous work. Secondly, SKNA calculation is combined with HRV analysis, providing a more comprehensive autonomic nervous system status evaluation framework. Lastly, a clinical experiment was conducted during a multi-sample anesthesia injection operation to verify the system’s reliability and effectiveness in a practical clinical scenario. In summary, the previous pilot study is strengthened with a new analog-front-end design and updated software, providing extensive theoretical, experimental, and statistical results to demonstrate the performance and effectiveness of the proposed system [21], [22], [23].