Heart Rate Variability analysis in patients undergoing local anesthesia

Shafqat, K. (2010). Heart Rate Variability analysis in patients undergoing local anesthesia. (Unpublished Doctoral thesis, City, University of London)

[img]
Preview
Text - Accepted Version
Download (23MB) | Preview

Abstract

The analysis of Heart Rate Variability (HRV), the beat to beat fluctuation in the heart rate, is a non-invasive technique with a main aim in gaining information about the autonomic neural regulation of the heart. Assessment of HRV has been shown to aid clinical diagnosis
and intervention strategies. However, there are quite a few conflicting reports on HRV that perhaps impede its use as a reliable clinical tool. The complex nature of different mechanisms that affect the HRV and the large number of signal processing techniques that have been used for HRV analysis are the contributing factors of these conflicting results. The aim of this study was to investigate for the first time the effect of HRV during
Brachial plexus block (local anaesthesia), applied using the axillary approach. The hypothesis was that, such investigation will enable the detection of possible changes in the dynamics of the cardiovascular system due to the intravenous introduction of anaesthetic drugs during local anaesthesia. For this purpose advanced HRV signals processing techniques were developed and evaluated on data collected before and after the application of the Brachial plexus block from fourteen patients undergoing local anaesthesia. Signal processing techniques for R-wave detection, signal representation, ectopic beat detection and detrending were first developed and validated with the help of simulated signals and physiological signals from Physionet data base. After the validation stage these methods were then used to analyse the data from the locally anaesthetised patients.

The ECG R-wave peak detection was carried out using the wavelet transform with first derivative of Gaussian smoothing function as the mother wavelet. The algorithm achieved accuracy and sensitivity of over 90%. The heart timing signal was used for the HRV signal representation and also for the correction of missing and/or ectopic beats. The results obtained from the ectopic beat correction algorithm showed that the algorithm managed to significantly reduce the error caused by missing and/or ectopic beats. Detrending of the HRV signal was carried out using the wavelet packet analysis algorithm which was specifically developed for this study. The respiration signal was also estimaited from the ECG signal using the ECG Derived Respiration (EDR) technique. In order
to take better account of slow respiration rates and/or irregular respiratory patterns in the HRV analysis, a new method for the estimation of the variable boundaries associated with the LF and the HF band of the HRV signal was implemented. This method relies on the frequency contents of both the HRV signal and the respiration signal and uses the cross-spectrum between these two signals to obtain the boundaries related to the HF band of the signal. The boundaries related to the LF band were defined using the HRV signal spectrum alone. The boundary estimation technique was applicable in all the spectral analysis methods that were used in this study.

After the pre-processing steps the clinical data was analysed using frequency and timefrequency analysis methods to obtain the parameters related to the HRV signals. Initially spectral analysis was carried out using the traditional non-parametric (Welch’s periodogram) and parametric (Autoregressive modelling) methods. Statistical analysis of the parameters obtained from both the non-parametric and the parametric methods showed significant decrease in the LF/HF ratio values within an hour of application of the block in nine out of fourteen patients. In order to overcome the inability of these methods to deal with non-stationary, time-frequency analysis techniques were used to further analyse the HRV signals. The three time-frequency analysis methods used were the ContinuousWavelet Transform (CWT), theWigner-Ville Distribution (SPWVD) and the Empirical Mode Decomposition (EMD). The analysis of the parameters estimated from these three techniques on the clinical data showed that the CWT and the EMD techniques have
performed equivalently, meaning that both these methods have detected significant decrease in thirteen out of fourteen patients for the ratio values after the application of the,anaesthetic block. The presence of interference terms has caused the degradation in the
performance of the SPWVD method and due to this reason it was only able to detect significant changes in the LF/HF ratio values in ten of the fourteen patients. The results
suggest that due to anxiety and/or adrenaline present in the local anaesthetic mixture, the LF/HF ratio values showed a transient increase shortly after the application of the block. After this transient increase the ratio values decreased considerably and remained low as compared to the values before the application of the block. This decrease could represent the shift of the sympathovagal balance towards parasympathetic predominance and/or inhabitation of sympathetic activity due to local anaesthesia. The use of timefrequency
analysis such as EMD and CWT could provide useful information about the changes caused in the dynamics of the cardiovascular system when a local anaesthetic
drug is administered in a patient.

Item Type: Thesis (Doctoral)
Subjects: R Medicine > R Medicine (General)
R Medicine > RD Surgery
Divisions: School of Engineering & Mathematical Sciences
URI: http://openaccess.city.ac.uk/id/eprint/18392

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year

View more statistics