Changes in finger temperature and blood flow in response to different frequencies of transcutaneous electroacupuncture at LI4 (hegu).
Interim analysis and ‘real life’ methodological issues: many factors, missing data and a multiplicity of measures
© David Mayor (1), Tony Steffert (2) and Ronakben Bhavsar (3)
1. School of Health and Social Work, University of Hertfordshire;
2. Music Computing Lab, Open University;
3. School of Computer Science, University of Hertfordshire
Background. Finger blood flow and temperature are often used as indices of autonomic (sympathetic) function. Transcutaneous electroacupuncture (TEA) is an increasingly used variant of electroacupuncture (EA). This is the fifth in a series of conference posters from a study investigating the effects of EA and TEA on heart/pulse rate variability (HRV/PRV), the electroencephalograph (EEG), and now blood flow.
Objectives. To assess how treatment factors – particularly TEA stimulation frequency (Hz) – contribute to changes in local blood flow, skin temperature and pulse transit time (PTT), and to explore associations between the outcome measures used.
Methods. In this pilot study, participants (N=17) each attended for a single session consisting of 10 consecutive 5-minute ‘slots’. In the second, fifth and eighth slots, TEA was applied bilaterally at LI4 (hegu) at three different frequencies (2.5 Hz, 10 Hz and 80 Hz), in counterbalanced order (all with 256 ?s pulse duration and at a ‘strong but comfortable’ intensity). Using finger photoplethysmography, with a thermistor on the same finger, the blood volume pulse (BVP) and temperature were monitored throughout. Electrocardiograph (ECG) signals were collected from wrist electrodes. Data on blood flow amplitude, pulse rate and transit time were derived from the BVP and ECG following standard procedures, including artefact processing. Mood was assessed at various time points using numerical rating scales and a short form of the Profile of Mood States questionnaire. Sample size estimates for the different measures and experimental factors were conducted as a basis for continuation of the study.
Results. TEA at 2.5 Hz consistently but not significantly resulted in greater fingertip blood flow, and 80 Hz in longer PTT, than at the other two stimulation frequencies (frequency effects on temperature were inconsistent, small and not significant). For most participants, the association between skin blood flow and temperature was significant and positive, with both tending to peak together shortly after TEA. However, over the session both tended to decrease, although several measures of mood and PRV improved. In contrast, overall session change in PTT was small, but group medians peaked in slots 5 and 8.
Conclusions. 2.5 Hz TEA is likely to enhance local skin blood flow. However, even if this question is apparently simple, its investigation turned out to be complicated. As we found before when studying HRV, stimulation frequency may be a less important factor than others such as the presence of muscle twitch or participants’ prior experience of related treatments. Thus in future our analysis will use multilevel modelling to take account of multiple factors and their interactions. In addition, we noted that although PTT may be helpful in assessing short-term changes in acupuncture-related research, this will only be so if high sampling rates are used. Further recruitment of participants is planned to consolidate our findings.
From the University of Hertfordshire, Professor Tim Watson (School of Health and Social Work) and Dr Na Helian (School of Computer Science) for supervisory advice; Dr Neil Spencer (Director, Statistical Services and Consultancy Unit) and Dr Alla Mashanova (Health and Human Sciences Research Institute) for their gentle statistical suggestions (which there was not always time to follow). Also: Jeroen Manten (Mind Media BV), for assistance with endless BioTrace and NeXus-10 queries; Zusana Kovacova (Herts EEG Biofeedback) for rescuing the study with a replacement temperature sensor; Dr Aleksander Valjamae (Decision, Emotion and Perception lab, Department of Behavioural Sciences and Learning, Linkoping University, Sweden) for assistance with Matlab pulse transit time processing, and of course our spouses/partners, and the participants in our study, without whom none of this would have been possible. Any errors in this report remain, of course, our own responsibility.
If you have any problems with this web site, please contact David Mayor at +44 (0) 1707 320782 or