Analysis and Interpretation of HRV Data with Particular. Principal Investigator Dorn VA Medical Center (OEF OIF) and Dorn Research Institute

Transcription

1 Analysis and Interpretation of HRV Data with Particular Reference to the Coherence Ratio and Application to Data from Research on Combat Veterans with PTSD Break out Session, 44 th Annual Scientific Meeting of the AAPB March 15, 2013 JP Ginsberg, Ph.D. Licensed Clinical Psychologist/Neuropsychologist and Licensed Clinical Psychologist/Neuropsychologist and Principal Investigator Dorn VA Medical Center (OEF OIF) and Dorn Research Institute

2 If we knew what we were doing, it wouldn t be called research, would it? Albert Einstein

3 "If we take in our hand any volume, let us ask: Does it contain any abstract reasoning concerning quantity or number? No. Does it contain any experimental reasoning concerning matter of fact and existence? No. Commit it then to the flames, for it can contain nothing but sophistry and illusion. DAVID HUME ( ), An Inquiry Concerning Human Understanding (1748)

4 I t i t I m trying to think, but nothing happens. Curley Joe

5 Grateful Acknowledgement is also made to collaborators from the Shirley L. Buchanan Neuroscience Laboratory (Human Studies): Donnie A. Powell, Ph.D., Director (dec) Liz Hamel Melanie E. Berry, M.S., B.C.B., B.C.B.S** Martin Durkin, M.D. Edward D. Ayers, Ph.D. Louisa Burriss, Ph.D. Andrew Pringle

6

7 The present study suggests that vagal control of HR is critical for conditioning and demonstrates convincingly that there is a relationship between concomitant cardiac deceleration and the rate of EB conditioning in human subjects The present data fit into a larger pattern suggesting that peripheral parasympathetic control is important for a range of different unrelated somatomotor and cognitive activities. Associative Learning is influenced by HRV

8

9 Task Force of The European Society of Cardiology and The North American Society of Pacing and Electrophysiology. (1996). Guidelines for Heart rate variability: Standards of measurement, physiological interpretation, and clinical use. European Heart Journal 17,

10

11

12 (Period) `` `

13

14

15

16 Note: All subjects (112 males, 148 females) underwent 24 h ambulatory Holter ECG monitoring.

17

18

19 Zucker, TL, Samuelson, KW, Muench, F, Greenberg, MA, Gevirtz, RN. (2009). The effects of respiratory sinus arrhythmia biofeedback on heart rate variability and posttraumatic stress disordersymptoms: symptoms: a pilot study. Appl Psychophysiol Biofeedback 34(2): Future research might benefit from assessing additional HRV frequency variables (in addition to SDNN, a time domain one), which may clarify how the two ANS branches may discretely influence HR activation and physiologically define the hypothetical ti lptsd subtypes.

20 Heart rate data was analyzed using the standard deviation of sequential interbeat intervals averaged over 5 min epochs (SDANN). Table 2 compares the average SDANN index score for the 20 veterans diagnosed with PTSD and the average SDANN for the ten subjects in Normal control group. The difference was statistically significant suggesting that the veterans diagnosed with PTSD display lower HRV when compared to those without PTSD. This difference represents a large, clinically significant effect (d = 1.89).

21 Cohen et al (1997) Cohen et al (2000a) Cohen et al (2000b) Frustaci et al (2010) Ginsberg et al (2010) Hauschildt et al (2011) Hauschildt et al (2011) Keary et al (2009) Lakusic et al (2007) Mellman TA (2004) Mitani et al(2006) Nishith et al (2003) Sack et al (2008) Shaikh et al (2012) Slewa Younan et al (2012) Song et al (2011) Tan et al (2009) Tan et al (2011) Zucker et al (2009)

22 Effect size estimates for HRV parameters for samples of PTSD compared to control samples HRV N Data N PTSD N Control Hedge's g p Parameters Sets Participants Participants Effect size (95% CI) effect size PTSD severity (2.9 to 5.22) HR (0.58 to 2.00) SDNN ( 1.26 to 0.03) TP ( 1.04 to 0.38) LF ( 3.06to 0.36) HF ( 3.43 to 1.11) LF/HF (1.20 to 3.02)

23 Billman, G. The LF/HF ratio does not accurately measure cardiac sympatho vagal balance. Frontiers in Physiology, 20 February 2013 doi: /fphys Table 1. Examples of the effects of varying cardiac sympathetic and parasympathetic nerve activity on LF/HF. Figure 1. An illustration of the possible non linear effects of varying cardiac sympathetic and cardiac parasympathetic nerve activity on LF/HF.

24 Changes in cardiac sympathetic and parasympathetic nerve activity are not reciprocal (i.e., increases in cardiac parasympathetic nerve activity are not always accompanied idwith ihcorresponding reductions in cardiac sympathetic nerve activity ii and vice versa). Physiological challenges, disease states, and interventions can elicit either complex non linear reciprocal or parallel changes in either division of the autonomic nervous system. The relationship between the effects of cardiac sympathetic and parasympathetic nerve activity on heart rate variability are not linear; mathematical complications arise due to the non linear relationship between R R interval and heart rate;. Similar changes in heart rate elicit much greater variability in R R interval at lower than at higher heart rates; it is difficult to separate the changes in HRV that arise from direct action of cardiac autonomic nerves from those changes that result indirectly from neurally induced changes in average heart rate. Division by average R R intervalhas been proposed as a method for correcting the Division by average R R interval has been proposed as a method for correcting the non linearities that arise from these physiological and mathematical influences on calculation and interpretation of HRV power.

25 Calculation of Coherence from the HRV Power Spectrum NOTE: Coherence Ratio can also be expressed as % Coherence: (Peak Power/Total Power)*100 McCraty, R., Atkinson, M., Tomasino, D., & Bradley, R.T. (2009). The Coherent Heart: Heartbrain interactions, psychophysiological coherence, and the emergence of system wide order. Integral Review 5, 2 (10 106).

26

27 Pre-Post HRV Data From One Participant EmWave data, HeartMath Institute Report using exported IBI file

28 Ratio % Pre Post HRVB Coherence Ratio (or %) and Total Power (ms 2 /Hz) in PTSD+ and PTSD Combat Veterans Pre and Post Coherence HRVB PTSD+ PTSD+ PTSD Ratio (%) Total Power PTSD (51.0) Non parametric tests. LF was only power band to significant increase Pre Post (4.6) (6.8) Pre HRVB Post HRVB 0.23 (19.0) Pre HRVB 1142 Variable Pre-Post x PTSD Pre-Post PTSD % Coherence Total HRV Power n/s ns ns 961 Post HRVB

29 HRV Power Spectrum Case 1 Peak kpower at Hz = ms Coherence Ratio calculated as: (Peak LF Power Hz)/(Total LF Power Peak Power) ms 2 /Hz Coherence ratio = 0.35 Biograph Infiniti PPG raw data exported to ADI LbCh LabChart and deartifacted d Frequency (Hz) 0.4

30 HRV Power Spectrum Case 2 Peak Power at Hz = 53.5 ms 2 /Hz ms 2 /H Hz ms s Coherence ratio = Frequency (Hz)

31 Median Median = = ms 2 ms 2 Median Median =.023 = 51.1 ms

32 HRV Report Case 1 Case 2 Length of recording = s Length of recording = s Total number of beats = 1062 Total number of beats = Normal (99.5%); Normal (80.5%); 48 Ectopic Ectopic(0.1%); 4 Artifact (0.4%) (1.8%); 462Artifact (17.7%) Max, Min NN = , ms Max, Min NN = , ms Range = ms Range = ms Mean NN = ms Mean NN = ms Avg BPM = 70.6 Avg BPM =79.7 SDNN = 67.8 ms; RMSSD = 50.5 SDNN = 81.7 ms; RMSSD = 65.2 VLF (DC 0.04Hz) = ms² VLF (DC 0.04Hz) = ms² LF peak = ms² LF peak = 53.5 ms² LF ( Hz) 04 0 = ms² LF ( Hz) 0.15Hz) = ms² HF ( Hz) = ms² HF ( Hz) = ms² Total power = ms² Total power = ms² LF nu =80.2; HF nu = LF nu = 62.2; 2 HF nu = 37.8 LF/HF = 4.05 LF/HF = 1.6 Coherence ratio = 0.35 Coherence ratio = 0.02

33 Case 1 Case 2 SDNN Total Power (ms 2 ) VLF Power (ms 2 ) LF Power (ms 2 ) HF Power (ms 2 ) LF/HF LF Peak Power (ms 2 ) Coherence Ratio

34 Pearson r (n = 33) LF Peak (ms 2 ) HRV and Coherence in Combat Veterans with PTSD SDNN (ms) Ttl Total VLF LF HF Power Power Power Power (ms 2 ) (ms 2 ) (ms 2 ) (ms 2 ) LF/HF Correlation Sig. (1 t) Coherence Ratio Correlation Sig. (1 t ) LF Peak does but Coherence does not associate with standard HRV frequency variables; however Coherence associates weakly withlf/hf which is known to be non linear linear, suggesting that there is a non linear aspect to Coherence as well

35 HRV and PTSD in Combat Veterans Pearson r (n = 33) CAPS SDNN (ms) Correlation.415 Sig. (1 tailed).008 Total Power (ms 2 ) Correlation.401 Sig. (1 tailed).010 LF Band Power (ms 2 ) Correlation.263 Sig. (1 tailed).069 HF Band Power (ms 2 ) Correlation.597 Sig. (1 tailed).000 LF/HF ratio Correlation.385 Sig. (1 tailed).013 LF Peak Power (ms 2 ) Correlation.285 Sig. (1 tailed) tild).054 Coherence Ratio Correlation.370 Sig. (1 tailed).017

36 HRV and Cognition in Combat Vets with PTSD Pearson r (n = 33) Digit Span Backward SDNN Correlation.435 Sig. (1 tailed).006 Total Power (ms 2 ) Correlation.513 Sig. (1 tailed).001 LF Power (ms 2 ) Correlation.536 Sig. (1 tailed).001 HF Power (ms 2 ) Correlation.488 Sig. (1 tailed).002 LF/HF Correlation.021 Sig. (1 tailed).454 LF Peak Power Correlation.519 (ms 2 ) Sig. (1 tailed) ld).001 Coherence Ratio Correlation.195 Sig (1 tailed) 139

37 Coherence Ratio and HRV in Combat Veterans with PTSD n = 33 LF Peak Power ms 2 r rho Coherence Correlation Ratio Sig. (1 tailed) Coherence has a significant correlation withlf Peak Power Coherence has a significant correlation with LF Peak Power non parametrically, but not parametrically, again suggesting that Coherence is distributed as a non linear function

38 a Non linearity of Coherence Ratio with LF Power c b (ms 2 )

39 Non linearity of Coherence Ratio with LF Power (cont d) Case SDNN Total VLF LF HF LF Peak Coherence Power Power Power Power Power a b c