October 1-3, 2004

Radisson Miyako Hotel

San Francisco, CA

A Retrospective evaluation of Acupuncture outcomes as a function of Heart rate Variability




It has been postulated that acupuncture affects the cardiovascular system through the autonomic nervous system

[1][2][3][4][5][6].  Using power spectral analysis, the ratio of low frequency to high frequency components of Heart Rate Variability (HRV) can be calculated.  This ratio reflects relative input from sympathetic and parasympathetic aspects of the autonomic nervous system (ANS).  Heart Rate Variability monitoring is a non-invasive method, which can be easily used in a clinical setting. In this study I evaluated Heart Rate Variability and some other physiological data retrospectively to see if any patterns developed that could shed light on the autonomic processes occurring during acupuncture treatment.




Materials and Methods:


 Patients coming to the clinic were evaluated and treated according to TCM. Patients were monitored using a “J and J Engineering Biofeedback Monitor.”  They were monitored for heart rate, standard deviation of the heart rate, heart rate variability (in three different spectra), skin resistance and peripheral temperature. There were a total of 17 patients who qualified for this study.  The patients were all referred or self-referred with various allopathic medical diagnoses.  They had to complete at least three biomonitoring sessions and acupuncture treatment sessions, and have a discernible outcome; i.e. they could not be “lost to follow-up,” and had to be able to give a coherent assessment of their treatment.  Written consent was obtained.  Monitors were placed after patient was supine. Needles were placed according to the principles of TCM diagnosis and treatment.  Recording was begun after needles were placed.  Recording was terminated and then needles were removed.  Data was collected for 30 minutes. Retrospectively their data was analyzed, after a series of acupuncture sessions had been completed. Data was analyzed by pasting raw data onto a spreadsheet, charted and a moving average taken.



A total of 17 patients were included in this study. There were 13 patients who demonstrated significant improvement (group 1, “Responders”). (Group 2, “Non-Responders”) had one patient with equivocal improvement and 3 who had no improvement whatsoever.  Significant improvement in this study meant a diminution in their chief complaint so that the patient reported the treatment a success. In this group there were patients with conditions that varied over a greater period of time (i.e. migraine.)  Obviously, for the treatment, or treatment series to be deemed a success in these cases, the symptoms had to decrease over the treatment course.  In some cases it meant disappearance of their symptoms altogether. Group 2 patients had no perceptible improvement or questionable improvement in spite of good follow through and attendance of appointments.  Several of the variables were charted, including skin resistance, the standard deviation of the intrabeat interval, LFR, HFR, and finally the LFR/ HFR ratio.  Other data were collected but not analyzed. (peripheral temperature VLF etc…)







Though other parameters changed during the course of acupuncture treatment, (the peripheral temperature, skin resistance, standard deviation of the heart rate variability) no other parameter showed a consistent correlation with successful treatment. In this pilot study it appears that calculating the LFR/HFR may show a decline in patients who respond positively to the acupuncture treatment setting. What the chart shows is the percentage of times the LFR/HFR ratio decreased by at least 25% for each patient. (For example: Patient 6 had four treatments.  During 3 out of the 4 treatments, he showed a decrease in his LFR/HFR ratio.  His score was a –75% meaning that 75% of the time he showed a decrease in his LFR/HFR ratio.) On average, the patients who responded to acupuncture showed a DECREASE in their LFR/HFR ratio 62% of the time.  The patients who did not respond, or where it was equivocal showed an average INCREASE in their LFR/HFR ratio 19% of the time.

A decline in the LFR to HFR ratio during an acupuncture treatment would indicate a relative decrease in the sympathetic nervous system inputs to the S.A. Node, and a relative increase in the parasympathetic input.






Heart rate variability has been established as a non-invasive tool that reveals the relative inputs of the parasympathetic and sympathetic nervous systems. The autonomic nervous system most likely plays a role in acupuncture’s effectiveness in some situations.   To be able to evaluate that effect, noninvasively and in real time, would be of tremendous interest clinically, and of even greater interest scientifically.  In this study, though patients were monitored during acupuncture treatment, there is no way to know if the changes in their LFR/HFR ratio were due to the acupuncture, since no controls were done. But looking at prior studies that have been done, it would not be surprising if a decrease in LFR/HFR ratio was a result of acupuncture. What we can say is that there may be some significance in the difference in subjects’ responses to the acupuncture clinic setting. 

If this phenomenon is a reproducible in a controlled, prospective study, it could shed light on the mechanisms involved in acupuncture’s efficacy.

Why is it of interest that acupuncture may activate the Autonomic Nervous System (ANS)?  If you increase input[i] from the parasympathetic nervous system, you have the result of increasing anti-oxidants, increasing co-enzyme Q10, increasing endorphins, increasing nitric oxide, increasing dopamine.  These factors all have a positive influence on health and well being. But, one of the absolutely crucial factors could be the decrease in stress hormones, most important, cortisol.  Cortisol has been determined to be detrimental to a person’s immune defenses, pain tolerance, and mood (depression.)  Furthermore, an increase in stress hormones leads to an inability to withstand insults to the brain (hypoxia, etc.) and can lead to accelerated aging of the hippocampus and the brain glia, which may be implicated in some Alzheimer’s sufferers.  For these reasons, an ability to modulate the sympathetic/parasympathetic inputs to the CNS/cardiovascular system could be an extremely important piece of the puzzle of acupuncture’s effectiveness.


What studies have been done looking at the sympathetic nervous system and acupuncture?  There have been a few recently.

1)      There was a study from China[ii] looking at acupuncture in coronary artery disease patients showing that the Low frequency components (sympathetic nervous input) changed during acupuncture, but the HF component (parasympathetic input) did not.

2)     There was a study from Sweden[iii], also using heart rate variability that showed an increase in parasympathetic and sympathetic inputs after Hoku (dorsal thenar muscle) stimulation. There was an increase in the parasympathetic component only with Lung 1 stimulation in the ear (cavum concha) for the stimulation period of 25 minutes and for 60 minutes afterwards.

3)     Study from Japan[iv], published in Anesthesiology, showing that preoperative intradermal acupuncture reduces postoperative pain, nausea and vomiting, analgesic requirement, and sympathoadrenal responses (as measured by cortisol and epinephrine concentrations.)

4)     In Germany[v] a group looked at the effects of needle acupuncture on ANS function in patients with minor depression or anxiety disorder.  Compared to placebo, the Acupuncture group showed a decrease in LF component and an increase in HF component, i.e. an overall decrease in LF/HFR ratio.  These results are similar to the results of my study (though I had no placebo group.)

5)     We know from a Swedish[vi] study that uterine artery blood flow impedance is decreased using electroacupuncture twice weekly for four weeks.   The flow was increased shortly after the eighth acupuncture treatment and for two weeks following.  It is suggested that this is due to central inhibition of the sympathetic activity.

6)     There is a related study, though not involving acupuncture, that is from Milan, Italy[vii] that shows that markers of autonomic regulation of the SA node (increase in sympathetic input) correlated significantly with cortisol levels.

7)     A study from China[viii] looked at the effect of magnetopuncture on sympathetic and parasympathetic nerve activities in healthy drivers, while driving, using HRV analysis, and found that the LF component (sympathetic) decreased significantly and HF component (parasympathetic) increased significantly.

8) In Sweden[ix] a group looked at using auricular acupuncture as a treatment in a prison psychiatric unit.  They found that cortisol levels were higher in the untreated group.  Inmates who were treated at least 25 times were prescribed fewer psycholeptic drugs than the untreated group. 

In summary, there is evidence that acupuncture may result in improved immunity, higher pain thresholds, improved mood, due to modulation in the ANS, and a lower cortisol level.  The fact that there are some subjects who do not seem to manifest this modulation, may be a clue of how to treat, or not treat with acupuncture, in a more advised way.  When I started looking at heartrate variability, I was not expecting to see this result.  I expected different responses in For example, one could analyze a treatment in real time and perhaps by altering the length of time or method of stimulation or even placement of needles to optimize the modulation of the ANS, this could be of enormous use clinically.  On the other hand, if a patient does not demonstrate a modulation in his ANS after a few treatments, one could suggest alternative types of therapies. In short, HRV may prove to be an adjunct to acupuncture treatment by virtue of its use as an efficient, non-invasive monitoring system. If prospective studies bear out the validity of this study’s findings HRV monitoring could streamline the acupuncture treatment process and aid in exploring acupuncture’s usefulness.




Kristen Sparrow,  M.D.

Private Practice

Sonoma and San Francisco,  California



[1]  M. Ernst, M.Lee Experimental Neurology 94, 1-10 1986


[2] S. Anderson, T.Lundeberg, Medical Hypothesis 45 (1995):3 p. 273-81

[3] S.Knardahl  Pain 75 (1998) 19-25


[4] D. Thomas, S. Collins Clinical rheumatology, 1992, 11 N01, 55-59


[5] E.Stener-Victorin et al Human Reproduction vol.11 no.6 pp 1314-1317, 1996

[6] M. Hammar et al Journal of Urology,  Vol161, 853-856 March 1999



[i] Singh RB, Kartik C, Otsuka K, Pella D,  Pella J.  Brain-heart connection and the risk of heart attack  Biomed Pharmacother. 2002; 56 Suppl 2: 257s-265s


[ii] Shi,X, Wang ZP, Liu KX, (Zhongguo Zhong Xi Yi Jie He Za Zhi) 1995 Sept; 15(9):536-8  [Effect of acupuncture on heart rate variability in coronary heart disease patients]


[iii] Haker, E, Egekvist.H, Bjerring P. Effect of sensory stimulation (acupuncture) on sympathetic and parasympathetic activities in healthy subjects. Journal of Autonomic Nervous System.2000 Feb 14; 79(1): 52-9


[iv] Kotani N, Hashimoto H, Sato Y, Sessler DI, Yoshioka H, Kitayama M, Yasui T, Matsuki A Preoperative intradermal acupuncture reduces postoperative pain, nausea and vomiting,  analgesic requirement, and sympathoadrenal responses.  Anesthesiology. 2001 Aug; 95(2): 349-56


[v] Agelink MW, Sanner D, Eich H, Pach J, Bertling R, Lemmer W, Klieser E, Lehmann E.  Does acupuncture influence the cardiac autonomic nervous system in patents with minor depression or onxiety disorders?  Fortschr Neurol Psychiatr. 2003 Mar; 71(3): 141-9



[vi] Elisabet Stener-Victorin, Urban Waldenstrom, Sven A Anderson andMatts Wikland Reduction of Blood Flow impedance in the uterine arteries of infertile women with electro-acupuncture Human Reproduction vol.11 no.6 pp 1314-1317, 1996


[vii] Lucini D, Norbiato G, Clerici M, Pagani M.  Hemodynamic and autonomic adjustments to real life stress conditions in humans  Hypertension. 2002 Jan; 39(1): 184-8


[viii] Li Z, Jiao K, Chen M, Wang C.  Effect of magnitopuncture on sympathetic and parasympathetic nerve activities in healthy drivers—assessment by power spectrum analysis of heart rate variability.  Eur J Appl Physiol. 2003 Jan; 88(4-5) 404-10. Epub 2002 Nov 15


[ix] Berman AH, Lundberg U. Auricular acupuncture in prison psychiatric units: a pilot study.   Acta Psychiatr Scand Suppl. 2002; (412): 152-7