- Email: [email protected]
The effects of therapeutic ultrasound on heart rate variability: A placebo controlled trial
Ultrasound in Med. & Biol., Vol. 31, No. 5, pp. 643– 648, 2005 Copyright © 2005 World Federation for Ultrasound in Medicine & Biology Printed in the USA. All rights reserved 0301-5629/05/$–see front matter
doi:10.1016/j.ultrasmedbio.2005.01.010
● Original Contribution THE EFFECTS OF THERAPEUTIC ULTRASOUND ON HEART RATE VARIABILITY: A PLACEBO CONTROLLED TRIAL VEDAT NACITARHAN,* HASAN ELDEN,* MUSTAFA KıSA,* ECE KAPTANOG˘ GLU,* and SEDAT NACITARHAN† *Cumhuriyet University, Medicine Faculty, Department of Physical Medicine and Rehabilitation, Sivas, Turkey; and † State Hospital, Department of Physical Medicine and Rehabilitation, Diyarbakır, Turkey (Received 15 November 2004; revised 24 January 2005; in final form 27 January 2005)
Abstract—The effect of therapeutic ultrasound (US) on nervous system is controversial and the effect on autonomic nervous system is not clear. Therefore, the present placebo-controlled trial was planned to investigate the effects of therapeutic US application on right-side stellate ganglion, by using analysis of heart rate variability (HRV). A total 12 healthy volunteers were included in the study. RR intervals were recorded for 5 min before and after the US application, in supine and sitting positions. All procedures were repeated in all participants with sham US one week later. The heart rate (HR) was obtained by time-domain analysis and low frequency (LF) power (%), high frequency (HF) power (%) and LF/HF ratio values were obtained by frequency-domain (power spectral density) analysis. After the US application, there was a decrease in the HR (p ⴝ 0.002) and the HF power (%) component (p ⴝ 0.015) in supine position and a decrease in HR (p ⴝ 0.002) and LF/HF ratio (p ⴝ 0.028) in sitting position. There was no significant difference after the sham US application. In conclusion, we observed that therapeutic US application on stellate ganglion causes alterations on HRV parameters. (E-mail: [email protected]) © 2005 World Federation for Ultrasound in Medicine & Biology. Key Words: Therapeutic ultrasound, Stellate ganglion, Heart rate variability.
action potential was reduced or completely lost in the sciatic nerves. Herrick (1953), in his study, suggested that US causes a selective heating and blockage after a critical level (about 45° to 46°C) for A-fibers in the nervous tissue. As the same result was obtained without US when the temperature was raised by an alternative method, Herrick (1953) concluded that US blocks the action potential via a thermal mechanism. Following studies, however, revealed contradictory reports about the effect of US on nerve conduction. Currier et al. (1978) and Halle et al. (1981) investigated the effect of US on lateral cutaneous branch of the radial nerve. They reported that the latencies were decreased, indicating increased speed of conduction, as the subcutaneous temperatures increased. Hong et al. (1988), in their study, searching, the effect of US therapy in rats with experimental compression neuropathy, reported that nerve recovery was facilitated by nonthermal US dose, whereas a thermal dose had the opposite effect. In another study, Hong (1991) investigated the effect of US on nerve conduction in patients with polyneuropathy and concluded that ultrasonic therapy with therapeutic dosage may cause a reversible conduction block. Moore et
INTRODUCTION Ultrasound (US) is defined as acoustic vibration with frequencies above the audible range. Since its introduction as a therapy over six decades ago, therapeutic US has developed into the most widely available and frequently used modality in practice of physical therapy (Robertson and Baker 2001; Warden and McMeeken 2002). In the literature, US is reported to have two main groups of effects, thermal and nonthermal (Baker et al. 2001; Demmink et al. 2003; van der Windt et al. 1999). The high absorption coefficients of large protein molecules mean that collagenous tissues may be heated preferentially, also including the nervous tissues (ter Haar 1999). Rosenberger (1950) found the temperature increase produced by US in sciatic nerves to be about twice that produced in neighboring tissues. Anderson et al. (1951) performed studies with US on spinal cords of rats and dogs and sciatic nerves of dogs and found that the spinal cords were irreversibly paralyzed and that the Address correspondence to: Vedat Nacitarhan, P.K. 743, 58140, Sivas, Turkey. E-mail: [email protected] 643
644
Ultrasound in Medicine and Biology
al. (2000) investigated the biophysical effects of US. They concluded that alterations in nerve latencies from ultrasound on healthy nerves appeared to be related to temperature changes induced by ultrasound’s thermal effects and not by nonthermal or mechanical effects. As it can be seen from the studies mentioned above, the results are contradictory and there is no consensus on the underlying mechanism. The heart is innervated through the efferent sympathetic and parasympathetic nerve fibers of the autonomic nervous system (ANS) that act on the intrinsic sinoatrial node of the heart (Wang et al. 2000). Vagal stimulation allows for beat-to-beat control of the heart rate (HR), whereas sympathetic stimulation has a more gradual effect on the HR (Pumprla et al. 2002; Wang et al. 2000). Physiological conditions illustrate the reciprocal relationship that exists between the two divisions of the ANS (Task Force of the European Society of Cardiology 1996; Wang et al. 2000). Heart rate variability (HRV) includes time and frequency domain methods (Task Force of the European Society of Cardiology 1996). Power spectral (frequency domain) analysis of HRV has been shown that harmonic oscillations in heart rate are concentrated into at least two distinct bands. That referred to as the low frequency band (LF: 0.04 to 0.15 Hz) is affected by the oscillatory rhythm of the baroreceptor system and is mediated by sympathetic activity, with some influence from vagal activity. In the other, the high frequency band (HF: 0.15 to 0.40 Hz), respiration is the primary rhythmic stimulus and it is mediated by changing levels of parasympathetic tone (Task Force of the European Society of Cardiology 1996). LF/HF ratio is interpreted to be a marker of sympathovagal balance. Power spectral analysis of HRV has shown to be a reliable noninvasive test for quantitative assessment of cardiovascular autonomic regulatory responses, providing a window reflecting the interaction of sympathetic and parasympathetic tone (Task Force of the European Society of Cardiology 1996). Cardiac sympathetic fibers originate from the stellate ganglia in conjunction with the superior and middle cervical ganglia and the first four or five thoracic sympathetic ganglia (Introna et al. 1995; Koyama et al. 2002). Previous reports revealed that block of right-side stellate ganglion decreased the sinus rate and log-transformations of the components of HRV in human (Koyama et al. 2002). A decrease in LF and HF power components after the right stellate ganglion blockage was found by Fujiki et al. (1999). Stellate ganglion blockage has beneficial effects in various clinical conditions such as complex regional pain syndromes, neuropathic painful disorders and peripheral vascular diseases. This is usually accomplished by pharmacological or surgical approaches (Buckley 2001).
Volume 31, Number 5, 2005
No study has, to our knowledge, reported the effect of therapeutic US on ANS via stellate ganglion. Based on these facts, we planned this placebo-controlled study to investigate the effects of therapeutic US by analysis of HRV, before and after therapeutic US application on right-side stellate ganglion. MATERIALS AND METHODS The study was designed as a placebo controlled trial. Subjects. This study was performed on 12 healthy volunteer subjects (four female, eight male, age range 23 to 28 y, mean 24.9). All subjects provided written informed consent and the study was approved by the ethical committee of Cumhuriyet University Medicine Faculty in Sivas, Turkey. Study protocol. The study was conducted in a silent room, at a temperature of 22 to 24°C, between 10:00 to 11:00 A.M. Subjects were abstaining from exercise and tea, caffeine and smoking intake for minimum of 1 h before study. Subjects were asked to rest for least 15 min in supine position after the electrodes were applied. Before US application, ECG recordings were made for 5 min in supine and sitting positions. Then the patients were asked to rest in supine position again. Sonogel was applied on the skin over right supraclavicular area, the sound head applied and oriented to right stellate ganglion through right supraclavicular region. Then the US apparatus (Sonoplus 992, Enraf-Nonius B.V., The Netherlands) was switched on in US group. The sound head was applied with circular movements. The ultrasound frequency was 1 MHz continuous and the intensity was 1 W/cm2. The radiation area was 5 cm2 and the application time was 5 min per subject. After the application, ECG recordings were repeated for 5 min again, both in supine and sitting positions. One week after the measurements, the study subjects were used as the placebo group. The sham US was applied without switching on the apparatus. All of the tests were performed in the same order and in the same manner with the US group. The study subjects were unaware of the type of the application (US or sham), while the operator was aware. However, the operator did not participate in the data analysis. ECG recording. Recordings were performed by using standard bipolar limb lead-I (negative electrode on right arm, positive electrode on left arm, ground electrode on left leg) with an ECG monitor (patient monitor, Ivy Biomedical System Inc, CT, USA). The analog data output of the ECG monitor was sampled at a sampling frequency of 1000 Hz, using an analog-to-digital converter. The sampled data were stored to disk for postprocessing. Detection of R-wave was done with peak
Ultrasound on heart rate variability ● V. NACITARHAN et al.
645
detection algorithm. The detected R waves were visually confirmed. RR interval was extracted in 1 ms resolution and saved consecutively. ECG recording was realized in each subject in four instances by 5 min, as pre- and postapplication supine and sitting values for analysis. Same procedure was performed for the placebo group. HRV analysis. All results were analyzed with HRV analysis software 1.1 for windows, developed by The Biomedical Signal Analysis Group, Department of Applied Physics, University of Kuopio, Finland (the software is distributed free of charge upon request in http:// venda.uku.fi/research/biosignal). The frequency-domain analysis of HRV was performed by an autoregressive model transform of the RR intervals. The time-domain measure (HR) of HRV and frequency-domain (power spectral analysis) measures [LF power (%), HF power (%), and LF/HF ratio] of HRV were calculated. Statistical analysis. The data were presented as mean ⫾ SD. As the data did not show normal distribution, nonparametric tests of comparison were used. Comparisons of the data obtained at supine and sitting positions were made by Wilcoxon (paired) signed-ranks test. Mann-Whitney test was used to compare the data among the groups. A value of p ⬍ 0.05 was considered to be significant. RESULTS HR and power spectral analysis parameters [LF power (%), HF power (%) and LF/HF ratio] of HRV of all subjects before and after application are shown in Figure 1 and Figure 2.
Fig. 2. The HR and HRV parameters of all subjects in sitting position pre- and postapplication
Comparison of pre- and postapplication data. Preand postapplication values were not significantly different in placebo group, either in supine or sitting position. In the US group, there were significant decreases in the HR (p ⫽ 0.002) and HF power (%) (p ⫽ 0.015) in the supine position and in HR (p ⫽ 0.002) and LF/HF ratio (p ⫽ 0.028) in the sitting position after the application (Table 1). The results of the comparison of pre- and postapplication data show that the US application on stellate ganglion decreased the sympathetic activity. Comparison of the data obtained in supine and sitting positions. There were significant differences both in the placebo and US groups between sitting and supine positions in HR, HF power (%) and LF/HF ratios (p ⫽ 0.002, p ⫽ 0.002, p ⫽ 0.003 in the placebo group, p ⫽ 0.002, p ⫽ 0.012, p ⫽ 0.006 in the US group, respectively) before the application. Significant differences in HR (p ⫽ 0.002), HF power (%) (p ⫽ 0.002) and LF/HF ratio (p ⫽ 0.023) were observed between the supine and sitting positions in the placebo group after application. However, there was a significant difference only in HR (p ⫽ 0.003) in the US group after the application between the sitting and supine positions (Table 2).
Fig. 1. The HR and HRV parameters of all subjects in supine position pre- and postapplication.
Comparison of the measured values in placebo and US groups. There was no difference between the groups in supine and sitting positions before the application. After the application, there were significant differences in HR and HF power (%) values between the groups in supine position (p ⫽ 0.043, p ⫽ 0.038, respectively) (Table 3).
646
Ultrasound in Medicine and Biology
Volume 31, Number 5, 2005
Table 1. Comparison of pre- and postapplication data Placebo Group
Supine
Sitting
HR LF HF LF/HF HR LF HF LF/HF
US Group
Before app†
After app†
p
Before app†
After app†
p
75 ⫾ 5 39 ⫾ 20 37 ⫾ 15 1⫾1 82 ⫾ 5 46 ⫾ 25 17 ⫾ 12 4⫾3
75 ⫾ 6 45 ⫾ 24 37 ⫾ 14 2⫾2 81 ⫾ 6 55 ⫾ 20 19 ⫾ 13 4⫾3
0.695 0.638 0.875 0.530 0.255 0.117 0.534 0.530
76 ⫾ 5 39 ⫾ 18 40 ⫾ 18 2⫾2 84 ⫾ 5 46 ⫾ 24 20 ⫾ 15 4⫾3
71 ⫾ 5 39 ⫾ 16 25 ⫾ 9 2⫾2 78 ⫾ 5 45 ⫾ 23 23 ⫾ 12 3⫾3
0.002* 0.937 0.015* 0.158 0.002* 0.388 0.182 0.028*
* Significant difference; † Application
DISCUSSION It is possible to find many studies in the literature investigating the effects of therapeutic US, a widely used agent in physical therapy. These studies are mostly on acute and chronic musculoskeletal disorders (Demmink et al. 2003; Robertson and Baker 2001; Speed 2001; van der Windt et al. 1999; Warden and McMeeken 2002). Goodman (1971) has reported therapeutic US to be effective on stellate ganglion, which is a unique study in the literature in this regard. There are some studies on peripheral nerves reporting that US blocks the neural transmission (Hong 1991; Portwood et al. 1987) or, on the contrary, that it increases the transmission velocity owing to its thermal effect (Kramer 1985; Moore et al. 2000). So, it is obvious that the effects of US on nervous system are controversial. Information on sympathovagal balance may be obtained by evaluating the cyclic variations in the heart rate via HRV analysis. RR interval is dependent on the sympathetic and parasympathetic nerves. The sympathovagal response of ANS, affected by pharmacological agent, may be traced by analysis of HRV. As the stellate ganglion gives sympathetic fibers to the heart, it is possible
to determine this effect on HRV analysis when stellate ganglion was blocked (Task Force of the European Society of Cardiology 1996). On the other hand, there is a lateralization in stellate ganglion functions. When the right stellate ganglion was blocked by a pharmacological agent, a variation in the HR and HRV parameters was determined (Koyama et al. 2002). In another study (Fujiki et al. 1999), no variation in HRV parameters was observed when the stellate ganglion blockage was performed on left side, while the right side blockage led to the variations in HRV parameters. In another study (Lobato et al. 2000) evaluating the functions of left ventricle after the stellate ganglion blockage, neither right nor left stellate ganglion blockage was found to be effective on left ventricular function. Either central (Vito et al. 2002) or spinal (Introna et al. 1995) way of sympathetic blockage have demonstrated a decrease in LF power (%) and HF power (%) components and it is reported that the power spectral density analysis of HRV provides a quantitative assessment of sympathovagal balance. In our study, there was an increase in HR and LF/HF ratio and a decrease in HF power (%) in supine
Table 2. Comparison of the data obtained in supine and sitting positions Before application
US
Placebo
HR LF HF LF/HF HR LF HF LF/HF
* Significant difference.
After application
Supine
Sitting
p
Supine
Sitting
p
76 ⫾ 5 39 ⫾ 18 40 ⫾ 18 2⫾2 75 ⫾ 5 39 ⫾ 20 37 ⫾ 15 1⫾1
84 ⫾ 5 46 ⫾ 24 20 ⫾ 15 4⫾3 82 ⫾ 5 46 ⫾ 25 17 ⫾ 12 4⫾3
0.002* 0.433 0.012* 0.006* 0.002* 0.388 0.002* 0.003*
71 ⫾ 5 39 ⫾ 16 25 ⫾ 9 2⫾2 75 ⫾ 6 45 ⫾ 24 37 ⫾ 14 2⫾2
78 ⫾ 5 45 ⫾ 23 23 ⫾ 12 3⫾3 81 ⫾ 6 55 ⫾ 20 19 ⫾ 13 4⫾3
0.003* 0.272 0.433 0.117 0.002* 0.071 0.002* 0.023*
Ultrasound on heart rate variability ● V. NACITARHAN et al.
647
Table 3. Comparison of the measured values in placebo and US groups Before application
Supine
Sitting
HR LF HF LF/HF HR LF HF LF/HF
After application
Placebo
US
p
Placebo
US
p
75 ⫾ 5 39 ⫾ 20 37 ⫾ 15 1⫾1 82 ⫾ 5 46 ⫾ 25 17 ⫾ 12 4⫾3
76 ⫾ 5 39 ⫾ 18 40 ⫾ 18 2⫾2 84 ⫾ 5 46 ⫾ 24 20 ⫾ 15 4⫾3
0.795 0.751 0.603 0.795 0.488 1.000 0.644 1.000
75 ⫾ 6 45 ⫾ 24 37 ⫾ 14 2⫾2 81 ⫾ 6 55 ⫾ 20 19 ⫾ 13 4⫾3
71 ⫾ 5 39 ⫾ 16 25 ⫾ 9 2⫾2 78 ⫾ 5 45 ⫾ 23 23 ⫾ 12 3⫾3
0.043* 0.488 0.038* 0.564 0.149 0.248 0.312 0.119
* Significant difference.
position compared with the sitting position, in both the US and placebo group before the application. This situation is related with the compensatory increase of sympathetic activation to prevent hypotension and peripheral pooling of blood in the lower extremities in the sitting-up position. The decrease in HF power (%) is in accordance with the decrease in parasympathetic tonus as a part of the adaptation. The increase in LF/HF ratio is the expression of this variation on behalf of the sympathetic component in autonomic nervous system, which is the normally expected physiological response in a healthy individual. After the US application, while the increase in HR was still significant, the increase in LF/HF ratio and the decrease in HF power (%) component was no longer significant in sitting position. On the other hand, after the application, the increase in HR and LF/HF ratio and the decrease in HF power (%) component were still significant in the placebo group. Although the increase in the HR was statistically significant after the US application, the basal HRs before and after the application were different. The HR, which was 76 ⫾ 5 in supine and 84 ⫾ 5 in sitting positions, decreased to 71 ⫾ 5 and 78 ⫾ 5 in supine and sitting positions, respectively, after the therapeutic US application. As can be traced from the results, either in supine or sitting positions, HR is found to be decreased compared with the preapplication values. Unfortunately, we could not compare the above observation, since no similar study was found in the literature on the subject. Both the decrease in HR and the absence of the increase in LF/HF ratio and the decrease in HF power (%) component in US group, different from that of the placebo group after the application, can be interpreted as “therapeutic US application decreases the effect of the sympathetic component of autonomic nervous system”. Koyama et al. (2002) have evaluated the effect of the autonomic nervous system on the heart by pharmacological blockage of the right stellate ganglion. They evaluated the sympathovagal tonus in supine position
and during head-up tilt test (HUTT) in eight patients with chronic pain, before and after right stellate ganglion blockage. While they have obtained normal physiological responses before blockage, HR and LF/HF ratio and HF power (%) components during HUTT did not significantly alter after the right stellate ganglion blockage procedure. So, they suggested that right stellate ganglion blockage suppresses cardiac sympathetic function. Fujiki et al. (1999) found a decrease in LF and HF power components after the right stellate ganglion blockage by 1% mepivacaine and concluded that the right-sided blockage causes a decrease in sympathetic and parasympathetic effect on the sinus node. The major difference between these studies and our study is the method of ganglion blockage, where we used a noninvasive method instead of pharmacological agents. The change in HRV component in our study is in concordance with these previous studies and either pharmacological blockage or US application (as in our study) seems to depress the sympathetic component. We take into consideration also the pressure/massage effects of the US probe in our study. The therapeutic effect of US was separated from a pressure effect on the stellate ganglion by using the same study subject as a placebo group by repeating the study one-week apart. In the placebo group, the increase in HR and LF/HF ratio and the decrease in HF power (%) component before the application was also obtained following the application, whereas it was different in US group. This outcome ruled out the pressure/ massage effects of US on our results. Another important aspect of our study was that there was no difference between the groups in supine and sitting positions before the application one week apart. So, we concluded that the effect of therapeutic US was reversible on stellate ganglion. This aspect was also in concordance with the study of Hong (1991). In conclusion, we believe that the application of therapeutic US on right-sided stellate ganglion would be a noninvasive therapeutic alternative of the pharmaco-
648
Ultrasound in Medicine and Biology
logical/surgical blockages, especially in the case of a pain with sympathetic hyperactivity, since our results demonstrated that the effects of US on right stellate ganglion led to variation in HRV parameters. However, as the present study was the first study, to our knowledge, investigating the HRV variation by application of therapeutic US on the stellate ganglion, it is clear that there is a need for further studies. Acknowledgments—The authors wish to thank the Biomedical Signal Analysis Group, Department of Applied Physics, University of Kuopio, Finland for HRV analysis software support.
REFERENCES Anderson TP, Wakim KG, Herrick JF, Bennett WA, Krusen FH. An experimental study of the effects of ultrasonic energy on the lower part of the spinal cord and peripheral nerves. Arch Phys Med 1951;32:71– 83. Baker KG, Robertson VJ, Duck FA. A review of therapeutic ultrasound: Biophysical effects. Phys Ther 2001;81:1351–1358. Buckley FP. Regional anesthesia with local anesthetics. In: Looser JD, Butler SH, Chapman CR, Turk DC, eds. Bonica’s management of pain. Philadelphia, PA: Lippincott Williams and Wilkins, 2001: 1893–1952. Currier DP, Greathouse D, Swift T. Sensory nerve conduction: effect of ultrasound. Arch Phys Med Rehabil 1978;59(4):181–185. Demmink JH, Helders PJM, Hobæk H, Enwemeka C. The variation of heating depth with therapeutic ultrasound frequency in physiotherapy. Ultrasound Med Biol 2003;29:113–118. Fujiki A, Masuda A, Inoue H. Effects of unilateral stellate ganglion block on the spectral characteristics of heart rate variability. Jpn Circ J 1999;63:854 – 858. Goodman CR. Treatment of shoulder-hand syndrome: combined ultrasonic application to stellate ganglion and physical medicine. NY State J Med 1971;71:559 –562. Halle JS, Scoville CR, Greathouse DG. Ultrasound’s effect on the conduction latency of the superficial radial nerve in man. Phys Ther 1981;61(3):345–350. Herrick JF. Temperatures produced in tissues by ultrasound: Experimental study using various technics. J Acoust Soc Am 1953;25: 12–16. Hong CZ, Liu HH, Yu J. Ultrasound thermotherapy effect on the recovery of nerve conduction in experimental compression neuropathy. Arch Phys Med Rehabil 1988;69:410 – 414.
Volume 31, Number 5, 2005 Hong CZ. Reversible nerve conduction block in patients with polyneuropathy after ultrasound thermotherapy at therapeutic dosage. Arch Phys Med Rehabil 1991;72:132–137. Introna R, Yodlowski E, Pruett J, et al. Sympathovagal effects of spinal anesthesia assessed by heart rate variability. Anesth Analg 1995; 80:315–321. Koyama S, Sato N, Nagashima K, et al. Effects of right stellate ganglion block on the autonomic nervous function of the heart. A study using the head-up tilt test. Circ J 2002;66:645– 648. Kramer JF. Effect of therapeutic ultrasound intensity on subcutaneous tissue temperature and ulnar nerve velocity. Am J Phys Med 1985; 64:1–9. Lobato EB, Kern KB, Paige GB, et al. Differential effects of right versus left stellate ganglion block on left ventricular function in humans: An echocardiographic analysis. J Clin Anesth 2000:12; 315–318. Moore JH, Gieck JH, Saliba EN, et al. The biophysical effects of ultrasound on median nerve distal latencies. Electromyogr Clin Neurophysiol 2000;40:169 –180. Portwood MM, Lieberman JS, Taylor RG. Ultrasound treatment of reflex sympathetic dystrophy. Arch Phys Med Rehabil 1987;68: 116 –118. Pumprla J, Howorka K, Groves D, et al. Functional assesment of heart rate variability: Physiological basis and pratical applications. Int J Cardiol 2002;84:1–14. Robertson VJ, Baker KG. A review of therapeutic ultrasound: Effectiveness studies. Phys Ther 2001;81:1339 –1350. Rosenberger H. über den wirkungsmechanismus der ultraschallbehandlung, insbesondere bei ischias und neuralgien. Der Chirurg 1950; 21:404 – 406. Speed CA. Therapeutic ultrasound in soft tissue lesions. Rheumatology 2001;40:1331–1336. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. Heart rate variability—standards of measurement, physiological interpretation, and clinical use. Eur Heart J 1996;17:354 –381. ter Haar G. Therapeutic Ultrasound. Eur J Ultrasound 1999;9:3–9. van der Windt DAWM, van der Heijden GJMG, van den Berg SGM, et al. Ultrasound therapy for musculoskeletal disorders: a systematic review. Pain 1999;81:257–271. Vito GD, Galloway SDR, Nimmo MA, et al. Effects of central sympathetic inhibition on heart rate variability during steady-state exercise in healthy humans. Clin Physiol & Func Im 2002;22:32–38. Wang YH, Huang TS, Lin JL, et al. Decreased autonomic nervous system activity as assessed by heart rate variability in patients with chronic tetraplegia. Arch Phys Med Rehabil 2000;81:1181–1184. Warden SJ, McMeeken JM. Ultrasound usage and dosage in sports physiotherapy. Ultrasound Med Biol 2002;28:1075–1080.
doi:10.1016/j.ultrasmedbio.2005.01.010
● Original Contribution THE EFFECTS OF THERAPEUTIC ULTRASOUND ON HEART RATE VARIABILITY: A PLACEBO CONTROLLED TRIAL VEDAT NACITARHAN,* HASAN ELDEN,* MUSTAFA KıSA,* ECE KAPTANOG˘ GLU,* and SEDAT NACITARHAN† *Cumhuriyet University, Medicine Faculty, Department of Physical Medicine and Rehabilitation, Sivas, Turkey; and † State Hospital, Department of Physical Medicine and Rehabilitation, Diyarbakır, Turkey (Received 15 November 2004; revised 24 January 2005; in final form 27 January 2005)
Abstract—The effect of therapeutic ultrasound (US) on nervous system is controversial and the effect on autonomic nervous system is not clear. Therefore, the present placebo-controlled trial was planned to investigate the effects of therapeutic US application on right-side stellate ganglion, by using analysis of heart rate variability (HRV). A total 12 healthy volunteers were included in the study. RR intervals were recorded for 5 min before and after the US application, in supine and sitting positions. All procedures were repeated in all participants with sham US one week later. The heart rate (HR) was obtained by time-domain analysis and low frequency (LF) power (%), high frequency (HF) power (%) and LF/HF ratio values were obtained by frequency-domain (power spectral density) analysis. After the US application, there was a decrease in the HR (p ⴝ 0.002) and the HF power (%) component (p ⴝ 0.015) in supine position and a decrease in HR (p ⴝ 0.002) and LF/HF ratio (p ⴝ 0.028) in sitting position. There was no significant difference after the sham US application. In conclusion, we observed that therapeutic US application on stellate ganglion causes alterations on HRV parameters. (E-mail: [email protected]) © 2005 World Federation for Ultrasound in Medicine & Biology. Key Words: Therapeutic ultrasound, Stellate ganglion, Heart rate variability.
action potential was reduced or completely lost in the sciatic nerves. Herrick (1953), in his study, suggested that US causes a selective heating and blockage after a critical level (about 45° to 46°C) for A-fibers in the nervous tissue. As the same result was obtained without US when the temperature was raised by an alternative method, Herrick (1953) concluded that US blocks the action potential via a thermal mechanism. Following studies, however, revealed contradictory reports about the effect of US on nerve conduction. Currier et al. (1978) and Halle et al. (1981) investigated the effect of US on lateral cutaneous branch of the radial nerve. They reported that the latencies were decreased, indicating increased speed of conduction, as the subcutaneous temperatures increased. Hong et al. (1988), in their study, searching, the effect of US therapy in rats with experimental compression neuropathy, reported that nerve recovery was facilitated by nonthermal US dose, whereas a thermal dose had the opposite effect. In another study, Hong (1991) investigated the effect of US on nerve conduction in patients with polyneuropathy and concluded that ultrasonic therapy with therapeutic dosage may cause a reversible conduction block. Moore et
INTRODUCTION Ultrasound (US) is defined as acoustic vibration with frequencies above the audible range. Since its introduction as a therapy over six decades ago, therapeutic US has developed into the most widely available and frequently used modality in practice of physical therapy (Robertson and Baker 2001; Warden and McMeeken 2002). In the literature, US is reported to have two main groups of effects, thermal and nonthermal (Baker et al. 2001; Demmink et al. 2003; van der Windt et al. 1999). The high absorption coefficients of large protein molecules mean that collagenous tissues may be heated preferentially, also including the nervous tissues (ter Haar 1999). Rosenberger (1950) found the temperature increase produced by US in sciatic nerves to be about twice that produced in neighboring tissues. Anderson et al. (1951) performed studies with US on spinal cords of rats and dogs and sciatic nerves of dogs and found that the spinal cords were irreversibly paralyzed and that the Address correspondence to: Vedat Nacitarhan, P.K. 743, 58140, Sivas, Turkey. E-mail: [email protected] 643
644
Ultrasound in Medicine and Biology
al. (2000) investigated the biophysical effects of US. They concluded that alterations in nerve latencies from ultrasound on healthy nerves appeared to be related to temperature changes induced by ultrasound’s thermal effects and not by nonthermal or mechanical effects. As it can be seen from the studies mentioned above, the results are contradictory and there is no consensus on the underlying mechanism. The heart is innervated through the efferent sympathetic and parasympathetic nerve fibers of the autonomic nervous system (ANS) that act on the intrinsic sinoatrial node of the heart (Wang et al. 2000). Vagal stimulation allows for beat-to-beat control of the heart rate (HR), whereas sympathetic stimulation has a more gradual effect on the HR (Pumprla et al. 2002; Wang et al. 2000). Physiological conditions illustrate the reciprocal relationship that exists between the two divisions of the ANS (Task Force of the European Society of Cardiology 1996; Wang et al. 2000). Heart rate variability (HRV) includes time and frequency domain methods (Task Force of the European Society of Cardiology 1996). Power spectral (frequency domain) analysis of HRV has been shown that harmonic oscillations in heart rate are concentrated into at least two distinct bands. That referred to as the low frequency band (LF: 0.04 to 0.15 Hz) is affected by the oscillatory rhythm of the baroreceptor system and is mediated by sympathetic activity, with some influence from vagal activity. In the other, the high frequency band (HF: 0.15 to 0.40 Hz), respiration is the primary rhythmic stimulus and it is mediated by changing levels of parasympathetic tone (Task Force of the European Society of Cardiology 1996). LF/HF ratio is interpreted to be a marker of sympathovagal balance. Power spectral analysis of HRV has shown to be a reliable noninvasive test for quantitative assessment of cardiovascular autonomic regulatory responses, providing a window reflecting the interaction of sympathetic and parasympathetic tone (Task Force of the European Society of Cardiology 1996). Cardiac sympathetic fibers originate from the stellate ganglia in conjunction with the superior and middle cervical ganglia and the first four or five thoracic sympathetic ganglia (Introna et al. 1995; Koyama et al. 2002). Previous reports revealed that block of right-side stellate ganglion decreased the sinus rate and log-transformations of the components of HRV in human (Koyama et al. 2002). A decrease in LF and HF power components after the right stellate ganglion blockage was found by Fujiki et al. (1999). Stellate ganglion blockage has beneficial effects in various clinical conditions such as complex regional pain syndromes, neuropathic painful disorders and peripheral vascular diseases. This is usually accomplished by pharmacological or surgical approaches (Buckley 2001).
Volume 31, Number 5, 2005
No study has, to our knowledge, reported the effect of therapeutic US on ANS via stellate ganglion. Based on these facts, we planned this placebo-controlled study to investigate the effects of therapeutic US by analysis of HRV, before and after therapeutic US application on right-side stellate ganglion. MATERIALS AND METHODS The study was designed as a placebo controlled trial. Subjects. This study was performed on 12 healthy volunteer subjects (four female, eight male, age range 23 to 28 y, mean 24.9). All subjects provided written informed consent and the study was approved by the ethical committee of Cumhuriyet University Medicine Faculty in Sivas, Turkey. Study protocol. The study was conducted in a silent room, at a temperature of 22 to 24°C, between 10:00 to 11:00 A.M. Subjects were abstaining from exercise and tea, caffeine and smoking intake for minimum of 1 h before study. Subjects were asked to rest for least 15 min in supine position after the electrodes were applied. Before US application, ECG recordings were made for 5 min in supine and sitting positions. Then the patients were asked to rest in supine position again. Sonogel was applied on the skin over right supraclavicular area, the sound head applied and oriented to right stellate ganglion through right supraclavicular region. Then the US apparatus (Sonoplus 992, Enraf-Nonius B.V., The Netherlands) was switched on in US group. The sound head was applied with circular movements. The ultrasound frequency was 1 MHz continuous and the intensity was 1 W/cm2. The radiation area was 5 cm2 and the application time was 5 min per subject. After the application, ECG recordings were repeated for 5 min again, both in supine and sitting positions. One week after the measurements, the study subjects were used as the placebo group. The sham US was applied without switching on the apparatus. All of the tests were performed in the same order and in the same manner with the US group. The study subjects were unaware of the type of the application (US or sham), while the operator was aware. However, the operator did not participate in the data analysis. ECG recording. Recordings were performed by using standard bipolar limb lead-I (negative electrode on right arm, positive electrode on left arm, ground electrode on left leg) with an ECG monitor (patient monitor, Ivy Biomedical System Inc, CT, USA). The analog data output of the ECG monitor was sampled at a sampling frequency of 1000 Hz, using an analog-to-digital converter. The sampled data were stored to disk for postprocessing. Detection of R-wave was done with peak
Ultrasound on heart rate variability ● V. NACITARHAN et al.
645
detection algorithm. The detected R waves were visually confirmed. RR interval was extracted in 1 ms resolution and saved consecutively. ECG recording was realized in each subject in four instances by 5 min, as pre- and postapplication supine and sitting values for analysis. Same procedure was performed for the placebo group. HRV analysis. All results were analyzed with HRV analysis software 1.1 for windows, developed by The Biomedical Signal Analysis Group, Department of Applied Physics, University of Kuopio, Finland (the software is distributed free of charge upon request in http:// venda.uku.fi/research/biosignal). The frequency-domain analysis of HRV was performed by an autoregressive model transform of the RR intervals. The time-domain measure (HR) of HRV and frequency-domain (power spectral analysis) measures [LF power (%), HF power (%), and LF/HF ratio] of HRV were calculated. Statistical analysis. The data were presented as mean ⫾ SD. As the data did not show normal distribution, nonparametric tests of comparison were used. Comparisons of the data obtained at supine and sitting positions were made by Wilcoxon (paired) signed-ranks test. Mann-Whitney test was used to compare the data among the groups. A value of p ⬍ 0.05 was considered to be significant. RESULTS HR and power spectral analysis parameters [LF power (%), HF power (%) and LF/HF ratio] of HRV of all subjects before and after application are shown in Figure 1 and Figure 2.
Fig. 2. The HR and HRV parameters of all subjects in sitting position pre- and postapplication
Comparison of pre- and postapplication data. Preand postapplication values were not significantly different in placebo group, either in supine or sitting position. In the US group, there were significant decreases in the HR (p ⫽ 0.002) and HF power (%) (p ⫽ 0.015) in the supine position and in HR (p ⫽ 0.002) and LF/HF ratio (p ⫽ 0.028) in the sitting position after the application (Table 1). The results of the comparison of pre- and postapplication data show that the US application on stellate ganglion decreased the sympathetic activity. Comparison of the data obtained in supine and sitting positions. There were significant differences both in the placebo and US groups between sitting and supine positions in HR, HF power (%) and LF/HF ratios (p ⫽ 0.002, p ⫽ 0.002, p ⫽ 0.003 in the placebo group, p ⫽ 0.002, p ⫽ 0.012, p ⫽ 0.006 in the US group, respectively) before the application. Significant differences in HR (p ⫽ 0.002), HF power (%) (p ⫽ 0.002) and LF/HF ratio (p ⫽ 0.023) were observed between the supine and sitting positions in the placebo group after application. However, there was a significant difference only in HR (p ⫽ 0.003) in the US group after the application between the sitting and supine positions (Table 2).
Fig. 1. The HR and HRV parameters of all subjects in supine position pre- and postapplication.
Comparison of the measured values in placebo and US groups. There was no difference between the groups in supine and sitting positions before the application. After the application, there were significant differences in HR and HF power (%) values between the groups in supine position (p ⫽ 0.043, p ⫽ 0.038, respectively) (Table 3).
646
Ultrasound in Medicine and Biology
Volume 31, Number 5, 2005
Table 1. Comparison of pre- and postapplication data Placebo Group
Supine
Sitting
HR LF HF LF/HF HR LF HF LF/HF
US Group
Before app†
After app†
p
Before app†
After app†
p
75 ⫾ 5 39 ⫾ 20 37 ⫾ 15 1⫾1 82 ⫾ 5 46 ⫾ 25 17 ⫾ 12 4⫾3
75 ⫾ 6 45 ⫾ 24 37 ⫾ 14 2⫾2 81 ⫾ 6 55 ⫾ 20 19 ⫾ 13 4⫾3
0.695 0.638 0.875 0.530 0.255 0.117 0.534 0.530
76 ⫾ 5 39 ⫾ 18 40 ⫾ 18 2⫾2 84 ⫾ 5 46 ⫾ 24 20 ⫾ 15 4⫾3
71 ⫾ 5 39 ⫾ 16 25 ⫾ 9 2⫾2 78 ⫾ 5 45 ⫾ 23 23 ⫾ 12 3⫾3
0.002* 0.937 0.015* 0.158 0.002* 0.388 0.182 0.028*
* Significant difference; † Application
DISCUSSION It is possible to find many studies in the literature investigating the effects of therapeutic US, a widely used agent in physical therapy. These studies are mostly on acute and chronic musculoskeletal disorders (Demmink et al. 2003; Robertson and Baker 2001; Speed 2001; van der Windt et al. 1999; Warden and McMeeken 2002). Goodman (1971) has reported therapeutic US to be effective on stellate ganglion, which is a unique study in the literature in this regard. There are some studies on peripheral nerves reporting that US blocks the neural transmission (Hong 1991; Portwood et al. 1987) or, on the contrary, that it increases the transmission velocity owing to its thermal effect (Kramer 1985; Moore et al. 2000). So, it is obvious that the effects of US on nervous system are controversial. Information on sympathovagal balance may be obtained by evaluating the cyclic variations in the heart rate via HRV analysis. RR interval is dependent on the sympathetic and parasympathetic nerves. The sympathovagal response of ANS, affected by pharmacological agent, may be traced by analysis of HRV. As the stellate ganglion gives sympathetic fibers to the heart, it is possible
to determine this effect on HRV analysis when stellate ganglion was blocked (Task Force of the European Society of Cardiology 1996). On the other hand, there is a lateralization in stellate ganglion functions. When the right stellate ganglion was blocked by a pharmacological agent, a variation in the HR and HRV parameters was determined (Koyama et al. 2002). In another study (Fujiki et al. 1999), no variation in HRV parameters was observed when the stellate ganglion blockage was performed on left side, while the right side blockage led to the variations in HRV parameters. In another study (Lobato et al. 2000) evaluating the functions of left ventricle after the stellate ganglion blockage, neither right nor left stellate ganglion blockage was found to be effective on left ventricular function. Either central (Vito et al. 2002) or spinal (Introna et al. 1995) way of sympathetic blockage have demonstrated a decrease in LF power (%) and HF power (%) components and it is reported that the power spectral density analysis of HRV provides a quantitative assessment of sympathovagal balance. In our study, there was an increase in HR and LF/HF ratio and a decrease in HF power (%) in supine
Table 2. Comparison of the data obtained in supine and sitting positions Before application
US
Placebo
HR LF HF LF/HF HR LF HF LF/HF
* Significant difference.
After application
Supine
Sitting
p
Supine
Sitting
p
76 ⫾ 5 39 ⫾ 18 40 ⫾ 18 2⫾2 75 ⫾ 5 39 ⫾ 20 37 ⫾ 15 1⫾1
84 ⫾ 5 46 ⫾ 24 20 ⫾ 15 4⫾3 82 ⫾ 5 46 ⫾ 25 17 ⫾ 12 4⫾3
0.002* 0.433 0.012* 0.006* 0.002* 0.388 0.002* 0.003*
71 ⫾ 5 39 ⫾ 16 25 ⫾ 9 2⫾2 75 ⫾ 6 45 ⫾ 24 37 ⫾ 14 2⫾2
78 ⫾ 5 45 ⫾ 23 23 ⫾ 12 3⫾3 81 ⫾ 6 55 ⫾ 20 19 ⫾ 13 4⫾3
0.003* 0.272 0.433 0.117 0.002* 0.071 0.002* 0.023*
Ultrasound on heart rate variability ● V. NACITARHAN et al.
647
Table 3. Comparison of the measured values in placebo and US groups Before application
Supine
Sitting
HR LF HF LF/HF HR LF HF LF/HF
After application
Placebo
US
p
Placebo
US
p
75 ⫾ 5 39 ⫾ 20 37 ⫾ 15 1⫾1 82 ⫾ 5 46 ⫾ 25 17 ⫾ 12 4⫾3
76 ⫾ 5 39 ⫾ 18 40 ⫾ 18 2⫾2 84 ⫾ 5 46 ⫾ 24 20 ⫾ 15 4⫾3
0.795 0.751 0.603 0.795 0.488 1.000 0.644 1.000
75 ⫾ 6 45 ⫾ 24 37 ⫾ 14 2⫾2 81 ⫾ 6 55 ⫾ 20 19 ⫾ 13 4⫾3
71 ⫾ 5 39 ⫾ 16 25 ⫾ 9 2⫾2 78 ⫾ 5 45 ⫾ 23 23 ⫾ 12 3⫾3
0.043* 0.488 0.038* 0.564 0.149 0.248 0.312 0.119
* Significant difference.
position compared with the sitting position, in both the US and placebo group before the application. This situation is related with the compensatory increase of sympathetic activation to prevent hypotension and peripheral pooling of blood in the lower extremities in the sitting-up position. The decrease in HF power (%) is in accordance with the decrease in parasympathetic tonus as a part of the adaptation. The increase in LF/HF ratio is the expression of this variation on behalf of the sympathetic component in autonomic nervous system, which is the normally expected physiological response in a healthy individual. After the US application, while the increase in HR was still significant, the increase in LF/HF ratio and the decrease in HF power (%) component was no longer significant in sitting position. On the other hand, after the application, the increase in HR and LF/HF ratio and the decrease in HF power (%) component were still significant in the placebo group. Although the increase in the HR was statistically significant after the US application, the basal HRs before and after the application were different. The HR, which was 76 ⫾ 5 in supine and 84 ⫾ 5 in sitting positions, decreased to 71 ⫾ 5 and 78 ⫾ 5 in supine and sitting positions, respectively, after the therapeutic US application. As can be traced from the results, either in supine or sitting positions, HR is found to be decreased compared with the preapplication values. Unfortunately, we could not compare the above observation, since no similar study was found in the literature on the subject. Both the decrease in HR and the absence of the increase in LF/HF ratio and the decrease in HF power (%) component in US group, different from that of the placebo group after the application, can be interpreted as “therapeutic US application decreases the effect of the sympathetic component of autonomic nervous system”. Koyama et al. (2002) have evaluated the effect of the autonomic nervous system on the heart by pharmacological blockage of the right stellate ganglion. They evaluated the sympathovagal tonus in supine position
and during head-up tilt test (HUTT) in eight patients with chronic pain, before and after right stellate ganglion blockage. While they have obtained normal physiological responses before blockage, HR and LF/HF ratio and HF power (%) components during HUTT did not significantly alter after the right stellate ganglion blockage procedure. So, they suggested that right stellate ganglion blockage suppresses cardiac sympathetic function. Fujiki et al. (1999) found a decrease in LF and HF power components after the right stellate ganglion blockage by 1% mepivacaine and concluded that the right-sided blockage causes a decrease in sympathetic and parasympathetic effect on the sinus node. The major difference between these studies and our study is the method of ganglion blockage, where we used a noninvasive method instead of pharmacological agents. The change in HRV component in our study is in concordance with these previous studies and either pharmacological blockage or US application (as in our study) seems to depress the sympathetic component. We take into consideration also the pressure/massage effects of the US probe in our study. The therapeutic effect of US was separated from a pressure effect on the stellate ganglion by using the same study subject as a placebo group by repeating the study one-week apart. In the placebo group, the increase in HR and LF/HF ratio and the decrease in HF power (%) component before the application was also obtained following the application, whereas it was different in US group. This outcome ruled out the pressure/ massage effects of US on our results. Another important aspect of our study was that there was no difference between the groups in supine and sitting positions before the application one week apart. So, we concluded that the effect of therapeutic US was reversible on stellate ganglion. This aspect was also in concordance with the study of Hong (1991). In conclusion, we believe that the application of therapeutic US on right-sided stellate ganglion would be a noninvasive therapeutic alternative of the pharmaco-
648
Ultrasound in Medicine and Biology
logical/surgical blockages, especially in the case of a pain with sympathetic hyperactivity, since our results demonstrated that the effects of US on right stellate ganglion led to variation in HRV parameters. However, as the present study was the first study, to our knowledge, investigating the HRV variation by application of therapeutic US on the stellate ganglion, it is clear that there is a need for further studies. Acknowledgments—The authors wish to thank the Biomedical Signal Analysis Group, Department of Applied Physics, University of Kuopio, Finland for HRV analysis software support.
REFERENCES Anderson TP, Wakim KG, Herrick JF, Bennett WA, Krusen FH. An experimental study of the effects of ultrasonic energy on the lower part of the spinal cord and peripheral nerves. Arch Phys Med 1951;32:71– 83. Baker KG, Robertson VJ, Duck FA. A review of therapeutic ultrasound: Biophysical effects. Phys Ther 2001;81:1351–1358. Buckley FP. Regional anesthesia with local anesthetics. In: Looser JD, Butler SH, Chapman CR, Turk DC, eds. Bonica’s management of pain. Philadelphia, PA: Lippincott Williams and Wilkins, 2001: 1893–1952. Currier DP, Greathouse D, Swift T. Sensory nerve conduction: effect of ultrasound. Arch Phys Med Rehabil 1978;59(4):181–185. Demmink JH, Helders PJM, Hobæk H, Enwemeka C. The variation of heating depth with therapeutic ultrasound frequency in physiotherapy. Ultrasound Med Biol 2003;29:113–118. Fujiki A, Masuda A, Inoue H. Effects of unilateral stellate ganglion block on the spectral characteristics of heart rate variability. Jpn Circ J 1999;63:854 – 858. Goodman CR. Treatment of shoulder-hand syndrome: combined ultrasonic application to stellate ganglion and physical medicine. NY State J Med 1971;71:559 –562. Halle JS, Scoville CR, Greathouse DG. Ultrasound’s effect on the conduction latency of the superficial radial nerve in man. Phys Ther 1981;61(3):345–350. Herrick JF. Temperatures produced in tissues by ultrasound: Experimental study using various technics. J Acoust Soc Am 1953;25: 12–16. Hong CZ, Liu HH, Yu J. Ultrasound thermotherapy effect on the recovery of nerve conduction in experimental compression neuropathy. Arch Phys Med Rehabil 1988;69:410 – 414.
Volume 31, Number 5, 2005 Hong CZ. Reversible nerve conduction block in patients with polyneuropathy after ultrasound thermotherapy at therapeutic dosage. Arch Phys Med Rehabil 1991;72:132–137. Introna R, Yodlowski E, Pruett J, et al. Sympathovagal effects of spinal anesthesia assessed by heart rate variability. Anesth Analg 1995; 80:315–321. Koyama S, Sato N, Nagashima K, et al. Effects of right stellate ganglion block on the autonomic nervous function of the heart. A study using the head-up tilt test. Circ J 2002;66:645– 648. Kramer JF. Effect of therapeutic ultrasound intensity on subcutaneous tissue temperature and ulnar nerve velocity. Am J Phys Med 1985; 64:1–9. Lobato EB, Kern KB, Paige GB, et al. Differential effects of right versus left stellate ganglion block on left ventricular function in humans: An echocardiographic analysis. J Clin Anesth 2000:12; 315–318. Moore JH, Gieck JH, Saliba EN, et al. The biophysical effects of ultrasound on median nerve distal latencies. Electromyogr Clin Neurophysiol 2000;40:169 –180. Portwood MM, Lieberman JS, Taylor RG. Ultrasound treatment of reflex sympathetic dystrophy. Arch Phys Med Rehabil 1987;68: 116 –118. Pumprla J, Howorka K, Groves D, et al. Functional assesment of heart rate variability: Physiological basis and pratical applications. Int J Cardiol 2002;84:1–14. Robertson VJ, Baker KG. A review of therapeutic ultrasound: Effectiveness studies. Phys Ther 2001;81:1339 –1350. Rosenberger H. über den wirkungsmechanismus der ultraschallbehandlung, insbesondere bei ischias und neuralgien. Der Chirurg 1950; 21:404 – 406. Speed CA. Therapeutic ultrasound in soft tissue lesions. Rheumatology 2001;40:1331–1336. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. Heart rate variability—standards of measurement, physiological interpretation, and clinical use. Eur Heart J 1996;17:354 –381. ter Haar G. Therapeutic Ultrasound. Eur J Ultrasound 1999;9:3–9. van der Windt DAWM, van der Heijden GJMG, van den Berg SGM, et al. Ultrasound therapy for musculoskeletal disorders: a systematic review. Pain 1999;81:257–271. Vito GD, Galloway SDR, Nimmo MA, et al. Effects of central sympathetic inhibition on heart rate variability during steady-state exercise in healthy humans. Clin Physiol & Func Im 2002;22:32–38. Wang YH, Huang TS, Lin JL, et al. Decreased autonomic nervous system activity as assessed by heart rate variability in patients with chronic tetraplegia. Arch Phys Med Rehabil 2000;81:1181–1184. Warden SJ, McMeeken JM. Ultrasound usage and dosage in sports physiotherapy. Ultrasound Med Biol 2002;28:1075–1080.