A Preliminary Study of the Effect of Acupuncture on Emotional Stress in Female Dysphonic Speakers

Elaine Y. L. Kwong and Edwin M.-L. Yiu, Sai Ying Pun, Hong Kong Summary: This study investigated the effect of acupuncture on emotional stress in subjects with phonotraumatic injuries. This study used a prospective randomized, placebo-controlled group design. The independent variable included the types of acupuncture (genuine vs sham) and the sampling time points (two pre-needling, one in the midway of needling, and two post-needling measurements). The dependent variable was the concentration of cortisol obtained from subjects saliva samples. Eighteen female subjects with phonotraumatic injuries were randomized to receive either genuine or sham acupuncture at the same acupoints during a 30-minute session. Saliva samples were collected from each subject at 10 minutes pre-needling, immediately pre-needling, mid-needling, immediately postneedling, and 10 minutes post-needling time points. The ndings suggested that the subjects salivary cortisol concentration did not reduce after acupuncture, and thus, acupuncture may not be able to reduce the emotional stress level in female dysphonic speakers. Key Words: Traditional Chinese medicineAcupunctureComplementary and medicineVoice disorders. Voice disorders often correlate with emotional stress.1,2 Individuals with vocal nodules or other voice problems usually have signicantly higher stress level than their normal counterparts.3 The psychological status of an individual was found to be correlated with his or her vocal condition.4 One of the psychological factors, emotional stress, would affect the autonomic nervous system and in turn lead to a number of physiological changes. Its impacts on voice include alteration of laryngeal secretions; increased tension in the head and neck area and especially in the laryngeal muscles; and increased likelihood of vocal misuse.3 These impacts may result in excessive phonotrauma, and thus, voice disorders with or without vocal fold pathologies. Stress is not only believed to be contributive to the development of voice problems, it may also delay the recovery process of voice problems.5 As stated earlier, stress responses often affect proper voicing, which is an essential component in managing or treating voice disorders.6 Stress could also adversely affect the immune system of an individual.7,8 Therefore, the healing of vocal fold lesion would, thus, be affected by emotional stress. Acupuncture has been shown to be effective in treating dysphonia caused by benign vocal pathologies.9 Separate studies have also shown that acupuncture is effective in reducing emotional stress level.10,11 Therefore, it may be possible that the effectiveness of acupuncture in treating benign vocal pathologies is related to, if not caused by, the reduction of stress level in the dysphonic speakers. This study set out to explore the effect of acupuncture on emotional stress. Salivary cortisol concentration has been found to be a physiological indicator of emotional stress level.5 The present study investigated the concentration level of salivary cortisol in individuals with phonotraumatic injuries following acupuncture. METHOD Participants A total of 18 female subjects were recruited from the Voice Research Laboratory at the University of Hong Kong. The subjects met the following selection criteria: (1) diagnosed by otolaryngologist to have phonotraumatic injuries, including vocal nodules, vocal polyps, vocal fold edema, or chronic laryngitis; (2) aged between 20 and 55 years; (3) had not received acupuncture for any condition within the preceding 10 years to ensure successful blinding by the placebo technique; (4) had no history of neurological condition to confound the effect of acupuncture; and (5) showed no signs of upper respiratory infection at the time of the study. Vocal fold edema refers to the edematous lesion that affects a diffused length of the vocal fold and is chronic in nature. Reinkes edema may be involved in subjects diagnosed to have vocal fold edema. Chronic laryngitis refers to the inammation to the vocal fold that is not related to infections and/or any signs of reux and is believed to have resulted from phonotrauma with reference to the patient history. The subjects were randomly assigned to experimental and placebo groups, which received genuine acupuncture and sham acupuncture, respectively. The mean age of the experimental group (33.56 years; standard deviation [SD] 10.08; range 2050 years) and placebo group (31.78 years; SD 4.87; range 2740 years) were not signicantly different (t(16) 0.477, P 0.64). The otolaryngological diagnoses of subjects in each group are summarized in Table 1. Analyzing tool Saliva samples were collected using a device called Salivette manufactured by Sarstedt, Germany (Figure 1). The device was made up of four main components, namely (1) a lid which keeps the device air- and water-tight; (2) a cotton swab for absorption of saliva; (3) a tube to suspend the cotton swab in position during centrifugation; and (4) a container which allows saliva to deposit after centrifugation. Acupuncture needles Needles used in acupuncture were made of stainless steel and were manufactured by Hua Tuo, China. Needles used in genuine acupuncture were 0.25 mm in diameter, 40 mm and 25 mm

Accepted for publication May 27, 2009. From the Division of Speech & Hearing Sciences, Voice Research Laboratory, The University of Hong Kong, Sai Ying Pun, Hong Kong. Address correspondence and reprint requests to Elaine Y.L. Kwong, Division of Speech & Hearing Sciences, Voice Research Laboratory, The University of Hong Kong, 5/F, Prince Philip Dental Hospital, 34 Hospital Road, Sai Ying Pun, Hong Kong. E-mail: jasperek@ graduate.hku.uk Journal of Voice, Vol. 24, No. 6, pp. 719-723 0892-1997/$36.00 2010 The Voice Foundation doi:10.1016/j.jvoice.2009.05.005

720

Journal of Voice, Vol. 24, No. 6, 2010

TABLE 1. Distribution of Otolaryngological Diagnoses of Subjects in the Experimental and Placebo Groups Number of Subjects Otolaryngological Diagnoses Vocal fold nodules Vocal fold polyp Vocal fold edema Chronic laryngitis Experimental Group (N 9) 7 0 1 1 Placebo Group (N 9) 6 2 0 1

in length Needles used in sham acupuncture were blunt needles with the dimensions of 0.25 mm in diameter and 25 mm in length (Figure 2). Procedures Timing of saliva sampling. Five saliva samples were obtained from each subject at ve time points: 1. 2. 3. 4. 5. Ten minutes before needling (PRE-10); Immediately before needle insertion (PRE-0); Fifteen minutes after needle insertion (MID); Immediately after needle removal (POST-0); and Ten minutes after needle removal (POST-10).

Collection of saliva sample. All sampling and acupuncture procedures were conducted at the Voice Research Laboratory, The University of Hong Kong. Subjects were required to refrain from eating and drinking (except water) for 3 hours before the collection of the rst saliva sample. It was to prevent any contamination of saliva, which might, in turn, lead to alteration of cortisol concentration. To collect saliva samples, subjects were required to gently chew the cotton swab of the Salivette tube (component 2 in Figure 1) for approximately 1 minute or until they report that the swab was soaked with saliva. The samples collected were kept in the Salivette tube and immediately frozen to a temperature of below 18 C to inhibit bacterial growth in it. Acupuncture. All acupuncture sessions were conducted by an acupuncturist registered in Hong Kong, who had more than 20 years experience in acupuncture. Subjects in the experimental group received genuine acupuncture with needles at the acupoints Hegu (LI4), Lieque (Lu7), Lianquan (Ren23), Renying (St9), and Zhaohai (Ki6), whereas subjects in the placebo group received sham acupuncture with blunt needles at the

FIGURE 1. Salivette tube and its components.

same acupoints. The acupoints were selected by three Traditional Chinese Medicine practitioners (each with over 20 years of experience) with reference to standard acupuncture textbooks, personal experiences, and also the protocol adopted by Yiu et al.9 During acupuncture, subjects were asked to lie comfortably in a supine position on a couch. The skin area around the acupoints was disinfected with 75% alcohol rst. After that, adhesive plastic stands were taped onto the skin surface at the acupoints. A needle guide tube was then inserted into the hole in the middle of each stand. For genuine acupuncture, guide tubes that were 30 mm in length were used for all acupoints and needles that were 40 mm in length were used for Hegu (LI4), Lianquan (Cv23), and Zhaohai (Ki6), whereas those that were 25 mm in length were used for Renying (St9) and Lieque (Lu7). Needles were inserted perpendicularly through the guide tubes to reach the subcutaneous tissue layer or until zhen gan (needling sensation) or de qi (qi sensation) was felt by the acupuncturist or reported by the subject (Figure 3). For sham acupuncture, the guide tubes and needles used for all acupoints were 47 and 25 mm in length, respectively. The blunt needles were also inserted perpendicularly through the guide tubes. After the blunt tip had reached skin surface at the acupoint, a plastic tag was inserted into the upper end of the guide tube to xate the blunt needle and to maintain the pressure exerted by the blunt needle to the skin surface and to simulate the feeling of needling (see Figure 4 for illustration of the sham needling). Both the experimental and placebo group subjects received one 30-minute session of acupuncture. Extra stimulations were applied to the acupoints once every 5 minutes to maintain the needling sensation (for the experimental group) and to simulate the typical acupuncture process (for the placebo group). Saliva samples were sent to the Saliva Cortisol Laboratory, Centre on Behavioral Health, The University of Hong Kong, for analyses. The Salivette tubes were rst centrifuged to obtain saliva uid from the cotton swab. The samples were then assayed for their cortisol concentration using an enzyme-linked immunoabsorbent assay kit (Salimetrics, Inc., State College, PA, USA). The assaying procedures were carried out according to that of Schwartz et al.12 In brief, the samples were brought back to room temperature, diluted with phosphate-buffered solution, and then mixed with tetramethylbenzide solution and stop solution of sulfuric acid. Salivary cortisol concentrations were then measured (in nanomole per liternmol/L) by obtaining the optical density of the end mixture.

Elaine Y.L. Kwong and Edwin M.-L. Yiu

Effect of Acupuncture on Emotional Stress

721

FIGURE 2. Guide tube, 40- and 25-mm needles for acupuncture. Data analyses The salivary cortisol concentrations at the PRE-10 and PRE0 time points were averaged for each subject to give a more representative pre-needling concentration (PREmean). The mean and SD of salivary cortisol concentration were calculated for each group (ie, experimental and placebo group) at each time point (ie, PREmean, MID, POST-0, and POST-10). Because of the small sample size, the assumptions of having normal distributions and equal variances were likely to be violated. Nonparametric tests were therefore employed to detect any statistical signicance. The changes in salivary cortisol concentration over time of each group were analyzed by comparing the differences between mean concentration at PREmean and that at the other time points using Wilcoxon signed rank tests. Group differences were analyzed using Mann-Whitney U tests for the changes in salivary cortisol level from the PREmean time point at the MID, POST-0, and POST-10 time points.

FIGURE 4. Illustration of sham acupuncture. time point to 1.28 nmol/L at the MID time point (Z 1.96; P 0.05).The cortisol level showed little changes at the postneedling time points (Post-0: Z 0.77, P 0.44; POST-10: Z 0.18, P 0.86). Interestingly, the placebo group also showed a signicant reduction in the mean salivary cortisol concentration during the acupuncture at the MID time point (1.72 nmole/L) when compared with PREmean time point (2.97 nmole/L; Z 2.19, P 0.03). This lower level of salivary cortisol (Table 2), however, did not reach the signicance level at the POST-0 (Z 1.6, P 0.11) and POST-10 (Z 1.72, P 0.09) time points when compared with the PREmean baseline. DISCUSSION The present study aimed to examine the effect of acupuncture on emotional stress in subjects with phonotraumatic injuries. Level of individuals emotional stress was measured by determining the concentration of salivary cortisol, a well-established measure in the eld of psychoneuroendocrinology.5 According to the literature, acupuncture is capable of reducing ones emotional stress level,10,11 and is reported to be capable of improving the vocal functions in individuals with benign vocal pathologies.9 It was, therefore, hypothesized that the present acupuncture protocol would reduce the emotional stress level, as indicated by the salivary cortisol concentration, in dysphonic subjects. The ndings showed that, both the experimental and placebo groups exhibited signicant reductions in salivary cortisol concentration during needling. Such signicant changes in the salivary cortisol concentration were hypothesized to be attributable to either (1) the reductions in emotional stress level following the genuine or sham acupuncture procedure, (2) an already elevated emotional stress level at the pre-needling

RESULTS The mean and SD of salivary cortisol concentration of the experimental and placebo groups are listed in Table 2. The experimental group showed a reduction in the mean salivary cortisol concentration from 2.19 nmol/L at the PREmean

FIGURE 3. Genuine acupuncture using foam stand, 25-mm needle and 30-mm guide tube.

722
TABLE 2. Mean and SD of Salivary Cortisol Concentration

Journal of Voice, Vol. 24, No. 6, 2010

Mean (SD) Concentration in nmol/L Group Experimental group (N 9) Placebo group (N 9) PREmean 2.19 (1.92) 2.87 (2.60) MID 1.28 (0.51) 1.72 (1.75) POST-0 2.41 (2.09) 1.77 (1.29) POST-10 2.30 (2.11) 1.80 (1.68)

SD Standard deviation. PREmean mean salivary cortisol concentration of saliva samples obtained 10 minutes before needling (PRE-10) and immediately before needle insertion (PRE0); MID 15 minutes after needle insertion; POST-0 immediately after needle removal; POST-10 10 minutes after needle removal.

time point in the rst place, or (3) a combination of both. As both the genuine and sham acupuncture procedures resulted in reduction in the salivary cortisol level, and as the subjects were not familiar with the procedures of acupuncture (one of the inclusion criteria was not receiving acupuncture for any condition within the preceding 10 years), elevated stress level before the procedure was not unexpected. It is well understood that state anxiety level13 could increase following unpleasant emotional arousal in facing threatening demands or dangers. The state anxiety level is generally reected in the emotional stress level, and thus, in the salivary cortisol concentration.14,15 Therefore, the reduction in salivary cortisol concentration from PREmean to MID in both groups may be interpreted as a result of eliminated state anxiety once needling had started. The reduction in the salivary cortisol concentration did not sustain after the needle removal (ie, at POST-0 and POST-10 time points) in the two groups as the cortisol levels of each group resembled those obtained at the PREmean time point. For both the experimental and placebo groups, the increases in mean salivary cortisol concentration after needling might be explained by the delayed effect of the heightened state anxiety level before needle removal.14,15 Again, as all the subjects were unfamiliar with acupuncture, the subjects anxiety level and the salivary cortisol concentration might have heightened before acupuncture needles were removed. The decrease in salivary cortisol concentration may lag the cessation of state anxiety by several minutes (eg, time lags ranging from 5 to 20 minutes were suggested by Kirschbaum and Hellhammer16). The increased salivary cortisol level at the POST-0 and POST-10 time points might be the result of the before-removal state anxiety. For the experimental group, in addition to the delayed effect of the before-removal state anxiety, the increases in salivary cortisol concentration at postneedling time points may also have resulted from the removal of acupuncture needle. In ordinary acupuncture, removal of needles involve overcoming the attractive forces between the needle and the connective tissues.17 Such a process would produce unpleasant sensation, such as pain, which subsequently would induce stress, and thus, increase salivary cortisol concentration in the experimental group subjects.5,18 One limitation of the present study was the small sample size. The inadequate sample size of 18 made the present study underpowered to detect both the across-group and within-group differences. Given the small effect size of 0.115 in the present

study, it was calculated that a study with at least a sample size of 106 (alpha level 0.05, power 0.8)19 would be powerful enough for any statistical signicance to be detected. A second limitation of the study might have arisen from the potential selection bias. The subjects baseline emotional stress level was not included as a selection criterion. It was possible that some, if not all, of the subjects recruited for the present study did not necessarily demonstrate high emotional stress even before receiving acupuncture. This might lead to a confounding oor effect, and thus, the negative results of the present study. It is therefore suggested that selection criteria regarding the psychological aspects, for instance, subjects perceived stress and anxiety level, may be included in future studies. Another limitation involved the inadequate follow-up time. The salivary cortisol concentration of the experimental group showed a tendency to decrease from POST-0 to POST-10. However, as no more saliva sample was collected since then, important changes might have been overlooked. It is suggested that more saliva samples may be collected for a longer follow-up period. Furthermore, it was also possible that one acupuncture session was not sufcient to result in signicant changes in emotional stress level. Studies may be designed to measure the change in cortisol concentration throughout the course of acupuncture treatment. CONCLUSION In summary, this article presents a preliminary study on the effect of acupuncture on emotional stress in subjects with phonotraumatic injuries. Emotional stress level was measured with the concentration of cortisol found in subjects saliva samples. Results from inferential statistics suggested that no group difference was found between the genuine and sham acupuncture. A signicant decrease in salivary cortisol concentration was found from the pre-needling to the mid-needling time points. However, the cortisol concentration returned to the pre-needling level after receiving acupuncture. To conclude, the ndings from the present study did not lend support to the hypothesis that acupuncture brought about reduction in emotional stress level in female dysphonic speakers. REFERENCES
1. Roy N, Bless DM, Heisey D. Personality and voice disorders: a multitrait-multidisorder analysis. J Voice. 2000;14:521548. 2. Roy N, Bless DM, Heisey D. Personality and voice disorders: a superfactor trait analysis. J Speech Lang Hear Res. 2000;43:749768.

Elaine Y.L. Kwong and Edwin M.-L. Yiu

Effect of Acupuncture on Emotional Stress

723

3. Goldman SL, Hargrave J, Hillman RE, Holmberg E, Gress C. Stress, anxiety, somatic complaints, and voice use in women with vocal nodules: preliminary ndings. Am J Speech Lang Pathol. 1996;5:4454. 4. Rosen CA, Sataloff RT. Psychology of Voice Disorders. San Diego: Singular Publishing Group, Inc.; 1997. 5. Kirschbaum C, Hellhammer DH. Salivary cortisol in psychoneuroendocrine research: Recent developments and applications. Psychoneuroendocrinology. 1994;19:313333. 6. Colton RH, Casper JK. Understanding Voice Problems. (2nd ed.). Baltimore: Williams & Wilkins; 1996. 7. Kiecolt-Glaser JK, Marucha PT, Malarkey WB, Mercado AM, Glaser R. Slowing of wound healing by psychological stress. Lancet. 1995;346:11941196. 8. Glaser R, Kiecolt-Glaser JK, Marucha P, MacCallum RC, Laskowski BF, Malarkey W. Stress-related changes in proinammatory cytokin production in wounds. Arch Gen Psychiatry. 1999;56:450456. 9. Yiu E, Xu JJ, Murry T, et al. A randomized treatment-placebo study of the effectiveness of acupuncture for benign vocal pathologies. J Voice. 2006;20:144156. 10. Middlekauff HR, Yu JL, Hui K. Acupuncture effects on reex responses to mental stress in humans. Am J Physiol Regul Integr Comp Physiol. 2001;280:14621468.

11. Knardahl S, Elam M, Olausson B, Gunnar WB. Sympathetic nerve activity after acupuncture in humans. Pain. 1998;75:1925. 12. Schwartz EB, Granger DA, Susman EJ, Gunnar MR, Laird B. Assessing salivary cortisol in studies of child development. Child Dev. 1998;69: 15031513. 13. Spielberger CD. State-trait Anxiety Inventory for Adults. Palo Alto: Mind Garden, Inc.; 1983. Sample set: manual, test, scoring key. 14. Pawlow LA, Jones GE. The impact of abbreviated progressive muscle relaxation on salivary cortisol. Biol Psychol. 2002;60:116. 15. Field T, Hernandez-Reif M, Diego M, Schanberg S, Kuhn C. Cortisol decreases and serotinin and dopamine increase following massage therapy. Int J Neurosci. 2005;115:13971413. 16. Kirschbaum C, Hellhammer DH. Salivary cortisol. In: Fink G, ed, Encyclopedia of Stress, Vol. 3. San Diego, CA: Academic Press; 2000:379383. 17. Langevin HM, Churchill DL, Cipolla MJ. Mechanical signaling through connective tissue: a mechanism for the therapeutic effect of acupuncture. J Fed Am Soc Exp Biol. 2001;15:22752282. 18. Carr DB, Goudas LC. Acute pain. Lancet. 1999;353:20512058. 19. Faul F, Erdfelder E, Lang A-G, Buchner A. G*Power 3: a exible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods. 2007;39:175191.