Ming-Feng Chen, MD, PhD,
1,4 §
Hsi-Ming Yu, MD, MS,
2
Shu-Fang Li, MT,
MS
3
Ta-Jung You, MD, MS,
2
1
Department of Internal Medicine, Show Chwan Memorial Hospital, No.542, Sec.1,
Chung-Shang Rd., Changhua city, Taiwan
2
Department of Chinese Medicine, TZU-AI General Hospital, Yulin country, Taiwan
3
Department of Clinical Laboratory Science, Tainan Municipal Hospital, Tainan city,
Taiwan
4
Institute of Chinese Medicine Science, China Medical University, Taichung city,
Taiwan
*These authors contributed equally to this work
§
Corresponding author
E-mail addresses:
MFC: cmffly@Gmail.com
HMY: ucmcmc@Gmail.com
TJY: ytj93@ms9.hinet.net
SFL: iris@ms6.url.com.tw

Abstract
Qi vacuity (QV), defined by traditional Chinese medicine, reflects a status with loss of energy in human body. An objective method to detect QV is not in practice until now. Automatic Reflective Diagnosis System (ARDK) is a device to detect bio-energy through measuring skin conductance of 24 special acupuncture points on both wrists and ankles. In this study we used ARDK to measure the skin conductance of 193 patients with QV and 89 sex- and age-matched healthy humans, to investigate whether it can be used in detecting QV. Patients with a diagnosis of
QV should have at least three or more of five symptoms/ signs, the severity of which was further scaled into 5 degrees, and scored from 0 to 4 points. We compared the difference in mean value of ARDK between patients with QV and healthy controls, and further investigated the correlation between the mean value of
ARDK and score of qi vacuity in patients with QV by linear regression analysis. We found that, the mean value of ARDK in patients with QV (30.2
± 16.8
A) were significantly lower than those in healthy controls (37.7
± 10.8
A; P < 0.001). A negative correlation was found between the mean value of ARDK and the score of qi deficiency (r-coefficient = - 0.61; P < 0.001). After adjusting age, the decreased mean value of ARDK in patients with QV showed a significant correlation with the score of QV. These results suggest that, the mean value of ARDK can reflect the severity of QV. It also suggests that the bio-energy of human body can be measured by skin conductance. ARDK may be used as a complementary method to detect QV.

Background
Qi vacuity (QV) is a syndrome defined by traditional Chinese medicine as a status with loss of energy in human body. The patients with QV usually have symptoms of tiredness, general malaise and loss of strength to speak loudly or for long-time. In some occasions, they may have spontaneous sweating abruptly during movement. Usually, the traditional Chinese medicine doctors make a diagnosis of
QV by observing the symptoms and sings in patients subjectively. A golden standard to diagnose QV is not in practice until now. A Professional Committee for the study of vacuity patterns and Age’s disease in the Society of Integrated Chinese and
Western Medicine proposed a criterion for diagnosis of QV (Sen and Wang, 1986). It is suggested that patients with a diagnosis of QV should have three or more of five major symptoms/ signs. The five major symptoms/ signs include fatigued spirit with lack of strength, shortness of breath accompanied by laziness to speak, spontaneous sweating, enlarged tongue with or without dental impressions, and a vacuous and weak pulse. However, the diagnosis of QV is still subjective and is not objective.
Automatic Reflex Detecting System (ARDK) is an electrical device, made by
Good News Natural Medicine Biotech Co., Ltd., Taichung, Taiwan, to detect the skin conductance between left hand and each one of 24 special acupuncture points of meridians on both wrists and ankles. The mean value of skin conductance from the above 24 acupuncture points, calculated by a computer system automatically, reflects the energy of human body. It is not known whether ARDK can be used as a tool in the detection of QV objectively. The aim of this study was to investigate whether the mean value of skin conductance, detected by ARDK, can reflect the severity of QV.
We used ARDK to detect the skin conductance between left hand and each one of 24 special acupuncture points in 193 patients with QV and 89 sex- and age-matched healthy humans without evident symptoms. We compared the difference in the mean value of skin conductance between patients with QV and healthy controls, and

further analyzed the correlation between the mean value of skin conductance and the severity of QV in those patients.
Materials and Methods
Subjects:
Eighty-nine healthy controls (age: 33.7 +/- 10.3 years old, mean +/- SD) and
193 patients (age: 30.2 +/- 16.8 years old) with a diagnosis of QV were involved in this study from August 1, 2006 to December 31, 2006.
The healthy controls (40 males and 49 females, age range: 22- 72 years old) were selected from the subjects who received a general health examination in the
Health Examination Center of Tainan Municipal Hospital. They had no special complaints during the past one month and no past history of chronic diseases including diabetes, cardiovascular disease, respiratory disease, liver disease, renal disease, autoimmune disease and malignancy. All of them were found normal in blood routine, urine routine, EKG and serum biochemistry examination including liver function, renal function, sugar and electrolytes.
The studied patients (83 males and 110 females ranging in age between 20-72 years) were selected from the outpatients in the Department of Integrated Chinese and Western medicine in Tainan Municipal Hospital. They all had a major complaint of tiredness and general malaise, and were diagnosed with QV. Among them, 108 were hepatitis B or C viral carriers with normal liver function. There were no significant differences in sex and age between studied patients and healthy controls.
Study procedure
The studied protocol was ethically approved by the institutional board of Tainan
Municipal Hospital. Patients with a chief complaint of tiredness and healthy controls were asked to joint the study. After signed the informed consent, each subject was diagnosed and evaluated the severity of QV by two experienced traditional Chinese medicine doctors. A well-trained technician measured their skin conductance using

an ARDK instrument in a room with constant temperature at 25 and humidity at
55%. The mean values of skin conductance in all subjects and the data of QV severity in patients were recorded and transferred to a statistician for analysis.
Method to diagnose and evaluate the severity of QV
The QV syndrome was diagnosed according to the criterion proposed by the
Society of Integrated Chinese and Western Medicine in China. Patients with three or more of the following symptoms/ signs including fatigued spirit with lack of strength, shortness of breath accompanied by laziness to speak, spontaneous sweating, enlarged tongue with or without dental impressions, and a vacuous and weak pulse were diagnosed with QV. In this study, we further scaled the severity of each symptom/ sign into 5 degrees with a score from 0 to 4. The summation of scores from 5 symptoms/signs of QV, denoted the severity of QV, and was named as the score of QV.
Procedure of ARDK test
The ARDK test was performed according to the standard operation procedure proposed by the manufacturer .
The studied subjects should avoid the intake of any drinks, such as coffee, tea, or drugs as beta-blockers, which may influence the activity of sympathetic nervous system, for more than 3 days before examination.
Each subject was asked to take a rest and relax for more than 20 minutes, and took off all metal accessories before examination. Skin conductance was measured in a room with temperature at 25 °C and humidity at 55%. Each subject was asked to lie on a bed and to hold a metal plate by left hand. Then, the technician used a small detector to measure the skin conductance over each one of 24 special acupoints in both wrists and ankles of each subject in sequence. Most of the detected acupoints were the original acupoints of 12 meridians. They were LU9, PC7, HT7, SI4, SJ4,
LI5, SP3, LR3, KI3, BL65, GB40 and ST40 in bilateral wrists or ankles. The equipment of ARDK, shown in Figure 1, included a metal plate, a detector and an

ARDK device. The metal plate was used to connect to the left hand of subject. The detector was used to touch the acupoints of subject. Both of the metal plate and detector were connected to the ARDK device, which could supply a weak electrical current with a constant voltage of 1.75V and gather about 50 values within 2 seconds of touching the skin. This device was further connected to a personal computer, which could calculate the mean value of skin conductance of 24 acupoints automatically soon after the examination.
Statistical Analysis
The difference between the skin conductance of healthy controls and that of patients with QV was analyzed by Student’s t-test with adjustment for alpha inflation.
Correlation between skin conductance and age in normal controls, and correlation between the decrease of skin conductance and the score of QV in studied patients were investigated by linear regression analysis. Further multiple regression analysis on the decrease of skin conductance with the scores of symptoms or signs as independent variables was performed by the Least square method.
Results
The mean total score of QV and mean value of ARDK in patients with QV and in healthy controls
The mean total score of QV in patients with QV (7.9
5.3 points) was significantly greater than that of healthy controls (1.5
1.4 points; P < 0.001). In contrast, the mean value of ARDK in patients with QV (30.2
± 16.8
A) was significantly lower than that in healthy controls (37.7
± 10.8
A; P < 0.001). On the other hand, there was no significant difference in the mean value of ARDK between male group (38.2
± 10.2
A) and female group (36.6
± 10.5
A) in healthy controls.
Similarly, there was also no significant difference in the mean value of ARDK

between male group (31.2
± 15.2
A) and female group (30.4
± 13.5
A) in patients with QV.
The correlation between the mean value of ARDK and age in healthy controls
As shown in figure 2, the mean value of ARDK in healthy controls decreased gradually with the increase of age. There was an inverse correlation between mean value of ARDK and age (r-coefficient = -0.7; P < 0.001) with the following equation: mean value of ARDK = 69 – (0.8 X age).
The correlation between the mean value of skin conductance and score of QV in studied patients
Figure3 shows that the mean value of ARDK decreased with the increase in score of QV in studied patients. There was an inverse correlation between the mean value of ARDK and the score of QV (r-coefficient = - 0.61; P < 0.001).
The correlation between the decreased value of ARDK and the score of QV in studied patients
The decreased value of ARDK was calculated by the subtraction of the real mean value of ARDK in each studied patient from the expected value of ARDK adjusted by age according to the equation as described in Figure 2. There was a positive correlation between the decreased value of ARDK and score of QV in studied patients (r-coefficient = 0.89; P<0.001) (Figure 4).
Multiple regression analysis on the decreased value of ARDK with the scores of 5 symptoms/ signs and age as independent variables
The multiple regression on the decreased value of ARDK with the scores of 5 symptoms/ signs and age as independent variable is given in Table 1. Only the scores of weak pulse (beta-coefficient = 0.422, P < 0.001) and fatigue (beta-coefficient =
0.317, P < 0.001) were significantly correlated with the decreased value of ARDK, whereas spontaneous sweating, weak voice, enlarged tongue with dental impressions and age were not significantly correlated.

Discussion
Skin electrical conductance test has been used widely by psychologists as an indicator for psychological stress (Jacobs et al., 1994; Vogele and Steptoe, 1992;
Kelsey, 1991; Pirrelli et al., 1989). Several studies indicated that the skin electrical conductance could be influenced by multiple psychological factors such as anxiety
( Benjamins et al., 1994; Hoehn- Saric et al., 1991)), phobia ( Braune et al., 1994;
Hoehn-Saric et al., 1989) and schizophrenia (Dawson et al., 1992). It has also been used as a tool for biofeedback training (Steptoe A and Greer K., 1980).
The skin electrical conductance test (ARDK) used in this study is different from the traditional skin electrical conductance test used by psychologists. Psychologists usually performed the skin electrical test between two different fingers of one hand in a subject. However, ARDK was used to detect the skin electrical conductance between the left hand and each one of 24 special acupoints in both wrists and ankles.
Most of the special acupoints are the original points that express the energy of
Meridians. The ARDK test is modified from the electro-dermal screening test (EDST) proposed by Dr. Voll who used the level and the balance of skin conductance among various acupoints of four limbs to evaluate the energy status of human body (Voll,
1975; 1980). EDST has been widely used by traditional medical doctors in the evaluation of energy condition of patients. However, the reliability of EDST in clinical application is still unknown. Most of EDST, used for traditional Chinese medicine, are inconvenient to be used due to narrow detector and inability to calculate the mean value of skin conductance automatically.
To develop an objective method to detect QV, we previously used a bio-energy detector (VGH-82A, Kolon, Taipei) to measure the skin conductance between both palms and soles of patients with QV (Yu et al., 1998). We found that, the mean value of skin conductance between palms and soles can reflect the severity of QV.

However, the skin conductance between palms and soles is easy to be influenced by the psychological stress. Moreover, the bio-energy detector (VGH-82A) cannot calculate the mean value of skin conductance automatically. Due to narrow sharp detector, it is inconvenient for medical doctors or technicians to use. In contrast,
ARDK with a wide detector and a function to calculate the mean value of skin conductance automatically is easy to be performed.
The present study showed that the mean total score of QV in patients with QV was significantly greater than that of healthy controls. In contrast, the mean value of
ARDK in patients with QV was significantly lower than that of healthy controls.
These results reveal that patients with QV may have higher score of QV but lower skin conductance than healthy controls. The result that the mean value of ARDK in healthy controls decreased with the increase of age, suggesting that age is an important factor in the decrease of skin conductance during natural aging process in the human body. Although gender might be a factor to influence the skin conductance, we did not find a significant difference in the mean value of ARDK between male and female groups in both healthy controls and studied patients.
The other result that the mean value of ARDK in patients with QV had a significantly inverse correlation with the score of QV, suggests that more severe QV patients had, they had lower skin conductance. The result that the decreased mean value of ARDK after age adjusted in patients with QV had markedly positive correlation with the score of QV, further indicates that the decreased mean value of
ARDK after age adjusted may reflect the severity of QV.
The reason behind why the mean value of skin conductance decreases with age in healthy humans and further decreases in patients with QV is still unknown.
Electrodermal activity (EDA) can be influenced by the epidermis hydration
(Makhnev, 1997), which is based on the activities of sweat glands and their innervations by the sympathetic nerve (Rathkolb and Gallei, 1993; Freedman et al.,

1994). The skin sympathetic nerve activity (SNA) is caused mainly by central command, and appears to be targeted to both sweat glands and to vascular smooth muscles, with the relative targeting being temperature dependent (Vissing et al., 1991;
Vissing, 1997; Sequeira et al., 1995). Therefore, it is possible that the decrease in the mean value of ARDK with age in healthy humans may be due to the degeneration of sweat glands or the decrease of SNA in aging process. Furthermore, the decreased skin conductance in patients with QV may be due to the decrease in activity of sweat glands resulted from the attenuation of SNA.
The result of multiple regressions on the decreased mean value of ARDK with the scores of 5 symptoms/ signs and age as independent variable showed that, only the degrees of fatigue and weak pulse have contributed to the decreased mean value of ARDK. Fatigue is a symptom commonly appearing in patients with chronic diseases including cystic fibrosis and Behcet’s disease. A decrease of skin conductance was found in patients with cystic fibrosis (Williamson et al., 1985) and
Behcet’s disease (Akyol et al., 2002). These results were similar to our findings that skin conductance was lower in patients with tiredness and inversely correlated with the severity of QV. The significant correlations between skin conductance and fatigue or weak pulse may be due to the decrease of sympathetic nervous activity.
The latter not only can induce the decrease of skin conductance via reduced secretion of sweat gland but also can induce fatigue and weak pulse resulted from the reduced left heart ventricle contractility according to the Frank-Starling law of the heart.
Why the scores of other symptoms (such as spontaneous sweating, weak voice and enlarged tongue) have not contributed to the decrease of skin conductance in patients with QV is not known. In general, patients with QV may have spontaneous sweating, which can increase the skin conductance. However, in this study we did not find a significant correlation between the score of spontaneous sweating and the decreased mean value of ARDK. The discrepancy may be due to the low incidence of

spontaneous sweating in studied subjects and the situation of subjects during ARDK measurement. Usually, abruptly spontaneous sweating appears in patients with severe
QV during movement. Only 23% of our studied subjects had a complaint of spontaneous sweating during movement. None of them had occurrence of spontaneous sweating during ARDK measurement under a relaxed status lying on a bed after taking a rest for more than 20 minutes. The facts that the skin conductance in patients with QV did not correlate with the score of weak voice or enlarged tongue may be due to a more complex mechanism which induced these two symptoms rather than the decrease of sympathetic nervous activity.
It is possible that the change of skin conductance in the specific meridian would represent the specific illness. In this study, most of studied patients were carriers of hepatitis B or C with normal liver function. We compared the skin conductance of liver meridian with those of other11 meridians. However, no significant difference was found in skin conductance between liver meridian and other meridians (data not shown). Therefore, we could not identify the above hypothesis.
So far there are no papers concerning about a device which can be used as a tool to help the diagnosis of QV. Our results suggest that bio-energy of human body can be measured by skin conductance. It also suggests that, ARDK may be used as a complementary method for physicians to make a diagnosis of QV in clinical practice.
Conclusion
A novel method, namely ARDK, was used to detect the skin conductance between the left hand and each one of 24 special acupoints in patients with QV and healthy controls. The studied patients were further evaluated for the severity of QV with a score method. Our results showed that, the mean value of ARDK decreased with the increase of age in healthy controls and further decreased in patients with QV.
The decreased mean value of ARDK after age adjusted in patients with QV had a significantly positive correlation with the severity of QV. Our results suggest the

possibility of ARDK to be used as a complementary method to detect QV objectively.
Abbreviations: ARDK, Automatic Reflective Diagnosis System; EDA,
Electro-dermal activity; QV, qi deficiency; SNA: skin sympathetic nerve activity.
Competing interests: The authors declare that they have no competing interests.
Authors’ contributions: AB: Ming-Feng Chen; MT: Hsi-Ming Yu; ES: Ta-Jung You;
JY: Shu-Fang Li.
Acknowledgments: We here thank Dr. Anita Mannikarottu, Ph. D, Research Asst.
Professor, Albany Medical College, Albany, NY. USA, for the revision of this paper.

References
1.
Akyol, M., Turaclar, U., Kececi, H., Ozcelik, S., Marufihah, M., Erdal, S., and
Polat M. Electrodermal activities and autonomic nervous system in Behcet's patients. Neurol. Sci. 23(2):55-8, 2002.
2.
Benjamins, C., Schuurs, A.H., and Hoogstraten J. Skin conductance,
Marlowe-Crowne defensiveness, and dental anxiety. Percept Mot Skills. 79(1 Pt
2):611-22, 1994.
3.
Braune, S., Albus, M., Frohler M., Hohn T., and Scheibe G..
Psychophysiological and biochemical change in patient with panic attacks in a defined stituational arousal. Europ. Arch. Psychiatry & Clinical Neuroscience.
224(2):86-92, 1994
4.
Dawson, M. E, Nuechterlein, K. H., and Schell, A.M. Electrodermal anomalies in recent-onset schizophrenia: relationships to symptoms and prognosis.
Schizophr. Bull. 18(2):295-311, 1992.
5.
Freedman, L. W., Scerbo A. S., Dawson M. E., Raine A, McClure W.O., and
Venables, P. H. The relationship of sweat gland count to electrodermal activity.
Psychophysiol. 31(2):196-200, 1994.
6.
Hoehn-saric, R., Mcheod, D.R., and Zimmerli, W. D. Somatic manifestations in woman with generalized anxiety disorder. Psychophysiological responses to psychological stress. Arch. General Psychiatry. 46(12):1113-1119, 1989.
7.
Hoehn-Saric, R., McLeod, D.R., and Zimmerli, W. D. Psychophysiological response patterns in panic disorder. Acta Psychiatr Scand. 83(1):4-11, 1991.
8.
Jacobs, S.C., Friedman, R., Parker, J.D., Tofler, G. H., Jimenez, A. H., Muller, J.
E., Benson, H., and Stone, P. H. Use of skin conductance changes during mental stress testing as an index of autonomic arousal in cardiovascular research. Am.
Heart J. 128(6 Pt 1):1170-7, 1994.

9.
Kelsey, R .M. Electrodermal liability and myocardial reactivity to stress.
Psychophysiology. 28(6): 619-631, 1991.
10.
Makhnev, V. P. Dynamics of electrodermal activity during changes in the hydration of the epidermis: theoretical model. Biofizika. 42(1):201-8, 1997
11.
Pirrelli, A., Nazzaro, P., Sabato, G., Valente, A., Mudon,i A., and Massaro, R.
Neurovegetative assessment in subjects with a risk for hypertension. Am J
Hypertension. 2(2 Pt 2):24S-28S, 1989.
12.
Rathkolb, O., and Gallei, L. Biosignals of dermal activity. Wien Klin
Wochenschr. 150(8):228-31, 1993.
13.
Sen, Z. I., and Wang, W. Z. The referential for standard for the differential diagnosis of Shi-pattern in traditional Chinese medicine. Chin J. Integrated
Traditional and Western Med.; 6(10):598, 1986 (In Chinese).
14.
Sequeira, H., Ba-M'Hamed, S., and Roy, J. C. Fronto-parietal control of electrodermal activity in the cat. J Auton Nerv. Syst. 53(2-3):103-14, 1995.
15.
Steptoe, A., and Greer, K. Relaxation and skin conductance feedback in the control of reactions to cognitive tasks. Biol Psychol. 10(2):127-38, 1980.
16.
Vogele, C., and Steptoe A. Emotional coping and tonic blood pressure as determinants of cardiovascular responses to mental stress. J. Hypertension.
10(9):1079-87, 1992.
17.
Voll, R. Twenty years of electroacupuncture diagnosis in Germany: A progress report. Am. J. Acup. 3:17-22, 1975.
18.
Voll, R. The phenomenon of medicine testing in electroacupuncture according to
Voll. Am. J. Acup. 8(2):97-104, 1980.
19.
Vissing, S. F., Scherrer, U., and Victor, R.G.. Stimulation of skin sympathetic nerve discharge by central command. Differential control of sympathetic outflow to skin and skeletal muscle during static exercise. Circ. Res. 69(1):228-38, 1991.

20.
Vissing, S. F. Differential activation of sympathetic discharge to skin and skeletal muscle in humans. Acta Physiol. Scand Suppl. 639:1-32, 1997.
21.
Williamson, P. S., Fowles, D. C., and Weinberger, M. Electrodermal potential and conductance measurements clinically discriminate between cystic fibrosis and control patients. Pediatr. Res. 19(8):810-4, 1985.
22.
Yu, H.M., Chang, H.H., Liou, S.Y., Li, S.F., Hou, M.M., and Chen, M.F. The correlation between skin electrical conductance and the score of qi vacuity. Am J
Chin Med. 26(3-4):283-90, 1998.

Figure Legends
Figure 1 The equipment of ARDK. The equipment of ARDK included a metal plate, a detector and an ARDK device. Both of the metal plate and the detector were connected to the ARDK device, which could supply a weak electrical current with a constant voltage of 1.75V and was further connected to a personal computer.
Figure 2 The correlation between mean value of ARDK and age in 89 healthy controls.
Figure 3 The correlation between mean value of ARDK and score of qi vacuity in 193 patients with qi vacuity.
Figure 4 The correlation between decreased value of ARDK and score of qi vacuity in
193 patients with qi vacuity. The decreased value of ARDK was calculated by the subtraction of the real mean value of ARDK in each studied patient from the expected value of ARDK adjusted by age.

Table1.
Multiple regression analysis on the decreased value of ARDK in patients with qi vacuity.
Independent variables -coefficient P value
Weak pulse
Fatigue
0.422
0.317
Spontaneous Sweating
Weak voice
0.100
0.079
Enlarged tongue with dental impressions 0.045
Age 0.001
0.001
0.004
0.063
0.330
0.36
0.97

Figure 1
Figure 1

Figure 2
Fig. 2

Figure 3
Fig. 3

Figure 4
Fig.4