2014 年 78 巻 8 号 p. 1924-1927
Background: The Yo/Yin concept is fundamental to making a Kampo (sho) diagnosis and may be deeply related to the autonomic nervous system. There is, however, little objective data to confirm the validity of these concepts.
Methods and Results: After diagnosis using standardized Kampo techniques, 20 men and 67 women (mean age, 52.4 years) for whom the prescribed Kampo medication was effective were judged to be correctly classified as Yo- (n=49) or Yin-sho (n=38) and enrolled. Autonomic nervous function was assessed at first visit using HRV obtained from 24-h Holter ECG. Nocturnal ultra low frequency-1 (ULF-1, 0.0001–0.0003 Hz) and ULF-2 (0.0003–0.003 Hz) were significantly higher in the Yin-sho than in the Yo-sho group (P=0.030, P=0.016), suggesting a higher variation of autonomic nervous activity according to sleep stage. On multivariate analysis BMI (≥23.0 kg/m2) and ULF-1 (≥1,150 ms2) were identified as independent factors associated with a differential diagnosis of Yo- or Yin-sho (odds ratio [OR], 11.63, P=0.002; OR, 0.30, P=0.038, respectively). When the sleep period was divided into 3 phases, the ULF-1 of the Yin-sho group was significantly higher than that of the Yo-sho group in the late phase of sleep (P=0.023).
Conclusions: On heart rate variability analysis there was a sleep stage-related difference in the autonomic nervous activity of patients treated with standard Yo- and Yin-sho Kampo medicines. (Circ J 2014; 78: 1924–1927)
Traditional Chinese medicine was imported from China around the sixth century, disseminated throughout Japan, and later modified into the domestic version currently in use called Kampo, which includes the mixing of and treatment with herbal medicines. The distinction between Yo-sho and Yin-sho (hyper-function and hypo-function) is one of the most fundamental concepts in making an appropriate diagnosis. For example, if patients remain active in spite of their disease condition, they are said to be in Yo-sho. Conversely, if patients are inactive, they are classified as Yin-sho. These concepts have been closely linked to autonomic nervous activity, with some papers reporting normalization of the autonomic nervous function in patients taking Kampo medicines.1–6
The analysis of heart rate variability (HRV) can be used as a non-invasive tool for the assessment of the autonomic nervous function and cardiovascular risk.7–9 In this pilot study, we, for the first time, utilized HRV to objectively investigate the association between autonomic nervous function and the use of Kampo medicines, especially Yo- and Yin-sho.
Kampo medicine was prescribed at the first visit to 182 patients (≥20 years old) who visited the Kampo Medicine Clinic of Kyushu University Hospital between January 2008 and March 2012. Prescription was based on a diagnosis using standard Kampo techniques that include a judgment of Yo- or Yin-sho according to the definitions in a standard Kampo textbook.10 If the prescription was considered effective at the next visit, the diagnosis of Yo- or Yin-sho for each patient was judged to be appropriate and they were allocated to a Yo-sho (group A) or a Yin-sho (group B) group. This classification of patients is standard in Kampo medicine, and is somewhat subjective even if the physician strictly follows the standard method. Using this method, 87 patients (20 men, 67 women; mean age, 52.4 years) were considered to be appropriately classified and were enrolled in the study. None of the patients met the exclusion criteria, which included sleep apnea syndrome, diagnosed on patient history; and symptomatic heart disease, including the use of a pacemaker. The study was conducted in accordance with the principles of the Declaration of Helsinki and was approved by the Ethics Committee of Kyushu University Hospital.
Electrocardiogram (ECG) MonitoringTwo-channel 24-h Holter ECG monitoring (Nihon Koden, Tokyo, Japan) was done before the start of Kampo treatment, and the signals were stored on a personal computer. Beat-by-beat cardiac cycle length data were obtained via off-line computer analysis. The maximum entropy spectral analysis method (Mem Calc; Suwa Trust, Tokyo, Japan) was used to calculate HRV11 in the frequency range 0.0001–0.5 Hz (total frequency; TF). TF was divided into 4 frequency bands of interest: ultra low frequency (ULF), 0.0001–0.003 Hz; very low frequency (LF), 0.003–0.04 Hz; LF, 0.04–0.15 Hz; and high frequency (HF), 0.15–0.40 Hz. The complexity of HRV over minutes to hours was examined by plotting the log-transformed spectral amplitude against the log-transformed frequency between 0.0001 and 0.01 Hz. The slope of this line was taken as an index of the fractal component of HRV.
Statistical AnalysisData are expressed as number (%), mean±SD, or median with quartiles (25–75%). Patient characteristics and the HRV measurements using Holter ECG were compared between the Yo-sho and Yin-sho groups using Fisher’s exact test for categorical variables and unpaired t-test, Mann-Whitney test for continuous variables (univariate analysis) and repeated-measures analysis of variance. Variables with a P<0.1 difference between Yo-sho and Yin-sho on univariate analysis were used in multivariate analysis to determine independent, significant predictors for the sho. Odds ratio (OR) and 95% confidence interval were calculated from the multiple logistic regression model after adjustment with each variable. All statistical analysis was done on a personal computer using SPSS for Windows (PASW 18, Tokyo).
Table 1 lists the baseline clinical characteristics of the enrolled patients, 49 of whom were assigned to group A (Yo-sho) and the remaining 38 to group B (Yin-sho) according to the effectiveness of the prescribed Kampo medicine, as described in the previous section. Body mass index (BMI) was slightly but significantly higher in group A than group B (P=0.015), but the other baseline characteristics were the same.
| Group A (n=49) | Group B (n=38) | P-value | |
|---|---|---|---|
| Sex (M/F) | 12/37 | 8/30 | 0.705 |
| Age (years) | 51.4±13.3 | 53.7±15.8 | 0.465 |
| BMI (kg/m2) | 21.7±3.2 | 20.2±2.4 | 0.015 |
| Temperature (°C) | 36.6±0.3 | 36.6±0.5 | 0.946 |
| Sleep time (min) | 452.4±83.0 | 455.3±112.2 | 0.891 |
| HT or DM | 9 (18.4) | 4 (10.5) | 0.309 |
Data given as mean±SD or n (%).
BMI, body mass index; DM, diabetes mellitus; group A, Yo-sho; group B, Yin-sho; HT, hypertension.
No significant difference in the HRV of groups A and B was found in the awake period. Mean ULF, however, was significantly higher in group B than in group A in the sleep period (P=0.009). Furthermore, even when ULF was divided into ULF-1 (0.0001–0.0003 Hz) and ULF-2 (0.0003–0.003 Hz), the mean frequency of both were significantly higher in group B than in group A (P=0.030, 0.016 respectively; Table 2).
| Awake | During sleep | |||||
|---|---|---|---|---|---|---|
| Group A (n=49) | Group B (n=38) | P-value | Group A (n=49) | Group B (n=38) | P-value | |
| TF (ms2) | 5,337.8±383.3 | 5,953.8±435.2 | 0.291 | 5,457.7±434.1 | 6,336.9±493.0 | 0.184 |
| ULF (ms2) | 3,471.0±261.7 | 3,988.6±297.1 | 0.195 | 2,679.0±230.5 | 3,616.1±261.8 | 0.009* |
| ULF-1 (ms2) | 1,106.0±109.9 | 1,474.2±124.8 | 0.030* | |||
| ULF-2 (ms2) | 1,573.0±153.8 | 2,141.9±174.2 | 0.016* | |||
| VLF (ms2) | 1,268.9±101.5 | 1,399.9±115.2 | 0.396 | 1,949.1± 240.4 | 2,289.3±273.0 | 0.352 |
| LF (ms2) | 365.4±39.0 | 402.1±44.3 | 0.535 | 498.8±62.4 | 462.5±70.9 | 0.702 |
| HF (ms2) | 123.2±20.8 | 156.8±23.6 | 0.287 | 374.8±60.7 | 335.2±68.9 | 0.668 |
| LF/HF | 3.9±0.4 | 3.8±0.4 | 0.956 | 2.1±0.3 | 2.1±0.3 | 0.918 |
Data given as mean±SD. ULF-1, 0.0001–0.0003 Hz; ULF-2, 0.0003–0.003 Hz; VLF, 0.003–0.04 Hz; LF, 0.04–0.15 Hz; HF, 0.15–0.4 Hz.
HF, high frequency; LF, low frequency; TF, total frequency; ULF, ultra low frequency; VLF, very low frequency.
To determine the relative influence of the background factors for predicting a patient being in group A or group B, univariate followed by multivariate analysis was done. Univariate analysis extracted BMI, ULF-1, and ULF-2 as significantly associated with a differential diagnosis of group A or group B. Only BMI and ULF-1 (OR, 11.63, P=0.002; OR, 0.30, P=0.038, respectively; Table 3) were extracted in multivariate analysis.
| Odds ratio | 95% CI | P-value | |
|---|---|---|---|
| Age (years) ≥60 | 0.504 | 0.169–1.432 | 0.205 |
| Male | 1.001 | 0.294–3.420 | 0.998 |
| BMI (kg/m2) ≥23 | 11.625 | 2.890–79.255 | 0.002* |
| ULF-1 (ms2) ≥1,150 | 0.299 | 0.090–0.906 | 0.038* |
| ULF-2 (ms2) ≥2,200 | 1.226 | 0.356–4.400 | 0.748 |
ULF-1, 0.0001–0.0003 Hz; ULF-2, 0.0003–0.003 Hz.
BMI, body mass index; CI, confidence interval; ULF, ultra-low frequency.
When the total sleep period was divided into early, middle, and late phases of equal length, the ULF-1 was the same in the early period, but became higher in group B than in group A in the middle and late phases (P=0.038). It eventually became significantly higher in the late phase of sleep (P=0.023; Figure).
Mean ultra-low frequency (ULF)-1 was the same in the early period, but became significantly higher in (
) group B (Yin-sho) than in (
) group A (Yo-sho) in the middle and late phases (P=0.038, between all group A and B, 2-way ANOVA;*
P=0.023 at late phase).
This is the first report to demonstrate that objective autonomic nervous balance expressed by HRV is clearly different during the sleep of patients diagnosed as Yo-sho or Yin-sho after strictly applying traditional Kampo diagnostic techniques.
Another research group also used HRV to investigate the effect of Kampo medicines on the autonomic nervous system, and reported an effect of ephedra on autonomic nervous modulation. The ingestion of ephedra dry tract tilted the sympathovagal balance toward increased sympathetic activity.12,13 No data, however, on the usefulness of HRV in making a Kampo diagnosis of Yo- or Yin-sho has been published to date. Also, several trials involving objective Kampo diagnosis of Oketsu syndrome have reported that α-sympathetic nervous activity is increased in the Oketsu state.14–17 Again, however, there were no objective data on the differences between Yo-sho and Yin-sho diagnosis, one of the most fundamental aspects of the diagnosis and the prescription of Kampo medicines. Although an increasing number of physicians, even in Western countries, have been using complementary and alternative medicine (CAM) including Kampo medicine in recent years18 and might have found in their experience that CAM includes the detection of factors affecting autonomic nervous activity to restore the disturbed balance, there was very little objective evidence to show that the diagnostic techniques of Kampo can truly be used to classify different types of patients in terms of autonomic nervous activity, using Western methodology. The present data have shown that HRV analysis would be a helpful tool that clinicians can use in daily medical practice to differentiate patients who need Yo-sho and Yin-sho medicine. A significant, objective difference was seen in autonomic nervous activity, especially during sleep and probably involving the sympathetic nervous system judging from the ULF band, determined using the relatively easy HRV technique. The present data show that this technique makes it feasible for clinicians to easily differentiate between the patients for whom Yo-sho or Yin-sho medicines would be beneficial.
We divided ULF into ULF-1 and ULF-2 because we are more interested in the lower frequency component of ULF-1, which represents 1–2 h of changes in autonomic nervous activity. Any changes in this length of time represent a change in the sleep cycle, because the sleep cycle appears to be around 90 min,19 which is included in the ULF-1 band. We have shown that the ULF-1 power during sleep is higher in the Yin-sho patients. ULF has been shown to be a marker of physical activity,20 and the sleep cycle is well known to be closely related to changes in autonomic nervous activity that directly affect HR and, hence, HRV. The increase in the power of ULF-1 of the Yin-sho patients could indicate that the swing in the autonomic nervous system of Yin-sho patients is larger than that of Yo-sho patients, indicating a relatively unstable state of the autonomic nervous system in this patient group. We consider this to be related to the basic physiological status of the patients and not related to an effect of the Kampo medicine that was prescribed.
From the point of view of Kampo, the Yin-sho patients are relatively inactive and the parasympathetic nerve is predominant. The present results, however, show that there is no direct increase in the parasympathetic nervous system expressed as an increase in the amplitude of the power of the HF band or a decrease in LF/HF. This indicates that the characteristics of Yin-sho patients lie not in high parasympathetic tones but in the fluctuating nature of parasympathetic control of HR during sleep. In general, non-rapid eye movement (non-REM) sleep usually appears relatively more in the early phase of sleep and REM sleep increases in the late phase.19 Given that the difference of the power in ULF-1 in the present study became prominent during the late phase of sleep, the difference in Yon-sho and Yin-sho may be related to REM sleep.
It is difficult in medical clinics to treat patients with medically unexplained or non-specific physical symptoms using Western-style medicine.21 Thus, these patients might be good candidates for Kampo medicine, given that most such patients have autonomic nervous imbalance and, empirically, Kampo has been shown to successfully restore the balance of the autonomic nervous system. The present HRV analysis will enable better understanding of the patient condition and make more objective Kampo diagnosis and treatment possible. We believe that accumulation of such knowledge is very important for the further understanding of Kampo medicine.
There were some limitations to this study. First, the number of participants was small, especially that of male patients. The fact, however, that differences in ULF power between Yo- and Yin-sho patients could be clearly observed suggests that this number may have been sufficient for this particular purpose. Larger prospective studies with more male patients are necessary to further clarify the differences and to obtain robust cut-offs. Second, we did not analyze the association of autonomic nervous parameters and other concepts of Kampo, such as Ki, Ketsu, and Sui. Another study specifically targeting these concepts would be useful for clarifying the impact of these concepts on the autonomic nervous system and on sleep.
Finally, we did not compare HRV before and after Kampo. To examine the changes in HRV after treatment is of fundamental importance to understanding the underlying mechanisms of the difference of HRV in Yo-sho and Yin-sho patients. Therefore, further experiments including a post-treatment analysis must be done to provide more detailed information on the relationship between Kampo therapy and autonomic nervous balance.
On HRV a difference was seen in the autonomic nervous balance of patients diagnosed as Yo- or Yin-sho, based on traditional Kampo methodology. This difference occurred during sleep, especially in the late phase. This simple HRV method would be useful for classifying patients into Yo or Yin groups to more efficiently and effectively prescribe Kampo medicine.
No conflict to disclose.
Grant: None.
