This is the first study using BIA to demonstrate an association between suidoku and body composition and to identify factors associated with suidoku. Many problems related to a lack of objectivity have been reported for Kampo medicine. Lo et al. showed low rates of agreement in tongue diagnoses, particularly differences in judgment about tongue color, and reported the usefulness of an automated tongue diagnosis system [6]. They also stated that tongue diagnosis is often biased by the diagnostic skill level, experience and color perception. Furthermore, Ishida et al. reported that a skin moisture deficiency might be an indicator of blood deficiency [12]. Although the establishment of objective indices is desired, reports based on objective evaluations of suidoku are still lacking. Although there have been some reports about factors associated with suidoku, they analyzed subjective symptoms. Because no studies have been published that have attempted to objectively quantify suidoku [17, 18], this study was done using a body composition analyzer to identify factors associated with suidoku that would be useful in making a Kanpo diagnosis.
About 60 % of our new patients who presented for treatment with kampo medicines were female, and other studies have similarly reported a high proportion of female patients reporting to Kampo clinics [27, 28]. In our study, the subjective symptom analysis was done only with female patients because of the small number of male patients and in consideration of the difference in physical size.
The logistic regression analysis found significantly lower Percent Body Fat, Muscle Mass Index of the Arms, Muscle Mass Index of the Trunk, VFA and BMI, in the splashing sound group than in the non-suidoku group. Terasawa et al. studied the physiology of the stomach and the amount of gastric juice in a cohort of patients with splashing sounds and reported on the mechanisms and conditions related to the manifestation of splashing sounds [29]. They found that splashing sounds require an appropriate amount of air and gastric juices and that the occurrence of splashing sounds is associated with weak abdominal strength and a drooping of the stomach antrum. It is possible that low VFA indicates gastroptosis because fat supporting the internal organs is low. Also, low Muscle Mass of the trunk, which includes the abdominal muscles, may indicate weak abdominal strength. Thus, our results support a previous finding that splashing sounds occur when oscillation of gastroptosis is done through weak abdominal musculature. Furthermore, many of the patients in our splashing sound group had low BMI or Percent of Body Fat, thus splashing sounds may mainly occur in people who are thin. Some studies have reported that BMI was lower in patients with gastroptosis than in these without gastroptosis subjects and tends to be more common in people who are thin [30], so there is the possibility that people who manifest the splashing sound may tend to have gastroptosis.
By ROC analysis, we identified VFA ≤ 5.4, BMI ≤ 19.2 kg/m2, Percent Body Fat ≤ 27.8 %, Muscle Mass Index of the Trunk ≤ 6.5 kg/m2, as being associated with splashing sound. Muscle Mass Index of the Arms was not a useful indicator because of the low sensitivity, specificity and AUC (Se: 42 %, Sp: 80 %, AUC: 0.65). On the other hand, BMI and Muscle Mass Index of the Legs might be associated with edema. However the number of patients, 12, in this group was small, thus the data of more patients will have to be gathered for future analysis.
In the analysis of our questionnaire, the percent of participants responding “cold hands and feet” and “Need electric blankets or warmers in winter” was 91.7 and 75.0 %, respectively. Both were significantly high, suggesting the possibility of an association between splashing sound and “cold”. Yamato et al. reported that the sensation of feeling cold is stronger in patients whose body weight, BMI, skinfold thickness, percent body fat and amount of fat are low [31]. They suggested that body weight loss in the form of fat free mass loss, leads to reduced metabolism and reduced blood flow velocity, which induced the sensation of cold. Similarly, Yoshino et al. in a study of the sense of cold and its trends reported that a decrease of heat production by reducing muscle mass results in decreased basal metabolism, which leads to cold sensations [27]. In our study, BMI, percent body fat and the Muscle Mass Index of the Trunk of the splashing sound group was lower than that of the non-suidoku group. The association of a sense of cold and suidoku has been recognized in clinical practice. It is possible that the “cold” symptoms are caused by the distribution of water, which is due to the disruption of water circulation by the decreasing metabolism. Our results support the association of cold and suidoku. Furthermore, 83.3 % of the respondents complained of “dry skin”, which may represent an uneven distribution of water: a lack of water compared to the amount normally present in skin. In Kampo, it is understood that intestinal function affects the skin condition. Some papers have reported that patients have improved skin condition after taking Kampo medicine for gastrointestinal weakness [32, 33]. It is possible that people with splashing sounds trend to have gastroptosis and skin symptoms caused by poor digestion. The improvement of gastrointestinal problems may lead to improved skin condition. In the edema group, we obtained a high response rate for the subjective symptom “In the morning, sometimes have stiffness in hands, joints, or body” (edema group and non-suidoku group 58.3 and 20.0 % respectively). It is said that edema is an uneven distribution of water in stroma, thus stiffness may similarly occur throughout the body. Suidoku primarily signifies an excessive or uneven distribution of water and is understood to represent an irregularity of water balance. In this study, we found that both the splashing sound and edema groups had symptoms related to an uneven distribution of water.
A splashing sound can be heard on palpation of the abdomen, an abdominal diagnosis technique used in Sho diagnosis that was developed in Japan and is unique to Japanese Kampo medicine [34]. It is an important examination technique that provides extremely valuable information for making a Sho diagnosis. It is sometimes viewed with suspicion because it depends on the sensitivity of the doctor's hands and the patient’s reaction to compression of the abdomen. Recently, the development of a simulator to teach palpation of the abdomen and abdominal diagnosis has been proposed. Such a simulator would be desirable as a means of helping standardize sense related factors and improving proficiency [35], but there remain problems with objectivity. Therefore, it is hoped that the indicators of and the questionnaire for suidoku of our study will be helpful to doctors who are short on diagnostic experience.
There were some limitations to our study. First, the number of patients studied was small. Also, we were only able to study female patients, so it will be necessary going forward to perform more detailed research to determine if similar tendencies are observed in male as were observed in the female patients in our study. Second, there was little information on the electronic medical records because it was a retrospective analysis. For a diagnosis of suidoku, both dental indentation and swelling of the tongue are necessary. If the tongue has only dental indentation, it might indicate “Ki-kyo”. Unfortunately, there was too little information on swelling to be of use in this study. We had to exclude 36 patients, including some suspected of suidoku, in order to accurately do an objective assessment. Third, we did not consider the effect of possible confounders. The possibility that BMI may be a confounder cannot be denied. We performed only univariate regression analysis in this study because we judged that multivariate regression analysis would not be useful because of the number of patients and independent variables. Fourth, we did not consider possible differences in the ability of the BIA instrument.
Although InBody730 can display water balance, suidoku has no relevance to water balance. Therefore, we think that the body composition values we extracted in this study are accurate and can be verified by studies using BIA analyzers with similar features to the one used in this study but that do not display water balance. It will be necessary perform future research to address these issues. Furthermore, we discussed a possible association of splashing sound and gastroptosis, but no objective data was obtained for this association, making further study necessary.