BACKGROUND: Kanemi Yusho was a food poisoning incident caused by rice bran oil that occurred in western Japan, particularly in northern Kyushu, in 1968. It is difficult to determine the symptoms in patients after many years since the occurrence. Techniques for measuring blood dioxins have advanced recently. More accurate data measurement has now become possible, and techniques for mass data analysis, such as data mining, have also advanced. It has also become possible to find unknown characteristics, even in an object group with elusive characteristics, by checking all the combinations in all the patients. AIM: There are already several reports on the incidence of symptoms in Yusho patients. These reports are limited to symptoms in a single period, and there has been no analysis taking into account the time that has elapsed. Here, we evaluated the relationship between recent and past symptoms and 2,3,4,7,8-penta-chlorodibenzofuran (PeCDF) levels in the same subject patient, in order to demonstrate the correlation between PeCDF levels and symptoms at a time point close to the incident. METHODS: Subjects were examined for symptoms of Yusho and had blood PeCDF levels measured, both recently and in the past. Combinations were extracted using association analysis of data mining technique for comparison, which had strong correlations between the presence or absence of symptoms in the medical examination, tests including blood test, dermatological examination, dental examination and ophthalmologic examination in recent years (2001-2004) and the blood PeCDF levels and those between the presence or absence of past (1986-1989) symptoms and recent blood PeCDF levels. RESULTS: Subjects with higher PeCDF levels were more likely to present with pigmentation, a symptom included in the diagnostic criteria for Yusho. Pigmentation was a commonly found symptom in the past. Past pigmentation was a common symptom in the present. DISCUSSION: PeCDF levels were measured recently and therefore should not be compared directly with past symptoms. However, among the symptoms included in the diagnostic criteria, past symptoms tended to have a stronger relationship with PeCDF levels than did recent symptoms. We suggest that the present PeCDF level is strongly related to the past symptoms since the present PeCDF level is correlated with the past PeCDF level due to constant emission rate. More specifically, if the past PeCDF level is strongly related to the past symptoms, it can indirectly be concluded that the present PeCDF level is also strongly related to the past symptoms. Thus, the present PeCDF level cannot be related to the past symptoms directly but can indirectly. CONCLUSIONS: Combining recent and past symptoms further demonstrated that clinical symptoms are strongly related to PeCDF toxicity. This may have resulted from the increase in symptoms with aging, however, it was demonstrated that the symptoms of each patient were relieved and become obscure.
|Number of pages||7|
|Journal||Fukuoka igaku zasshi = Hukuoka acta medica|
|Publication status||Published - May 2007|
All Science Journal Classification (ASJC) codes