TAKESAKI Toshiro Cancer Institute of Aichi Prefecture, Researcher, 研究所, 研究員 (50227013)
NAGAO Toru Aichigakuin Unversity, Oral Maxillofacial Surgery, Lecturer, 歯学部, 講師 (90261007)
IKEDA Noriaki International Medical Center of Japan, Medical Officer, 地域医療研究室, 研究員 (30150791)
RAZAT I.A. University of Malaya, Oral Medicine and Periodontology, Professor, 歯学部, 教授
ITO Yshinori Fujita Health University School of Health Science, Professor, 衛生学部, 教授 (50087665)
RAZAT I.A. マラヤ大学, 歯学部, 教授
ZAIN R.B. マラヤ大学, 歯学部口腔病理学教室, 助教授
宮崎 秀夫 新潟大学, 歯学部予防歯科学講座, 教授 (00157629)
|Budget Amount *help
¥6,600,000 (Direct Cost : ¥6,600,000)
Fiscal Year 1998 : ¥3,100,000 (Direct Cost : ¥3,100,000)
Fiscal Year 1997 : ¥3,500,000 (Direct Cost : ¥3,500,000)
For the selected subjects, epidemiological data of individual habit characteristics such as smoking, quid chewing and alcohol drinking were obtained from the same questionnaires recorded in the survey forms as for screening program. Venous blood samples were taken from subjects. Informed consent was obtained from all subjects (All selected subjects consented to this procedure by singing or thumb-printing the prepared consent form). The blood samples were centrifuged immediately to separate the serum from the cellular components. The serum samples were stored in -70℃ in freezer in University of Malaya prior to analysis of the serum micronutrients. The serum concentrations of retinol, α-tocopherol, lycopene, zeaxianthin/lutein , criptoxanthin, α-carotene, and β-carotene were determined using high performance liquid chromatography.
The diagnostic criteria for oral precancer (leukoplakia, erythroplakia and oral submucos fibrosis were based on the recommendation by WHO in 1998. The diagnosti
c criteria for lesions were also based on WHO in 1998 and criteria described by Axcell which were in accordance with the WHO's application of the International Classification of Diseases to Dentistry and Stomatology.
The prevalence of oral mucosal lesions in Indians was 40.5% (N=255) . Oral cancer prevalence was 0.2% (1) , leukoplakia 10.9% (95) , lichen planus 0.7% (21) , oral submucous fibrosis 0.3% (13) , erythroplakia 0.5% (3) , chewer's mucosa 20.7% (133) and other non-OPC lesions 0.4% (13) . A multivariate analysis using logistic regression with OPC as the dependent variable showed significance for 2 independent variable namely, quid chewing and serum levels of α-carotene with 30.2 and 0.5 of odds ratio, respectively.
In indigenous people in Sarawaku, the prevalence of oral mucosal lesions was 30.1% (N=203) . There were 23 (0.7%) subjects with OPC, 88 (14.3%) with chewer's mucosa and 129 (20.0%) with other non-OPC lesions. The OPC lesions consisted of 2 (0.3%) cases of oral cancer, 20 (0.3%) cases of oral leukoplakia, 1 (0.2%) case of oral submucous fibrosis.
In Malays (N=276) , the prevalence of oral mucosal lesions was 20.3%. There were 4 (0.4%) subjects with OPC, 30 (10.3%) with chewer's mucosa and 21 (20.0%) with other non-OPC lesions. The OPC lesions consist of 2 (0.3%) cases of oral leukoplakia, 2 (0.2%) cases of oral lichen planus.
The logistic regression analyses showed that the Indians are 8 times more likely to have OPC as compared to the Indigenous people. When oral leukoplakia was used as the dependent variable, significance for 2 independent variables were also evident namely, the Malays ethnicity and quid chewing. Similar to OPC, support for the bivariate analyses of positive correlation between oral leukoplakia and quid chewing was observed where subjects with quid chewing habit are significantly more likely to suffer from oral leukoplakia (coefficient=0.81), i. e., about 123 times more likely to have oral leukoplakia (odds ratio=120.6). This mulitvariate analysis also showed that Malays are significantly less likely to suffer from oral leukoplakia as compared to Indigenous people (coefficient=-0.245), i. e., the Malays are one-tenth times less to suffer from oral leukoplakia as compared to Indigenous people (odds ratio=0.106). The independent variable used here can only account for about 60% of oral leukoplakia as the dependent variable (Pseudo R2=0.603). Less