The purpose of this study was to investigate whether a relationship exists between mortality rates and serum antioxidant levels among Japanese inhabitants. The follow-up subjects, who participated in comprehensive health examinations, consisted of 2444 inhabitants (949 males and 1495 females) of a rural area in Hokkaido, Japan. Between 1991 and December 2000, 146 subjects (94 males and 52 females) died, with cancer accounting for 76 of these deaths (48 males and 28 females). Serum samples at fasting were collected at entry into the study, and serum levels of β- and α-carotenes, lycopene, β-cryptoxanthin, canthaxanthin, zeaxanthin/lutein, tocopherols, and retinol were measured separately by high-performance liquid chromatography (HPLC). The statistical analyses were conducted using the Cox proportional hazard model. Age- and gender-adjusted hazard ratios of the groups with high serum levels of lycopene, β-carotene, zeaxanthin/lutein, and total carotenoids compared to those with low serum levels were 0.36 (95% C.I: 0.19-0.69), 0.53 (0.29-0.95), 0.73 (0.43-1.25), and 0.52 (0.30-0.92) for cancers of all sites, and 0.44 (95% C.I: 0.28-0.69), 0.59 (0.39-0.90), 0.61 (0.40-0.93), and 0.50 (0.33-0.76) for all causes, respectively. Similar results were found after adjusting for gender, age, smoking habits, alcohol consumption, and serum levels of total cholesterol and glutamic pyruvic transaminase (GPT) activity. Moreover, after excluding mortality within the first three years of follow-up, the hazard ratios of subjects with high serum levels of lycopene, total carotenes, and total carotenoids were significantly and inversely associated with subsequent mortality from all causes and cancers of all sites after adjusting for gender, age, and serum levels of total cholesterol, α-tocopherol, and retinol. These results suggest that high serum levels of antioxidants, such as lycopene, β-carotene and zeaxanthin/lutein, play roles in preventing death from cancer and from all causes. However, high serum levels of tocopherols and retinol did not demonstrate clear associations with either low mortality rates from all causes or cancer of all sites.
|Number of pages||14|
|Journal||International Journal for Vitamin and Nutrition Research|
|Publication status||Published - 2002|
All Science Journal Classification (ASJC) codes
- Medicine (miscellaneous)
- Endocrinology, Diabetes and Metabolism
- Nutrition and Dietetics