Natto Intake is Inversely Associated with Osteoporotic Fracture Risk in Postmenopausal Japanese Women

ABSTRACT Background The direct association between intake of Japanese fermented soybeans, namely natto, and bone mineral density (BMD) is known. However, the association with osteoporotic fractures has not been studied. Objective This study aimed to investigate whether habitual natto intake is associated with a risk of osteoporotic fractures. Methods This prospective cohort study included 1417 postmenopausal Japanese women who were enrolled in the Japanese Population-Based Osteoporosis cohort study in 1996, 1999, 2002, and 2006 and were aged ≥45 y at baseline. The intake of natto, tofu, and other soybean products was surveyed with use of a FFQ at baseline. Fractures were ascertained in follow-up surveys conducted in 1999, 2002, 2006, and 2011/2012. Osteoporotic fracture was the primary outcome and was defined as a clinical fracture occurring without strong external force, diagnosed with radiographs by a medical doctor. HRs with 95% CIs were estimated with Cox proportional hazard models. Results During the 17,699 person-years of follow-up (median, 15.2 y), 172 women experienced osteoporotic fractures. After adjustment for age and BMD at the total hip, the HRs compared with those of < 1 pack (approximately 40 g)/wk natto intake were 0.72 (95% CI: 0.52, 0.98) and 0.51 (95% CI: 0.30, 0.87) for 1–6 and ≥7 packs/wk, respectively. After further adjustment for BMI, history of osteoporotic fractures, history of myocardial infarction or stroke, diabetes mellitus, current smoking, alcohol intake, frequency of tofu and other soybean product intakes, and dietary calcium intake, the HRs were 0.79 (95% CI: 0.56, 1.10) and 0.56 (95% CI: 0.32, 0.99) for 1–6 and ≥7 packs/wk, respectively. Frequency of tofu or other soybean product intakes had no association with the risk of osteoporotic fractures. Conclusions Habitual natto intake may be associated with a reduced risk of osteoporotic fractures independent of confounding factors, including BMD, in Japanese postmenopausal women. This trial was registered at umin.ac.jp as UMIN 000032869.