Osteoporosis and cognitive impairment, which are highly prevalent conditions in elderly populations, share several risk factors. This study aims at evaluating the association of bone mineral density (BMD) with prevalent and incident cognitive impairment after a 3-year follow-up. We studied 655 community-dwelling women aged 65+ participating in the InCHIANTI study, who had been followed for 3 years. Total, trabecular, and cortical BMD were estimated by peripheral quantitative computed tomography using standard transverse scans at 4 and 38 % of the tibial length. Cognitive performance was evaluated using the Mini-Mental State Examination and the Trail Making Tests (TMT) A and B; a MMSE score <24 was adopted to define cognitive impairment. The TMT A-B score was calculated as the difference between TMT-A and TMT-B times (ΔTMT). The association of cognitive performance after 3 years with baseline indices of BMD was assessed by logistic and linear regression analyses. Cortical, but not trabecular, BMD was independently associated with incident cognitive impairment (OR 0.93, 95 % CI 0.88-0.98; P = 0.012), worsening cognitive performance (OR 0.96, 95 % CI 0.92-0.98; P = 0.039), and worsening performance in ΔTMT (OR 0.96, 95 % CI 0.92-0.99; P = 0.047). Increasing cortical BMD tertiles was associated with decreasing probability of incident cognitive impairment (P for linear trend =0.001), worsening cognitive performance (P = 0.013), and a worsening performance below the median value (P for linear trend <0.0001). In older women, low BMD might represent an independent and early marker of subsequent cognitive impairment. Physicians should assess and monitor cognitive performance in the routine management of elderly women with osteoporosis.
Keywords: Bone mineral density; Cognitive performance; Elderly; Epidemiology.