Abstract P357: Trajectories of Stroke Risk Factors Before Stroke Onset With a 24-year Follow-up of Japanese People Living in an Urban Area: The Suita Study

Introduction: Many studies investigated associations of stroke risk factors assessed at one point in time with stroke onset. However, few studies investigated growth curves (trajectories) of the stroke risk factors assessed at multiple points in time before stroke onset. Hypothesis: We assessed the hypothesis that people with stroke, compared with their controls, would have higher values of the stroke risk factors at multiple points in time and higher change rates by year (slopes) of those factors before stroke onset. Methods: The present study used a nested case-control design based on the Suita study that is a cohort study launched in 1989 with 15,746 community-dwelling participants in an urban city, Suita, Japan. During a 24-year follow-up, 201 cases (43.8% in women) were identified. Corresponding to the cases, 2010 controls (51.5% in women) matched by age (± 4 years) were identified by incidence density sampling. As the stroke risk factors, we included systolic blood pressure (SBP), diastolic BP (DBP), blood glucose level, body mass index, waist circumference, non-HDL cholesterol, and triglyceride assessed every two years on health check-ups. Linear mixed models were performed to investigate adjusted mean differences in the stroke risk factors between the stroke cases and the controls at 20, 15, 10, 5 years and the last assessment before stroke onset. Mean differences in slopes of the stroke risk factors before stroke onset were also investigated. Results: The linear mixed models showed that SBP and DBP were significantly higher in the cases than the controls at any points in time before stroke onset (the adjusted mean differences [95% CI] at the last assessment before stroke onset: 4.52 [1.4, 7.64] mmHg for SBP and 2.93 [1.23, 4.63] mmHg for DBP). However, slopes of SBP and DBP were not significantly different between the cases and the controls. We observed no significant difference between the cases and the controls in blood glucose level at 20 years before stroke onset (the adjusted mean difference: 2.24 [-0.26, 4.74] mg/dl). However, compared with the controls, the cases significantly had a steeper slop of blood glucose increase (the adjusted mean difference in the slope per year: 0.21 [0.02, 0.40] mg/dl; and the adjusted mean differences at 15, 10, and 5 years and the last assessment before stroke onset: 3.29 [1.31, 5.26], 4.34 [2.51, 6.16], 5.39 [3.25, 7.52], 6.44 [3.69, 9.19] mg/dl, respectively). We observed no significant differences of means and slopes in BMI, waist circumference, non-HDL cholesterol, and triglyceride between the cases and the controls. Conclusions: In conclusions, multiple assessments of the stroke risk factors on health check-ups can be useful to early identify people who will be likely to develop stroke by observing SBP and DBP at any points in time, and the trajectory of blood glucose level.