2015 Volume 43 Issue 2 Pages 110-117
UCAS Japan was conducted to clarify the natural course of unruptured cerebral aneurysms.
Method: According to a pre-specified protocol, 5, 720 patients with 6, 697 aneurysms were prospectively registered and rupture risks were analyzed.
Results: During 11, 660 aneurysm-years of follow-up, 111 aneurysms ruptured. The total annual rupture rate was 0.95%. A multivariate Cox regression model showed that size, location, and shape (association with bleb formation) were significantly associated with rupture. UCAS data were included in the meta-analysis used to develop the PHASES scoring system. In this scoring system, aneurysm rupture risks for the first 5 years can be estimated according to population, hypertension, age, size, earlier subarachnoid hemorrhage history, and aneurysm site.
Discussion: To conduct and publish large-scale cohort studies, it is important to: 1) never publish main results before the final paper, 2) fix the protocol and analysis plan prior to the study, 3) establish good collaboration with a statistician, and 4) raise a simple but important research question.