Selecting subjects for ultrasonographic screening for aneurysms of the abdominal aorta: four different strategies

Background We studied whether the effectiveness of ultrasound screening for abdominal aortic aneurysms could be increased by preselecting high-risk su bjects, based on the presence of risk indicators for the disease. Methods In a population-based screening programme for abdominal aortic aneu rysms among 5328 subjects living in Rotterdam, The Netherlands, we studied four different strategies to select subjects for ultrasound screening of th e abdominal aorta, based on risk indicators for abdominal aortic aneurysm d isease. Risk indicators used in each strategy were entered in a logistic re gression model to predict the probability of an individual having an abdomi nal aortic aneurysm. Using several cutoff values for the probability of a s ubject having an aneurysm for each strategy, we estimated the proportion of subjects that should be referred for ultrasound screening and the proporti on of aneurysms that would be diagnosed by each strategy (sensitivity). Results When a probability of 1.5% of having an aneurysm is chosen as the c utoff point above which ultrasound screening is indicated, the proportion o f subjects that would be referred for screening ranged from 36% (first stra tegy) to approximately 50% (other strategies), while 80% (first strategy) t o approximately 94% (other three strategies) of all aneurysms would be dete cted. Conclusion Effectiveness in screening for abdominal aortic aneurysms can be increased by selecting subjects by means of a short medical questionnaire, filled out by the screening candidates, including questions on medical his tory.