Abstract
Three percent of the population harbors an intracranial aneurysm. A minority of these aneurysms will rupture and cause a subarachnoid hemorrhage (SAH). SAH is a devastating disease with high case fatality and morbidity. A major contributor to the poor outcome after SAH is delayed cerebral ischemia (DCI). About 30% of
... read more
the SAH patients develop DCI, but it is difficult to predict which patients. This thesis focuses on new insights on the incidence of SAH, the etiological pathway of aneurysm rupture, and early prediction of DCI. We performed a meta-analysis including 52 studies on the incidence of SAH. Overall, the incidence was 9.2 per 100,000 person-years, whereas incidence rates were doubled in Japan and Finland. Another main finding was that at young ages the incidence is higher in men, whereas after the age of 55 years it is higher in women. In another study, we compared several morphological characteristics on CT-angiograms in patients with ruptured and unruptured aneurysms. The main finding was that direction of flow into the aneurysm and nonspherical shape are potential new risk factors for aneurysm rupture. In DCI, we can distinguish between two infarct patterns: single infarction and infarction in multiple cerebral vascular territories. By comparing the risk profiles of SAH patients we found that these different infarct patterns are not distinct disease entities, but different degrees of the same pathophysiological process. Up to now it is not possible to predict easily and accurately which patients will be affected with DCI. Established predictors of DCI are large amount of subarachnoid blood and poor clinical condition on admission. In a review we systematically summarized the evidence on other predictors. We included 52 studies on early clinical, laboratory and radiological predictors routinely available after SAH and categorised the studies according to methodological quality. Our main finding was the strong evidence for smoking as a predictor of DCI. For several suggested predictors more research is needed to assess whether or not they are predictors of DCI. Early systemic inflammatory response syndrome, hyperglycaemia on admission, hydrocephalus, and history of diabetes may prove to have additional predictive value. Next, we analysed and compared multiple predictive models for development of DCI in patients with SAH. We developed a practical risk chart for the development of DCI. The best model to predict DCI consisted of 4 predictors present on admission and easy to assess: clinical condition, amount of cisternal blood, amount of intraventricular blood, and age. Additional predictors were smoking and hyperglycaemia, but both contributed little to improvement of the model performance. Low or high risk of DCI can be reliably predicted by a simple risk chart including these 4 predictors. The risk chart presents an absolute risk between 12 and 61% for the individual patient. In the same study, we performed a validation study in a more recent SAH population showing an equal area under de receiver curve of 0.69. The risk chart can be used for decision making regarding intensity of monitoring and treatment in patients with SAH
show less
