• Rachel Kleinloog, Nikki de Mul, Bon H Verweij, Jan Andries Post, RinkelGabriel J EGJEBrain Center Rudolf Magnus, Depart-ment of Neurology and Neurosurgery, University Medical Center Utrecht, Utrecht, The Netherlands., and Ynte M Ruigrok.
• Brain Center Rudolf Magnus, Depart-ment of Neurology and Neurosurgery, University Medical Center Utrecht, Utrecht, The Netherlands.
• Neurosurgery. 2018 Apr 1; 82 (4): 431-440.

BackgroundIntracranial aneurysm rupture prediction is poor, with only a few risk factors for rupture identified and used in clinical practice.ObjectiveTo provide an overview of all the risk factors (including genetic, molecular, morphological, and hemodynamic factors) that have potential for use in clinical practice.MethodsWe systematically searched PubMed and EMBASE and focused on factors that can be easily assessed in clinical practice, might be used for rupture prediction in clinical practice, and/or are potential targets for further research. Studies were categorized according to methodological quality, and a meta-analysis was performed, if possible.ResultsWe included 102 studies describing 144 risk factors that fulfilled predefined criteria. There was strong evidence for the morphological factors irregular shape (studied in 4 prospective cohort studies of high-quality, pooled odds ratio [OR] of 4.8 [95% confidence interval 2.7-8.7]), aspect ratio (pooled OR 10.2 [4.3-24.6]), size ratio, bottleneck factor, and height-to-width ratio to increase rupture risk. Moderate level of evidence was found for presence of contact with the perianeurysmal environment (pooled OR 3.5 [1.4-8.4]), unbalanced nature of this contact (pooled OR 17.8 [8.3-38.5]), volume-to-ostium ratio, and direction of the aneurysm dome (pooled OR 1.5 [1.2-1.9]).ConclusionIrregular aneurysm shape was identified as a risk factor with potential for use in clinical practice. The risk factors aspect ratio, size ratio, bottleneck factor, height-to-width ratio, contact with the perianeurysmal environment, volume-to-ostium ratio, and dome-direction should first be confirmed in multivariate analysis and incorporated in prediction models.

11 keywords...

hide…