Background:

the stratification of the intracranial aneurysms rupture risk is important to decide the strategy before those patients with aneurysms that are incidental or asymptomatic. There is no consensus to determine the performance of surgical intervention or medical follow-up of these patients.

Objective:

to develop a predictive instrument for incidental intracranial aneurysm rupture.

Methods:

a sample of 152 patients diagnosed by computed tomography angiography of ruptured (n=138) and unruptured (n=22) saccular intracranial aneurysms was included. The 160 images of intracranial aneurysms were studied. The 152 patients were randomly divided into a development group consisting of 95 patients, 100 aneurysm images, and a validation group consisting of 57 patients, 60 aneurysm images. Measurements and segmentations of the aneurysms were performed; nine morphological factors were obtained. A multivariate combination was performed, using multiple logistic regression, which expressed six predictive demographic, clinical and morphological factors obtained from the clinical records of the patients. The selection for inclusion of the factors was made from a consensus of 15 experts with more than 15 years of experience in the subject. A representative nomogram of the model with the significant predictors was made. Calibration and accuracy of the predictive instrument represented by a model and its nomogram were evaluated.

Results:

the instrument was made up of five predictors that were statistically significant associated with breakage in the multivariate analysis: female sex, aspect ratio, the greatest width of the dome, volume, and non-sphericity index. The nomogram showed good calibration and discrimination (training group: area under the curve = 99%; validation group area under the curve = 99% ).

Conclusions:

the predictive instrument, validated and represented by the nomogram, is a useful model to stratify the risk of aneurysm rupture. It can be used to monitor aneurysms considered to be of lower risk.

        · resumen en Español     · texto en Español     · Español ( pdf )