Purpose: To investigate the association of middle cerebral artery (MCA) bifurcation aneurysms with bifurcation morphology.

Materials and methods: 205 patients were enrolled, including 61 patients with MCA bifurcation aneurysms and 144 non-aneurysmal subjects. Aneurysmal cases were divided into types C (aneurysm neck on extension of the parent artery centerline) and D (deviating neck). The radius of the parent artery M1 (R_P), smaller (R_S) and larger (R_L) branches, smaller (φ_S) and larger (φ_L) lateral angles, bifurcation angle, and arterial tortuosity (T_S) and (T_L) were analyzed. Logistic regression and receiver operator characteristic (ROC) curve analysis were performed to identify risk factors and predictive values for MCA aneurysm presence and types. Results: In aneurysmal MCA bifurcations, bifurcating angle, T_S, T_L and R_L were significantly larger (P<0.01) while φ_S was smaller (P<0.001) than those in controls. The bifurcation angle, T_S and LogitP were better morphological parameters for predicting MCA aneurysm presence with the AUC of 0.795, 0.932 and 0.951, respectively. Significant (P<0.05) differences were observed in the bifurcation angle, φ_L, R_P, R_L and T_L between types C and D aneurysmal bifurcations. T_L was an independent factor in discriminating types C from D aneurysms with an AUC of 0.802.

Conclusions: Bifurcation angle and arterial tortuosity from the parent artery to the branch forming a smaller angle with the parent artery have a higher value in distinguishing MCA aneurysmal from non-aneurysmal bifurcations, and the tortuosity from the parent artery to the contralateral branch is the best indicator for distinguishing types C from D aneurysmal bifurcations.