Quality Assurance During Brain Aneurysm Microsurgery—Operative Error Teaching

Marcelo Magaldi Oliveira, Carlos Eduardo Ferrarez, Renan Lovato, Pollyana Vieira Costa, Jose Augusto Malheiros, Leonardo Avellar, Manuel Granja, Eric Sauvageau, Carla Machado, Ricardo Hanel

World Neurosurgery, Available online 7 June 2019


Introduction Quality assurance (QA) is a way to prevent mistakes in advance. Although it has been previously reported for surgical setup, there is no effective approach for minimizing microsurgical technical errors before an operation is done. Neurosurgery resident operative errors during brain aneurysm surgery could be foreseen by practicing in an ex vivo hybrid simulator with microscopic fluorescein vessel flow image.

Methods Five vascular neurosurgeons and 8 junior/senior neurosurgical residents voluntarily joined this research initiative. The following methodology was adopted: 1) Identification of the 7 most-common resident operative performance errors during brain aneurysm surgery; 2) Design of exercises to prevent common mistakes in brain aneurysm microsurgery using a placenta simulator; and 3) Blinded staff neurosurgeon evaluation of resident performance during real brain aneurysm microsurgery.

Results All key steps to perform such intervention were accomplished with a simulator that uses 2 placentas, a synthetic cranium, and microscopic fluorescein vessel flow image. Neurosurgery residents trained in this model had better surgical performance with fewer perioperative mistakes (P < 0.05). Fine microsurgical dissection of the arachnoid membrane and aneurysm sac were the most commonly improved tasks among the 7 common operative mistakes. Brain parenchyma traction with secondary bleeding was the only error not prevented after previous simulator training.

Conclusions There was a left-shift on the quality assurance line with residents who practiced brain aneurysm microsurgical errors in an ex vivo model. A multicentric prospective study is necessary to confirm the hypothesis that real operative error could be reduced after training in a realistic simulator.