Introduction: Despite external ventricular drain (EVD) placement is known to be inaccurate, a few studies suggest that the conventional technique should be modified. Our goal was to assess if the ipsilateral medial canthus (IMC) should still be considered as the best external landmark or if other ones should be preferred.

Methods: The preoperative 3D-CT scan of 45 consecutive patients requiring EVD placement were selected if no medial shift was present and served to reconstruct virtual trajectories on a radiological workstation. All trajectories were drawn to pass through the Kocher’s point, considered as entry point and were all directed to the tragus in the coronal plane. They were then all directed variably in the sagital plane, respectively to the IMC, the nasion and the contralateral medial canthus (CMC) with the goal to determine their relationship with the frontal horn of the lateral ventricle. The ideal trajectory was subsequently defined as targeting the middle of the foramen of Monro and projected on the skin in the sagital plane.

Results: Trajectories directed to IMC, the nasion, and the CMC reached the frontal horn of the lateral ventricle respectively in 49%, 95%, and 93% of cases (p<0,001). These rates decreased respectively to 7%, 87%, and 87% when the Evans Index was lower than 0.2 (p<0,001). The projection on the skin of the 45 ideal trajectories was distributed along a gaussian curve centered 0.8mm away from the nasion on the ipsilateral side.

Conclusions: External landmarks should be adapted for every patient by reconstructing preoperatively the ideal trajectory on 3D-CT scan to maximize the accuracy of EVD placement. Otherwise, the nasion should be preferred to the IMC, giving the highest probability to target the frontal horn of the ventricle.

Patient Care: By improving the technique, we could improve the accuracy of EVD placement. The risk of misplacement and the number of complication could so be reduced for our patients.

Learning Objectives: By the end of this session, participants should be able to: 1)place EVD more accurately, 2)know the importance of reconstructing virtual trajectories before surgery in each case.

References: Muirhead et al, Br J Neurosurg, 2012 Oct Toma et al, Neurosurgery, 2009