Neurosurg Focus 26 (5):E20, 2009

Minimally invasive superficial temporal artery to middle cerebral artery bypass through a minicraniotomy: benefit of three-dimensional virtual reality planning using magnetic resonance angiography

Ge r r i t Fi s c h e r , M.D.,1 Ax e l St ad i e , M.D.,1 Ei k e Sc h w a n d t , M.D.,1 Jo a c h i m Ga w e h n , M.D.,2 St e p h a n Bo o r , M.D.,2 Ju e r g e n Ma r x , M.D.,3 a n d Jo a c h i m Oe r t e l , M.D.1 1Neurochirurgische Klinik und Poliklinik, 2Institut für Neuroradiologie, and 3Klinik und Poliklinik für Neu- rologie, Universitaetsmedizin, Johannes-Gutenberg-Universitaet, Mainz, Germany

Object. The aim of the authors in this study was to introduce a minimally invasive superficial temporal artery to middle cerebral artery (STA-MCA) bypass surgery by the preselection of appropriate donor and recipient branches in a 3D virtual reality setting based on 3-T MR angiography data. Methods. An STA-MCA anastomosis was performed in each of 5 patients. Before surgery, 3-T MR imaging was performed with 3D magnetization-prepared rapid acquisition gradient echo sequences, and a high-resolution CT 3D dataset was obtained. Image fusion and the construction of a 3D virtual reality model of each patient were completed. Results. In the 3D virtual reality setting, the skin surface, skull surface, and extra- and intracranial arteries as well as the cortical brain surface could be displayed in detail. The surgical approach was successfully visualized in virtual reality. The anatomical relationship of structures of interest could be evaluated based on different values of translucency in all cases. The closest point of the appropriate donor branch of the STA and the most suitable recipi- ent M3 or M4 segment could be calculated with high accuracy preoperatively and determined as the center point of the following minicraniotomy. Localization of the craniotomy and the skin incision on top of the STA branch was calculated with the system, and these data were transferred onto the patient’s skin before surgery. In all cases the preselected arteries could be found intraoperatively in exact agreement with the preoperative planning data. Suc- cessful extracranial-intracranial bypass surgery was achieved without stereotactic neuronavigation via a preselected minimally invasive approach in all cases. Subsequent enlargement of the craniotomy was not necessary. Perioperative complications were not observed. All bypasses remained patent on follow-up. Conclusions. With the application of a 3D virtual reality planning system, the extent of skin incision and tissue trauma as well as the size of the bone flap was minimal. The closest point of the appropriate donor branch of the STA and the most suitable recipient M3 or M4 segment could be preoperatively determined with high accuracy so that the STA-MCA bypass could be safely and effectively performed through an optimally located minicraniotomy with a mean diameter of 22 mm without the need for stereotactic guidance. (DOI: 10.3171/2009.2.FOCUS0917)

Ke y Wo r d s • extracranial-intracranial bypass • magnetic resonance angiography • Dextroscope • three-dimensional virtual reality planning

h e STA-MCA bypass is an appealing microanasto- ment hemispheric perfusion in major vessel occlusion motic procedure initially developed and described with hemodynamic insufficiency.1,2,4,5,8,10,14,15,18 Although by Yaşargil16 and colleagues17 in 1967. At present, the microanastomosis itself is a highly sophisticated pro- thereT are 3 overall accepted indications: to achieve per- cedure on small structures, the need to intraoperatively fusion in distal MCA branches when iatrogenic vessel evaluate suitable recipient vessels requires a consider- occlusion cannot be avoided for various reasons, to im- able frontotemporal craniotomy to expose a wide area of prove brain perfusion in moyamoya disease, and to aug- the distal sylvian fissure. The introduction of 3-T MR imaging in the clinical Abbreviations used in this paper: ACoA = anterior communicat- diagnostic routine at major neurovascular centers has ing artery; AVM = arteriovenous malformation; ICA = internal tremendously improved the preoperative evaluation of carotid artery; STA-MCA = superficial temporal artery–middle potentially suitable recipient arteries.6 Nevertheless, the cerebral artery; TIA = transient ischemic attack. preoperative selection of appropriate donor and recipient

Neurosurg. Focus / Volume 26 / May 2009 1

Unauthenticated | Downloaded 10/03/21 08:27 PM UTC G. Fischer et al.

Fig. 1. Preoperative selection of the appropriate donor and recipient vessels. A: A 3-T MR angiogram with conventional 3D reconstruction and preselected vessels (yellow arrow). B: A 3D reconstruction image obtained using the Dextroscope, showing the superimposition of various structures: the STA, brain cortex, cortical MCA branches, and veins. Circle marks the proximity of appropriate MCA and STA branches. C: A 3D image depicting the simulated localization of the craniotomy. vessels, and thus the tailoring of an ideally located cran- quisition gradient echo (MPRAGE) sequences (TR 24 iotomy based on MR imaging and digital subtraction an- msec, TE 4.4 msec, flip angle 15°, matrix 640 × 564, slice giography data, remain challenging. In the current study thickness 0.49 mm, slice gap 0.49 mm) were created. we describe the application of a 3D virtual reality plan- Computed tomography 3D datasets with a slice thickness ning system based on 3-T MR angiography and its conse- of 0.5 mm were obtained to assess bony structures. Radio- quences for preoperative planning and an intraoperative logical data were integrated into a 3D virtual reality plan- strategy without stereotactic neuronavigation guidance. ning system (Dextroscope). The planning tool, its appli- cations, and the advantages and disadvantages in general 2,3 Methods neurosurgery have been discussed in detail elsewhere. In brief, 3-T MR images, time-of-flight MR angiograms, A 3D virtual reality planning tool (Dextroscope, and cranial CT images were fused, and a 3D virtual re- Volume Interactions, Bracco AMT, Inc.) was applied in 5 ality model was generated. Skin surface, skull surface, consecutive STA-MCA bypass procedures between July and extra- and intracranial arteries as well as the cortical and December 2008. The system’s influence on the surgi- brain surface were displayed in detail in a 3D configura- cal procedure was evaluated with special attention to the tion. The surgical approach was visualized virtually, and selection of the recipient vessel, localization of the donor the anatomical relationships of structures of interest were artery, and size and location of the craniotomy. presented by different values of translucency (Fig. 1). Anatomical findings in each individual were ana- Patient Population lyzed in detail, and the surgical steps of the upcoming All patients were men between the ages of 56 and procedure were simulated. The optimal STA branch for a 67 years (average age 62 years) who had suffered an ICA donor vessel was identified, as was its course in relation to occlusion (3 left side, 2 right side). Of these 5 patients, anatomical landmarks and auxiliary lines. The optimum only 1 had a history of a small MCA infarct. This patient recipient vessel, usually the M3 or M4 segment, was cho- and 3 others suffered from recurrent TIA symptoms. An sen based mainly on its caliber and superficial location. impaired O2 reserve capacity was found using duplex ul- The size and location of the craniotomy as well as the trasonography under CO2 exposition in all 4 patients. The length of the skin incision were determined by the prox- fifth patient was subjected to deep brain stimulation of imity of the selected donor and recipient vessels. With the subthalamic nucleus for essential tremor. In the work- the system’s “surface ruler” mode, we calculated the size up for the procedure, occlusion of the right ICA and a 10- and localization of the calculated craniotomy as well as mm ACoA