Microsurgical anatomy of safe entry zones to the brainstem · 2017-10-17 · distance of 4.64 mm...

laboratory investigation

J neurosurg October 9, 2015 1

HISTORICALLY, the surgical management of intrinsic brainstem lesions has been controversial. The sur-gical extirpation of focal gliomas, cavernous mal-

formations, or hemangioblastomas within the brainstem has caused heated discussions in scientific meetings and the literature. In 1939, Bailey et al.3 declared this subject to be a pessimistic chapter in neurosurgery; 30 years lat-er, Matson and Ingraham26 would still claim such lesions were inoperable. However, in 1971, Lassiter et al.25 were among the first to advocate surgical intervention. By 1986, Epstein and McCleary reported that surgery was feasible with reasonable morbidity and mortality.15 Concurrent with Epstein and McCleary’s report, Raimondi would ra-tionally state that to have the child merely survive (i.e., with severe neurological deficits) is no justification for surgery.33 The development and improvement of complex

skull base surgical approaches and incremental advances in neuroimaging, parallel to image-guided surgery, al-lowed a few authors to safely and effectively resect lesions in the brainstem.5,23,32

Knowledge of different skull base exposures, gained through laboratory dissections, allows neurosurgeons to approach lesions in the brainstem. Nevertheless, the brain-stem, roughly the size of the human thumb, contains a rich concentration of nuclei and fibers in a small sectional area, resulting in a high likelihood of morbidity after ma-nipulation. Awareness of the main safe entry zones on the brainstem is key to reducing morbidity for any lesion that does not emerge to the pial or ependymal surface. Such zones represent entry points and trajectories where elo-quent structures and perforators are sparse and where a neurotomy would cause the least possible damage.

abbreviations CN = cranial nerve; mini-OZ = mini-modified orbitozygomatic approach; PCA = posterior cerebral artery; P2P = posterior P2; SCA = superior cerebellar artery; TAPS = transanterior perforating substance.submitted August 21, 2014. accepted April 8, 2015.include when citing Published online October 9, 2015; DOI: 10.3171/2015.4.JNS141945.

Microsurgical anatomy of safe entry zones to the brainstemdaniel d. cavalcanti, md,1 mark c. preul, md,2 m. yashar s. Kalani, md, phd,2 and robert F. spetzler, md2

1Department of Neurosurgery, Paulo Niemeyer State Brain Institute, Rio de Janeiro, Brazil; and 2Division of Neurological Surgery, Barrow Neurological Institute, St. Joseph’s Hospital and Medical Center, Phoenix, Arizona

obJect The aim of this study was to enhance the planning and use of microsurgical resection techniques for intrinsic brainstem lesions by better defining anatomical safe entry zones.methods Five cadaveric heads were dissected using 10 surgical approaches per head. Stepwise dissections focused on the actual areas of brainstem surface that were exposed through each approach and an analysis of the structures found, as well as which safe entry zones were accessible via each of the 10 surgical windows.results Thirteen safe entry zones have been reported and validated for approaching lesions in the brainstem, including the anterior mesencephalic zone, lateral mesencephalic sulcus, intercollicular region, peritrigeminal zone, su-pratrigeminal zone, lateral pontine zone, supracollicular zone, infracollicular zone, median sulcus of the fourth ventricle, anterolateral and posterior median sulci of the medulla, olivary zone, and lateral medullary zone. A discussion of the approaches, anatomy, and limitations of these entry zones is included.conclusions A detailed understanding of the anatomy, area of exposure, and safe entry zones for each major ap-proach allows for improved surgical planning and dissemination of the techniques required to successfully resect intrinsic brainstem lesions.http://thejns.org/doi/abs/10.3171/2015.4.JNS141945Key words brainstem; cavernous malformation; microsurgery; safe entry zones; surgical approaches; surgical anatomy

©AANS, 2015

d. d. cavalcanti et al.

The aim of our study was to enhance the planning and use of microsurgical resection techniques for intrinsic brainstem lesions. We examined 13 safe entry zones on the brainstem, which have been described in the litera-ture, and used detailed cadaveric dissections to evaluate the main surgical approaches currently employed to man-age intrinsic brainstem lesions. Through detailed dissec-tion images of these approaches, we visually demonstrate what can be seen on the brainstem surface through each of these corridors and delineate the safe entry zones provid-ed by each approach. It is critical to note that large lesions may distort safe entry zones and that neurophysiological monitoring is a critical adjunct in these cases.36

methodsWe searched MEDLINE and Google Scholar for stud-

ies containing terms related to the surgical management of brainstem lesions to determine the most frequently discussed safe entry zones. Thirteen zones were selected: 1) anterior mesencephalic zone, 2) lateral mesencephalic sulcus, 3) intercollicular region, 4) peritrigeminal zone, 5) supratrigeminal zone, 6) lateral pontine zone, 7) supracol-licular zone, 8) infracollicular zone, 9) median sulcus of the fourth ventricle, 10) anterolateral and 11) posterior me-dian sulci of the medulla, 12) olivary zone, and 13) lateral medullary zone.5,9,14,20,23,35

Five human cadaveric heads, formalin-fixed and inject-ed with colored silicone rubber, were carefully dissected in a simulated surgical environment at the Skull Base Laboratory of the Barrow Neurological Institute in Phoe-nix, Arizona. With the heads fixed with Mayfield clamps, 10 surgical approaches were performed on each head: 1) orbitozygomatic, 2) subtemporal, 3) subtemporal transten-torial, 4) anterior petrosectomy, 5) suboccipital telovelar, 6) median supracerebellar infratentorial, 7) extreme lat-eral supracerebellar infratentorial, 8) retrosigmoid, 9) far lateral, and 10) retrolabyrinthine. The order in which the approaches were performed was altered for each head to obtain clear images of each approach. Every step was photographically registered, and the resultant surgical ex-posure on the brainstem surface for each approach was analyzed for both identifiable anatomical structures and the safe entry zones that were available.

resultssafe entry Zones

The safe entry zones discussed can be divided into 3 main regions: midbrain (Fig. 1), pons (Fig. 2), and medulla oblongata (Fig. 3).

MidbrainAnterior Mesencephalic ZoneLesions involving the anterior midbrain can be ac-

cessed through a limited area on the cerebral peduncle bounded medially by the oculomotor tract and nerve and laterally by the corticospinal tract (Fig. 1A and B). Such a narrow corridor takes advantage of the distribution of corticospinal tract fibers mainly in the intermediate three-fifths of the peduncle and the fact that the red nucleus and

the nigrostriatal circuit are in a deep medial location.5 The entry point inside the interpeduncular cistern is limited superiorly by the posterior cerebral artery (PCA) and in-feriorly by the main trunk of the superior cerebellar artery (SCA) (Figs. 4F and H and 5F).

Lateral Mesencephalic SulcusHidden by the lateral mesencephalic vein, the lat-

eral mesencephalic sulcus separates the peduncular and tegmental surfaces of the midbrain facing the middle incisural space.30 The lateral mesencephalic sulcus ex-tends downward in a concave fashion from the medial geniculate body to the pontomesencephalic sulcus (Fig. 1A and C). Crossing the sulcus are the posterior P2 seg-ment (P2P) superiorly, the medial posterior choroidal ar-tery centrally, and the cerebellomesencephalic segments of the SCA, trochlear nerve, and tentorial edge inferiorly (Fig. 5G and H).

The safe entry zone runs between the substantia nigra anterolaterally and the medial lemniscus posteriorly. The oculomotor nerve fibers crossing from the red nucleus to the substantia nigra impose an anteromedial limit to dis-section. Recalde et al.35 found that the average total length of the sulcus was 9.6 mm (range 7.4–13.3 mm) with an av-erage working-channel length of 8.0 mm (range 4.9–11.7 mm).

Intercollicular RegionThe quadrigeminal plate or tectum is composed of 2

superior and 2 inferior rounded eminences, designated as superior colliculi and inferior colliculi, respectively, which represent the dorsal surface of the midbrain (Fig. 1A and D).

The superior colliculi are active in the visual system; they are critical to visual fixation and saccadic eye move-ments.19,28 They are connected to each lateral geniculate body by a superior brachium, a path in which the retino-tectal fibers run.21 Additionally, spinotectal and cortico-tectal fibers reach the superior colliculi, while tectospinal, tectothalamic, and tectocortical tracts leave these struc-tures.

The inferior colliculi are part of the auditory system. They receive fibers from the contralateral cochlear nucle-us, dorsal and ventral nuclei of the lateral lemniscus, con-tralateral and ipsilateral superior olive, ipsilateral medial superior olive, and descending projections from sensory areas through the corticollicular neurons.8,27 Commissural fibers connect 1 inferior colliculi to another.27 The inferior colliculi extends laterally through the inferior brachium to the medial geniculate body of the thalamus, which proj-ects to the primary auditory cortex.

The most appropriate area for a small neurotomy has been described as the intercollicular region, because of its sparseness of fibers (Fig. 1A and D). Bricolo and Turazzi5 first suggested this corridor, which has been further sup-ported by other studies.9,22,34

PonsPeritrigeminal ZoneThe anterolateral surface of the pons has tradition-

ally been considered a safe zone for entering the brain-stem.9,17,32 Using the white fiber dissection technique, Re-calde et al.35 were able to clearly quantify a safe trajectory

J neurosurg October 9, 20152

surgical approaches to the brainstem

Fig. 1. Safe entry zones of the midbrain. a: Cross section of the midbrain at the level of the cerebral peduncle revealing its 3 safe entry zones (white dashed lines): the lateral mesencephalic sulcus (LMS), the intercollicular region (ICR), and the anterior mesencephalic zone (AMZ). b: Anterior view of a dissected brainstem revealing the AMZ (white dashed line). c: Right postero-lateral view of the brainstem showing the LMS (black dashed line). d: Posterior exposure of the brainstem showing the ICR (black dashed line). Cerebral ped. = cerebral peduncle; CST = corticospinal tract; inf. = inferior; interped. fossa = interpeduncular fossa; mid. cerebell. ped. = middle cerebellar peduncle; pit. stalk = pituitary stalk; post. = posterior; rhomb. fossa = rhomboid fossa; sup. = superior; sup. cerebell. ped. = superior cerebellar peduncle.

Fig. 2. Safe entry zones of the pons. a: Cross section of the pons demonstrating the peritrigeminal zone (PTZ; arrow). b: Three entry zones can be used on the rhomboid fossa: the supracollicular zone (SCZ; arrow), the infracollicular zone (ICZ; arrow), and the median sulcus of the fourth ventricle (MS; dashed line). c: The arrows represent the safe entry zones for excising lateral and anterolateral pontine lesions: the supratrigeminal zone (STZ), the PTZ, and the lateral pontine zone (LPZ). Ant. med. fissure = anterior median fissure; med. long. fascicle = medial longitudinal fascicle.



in front of the trigeminal nerve (cranial nerve V [CN V]) entry zone, lateral to the corticospinal tract and anterior to the motor and sensory nuclei of the trigeminal nerve (Fig. 2A and B). On the axial plane, they found a mean distance of 4.64 mm (range 3.8–5.6 mm) between CN V and the corticospinal tract, and a mean depth of dissection of 11.2 mm (range 9.5–13.1 mm) to the trigeminal nuclei. The fibers of CNs VI, VII, and VIII run downward and are located posterior to the trigeminal nuclei.

Supratrigeminal ZoneA second entry point that has been used to manage an-

teriorly placed lesions in the pons is located just above the trigeminal root entry zone on the middle cerebellar pe-duncle (Fig. 2B).20 Taking advantage of the posterolateral location of the middle cerebellar peduncle and the thick pontine transverse fibers, it is possible to carefully dissect along these fibers, medially or anteromedially, posterior to the trajectory of the corticospinal tract.

Lateral Pontine ZoneIn 1982, Baghai et al.2 recommended a safe corridor on

the junction between the middle cerebellar peduncle and the pons and between the trigeminal and the facial-ves-tibulocochlear complex root entry zones (Fig. 2B). Other authors have supported using the narrow corridor of the lateral pontine zone, but it restrains vertical manipulation.5

Supracollicular and Infracollicular ZonesThe rhomboid fossa hides several structures whose ma-

nipulation may result in increased morbidity. Superficial landmarks help protect crucial structures located at the

depth of the fourth ventricular floor. At the floor of the fourth ventricle, the facial nerve passes around the nucleus of the abducens nerve; this combined structure is called the facial colliculus. The medial longitudinal fascicle lies parallel to the median sulcus. Similarly, the nuclei of CNs X and XII are located just caudal to the striae medullaris.

After examining the topographic anatomy of the facial colliculi and every potential sign of neurological deficit presented after manipulating the rhomboid fossa, Kyo-shima et al.23 described 2 safe zones through this region that minimally displace surrounding neural structures (Fig. 2C, with specific reference to SCZ and ICZ). The first safe zone is the suprafacial triangle, which is delim-ited caudally by the facial nerve, laterally by the cerebel-lar peduncles, and medially by the medial longitudinal fascicle. The second safe zone is described by the edges of the infrafacial triangle, which are the striae medullaris caudally, the facial nerve laterally, and the medial longitu-dinal fascicle medially.

However, Strauss et al.37 stressed the variability of the striae medullaris and the possible damage to either the tri-geminal motor nucleus or the nuclei of the lower cranial nerves when using the suprafacial or infrafacial triangles, respectively. They conducted a detailed morphometric study to measure the dimensions of the facial colliculus and its distance to the midline, decussation of CN IV, and vagal and hypoglossal trigones. They defined a parame-dian supracollicular area that measured 13.8 mm verti-cally between the facial colliculus and the decussation of

Fig. 3. Safe entry zones of the medulla oblongata. a: Cross section of the medulla revealing the 3 safe entry zones: the antero-lateral sulcus (ALS; arrow), the olivary zone (OZ; dashed line), and the posterolateral lateral medullary zone (LMZ; dashed ar-row). b: Anterolateral view of a dissected brainstem showing the site for entering the olive (dashed line) and the entry site (arrow) on the ALS just below the root of the hypoglossal nerve. c: Posterior view of the brainstem revealing the posterior median sulcus (PMS) and LMZ safe zones. Ant. med. fissure = anterior median fissure.



CN IV, 0.6 mm from the midline. The trigeminal motor nucleus limited the approach laterally, being located 6.3 mm from the midline.

A paramedian infracollicular area can be tailored be-tween the projection of the facial nerve fibers on the facial colliculus and the superior limits of the nucleus of the hy-poglossal nerve and the dorsal nucleus of the vagal nerve, extending a mean distance of 9.2 mm vertically and ap-proximately 0.3 mm from the midline.37

Median Sulcus of the Fourth VentricleAn approach through the midline, between the projec-

tion of the CN VI nuclei on the surface and the projection of the CN III nuclei on the midbrain surface, was pro-posed by Bricolo et al.,5 taking advantage of the sparse-ness of crossing fibers (Fig. 2C). Even the slightest lateral

retraction may provoke extraocular movement disorders caused by damage to the medial longitudinal fascicle.

Medulla OblongataAnterolateral SulcusJust lateral to the pyramid, the rootlets of the hypoglos-

sal nerve leave the brainstem on the anterolateral sulcus. The short space between these rootlets and those of the C-1 nerve coincides with the decussation of the cortico-spinal tract.9 A paramedian oblique dissection may avoid the corticospinal tract and address lesions of the anterior lower medullary region (Fig. 3A and B).

Posterior Median SulcusA neurotomy on the median sulcus provides a corri-

Fig. 4. Orbitozygomatic approach. a: A hemicoronal incision begins in the preauricular area and curves forward at the mid-line. b: The skin flap is reflected along with the temporal fascia. A subperiosteal dissection reveals the zygomatic process of the frontal bone, the frontal process of the zygomatic bone, and the zygomatic arch. c: A pterional craniotomy is prepared after detaching the temporal muscle. d: Magnified view after elevating the pterional flap and completing the orbitozygomatic oste-otomy. e: The 2-piece craniotomy flap. F: The surgical window provided by this approach, centered on the exit of the oculomotor nerve from the cerebral peduncle. The anterior mesencephalic zone (AMZ) is shown by the dashed line. g: Cadaveric dissection demonstrating the AMZ, limited superiorly by the posterior cerebral artery, inferiorly by the superior cerebellar artery, medially by the oculomotor nerve, and laterally by the projection of the corticospinal tract fibers on the surface. h: Shaded area represents the area of exposure provided by the orbitozygomatic approach. i: Preoperative T2-weighted MR image in a 3-year-old male reveals the characteristic imaging findings of a cavernous malformation. The lesion was approached via a left-modified orbitozy-gomatic approach via the anterior mesencephalic sulcus. J: Postoperative T2-weighted MR image reveals gross-total resection of the lesion. AICA = anterior inferior cerebellar artery; BA = basilar artery; frontozyg. = frontozygomatic; ICA = internal carotid artery; m. = muscle; M1 = M1 segment of the middle cerebral artery; P2A = anterior part of the P2 segment of the posterior cerebral artery; PCA (P1) = posterior cerebral artery (P1 segment); PCoA = posterior communicating artery; proc. = process; SCA = superior cerebellar artery; zyg. = zygomatic.



dor near the center of the medulla.5,9 Below the obex and restricted laterally by the clava, which covers the gracile nucleus, a surgeon can approach posteriorly placed lesions in a fashion similar to the traditional approach for intra-medullary spinal cord lesions (Fig. 3C).

Olivary ZoneThe olives are marked oval eminences on the antero-

lateral surface of the medulla, limited medially by the anterolateral sulcus and the pyramids and posteriorly by the posterolateral sulcus. In a cross section at the level of the inferior olivary nucleus, fibers of the hypoglossal nerve separate it from the corticospinal tract running within the pyramids. The olives are also limited medially by hypo-glossal nerve fibers and the medial lemniscus and are lim-

Fig. 5. Subtemporal approach. a: Photograph showing the site of skin incision. b: A bur hole is placed just above the root of the zygomatic arch. c: A square craniotomy is made extending two-thirds in front of the bur hole and one-third behind it. The inferior edge of the craniotomy is then drilled flush with the middle fossa floor. d: The dura is elevated, exposing the temporal lobe. e: The microdisection is carried between the basal surface of the temporal lobe and the tentorial edge, through the arachnoid of the ambient cistern. F: The area of exposure provided by this approach, including part of the anterior and the lateral incisural spaces, leads to the entire lateral midbrain surface. g: Magnified view of the lateral midbrain though the opened interpeduncular and ambient cisterns. This approach also exposes 2 safe entry zones: the anterior mesencephalic zone (AMZ; green dashed line) and the lateral mesencephalic sulcus (LMS; green dashed line seen in H). h: Cutting the tentorium significantly improves visualization of the pontomesencephalic junction (PMJ) and the lateral upper pons. i: Progressive increases to a simple subtemporal approach in the area and length of exposure after adding a transtentorial extension and then an anterior petrosectomy. ICA = internal carotid artery; medial post. choroidal a. = medial posterior choroidal artery; P2A = anterior part of the P2 segment of the posterior cerebral artery; PCoA = posterior communicating artery; quad. = quadrangular; SCA = superior cerebellar artery; tent. = tentorial.



ited posteriorly mainly by the tectospinal and spinotha-lamic tracts.

Recalde et al.35 identified a safe depth of dissection via the olive, ranging from 4.7 to 6.9 mm, with a vertical length of 13.5 mm (Fig. 3).

Lateral Medullary Zone “Inferior Cerebellar Pedun-cle Approach”

Akin to the lateral pontine zone in the pons, the lateral medulla has recently been shown by Deshmukh et al. to be a relatively safe entry zone for resection of dorsolateral medullary lesions.14 These authors recently reported their experience with 4 patients whose lesions were approached through the foramen of Luschka with an incision in the inferior cerebellar peduncle and whose outcomes were excellent.14 We call the access through this approach the lateral medullary zone. Via a retrosigmoid approach, the foramen of Luschka is opened, and the origins of CNs IX and X are identified. Then a small vertical incision is made in the inferior cerebellar peduncle inferior to the cochlear nuclei and posterior to the origin of CNs IX and X.

surgical approachesOrbitozygomatic

For neurosurgeons to perform the orbitozygomatic approach, the patient is placed supine and the head is el-evated above the level of the heart, rotated 15° to 30° con-tralaterally, and the neck is slightly extended (Fig. 4A). A hemicoronal skin incision is performed, the skin flap is reflected, and the galeal flap is dissected and prepared. An interfascial dissection is performed, exposing the zygo-matic process of the frontal bone and the frontal process of the zygomatic bone (Fig. 4B). The temporal fascia is ele-vated, and a subperiosteal dissection is performed, reveal-ing the zygomatic arch. The temporal muscle is then el-evated, and a pterional craniotomy is tailored as proposed by Yaşargil39 (Fig. 4C). The orbitozygomatic osteotomy is then completed as a second step, returning the temporal muscle to its original position in a sequence of 6 cuts as modified by Zabramski et al.40 (Fig. 4D and E). The dura is opened with an arciform incision.

The sylvian fissure is opened, and the dissection is car-ried proximally, following the M1 segment of the middle cerebral artery toward the carotid and chiasmatic cisterns. These cisterns are opened, and the focus of the dissection is then turned to the carotico-oculomotor triangle, which is entered in order to reach the crural and interpeduncu-lar cisterns. Their arachnoid trabeculae are also opened widely. Tracking up the oculomotor nerve to its entry zone leads the surgeon to the cerebral peduncle (Fig. 4F). Possible lesions emerging on the anterior surface of the midbrain, pontomesencephalic junction, and upper pons can be accessed (Fig. 4G). This approach also provides a straight path to the aforementioned anterior mesencephal-ic safe entry zone between the PCA and SCA, laterally to CN III (Fig. 4H). Clinically, the ventral approaches are often avoided because of the rich motor tracts that travel ventrally in the brainstem; nonetheless, in select cases, anterior approaches may be used to resect lesions in the midbrain (Fig. 4I–J).

Depending on the anteroposterior extension of the le-

sion, as well as its closest point to the surface on the ce-rebral peduncle, a contralateral orbitozygomatic approach offers the best approach for complete resection of a mid-brain lesion. Opening the Liliequist membrane allows the surgeon to reach the interpeduncular cistern, a very com-plex region demanding extremely delicate dissection. This small region is populated by the basilar artery bifurcation, both PCAs and SCAs, the medial posterior choroidal ar-tery, and the thalamoperforating arteries and direct perfo-rators of the proximal SCA, restricting surgical freedom significantly.

Lesions of the thalamomesencephalic junction are ex-tremely challenging to expose. After widely dissecting the sylvian fissure, a small corridor is developed through the lenticulostriate arteries, between the posterior limits of both the gyrus rectus and the medial orbital gyrus, and the M1. We have described this approach—the transante-rior perforating substance (TAPS) approach—previously, and it demands a small opening on the anterior perforat-ing substance, posterior to the olfactory stria and near the optic tract.16

SubtemporalFor neurosurgeons to perform the subtemporal ap-

proach, the patient is placed in a lateral decubitus posi-tion, and the sagittal suture is kept parallel to the floor. A straight vertical incision is placed in front of the tragus, and the temporal fascia and muscle are opened in the same fashion, exposing mainly the temporal bone (Fig. 5A). A bur hole is made just above the root of the zygomatic arch, and the width of the craniotomy is tailored accord-ing to a preoperative plan for the best angle of approach to the brainstem (Fig. 5B and C). The dura is opened in a U-shaped fashion, and the dissection is carried below the base of the temporal lobe until the tentorial edge is exposed (Fig. 5D and E). The anterior and middle inci-sural spaces are then visible. The arachnoid of the ambi-ent, crural, and interpeduncular cisterns is opened widely (Fig. 5F). The lateral surface of the midbrain, down to the pontomesencephalic junction, occupies most of the opera-tive field, with the anterior P2 segment (P2A crural) and P2P (ambient) segments of the PCA and the middle poste-rior choroidal artery crossing the field. The lateral mesen-cephalic vein frequently covers the lateral mesencephalic sulcus (Fig. 5G). This corridor yields an oblique access to the anterior mesencephalic zone and a straight route to the lateral mesencephalic sulcus (Fig. 5H and I). When pos-sible, care must be exercised to avoid traversing the floor of the fourth ventricle. Approaches through the floor of the fourth ventricle can result in significant lower cranial nerve deficits and respiratory and swallowing difficulties.

Subtemporal TranstentorialA transtentorial extension to the subtemporal approach

is obtained by sectioning and retracting the tentorium just before the trochlear nerve pierces it. The pontomesence-phalic junction and an anterolateral perspective of the up-per pons can then be appreciated, in addition to the SCA running in the direction of the quadrigeminal cistern (Fig. 5H).



Anterior PetrosectomyAn anterior petrosectomy or Kawase approach can be

added to the subtemporal approach or performed alone when aiming for a more anterolateral view of the pons from above in order to focus on lesions anterior to the tri-geminal nerve (Fig. 5I). The petrous bone in front of the internal acoustic canal is drilled away; the exposure is lim-ited anteriorly by the V3 segment of the trigeminal nerve and laterally by the greater superficial petrosal nerve.

Suboccipital TelovelarFor neurosurgeons to perform the suboccipital telove-

lar approach, the patient is placed in a prone position, the head is flexed, and a straight median incision is made to expose the myofascial layer, which is opened on the mid-line and retracted laterally (Fig. 6A). The suboccipital area is exposed together with the C-1 posterior arch. A median suboccipital craniotomy large enough to retract both cer-ebellar tonsils superolaterally is tailored using 2 bur holes, which are placed laterally at the superior edge (Fig. 6B and C). A laminoplasty of the C-1 arch may be performed to significantly extend the vertical angle of approach.13

The dura is opened in a Y-shaped fashion to preserve the occipital sinus. Opening the dura exposes the cerebel-lar tonsils, which hide the cerebellomedullary fissure (Fig. 6D). By slightly retracting the tonsils and carefully push-ing the telovelotonsillar segment of the posterior inferior cerebellar arteries laterally, it is possible to reach and open the tela choroidea. The tela choroidea, together with the inferior medullary velum, constitute the inferior half of the roof of the fourth ventricle (Fig. 6E).29 The telovelar junc-tion attaches medially to the nodule and extends laterally into the lateral recess. Dividing the tela choroidea bilater-ally is enough to bring the whole rhomboid fossa into the surgical field, as well as the lateral recesses (Fig. 6F). The safe zones above and below the facial colliculus are vis-ible, as is the superior half of the median sulcus (Fig. 6G). Opening the inferior velum will expose the superior half of the ventricular roof beside the superolateral recesses.

Median Supracerebellar InfratentorialThe setting for the median supracerebellar infratento-

rial approach is very similar to the suboccipital telovelar approach described above, except that the skin incision is extended a bit cranially to ensure that the craniotomy ex-tends to a point just above the transverse sinuses (Fig. 7A and B). Doing so allows the tentorium to be intermittently retracted. The dura is again opened in a Y-shaped fashion, and the dissection is begun via the tentorial surface of the cerebellum, opening the supracerebellar cistern and coagu-lating as few bridging veins as possible (Fig. 7C and D). The quadrigeminal cistern sits deeply in the dissection bed. It contains a complex venous structure that covers the pi-neal gland and includes the internal cerebral and internal occipital veins, the basal veins of Rosenthal, and the vein of the cerebellomesencephalic fissure (Fig. 7E and F). Con-tinuing the dissection more deeply leads the neurosurgeon to the quadrigeminal plate, containing the superior and in-ferior colliculi and the intercollicular safe zone (Fig. 7G and H). The pulvinar and the tentorial edge laterally limit the exposure of both the superior and inferior brachia.

Extreme Lateral Supracerebellar InfratentorialFor neurosurgeons to perform the extreme lateral su-

pracerebellar infratentorial approach, the patient is placed in a lateral decubitus position, and the head is slightly flexed and rotated ipsilaterally. A retroauricular straight skin incision is made, the myofascial layer is elevated, and the flap is retracted anteriorly (Fig. 8A). A bur hole is placed just above the asterion on the parietomastoid su-ture, and a modified retromastoid craniotomy is tailored that extends across the transverse sinus, which eases surgi-cal exposure through its intermittent retraction. The dura is opened using an inverted T-shaped incision, leaving a dural base along the transverse sinus and another along the sigmoid sinus. The dissection is made by cutting the arachnoid along the tentorial surface of the cerebellum, coagulating as few bridging veins as possible to reach the cerebellomesencephalic fissure (Fig. 8B).

The ambient cistern is then opened, exposing the troch-lear nerve running along the posterolateral midbrain, fre-quently close to the SCA (Fig. 8C). Turning the focus to the midline and opening the quadrigeminal cistern pro-vides an oblique approach to the collicular region, main-ly to the inferior colliculus. The trochlear nerve can be traced to its origin laterally, below the inferior colliculus (Fig. 8D). Retracting the cerebellum closer to the transition between the tentorial and petrosal surfaces offers a more lateral perspective of the midbrain, centered on the lateral mesencephalic sulcus. Figure 8E reveals the window on the brainstem provided by this important approach. The supracerebellar infratentorial approach is a robust ap-proach; a more superior view can be obtained by cutting the tentorium. This addition allows the surgeon to attack lesions in the medial temporal lobe and thalamus. Figure 8F and G demonstrate a lesion resected via the extreme lateral supracerebellar infratentorial approach.

RetrosigmoidFor neurosurgeons to perform the retrosigmoid ap-

proach, the patient is placed in a lateral position with the head slightly flexed and rotated toward the ipsilateral side (Fig. 9A). A straight skin incision is placed 2 finger-breadths behind the earlobe, exposing the myofascial lay-er. The myofascial layer is elevated, and the intersection of the parietal, temporal, and occipital bones is exposed using a self-retractor. One bur hole located just above the asterion on the parietomastoid suture is enough to perform a retromastoid craniotomy, with the anterior edge level with the posterior margin of the sigmoid sinus (Fig. 9B and C). The dura is opened using an inverted T-shaped incision, leaving a dural base along the transverse and the sigmoid sinuses (Fig. 9D). The microsurgical dissec-tion starts along the petrosal surface of the cerebellum, reaching the cerebellopontine cistern (Fig. 9E). Arachnoid strands around the superior petrosal vein and its tributar-ies as well as around the trigeminal, facial, and vestibulo-cochlear nerves are cut with regard to the trajectories of the SCA and anterior inferior cerebellar artery. It is then possible to expose the middle cerebellar peduncle and the lateral pons, which contain 3 safe zones around the CN V entry zone (Fig. 9F and G), as described above. We use the retrosigmoid approach more commonly than any other ap-



Fig. 6. Median suboccipital telovelar approach. a: A midline vertical skin incision is made. b: Two bur holes are made laterally to the inion just below the transverse sinuses, and a suboccipital craniotomy is tailored. c: The dura is exposed. The posterior arch of C-1 was opened to enhance the vertical angle of approach to the rhomboid fossa. d: The dura is opened, exposing cerebellar structures such as the cerebellar tonsils and the uvula, as well as the cerebellomedullary fissure. e: An opened foramen of Magendie, limited laterally by the tela choroidea, the tonsils, and the posterior inferior cerebellar arteries. F: Dissecting the tela choroidea and the inferior medullary velum elegantly exposes the whole rhomboid fossa and cerebral aque-duct, without the associated risks of splitting the vermis. g: The area on the dorsal brainstem exposed by the telovelar approach is represented by the shaded area. h: Preoperative T2-weighted MR image in this 55-year-old male reveals a hyperintense lesion surrounded by a hypointense hemosiderin halo in the dorsal pons. The lesion was approached via a suboccipital approach with gross-total resection of the lesion. i: Postoperative T2-weighted MR image reveals the resection cavity. PICA = posteroinferior cerebellar artery; sup. = superior.



Fig. 7. Median supracerebellar infratentorial approach. a and b: The skin incision and craniotomy are extended farther cranially than in the median suboccipital approach to expose the transverse sinuses. c and d: After opening the dura in the customary Y-shape, the dissection is begun at the ten-torial surface of the cerebellum. e: Most of the posterior incisural space is occupied by the vein of Galen complex. F: The pineal gland is exposed by dissecting downward. g: The objective is to expose the quadrigeminal plate below the pineal region at the base of the posterior incisural space. The in-tercollicular region (ICR) is shown in the center of the field. h: The shaded area represents the area of the dorsal brainstem exposed by this approach. IC = inferior colliculus; mesenceph. = mesencephalic; PCA = posterior cerebral artery; PG = pineal gland; SC = superior colliculus; v. = vein.



proach. The posterolateral view afforded by this approach allows the surgeon to remove lesions while creating few motor deficits, albeit with significant, often temporary, sensory deficits (Fig. 9H–I). The use of an endoscope can allow surgeons to visualize the lateral pons without the need for the more extensive far-lateral approach.

Far LateralFor the far-lateral approach, the patient is placed in the

park-bench position, and a classic hockey-stick skin inci-sion allows the surgeon to promptly identify each muscu-lar layer down to the suboccipital triangle (Fig. 10A).24 A straight incision is faster but should be reserved for sur-geons who are experienced with it. The very first layer exposed is composed of the sternocleidomastoid and tra-pezius muscles, which when retracted expose the splenius capitis muscle. The next layer comprises the longissimus capitis muscle laterally and the semispinalis capitis mus-

Fig. 8. Extreme lateral supracerebellar infratentorial approach. a: The patient is placed in a lateral position with the head fixed in a slight flexion and ipsilateral rotation. A straight incision is made in the retroauricular region. b: A retrosigmoid craniotomy is tailored to extend just above the transverse sinus; however, the dissection is carried superior to the tentorial surface of the cer-ebellum. c: After coagulating as few bridging veins as possible, and dissecting the arachnoid around the cerebellomesencephalic fissure and the superior petrosal vein, the neurosurgeon reaches the ambient cistern and the lateral segment of the quadrigeminal cistern. The trochlear nerve is seen crossing the cistern, along the superior cerebellar artery (SCA) and distal branches of the posterior cerebral artery. d: Magnified posterolateral view of the midbrain revealing the exit of the trochlear nerve inferolaterally to the inferior colliculus (IC). e: The shaded area corresponds to the area of exposure on the posterolateral brainstem gained by this approach. F and g: This 56-year-old male presented to our service with multiple hemorrhages attributable to this thalamomes-encephalic cavernous malformation. F: Preoperative T2-weighted MR image reveals the cavernous malformation. The lesion was approached via an extreme lateral supracerebellar infratentorial approach, and the lesion was resected via the lateral mesence-phalic sulcus. G: Postoperative MR image reveals gross-total resection of the lesion. Lat. mesenceph. v. = lateral mesencephalic vein; LMS = lateral mesencephalic sulcus; transv. = transverse.



cle, which when elevated bring the suboccipital triangle into the surgical field.

The superior oblique, inferior oblique, and rectus capitis posterior major muscles limit the triangle that protects the V3 segment of the vertebral artery (Fig. 10B). The rectus

capitis posterior minor muscle is located medially to the rectus capitis posterior major muscle. Both of these mus-cles are elevated, exposing the C-1 posterior arch and the vertebral artery (Fig. 10C). The ipsilateral half of the C-1 posterior arch is removed after subperiosteal dissection of

Fig. 9. Retrosigmoid approach. a: The cadaver head is placed in the lateral position. The skin incision is usually placed 2 finger-breadths behind the pinna. b: A bur hole is drilled just cranial to the asterion, on the parietomastoid suture. c: The craniotomy is started with a C-shaped sawing line. The posterior edge of the sigmoid sinus as well as the transverse-sigmoid junction is skel-etonized. d: The dura is opened and the cerebellar surface is exposed. e: The microsurgical dissection starts along the petrosal surface of the cerebellum, reaching the cerebellopontine cistern. The approach also provides access to the supratrigeminal (STZ) and the lateral pontine (LPZ) safe zones but provides a suboptimal angle of attack for the peritrigeminal zone. F: Lesions abutting the middle cerebellar peduncle or the lateral pontine surface can be resected using this route. g: The shaded area represents the area of exposure on the lateral brainstem produced by the retrosigmoid approach. h and i: This pontomedullary cavernous malformation in a 10-year-old male was approached via a retrosigmoid approach through the lateral medullary zone. H: Preopera-tive T2-weighted MR image demonstrates the lesion. I: Postoperative T2-weighted MR image reveals gross-total resection of this lesion. AICA = anteroinferior cerebellar artery; LPZ = lateral pontine zone; mid. cerebell. ped. = middle cerebellar peduncle; PICA = posterior inferior cerebellar artery; STZ = supratrigeminal zone; sup. = superior; transv. = transverse; v = vein.



the vertebral artery. The posterior root of the transverse foramen may be drilled away to mobilize the vertebral ar-tery, although this is not required for most brainstem le-sions.

A lateral suboccipital craniotomy is then tailored to the craniocaudal extension of the lesion, exposing the poste-rior margin of the sigmoid sinus (Fig. 10D). The posterior third of the occipital condyle may also be drilled away, preserving the hypoglossal canal and improving the angle of attack for lesions in the anterolateral medulla. The dura is then opened in a curvilinear fashion, starting close to

midline and curving toward the transverse sigmoid junc-tion (Fig. 10E).

The arachnoid over the cisterna magna is opened, as well as the lateral cerebellomedullary cistern. The V4 seg-ment of the vertebral artery can be seen coursing laterally to the spinal accessory nerve and in front of the hypoglos-sal rootlets, where the posterior inferior cerebellar artery arises (Fig. 10F). A wide window is opened to reach the posterolateral medullary surface and the cervicomedul-lary junction. The accessory nerve, together with the glos-sopharyngeal and the vagus, enter the jugular foramen.

Fig. 10. Far-lateral approach. a: The classic hockey-stick incision with the patient in the park-bench position. b and c: A meticu-lous muscle dissection exposes the suboccipital triangle, covering the V3 segment of the vertebral artery (VA). d: A lateral suboc-cipital craniotomy and a C-1 laminotomy. The posterior root of the transverse foramen may be drilled away to mobilize the vertebral artery, a step not required for most brainstem lesions. e: The posterior third of the occipital condyle may also be drilled away, widening the angle of attack for anterolateral medullary lesions. The dura is then opened. F: The posterolateral medulla is ex-posed for access to superficial lesions, including exposure of the trajectory of the lower cranial nerves to the jugular foramen, the vertebral artery, and posterior inferior cerebellar artery. The cervicomedullary junction is also exposed. Changing the angle of the microscope allows visualization of the anterolateral surface of the medulla. g: The shaded area correlates with the area of expo-sure on the brainstem accessed by the far-lateral approach. h: T1-weighted postcontrast MR image of the brain in a 67-year-old female reveals a lesion consistent with a cavernous malformation. The patient underwent a right far-lateral approach and resection of the lesion via the lateral pontine zone. i: Postoperative T2-weighted MR image reveals gross-total resection of the lesion. ALS = anterolateral sulcus safe zone; C2 spin. proc. = C-2 spinous process; inf. oblique m. = inferior oblique muscle; PICA = posterior inferior cerebellar artery; post. = posterior; rectus cap. post. major m. = rectus capitis posterior major muscle; sup. oblique m. = superior oblique muscle; transv. proc. of C1 = transverse process of C-1; TS/SS junction = transverse sinus/sigmoid sinus junction; V2 = V2 segment of the vertebral artery; V3 = V3 segment of the vertebral artery; V4 = V4 segment of the vertebral artery.



The arachnoid covering the cerebellopontine cistern can be divided to approach the pontine surface. Continuing the dissection ventrally leads to the premedullary cistern, al-lowing access to the anterolateral sulcus and the olivary zone (Fig. 10G). Figure 10H and I illustrates removal of a lesion via the far lateral approach.

RetrolabyrinthineFor neurosurgeons to perform the retrolabyrinthine ap-

proach, the patient can be positioned supine using a thick roll under the ipsilateral shoulder, rotating the head con-tralaterally and flexing it. The surgery starts with a ret-roauricular C-shaped incision, followed by elevation and anterior retraction of the myofascial flap to expose the spine of Henle and the posterior margin of the external acoustic canal (Fig. 11A). Macewen’s triangle is a land-mark for identifying the mastoid antrum 1.5 cm below the surface and is limited superiorly by the suprameatal crest, anteroinferiorly by a line running along the superior and posterior margins of the external auditory canal and cross-ing the spine of Henle, and posteriorly by a tangential line from the posterior margin of the canal crossing the first line.

A mastoidectomy is the first step to this transpetrosal presigmoid approach. The suprameatal crest is the supe-rior limit for the initial drilling, the posterior wall of the canal is the anterior limit, the mastoid tip is the inferior limit, and the sigmoid sinus is the posterior limit. The drilling should start by drawing an inverted L over the cortical bone, exposing the first air cells, and employing a uniform depth of dissection (Fig. 11B). The next objec-tive is to reach the mastoid antrum cranially and expose the Trautmann’s triangle caudally (Fig. 11C). This triangle represents an area of the posterior fossa dura limited by the sigmoid sinus posteriorly; the jugular bulb inferiorly; and the superior petrosal sinus, middle fossa dura, and otic capsule superiorly (Fig. 11D).

The dura is opened along the sinuses and retracted an-teriorly. The petrosal surface of the cerebellum is initially exposed; however, after cerebrospinal fluid is drained from the cerebellopontine cistern, the surgeon can observe the flocculus behind the CN VII/VIII complex and the anterior inferior cerebellar artery (Fig. 11E and F). After dissecting the cistern widely, the surface of the middle cer-ebellar peduncle and the root entry zone of the trigeminal nerve can be seen (Fig. 11F). Its smaller motor root exits the pons superomedial to the larger sensory root. This ap-proach provides a significantly more lateral approach to

the supratrigeminal, peritrigeminal, and lateral pontine safe zones (Fig. 11G).

discussionThe brainstem had been considered an untouchable re-

gion for decades, protected anteriorly by the clivus, later-ally by the petrous bones, posteriorly by the cerebellum, and superiorly by the diencephalon. Because it harbors significant cranial nerve nuclei and tracts in a small sec-tional area, any manipulation can cause an elevated risk of morbidity.

Almost concurrently, Pool in 196831 and Lassiter et al. in 197125 reported their pioneering surgical series that of-fered a surgical option for managing intrinsic brainstem pathology. The following decades were marked by strong discussions on the indications and feasibility of brainstem surgery. But the expertise gained by select groups of neu-rosurgeons using an essential combination of progress in microsurgical instruments and techniques, skull base sur-gery, surgical planning with magnetic resonance imaging and image guidance, neuroanesthesia and neurointensive care, and intraoperative monitoring produced significantly better results and outcomes for patients with brainstem le-sions. For instance, Bricolo et al.5,6 have published exten-sive work on brainstem tumors, introducing and discuss-ing different safe entry zones. Similarly, the senior author of this paper (R.F.S.) has published a number of studies on the evolution of surgery for cavernous malformations of the brainstem.1,11,32 Finally, other select groups have con-tributed meaningful additions to the surgical management of brainstem hemangioblastomas.38,41

Experience in multiple surgical routes is key to select-ing the right approach to different locations within the brainstem. A detailed review of the lesion on thin-cut T1-, T2-, and susceptibility-weighted MRI sequences is required. Image guidance supports anatomical and radio-logical knowledge, ensuring a perfect trajectory to the le-sion. Nevertheless, whenever lesions do not rise to the pial or ependymal surface, it is essential to have a fundamen-tal understanding of the concept of safe entry zones. Such landmarks have been described in surgical series and neu-roanatomy studies (Tables 1 and 2). They represent small regions between vital neural structures and take advantage of the sparseness of perforators.2,5,20,23 Moreover, manipu-lation through these corridors is believed to minimize deficits when performed by experienced surgeons.

It is important to note that large intrinsic lesions may

table 1. surgical approaches to the brainstem according to lesion location

Lesion Location Anterior Lateral Posterior

Midbrain OZ, mini-OZ, PT Anterolateral: OZ, mini-OZ, ST Median SCITPosterolateral: paramedian or extreme lateral SCIT

Pons ST ± TT ± AP, RL, RS RS SOTV ± C-1 laminoplastyMedulla FL Upper Medulla: FL, RS SOTV

Lower Medulla: FL

AP = anterior petrosectomy; FL = far lateral; OZ = orbitozygomatic; PT = pterional; RL = retrolabyrinthine; RS = retrosigmoid; SCIT = supracer-ebellar infratentorial; SOTV = suboccipital telovelar; ST = subtemporal; TT = transtentorial.



Fig. 11. Retrolabyrinthine approach. a: The patient is positioned supine, and the head is rotated toward the contralateral side. A retroauricular C-shaped incision is placed on the left side; the skin flap is reflected forward. Exposure of the lateral surface of the mastoid bone with its landmarks is shown. b: Initial mastoid cortical drilling initiated just posterior to the spine of Henle. The mastoidectomy starts by drilling a straight cut along the temporal line aiming toward the sinodural angle. A second perpendicular line is drilled from the initial point toward the mastoid tip, bounded by the external auditory canal anteriorly. c: The mastoid cortex was removed and the air cells are open. Koerner’s septum is thinned, unveiling the mastoid antrum, which is the largest mastoid cell. d: The semicircular canals are identified. The dura in front of the sigmoid sinus and above the jugular tubercle delineates Trautmann’s triangle, e: The posterior fossa dura is then opened, taking into consideration the anatomy of the endolymphatic sac. The petrosal surface of the cerebellum comes to view, with the flocculus behind the CN VII/VIII complex. The distal anterior inferior cerebellar artery is also noted in relation to such complex. F: This presigmoid approach provides a straighter route to the lateral surface of the pons, although it results in a limited view and requires time-consuming drilling. The superior and middle neurovascular complexes of the cerebellopontine angle are exposed. The trigeminal nerve root entry zone determines 3 safe entry zones on the lateral pons: the lateral pontine (LPZ), peritrigeminal (PTZ), and supratrigeminal (STZ) zones. g: The area of exposure on the lateral pons that is provided by the retrolabyrinthine approach is shown by the shaded area. LSC = lateral semicircular canal; m. = muscle; mid. = middle; mid. cerebell. ped. = middle cerebellar peduncle; PSC = posterior semicircular canal; SSC = superior semicircular canal; supramast. = supramastoid.



distort the present surgical windows and their landmarks and safe entry zones provided by each approach.

midbrain lesionsBoth pterional and orbitozygomatic approaches provide

a straightforward route to the ipsilateral cerebral peduncle (Table 1). The orbitozygomatic approach offers the widest vertical angle of attack and surgical freedom in the pe-duncular region. Lesions surfacing anteriorly on the pe-duncle, pontomesencephalic junction, or upper pons can be managed through these approaches. The anterior mes-encephalic zone can also be nicely exposed when neither color nor volume alteration is visible on the pial surface. We currently use keyhole approaches rather than exten-sive dissections and large craniotomies whenever possible. The mini-modified orbitozygomatic (mini-OZ) approach generally replaces full orbitozygomatic craniotomies. The mini-OZ requires a significantly smaller skin incision, just behind the hairline, and the bone opening is reduced to a 3-cm bone piece, starting just lateral to the supraorbital notch and extending to a point just lateral to the fronto-zygomatic suture (Fig. 12). Our group described the same anatomical exposure when quantitatively comparing the mini-OZ, pterional, and orbitozygomatic approaches.18 Another interesting option is the transciliary supraorbital approach, which utilizes a discreet eyebrow incision and provides a similar bone opening and corridor as that of the mini-OZ (Fig. 12 inset).10

The subtemporal approach exposes the entire lateral surface of the midbrain. With proper planning, the crani-otomy will provide an adequate anterior or posterior view of the middle incisural space. However, the extreme lateral supracerebellar infratentorial approach provides an angled view of the posterolateral surface of the midbrain, without the drawbacks of a posterior subtemporal microdissection and retraction. The 2-point method may aid such decision making when approaching lesions on the midbrain teg-mentum.7 Lesions that are deeper and closer to the lateral surface should be approached through the lateral mesen-

cephalic sulcus. The long axis of the lesion will dictate whether the subtemporal or extreme lateral supracerebel-lar infratentorial corridors are used. Lesions within the tegmentum that are closer to the ventral surface may be best approached through the anterior mesencephalic zone via a pterional or orbitozygomatic approach.

Finally, tectal plate lesions may be reached through any variation of the supracerebellar infratentorial approach, de-pending on where the lesion is closest to the pial surface.12 The median supracerebellar infratentorial approach is em-ployed for midline lesions or when a dissection through the intercollicular region is necessary. The extreme lateral approach is reserved for lateral lesions on the tectal plate that occur on the surface of either the superior or inferior colliculus. This approach provides a wide horizontal angle of approach to any lesion on the posterolateral midbrain surface including the lateral mesencephalic sulcus. The exit of the trochlear nerve is an important landmark; dis-secting this nerve within the ambient cistern medial to its exit within the quadrigeminal cistern will lead to the infe-rior colliculus, which is superomedially placed.

pontine lesionsAs previously mentioned, the pterional and the orbito-

zygomatic approaches offer some exposure of the ventral upper pons. The transtentorial extension of the subtempo-

table 2. accessible safe entry zones by surgical approach

Approach Safe Entry Zones

Orbitozygomatic AMZSubtemporal AMZSubtemporal transtentorial AMZ, STZAnterior petrosectomy AMZ, STZ, PTZSuboccipital telovelar MSMedian SCIT LMS, IC, SC, IFExtreme lateral SCIT LMS, IC, SC, IFRetrosigmoid LMS, STZ, PTZ, LPZ, AL, PM, LMZFar lateral AL, PM, LMZ, olivaryRetrolabyrinthine LMS, STZ, PTZ, LPZ, AL, PM, LMZ,

olivary

AL = anterolateral sulcus of medulla; AMZ = anterior mesencephalic zone; IC = intercollicular; IF = infracolliclular; LMS = lateral mesencephalic sulcus; LMZ = lateral medullary zone; LPZ = lateral pontine zone; MS = median sulcus of fourth ventricle; PM = posterior median sulcus of medulla; PTZ = peritrigemi-nal zone; SC = supracollicular; SCIT = supracerebellar infratentorial; STZ = supratrigeminal zone.

Fig. 12. Keyhole alternatives to the classic orbitozygomatic (OZ) ap-proach to the brainstem, which demand less myofascial dissection and provide a significantly smaller bony opening but similar exposure of the basal cisterns compared with the classic OZ approach. In these 2 alternatives, 2 different skin incisions can be used, but the bony open-ings are essentially the same. The mini-modified OZ (mini-OZ) approach is shown with its reduced skin incision; the inset illustration shows the transciliary supraorbital approach with its discreet eyebrow incision. Reproduced with permission from Barrow Neurological Institute, Phoe-nix, Arizona.



ral approach allows exposure of the lateral upper pons as well as the pontomesencephalic junction. However, the tri-geminal nerve entry zone is actually the major landmark in approaching most of the anterolateral pontine lesions. Thus, approaches that aim toward the emergence of CN V represent the main routes to managing pontine lesions.

Despite the complexity of dissection and drilling work, the anterior petrosectomy offers the best direct anterior view of the peritrigeminal zone. The retrolabyrinthine approach is also time consuming and provides a limited posterolateral view of the CN V entry zone. However, the retrosigmoid approach is the workhorse for managing ante-rior and lateral pontine and high-riding medullary lesions, while providing a more posterolateral exposure of the pons and the middle cerebellar peduncle than other approaches (Fig. 13). Additionally, its widespread use in managing ves-tibular schwannomas and neurovascular conflicts provides experience for those who employ it to attack pontine lesions directly or through the peritrigeminal, supratrigeminal, and lateral pontine safe zones.

Dorsal pontine lesions abutting or close to the rhomboid fossa can be reached through a suboccipital telovelar ap-proach. The depth of the rhomboid fossa is rich in nuclei and tracts, which limits free manipulation. The facial col-liculus is the major landmark and guides the surgeon to the supracollicular and infracollicular safe zones when the lesion does not reach the surface. Lastly, cranial lesions can be resected through the median sulcus.

medullary lesionsThe far-lateral approach is chosen to manage anterolat-

eral lesions within the medulla. Cranial lesions not abut-ting the pial surface are reached through the olivary safe

zone. Otherwise, caudal lesions are approached through the anterolateral sulcus between the emergence of the hy-poglossal and C-1 spinal nerve rootlets.

Lateral lesions close to the pontomedullary junction re-quire mainly a retrosigmoid approach. This craniotomy can also be employed to manage some lateral caudal lesions if the long axis of the lesion is amenable to the oblique angle of view provided by the retrosigmoid approach (Fig. 13).4 Deeper dorsolateral lesions can be managed by tailoring an incision over the lateral medullary zone. Finally, posterior lesions can be directly accessed via a median suboccipital approach. The neurosurgeon should avoid any manipula-tion on the calamus scriptorius, an extremely eloquent re-gion, populated by the nuclei of the lower cranial nerves. A posterior midline medullary incision below the obex is advised instead.

a note of cautionBrainstem lesions represent some of the most challeng-

ing entities faced by neurosurgeons. In this article, we have summarized the approaches and safe entry zones into the brainstem. However, no region in the brainstem is truly “safe” for entry. The zones presented in this paper represent the regions through which lesions can be accessed with the least morbidity. As clinical experience from our group has demonstrated, surgery in the brainstem is associated with a high rate of temporary, but real, deficits.1 Practitioners with limited experience operating in the posterior fossa and the brainstem should consider referring these patients to high-volume centers with the requisite experience.

conclusionsProgress in medical technology combined with the

experience and dissemination of knowledge regarding microsurgery and brainstem pathology changed the para-digm that the brainstem represented a “no man’s land” for neurosurgeons.

Our revisit of the main safe entry zones provided by each major surgical approach to the brainstem, which in-cludes detailed dissection images and shows areas of ex-posure, may allow better planning to approach lesions of the brainstem and may help disseminate the techniques for successful resection of intrinsic brainstem lesions.

acknowledgmentsThe work described in this paper was funded by the Barrow

Neurological Foundation and by the Newsome Chair in Neurosur-gery Research held by Dr. Preul.

references 1. Abla AA, Lekovic GP, Turner JD, de Oliveira JG, Porter

R, Spetzler RF: Advances in the treatment and outcome of brainstem cavernous malformation surgery: a single-center case series of 300 surgically treated patients. Neurosurgery 68:403–415, 2011

2. Baghai P, Vries JK, Bechtel PC: Retromastoid approach for biopsy of brain stem tumors. Neurosurgery 10:574–579, 1982

3. Bailey P, Buchanan DN, Bucy PC: Intracranial Tumors of Infancy and Childhood. Chicago: University of Chicago Press, 1939

4. Bricolo A: Brainstem tumors, in Sindou M (ed): Practical Handbook of Neurosurgery. Vienna: Springer, 2009

Fig. 13. Illustration of a retrosigmoid craniotomy and microsurgical resection of an intrinsic lesion in the posterolateral medulla. The floc-culus and choroid plexus protruding through the foramen of Luschka are mobilized, and the lateral medullary zone is entered to gain access to the lesion. Reproduced with permission from Barrow Neurological Insti-tute, Phoenix, Arizona. First published in Deshmukh et al: J Neurosurg 121:723–9, 2014.



5. Bricolo A, Turazzi S: Surgery for gliomas and other mass le-sions of the brainstem. Adv Tech Stand Neurosurg 22:261–341, 1995

6. Bricolo A, Turazzi S, Cristofori L, Talacchi A: Direct sur-gery for brainstem tumours. Acta Neurochir Suppl (Wien) 53:148–158, 1991

7. Brown AP, Thompson BG, Spetzler RF: The two-point method: Evaluating brainstem lesions. Barrow Quarterly 12:20–24, 1996

8. Brunso-Bechtold JK, Thompson GC, Masterton RB: HRP study of the organization of auditory afferents ascending to central nucleus of inferior colliculus in cat. J Comp Neurol 197:705–722, 1981

9. Cantore G, Missori P, Santoro A: Cavernous angiomas of the brain stem. Intra-axial anatomical pitfalls and surgical strate-gies. Surg Neurol 52:84–94, 1999

10. Cavalcanti DD, García-González U, Agrawal A, Crawford NR, Tavares PL, Spetzler RF, et al: Quantitative anatomic study of the transciliary supraorbital approach: benefits of additional orbital osteotomy? Neurosurgery 66 (6 Suppl Operative):205–210, 2010

11. Cavalcanti DD, Kalani MY, Martirosyan NL, Eales J, Spetz-ler RF, Preul MC: Cerebral cavernous malformations: from genes to proteins to disease. J Neurosurg 116:122–132, 2012

12. de Oliveira JG, Lekovic GP, Safavi-Abbasi S, Reis CV, Hanel RA, Porter RW, et al: Supracerebellar infratentorial approach to cavernous malformations of the brainstem: surgical vari-ants and clinical experience with 45 patients. Neurosurgery 66:389–399, 2010

13. Deshmukh VR, Figueiredo EG, Deshmukh P, Crawford NR, Preul MC, Spetzler RF: Quantification and comparison of telovelar and transvermian approaches to the fourth ventricle. Neurosurgery 58:ONS-202–ONS-207, 2006

14. Deshmukh VR, Rangel-Castilla L, Spetzler RF: Lateral infe-rior cerebellar peduncle approach to dorsolateral medullary cavernous malformation. J Neurosurg 121:723–729, 2014

15. Epstein F, McCleary EL: Intrinsic brain-stem tumors of childhood: surgical indications. J Neurosurg 64:11–15, 1986

16. Feiz-Erfan I, Horn EM, Spetzler RF: Transanterior perforat-ing substance approach to the thalamomesencephalic junc-tion. Neurosurgery 63:ONS69–ONS72, 2008

17. Ferroli P, Sinisi M, Franzini A, Giombini S, Solero CL, Broggi G: Brainstem cavernomas: long-term results of mi-crosurgical resection in 52 patients. Neurosurgery 56:1203–1214, 2005

18. Figueiredo EG, Deshmukh V, Nakaji P, Deshmukh P, Cru-sius MU, Crawford N, et al: An anatomical evaluation of the mini-supraorbital approach and comparison with standard craniotomies. Neurosurgery 59:ONS212–ONS220, 2006

19. Fischer B, Weber H: Express saccades and visual attention. Behav Brain Sci 16:553–567, 1993

20. Hebb MO, Spetzler RF: Lateral transpeduncular approach to intrinsic lesions of the rostral pons. Neurosurgery 66 (3 Suppl Operative):26–29, 2010

21. Hubel DH, LeVay S, Wiesel TN: Mode of termination of retinotectal fibers in macaque monkey: an autoradiographic study. Brain Res 96:25–40, 1975

22. Kumar RSV: Tuberculous brain stem abscesses in children. J Pediatr Neurol 2:101–106, 2004

23. Kyoshima K, Kobayashi S, Gibo H, Kuroyanagi T: A study of safe entry zones via the floor of the fourth ventricle for brain-stem lesions. Report of three cases. J Neurosurg 78:987–993, 1993

24. Lanzino G, Paolini S, Spetzler RF: Far-lateral approach to the craniocervical junction. Neurosurgery 57 (4 Sup-pl):367–371, 2005

25. Lassiter KR, Alexander E Jr, Davis CH Jr, Kelly DL Jr: Sur-gical treatment of brain stem gliomas. J Neurosurg 34:719–725, 1971

26. Matson DD, Ingraham FD Neurosurgery of Infancy and Childhood, ed 2. Springfield, IL: Thomas, 1969

27. Merchán MA, Saldaña E, Plaza I: Dorsal nucleus of the lat-

eral lemniscus in the rat: concentric organization and tono-topic projection to the inferior colliculus. J Comp Neurol 342:259–278, 1994

28. Munoz DP, Wurtz RH: Fixation cells in monkey superior col-liculus. I. Characteristics of cell discharge. J Neurophysiol 70:559–575, 1993

29. Mussi AC, Rhoton AL Jr: Telovelar approach to the fourth ventricle: microsurgical anatomy. J Neurosurg 92:812–823, 2000

30. Ono M, Ono M, Rhoton AL Jr, Barry M: Microsurgical anatomy of the region of the tentorial incisura. J Neurosurg 60:365–399, 1984

31. Pool JL: Gliomas in the region of the brain stem. J Neuro-surg 29:164–167, 1968

32. Porter RW, Detwiler PW, Spetzler RF, Lawton MT, Baskin JJ, Derksen PT, et al: Cavernous malformations of the brain-stem: experience with 100 patients. J Neurosurg 90:50–58, 1999

33. Raimondi AJ: Pediatric Neurosurgery: Theoretical Prin-ciples–Art of Surgical Techniques. New York: Springer, 1987

34. Ramina R, Mattei TA, de Aguiar PH, Meneses MS, Ferraz VR, Aires R, et al: Surgical management of brainstem cav-ernous malformations. Neurol Sci 32:1013–1028, 2011

35. Recalde RJ, Figueiredo EG, de Oliveira E: Microsurgical anatomy of the safe entry zones on the anterolateral brain-stem related to surgical approaches to cavernous malforma-tions. Neurosurgery 62 (3 Suppl 1):9–17, 2008

36. Sala F, Manganotti P, Tramontano V, Bricolo A, Gerosa M: Monitoring of motor pathways during brain stem surgery: what we have achieved and what we still miss? Neurophysiol Clin 37:399–406, 2007

37. Strauss C, Lütjen-Drecoll E, Fahlbusch R: Pericollicular sur-gical approaches to the rhomboid fossa. Part I. Anatomical basis. J Neurosurg 87:893–899, 1997

38. Wind JJ, Bakhtian KD, Sweet JA, Mehta GU, Thawani JP, Asthagiri AR, et al: Long-term outcome after resection of brainstem hemangioblastomas in von Hippel-Lindau disease. J Neurosurg 114:1312–1318, 2011

39. Yaşargil MG: Microsurgical Anatomy of the Basal Cis-terns and Vessels of the Brain, Diagnostic Studies, Gener-al Operative Techniques and Pathological Considerations of the Intracranial Aneurysms. Stuttgart: Thieme, 1984

40. Zabramski JM, Kiriş T, Sankhla SK, Cabiol J, Spetzler RF: Orbitozygomatic craniotomy. Technical note. J Neurosurg 89:336–341, 1998

41. Zhou LF, Du G, Mao Y, Zhang R: Diagnosis and surgical treatment of brainstem hemangioblastomas. Surg Neurol 63:307–316, 2005

disclosure The authors report no conflict of interest concerning the materi-als or methods used in this study or the findings specified in this paper.

author contributionsConception and design: Cavalcanti. Acquisition of data: Caval-canti. Analysis and interpretation of data: Cavalcanti. Drafting the article: Cavalcanti, Kalani. Critically revising the article: Preul, Kalani, Spetzler. Reviewed submitted version of manuscript: Preul, Kalani, Spetzler. Administrative/technical/material support: Preul. Study supervision: Preul, Spetzler.

correspondenceDaniel D. Cavalcanti, Departamento de Neurocirurgia, Instituto Estadual do Cérebro Paulo Niemeyer, Rua do Resende, 156, Rio de Janeiro, RJ, Brazil 20231-092. email: [email protected].


Endoscopic Endonasal Transclival Resection of a Brainstem Cavernoma: A DetailedAccount of Our Technique and Comparison with the Literature

Stefan Linsler and Joachim Oertel

-OBJECTIVE: To report a technique of endoscopic trans-clival resection of a hemorrhagic brainstem cavernousmalformation manifesting in the ventral pons.

-METHODS: A 29-year-old woman presented with numb-ness and tingling of the right arm and leg and loss of finemotor control. Magnetic resonance imaging revealed acavernoma in the ventromedial brainstem on the ventralsurface. A purely endoscopic, endonasal, transclivalapproach was used to resect this cavernoma. Computedtomography/magnetic resonance imaging merged naviga-tion (StealthStation, Medtronic) was used.

-RESULTS: The patient had no neurologic deficits post-operatively. The motor control loss and tingling dis-appeared. She did not experience any complications.Cerebrospinal fluid leakage appeared to result from usingthe very small opening of the skull base and dura mater andwas the reason for the use of a lumbar drain for severaldays. At the 6-week follow-up examination, the patientwas in excellent condition with no neurologic deficits andhad returned to her full-time job.

-CONCLUSIONS: Successful endoscopic, endonasal,transclival resection of a brainstem cavernous malforma-tion was described. This patient experienced improvementin neurologic symptoms after surgery without morbidity.Technologic advances in endoscopic skull base ap-proaches provide access to lesions of the brainstem thatpreviously required more invasive approaches. The endo-nasal transclival approach provides the most direct routeto ventral pontine lesions. Early intervention in brainstemcavernous malformation is indicated and should be

performed with an individualized approach taking intoconsideration the possible complications.

INTRODUCTION

B rainstem cavernous malformations (BSCMs) exhibit morefrequent symptomatic hemorrhages than supratentorialcavernous malformations.1 Generally, BSCMs represent

about 10% of all vascular lesions of the brain.2e4 They occurequally among male and female patients and usually manifestbetween the ages of 20 and 40 years.2e4 Patients withcavernous malformations can present with headaches, seizures,or neurologic deficits, or the lesions can be found incidentally.BSCMs in particular pose a difficult problem for surgeons andpatients. The natural history of cavernous malformations hasbeen studied in detail and is still debated.2,4,5 When hemorrhagesoccur, cavernous malformations have a high risk of rebleeding,with rebleeding rates of BSCM of 5%e35% per year.1,6

Surgical intervention is indicated in patients with hemorrhagesand a BSCM that manifests on the pial surface.3 The surgicalapproach is dictated by the location of the cavernousmalformation within the brainstem. The main goal is tominimize the amount of healthy tissue that must be traversedto achieve a complete resection. The risks and benefits ofsurgical treatment must be weighed against possible morbidityresulting from surgery. However, brainstem hemorrhages andrepeat hemorrhages can also result in a very poor condition.7

Although surgical resection of BSCM is challenging, as a resultof modern magnetic resonance imaging (MRI), excellentneuronavigation systems, and new surgical tools in the lastdecade, surgery is considered more frequently.

Key words- Brain stem- Cavernous malformation- Endonasal- Endoscopy- Transclival

Abbreviations and AcronymsBSCM: Brainstem cavernous malformationCSF: Cerebrospinal fluidCT: Computed tomographyMRI: Magnetic resonance imaging

Klinik für Neurochirurgie, Universitätsklinikum des Saarlandes, Homburg, Germany

To whom correspondence should be addressed: Stefan Linsler, M.D.[E-mail: [email protected]]

Citation: World Neurosurg. (2015) 84, 6:2064-2071.http://dx.doi.org/10.1016/j.wneu.2015.08.029

Supplementary digital content available online.

Journal homepage: www.WORLDNEUROSURGERY.org

Available online: www.sciencedirect.com

1878-8750/$ - see front matter ª 2015 Elsevier Inc. All rights reserved.

2064 www.SCIENCEDIRECT.com WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2015.08.029

Technical Note

http://crossmark.crossref.org/dialog/?doi=10.1016/j.wneu.2015.08.029&domain=pdf

mailto:[email protected]

http://dx.doi.org/10.1016/j.wneu.2015.08.029

http://www.WORLDNEUROSURGERY.org

www.sciencedirect.com/science/journal/18788750



Endoscopic procedures are increasingly used for skull base sur-gery. Many publications on endonasal endoscopic surgery stressthe less invasive nature of these techniques8e15 and the feasi-bility of approaching the clivus.13,16 We report our technique ofendoscopic transclival resection of a BSCM manifesting on theventral pons and compare our case with related cases reportedmore recently in the literature.17e20 To our knowledge, this is thefirst reported successful resection of a BSCM via an endoscopictransnasal transclival approach without any postoperative com-plications and with no neurologic deficits.

CASE REPORT

Presentation of Clinical CaseA 29-year-old woman presented with numbness and tingling ofthe right arm and leg. She also described loss of fine motorcontrol of the right hand and transient diplopia. She reportedexperiencing a transient headache 2 weeks ago. MRI performedafter the patient arrived in our department revealed an enhancinglesion in the ventromedial brainstem on the ventral surface(Figure 1).

It was decided to offer the patient surgical resection because thiswas the first episode of symptomatic bleeding, and there was ahigh risk of rebleeding resulting in a deteriorating condition. Weconsidered approaching the lesion using an endoscopic, endo-nasal, transclival approach (Video I). We believed that we wouldminimize the amount of intact neural tissue with this approach.Via an endonasal transclival approach, we had a direct angle tothe ventral brainstem and the lesion, which was closed to thesurface (Figure 1).

Surgical Setting and TechniqueThe patient was placed supine with the upper part of the bodyslightly elevated to about 20! and the head tilted to the left. The

patient’s head was fixed with a 3-pinhead-fixation system; general anes-thesia was administered with oro-tracheal intubation. Lateral fluoroscopy(C-arm) was used for intraoperativeimaging. An intraoperative computedtomography (CT) scan with SiemensCT suite (Siemens Healthcare GmbH,

Erlangen, Germany) was used for MRI/CT-based merged neuro-navigation with StealthAir System (Medtronic, Minneapolis,Minnesota, USA) to achieve the most accurate navigation in thehigh-risk area of the lesion in the brainstem (Figure 2). Our surgicalsetting included intraoperative neurophysiologic monitoringof somatosensory evoked potentials and auditory evokedpotentials.

Video available atWORLDNEUROSURGERY.org

Figure 1. Preoperative magnetic resonance imaging revealed anenhancing lesion in the ventromedial brainstem close to the ventralsurface in sagittal (A) and axial (B) views. In the axial view, the basilarartery is in the middle of the ventral surface of the brainstem in front ofthe lesion. Axial T2 fluid attenuation inversion recovery image (C)indicated a cavernous malformation most likely. The arrows demonstratethe most direct approach and angle to the lesion with minimal opening ofthe brainstem surface via the endonasal transclival route.

WORLD NEUROSURGERY 84 [6]: 2064-2071, DECEMBER 2015 www.WORLDNEUROSURGERY.org 2065

TECHNICAL NOTE


The endoscopic equipment consisted of a series of various rigid-rod lenses Hopkins optics, a Xenon cold light source, a digital 1-chip camera, a high-resolution video monitor screen, and a digitalrecording system (AIDA; KARL STORZ GmbH & Co., Tuttlingen,Germany).

The nose and the nasal cavities were prepared with applica-tion of a nasal decongestant and an alcohol-based disinfec-tant. Mepivacaine with 1:100,000 epinephrine was injectedinto nasal mucosa for hemostasis. The nasal approach wasperformed by the authors. An endonasal approach to the cli-vus was used initially with the same technique as reported

previously by the authors in a large series of sellarpathologies.15,21

RESULTS

Detailed Account of Surgical Approach and TechniqueThe nasal approach was performed by the authors withoutpreparation of a nasoseptal flap. Initially, we decided to use amononostril endonasal approach to the clivus with the sametechnique as we reported previously in a large series of sellarpathologies.15,21 Via the right mononostril approach with careful

Figure 2. Use of merged neuronavigation with magnetic resonance imaging (A) and computed tomography (B). Weachieved excellent accuracy of neuronavigation using merged images at the skull base and brainstem.


TECHNICAL NOTE



insertion of the speculum under direct endoscopic control, wepassed the inferomedial aspect of the middle turbinate underlateral fluoroscopy to the sellar floor until the sphenoid sinus wasreached. The nose was carefully dilated in several steps. Withthis use of the speculum, the nasal mucosa can be preservedthrough the whole surgical procedure, and almost any mucosalbleeding can be prevented. After exploration of both sphenoidostia, the anterior face of the sphenoid sinus was removed withKerrison rongeurs to expose the complete sphenoid sinus andclivus. CT/MRI merged neuronavigation and anatomic landmarkswere used to identify the carotid arteries bilaterally. Because of anarrow corridor of the clivus between the internal carotid arteries(Figure 3), we had to change to a binostril approach without aspeculum. Otherwise, the surgeon was unable to handle theendoscope and instruments effectively and safely within thisnarrow space. A 2-handed technique with a fixed endoscopewas used during the procedure in this case. The clivus was drilleddown to the dura mater in the midline with high-speed electricdiamond drill. The sellar floor and the basion were the superiorand inferior limitation of the bone resection. The lateral extent ofdrilling was limited by the petrous part of the internal carotid ar-teries bilaterally. Dura mater was opened in the midline. Thebasilar artery could be directly identified slightly diverted to theleft as expected in the neuronavigation. Using image-guidednavigation, the bleeding was located to the right of the basilarartery. A small corticectomy was made, and dark blood drained

out of the cavernoma cavity without manipulation of the perfo-rating arteries of the brainstem. With gentle suction anddissection, the cavernoma was completely removed. A 0! endo-scope was used for inspection of the cavity as far as possible.There was no remnant cavernoma. After confirmation of excel-lent hemostasis, closure of the dura mater was performed withautologous periumbilical fat graft, TachoSil, and fibrin glue. Noattempt was made to reconstruct the clival bone. The bone anddural opening was very small, so we decided not to use anasoseptal flap for additional closure. After repositioning of thenasal septum, nasal packing was placed. After the procedure, alumbar drain was inserted directly and was continued for 5 days.An intraoperative CT scan was performed at the end of theprocedure for resection control (Figure 4). Surgical time was 189minutes. Histopathologic analysis revealed a cavernousmalformation as demonstrated in Figure 5.

Postoperative CourseThe patient was extubated immediately after the procedure andadmitted to the intensive care unit for neurologic monitoring for24 hours. She was speaking and tolerating a regular diet onpostoperative day 1. The patient reported reduction of thenumbness and tingling of the right arm and leg within 12 hourspostoperatively. No new neurologic deficits were detectedpostoperatively.

Lumbar drainage was continued postoperatively for 5 days withbed rest. After removing the lumbar drain, the patient wasmobilized successfully. She had no ataxia and no fine motor skill

Figure 3. The narrow corridor (green arrow) of the clivus between theinternal carotid arteries is only approximately 12 mm. A binostrilapproach was used for better handling and control of surgicalinstruments.

Figure 4. Intraoperative computed tomography scan for control ofresection at the end of the procedure revealing total resection of thebleeding and cavernoma. The image also shows the very limited openingof the clivus to the target of only 7 mm (arrow).


TECHNICAL NOTE


deficits. No cerebrospinal fluid (CSF) leak was appreciated.Postoperative MRI demonstrated complete resection of thecavernous malformation and bleeding (Figure 6). The patient wasdischarged to home on postoperative day 7 in good conditionwithout any neurologic deficits.

At 6-week follow-up examination, the patient presented inexcellent condition without any neurologic symptoms. No CSFrhinorrhea had developed. She had returned to her full-time job aslaboratory technical assistant.

Before the present case, 4 comparable case reports were pub-lished in the literature. Two patients had persistent CSF leakagepostoperatively and underwent revision surgery using differentsurgical techniques. Remnant cavernous malformation wasdetected on follow-up MRI in 1 case. No morbidity was reported.The reported cases from the literature are compared with thepresent case in Table 1.

DISCUSSIONMany studies have been performed of BSCMs to understand thepathology better and to decide when which kind of surgery isindicated. Some studies showed that the rates of hemorrhageare similar for cavernous malformations in the supratentorial andinfratentorial compartments.6 However, 1 series showed that therates of hemorrhage for infratentorial lesions were 30 timesgreater in prospectively followed patients with BSCMs having a5% per years per lesion rate of hemorrhage.5 Fritschi et al.22

reported a series of 41 patients. Their retrospective seriesestimated a minimum bleeding rate of 2.7% per year and anaverage rebleeding rate of 21% per year per lesion. Of 12patients who died after a hemorrhage, 5 died after the firsthemorrhage, and 7 died after subsequent hemorrhages.

BSCMs are difficult to treat because of their location in aneloquent area. Resection of BSCMs is associated with significantmorbidity. However, total resection is curative. Additionally,cavernomas displace, rather than invade—especially afterbleeding with a hemorrhagic edge—,nervous tissue, making itpossible to resect lesions that manifest on the pial surface withminimal risk of a persistent neurologic deficits. Surgical treat-ment should be considered in every BSCM with an individualizedapproach that takes in consideration the possible complications.

The ideal approach to the brainstem and the lesion provides thebest visualization, minimizes retraction, and provides a goodworking angle to the cavernoma resulting in good postoperativeresults with reduction of the surgical morbidity. Many surgicalapproaches have been proposed to access cavernomas in thislocation, including the lateral transpeduncular route,23 thepresigmoid approach,24 the extradural25 and intradural26

transpetrosal approaches, the retrosigmoid or extendedretrosigmoid approach,27 and the combined retrosigmoid-subtemporal approach.28,29 All of these approaches provide ac-cess to the anterolateral brainstem but are suboptimal foraddressing lesions that manifest on the pontine surface at or nearthe midline. The lateral or posterolateral approaches, particularlythe transpetrosal approaches, require extensive drilling of boneand manipulation and retraction of cranial nerves and vascularstructures. None of these approaches provide a direct workingangle to a BSCM on the ventromedial pons. Also, large newseries show up to a 30% risk of motor deficits after resection ofpontomedullary junction cavernomas when approached laterally1

caused by irrigation of the corticospinal tracts that runlongitudinally in the ventromedial pons.

Reisch et al.30 presented 2 cases of anterior cavernousmalformations resected using a microscopic technique througha transoral transclival technique, allowing direct access to theventral brainstem. Both patients had an improved neurologicexamination relative to their baseline at the 3-month follow-upevaluation, although 1 patient required reoperation for a CSFleak. This technique has been used infrequently since then.

At the present time, many lesions of the skull base can beaccessed and removed via an endoscopic endonasal approach. Itis important to determine preoperatively that the entire lesion isaccessible via an endoscopic approach and that the surgical toolsare adequate. The surgeon has to balance the advantages of adirect vector to the lesion and minimizing surgical trauma and theknown risks of CSF leaks and potentially dealing with bleedingwith endoscopic instruments.

New developments in endoscopic techniques have made itpossible to perform a transclival approach with endoscopicvisualization. The endoscopic, transnasal, transclival approach toBSCMs has previously been described in case reports.17e20 Inthe presented cases, 2 of 4 patients had persistent CSF leakageand underwent revision using different surgical techniques thandescribed in the case reports. In 1 patient, there was a residualcavernous malformation on follow-up MRI. Nayak et al.19

described endoscopic resection of BSCM in 4 patients viaendonasal and retrosigmoidal approaches. They pointed outthat they will consider prophylactic CSF drainage in futurecases using an endonasal transclival approach. In Table 1, wecompare the published cases and our case. In general, theauthors pointed out the importance of significant experience of

Figure 5. Histopathologic analysis with elastica van Gieson stainingrevealed typical findings of a cavernous malformation with largepathologic vessels with fibrotic vascular wall and hemosiderin in thesurrounding tissue in the resected tissue (magnification " 100).


TECHNICAL NOTE



the surgeon with respect to relevant anatomy, endoscopictechniques, and endoscopic instruments.17e19

The biggest concern we had before surgery using an endoscopicskull base approach was the risk of CSF leak. CSF leak is still the

most common complication in endoscopic skull base surgery,and although these has been dramatic improvement, thiscomplication is underreported in the literature. Since we beganour endoscopic technique, we have improved our CSF leak rate

Figure 6. Postoperative magnetic resonance imaging revealed totalresection of the brainstem cavernous malformation in sagittal (A) and axial(B) views. Postoperative computed tomography scan (C) demonstrates thevery small opening of the clivus for intradural handling for cavernomaremoval.

Table 1. Comparison of Clinical Course, Complications, and Dimension of Lesion in Reported Cases of Endonasal Transclival Resectionof Brainstem Cavernous Malformations

Case 1 (Sanborn et al., 201220) Case 2 (Kimball et al., 201218) Case 3 (Dallan et al., 201517) Case 4 (Present Case)

Age (years)/sex 17/male 59/female 15/male 29/female

Presenting symptoms Acute onset of headache, facialnumbness, left-sided hemiparesis,right sixth cranial nerve palsy,dysphagia

Intermittent dysarthria, rightfacial weakness, left arm and legweakness, loss of fine motorcontrol

Acute onset of headache,vomiting, diplopia, hearing loss,facial palsy

Numbness and tingling of rightarm and leg, loss of fine motorcontrol of right hand, transientdiplopia, transient headache

Dimensions (mm) 17 " 12 20 " 20 " 20 10 " 10 20 " 18 " 22

Complications CSF leak CSF leak Small residual of CM None

Postoperative neurologicdeficits

Worsening of left-side motorfunction, restricted horizontalgaze; improved in follow-up

Remained at preoperativeneurologic baseline; improved infollow-up

Recovered after 2 years None

CSF, cerebrospinal fluid; CM, cavernous malformation.


TECHNICAL NOTE


with application of an autologous periumbilical fat graft for duralclosure and use of lumbar drainage after dural repair in all casesof endonasal surgery. Aggressive lumbar CSF drainage is neededfor at least 5 days postoperatively. Otherwise, the risk of apersistent CSF leak is very high independent of the appliedtechnique for dural and skull base repair. A second method tominimize the risk of CSF leakage is the reduction of bone anddural opening. We used the merged CT/MRI neuronavigation toaccess the cavernoma via a minimized opening of the clivus anda small opening of the dura mater. The opening of the clivus isonly 7 mm (Figure 4). In consequence of the very small boneopening and consequently small dural opening, we had a lowrisk of CSF leakage postoperatively, and there was no need fora nasoseptal flap.

Significant bleeding after opening of the premontane dura materas described before20,31 did not occur. In general, the control ofbleeding during the approach to the brainstem is essential for asuccessful resection of BSCM or other lesions of the brainstemunder endoscopic view. Manipulation with instruments and

especially bipolar electrocautery near the brainstem or into thebrainstem must be minimized to avoid operative morbidity.

CONCLUSIONSWe have described in detail our surgical technique of endoscopic,endonasal, transclival resection of a BSCM manifesting on theventral surface of the pons. The endonasal transclival approachprovides the most direct route to ventral pontine lesions. A majorstep to prevent postoperative persistent CSF leakage is theinsertion of a lumbar drain for several days and a minimalizedopening of the skull base and dura mater. Early intervention inBSCM is indicated and should be performedwith an individualizedapproach taking into consideration the possible complications.

ACKNOWLEDGMENTSWe thank Y. J. Kim for providing the histopathologic image of theremoved brainstem cavernous malformation in the presentedcase.

REFERENCES1. Abla AA, Lekovic GP, Turner JD, de Oliveira JG,

Porter R, Spetzler RF. Advances in the treatmentand outcome of brainstem cavernous malforma-tion surgery: a single-center case series of 300surgically treated patients. Neurosurgery. 2011;68:403e414: discussion 414-415.

2. Barrow DL. Intraoperative misadventures:complication avoidance and management inaneurysm surgery. Clin Neurosurg. 2011;58:93e109.

3. Gross BA, Batjer HH, Awad IA, Bendok BR.Cavernous malformations of the basal ganglia andthalamus. Neurosurgery. 2009;65:7e18: discussion18-19.

4. Moriarity JL, Clatterbuck RE, Rigamonti D. Thenatural history of cavernous malformations. Neu-rosurg Clin N Am. 1999;10:411e417.

5. Porter PJ, Willinsky RA, Harper W, Wallace MC.Cerebral cavernous malformations: natural his-tory and prognosis after clinical deteriorationwith or without hemorrhage. J Neurosurg. 1997;87:190e197.

6. Kondziolka D, Lunsford LD, Flickinger JC,Kestle JR. Reduction of hemorrhage risk afterstereotactic radiosurgery for cavernous malfor-mations. J Neurosurg. 1995;83:825e831.

7. Samii M, Eghbal R, Carvalho GA, Matthies C.Surgical management of brainstem cavernomas.J Neurosurg. 2001;95:825e832.

8. Cappabianca P, Alfieri A, Colao A, et al.Endoscopic endonasal transsphenoidal surgeryin recurrent and residual pituitary adenomas:technical note. Minim Invasive Neurosurg. 2000;43:38e43.

9. Cappabianca P, Cavallo LM, Esposito F, DeDivitiis O, Messina A, De Divitiis E. Extendedendoscopic endonasal approach to the midlineskull base: the evolving role of transsphenoidalsurgery. Adv Tech Stand Neurosurg. 2008;33:151e199.

10. de Divitiis E, Cappabianca P. Endoscopic endo-nasal transsphenoidal surgery. Adv Tech StandNeurosurg. 2002;27:137e177.

11. Jho HD. Endoscopic pituitary surgery. Pituitary.1999;2:139e154.

12. Jho HD, Alfieri A. Endoscopic endonasal pituitarysurgery: evolution of surgical technique andequipment in 150 operations. Minim Invasive Neu-rosurg. 2001;44:1e12.

13. Kassam AB, Gardner P, Snyderman C, Mintz A,Carrau R. Expanded endonasal approach: fullyendoscopic, completely transnasal approach to themiddle third of the clivus, petrous bone, middlecranial fossa, and infratemporal fossa. NeurosurgFocus. 2005;19:E6.

14. Kassam AB, Prevedello DM, Carrau RL, et al.Endoscopic endonasal skull base surgery: analysisof complications in the authors’ initial 800 pa-tients. J Neurosurg. 2011;114:1544e1568.

15. Linsler S, Gaab MR, Oertel J. Endoscopic endo-nasal transsphenoidal approach to sellar lesions: adetailed account of our mononostril technique.J Neurol Surg B Skull Base. 2013;74:146e154.

16. Jho HD, Ha HG. Endoscopic endonasal skull basesurgery: part 3—the clivus and posterior fossa.Minim Invasive Neurosurg. 2004;47:16e23.

17. Dallan I, Battaglia P, de Notaris M, Caniglia M,Turri-Zanoni M. Endoscopic endonasal transclivalapproach to a pontine cavernous malformation:case report. Int J Pediatr Otorhinolaryngol. 2015;79:1584e1588.

18. Kimball MM, Lewis SB, Werning JW, Mocco JD.Resection of a pontine cavernous malformationvia an endoscopic endonasal approach: a casereport. Neurosurgery. 2012;71:186e193: discussion193-194.

19. Nayak NR, Thawani JP, Sanborn MR, Storm PB,Lee JY. Endoscopic approaches to brainstemcavernous malformations: case series and reviewof the literature. Surg Neurol Int. 2015;6:68.

20. Sanborn MR, Kramarz MJ, Storm PB, Adappa ND,Palmer JN, Lee JY. Endoscopic, endonasal,transclival resection of a pontine cavernoma: casereport. Neurosurgery. 2012;71:198e203.

21. Oertel J, Gaab MR, Tschan CA, Linsler S.Mononostril endoscopic transsphenoidalapproach to sellar and peri-sellar lesions: per-sonal experience and literature review. Br JNeurosurg. 2015;29:532e537.

22. Fritschi JA, Reulen HJ, Spetzler RF, Zabramski JM.Cavernous malformations of the brain-stem—areview of 139 cases. Acta Neurochir. 1994;130:35e46.

23. Hebb MO, Spetzler RF. Lateral transpeduncularapproach to intrinsic lesions of the rostral pons.Neurosurgery. 2010;66:26e29.

24. Hauck EF, Barnett SL, White JA, Samson D. Thepresigmoid approach to anterolateral pontinecavernomas. J Neurosurg. 2010;113:701e708.

25. Francois P, Ben Ismail M, Hamel O, Bataille B,Jan M, Velut S. Anterior transpetrosal and sub-temporal transtentorial approaches for pontinecavernomas. Acta Neurochir. 2010;152:1321e1329.

26. Steiger HJ, Hanggi D, Stummer W, Winkler PA.Transpetrosal approach—response. J Neurosurg.2006;105:337e338.

27. Quinones-Hinojosa A, Chang EF, Lawton MT. Theextended retrosigmoid approach: an alternative toradical cranial base approaches for posterior fossalesions. Neurosurgery. 2006;58:ONS-208eONS-214:discussion ONS-214.

28. Bertalanffy H. Presigmoid approach for cavernousangioma in the pons—technical note. Neurol Medchir (Tokyo). 2002;42:97e98.

29. Bertalanffy H, Benes L, Miyazawa T, Alberti O,Siegel AM, Sure U. Cerebral cavernomas in theadult. Review of the literature and analysis of 72surgically treated patients. Neurosurg Rev. 2002;25:1e53.


TECHNICAL NOTE

http://refhub.elsevier.com/S1878-8750(15)01045-1/sref1

















































































































































30. Reisch R, Bettag M, Perneczky A. Transoraltransclival removal of anteriorly placed cavernousmalformations of the brainstem. Surg Neurol. 2001;56:106e115.

31. Esposito F, Becker DP, Villablanca JP, Kelly DF.Endonasal transsphenoidal transclival removal of

prepontine epidermoid tumors: technical note.Neurosurgery. 2005;56:E443: discussion E443.

J.O. is a consultant to KARL STORZ GmbH & Co.

Received: 28 June 2015; accepted: 25 August 2015

Citation: World Neurosurg. (2015) 84, 6:2064-2071.http://dx.doi.org/10.1016/j.wneu.2015.08.029

Journal homepage: www.WORLDNEUROSURGERY.org

Available online: www.sciencedirect.com

1878-8750/$ - see front matter ª 2015 Elsevier Inc.All rights reserved.


TECHNICAL NOTE














J Neurosurg Volume 122 • March 2015

cliNical articleJ Neurosurg 122:653–662, 2015

abbreviatioNs BSCM = brainstem CM; CM = cavernous malformation; CST = corticospinal tract; DTI = diffusion tensor imaging; DTT = diffusion tensor tractography; FA = fractional anisotropy; ICP = inferior cerebral peduncle; ML = medial lemniscus; MLF = medial longitudinal fasciculus; mRS = modified Rankin Scale.submitted April 9, 2013. accepted November 18, 2014.iNclude wheN citiNg Published online January 9, 2015; DOI: 10.3171/2014.11.JNS13680.disclosure The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

The utility of preoperative diffusion tensor imaging in the surgical management of brainstem cavernous malformationsbruno c. Flores, md,1 anthony r. whittemore, md, phd,2 duke s. samson, md,1 and samuel l. barnett, md1

Departments of 1Neurological Surgery and 2Radiology, University of Texas Southwestern Medical Center, Dallas, Texas

obJect Resection of brainstem cavernous malformations (BSCMs) may reduce the risk of stepwise neurological dete-rioration secondary to hemorrhage, but the morbidity of surgery remains high. Diffusion tensor imaging (DTI) and diffu-sion tensor tractography (DTT) are neuroimaging techniques that may assist in the complex surgical planning necessary for these lesions. The authors evaluate the utility of preoperative DTI and DTT in the surgical management of BSCMs and their correlation with functional outcome.methods A retrospective review was conducted to identify patients who underwent resection of a BSCM between 2007 and 2012. All patients had preoperative DTI/DTT studies and a minimum of 6 months of clinical and radiographic follow-up. Five major fiber tracts were evaluated preoperatively using the DTI/DTT protocol: 1) corticospinal tract, 2) medial lemniscus and medial longitudinal fasciculus, 3) inferior cerebellar peduncle, 4) middle cerebellar peduncle, and 5) superior cerebellar peduncle. Scores were applied according to the degree of distortion seen, and the sum of scores was used for analysis. Functional outcomes were measured at hospital admission, discharge, and last clinic visit using modified Rankin Scale (mRS) scores.results Eleven patients who underwent resection of a BSCM and preoperative DTI were identified. The mean age at presentation was 49 years, with a male-to-female ratio of 1.75:1. Cranial nerve deficit was the most common presenting symptom (81.8%), followed by cerebellar signs or gait/balance difficulties (54.5%) and hemibody anesthesia (27.2%). The majority of the lesions were located within the pons (54.5%). The mean diameter and estimated volume of lesions were 1.21 cm and 1.93 cm3, respectively. Using DTI and DTT, 9 patients (82%) were found to have involvement of 2 or more major fiber tracts; the corticospinal tract and medial lemniscus/medial longitudinal fasciculus were the most commonly affected. In 2 patients with BSCMs without pial presentation, DTI/DTT findings were important in the selection of the surgical approach. In 2 other patients, the results from preoperative DTI/DTT were important for selection of brainstem entry zones. All 11 patients underwent gross-total resection of their BSCMs. After a mean postoperative follow-up dura-tion of 32.04 months, all 11 patients had excellent or good outcome (mRS Score 0–3) at the time of last outpatient clinic evaluation. DTI score did not correlate with long-term outcome.coNclusioNs Preoperative DTI and DTT should be considered in the resection of symptomatic BSCMs. These imaging studies may influence the selection of surgical approach or brainstem entry zones, especially in deep-seated lesions without pial or ependymal presentation. DTI/DTT findings may allow for more aggressive management of lesions previously considered surgically inaccessible. Preoperative DTI/DTT changes do not appear to correlate with functional postoperative outcome in long-term follow-up.http://thejns.org/doi/abs/10.3171/2014.11.JNS13680Key words cavernous malformations; brainstem; diffusion tensor imaging; diffusion tensor tractography; resection; outcome; vascular disorders

653©AANS, 2015

b. c. Flores et al.

CAVERNOUS malformations (CMs) are uncommon lesions, with a estimated prevalence of 0.4%–0.8%.26,31,46,53 They represent 10%–20% of all CNS

vascular malformations, with the majority of cases involv-ing supratentorial structures. Approximately 10%–35% of CMs arise in the brainstem16,28 and may be associated with initial hemorrhage and rebleeding rates higher than those occurring with supratentorial and spinal cord lesions. Re-section may reduce the risk of a stepwise neurological deterioration, but the complexity of brainstem anatomy makes this task technically demanding and is associated with the risk of additional brainstem injury.28

Several adjuvant techniques have been implemented in the surgical management of brainstem lesions, such as functional MRI, intraoperative frameless stereotactic nav-igation, and intraoperative MRI. None of those methods can precisely visualize white matter tracts or define essen-tial aspects of tract location or displacement.10,11 Diffusion tensor imaging (DTI) and diffusion tensor tractography (DTT) are promising neuroimaging techniques that help overcome some of those limitations.6,8,10,11,33,36 They have become useful clinical tools that can delineate function-ally important white matter tracts for surgical planning.6 Tractography based on diffusion MRI explores the cor-relation between water diffusion and brain structure to delineate the course of white matter pathways. Diffusion tensor tractography (DTT) follows a white matter tract from voxel to voxel in 3 dimensions by assuming that the direction of least restricted diffusion corresponds to the orientation of axons. As a result of this predictable pattern of diffusion, which is orientation dependent (or anisotro-pic), location and orientation of white matter tracts can be determined.6,13,14,33,34,50

Several articles have described the implementation of DTI for a multitude of neurological diseases and neuro-anatomical studies;8,10,14,22,38,42,50,58,59 however, none has addressed its specific application in the management of brainstem cavernous malformations (BSCMs).

The goals of the present study were to determine if preoperative DTI and DTT can clearly identify relevant fiber tracts adjacent to BSCMs and to evaluate their roles in surgical planning. We analyze the correlation of DTI and DTT findings with known microsurgical anatomy of the brainstem safe entry zones, and their implications for the selection of surgical approaches. Finally, we attempt to correlate preoperative changes in DTI and DTT and func-tional outcome after resection of these lesions.

methodspatients

All patients who were evaluated for a BSCM at the University of Texas Southwestern Medical Center be-tween July 2007 and July 2012 were screened for inclusion into the study. Eleven patients who underwent resection of a symptomatic BSCM and had a preoperative DTI scan were identified. Retrospective chart review was used to identify patients with a minimum of 6 months of clinical and radiographic follow-up. Surgical intervention was con-sidered in cases of 1) BSCMs with pial presentation, inde-pendent of the number of hemorrhages preoperatively; or

2) history of BSCMs and progressive neurological deficits, with or without radiographic evidence of rehemorrhage. Detailed neurological examination and brain MRI gra-dient echo and gadolinium contrast-enhanced sequences were obtained. All patients were reassessed daily during hospital admission, then routinely during outpatient clinic visits for a minimum of 6 months postoperatively. Retro-spective analysis and data collection were approved by the institutional review board at our institution.

conventional mr and dtiAll patients underwent initial brain MRI with and

without gadolinium contrast as part of their diagnostic workup. Once the patients were selected for surgical inter-vention, additional brain MRI with fast imaging employ-ing steady state acquisition (FIESTA) and DTI protocol were performed (Signa HDxt 3.0T, GE Medical Systems; FOV 16.0 cm, matrix size 320 × 256, slice thickness 2.00 mm). DTI was performed using single-shot spin-echo pla-nar imaging (TR 8.500–12,000 msec, TE 75–85 msec, matrix 128 × 128, slice thickness 2.4 mm, gap 0 mm). Nineteen diffusion directions at b = 1000 sec/mm2 were acquired in addition to b = 0 images. All images were re-viewed by one of the senior authors (A.R.W.). A default fractional anisotropy (FA) threshold of 0.20 and minimum fiber length of 50 mm were used for construction of DTT, as validated by Kunimatsu et al.10,34,38 Patterns of fiber tract alterations (based on morphological appearance on color map FA threshold and reconstructed 3D tractogra-phy) were classified into 4 groups, as described by Lazar et al. (Table 1).33,36 Areas of interest were limited to the neuroanatomical region of the brainstem correlating to the BSCM site and its hemorrhage. The following major fiber tracts were selected for evaluation: 1) corticospinal tract (CST), 2) medial lemniscus (ML) and medial longitudinal fasciculus (MLF), 3) inferior cerebellar peduncle (ICP), 4) middle cerebellar peduncle, and 5) superior cerebellar peduncle (Fig. 1). Scores of 0–4 were applied according to the degree of distortion seen in each of the involved tracts, and the sum of individual fiber tract scores was used for univariate analysis.

Initial postoperative imaging follow-up consisted of noncontrasted head CT scanning, completed up to 24 hours after resection. Postoperative brain MRI with gado-linium contrast was obtained during the hospital stay or

TABLE 1. Classification and grading system for fiber tract changes in DTI*

Description Score

Normal 0Deviated (tract was in abnormal location &/or direction due

to mass effect of the lesion)2

Deformed (partial defect or interruption in parts of tract while the rest of it was identifiable on directional color maps & DTT)

3

Interrupted (tract was discontinuous or defective com-pletely on FA color maps & DTT)

4

* As described in the study by Kovanlikaya et al. Modified from Lazar et al.


DTI in the surgical management of brainstem cavernomas

after discharge, but no longer than 6 months postopera-tively. If no evidence of a recurrent or residual lesion was identified on initial brain MRI, this study was repeated at 6-month intervals for the first 2 postoperative years and at 2-year intervals thereafter.

Anatomical Location and Hemorrhage StatusCavernous malformation location was stratified into 5

categories: mesencephalic, pontomesencephalic, pontine, pontomedullary, and medullary. Sizes were measured on thin-slice T1- or T2-weighted sequences, including the extralesional hematoma cavity (if present). The volume (including BSCM and surrounding hematoma) was calcu-lated assuming an approximate ellipsoid lesional shape: (4/3) × p × r1 × r2 × r3.

Clinical criteria to determine evidence of hemorrhage were 1) acute onset of neurological deficit that persisted for more than 24 hours and 2) deterioration of preexisting symptoms or development of new neurological deficits in patients with previously ruptured BSCMs.

Surgical Principles and TechniqueSeveral surgical approaches were used (Fig. 2). To

minimize local tissue trauma, the hemosiderin-stained surrounding parenchyma is not routinely resected in our practice. Developmental venous anomaly was preserved in all identified cases.1,2,18–21,49 The 2-point method was used when feasible to guide the surgical decision–making process, as described elsewhere.1,2,20 The selected surgical approach was correlated with the results from preopera-tive DTI/DTT and the initial surgical plan adjusted if the findings showed discrepant results or potential risk to the integrity of major long fiber tracts. Somatosensory, mo-tor, and brainstem auditory evoked potential monitoring, as well as electromyographic recordings of specific cra-nial nerves, were performed selectively based on CM lo-cation. A perioperative lumbar drain was used only once for adequate brain relaxation (subtemporal approach for a mesencephalic CM). Histopathological confirmation was obtained for all surgical specimens.

Fig. 1. Cross-sectional brainstem anatomy highlighting the major fiber tracts and their relationship to adjacent neuroanatomical structures. CB = corti-cobulbar tract; CS = CST; FP = frontopontine fibers; ICP = inferior cerebellar peduncle; MCP = medial cerebellar peduncle; ML = medial lemniscus; MLF = medial longitudinal fasciculus; N X = dorsal motor nucleus of vagal nerve; N XII = hypoglossal nucleus; P = pyramid; PTOP = parietotemporooccipito-pontine tract; SCP = superior cerebellar peduncle; VI = cranial nerve VI (nucleus); VII = cranial nerve VII (nucleus). Copyright Suzanne Truex. Published with permission.

J Neurosurg Volume 122 • March 2015 655

b. c. Flores et al.

Postoperative Follow-Up and OutcomeClinical and/or radiographic follow-up was obtained

at the first postoperative clinic evaluation and subsequent clinic visits. Neurological examination findings at the time of discharge and the first and most recent postoperative visits were reviewed. Functional outcomes were measured independently by 2 separate observers at the time of admis-sion, discharge date, and last clinic visit using the modified Rankin Scale (mRS) score.3,30,51,56 Excellent (mRS Score 0–1), good (mRS Score 2–3), or bad (mRS Score 4–6) out-come was determined preoperatively and at the time of the last clinic visit. No patients were lost to follow-up.

Statistical AnalysisPatient data including age, sex, medical and neurosurgi-

cal history, number of hemorrhages before surgical inter-vention, and length of hospital and inpatient rehabilitation stays were collected. Mean, standard deviation, and range were determined for all the demographic data analyzed. Fisher’s exact test, independent samples t-test, or Mann-Whitney U-test was performed as indicated for categori-cal and continuous variables, using available commercial software (version 20.0, IBM SPSS Statistics, IBM Corp.). The Spearman correlation coefficient test (rs) was used for analysis of nonparametric data, with bivariate correlations obtained to quantify the degree of association between in-

dependent continuous variables. A test probability value < 5% was considered significant.

resultsPatient Characteristics and Presentation

Between November 2007 and March 2012, 11 patients with surgically treated BSCMs were identified and ful-filled the enrollment criteria (Tables 2 and 3). The mean age at presentation was 49 years, with a male-to-female ratio of 1.75:1. Cranial nerve deficits were the most com-mon presenting symptoms (81.8%), followed by cerebel-lar signs and gait/balance difficulties (54.5%) and hemi-body numbness/hemianesthesia (27.2%). Headaches were infrequent complaints at initial presentation (9%); only 1 patient (with a left posterior pontine tegmental lesion) pre-sented with hemiparesis.

Lesion Characteristics and LocationThe majority of the lesions were located within the

pons (72.7%). The remaining lesions were mesencephalic (n = 2) and medullary (n = 1). Seven patients (63.6%) un-derwent resection of the CM after a single symptomatic event; 3 patients (27.3%) were submitted to resection after a second symptomatic event. One patient with a CM lo-cated on the posterolateral aspect of the midbrain (epicen-ter on lateral mesencephalic sulcus) had 3 hemorrhages before surgical intervention (including a remote episode 10 years before his presentation to our institution).

The mean size of symptomatic lesions was 1.21 cm, and the mean estimated volume was 1.93 cm3. No statisti-cal difference was seen for mean size and volume between patients with 1 or 2 or more hemorrhagic episodes (p = 0.842 and p = 0.963, respectively). The mean elapsed times from symptom onset to surgical intervention were 103.86 days and 120.50 days for patients with 1 or more than 1 hemorrhagic episode, respectively.

Diffusion Tensor Imaging and Diffusion Tensor Tractography

The mean time from DTI data acquisition to operative intervention was 24.45 days. All patients had changes in DTI and DTT involving at least 1 major fiber tract; 82% showed involvement of 2 or more fiber tracts. Overall, a total of 32 fiber tracts in all 11 patients were found to have some degree of disturbance caused by a BSCM and as-sociated hemorrhage. CST and ML/MLF were the most commonly affected (n = 8 and n = 9, respectively). The majority of changes were displacement (n = 14) or partial interruption and distortion (n = 14). Complete interruption was seen in 4 cases (2 pontine, 1 pontomedullary, and 1 medullary lesion), 3 of those involving the CST (Table 4).

The results obtained from DTI and DTT were cross-referenced with the planned surgical approach. Four cases with radiographic evidence of pial presentation on preoperative MRI underwent surgery through the origi-nally selected surgical approach. This paramount prin-ciple of resection of brainstem lesions is associated with a better postoperative neurological outcome and lower incidence of new neurological deficits, in part because it minimizes normal brainstem tissue retraction, avoids ex-

Fig. 2. Schematic representation of the different neurosurgical ap-proaches used for the resection of symptomatic BSCMs. Copyright Suzanne Truex. Published with permission.



tensive parenchymal surgical corridors, and utilizes the local surface changes caused by the underlying lesion and hemorrhage as a reliable marker for entry zone se-lection.1,2,7,9,21,27,28,45,52,60 The pial presentation on MRI cor-related with DTI/DTT findings (fiber tract deviation or interruption at the selected brainstem entry point) as well as the intraoperative findings during initial microscopic dissection. In 4 patients with no clear evidence of pial presentation on preoperative MRI studies, the decision-making process was influenced by the results of DTI/DTT. A 50-year-old patient presenting after second hem-orrhage from a known right pontine cavernous malforma-tion underwent preoperative MRI, which showed possible pial presentation both on the anterolateral aspect of the pontine tegmentum and on the floor of the fourth ventricle (Fig. 3A). DTI/DTT revealed displaced and distorted ML

and ICP overlying the posterior aspect of that lesion and a preserved but medially deviated right CST anteriorly (Fig. 3B and C). Those findings favored a presigmoid approach over a suboccipital transventricular or transvermian for resection. The patient underwent gross-total resection of her CM with a peritrigeminal area brainstem entry point (Fig. 3D). In a similar situation, a 15-year-old boy with a posteriorly located left pontomedullary junction cavern-ous malformation diagnosed after new onset of left facial weakness and hemifacial numbness (Fig. 4A) underwent a far-lateral transcondylar transpeduncular approach in-stead of a transventricular/posterior approach after DTI/DTT showed the cavernous malformation to displace the ML, ICP, and the facial colliculus posteromedially (Fig. 4B and C). Postoperative MRI showed gross-total resec-tion (Fig. 4D). He had no new postoperative neurological symptoms. Two other patients (a 48-year-old woman with a right rostral anterior pons and a 59-year-old man with a right upper pyramidal/medullary CM) harbored lesions without pial presentation in which the results from preop-erative DTI/DTT were important for selection of brain-stem entry zones.

Surgical ApproachesFour patients (36.3%) underwent a far-lateral approach

and its variations for resection of lesions located at the pons or pontomedullary junction. Three patients (27.4%) had large CMs abutting the floor of the fourth ventricle, with intraventricular pial presentation, and underwent a midline suboccipital craniotomy. Table 5 summarizes the remaining surgical approaches used for the resection of the BSCMs.

OutcomesAll 11 patients had gross-total resection of their BSCMs.

The mean overall hospital length of stay was 13.09 days. Six patients (54.5%) were discharged home after recov-ering from the surgical procedure; 5 patients (45.5%) re-quired admission to an inpatient rehabilitation facility, with a mean length of stay of 28.60 days; all 5 patients

TABLE 2. Brainstem cavernous malformations: general patient demographic data

Case No.

Age (yrs), Sex

No. of Preop Hemorrhages Location

Anteroposterior Size (mm)

Volume (cm3)

LOS (days)

LOS Rehab (days) Surgical Approach

Follow-Up (mos)

Initial mRS Score

Last mRS Score

1 40, F 1 Pons 12 1.04 5 — Far lateral 59.00 1 02 44, F 1 Pons 5 0.07 7 — Transvermian 53.60 2 13 48, F 2 Pons 6 0.20 7 — Transpetrosal presigmoid 45.93 1 14 15, M 1 Pons 12 0.97 6 — Far lateral transcondylar 39.83 1 25 70, M 1 Pons 9 0.62 21 22 Telovelar 35.90 1 26 29, M 2 Pons 19 5.69 8 — Telovelar 32.37 2 17 67, F 1 Midbrain 12 1.04 38 23 Transsylvian 28.17 0 28 62, M 3 Midbrain 10 0.58 17 57 Subtemporal 22.20 3 29 55, M 1 Pons 6 0.53 8 29 Far lateral 20.47 1 210 50, M 2 Pons 29 9.84 6 12 Transpetrosal presigmoid 16.30 2 211 59, M 1 Medulla 13 0.67 21 — Far lateral transcondylar 6.00 2 1

LOS = length of stay; — = not applicable.

TABLE 3. Descriptive analysis of 11 patients who underwent resection of symptomatic BSCMs

Variable Minimum Maximum Mean SD

Age (yrs) 15 70 49.00 16.535Time btwn diagnosis/

presentation and surgery in days

39 224 109.91 56.749

Bleeding episodes until surgery

1 3 1.45 0.688

Anteroposterior size (mm)

5 29 12.09 6.877

Volume (cm3) 0.07 9.84 1.932 3.045Time btwn DTI acqui-

sition and surgery in days

1 69 24.45 21.135

Hospital LOS (days) 5 38 13.09 10.261Rehab LOS (days) 12 57 28.06 17.009Follow-up (mos) 6 58 32.04 16.209Initial mRS score 0 3 1.45 0.820Last mRS score 0 2 1.45 0.688


b. c. Flores et al.

were eventually discharged home. Need for inpatient reha-bilitation admission postoperatively did not predict worse outcome (p = 0.137). As observed in various other stud-ies,1,2,9,16,18,20,24,26,28,60 most patients had transient worsening of preoperative neurological deficits or development of new neurological symptoms; nonetheless, all 11 patients had improvement in neurological function by the first postoperative outpatient visit; the improvement continued over the first 24 months. The mean length of postoperative

follow-up was 32.04 months. There were no deaths during long-term follow-up.

Two patients developed postoperative CSF leak. One of the cases resolved after lumbar drain placement and temporary CSF diversion; the second patient required re-operation for an enlarging pseudomeningocele and wound infection with early meningitis. One patient required tem-porary tracheostomy/gastrostomy tube placement that was discontinued 3 months postoperatively (Table 6).

Five patients (45.5%) had excellent outcomes (mRS

TABLE 4. DTI/DTT changes and attributed DTI score stratified by brainstem location

Case No.

Brainstem Location

DTI ScoreMedulla Oblongata Pons Midbrain

ICP CST ML/MLF ICP SCP MCP CST ML/MLF SCP CST ML

1 — — — 0 0 3 4 2 — — — 92 — — — 0 0 0 0 2 — — — 23 — — — 0 0 0 3 0 — — — 34 — — — 2 2 0 0 2 — — — 65 — — — 0 3 2 0 3 — — — 86 — — — 3 2 2 2 4 — — — 137 — — — — — — — — 3 3 3 98 — — — — — — — — 2 2 2 69 — — — 0 0 3 4 0 — — — 710 — — — 3 2 3 2 3 — — — 1311 3 4 3 — — — — — — — — 10

ICP = inferior cerebellar peduncle; MCP = middle cerebellar peduncle; SCP = superior cerebellar peduncle.

Fig. 3. Case 10. Images obtained in a 50-year-old patient presenting with new right-sided cranial nerve V, VII, and VIII deficits after a second hemorrhage from a known right pontine CM. A: Preoperative MR image showing possible pial presentation both on the anterolateral aspect of the pontine tegmentum and on the floor of the fourth ventricle. b and C: DTI (B) and DTT (C) revealing displaced and distorted ML/MLF and ICP overlying the posterior aspect of that lesion, and a preserved but medially deviated right CST anteriorly. D: The patient underwent a pre-sigmoid approach with a peritrigeminal area brainstem entry point with gross-total resection of his cavernous malformation.

Fig. 4. Case 4. A: MR image obtained in a 15-year-old patient with a posteriorly located left pontomedullary junction cavernous malformation, diagnosed after new onset of left facial weakness and hemifacial numb-ness. B and C: He underwent a far lateral transcondylar transpeduncu-lar approach after DTI (B) and DTT (C) showed the cavernous malforma-tion displacing the ML, ICP, and the facial colliculus posteromedially. D: Postoperative MR image showing gross-total resection.



Score 0–1) at the time of their last outpatient clinic evalu-ation. Six patients (54.5%) had good outcomes (mRS Scores 2–3). No patients had an mRS score of 4 or 5 at last follow-up. Compared with preoperative assessment, 5 pa-tients (45.5%) had improvement of neurological function; 2 patients (18.2%) remained stable, and 4 patients (36.3%) had some degree of neurological decline postoperatively. Except for the number of preoperative bleeding episodes (p = 0.047, 95% CI 0.015–1.413) and initial mRS score (p = 0.021, 95% CI 0.210–2.005), no other independent variable was associated with long-term outcome.

The mean DTI scores for patients with a stable/im-proved or worse mRS score postoperatively were 8.00 and 7.50, respectively (p = 0.724). The results were similar af-ter stratification by brainstem location for pontine lesions (p = 0.687); small sample size prevented statistical analy-sis of mesencephalic lesions. The only patient with a med-ullary BSCM had overall improvement and mRS Score 1 postoperatively, despite significant changes seen on DTI.

DiscussionCavernous malformations are discrete, lobulated,

well-circumscribed lesions consisting of dilated and thin-walled capillaries with simple endothelial lining and vari-ably thin fibrous adventitia. The typical CM has no inter-vening brain parenchyma between the vascular channels.56 Although uncommon, BSCMs are associated with signifi-cant morbidity and frequent development of irreversible neurological deficits caused by hemorrhage into a confined but critical neuroanatomical location. New or worsening neurological symptoms are also associated with resection. Surgical morbidity and mortality have been estimated to be 5%–27.7% and 0%–6.3%, respectively.9 There is an even higher risk of early but transient morbidity (as high as 57%–86%).24 Despite these facts, resection represents the best treatment option for accessible brainstem lesions once they bleed, because it may prevent neurological de-cline due to recurrent hemorrhages.

Natural historyThere is variable risk of hemorrhage reported in the liter-

ature, partially due to preselection of patients in retrospec-

tive studies and variations in the definition of hemorrhagic episodes.18 An overall hemorrhage rate of 2.4%–4.6% per patient-year appears to be an adequate estimate from vari-ous natural history studies,15,26,31,32,35,41,48,49,53 with higher numbers (2.6%–7% per patient-year) obtained from surgi-cal series.1,2,16,18–20,24,25,27,28,49,60 After an initial event, bleed-ing rates are generally higher (5%–34.7%).1,2,16,18,20,24,25,28,49,60

Study PopulationThe study population was slightly older (mean age 49

years) than reported in surgical literature. Most lesions were located in the pons; this neuroanatomical distribu-tion of BSCMs in our study is similar to the one reported by Dukatz et al.16 Headaches were usually not a predomi-nant complaint and cranial nerve deficits were the most prevalent presentation. Postoperatively, clinical symptoms improved or remained stable in 63.7% of the patients. Four patients’ conditions (36.3%) deteriorated or they devel-oped new neurological deficit; overall, these results are similar to those described in the literature.1,16,19,60 Patients who improved or remained stable had more hemorrhages and higher initial mRS scores preoperatively, which did not reflect negatively on final functional outcome. All 11 patients had mRS scores of 2 or lower by their last postop-erative visit, despite a 63% incidence of transient worsen-ing on neurological examination. This deterioration often involved new or worsening cranial nerve deficits and has been noted by other authors.1,18,24,28,49,60 No rehemorrhages were documented over a mean follow-up of 32 months.

dti/dtt dataSince the advent of DTI 2 decades ago, there has been

significant interest in its potential clinical applications.4,5,29,

33,39,40,47,57 The physical principles and rationale of this tech-nique have been extensively debated, and a more profound discussion is beyond the scope of the current study.6,8,12–14,

TABLE 5. Surgical approaches for resection of BSCMs

Surgical Approach No. of Patients

Far lateral 1 Transcondylar 3Suboccipital Transvermian 1 Telovelar 2Transpetrosal presigmoid 2Frontotemporal transsylvian 1Supracerebellar-infratentorial* 1Subtemporal 1Total 12

* Aborted due to inadequate exposure and prominent venous anatomy precluding safe resection.

TABLE 6. Perioperative complications in 11 patients undergoing resection of BSCMs

Complications No. of Patients

New transient CN deficit 6 CN III 1 CN VI 3 CN VII 1 CN VIII 1New permanent CN deficit 3 CN VII 2 CN VIII 1New motor deficit 1New sensory deficit 1Thalamic ischemic infarct 1Need for tracheostomy &/or gastrostomy tube

placement1

CSF leak 2Meningitis 1Ventilator-associated pneumonia 1


b. c. Flores et al.

33,34,37,38,43,50,57,61 Wu et al. showed that DTI-based functional neuronavigation contributed to maximal safe resection of cerebral gliomas and decreased postoperative motor defi-cits while increasing high-quality survival. Lower rates of postoperative deficit and dependency as well as higher median survival rates were reported in 118 patients who underwent surgery with DTI-integrated neuronavigation compared with 120 patients who underwent surgery with standard techniques.61 Others have shown good concor-dance between tractography results and intraoperative di-rect electric stimulations.37

Despite good data on DTI use for supratentorial lesions, there have been only a limited number of studies regarding its application in brainstem pathologies, partially because of known technical difficulties intrinsic to this location. Kovanlikaya et al. investigated the role of DTI and DTT on the CST alterations caused by space-occupying lesions in the brainstem before and after resection in 14 patients.33 All of the patients with normal CST on DTT presented without motor deficit on neurological examination. DTT was shown to have high sensitivity (100%) and negative predictive value (100%), but high false-positive results (positive predictive value of 42.9%) in preoperative evalu-ation, related to lesion artifact. In 2 separate articles,10,11 Chen et al. acknowledged that, compared with conven-tional MRI, DTI and DTT provided superior quantifica-tion and visualization of lesion involvement in eloquent fi-ber tracts of the brainstem. Moreover, DTI and DTT were found to be beneficial for white matter recognition in the neurosurgical planning and postoperative assessment of brainstem lesions.

In our study, 82% of the patients presented with DTI changes involving 2 or more major fiber tracts. Most dis-turbances were displacements or partial interruptions and involved the CST and ML/MLF complex. Sensitivity and negative predictive value were excellent and similar to the ones presented by Kovanlikaya et al.33 Positive pre-dictive values as low as 37.5% were also observed in our study. Preoperative DTI was not shown to correlate with postoperative outcome, as determined by the mRS score. These findings suggest that preoperative DTI/DTT may not be a useful tool for prediction of neurological deficits or long-term outcome after resection of BSCMs. Part of the explanation could reside in the fact that the majority of preoperative symptoms in our population were cranial nerve deficits (81.8%), which are not well delineated by DTI/DTT. Another explanation derives from limitations in the technique itself. The hemosiderin rim is well known to cause local modifications in tissue anisotropy and to falsely disrupt tractography maps.6,13,37 Additionally, echo planar imaging–based DTI is hindered by susceptibil-ity distortions at the proximity of the skull base air-filled spaces, partially secondary to pulsatile cardiac and respi-ratory motion artifacts.

DTI could represent, however, a valuable tool in preop-erative and intraoperative planning for resection of specific subtypes of lesions. Cavernous malformations are thought to displace rather than infiltrate surrounding structures.1 Some neurological deficits observed preoperatively could be attributed to tissue expansion and displacement caused by blood contained in the hemorrhagic cavity; resorption

of blood products correlates positively with neurological improvement and is one of the reasons why some authors recommend resection in a delayed fashion.9,18,21,45,60 This delay allows for partial liquefaction of the hematoma, which facilitates resection and minimizes surgery-related trauma. In cases in which the BSCM has pial or ependy-mal presentation, the lesion itself provides the surgical ac-cess route.52 In our series, 4 patients with preoperative im-aging studies suggesting pial presentation had concordant intraoperative findings; more importantly, DTI and DTT in these cases confirmed displacement of major fiber tracts by the lesion, and no overlying tract was identified on the planned brainstem entry zone. In 2 patients with pontine lesions without pial presentation, DTI information was im-portant for approach selection. On 2 other occasions, deep-seated lesions were removed through safety entry zones selected based on preoperative DTI/DTT data.

To our knowledge, this is the largest study published to date involving brainstem cavernous malformations and DTI. Our results confirm the feasibility of DTI and DTT on preoperative assessment of patients with brainstem space-occupying lesions such as BSCMs. In selected cases, DTI/DTT could provide useful information preoperatively that may influence surgical results in deep-seated brainstem lesions without pial or ependymal presentation. In this subtype of BSCMs, DTI/DTT may assist in the selection of the surgical approach and brainstem entry zone. In the future, intraoperative DTT visualization for neurosurgical planning can be integrated with the neurosurgical micro-scope and real-time querying tractography obtained,17,23,44 maximizing resection while preserving structure and, sub-sequently, function. DTI/DTT could be a potential tool for preoperative counseling of patients with critically located lesions, anticipating neurological deficits that—even tran-siently—can impact negatively on quality of life. The long follow-up demonstrated well the remarkable functional recovery those patients can attain after an initial expect-ed postoperative neurological decline. Dukatz et al., in a study published in 2011, emphasized the importance of the quality-of-life assessment after resection of BSCMs.16 In that study, 87% of the individuals who underwent surgical treatment for BSCMs pursued the same professional activ-ity as before surgery at last follow-up. Of all 71 patients, 58 (82%) felt good or better after surgery and 63 (89%) stated that they had experienced emotional relief postoperatively.

Several factors need to be taken into account when evaluating the results of this study. The small sample size and the retrospective nature of our data collection incur a strong selection bias. The absence of a control group is an important limiting factor for the analysis of our surgi-cal results. We did not routinely obtain postoperative DTI in our patients; different authors have shown improved DTI signal or correction of distortion on major fiber tracts postoperatively, with possible correlation of those findings with final neurological symptoms.10,33,50 Finally, there are limitations of the technique itself. DTI is still highly op-erator dependent and subject to interobserver variability. Its information is limited to long fiber tract alterations, and the technique does not provide information on brain-stem substructures such as nuclei, their interconnecting fibers, and cranial nerve trajectories. Artifacts caused by



the BSCM itself, hemorrhage, and hemosiderin deposits as well as surrounding blood vessel pulsations can limit adequate imaging acquisition. Some of these obstacles can be overcome with ongoing improvements in the technique, such as high angular resolution diffusion imaging and q-ball reconstruction of high angular resolution diffusion imaging information.6,37 Tract-based spatial statistics is a promising approach that can improve the sensitivity, ob-jectivity, and interpretability of the analysis of multisubject diffusion imaging studies.54,55 However, its use for large space-occupying lesions on the brainstem is still limited.

ConclusionsDiffusion tensor imaging and DTT are potentially use-

ful preoperative tools that could be considered when eval-uating patients with deep-seated, symptomatic BSCMs. In patients in whom no clear radiographic evidence of pial or ependymal presentation is seen, DTI and DTT could influ-ence the selection of surgical approach or brainstem entry zone. In our series, preoperative DTI/DTT changes did not appear to correlate with patients’ functional postoperative outcome on long-term follow-up.

AcknowledgmentsWe thank Suzanne “Jorlam” Truex, gifted Medical Illustrator,

for her unique ability of transforming chaotic thoughts into beauti-ful harmonic pictures on a piece of paper.

References 1. Abla AA, Lekovic GP, Turner JD, de Oliveira JG, Porter

R, Spetzler RF: Advances in the treatment and outcome of brainstem cavernous malformation surgery: a single-center case series of 300 surgically treated patients. Neurosurgery 68:403–415, 2011

2. Abla AA, Turner JD, Mitha AP, Lekovic G, Spetzler RF: Surgical approaches to brainstem cavernous malformations. Neurosurg Focus 29(3):E8, 2010

3. Banks JL, Marotta CA: Outcomes validity and reliability of the modified Rankin scale: implications for stroke clinical trials: a literature review and synthesis. Stroke 38:1091–1096, 2007

4. Basser PJ, Mattiello J, LeBihan D: MR diffusion tensor spec-troscopy and imaging. Biophys J 66:259–267, 1994

5. Beaulieu C, Allen PS: Determinants of anisotropic water dif-fusion in nerves. Magn Reson Med 31:394–400, 1994

6. Berman J: Diffusion MR tractography as a tool for surgical planning. Magn Reson Imaging Clin N Am 17:205–214, 2009

7. Brown AP, Thompson BG, Spetzler RF: The two-point meth-od: evaluating brain stem lesions. BNI Q 12:20–24, 1996

8. Cao Z, Lv J, Wei X, Quan W: Appliance of preoperative dif-fusion tensor imaging and fiber tractography in patients with brainstem lesions. Neurol India 58:886–890, 2010

9. Chen L, Zhao Y, Zhou L, Zhu W, Pan Z, Mao Y: Surgical strategies in treating brainstem cavernous malformations. Neurosurgery 68:609–621, 2011

10. Chen X, Weigel D, Ganslandt O, Buchfelder M, Nimsky C: Diffusion tensor imaging and white matter tractography in patients with brainstem lesions. Acta Neurochir (Wien) 149:1117–1131, 2007

11. Chen X, Weigel D, Ganslandt O, Fahlbusch R, Buchfelder M, Nimsky C: Diffusion tensor-based fiber tracking and in-traoperative neuronavigation for the resection of a brainstem cavernous angioma. Surg Neurol 68:285–291, 2007

12. Clark CA, Barrick TR, Murphy MM, Bell BA: White matter fiber tracking in patients with space-occupying lesions of the brain: a new technique for neurosurgical planning? Neuro-image 20:1601–1608, 2003

13. DaSilva AFM, Tuch DS, Wiegell MR, Hadjikhani N: A prim-er on diffusion tensor imaging of anatomical substructures. Neurosurg Focus 15(1):E4, 2003

14. de Souza JM, Gasparetto E: Better definition for brainstem tumors: advances in diffusion tensor imaging. World Neuro-surg 74:5–6, 2010

15. Detwiler PW, Porter RW, Zabramski JM, Spetzler RF: De novo formation of a central nervous system cavernous malformation: implications for predicting risk of hemor-rhage. Case report and review of the literature. J Neurosurg 87:629–632, 1997

16. Dukatz T, Sarnthein J, Sitter H, Bozinov O, Benes L, Sure U, et al: Quality of life after brainstem cavernoma surgery in 71 patients. Neurosurgery 69:689–695, 2011

17. Elhawary H, Liu H, Patel P, Norton I, Rigolo L, Papademetris X, et al: Intraoperative real-time querying of white matter tracts during frameless stereotactic neuronavigation. Neuro-surgery 68:506–516, 2011

18. Ferroli P, Sinisi M, Franzini A, Giombini S, Solero CL, Broggi G: Brainstem cavernomas: long-term results of mi-crosurgical resection in 52 patients. Neurosurgery 56:1203–1214, 2005

19. Fritschi JA, Reulen HJ, Spetzler RF, Zabramski JM: Cavern-ous malformations of the brain stem. A review of 139 cases. Acta Neurochir (Wien) 130:35–46, 1994

20. Garrett M, Spetzler RF: Surgical treatment of brainstem cav-ernous malformations. Surg Neurol 72 Suppl 2:S3–S10, 2009

21. Giliberto G, Lanzino DJ, Diehn FE, Factor D, Flemming KD, Lanzino G: Brainstem cavernous malformations: anatomi-cal, clinical, and surgical considerations. Neurosurg Focus 29(3):E9, 2010

22. Giussani C, Poliakov A, Ferri RT, Plawner LL, Browd SR, Shaw DWW, et al: DTI fiber tracking to differentiate demy-elinating diseases from diffuse brain stem glioma. Neuroim-age 52:217–223, 2010

23. Golby AJ, Kindlmann G, Norton I, Yarmarkovich A, Pieper S, Kikinis R: Interactive diffusion tensor tractography visu-alization for neurosurgical planning. Neurosurgery 68:496–505, 2011

24. Gross BA, Batjer HH, Awad IA, Bendok BR: Brainstem cav-ernous malformations. Neurosurgery 64:E805–E818, 2009

25. Gross BA, Batjer HH, Awad IA, Bendok BR, Du R: Brain-stem cavernous malformations: 1390 surgical cases from the literature. World Neurosurg 80:89–93, 2013

26. Gross BA, Lin N, Du R, Day AL: The natural history of intracranial cavernous malformations. Neurosurg Focus 30(6):E24, 2011

27. Hauck EF, Barnett SL, White JA, Samson D: The presigmoid approach to anterolateral pontine cavernomas. Clinical ar-ticle. J Neurosurg 113:701–708, 2010

28. Hauck EF, Barnett SL, White JA, Samson D: Symptomatic brainstem cavernomas. Neurosurgery 64:61–71, 2009

29. Henkelman RM, Stanisz GJ, Kim JK, Bronskill MJ: An-isotropy of NMR properties of tissues. Magn Reson Med 32:592–601, 1994

30. Huybrechts KF, Caro JJ: The Barthel Index and modified Rankin Scale as prognostic tools for long-term outcomes after stroke: a qualitative review of the literature. Curr Med Res Opin 23:1627–1636, 2007

31. Kim DS, Park YG, Choi JU, Chung SS, Lee KC: An analy-sis of the natural history of cavernous malformations. Surg Neurol 48:9–18, 1997

32. Kondziolka D, Lunsford LD, Kestle JR: The natural history of cerebral cavernous malformations. J Neurosurg 83:820–824, 1995


b. c. Flores et al.

33. Kovanlikaya I, Firat Z, Kovanlikaya A, Uluğ AM, Cihangiro-glu MM, John M, et al: Assessment of the corticospinal tract alterations before and after resection of brainstem lesions using Diffusion Tensor Imaging (DTI) and tractography at 3T. Eur J Radiol 77:383–391, 2011

34. Kunimatsu A, Aoki S, Masutani Y, Abe O, Hayashi N, Mori H, et al: The optimal trackability threshold of fractional an-isotropy for diffusion tensor tractography of the corticospinal tract. Magn Reson Med Sci 3:11–17, 2004

35. Kupersmith MJ, Kalish H, Epstein F, Yu G, Berenstein A, Woo H, et al: Natural history of brainstem cavernous malfor-mations. Neurosurgery 48:47–54, 2001

36. Lazar M, Alexander AL, Thottakara PJ, Badie B, Field AS: White matter reorganization after surgical resection of brain tumors and vascular malformations. AJNR Am J Neurora-diol 27:1258–1271, 2006

37. Leclercq D, Delmaire C, de Champfleur NM, Chiras J, Lehéricy S: Diffusion tractography: methods, validation and applications in patients with neurosurgical lesions. Neuro-surg Clin N Am 22:253–268, ix, 2011

38. Lui YW, Law M, Chacko-Mathew J, Babb JS, Tuvia K, Al-len JC, et al: Brainstem corticospinal tract diffusion tensor imaging in patients with primary posterior fossa neoplasms stratified by tumor type: a study of association with motor weakness and outcome. Neurosurgery 61:1199–1208, 2007

39. Mori S, Crain BJ, Chacko VP, van Zijl PC: Three-dimension-al tracking of axonal projections in the brain by magnetic resonance imaging. Ann Neurol 45:265–269, 1999

40. Mori S, van Zijl PC: Diffusion weighting by the trace of the diffusion tensor within a single scan. Magn Reson Med 33:41–52, 1995

41. Moriarity JL, Wetzel M, Clatterbuck RE, Javedan S, Shep-pard JM, Hoenig-Rigamonti K, et al: The natural history of cavernous malformations: a prospective study of 68 patients. Neurosurgery 44:1166–1173, 1999

42. Newcombe VFJ, Williams GB, Scoffings D, Cross J, Car-penter TA, Pickard JD, et al: Aetiological differences in neuroanatomy of the vegetative state: insights from diffusion tensor imaging and functional implications. J Neurol Neuro-surg Psychiatry 81:552–561, 2010

43. Niizuma K, Fujimura M, Kumabe T, Higano S, Tominaga T: Surgical treatment of paraventricular cavernous angioma: fibre tracking for visualizing the corticospinal tract and de-termining surgical approach. J Clin Neurosci 13:1028–1032, 2006

44. Nimsky C: Intraoperative acquisition of fMRI and DTI. Neu-rosurg Clin N Am 22:269–277, ix, 2011

45. Ohue S, Fukushima T, Kumon Y, Ohnishi T, Friedman AH: Surgical management of brainstem cavernomas: selection of approaches and microsurgical techniques. Neurosurg Rev 33:315–324, 2010

46. Otten P, Pizzolato GP, Rilliet B, Berney J: [131 cases of cavernous angioma (cavernomas) of the CNS, discovered by retrospective analysis of 24,535 autopsies.] Neurochir 35:82–83, 128–131, 1989 (Fr)

47. Pierpaoli C, Jezzard P, Basser PJ, Barnett A, Di Chiro G: Diffusion tensor MR imaging of the human brain. Radiology 201:637–648, 1996

48. Porter PJ, Willinsky RA, Harper W, Wallace MC: Cerebral cavernous malformations: natural history and prognosis after clinical deterioration with or without hemorrhage. J Neuro-surg 87:190–197, 1997

49. Porter RW, Detwiler PW, Spetzler RF, Lawton MT, Baskin JJ, Derksen PT, et al: Cavernous malformations of the brain-stem: experience with 100 patients. J Neurosurg 90:50–58, 1999

50. Prabhu SP, Ng S, Vajapeyam S, Kieran MW, Pollack IF, Geyer R, et al: DTI assessment of the brainstem white mat-ter tracts in pediatric BSG before and after therapy: a report from the Pediatric Brain Tumor Consortium. Childs Nerv Syst 27:11–18, 2011

51. Rankin J: Cerebral vascular accidents in patients over the age of 60. II. Prognosis. Scott Med J 2:200–215, 1957

52. Recalde RJ, Figueiredo EG, de Oliveira E: Microsurgical anatomy of the safe entry zones on the anterolateral brain-stem related to surgical approaches to cavernous malforma-tions. Neurosurgery 62 (3 Suppl 1):9–17, 2008

53. Robinson JR, Awad IA, Little JR: Natural history of the cav-ernous angioma. J Neurosurg 75:709–714, 1991

54. Smith SM, Jenkinson M, Johansen-Berg H, Rueckert D, Nichols TE, Mackay CE, et al: Tract-based spatial statistics: voxelwise analysis of multi-subject diffusion data. Neuroim-age 31:1487–1505, 2006

55. Smith SM, Jenkinson M, Woolrich MW, Beckmann CF, Beh-rens TEJ, Johansen-Berg H, et al: Advances in functional and structural MR image analysis and implementation as FSL. Neuroimage 23 (Suppl 1):S208–S219, 2004

56. Steinberg GK, Marks MP: Lesions mimicking cavernous malformations, in Awad IA, Barrow DL (eds): Cavernous Malformations. Park Ridge, IL: American Association of Neurological Surgeons, 1993, pp 151–162

57. Stieltjes B, Kaufmann WE, van Zijl PC, Fredericksen K, Pearlson GD, Solaiyappan M, et al: Diffusion tensor imaging and axonal tracking in the human brainstem. Neuroimage 14:723–735, 2001

58. Unger MM, Belke M, Menzler K, Heverhagen JT, Keil B, Stiasny-Kolster K, et al: Diffusion tensor imaging in idio-pathic REM sleep behavior disorder reveals microstructural changes in the brainstem, substantia nigra, olfactory region, and other brain regions. Sleep 33:767–773, 2010

59. Upadhyay J, Knudsen J, Anderson J, Becerra L, Borsook D: Noninvasive mapping of human trigeminal brainstem path-ways. Magn Reson Med 60:1037–1046, 2008

60. Wang CC, Liu A, Zhang JT, Sun B, Zhao YL: Surgical man-agement of brain-stem cavernous malformations: report of 137 cases. Surg Neurol 59:444–454, 2003

61. Wu JS, Zhou LF, Tang WJ, Mao Y, Hu J, Song YY, et al: Clinical evaluation and follow-up outcome of diffusion ten-sor imaging-based functional neuronavigation: a prospective, controlled study in patients with gliomas involving pyramidal tracts. Neurosurgery 61:935–949, 2007

Author ContributionsConception and design: Flores, Barnett. Acquisition of data: Flores, Whittemore, Barnett. Analysis and interpretation of data: all authors. Drafting the article: Flores, Samson, Barnett. Critically revising the article: all authors. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Flores. Statistical analysis: Flores. Administrative/technical/material support: Whittemore. Study supervision: Flores, Samson, Barnett.

CorrespondenceBruno C. Flores, Department of Neurological Surgery, University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd., Mail Code 8855, Dallas, TX 75390. email: [email protected].


Microsurgical anatomy of safe entry zones to the brainstem · 2017-10-17 · distance of 4.64 mm...

Documents

Transcript of Microsurgical anatomy of safe entry zones to the brainstem · 2017-10-17 · distance of 4.64 mm...