Retraction Without Retractors

Brain retraction is bad. It can raise brain tissue pressure, reduce cerebral perfusion locally, hide critical anatomy, and injure neurovascular structures. However, subarachnoid corridors of the brain are often too narrow to be navigated without some retraction. Therefore, it needs to be applied discriminately and with finesse.

The sucker and suction hand have an underappreciated role as a roving retractor. While drying the surgical field, the sucker also applies countertraction at the point of dissection with its tip and gentle pressure to the brain with its shaft. The bullet-tipped No. 7 or No. 5 microsuction is smooth and atraumatic. Suction strength is regulated by rolling the thumb forward to cover the keyhole fenestration at the thumb grip for more suction, or backward to uncover the fenestration for less suction. The thumb rests in a middle position that partially covers the hole, and a constant whistling noise should be heard at all times. When the whistling disappears, the thumb may be covering the fenestration and suction may draw in adjacent structures. The sucker is malleable; the shaft is straightened to follow the dissection plane and curved gently to the hand. Connecting the sucker to soft silicone tubing keeps it mobile in the hand, whereas stiff plastic tubing creates resistance.

A properly adjusted sucker naturally complements the dissecting instrument. These instruments lie directly opposite one another at the depth of the field; the sucker provides microretraction for each maneuver of the dissecting instrument, and, unlike a fixed retractor blade, the sucker adjusts constantly to the dissection. Lateral pressure with the sucker provides countertraction when cutting tissue, and retracts tissues to facilitate visualizing the dissection plane. The sucker can cross to the side of the dissecting instrument to apply contralateral pressure. A dynamic suction hand substantially reduces the need for a fixed retractor.

In addition to microretraction with the sucker tip, the sucker shaft can function as a slim retractor blade. Instead of positioning the shaft in the dissection corridor, laying the shaft against the brain gently retracts it and opens the corridor like a funnel. The position of the tip is not affected by this lateral hand movement. Like sucker tip retraction, shaft retraction is also dynamic and adapts to the changing needs of the dissection. Dissecting instruments in the dominant hand, like the microscissors or bipolar forceps, can also function as retractors. The shafts of these instruments can retract their side of the surgical corridor by gently lying against brain, opening the other side of the funnel. Some dissection maneuvers do not allow the microscissors or bipolar forceps to double as retractors, but most maneuvers allow the shaft to pivot around the instrument’s tips and generate some retraction pressure.

Retraction with Retractors

Fixed brain retractors are used very sparingly. A basic Greenberg retractor system has two C-clamps that attach to the Mayfield head holder with their posts pointing toward the vertex and the seated neurosurgeon, and with the C-clamps fixed as close to the surgical field as possible. Clamp posts in this position eliminate the need for extender bars that clutter the working area. The Greenberg retractor is mounted on the posts with the flexible arm arcing up from beneath the surgical field in a gentle curve. Greenberg arms that arc down from above the field often interfere with the hands and can be bumped. Retractor blades that are rounded across their width have a more gentle pressure profile against the brain. The blade length from tip to shoulder is minimized to lower clearance above the brain. The brain is irrigated and covered with Telfa strips to keep the blades from directly touching brain.

Retractors should “hold” brain tissue that has already been thoroughly dissected. Extensive preliminary dissection minimizes any “pull” on brain tissue. Maneuvers that slacken the brain also minimize retraction pressure, like evacuating cisternal cerebrospinal fluid (CSF), fenestrating the lamina terminalis, opening the membrane of Liliequist to communicate with posterior fossa cisterns, and lowering external ventricular drains. Lumbar drains are not used during aneurysm surgery because other points of access to CSF are readily available. Mannitol (1 g/kg) is routinely given to dehydrate brain tissue, and Decadron (10 mg) is given to minimize edema from retraction.

Mobilizing Brain

Retractors move brain, but brain prefers not to be moved. Therefore, the amount of retraction is minimized by skull-base approaches that remove bone along the skull base instead. Drilling the sphenoid wing with the pterional approach or the occipital condyle with the far lateral approach widens the surgical corridor under the brain and reduces retraction. Gravity also minimizes retraction. Patient and head position with some approaches will eliminate the need for retractors, like the anterior interhemispheric approach performed with the patient’s head turned laterally 90 degrees and gravity retracting the dependent hemisphere. Similarly, gravity pulls down on the cerebellum during a supracerebellar-infratentorial approach performed with the patient in the sitting position, opening the plane to the pineal region, ambient cistern, and midbrain. Even with more basic approaches like the pterional approach, head extension allows gravity to open the plane between the anterior skull base and inferior frontal lobe, and head rotation vertically aligns the sylvian fissure to allow gravity to pull the frontal and temporal lobes to opposite sides of the fissure.

An escape hatch must be prepared during the craniotomy for brain that will be mobilized later. For example, retraction of the temporal lobe during the transsylvian-pretemporal approach to the basilar bifurcation requires drilling the temporal squamosal bone inferiorly until it is flush with the middle fossa floor, and posteriorly beyond the zygomatic root. Without this egress, retraction would compress temporal lobe against a ledge of bone. Mobilized brain needs complete freedom from arachnoid adhesions that might resist retraction. For example, arachnoid of the sylvian cistern couples the frontal and temporal lobes and resists frontal retraction; arachnoid of the chiasmatic cistern tethers the frontal lobe and optic nerve and resists frontal retraction; and arachnoid of the crural cistern couples the deep frontal and temporal lobes and resists temporal lobe retraction. Subarachnoid dissection removes this resistance before placing a retractor.

Small arteries can also resist retraction. The anterior temporal artery (ATA) can adhere to the temporal lobe; the recurrent artery of Heubner can adhere to the inferior frontal lobe; and the posterior inferior cerebellar artery (PICA) can adhere to the cerebellar tonsil. Failure to release these adhesions can injure or avulse the artery during retraction. Arteries should not be placed behind a retractor blade because they can be occluded by retraction pressure; they should remain in full view and the dissection should progress around them. Retraction can injure bridging veins, particularly those at the temporal pole, tentorium, and inter-hemispheric fissure. Bridging veins are preserved whenever possible, but aneurysm exposure can sometimes require their sacrifice. Some veins are sacred because they have scant collateral connections and their sacrifice can cause venous infarctions, including veins along the middle third of the superior sagittal sinus (SSS) and the vein of Labbé. Other veins can be taken because of their extensive collateral connections, including the temporal polar vein bridging to sphenoparietal or cavernous sinus, and superior cerebellar and vermian veins bridging to tentorial sinuses. Failure to sacrifice a bridging vein can result in its avulsion with retraction, which can cause brisk bleeding from a venous sinus and be difficult to control. When a vein must be divided, it should be interrupted only at one point to preserve its retrograde connections to collateral veins. Arachnoid granulations also resist retraction. Granulations along the dura of the middle cranial fossa floor and along the SSS can be avulsed with retraction of the temporal pole and medial frontal lobe, respectively. It is easier to release these adhesions before retracting than to chase venous bleeding after retracting.

Most importantly, retraction can avulse an aneurysm’s dome. Aneurysms with intraparenchymal hemorrhage often adhere to that portion of brain. Other aneurysms have notorious points of attachment: a superiorly projecting ophthalmic artery aneurysm adheres to the frontal lobe; an inferiorly projecting anterior communicating artery (ACoA) aneurysm adheres to the optic nerve or chiasm; and a laterally projecting posterior communicating artery (PCoA) aneurysm adheres to the temporal lobe. Retraction early in the dissection of these aneurysms can precipitate intraoperative rupture before establishing proximal control or identifying the aneurysm. These specific retraction moves are avoided with their respective aneurysms. In general, the safest retraction with a ruptured aneurysm is a retraction that is avoided completely.