Midline Shift Due to Acute Intracranial Hemorrhage in the Surgical ICU: Bedside or OR for Decompression?

Patients on anticoagulants are at greater risk of acute intracranial bleeding, a neurosurgical emergency that benefits from immediate attention. While relatively uncommon, a patient who presents with an acute or evolving intracranial hemorrhage is traditionally transferred to the closest OR for immediate evacuation, regardless of whether the patient is at home, on the medical ward, or in the ICU. In this article, we highlight a recent debate on whether a 35-year-old female with peripartum cardiomyopathy, anticoagulated for right and left ventricular assist devices (VAD) who presented with acute intracranial bleeding while in the ICU, should have been operated on at the bedside or transferred to an OR.

Briefly, nine days following VAD placement, the patient developed progressive neurological deterioration and was diagnosed with a large left-sided acute subdural hematoma with midline shift on computer tomography (CT). Anticoagulation was immediately reversed, and the patient was transferred to the OR for emergent craniectomy and evacuation. At the time of the operation while in the OR, surgical hemostasis was observed following the evacuation. An immediate postoperative CT showed worsening cerebral edema, re-accumulation of the subdural hematoma, and a worsening midline shift. The patient was transferred to the ICU where she began to show signs of increased intracranial pressure, hypertension, bradycardia, and dilated and fixed pupils. The neurosurgical team informed her family, who wanted to pursue another attempt at surgical evacuation, and at which point the care team needed to decide whether to return to the OR for immediate evacuation or perform the procedure at the patient’s bedside.

Below, we present the potential risks and benefits of performing neurosurgical interventions at a patient’s bedside versus in the OR. It will always be important to determine the patient’s evolving medical condition and the urgent need for surgical intervention, the complexity of transferring a patient from the ICU to the OR, hospital layout, the willingness and comfort level of the neurosurgical team to operate outside the OR, the availability of trained support staff, and resource availability, including sterile instruments, appropriate lighting, and the ability to correctly position the patient. Ultimately, deciding where to perform emergent neurosurgery is a judgment call that ideally can be made by the team caring for the patient with input from neurosurgeons, anesthesiologists, and intensivists.

We found ourselves needing to match the patient’s immediate needs with our abilities as clinicians and available resources in that moment. At that time, the clinical care team agreed immediate neurosurgical intervention was necessary to relieve an expanding subdural hematoma in a patient with bilaterally dilated and fixed pupils who was completely reliant on full life support, including mechanical ventilation and right and left heart mechanical circulatory support in the form of a permanent implanted left VAD and a temporary percutaneous right VAD, albeit improving coagulopathy, whose family wanted everything done to save her life. While our abilities to intervene in the ICU setting were unknown, and resource availability severely limited (i.e., surgical instruments, adequate lighting, ability to position the patient, access to support staff), we also knew from experience that transferring a patient on full life support from the ICU down a floor and onto an OR table could have taken as long as an hour.

Practice patterns in medicine are often defined by “how we’ve always done it,” and while there may be many advantages of transferring this patient to the OR, we may not always have that luxury when prioritizing preserving brain function. We suggest considering making medical decisions not based on rote tradition but rather by assessing what is possible to meet a patient’s need at a particular time. Yes, the lighting is better in the OR. Yes, the OR is more comfortable and affords the surgical team better access. And yes, all the “stuff” is available in the OR. This is only one side of the equation. On the other side is this patient’s need for immediate life-saving evacuation, and by transferring this patient to the OR, we surmise that that need would not have been met.

Expanding intracranial hemorrhage is a neurosurgical emergency that often requires immediate surgical decompression. While timing is critical, appropriate instrumentation, lighting, and magnification are required to ensure a favorable patient outcome. Historically, the latter may have been compromised to achieve a rapid decompression, but in today’s modern hospitals, such patients should undergo surgical treatment in the OR and not at the bedside in the ICU. Timing of surgery should be quick, but how fast is not defined. In our estimation, the urgency of this case lies somewhere in between a bedside tracheostomy for a compromised airway and 30 minutes.

While the care team agreed that emergent evacuation was of paramount importance, remaining cognizant of perioperative sequalae is foundational to the success of surgery. These include cerebral swelling requiring resection, ongoing cerebral hemorrhage, or injury to a major sinus, all of which are ideally dealt with in an OR setting where access to bipolar cautery, a surgical microscope, gel foam, flow-seal, and other hemostatic agents is immediately available.

From the anesthetic perspective, performing emergent bedside craniotomy in this patient would have been fraught with challenges. The basic ergonomics that we are used to in the OR would have been limited by biventricular pumps, rapid-infusing warmers needed for ongoing cerebral hemorrhage, surgical equipment, a transesophageal echocardiography machine to guide her biventricular failure, and the additional personnel needed for care beyond what is typical (i.e., perfusion/ECMO specialists, LVAD coordinators, blood bank delivery staff). The lack of real-time ETCO₂ monitoring and rapid arterial blood gas analysis in the ICU would have made management of elevated ICP and rapid blood loss difficult. As Dennis and Gunter state in their critical review, the risk of massive blood loss should be considered a contraindication to bedside surgery (J Am Coll Surg 2013;216:858-65).

Finally, as patient complexity increases, the need for clear communication is of paramount importance. In a hectic environment, communication and care team focus break down when efforts are expended to quiet the room, limit foot traffic, and find medications and instruments in unfamiliar places. It is likely for these reasons that only a limited array of controlled bedside procedures (e.g., percutaneous tracheostomy and damage control laparotomy) have been described in the literature backed by robust institutional protocols (J Am Coll Surg 2013;216:858-65).

The evacuation was performed at the patient’s ICU bedside. We arranged for sterile instruments and better lighting to be delivered from the OR. Anesthesiologists and intensivists worked to manage the patient’s complex hemodynamic needs and reverse non-surgical coagulopathy, while the surgical team intervened at the bedside to evacuate the hematoma. Over the course of the next few days, some neurologic function returned, and the patient was able to follow commands on her left side, while withdrawing from pain on her right. Despite clear progress and the chance of continued meaningful neurological recovery, the patient’s family decided to withdraw care.

The clinical care team agreed that coordinating the care of this patient between the OR and ICU was extremely complex. Her hemodynamic needs, coagulopathy, and case urgency, not to mention having to consider hospital layout, pushed our team to the very edge of our practice abilities. As patients continue to present at the edge of our practice abilities, and as our life-sustaining technologies and innovations continue to expand, more of us will undoubtedly be faced with similar decisions. We recommend open lines of communication between disciplines as well as a pre-mortem practice to mitigate poor outcomes (asamonitor.pub/3sgneGs).