History

An 8-year-old 19-kg (42-lb) spayed female English Springer Spaniel was examined by the neurology service at the Cummings School of Veterinary Medicine at Tufts University because of behavioral changes, weakness of the left forelimb, circling to the right, and 2 seizures during the previous month. The dog had been treated with phenobarbital (64.8 mg, PO, q 12 h).

Neurologic examination revealed intermittent circling to the right, depression of the postural reactions on the left side, and an increase in the patellar reflex on the left side. Findings were indicative of a lesion localized in the right cerebral cortex. A CBC and serum biochemistry profile were performed; abnormalities included neutrophilia (13,572 neutrophils/μL; reference range, 2,800 to 11,500 neutrophils/μL); lymphopenia (590 lymphocytes/μL; reference range, 1,000 to 4,800 lymphocytes/μL); hypochloremia (104 mEq/L; reference range, 106 to 126 mEq/L); high alkaline phosphatase (1,097 U/L; reference range, 12 to 121 U/L), alanine aminotransferase (166 U/L; reference range, 18 to 86 U/L), and γ-glutamyltransferase (36 U/L; reference range, 2 to 10 U/L) activities; and low creatinine concentration (0.5 mg/dL; reference range, 0.6 to 2.0 mg/dL). Thoracic radiography revealed mild right-sided cardiomegaly and a mild diffuse pulmonary interstitial pattern.

Magnetic resonance imaging revealed a space-occupying lesion in the right temporal lobe. The lesion was sessile, well demarcated, and homogeneously contrast enhancing with a broad-based attachment to the skull. It measured approximately 20 × 16 × 12 mm, was associated with a substantial amount of peritumoral edema in the surrounding white matter, and had caused a substantial shift of the midline to the left side. The location and appearance of the mass, combined with detection of a dural tail sign, indicated that it was most likely a meningioma. The dog was treated with mannitol (0.9 mg/kg [0.41 mg/lb], IV), and administration of prednisone (1 mg/kg [0.45 mg/lb], PO, q 12 h) was begun to alleviate the edema associated with the mass. Craniectomy with tumor resection was scheduled for the following week.

At the time of the proposed tumor resection, the dog was assigned an American Society of Anesthesiologists status of III on the basis of the underlying disease. The dog was premedicated with hydromorphone (0.1 mg/kg [0.045 mg/lb], IV) and midazolam (0.1 mg/kg, IV), and 100% oxygen was delivered via a face mask. Five minutes after premedications were administered, anesthesia was induced with propofol (2 mg/kg [0.9 mg/lb], IV, to effect). A 10-mm, wirereinforced, sterile, cuffed endotracheal tube was inserted in the trachea under direct laryngoscopic visualization, and the tip of the endotracheal tube was advanced to the level of the thoracic inlet. Sevoflurane (initial vaporizer setting, 2%) delivered in oxygen (initial flow, 1 L/min) via a circle anesthetic circuit was used to maintain anesthesia during the surgical preparation and instrumentation period. During surgical preparation, the patient was hand-ventilated to maintain end-tidal partial pressure of CO₂ (PETCO₂), measured by means of a side-stream capnograph,^a between 35 and 40 mm Hg. Care was taken to avoid exceeding a peak inspiratory pressure of 20 cm of water. A 20-gauge catheter was placed percutaneously into the left dorsal pedal artery and connected to a disposable tranducer^b to allow for direct measurement of blood pressure. An 18-gauge catheter was also placed percutaneously into the right saphenous vein. The patient was connected to a multi-purpose monitor^c consisting of an electrocardiograph, pulse oximeter, capnograph, oscillometric blood pressure monitor, and direct blood pressure monitor. Rectal temperature was maintained between 37.4° and 37.7°C (99.3° and 99.8°F) throughout the procedure with a convective air warming system^d and a circulating hot water blanket.^e Lactated Ringer's solution was administered at a rate of 10 mL/kg/h (4.5 mL/lb/h), IV.

In the operating room, the dog was positioned in sternal recumbency, with the nose placed in a specially designed positioner that allowed access to the skull while ensuring head stability throughout the procedure. The patient was mechanically ventilated at a rate of 12 breaths/min, and PETCO₂ was maintained between 38 and 42 mm Hg. Fifteen minutes before the onset of surgery, continuous rate infusions of fentanyl (0.075 μg/kg/min [0.034 μg/lb/min]) and propofol (0.1 mg/kg/min) were begun, and the delivered concentration of sevoflurane was maintained between 0% and 1% for the remainder of the procedure. The patient was also given methylprednisolone sodium succinate (26 mg/kg [11.8 mg/lb], IV) over 10 minutes. Heart rate remained between 65 and 108 beats/min. Systolic, diastolic, and mean blood pressures, measured directly, remained between 100 and 115 mm Hg, 60 and 62 mm Hg, and 70 and 72 mm Hg, respectively.

One hour after the start of surgery (2 hours and 35 minutes after the start of anesthesia), the mass was removed. Shortly after this, the surgeon reported that normal brain tissue had started to herniate through the skull incision. The patient's systolic blood pressure increased to 168 mm Hg, and the heart rate decreased to 48 beats/min.

Question

What is the term used to refer to the physiologic response that the dog was having? What is an appropriate emergency treatment plan for this patient?

Answer

The increase in systolic blood pressure and decrease in heart rate were consistent with development of the Cushing reflex secondary to a dangerous increase in intracranial pressure. Anesthetic management consisted of increasing the ventilation rate to reduce PETCO₂ to 34 to 35 mm Hg, with the aim of decreasing arterial partial pressure of CO₂ (PaCO₂) to 30 mm Hg. An arterial blood sample was obtained and submitted for blood gas analysis to ensure adequate arterial oxygen content and to determine the gradient between PETCO₂ and PaCO₂. At this time, PETCO₂ was 34 mm Hg, PaCO₂ was 30 mm Hg, arterial partial pressure of O₂ (PaO₂) was 498.3 mm Hg, bicarbonate concentration was 18.7 mmol/L, base excess was −4.0 mmol/L, and arterial oxygen saturation was 99.9%. The patient's jugular veins were checked to ensure that they had not become occluded because of positioning, and mannitol (1 g/kg [0.45 g/lb), IV) was administered over 20 minutes while the surgeon applied cold sterile saline (0.9% NaCl) solution to the area of the brain that was swelling. Administration of sevoflurane was discontinued, and anesthesia was maintained with IV administration of fentanyl and propofol.

Within 15 minutes, the dog's systolic blood pressure had begun to decrease while the dog's heart rate had begun to increase. In addition, the surgeon reported that swelling of the brain was subsiding. Ventilator settings were readjusted to achieve a PETCO₂ of 38 to 42 mm Hg, and sevoflurane was administered at a concentration of 0.5%. The remainder of the procedure was unremarkable. Histologic analysis of the mass revealed a malignant meningioma. After recovery from the procedure, the dog underwent radiation therapy (16 sessions of 3 Gy/session) and was doing well 6 months after surgery.

Discussion

In humans, sudden swelling of the brain through a surgical incision is known as acute open brain herniation^1,2 and has been reported in patients undergoing neurosurgery for elective tumor resection and in patients undergoing emergency decompression procedures for treatment of traumatic brain injury. To the authors' knowledge, acute open brain herniation in dogs has not been reported previously. Potential reasons for acute open brain herniation include jugular vein occlusion, hypercapnia, hypoxemia, and subarachnoid or intraventricular hemorrhage. However, in most patients, an underlying cause is not identified. Treatment for acute open brain herniation, regardless of the underlying cause, is directed at decreasing intracranial pressure and cerebral edema.

Intracranial pressure is the pressure exerted by tissues and fluids within the skull and, under normal circumstances, ranges from 5 to 12 mm Hg. When the skull is closed, the intracranial pressure is related to the volume of the intracranial contents and will increase as the volume of the intracranial contents increases. Volume of the intracranial contents is a function of the volume of the brain itself and the volume of the CSF, blood, and any masses within the calvarium. If the volume of any of these individual components increases without a concomitant decrease in the volume of the other components, intracranial pressure will increase. For example, when IV volume expands, as, for instance, during IV fluid administration, compensatory mechanisms maintain intracranial pressure by decreasing CSF production, increasing outflow of CSF through the foramen magnum into the spinal cord, or shunting venous blood away from the brain. Above a certain threshold intracranial volume, however, the brain is no longer able to compensate, and intracranial pressure will increase dramatically.

An increase in intracranial pressure may decrease cerebral perfusion pressure, which can lead to brain ischemia and damage. Cerebral perfusion pressure is the difference between the force driving blood into the brain (arterial blood pressure) and the force resisting movement of blood into the brain (intracranial pressure). It is calculated by subtracting intracranial pressure from mean arterial pressure. Thus, when intracranial pressure is high, the cerebral perfusion pressure is low, and cerebral blood flow (ie, perfusion of the brain with arterial blood) decreases, reducing delivery of oxygen to the brain. Reduced delivery of oxygen to the brain stimulates release of catecholamines, which results in an increase in mean arterial pressure. However, this increase in mean arterial pressure triggers pressure receptors in the aortic arch and carotid body, causing the heart rate to slow (baroreceptor reflex), giving rise to the Cushing reflex.

In patients who are anesthetized and are known to have neurologic disease, the Cushing reflex is an indication of a potentially life-threatening decrease in cerebral perfusion pressure, and emergency treatments should be instituted. Strategies to rapidly reduce intracranial pressure in patients that are anesthetized include hyperventilation, IV administration of hyperosmolar agents, and a reduction in the delivered concentration of any volatile anesthetic.³ These strategies take advantage of the fact that cerebral blood vessels constrict or dilate in response to alterations in PaO₂ and PaCO₂. In particular, a decrease in PaCO₂ will cause a constriction of cerebral blood vessels. Thus, hyperventilation, by reducing PaCO₂, can cause a rapid decrease in intracranial pressure. For example, a 10 mm Hg reduction in PaCO₂ can reduce intracranial pressure by up to 30% within 15 seconds.⁴ However, additional decreases in cerebral blood flow do not occur when PaCO₂ is < 25 mm Hg, and it has been suggested that a PaCO₂ < 30 mm Hg in compromised brain tissue may lead to additional brain injury.^5,6 Therefore, in emergency situations, a target PaCO₂ of 30 mm Hg or target PETCO₂ of 35 mm Hg should be used. Because prolonged periods of hypocapnia have been shown to worsen outcome in patients with traumatic brain injury,⁷ hyperventilation should only be instituted for short periods.

Volatile anesthetics have been shown to have a dose-related effect on intracranial pressure. At concentrations of 1 to 1.5 times the minimal alveolar concentration, the vasodilatory effects of volatile anesthetics lead to increases in intracranial pressure and decreases in cerebral perfusion pressure. These vasodilatory effects are not clinically important in animals without brain disease because they are able to compensate for the increase in blood volume. The dog in the present report was premedicated with hydromorphone and midazolam (an opioid and a benzodiazepine), and anesthesia was maintained with fentanyl and propofol (an opioid and a hypnotic). Use of opioid-benzodiazepine or opioid-hypnotic combinations allows for a reduction in doses needed for induction and maintenance of anesthesia with minimal effects on arterial blood pressure. Sevoflurane was administered at a low concentration in this dog to ensure sufficient anesthetic depth. If the dog had had signs of a high intracranial pressure prior to anesthesia, we would have omitted use of sevoflurane and maintained anesthesia by means of IV drug administration alone. In humans undergoing neurosurgical procedures, total IV anesthesia has been shown to maintain cerebral autoregulation better than use of inhalant anesthetics.^8,9 In addition, a previous study¹⁰ has suggested that propofol may have various neuroprotective properties.

Hyperosmolar agents that are commonly used include mannitol and hypertonic saline (7.2% NaCl) solution. Both of these agents work by causing movement of water from the intracellular and interstitial spaces of the brain into the intravascular space. For both of these agents, their effectiveness in reducing intracranial pressure depends on the fact that they do not permeate the blood-brain barrier. For any substance, permeability across the blood-brain barrier is based on the substance's reflection coefficient,^11–15 with solutions that have a reflection coefficient ≤ 1 less able to cross the blood-brain barrier. Mannitol has a reflection coefficient of 0.9 and an osmolarity of 1,372 mOsm/L.^11–13 Hypertonic saline solution has a reflection coefficient of 1.0 and an osmolarity of 2,464 mOsm/L.^14,15 Use of either mannitol or hypertonic saline solution is controversial when there is a possibility that integrity of the blood-brain barrier has been disrupted because leakage of either substance into the brain interstitium could worsen cerebral edema. In addition, mannitol is a potent diuretic, as it is freely filtered by the glomerulus with < 10% reabsorbed by the renal tubules. Because of this, fluid homeostasis needs to be closely monitored in patients receiving mannitol to prevent dehydration. On the other hand, administration of multiple boluses of hypertonic saline solution has been shown to cause electrolyte abnormalities.

Other strategies to reduce intracranial pressure in patients that are anesthetized include ensuring that the jugular veins are not occluded because occlusion could prevent blood from leaving the brain, maintaining adequate oxygenation, and ensuring that blood pressure is adequate. Finally, ensuring a smooth anesthetic recovery is vital, as emergence excitement or pain can lead to an increase in sympathetic stimulation and a subsequent increase in intracranial pressure. This can be accomplished by providing adequate perioperative analgesics, allowing recovery in a low-stimulus environment, and administering small doses of an opioid, benzodiazepine, or other hypnotic agent if emergence delirium appears to be occurring.

In the dog described in the present report, there was no clear indication of the inciting cause for the acute open brain herniation, although herniation occurred just after removal of the mass. Actions taken to reverse the herniation included removal of any potential contributing factors (ie, discontinuing inhalant anesthesia, decreasing PaCO₂, and ensuring that the jugular veins were not occluded) and direct measures to reverse brain swelling (ie, IV administration of a hyperosmolar agent and topical administration of cold saline solution). When the potential outcome of an anesthetic complication is disastrous, a multiple treatment approach is beneficial.