Portal de Periódicos da CAPES

Probable Venous Air Embolism Associated with Removal of the Mayfield Skull Clamp

1995; Lippincott Williams & Wilkins; Volume: 80; Issue: 5 Linguagem: Inglês

10.1097/00000539-199505000-00036

1526-7598

Francisco Grinberg, Thomas Slaughter, Brian J. McGrath,

Airway Management and Intubation Techniques

Venous air embolism (VAE) is recognized as a serious complication occurring during neurosurgical procedures requiring craniotomy. Most often, embolic events occur in the sitting position and are temporally associated with surgical entry into the venous structures and marrow spaces of the skull [1]. This report describes the probable occurrence of VAE occurring at the conclusion of surgery in the supine position after removal of the Mayfield skull clamp fixating pins. Case Report A 61-yr-old, 65-kg male was admitted to the neurosurgery service for a right temporal lobectomy for tumor. The past medical history was notable for a small cell carcinoma of the pyriform sinus (T2-N2) treated with radiation and chemotherapy 2 yr prior to admission. Preoperative physical examination, laboratory evaluation, and chest radiograph were unremarkable. The electrocardiogram revealed a normal sinus rhythm with an incomplete left bundle-branch block. In the operating room, two 16-gauge peripheral intravenous catheters and a radial arterial catheter were inserted. Anesthesia was induced with thiopental (4 mg/kg), fentanyl (4 micro gram/kg), and midazolam (2 mg) intravenously. Pancuronium (0.1 mg/kg) was administered for muscle relaxation. After applying the Mayfield skull clamp, the patient was placed in the supine position with a 15 degrees right shoulder elevation and a 45 degrees left head rotation. Anesthesia was maintained with fentanyl, isoflurane, and oxygen. The end-tidal PCO2 was monitored with an Ohmeda respiratory gas monitor and maintained at 25 mm Hg by hyperventilation. Vancomycin (1 g), furosemide (0.5 mg/kg), and mannitol (0.75 g/kg) were administered prior to craniotomy. Methylprednisolone (125 mg) was administered at 1-h intervals. Surgery proceeded without incident for 5.5 h. At the end of the procedure, after closure of the incision site, the head was lifted several centimeters above the horizontal plane as the Mayfield fixating pins were removed from the skull. Prior to pin removal, the patient was hemodynamically stable with a blood pressure of 110/70 mm Hg, heart rate of 70 bpm, and an end-tidal PCO2 of 32 mm Hg. Immediately after pin removal, the patient developed a sinus tachycardia at a rate of 130 bpm and the systolic blood pressure decreased to 50 mm Hg. The end-tidal PCO2 precipitously decreased to 13 mm Hg. During the 5 min after the event, the heart rate slowed to 56 bpm and the patient developed 3-4 mm of ST segment increase in lead V5. Nitrous oxide and isoflurane were discontinued immediately upon initiation of the tachycardia, and 100% oxygen was administered. The patient was placed in the Trendelenburg position and lactated Ringer's solution was rapidly infused. Boluses of phenylephrine (100 micro gram/dose) were ineffective in restoring the systemic arterial pressure. Epinephrine (10 micro gram/dose; total dose 50 micro gram) increased the arterial pressure to 125/70 mm Hg. The ST segment increase resolved over a period of 10 min as the systemic arterial pressure returned to preevent levels. To optimize myocardial perfusion, an infusion of nitroglycerin at 25 micro gram centered dot kg-1 centered dot min-1 was administered after stabilization of arterial blood pressure. Chest radiographs obtained intra- and postoperatively were unremarkable. The patient was transferred to the surgical intensive care unit where serial electrocardiograms and measurements of myocardial enzymes remained unchanged. The patient's recovery from the procedure was subsequently uneventful, and no cardiac or neurologic damage was demonstrable on follow-up visits. Discussion The incidence of VAE during neurosurgical procedures ranges from 10% to 47%, dependent upon patient position and the sensitivity of the detection system [1]. Air may be entrained by the venous circulation from any site at which atmospheric pressure exceeds that of the local venous pressure. The risk of entraining air increases at sites where anatomic structures "stent" open the veins preventing their collapse, i.e., the venous sinuses, epidural veins, and marrow cavities. VAE originating from sources external to the operative site occurs infrequently. Four prior reports [2-5] suggest the association of VAE with the pin insertion sites of head-fixating devices. Failure to adequately secure the head in a pin-type fixating device would appear to increase the risk of VAE. In two of three patients described by Wilkins and Albin [2], movement of the patient's head within the head-fixating device preceded the occurrence of VAE. Furthermore, Pang [3] reported a case in which placement of a pin-type head-holder in a 7-yr-old patient with a cerebellar astrocytoma produced a comminuted depressed skull fracture with subsequent VAE [3]. In contrast to prior reports, examination of our patient in the postoperative period failed to demonstrate any evidence of traumatic injury at the pin insertion sites. Furthermore, the patient's head was firmly secured in the Mayfield device throughout the surgery. Traumatic displacement of the head-fixating device is not, however, a necessary prerequisite to VAE. Embolism may occur in the absence of trauma and in fact may occur around the pin, in situ, with the head-fixating device remaining in place, as described by Cabezudo et al. [4] and DeLange et al. [5]. The signs and symptoms of VAE vary depending on the total volume of air embolized and the rate of air entrainment [1]. At the time of hemodynamic decompensation, a precordial Doppler monitor was not being used in our patient. Although each of the prior reports of VAE associated with pin insertion sites occurred in the sitting or semisitting position, our patient was in a supine position throughout the operative period. Immediately prior to hemodynamic decompensation, however, we observed that the patient's head was lifted several centimeters above the horizontal plane as the Mayfield-fixating device was removed and an occlusive dressing was applied. Although one could speculate that a primary cardiac event might generate similar diagnostic signs, the sudden decrease in end-tidal PCO2 associated with profound hemodynamic collapse and the lack of concomitant evidence for myocardial injury support the diagnosis of VAE. Although not as sensitive a diagnostic sign as end-tidal nitrogen, a sudden decrease in end-tidal PCO2 is a characteristic marker of VAE [6,7]. ST segment increase was not apparent until several minutes after development of the tachycardia and hypotension. Furthermore, postoperative evaluation failed to detect any evidence of myocardial injury. Whether the ST-segment increase occurring in this patient resulted secondary to hypotension or to air in a coronary artery could not be determined; however, given that 27% of the general adult population retain a probe patent foramen ovale, paradoxical air embolism resulting in temporary occlusion of a coronary artery is possible [8,9]. Early detection and treatment of VAE has been the key to improved patient survival. The case presented describes the probable occurrence of VAE in the supine position after removal of Mayfield fixating pins from the skull at the conclusion of surgery. Penetration of head-fixating pins into a periosteal sinus or vein provides a direct source for VAE. This case clearly demonstrates the necessity of removing headfixating devices in the supine position and of monitoring for VAE throughout the entire intraoperative period until all potential sites for air entry into the circulation have been occluded.