NEUROSURGICAL procedures are often performed in the prone, sitting, or lateral position for improved surgical access. Cardiopulmonary arrest while in such a position creates a challenge for both the anesthesiologist and surgeon. We present two cases of cardiopulmonary arrest and unsuccessful attempts at resuscitation after repositioning supine. In both cases, the surgical site of bleeding became inaccessible with repositioning, leading to the question of whether prone resuscitation would have provided a better alternative.
A 21-yr-old woman with a medical history significant for neurofibromatosis type 2 and multiple neurofibromas presented with a right jugular foramen tumor and was scheduled to have a skullbase craniotomy for tumor debulking. After induction of anesthesia, the patient was placed in a left lateral decubitus position. During the surgical procedure, the sigmoid sinus was disrupted, leading to significant blood loss and hypotension. Blood products were delivered via a rapid infusion device and vasopressors were administered. Attempts to gain further access to the bleeding site resulted in two additional small lacerations of the transverse-sigmoid junction. Attempts to suture these and ligate the sigmoid sinus were unsuccessful. Despite the compromised sinus, there was no evidence of venous air embolism, as judged by end-tidal carbon dioxide and peak inspiratory pressures. The patient’s hemodynamics continued to deteriorate, and she was placed in the supine position for cardiopulmonary resuscitation (CPR). After turning the patient to the supine position, the surgical site became inaccessible and the patient continued to bleed briskly from the wounds. The patient developed pulseless electrical activity and resuscitation efforts were unsuccessful.
A 69-yr-old woman with renal cell carcinoma that metastasized to the lumbar spine and encroached on the neural foramen presented for anterior and posterior lumbar corpectomy and fusion. Her surgical history was significant for nephrectomy, and she had undergone embolization of the mass 1 d before surgery. After induction of anesthesia, the patient was positioned prone, and the posterior portion of the planned procedure was uneventful. The patient was then placed in a lateral park-bench position for the anterior fusion. Upon scraping of the second lumbar endplate, a small tear of a contralateral segmental vessel off the aorta occurred. Blood loss was significant and the patient was treated with fluids, blood products, and vasopressors. The aorta was cross-clamped and the vessel repaired. After repair, the patient’s hemodynamics stabilized and the surgical field remained dry. Approximately 10 min later, multiple sites of oozing were noted from the retroperitoneal fat, from the site of vertebral corpectomy and muscle. The patient’s blood pressure began to decrease and required ongoing volume resuscitation, a norepinephrine infusion, and a 50-μg bolus of epinephrine. As her blood pressure improved, the bleeding increased and the patient developed pulseless electrical activity arrest. The wounds were packed to tamponade the bleeding, and the patient was positioned supine for CPR. Access to the source of bleeding was impossible in the supine position—during resuscitation, blood was noted to be exiting the surgical site. After 20 min of unsuccessful CPR, resuscitation efforts were ceased.
Neurosurgical procedures often involve various nonsupine positions, including prone, sitting, and lateral decubitus. Although intraoperative cardiac arrest is fortunately an uncommon event, when it occurs with the patient in a position other than supine, access to a bleeding operative site can be difficult or impossible. In addition, an open wound, time required to reposition the patient, and an unstable spine may lead to a deleterious outcome. Alternative methods for CPR may therefore be desirable.
Cardiac resuscitation was first described by Kouwenhoven et al. in 1960 and has remained essentially unchanged for more than 45 yr.1It was originally thought that blood circulates based on direct compression of the heart against the vertebral column. Rudikoff et al. challenged this theory by demonstrating that alterations in intrathoracic pressure lead to blood flow and direct cardiac compression is not required.2This “thoracic pump mechanism” suggests that the heart is merely a conduit for blood flow. In 1989, McNeil3proposed prone CPR as an alternative to supine CPR for out-of-hospital arrests to address problems with traditional CPR. Such problems include reluctance of providers to perform mouth-to-mouth ventilation, risk of gastric distention and aspiration, and difficulty in learning and retention of the procedure.4
Prone resuscitation by reverse precordial compressions was first described in two case reports by Sun et al. in 1992.5The first procedure was a posterior fossa craniotomy in which the transverse sinus was torn; the second was a cervical spine procedure in which an obstructed endotracheal tube led to arrest. The resuscitative method involved placing the left clenched fist under the sternum while compressing the mid-thoracic spine with the right hand. This produced arterial systolic pressures of 160 mmHg in one case and 200 mmHg in another. In both cases, a life-sustaining rhythm and blood pressure were obtained within 6 min. In 1994, Tobias et al. described a modified technique during spinal surgery with an open wound.6In this case, the surgeon compressed the thoracic cage with a palm placed over the scapula on each side of the spinal wound. The patient was on a frame that supported his anterior thorax so counterpressure on the sternum was not necessary. Systolic blood pressures of 80-90 mmHg were noted, and the patient was successfully resuscitated after 7 min of asystole.
Based on McNeil’s description of CPR in the prone position and the case reports of intraoperative prone CPR, Mazer et al. performed a pilot study of CPR in the prone position.7The study compared traditional CPR and reverse CPR with the primary endpoint of systolic blood pressure. They revealed that reverse CPR dramatically improved systolic blood pressure and mean arterial pressure (mean improvements 23 ± 14 mmHg and 14 ± 11 mmHg, respectively). Brown et al. performed a systematic review of 22 cases of intubated hospital patients who received prone CPR, 10 of whom survived.8
The 2005 American Heart Association Guidelines for CPR and Emergency Cardiovascular Care recommend that prone CPR be considered when the patient cannot be placed in the supine position, particularly for the hospitalized patient with an advanced airway.9The recent patients at our institution were positioned lateral decubitus at the time of cardiopulmonary arrest; we have found no reports describing CPR in the lateral position. In the cases we describe, placing the patient in the prone versus the supine position for resuscitation may have provided improved access to the surgical site and at least preserved the potential for surgical correction of the problem. Even in arrests not involving exsanguination, the studies of Mazer et al. and Wei et al. suggest that prone CPR may be advantageous in generating higher blood pressures compared with the supine position.7,10
These cases suggest that the prone position should be considered as the optimal choice for CPR in certain limited circumstances, even if the supine position is achievable. As such, further research on prone CPR is warranted.