Detailed review of unusual cases is a cornerstone of anesthesiology education. Each month, the AQI-AIRS Steering Committee will abstract a case and provide a detailed discussion based on a submission to the national Anesthesia Incident Reporting System. Feedback regarding this item can be sent by email to Report incidents to

Case 2013-11: Automatic or Manual?

A 75-year-old man presented for 3-level resection of lumbar spinal osteophytes and subsequent fusion. The patient had no known medication allergies. His past history was significant for stable coronary artery disease, hypertension, type-2 diabetes and spondylolithesis. He was taking a beta-blocking agent and a statin. Exercise tolerance was limited by back pain. Preoperative hematocrit was 31 percent. Transthoracic echocardiography with dobutamine challenge was negative for inducible ischemia and demonstrated an ejection fraction of 33 percent. Induction of general anesthesia and endotracheal intubation were unremarkable, and an arterial line and central line were placed prior to turning the patient prone for surgery. The first four hours of surgery were notable only for persistent tachycardia (90-100 bpm), treated with fluids, narcotics and esmolol. Estimated blood loss was 600 mL, consistent with hematocrit of 27 percent and normal lactate on arterial blood gas assessment. Heart rate increased to more than 110 bpm as surgical closure was completed. Blood pressure and end-tidal CO2 fell, unresponsive to phenylephrine and ephedrine administration, and ST segment abnormalities appeared.

The patient was turned supine onto a gurney and external cardiac compressions begun for pulseless electrical activity (PEA). Multiple doses of epinephrine and several minutes of CPR led to return of spontaneous circulation. An emergency manual present in the O.R. was consulted by the anesthesia team during the course of resuscitation. Treatable causes of PEA were sequentially ruled out. Transesophageal echocardiography (TEE) revealed a dilated, globally hypokinetic heart. Blood gas analysis revealed a hematocrit of 26 percent. The patient received two units of red blood cells and was maintained on an epinephrine infusion to the ICU. The patient subsequently made a slow but steady recovery and was eventually discharged from the ICU neurologically intact.


The cause of this patient’s cardiac arrest was not immediately obvious, but consideration of the “8Hs and 8Ts” would indicate the possibilities (Table 1). Blood loss was an obvious consideration, but less likely given the relatively small drop in hematocrit and lack of metabolic acidosis seen on the arterial blood gas taken at the end of surgery. Hypoxia was ruled out by pulse oximetry and hypothermia by continuous core temperature measurement. Acidosis was unlikely, given the ABG values. There was no ECG evidence of hyperkalemia and no inciting event for a sudden change (i.e., succinylcholine administration). The patient’s potassium level had been normal preoperatively, making hypokalemia unlikely.

Mechanical causes of arrest – tension pneumothorax or pericardial tamponade – might have resulted from surgical misadventure or as a complication of central line placement, but were less likely given the timing. Tamponade was definitively ruled out by echocardiography. Tension pneumothorax was deemed unlikely based on no change in ventilation parameters and continued equal and bilateral breath sounds; had circulation not returned, a chest radiograph to definitively eliminate this possibility would have been in order. In situations with a higher risk (i.e., arrest occurring within minutes after line placement; surgeries in or near the thorax; perceived difference in breath sounds), placement of a tube thoracostomy would have been indicated.

Air embolus was unlikely during surgical closure. Toxins are a specific concern in the O.R., but the patient had received no medications immediately prior to deterioration. There were no signs of an allergic reaction. Pulmonary embolus from a deep venous thrombosis was a possibility, as in any long orthopedic procedure. The patient had received perioperative prophylaxis with subcutaneous heparin, but this does not reduce the risk to zero. TEE did not demonstrate right heart dilation relative to the left side, and there were no changes in oxygenation until arrest occurred. Neither of these findings are definitive, however, and PE can occur transiently and then resolve as the clot breaks up or lyses during CPR. Further work-up for this condition is warranted, depending on the patient’s subsequent course. Air embolus can occur during any spine surgery,1  but is less likely during closure than during dissection; the TEE findings also make this diagnosis unlikely.

The remaining, and most likely, possibility is an episode of myocardial ischemia. Multiple factors suggest this, including the patient’s age and past history, the persistent intraoperative tachycardia, the ST segment changes seen prior to PEA, the globally hypokinetic heart and the response to epinephrine. Like many cardiac arrests in older, comorbid patients, the exact pathophysiology of the event may remain somewhat elusive. Coronary thrombosis is a possibility, as is coronary spasm or global ischemia and dysfunction due to a momentary mismatch of oxygen supply and demand. A transient pulmonary embolus could have initiated cardiac failure which then persisted. External cardiac compressions can cause direct myocardial injury, pericardial tamponade and pneumothorax, all of which can confound diagnostic efforts following return of spontaneous circulation. Further diagnostic work, up to and including urgent percutaneous coronary intervention, would be strongly indicated in this patient.


Cardiac arrest occurs rarely in the O.R., in part perhaps because anesthesiologists are expert at early identification and treatment of hemodynamic instability. Unadjusted data from the National Anesthesia Clinical Outcomes Registry suggests a risk of cardiac arrest occurring prior to PACU discharge of about 0.06 percent, or 1 per 1,800 cases. The newly developed Anesthesia-ACLS course, offered in the ASA Education Center provides a self-paced, online curriculum for understanding key features of intraoperative cardiac life support. This course suggests an organized approach to the unstable patient, and emphasizes likely causes of arrest and the most efficient treatments available in the O.R. Airway misadventures and hypoxia are more likely causes than in out-of-hospital arrests, and protocols for advanced cardiac life support (ACLS) should be modified accordingly. Compared to out-of-O.R. arrests, blood loss is a more likely cause, and primary cardiac events – this case notwithstanding – are less common. The early use of TEE as a diagnostic aid is strongly recommended as an advantage that anesthesiologists have over most other first-responders. Outside the O.R., or in smaller hospitals without TEE, the emerging FEER exam (Focused Emergency Echocardiography in Resuscitation) based on surface ultrasound is likely to be a component of the next edition of ACLS.

Noteworthy in this case report was the use of an emergency manual to guide the anesthesia team. The reporter noted that the anesthesia resident was assigned to consult the manual and read off the recommendations while the anesthesia attending directed the code response. Use of emergency manuals – and strategic positioning on anesthesia and code carts throughout the perioperative area – is a recommendation of the Anesthesia Patient Safety Foundation. Mobile and online versions are also under development. More information can be found at:, and at The purpose of the manual is to help responders avoid diagnostic fixation and errors of omission in the face of increasing patient and procedural complexity. Like the crisis checklists used in the aviation industry, the emergency manual offers a framework for strategic thinking about a given patient problem and provides decision support for the over-stressed provider at a time when seconds count.2  O.R.s and anesthesia practices not yet using emergency manuals are strongly advised to consider providing them in the perioperative area.


Venous air embolism during lumbar laminectomy in the prone position: report of three cases
Anesth Analg
A simulation-based trial of surgical-crisis checklists
N Engl J Med