Cardiac Arrest In The OR

My hospital routinely runs mock codes in a variety of settings to make sure that all providers can quickly and efficiently act as a team if the worst happens. This includes our operating rooms. Cardiac arrest under anesthesia in the OR is fortunately rare.  The total number of OR procedures performed in the US last year was 51.4 million. The incidence of intraoperative cardiac arrest has been quoted as less than 0.1%. However, a small percentage of a large number is still a fairly significant number. Getting prepared for rare events is often more challenging than preparing for regular events. When a rare event happens the shock and fear induced by the event can erase protocol from the most prepared mind. Let me share some of the things that I’ve learned over almost 36 years of OR emergencies.

Always Assume An Alarm Is Real Until Proven Otherwise

Recognizing an arrest in the OR may not be easy. The vast majority of times that our alarms go off in the OR, they are false alarms. A lead has become disconnected mimicking asystole, the cautery device is producing interence that looks like V-tach, the pulse oximeter is off the finger, the end-tidal-CO2 sensor has gotten plugged with moisture. I could go on. However, and this is a huge “however”, sometimes the alarms are real.

I was once on a third world medical volunteer trip where the anesthesiologist spent precious minutes troubleshooting the loss of a pulse oximeter reading thinking he had a mechanical failure, when what he really had was a patient in cardiac arrest. We did successfully resuscitate the patient.

The trap of ignoring an alarm can happen to anyone. Never assume that an alarm is false. I tell my students “don’t just look at your monitors, look at your patient”. Feel for a pulse, listen for heart tones and breath sounds, look at patient color. The devices are there to help us, not replace us. We are the most important monitor in the room.

Warning Signs of Intraoperative Cardiac Arrest

So, what are some of the monitor signs that our patient has either arrested or could potentially arrest in the OR?

  • EKG with pulseless rhythmVentricular Tachycardia [V-Tach]
  • Ventricular Fibrillation [V-Fib]
  • Severe Bradycardia
  • Asystole
  • Loss of Carotid pulse > 10 seconds
  • Loss of End Tidal CO2 on capnograph
  • Loss of Arterial line tracing
  • Loss of waveform on pulse oximeter

If you see something worrisome on the monitors, start with your patient.

Tips About Managing Cardiac Arrest In The OR

Emergency Check Lists

I’m not going to review the BLS/ACLS guidelines here. Needless to say preparation involves knowing those protocols. As I said earlier, however, fear can erase a protocol from even the best prepared mind. My OR has printed and laminated check lists for management of serious complications which fortunately rarely happen. These include cardiac arrest, malignant hyperthermia, severe hemorrhage, airway fire, among others. Being able to reference these lists in an emergency is helpful in making sure a critical step is not forgotten. Consider making lists such as these available.

Work The Differential diagnosis

Again, I’m not going to review all of the possible causes of cardiac arrest, which of course include preexisting comorbid patient conditions. But I am going to remind you of rare events that are more unique to the perioperative environment and which must be thought about to be diagnosed. These events include:

  • respiratory arrest from opioid overdose
  • hypoxia from loss of the airway, accidental extubation, laryngospasm
  • reaction to anesthesia: allergic, malignant hyperthermia, sensitivity
  • anesthesia too deep for level of stimulation
  • vasoactive infusion running too fast
  • complications of surgery, such as acute blood loss or organ injury
  • air embolism or pulmonary embolism
  • tension pneumothorax, either from surgical complication or barotrauma
  • cardiac tamponade, from central venous access or surgical complication

If you don’t imagine something could happen, you won’t diagnose it quickly enough to help. Even more so than managing cardiac arrest out of hospital or the ward, consider the unique things that can cause cardiac arrest in the OR environment. If you don’t think of it, you won’t diagnose it in a timely fashion.

Have The Gurney Outside the OR for all Prone Cases

It is easier to turn the prone patient supine onto a gurney, than it is to rotate them on the OR table. Time is critical and you do not want to wait for a gurney to be brought from preop or recovery room. Always keep the gurney outside the OR if your patient is prone.

Call for crash cart/defibrillator early

Call for the crash cart and defibrillator as soon as you become concerned that the patient is becoming unstable. Even if you never need it, having it close by will save critical minutes if the patient does decompensate. It will also prevent you from having to send one of your assistants, such as your OR nurse, out of the room just when you need him/her most.

asking for help is a sign of strength

We all have a tendency to wait before asking for help. We believe we can manage the issue ourselves because most of the time we can. Denial is common: surely this can’t be happening —it must be a mechanical malfunction. We may be worried someone will think poorly of us if we ask for assistance.

We have all experienced that queasy feeling in our gut that things are happening too fast,  that we aren’t quite in control, or that we don’t know exactly what is going on. Listen to that sixth sense and call for back-up. Make sure your surgeon and your OR nurse know you are worried. If you end up not needing the help, thank them for coming and send them away. I tell my students that knowing when to ask for help is a sign of strength, not a sign of weakness.

manage the anesthetic

Don’t forget to go to 100% oxygen and turn off or down the anesthetic agent as appropriate. In a crisis, little but important things often get overlooked. It’s very easy to forget to turn the anesthetic back on. If the resuscitation is successful don’t forget to now provide some level of anesthesia for pain control and amnesia until the skin incision is closed.

Who Is Running This Code?

Many us have responded to a cardiac arrest situation where  there were too many people in the room. Multiple providers, from surgeons, to anesthesiologists, to intensivists were all shouting orders. Sometimes in the after-action discussion it turns out that no one person was officially running the code. Key decisions during codes must be made where timing is critical. Someone has to have the big picture. Make sure you know who is running the code or the resuscitation. Don’t be afraid to ask if you don’t know.

In the OR, the anesthesiologist often runs the code. However, the anesthesiologist will often be tempted to also start extra IVs, give medications, etc. because that’s what we do everyday. It may be our anesthetic, making us feel like the anesthetic is still the priority. Optimally, the person running the code is running the code and not performing other duties. Its hard to stand back and not do the things that we do every day — and if you don’t have a lot of helpers it may be unavoidable. But it’s imperative that the provider running the code not get distracted by tasks that will prevent him or her from seeing the bigger picture while running the code.

Who Should Do Chest Compressions?

The anesthesia team is going to be tempted to do the chest compressions. For an abdominal or chest case this is not advisable. You won’t be able to compress the chest in the optimal location. It will interfere with your resuscitation and running the code.

In the OR, it’s okay to ask a surgeon to do the chest compressions on their side of the sterile drape since they will be most likely have better access to the correct spot on the sternum. This allows the anesthesia team more space to resuscitate the patient, as well as less distraction from folks trading places as they place lines, give drugs, etc.

Now, there are also important things that the surgeon needs to do. Control of bleeding comes to mind. Don’t forget you can have someone else other than the surgeon gown and glove quickly to relieve whoever is doing the initial chest compressions. In this circumstance a five minute scrub is not needed: just gown and glove.

Consider lowering the drape to the extent that it is easier to communicate between surgeons and anesthesia team.

Don’t Forget Ongoing Blood Loss!

Speaking of blood loss, don’t forget that it may be ongoing during the code. I was once performing anesthesia for a teenage boy having a T3 to L4 thoracolumbar fusion. Midway through the case the patient arrested. We thought that an air embolism had occurred and we immediately turned the patient supine on a gurney, performed CPR and ACLS  and successfully resuscitated him. However after about 10 minutes of stability the patient once again became hypotensive and developed pulseless electrical activity. In our focus on the initial arrest and relief at the “save” we had not paid attention to the ongoing bleeding from the boys still fully open back wound. He had bled several units of blood into the sheets during the code. Rapid transfusion again resuscitated the patient and he went on to recover fully.

Never forget that when surgery is being performed, other things related to the surgery,  even those unrelated to the initial arrest, may deteriorate during the CPR. Once again: big picture.

check the defibrillator pads

In the OR, the chest area may very well be wet with blood, sweat, and other secretions. During CPR, the defibrillator pads may come lose. Keep checking for a good connection before shocking if the code has gone on for a while.

communication is key

If it’s too noisy to hear, ask for quiet. It’s okay to ask for unneeded people to leave. Providing a verbal summary of resuscitative steps periodically can be helpful to keep everyone in the loop and let others point out suggestions on other things to do/try. Ask for suggestions as the code progresses. Many coordinated brains in an emergency are often better than one.

How Do You Know CPR Is Effective?

Start With Your Patient

Look at your patient’s color. Is it improving? Can you feel a pulse with the compressions? Can you hear breath sounds with ventilation?

End Tidal Carbon Dioxide monitoring

It is standard of care in the modern OR to monitor capnography. End tidal carbon dioxide, is being recognized as a critical indicator of adequacy of chest compressions and also return of spontaneous circulation (ROSC). Kalenda was the first to report this observation in 1978 and observed that this was helpful in assessing rescuer exhaustion. He noted that performing CPR  improved ETCO2. Over time, as the rescuer got tired, ETCO2 decreased again. Replacing the fatigue rescuer with a fresh rescuer resulted in an improvement in ETCO2. Kalenda also reported the connection between improved ETCO2 and ROSC.

Why does this happen? During a normal ventilatory cycle, we inhale air bringing oxygen into the lungs . Oxygen is picked up and delivered to the tissues via the bloodstream. The blood in the periphery drops off oxygen and picks up waste product carbon dioxide, delivering it back to the lungs where we exhale it. If cardiac output drops and there is inadequate circulation, blood and tissue oxygenation will drop and tissue carbon dioxide will rise. However, carbon dioxide delivery back to the lungs will drop. What you will see on your capnograph is that the end-tidal CO2 will abruptly drop.

I see this acute drop of ETCO2 during anesthesia if my patient suddenly becomes hypotensive. As I treat the hypotension, end tidal CO2 rises again, reassuring me that cardiac output is improving.

Keep an eye on the CO2 as an indication of good CPR or of return to spontaneous circulation. It should at least be greater than 10mmHg. Similarly we can use our capnograph to assess CPR. End Tidal CO2 > 20mm Hg on capnography indicates adequate CPR or ROSC. If it’s less than that, improve your compressions.

Of note, an ETCO2 <10mm Hg after 20 min of standard ACLS is associated with 100% failure of ROSC.

Intravascular Monitors

During chest compressions, aim for diastolic pressure of 30-40mm Hg by arterial line catheter at time of full compressions. Coronary perfusion pressure (CPP) should be greater than 15 mmHg. (CPP= Diastolic BP- CVP)


Consider using a portable ultrasound or echocardiogram machine to identify cardiac motion. Our ability to feel a very weak pulse is limited. Sometimes what we think is pulseless electrical activity is really severe hypotension and poor perfusion, with persistent cardiac motion. A patients whose heart is still beating, even if we can’t feel the pulse, has a much higher chance of survival. Placing the ultrasound sub-xiphoid or parasternal to aim toward the heart can tell you if there is still cardiac motion.


Cardiac arrest in the OR is challenging both to recognize quickly as well as to treat. You must be able to work the causal diagnosis in the face of some potentially complex anesthetic, surgical and comorbid combinations. At the same time  you must perform a resuscitation, often in the face of ongoing surgery. To do so successfully requires active imagination, flexibility, teamwork, excellent communication skills, knowledge of protocols, ability to actively assess your patient, and the use of sophisticated monitoring tools.

May the force be with you

Christine Whitten MD

Further Reading

  • Salen P, Melniker L, Chooljian C, Rose JS, Alteveer J, Reed J, Heller M. Does the presence or absence of sonographically identified cardiac activity predict resuscitation outcomes of cardiac arrest patients? Am J Emerg Med. 2005 Jul; 23(4):459-62.
  • Jan M. Shoenberger, Kristy Massopust, Sean O. Henderson. The Use of Bedside Ultrasound in Cardiac Arrest. Cal J Emerg Med. 2007 May; 8(2): 47–50.
  • Kodali BS1, Urman RD1. Capnography during cardiopulmonary resuscitation: Current evidence and future directions. J Emerg Trauma Shock. 2014 Oct;7(4):332-40.
  • Kalenda, Z. The capnogram as a guide to the efficacy of cardiac massage. Resuscitation, 1978 6, 259-263.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.