[soundcloud_ultimate track=https://soundcloud.com/adamrodman/episode-2-full-code]

In episode 2 of Bedside Rounds (though still technically untitled), I talk some about the myths and realities of resuscitation in the hospital, and how the media influences how doctors and patients approach these important conversations.

If you’re ever admitted to the hospital, it’s almost a certainty that you will have the discussion with your admitting physician that we refer to, rather antiseptically I feel, as “clarifying your code status.” The discussion goes something like this:

“Ms. Smith, this is a question I ask all my patients who come into the hospital, and it doesn’t necessarily mean I think anything bad is going to happen, but if it does, I want to respect your wishes.  If your heart were to stop, and you were to essentially be dead, would you want us to attempt chest compressions, potentially insert a tube down your throat to breath for you, and possibly attempt to shock you in an attempt to bring you back to life?”

If they say yes, we call this “full code,” a no is a DNR/DNI, or do not resuscitate, do not intubate. The details vary depending on the patient and the clinical situation that we have to “clarify” – it goes very different with a 20-year old woman in for appendicitis from the elderly, bed-bound Alzheimer’s patient who is in with pneumonia,  but the core of the conversation remains the same – if you die, do you want us to try and bring you back? It seems like a no-brainer, one of the victories of modern medicine.

Now, I want to talk about my first experience actually being in a code, which happened during my intern year of residency (and for those of your keeping track, my intern year ended two weeks ago). A code blue was called overhead – the pretty much universal hospital designation for a patient requiring immediate resuscitation – and I realized that I was only four rooms away. The whole experience was surreal. I remember the patient – bloated belly, yellow skin, clearly had suffered from liver failure. I remember doing chest compressions and feeling his ribs break, and thinking how life-like the simulators I had trained on in medical school actually were. I remember the smell – his picked at dinner was sitting beside me as I pushed on his chest. I remember how calm my colleague running the code seemed. And I remember how long the entire process seemed, even though it was only about 20 minutes, and how tired, both physically and emotionally, I was after it. And I should mention, the patient was not resuscitated.

Now I want to contrast that to this clip from ER. To set the scene, there are three doctors  standing around the patient, who has just become unresponsive.

So one doctor has noticed that his rhythm on the telemetry monitor is in normal sinus, making this pulseless electrical activity, and not, correctly, not fib. Someone’s hands come out of nowhere and start bagging the patient. The main physician does a first bump or two on his chest. And that’s it, the patient gasps and he’s back

As an aside, I found this video from a youtube account called “maleresuscitation”, which at current count has 38 videos from various TV shows showing attempts at resuscitation, proof you can find anything online.

TV is obviously fantasy, but it can subtly shape the way that we think. The problem is, I think, that when I have the code status talk with my patients, they’ve already seen far more codes than I, and these codes tend to go a lot differently than what normally happens in the hospital.

How differently? A study in the New England Journal of Medicine in 1996 attempted to answer that question. I wouldn’t normally put my listeners through a methods section of the NEJM, but this part is too great to leave out. And I quote:

“We watched all the episodes of the television programs ER and Chicago hope during the 1994-1995 viewing season and 50 consecutive episodes of Rescue 911 broadcast over a three month period in 1995.”

Couch potato as medical research. Awesome. The most interesting finding is that TV doctors are a lot better at their job than their real life counterparts. On TV, the short-term survival rate is 75%, and no patients are shown dying afterwards. In reality, the survival rate of a cardiac arrest at home is ~10%, and in the hospital, only 17% of people with an arrest survive to discharge.

Maybe even more interesting was the underlying cause of the code. In reality, 75-95% of arrests are due to underlying cardiac disease. In the 97 episodes reviewed, there were 60 occurrences of CPR. Over half were due to various factors like gunshots, drowning, or surgery. Only 28 were cardiac arrests and many were from unusual causes, like lightning strikes, hypothermia, and pericarditis from lupus erythematosus. And only seven of those 60 had any underling illnesses.

The study found a focus on miracles. On Rescue 911, a show which recreates real emergency situations with actors, the word was used to describe 10 of 18 instances of CPR, all of which were successful.

But what I think the ultimate oversight is, is that there are only two outcomes shown on television – full recovery, or in a minority of situations, death. There was only one case of a long term disability shown in patients who were resuscitated, which was a young man who had a mild dysarthria. In reality, more than half the survivors – and these are low rates of survivors anyway – have varying degrees of brain damage.

As the article concluded, “If CPR were a benign, risk-free procedure that offered a good hope of long term survival in the face of otherwise certain death, few people would ever choose to have medical personnel withhold resuscitation”. The reality, of course, is far more complicated.

So when this “clarifying code status” conversation happens on admission – usually at a very stressful and emotional time in the patient’s and their family’s life – the physician and patient might not even be on the same page, since our experiences of what constitutes a code are drastically different.

So how can we help bridge that gap? One excellent way is the POLST program here in Oregon, and now in many states nationwide, which among other things encourages patients to talk about end-of-life decisions with their primary care providers, prior to a crisis situation happening, which is a great topic for another podcast.

But if a patient doesn’t have a POLST, that doesn’t help much when it’s 2 AM, and it’s just me and a patient in a dark hospital room. How can I help educate the patient about the risks and benefits of even attempting CPR?

So something that I’m jazzed about is an outcome prediction tool recently published in JAMA, called the GO-FAR tool (Good Outcomes Following Attempted Resuscitation). The tool was developed from looking at the characteristics and outcomes of 51,000 patients across 366 hospitals in the US, and uses 11 variables to determine the chance of a good neurologic outcome at hospital admission, defined as “the patient being conscious, alert, and able to work, but might have mild neurologic and psychological deficits”.

Maybe GO-FAR or another prediction tool will help us to change that code conversation, and give patient’s a reasonable expectation of their chances of recovering from a code. Maybe it will help to give the code talk, and say, “Ms. Smith, since you have metastatic breast cancer and pneumonia, your chance of having a good outcome from a code is less than 1%”. I don’t know.

But the editorial accompanying the piece raises some interesting questions about the ubiquity of the code conversation in the hospital world. Basically, it argues that cardiopulmonary resuscitation should be treated like any other medical procedure, like, say, dialysis or heart surgery, with specific indications, risks, benefits, moral, religious and social implications, and be subjected to rigorous quality improvement; basically, that the sole indication for CPR is NOT just the fact that the patient is having a cardiac arrest.

Would this improve patient care and reduce suffering? I don’t know. But after all, the first precept of medicine is primum non nocere – above all, do no harm.

Bibliography

Diem, Lantos, and Tulsky, Cardiopulmonary Resuscitation on Television – Miracles and Misinformation, NEJM June 1996

Ebell et al, Development and Validation of the Good Outcome Following Attempted Resuscitation Score to Predict Neurologically Intact Survival After In-Hospital Cardiopulmonary Resuscitation, JAMA Internal Medicine, November 11, 2013

Young, Bryan, Neurologic Prognosis after Cardiac Arrest, NEJM, August 6, 2009.

Berger, J, “Improving Quality Improvement for Cardiopulmonary Resuscitation,” Nov 11, 2013