Guest post: Coding an out-of-hospital cardiac arrest

by Howard Rodenberg, MD, MPH, CCDS

The art of medicine, (and, for that matter, the art of life) is most often taught through what we call “pearls of wisdom.” These are those succinct nuggets of information learned through life experiences, tidbits of knowledge not found in books.

In healthcare, some pearls are metaphorical. For example, the adage “When you hear hoofbeats, look for horses, not zebras,” is a reminder that common things happen commonly, and not everyone has a publication-worthy illness. Others are more concrete. “If someone usually walks, talks, eats, drinks, looks, and smells normal, and now they don’t, you can’t send them home” is good advice in determining the need for admission.

Another group of pearls are more definitive. A preceptor during my Emergency Medicine residency quipped, “You’re either alive or dead. There’s nothing in between.” (This was an attending physician without any ex-wives, which does cast some doubt on his opinion.)

This latter reflection came back to me recently as I received a note from Diana Matysik, a CDI supervisor for Ascension Health in Duluth, Minnesota, who asked a question near and dear to my emergency department (ED) heart. If a patient suffers an out-of-hospital cardiac arrest and is resuscitated before arrival in the ED, should the scenario be described with a Z code?

(Duluth is pretty far north, sitting on the shores of Lake Superior. I understand that summer is scheduled for the third week of August, followed by winter, more winter, and worst winter. That takes us through September. And as you may know, Duluth also is the hometown of one Robert Zimmerman, also known as Bob Dylan. Early in his career, he shared the stage with Bobby Vee, best known for a tune called “The Night has a Thousand Eyes,” to which the Beloved Dental Empress refers me when discussing the Rules of the Relationship, along with the Beatle’s “Happiness is a Warm Gun.” I think there a message there.)

The place to start the discussion is to look at how we should code cardiac arrest. The best summary we have is a note in Coding Clinic, second quarter 1988, which described four scenarios in which one might code cardiac arrest:

• If the patient arrives in the hospital’s emergency service unit in a state of cardiac arrest, cannot be resuscitated or only briefly resuscitated, and is pronounced dead with the underlying cause of the cardiac arrest not established (cause unknown), code 427.5 is assigned as the diagnosis.
• If the patient arrives at the hospital in a state of cardiac arrest, is resuscitated, and is admitted as an inpatient but dies before the underlying cause of the cardiac arrest is established (cause unknown), code 427.5 is assigned as the principal diagnosis.
• Code 427.5, Cardiac arrest, may be used as a secondary code in the following instances:
  
  • The patient arrives in the hospital’s emergency service unit in a state of cardiac arrest and is resuscitated (and admitted) with the condition prompting the cardiac arrest known, such as ventricular tachycardia or trauma. The condition causing the cardiac arrest is sequenced first followed by code 427.5, Cardiac arrest.
  • When cardiac arrest occurs during the course of hospitalization and the patient is resuscitated, code 427.5 may be used as a secondary code except as outlined in the exclusion note under category 427.

When the physician records cardiac arrest to indicate an inpatient death, do not assign code 427.5 when the underlying cause or contributing cause of death is known since the Uniform Hospital Discharge Data Set (UHDDS) has a separate item for reporting deaths occurring during an inpatient stay.

(Note that this Coding Clinic is from the ICD-9-CM era. Today, we would translate these scenarios into ICD-10-CM codes I46.2, I46.8, or I46.9 accordingly if the cause of the arrest is known or unknown.)

The scenario Coding Clinic seems to leave out is when a patient comes in with cardiac arrest, is resuscitated, doesn’t die, and you never find out the cause. In that case, I think code I46.9, Cardiac arrest, cause unspecified, would be an appropriate principal diagnosis.

(Side note: Erica Remer, MD, FACEP, CCDS, president and founder of Erica Remer, MD, Inc., penned a nice review of the use of cardiac arrest codes in the acute setting for ICD-10 Monitor. Highly recommend.)

The situation described by my colleague from the frozen tundra is different. In this event, a patient found in prehospital cardiac arrest is fully resuscitated prior to arrival in the hospital, to the extent that active resuscitation is not in process while crossing the hospital doors. Could you use a Z code, specifically Z86.74, Personal history of sudden cardiac arrest, to represent the out of hospital arrest in someone who’s already resuscitated?

I think you can, but with a specific caution. Just like there’s a plain reading to the terms “alive” and “dead,” I think there’s a plain reading to the phrase “personal history of…” within the Z codes. “Personal history of…” means these conditions are in the past, and more specifically, not directly impactful to the current admission.

This is one of those times that Coding World and Clinical World collide. There is no question that a history of a particular condition may be relevant for the treating provider. If I know a patient has a history of breast cancer, and she presents with a chronic cough, you bet I’m getting the chest x-ray to be worried about a possible recurrence with pulmonary metastases. But from a coding standpoint, if the same patient gets admitted for pneumonia, does that personal history really matter? Probably not. It’s over and done with in respect to its impact on the current hospital stay.

So, the caution about using the Z code for “personal history of sudden cardiac arrest” would be the proximity of the arrest to the admission. If the patient is freshly resuscitated and brought to the hospital, I don’t think the Z code would be appropriate, as the evaluation and care of the prehospital arrest is still in process. However, during subsequent admissions (or office visits, for that matter), the Z code would be entirely appropriate for use, as the acute episode of cardiac arrest precipitating the index admission to the hospital is resolved.

(As an aside, does it strike anyone else as ironic that if a patient has a cardiac arrest in the hospital and dies, it’s not recorded as an MCC? Intuitively, you would think that death would be the ultimate MCC. But it’s probably true that the need for additional monitoring, evaluation, treatment, nursing care, or further length of stay end at that point, at least in the earthly realm. I suppose the actuaries win on this one. Just another pearl for you.)

Editor’s note: Rodenberg is the adult physician advisor for CDI at Baptist Health in Jacksonville, Florida. Contact him at howard.rodenberg@bmcjax.com or follow his personal blog at writingwithscissors.blogspot.com. Advice given is general. Readers should consult professional counsel for specific legal, ethical, clinical, or coding questions. Opinions expressed are that of the author and do not represent HCPro or ACDIS.