The issue is with the terminology and the coding implications.

Medscape recently highlighted an article from Neurocritical Care, published March 16, by Frontera, Melmed, Fang, et al., titled Toxic-Metabolic Encephalopathy in Hospitalized Patients with COVID-19. When I read the article, alarm bells started blaring in my brain. I am writing this article so that if you start seeing your providers documenting hypoxic-ischemic encephalopathy (HIE) in your patients, you can forward it to them and shut down the practice immediately.

The article presented the fact that many patients with COVID-19 experience encephalopathy. Their retrospective, multi-center, observational study showed that nearly one in eight patients had encephalopathy not attributable to the effects of sedative medications. Encephalopathy in COVID-19 imparted an increased risk of mortality of 24 percent. It was also associated with a longer length of stay and an increased likelihood of not being discharged to home.

There is no dispute with the substance of the article. The issue is with the terminology and the coding implications.

The National Institute of Neurological Disorders and Strokes (NINDS) defines encephalopathy as a term for any diffuse disease of the brain that alters function or structure. It has myriad causes, including infection, metabolic or mitochondrial dysfunction, toxins, trauma, poor nutrition, hypoxia, or hypoperfusion of the brain. The hallmark is altered mental status, either in level of consciousness or impaired cognition.

As a review, there are different types of encephalopathy:

  • Metabolic encephalopathy is the most common pathway. Encephalopathy from sepsis is categorized as metabolic. It is not caused by an infectious agent directly compromising brain tissue – that would be considered an encephalitis.

Also landing in this bucket are electrolyte disturbances, hypoglycemia, hypoxia, and mitochondrial dysfunction. Hypoperfusion and increased intracranial pressure are also metabolic-based.

Metabolic encephalopathy is a major comorbid condition or complication (MCC).

  • Toxic encephalopathy and toxic-metabolic encephalopathy (for the clinically indecisive) include exposure to drugs, such as overdoses; toxins, such as heavy metals; solvents, chemicals, and uremia. Most of the time, another code is necessary to identify the culprit medication or toxic agent.

If a medication is appropriately dosed and the intention is depressed level of consciousness, that would not be considered toxic encephalopathy. In that case, the alteration of consciousness is integral to the medication administration.

Toxic encephalopathy also risk-adjusts as an MCC.

  • There are specific types, such as hypertensive, hepatic, or Wernicke’s encephalopathies. There is no rhyme or reason as to why those are only comorbid conditions or complications (CCs).
  • Unspecified and other encephalopathy were downgraded to CCs a few years ago.

As usual, the crux of the matter is in the details. The article uses two terms that must be discussed further.

  • They have bundled all encephalopathy into toxic-metabolic encephalopathy (TME). The code used for “toxic-metabolic encephalopathy” is G92, Toxic encephalopathy. There is an instruction to code first, if applicable, the drug or toxin that is inducing the encephalopathy.

The prevalence of different etiologies of TME is found in Figure 2. A total of 28 percent of the 559 patients with encephalopathy in this study had uremia listed as the etiology (uremia is often referred to by providers as causing a TME), and 6.2 percent had formal toxic encephalopathy (supratherapeutic, intoxication, withdrawal, poisoning). The vast majority of patients (61.8 percent) had metabolic encephalopathy with etiologies such as sepsis, hypoxia, hypo/hypernatremia, fever, acidosis, hyper/hypoglycemia, and hypothermia.

  • The really problematic term was hypoxic-ischemic encephalopathy (HIE), which the authors defined as a global cerebral insult due to oxygen deprivation to the brain or lack of perfusion to the brain caused by systemic hypoxemia, hypotension, or cardiac arrest. This does accurately describe the mechanism of injury. The issue is that there is a specific neonatal entity known by this name and referred to as HIE. The ICD-10-CM code for HIE is P91.6-, subdivided by severity. Perinatal codes are to be used in the first 28 days of life. A condition that arises in adulthood is never coded with a P code. Although “hypoxic-ischemic” describes the mechanism of how the encephalopathy arises in these patients, it will result in a nightmare dilemma for the coders.

This is not just semantics. COVID-19 patients land in the Respiratory Infections and Inflammations MS-DRG. Toxic or metabolic encephalopathy serves as an MCC and elevates the encounter to the highest tier. This article lends credence to these patients belonging in the highest tier, demonstrating an increased consumption of resources by having a longer length of stay and an increased risk of in-hospital mortality. If providers read this article and adopt the terminology, coders are going to have to query repeatedly to get codable verbiage. HIE will not be acceptable in this patient population, and the claims will be rejected before they go anywhere. Coders will be forced to query incessantly.

Furthermore, if the TME verbiage predominates, payers may start rejecting the diagnosis because there is no toxin/medication code coded first. There will be an epidemic of payer denials trying to remove the (possible sole) risk-adjusting, tier-elevating MCC from the diagnosis list.

If your providers read the Neurocritical Care article and jump on the TME/HIE bandwagon, here is my advice. Have them read this article so they understand the implications. Then, let’s give them verbiage that makes sense to them.

  • If the final pathway to the encephalopathy is metabolic, use the term “metabolic encephalopathy.” Conditions that lead to metabolic encephalopathy are decreased perfusion, hypoxia, electrolyte or glucose disturbances, and sepsis. Many conditions take this common pathway.
  • If toxins or medications are a factor, document which medication or toxin is causative and whether it is an excess, underdosing, or just an undesirable adverse effect of the compound. If there is no toxin or medication, reconsider whether TME is the correct terminology. It is probably metabolic encephalopathy instead.
  • I recommend that providers who want to use the expression “hypoxic-ischemic encephalopathy” in an adult COVID-19 patient resist the urge and document “metabolic encephalopathy due to hypoxia” (or ischemia, whichever is your intent).
  • If your electronic medical record has the functionality to do this, design an acronym expansion that takes “HIE” and “hypoxic-ischemic encephalopathy” to “metabolic encephalopathy due to hypoxia and/or ischemia” in adult patients. If the patient is 28 days old or younger, documented “HIE” should go to P91.6-. Patients who have perinatal conditions that persist beyond the neonatal period may be coded with the perinatal code regardless of age, so if HIE is documented in the pediatric population beyond 28 days, the coder/CDI professional should assess for clinical validity. If HIE causes cerebral palsy (CP), the HIE often falls off the clinician’s list, and the specified CP is documented and captured instead.

I am completely supportive of the conclusion that encephalopathy is a big problem in hospitalized COVID-19 patients, and documenting that encephalopathy specifically and accurately is crucial. I just hope this Neurocritical Care article won’t cause a coding-clinical disconnect.


Programming Note: Listen to Dr. Erica Remer every Tuesday when she co-hosts Talk Ten Tuesdays with Chuck Buck, 10 a.m. Eastern.

Share This Article