Guest post: CRS and HLH

by Howard Rodenberg, MD, MPH, CCDS

Sometimes writing for the ACDIS Blog feels like screaming into a pillow…you do it for your own benefit, because there’s no one to hear you. I’m very grateful that Brian Murphy, Melissa Varnavas, Linnea Archibald, and Carolyn Riel let me continue to indulge in what might best be described as a vanity product. So, when I do get feedback, it’s kind of a red-letter day.

Janie Brown of Community Health Network in Indianapolis wrote to me with her suggestion for a CDI theme song. She suggests “You Can Call Me Al,” the key lyrics of which are:

A man walks down the street

It's a street in a strange world

Maybe it's the third world

Maybe it's his first time around

Doesn't speak the language

He holds no currency

He is a foreign man

This isn’t a bad option, as it sounds like all of us the first time we saw ICD-10. My research continues, however. I found that on Lyrics.com, you can enter a term and see where it shows up in a song. For “documentation,” there were 18 options. The closest I could find that might work is a 1999 song called “Centrifuge” by an emo band called Rainer Maria:

Expert opinions

Amount to rumors,

As far as you're concerned.

 

And documentation

Of your medical condition

Won't do us right,

Won't help you sleep tonight.

It’s not exactly spot on, but close. And later in the song it mentions Rohypnol, Risperdal, and Progesterone by name, so the specificity’s there.

Unfortunately, everything comes with a price, and after the pleasant musical diversion, Janie asked for some thoughts on the relationship between cytokine storm and hemophagocytic lymphohistiocytosis (HLH). The question has special relevance today as it seems that one of the final pathways in patients with COVID-19 is cytokine release syndrome (CRS, also called a “cytokine storm”) with life-threatening pulmonary involvement and death.

Cytokine storm isn’t new. It’s a diagnosis that’s been around for a while, and CMS has finally given us a code for it within the fiscal year 2021 Inpatient Prospective Payment System (IPPS) proposed rule. Cytokine storm is a fulminant form of systemic inflammation, where activated white blood cells promoted by infection or other bodily insult such as trauma or drug toxicity release cytokines, chemicals which prompt an additional inflammatory response. The white cells involved in the continued response release more cytokines, both perpetuating and magnifying the problem. Ultimately, cytokine storm leads to multi-organ failure; specifically, with COVID-19, the hyperinflammatory state leads to infiltration of lung and heart tissue with white blood cell forms, resulting in acute respiratory distress syndrome (ARDS) and cardiac failure.

(As an historical note, cytokine storm has been theorized as a main cause of mortality during the 1918 influenza pandemic as a way of explaining the unexpectedly high death rates of the young and healthy—those who had the most active immune systems and were most at risk of a hyperinflammatory state.)

HLH is a life-threatening manifestation of an uncontrolled inflammatory response. There are inherited and acquired forms. Inherited (primary) HLH is a result of a genetic mutation that weakens the ability of the immune system to kill invaders; acquired (secondary) forms may be provoked from certain viruses (Epstein-Barr virus is the classic association), immunosuppressive drugs or conditions, and cancer chemotherapy. The basic mechanism is that as the white cells are unable to fight infection properly, there is no signal to the body that the infection is contained, and the inflammatory process should be checked. The unrelenting rush of inflammatory mediators provokes a “cytokine storm.”

The question posed by Ms. Brown and her colleagues is if cytokine storm can be used as a lead-in to query for HLH. The first issue to resolve here is to verify that HLH is not inherent to cytokine storm; if so, The Official Guidelines for Coding and Reporting would tell us that we cannot code the condition separately, so a query would add little to the record. In looking at the literature, it’s fairly clear that while cytokine storm is an end result of HLH, the latter is not inherent to the clinical definition of the former, but simply one of the conditions that can precipitate the tempest.

The best analogy I can find is to think about coding for metabolic acidosis in sepsis. With rare exceptions (perhaps septic shock), metabolic acidosis may be common finding in sepsis, but it is not inherent to the case definition whether you are a fan of Sepsis-2 or a Sepsis-3. While metabolic acidosis may be a manifestation of sepsis, neither Sepsis-2 nor Sepsis-3/sequential organ failure assessment (SOFA) scores use anion gaps, lactate measurements, or arterial pH and bicarbonate levels as part of their criteria. So, when acidosis is present, it would not be considered inherent to sepsis, and in my opinion can and should be coded separately. Similarly, the diagnosis of cytokine storm does not require the presence of HLH to establish the diagnosis.

(I would, however, have some doubt if it would work the other way. If you had HLH as a principal diagnosis, isn’t the resultant cytokine storm an inherent part of HLH? The answer to that one might be yes. It’s all about the sequencing.)

If we can ask about HLH, what are the clinical indicators we can use to justify the query? Like any clinical diagnosis, there are lots of signs and symptoms, but few are universal or definitive. Manifestations include pancytopenia, elevated liver enzymes, low albumin, and elevated inflammatory markers including erythrocyte sedimentation rate (sed rate, or ESR), c-reactive protein (CRP), and ferretin. D-dimer may be elevated and fibrinogen low, suggesting a clotting abnormality as well. Definitive diagnosis of primary HLH requires molecular identification of genetic abnormalities; secondary HLH can be confirmed through bone marrow biopsy, measures of white blood cell activity, or assessment of Interleukin-2 (IL-2), none of which are commonly found within the medical record.

Fortunately, we do have an easier way to categorize the clinical indicators that might prompt a query. An “H Score” has been developed to predict the probability of HLH in the adult patient. I don’t think this is a calculation that needs to be built into a medical record, nor anything the CDI specialist should memorize for daily use. (You can find the “H Score,” as well as a host of other useful tools, on MDCalc, a free app for your desktop or phone.) But I do think it’s probably worth keeping in the CDI toolbox for those patients with COVID-19 in the ICU, and especially when CRS or cytokine storm is mentioned in the record. As the probability of HLH rises with the H Score, the more clinical indicators and support there is for the query.

One more point: do we really need to chase down this diagnosis? You could make the argument that in these patients, if COVID-19 is the principal diagnosis and the patient has underlying HLH, surely there’s an MCC such as sepsis or respiratory failure on the chart, so the HLH diagnosis (as a CC) may have little meaning in terms of relative weight, length of stay, or reimbursement under the MS-DRG system. However, I would contend that the diagnosis is still worth an ask, as its presence may influence other parameters such as severity of illness and risk of mortality scores. Capturing this information clearly falls within the realm of CDI work.

Let’s say you’ve identified the patient at high risk of HLH, done an H Score, asked the query, and had a positive response. Now what? This is another one of those times that ICD-10 doesn’t speak clinical language, and there’s no obvious way to differentiate primary from secondary HLH. One has to look harder within the codes and translate. Assuming your encoder points you the same direction mine does with the keyword of “HLH,” I would advocate using code D76.1, Macrophage activation, to refer to primary HLH (additional notes indicate that this code encompasses “familial hemophagocytic reticulosis”). Code D76.2, Macrophage activation, due to infection (or hemophagocytic, infection-associated), would seem to be the right code for secondary HLH linked to COVID-19.

I want to thank Ms. Brown for her note, and hope others will continue to write. Responding to questions helps me learn things as well, and as the Beloved Dental Empress reminds me on a daily basis, I’ve got a LOT to learn. And yet she keeps me around. Go figure.

Happy Quarantine!

Selected references:

• Fardet L, Galicier L, Lambotte O, et al. Development and Validation of the H Score, a Score for the Diagnosis of Reactive Hemophagocytic Syndrome. Arthritis and Rheumatology 2014; 66(9):2613-30.
• Mehta P, McAuley DF, Brown M, et al. COVID-19: Consider Cytokine Storm Syndromes and Immunosuppression. Lancet 2020; (395):1033-34.

 

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.