2024 Q1 Coding Clinic Reinforces “As Many Codes as it Takes” Notion

I am overdue to give my comments on the 2024 American Hospital Association (AHA) Coding Clinic published for the first quarter. I really appreciate this Coding Clinic, because it gives reminders of general coding rules.

One of the things I have always loved about ICD-10-CM reflects my mantra: “as many codes as it takes.”

• The revisions to the ICD-10-CM Coding Guidelines include the following:
  • The additions of “other surgical site” for post-procedural and obstetrical surgery in the context of sepsis. This means that there should be at least four codes for sepsis in the setting of a procedure:
    • A code for the site of infection, specifying the area involved, like superficial or deep incisional surgical site;
    • The code to specify that there is post-procedural sepsis, T81.44 or O86.04;
    •  A code to identify the infectious agent, if possible;
    • R65.2- for severe sepsis, because all sepsis is now “the condition formerly known as severe sepsis;” and
    • And at least one code to detail what the sepsis-related organ dysfunction was.
• The answer to a question about how to code “acute prostatitis and cystitis” reminds us that you need both the code for prostatocystitis, which establishes the site, and N41.0, Acute prostatitis, to establish the acuity.
• On page 20, the guidance explains that a postprocedural intra-abdominal abscess would take 2 codes – T81.43XA, Infection following a procedure, organ and space surgical site, initial encounter, and K65.1, Peritoneal abscess – to fully flesh it out.

There were several recent questions that arose regarding what “other” or “other specified” codes are meant for conditions for which the provider gives details and often linkage, but with no specific code for the etiology. For instance, chorioamniotic separation goes to O41.8X30, Other specified disorders of amniotic fluid and membranes, third trimester, not applicable or unspecified; and E27.49, Other adrenocortical insufficiency, is one of the codes needed to capture the condition of hypothalamic pituitary adrenal axis insufficiency.

A question on page 17 regarding the coding of the verbiage “rheumatoid arthritis with inflammatory polyarthropathy” points out to me that you need to take into consideration where in the classification a code appears. M06.4, Inflammatory polyarthropathy, lives under the umbrella of M06, titled Other rheumatoid arthritis. It probably would have been better if it were titled Inflammatory rheumatoid polyarthropathy. However, I (not Coding Clinic) would also suggest that if the provider had specified “rheumatoid factor positive RA with inflammatory polyarthropathy,” the more accurate code would be M05.79, Rheumatoid arthritis with rheumatoid factor of multiple sites without organ or systems involvement.

There was a set of questions regarding dural tears that I am sure is driving some of you crazy. My experience is that some quality and clinical documentation integrity (CDI) teams try to get their providers to perform contortions to get out of triggering patient safety indicators and complications. They think that the magic words to preclude a complication from being considered a complication are “inherent to” or “integral to.” However, on pages 20 and 21, Coding Clinic is pointing out that regardless of whether the circumstances would have led to a dural tear for anyone performing that procedure, or whether there is thinning, scarring, adhesions, or stenosis, the condition is clinically significant and should be documented and captured as an accidental puncture or laceration. I agree with their assessment.

Pages 23-24 posed an interesting question. A patient presented to the ED with an exacerbation of asthma and did not have access to their albuterol inhaler. Coding Clinic explained that you don’t use an underdosing code for as-needed medications. The code they recommend is Z91.198, Patient’s noncompliance with other medical treatment and regimen for other reason. It makes me think that I might have been using the wrong code for underdosing of antipyretics (to reduce fever). It is still clinically significant and should be recorded (it might be the reason that mom dragged the baby out in the middle of the night – giving too low a dose of acetaminophen), but since it isn’t a long-term drug or prescribed course, like 10 days of antibiotics, it isn’t a “medication regimen.”

There was also a question about transaminitis and hyperbilirubinemia being documented as “acute liver injury due to metastatic liver disease and chemotherapy.” The indexing of “acute liver injury” goes to a trauma code in S36. The coder recognized that this is not correct. Coding Clinic advises using K71.8, Toxic liver disease with other disorders of liver and then codes for the metastasis and adverse effect of the antineoplastic drugs. Since the etiology is known and there is no distinct combination code, K71.8 is indicated, not K71.9, Toxic liver disease, unspecified. The next question also tackles the use of a trauma code – if there is no trauma, you shouldn’t use a trauma code (an S-T code). Medical intervention misadventures are not considered “trauma.”

The next one is a head-scratcher to me. On page 28, a patient with coronary artery disease and a bypass graft (CABG), presents to the ED with chest pain over three days and is diagnosed with a non-ST-segment elevation myocardial infarction (NSTEMI), likely from the stenosis of a vein graft. They recommend I24.4, NSTEMI, as the principal diagnosis and I25.10, Atherosclerotic heart disease of native coronary artery without angina pectoris, as the codes. Their reasoning is that “it would be inappropriate to assign a code for angina in the setting of an MI.”

Am I to infer that if the patient hadn’t suffered an MI, you would have used the code indicating “of CAD graft” with “unstable angina pectoris?” What if the patient had chronic stable angina prior to the MI? Are you not allowed to include that information in the coding? I think what they were trying to convey is that the NSTEMI was the progression of the atherosclerosis of autologous vein coronary artery bypass graft with unstable angina, and that once you had the MI, it superseded the CAD in that distribution. In addition, the patient was found to have other CAD, which was not felt to be the etiology of the chest pain.

I know that coding is based on documentation. We want the codes to reflect the conditions the patient has, so if the coding based on documentation doesn’t tell the story accurately, either the documentation or the coding must be adjusted. And…with as many codes as it takes.

As always, I recommend you read the Coding Clinic in its entirety on your own.

Programming Note: Listen live to Dr. Erica Remer as she reports this story during today’s Talk Ten Tuesdays broadcast at 10 a.m. EST.