By Robert S. Gold, MD
I have been musing recently about things I’ve written for this journal over the past years. Hard to believe I’ve been doing monthly educational articles regarding the clinical aspects of coding since about 2002.
I know that a lot of my pieces had relatively universal appeal. Some had been considered outrageous and seemingly destructive, depending on the view of the reader. But time has proven they were right then, and they are still right in ICD-10. Return with us now to those thrilling days of yesteryear–the Lone Ranger rides again.
SIRS
Back in 2002, I objected that the codes for "sepsis" and "septicemia" were the same codes–and they were all described as "septicemia."
In the article, I emphasized that sepsis was a condition that resulted when the body suffered the consequences of a localized infection mediated by chemicals that were released into the bloodstream, but that the infection was a local event. I pointed out that septicemia was an infection of the bloodstream itself–and that both could exist simultaneously, but they were different animals and needed different diagnosis codes to describe them.
Then, in 2003, it was finally published, after adoption of the 995.9x series of codes, that septicemia and sepsis were different (AHA’s Coding Clinic, Fourth Quarter 2003, pp. 79-81) and that the new codes would identify that distinction.
Okay, that being somewhat resolved, I pursued the issue that SIRS plus infection is NOT sepsis in 2009. Why? Because the combination of codes for SIRS plus infection, without or with organ failure, led to a massive proliferation of coding "sepsis" cases when there was no sepsis–and often when there was no SIRS. Nothing changed in the rules or definitions.
I brought along some of the world’s most renowned specialists in infectious disease and critical care who had supported for years (since 2001, actually) that SIRS plus infection is NOT pathognomonic of sepsis. Nothing happened.
Finally, with the coming of ICD-10-CM, the equivalent of 995.91 (SIRS plus infection without organ failure) disappeared. "Sepsis" is now "sepsis." You need the word "sepsis" to code "sepsis." All is right with the world, right?
Nope. Not a chance. Now, instead of all of the codes being "septicemia" codes, they are all "sepsis" codes–and the only "septicemia" code we have is for plague. (There’s actually one for meningococcemia, which is infection of the bloodstream with the meningococcal organism.)
So we have no other codes for septicemia when all of the codes had been for reporting septicemia up until now. If you look up "septicemia," you get A41.9 (sepsis, unspecified organism), equivalent to the 038.9 (unspecified septicemia) of ICD-9-CM.
So sepsis is septicemia again–after all of our work to distinguish that the two are different. And though septicemia is defined as infection of the bloodstream, we have no codes for bloodstream infection in ICD-10-CM except catheter-related bloodstream infection (T80.211-). And infectious disease physicians are calling these "bacteremia," so there’s no chance of determining what the patient has through analysis of diagnosis codes.
One step forward, two steps back.
Syncope
In November 2002, I wrote an article on syncope. Here I spoke of the myriad of conditions that could led to the symptom of passing out. (Remember, syncope means that the patient actually passed out. When we see "near syncope" written and try to code it, the encoder sees the word "syncope" and assigns that code, recognizing that "near" and "pre-" are nonessential modifiers, so the patient didn’t actually have to pass out.)
I talked about neurogenic syncope causes and cardiogenic causes. I spoke of volume changes (hypovolemia) and autonomic nerve dysfunction and arrhythmias such as bradycardia. In ICD-9-CM, everything went to 780.2 (syncope and collapse) without additional specifics being provided by the physician and the "due to" cause of the syncope if a cause could be found.
Yes, there were syncopes due to lumbar puncture and complicating delivery and such. The arrhythmia codes were arrhythmia codes, and the syncope part disappeared.
Well, in ICD-10-CM, all of the syncopes are now R55 (syncope and collapse). Whether it was a cardiogenic or neurogenic cause or it was attributed specifically to an arrhythmia or to heat or a coughing episode (which is really neurogenic, but it has a code of its own at R05 [cough]), there’s no improvement without the physician getting involved. And the doctor must identify the cause of the syncope and make the link so that the other diagnosis would be the principal diagnosis, not the syncope.
In this article, I noted that the term "orthostatic hypotension" was usually a symptom when provided by the physician and not a diagnosis, but was assigned 458.0 (orthostatic hypotension). That’s when the patient’s blood pressure drops with change in opposition from lying to sitting or standing and causes decreased blood supply to the brain, leading to the patient becoming dizzy or passing out.
It should be called orthostatic changes in vital signs, but the docs and nurses call it orthostatic hypotension. It’s a symptom, and it’s always due to something or other. In ICD-10-CM, we again have a breakdown of codes with I95.0 (idiopathic hypotension) and I95.1 (orthostatic hypotension), which is the equivalent of 458.0 and the code for the symptom complex of orthostatic changes in vital signs.
We also have I95.2 (orthostatic hypotension due to drugs), as often happens with patients on beta blockers; I95.3 (hypotension of hemodialysis); and I95.89 (other specified cause of orthostatic blood pressure changes).
But it’s still not a diagnosis. We still have no better idea about the pathophysiology of the patient’s syncopal episode, with or without measured hypotension. We need this from the doc. That’s the conclusion I wrote in 2002:
Interactions between or among drugs can cause instability of the arteries and veins. The physician might have to change the patient’s beta blocker dosage or switch to a non-beta-blocker drug for treatment of the patient’s hypertension. A patient might have intrinsic autonomic nerve dysfunction, where the arteries and veins cannot maintain the pressures that they normally exert on the column of blood in them, or they can’t respond quickly enough to changes in position. In all of these, the patient stands up and falls down. Immediate testing of pulse rate and blood pressure on position change demonstrates "orthostatic" changes.
Whether it’s dehydration, autonomic dysfunction of diabetes, sick sinus syndrome, or aortic stenosis, the coder must recognize that most of the time the physician knows the cause of the syncope and makes some effort to document that cause. Most of the time, however, the cause of the syncope is not clear. Keep this column nearby and refer to it when you see syncope or orthostatic hypotension documented. If you see one of these causes, a clinically oriented query couldn’t hurt.
Hypertension
I10 is the ICD-10-CM code for hypertension, whether benign or malignant. Too easy, right? I objected, as I seem to do a lot, and wrote my objections and got together with some of the premier physicians in nephrology and hypertension.
We agreed that malignant hypertension kills patients and we must have a code set to demonstrate this potentially lethal condition. Word got to the code gurus in the Coordination and Maintenance Committee, and we all must be ready for this change, though it doesn’t exist–yet.
I’m sure you have all seen documentation of such things as hypertensive emergency, hypertensive urgency, and hypertensive crisis, right? Well, this is a situation where acute onset of exceptionally high blood pressure levels can cause target organ damage, such as hypertensive encephalopathy or hypertensive stroke or seizure, acute renal failure, or acute pulmonary edema.
Blood pressures in the range of 220/110 or higher (it can be lower in children) must be treated quickly to avoid death of the patient. Existence of this situation is called a hypertensive crisis. It’s not slight elevations in blood pressure that happen when you run a half block–it’s real, serious stuff.
If there is target organ damage, as above, it should be referred to as hypertensive emergency. If it is identified that no target organ damage has occurred, the incident was a situation of hypertensive urgency. In future updates, there will be an I16 code for hypertensive crisis that requires specificity with the fifth character to distinguish between hypertensive emergency and hypertensive urgency. They got this almost right.
Instructions will be there to determine if the patient’s hypertensive crisis was associated with endocrine-induced hypertension or renal artery stenosis or other secondary cause of hypertension, or if it was essential hypertension. But the instructions are not there (yet) to "code also" the target organ damage that justifies coding the hypertensive emergency. Maybe one day they will be.
Editor’s note: Dr. Gold is CEO of DCBA, Inc., a consulting firm in Atlanta that provides physician-to-physician CDI programs, including needs for ICD-10. Contact him at 770-216-9691 or[email protected].
If you have a specific procedure or condition you would like Dr. Gold to address in his column, contact Editor Steven Andrews at
[email protected].
This article was originally published in the November issue of Briefings on Coding Compliance Strategies.
HCPro.com – JustCoding News: Inpatient
