Once upon a midnight dreary, as I labored on a query
As I nodded, nearly napping (that darn G47.411 again)
Suddenly there came a tapping
As of someone gently rapping, rapping at my cubicle wall
Tis just my manager, I muttered, coming for the query
Only this and nothing more
The silken, sad, muffled rustling of each chart that I sat shuffling
Thrilled me—filled me with fantastic terrors never felt before (oh wait, that’s just F41.1);
So that now, to still the beating of my heart, I stood repeating,
“Tis just my manager coming for the query
Just my manager looking for the query”
This it is and nothing more.
Presently my soul grew stronger; hesitating then no longer,
“Jill,” said I, “or Shannon, truly your forgiveness I implore;
But you know my lack of hearing (H90.0) caused by this incessant ringing (H93.13)
Left me thinking there was no one tapping at my cubicle door,
I scarce was sure I heard you”—here I opened wide the door—
Darkness there and nothing more.
Deep into that darkness peering, long I stood there wondering, fearing (more F41.1),
Doubting, dreaming dreams no mortal ever dared to dream before;
But the silence was unbroken (or am I just suffering H91.23?), and the stillness gave no token,
Until I heard this one word spoken—ICD-10
This I whispered and an echo murmured back—ICD-10
Merely this and nothing more.
Back into the chamber turning, all my soul within me burning (maybe it’s really R12 and not my soul),
Soon again I heard a tapping somewhat louder than before.
Open here I flung the door, to see a Raven of the days of yore;
Perched above the neighboring cubicle door
Perched contently above my next-door neighbor’s cubicle door
Then he spoke—ICD-10—and nothing more.
Wonderful, I muttered, now I’m seeing birds, I shuttered
I’ve started hallucinating a creepy bird of yore (R44.1, too bad it doesn’t specify what I’m seeing)
I really want to write this query
So I can code this record, I am weary (R53.83)
Weary of worrying when ICD-10 will be implemented
Quoth the raven, 2015
This he said and nothing more.
Listen bird, I said with feeling, please don’t let me be dreaming
Please tell me that ICD-10 is really coming soon
Currently our codes are lacking
We can’t even tell what’s catching
We don’t know what diseases we’re not tracking
Quoth the raven, 2015
This he said and nothing more.
“Be that word our sign of parting, bird my friend!” I grinned, upstarting—
“Get thee back into the tempest and spread the word to every coder!
Leave one black plume as a token of that joy thy beak hath spoken!
Leave my hopefulness unbroken!—quit the perch above my neighbors door!
Take thy beak from out my heart (S26.19, W61.99XA), and take thy form from off my neighbor’s door!”
Quoth the raven, 2015
This he said and nothing more.
Ewww, I don’t think so. Your feet stink. What could be causing that foul odor to emanate from your feet?
Sweat by itself is odorless, but when it comes into contact with bacteria it begins to smell bad. Usually, the bacteria are harmless, but some, such as necrotizing fasciitis, can eat through the skin.
Maybe you have necrotizing fasciitis. How would we code that in ICD-10-CM?
If you look up fasciitis, necrotizing, in the ICD-10-CM Alphabetic Index, you are directed to code M72.6. Pretty easy at first glance. We only have one code, so we don’t need to know what area the necrotizing fasciitis affected.
We do, however, need to use additional code (B95.-, B96.-) to identify causative organism. Is the necrotizing fasciitis caused by a form of Streptococcus, Staphylococcus, and Enterococcus or by some other bacterial agent, such as E. coli or Klebsiella pneumonia?
Categories B95 and B96 do include unspecified options in case the physician doesn’t know whether the patient’s necrotizing fasciitis is caused by group A strep or group B strep. And you can use those unspecified codes. They are completely legitimate. The ICD-10-CM Official Guidelines for Coding and Reporting even tell us that performing a medically unnecessary lab test just to arrive at a more specific code is inappropriate. (It’s guideline I.B.18 in case you’re interested.)
Perhaps you have gangrene instead of necrotizing fasciitis. Gangrene reeks. The ICD-10-CM codes for gangrene are more specific than those for necrotizing fasciitis. We need to know where the gangrene is before we can code it.
ICD-10-CM does not include a separate listing for gangrene of the feet. It falls under gangrene of the upper and lower extremities, which leads to I96 (gangrene, not elsewhere classified).
Before you report I96, make sure you review the Excludes1 note. Excludes1 in ICD-10-CM means the two conditions are mutually exclusive. A patient can’t have both at the same time. Under I96, the Excludes1 note specifies:
- Gangrene in atherosclerosis of native arteries of the extremities (I70.26)
- Gangrene in diabetes mellitus (E08-E13)
- Gangrene in hernia (K40.1, K40.4, K41.1, K41.4, K42.1, K43.1-, K44.1, K45.1, K46.1)
- Gangrene in other peripheral vascular diseases (I73.-)
- Gangrene of certain specified sites – see Alphabetical Index
- Gas gangrene (A48.0)
- Pyoderma gangrenosum (L88)
If a patient suffers from atherosclerosis of native arteries of the extremities or diabetes mellitus, you may need to query the provider to make sure the two conditions aren’t causally related. In ICD-9-CM, we can assume a relationship between diabetes mellitus and gangrene if the physician does not document any other cause of the gangrene.
Coders cannot report the combination codes for diabetes and a complication, such as gangrene, unless the physician specifically documents the relationship between the diabetes and the condition.
Maybe the cause of your foul feet is something a little less life-threatening. You could just suffer from hyperhidrosis (aka sweaty feet).
The default code in ICD-10-CM is R61 (generalized hyperhidrosis). However, again we find some important Exludes1 notes:
- Focal (primary) (secondary) hyperhidrosis (L74.5-)
- Frey’s syndrome (L74.52)
- Localized (primary) (secondary) hyperhidrosis (L74.5-)
These conditions are all more specific than the generalized hyperhidrosis.
When excessive sweating affects the hands, feet, and armpits, it is called primary or focal (local) hyperhidrosis. Since we know the hyperhidrosis is of the feet, we would look to L74.5-.
The only code in this subcategory that applies is L74.513 (primary focal hyperhidrosis, soles). Again, the physician will need to documents that specific information in the medical record or you may need to query.
In the meantime, might I suggest some extra strength Odor Eaters?
The AHA publishes Coding Clinic on a quarterly basis and addresses questions submitted to Coding Clinic by stakeholders in the healthcare industry.
Coders, as well as payers, use Coding Clinic advice to determine accurate code assignment.
However, Coding Clinic does not provide clinical criteria for establishing diagnoses and has no authority to provide clinical definitions, nor does it replace physician documentation regarding the clinical significance of a patient’s condition.
Coding Clinic does provide guidance on what codes can be reported based on already-documented diagnoses.
Coding Clinic information may still be useful to understand clinical clues regarding signs or symptoms that may be integral (or not) to a condition.
The Coding Clinic editorial staff is not updating past guidance for ICD-9-CM to ICD-10, just like CMS is not transitioning all National Coverage Determinations from ICD-9-CM to ICD-10.
Coding Clinic staff members are only answering ICD-10 questions going forward, but the editors can only answer question that you ask. So if you are practicing (or learning) ICD-10 and something seems wrong or you’re confused, submit a question to Coding Clinic.
They don’t guarantee they will answer your question, but odds are, someone else has the same question. The more questions we raise before implementation, the more information we will have when we start assigning ICD-10 codes for real. Remember, though, that the Coding Clinic editors are new to ICD-10 as well. Some of the answers they publish now may need to be updated as more information becomes available and coders actually use the new codes.
Poor Mr. Frank N. Stein. He just coughed up a lung. Literally.
First, what diagnosis code would we report for Frank? Well, he’s here for a transplant, so lung transplant would be his principal diagnosis, right? No.
The lung transplant is being used to treat a condition, such as COPD or cystic fibrosis. Remember that the principal diagnosis is the condition established after study to be chiefly responsible for occasioning the admission of the patient to the hospital for care. Frank wound up in the hospital because he has some disease or condition of the lung that resulted in him coughing one up.
He could have bronchitis, pneumonia, COPD, or perhaps necrotic lung tissue. You’ll need to read the physician’s documentation to find out the underlying condition that caused Frank’s admission. You’ll also need to make sure the diagnosis supports the medical necessity of the lung transplant.
Let’s take Frank to the operating room and see what Dr. Shelley is going to do to fix this lung problem. Dr. Shelley documents that she implanted a new left lung using a donated lung from a cadaver.
Our first character will be 0 (zero) for the Medical and Surgical section. Our second character denotes the body system—in Frank’s case the respiratory system. Our second character is B.
The third character is the root operation. What is the objective of the procedure? Dr. Shelley documents “implantation,” which is not an ICD-10-PCS root operation. However, we do have root operation Transplantation.
The official ICD-10-PCS definition is “putting in or on all or a portion of a living body part taken from another individual or animal to physically take the place and/or function of all or a portion of a similar body part.” That sounds like the procedure Dr. Shelley is performing, so we’ll going with Transplantation (third character Y) as our root operation.
Remember, it’s up to us to determine the root operation. We need to apply the root operation definitions to find the one that matches the intent of the procedure. Just because a physician calls a procedure an Insertion doesn’t mean that’s the root operation we’re going to use.
We have enough information to head to an ICD-10-PCS table: 0BY.
The table gives us nine choices for a specific body part. Wait, we only have two lungs (two codes) and possibly a bilateral option. That’s three choices tops. How did we end up with nine?
Think back to your anatomy and physiology. The lungs are made up of lobes—two on the left and three on the right. Each lobe of each lung has its own body part value in ICD-10-PCS. So our body part choices are:
- C, upper lung lobe, right
- D, middle lung lobe, right
- F, lower lung lobe, right
- G, upper lung lobe, left
- H, lung lingula
- J, lower lung lobe, left
- K, lung, right
- L, lung, left
- M, lungs, bilateral
We know Dr. Shelley is transplanting the entire left lung, so our fourth character will be L.
Our next two characters are pretty easy because we only have one choice for each. The approach is always open and no device is ever used. Quick aside, device has a very specific meaning in ICD-10-PCS. You only code a device when it remains in the patient after the procedure is completed. So stents are devices, but sutures and temporary wound drains are not.
That brings us to the seventh character: the qualifier. For transplants, the qualifier denotes the type of tissue used in the transplant. For lung transplants, the tissue can be:
- 0, allogeneic (non-identical donor of the same species)
- 1, syngeneic (an identical member of the same species, i.e., an identical twin)
- 2, zooplastic (from a different species)
Dr. Shelley documented that the donor organ came from a cadaver and did not indicate that the donor was Frank’s identical twin brother, so our seventh character is 0.
Our final code is 0BYL0Z0 (Transplantation of left lung, allogeneic, open approach).
That code contains quite a bit of information: the organ, the laterality, the procedure, how it was done, where the donor organ came from.
In ICD-9-CM Volume 3, we need two codes to get all of that information: one code for the lung transplant (33.51, unilateral lung transplant) and one for the donor source (00.93, transplant from cadaver).
One thing we don’t have in ICD-9-CM is the laterality. We don’t know which lung was involved. Why would that matter? Well, think about Frank’s future medical care. Perhaps he develops lung cancer. You’ll be able to tell whether it’s the transplanted lung or his own lung involved.
Maybe Frank will need another lung transplant down the road (he’s a pretty unlucky guy). The physician will be able to tell whether he or she is replacing an original lung or a lung Frank picked up. That could provide information about the progression of the diseased that caused Frank’s need for a new lung in the first place. It could also provide information about the lifespan not just of the transplant recipient, but of the transplanted tissue itself.
And all of that could (and hopefully will) lead to better treatment.
If I hear one more person poke fun at ICD-10-CM code V97.33XD (sucked into a jet engine, subsequent encounter), I am going to develop a very strong case of R45.850. (That’s homicidal ideation in case you don’t have your code book handy.)
First of all, most of the people making fun of this code don’t actually understand what the code is conveying. See the New York Times, an Alabama physicians group, Healthcare Dive, The Boston Globe, and on and on and on.
The subsequent encounter part is not saying the person was sucked into a jet engine twice (what are the odds of that?). It’s telling us that the person is being seen for a subsequent encounter for injuries suffered when he or she was sucked into the jet engine. (And you can indeed survive being sucked into a jet engine as long as you are not on Lost.)
The seventh character is one of the main new concepts in ICD-10-CM. Maybe we need to do a better job of explaining what it means.
In most cases the seventh character indicates the episode of care. If the patient is receiving active treatment, you use seventh character A in most cases.
If the patient is being seen for routine follow up, the seventh character becomes D, again in most cases.
When the patient develops a complication or a condition that arises as a direct result of a condition, that’s a sequela reported with seventh character S (always).
Fracture codes have some additional seventh characters for nonunions, malunions, delayed healing, and open fractures. Most injury codes only give you three choices: A, D, and S.
Do the physicians at your organization know what the seventh character actually means? If not, here’s a perfect example you can use to explain it. V97.33XD doesn’t mean sucked into a jet engine twice. It means the patient is actually recovering from injuries sustained by his or her sole encounter with a jet engine.
The second reason this example drives me up a wall is because V97.33XD is an external cause code. It will never, ever, ever be the principal diagnosis. And you will almost never hold a claim for an external cause code. The only time you would is if a specific payer requires them. The ICD-10-CM Official Guidelines for Coding and Reporting even state that there is no national requirement for reporting external cause codes. If you don’t report them now, you won’t report them in ICD-10-CM.
So why are we spending time talking about whether the patient was bitten by a chicken or a duck? It’s not going to change your payment. It’s probably not going to affect patient care. Although, if you are bitten by a wild animal, the physician may be more likely to worry that you’ll contract rabies.
Unfortunately too many physicians (and Congress members) rely on the mainstream uninformed media for information about ICD-10. We need to focus the discussion on what actually matters, not on what makes the best punchline.
For example, ICD-9-CM has no code for Ebola. You may have heard of Ebola and, depending on where you are in the U.S., you may indeed be coding for Ebola. Outside of the running tally in the mainstream media, how can you tell how many people in the U.S. have Ebola? Right now, you can’t. In ICD-10-CM you will be able to track that number.
If a patient breaks his leg, which leg is involved? In ICD-9-CM, we have no idea. In ICD-10-CM, we will not only know which leg is broken, but we’ll also be able to follow that patient’s progress with the seventh character. We’ll be able to see how many times the patient came back for active treatment, how often he was in for routine follow up, and whether he had any late effects from the break. The main code stays the same. Only the seventh character changes.
We have less than a year until ICD-10 implementation. We need to take the initiative to get the right information out about the benefits of ICD-10. Don’t be shy. Share the real story.
You’ve probably heard that you need to beef up your clinical knowledge for ICD-10-CM and ICD-10-PCS coding. And you’re probably wondering when you have time to do that.
We’re here to help. We’ve researched 140 clinical conditions and collected them in JustCoding’s Clinical Conditions Encyclopedia. Each entry includes clinical information, such as signs and symptoms, types or stages, and treatment, along with coding and documentation for both ICD-9-CM and ICD-10-CM.
Mad cow disease, more properly called Creutzfeldt-Jakob disease (CJD), is one of the conditions the encyclopedia covers. CJD is divided into three major categories:
- Sporadic CJD
- Hereditary CJD
- Acquired CJD
On October 1, 2008, the Cooperating Parties revised and expanded the ICD-9-CM codes for CJD to include:
- Variant Creutzfeldt-Jakob disease: Report code 046.11
- Other and unspecified Creutzfeldt-Jakob disease: Report code 046.19
- Dementia in conditions classified elsewhere without behavioral disturbance: Report code 294.10
- Dementia in conditions classified elsewhere with behavioral disturbance: Report code 294.11
The revised code description for code 046 includes slow virus infections and prion diseases of the central nervous system. The fifth digit represents the form of the disease. Currently, only variant CJD has a fifth digit of a 1. Report code 046.19 for all other forms.
When a patient develops dementia with or without behavioral disturbance, report an additional code to represent the dementia.
When we get to ICD-10-CM, our coding will look like this:
- Creutzfeldt-Jakob disease, unspecified, or Jakob-Creutzfeldt disease, unspecified: A81.00
- Variant Creutzfeldt-Jakob disease: A81.01
- Other Creutzfeldt-Jakob disease including familial Creutzfeldt-Jakob disease, iatrogenic Creutzfeldt-Jakob disease, sporadic Creutzfeldt-Jakob disease, and subacute spongiform encephalopathy (with dementia): A81.09
Category A81 (atypical virus infections of central nervous system) still includes diseases of the central nervous system caused by prions.
And we will still use an additional code to identify:
- Dementia with behavioral disturbance (F02.81)
- Dementia without behavioral disturbance (F02.80)
While you’re unlikely to see a large number CJD diagnoses, you probably code for patients with anemia, hypertension, heart failure, migraine, stroke, and syncope. All of those conditions are part of the encyclopedia as well.
You’ll also find some less common conditions such as hungry bone syndrome, blue diaper syndrome, and kabuki syndrome.
Order your copy of JustCoding’s Clinical Conditions Encyclopedia today!
Jeff, an 18-year-old male, came into the Fix ‘Em Up Clinic with complete left oculomotor palsy. Jeff stated he had a severe, throbbing headache around his eye yesterday. He took some aspirin and went to sleep.
When he woke up this morning, he had troubled seeing out of his left eye because his left eyelid was drooping so much. He says overall his vision is blurry and the area around his left eye is very painful.
Fortunately for Jeff, Dr. Achy is on the case. Dr. Achy documented that Jeff’s left pupil was 6 mm in diameter and reactive to light. She performed a neurological exam and ordered x-rays, both of which were normal.
To be on the safe side, Dr. Achy ordered an MRI and magnetic resonance angiography to rule out really bad things such as tumors and aneurysms. Again, the tests came back normal. Dr. Achy also ruled out glaucoma.
Dr. Achy prescribed two aspirin and Jeff’s symptoms almost completely resolved. It’s a miracle, but what on earth is wrong with Jeff?
He has an ophthalmoplegic migraine, a very rare headache that tends to occur in younger adults.
The pain centers around one eye and is usually less intense than in a standard migraine. It may be accompanied by vomiting, double vision, a droopy eyelid, and paralysis of eye muscles. Attacks can last from hours to months.
In ICD-9-CM, we would report code 346.20 (variants of migraine, not elsewhere classified, without mention of intractable migraine without mention of status migrainosus). That doesn’t really tell us much about Jeff’s specific condition. “Without mention” appears twice in the code description and it’s a not elsewhere classified code—we have information, but no specific code available.
ICD-10-CM resolves part of that problem. We will have a code for ophthalmoplegic migraine. In fact, we have two:
- G43.B0, ophthalmoplegic migraine, not intractable
- G43.B1, ophthalmoplegic migraine, intractable
We know the migraine is not intractable because it resolved with standard treatment. First, Dr. Achy didn’t document it as intractable. Second, “intractable” means the headache doesn’t resolve with the usual treatment. Jeff’s condition improved after he took some aspirin.
That gives us G43.B0.
If you see a lot of migraine patients in your facility or organization, check the physician documentation to make sure he or she is noting the type of migraine, e.g., with aura, without aura, with status migrainosus, persistent.
ICD-10-CM includes much more detailed choices for migraines so if your physicians aren’t documented to the level of specificity you need in ICD-10-CM, start working with them now. Implementation is less than a year away.
Saturday marks the 165th anniversary of the death of Edgar Allan Poe, but sadly, we’re still not sure of the cause of Poe’s demise. We do, however, have plenty of theories.
If Poe, who couldn’t really hold his liquor, did succumb to alcohol poisoning, how would we code it? First, we head to the Table of Drugs and Chemicals, then find alcohol and its 28 related entries. Apparently you can be poisoned by a lot of different types of alcohol.
In Poe’s case, it was probably ethanol (T51.0X-). In fact, 13 of the terms under alcohol in the Table of Drugs direct you to T51.0X-.
Now we need the intent behind the poisoning. We have specific codes for:
- Accidental (unintentional)
- Intentional self-harm
We also need a seventh character to denote the encounter. We have three choices:
- A, initial encounter
- D, subsequent encounter
- S, sequela
Most likely, the alcohol poisoning (if that’s what put an end to the poet) was accidental, so we would report T51.0X1A.
However, we are faced with a more sinister possibility. Some have speculated that Poe was actually the victim of cooping, a method of voter fraud practiced by gangs in the 19th century. The gangs would kidnap people and take them around to various polling places so they could vote multiple times. After each successful casting of the ballot, the gang would reward their victim with alcohol.
You might be able to make a case for this being an assault, although that is probably a tough sell. You could also make a case for undetermined.
You’ll also find the following note under T51.01X-:
- Acute alcohol intoxication or ‘hangover’ effects (F10.129, F10.229, F10.929)
- Drunkenness (F10.129, F10.229, F10.929)
- Pathological alcohol intoxication (F10.129, F10.229, F10.929)
An Excludes2 note tells us that the conditions listed are not part of T50.1X-, but a patient could be suffering from one of these conditions as well. So if Poe’s doctor documented pathological alcohol intoxication, we would code it in addition to the alcohol poisoning.
Not everyone is convinced that Poe died from too much drink. Another popular theory is rabies. Many of Poe’s symptoms fit. He was admitted to a hospital due to “lethargy and confusion.”
Once admitted, his condition rapidly declined and he exhibited:
- Visual hallucinations
- Wide variations in pulse rate and rapid
- Shallow breathing
Four days after being admitted, Poe died. Four days is the median survival rate after the onset of serious rabies symptoms.
Rabies, however, is not a perfect fit. Poe did not display hydrophobia, which commonly afflicts those with rabies. He also didn’t have signs of an animal bite.
ICD-10-CM offers three codes for rabies:
- A82.0, sylvatic rabies
- A82.1, urban rabies
- A82.9, rabies, unspecified
We would probably go with unspecified rabies because we don’t know where Poe was infected (if he even was). The author was missing for several days before he was found lying in the street incoherent.
Perhaps flu felled the master of the macabre. Poe may have been suffering from influenza, which may have turned into pneumonia. Poe is reported to have been ill before he left Richmond, Virginia, to travel to Philadelphia. A high fever might account for his hallucinations and his confusion and the rain in Baltimore may explain his second-hand clothes.
If this was indeed Poe’s cause of death, we would code both the influenza and the pneumonia. And we can probably code it with a single combination code.
Poe’s pneumonia was likely either bacterial or viral. Code categories J13 (pneumonia due to Streptococcus pneumonia), J14 (pneumonia due to Hemophilus influenza), J15 (bacterial pneumonia, not elsewhere classified), and J16 (pneumonia due to other infectious organisms, not elsewhere classified) all instruct codes to first code any associated influenza.
However, when you look up “influenza” in the ICD-10-CM Alphabetic Index, you will find a listing for “influenza with pneumonia.” The index entry leads to you a very non-specific code: J11.00 (influenza due to unidentified influenza virus with unspecified type of pneumonia).
That’s probably the code we need to report because we don’t have any additional information (and we can’t query the physician. At this point, he’s dead too).
Keep in mind that the ICD-10-CM Official Guidelines for Coding and Reporting tell us it’s okay to report unspecified codes. Guideline B.18 in fact uses pneumonia as an example:
When sufficient clinical information isn’t known or available about a particular health condition to assign a more specific code, it is acceptable to report the appropriate “unspecified” code (e.g., a diagnosis of pneumonia has been determined, but not the specific type).
The guidelines also state that the physician shouldn’t perform tests just for more specificity.
Poe’s official cause of death, the one on his death certificate, is phrenitis, or swelling of the brain.
If you look up “phrenitis” in the ICD-10-CM Alphabetic Index, you’re directed to “see encephalitis.”
Not surprisingly, we find a plethora of choices, including alcoholic encephalopathy (G31.2). We would need the physician to document alcoholic encephalopathy before we can code it.
Sadly Poe’s medical records have gone missing and shall be seen nevermore.
Today we’re going to look at a real-life injury, but I will say up front that I don’t have all of the information about this particular patient’s injuries.
Let’s start at the very beginning. A diffuse axonal injury is a severe traumatic brain injury or severe closed head injury. That tells us we will not be coding for an open head injury. Diffuse also tells us that the injury does not involve one specific area of the brain like a focal brain injury (ICD-10-CM includes specific codes for focal injuries). Instead a diffuse axonal injury affects a widespread area of the brain.
How will we look up a diffuse axonal injury in the ICD-10-CM Alphabetic Index? The place to start is Injury. We have pages and pages of injury codes in ICD-10-CM.
Would you look under “axonal” or “brain”? If you said brain, you’re correct. I tried looking up Injury, axonal and got nowhere. I also tried diffuse axonal with no luck.
Under injury, brain (traumatic), we have an entry for diffuse (axonal), which directs us to S06.2X-.
A quick aside before we head to the Tabular List. As with ICD-9-CM, in ICD-10-CM when words appear in parentheses, they are nonessential modifiers. That means we don’t need them to code, but they help tell us we’re in the right place.
Off to the Tabular List to identify our additional characters. We find the following choices, broken down by the length of any associated loss of consciousness (LOC):
- S06.2X0-, diffuse traumatic brain injury without loss of consciousness
- S06.2X1-, diffuse traumatic brain injury with loss of consciousness of 30 minutes or less
- S06.2X2-, diffuse traumatic brain injury with loss of consciousness of 31 minutes to 59 minutes
- S06.2X3-, diffuse traumatic brain injury with loss of consciousness of 1 hour to 5 hours 59 minutes
- S06.2X4-, diffuse traumatic brain injury with loss of consciousness of 6 hours to 24 hours
- S06.2X5-, diffuse traumatic brain injury with loss of consciousness greater than 24 hours with return to pre-existing conscious levels
- S06.2X6-, diffuse traumatic brain injury with loss of consciousness greater than 24 hours without return to pre-existing conscious level with patient surviving
- S06.2X7-, diffuse traumatic brain injury with loss of consciousness of any duration with death due to brain injury prior to regaining consciousness
- S06.2X8-, diffuse traumatic brain injury with loss of consciousness of any duration with death due to other cause prior to regaining consciousness
- S06.2X9-, diffuse traumatic brain injury with loss of consciousness of unspecified duration
Note that these are not complete codes. We need a seventh character to indicate the encounter. You’ll need to go back to the beginning of category S06 (intracranial injury) to find that information.
Category S06 also includes these notes:
Code also any associated:
- Open wound of head (S01.-)
- Skull fracture (S02.-)
We face two challenges whenselecting the correct code. First, does the documentation specify the length of the LOC? Based on information in the media right now, we’re looking at an LOC of greater than 24 hours. That eliminates the first five codes.
And it brings us to the second challenge—how well did the patient return to pre-existing conscious level? We may not know the final amount of recovery at the time of discharge (assuming the patient survives—the prognosis for diffuse axonal injuries isn’t very good). Brain injuries are complex and can take a significant amount of time to heal.
The code we assign at discharge will be based on the patient’s status at that time.
The Workgroup for Electronic Data Interchange’s (WEDI) most recent survey on ICD-10 readiness included a particularly alarming (at least to me) statistic: Approximately 25% of the 324 providers who responded to the survey plan only to do crosswalking from ICD-9 to ICD-10.
Here’s why. General Equivalence Mapping (GEMs) like those CMS posts and other crosswalks are tools to help coders and, like any tool, they have their good points and bad points. Good point: they give coders a place to start. Bad point: they don’t get you to the final code.
Put another way: GEMs will get you to the correct neighborhood, but they won’t get you to the front door.
CMS even admits the limitations of the GEMs, stating: “There is no simple crosswalk from I-9 to I-10 in the GEM files.”
In some cases, you will find a one-to-one correlation from ICD-9-CM to ICD-10-CM (forget about it when it comes to ICD-9-CM procedure codes and ICD-10-PCS).
For example, Salmonella meningitis translates cleanly from 003.21 in ICD-9-CM to A02.21 in ICD-10-CM. Those cases are the exception. We wouldn’t need to move to ICD-10-CM if all codes were one-to-one matches.
In many cases, one ICD-9-CM code translates to multiple ICD-10-CM codes. Consider ICD-9-CM code 733.93 (stress fracture of tibia or fibula).
ICD-10-CM contains five choices that could match 733.93:
- M84.361-, stress fracture, right tibia
- M84.362-, stress fracture, left tibia
- M84.363-, stress fracture, right fibula
- M84.364-, stress fracture, left fibula
- M84.369-, stress fracture, unspecified tibia and fibula
Odds are, the GEM will take you to M84.369, which really tells you nothing about the actual fracture. No laterality, no specific bone.
It’s also not a valid code as is. It requires one of these seventh character to denote the encounter:
- A, initial encounter for fracture
- D, subsequent encounter for fracture with routine healing
- G, subsequent encounter for fracture with delayed healing
- K, subsequent encounter for fracture with nonunion
- P, subsequent encounter for fracture with malunion
- S, sequela
That information is not included in the ICD-9-CM code you’re crosswalking.
GEMs have other drawbacks as well. By now, you’ve probably heard that diabetes codes in ICD-10-CM no longer specify controlled or uncontrolled. ICD-10-CM also includes five subcategories of diabetes codes compared to two in ICD-9-CM. No way those codes will match up.
ICD-10-CM also includes different ways of classifying conditions. Consider complications of pregnancy. ICD-9-CM classifies them as:
- Unspecified episode of care
ICD-10-CM codes instead classify them by trimester:
- First trimester
- Second trimester
- Third trimester
- Unspecified trimester
An antepartum episode of care could fall into any of those trimesters. The OB/GYN is probably documenting somewhere in the patient’s record which trimester she is in. By just using a crosswalk, you lose that detail and likely end up with an unspecified code.
Unspecified codes aren’t the end of the world, but they do defeat part of the purpose for moving to ICD-10-CM. They could also result in denials. We don’t know for sure yet, but why take the chance?
ICD-10-CM includes a few different guidelines from ICD-9-CM (not many, but you still need to know them). GEMs won’t take those guideline changes into account. And you may or may not end up at the correct combination codes (ICD-10-CM includes a lot of those).
The best way to learn to code in ICD-10-CM is to actually code the record using the ICD-10-CM Manual. In fact, almost half of the providers responding to the WEDI survey plan to do just that. If your organization is using just GEMs, you may need to do a little work on your own or explain why GEMs alone aren’t the answer. It’s more work now, but it will pay off after the transition.