How Do We Code Outpatient Surgeries-Turned-Hospitalizations?

Last week I wrote about when surgery should be performed as an inpatient, when it can be outpatient, and how to utilize observation services in the context of surgical procedures. This week, I want to tackle how to document and code such encounters compliantly. I’d like to thank Colleen Ejak, Kathy Murchland, Damon Schmelzle, and ICD10monitor’s Laurie Johnson and Patty Chua for their sage advice.

Let’s use as our example a Medicare patient with chronic cholecystitis who has a laparoscopic cholecystectomy. The procedure is not on the Inpatient-Only (IPO) List, so, barring extenuating circumstances like having high-risk comorbidities, it will be done as an outpatient. For right now, let’s stipulate that it is being done in a hospital-based outpatient surgery department.

Undergoing this procedure as an outpatient surgery, the patient is expected to be in the recovery room for a limited amount of time and then discharged to home. The first-listed diagnosis is K81.1, Chronic cholecystitis, and the procedure is billed as with Current Procedural Terminology (CPT®) Code 47562, Laparoscopy, surgical; cholecystectomy.

Let’s say that the surgery was scheduled as the last procedure of the day, and ended up getting bumped so late that the surgeon takes pity on the patient and says, “Let’s keep you overnight to watch you.” Without any complications or unusual issues with recovery, an overnight stay for convenience is just an extended recovery or overnight/outpatient in a bed. These do not constitute observation services.  Also, an observation stay cannot be scheduled in advance. Dr. Ronald Hirsch’s saying is that if your provider is scheduling observation prior to the procedure, they are scheduling a complication – and maybe you should find someone else to do the surgery.

Now, say our newly gallbladder-less friend begins vomiting from the anesthesia and just can’t stop. They now have the complication of intractable vomiting. We have every hope that this will stop within about 24 hours, so placing the patient in outpatient status for observation services (OBS) is reasonable. ICD-10-CM Guidelines for Coding and Reporting, IV.A.2., states that the reason for the surgery is the first reported diagnosis (it is the reason for the encounter), followed by codes for the complications as secondary diagnoses.

Therefore, her first-listed diagnosis (the outpatient correlates to the principal diagnosis (PDx) for an inpatient admission) is K81.1, Chronic cholecystitis. But what would the secondary diagnosis/diagnoses be?

Before we tackle the secondary diagnoses, let’s address the procedure. Since the procedure and the observation stay are in the same hospital, the encounters are combined (for the technical component). The 47562, Laparoscopic cholecystectomy, would also be coded on the outpatient claim with observation hours. But it does not get coded if the entities were not related; e.g., outpatient surgery in an ambulatory surgical center (ASC) not affiliated with the hospital to which the patient was transferred (I’ll get to this scenario in a moment).

The issue is vomiting, so a secondary diagnosis would be R11.10, Vomiting, unspecified. But does that tell the whole story? Especially if the procedure is not coded as in the ASC/observation case, how can we tell that the patient had an operation?

If the procedure were an appendectomy, you would use K91.0, Vomiting following gastrointestinal surgery. This is a combination code – it informs us that there is vomiting and the circumstances that elicited the vomiting (that it is post-surgical). Cholecystectomy isn’t gastrointestinal surgery – it is on the biliary tract. It is in the digestive system, but it isn’t gastrointestinal. The fact that the gastrointestinal Procedure Coding System falls into 0D and gallbladder procedures fall into 0F supports this position.

I posit that ICD-10-CM code K91.89 and other postprocedural complications and disorders of the digestive system should be assigned. R11.10 would be an “additional code, if applicable, to further specify disorder.” K91.89 gives that piece of information that the patient underwent a surgical procedure, and the issue is a postprocedural complication. R11.10 indicates what the postprocedural complication was.

An aside – K91.5, Post-cholecystectomy syndrome (PCS), is a condition in which a patient undergoes a cholecystectomy and the symptoms that elicited the surgical intervention persist or recur – or new symptoms normally attributed to the gallbladder arise. Right upper-quadrant abdominal pain and dyspepsia are common. It is considered early if it occurs in the postoperative period, but it can manifest after months or years in the late variant. If the provider documented “post-cholecystectomy syndrome” for our exemplar, you would pick this code up. If they document “intractable vomiting” or “persistent vomiting post-anesthesia,” PCS would not be appropriate.

The next scenario is that the same outpatient surgery patient placed in observation with intractable vomiting gets so dehydrated that they ends up in renal failure; in this case, it is clear they are going to cross a second midnight, so the provider converts them to inpatient. According to II. I. 2., “When a patient is admitted to observation to monitor a condition (or complication) that develops following outpatient surgery, and then is subsequently admitted as an inpatient of the same hospital, hospitals should apply the Uniform Hospital Discharge Data Set (UHDDS) definition of PDx as ‘that condition established after study to be chiefly responsible for occasioning the admission of the patient to the hospital for care.’”

The acute kidney injury (AKI) is the reason they are being admitted, so you would think N17.9, Acute kidney failure, unspecified, could be the PDx. However, I would still say that there is a direct line from the postoperative complication (vomiting) to the dehydration causing AKI, and therefore, K91.89 is still the PDx.  E86.0, Dehydration, N17.9, and R11.10 would be secondary diagnoses.

The other twist is that the procedure gets converted into ICD-10-PCS. It will likely drive the Diagnosis-Related Group (DRG) and is included because it is part of the episode of care, according to the three-day payment window rule.

We have to digress for a moment again. This surgery was done in a hospital-based outpatient setting. For Medicare patients, the technical component of all outpatient diagnostic services and therapeutic services considered related, within the three days preceding, get bundled into an inpatient admission. This is called the three-day payment window rule or policy and applies to entities wholly owned or wholly operated by the hospital. This may encompass more than 72 hours because it is according to the calendar day.

There is an exclusion if the hospital and other Part B entity are both owned by a third party, such as a comprehensive healthcare system. It also doesn’t apply if the entities are not related in any way. Psychiatric hospitals and units, inpatient rehabilitation hospitals and units, long-term care hospitals, children’s hospitals, and cancer hospitals have a one-day payment window.

The final setup is directly admitting a patient from outpatient surgery. Let’s use a different example this time because there are three scenarios to discuss. I’m choosing acute appendicitis seen in the emergency department. If the patient is sick and/or has significant comorbidities and risk, they may get admitted inpatient prior to the surgery, even though appendectomy is not on the IPO List. However, sometimes, the surgeon takes the patient to the operating suite intending for it to be an outpatient procedure, and then has to pivot after surgery.

• Scenario 1: The patient has a complication after or during surgery. Let’s pick accidental puncture and laceration of the bowel with contents spillage. K91.71, Accidental puncture and laceration of a digestive system organ or structure during a digestive system procedure is the PDx, as per II. J. Admission from Outpatient Surgery. If the reason for the inpatient admission is a complication, that is assigned as the PDx. If the complication is from T80-T88 or is too generic and lacks specificity, an additional code for the specific complication is indicated. The condition that elicited the procedure is a secondary diagnosis. The procedure is included on the claim in the ICD-10-PCS form.

• Scenario 2: During the procedure, it was determined that the appendicitis was more complicated than the imaging suggested. There was perforation or a poorly visualized abscess, and the surgeon realized that it was going to take several days of antibiotics and monitoring to ensure a good outcome. They write admission orders and convert the outpatient surgery to an inpatient admission. The PDx is the reason for the operation, found in K35.-, Acute appendicitis, with specificity (generalized or localized peritonitis, abscess, perforation, and/or gangrene). II.J. stipulates: “if no complication, or other condition, is documented as the reason for the inpatient admission, assign the reason for the outpatient surgery as the PDx.”
  
  If the hospital owns the outpatient surgical center (or ASC, for an appropriate procedure), the procedure would be coded. But if the hospital doesn’t own it, how do you convey that the patient is postoperative? Could you use Z90.49, Acquired absence of other specified parts of the digestive tract, to signify that the patient had undergone an appendectomy? This type of Z code, or status code, indicates “that a patient has the sequelae or residual of a past disease or condition” [I.C.21.3)]. This isn’t a “past disease or condition” yet. Z90.49 (when relevant) will be used in the future, but it is not appropriate until the patient has completely healed.
  
  What I am going to say now may be controversial. I think you would use Z48.815, Encounter for surgical aftercare following surgery on the digestive system. This conveys that the patient is still in the recovery phase after a procedure. II.21.c.7 instructs that this type of code be used “when the patient requires continued care during the healing or recovery phase.” The American Hospital Association’s ICD-10-CM and -PCS Coding Handbook says: that aftercare codes “can be used occasionally as additional codes when aftercare is provided during an encounter for treatment of an unrelated condition but no applicable diagnosis code is available. Aftercare codes should be used in conjunction with any other aftercare or diagnosis code(s) to provide better detail on the specifics of an aftercare visit…” We have the details of the underlying condition that caused the surgery, but we don’t, as yet, have a code indicating that surgery was done for that condition and the patient is in the recovery phase.
  
  If the hospital owned both entities, the payment episode was continuous, and the procedure is enfolded into the encounter. If the hospital does not, then the payment episode is interrupted, and Z48.815 could communicate that the patient was post-surgical. Especially now that we treat some acute appendicitis with antibiotics instead of cold, hard steel, we definitely need a mechanism to signal that the patient has been operated upon.
• Scenario 3: The patient has a comorbidity that occasionally flares up, and it does so in the recovery room or has a new condition not felt to be directly due to the surgery, per se. Perhaps the provider explicitly links it to something other than the surgery (e.g., acute-on-chronic hypercapnic respiratory failure due to exacerbation of COPD). Don’t get confused like clinicians do – just because it occurs in the postoperative period doesn’t make it cause-and-effect, a requisite for recognizing a complication for ICD-10-CM.
  
  As per II.J., “If the reason for the inpatient admission is another condition unrelated to the surgery, assign the unrelated condition as the PDx.” J44.1, Chronic obstructive pulmonary disease with (acute) exacerbation, would be the PDx; J96.22, Acute and chronic respiratory failure with hypercapnia and K35.30, Acute appendicitis with localized peritonitis without perforation or gangrene, would be secondary diagnoses, and 0DTJ4ZZ, Resection of Appendix, Percutaneous Endoscopic Approach, would be the principal procedure.
  
  If the coder is unsure, they may need to query. For instance, a query to ascertain whether the provider believes the new-onset atrial fibrillation is a postprocedural complication or unrelated may be necessary (i.e., is it I97.191, Other postprocedural cardiac functional disturbances following other surgery with I48.91, Unspecified atrial fibrillation, or just I48.91?).
  
  Along those lines, an unrelated medical condition arising in recovery resulting in the provision of observation services would be first-listed, the condition causing the surgery would be a secondary diagnosis, and the procedure would be coded only if the observation site was owned by the same hospital as the outpatient surgical center. Again, I endorse Z48.815, Encounter for surgical aftercare following surgery on the digestive system to indicate the postoperative status if the surgical site was not owned by the hospital providing the observation services. There is a whole set of surgical aftercare codes available in Z48.8-.

Bottom line, here’s my flow chart:

•   Intraoperative or postprocedural complication
  • Observation stay
    • Reason for surgery is the first-listed diagnosisComplication code secondaryAdditional code describing complication, if applicable(Other comorbidities’ secondary diagnoses, if applicable)Hospital owns both outpatient surgery center and observation à CPT codeIndependent outpatient/ASC and hospital observation à No procedure code. The fact that the patient is post-op is inherent in the complication code.
    Observation to inpatient conversion
    • Reason for conversion to inpatient is PDx (likely will be the complication)Reason for surgery is secondary diagnosisAdditional code describing complication, if applicable(Other comorbidities’ secondary diagnoses, if applicable)Hospital owns both outpatient surgery center and observation à ICD-10-PCS codeIndependent outpatient/ASC and hospital observation à No procedure code. The fact that the patient is post-op is inherent in the complication code.
  • Inpatient admission directly
    • Complication code is PDx
    • Reason for surgery is secondary diagnosis
    • Additional code describing complication, if applicable
    • (Other comorbidities secondary diagnoses, if applicable)
    • Hospital owns outpatient surgery center à ICD-10-PCS code
    • Independent outpatient/ASC and hospital observation à No procedure code. The fact that the patient is post-op is inherent in the complication code.
• Needs further hospital-level care following procedure without complication
  • Observation stay
    • Reason for surgery is first-listed diagnosis
    • Other comorbidities secondary diagnoses, if applicable
    • Hospital owns both outpatient surgery center and observation à CPT code
    • Independent outpatient/ASC and hospital observation à No procedure code and no complication  code. Use Z48.815, Encounter for surgical aftercare following surgery on the digestive system, as additional diagnosis.
  • Observation to inpatient conversion
    • Reason for conversion to inpatient is PDx (according to UHDDS – probably the reason for surgery)
    • Reason for surgery is either primary or secondary diagnosis
    • Other comorbidities’ secondary diagnoses, if applicable
    • Hospital owns both outpatient surgery center and observation à ICD-10-PCS code
    • Independent outpatient/ASC and hospital observation à No procedure code and no complication code. Use Z48.815, Encounter for surgical aftercare following surgery on the digestive system, as additional diagnosis.
  • IP admission
    • PDx according to UHDDS (probably the reason for surgery)
    • Reason for surgery is either primary or secondary diagnosis
    • Other comorbidities’ secondary diagnoses, if applicable
    • Hospital owns outpatient surgery center à ICD-10-PCS code
    • Independent outpatient/ASC and hospital observation à No procedure code and no complication code. Use Z48.815, Encounter for surgical aftercare following surgery on the digestive system, as additional diagnosis.
• Unrelated medical condition arises that is not a postoperative complication and requires observation or inpatient (if the documentation is not clear, may need a query to determine if complication or unrelated)
  • Observation stay
    • Unrelated medical condition causing observation is first-listed diagnosisReason for surgery is secondary diagnosisOther comorbidities’ secondary diagnoses, if applicableHospital owns both outpatient surgery center and observation à CPT codeIndependent outpatient/ASC and hospital observation à No procedure code and no observation code. Use Z48.815, Encounter for surgical aftercare following surgery on the digestive system, as additional diagnosis.
      
      
    Observation to inpatient conversion
    • Reason for conversion to inpatient is PDx (according to UHDDS – in this case, it is likely the unrelated medical condition)Reason for surgery is secondary diagnosisOther comorbidities’ secondary diagnoses, if applicableHospital owns both outpatient surgery center and observation à ICD-10-PCS codeIndependent outpatient/ASC and hospital observation à No procedure code and no complication code. Use Z48.815, Encounter for surgical aftercare following surgery on the digestive system, as additional diagnosis.
      
  • IP admission
    • Unrelated medical condition is PDx according to UHDDS
    • Reason for surgery is secondary diagnosis
    • Other comorbidities’ secondary diagnoses, if applicable
    • Hospital owns outpatient surgery center à ICD-10-PCS code
    • Independent outpatient/ASC and hospital observation à No procedure code and no complication code. Use Z48.815, Encounter for surgical aftercare following surgery on the digestive system, as additional diagnosis.

This was a fascinating thought exercise! In essence, you are telling the story of the surgery and what happened during recovery in codes. You need the provider to document so you understand the course of events and whether there was a complication or not. As far as how it gets put on the claims and gets billed, or what condition codes or modifiers you use, you need a real person from revenue cycle to counsel you on that!I hope you found this as interesting as I did. Let me know if you agree, or if it changes your practice (icd10md@outlook.com).