Documentation of chronic kidney disease (CKD) and its stage are crucial for correct coding, which affects hospital revenue and severity of illness classification. Precise diagnosis and reporting of CKD stage improve the accuracy of our national health care database used for research and for projections of national health care needs.
The 2012 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease published by the National Kidney Foundation (NKF) is the authoritative, consensus standard for the diagnosis, classification and management of CKD. The guideline identifies five stages of CKD (listed in Table 1), which are defined by either glomerular filtration rate (GFR) decreased to less than 60 mL/min/1.73 m2 or by presence of certain kidney damage markers (listed in Table 2) for more than three months.
The diagnosis of CKD stage 1 or 2 requires the presence of one or more of the markers of kidney damage. Therefore, a patient with a GFR greater than or equal to 60 mL/min/1.73 m2 does not have CKD if there are no markers of kidney damage. Stages 3, 4, and 5 are based on GFR cutoffs alone, although such patients may also have markers of kidney damage.
Lab reports of creatinine values typically include the calculated GFR for both African-American and non-African-American patients. The calculated GFR is proportional to serum creatinine but also depends on age, race, and gender. The NKF recommends using the CKD-EPI Creatinine Equation, published in 2009, to estimate GFR (available on the NKF website). It is not necessary to perform a 24-hour urine collection to measure creatinine clearance, even though this may be a more precise measure of GFR if done correctly.
The stage of CKD can only be correctly assigned when GFR and therefore creatinine levels are at a stable baseline. Staging cannot be performed when the creatinine is in flux, such as during an inpatient admission. Renal function must be stabilized before staging.
The currently preferred clinical terminology of “chronic kidney disease” or CKD should be used in diagnostic documentation. Clinicians should avoid nonspecific, imprecise terminology such as “renal insufficiency” or “chronic renal insufficiency (CRI)” as the correct codes for CKD will not be assigned.
End-stage renal disease (ESRD) represents the culmination of progressive stage 5 CKD. It has been a long-standing clinical practice to classify patients with CKD who require either dialysis or transplantation as ESRD. However, ICD-10-CM defines ESRD (code N18.6) as “dialysis-dependent stage 5 CKD.” Dialysis dependence means that the requirement for dialysis is expected to last three months or more.
A significant conflict exists between the clinical documentation standards and ICD-10-CM coding practices for a patient who has had a kidney transplant. Clinically, transplant patients are considered to have ESRD even though the transplanted kidney may be functioning well.
On the other hand, ICD-10-CM requires the transplant status code Z94.0 plus a code for the current CKD stage of the transplanted kidney. All transplant patients are considered to have CKD so any of the five stages will apply, based on the GFR. As an example, the code for a patient with a GFR of 50 mL/min/1.73 m2 after transplant would be N18.3 (CKD-3) and Z94.0. The code for ESRD (N18.6) cannot be assigned because it requires chronic dialysis dependence If the transplant patient eventually became dependent on dialysis again, code N18.6 could then be assigned together with Z94.0.
In summary, correct documentation of CKD stage is crucial for accurate coding, hospital reimbursement, and severity of illness classification. Clinicians should always review the calculated GFR associated with the creatinine level on clinical lab reports and remember that CKD stage can only be determined when renal function (GFR and creatinine levels) are stable, unless a recent baseline measurement is available.
CKD stages are defined by either a GFR below 60 mL/min/1.73 m2 or by the presence of certain kidney damage markers for more than three months. ESRD is defined clinically as either dialysis dependence or kidney transplant status, but the ICD-10-CM definition requires dialysis dependence. Transplant patients are identified with the transplant status code (Z94.0) plus a code for the current CKD stage of the transplanted kidney.
Ask Dr. Pinson
Q: A patient with alcohol dependency was admitted for alcohol intoxication with confusion and disorientation. Blood alcohol level was 412 mg/dL. Subsequently he began experiencing signs and symptoms of delirium tremens (DTs) and was transferred to the ICU. In your opinion, is it appropriate to query for toxic encephalopathy due to alcohol intoxication?
A: Clinicians should be aware of the coding implications of toxic encephalopathy due to alcohol. Yes, a clarification query may be submitted. Without a query, the code assignment would be:
1. Principal diagnosis code: F10.229, alcohol dependence with intoxication (present on admission [POA] indicator = yes)
2. Secondary diagnosis code: F10.231, alcohol dependence with withdrawal delirium (DTs) (POA = no, since it occurred after admission)
The diagnosis-related group (DRG) is 897.
If the patient's condition is clarified as toxic encephalopathy due to alcohol, then the code assignment would change to:
1. Principal diagnosis code: T51.0X4A, toxic effect of ethanol, initial encounter (POA = yes)
2. Secondary diagnosis codes: G92, toxic encephalopathy (POA = yes), followed by F10.229 (POA = yes), then F10.231 (POA = no)
The resulting DRG is 917. Clinicians should also be aware that documentation of “alcoholic encephalopathy” results in assignment of code G31.2 for a chronic degenerative brain condition (Wernicke-Korsakoff) rather than acute encephalopathy due to alcohol.
Q: I was disappointed by your column “Linking Severe Sepsis and Hyperlactatemia” in the June 2018 ACP Hospitalist. You seem comfortable with preserving the Sepsis-2 definition, instead of adopting the 2016 Sepsis-3 criteria that I believe should be embraced by the medical community. Hyperlactatemia is not one of the Sequential Organ Failure Assessment (SOFA) organ dysfunction parameters in the Sepsis-3 definition.
A: Thank you for your concern about the Sepsis-2/Sepsis-3 controversy. The June edition of Coding Corner was written in response to a specific question from a reader working at a facility where, whether we agree with it or not, the Sepsis-2 definition is still used—a not uncommon situation at this time.
The article was devoted to the diagnostic, documentation, and coding nuances of hyperlactatemia pursuant to the ICD-10-CM Official Guidelines for Coding and Reporting, the Surviving Sepsis Campaign (SSC), CMS's Hospital Inpatient Quality Reporting (IQR) system, and Sepsis-2.
I believe we agree on the authoritativeness of Sepsis-3. As I wrote in the June article:
“Another issue is that Sepsis-2 definitions and criteria (systemic inflammatory response syndrome [SIRS] due to infection) are no longer the authoritative clinical diagnostic standard for sepsis recognized by SSC, having been replaced by Sepsis-3, which was published in February 2016 and adopted by SSC in March 2017. Appeals of auditor denials based on Sepsis-2 criteria are unlikely to be successful, but all such cases should be reviewed for the possibility of using Sepsis-3 diagnostic criteria as the basis for appeal.”
It is widely recognized that the SIRS criteria are overly sensitive, leading to overdiagnosis of sepsis in patients who are not truly “septic.” We should also acknowledge that many of our colleagues do not yet trust the sensitivity of Sepsis-3 and qSOFA without prospective outcomes research and are concerned about underrecognition of truly septic patients.
Lastly, the CMS IQR measure for management of severe sepsis still follows selected SIRS diagnostic and severe sepsis criteria (not Sepsis-3/SOFA), so clinicians must utilize these Sepsis-2 criteria for initiating a severe sepsis management bundle to demonstrate quality care.
This is certainly an unsettled period in the evolution of our definitions, diagnostic tools, and effective management of sepsis. I hope this discussion offers some clarity for all our readers.