• Home
    • » e-Newsletters

Specificity key to simplifying sepsis coding in ICD-10-CM

JustCoding News: Inpatient, November 20, 2013

Want to receive articles like this one in your inbox? Subscribe to JustCoding News: Inpatient!

Coders may find assigning codes for sepsis somewhat easier in ICD-10-CM, but they will still face some challenges. The first of those challenges, and probably the biggest, centers on physician documentation.

Coders know that physicians often use the terms “bacteremia,” “septicemia,” and “sepsis” interchangeably, even though those terms refer to different clinical conditions. ICD-9-CM contains separate codes for each of those conditions. Physicians also use the nonspecific term “urosepsis.”

Part of the difficulty is that physicians aren’t familiar with either the various ICD-9-CM or ICD-10-CM definitions for these conditions or the coding guidelines for coding sepsis, says Ann Barta, MSA, RHIA, CDIP, director of practice excellence for AHIMA in Chicago. “That really makes it a challenge for coders.”

Does the patient have septicemia or sepsis? Is it really bacteremia or possibly a localized infection, such as a urinary tract infection (UTI)? “Each of those conditions results in a different code assignment whether I’m in ICD-9-CM or ICD-10-CM,” Barta says.

Another problem arises when the physician documents all of the signs and symptoms of sepsis without documenting a diagnosis of sepsis.

“If I am reading the chart as a coder, I can tell from the patient’s signs and symptoms that this patient really meets the clinical criteria for sepsis, but the physician is not actually giving me the diagnosis,” Barta says. “When this situation occurs, a coder currently queries the physician for clarification in ICD-9-CM and will continue to have to query with the transition to ICD-10-CM. The clinical picture of sepsis isn’t going to change.”

Although the actual codes in ICD-10-CM will look different from ICD-9-CM codes, coders will still look for the same information in the documentation, Barta says. However, coders will need to become familiar with some new codes and new guidelines in ICD-10-CM.


ICD-9-CM contains codes for septicemia, which is the presence of bacteria in the bloodstream causing the patient to exhibit symptoms.

Coders currently have seven series of codes for septicemia:
  • 038.0, streptococcal septicemia
  • 038.1x, staphylococcal septicemia
  • 038.2, pneumococcal septicemia [Streptococcus pneumoniae septicemia]
  • 038.3, septicemia due to anaerobes
  • 038.4x, septicemia due to other gram-negative organisms
  • 038.8, other specified septicemias
  • 038.9, unspecified septicemia

ICD-9-CM includes a guideline that tells coders to report the underlying infection—which is usually the septicemia--first, Barta says. Coders also report the code for the sepsis, which is usually 995.91. Coders need two codes to report sepsis in ICD-9-CM.

“In ICD-10-CM, coders will not have codes with septicemia in the code title but rather will be assigned as sepsis since it is the body’s systemic reaction to infection,” says Shannon E. McCall, RHIA, CCS, CCS-P, CPC, CCDS, director of coding and HIM for HCPro, Inc., in Danvers, Mass.

If a coder looks up streptococcal septicemia in ICD-10-CM, it provides a cross-reference to see-Sepsis, where a specific code from category A40 (streptococcal sepsis) would be assigned. Coders then have these additional choices:

  • A40.0, sepsis due to streptococcus, group A
  • A40.1, sepsis due to streptococcus, group B
  • A40.3, sepsis due to streptococcus pneumoniae
  • A40.8, other streptococcal sepsis
  • A40.9, streptococcal sepsis, unspecified

ICD-10-CM category A40.- instructs coders to code first other categories that identify sepsis directly related to procedures or in pregnancy-related situations:

  • Postprocedural streptococcal sepsis (T81.4)
  • Streptococcal sepsis during labor (O75.3)
  • Streptococcal sepsis following abortion or ectopic or molar pregnancy (O03-O07, O08.0)
  • Streptococcal sepsis following immunization (T88.0)
  • Streptococcal sepsis following infusion, transfusion, or therapeutic injection (T80.2-)

The A40.- series of codes also has an Excludes1 note, which tells coders not to report the following conditions using a code from A40.-:

  • Neonatal (P36.0-P36.1)
  • Puerperal sepsis (O85)
  • Sepsis due to streptococcus, group D (A41.81)

“In ICD-10-CM, I am still coding the underlying infection, the streptococcal sepsis, but I don’t have an equivalent code to 995.91 because we no longer classify septicemia in ICD-10-CM,” Barta says. This change will result in coders understanding that only one code will be assigned for the diagnosis of sepsis in ICD-10-CM.

“Once coders get used to it, it should be easier because it’s one less term that physicians may document,” Barta says. If a physician documents septicemia, the default code in ICD-10-CM is A41.9, sepsis, unspecified organism.

Severe sepsis

As sepsis progresses, a patient’s organs begin to malfunction or could shut down. This could indicate the patient has severe sepsis. However, not all patients with sepsis will progress to severe sepsis, McCall says.

In order to code severe sepsis, coders must see documentation that supports the diagnosis in the medical record—or query the provider if necessary—when clinical indicators support the more specific diagnosis. Coders need to make sure the physician ties the organ failure/dysfunction directly to the sepsis, says McCall. When other conditions cause organ failure, it would not be appropriate to assign a code for severe sepsis.

If the physician appropriately documents severe sepsis, report either ICD-10-CM code R65.21 (severe sepsis without septic shock) or R65.22 (severe sepsis with septic shock) in addition to the sepsis code (A40.- or A41.-). Do not report a code from series R65.2- alone, McCall says.

The ICD-10-CM coding guideline for severe sepsis basically provides the same guidance as the current ICD-9-CM severe sepsis guideline, Barta says. ICD-10-CM, like ICD-9-CM, instructs coders to first report the underlying systemic infection, which is quite often sepsis, along with the code for severe sepsis and any acute organ dysfunction resulting from the severe sepsis.

The only real difference, Barta says, is that ICD-10-CM includes a combination code for severe sepsis with septic shock. In ICD-9-CM, coders report two codes for severe sepsis and septic shock.

The combination code in ICD-10-CM is able to link severe sepsis to septic shock since a patient cannot clinically have septic shock without severe sepsis. This concept was addressed in the ICD-9-CM Official Guidelines, which was likely the reasoning behind creating the combination code.


Another common problem area in ICD-9-CM is urosepsis. Physicians frequently use this term to refer to a systemic inflammatory response initiating from a urinary source, but without further detail in ICD-9-CM this diagnosis defaults to a UTI with code 599.0, McCall says.

Urosepsis is very ambiguous, Barta says. Coders often query physicians to clarify whether they are documenting a UTI or sepsis. “If I don’t get that clarification from the physician or, if as I am clinically reading the chart, I don’t think that patient has the symptoms that go with sepsis, I can code it to 599.0,” Barta says.

ICD-10-CM does not include a default code for urosepsis. “I think coders will be very glad because urosepsis has always been very confusing,” Barta says.

The ICD-10-CM Official Guidelines for Coding and Reporting state:

The term urosepsis is a nonspecific term. It is not to be considered synonymous with sepsis. It has no default code in the Alphabetic Index. Should a provider use this term, he/she must be queried for clarification.

“ICD-10-CM forces physicians to specify whether the patient has a UTI or a systemic inflammatory response from a urinary source,” McCall says. “If the physician doesn’t specify, it will result in a query 100% of the time.”

The change is great for coders, Barta says, because it eliminates the term urosepsis. For those physicians who still document urosepsis without further clarification, it will mean more queries for clarification. Hopefully those physicians will get on track,” Barta says. “For years we have been trying to get physicians to stop documenting urosepsis, because it really is not a recognized diagnosis in the coding classification system.”

Editor’s Note: This article was originally published in the November issue of Briefings on Coding Compliance Strategies. Email your questions to Senior Managing Editor Michelle A. Leppert, CPC, at mleppert@hcpro.com.

Want to receive articles like this one in your inbox? Subscribe to JustCoding News: Inpatient!

Most Popular