September 12, 2017

Warning: All Sepsis is Severe Sepsis

In the coding and clinical documentation community, we are still trying to sort out sepsis. In my previous article on this topic (, I made some recommendations on how to approach sepsis. We need to revisit this.

We have now had some time to live with the Sepsis-3 criteria, established by the Third International Consensus Definitions for Sepsis and Septic Shock published in the Journal of the American Medical Association (JAMA). In January 2017, the Surviving Sepsis Campaign: International Guidelines for Management of Severe Sepsis and Septic Shock: 2016 (SSC-2016) was issued, and I think it since really has been flying under the radar.

SSC-2016 has three notable changes:

  1. The details regarding how to make the diagnosis of sepsis are not laid out in the article. SSC-2012 had Table 1, Diagnostic Criteria for Sepsis, but SSC-2016 does not have a correlate. The current guidelines focus on recommendations for treatment. It is unclear to me whether the authors intend readers to refer back to the SSC-2012 publication for the diagnostic criteria or whether they advocate migrating to the Sepsis-3 criteria.

One of the main reasons we are in this pickle is that a grave disservice was done to the clinical criteria by reducing them merely to the SIRS (systemic inflammatory response syndrome) criteria. The definition of sepsis was “presumed or confirmed infection plus systemic manifestations of infection,” and Sepsis-2 included inflammatory, hemodynamic, organ dysfunction, and tissue perfusion variables, in addition to the general variables. Although they are very simple to recall and apply, the problem with using SIRS criteria exclusively is that they are so incredibly nonspecific, and consequently, patients were frequently labeled as septic inappropriately.

  1. The definitions of sepsis and septic shock are identical to the definitions found in Sepsis-3:

“Sepsis is life-threatening organ dysfunction caused by a dysregulated host response to infection. Septic shock is a subset of sepsis with circulatory and cellular/metabolic dysfunction associated with a higher risk of mortality.”

The phrase “severe sepsis” is no longer in the lexicon, having been subsumed by “sepsis.” This is highly consequential, as we will touch on shortly.

  1. There are subtle changes in the recommendations from 2012 as compared to 2016, and these are outlined in a comparison table found in Appendix 2. The most significant changes include removing references to early goal-directed therapy (which was felt to not be valid) and recommending dynamic variable measurements for fluid responsiveness assessment.

The remaining changes seem to be less substantive, addressing strength of recommendations (“recommend” versus “suggest,” as opposed to by grade). Epinephrine is no longer the definitive second-line vasopressor, and low-molecular-weight heparin is recommended over unfractionated heparin.

The definitions for SSC-2016 and Sepsis-3 may have aligned, but their semantics do not correspond to ICD-10-CM’s verbiage and codes. ICD-10-CM has three choices of codes for sepsis, whereas there are now only two options for conditions: sepsis and septic shock.

The guidance in Sepsis-3 was to code R65.20, Severe sepsis, whenever a provider diagnosed “sepsis.” Although this instruction isn’t compliant, the implication is that there are really only two categories now: severe sepsis without, and with, septic shock. We will always need a “sepsis” code, in addition, because it establishes the organism (e.g., A41.51, Sepsis due to Escherichia coli), but my interpretation is that in the future, we will no longer see a sepsis code used alone.

The challenge here is twofold. Providers need to apply the diagnosis of sepsis (previously known as severe sepsis) only to sick patients, as was always intended, and then they need to document such that coding of R65.20 is appropriate, compliant, and denial-proof.

Here is the information I would provide the practitioners to help them:

  • Sepsis is life-threatening organ dysfunction caused by a dysregulated host response to infection manifested by a constellation of clinical signs and symptoms.
    • This patient is, and looks, ill! He or she is sicker than the average patient with the same garden-variety infection. In the physical examination, you should be describing a sick patient: “looks toxic,” “ill-appearing,” “in extremis,” “looks acutely sick,” “in acute distress,” etc.
    • A best practice is to detail sepsis-related acute organ dysfunction and manifestation. I recommend electronic assistance. Create a macro or template that lists suitable choices:

Sepsis with sepsis-related acute organ dysfunction (i.e., severe sepsis) of:
           Check box                                                     Text output
□ Hypoxemia; pO2 ____                     →      Hypoxemia with pO2 of 62
□ Acute resp fail; pO2 ____               →       Acute respiratory failure with pO2 of 55                 
□ ileus                                                  →       GI dysfunction of paralytic ileus
□ AKI; Cr. ____ ; baseline ____           →       AKI with Cr of 1.8 with baseline of 1.1
□ lactic acidosis; lactate ____          →       Tissue hypoperfusion with lactic acidosis, lactate4.7 mmol/L     
Even the expanded SOFA (quick Sequential [Sepsis-Related] Organ Failure Assessment) does not include every marker of organ dysfunction. You can anticipate that the insurers are going to try to deny “liver dysfunction” when only the AST/ALT rise with normal bilirubin. Include a generic option with specification of organ dysfunction as manifested by each patient.

  • It is in the patient’s best interests to consider sepsis early and rule it out as opposed to coming to the party late, incurring a higher morbidity and mortality.
    • Use an uncertain diagnosis if you are unsure (“possible,” “probable,” “likely,” etc.)
    • Attend to the SEP-1 Core Measures. I recommend setting up the electronic record so if a provider pulls up that sepsis macro, he or she is prompted to address core measures.
  • qSOFA (quick Sequential [Sepsis-Related] Organ Failure Assessment):
    • Does not require laboratory tests, like SOFA, and can be assessed quickly
    • Is a tool to help identify patients likely to have a prolonged ICU stay or to die. It is not the gold standard for diagnosing sepsis (otherwise known as “severe sepsis”)
    • Consists of:
      • Altered mental status (best documented as “encephalopathy” to address SOI/ROM/MCC)
      • Systolic BP ≤ 100 mm Hg (documented as “hypotension” or “shock,” depending)
      • Respiratory rate ≥ 22/min (take yourself; document as tachypnea)
    • Documentation is essential to prevent clinical validity denials.
      • Have diagnosis appear when made, as treated, when resolved, and in discharge summary. If the patient initially appeared septic in the ED but rapidly improved, and you decide it was not sepsis, rule it out and remove the diagnosis.
      • Documentation should be consistent from day to day and provider to provider
      • Support your diagnosis with your clinical indicators
    • Clinical validation queries are not meant to impugn your clinical judgement. It is a signal of a incongruity between documented diagnoses and clinical indicators. You have one of two choices:
      • Reconsider whether the condition is valid. If it is not, document that it is ruled out, and remove it from your diagnosis list.
      • If it is valid, the clinical support may not be evident, so you should beef up your documentation, providing the clinical indicators that have led you to your justifiable conclusion.

The geometric length of stay (GLOS) for MS-DRG 872, Septicemia or severe sepsis without mechanical ventilation is greater than 96 hours; without a MCC it is 3.9 days. MS-DRG 871 (with MCC) has a GLOS of 5.0 days, and MS-DRG 870 (with mechanical ventilation greater than 96 hours) has a GLOS of 12.6 days. If the patient survives, it is reasonable to speculate that sepsis (formerly known as “severe sepsis”) might not be a valid diagnosis if the length of stay was two days or less.

If your institution developed internal guidelines based on the Sepsis-2 criteria, it may be time to revisit and revamp them. I think there has been significant convergence, and third-party payers are migrating to Sepsis-3.

Late adopters may be ready to shift. I know I have.

PROGRAM NOTE: Hear Dr. Remer discuss Sepsis-3 criteria live on Talk Ten Tuesdays today at 10 a.m. ET.
Erica E. Remer, MD, FACEP, CCDS

Erica Remer, MD, FACEP, CCDS has a unique perspective as a practicing emergency physician for 25 years, with extensive coding, CDI, and ICD-10 expertise. As Physician Advisor for University Hospitals Health System in Cleveland, Ohio for four years, she trained 2,700 providers in ICD-10, closed hundreds of queries, fought numerous DRG clinical determination and medical necessity denials, and educated CDI specialists and healthcare providers with engaging, case-based presentations. She transitioned to independent consulting in July, 2016. Dr. Remer is a member of the ICD10monitor editorial board and co-hosts Talk Ten Tuesdays.

Related Stories

  • COPD, Sepsis Among Conditions Creating Coding Dilemmas in New AHA Coding Clinic
    AHA Fourth Quarter Coding Clinic identifies problematic diagnosis codes.Hopefully the subject of the most recently published Coding Clinic will not be too scary to discuss. It is effective for discharges from Oct. 1, 2017 forward. The majority of this issue…
  • Learning How to Avoid Sepsis Denials
    In short, you can avoid sepsis denials when documentation in the patient encounter shows a clear delineation of a non-systemic infection. This is the only ironclad defense that will withstand any and all scrutiny. Sepsis clinical parameters are good for…
  • Demystifying the Pathogenesis of Sepsis
    I want to begin with a coding scenario: sepsis and pneumonia are documented, and the coder captures these conditions and assigns MS-DRG 871 and APR-DRG 720 with severity of illness (SOI) 2.  Sepsis presents challenges for coders as well as…