Sepsis-3: The world turned upside down

New definitions of sepsis and septic shock represent a radical departure from previous criteria.


It may not have the same implications as Cornwallis' surrender to George Washington at Yorktown in 1781, but Feb. 23, 2016, marked a revolutionary transformation of the concept of sepsis. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) (1), published on that date in the Journal of the American Medical Association (JAMA), represent a radical departure from the prior sepsis definitions in 1991 and 2001 and subsequent Surviving Sepsis Campaign (SSC) guidelines through 2015.

Photo by Thinkstock
Photo by Thinkstock

As discussed last month in this column, the first sepsis definition conference was convened in 1991, and the results were published in 1992 (2). The conference defined sepsis as systemic inflammatory response syndrome (SIRS) due to infection, and SIRS was defined as the presence of more than 1 of 4 findings:

  • Body temperature >38.0 °C or <36.0 °C
  • Heart rate >90 beats/min
  • Tachypnea >20 breaths/min or hyperventilation with PaCO2 <32 mm Hg
  • White blood cell (WBC) count >12,000 cells/mm3; or <4,000 cells/mm3;

A second International Sepsis Definitions Conference was convened in 2001, and the results were published in 2003 (3). This consensus retained the definitions of sepsis as SIRS due to infection (presumed or confirmed) and of severe sepsis as sepsis associated with acute organ dysfunction. The criteria defining SIRS (provided in last month's column) were greatly expanded from the 1991 original. Since 2003, the Surviving Sepsis Campaign (SSC) literature, including the 2012 SSC guidelines (4) and 2015 Sepsis Bundles (5), has consistently reaffirmed the 2001 criteria while offering some expanded details.

Now we have Sepsis-3. Citing the “inadequate specificity and sensitivity of the systemic inflammatory response syndrome (SIRS) criteria,” the new Sepsis-3 consensus abandons the concept of sepsis as SIRS due to infection, which has been the diagnostic standard for the last 25 years. Instead, it states that “Sepsis should be defined as life-threatening organ dysfunction caused by a dysregulated host response to infection [suspected or confirmed].” SIRS will no longer be the basis for diagnosis.

According to Sepsis-3, either of 2 standards may be used for identifying organ dysfunction. One option is the Sequential [Sepsis-related] Organ Failure Assessment (SOFA) score with organ dysfunction representing sepsis defined as an increase in the SOFA score of 2 points or more. The SOFA grades the function of 6 organ systems on a scale of 0 to 4 depending on the degree of dysfunction using objective measurements (Table). The baseline SOFA score is assumed to be 0 in patients not known to have preexisting organ dysfunction.

Sepsis-3 also recommends a new bedside clinical approach to identify patients with high risk of adverse outcomes called quickSOFA (qSOFA). These patients are identified under qSOFA by the presence of 2 or more of 3 clinical criteria: altered mentation, respiratory rate ≥22 breaths/min, and systolic blood pressure ≤100 mm Hg. The Sepsis-3 writers offer qSOFA as a useful tool, but note that it does not substitute for the SOFA criteria in defining sepsis.

The new definition of septic shock is quite strict: “persisting hypotension requiring vasopressors to maintain MAP [mean arterial pressure] >65 mmHg and having a serum lactate level >2 mmol/L (18 mg/dL) despite adequate volume resuscitation.” The requirement for both vasopressor-sustained MAP and an elevated lactate level seems extreme to me and perhaps inconsistent with other concepts and definitions of shock states.

Sepsis-3 also said that “the term severe sepsis was redundant,” adopting the idea that sepsis without organ dysfunction does not exist. Henceforth, all cases of sepsis will be classified as what was formerly severe sepsis with or without septic shock; there will no longer be any cases of sepsis without organ dysfunction. It will be interesting to see the reaction of the medical community in general to this concept.

One problem already is that the Sepsis-3 definitions are inconsistent with the ICD-10-CM Official Guidelines for Coding and Reporting (OCG). Adherence to these guidelines when assigning diagnosis codes is required under the Health Insurance Portability and Accountability Act (HIPAA). OCG clearly distinguishes between sepsis without organ dysfunction and sepsis with organ dysfunction, whereas Sepsis-3 does not. If hospitals begin reporting only cases of sepsis with organ dysfunction and no cases of sepsis without organ dysfunction, the entire process of coding, reimbursement, quality analysis, and regulatory oversight pertinent to sepsis will be disrupted.

The Sepsis-3 consensus also recommends the “primary” ICD-10 codes to be used pursuant to the new definitions. The document identifies code R65.20 for “sepsis” and R65.21 for “septic shock.” However, these codes are not “primary” sepsis codes at all. ICD-10-CM and the OCG require a primary code for sepsis (e.g., A41.9, unspecified organism or B37.7, Candida sepsis) to be sequenced first, followed by code R65.20 to identify severe sepsis without septic shock or by R65.21 when septic shock is identified.

Therefore, at the current time, following these new Sepsis-3 definitions will leave the expectations and practices for national coding and reporting requirements unmet. The national health care database will become inconsistent with all prior information, disrupting research and national health care trends and planning. Of course ICD-10-CM and the OCG can be modified on Oct. 1 each year, but the responsible parties are typically reluctant to make such radical changes.

The 19 members of the Sepsis-3 task force were selected for their “scientific expertise in sepsis epidemiology, clinical trials, and basic or translational research,” the paper said, and the consensus has been endorsed by 31 professional societies worldwide, including the American Thoracic Society, the Society of Critical Care Medicine, the Shock Society, and the Surgical Infection Society. In the same JAMA issue, Sepsis-3 was accompanied by 2 studies purporting to substantiate these new definitions and criteria.

This new sepsis definition is clearly based on the latest research and understanding of the pathobiology of sepsis, as the consensus claims, but does it accurately reflect the clinical reality and imperatives of early recognition and treatment? Even the Sepsis-2 (2001) criteria are not consistently utilized today, and sepsis still goes unrecognized. Clinicians have seen many cases of “sepsis” without organ dysfunction that have progressed to severe sepsis with organ dysfunction or septic shock, but the Sepsis-3 consensus disputes the existence of this continuum.

The Sepsis-3 consensus asserts that these new definitions and criteria will “facilitate earlier recognition and more timely management of patients with sepsis or at risk of developing sepsis.” That statement seems counterintuitive when the bar for the diagnosis of sepsis has been raised substantially, when sepsis without organ failure is not acknowledged, and when no criteria for “risk” have been offered. Perhaps using the standard of 2 out of the 3 qSOFA criteria will result in diagnostic accuracy comparable to the Sepsis-2 definitions and will result in better outcomes, but perhaps not.

How will other professional societies, the medical community in general, payers, and regulatory agencies respond to these radical changes? The Sepsis-3 consensus concludes that the new definitions “should replace previous definitions” (without specifically stating that they “do” replace them) and that the process “remains a work in progress.” Is it definitive or subject to comment and change? Could or should clinicians have the discretion to continue using Sepsis-2 criteria based on their clinical judgment while awaiting the health care community's reaction to these new definitions? All of this will play out in the real-world setting of everyday medical practice, and only time will tell.