Guest Post: Organizations respond to sepsis changes

by James S. Kennedy, MD, CCS, CDIP

This October celebrates the eight month anniversary of the February release of the intellectually stimulating, yet controversial, third international consensus definitions for sepsis and septic shock (Sepsis-3). This new definition has generated significant discussions among those invested in ICD-10-CM-based documentation and coding integrity and compliance.

As we negotiate the fiscal year (FY) 2017 ICD-10-CM Official Guidelines for Coding and Reporting and revisions to the CMS Severe Sepsis/Septic Shock Core Measure (SEP-1), we in coding compliance must develop strategies that preserve the clinical integrity of the definition and diagnosis of sepsis in clinical care, support the proper cohort selection for SEP-1, and compliantly assign defendable ICD-10-CM codes based on provider documentation and coding conventions.

Authored by 19 critical care physicians and endorsed by many critical care societies, including the United States-based Society for Critical Care Medicine and the American Association of Critical Care Nurses, Sepsis-3’s goals were to better differentiate sepsis from uncomplicated infections, and to update definitions of sepsis and septic shock as to be consistent with improved understanding of their pathobiology. In recognizing that sepsis is a clinical syndrome without a validated diagnostic test, the committee sought to promulgate clinical criteria that could be standardized as to meet their objectives.

In so doing, the Sepsis-3 committee redefined sepsis as a “as life-threatening organ dysfunction caused by a dysregulated host response to infection,” which eliminated the concept of sepsis as a systemic inflammatory responses syndrome due to infection which was established with SEP-1 in 1991.

SEP-1 was built into ICD-9-CM in 2001 (though systemic inflammatory response syndrome (SIRS) due to infection cannot be coded as sepsis in ICD-10-CM), required only two out of four simple criteria (temperature above 101° F, WBC count over 12,000, tachycardia, and tachypnea), and did not require organ dysfunction to be present. The new Sepsis-3 also removed the SEP-1 term “severe sepsis,” which is sepsis with acute organ dysfunction.

In identifying a “life-threatening organ dysfunction” for the purpose of diagnosing sepsis, Sepsis-3 changed the Sepsis-related Organ Failure Assessment (SOFA) score to two or more. On the other hand, the CMS SEP-1 Core Measure, severe sepsis, and septic shock bundle, use different criteria than SOFA in defining severe sepsis (criteria available here), and relies on the documentation and coding of the word “severe sepsis,” no matter how its defined.

While agreeing with Sepsis-3 in concept, the Surviving Sepsis Campaign (SSC) rebutted that other clinical indictors of organ dysfunction besides SOFA, such as a lactate level over two, ileus, or sepsis-induced hypotension, should also meet the new criteria for sepsis. View the SOFA criteria and the SSC clarification.

One would think that this thoughtful conclusion would be welcomed by clinicians, clinical documentation improvement (CDI) specialists, and coders alike, much like the acceptance of the 2012 Academy of Nutrition and Dietetics and American Society for Parenteral and Enteral Nutrition’s criteria for adult malnutrition, or the 2013 Kidney Disease: Improving Global Outcomes’ criteria for acute kidney injury. Not so. Almost immediately upon the publication of Sepsis-3, Steven Simpson, MD, of the Division of Critical Care Medicine at the University of Kansas in Lawrence, Kansas, wrote a strong rebuttal and advocated that clinicians not adopt the new definition. Read his reply here.

The comment and response section for the July 26, 2016 issue of the Journal of the American Medical Association contained similar rebuttals, including a statement by Lemeneh Tefera, MD, of CMS in Windsor Mill, Maryland, that states, “although the task force’s definition structure may identify patients with the highest likelihood of poor outcomes, it does not clearly identify patients in the early stages of sepsis when rapid resuscitation provides the greatest patient benefit and improves survival.” Read his comment here.

The Advisory Board of the American Clinical Documentation Improvement Specialists also urged caution; read their position paper. As of September 5, AHIMA sponsored only one article in its CodeWrite newsletter, available to AHIMA members, which focused on the new definition and emphasized that background material published in Coding Clinic cannot be used as clinical criteria for code assignment.

I’m not aware of any forthcoming changes to ICD-10-CM as a result of Sepsis-3; if there are, we won’t see them until at least October 1, 2017. Therefore, at time of press, unless Coding Clinic publishes advice that changes the landscape, we now have a Sepsis-3 definition in a Sepsis-1, Sepsis-2, SSC, or ICD-10-CM definition, documentation, or coding environment.

Editor’s note: This post is an excerpt from an article originally published in JustCoding. Click here to read the full version.