You’ve heard us on our “soapbox” about sepsis several times. We like to point out that 2 significant drivers may have changed the landscape and created a population of “sepsis” that has a lesser mortality rate. CMS data clearly show that a few years back there was a significant change in the number of pneumonia cases and sepsis cases (Lindenauer 2012). The number of pneumonias decreased and sepsis cases increased. Why? Hospitals (often with the help of the third party "clinical documentation programs" that help billing) began using the definition of sepsis that required two of the SIRS criteria in a patient suspected of having infection in order to have a DRG for sepsis. You could have a patient with pneumonia happily pushing his IV pole up and down the hallway who meets those threshold criteria and now gets a label of "sepsis", which of course gets a significantly higher reimbursement than pneumonia. The same research group showed how such coding changes have affected reported mortality rates for both pneumonia and sepsis (Rothberg 2014). They note that such shifts in coding reduce the pneumonia mortality rate (because sicker patients are shifted to the sepsis or respiratory failure diagnosis codes) and also reduce the mortality rate for sepsis (because the newly added patients are less sick than the average sepsis patient).
Also, CMS readmission penalties apply to pneumonia but not to sepsis. So there was additional pressure on hospitals to see if patients met the sepsis "criteria". And, as pointed out in the recent coding impact article (Rothberg 2014) the Surviving Sepsis Campaign may have increased awareness of sepsis, resulting in more patients with pneumonia being coded as sepsis since they met the SIRS criteria requirement.
So it’s a great relief that the new consensus definition of sepsis does away with the SIRS criteria. A consensus panel has arrived at new definitions for sepsis and septic shock, marking the first refinement in definition in over 15 years (Singer 2016). The JAMA article by Singer et al was accompanied by 2 articles describing the consensus process and development of clinical criteria (Shankar-Hari 2016, Seymour 2016) and an editorial (Abraham 2016). There was apparently a unanimous agreement to eliminate use of the SIRS criteria.
In the new definition sepsis is defined as “life-threatening organ dysfunction caused by a dysregulated host response to infection” and the clinical criteria for diagnosis are organ dysfunction represented by an increase in the Sequential [Sepsis-related] Organ Failure Assessment (SOFA) score of 2 points or more. The SOFA score is familiar to those who work in ICU’s and has apparently been a useful predictor of mortality.
Because the SOFA score requires measurement of lab parameters and is most easily calculated in ICU patients, the panel came up with the qSOFA (“quick” SOFA) score to be used in patients outside the ICU to quickly identify patients with sepsis without the need for lab data. The qSOFA is based on 3 criteria:
(1) a respiratory rate ≥22
(2) altered mentation (Glasgow Coma Scale score of 13 or less) and
(3) systolic blood pressure ≤100 mm Hg.
The new definition for septic shock requires:
(1) hypotension not responsive to fluid resuscitation
(2) requiring vasopressors to maintain a mean arterial BP ≥65 mm Hg, and
(3) serum lactate level >2mmol/L (18 mg/dL)
Models show those meeting the new definition of sepsis have an in-hospital mortality greater than 10% and those meeting the septic shock definition have a mortality greater than 40%.
While we applaud the new definition in that it should exclude those cases of mild infections that simply met 2 SIRS criteria, it will likely take some time before the new definition is integrated into ICD-9 or ICD-10 coding. This will obviously complicate measurement in ongoing pay-for-performance programs or public quality reporting programs using sepsis as a parameter. And it may complicate some ongoing research trials. Probably most importantly, it will impact many of the early warning scores we’ve discussed in many of our prior columns on early recognition of sepsis.
One fact we were, frankly, unaware of before this new paper was published is that culture-positive sepsis is observed in only 30% to 40% of cases. Makes you wonder how many more of the cases of “sepsis” under the old definition were actually related to infection.
We still need to focus on early recognition, timely antibiotics and adequate fluid resuscitation to reduce morbidity and mortality from “sepsis”. But hopefully these new definitions are steps in the right direction and will help us determine better management in the future.
Our other columns on sepsis:
Lindenauer PK, Lagu T, Shieh M-S, et al. Association of Diagnostic Coding With Trends in Hospitalizations and Mortality of Patients With Pneumonia, 2003-2009. JAMA 2012; 307(13): 1405-1413
Rothberg MB, Pekow PS, Priya A, Lindenauer PK. Variation in Diagnostic Coding of Patients With Pneumonia and Its Association With Hospital Risk-Standardized Mortality Rates: A Cross-sectional Analysis. Ann Intern Med 2014; 160(6): 380-388
Singer M, Deutschman CS, Seymour CW, et al. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA 2016; 315(8): 801-810
Seymour CW, Liu VX, Iwashyna TJ, et al. Assessment of Clinical Criteria for Sepsis For the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA 2016; 315(8): 762-774
Shankar-Hari M, Phillips GS, Levy ML, et al. Developing a New Definition and Assessing New Clinical Criteria for Septic Shock for the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA 2016; 315(8): 775-787
Abraham E. New Definitions for Sepsis and Septic Shock. Continuing Evolution but With Much Still to Be Done (editorial). JAMA 2016; 315(8): 757-759