A new retrospective cohort study compared the performance of the systemic inflammatory response syndrome (SIRS) criteria, the sepsis-related organ failure assessment (SOFA) score, and the quick SOFA (qSOFA).
Researchers used data from a database of 30,239 U.S. adults ages 45 years or older. During a follow-up period from Feb. 5, 2003, to Dec. 31, 2012, 2,593 of the participants were hospitalized for an infection for the first time. The study authors calculated the scores and criteria for the hospitalized, infected patients and assessed the association with in-hospital mortality and one-year mortality. Results were published in The Lancet Infectious Diseases on March 3.
Of the 2,593 patients, 1,526 met SIRS criteria, 1,080 met SOFA criteria, and 378 met qSOFA criteria. In-hospital mortality was highest in patients who met qSOFA criteria (23%), followed by SOFA (13%) and SIRS (9%). The same was true for the outcome of one-year mortality. The number of deaths per 100 person-years was 29.4 in the qSOFA group (95% CI, 22.3 to 38.7) compared to 22.6 in the SOFA group (95% CI, 19.2 to 26.6) and 14.7 in the SIRS group (95% CI, 12.5 to 17.2).
“Our findings support the use of the SOFA and qSOFA classifications to identify patients with infection who are at elevated risk of poor outcomes,” the study authors concluded. The results also show that the different scores identify different, overlapping groups of patients, with SOFA and qSOFA being elevated disproportionately in older patients with comorbidities, the authors said. The study also evaluated incidence over time and found that incidence of meeting qSOFA and SOFA criteria is increasing more than SIRS, which has been shown by previous research and may result from changes in coding and measurement practices.
“Our results do not address the feasibility of incorporating the Sepsis-3 classifications into routine clinical care or the potential effect on patient outcomes,” the authors cautioned. An accompanying editorial praised the study as a classic among the many recent analyses of the new sepsis criteria but offered additional cautions. “The retrospective analysis of hospital data might be more clinically applicable to the provider reviewing the past days' worth of data on a Monday morning than to the emergency provider needing to make rapid decisions using available data,” the editorialists wrote.