Patient population

Study inclusion criteria were: 1) adult ED patients (age 18 years or older) meeting sepsis criteria. Sepsis was defined as the presence of an infection (defined both by the clinical treating team and through confirmatory case review) and the presence of at least 2 Systemic Inflammatory Response Syndrome (SIRS) criteria.[7]. Furthermore, for the derivation set, the septic patients had to survive the septic event for the 90-day RNA blood draw to occur. Exclusion criteria were inability to obtain written informed consent and absence of an infection. We preferentially recruited patients meeting severe sepsis (sepsis plus organ dysfunction) or septic shock (sepsis plus systolic blood pressure (SBP) < 90 mmHg after a 20 cc/kg fluid challenge).

Five patients with sepsis were included in the derivation set. We collected blood during their ED presentation for sepsis, as well as at least 90 days later (while they were out of the hospital), after recovery from their illness. This approach allowed patients to serve as their own internal control by comparing an individual to themselves during sepsis and non-sepsis states. During the 90-day follow-up draw, patients needed to be without infection which was defined as fulfilling following criteria: 1) no antibiotics either current or within the past 2 weeks; 2) no reported fever or temperature > 100.4°F; and 3) no hospitalization due to an infection within the last two weeks prior to the follow-up blood draw. This subset population was the basis for the discovery set using deep RNA-sequencing (Supplemental Figure 1).

The validation set consisted of 18 patients with sepsis (including severe sepsis and septic shock) and 25 non-infected ED “controls”. Controls were non-infected patients from the ED. We selected patients from our general ED population, including patients with comorbid diseases in order to have a control cohort that had a similar demographic as the sepsis population.