Between 6 and 16% of patients presenting to hospital emergency departments have infections, with half of these having signs of systemic inflammation (known as 'sepsis'). A second issue is that, at time of presentation, it can be difficult to determine who has inflammation as a result of infection and who does not.

Some of the patients with infections will deteriorate to organ failure ('severe sepsis') including failure of the heart and blood vessels to maintain normal blood pressure ('septic shock'). Septic shock as arguably the most dangerous form of severe sepsis is associated with a significant mortality, which can be reduced by early intervention. However identifying those patients who are at high risk of deteriorating to septic shock can be difficult on initial presentation to hospital, and thus these patients risk being 'triaged' to an inappropriate level of care and/or missing the crucial early interventions which can modify mortality. Equally failure to identify which patients have underlying infections can lead to potential inappropriate targeting of antibiotics. Existing clinical and laboratory tests are often unable to accurately identify those patients with infection, and those who are likely to deteriorate to severe sepsis and septic shock.

Investigators in this group have recently identified several signatures of immune system activation which predict those patients who are likely to deteriorate, and which patients with suspected infection subsequently have this confirmed. Such tests would have major benefits for the management of patients with early suspected infection and sepsis if they can be translated into a test usable in everyday clinical practice. This study aims to determine the prevalence of these markers in a cohort of patients admitted with suspected sepsis, and their predictive ability for developing established septic shock. From this investigators aim to derive an optimal test, to be tested in a validation cohort (ExPRES-Sepsis II) which will be suitable for everyday clinical practice, and thus take the next step towards developing a market-ready test.

Study hypothesis is:

Measurement of markers of immune activation will allow i) Risk stratification for deterioration into severe sepsis ii) Risk stratification for death amongst patients presenting with sepsis iii) Identification of patients with confirmed sepsis

Inclusion Criteria:

• Age >16 (>18 in England)
• SIRS criteria met (2 or more of White Cell Count (WCC) >11 or <4, Heart Rate (HR) >90, Respiratory Rate (RR) >20 or temp >38 or <36oC)
• Clinical suspicion of sepsis (cultures taken or antibiotics started)
• Enrolled within 12 hours of hospital admission

Exclusion Criteria:

• Acute pancreatitis
• Septic shock at time of enrolment
• Severe organ failure at time of enrolment (immediate requirement for ventilation, vasopressor or renal replacement therapy)
• Haematological malignancy
• Recent chemotherapy (past 2 weeks)
• Myelodysplastic syndromes
• Known neutropaenia
• HIV infection
• Pregnancy
• Blood transfusion >4 units in past week
• Oral Corticosteroids for >24 hours prior to enrolment
• Decision not for active therapy/ palliative care at admission
• Lacking in capacity to consent and nearest relative/welfare guardian not available for consultation and proxy consent (Scotland only)

Patients in the initial 12 hours following presentation to hospital with suspected sepsis, i.e. patients with signs of systemic inflammation (i.e. meeting criteria for SIRS) where the treating clinician takes microbial samples and/or starts empiric antibiotics.