The proposal is aimed at identifying genetic factors that determine the incidence and severity of, and the outcome from life-threatening infections (severe sepsis/septic shock) in patients admitted to High Dependency Units (HDUs) or Intensive Care Units (ICUs) with pneumonia which developed outside the hospital (community acquired pneumonia - CAP) or contamination of the abdominal cavity with faeces due to a leak in the bowel (faecal peritonitis). This will require the acquisition of a large, high quality resource of genetic material (DNA), plasma, urine, white blood cells and clinical information from well characterized groups of similar patients with, or at risk for, severe sepsis/septic shock. The principal objective is to perform studies which are sufficiently large to establish beyond doubt the influence of a series of selected "candidate" genes on the development, progress and outcome of sepsis.

The investigators plan to recruit 2,000 cases of community acquired pneumonia (CAP) and 2,000 cases of faecal peritonitis (FP) from 30 UK ICUs and HDUs (members of the UK Critical Care Genomics group - UKCCG). The large number of patients is required to satisfy the power calculations based upon the predicted allele frequencies of candidate genes and the level of functional expression of the gene polymorphisms.

If a patient is eligible, written, informed consent will be obtained either from the patient or, if the patient is incompetent via the patient's legal representative. Patients will be characterized clinically in terms of admission diagnosis, severity of illness (APACHE II), organ failures (SOFA) and final outcome (ICU and hospital mortality, death or survival 6 months following ICU admission. Date of death will be recorded when appropriate). Clinical status will be assessed daily for days 1, 2, 3, 5 and 7 of ICU admission using the Sepsis criteria, SOFA score, microbiological culture results and antibiotic therapy.

Selected ICUs will, following consent, also undertake blood and urine sampling on days 1, 3 and 5 for genomic, proteomic and metabolomic studies. Information will be recorded on a bar-coded paper clinical report form (CRF) at the bedside. The CRFs will be securely stored locally and copied to the research coordinator, where they will be archived. The data will be entered independently by the research coordinator and one of the investigators into a secure, central web-based electronic database for storage of clinical data and the calculation of derived values. The patient codes for genetic analysis will be derived directly from the clinical database. Those undertaking genotyping will be blinded to the clinical details and these two databases will be brought together at the time of analysis only under the direct supervision of one of the principal investigators.

Inclusion Criteria:

• Patient, or legal representative, is able to give informed consent
• Male or female of 18 years or more
• Patient admitted to ICU/HDU with faecal peritonitis or community acquired pneumonia

Exclusion Criteria:

• Patient or legal representative is unwilling to consent
• Patient is under the age of 18 years
• Patient is already enrolled in an interventional study
• Patient is immunocompromised
• Patient is pregnant
• There is an advance directive to withhold or withdraw life-sustaining treatment or patient is admitted for palliative care only.

Patients admitted to Intensive Care Units or High Dependency Units in the UK with faecal peritonitis or community acquired pneumonia