To determine the association between site of infection and acute organ dysfunction at the time of admission in ICU patients.
Using the PROJECT IMPACT (PI) database, 69 medical patients with infection on ICU admission from July 2003 to February 2004 were identified. All were admitted to the ICU from the emergency department or were transferred from another ED. Patients were divided into two groups: a)acute organ dysfunction (by predetermined criteria)on admission (OD, N=41) and b)no acute organ dysfunction on admission(NOD, N=28). The presence of organ dysfunction was examined relative to site of infection (PI criteria) for patients with sinle sites of infection (33 OD, 24 NOD) and for all patients with infection (patients with both single and multiple sites of infection) (41 OD, 28 NOD)using the Fisher’s exact test.
41/69 (59%) patients admitted to the ICU with infection had organ dysfunction at the time of admission. No sites of infection, alone or in combination, were associated with a greater incidence of organ dysfunction at admission. Chest infection was present in a significantly greater percentage of patients without vs with OD (TableSite of InfectionOD groupNOD groupP valueABlood stream infection only5/331/24NSBlood stream infection with other infections10/413/28NSBVascular access infection only1/330/24NSVascular access infection w/other infection2/410/28NSCBlood stream (BS) or vascular access (VA) infection6/331/24NSBS or VA along with other infections5/413/28NSDUrinary tract infection only11/335/24NSUrinary tract infection with other infection16/417/28NSEChest infections only7/3315/24<0.003Chest infections with other infections10/4118/280.001). This was not explained by acuity as APACHE II(SD) was similar for patients with chest infection alone with and without(OD): 17.6 (5.07) vs 20.4 (12.01); chest infection alone without OD vs all patients with OD: (17.6 (5.07) vs 20.29 (7.79); and all chest infection vs all other patients regardless of OD status: 17.46 (5.71) vs 19.24 (8.05).
Organ dysfunction in infected patients is common at the time of admission to the ICU, but site of infection does not correlate with the presence of dysfunction. The presence of chest infection appears to be associated with a lower likelihood of organ dysfunction at time of admission.
Early identification of organ dysfunction in patients admitted to the ICU may impact upon treatment and outcome and should be pursued in patients with infection at the time of admission regardless of site.
S. Patel, None.