(P78) Predictive Factors of Infection Source on the Combined Chest and Abdominal Computed Tomography in Febrile Elderly Patients with No Clinical Clue in the Emergency Department: An Observational Study

Objectives: The purpose of this study is to evaluate the utility of the combined chest and abdominal computed tomography (CT) in febrile elderly patients with no clinical clue in the emergency department (ED) and to find predictors that can identify infection source on the CT.

Background: In patients with unclear causes of fever, if vital signs are unstable or if severe infection and sepsis are suspected due to high levels of inflammatory markers including C-reactive protein (CRP) and procalcitonin (PCT), CT is performed to clearly identify the cause of fever and use of appropriate antibiotics. the authors investigated whether combined chest and abdominal CT (systemic CT) scan performed in febrile elderly patients without any clinical clue to the cause of fever in the ED was useful in finding infection source, and there is factor that can predict the presence of infection source on the systemic CT.

Methods: We reviewed medical records of febrile elderly patients without any clinical clue who underwent combined chest and abdominal CT in the ED between January 2017 to December 2018. Fever in the elderly patients was defined as tympanic temperature ≥ 37.2℃ or increase in body temperature ≥ 1.3℃, and no clinical clue of fever was defined as no findings suggestive of the cause of fever in medical history taking, review of systems, physical examination, and emergency diagnostic tests including various blood tests, urinalysis and simple chest radiography.

Results: Total 110 febrile elderly patients without clinical clue who underwent systemic CT in the ED were included. 56 patients were classified as negative infection source on the CT group, and 54 patients classified as positive infection source group. The rates of identifying infection source by systemic CT performed in febrile elderly patients without clinical clue in the ED was 49.1%. From multivariate logistic regression, the underlying diabetes mellitus [odds ratio (OR) = 3.097, 95% confidence interval (CI): 1.146–8.364, p-value = 0.026], malignancy [OR= 12.973, 95% CI:2.433–69.168, p-value = 0.003], CRP (≥5.16 mg/L) [OR= 9.594, 95% CI: 3.157–29.159, p-value < 0.001] and q-SOFA score (≥2 point) [OR= 8.818, 95% CI: 2.597–29.947, p-value = 0.001] were significant independent predictors of the positive infection source on the combined chest and abdominal CT in febrile elderly patients without any clinical clue in the ED.

Conclusions: For febrile elderly patients with no clinical clues in the ED, with underlying diabetes, malignancy, CRP is greater than 5.16 mg/L, or q-SOFA score of 2 or higher, combined chest and abdominal CT to identify infection source can be considered when clinical benefit is judged to be high.

References (Optional):

Funding (Optional): none