Portal phase alone is equivalent to multiphasic phase for CT diagnosis of acute non-traumatic pains in an emergency context.

Abstract

To evaluate radiological diagnosis concordance between a simplified and a multiphasic computed tomography (MCT) protocol for patients presenting acute non-traumatic abdominal pains (ANTAE). During five consecutive months, all patients admitted in an emergency department for ANTAE were retrospectively included if they underwent MCT, including at least pre-contrast phase, late arterial phase (LAP), and portal venous phase (PVP). Clinical cases of suspected hemorrhagic conditions were secondarily excluded. For the study, two image sets, pre-contrast phase + LAP + PVP ± late phase called S1 and PVP alone called S2, were reviewed independently to give the most appropriate diagnosis with 5-point confidence scale. Diagnosis concordance and radiation dose were compared for each set of protocol by chi-square test. Linear mixed model was used to assess changes of diagnostic confidence and radiation dose. All in all, 196 patients were included. The kappa coefficient between S1 and S2 was excellent (98.5%, CI95% 95.6-99.7). Three errors due to an inappropriate protocol were observed (1.5%; CI95% = - 0.2 to 3.2%), 2 related to biliary tract obstruction causes and one due to gastric bleeding not suspected on clinical data. S2 was associated with a 61% decrease of the radiation dose (p = 0.01) with a mild decrease of the confidence scale (4.54 ± 0.05 versus 4.74 ± 0.03, p = 0.001). Using PVP-CT alone or MCT is equivalent for the diagnosis of ANTAE if suspected acute hemorrhages are excluded. A simplified CT protocol is associated with a dose decrease of 61%.