CT diagnosis of Fitz-Hugh and Curtis syndrome: value of the arterial phase scan

Abstract

Objective: We wanted to evaluate the role of the arterial phase (AP) together with the portal venous phase (PP) scans in the diagnosis of Fitz-Hugh-Curtis syndrome (FHCS) with using computed tomography (CT).

Materials and methods: Twenty-five patients with FHCS and 25 women presenting with non-specifically diagnosed acute abdominal pain and who underwent biphasic CT examinations were evaluated. The AP scan included the upper abdomen, and the PP scan included the whole abdomen. Two radiologists blindly and retrospectively reviewed the PP scans first and then they reviewed the AP plus PP scans. The diagnostic accuracy of FHCS on each image set was compared for each reader by analyzing the area under the receiver operating characteristic curve (Az). Weighted kappa (wk) statistics were used to measure the interobserver agreement for the presence of CT signs of the pelvic inflammatory disease (PID) on the PP images and FHCS as the diagnosis based on the increased perihepatic enhancement on both sets of images.

Results: The individual diagnostic accuracy of FHCS was higher on the biphasic images (Az = 0.905 and 0.942 for reader 1 and 2, respectively) than on the PP images alone (Az = 0.806 and 0.706, respectively). The interobserver agreement for the presence of PID on the PP images was moderate (wk = 0.530). The interobserver agreement for FHCS as the diagnosis was moderate on only the PP images (wk = 0.413), but it was substantial on the biphasic images (wk = 0.719).

Conclusion: Inclusion of the AP scan is helpful to depict the increased perihepatic enhancement, and it improves the diagnostic accuracy of FHCS on CT.

Fig. 1

Receiver operating characteristic curves for reader 1 (A) and reader 2 (B) for the diagnosis of Fitz-Hugh-Curtis syndrome. The accuracy of the arterial phase plus portal venous phase set showed high statistical significance compared with using only the portal venous phase set for reader 2 (p = 0.0003) (B), and it nearly approached statistical significance for reader 1 (p = 0.0516) (A). The overall accuracy was superior for the arterial phase plus portal venous phase image set in respect to using only the portal venous phase set.

Fig. 2

Axial contrast-enhanced CT scan in a 24-year-old woman with right upper quadrant pain and fever, which is a true positive example of Fitz-Hugh-Curtis syndrome. A. The arterial phase scan reveals conspicuous homogenously increased perihepatic enhancement on the right lobe of the liver. B. Portal venous phase scan reveals inconspicuous enhancement.

Fig. 3

Axial contrast-enhanced CT scan in a 40-year-old woman with right upper quadrant pain and fever, which is a true positive example of Fitz-Hugh-Curtis syndrome. A. Arterial phase scan reveals conspicuous increased perihepatic enhancement on the right lobe of the liver. B. Portal venous phase scan reveals conspicuous identical enhancement.

Fig. 4

Axial contrast-enhanced CT scan in a 25-year-old woman with mild pelvic inflammatory disease as the final diagnosis, which is a false positive example of Fitz-Hugh-Curtis syndrome. A. Arterial phase scan reveals the increased perihepatic enhancement at the anterior portion of the right lobe of the liver. B. Perihepatic enhancement cannot be seen on the portal venous phase scan.

Fig. 5

Axial contrast-enhanced CT scan in a 49-year-old woman with right upper quadrant pain and fever, which is a false negative example of Fitz-Hugh-Curtis syndrome. A. Arterial phase scan reveals a diffuse heterogenous perihepatic enhancement on the entire liver. B. Follow up arterial phase scan 2 weeks later reveals no enhancement.