Objective: The objective of this study was to prospectively assess the added value of gadolinium-enhanced and diffusion-weighted (DWI) magnetic resonance imaging (MRI) to pelvic T2-weighted MRI for diagnosis of deep pelvic abscesses.
Methods: Twenty-nine patients with suspected abscess and a control group of 43 patients underwent T2-weighted, gadolinium-enhanced, T1-weighted and DWI magnetic resonance sequences. Three readers (R1, R2, R3) scored likelihood of abscess on standard MRI, standard MRI + gadolinium, and standard MRI + DWI.
Results: Twenty-nine patients had 36 abscesses. On standard MRI, R1 achieved area under the receiver operating characteristic (ROC) curve (AUC) of 0.97, sensitivity 92%, and specificity 100%. For R2, these figures were 0.87, 81%, and 100%, and for R3, these were 0.85, 83%, and 79%. After gadolinium, R2 improved AUC to 0.97 (P = 0.005), and R3 to 0.95 (P = 0.006). Standard MRI + DWI yielded improved AUC for all readers (P = 0.15, 0.001, and 0.001 for R1, R2, R3, respectively).
Conclusions: Addition of gadolinium or DWI to T2-weighted MRI improves performances for the diagnosis of deep pelvic abscess. Diffusion-weighted imaging may replace gadolinium.
From the *Department of Radiology, Hôpital Robert Debré, Reims; †Department of Radiology, Hôpital Lariboisière, Paris; and Departments of ‡Statistics and §Obstetrics and Gynaecology, Hôpital Robert Debré, Reims, France.
Received for publication September 11, 2012; accepted February 6, 2013.
Reprints: Truong Luong Francis Nguyen, MD, Department of Radiology, Hôpital Robert Debré, 51092 Reims Cedex France (e-mail: email@example.com).
The authors have not received any funding for this research.
The authors have no conflict of interest to report.