Non-invasive detection of infection in acute pancreatic and acute necrotic collections with diffusion-weighted magnetic resonance imaging: preliminary findings.
- Resource Type
- Academic Journal
- Authors
- Islim F; Department of Radiology, Istanbul Bakirkoy, Dr. Sadi Konuk Training and Research Hospital, Istanbul, Turkey, fislim@yahoo.com.; Salik AE; Bayramoglu S; Guven K; Alis H; Turhan AN
- Source
- Publisher: Springer International Country of Publication: United States NLM ID: 9303672 Publication Model: Print Cited Medium: Internet ISSN: 1432-0509 (Electronic) Linking ISSN: 09428925 NLM ISO Abbreviation: Abdom Imaging Subsets: MEDLINE
- Subject
- Language
- English
Purpose: The purpose of this study was to evaluate the contribution of diffusion-weighted magnetic resonance imaging (DW-MRI) to the detection of infection in acute pancreatitis-related collections.
Methods: A total of 21 DW-MRI, and computed tomography (CT) were performed on 20 patients diagnosed as acute pancreatitis with acute peri-pancreatic fluid or necrotic collections. Collections were classified as infected or sterile according to the culture and follow-up results. Collections with gas bubbles on CT images were considered to be infected. Collections with peripheral bright signals on DW-MRI images were considered to be positive, whereas those without signals were considered to be negative. Apparent diffusion coefficient (ADC) values of the peripheral and central parts of the collections were measured. Student's t test was used to compare the means of ADC values of independent groups.
Results: Apart from one false positive result, the presence of infection was detected by DW-MRI with 95.2% accuracy. The sensitivity and accuracy of DW-MRI were higher than CT for the detection of infection. The ADC values in the central parts of the collections were significantly different between the infected and sterile groups.
Conclusion: DW-MRI can be used as a non-invasive technique for the detection of infection in acute pancreatitis-associated collections.