OBJECTIVE: To compare diagnostic accuracy and interobserver reliability of Magnetic Resonance Imaging (MRI) and Multidetector Computed Tomography (MDCT) in the detection of acute pelvic fractures. MATERIALS AND METHODS: In 38 consecutive patients (mean age 74.7 years) with a positive finding of anterior pelvis fracture in the conventional X-rays, pelvic MRI and MDCT were performed for further evaluation of pelvic ring fractures. Two radiologists independently read all data sets. Sensitivity and specificity were calculated based on mean scores for each method. Sensitivities of CT and MRI were compared using a paired proportion test (McNemar). Diagnostic validity of both methods was assessed by the interobserver variability using kappa statistics. Combined clinical data and findings from all imaging studies served as the reference standard. RESULTS: 122 fractures were identified in the reference standard (37 sacral, 58 pubic, 22 acetabular, 1 ischial, 4 ilial). On average, MRI detected 96.3% whereas CT detected 77% of all fractures. With regard to sensitivity, MRI proved to be significantly better compared to MDCT (observer 1, p=0.0009; observer 2, p=0.0003 by observer 2). In particular, MRI performed better in the depiction of sacral fractures, reaching a sensitivity of 98.6% compared to 66.1% at CT. The interobserver variability was determined to be very good (k=0.955 for MRI and 0.902 for MDCT). CONCLUSION: MRI reaches a significantly higher sensitivity than CT in the detection of acute pelvic fractures, particularly of the sacrum. Especially in elderly patients with suspicion of a sacral fracture and negative results at MDCT, MRI may be considered as the next step in diagnostic workup.
OBJECTIVE: To compare diagnostic accuracy and interobserver reliability of Magnetic Resonance Imaging (MRI) and Multidetector Computed Tomography (MDCT) in the detection of acute pelvic fractures. MATERIALS AND METHODS: In 38 consecutive patients (mean age 74.7 years) with a positive finding of anterior pelvis fracture in the conventional X-rays, pelvic MRI and MDCT were performed for further evaluation of pelvic ring fractures. Two radiologists independently read all data sets. Sensitivity and specificity were calculated based on mean scores for each method. Sensitivities of CT and MRI were compared using a paired proportion test (McNemar). Diagnostic validity of both methods was assessed by the interobserver variability using kappa statistics. Combined clinical data and findings from all imaging studies served as the reference standard. RESULTS: 122 fractures were identified in the reference standard (37 sacral, 58 pubic, 22 acetabular, 1 ischial, 4 ilial). On average, MRI detected 96.3% whereas CT detected 77% of all fractures. With regard to sensitivity, MRI proved to be significantly better compared to MDCT (observer 1, p=0.0009; observer 2, p=0.0003 by observer 2). In particular, MRI performed better in the depiction of sacral fractures, reaching a sensitivity of 98.6% compared to 66.1% at CT. The interobserver variability was determined to be very good (k=0.955 for MRI and 0.902 for MDCT). CONCLUSION: MRI reaches a significantly higher sensitivity than CT in the detection of acute pelvic fractures, particularly of the sacrum. Especially in elderly patients with suspicion of a sacral fracture and negative results at MDCT, MRI may be considered as the next step in diagnostic workup.
Authors: Andrew B Ross; Brian Y Chan; Paul H Yi; Michael D Repplinger; David J Vanness; Kenneth S Lee Journal: Skeletal Radiol Date: 2018-06-18 Impact factor: 2.199
Authors: Thomas Mendel; Bernhard Wilhelm Ullrich; Gunther Olaf Hofmann; Philipp Schenk; Felix Goehre; Stefan Schwan; Friederike Klauke Journal: Eur J Trauma Emerg Surg Date: 2020-09-02 Impact factor: 3.693
Authors: W Lehmann; M Hoffmann; D Briem; L Grossterlinden; J P Petersen; M Priemel; P Pogoda; A Ruecker; J M Rueger Journal: Eur J Trauma Emerg Surg Date: 2012-09-25 Impact factor: 3.693
Authors: F O Henes; M Groth; T A Bley; M Regier; J V Nüchtern; H Ittrich; A Treszl; G Adam; P Bannas Journal: Eur Radiol Date: 2012-05-15 Impact factor: 5.315