OBJECTIVES: The purpose of this study was to review the MRI examinations of a large group of low-energy trauma patients in whom pelvic MRI had detected radiographically occult fractures, in order to characterize prevailing fracture patterns and determine how often co-existing proximal femoral and pelvic fractures were observed. METHODS: All patients having pelvic MRI over 5 years were identified. Word-search software selected 269 MRI reports containing the term ‘fracture’. Further scrutiny identified 168 with diagnosis of fracture. MRI request and imaging record review identified 102 low-energy trauma cases that had MRI for clinical suspicion of fracture despite normal radiographs. Sixty-six cases were excluded for the following reasons: no expressed clinical suspicion of occult fracture; history suggesting high-energy trauma;skeletal co-morbidity hindering acute fracture identification;interval more than 2 weeks between radiographs and MRI. The 102 study MRI examinations, which employed a limited two-sequence protocol, were reviewed. Any fracture that had not been appreciated on radiographs was recorded and characterized as femoral, pelvic, or co-existing femoral and pelvic fractures. RESULTS: The 102 study cases had a median age of 82 years.The median interval between pelvic radiographs and MRI was 3 days. MRI showed undiagnosed femoral fracture in 48/102 cases (47.1%), sacral fracture in 41/102 (40.2%),and pubic fracture in 55/102 (53.9%). In 11/102 cases(10.8%), MRI showed undiagnosed fractures of both proximal femur and pelvic ring (seven sacral, six pubic bone, two other site fractures). In 10/11 cases with coexisting femoral and pelvic fractures, the femoral fracture was incomplete. CONCLUSIONS: Limited pelvic MRI found a high prevalence of radiographically occult femoral and pelvic fractures in low-energy trauma patients, with clinical suspicion of fracture despite normal radiographs. Co-existing occult femoral and pelvic ring fractures were commonly observed, and in such cases, the femoral fracture was likely to be incomplete and multiple pelvic fractures were typically present.
OBJECTIVES: The purpose of this study was to review the MRI examinations of a large group of low-energy traumapatients in whom pelvic MRI had detected radiographically occult fractures, in order to characterize prevailing fracture patterns and determine how often co-existing proximal femoral and pelvic fractures were observed. METHODS: All patients having pelvic MRI over 5 years were identified. Word-search software selected 269 MRI reports containing the term ‘fracture’. Further scrutiny identified 168 with diagnosis of fracture. MRI request and imaging record review identified 102 low-energy trauma cases that had MRI for clinical suspicion of fracture despite normal radiographs. Sixty-six cases were excluded for the following reasons: no expressed clinical suspicion of occult fracture; history suggesting high-energy trauma;skeletal co-morbidity hindering acute fracture identification;interval more than 2 weeks between radiographs and MRI. The 102 study MRI examinations, which employed a limited two-sequence protocol, were reviewed. Any fracture that had not been appreciated on radiographs was recorded and characterized as femoral, pelvic, or co-existing femoral and pelvic fractures. RESULTS: The 102 study cases had a median age of 82 years.The median interval between pelvic radiographs and MRI was 3 days. MRI showed undiagnosed femoral fracture in 48/102 cases (47.1%), sacral fracture in 41/102 (40.2%),and pubic fracture in 55/102 (53.9%). In 11/102 cases(10.8%), MRI showed undiagnosed fractures of both proximal femur and pelvic ring (seven sacral, six pubic bone, two other site fractures). In 10/11 cases with coexisting femoral and pelvic fractures, the femoral fracture was incomplete. CONCLUSIONS: Limited pelvic MRI found a high prevalence of radiographically occult femoral and pelvic fractures in low-energy traumapatients, with clinical suspicion of fracture despite normal radiographs. Co-existing occult femoral and pelvic ring fractures were commonly observed, and in such cases, the femoral fracture was likely to be incomplete and multiple pelvic fractures were typically present.
Authors: Karen M Verbeeten; Kirstine Lintrup Hermann; Maria Hasselqvist; Gunnar S Lausten; Per Joergensen; Claus Munk Jensen; Henrik S Thomsen Journal: Eur Radiol Date: 2004-07-27 Impact factor: 5.315