Abstract
Purpose of the study: These fractures, and the patients, are generally unstable. Mortality associated with these fractures remains high. It is mainly due to the haemorrhagic risk of the presacral venous plexus and the iliac system. Different techniques have been described to control the haemorrhage: pelvic girdle, embolisation, ligature of the iliac arteries, pelvic packing, pelvis clamp or external fixator. Our objective was to analyse our series of fractures of this type in order to optimise patient outcome.
Material and methods: A prospective study was undertaken from January 2003 to December 2006. Among 450 multiple injury patients, 68 presented an unstable fracture of the pelvis, type B or C. The 38 patients included in this series were haemodynamically unstable. The mean ISS for these patients was 53, mean age 38.6 years (range 24–51). Fractures were diagnosed on plain x-rays of the pelvis, ap view, completed by a total body scan.
Results: All patients were victims of high-energy traffic accidents and were managed using the ATLS protocol. Five patients died early despite intensive care. The patients were divided into three groups: group X: 19 patients treated with a first-intention external fixator, with or without arteriography, 18 patients survived, 94%; group Y: 8 patients treated with a first-intention external fixator with arteriography and followed by laparotomy, 7 patients survived, 87%; group Z: 6 patients had laparotomy without an external fixator, 6 patients died, 100% mortality.
Conclusion: In our experience, the best way to control bleeding associated with unstable fractures of the pelvis is as follows: pelvic girdle at the scene of the accident to the emergency room, emergency external fixation followed by laparotomy if the ultrasound is positive. False positives occur due to suffusion of the retroperitoneal haematoma. Emergency laparotomy without prior external fixation of the pelvis lead to 100% mortality in our series. Similarly pelvic packing or the retroperitoneal approach cannot be proposed without exploration.
Correspondence should be addressed to Ghislaine Patte at sofcot@sofcot.fr