Effect of angioembolisation versus surgical packingon mortality in traumatic pelvic haemorrhage： Asystematic review and meta-analysis

摘要

BACKGROUND： The management of complex pattern of bleeding associated with pelvic traumaremains a big challenge for trauma surgeons. We aimed to conduct a comprehensive meta-analysis tocompare the outcomes of angioembolisation and pelvic packing in patients with pelvic trauma.METHODS： We conducted a systematic search of electronic information sources, includingMEDLINE; EMBASE; CINAHL; the CENTRAL; the World Health Organization International ClinicalTrials Registry; ClinicalTrials.gov; ISRCTN Register, and bibliographic reference lists. The primaryoutcome was defined as mortality. Combined overall effect sizes were calculated using randomeffectsmodels. Results are reported as the odds ratio （OR） and 95% confi dence interval （CI）.RESULTS： We identified 3 observational studies reporting a total of 120 patients undergoingangioembolisation （n=60） or pelvic packing （n=60） for pelvic trauma. Reporting of the Injury SeverityScore （ISS） was variable, with higher ISS in the pelvic packing group. The risk of bias was low intwo studies, and moderate in one. The pooled analysis demonstrated that angioembolisation did notsignifi cantly reduce mortality in patients with pelvic trauma compared to surgery （OR=1.99; 95% CI=0.83-4.78, P=0.12）. There was mild between-study heterogeneity （I2=0%, P=0.65）.CONCLUSION： Our analysis found no signi f icant di fference in mortal i ty betweenangioembolisation and pelvic packing in patients with traumatic pelvic haemorrhage. The currentlevel of evidence in this context is very limited and insuffi cient to support the superiority of a treatmentmodality. Future research is required.