A quantitative analysis of the injury severity of 40 patients with open pelvic fractures was performed. Data were analyzed using the Statistical-Analysis-System (SAS Institute Inc., Cary, NC) with regard to patients' age, gender, trauma mechanism, classification and nature of the pelvic injury, associated lesions, and mortality. Trauma-scoring included the Hannover-Polytrauma-Score [11], the Pelvis-Fracture-Scale [2] and the Pelvis-Score [19]. Five patients died. The survivors had a mean Polytrauma-Score of 34.1, the nonsurvivors had a mean Polytrauma-Score of 44.6 (p = 0.7; Mann-Whitney-test). The nonsurvivors required highly significant more units of blood transfusions than the survivors (49.4 vs. 14.1; p = 0.003; Mann-Whitney-test). The loss of blood was related to the severity of the injury according to the Polytrauma-Score. There was no significant difference in the Pelvis-Fracture-Scale of survivors and nonsurvivors. Neither there was a significant correlation between the fracture type and the Polytrauma-Score nor between the fracture type and the mortality. The Pelvis-Score--with the variable "bleeding" defined as "major vessel lesion"--was significantly higher in the nonsurvivors than in the survivors (12.0 vs. 4.9; (p = 0.04; Mann-Whitney-test). In conclusion, in this retrospective study the Polytrauma-Score and the Pelvis-Score proved effective regarding some important aspects of the prognostic estimation of the general injury and the pelvic trauma, respectively. The Hannover-Pelvis-Fracture-Scale allows an exact documentation of the pelvic trauma as it pays proper attention not only to the fracture classification but especially to the soft tissue damage.