Associated injuries and not fracture instability predict mortality in pelvic fractures: a prospective study of 100 patients.

Background: Whether pelvic fracture instability is correlated to mortality in blunt multiply-injured trauma patients is debatable. This is the first prospective study on patients with pelvic fractures aiming at finding whether pelvic fracture type affects mortality.

Methods: There were 100 consecutive patients (77 males, mean age of 31 [3-73] years) studied between September 2003 and October 2004. Data were collected regarding mechanism of injury, associated injuries, Injury Severity Score (ISS), Revised Trauma Score, blood transfusions, and mortality. The fractures were classified according to instability, where type O is stable, type R is rotationally unstable, and type RV is both rotationally and vertically unstable. Because a pure acetabular fracture is a single break in the pelvic ring, we classified it as type O. Computer tomography was used for fracture classification in 73 patients and plain X-rays in 27 patients.

Results: There were 77 fractures caused by road traffic collisions. Type O fractures (n = 63) had lower median ISS (13 [4-48]) than type R (n = 19) (18 [9-75]) and type RV (n = 18) (18 [6-66]) (p = 0.019, Kruskall Wallis). There was no significant difference in ISS between type R and RV fractures. A logistic regression model has shown that ISS was the only significant factor that predicts mortality.

Conclusions: ISS is the most important predictor in defining mortality in patients with pelvic fracture and not the type of pelvic instability.