The impact of pelvic and lower extremity fractures on the incidence of lower extremity deep vein thrombosis in high-risk trauma patients

Lower extremity fractures (LEFx) and pelvic fractures (PFx) are believed to increase the risk of lower extremity deep vein thrombosis (LEDVT). We studied trauma patients at high risk for LEDVT to determine whether an increased incidence of LEDVT was associated with LEFx and/or PFx. From January 1995 through December 1997 4163 trauma patients were admitted to our Level I trauma center. One thousand ninety-three patients at high risk for LEDVT were screened with serial lower extremity venous duplex ultrasound. Their medical records were retrospectively reviewed for demographics, mechanism of injury, and fracture data. The occurrence of LEDVT, pulmonary embolus, and LEDVT prophylaxis and treatment were noted. The incidence of LEDVT in the fracture group (Fx) was compared with that in the nonfracture group (NFx) using chi-square analysis and logistic regression. Statistical significance was set at less than or equal to0.05. Complete data were available for 1059 of 1093 patients. Five hundred sixty-nine (53.73%) patients had PFx and/or LEFx, 151 (14.26%) patients had PFx only, 317 (29.3%) patients had LEFx only, and 101 (9.54%) patients had both PFx and LEFx. Four hundred ninety (46.27%) patients had NFx. In 1059 patients LEDVT was detected in 125 (11.8%). Sixty-three patients in the Fx groups developed LEDVT (50.4%): 19 (15.2%) PFx patients, 15 (12.0%) PFx/LEFx patients, and 29 (23.2%) LEFx patients. Sixty-two (49.6%) NFx patients developed LEDVT. LEDVT incidence was not significantly different between the Fx and NFx groups or among the PFx, LEFx, and PFx/LEFx groups (P = 0.317). Nine patients developed pulmonary embolism: four NFx patients, two LEFx patients, two PFx patients, and one PFx/LEFx patient. Significant predictors of LEDVT were age and hospital length of stay. Mean age in patients with LEDVT was 47.58 years and in patients without LEDVT it was 40.89 years (P < 0.001). Mean hospital length of stay in patients with LEDVT was 29.81 days and in patients without LEDVT it was 16.84 days. The power of this study to detect differences representing medium effect sizes was greater than 90 per cent. We conclude that LEFx and/or PFx was not associated with an increased incidence of LEDVT in trauma patients at high risk for LEDVT. Lower extremity venous duplex ultrasound needs to be performed in both Fx and NFx groups to detect LEDVTs.