Session III - Polytrauma Fracture Healing

Timing of Femur Fracture Fixation in Patients with Head Injuries

Jeffrey Mark Smith, MD; Torin J. Cunningham, MD, University of California San Diego, San Diego, CA

Purpose: To determine whether surgical fixation of femoral (neck, intertrochanteric, subtrochanteric, or shaft) fractures within 24 hours of admission increases the risk of adverse neurologic or pulmonary complications in patients with coincident head injuries.

Methods: A retrospective review of the records of all patients admitted to a Level I trauma center between 1992 and 1997 with associated head injury and femur fracture were reviewed. Patients were excluded if they had a distal femur fracture, if they were transferred from another facility, if they were transferred to another facility prior to fracture fixation, or if they expired prior to the decision to perform fracture stabilization. 77 patients met the criteria for inclusion. The patients were divided into groups based upon the timing for surgical femur fracture stabilization; either <24 hours or >24 hours. The groups were compared for demographic data, as well as neurologic and pulmonary complications, ICU and hospital length of stay.

Results: Of the 77 patients, the age (35.5 vs. 27.2 years, p=0.014), ISS (24.2 vs. 19.8, p=0.046) and the number of positive head CT scans (31% vs. 10%, p=0.028) was significantly greater in the late fixation group. The late fixation group also had significantly lower admission GCS scores (12.4 vs. 14.0, p=0.023). The early fixation group received greater amounts of intraoperative crystalloid (4470cc vs. 3151cc, p=0.047), had a greater estimated blood loss (896cc vs. 468cc, p=0.008), and experienced a higher incidence of hypotension (47% vs. 11%, p=0.013). Length of ICU and hospital stay was significantly longer (1.3 and 7.1 days vs. 6.9 and 23.6 days respectively, p=0.0002 and p=0.0001) in the late fixation group. This group had a significantly greater occurrence of pulmonary (27% vs. 6%, p=0.034) and neurologic complications (13% vs. 0%, p=0.038). Multivariate analysis of neurologic complications could not be carried out since no patients in the early fixation group had CNS complications. Bivariate regression of pulmonary complications showed borderline significant benefit of early fixation (OR=0.21, p=0.057), while greater severity of the original head injury predicted an increase in adverse pulmonary events (OR=3.35, p=0.059). There were no patients who demonstrated either an increase in ICP >20mmHg or progression of their original neurological insult.

Discussion: Several studies have looked at the pulmonary benefits with early femur fracture stabilization. The effects upon the neurologic outcome continue to be controversial. Most of the neurosurgical literature supports a worse outcome with early fracture fixation, whereas the orthopaedic literature continues to find little correlation.

Conclusion: We were unable to demonstrate that early fracture fixation increases the likelihood or severity of CNS complications. We found only that the patient's head injury had a significant impact on whether the patient received early versus late fixation. We recommend that neither hypothesis be accepted until a more controlled study population is available either through appropriate prospective protocols or through a multi-center study.