Session IV Scientific Basis for Fracture Care

The Effects of the Skeletal Injury in Multiple Trauma

Bruce H. Ziran MD, T. Le MD, W. Fallon MD, J.H. Wilber MD

University of Pittsburgh Medical Center, Pittsburgh, PA

Purpose: Scoring systems have been developed (ISS & AIS) to characterize the contribution of the skeletal system with regards to morbidity and mortality but the differentiation between single and multiple skeletal injuries is not well elucidated. An increased number of fractures may result in greater morbidity or mortality in multiply injured patients (especially with concomitant pulmonary injury). The hypothesis of the present study was that the quantity of skeletal injuries (and the timing to fixation) increases the mortality or pulmonary morbidity in patients with and without chest injuries.

Methods: The study design was retrospective using a strict statistical multiple logistic regression analysis. The trauma registry from the authors' institution was reviewed for the year 1991 looking specifically at adult (>16 years) patients with significant chest (AIS-chest3) and/or skeletal injuries(AIS-ext3). Skeletal injury was quantified by determining the presence or absence of a fracture in specific body regions (humeri, forearm, femur, tibia, spine, pelvis) for a maximum of 10 skeletal injuries. Since very few patients had greater than four skeletal injuries, patients with three or more injuries were pooled to facilitate statistical analysis (0, 1, 2, or 3 skeletal injuries). The timing of fixation for fractures was categorized as <24 hrs, <48 hrs, <72 hrs, <5 days, >5 days, or no fixation. Chest injuries were recorded based on the documentation of injuries at admission and during the initial hospital evaluation. The pulmonary morbidity was based on the accepted list of complications reported in the literature. The establishment of pulmonary complications required both radiographic and clinical data. To prevent the confounding effects of head, chest, and abdominal injuries, these variables were controlled in the logistic model.

Results: There were a total 226 patients stratified into three groups analyzed according to the presence or absence of a chest and/or skeletal injury: those without skeletal injury (Group NSI, n=59), those without chest injuries (Group NCI, n=108), and those with both skeletal and chest injuries (Group B, n=59). The hypothesis that the quantity of skeletal injury (and time fixation) increase pulmonary morbidity and mortality in multiply injured patients was inconsistently supported.

Pulmonary Complications: When all patients (groups NCI, NSI, B) were pooled together, greater chest injury (p<0.0008), greater skeletal injury (p<0.02) and the delayed fixation(p<0.04) were associated with increased risk of developing a pulmonary complication. In the group of patients without a chest injury (NCI), this risk was associated with greater head injury (p<0.005) and greater skeletal injury (p<0.04) while in the group without a skeletal injury (NSI) only chest injury demonstrated (p<0.05) significance. When both skeletal and chest injuries (B) were present, greater head injury (p<0.03) and fixation time (p<0.03) increased the risk of developing a pulmonary complication.

Mortality: With all patients pooled (NCI, NSI, B), head injury (p<0.02), abdominal injury (p<0.0l2), and fixation time (p<0.01) were risk factors. In patients without a chest injury (NCI), none of the indexed variable were associated with mortality. In patients without a skeletal injury (NSI), greater head injury (p<0.01), greater chest injury (p<0.0l) and greater abdominal injury (p<0.04) were risk factors for mortality. When both chest and skeletal injuries were present (B), only head injury (p<0.0003) was associated with mortality. We were unable to demonstrate a correlation between the independent variables (skeletal injuries and fixation time) and vent days, icu days, or length of stay. The prevalence of mortality and a pulmonary complications were compared between groups NCI (no chest), NSI (no skeletal) and B (both). Group NCI (no chest) had less pulmonary complications (p<0.004) (B and NSI). When examining mortality group NCI (no chest) had less mortality than groups NSI (no skeletal) and B (both).

Discussion: In terms of orthopedic parameters the following findings were noted: In the absence of a significant chest injury the quantity of skeletal injury has an impact on pulmonary morbidity but not on mortality. In the presence of a chest injury, time to fixation of pelvic, spinal, and femoral fractures were associated with pulmonary morbidity but not on mortality. If all patients are pooled together then both the quantity of skeletal injury and time to fixation were associated with pulmonary morbidity and only fixation time was associated with mortality. The inconsistent significance of the skeletal injury may be secondary to the more powerful effects of head and chest injuries (head injury was significant in most groups analyzed). When comparing groups with and without extremity and chest injuries, it appears that the combination of chest and skeletal injury together do not impose a greater risk for pulmonary morbidity or mortality than chest injury alone, and that these outcomes may be more determined by other factors (head, abdominal, and chest injuries) as supported by previous literature. Previous studies examining the timing to fixation have not accurately accounted for confounding variables, nor have any studies examined the effects of the number of skeletal injuries.

Conclusion: The combination of skeletal and chest injuries do not amplify the pulmonary morbidity and mortality as compared to chest injury alone. The quantity of the skeletal injury and the time to fixation of structures affecting mobilization have a significant effect on pulmonary morbidity, but not on mortality. Better scientific studies on the effects of skeletal injury and timing to fixation in relation to pulmonary morbidity and mortality are required.