Morbidity and mortality in elderly trauma patients.J Trauma. 1999 Apr; 46(4):702-6.JT
Despite an increasing incidence, relatively few studies have examined the factors that predict morbidity and mortality in older patients and several reports have found standard predictors such as the Injury Severity Score to be less useful in this patient population. Similarly, the effect of skeletal injury has not been examined with regard to complications and mortality. The purpose of this study was to review a large multicenter experience with elderly trauma patients to isolate factors that might predict morbidity and mortality. The potential effect of skeletal long-bone injury was of particular interest.
The charts of all patients older than 60 years who were admitted to one of four Level I trauma centers after sustaining blunt trauma were reviewed. Mechanisms of injury included in the study were motor vehicle crash, pedestrian struck, fall from a height, and crush injury. Slip-and-fall injuries were excluded. A total of 326 patients met inclusion criteria. Variables studied included age, sex, mechanism of injury, Injury Severity Score (ISS), Revised Trauma Score, Glasgow Coma Scale (GCS) score, blood transfusion, fluid resuscitation, surgery performed (laparotomy, long-bone fracture stabilization, both), and timing of surgery. Outcome variables measured included incidence of adult respiratory distress syndrome, pneumonia, sepsis, myocardial infarction, deep venous thromboembolism, gastrointestinal complications, and death. chi2, logistic regression, t test, and nonparametric analyses were done as appropriate for the type of variable.
The average age of the patients was 72.2+/-8 years. Overall, 59 patients (18.1%) died, of whom 52 of 59 survived at least 24 hours. Statistical significance for continuous variables (p < 0.05) using univariate analysis was reached for the following factors for the patients who died: higher ISS (33.1 vs. 16.4), lower GCS score (11.5 vs. 13.9), greater transfusion requirement (10.9 vs. 2.9 U), and more fluid infused (12.4 vs. 4.9 L). Logistic regression analysis was performed to determine the factors that predicted mortality. They included (odds ratios and p values in parentheses) transfusion (1.11, p = 0.01), ISS (1.04, p = 0.008), GCS score (0.87, p = 0.007), and fluid requirement (1.06, p = 0.06). Regarding surgery, orthopedic surgery alone had an odds ratio of 0.53, indicating that orthopedic patients was less likely to die than patients who did not undergo any surgery. Patients who underwent only a general surgical procedure were 2.5 times more likely to die (p = 0.03) and patients who underwent both general and orthopedic procedures were 1.5 times more likely to die (p = 0.32) than patients who did not require surgery. Early (< or =24 hours) versus late (>24 hours) surgery for bony stabilization did not have a statistical effect on mortality (11% early vs. 18% late). Two patients in need of bony stabilization, however, died before these procedures were performed. With regard to complications, regression analysis revealed that ISS predicted adult respiratory distress syndrome, pneumonia, sepsis, and gastrointestinal complications; fluid transfusion predicted myocardial infusion; and need for surgery and transfusion requirements predicted sepsis. These complications, in turn, were significant risk factors for mortality. This large series of elderly patients demonstrates that mortality correlates closely with ISS and is influenced by blood and fluid requirements and by GCS score. The institution-specific mortality was the same when adjusted for ISS. The need for orthopedic surgery and the timing of the surgery was not a risk factor for systemic complications or mortality in this series.
Mortality is predicted by ISS and by complications in older patients. Seventy-seven percent of the orthopedic injuries were stabilized early, but the timing of surgery did not have any statistical effect on the incidence of complications or mortality. (ABSTRACT TRUNCA