Predictors of blunt abdominal aortic injury in trauma patients and mortality analysis

ObjectiveBlunt abdominal aortic injury (BAAI) occurs in less than 0.1% of blunt traumas. A previous multi-institutional study found an associated mortality rate of 39%. We sought to identify risk factors for BAAI and risk factors for mortality in patients with BAAI using a large national database. We hypothesized that an Injury Severity Score of 25 or greater, and thoracic trauma would both increase the risk of mortality in patients with BAAI.MethodsThe Trauma Quality Improvement Program (2010-2016) was queried for individuals with blunt trauma. Patients with and without BAAI were compared. Covariates were included in a multivariable logistic regression model to determine mechanisms of injury, examination findings, and concomitant injuries associated with increased risk for BAAI. An additional multivariable analysis was performed for mortality in patients with BAAI.ResultsFrom 1,056,633 blunt trauma admissions, 1012 (0.1%) had BAAI. The most common mechanism of injury was motor vehicle accident (MVA; 57.5%). More than one-half the patients had at least one rib fracture (54.0%), or a spine fracture (53.9%), whereas 20.8% had hypotension on admission and 7.8% had a trunk abrasion. The average length of stay was 13.4 days and 24.6% required laparotomy, with 6.6% receiving an endovascular repair and 2.9% an open repair. The risk of death in those treated with endovascular vs open repair was similar (P = .28). On multivariable analysis, MVA was the mechanism associated with the highest risk of BAAI (odds ratio [OR], 4.68; 95% confidence interval [CI], 3.87-5.65; P < .001) followed by pedestrian struck (OR, 4.54; 95% CI, 3.47-5.92; P < .001). Other factors associated with BAAI included hypotension on admission (OR, 3.87; 95% CI, 3.21-4.66; P < .001), hemopneumothorax (OR, 3.67; 95% CI, 1.16-11.58; P < .001), abrasion to the trunk (OR, 1.49; 95% CI, 1.15-1.94; P = .003), and rib fracture (OR, 1.46; 95% CI, 1.25-1.70; P < .001). The overall mortality rate was 28.0%. Of the variables examined, the strongest risk factor associated with mortality in patients with BAAI was hemopneumothorax (OR, 12.49; 95% CI, 1.25-124.84; P = .03) followed by inferior vena cava (IVC) injury (OR, 12.05; 95% CI, 2.80-51.80; P < .001).ConclusionsIn the largest nationwide series to date, BAAI continues to have a high mortality rate with hemopneumothorax and IVC injury associated with the highest risk for mortality. The mechanism most strongly associated with BAAI is MVA followed by pedestrian struck. Other risk factors for BAAI include rib fracture and trunk abrasion. Providers must maintain a high suspicion of injury for BAAI when these mechanisms of injury, physical examination or imaging findings are encountered.     

Predictors of patients who will develop prolonged occult hypoperfusion following blunt trauma

BACKGROUND: Prolonged occult hypoperfusion or POH (serum lactate >2.4 mmol/L persisting >12 hours from admission) represents a reversible risk factor for adverse outcomes following traumatic injury. We hypothesized that patients at increased risk for POH could be identified at the time of admission. METHODS: Prospective data from adult trauma admissions between January 1, 1998 and December 31, 2000 were analyzed. Potential risk factors for POH were determined by univariate analysis (p < or =0.10= significant). Significant factors were tested in a logistic regression model (LR) (p < or =0.05= significant). The predictive ability of the LR was tested by receiver operating curve (ROC) analysis (p < or =0.05= significant). RESULTS: Three hundred seventy-eight patients were analyzed, 129 with POH. Injury Severity Score (ISS), emergency department Glasgow Coma Scale score, hypotension, and the individual Abbreviated Injury Scale score (AIS) for Head (H), Abdominal/Pelvic Viscera (A) and Pelvis/Bony Extremity (P) were significantly associated with POH. LR demonstrated that ISS, A-AIS > or =3 and P-AIS > or =3 were independent predictors of POH (p <0.05). ROC analysis of the LR equation was statistically significant (Area=0.69, p <0.001). CONCLUSIONS: We identified factors at admission that placed patients at higher risk for developing POH. Select patients may benefit from rapid, aggressive monitoring and resuscitation, possibly preventing POH and its associated morbidity and mortality.     

Systemic inflammatory response syndrome score at admission independently predicts infection in blunt trauma patients

BACKGROUND: Systemic inflammatory response syndrome (SIRS) score has been demonstrated to be an accurate predictor of outcome in critical surgical illness. To our knowledge, there is a paucity of data using SIRS score as a tool to predict posttraumatic infection. Our goal was to determine whether the severity of SIRS score at admission is an accurate predictor of infection in trauma patients. METHODS: Prospective data were collected on 4,887 blunt trauma patients admitted to a primary adult resource center designated trauma center over an 18-month period. Patients were stratified by age and Injury Severity Score (ISS). SIRS score was calculated at admission. SIRS was defined as an SIRS score > or = 2. Each patient was screened for infection by an infectious disease specialist. Those at high risk for infection were then monitored daily throughout their hospitalization. Centers for Disease Control and Prevention guidelines were used to diagnose infection. RESULTS: Of the 4,887 patients, 1,850 (38%) were admitted > 24 hours and evaluated for subsequent infection (mean ISS, 16 +/- 9; mean age, 43 +/- 19, SD). Thirty-one percent (577) of the patients acquired an infection. The mean hospital length of stay (20.2 days vs. 6.5 days) and mortality (7.8% vs. 2.7%) were significantly greater in the infected group (p < 0.001). Of the four SIRS variables (temperature, heart rate, white blood cell count, and respiratory rate), hypothermia and leukocytosis were the most significant predictors of infection (p < 0.001) when adjusted for age and ISS. SIRS scores of > or = 2 were increasingly predictive of infection when analyzed by multiple logistic regression analysis. CONCLUSION: An admission SIRS score of > or = 2 is a significant independent predictor of infection and outcome in blunt trauma. Daily SIRS scores may be a meaningful method of assessing postinjury risk of infection, and may initiate earlier diagnostic intervention for determination of infection.     

Invasive and non-invasive physiological monitoring of blunt trauma patients in the early period after emergency admission

Pulmonary artery catheterization is usually not available to critically injured patients before admission to the intensive care unit, where action to correct values derived from such monitoring may be too late. Methods allowing hemodynamic monitoring during the early stages after trauma need to be explored. We used non-invasive monitoring systems (bioimpedance cardiac output monitoring, pulse oximetry and transcutaneous oximetry) to evaluate early temporal hemodynamic patterns after blunt trauma, and compared these to invasive PA monitoring. We included prospectively 134 patients monitored shortly after admission to the emergency department. The non-invasive impedance cardiac output estimations under extenuating emergency conditions approximated those of the thermodilution method: r = 0.83, r2 = 0.69, P<0.001; bias and precision were -0.02+/-0.78 l/min/m2. In the intensive care unit, these values improved further to: r = 0.91, r2 = 0.83, P<0.001; bias and precision = 0.36+/-0.59 l/min/m2. Monitoring revealed episodes of hypotension, low cardiac index, arterial hemoglobin desaturation, low transcutaneous oxygen and high transcutaneous carbon dioxide tensions, and low oxygen consumption during initial resuscitation. Low flow and poor tissue perfusion were more pronounced in non-survivors by both methods. Multicomponent non-invasive monitoring systems give continuous on-line, real-time displays of physiological data that allow early recognition of circulatory dysfunction. Such systems provide information similar to that provided by the invasive thermodilution method, and are easier and safer to use.     

Predictors of blinding or serious eye injury in blunt trauma.

Multivariate analysis was used to identify factors predicting injury and visual outcome in 94 blunt trauma patients evaluated for eye injuries among 6700 admissions to a level I trauma center over a 29-month period. Patients with penetrating eye injuries were excluded from this review. Eye injury was detected in 93% or 87 of the patients evaluated. Seven percent of eye injuries resulted in blindness, 22% were serious (visual acuity between 20/40 and 20/200 or eye injury requiring surgery), and 71% were temporary (final visual acuity of 20/40 or better). The presence of an afferent pupillary defect or a nonreactive pupil was the most important factor in predicting the severity of eye injury (p = 0.0023), followed by facial fractures (p = 0.0084), and no eye opening or eye opening to pain within the Glasgow Coma Scale (p = 0.02). Eye injury is an infrequent complication of blunt trauma. Appropriate consultation for evaluation of this problem can be obtained based on findings from the initial history and screening physical examination.     

Implications of admission hypothermia in trauma patients

Hypothermia is common after severe injury, and has been associated with an increased mortality rate in patients stratified by anatomic indices of injury severity. In this retrospective study of 173 patients, early post-traumatic hypothermia was found to correlate with physiologic indicators of volume deficit, independently of the amount of intravenous fluid received. There was no correlation found between admission core temperature and time from injury, blood alcohol, or presence of severe closed head injury. Hypothermic patients (less than 35 degrees C) had a lower predicted probability of survival and a higher mortality rate than euthermic patients (greater than or equal to 35 degrees C). However, when patients were stratified by physiologic and anatomic indicators of injury severity, mortality rates among the euthermic and hypothermic patients were not significantly different. Early post-traumatic hypothermia does not appear to exert an independent effect upon outcome.     

Predictors of mortality in trauma patients

The purpose of this study was to ascertain risk factors for death from trauma. The large cohort allows for simultaneous evaluation of known mortality risk factors along with controlling for factors to assess the influence of each independently. Individually, base deficit, temperature, hypotension, age, and injury severity have been shown to be associated with an increased risk of death. However, in the English literature, there is no data on the independent predictive power and interaction of these risk factors. A review of trauma registry parameters from 1995 to 2000 was used. Demographics, injury severity, physiological and hematological parameters, and time data were evaluated in a univariate analysis. Variables significantly associated with mortality were entered into a stepwise backward multiple logistic regression. There were 1276 deaths (8.9%) with 25 per cent of the deaths within 3 hours. The top four predictors of mortality in this group were partial thromboplastin time (OR 3.37, 95% CI: 2.51-4.52), positive head computed tomography result (OR, 2.47; 95% CI, 1.95-3.04), initial hemoglobin (OR, 1.69; 95% CI, 1.23-2.31), base deficit (OR, 1.62; 95% CI, 1.29-2.04), and trauma resuscitation bay systolic blood pressure (OR, 1.45; 95% CI, 1.11-1.88). We conclude that prognostic indicators of all-cause mortality after trauma, which remain independent in the presence of all other factors and are potentially treatable, included low hemoglobin, elevated prothrombin and partial thromboplastin time, low scene and trauma bay systolic pressure, and elevated base deficit. The independent indicators of mortality, which are untreatable, included head injury, increasing age, and Injury Severity Score.     

Predictors of mortality, length of stay and discharge destination in blunt trauma

BACKGROUND: The present study explored a range of variables to identify predictors of mortality and morbidity and to develop prediction models based on these variables. METHODS: Tools for predicting mortality, hospital length of stay and a patient's destination post-hospital discharge were developed using logistic regression in one dataset (design) and evaluated for prediction performance in a separate dataset (validation). The performance of the mortality model was compared to the trauma and injury severity score (TRISS) and a severity characterization of trauma (ASCOT). RESULTS: The profile of variables contributing to the final prediction models developed from the design dataset varied across the different outcomes of interest although age, injury severity score, development of complications and triage category were common predictors of all three outcomes. The performance of the new mortality prediction model was superior to both TRISS and ASCOT in the validation dataset. Overall, the new models did not meet the prespecified performance criteria. CONCLUSIONS: The present study identified key predictors of mortality and morbidity (length of hospital stay and discharge destination). The newly developed mortality model out-performed published trauma scoring methods. However, further development and trial of the new prediction models is required before implementation as definitive audit and benchmarking tools could be recommended.     

Abdominal ultrasound examination in pregnant blunt trauma patients

BACKGROUND: The ability of abdominal ultrasound to detect intraperitoneal fluid in the pregnant trauma patient has been questioned. METHODS: Pregnant blunt trauma patients admitted to a Level I trauma center during an 8-year period were reviewed. Ultrasound examinations were used to detect intraperitoneal fluid and considered positive if such fluid was identified. RESULTS: One hundred twenty-seven (61%) of 208 pregnant patients had abdominal ultrasound during initial evaluation in the emergency department. Seven patients had intra-abdominal injuries, and six had documented hemoperitoneum. Ultrasound identified intraperitoneal fluid in five of these six patients (sensitivity, 83%; 95% confidence interval, 36-100%). In the 120 patients without intra-abdominal injury, ultrasound was negative in 117 (specificity, 98%; 95% confidence interval, 93-100%). The three patients without intra-abdominal injury but with a positive ultrasound had the following: serous intraperitoneal fluid and no injuries at laparotomy (one) and uneventful clinical courses of observation (two). CONCLUSION: The sensitivity and specificity of abdominal ultrasonography in pregnant trauma patients is similar to that seen in nonpregnant patients. Occasional false negatives occur and a negative initial examination should not be used as conclusive evidence that intra-abdominal injury is not present. Ultrasound has the advantages of no radiation exposure.     

Massive transfusion: outcome in blunt trauma patients

Over a 54-month period 6,142 patients were consecutively admitted to our Level I trauma center. Ninety-two blunt trauma patients required massive transfusion (MT) of 20 or more units of packed red blood cells (range, 20-126). Eighty-two per cent of all transfused blood was given within 24 hours of admission. Forty-eight patients (52%) were long-term survivors. Twenty-six patients died (28%) within 24 hours and 21 of these exsanguinated. Eighteen patients died greater than 24 hours: nine (50%) died from multiple organ failure, and nine (50%) died from severe closed head injury (CHI). Clinical predictors of increased mortality were: shock on admission, closed head injury, and age. Forty-three survivors were followed for a mean of 2.5 years (range, 1-5 years). No patient died during followup. All patients were home at 1 year; only four patients required continued medical assistance. Thirty-two patients (74%) returned to work. We conclude that: 1) blunt and penetrating trauma patients receiving MT have similar survival rates of 50%; 2) shock, closed head injury, and age predict increased mortality but do not preclude survival; 3) long-term outcome in blunt patients requiring MT is excellent. Post-discharge death is rare and 3/4 of the survivors return to work, justifying the high cost of acute care.     

Screening for blunt cerebrovascular injuries in pediatric trauma patients

BackgroundAdult imaging for blunt cerebrovascular injuries (BCVI) is based on the Denver and Memphis screening criteria where CT angiogram (CTA) is performed for any one of the criteria being positive. These guidelines have been extrapolated to the pediatric population. We hypothesize that the current adult criteria applied to pediatrics lead to unnecessary CTA in pediatric trauma patients.Study designAt our center, a 9-year retrospective study revealed that strict adherence to the Denver and Memphis criteria would have resulted in 332 unnecessary CTAs out of 2795 trauma patients with only 0.3% positive for BCVI. We also conducted a retrospective chart review of 776,355 pediatric trauma patients in the National Trauma Data Bank (NTDB) from 2007 to 2014. Data collection included children between ages 0 and 18, ICD-9 search for blunt cerebrovascular injury, and ICD-9 codes that applied to both Denver and Memphis criteria.ResultsOf 776,355 pediatric trauma activations, 81,294 pediatric patients in the NTDB fit the Denver/Memphis criteria for screening CTA neck or angiography based on ICD-9 codes, while only 2136 patients suffered BCVI. Strict utilization of the Denver/Memphis criteria would have led to a negative CTA in 79,158 (97.4%) patients. Multivariate regression analysis indicates that patients with skull base fracture, cervical spine fractures, cervical spine fracture with cervical cord injury, traumatic jugular venous injury, and cranial nerve injury should be considered part of the screening criteria for BCVI.ConclusionOur study suggests the Denver and Memphis criteria are inadequate screening criteria for CTA looking for BCVI in the pediatric blunt trauma population. New criteria are needed to adequately indicate the need for CT angiography in the pediatric trauma population.Level of evidenceIV.     

Cardiac contusion in pediatric patients with blunt thoracic trauma

To investigate the prevalence of myocardial contusion associated with blunt chest trauma in the pediatric age group, all patients admitted to our institution during a 6-month period with blunt thoracic trauma severe enough to produce a pulmonary contusion or rib fracture were prospectively evaluated. Cardiac evaluation was undertaken, including a multiple-gated acquisition (MUGA) cardiac scan, serial electrocardiograms (ECG), and serum creatine phosphokinase (CPK) and CPK isoenzymes. Seven patients, ranging in age from 2 1/2 to 18 years, with rib fractures or pulmonary contusion by chest roentgenograph were identified. One patient was injured as a passenger in a motor vehicle accident, five were struck by automobiles as pedestrians, and one sustained traumatic asphyxia when a car, supported by a jack, fell on his chest. All had at least one other major organ system injured. All patients had pulmonary contusions as determined by chest radiograph, and two had associated rib fractures. In 43% (three of seven) of patients, a significant cardiac contusion was identified, defined by abnormal right or left ventricular wall motion and a decreased ejection fraction on MUGA scan, and confirmed by an increase in cardiac enzymes and isoenzymes. However, in contrast with adults, no patients had ECG abnormalities. This limited series suggests that cardiac contusion may occur frequently in pediatric patients who have suffered from blunt thoracic trauma significant enough to result in pulmonary contusion. An MUGA scan provides a rapid, noninvasive assessment of cardiac damage in this setting. Further studies will be required to determine the clinical significance and long-term consequences of traumatic myocardial damage in the pediatric population.     

Outcomes after rib fractures in geriatric blunt trauma patients

Introduction

Rib fractures after blunt trauma contribute substantially to morbidity and mortality in the elderly.

Fibrinogen level on admission is a predictor for massive transfusion in patients with severe blunt trauma: Analyses of a retrospective multicentre observational study

IntroductionIn the early phase of trauma, fibrinogen (Fbg) plays an important role in clot formation. However, to the best of our knowledge, few studies have analysed methods of predicting the need for massive transfusion (MT) based on Fbg levels using multiple logistic regression. Therefore, the present study aimed to evaluate whether Fbg levels on admission can be used to predict the need for MT in patients with trauma.MethodsWe conducted a retrospective multicentre observational study. Patients with blunt trauma with ISS ≥16 who were admitted to 15 tertiary emergency and critical care centres in Japan participating in the J-OCTET were enrolled in the present study. MT was defined as the transfusion of packed red blood cells (PRBC) ≥10 units or death caused by bleeding within 24 h after admission. Patients were divided into non-MT and MT groups. Multiple logistic-regression analysis was used to assess the predictive value of the variables age, sex, vital signs, Glasgow Coma Scale (GCS) score, and Fbg levels for MT. We also evaluated the discrimination threshold of MT prediction via receiver operating characteristic curve (ROC) analysis for each variable.ResultsHigher heart rate (HR; per 10 beats per minutes [bpm]), systolic blood pressure (SBP; per 10 mm Hg), GCS, and Fbg levels (per 10 mg/dL) were independent predictors of MT (odds ratio [OR] 1.480, 95% confidence interval [CI] 1.326–1.668; OR 0.851, 95% CI 0.789–0.914; OR 0.907, 95% CI 0.855–0.962; and OR 0.931, 95% CI 0.898–0.963, respectively). The optimal cut-off values for HR, SBP, GCS, and Fbg levels were ≥100 bpm (sensitivity 62.4%, specificity 79.8%), ≤120 mm Hg (sensitivity 61.5%, specificity 70.5%), ≤12 points (sensitivity 63.3%, specificity 63.6%), and ≤190 mg/dL (sensitivity 55.1%, specificity 78.6%), respectively.ConclusionsOur findings suggest that vital signs, GCS, and decreased Fbg levels can be regarded as predictors of MT. Therefore, future studies should consider Fbg levels when devising models for the prediction of MT.