Shea C. Gregg

Management of the Most Severely Injured Spleen A Multicenter Study of the Research Consortium of New England Centers for Trauma (ReCONECT)

OBJECTIVE To determine the rate and predictors of failure of nonoperative management (NOM) in grade IV and V blunt splenic injuries (BSI). DESIGN Retrospective case series. SETTING Fourteen trauma centers in New England. PATIENTS A total of 388 adult patients with a grade IV or V BSI who were admitted between January 1, 2001, and August 31, 2008. MAIN OUTCOME MEASURES Failure of NOM (f-NOM). RESULTS A total of 164 patients (42%) were operated on immediately. Of the remaining 224 who were offered a trial of NOM, the treatment failed in 85 patients (38%). At the end, 64% of patients required surgery. Multivariate analysis identified 2 independent predictors of f-NOM: grade V BSI and the presence of a brain injury. The likelihood of f-NOM was 32% if no predictor was present, 56% if 1 was present, and 100% if both were present. The mortality of patients for whom NOM failed was almost 7-fold higher than those with successful NOM (4.7% vs 0.7%; P = .07). CONCLUSIONS Nearly two-thirds of patients with grade IV or V BSI require surgery. A grade V BSI and brain injury predict failure of NOM. This data must be taken into account when generalizations are made about the overall high success rates of NOM, which do not represent severe BSI.

Portal Venous Gas Detected by Computed Tomography: Is Surgery Imperative?

Background: Portal vein gas (PVG) has historically been associated with mortality rates of 75% or higher and mandatory abdominal exploration. The following study reassesses the clinical significance of gas in the portal venous system detected by computed tomography (CT) and reevaluates the need for surgical intervention in that setting. Methods: We performed a retrospective chart review of 26 patients presenting with gas in the portal venous system as imaged by CT between August 1, 1993 and April 8, 2001. Characteristics assessed included age, sex, clinical presentation and course, additional CT findings, diagnosis, surgical versus non-surgical management, surgical findings, pathologic correlation, and mortality. Results: Of the 26 patients who had gas in the portal venous system by CT, 11 underwent surgery to determine underlying pathology. Of those who underwent surgery, 8 survived to be discharged while 3 died (73% surgical survival). The survivors presented with gastrointestinal related signs/symptoms and were found at the time of surgery to have adhesions, non-surgically treated inflammation, unclear pathology, or resectable lesions. Those who died all demonstrated marked bowel ischemia intraoperatively. Among patients who did not undergo surgery, 9 survived and 6 died (60% nonsurgical survival). Those who survived presented either asymptomatically or with pathology including acute pancreatitis, recent hypotensive episode, uncomplicated diverticulitis, unresectable colon cancer, or infectious etiology. Those who died were poor surgical candidates or status postvascular procedure. The overall survival in patients with portal venous gas by CT was 65%. Conclusion: Portal venous gas on CT is associated with a wide range of pathologies that do not necessarily warrant surgical management. Clinical presentation should be correlated with the presence of PVG and other associated CT findings prior to deciding whether surgery should be performed.

Impact of Socioethnic Factors on Outcomes Following Traumatic Brain Injury

BACKGROUND Ethnic minorities and low income families tend to be in poorer health and have worse outcomes for a spectrum of diseases. Health care provider bias has been reported to potentially affect the distribution of care away from poorer communities, minorities, and patients with a history of substance abuse. Trauma is perceived as a disease of the poor and medically underserved. Minorities are overrepresented in low income populations and are also less likely to possess health insurance leading to a potential overlapping effect. Traumatic brain injury (TBI) is a predominant cause of mortality and long-term morbidity, which imposes a considerable social and financial burden. We therefore sought to determine the independent effect on outcome after TBI from race, insurance status, intoxication on presentation, and median income. METHODS A 5-year retrospective chart review of admitted trauma patients aged 18 years and older to a Level I trauma center. Zip code of residency was a surrogate marker for socioeconomic status, because median income for each zip code is available from the US Census. Charts review included race, insurance status, mechanisms of trauma, and injuries sustained. Outcomes were placement of tracheostomy, hospital length of stay (HLOS), leaving Against Medical Advice (AMA), and discharge to home versus rehabilitation and mortality. RESULTS A total of 3,101 TBI patients were included in the analyses. Multivariable logistic and proportional hazard regression analyses were undertaken adjusting for age, gender, Injury Severity Score, and mechanism. Rates of tracheostomy placement were unaffected by race, median income, or insurance status. Race and median income did not affect HLOS, but private insurance was associated with shorter HLOS and intoxication was associated with longer HLOS. Neither race nor intoxication affected rates of AMA, but higher income and private insurance was associated with lower rates of AMA. Non-Caucasian race and lack of insurance had significantly lower likelihood of placement in a rehabilitation center. Mortality was unaffected by race, increased in intoxicated patients, was variably affected by median income, and was lowest in patients with private insurance. CONCLUSIONS An extremely complex interplay exists between socioethnic factors and outcomes after TBI. Few physicians would claim overt discrimination. Tracheostomy, the factor most directed by the surgeon, was unbiased by race, income, or insurance status. The likelihood of placement in a rehabilitation center was significantly impacted by both race and insurance status. Future prospective studies are needed to better address causation.

The development of a urinary tract infection is associated with increased mortality in trauma patients.

BACKGROUND In October 2008, Medicare and Medicaid stopped paying for care associated with catheter-related urinary tract infections (UTIs). Although most clinicians agree UTIs are detrimental, there are little data to support this belief. METHODS This is a retrospective review of trauma registry data from a Level I trauma center between 2003 and 2008. Two proportional hazards regressions were used for analyses. The first predicted acquisition of UTI as a function of indwelling urinary catheter use, adjusting for age, diabetes, gender, and injury severity. The second predicted hospital mortality as a function of UTI, covarying for age, gender, chronic obstructive pulmonary disease, congestive heart failure, hypertension, diabetes, pneumonia, and injury severity. RESULTS After excluding patients who stayed in the hospital <3 days and those with a UTI on arrival, 5,736 patients were included in the study. Of these patients, 680 (11.9%) met criteria for a UTI, with 487 (71.6%) indwelling urinary catheter-related infections. Predictors of UTI included the interaction between age and gender (p = 0.0018), Injury Severity Score (p = 0.0021), and indwelling urinary catheter use (p < 0.001). The development of a UTI predicted the risk of in-hospital death as a patients age increased (p = 0.002). Similar results were seen when only catheter-associated UTIs are included in the analysis. CONCLUSIONS Indwelling urinary catheter use is connected to the development of UTIs, and these infections are associated with a greater mortality as the age of a trauma patients increases.

Undiagnosed medical comorbidities in the uninsured: A significant predictor of mortality following trauma

BACKGROUND Lack of health care insurance has been correlated with increased mortality after trauma. Medical comorbidities significantly affect trauma outcomes. Access to health care and thereby being diagnosed with a pretrauma comorbidity is highly dependent on insurance status. The objective of this study was to determine whether rates of diagnosed or undiagnosed preexisting comorbidities significantly contribute to disparities in mortality rates observed between insured and uninsured trauma patients. METHODS Review of trauma patients admitted to a Level I trauma center during a 5-year period. Data extracted from the registry included age, sex, Injury Severity Score (ISS), comorbidities, mortality, and insurance status. Multivariate logistic regression analysis was performed using age, sex, and insurance status to predict comorbidities and age, sex, ISS, and insurance status to predict mortality. RESULTS Insured patients were older (54 years vs. 38, p < 0.001) and more likely female (41.3% vs. 22.5%, p < 0.001). When adjusting for age and sex, insured patients were more likely to have a pretrauma diagnosis of coronary artery disease (odds ratio [OR], 2.09; 95% confidence interval [CI], 1.54–2.83), diabetes mellitus (OR, 2.09; 95% CI, 1.61–2.72), hypertension (OR, 1.97; 95% CI, 1.65–2.35), asthma/emphysema (OR, 1.64; 95% CI, 1.32–2.04), neurologic problems (OR, 1.79; 95% CI, 1.31–2.44), and gastroesophageal reflux disease (OR, 2.03; 95% CI, 1.33–3.11), compared with patients without insurance. In the analysis to predict mortality, having insurance was protective (OR, 0.57; 95% CI, 0.45–0.71). Among patients with no diagnosed comorbidities, insured patients had the lowest mortality risk (OR, 0.5; 95% CI, 0.38–0.67). When analyzing only patients with diagnosed comorbidities, insurance status had no impact on mortality risk (OR, 0.81; 95% CI, 0.53–1.22). CONCLUSION Undiagnosed preexisting comorbidities play a crucial role in determining outcomes following trauma. Diagnosis of medical comorbidities may be a marker of access to health care and may be associated with treatment, which may explain the gap in mortality rates between insured and uninsured trauma patients. LEVEL OF EVIDENCE Prognostic/epidemiologic study, level III.

Infections after trauma are associated with subsequent cardiac injury.

BACKGROUND Trauma produces profound inflammatory and immune responses. A second hit such as an infection further disrupts the inflammatory cascade. Inflammatory responses, following traumatic injuries, infections, or both, are emerging as biologic mediators of cardiac disease including myocardial ischemia and infarction. Inflammation-induced and stress-related cardiac damage are increasingly recognized in patients with critical illness. It is believed that cardiac dysfunction is the result of alterations in the inflammatory and immune cascades. Urinary tract infections (UTIs) and ventilator-associated pneumonia (VAP) are associated with increased mortality in trauma patients. UTIs and VAPs induced inflammatory responses. We postulate that increased mortality seen in trauma patients with infections is caused by increased rates of cardiac injury. METHODS This is a retrospective review of prospectively collected data. All trauma patients admitted to the intensive care unit at our Level I trauma center during 5 years were included in the analysis. Proportional hazard regression analysis was performed to predict suspicion of cardiac injury (troponin ordered), any cardiac injury (troponin > 0.15 ng/mL), or severe cardiac injury (troponin > 1 ng/mL) using age, sex, Injury Severity Score (ISS), pulmonary disease (chronic obstructive pulmonary disease), heart failure, hypertension, diabetes, and the presence of a UTI or VAP. A similar proportion hazard regression was performed to predict mortality. RESULTS In the model to predict any cardiac injury, chronic obstructive pulmonary disease (hazards ratio [HR], 1.9; p = 0.02), ISS (HR, 1.01; p = 0.04), VAP (HR, 5.6; p < 0.01), and UTI (HR, 2.4; p = 0.03) were significant. Neither VAP nor UTI predicted severe cardiac injury. In the model to predict death, any cardiac injury was not associated with mortality, but severe cardiac injury and UTI were associated with mortality as age increased. CONCLUSION Infectious complications have been associated with increased mortality in trauma patients. Our data demonstrate that development of VAP or UTI is associated with an increased risk of developing cardiac injury in trauma patients, which may contribute to subsequent increased mortality. LEVEL OF EVIDENCE Prognostic/epidemiologic study, level III.

Empiric antibiotics pending bronchoalveolar lavage data in patients without pneumonia significantly alters the flora, but not the resistance profile, if a subsequent pneumonia develops

INTRODUCTION Ventilator-associated pneumonia (VAP) occurs in up to 25% of mechanically ventilated patients, with an associated mortality up to 50%. Early diagnosis and appropriate empiric antibiotic coverage of VAP are crucial. Given the multitude of noninfectious clinical and radiographic anomalies within trauma patients, microbiology from bronchioalveolar lavage (BAL) is often needed. Empiric antibiotics are administered while awaiting BAL culture data. Little is known about the effects of these empiric antibiotics on patients with negative BAL microbiology if a subsequent VAP occurs during the same hospital course. METHODS This is a retrospective chart review of intubated trauma patients undergoing BAL for suspected pneumonia over a 3-y period at a Level 1 trauma center. All patients with suspected VAP undergoing a BAL receive empiric antibiotics. If microbiology data are negative at 72 h, all antibiotics are stopped; however, if the BAL returns with ≥10(5) colony-forming units per milliliter, the diagnosis of VAP is confirmed. We divided patients into three groups. Group 1 consisted of patients in whom the initial BAL was positive for VAP. Group 2 consisted of patients with an initial negative BAL, who subsequently developed VAP at a later point in the hospital course. Group 3 consisted of patients with negative BAL who did not develop a subsequent VAP. RESULTS We obtained 499 BAL specimens in 185 patients over the 3-y period. A total of 14 patients with 23 BAL specimens initially negative for VAP subsequently developed VAP later during the same hospital stay. These patients did not have an increase in the hospital length of stay, intensive care unit days, ventilator days, or mortality compared with those who had a positive culture on the first suspicion of VAP. There was a significant increase in the percentage of Enterobacter (21% versus 8%) and Morganella (8% versus 0%) as the causative organism in these 14 patients when the VAP occurred. Furthermore, the profile of the top two organisms in each group changed. Enterobacter (21%) and Pseudomonas (17%) were the principal organisms in the initial BAL-negative group, whereas the two predominant strains in the initial positive BAL group were methicillin-sensitive Staphylococcus aureus (21%) and Haemophilus influenza (11%). Interestingly, methicillin-resistant S. aureus remained the third most common organism in both groups. Empiric antibiotics also did not seem to induce the growth of multidrug-resistant organisms, and there was no increased rate of secondary infections such as Clostridium difficile. CONCLUSIONS Ventilator-associated pneumonia remains a significant cause of morbidity and mortality in mechanically ventilated trauma patients. The diagnosis and treatment of VAP continue to be challenging. Once clinically suspected, empiric coverage decreases morbidity and mortality. Our data demonstrate that patients who receive empiric coverage exhibit a significantly different microbiologic profile compared with those who had an initial positive BAL culture. Initial empiric antibiotics in BAL-negative patients were not associated with an increase in multidrug-resistant organisms, hospital, or intensive care unit length of stay, ventilator days, and mortality or secondary infections.

Teaching leadership in trauma resuscitation: Immediate feedback from a real-time, competency-based evaluation tool shows long-term improvement in resident performance.

BACKGROUND Limited data exist on how to develop resident leadership and communication skills during actual trauma resuscitations. METHODS An evaluation tool was developed to grade senior resident performance as the team leader during full-trauma-team activations. Thirty actions that demonstrated the Accreditation Council for Graduate Medical Education core competencies were graded on a Likert scale of 1 (poor) to 5 (exceptional). These actions were grouped by their respective core competencies on 5 × 7-inch index cards. In Phase 1, baseline performance scores were obtained. In Phase 2, trauma-focused communication in-services were conducted early in the academic year, and immediate, personalized feedback sessions were performed after resuscitations based on the evaluation tool. In Phase 3, residents received only evaluation-based feedback following resuscitations. RESULTS In Phase 1 (October 2009 to April 2010), 27 evaluations were performed on 10 residents. In Phase 2 (April 2010 to October 2010), 28 evaluations were performed on nine residents. In Phase 3 (October 2010 to January 2012), 44 evaluations were performed on 13 residents. Total scores improved significantly between Phases 1 and 2 (p = 0.003) and remained elevated throughout Phase 3. When analyzing performance by competency, significant improvement between Phases 1 and 2 (p < 0.05) was seen in all competencies (patient care, knowledge, system-based practice, practice-based learning) with the exception of “communication and professionalism” (p = 0.56). Statistically similar scores were observed between Phases 2 and 3 in all competencies with the exception of “medical knowledge,” which showed ongoing significant improvement (p = 0.003). CONCLUSIONS Directed resident feedback sessions utilizing data from a real-time, competency-based evaluation tool have allowed us to improve our residents’ abilities to lead trauma resuscitations over a 30-month period. Given pressures to maximize clinical educational opportunities among work-hour constraints, such a model may help decrease the need for costly simulation-based training. LEVEL OF EVIDENCE Therapeutic study, level III.

Trauma morning report is the ideal environment to teach and evaluate resident communication and sign-outs in the 80 hour work week

BACKGROUND The 80h work week has raised concerns that complications may increase due to multiple sign-outs or poor communication. Trauma Surgery manages complex trauma and acute care surgical patients with rapidly changing physiology, clinical demands and a large volume of data that must be communicated to render safe, effective patient care. Trauma Morning Report format may offer the ideal situation to study and teach sign-outs and resident communication. MATERIALS AND METHODS Surgery Residents were assessed on a 1-5 scale for their ability to communicate to their fellow residents. This consisted of 10 critical points of the presentation, treatment and workup from the previous nights trauma admissions. Scores were grouped into three areas. Each area was scored out of 15. Area 1 consisted of Initial patient presentation. Area 2 consisted of events in the trauma bay. Area 3 assessed clarity of language and ability to communicate to their fellow residents. The residents were assessed for inclusion of pertinent positive and negative findings, as well as overall clarity of communication. In phase 1, residents were unaware of the evaluation process. Phase 2 followed a series of resident education session about effective communication, sign-out techniques and delineation of evaluation criteria. Phase 3 was a resident-blinded phase which evaluated the sustainability of the improvements in resident communication. RESULTS 50 patient presentations in phase 1, 200 in phase 2, and 50 presentations in phase 3 were evaluated. Comparisons were made between the Phase 1 and Phase 2 evaluations. Area 1 (initial events) improved from 6.18 to 12.4 out of 15 (p<0.0001). Area 2 (events in the trauma bay) improved from 9.78 to 16.53 (p<0.0077). Area 3 (communication and language) improved from 8.36 to 12.22 out of 15 (P<0.001). Phase 2 to Phase 3 evaluations were similar, showing no deterioration of skills. CONCLUSIONS Trauma Surgery manages complex surgical patients, with rapidly changing physiologic and clinical demands. Trauma Morning Report, with diverse attendance including surgical attendings and residents in various training years, is the ideal venue for real-time teaching and evaluation of sign-outs and reinforcing good communication skills in residents.

Peritoneal free air due to evacuation of pneumobilia in blunt abdominal trauma

Pneumobilia is mostly observed on computed tomography (CT) following surgical biliary-enteric anastomosis and biliary manipulation through endoscopic procedures. Although pneumobilia can be seen in pathological conditions, post-surgical pneumobilia is typically not associated with morbidity. In the present article, we report a case in which blunt abdominal trauma led to the evacuation of pre-existing pneumobilia causing pneumoperitoneum. Given that the subsequent laparotomy proved to be non-therapeutic, this report adds to the few cases of intra-peritoneal free air not helped by surgical intervention.