86.15 Is Routine Echocardiography Useful in Evaluating Blunt Cardiac Injuries?

Posted on January 26, 2017

R. Uhlich1, P. Hu1, L. D. Raff1, J. D. Kerby1, P. L. Bosarge1  1University Of Alabama at Birmingham,Acute Care Surgery/Surgery/Medicine,Birmingham, Alabama, USA

Introduction:  The use of echocardiography (ECHO) in the trauma patients with suspected blunt cardiac injury (BCI) has been suggested for assessment of arrhythmia or unexplained hypotension.  Despite this recommendation, the diagnostic value of echocardiography is poorly defined.  The purpose of this study was to assess the utility of obtaining diagnostic echocardiography in the patient with BCI. 

Methods:  A retrospective study was conducted at a single verified American College of Surgeons Level I trauma center from June 2014 to July 2016. The study population was limited to trauma patients diagnosed with BCI using electrocardiogram who had an arrhythmia noted. Demographic and medical history data were collected on all patients.  The primary outcome of interest was need for cardiac intervention in the acute hospitalization to include cardiac surgery or cardiac catheterization in any form; secondary outcomes included echocardiography findings and mortality at 48 hours and 30 days.

Results:  BCI was diagnosed in 89 patients over a two-year study period.  ECHO was obtained in 57 of these patients. Of the 57 patients with ECHO performed, only one patient (1.8%) demonstrated significant cardiac wall motion abnormalities (ejection fraction=35%), who had a prior history of congestive heart failure. Of the remaining patients, left ventricular ejection fraction averaged 53.7% (+/- 5.2). No patient required cardiac surgery or cardiac catheterization.  One death (1.8%) occurred within 48 hours of admission; Overall 30-day mortality was 12.5%. No deaths were attributed to cardiac causes. 

Conclusion:  Despite recommendations of obtaining ECHO for patients with arrhythmias related to BCI, echocardiography adds little additional information to the overall care of these patients.  Patients with BCI rarely need cardiac intervention in the absence of associated hypotension suggesting valvular compromise or potential cardiac rupture.  Echocardiography should be limited to only BCI patients who present with hypotension.

 

86.14 Pulmonary Contusions In Elderly Blunt Trauma Are Infrequently Seen On CXR And Are Highly Morbid

Posted on January 26, 2017

A. Bader1, M. Morris1, J. A. Vosswinkel1, J. E. McCormack1, E. C. Huang1, R. S. Jawa1  1Stony Brook University Medical Center,Stony Brook, NY, USA

Introduction:  In patients with blunt chest trauma, pulmonary contusions are variably identified. However, there is limited research on the outcomes of elderly patients with pulmonary contusion.    

Methods:  We retrospectively reviewed the trauma registry for all admissions aged ≥65 years, admitted following blunt trauma with a thoracic injury.  Emergency Room deaths were excluded. The medical records of patients with pulmonary contusions were subsequently reviewed for additional details.

Results: There were 960 patients age ≥65 years admitted with thoracic trauma, of which 180 had  pulmonary contusions (PC) and 780 had no pulmonary contusion (NO). The major mechanisms of injury were MVC/MCC (52.22% PC, 35.64% NO, p<0.001) followed by falls (38.89% PC, 58.72% NO, p<0.001). Rib fractures were present in 80% of PC and 73.5% of NO patients, p=0.09. Hemothorax/pneumothorax was more prevalent in those with pulmonary contusions (44.44% PC vs 19.23% NO, p<0.001). While 98.3% of PC patients had chest AIS≥3, 41.9% of NO patients had chest AIS≥3, p<0.001. Hospitalization outcomes are presented in the table. Chart review of PC patients noted that pulmonary contusion was identified in only 34/180 patients on initial CXR. An additional 22 patients were noted to have pulmonary contusion on a subsequent CXR. A CT thorax was performed in 174 patients within 24 hours of admission. This CT scan identified the pulmonary contusion. Further, rib fractures were identified in 80% of PC patients.

Conclusion: Pulmonary contusions in the elderly blunt trauma population were infrequently identified on CXR.  They are associated with severe chest injury. Their presence is associated with substantial morbidity and mortality. The data suggest the need for increased vigilance for pulmonary contusion such as early chest CT scan performance in this population. Further study is warranted. 

86.13 Quality Improvement Process Lowers Effective Radiation Dosage in Pediatric Trauma Imaging

Posted on January 26, 2017

H. Naseem2, P. Montgomery3, K. D. Bass1,2  1State University Of New York At Buffalo,Department Of Surgery, Jacobs School Of Medicine And Biomedical Sciences,Buffalo, NY, USA 2Women And Children’s Hospital Of Buffalo,Department Of Pediatric Surgery,Buffalo, NY, USA 3Women And Children’s Hospital Of Buffalo,Department Of Pediatric Radiology,Buffalo, NY, USA

Introduction: Radiation safety has been a concern in pediatric trauma with recent efforts at reducing computed tomography (CT) imaging. A previous study at our institurion showed an inverse relationship between age group and effective dose (ED).  Our purpose was to evaluate current data for changes in the trend.

Methods: Retrospective review of the institutional trauma registry for patients who received computed tomography (CT) imaging from 1/1/2013 to 6/30/2014. Variables collected were age, gender, CT dose length product and body area scanned. Effective dose (ED) was calculated for five age groups. Factorial ANOVA analysis was used to calculate statistical significance. 

Results: There were 312 patients that met criteria and 415 CT scans: 86 abdomen/pelvis, 21 chest, 36 facial, 259 head, and 13 neck scans. The ED for chest scans was 6.52±4.38 mSv and 0.57±0.38 mSv for facial scans (p<0.001). The ED was highest in the large child age group with a mean of 3.14±3.34 mSv and lowest in the toddler age group with a mean of 1.50±0.98 (p<0.001). In comparison to data from 2008-2011, the overall ED was lower across all age groups and showed an inverse relationship to the previous trend of higher ED in younger patients, especially noted in CT abdomen/pelvis (Fig 1).

Conclusion: There was a direct relationship between age and effective dose contradictory to the previous data. Overall effective dose radiation was also lower for CT scans compared to previous data. Concerted efforts of justification, optimization, and documentation lowered effective dose radiation in pediatric trauma patients.

 

86.11 Ethnic Disparities in Traumatic Brain Injury Care Referral in a Hispanic-Majority Population

Posted on January 26, 2017

H. C. Budnick1, A. Tyroch1, S. Milan1  1Texas Tech University Health Sciences Center In El Paso Paul L. Foster School Of Medicine,Department Of Surgery,El Paso, TX, USA

Introduction:

Traumatic brain injury (TBI) is a leading cause of death in the United States and the largest and most swiftly growing population of these injuries in the United States is among the Hispanic population. Functional outcomes for TBI cases can be significantly improved by post-hospitalization rehabilitation including intensive physical, occupational, and cognitive rehabilitation. This treatment is usually accomplished by discharge to post-hospitalization care following the acute period. In studying the referral to these facilities, Hispanics have been shown to have the lowest physician referral rate nationally. However, this relationship has not been studied in a population where Hispanics are by far the majority. This study seeks to determine if differences exist between ethnic groups in referral of TBI patients to post-hospitalization care in the Hispanic-majority population of El Paso, Texas. 

Methods:
This study included 1,124 patients over the age of 18 who presented to University Medical Center in El Paso, Texas between the years of 2005-2015 with acute TBI. The patients’ age, sex, race, residence, admission GCS, GCS-Motor, Injury Severity Score (ISS), ICU and hospital length of stay (LOS), mechanism of injury, and discharge referral were extracted. The data was analyzed in univariate and multivariate analysis using SPSS.

Results:
The discharge disposition was found to be significantly different between the Hispanic and the non-Hispanic populations. 70.2% of Hispanic patients were sent home without post-hospitalization care whereas only 53.5% of the non-Hispanic patients were sent home. Hispanics were also sent to acute care facilities 6.9% of the time and to rehabilitation centers 18.5% of the time compared to non-Hispanics who were sent to acute care facilities 10.8% of the time and to rehabilitation 27.5% of the time. Further, the ages of presentation, mechanism of injury, LOS, ISS, GCS, and GCS-M were comparable between the ethnic groups.

Conclusion:
The Hispanic population has been shown to be discharged to post-hospitalization care facilities at a lower rate as compared to non-Hispanic populations. This remains true even where the overwhelming majority of the population is Hispanic such as El Paso, Texas. Further, when risk factors for poor outcomes were stratified by ethnicity, there was no appreciable difference. This suggests that TBI patients of comparable traumatic severity and functional outcome probability but different ethnicities are discharged without further care at different rates.
 

86.10 CPR after Brain Injury caused Traumatic Cardiac Arrest is a Futile Endeavor

Posted on January 26, 2017

L. D. Raff1, P. Hu1, R. Uhlich1, J. D. Kerby1, P. L. Bosarge1  1University Of Alabama at Birmingham,Acute Care Surgery/ Surgery/Medicine,Birmingham, Alabama, USA

Introduction:  The use of cardiopulmonary resuscitation (CPR) following traumatic arrest historically leads to dismal survival rates. Recently, major critical care organizations have endorsed that interventions should be considered inappropriate when there is no reasonable expectation that the patient will improve sufficiently to survive outside the acute care setting or have meaningful neurological recovery.  To date, very little information has been published to suggest if patients with traumatic brain injury (TBI) benefit from CPR and what functional outcomes can be expected if the patient has return of spontaneous circulation (ROSC).  The purpose of this study is to determine the outcome of patients with obvious traumatic brain injury that required CPR.

Methods:  We identified all adult trauma patients that presented with traumatic cardiac arrest from June 1, 2014 to August 1, 2016. Data regarding demographics, mechanism of injury, cardiac rhythm, resuscitation duration, resuscitation procedures, blood product utilization, mortality data, and disposition from hospital were collected. 

Results:  Among 183 cases of CPR due to traumatic arrest, 21 were identified as having an obvious traumatic brain injury.  Of those 21 patients, 18 (85.7%) had penetrating trauma to the head.  The mean age was 39.7 (± 17.4) years and 38.1% were women.  The median prehospital CPR time was 10 (0-50) minutes; the median hospital CPR time was 9 (0-59) minutes; and the median total CPR time was 23 (2-89) minutes.  ROSC occurred in only 2 patients (9.5%) who had a median CPR time of 4 (2-6) minutes.  No patient that presented in asystole had ROSC (0 of 10) versus 20% of those patients that presented in any other cardiac rhythm other than asystole (p=0.065).  Of the two patients that had ROSC, both patients were subsequently declared brain dead yielding no survivors for this patient group.

Conclusion:  In patients with obvious TBI that present in traumatic arrest, overall ROSC is poor.  When ROSC occurs overall mortality remains high as these patients do not survive the neurological insult.   When patients present in traumatic arrest with asystole, efforts to continue resuscitation should be terminated.  Performance of CPR in patients with obvious TBI upon hospital presentation should be considered futile interventions.

 

86.09 Characterization of Vascular Anatomy for REBOA and Endovascular Procedures

Posted on January 26, 2017

N. C. Wang1, P. E. Rabban1, X. Yan2, R. L. Goulson1, B. A. Derstine1, G. L. Su1, H. Lee2, J. L. Eliason1, S. C. Wang1  1University Of Michigan,Surgery,Ann Arbor, MI, USA 2University Of Michigan,Electrical Engineering & Computer Science,Ann Arbor, MI, USA

Introduction: Non-compressible torso hemorrhage is a major cause of mortality in battlefield as well as civilian trauma settings. Resuscitative endovascular balloon occlusion of the aorta (REBOA) has been pioneered by the military to stabilize patients. This technique is starting to be used in civilian trauma settings for the management of torso hemorrhage. To support the design of catheter enhancements and inform medical personnel of optimal balloon placement without fluoroscopy, a broad survey of the variation of aortic geometry is needed.

Methods:  This study measured aortic geometry using the computed tomography (CT) scans of 1769 trauma patients between the ages of 18-50. A custom set of MATLAB algorithms was used to semi-automatically process the aorta for each scans. The centerline and radii of the aorta were measured for these scans from the femoral artery at the level of each femoral head, through the bifurcation, and up into the aortic arch. Additionally vascular landmarks were placed including the aortic bifurcation (AoBi), kidney, celiac, SMA, and left subclavian branches. These landmarks define the aortic zones, zone I (left subclavian to celiac), zone II (celiac to lowest renal artery), and zone III (renal artery to AoBi)

Results: Within our population, the median length (interquartile range in parentheses) of zone I was 223.7 mm (210-237 mm). Zone III was a significantly smaller region, with a length of 87.0 mm (76-96 mm). The distance from the left femoral artery at the femoral head to the AoBi was 195.8 mm (186-206 mm) in men, and 193.0 mm (184-202 mm) in women, with the distance from the right femoral artery to the AoBi being slightly longer than the left at 200.7 mm (191-210 mm) in men and 198.6 mm (189-208 mm) in women.

            The median luminal diameters of the left and right femoral artery were 6.07 mm (5-7 mm) and 6.11 (5-7 mm) respectively. Aortic diameter was largest near the left subclavian, 20.5 mm (19-22 mm). The diameter decreased down the aorta to 18.6 mm (17-21 mm) at the celiac branch, 16.7 mm (15-19 mm) at the lowest renal artery, and 14.3 mm (13-16 mm) at the aortic bifurcation.

Conclusion: Overall, there is significant variation within the population in terms of vascular anatomy. As REBOA is being advocated for use in zone III, to control lower abdominal hemorrhage, it’s important to understand the size and lengths of the vasculature to ensure safe placement within the intended zone. Current catheters range in size up to 14 Fr (4.6 mm), increasing risk of damage to small femoral arteries; recent experience in a Japanese civilian population reported a high incidence of flow occlusion to the lower extremity. The current report may aid in the development and clinical application of novel endovascular devices. 

86.08 Characterizing the Relationship Between Age and Venous Thromboembolism in Adult Trauma Patients

Posted on January 26, 2017

A. J. Nastasi1,2, J. K. Canner1, B. D. Lau1, M. B. Streiff3, J. K. Aboagye1, K. J. Van Arendonk1, P. S. Kraus6, D. B. Hobson5, D. Shaffer5, E. R. Haut1,4  1Johns Hopkins University,Surgery,Baltimore, MD, USA 2Johns Hopkins University,Epidemiology,Baltimore, MD, USA 3Johns Hopkins University,Hematology,Baltimore, MD, USA 4Johns Hopkins University,Health Policy And Management,Baltimore, MD, USA 5Johns Hopkins University,Nursing,Baltimore, MD, USA 6Johns Hopkins University,Pharmacy,Baltimore, MD, USA

Introduction:
Venous thromboembolism (VTE) is a great burden in trauma; however, current guidelines lack recommendations regarding the prevention of VTE in older adult trauma patients. Furthermore, the appropriate method of modeling age in VTE models is currently unclear.

Methods:
3,598,881 patients between the years 2008 and 2014 in the National Trauma Data Bank (NTDB) and 505,231 patients between 2009-2013 from the National Inpatient Sample (NIS) were analyzed. Multiple logistic regression of VTE on age was performed. Based on unadjusted VTE incidence, age was modeled as a linear spline with a knot at age 65.

Results:
In the NTDB, 34,202 (0.95%) patients were diagnosed with VTE while 1,709,881 (47.5%) patients were ≥65 years. In both the fully adjusted NTDB and NIS model, age was positively associated with VTE incidence until age 65 (NTDB: aOR 1.018, 95% CI 1.017 – 1.019, p < 0.001; NIS: aOR 1.025, 95% CI 1.022 – 1.027, p < 0.001). In patients ≥65 years, age was inversely associated with VTE in the NTDB model (aOR 0.995, 95% CI 0.992 – 0.999, p = 0.006) and not associated with VTE risk in the NIS model (aOR 0.998, 95% CI 0.994 – 1.002, p = 0.26).

Conclusion:
VTE risk in adult trauma patients appears to steadily increase with age until 65 years, after which risk appears to level off or even slightly decrease. These findings should be considered when creating standardized guidelines for VTE prevention in older adults as well when modeling age in VTE models of adult trauma patients.
 

 

 

 

86.07 BURDEN AND CHARACTERISTICS OF GLASS TABLE INJURIES

Posted on January 26, 2017

D. Chauhan1, C. Villegas1, R. Bueser1, S. Bonne1, D. Livingston1  1Rutgers – New Jersey Medical School,Department Of Trauma/critical Care,Newark, NJ, USA

Introduction:
Our trauma center has observed an increase in children with severe injuries from glass tables. This mechanism of injury is not well described. The goals of this study were to describe the burden of glass table injuries using the National Electronic Injury Surveillance System (NEISS) dataset and compare it to our data from a level 1 urban trauma center.

Methods:
The NEISS dataset from 2009 to 2015 was reviewed for glass table injuries. Data on demographics, injury severity, its description and outcomes were extracted. Cases were divided as they related to the glass shattering: definite, probable and not due to faulty glass. Similarly, our trauma registry was queried for all patients injured involving breaking glass tables from 2001 to 2016. An online search of 3 furniture websites was performed for details of the glass and safety information provided to consumers.  

Results:
3241 patients were reviewed in the NEISS data; 1151 definitely and 665 probably sustained injury due to the glass itself. Defined criteria (e.g. torso penetration or shock) to classify injury severity were developed and each injury was classified as mild or severe.  265 injuries were severe, 233 (88%) of which were due to faulty class.  There was a bimodal distribution in age with peaks at 2 and 22 years.  Demographics, injury pattern and disposition are shown in the Table and is compared to the 24 patients treated at our trauma center. No websites provided any safety instructions for glass tables. 113/300 (38%) tables examined had no information if the glass was tempered or on glass thickness. Currently there are no quality requirements for glass tables in the United States. 

Conclusion:
Glass table injuries are not uncommon, occasionally lethal and preventable. The burden is real, likely under reported and costly. Children are especially at risk. Warnings to consumers and enactment of glass standards by the Consumer Product Safety Commission is warranted. 
 

86.06 Current Practice Patterns and Burnout of Trauma and Acute Care Surgeons

Posted on January 26, 2017

N. Droz2, P. Parikh2, M. Whitmill2, K. M. Hendershot1  1University Of Alabama at Birmingham,Acute Care Surgery,Birmingham, Alabama, USA 2Wright State University,Trauma And Acute Care Surgery,Dayton, OH, USA

Introduction:

Providing 24/7 care for our patients and supervision for our residents/fellows is a cornerstone in the Trauma/Acute Care Surgeons (T/ACS) work-life.  Our previous work defined what the T/ACS current practice pattern is (majority in a group practice with shared responsibilities; majority take in-house call in 24 hour shifts and 3-7 calls per month; majority staying part or all of post-call day).

The purpose of our current study is to look at the T/ACS attitudes regarding their practice patterns, specifically related to their call schedule and post-call day.  Issues such as post-call fatigue and burnout related to their work schedule are also explored.

Methods:

An IRB-approved electronic survey was distributed nationally to Eastern Association for the Surgery of Trauma members.  Participants were asked about attitudes related to their call schedule and coverage they provide while on call.  They were also asked about fatigue and burnout related to their work schedule.

Results:

A total of 274 participants were analyzed (response rate 20%).  The majority like their call schedule structure and length of their call shifts (62% and 66%, respectively).  The scope of practice was liked by 77% with 14% not liking the elective surgery aspect of the practice.  The majority (86%) covers all trauma, emergency general surgery (EGS), and surgical critical care while on call and 75% feel this is an adequate amount of work to cover while on call.  The majority (83%) think they should get paid for trauma/EGS call.

 

Although 75% state they are able to get some rest while on call, 56% are “very tired” post-call and 29% have fallen asleep while driving post-call.  The majority (71%) is concerned about fatigue post-call; 67% are concerned about being over-worked, and 72% are concerned about burnout.  A change in their practice pattern could help with fatigue and feelings of being overworked according to 72% of participants.

Conclusion:

Despite the majority of T/ACS expressing concern about post-call fatigue, being overworked, and feelings of burnout, less than 20% have developed or implemented any innovative strategies to change the call structure and post-call day.  Change is often difficult, so trying to think outside the box and develop novel approaches to attendings’ practice patterns should be encouraged and shared with the larger trauma community.

 

86.05 Race and Insurance Status as Predictors of Outcomes and Care in Severe Thoracic Trauma

Posted on January 26, 2017

R. A. Rauh1, T. J. Zens1, G. Leverson1, M. V. Beems1, S. K. Agarwal1  1University Of Wisconsin,Trauma And Acute Care Surgery,Madison, WI, USA

Introduction:   

Healthcare disparities based on race and socioeconomic status have been documented in the literature; however, data on how these factors effect outcomes in patients experiencing severe thoracic trauma is lacking. This study aims to identify potential disparities in treatment and outcomes in this patient population.

Methods:

The National Trauma Data Bank was queried for all rib fracture patients with ISS scores>15 between 2007-2012. A univariate and multivarite logistic regression model was run which controlled for patient co-morbidities, age, ISS, and associated injuries.  Patient outcomes in length of stay, mortality, discharge disposition, and in hospital procedures were compared between patients of varying race and insurance status to white and privately insured patients, respectively.

Results: 

A cohort of 69,424 patients were selected for analysis.  87.1% of patients were white, 10.2% African American and 1.98% Asian. 14.2% of patients were covered by private insurance vs. 30.1% by Medicare and 21.5% by Medicaid.  34.1% were uninsured.  Uninsured (OR = 1.753; CI = 1.468- 2.094), Medicaid (OR = 1.568; CI = 1.295-1.898), and Medicare (OR = 2.768; CI = 2.313-3.313) patients had higher in-hospital mortality than privately insured patients. Uninsured patients (OR = 0.804; CI = 0.745, 0.867) were less likely to exceed the median hospital stay, while Medicaid (OR = 1.445 CI = 1.331-1.568) and African American patients (OR = 1.144, CI= 1.083-1.208)  were more likely exceed the median hospital stay than those privately insured.  Medicare (OR = 1.103; CI = 1.004-1.212) and Medicaid (OR = 1.328; CI =1.210-1.458) patients were more likely to receive an epidural during the course of care than privately insured patients, but there were no other statistically significant differences with regards to race or insurance status. Medicaid (OR=1.330; CI = 1.216-1.453) and African American patients (OR = 1.081; CI= 1.018-1.148) were more likely to require mechanical ventilation than privately insured or White patients.  Finally, uninsured patients (OR=0.572; CI = 0.505-0648) were less likely to receive continuing medical care after hospitalization in a nursing facility or acute care rehab center.  In contrast, Medicaid (OR=1.412; CI = 1.249-1.595) and Medicare (OR = 3.661; CI = 3.252- 4.121) patients were more likely to be discharged one of these facilities.

Conclusion: 

When examining healthcare disparities among thoracic trauma patients, we documented less significant differences among racial groups than among insurance statuses.  Overall, we found the uninsured were more likely to be discharged early to their homes while Medicare and Medicaid patients were more likely to be discharged to a care facilities such as nursing homes or acute care hospitals. We also found the privately insured had lower mortality than Medicare, Uninsured and Medicare patients.  Further research is needed on whether changes implemented by Affordable Care Act have helped to eliminate this disparities.

86.04 Time to Surgical Source Control in Intra-Abdominal Infections

Posted on January 26, 2017

R. Chang1, M. Scerbo1, L. Moore1, A. Macaluso1, C. Wade1, J. Holcomb1  1University Of Texas Health Science Center At Houston,Surgery,Houston, TX, USA

Background: Although many infections can be treated with antibiotic therapy alone, intra-abdominal infections (IAI) often require surgical intervention to achieve adequate source control. Time to initiation of appropriate antibiotic therapy is a well-described quality metric in the treatment of life-threatening infections (sepsis), but time to operation for source control has not been amply investigated for surgical sources of infection. We hypothesized that decreased time to laparotomy (TTL) to achieve surgical source control was associated with improved outcomes in patients presenting with IAI.

Methods: Billing codes were used to identify adult patients who underwent laparotomy from 2011-2015 at a single center. These were screened to identify patients who presented to the emergency department (ED) with IAI, underwent laparotomy for source control, and had hospital stay >24 hours. TTL was defined as the time from ED triage to initiation of laparotomy. The SOFA score was calculated using parameters obtained in the ED. The primary outcome was survival to hospital discharge; the secondary outcome was ICU-free days. Using SOFA score as a covariate, we constructed multivariable logistic and linear regression models to test the hypothesis that decreased TTL was associated with increased survival and increased ICU-free days respectively.

Results: Of the 54 patients included for analysis, 46 (85%) survived to hospital discharge. Overall incidence of sepsis (defined as change in baseline SOFA ≥2) was 57%. Median ICU-free days was 26 with interquartile range of 15 to 30. Survivors had lower SOFA scores (median 2 vs 7, p<0.01) but similar TTL (median 16 vs 17 hours, p>0.05) compared to non-survivors. For patients with sepsis, TTL was also similar between survivors and non-survivors (median 15 vs 17 hours, p>0.05). Perforated hollow viscus accounted for 54% of infectious sources (colorectal 20%, small bowel 17%, stomach 17%), and intra-abdominal abscess accounted for 46%.

Decreased TTL was not associated with improved survival (odds ratio 1.00, 95% confidence interval [CI] 0.98 – 1.02) on multivariable logistic regression, but was significantly associated with increased ICU-free days (relative risk -0.05, 95% CI -0.10 to -0.01) on multivariable linear regression.

Conclusion: Although there was no difference in mortality, decreased TTL was associated with increased ICU-free days in patients presenting with IAI requiring laparotomy. Despite the emphasis on time to initiation of antibiotic therapy, comparatively little attention has been paid to time to surgical source control, even though both are needed to treat certain cases of IAI.

86.03 Laboratory versus clinically-evident coagulopathy: results from PROHS

Posted on January 26, 2017

R. Chang1, E. Fox1, T. Greene1, M. Swartz1, S. DeSantis1, D. Stein6, E. Bulger4, S. Melton8, M. Goodman2, M. Schreiber5, M. Zielinski3, T. O’Keeffe9, K. Inaba7, J. Tomasek1, J. Podbielski1, C. Wade1, J. Holcomb1  1University Of Texas Health Science Center At Houston,Houston, TX, USA 2University Of Cincinnati,Cincinnati, OH, USA 3Mayo Clinic,Rochester, MN, USA 4University Of Washington,Seattle, WA, USA 5Oregon Health And Science University,Portland, OR, USA 6University Of Maryland,Baltimore, MD, USA 7University Of Southern California,Los Angeles, CA, USA 8University Of Alabama At Birmingham,Birmingham, AL, USA 9University Of Arizona Medical Center,Tuscon, AZ, USA

Introduction: Laboratory evidence of coagulopathy is observed in 25% of severely injured trauma patients, but clinically-evident coagulopathy (CC) is not well-described. This study investigates the characteristics of CC and seeks to identify any potentially modifiable prehospital risk factors of CC.

 

Methods: The Prehospital Resuscitation on Helicopters Study (PROHS) was a prospective observational study of adult trauma patients transported by helicopter from the scene to one of nine Level 1 trauma centers in 2015. Predefined highest-risk criteria were any of the following during helicopter transport: heart rate >120 bpm, SBP ≤90 mmHg, penetrating truncal injury, tourniquet application, pelvic binder application, or intubation. Patients meeting any highest-risk criteria were divided into 2 groups based on presence of CC, defined as surgeon-confirmed bleeding from uninjured sites or injured sites not controllable by normal means (e.g. sutures). Purposeful multiple logistic regression was performed to identify potentially modifiable prehospital risk factors of CC.

 

Results: Of the 2341 patients enrolled, 1058 (45%) met highest risk criteria and were divided into CC (n=43, 4%) and not CC (n=1015, 96%) groups. CC patients were older (median age 50 vs 38), more severely injured (median ISS 30 vs 17), and were more likely to have had penetrating trauma (33% vs 19%), prehospital RBCs and/or plasma (56% vs 12%), and laboratory evidence of coagulopathy on admission (86% vs 46%) (all p<0.05). Prehospital crystalloid volumes were similar (median 200 vs 250ml), and transfusion ratios were balanced. CC patients had increased mortality at 30 days (60% vs 15%, p<0.01); although the leading cause of death was TBI in both groups (54% vs 66%), exsanguination was increased in CC patients (38% vs 18%, p<0.01). Transport time, prehospital RBC or plasma units, and crystalloid volume were not significant predictors of CC on multiple logistic regression after controlling for age, ISS, mechanism, admission GCS, and availability of prehospital blood products.

 

Conclusion:

Despite the relatively common finding of laboratory evidence of coagulopathy, CC was rare (4%) but associated with substantial mortality. No obvious modifiable prehospital risk factors of CC were identified.

86.02 Do Traffic Law Violators Have Differing Attitudes About Their Driving Behaviors?

Posted on January 26, 2017

J. A. Vosswinkel1, K. L. Ladowski1, J. E. McCormack1, R. S. Jawa1  1Stony Brook University Medical Center,Stony Brook, NY, USA

Introduction: Despite advances in engineering, motor vehicle crashes remain a leading cause of injury morbidity and mortality, due in great part to driver behaviors such as speeding and inattention.

Methods: In 2015, the County’s Traffic & Parking Violation Agency began offering a 3 hour traffic violator course as part of a plea deal to reduce points/fines. Course instructors include representatives from agencies with a vested interest in traffic safety; county police department, defense lawyer and/or judge, trauma center, and a local human service agency. The course is divided into 2 sections based on type of violation received: Dangerous Driver (DD) program for speeding and/or aggressive driving and Inattentive Driver (ID) program for cell phone violations. Both courses cover similar content including traffic laws, judicial consequences of unsafe driving, and emotional/physical consequences of unsafe driving. After the course, participants are given an anonymous post-then-pre survey about their driving behavior. The Likert-type answers are coded numerically (1=Not At All, 2=A little, 3=somewhat, 4=A lot). Retrospective data analysis was performed using Student’s t-test.

Results: There were 214 surveys (139DD, 75 ID) collected from 11 classes (7DD, 4ID). We analyzed 5 key questions about their driving behavior: 1.Worried it could result in legal consequences; 2.Worried it could cause injury; 3.Want to change driving behavior to avoid legal consequences; 4.Want to change driving behavior to avoid causing injury; 5.Believe can improve driving behavior. Results presented below. Of note, the final 3 columns compare the mean change in attitudes between the DD and ID program. Furthermore, the DD overall scores were significantly higher both pre (DD=14.5±3.2, ID=12.7±3.1) and post (DD=17.4±3.4, ID=16.0±3.5)(p <.001), but there was no significant difference in the overall score increases between groups (DD=2.6±4.4, ID=2.9±3.8, p=0.60).

Conclusion: On course completion, both groups were more likely to agree with statements that worried about their driving behavior and more likely wanted to change their driving behavior. Although both group’s overall scores increased similarly, there were underlying differences in the attitudes of driving behavior between the DD and ID groups.  Overall, ID were less likely to worry about their driving behavior and less strongly felt they needed to change their driving behavior compared to the DD both before and after the class. These findings are similar to other studies that have concluded that drivers who operate cell phones tend to overestimate their driving ability and underestimate the demands of driving.  Further study is warranted.

86.01 Laproscopic Cholecystectomy In The Acute Care Surgery Model: Risk Factors And Complications

Posted on January 26, 2017

E. Sweet1, E. Seabold1, K. Herzing1, R. Markert1, A. Gans1, A. Ekeh1  1Wright State University,Surgery,Dayton, OH, USA

Introduction:
The Acute Care Surgery (ACS) model has been widely popularized over the last decade – fusing the care of Trauma and Emergency General Surgery patients. Laparoscopic Cholecystectomies (LC) are commonly performed by ACS teams typically for acute indications admitted from the Emergency Department. We reviewed LCs performed by an ACS service with > 3000 Trauma annual admissions, focusing on outcomes and risk factors for complications in the emergent setting.

Methods:
All patients who underwent LC on our ACS service over a 26 month period (Jan 2014-Feb 2016) were identified. Data including demographic data, BMI, indications for surgery, time of day of surgery (am or pm), surgeon years of experience, rate of conversion to open, bile leaks, major biliary injury and other complications were collected. Risk factors for complications were analyzed using Chi-squared and Mann-U Whitney tests.

Results:
There were 547 patients who had LC in the studied period (70.2% female, mean age 46 years, meanBMI 32.4 kg/m2) performed by 11 surgeons. Indications for surgery included Acute Cholecystitis(46.8%), Symptomatic Cholelithiasis (25.2%) and Gallstone pancreatitis (6.6%) Mean surgery time was 79±50 mins and 5.7% of cases were performed "after hours." Conversion to open rate was 6%. Minor bile leaks were present in 3.8%, retained stones in 1.1%, post-op bleeding in 1.1% and major duct injury in 0.9%. Statistical analysis did not identify any risk factors for bile leaks, majorbiliary injury or other complications.

Conclusion:
ACS services are capable of performing a high volume of LCs with low complication and conversion to open rates. The majority of LCs were for emergent indications. No correlations between complications and patient age, gender, BMI, indications for surgery, surgeon experience or time of day of operation were found. The ACS model is well suited to address needs of patients acute biliary disease. 
 

85.20 Nationwide Evaluation of Pediatric Non-Cardiac Thoracic Trauma

Posted on January 26, 2017

I. I. Maizlin1, R. T. Russell1  1University Of Alabama at Birmingham,Pediatric Surgery,Birmingham, Alabama, USA

Introduction:  Trauma is the single greatest cause of mortality in the pediatric population. While chest trauma accounts for less than 10% of trauma affecting children, it is relevant because of the considerable mortality associated with it. We aim to identify nationwide trends in circumstances and clinical outcomes resulting from pediatric non-cardiac thoracic accidents.  

Methods: The National Trauma Database (2010-2012) was reviewed for patients ≤19 years of age admitted with diagnosis of non-cardiac thoracic trauma. Patients were stratified into 3 age groups: 1-9 years, 10-16 years, 17-19 years. Demographics, patterns of injury, and outcomes were evaluated, with chi-square and ANOVA tests used for analysis. 

Results: 59,027 children (67.4% male, 59.0% white) were admitted with thoracic injuries, with mean age of 14.2 ± 5.5 years. 68.4% of the children were injured in motor vehicle accidents, 17.2% in assaults, and 6.6% in falls. 43.1% of the accidents resulted in lung contusions, 16.7% pneumo- or hemothorax, 9.0% rib fractures, 7.4% open chest wounds. Mean hospital stay was 6.7 days (compared to 5.0 days in the overall pediatric trauma population), with 45.4% of the patients admitted to the ICU, and 23% requiring ventilator support. As compared to the overall national trauma mortality rate in this population of 2.7%, children experiencing thoracic trauma had a much higher mortality rates of 7.3%. When evaluating etiologies by age group (Table 1), youngest patients were more likely to suffer thoracic trauma as a result of accidents and falls, while trauma in the oldest patients was more likely to be caused by assaults and self-inflicted injuries. Compared to younger age groups, patients 17-19 years old were most common group to present with thoracic trauma, with the highest mean thoracic Abbreviated Injury Score (AIS) and the longest mean hospital stay (7.22 days, p<0.001). However, the youngest group had the highest rate of ICU admissions (48.7%, p=0.007), greatest rate of ventilator requirements (24.6%, p=0.011) and highest associated mortality (8.3%, p=0.001). Despite no difference in Injury Severity Score (16.03 vs. 16.10 vs. 16.77, p=0.280), the youngest group was also associated with higher rates of concurrent head trauma (18.6% vs. 13.8% vs. 12.2%, p<0.001) and higher mean head AIS (3.32 vs. 3.17 vs. 3.14, p=0.002). 

Conclusion:  Thoracic trauma results in a significant number of pediatric injuries and trauma-related admissions, especially in the 17-19 year old age-group. However, children below age 9 were most likely to suffer from associated morbidities and mortality, possibly due to a higher rate of concurrent head traumas.
 

85.19 Emergency Abdominal Surgery: Is it Time to Move to Laproscopic Approach?

Posted on January 26, 2017

A. CHEAITO1, A. CHEAITO1  1University Of California – Los Angeles,General Surgery,Los Angeles, CA, USA

Background: Emergent abdominal surgery carries a considerable risk of mortality and postoperative complications. Population-based studies evaluating laparoscopy and outcomes compared with open surgery have concentrated on elective settings. As such, data assessing emergent laparoscopic abdominal surgeries are limited. Our goal was to evaluate the current usage and outcomes of laparoscopic surgery in the emergent setting at a single tertiary academic center.

 

Materials and

Methods: We report a retrospective review of 165 patients who underwent emergent surgery over a 3 year period. Demographics, perioperative clinical variables evaluated. Primary outcomes (30 day mortality) and secondary outcomes (length of hospitalization, prolonged ileus, wounds infection, pneumonia, sepsis, need for secondary procedure, median operative time, conversion rate, and cost) were evaluated.

 

Results: A total of 16 patients died within 30 days of surgery. In all, 58 of the patients were treated with laproscopy with a 30 day mortality of 6.9% vs 11.2% for open surgery (P=0.4). 3.4% patients had pneumonia in the laproscopy group compared to 16.8% in the open group (p=0.012); 55.1 % of the open group required secondary intervention compared to only 27.6% in the laproscopy group (P=0.01). 100% of all laproscopic cases had all layers of their wound closed compared to 77.6% of open cases (P=0.01). More patients in the open group required blood transfusions compared to the laproscopic group (28% vs 8.6%, P=0.05).

Discussion: Laparoscopy has begun to be the preferred method to manage emergent surgical abdomen, but only few reports are available actually. Our analysis revealed lthat only 50% of emergent abdominal surgeries are performed laparoscopically. Outcomes following laparoscopic surgery in this setting resulted in reduced mortality, length of stay, lower complication rates, and less discharges to skilled nursing facilities. Increased adoption of laparoscopy in the emergent setting should be considered.

85.18 Reversal of Antiplatelet Therapy in Traumatic Intracranial Hemorrhage: Does Timing Matter?

Posted on January 26, 2017

A. Malik2, M. Messina3, U. Pandya1  1Grant Medical Center,Trauma Services,Columbus, OH, USA 2Northeastern Ohio Medical University,Rootstown, OH, USA 3Ohio University,Heritage College Of Medicine,Athens, OH, USA

Introduction: The utility of antiplatelet therapy reversal with platelet transfusion in patients with traumatic intracranial hemorrhage remains controversial.  Several studies have examined this topic but few have investigated whether the timing of platelet transfusion has any effect on outcomes.

Methods: Medical records of all patients admitted to a level 1 trauma center from 1/1/14 to 3/31/16 with blunt traumatic intracranial hemorrhage who were taking pre injury antiplatelet therapy were retrospectively analyzed.  Patients on concurrent pre injury anticoagulant therapy were excluded.  Per institutional guideline, patients on pre injury clopidogrel received 2 doses of platelets while patients on pre injury aspirin received 1 dose of platelets.  Time from hospital presentation to start of platelet transfusion was determined and patients were categorized as either receiving early transfusion (≤ 240 minutes) or late transfusion (> 240 minutes).  Primary outcomes of interest included intracranial hematoma expansion and in-hospital mortality.  Presence of hematoma expansion was determined by radiologic interpretation comparing initial and repeat head CT imaging.   P-values < 0.05 were considered statistically significant.  Multivariate regression analysis was used to control for potentially confounding variables.

Results:A total of 276 patients met inclusion criteria with 97 in the early transfusion group and 151 in the late transfusion group.  There were no significant differences in age, head abbreviated injury scale, or gender between the groups.  Patients in the early group had a significantly higher injury severity score (18.7 ± 8.79 versus 15.0 ± 7.90, p=0.0006) and significantly lower admission Glasgow coma scale (13.1 ± 3.83 versus 14.2 ± 2.57, p= 0.003).   The early transfusion group had a higher rate of hematoma expansion (36.9% versus 18.8%) and a higher mortality (22.7% versus 5.3%).  After multivariate regression analysis, however, there was no significant difference in hematoma expansion or mortality between the early and late transfusion groups.

Conclusion:After correcting for confounders, patients on pre injury antiplatelet therapy who received early platelet transfusion after traumatic intracranial hemorrhage did not have significantly different rates of mortality or hematoma expansion than patients who received later platelet transfusion.  Early platelet transfusion may not be beneficial in this patient population.

 

85.17 Primary Non-Hodgkin Lymphoma of the Gallbladder: Characteristics and Outcomes of 106 Patients

Posted on January 26, 2017

A. Ayub1, S. Rehmani1, A. Al Ayoubi1, W. Raad1, J. McGinty2, G. Kim2, F. Y. Bhora1  1Mount Sinai School Of Medicine,Division Of Thoracic Surgery / Department Of Surgery / Mount Sinai West,New York, NY, USA 2Mount Sinai School Of Medicine,Department Of Surgery / Mount Sinai West,New York, NY, USA

Introduction:  Primary Non-Hodgkin lymphoma of the gallbladder (PNHLGB) is extremely rare with limited available data. In this study, we sought to evaluate the clinical features and outcomes of patients with PNHLGB utilizing a population-based database.

Methods:  Surveillance, Epidemiology, and End Results (SEER) database was queried for patients with gallbladder cancer between 1973 and 2013. Only patients with histologically proven PNHLGB were included.  Demographics, tumor characteristics, and outcomes were assessed. 

Results: Of 150 gallbladder lymphomas in the SEER database, 106 cases had PNHLGB and were included in the study. The mean age at diagnosis was 71 (±15) years. PNHLGB primarily afflicted whites (92%) with a male: female of 1.03: 1. Diffuse large B-cell lymphoma (DLBCL) was the most common histological subtype (33%); majority (61%) had loco-regional disease. Surgical resection was performed in 85% cases. Median overall survival of the whole cohort was 41 months with a 5-year survival rate of 40%. In multivariate analysis, increasing age at diagnosis (p<0.001) was associated with increased hazards of death, surgical resection had a protective effect (p=0.007), while gender, race, tumor histology and disease stage were not associated with overall survival (Figure). 

Conclusion: This study represents the largest series of PNHLGB to be reported. Compared to other gastrointestinal lymphomas reported in the literature, PNHLGB appears to have worse prognosis and surgical resection provides survival benefit. Further studies with information regarding adjunctive therapies are warranted.  

 

85.16 Outcomes of Cholecystectomy in Patients with End Stage Renal Disease

Posted on January 26, 2017

I. Olorundare1, S. DiBrito1, C. Holscher1, C. Haugen1, D. Segev1  1Johns Hopkins University School Of Medicine,Surgery,Baltimore, MD, USA

Introduction:  Patients diagnosed with End Stage Renal Disease (ESRD) are a growing population who are at risk of cholelithiasis and cholecystitis and may require surgical management. Despite this, previous studies of outcomes in this population have been limited by small sample size and a lack of generalizability. We studied outcomes of ESRD patients following cholecystectomy in a large nationally representative database

Methods: We used the Nationwide Inpatient Sample to study 40,765 ESRD and 5.4 million non-ESRD patients who underwent cholecystectomy from 2000-2011. Postoperative complications were defined by ICD-9 codes. Mortality, complication rates, LOS, and hospital costs were compared using hierarchical logistic regression, hierarchical negative binomial regression, and mixed effects log-linear models respectively.

Results:ESRD patients had significantly higher mortality and postoperative complication rates than non-ESRD peers (5.0% vs 0.7%, p<0.001) and (23.1% vs 12.8%, p<0.001) respectively on primary admission. After accounting for patient and hospital level factors, ESRD patients had a greater risk of mortality (OR 4.03, 95% CI 3.08 – 5.26) and postoperative complications (OR 2.42, 95% CI 2.09 – 2.81). In particular, they were at a greater risk of infectious (OR 2.98, 95% CI 2.68 – 3.32), mechanical wound (OR 2.21, 95% CI 1.82 – 2.69), and intraoperative complications (OR 1.53, 95% CI 1.32 – 1.78). Median length of stay (LOS) was longer in ESRD patients (8 vs 3 days, p<0.001) as were median hospital costs ($17169 vs $8762, p<0.001). In adjusted analysis, ESRD patients were at significantly greater risk of extended LOS (RR 1.48, 95% CI 1.45 – 1.50) and higher costs (Ratio 1.36, 95% CI 1.34 – 1.39).

Conclusion:ESRD patients experience higher postoperative mortality, complication rates, hospital costs and an extended length of stay following cholecystectomy when compared to non-ESRD peers. Interventions targeting better control of postoperative wound and infectious complications may allow for improvement in overall outcomes of ESRD patients following cholecystectomy. 

 

85.14 A Comparison of Low Tidal Volume Ventilation to Airway Pressure Release Ventilation in ARDS Patients

Posted on January 26, 2017

K. L. Haines1, H. S. Jung1, S. K. Agarwal1  1University Of Wisconsin,Surgery,Madison, WI, USA

Introduction:  Acute respiratory distress syndrome (ARDS) is currently associated with 30-40% mortality and responsible for approximately 75,000 deaths in the United States yearly. Low tidal volume ventilation [the Acute Respiratory Distress Syndrome Network (ARDSNet) strategy] and Airway Pressure Release Ventilation (APRV) are routinely used for these patients.  Past studies comparing known injurious ventilator strategies have shown increases in cytokines after 1 hour of exposure with a return to baseline between 1-6 hours after re-institution of a protective strategy.  This study was undertaken to determine if these modes could be compared at this rate in the same critically ill patients.

Methods:  This was a prospective blinded randomized comparison trial of ARDSNet and APRV in clinically stable consented subjects with ARDS for less than 7 days prior to enrollment.  Patients were randomized to APRV followed by ARDSNet or ARDSNet followed by APRV. Arterial blood gasses and physiologic parameters were collected for analysis prior to intervention, 6 hours, and 12 hours.  Patients with respiratory failure for greater than 14 days, a diagnosis of ARDS for greater than 7 days, or history of lung disease prior to evaluation were excluded.

Results:15 patients were screened for the study, 6 qualified, 4 were able to complete the study and be randomized.  No patients in this trail had symptoms of Left sided heart failure and none were on vasopressors at any point in the trial.  Patients were on the ventilator for 3±1 day prior to study initiation.  Pre-intervention measures were RR 16±3, PEEP 8±2, PIP 24±4, Plateau Pressure 20±5, MAP 89±13, Temp 38±0.1, and Riker 4±1.  There was no difference in Fi02 48±10 for any patient throughout the intervention and oxygen saturation improved or was unchanged in all patients throughout the trial period.  Respiratory rate was unchanged throughout the trial in all patients. No adverse events occurred in this study. Data are mean ± STDEV unless otherwise stated.  

Conclusion: Data from the trial shows that ARDSNet and APRV can both be utilized quickly and safely back to back in surgical critical care patients with ARDS despite being known injurious ventilator strategies.  Not one patient was harmed by this intervention, and all except one patients oxygenation parameters improved.  This will allow further studies to evaluate biomarkers of lung injury in the same surgical critical care patients back to back to see how these ventilator strategies correlate with the degree of lung injury on the molecular level.   

 

« Previous Page
Next Page »