15.10 End of Life Decision Making for Geriatric Trauma Intensive Care Patients

Posted on January 18, 2018

M. Wooster4, A. Stassi5, J. Kurtz3, J. Hill2, M. Bonta6, M. C. Spalding2  2Grant Medical Center,Trauma And Acute Care Surgery,Columbus, OH, USA 3Doctor’s Hospital,General Surgery,Columbus, OH, USA 4Indiana University School Of Medicine,Trauma And Acute Care Surgery,Indianapolis, IN, USA 5University Of South Carolina, Palmetto Health-Richland,Trauma And Acute Care Surgery,Columbia, SC, USA 6Riverside Methodist Hospital,Trauma And Acute Care Surgery,Columbus, OH, USA

Introduction:  The geriatric trauma population is growing and fraught with poor physiological response to injury and high mortality rates. We investigated end of life (EOL) decision making of geriatric trauma patients. We hypothesize that age, religion, injury severity score (ISS), decision maker, pre-existing medical conditions, living wills/advanced directives/do not resuscitate status, and in-hospital complications will affect decision making regarding continued life support (CLS) versus withdrawal of care (WOC). 

Methods:  We performed a retrospective review of geriatric trauma patients at a level I and level II trauma center from January 1, 2007 to December 31, 2014. 274 patients met inclusion criteria with 144 patients undergoing CLS and 130 WOC.

Results: 35,747 geriatric trauma patients were admitted. Age, Catholicism, insurance type, massive transfusion protocol, antithrombotic therapy, ventilator days, ICU length of stay (LOS), and overall LOS were found to be statistically significant (p<0.05) predictors of WOC. After logistic regression, insurance type and Injury Severity Score were found to be significant (p=0.013/0.045). WOC patients had shorter time to palliative consultation. Patients with geriatrics consultation were 16.1 times more likely to undergo CLS (p=0.026). There was no difference in outcomes relative to patients advanced directives/living will/do not resuscitate status prior to hospital admission. However, 16% (44/274) of patients who underwent CLS or WOC had an advanced directive/living will/do not resuscitate status prior to hospital admission eventually progressed to WOC.

Conclusion: Our study examined the complex nature of EOL decisions and revealed difficulty in discerning progression to WOC versus CLS based on demographics, pre-hospital, and in-hospital factors. We also observed an apparent disconnect between the patient's wishes via living wills/advanced directives/do not resuscitate orders and fulfillment during EOL decision-making. Both geriatric and palliative care consultations are encouraged and may influence end of life decision making in geriatric trauma patients.

 

15.09 Blunt Cerebrovascular Injury: Does Early Therapy Alter Injury Grade?

Posted on January 18, 2018

A. Kaple2, I. Catanescu1, M. C. Spalding1  1Grant Medical Center,Trauma,Columbus, OHIO, USA 2Ohio University,Heritage College Of Osteopathic Medicine,Dublin, OHIO, USA

Introduction:  Blunt cerebrovascular injury (BCVI) affects 1-2% of all traumas and leads to increased risk of stroke and neurological sequelae if not treated. However, many cases of BCVIs occur in a poly-trauma setting, delaying the initiation of antiplatelet therapies (APT). Such cases include comorbidities like solid organ injury and traumatic brain injury. Though studies have suggested that it is safe to start APT in certain cases, there is a lack of data in regards to timing of therapy initiation. The purpose of our study was to analyze the change in grade of BCVI as a function of initiation of APT.
 

Methods:  This was a retrospective study of blunt traumas with radiographic BCVI diagnosis performed at a level one trauma center from October 2016 to July 2017. Initially, the cohort included 115 patients. Exclusion criteria was defined as; injuries by a penetrating mechanism, atherosclerotic vessels, or confounding artifact on imaging. 104 blunt trauma patients with 153 total blood vessel injuries comprised the study population. Variables analyzed included; neurological exam, medication used for APT, time to initiate treatment, and angiographic findings. Primary outcomes were; death, stroke, resolution or progression of BCVI. Secondary outcomes included; hospital and ICU stay, DVT, sepsis, and cardiac arrest. We defined early treatment as an initiation under 48 hours, between 2-10 days, and greater than 10 days. Patients were organized by Grade of BCVI, and then compared between different treatment initiation times.
 

Results: Out of 153 BCVIs, 58.2% were Grade 1, 17.6% were Grade 2, 15.7% were Grade 3, 8.5% were Grade 4, and no Grade 5 injuries were encountered. There was a significantly higher mortality for patients with a Grade 4 BCVI (p < 0.05). Regarding the outcomes of Grade 1 BCVIs, there were significant differences when compared to other grades (p < 0.05).  However, there was no statistical significance in the timing of treatment versus BCVI progression (p=0.73). For BCVIs treated under 48 hours, 59.6% improved.  When treated between 2 and 10 days, 56.3% of BCVIs improved. BCVIs treated after 10 days had an improvement rate of 66.7%. Treatment arms were no different between those injuries that remained the same and those that were not treated (Table 1). 
 

Conclusion: Our study found that Grade 4 BCVI mortality was statistically significant, as well as Grade 1 BCVIs and outcomes. However, when we analyzed BCVI progression, we found that there was no statistical significance between progression and early treatment time. It appears that early treatment may not need to be initiated promptly; however, we acknowledge a limitation is that this calculation is underpowered.  Future research will continue to compile BCVI data to enhance our sample size so that a potentially efficacious time period is found to initiate APT.    

 

15.04 The Positive Impact of Methadone Treatment on Trauma Patient Outcomes

Posted on January 18, 2018

S. M. Miller1, S. N. Lueckel1, D. S. Hefferenan1, A. H. Stephen1, M. D. Connolly1, T. Kheirbek1, W. G. Cioffi1, C. A. Adams1, S. F. Monaghan1  1Brown University School Of Medicine,Surgery,Providence, RI, USA

Introduction:  Each day 78 people die from opioid-related overdoses in the United States. With heightened public awareness, the number of people in methadone treatment programs has increased. Methadone treatment was not intended to be a chronic medication and we predict methadone treatment will be associated with adverse outcomes in trauma patients. 

Methods:  The trauma registry of a single level-one trauma center was queried between 2011 and 2016 for patients who were tested for drug use and were grouped based on their methadone use. First demographic and outcome measures were compared among all patients. Then, case-control matching (2 controls for every case) was then performed for between groups, matching for age, gender, Glasgow coma scale (GCS), and injury severity score (ISS). Regression analysis was used to identify variables affecting patient outcomes. Alpha was set to 0.05. 

Results:6848 patients tested for drugs on admission were identified from the trauma registry; 175 were in the methadone group and 6673 were controls. Patients on methadone were younger (43 years vs 52, p<.001) but had similar gender, racial and ethnicity group distributions. There was no significant difference in mechanism of injury, ISS, or GCS on admission. Methadone patients were more likely to have a psychiatric illness (29% vs 17%, p<.001), to smoke (62% vs 31, p<.001) and to use illegal drugs (90% vs 63%, p<.001), while they were less likely to have hypertension (15% vs 32%, p<.001), diabetes (6% vs 11%, p<.05), and congestive heart failure (2% vs 5%, p<.05). The hospital mortality was lower in the methadone group (3% vs 6%, p<.05). Case-control matching yielded a cohort of 509 patients, 170 of whom were on methadone. In the matched sample (with similar age, gender, GCS and ISS), methadone patients were more likely to have a psychiatric illness (30% vs 7%, p<.001), to smoke (62% vs 45%, p<.001) and to use illegal drugs (89% vs 68%, p<.001). Similarly, methadone patients demonstrated lower mortality (2% vs 17%, p<.001) but were observed to have longer lengths of stay in the hospital (9 days vs 7, p<.05). In addition, patients receiving methadone treatment were less likely to be discharged home with no services (51% vs 82%, p<.001). Regression analyses revealed that methadone patients had lower mortality (OR = 21, 95% CI 5.5-79, p<.001) when adjusting for patient and injury characteristics. 

Conclusion: Counter to our hypothesis, patients on methadone were more likely to survive than those not taking methadone. Chronic narcotics may have a salutary effect on injured-induced immune-inflammatory activation. However, patients on methadone were hospitalized for two days longer. This potentially speaks to difficulty in placing patients with services due to the methadone use. 

 

15.07 National Trends in Use and Outcomes of Nonoperative Management versus Splenectomy at Trauma Hospitals

Posted on January 18, 2018

T. Bongiovanni1, A. Stey1, A. Conroy1, C. Wybourn1, R. A. Callcut1  1University Of California – San Francisco,Zuckerberg San Francisco General Hospital, Department Of Surgery, General And Trauma Surgery,San Francisco, CA, USA

Introduction: In 2003, national guidelines were first published recommending potential benefit to non-operative management for hemodynamically stable patients suffering splenic injury.  In 2012, updated guidelines supported extension of non-operative therapy to higher-grade injuries and older patients in the presence of hemodynamic stability.  This study investigates the adoption of non-operative therapy by examining national trends and associated outcomes.

Methods:  The National Trauma Data Bank National Sample Program weighted file was used to conduct an observational and serial cross-sectional cohort study between January 1, 2008 and December 31, 2012, identifying hospitalizations during which a patient greater than 12 years old was diagnosed with a traumatic splenic injury. 

Results: Among the almost 3.5 million unique patients in the database, there were 47,212 splenic injuries documented from the years 2008-2012 (69% men, mean [SD] 37.8 [18.1] years) for traumatic splenic injury, of which 9,961 (21%) underwent operative intervention.

Interestingly, there was as overall decrease in reporting of splenic injuries by 2011 and 2012, though there was no change in OR use (210 per 1000 injuries in 2008 vs 220 per 1000 injuries in 2012).  Over the 5 year study period, there was no improvement in the mean length of stay (11.5 days in 2008, 11.0 days in 2012) or in the number of ICU days (4.81 days in 2008, 5.13 days in 2012). However, the rates of transfusion have increased dramatically from 2008 to 2012 (FFP transfusion 3.0% to 8.2%, p<0.001, platelet transfusion 1.4% to 4.8%, p<0.001, pRBC 9.3% to 18.7%, p<0.001). 

In multivariate regression, controlling for age, injury severity score, GCS upon arrival, transfusions of FFP, platelets, prbcs, race, and tachycardia or hypotension in the emergency department, there was no significant difference in survival among each year of analysis. 

Conclusion: Within 5 years of the initial recommendations for non-operative therapy, the rate of surgical intervention had plateaued and remained stable in the subsequent years 2008-2012.  However, the rate of transfusion has continued to climb suggesting that patient exposure to blood products has increased while attempting splenic preservation.  Further investigation should be done to better elucidate the reasons for increased transfusions requirement, and possible delayed care in these patients. 
 

15.08 Outcomes After TBI in Patients on P2Y12 Inhibitors: Is There a Need for Platelet Transfusion?

Posted on January 18, 2018

F. S. Jehan1, M. Zeeshan1, A. Jain1, T. O’Keeffe1, N. Kulvatunyou1, A. Tang1, L. Gries1, E. Zakaria1, B. Joseph1  1University Of Arizona,Tucson, AZ, USA

Introduction:
A significant portion of patients sustaining traumatic brain injury (TBI) are on antiplatelet medications. The role of the cyclooxygenase inhibitor (Aspirin) is well studied; however, the reversal of P2Y12 inhibitors after intracranial hemorrhage remains unclear. The aim of our study is to evaluate outcomes after traumatic brain injury in patients who are on preinjury P2Y12 inhibitors.

Methods:
We analyzed our prospectively maintained traumatic brain injury database from 2014-2106 and included all patients with intracranial hemorrhage (ICH) who were on P2Y12 inhibitors (Clopidogrel, Prasugrel, Ticagrelor). Regression analysis was performed adjusting for the age, gender, race, admission Glasgow coma scale (GCS) score, transfusion of blood products, severity of injury, type and size of ICH. Outcome measures included progression of ICH, adverse discharge disposition (SNiF), and mortality.

Results:
A total 243 patients with ICH were on preinjury P2Y12 inhibitor met our inclusion criteria and were analyzed. Mean age was 55 + 18 years, 58% were males and 60% were white while the median [IQR] ISS was 14[9-22]. 74% received platelet transfusion after admission. The mean units of platelet transfusion were 1.6 + 2 units. On regression analysis after controlling for confounders, patients who received platelet transfusion had lower rate of progression of ICH on repeat head CT scan (OR: 0.77; 95%CI [0.4-0.8], p=0.01), and decreased rate of neurosurgical intervention (OR: 0.86; 95%CI [0.32-0.9], p=0.03) compared to those who did not. Overall mortality was 11%. In addition, patients on P2Y12 inhibitors who received platelet transfusion had lower odds of discharge to a skilled nursing facility SNiF (OR: 0.71; 95%CI [0.5-0.0.8], p=0.02) and mortality (OR: 0.85; 95%CI [0.44-0.91], p=0.02) as well compared to those patients who did not receive platelet transfusion. 

Conclusion:
Platelet transfusion after traumatic ICH in patients on P2Y12 inhibitors is associated with decreased risk of progression and neurosurgical intervention after traumatic intracranial hemorrhage. In addition, patients with platelet transfusion had lower mortality and were less likely to be discharged to a SNiF. Further randomized studies are required to unify the practice of platelet transfusion after ICH in patients on P2Y12 inhibitors to improve outcomes.
 

15.02 Obese Patients Have a Higher Need for Dialysis After Trauma

Posted on January 18, 2018

A. Grigorian1, N. T. Nguyen1, B. Smith1, B. J. Williams1, S. Schubl1, V. Joe1, D. Elfenbein1, J. Nahmias1  1University Of California – Irvine,Division Of Trauma, Burns & Surgical Critical Care,Orange, CA, USA

Introduction: Obesity is a well-known risk factor for diabetes and hypertension which are the leading causes of end-stage renal disease (ESRD). Obesity is also a risk factor for the development of acute kidney injury (AKI). The effect of obesity on the need for dialysis in trauma has not been elucidated. We hypothesized that patients with a higher body mass index (BMI) will have a higher risk for need of dialysis after trauma.

Methods: This was a retrospective analysis using the National Trauma Data Bank. We included all patients 18 years of age and older. Patients were grouped based on their BMI: normal (18.5-24.99 kg/m2), obese (30-34.99 kg/m2), severely obese (35-39.99 kg/m2) and morbidly obese (> 40 kg/m2). The primary outcome was the need for dialysis. Patients with chronic renal failure were excluded from the analysis since a high proportion of these patients may have been on dialysis prior to their admission. We performed a multivariate linear regression analysis after controlling for significant cofactors.

Results: There were 1,221,990 patients included in the study. The obese group differed from the normal BMI group by age (median, 52.0 vs 38.0), history of diabetes (17.7% vs 6.8%), amount of traumatic brain injury (27.6% vs 30.5%) and lower extremity injury (26.2% vs 23.8%) but no difference in injury severity score (p>0.05). The severely obese group were older (median, 53.0 vs 38.0), had more ESRD (1.5% vs 1.1%) and hypertension (41.6% vs 24.6%). Morbidly obese patients were older (median, 50.0 vs 38.0) and had more lower extremity injuries (30.6% vs 23.8%). There was no difference among groups in regards to ICU stay and ventilatory days (p>0.05). Morbidly obese patients had a higher incidence of rhabdomyolysis (0.1% vs 0.02%), AKI (1.1% vs 0.4%) and mortality (3.1% vs 2.8%). After adjusting for covariates, we found that BMI > 30 kg/m2 (Odds ratio [OR]=1.21, confidence intervals [CI] 1.10-1.33, p<0.001), BMI > 35 kg/m2 (OR=1.50, CI=1.34-1.80, p<0.001) and > 40 kg/m2 (OR=1.84, CI=1.64-2.06, p<0.001) had a stepwise increased need for dialysis after trauma.

Conclusion: Trauma patients with a BMI > 30 kg/m2 are associated with increased risk for dialysis in a large database. This holds true even after controlling for multiple well-known risk factors for acute renal failure in trauma patients. Aggressive screening and treatment of obese trauma patients may help prevent acute renal failure requiring dialysis.

 

15.03 Supratherapeutic INR in the Elderly Trauma Patient: Is It Lethal?

Posted on January 18, 2018

D. Sharma1, L. Sadri1, A. Rogers1, G. Filosa1, Q. Yan1, R. Shadis1, R. Josloff1, T. Vu1  1Abington Memorial Hospital,Abington, PA, USA

Introduction:  Elderly patients (>65 years) often present to the trauma bay on anticoagulants with an elevated INR. Among these patients, traumatic brain injury (TBI) is a common mechanism of injury. We aim to investigate if elderly patients presenting with supratherapeutic INRs have increased mortality compared to those with therapeutic and subtherapeutic INRs after blunt trauma. For patients with TBI, we will also determine if a supratherapeutic INR has higher risk of mortality.

Methods:  A retrospective chart review was performed for patients on the trauma service from 2010 to 2015 at Abington Jefferson Hospital, a level 2 trauma center. Elderly patients on anticoagulation with blunt traumatic injury were divided into three cohorts based on INR: subtherapeutic (< 2.0), therapeutic (2.0-3.5), and supratherapeutic INR (>3.5). The primary outcome of mortality and relative risk (RR) was determined for each group, with the therapeutic group serving as the control. The data was then stratified by mechanism of injury (TBI versus other polytrauma) and mortality and relative risk was reported by INR cohorts.

Results: 

Seven hundred and forty-seven patients were included. In this group, 189 patients were subtherapeutic (25%), 440 were therapeutic (59%), and 118 were supratherapeutic (16%). There was no statistically significant difference in mortality rates between the subtherapeutic group and therapeutic group (RR: 0.58; 95% CI: 0.24-1.40; P = 0.23). However, compared to the therapeutic group, the supratherapeutic group had a statistically significant increase in mortality (RR: 2.18; 95% CI: 1.16-4.07; P= 0.015).  

Of the 220 patients with TBI, the mortality of the subtherapeutic (N = 53), therapeutic (N = 123) and supratherapeutic group (N = 26) was 1.9%, 12.2% and 46.2%, respectively. The RR of death of the subtherapeutic group compared to therapeutic group was 0.15 and not statistically significant (95% CI: 0.02-1.14; P = 0.067). However, compared to the therapeutic group, the supratherapeutic group had a significantly higher risk of mortality (RR: 3.78; 95% CI: 2.02-7.11; P < 0.0001).  

Of 545 patients without TBI, the mortality of the subtherapeutic (N = 136), therapeutic (N = 317) and supratherapeutic groups (N = 92) were 3.7%, 2.8% and 2.2%, respectively. Compared to the therapeutic group, the RR of death was not statistically significant for the subtherapeutic (p=0.64) or supratherapeutic group (P = 0.73).

 

Conclusion: Elderly trauma patients with supratherapeutic INRs have a significantly higher risk of death during hospitalization than those with therapeutic or subtherapeutic INRs. Furthermore, those with traumatic brain injury and supratherapeutic INRs also have a significantly higher risk of death. Therefore, elderly patients on anticoagulants with supratherapeutic INRs warrant purposeful and aggressive monitoring given the increased risk of mortality following blunt traumatic injury.
 

15.01 Rigid Sigmoidoscopy is Diagnostically Superior to CT for Penetrating Rectal Injury

Posted on January 18, 2018

M. J. Chaudhary1, R. Smith2, G. Victorino1  1UCSF-East Bay,Surgery,Oakland, CA, USA 2Emory University,Surgery,Atlanta, GA, USA

Introduction:
Computed tomography (CT) is commonly used to evaluate penetrating pelvic organ injury. Rigid sigmoidoscopy may be used as an adjunct in identifying penetrating rectal injury but its sensitivity compared to CT remains unknown. The purposes of this study were: (1) to determine the clinical utility of pelvic computed tomography (CT) in identifying the need for operative intervention after penetrating pelvic trauma, and (2) to determine if rigid sigmoidoscopy, cystogram or retrograde urethrogram improve the diagnostic yield of penetrating pelvic organ injury.

Methods:
We conducted a retrospective review of the trauma registry at our university-affiliated trauma center between January 1999 and December 2016. All patients with penetrating pelvic trauma, who had a CT of the pelvis prior to any potential operative intervention, were included. Operative reports were used to calculate the sensitivity, specificity, negative predictive value (NPV), and positive predictive values (PPV) for CT and rigid sigmoidoscopy in identifying pelvic organ injury.

Results:
During the study period, 160 patients were treated for penetrating pelvic trauma. Overall mortality after penetrating pelvic injury (including combined body compartment trauma) was 16% (26/160). Bladder injuries comprised the majority of injuries (n=86, 54%), followed by injuries to the ureter, blood vessels, and rectum, respectively. Out of the 160 patients with penetrating pelvic trauma, 37% (59/160) underwent preoperative CT scans and 19% (31/160) underwent rigid sigmoidoscopy. A comparison of the sensitivity, specificity, PPV, and NPV of CT and rigid sigmoidoscopy for penetrating rectal injury is attached.

Rigid sigmoidoscopy identified 71% (5/7) of rectal injuries missed by CT. For the remaining two missed injuries, in one case rigid sigmoidoscopy failed to identify an injury and in the other rigid sigmoidoscopy was not performed. CT had a sensitivity of 66%, specificity of 98%, PPV of 67% and NPV of 95% for bladder injury. Cystogram or retrograde urethrogram (RUG) was performed in 3% (5/160) of patients. Cystogram and RUG used in isolation or combination had 100% sensitivity, specificity, NPV and PPV for bladder injury. However, these adjuncts did not identify any injuries missed on CT.

Conclusion:
CT of the pelvis in clinically suitable patients with penetrating pelvic trauma has a low sensitivity and NPV for diagnosing operatively significant rectal or bladder injury. Rigid sigmoidoscopy increases the diagnostic yield for penetrating rectal injury requiring operative intervention. When clinical concern for rectal injury exists following penetrating trauma in the absence of CT findings, rigid sigmoidoscopy is warranted. 

13.16 Stop the Radiation: Limiting Chest CT scans in the Pediatric Trauma Patient

Posted on January 18, 2018

S. Azari2, T. Hoover1, M. Browne1,2  1Lehigh Valley Health Network,Pediatric Surgical Specialties,Allentown, PA, USA 2University Of South Florida College Of Medicine,Tampa, FL, USA

Introduction: National attention has been drawn to decreasing pediatric radiation exposure with a push to “image gently”, however there are currently no national pediatric CT guidelines.  CT scans aid in the evaluation and treatment of pediatric patients; unfortunately, they expose the child to a considerable amount of radiation.  This creates a challenge for physicians, especially those caring for the acute pediatric trauma patient.  Due to the flexibility of the pediatric chest wall, the incidence of thoracic injury with blunt trauma is low.  We hypothesize that chest CT scans after a normal chest xray will not add clinically relevant information to justify the risk of the radiation.  

Methods:  A retrospective chart review of all level 1 trauma patients < 15 years of age who were evaluated at our pediatric trauma center between January 2013 and June 2016 was performed.  Using our database and chart review, patients who had a chest CT scan during their initial evaluation were reviewed for demographics, mechanism of injury, radiological results, and change in management based on those results.  Patients were excluded if their radiological evaluations were performed at an outside facility; no radiological chest evaluation was preformed; or if their mechanism of injury was drowning.

Results: There were 257 patients who met our inclusion criteria.  Eighty-two percent (211/257) had a chest xray.  Though 44% (114/257) had a chest CT scan; only 60% (68/114) of those patients had a chest xray prior to CT.  Of those patients, 74% (50/68) had a normal x-ray. Thirty percent (15/50) of the chest CTs done after a normal x-ray had an abnormal result.  Only 1 patient (2%) had a result which changed clinical management.

Conclusion: Though chest CT scans increase abnormal diagnoses, the chance of their results changing clinical management is very low.  Chest CTs should be consider unnecessary when the chest xray is normal.  

 

13.18 A Ten Year Review of Firework-Related Injuries Treated at a Regional Pediatric Burn Center

Posted on January 18, 2018

P. H. Chang2,4, D. Toplauffe1, S. Wang1, S. Romo1, K. Hannigan1, R. Sheridan1,3  1Shriners Hospitals For Children-Boston,Boston, MA, USA 2Shriners Hospitals For Children-Cincinnati,Cincinnati, OH, USA 3Massachusetts General Hospital,Department Of Surgery,Boston, MA, USA 4University Of Cincinnati,Division Of Plastics/Burn Surgery,Cincinnati, OH, USA

Introduction:
In 2015, 11,900 firework-related injuries were reported in the United States. Laws regulating the use of consumer fireworks vary from state to state in our region. However, it is common practice for consumers to cross state borders to purchase fireworks illegal in their state. The objective of this study is to describe the population of patients treated for injuries involving fireworks at a single regional pediatric burn center.

Methods:
A retrospective chart review was conducted to analyze all patients aged 0-18 years admitted to our regional pediatric burn hospital with a firework-related injury between 2006 and 2015. Data collected included demographics, total body surface area (TBSA) involved, location of burn, state in which the injury occurred, and whether sparklers, firecrackers, or aerial fireworks were involved. 

Results:
Of the 61 patients who met the inclusion criteria for review, four times as many patients were males than females. The mean age of the study sample was 10.53 ± 5.42 years (range: 0.52-17.9 years) and the mean TBSA was 3% ± 7%. More than half of these patients were from MA (66%), while the other injuries occurred in: NH (21%), VT (7%), and less than 4% in NY, CT, ME, and VA. Seventy-one percent of these patients had to be admitted as inpatients for treatment. At least 40% of injuries were to critical areas (i.e. face, hands, feet, genitalia). Aerial fireworks were involved in 46% of these injuries, while sparklers and firecrackers were each involved in 28%.  

Conclusion:
Fireworks pose a serious danger to children in every state, regardless of mandated state legislation pertaining to fireworks sales. Sparklers, which are legal in six of the seven states included in our review, were responsible for more than a quarter of the injuries treated. Moreover, preliminary data suggests that laws regarding firework sales are not being properly implemented. Fireworks are illegal in the state of MA; however, 40 of the patients referred to our facility due to firework-related injuries were injured in MA. Additionally, although firecrackers are illegal in all of the states in which these injuries occurred, they were involved in more than a quarter of the injuries reviewed.   Over the past ten years, our pediatric burn center has treated numerous children injured due to fireworks. Our research demonstrates a need for clinicians and lawmakers to work together to help enact legislation limiting the sales and use of fireworks.
 

13.13 Examination of Postoperative Length of Stay Following Common Procedures in ACS-NSQIP Pediatric

Posted on January 18, 2018

D. Papandria1, Y. V. Sebastião1, K. J. Deans1, K. A. Diefenbach1, P. C. Minneci1  1Nationwide Children’s Hospital,Pediatric Surgery,Columbus, OH, USA

Introduction:

Though frequently reported in comparative studies, there are few reports describing typical postoperative length of stay (LOS) associated with commonly performed operations in the pediatric population. The objective of this study was to identify ranges of postoperative LOS for common pediatric procedures using a large multi-institutional database.

 

Methods:

A retrospective analysis of the ACS-NSQIP Pediatric Public Use File (2013-2015) was performed. General surgical procedures were grouped using Current Procedural Terminology codes (CPTs). The most frequently performed procedures were identified and analyzed. These included: laparoscopic appendectomy (LA), laparoscopic cholecystectomy (LC), laparoscopic gastrostomy (LG), laparoscopic esophagogastric fundoplication (LF), thoracoscopic repair of pectus excavatum (TPE), appendectomy (OA), enterostomy closure (OEC), gastrostomy closure (OGC), and bowel resection (OBR). Patients < 6 months or > 18 years of age and those receiving unrelated major concurrent surgical procedures were excluded, as were day-of-surgery discharges and inpatient deaths. Postoperative LOS was examined for each procedure, including multivariable analysis of patient preoperative risk factors for postoperative LOS > 75th percentile.

 

Results:

A total of 29, 557 cases were identified (median age: 7 years; 57% male; 73% white), and included procedure subgroups ranging from 505 (OBR) to 19,260 (LA) cases. Procedure-specific median postoperative LOS (75th percentile; 90th perecentile) were: LA 1d (2d; 5d); LC 1d (1d; 2d); LG 2d (2d, 4d); LF 3d (4d, 6d);  TPE 4d (5d, 6d);  OA 3d ( 6d, 9d);  OEC 4d (6d, 10d);  OGC 1d (1d, 2d); and OBR 6d (10d, 20d)(Fig. 1). Preoperative risk factors for high postoperative LOS varied by procedure and included patient demographics (age, race), admission factors (inpatient classification, admission from Emergency Dept.), case characteristics (emergent designation, ASA class III / IV), and comorbidities (sepsis, developmental delay, neurologic disease). No single risk factor reached statistical significance for more than six of the procedures.

 

Conclusion:

The range of postoperative LOS for commonly performed procedures varies considerably across procedures. Risk factors for high postoperative LOS also varied by procedure. These results may be a useful reference for benchmarking and resource utilization analyses at the institutional and health systems levels.

13.10 Approaches and Safety Profile of Surgical Treatment of Velopharyngeal Insufficiency Using NSQIP

Posted on January 18, 2018

A. D. Chen1, B. N. Tran1, Q. Z. Ruan1, B. T. Lee1, O. Ganor2  1Beth Israel Deaconess Medical Center,Plastic And Reconstructive Surgery,Boston, MA, USA 2Boston’s Children Hospital,Plastic And Reconstructive Surgery,Boston, MA, USA

Introduction: Velopharyngeal insufficiency (VPI) often manifests after cleft repair or adenoidectomy as a result of an occult palatal problem, which can result in hypernasal speech and nasal air emission. This study aims to study the outcomes of different techniques for VPI correction using the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP).  

Methods: VPI cases from 2012-2015 were identified.  Patients were subdivided in 4 unique cohorts: (1) palatoplasty, (2) pharyngeal flap and sphincter pharyngoplasty, (3) lengthening, and (4) others including tissue excision and rearrangement, dermal grafts or fillers. Group characteristics and postoperative outcomes were compared using chi-square test for categorical variables and one-way ANOVA for continuous variables.

Results:  There were 591 VPI cases identified, 83 in group1, 359 in group 2, 40 in group 3, and 109 in group 4. The average age of repair was 7.9 with palatoplasty and pharyngeal flap done at a later time. More Asian patients received lengthening compared to other techniques. The longest operating time (108 minutes) was noted in lengthening group while the longest length of stay (2 days) was seen in the palatoplasty group.  Pediatric plastics performed the majority of the palatoplasty and lengthening cases whereas pediatric ENT performed most of the pharyngeal flap and local tissue rearrangement. Overall complication rate was 2%, with palatoplasty group had the lowest rate. Subgroup analysis comparing flap and sphincter techniques showed more complications in the pharyngeal flap group, however, these trends were not statistically significant.

Conclusion: Repairing a VPI can be done safely and effectively using different surgical approaches depending on the extent of the defect. A small gap causing a mild VPI will probably require a secondary palatoplasty, local tissue rearrangement or lengthening while a wide gap mandates bringing extra tissue to narrow it. Timely correction is crucial to facilitate proper phonation in children of developmental age. 

13.05 Blunt Renal Injury in Children: Do National Trends in Management Follow Recent Literature?

Posted on January 18, 2018

R. Sola1, T. A. Oyetunji1, K. D. Graziano2, S. D. St. Peter1  1Children’s Mercy Hospital- University Of Missouri Kansas City,Surgery,Kansas City, MO, USA 2University Of Arizona,College Of Medicine/ Department Of Surgery,Phoenix, AZ, USA

Introduction:  Published prospective observational data suggest children with blunt renal injury can be managed without bedrest, the use of catheters or antibiotics, and follow up imaging can be reserved to those with urinary extravasation. We wanted to investigate the current practice patterns of major children’s’ hospital to identify variation and areas for improvement with the use of evidence based protocols. 

Methods:  Data from 2006 to 2015 were requested from the Pediatric Health Information System. Patients were included based on the International Classification of Disease Ninth Revision (ICD-9) coding for blunt renal injury. Children were excluded if they had concomitant major thoracic or abdominal injury, diagnosis of renal injury without computerized tomography (CT) imaging confirmation, length of stay greater than seven days, underwent laparotomy and those that were intubated. Demographics, need for further imaging, and hospital outcomes were analyzed.

Results: During the study period, 1487 children were found to have a blunt renal injury. A total of 638 children were identified after excluding those that did not meet our inclusion criteria.  Median age was 12 years old (8,14). There were 474 (74%) males and 386 (61%) were white.  Median length of stay was 3 days (2,4).  Foley catheters were placed in 93 (15%) children and 157 (25%) were given antibiotics during their hospital course.  Two or more CT scans were performed in 376 (59%) children during their hospitalization.  

Conclusion: Children with blunt renal injury appear to be utilizing excess healthcare resources compared to published recommendations.  Further studies implementing an evidence based protocol would allow for the reduction of Foley catheters, antibiotics and CT scans. 

 

13.07 The Effect of Hospital Volume on Patient Outcomes for Pyloric Stenosis

Posted on January 18, 2018

C. Tom1, C. Niino2, A. D. Lee2, E. Saab2, S. Friedlander3, S. L. Lee1,2,3  1Harbor-University Of California Los Angeles Medical Center,Department Of Surgery,Torrance, CA, USA 2University Of California Los Angeles,Department Of Surgery,Los Angeles, CA, USA 3Los Angeles Biomedical Research Institute,Torrance, CA, USA

Introduction: For many surgical operations, there is a well-established relationship between surgical volume and outcome.  In the field of pediatric surgery, this has been shown to be true for conditions requiring complex operations, however, for common conditions this relationship is less clear. This study investigated the relationship between hospital volume and surgical outcomes for infants affected by hypertrophic pyloric stenosis. 

Methods:  Kid’s Inpatient Database (KID) was used to identify patients with congenital hypertrophic pyloric stenosis who underwent pyloromyotomy for years 2003, 2006, 2009, and 2012. Surgical outcomes were measured by length of stay (LOS), complication rates, mortality, and cost. Hospitals were stratified based on case volume. Low-volume hospitals had the lowest quartile of patients treated per year, medium-volume hospitals had the middle two quartiles, and high-volume hospitals had the highest quartile of patients. 

Results: A total of 2,234 hospitals treated 51,792 patients with pyloric stenosis. The majority of hospitals were low-volume (n=1,834), while only 51 were high-volume. The overall mortality rate was 0.1% and the median length of stay was 2 days. Females were associated with higher complication rates. Results of multivariate analysis are summarized in table. High-volume hospitals were associated with lower complication rate and increased cost compared to medium- or low-volume hospitals. There were no differences in mortality or LOS. 

Conclusion: Using national data, we found that patients with hypertrophic pyloric stenosis treated at high-volume hospitals have improved outcomes despite higher costs. This indicates a benefit to receiving treatment for pyloric stenosis at a high-volume hospital. 

 

13.03 The Role of LFTs in the Evaluation of Blunt Trauma in Pediatric Trauma Patients: Are They Necessary?

Posted on January 18, 2018

S. F. Rosati1, B. A. Borg1, P. Kato1, A. Husseini1, L. Donoghue1, C. Shanti1  1Children’s Hospital Of Michigan,Pediatric Surgery,Detroit, MI, USA

Introduction:  Injury is the leading cause of morbidity and mortality in children over one year; over 90% are the result of blunt trauma. Diagnostic aids to detect intra-abdominal injuries (IAIs) of the liver and spleen include abdomen/pelvis CT scans (AP CT) and liver function tests (LFTs). Historically, elevated LFTs have been used as a marker for when to obtain AP CTs. Our objectives were to evaluate the number of clinically significant injuries (defined as Grade IV or V) found using AP CT, and if there was a correlation to elevated LFTs.

Methods:  This is a retrospective review of pediatric patients (<18 years) evaluated at our Pediatric Level I trauma center from 1/1/15-12/31/16, who suffered blunt trauma. Variables included age, gender, injury severity score (ISS), LFTs, AP CTs and IAI with grades. 

Results: 1138 children were evaluated: 63% male, 37% female, with ages from 6 wks – 18 yrs (mean 5.25 yrs) and ISS from 0-45 (mean 5.4).  38% of patients (pts) had LFTs, 5% had an AP CT, 37% had IAI (1.8% overall). In the 62% of pts without LFTs, there were 16 AP CTs, 4 IAI, 0 significant; in pts with LFTs 0-100 (33%), there were 27 AP CTs, 7 IAI, 1 significant; in pts with LFTs 101-200 (2.8%), there were 7 AP CTs, 1 IAI, not significant; in pts with LFTs 201-300 (0.8%), there were 6 AP CTs, 5 IAI, 0 significant; in pts with LFTs 301-400 (0.4%), there were 2 AP CTs, 1 IAI, significant; in pts with LFTs 401-500 (0.01%), there was 1 AP CT, 1 IAI, not significant; in pts with LFTs > 500 (0.7%), there were 8 AP CTs, 2 IAI, 0 significant. 

Conclusion: In this limited review, a fraction of pts required an AP CT. While 33% of them were found to have IAI on CT, only 1% was clinically significant. Elevated LFTs do not correlate with the severity of IAI. LFTs alone may be a poor screening lab to determine need for an AP CT. We propose developing a different screening approach to our pts besides LFTs to determine need for AP CT.

 

13.02 The Management of Blunt Traumatic Retroperitoneal Hematomas in Children.

Posted on January 18, 2018

P. Dasari1, G. P. Wools2, L. S. Burkhalter2, F. G. Qureshi1,2  1University Of Texas Southwestern Medical Center,Pediatric Surgery,Dallas, TX, USA 2Children’s Medical Center,Pediatric Surgery,Dallas, Tx, USA

Introduction:
Management of blunt traumatic Retroperitoneal Hematomas (RPH) in adults is dependent on anatomical classification.  Zone 1 is central, contains the aorta, inferior vena cava, renal vessel origins, partial duodenum/pancreas and requires mandatory exploration. Zone 2 includes the paranephric areas, renal vessels, kidneys, ureters, adrenals/colon and is explored for expanding hematoma.  Zone 3 includes iliac vessels, distal ureters, sigmoid/ rectum and may need surgical or radiologic interventional. This strategy has been used in children but has not been studied.  The aim of this study is to evaluate the management and outcome of children with retroperitoneal hematomas after blunt trauma. 

Methods:
With IRB approval, 10 year (2007-2016) retrospective review of all children with RPH from blunt trauma was performed.  RPH zone was determined by imaging and/or operative findings. Mechanism of injury, laparotomy, RPH explorations, and outcomes were collected.  Descriptive statistics provided mean, standard deviation, median and range. Comparative statistics identified univariate correlations using Fischer’s exact test. 

Results:
We identified 32 patients (84% male, mean age 10±4) with 43 RPH injuries, 14 zone 1, 25 zone 2 and 4 zone 3 injuries (table 1). Mechanisms included motor vehicle collision (75%), struck by object (19%), and pedestrian struck (6%). Nine (28%) patients were unstable on arrival and two expired in the emergency room. Laparotomy was performed in 17 patients, 10 immediately for instability, shock or peritonitis. 13 (30%) RPH zone injuries were explored; two zone 1, nine zone 2 and two zone 3. Four zone explorations required intervention: none in zone 1, four zone 2 (three nephrectomies, one packing) and none in zone 3.  RPH exploration had no post-operative surgical complications. Overall mortality was five (16%): two zone 1 before laparotomy (traumatic brain injury, TBI); two zone 1 after laparotomy (TBI and uncontrolled liver hemorrhage); and one zone 2 after laparotomy from chest injury.  Mortality was higher in unstable patients (p=0.0006). No mortality occurred from RPH exsanguination and RPH exploration did not impact mortality.

Conclusion:
Only a third of pediatric RPH injuries were explored which identified injuries requiring intervention in zone 2 but not zone 1 or 3.  RPH injury in children may require a different treatment paradigm compared to adults. Zone 1 injuries in an otherwise stable pediatric patient without peritonitis may benefit from non-operative management. Further larger scale studies will be required to understand the role of surgical intervention in RPH injury in children.
 

13.01 Underutilization of the Organ Injury Scaling System in a Pediatric Trauma Center

Posted on January 18, 2018

K. B. Savoie1, N. Jain2, R. F. Williams1  1University Of Tennessee Health Science Center,Department Of Surgery,Memphis, TN, USA 2University Of Tennessee Health Science Center,College Of Medicine,Memphis, TN, USA

Introduction:
The value of the American Association for the Surgery of Trauma (AAST) Organ Injury Scaling (OIS) system has been beneficial in managing solid organ injuries in adults. However, OIS may not correlate with pediatric solid organ injuries and thus may be inconsistently used at pediatric institutions. We hypothesized that radiologists inconsistently assign OIS grades for pediatric blunt solid organ injuries.

Methods:
All patients with blunt liver, spleen, and kidney organ injuries from a January 2009 to December 2014 at an urban tertiary pediatric hospital were identified from an institutional trauma database. Demographic information, imaging, radiologic grade of injury, and surgical grade of injury were collected.  Spearman’s correlation and weighted Kappa was used to evaluate radiologist and surgeon’s grading agreement of the injuries. 

Results:
A total of 352 patients were identified; OIS grading was assigned to 73% of patients; 37% had grading by a radiologist and 66% by a surgeon. Liver: 128/179 injuries were graded. 56 patients had grading by both radiologists and surgeons with a Spearman correlation of 0.70 and a weighted kappa of 0.59 (figure). OIS was associated with overall need for intervention (p <0.01) and specifically for need for transfusion (p <0.01) and operative intervention (p = 0.02); it was not associated with need for angiography. Spleen: 97/126 injuries were graded. 41 patients had grading by both radiologist and surgeons with a Spearman correlation of 0.93 and a weighted kappa of 0.86. Kidney: 30/47 injuries were graded. 9 patients had grading by both radiologist and surgeons with a Spearman correlation of 0.82 and a weighted kappa of 0.67. For spleen and renal injuries there was no correlation between OIS grade and need for overall intervention (spleen p=0.12, renal p=0.23) or specific types of intervention. There was no correlation between grade and complications for any type of injury.

Conclusion:
Pediatric surgeons utilized OIS more frequently than pediatric radiologists; there was higher correlation for spleen and renal injuries. Although OIS was associated with need for intervention in liver injuries, it was not associated with interventions for spleen and renal or for complications for any type of injury. Efforts to increase utilization or the development of a pediatric specific grading system may help standardize care for pediatric trauma patients.
 

12.20 Hemorrhage After On-ECMO Repair of CDH is Equivalent for Muscle Flap and Prosthetic Patch

Posted on January 18, 2018

H. Nolan1, E. Aydin1, J. Frischer1, J. L. Peiro1, B. Rymeski1, F. Lim1  1Cincinnati Children’s Hospital Medical Center,Cincinnati, OH, USA

Introduction: The defect in severe congenital diaphragmatic hernia (CDH) often requires a prosthetic patch (patch) or muscle flap (flap) repair. The patch is easy to use but is synthetic, while the flap’s autologous tissue dissection has potential for increased bleeding. Hemorrhage can be further exaggerated when maintained on therapeutic anticoagulation for extracorporeal membrane oxygenation (ECMO), especially if clinical status demands on-ECMO repair. The purpose of this study was to assess bleeding complications for on-ECMO patch compared to flap repair of CDH.

Methods: We retrospectively reviewed on-ECMO CDH repairs from 2010-2016 at a single academic children’s hospital (IRB2017-2322). Exclusions included incomplete records or concomitant procedures that could result in additional blood loss. Patients were grouped by repair type and bleeding complications were captured with intra-operative blood loss, 48-hour re-operation rates for bleeding, and 48-hour post-operative blood product use.

Results: Twenty-nine patients met criteria for analysis. Thirteen (44.8%) had patch repair and 16 (55.2%) had flap repair. Eight (62.5%) of the patch and 13 (81.2%) of the flap group were left-sided defects (p=0.223). All had Type C or D defects with comparable distribution (Type C: patch 56%, flap 54%, p=0.596). There was no difference in mean gestational age at delivery (patch 37.5±0.9 weeks, flap 37.2±1.3 weeks, p=0.390) or mean age at time of repair (patch 7.46±6.6 days, flap 6.00±4.3 days, p=0.476). Both had similar total ECMO duration (patch 361.4±167.1 hours, flap 277.1±149.4 hours, p=0.170) and time from repair to decannulation (patch 7.77±6.0 days, flap 7.00±6.0 days, p=0.734). Only one patient in each group was decannulated within 48 hours of repair for bleeding. Seven patch patients (53.8%) and 9 flap patients (56.2%) survived to discharge (p=0.596).

 

Estimated intra-operative blood loss was equivalent (patch 35.3±53.9 mL, flap 24.2±18.4 mL, p=0.443). One patient (7.6%) in the patch group and two patients (12.5%) in the flap group required re-operation for bleeding (p=0.580). Transfusion requirements in the re-operative group were no different for the patch compared to the flap repair (282.0 mL/kg vs 208.5±21.9 mL/kg, p=0.054). Transfusion requirements for those who did not require a reoperation were also similar (patch 120.7±111.7 mL/kg, flap 118.4±89.9 mL/kg, p=0.561).

Conclusions: Our study demonstrates the feasibility of CDH repair while on ECMO for both flap and patch techniques. Bleeding risks were no different between the two groups with regard to estimated blood loss, reoperation rates, and post-operative transfusions.

12.18 Same Day Discharge vs Observation For Uncomplicated Laparoscopic Appendectomy: A Prospective Cohort

Posted on January 18, 2018

K. Gee1, S. Ngo1, A. Beres1  1University Of Texas Southwestern Medical Center,Department Of Surgery, Division Of Pediatric Surgery,Dallas, TX, USA

Introduction:  Appendicitis remains the most common gastrointestinal pediatric surgical emergency. With the introduction of laparoscopic techniques in the 1990s, recovery, pain and hospital stay after laparoscopic procedures have been significantly reduced. Through 2015 our institution routinely admitted uncomplicated appendicitis patients for overnight observation after laparoscopic appendectomy. Given the increasing body of evidence suggesting the safety and feasibility of same day discharge after uncomplicated appendectomies we elected to perform a prospective study evaluating the complication rates of same day discharge appendectomies compared to overnight observation.

Methods:  After IRB approval, all pediatric patients who underwent laparoscopic appendectomies for uncomplicated appendicitis in 2016 were observed. Decision for same day discharge was based on surgeon preference and parental agreement. Data regarding demographics, admission and discharge times and outcomes of complications, readmissions, return to the ED and non-scheduled clinic visits were collected and analyzing using chi-square and multivariate regression.

Results: A total of 1321 appendectomies were performed during the study period; 849 were uncomplicated, of which 382 were discharged same day and 467 were admitted for overnight observation. Univariate analysis revealed no statistical difference between readmission rates for same day vs observation (2 vs 6 patients, p=0.21) or in emergency department visits (22 vs 27 patients, p=0.98). There was also no difference between the number of surgical site infections or the number of patients who required an extra clinic visit. On multivariate logistic regression, controlling for age, gender and discharge from PACU vs floor, there was a significant difference only for calls related to pain favoring those who went home same day (OR=0.88, p value 0.008).

Conclusion: Same day discharge for laparoscopic non-complicated appendectomy is a safe and feasible alternative to post-operative admission and observation. In our prospective study of 849 patients there were no differences in outcomes between the two groups. This has the potential to yield significant healthcare cost savings.

12.10 ECMO Duration Predicts Survival in Congenital Diaphragmatic Hernia

Posted on January 18, 2018

S. M. Deeney1, D. D. Bensard1, T. M. Crombleholme1  1Children’s Hospital Colorado,Department Of Pediatric Surgery,Aurora, CO, USA

Introduction:
Physicians caring for patients supported on extracorporeal membrane oxygenation (ECMO) with congenital diaphragmatic hernia (CDH) may wish to know the chance of survival based on time on ECMO. There are limited data reporting the predicted survival outcomes of these patients as a function of ECMO support duration. We aim to describe survival rates in patients with CDH repaired on ECMO in relation to their duration of ECMO support.

Methods:
Retrospective patient data of all patients who underwent repair of CDH while on ECMO from 2008 through 2015 was collected at our institution. Statistical analysis was by logistic regression analysis and chi square, p<0.05.

Results:
There were 22 patients with 10 surviving to discharge. The total number of time spent on ECMO predicted survival to discharge in CDH patients (p=0.006). For every additional day on ECMO, the odds of survival changed by a factor of 0.86 (0.75-0.99). The odds of survival was 50% after 12 days on ECMO, and 25% after 21 days. There was no survival in our patients after 26 days on ECMO.

Conclusion:
Duration of ECMO support predicts survival in patients with CDH repaired while on ECMO, with low survival after 3 weeks and no survivors after 4 weeks. This information may be useful in guiding goals of care conversations.
 

« Previous Page
Next Page »