53.17 Association of Survival with Admission to Trauma Centers with Extracorporeal Membrane Oxygenation

Posted on January 18, 2018

K. Carlson1, N. K. Dhillon1, G. Liao1, C. Colovos1, R. Chung1, D. R. Margulies1, E. J. Ley1, G. Barmparas1  1Cedars-Sinai Medical Center,Los Angeles, CA, USA

Introduction:  Traditional metrics to evaluate quality of care among trauma centers are inconsistent. Evaluating access to additional resources might offer more useful metrics. We aimed to characterize outcomes of trauma patients undergoing extracorporeal membrane oxygenation (ECMO) and to assess whether trauma centers with ECMO capabilities have improved overall survival.

Methods:  Patients receiving ECMO therapy at Level I and II centers from 2007 to 2011 were selected from the National Trauma Data Bank. A logistic regression was utilized to calculate the adjusted odds ratio (AOR) for mortality between patients admitted to centers with ECMO capabilities to those admitted to centers with no such capabilities. 

Results: A total of 97 patients admitted to 37 centers were included. The median age was 25 years and 76% were male. Injury severity score was high (median 25). Initiation of ECMO ranged from day 0 to 90 from admission. ARDS was present in 52%. Overall mortality was 43%. The 37 centers with ECMO capabilities were mostly Level I (94%), and academic (90%). Compared to patients admitted to Level I and II centers with no ECMO capabilities, those admitted to centers with ECMO capabilities had a significantly lower overall mortality (AOR: 0.86, p<0.01).

Conclusion: Although the number of trauma patients who require ECMO is small, admission to trauma centers with access to ECMO is associated with improved survival. This survival advantage may reflet the availability of advanced therapies for critically ill trauma patients. Access to ECMO could be considered one of the quality metrics for trauma centers. 
 

53.18 Comparing Complication Rates of Chest Tube Placement in Trauma Patients

Posted on January 18, 2018

C. W. Jones1, R. L. Griffin2, G. McGwin2, J. Jansen1, J. D. Kerby1, P. L. Bosarge1  1University Of Alabama at Birmingham,Department Of Surgery, Division Of Acute Care Surgery,Birmingham, Alabama, USA 2University Of Alabama at Birmingham,Department Of Epidemiology,Birmingham, Alabama, USA

INTRODUCTION Thoracic injury accounts for 25% of all trauma deaths. While tube thoracostomy can be lifesaving it is also a source of preventable morbidity. Malpositioned chest tubes, the most commonly reported complication, lead to retained hemothorax or pneumothorax, and can result in the need for subsequent procedures including placement of a second chest tube or more invasive surgical procedures to access the pleural space. The goal of this study was to compare complications of chest tube placement among trauma patients whose chest tubes were placed at outside institutions prior to patient transfer with those placed at the trauma center.

METHODS Trauma patients directly admitted to an academic, Level-I trauma center between 2004 and 2013 who underwent chest tube placement prior to arrival at the trauma center were matched to patients admitted to the same trauma center who had a chest tube first placed at that center. Patients were matched on year of admission, age±5 years, injury mechanism, and Injury Severity Score ± 5. Medical record review was conducted to collect data on complications including empyema, residual hemothorax, residual pneumothorax, malposition, placement of a second chest tube, and use of VATS. The trauma registry was used to collect information on clinical outcomes (i.e., thoracotomy, pneumonia, death after 24 hours, hospital length of stay, days in the ICU, days on ventilator support). A paired t-test compared continuous outcomes, and a conditional logistic regression compared the likelihood of complications and death between groups.

RESULTS From 2004-2013, a total of 4216 patients had a chest tube first placed in trauma center, and 364 patients had a chest tube placed outside of the trauma center. At the time of this abstract, chart abstraction was completed on 151 of these 364 patients, all of whom matched to a patient with a chest tube placed at the trauma center. Patients with a chest tube placed outside of the trauma center had shorter hospital length of stay (17.3 vs 22.1 days, p=0.0339) and days on ventilator support (13.1 vs 17.6, p=0.0406). These patients, though, had increased likelihood of malposition (OR 5.26, 95% CI 2.86-10.00), residual hemothorax (OR 5.88, 95% CI 3.03-11.11), residual pneumothorax (OR 6.67, 95% CI 3.57-12.50), as well as having a second chest tube placed (OR 3.45, 95% CI 2.08-5.56). However, patients with a chest tube placed outside of the trauma center were 67% less likely to get pneumonia (OR 0.33, 95% CI 0.13-0.84). There was no difference in empyema, need for VATS, thoracotomy, or death.

CONCLUSIONS These early data suggest an increased complication rate for patients transferred from another facility; however, the reason for this increase is not yet definitive. Future research is needed to examine the reason for the observed increase, whether it be related to training of the personnel at non-trauma institutions or characteristics related to the patient or their injury.

 

53.19 Age Should Not Preclude Elderly Trauma Patients from Undergoing a Percutaneous Tracheostomy

Posted on January 18, 2018

K. Carlson1, N. K. Dhillon1, P. Ng1, N. T. Linaval1, G. M. Thomsen1, D. R. Margulies1, E. J. Ley1, G. Barmparas1  1Cedars-Sinai Medical Center,Los Angeles, CA, USA

Introduction:  Performance of percutaneous tracheostomy (PT) in intensive care unit (ICU) trauma patients with prolonged ventilatory support is associated with decreased ventilation days. Nonetheless, clinicians might hesitate to perform this procedure on elderly patients due to presumed higher overall mortality risk and to avoid unnecessary interventions. The purpose of this study was to investigate whether elderly patients are less likely to undergo PT and whether this has an impact on mortality.

Methods:  Patients 18 years or older with at least 48 hours on the ventilator were selected from the National Trauma Databank research datasets 2007 to 2015. Transferred patients and patients who underwent PT placement within 48 hours or after 30 days from admission were excluded. Patients were divided based on age:  ≤80 years (YOUNG) and > 80 years old (OLD) and were compared using standard statistical tools. The primary outcome was mortality. To account for the timing of mortality, a Cox regression model with a time dependent variable was utilized to calculate the adjusted hazard ratio (AHR) for mortality between those receiving a PT placement and those who did not.

Results: Over the 9-year study period 214,045 patients met inclusion criteria. Of those, 13,954 (6.5%) were older than 80 years. OLD patients were significantly less likely to undergo a PT (16.1% vs. 23.8%, p<0.01). Among those undergoing a PT, OLD had a longer duration of ventilatory support prior to the procedure (median: 10 vs. 9 days, p<0.01), however, there was no significant difference in the post-PT ventilation days (median: 7 vs. 7 days, p=0.82). The overall mortality was significantly higher in OLD patients (41.8% vs. 15.6%, p<0.01). In the YOUNG cohort, those undergoing a PT had a significantly lower overall mortality (6.6% vs. 18.4%, p<0.01) compared to those with no PT. Similarly, in the OLD cohort, PT was associated with significantly lower mortality (16.3% vs. 46.7%, p<0.01). In a Cox regression model adjusting for gender, injury severity score (ISS), admission Glasgow Coma Scale (GCS) score, and admission systolic blood pressure, the AHR for mortality for younger patients receiving tracheostomy was 0.43 (adjusted p<0.01) compared to those not receiving a tracheostomy. The AHR for elderly patients was lower, at 0.38 (adjusted p<0.01).

Conclusion: In ventilated trauma patients, percutaneous tracheostomy is associated with a higher overall survival and this survival benefit is more profound in elderly patients. Delaying or even avoiding this procedure in elderly patients might not be justified. 

53.15 Beyond Mortality: Low Education Associated with Poor Long-term Physical & Mental Health After Trauma

Posted on January 18, 2018

S. Shah1, S. S. Al Rafai1, J. P. Herrera-Escobar1, M. Jarman1, A. Geada2, J. M. Lee3, K. Brasel4, H. M. Kaafarani3, G. Kasotakis2, G. Velmahos3, A. Salim1, A. H. Haider1, D. Nehra1  1Brigham And Women’s Hospital,Boston, MA, USA 2Boston University,Boston, MA, USA 3Massachusetts General Hospital,Boston, MA, USA 4Oregon Health And Science University,Portland, OR, USA

Introduction:  It has been hypothesized that educational level is associated with long-term outcomes after trauma. Patients with a lower level of education may be at risk for less involved follow-up care and may feel less empowered to seek all possible avenues for functional recovery. Our objective was to determine the association between education and both physical and mental composites of quality of life 6 or 12 months after injury.  

Methods:  Trauma patients with an Injury Severity Score (ISS) ≥9 were identified using the institutional trauma registries of three Level I trauma centers and contacted 6- or 12-months post-injury to participate in a telephone interview. Patients were asked to complete the Short-Form 12 (SF-12) questionnaire which is a validated Health-related Quality of Life tool used to assess both mental and physical health. SF-12 scores are represented as t-scores with a population mean of 50 and a standard deviation of 10, in which 0 represents the lowest level of health and 100 the highest. Multivariate logistic regression models adjusted for age, sex, insurance, number of comorbidities, ICU admission, placement on ventilator, injury cause, ISS, alcoholism, smoking status and discharge were used to determine the effect of a low (high school or lower, LE) as compared to high (higher than high school, HE) education on long-term physical and mental health.

Results: A total of 555 patients were included in this study of whom 254 (46%) had a LE. Mean age of patients with a LE was 50 (SD 20.9) and 58 (SD 20) for patients with a HE. Mean ISS was 14 for both groups. Upon adjusted analyses, mean SF-12 physical composite score was lower for patients with a LE [38.9 (SD: 11.6)] as compared to patients with a HE [44.3 (SD: 10.9)] (p value: 0.001). Similarly, mean SF-12 mental composite score was lower for patients with a LE [47.4 (SD: 11.8)] as compared to patients with a HE [51.7 (SD: 11)] (p value: 0.001). After adjusting for confounders, educational level was found to be an independent predictor of long-term physical and mental health; specifically, patients with a LE had significantly lower SF-12 physical [β: -6.16 (95% CI: -8.01 to -4.31)] and mental [β: -2.48 (95% CI: -4.60 to -0.35)] composite scores compared to patients with a HE.

Conclusion: A lower educational level at the time of traumatic injury is associated with poor long-term mental and physical health. This finding deserves suggests that there may be a role for adapting the available resources (i.e., rehabilitation, financial and family support programs) to the capacity and needs including educational level of individual patients.

 

53.16 Family Satisfaction in the Trauma and Surgical Intensive Care Unit: Another Important Quality Measure

Posted on January 18, 2018

T. Maxim1, A. Alvarez1, Y. Hojberg1, D. Antoku1, C. Moneme1, A. Singleton1, C. Park1, M. Lewis1, K. Inaba1, D. Demetriades1, K. Matsushima1  1University Of Southern California,Los Angeles, CA, USA

Introduction:  A growing body of research has explored patient satisfaction as one of the healthcare quality measures. Family is also a key component of the healing team for critically-ill patients. Therefore, it may follow that a highly-satisfied family member will be better prepared to provide support for their loved one, participate in the decision-making process, and positively influence patient outcomes. To date, scarce data is available in regards to family experience in the trauma and surgical intensive care unit (TSICU). The purpose of this study was to describe and analyze the results of a family satisfaction survey conducted in the TSICU.

Methods:  A prospective series of patients in the TSICU and their family members at a Level 1 Trauma Center were invited to participate in this study after 72 hours of ICU stay. Family members were instructed to complete the Family Satisfaction in the ICU (FS-ICU) questionnaire, a validated survey measuring family satisfaction in the domains of satisfaction with care and decision-making. Data collection spanned from April 2016 to July 2017. Patient characteristics were compiled from the medical record. Quantitative analysis was performed using a 5-point Likert score, converted to a scale of 0 (poor) to 100 (excellent).

Results: A total of 103 family members submitted responses. Respondents were mostly young (median age 41, interquartile range 29-56) and female (75%). Forty five percent had been involved as family of an ICU patient in the past, and 60.8% reported living with the patient. Language fluency was 44.6% English-only, 31.7% Spanish-only, and 23.8% bilingual. Overall satisfaction with care was high. Mean scores ± standard deviation for the TSICU staff’s concern/caring for the patient and respondent were 86.8 ± 21.5 and 82.8 ± 21.8 respectively. Pain (85.8 ± 21.7) and agitation management (81.0 ± 22.9) earned high scores as well. Respondents were mostly satisfied with their level of inclusion in the decision making process (76.3 ± 24.5) and frequency of communication with ICU nursing staff (79.2 ± 22.8), but less satisfied with the frequency of communication with physicians (70.7 ± 27.4). Language translation at the medical center scored 73.2 ± 31.2, and satisfaction was lowest with the atmosphere of the waiting room (64.8 ± 31.4).

Conclusion: While overall family satisfaction with the care provided to patients in the TSICU is high, opportunities for improvement were noted in the frequency of communication between physicians and family. Given the large cohort of exclusively Spanish-speaking family members, language translation services at the medical center are another area for improvement.

53.13 Selecting Patients for Early Tracheostomy After Spinal Cord Injury

Posted on January 18, 2018

D. Scantling1, E. Gleeson1, J. Fazendin1, A. Galvez1, A. Teichman1, J. Eakins2, B. McCracken1  1Drexel University/Hahnemann University Hospital,Surgery,Philadelphia, PA, USA 2AtlantiCare Regional Medical Center,Trauma Surgery,Atlantic City, NJ, USA

Introduction:
Cervical spinal cord injuries (CSCI) often necessitate ventilator dependence. Although endotracheal tubes are the initial airway access of choice, they convey substantial morbidity and tracheostomy is recommended if ventilator support is anticipated to be 7 days. Identifying patients who will need this duration of support and performing early tracheostomy could prevent substantial morbidity, hospital costs and even survival. Prior attempts to identify these patients have been lacking and this provides little data to discuss with consenting family.

Methods:
A retrospective review of the Pennsylvania Trauma Outcome Study (PTOS) database was performed identifying patients with both a cervical spine fracture and CSCI from 2005-2014. 2,339 patients initially met this description. Patients were excluded for incomplete data, never requiring any ventilator support or death within 6 days. Patients with C1-4 CSCI were included. 223 patients met these metrics. Cohorts were created based on type of CSCI and those needing ventilator support for ≥7 days. ISS, AIS, age, GCS, LOS, ICU LOS, mortality, tracheostomy were evaluated. Analysis was accomplished using confidence intervals, Mann-Whitney U tests and Chi Square.

Results:
Of 223 patients meeting inclusion criteria, 142 had complete C1-C4 CSCI.133/142 required ≥7 ventilator days (93.7%, 95% CI 88.3% to 97.1%) and 120/142 underwent tracheostomy (84.5%, 95% CI 77.5% to 90%). Of those with ≥7 ventilator days, mean highest level of fracture was 3.35, ISS was 55.8, mean ventilator days were 30.54, GCS was 7.60  and GCS motor was 2.59. 20 died and 119 had a surgical airway placed (89.5%, 95% CI 83.0% to 94.1%). Of those with ≤6 ventilator days, mean highest associated fracture level was 4.67, ISS was 35.11, age was 46.33, mean ventilator days was 2.11 (range 1-4). Mean GCS was 14.25 and all patients had a motor score of 6. One underwent tracheostomy. Significant differences were identified between groups for highest level of fracture, ISS, GCS and GCS motor score (p=0.012, p=0.030, p=0.0001, p=0.0001). AIS head, face and neck were not significant (p=0.803, p=0.412 and p=0.624). Compared to previously published data, our methodology identified significantly more patients receiving tracheostomy (84.5% vs 64.2%, p=0.007).

Conclusion:
This data should be used to guide family discussions to promote very early tracheostomy for patients with complete C1-C4 injuries associated with a cervical spine fracture, require any ventilator support and are expected to survive for a week. We anticipate decreases in morbidity, length of stay and hospitalization costs amongst these patients. 

53.14 Trauma/Acute Care Surgeries Performed in the Emergency Room Impact on the Timeliness.

Posted on January 18, 2018

K. Ito1, K. Nakazawa1, T. Nagao1, H. Chiba1, T. Fujita1  1Teikyo University Hospital Trauma And Resuscitation Center,Department Of Emergency Medicine, Division Of Acute Care Surgery, Teikyo University School Of Medicine,Tokyo, , Japan

Introduction:  Timely surgical interventions are important for patients with trauma or acute general surgical diseases. At our institution, we have emergency rooms (ERs) with the operating room (OR) set-up which allow surgeons to perform thoracotomy and/or laparotomy without transferring patients to the OR. This practice pattern is not standard in the United States or other western countries. We conducted this study to assess the overall outcomes of this practice, as well as, to test the hypothesis that the ERs with OR set-up improve the timeliness of surgery for trauma / acutely ill general surgical patients.

Methods:  Patients who underwent emergent surgeries by our acute care surgery group (4/2013 – 6/2017) were reviewed. Patients’ demographics, diagnoses, the location of operation (ER vs regular OR),  type of operations, time from admission to operation, preoperative interventions, postoperative outcomes, and in-hospital mortality were analyzed. These data were compared with patients who underwent surgery in the ER (ER group) and patients who underwent surgery in the OR (OR group). Parametric data were analyzed by Chi-square test. Non-Parametric data were analyzed by Mann-Whitney U test.

Results: There were 322 consecutive patients (105 traumas [33%] and 217 emergent general surgeries [67%]) who met inclusion criteria. Among them, there were 68 patients who underwent surgery in the ER (21%, ER group) and 254 patients in the OR (79%, OR group). Compared to the OR group, The ER group had more trauma patients (74% vs 24%, p<0.001). The time from admission to operation was shorter in the ER group than the OR group (median 57 minutes [range 4 – 1069] vs 170 minutes [range 25 – 1320], p<0.001). For trauma patients, the Injury Severity Score was higher in the ER group than the OR group (median 34 [range 1 – 59] vs 9 [4 – 45], p<0.001). The in-hospital mortality rate was higher in the ER group than the OR group (42% vs 13%, p<0.001). Details of surgeries in the ER were shown on the Table 1.

Conclusion: The ERs with the OR set-up can allow surgeons to start surgery quicker. These surgeries tended to be performed in sicker patients and likely associated with higher. Further study is warranted for selecting patients who benefit best from this approach.

 

53.11 Elderly Pedestrians Struck by Vehicles Are More Likely to Be Admitted in the Morning

Posted on January 18, 2018

D. C. Patel1, T. Li1, N. K. Dhillon1, N. T. Linaval1, L. Kirillova1, D. R. Margulies1, E. J. Ley1, G. Barmparas1  1Cedars-Sinai Medical Center,Los Angeles, CA, USA

Introduction:  Elderly patients are at high risk for mortality following traumatic injury and prevention is of paramount importance. This study’s aim was to identify specific times of the day during which the elderly are more likely to be injured.

Methods:  The Los Angeles County TEMIS database was retrospectively reviewed for all pedestrians ≥ 18 years who were struck by vehicles over a 16-year period (2000-2015). Elderly (≥ 65 years) patients were compared to non-elderly (18-64 years) with respect to the time of admission and mortality.

Results: Of 36,358 patients, 4,871 (13%) were elderly. The proportion of elderly pedestrian injuries was highest between 6-11am (average 21%), peaking between 10-11 am (23%). This proportion started to abruptly drop at noon, reaching the lowest levels between 2-3 am (2%). Compared to their non-elderly counterparts, the elderly were more likely to have a Glasgow Coma Scale ≤ 8 (11% vs. 7%, p < 0.01), a systolic blood pressure < 90 mmHg (6% vs. 3%, p < 0.01), and a higher Injury Severity Score (median 9 vs. 5, p < 0.01). Overall mortality was 6%, significantly higher in the elderly (14% vs. 4%, p<0.01). Elderly pedestrians had almost a 7-fold higher adjusted odds for death (AOR: 6.8, p<0.01).

Conclusion: Elderly pedestrians struck by vehicles are more likely to be admitted during the morning hours and their mortality risk is high. Preventative strategies with lower speed limits or high surveillance during morning hours in highly populated areas may result in a decreased incidence of these injuries.

53.12 The Depth of Sternal Fracture Displacement is Not Associated with Blunt Cardiac Injury

Posted on January 18, 2018

L. Heidelberg1, R. Uhlich1, P. Bosarge1, J. Kerby1, P. Hu1  1University Of Alabama at Birmingham,Acute Care Surgery,Birmingham, Alabama, USA

Introduction:
Despite little evidence to support an association of sternal fractures with blunt cardiac injury (BCI), displaced sternal fractures are viewed as more severe and more likely to result in BCI. Common recommendations include inpatient evaluation for observation and additional diagnostic screening. Little information exists regarding the depth and severity of sternal fracture displacement and it remains unclear if there is increased risk of BCI with increasing sternal fracture depth. The purpose of this study was to quantify fracture severity by the degree of displacement and evaluate the association of fracture severity with BCI.

Methods:
A retrospective review was performed at an American College of Surgeons verified level 1 trauma center from 2011-2014. All adult patients admitted to the trauma surgery service were eligible for inclusion, with patients excluded for pregnancy, age <18 years old, or lack of imaging. Patients with sternal fracture were identified from the trauma registry using ICD-10 codes. Sternal fracture displacement was measured by posterior displacement in the axial plane of computerized tomography of the chest. Fracture displacement was defined as mild (>0 mm, <5 mm), moderate (≥5 mm, <10 mm), or severe (≥10 mm). BCI was diagnosed using electrocardiogram or echocardiograph and graded according to standard AAST grading.  Analysis was performed using χ2 and Student's t-test or one-way ANOVA for categorical and continuous variables respectively. Multivariate regression analysis was subsequently performed to assess the association of sternal fracture displacement with BCI. The primary outcome of interest was the association of BCI with severity of sternal fracture displacement. 

Results:

235 patients with sternal fractures were identified and eligible for inclusion during the study period. 45% of patients suffered a displaced fracture, with the majority of these being mild (65.1%) or moderate (24.5%) in severity. Only 10.4% of patients had severely displaced fractures. Overall, 42.6% of patients were diagnosed with BCI. Of patients with BCI, there was no difference in mean fracture displacement when compared to patients without BCI (2.4 vs 1.6 mm, p=0.07)[Office1] . There was no significant increase in BCI with sternal fracture displacement when compared to patients with non-displaced fractures (44.3% vs 41.1%, p=0.62)[Office2] . While among patients with displaced fractures, the incidence of BCI increased with increasing severity (39.1% vs 50.0% vs 63.6%, p=0.25), this was not significant. Further, neither fracture displacement (OR 1.10, CI 95% 0.65-1.88) nor severe displacement (OR 2.34, CI 95% 0.64-8.54) were associated with significantly increased risk of BCI on multivariate analysis.

Conclusion:
There is no significant association between the depth of sternal fracture displacement and BCI. Further evaluation and management for BCI should be reserved in the absence of additional symptoms or findings.

53.09 A Review of Hydroxocobalamin Use in Patients with Inhalation Injury at a Regional Burn Center

Posted on January 18, 2018

J. S. Vazquez1,2, L. S. Johnson1,2, T. E. Travis1,2, L. T. Moffatt2, J. W. Shupp1,2  1MedStar Washington Hospital Center,The Burn Center, Department Of Surgery,Washington, DC, USA 2Firefighters’ Burn And Surgical Research Laboratory,Washington, DC, USA

Introduction:
Hydrogen cyanide is produced during combustion of several different types of household materials such as synthetic polymers. This small lipid soluble molecule inhibits oxidative phosphorylation and produces a severe lactic acidosis in affected patients. Early identification of cyanide (CN) toxicity allows for treatment with Hydroxocobalamin, a CN binding agent that forms cyanocobalamin, a non-toxic substance excreted by the kidneys. However, even though the medication has a mild safety profile, its high cost should guide judicious use. We evaluated the appropriateness of a single burn center’s administration of Hydroxocobalamin in patients with inhalation injury suspected of having CN intoxication.

Methods:
In this single center retrospective study, pharmacy records where queried for patients that received Hydroxocobalamin between January 2014 through June 2017. Twenty-seven patients received treatment based on the suspicion of inhalation injury and CN toxicity and where included in this study. The electronic medical record of those patients was queried to collect data regarding survival status, clinical parameters, and details surrounding Hydroxocobalamin administration. Blood CN levels were drawn either prior to or immediately after Hydroxocobalamin administration in 20 patients. Clinical criteria for out of hospital administration of Hydroxocobalamin by EMS was decreased GCS in the setting of inhalation injury. As for patients arriving to the trauma bay clinical criteria for physician administration of Hydroxocobalamin was acidosis out of proportion to CO2 retention and an elevated CO level in the setting of inhalation injury.

Results:
A total of 27 patients (mean age 48.9 years; 17 men) were treated with Hydroxocobalamin based on the suspicion of CN poisoning in the setting of inhalation injury. House fires (74%) were the most common cause injury overall. Sixteen patients had GCS <8. A concomitant burn injury was present in 78% of patients (%TBSA 0-95). After correcting for smoking status, 70% of patients had CN levels above normal range (Max 2.79mg/L). Presence of an elevated CN level was used as a surrogate for accuracy of administration. All patients arrived at the burn center within the cyanide half-life. Mortality was 37% in this patient cohort.

Conclusion:

A high index of suspicion should be maintained for the presence of CN toxicity in patient with concomitant burn and inhalation injury. Treatment for CN poisoning should be initiated in any patients with inhalation injury, unexplainable lactic acidosis, or impaired consciousness that cannot be explained only by CO poisoning. In this study, we demonstrated that the medication was administered correctly in 70% of patients. Given the broad therapeutic index of Hydroxocobalamin and the mortality associated with cyanide toxicity, this medication should be used when the correct clinical scenario is present, although the high cost might be a barrier for some centers around the world.

53.10 The Bigger They Are, the Harder They Fall: Obesity and Severity of Proximal Humerus Fractures

Posted on January 18, 2018

R. Belayneh1,2, J. Haglin1, A. Lott1, S. Konda1, K. A. Egol1  1New York University School Of Medicine,New York, NY, USA 2Howard University College Of Medicine,Washington, DC, USA

Introduction: The prevalence of obesity in adults has increased significantly in the United States and worldwide. It has been extensively reported in the literature to cause not only medical problems, but musculoskeletal issues as well. In addition to being associated with high rates of osteoarthritis as compared to normal weight populations, obese patients have a 48% increased risk of trauma, including minor injuries and fractures.  Obesity is also associated with an increased risk of injury to the upper limbs resulting from falls from an individual’s own height, including proximal humerus fractures, which account for 4-6% of all fractures. The purpose of this study is to evaluate if there is a relationship between obesity and proximal humerus fracture characteristics.

Methods:   : Proximal humerus fractures at one academic medical center were prospectively followed. Fractures were classified according to the international AO-Müller/Orthopedic Trauma Association (AO/OTA) classification in order to determine their severity. All Type OTA 11-A proximal humerus fractures were categorized as less severe and type OTA 11-B or 11-C were categorized as more severe. Patients’ Body Mass Indexes (BMI) were calculated and used to identify two groups, BMI ≥ 30 kg/cm (obese) and < 30 kg/cm (non-obese). Variables such as age, gender, height, weight, Charlson Comorbity Index (CCI), AO classification; number of complications, latest follow-up shoulder range of motion (ROM), latest follow-up Disabilities of the Arm, Shoulder, and Hand (DASH) survey scores were also recorded. Independent t-tests were used for statistical analysis of continuous variables and χ2 tests for categorical variables. Regression analysis was performed to determine if BMI was a predictor of severity of fractures as determined by the AO classification. Statistical significance was considered as  p<0.05.

Results: Overall, 190 patients who sustained proximal humerus fractures were available for analysis where the average age at time of injury was 59.6±13.9 years. There were 56 OTA 11-A, 67 OTA 11-B, and 67 OTA 11-C fracture types. 58 patients (30.5%) were obese, while 132 patients (69.5%) were non-obese. No significant differences were seen between groups in regards to age, gender, height, CCI, complication rates, or functional and clinical outcomes as determined by follow-up DASH scores and shoulder ROM, respectively. Statistical analysis also demonstrated that obese patients had greater fracture severity per the AO classification (P=0.025).

Conclusion: Based on the results of this study, obesity is associated with more severe fractures of the proximal humerus as determined by the AO/OTA classification. However, there are no differences outcomes or complication rates between obese patients and non-obese patients. With increasing rates of obesity, this relationship may have important epidemiological implications in the future, including predicting proximal humerus fracture burden and severity in society. Additionally, orthopaedic surgeons should be reassured that performing proximal humerus fixation in obese patients yields similar outcomes and complication rates to non-obese patients.
 

53.07 Impact of Red Blood Cell Transfusion in Severe Pediatric Thermal Injury

Posted on January 18, 2018

A. Nordin1, N. Shah2, R. Devine1, R. Fabia1, R. K. Thakkar1  1Nationwide Children’s Hospital,Department Of Pediatric Surgery,Columbus, OH, USA 2Ohio State University College Of Medicine,Columbus, OH, USA

Introduction:
Blood transfusions in adult trauma patients increase the risk of nosocomial infections, and similar results have been found in adult burn patients. However, the literature regarding transfusion practices and their consequences in pediatric burns is limited. We therefore sought to determine the impact of packed red blood cell (pRBC) transfusions in pediatric patients with severe thermal injury, and hypothesized that pRBC administration would be associated with increased risk of infection and greater length of stay.

Methods:
We utilized our institutional trauma registry to identify all patients 0-18 years old admitted for burns 10% total body surface area (TBSA) or greater between 2007and 2015. Data points collected included demographics, mechanism of injury, TBSA and clinical outcomes including overall and intensive care unit (ICU) length of stay (LOS), number of operative procedures, and the number of nosocomial infections. Nosocomial infections were defined as positive cultures treated with a complete course of antibiotics. The volume of pRBC administered was also analyzed, and patients who received transfusions were compared against those who did not. Variables were analyzed using either student’s t-test or chi square analysis as appropriate.

Results:
142 patients were identified during the study period, and 15 were excluded for incomplete records. Of the remaining 127, 39 (30.7%) received pRBC transfusions and 88 (69.3%) did not require transfusion. The two groups of patients were not significantly different in terms of age, sex or weight. Mean TBSA in the transfused group was 28.7%, as compared to 14.9% among those not transfused (p < 0.001). The volume of blood transfused increased with TBSA burned. Transfused patients had significantly longer overall LOS (45.5 v 11.7 days; p < 0.001), ICU LOS (18.2 v 1.1 days; p < 0.001), and increased ventilator days (8.2 v 0.6 days; p < 0.001).The average number of operations (4.5 v 0.6; p < 0.001) and infections (3.5 v 0.2; p < 0.001) also increased relative to non-transfused patients. There were no mortalities in either group.

Conclusion:
Transfusions in pediatric burn patients are associated with adverse outcomes, including prolonged hospital and ICU LOS and increased infections. However, pRBC transfusion increased with increasing TBSA, suggesting a possible confounding factor. Our results add to existing literature on the prevalence and impact of blood transfusions in severe pediatric burns, and may have implications for the resuscitation and management of children with severe thermal injuries.
 

53.08 Use of Serum Amylase Levels to Diagnose Traumatic Pancreatic Injury

Posted on January 18, 2018

R. Uhlich1, J. Kerby1, P. Hu1, P. Bosarge1  1University Of Alabama at Birmingham,Acute Care Surgery,Birmingham, Alabama, USA

Introduction:
Pancreatic injury is a rare, although potentially devastating consequence of trauma. Diagnosis of and appropriate grading of these injuries remains challenging, with variable sensitivity reported for CT. Other methods, such as magnetic resonance cholangiopancreatography (MRCP) or endoscopic retrograde cholangiopancreatography (ERCP) may improve diagnostic accuracy, however are limited to specialized centers and take longer to perform. Serum amylase has been suggested for use as an adjunctive test to help identify patients with pancreatic injury, but remains controversial. We sought to evaluate its role in the diagnosis and management of pancreatic trauma.

Methods:
A retrospective case control study was performed at an American College of Surgeons verified level 1 trauma center from 2011-2017. All adult patients admitted to the trauma surgery service were eligible for inclusion, while patients with pregnancy or age <18 years old were excluded. Patients with pancreatic injury were identified from the trauma registry using ICD-10 codes. Corresponding controls with thoracic trauma, but without pancreatic or hollow viscus injury, were identified and matched using injury severity score (ISS), age, then gender in a 1:1 fashion. Pancreatic injuries were graded according to the AAST guidelines (Grades 1-5), with major injury identified as ≥ grade 3 (pancreatic ductal injury). Serum amylase levels were recorded from admission and throughout hospitalization. Hyperamylasemia was defined as a serum amylase >103 U/L, according to institutional standard. Analysis was performed using χ2 or Student's t-test for categorical and continuous variables respectively. The primary outcome of interest was admission serum amylase level. Secondary outcomes included serum amylase levels in isolated pancreatic injury and grade of pancreatic injury with elevated serum amylase.

Results:
51 patients with pancreatic injury and 51 corresponding controls were identified. Admission hyperamylasemia was identified in 18 patients with pancreatic injury and 2 controls (p<0.001). Average admission serum amylase levels were significantly increased in patients with pancreatic injury compared to without (122.24±136.37 vs 53.90±60.59, p=0.002), which persisted when controlling for hollow viscus injury (145.93±171.13 vs 48.93±32.35, p=0.006). Among patients with pancreatic injury, 36% (18/51) had hyperamylasemia on admission. Hyperamylasemia did not predict major pancreatic injury (50.0% vs 40.6%, p=0.57), need for operative intervention (100% vs 84.4%, p=0.15), or pancreatic resection (38.9% vs 37.5%, p=0.94) when compared to patients with normal admission amylase. 

Conclusion:
Pancreatic injury results in higher mean levels of admission serum amylase. However, measurement of serum amylase is of questionable clinical value as hyperamylasemia fails to differentiate major and minor pancreatic trauma or need for operative intervention. 
 

53.06 What’s in a name? Provider perception of injured John Doe patients

Posted on January 18, 2018

C. F. Janowak1, S. K. Agarwal2, B. L. Zarzaur3  1University Of Cincinnati,Trauma And Surgical Critical Care,Cincinnati, OH, USA 2University Of Wisconsin,Department Of Surgery,Madison, WI, USA 3Indiana University School Of Medicine,Department Of Surgery,Indianapolis, IN, USA

Introduction:
We previously demonstrated that unidentified, aliased patients, John Doe’s (DOE) are one of the highest risk and medically fragile populations of injured patients.  Aliasing can result in misplaced information and confusion that must be overcome by healthcare professionals.  DOE alias use is institutionally dependent and not uniform.  We sought to determine if healthcare practitioners experience confusion and deliver compromised care by caring for injured DOE patients.

Methods:
After obtaining institutional review board (IRB) approval we surveyed critical care nurses, nurse practitioners, resident physicians and surgeons who care for DOE patients at two academic level I trauma centers with separate DOE alias practices.  Surveys asked whether caring for DOE patients created possible or actual confusion, and possible or actual patient care errors.  In one institution (System 1) only unidentified patients were given an alias that was reconciled when information became available.  In the other (System 2) all trauma patients were admitted with an alias that was reconciled within 24 hours.  Respondents were invited to complete an anonymous questionnaire regarding the care for DOE patients.  Results were analyzed with Wilcoxon rank-sum tests and significance was assessed at a level of 0.05.

Results:
Out of 176 total respondents, 120 (68.2%) reported from System 1, and 56 (31.8%) from System 2.  Overall 53.1% reported that DOE alias use can cause serious confusion possibly resulting in errors affecting patient care.  Specifically, 31.3% reported experiencing actual confusion, although only 4% reported actual errors.  Nurses had significantly higher perceived risk of confusion in the system of all DOE versus selective DOE assignment (35.4% vs. 8.2%, p < 0.01).  Resident physicians reported significantly more frequent actual mistakes within the System 2 versus System 1 (61.9% vs. 17.5%, p < 0.01) despite finding no significant difference in resident perception of confusion (54.5% vs. 37.5% respectively, p = 0.2).

Conclusion:
Our study sheds light on clinical consequences of EMR use and aliases for end users.  We show that nurses perceive there are greater potential complications associated with DOE aliases use, and this varies dependent upon the system used for managing unidentified patients.  Provider confusion, risk for error, and patient safety should be considered in DOE alias use.
 

53.04 Treatment of ICU Delirium with QTc Prolonging Medications Does Not Lead to Cardiac Arrhythmias

Posted on January 18, 2018

J. Zakko1, A. Francis2, C. V. Murphy2, D. A. Eiferman1  1Ohio State University Wexner Medical Center,Department Of Surgery,Columbus, OH, USA 2Ohio State University Wexner Medical Center,Department Of Pharmacy,Columbus, OH, USA

Introduction:  ICU delirium is common in the SICU population and many patients are treated with medications that can cause QTc prolongation, which is a risk factor for the development of Torsades de Pointes (TdP). Serial ECGs are often ordered in this population to assess for QTc prolongation, and effective medications are routinely discontinued due to an increase in QTc for fear of progression to TdP. There is limited data available to determine the risk of TdP in the surgical intensive care unit (SICU) patient population as well as any morbidity from QTc prolongation. This study aims to determine if QTc prolongation is associated with development of TdP.

Methods: A single-center retrospective cohort study was conducted to evaluate QTc prolongation and development of TdP amongst non-cardiac SICU patients being treated for ICU delirium at a large academic medical center. Delirium treatment included at least one of the following medications: haloperidol, risperidone, quetiapine, or olanzapine. QTc prolongation was defined as QTc > 500 milliseconds or >20% increase from baseline. Exclusion criteria included ventricular pacing, bundle branch blocks, incarceration, pregnancy, patients on select antipsychotics prior to admission, and congenital long QT syndrome. The primary outcome was to determine prevalence of QTc prolongation and TdP. Secondary outcomes included SICU mortality and risk factors for QTc prolongation. Univariate and multivariate logistic regression models were constructed for assessment of the outcomes.

Results: Eighty patients were eligible for evaluation. Eight (10%) patients had QTc prolongation. There were no cases of TdP. Assessing patient demographics, concomitant antiarrhythmic and antidepressant use, history of heart disease, diuretic use, hepatic dysfunction, SICU length of stay, length of delirium treatment, and SICU mortality in univariate and multivariate analysis yielded no statistically significant association with development of QTc prolongation, morbidity, or mortality.

Conclusion: Among SICU patients receiving treatment for ICU delirium, the frequency of QTc prolongation was only 10% in our cohort with no cases of TdP. Furthermore, univariate and multivariate analysis did not demonstrate a significant correlation between QTc prolongation and morbidity or mortality. These results suggest that QTc prolongation due to medications used to treat delirium does not lead to the development of cardiac arrhythmias. Furthermore, we propose that serial ECG monitoring may be overused and the subsequent discontinuation of delirium-treating medications may not be necessary in this patient population.

53.05 The Epidemiology of Injuries and Related Surgical Intervention in Aftermath of Tornados in America

Posted on January 18, 2018

M. Rajaei1, R. Griffin2, P. Hu1, T. Swain2, J. Kerby1  1UAB,Division Of Acute Care Surgery, Department Of Surgery, School Of Medicine,Birmingham, ALABAMA, USA 2UAB,Department Of Epidemiology, School Of Public Health,Birmingham, ALABAMA, USA

Introduction: Since 1900, natural disasters in the United States have affected more than 26.5 million people and caused almost $737 billion in damage. Tornadoes are common in North America, where the majority of the 1,000 annual tornadoes occur. These disasters result in hundreds of deaths and further injuries every year. Studying the most common injuries and associated operative interventions in the aftermath of the disasters contributes to more efficient mass casualty management. The purpose of this study was to describe the most prevalent injuries and operative procedures perform on victims of tornadoes in the United States.

Methods: A cross-sectional study was performed using National Trauma Data Bank. Tornado-related injuries were identified using ICD-9 codes. Patient information was collected on demographic, injury location, procedures, and clinical characteristics. Injury severity was measured by Abbreviated Injury Scale (AIS), Injury Severity Score (ISS), and Glasgow Coma Scale (GCS).

Results: From 2003-2013, 1,059 individual-injuries were included. Age was 44±39 years with 53% female. The length of hospital and ICU stay were 10±23 and 8±16 days respectively. The average ISS and GCS were 17±22 and 13±5, with 24.6% having a GCS < 8. Most common injury locations were thorax (53.3%), head (52.2%), lower (51.7%) and upper (53.6%) extremities. Severe injuries happened most commonly in the thorax (40.0%), head (15.2%), and lower extremities (15.2%). Head (85.6%), thorax (79.9%), and spinal (64.3%) injuries were the most common injuries among fatal cases. Severe thoracic injuries occurred nearly twice as common as severe head injuries in patient who didn’t survive the incidents (66.3% vs. 35.9%). Patients were equally admitted to the ICU (35.6%) or regular floor (31.5%) with 20.1% requiring operative intervention prior to admission. The most common procedures were performed on skin/ soft tissue (17.9%); fractures/ dislocations (16.9%); injured vessels (11.0%); chest (7.9%); and muscle/ tendons (5.6%).

Conclusion: ~~Head and thoracic injuries were most common in tornadoes, with the latter being the most prevalent severe injury associated with mortality. While more study is needed, these data could potentially inform disaster preparedness and injury prevention approaches.

53.03 Prehospital Blunt Traumatic Cardiac Arrest: Is It Worth the Resuscitative Investment?

Posted on January 18, 2018

N. R. Manley1, J. Holley1,2, J. Martin2, T. Stavely1, M. Croce1, P. E. Fischer1  1University Of Tennessee Health Science Center,Surgery,Memphis, TENNESSEE, USA 2Memphis Fire Department,Memphis, TENNESSEE, USA

Introduction:  Prehospital resuscitation of patients with blunt traumatic cardiac arrest (BTCA) is known to have extremely poor outcomes when success is considered hospital discharge. However, the results are unknown when survival is defined as hospital discharge or organ donation. Prehospital protocols must be adjusted to decrease futile resource utilization while not sacrificing potential survivors or donors, as one donor can potentially provide 7 organs. We sought to identify factors associated with survival (discharge or organ donation) of patients with BTCA.

Methods:  All adult patients with BTCA transported by our city EMS for 2013-2015 were included.  Data on demographics, prehospital resuscitation, prehospital injury assessment, hospital course and outcomes were collected and analyzed.  Survival was considered either hospital discharge or consideration for organ donation.  

Results:  There were 61 patients with BTCA and overall survival was 5% (1 discharge, 2 organ donation). Two kidneys and one liver were successfully procured allowing 3 transplants.  There was no difference in survivors regarding mean prehospital CPR time (25 vs 24 min) or initial cardiac rhythm.  On prehospital assessment, 36 patients (59%) had documented head trauma and 2 survived (8%).  22 patients (36%) had torso trauma and none survived.  All patients required extensive prehospital resources with an average ambulance service time of 103 minutes. 

Conclusion:  Prehospital resuscitation of BTCA patients with evidence of torso trauma appears futile and these patients should be declared in the field.  However, BTCA patients with no apparent trauma or injuries isolated to the head/face have a small chance of survival to discharge or organ donation and resuscitation should be attempted until further study can elucidate more prehospital predictors of death.

 

52.21 Increasing “Off-label” use of GnRH Agonists among Pediatric Patients in the United States

Posted on January 18, 2018

C. M. Lopez1, D. Solomon1, R. A. Cowles1, D. E. Ozgediz1, D. H. Stitelman1, M. G. Caty1, E. R. Christison-Lagay1  1Yale University School Of Medicine,Department Of Surgery, Section Of Pediatric Surgery,New Haven, CT, USA

Introduction: Gonadotropin releasing hormone (GnRH) agonists are FDA approved for the treatment of precocious puberty. Therapy consists of either histrelin acetate (Supprelin, Endo Pharmaceuticals), a surgically implanted device seen as the preferred form of treatment, or leuprolide acetate (Lupron Depot, Abbvie) injections. In recent years, the use of these agents has been extended to include the “off-label” treatment of a heterogeneous group of pediatric conditions with normally timed puberty including hyperandrogenism, endometriosis, short stature, and gender dysphoria. We sought to investigate the trends in the “off-label” use of GnRH agonists among pediatric patients, hypothesizing that “off-label” usage would preferentially favor those patients covered by commercial insurers.  

Methods: We analyzed data on the use of Supprelin and Lupron reported to the Pediatric Health Information System (PHIS) from 2013 to 2016.  Demographic information and payer status were collected. ICD-9 and ICD-10 diagnostic codes were analyzed to determine whether the indication for therapy was for precocious puberty (“on-label use”) or for other diagnostic codes (“off-label use”).

Results: 39 children’s hospitals within the PHIS provided outpatient surgical and billing data on the administration of GnRH agonist therapies during the study period. During this period, the annual number of unique pediatric patients treated with GnRH agonists for precocious puberty increased modestly, from 283 to 303; meanwhile, the number of procedures for an off-label indication more than tripled from 39 to 125. The increasing use of the Supprelin implants to treat short stature and gender dysphoria were largely responsible for this increase in off-label use, the former increasing from 10 to 46 patients and the latter from 3 to 57 patients over the study period.

When compared to the distribution of patients treated for precocious puberty, commercially insured patients were more likely to receive GnRH agonists for an off-label indication (p=0.047) than those qualifying for public insurance. This disparity was more pronounced in those in whom GnRH agonists were being used to treat short stature or gender dysphoria (p=0.028).

Conclusion: From 2013 to 2016, off-label use of GnRH agonists in children increased more than threefold with the greatest increases seen in the treatment of short stature and gender dysphoria. Patients treated for these conditions were more likely to be privately insured than the larger cohort of patients receiving on-label treatment for precocious puberty, thus identifying a potential disparity in access to care which requires future investigation.  

53.02 Impact of Aggressive Treatments in Trauma: Using the Emergency Department Thoracotomy to Death Ratio

Posted on January 18, 2018

D. C. Patel1, N. K. Dhillon1, A. Ko1, C. Colovos1, N. Melo1, D. R. Margulies1, E. J. Ley1, G. Barmparas1  1Cedars-Sinai Medical Center,Los Angeles, CA, USA

Introduction:  More than 50 years after its introduction, Emergency Department Thoracotomy (EDT) remains a heavily debated procedure due to the absence of high level evidence in its support. We have previously shown that there is significant variation among trauma centers in performing EDT for trauma patients in extremis, with liberal such centers performing additional, unnecessary procedures. We sought to explore the correlation between trauma center practices in regards to the performance of EDT and survival of patients admitted to these centers. We hypothesized that centers that are liberal in performing EDT would not necessarily have increased survival.

Methods:  Level I and II trauma centers contributing data to the National Trauma Data Bank between 2007 to 2011 were included. Centers with < 200 subjects reported and ≤ 25 ED deaths (EDD) during the study period were excluded. The data was aggregated and the counts for EDT and EDD were calculated for each center. All centers were then divided into quartiles based on the ratio of EDT:EDD. A multivariate logistic regression model was then utilized to calculate the adjusted odds ratio (AOR) for mortality for patients admitted to each quartile. Patients admitted to centers with the lowest quartile (Q1) were used as the reference group. The primary outcome was overall mortality and mortality among subgroup of patients, including those with penetrating trauma and those with critical injuries (ISS≥25).

Results: A total of 174 trauma centers admitting 1,432,811 subjects were included. The median EDT:EDD ratio ranged from 0 for Q1 to 17.6% for centers in the highest quartile (Q4). Q4 centers were more likely to be Level I (Q4: 53% vs. Q1: 44%, p=0.03) and an academic center (Q4: 67% vs. Q1: 42%, p=0.05). Compared to patients admitted to Q1 centers, those admitted to Q4 centers had a significantly higher adjusted mortality (AOR: 1.06, p<0.01). This difference applied also to the subgroup of patients with a penetrating injury (AOR: 1.21, p<0.01). There was no difference in mortality for patients with critical injuries (AOR: 1.02, p=0.52).

Conclusion: Trauma centers where emergency department thoracotomy is liberally performed for trauma patients in extremis had higher adjusted mortality compared to less liberal centers. This paradoxical finding might be explained in part by triaging patients at extremely high risk for mortality to these centers that are liberal in performing those procedures. Further investigation of this phenomenon is required to identify areas for potential improvement and standardization of the management of the trauma patient in extremis, avoiding unnecessary interventions.
 

52.18 The Impact of Sociodemographic and Hospital-Related Factors on Length of Stay After Pyloromyotomy

Posted on January 18, 2018

M. Joseph1, E. Hamilton1, K. Tsao1, M. T. Austin1,2  1McGovern Medical School At The University Of Texas Health Science Center At Houston,Pediatric Surgery,Houston, TX, USA 2University Of Texas MD Anderson Cancer Center,Surgical Oncology,Houston, TX, USA

Introduction: Hypertrophic pyloric stenosis (HPS) is a common surgical problem in infants and pyloromyotomy is the curative treatment of choice in this population. Hospitalizations following pyloromyotomy for HPS are usually short, often less than one day.  Recent research has focused on evaluating the presence of health disparities in children and their impact on clinical outcomes.  The purpose of this study was to evaluate the association sociodemographic and hospital factors with post-operative length of stay (LOS) in patients with HPS.

Methods: We identified all patients age < 1 year old hospitalized with a primary diagnosis of HPS in 2006, 2009 and 2012 using the Health Care Utilization Project-Kids Inpatient Database. We included all patients who received an open or laparoscopic pyloromyotomy during the hospital admission. Patient demographics, hospital characteristics and clinical outcomes were collected.  All patients with secondary procedures during the same admission were excluded. Prolonged post-operative length of stay (LOS) was defined as greater than 1 day.  Neonatal age is defined as <= 28 days in the KIDS database.  Binary logistic regression was used to calculate odds ratios (OR) with 95% confidence intervals using SPSS version 24.

Results: A total of 12,401 cases were identified with a diagnosis of HPS. Most (n=7387, 60%) were non-Hispanic White (NHW) followed by Hispanic (n=3159, 26%), non-Hispanic Black (NHB) (n=923, 7%) and Other (n=932, 7%).  The majority had public insurance (n=7551, 61%) followed by private insurance (n=4076 33%) with only 3% uninsured/self-pay.  The median post-operative length of stay was 1.00 + 0.93 days.  By multivariate analysis, the presence of fluid/ electrolyte disorders (OR=1.56, 95%CI:1.41-1.72), neonatal age (OR=1.38, 95%CI:1.26-1.51), NHB and Other race/ethnicity (OR=1.34, 95%CI: 1.15-1.62; OR=1.32, 95%CI:1.12-1.55, respectively), and rural hospital location (OR= 1.67, 95%CI:1.29-2.16) were all independently associated with prolonged post-operative LOS.  Compared to the North, all hospital regions were associated with decreased post-operative LOS: Midwest (OR= 0.72, 95% CI: 0.63-0.82), South (OR= 0.87, 95% CI: 0.77-0.99) and West (OR= 0.83, 95% CI: 0.72-0.94). Insurance status and zip-code based income quartile were not statistically significant predictors of prolonged post-operative LOS.

Conclusion: NHB and Other race/ethnicity were independently associated with prolonged post-operative LOS following pyloromytomy for HPS. Other hospital-related factors were important including rural versus urban location and geographic region.  Further research is needed to better describe and address disparities in the clinical management and outcome of children with HPS.

 

« Previous Page
Next Page »