Deep Learning for FAST Quality Assessment.

OBJECTIVES: To determine the feasibility of using a deep learning (DL) algorithm to assess the quality of focused assessment with sonography in trauma (FAST) exams. METHODS: Our dataset consists of 441 FAST exams, classified as good-quality or poor-quality, with 3161 videos. We first used convolutional neural networks (CNNs), pretrained on the Imagenet dataset and fine-tuned on the FAST dataset. Second, we trained a CNN autoencoder to compress FAST images, with a 20-1 compression ratio. The compressed codes were input to a two-layer classifier network. To train the networks, each video was labeled with the quality of the exam, and the frames were labeled with the quality of the video. For inference, a video was classified as poor-quality if half the frames were classified as poor-quality by the network, and an exam was classified as poor-quality if half the videos were classified as poor-quality. RESULTS: The results with the encoder-classifier networks were much better than the transfer learning results with CNNs. This was primarily because the Imagenet dataset is not a good match for the ultrasound quality assessment problem. The DL models produced video sensitivities and specificities of 99% and 98% on held-out test sets. CONCLUSIONS: Using an autoencoder to compress FAST images is a very effective way to obtain features that can be used to predict exam quality. These features are more suitable than those obtained from CNNs pretrained on Imagenet.

Impact of Transfusing Packed Red Blood Cells Through a Rapid Infuser on Potassium Levels.

BACKGROUND: Hyperkalemia is associated with the rapid transfusion of packed red blood cells in trauma patients. Rapid infusers can infuse blood up to 500 ml/min. OBJECTIVE: This study aimed to determine whether infusing packed red blood cells through a rapid infuser impacts the potassium levels of the infused blood. METHODS: Two baseline samples were obtained to measure potassium and hemolysis scores in 12 units of expired blood prior to infusion. The blood was then infused via the Belmont Rapid Infuser into collection bags at varying infusion rates (50, 100, 250, and 500 ml/min) utilizing different gauge catheter sizes (18-gauge, 16-gauge, and Cordis catheter). Two postinfusion blood samples were collected and tested for potassium and hemolysis scores and compared with preinfusion values. This process was then repeated with fresh blood. RESULTS: The potassium levels of the samples taken from each unit prior to infusion (average difference 0.245) and after infusion (average difference 0.08) correlated well. There was no difference in potassium levels pre- and postinfusion at any infusion rate after accounting for catheter size and age of blood. The median potassium level of the fresh blood was 5.025 prior to infusion and 4.875 after infusion. The median potassium level of the expired blood was 16.05 prior to infusion and 16.4 postinfusion. There was no significant difference in the hemolysis scores between the preinfusion and postinfusion samples. CONCLUSIONS: Hyperkalemia in trauma patients undergoing massive transfusions is not a result of mechanical hemolysis from the high rates of blood infusion.

Isolated Rib Fractures in Elderly Falls: Not As Deadly As We Think.

BACKGROUND: Rib fractures in elderly patients have been associated with high morbidity and mortality; however, many of these patients had substantial mechanisms of injury, which may have contributed to these high rates. OBJECTIVE: The purpose of this study was to determine the morbidity and mortality of elderly patients with isolated rib fractures who fell from standing. METHODS: A single-institution retrospective study was conducted in a Level I trauma center using the trauma registry and a separate elderly fall from standing database. Admitted patients 65 years or older who presented with rib fractures after a fall from January 2013 to June 2017 were included. Patients with a nonthoracic Abbreviated Injury Scale score greater than 2 were excluded from the study. RESULTS: Of 129 patients with isolated rib fracture, 94% (n = 121) had comorbidities and 71% (n = 92) had two or more comorbidities. Almost half (41.9%; n = 54) were taking antiplatelet and anticoagulant medications, 78.3% (n = 101) were caused by a mechanical fall, and 7% (n = 9) were caused by syncope. Data showed 72.9% (n = 94) had three or more rib fractures. The mortality rate of patients was 3.9% (n = 5). Three patients had dementia at death, four had do-not-resuscitate order, and only two deaths were directly related to pulmonary status. Patients who developed pneumonia (14.7%; n = 19) and required mechanical ventilation for a median of 11 days (3.9%; n = 5) were fewer than those in in previous studies. CONCLUSION: The morbidity and mortality associated with rib fractures are significantly less than reported in the literature when additional injuries are excluded.

The Effect of Opioid Dependence on Firearm Injury Treatment Outcomes: A Nationwide Analysis.

BACKGROUND: Both the opioid and gun violence epidemics are recurrent public health issues in the United States. We sought to determine the effect of opioid dependence on gunshot injury treatment outcomes. MATERIALS AND METHODS: Using the 2016 National Readmission Database, patients were included if they had a principal diagnosis of firearm injury. Opioid dependence was identified using appropriate International Classification of Diseases, 10th Revision, Clinical Modification codes. The primary outcome was 30-day all-cause readmission. Secondary outcomes were in-hospital and 1-year mortality, resource utilization, and most common reasons for admission and readmission. Confounders were adjusted for using multivariate regression analysis. RESULTS: A total of 31,303 patients were included, 695 of whom were opioid dependent. Opioid-dependent patients were more likely to be young (35.1 y, range: 33.4-36.7 y) and male (89.9%) compared with patients without opioid dependence. Opioid dependence was associated with higher 30-day readmission rates (adjusted odds ratio [aOR]: 1.67, 95% confidence interval [CI]: 1.12-2.50, P = 0.01). However, opioid dependence was associated with lower in-hospital (aOR: 0.16, CI: 0.07-0.38, P < 0.01) and 1-year (aOR: 0.15, CI: 0.06-0.38, P < 0.01) mortality, longer mean length of stay (adjusted mean difference [aMD]: 2.09 d, CI: 0.43-3.76, P = 0.03), and total hospitalization costs (aMD: $6,318, CI: $ 257-$12,380, P = 0.04). Both groups had similar total hospitalization charges (aMD: $$10,491, CI: -$12,618-$33,600, P-value = 0.37). CONCLUSIONS: Opioid dependence leads to higher rates of 30-day readmission and resource utilization among patients with firearm injuries. However, the in-hospital and 1-year mortality rates are lower among patients with opioid dependence secondary to lower injury acuity.

A quality assessment tool for focused abdominal sonography for trauma examinations using artificial intelligence.

BACKGROUND: Current tools to review focused abdominal sonography for trauma (FAST) images for quality have poorly defined grading criteria or are developed to grade the skills of the sonographer and not the examination. The purpose of this study is to establish a grading system with substantial agreement among coders, thereby enabling the development of an automated assessment tool for FAST examinations using artificial intelligence (AI). METHODS: Five coders labeled a set of FAST clips. Each coder was responsible for a different subset of clips (10% of the clips were labeled in triplicate to evaluate intercoder reliability). The clips were labeled with a quality score from 1 (lowest quality) to 5 (highest quality). Clips of 3 or greater were considered passing. An AI training model was developed to score the quality of the FAST examination. The clips were split into a training set, a validation set, and a test set. The predicted scores were rounded to the nearest quality level to distinguish passing from failing clips. RESULTS: A total of 1,514 qualified clips (1,399 passing and 115 failing clips) were evaluated in the final data set. This final data set had a 94% agreement between pairs of coders on the pass/fail prediction, and the set had a Krippendorff α of 66%. The decision threshold can be tuned to achieve the desired tradeoff between precision and sensitivity. Without using the AI model, a reviewer would, on average, examine roughly 25 clips for every 1 failing clip identified. In contrast, using our model with a decision threshold of 0.015, a reviewer would examine roughly five clips for every one failing clip - a fivefold reduction in clips reviewed while still correctly identifying 85% of passing clips. CONCLUSION: Integration of AI holds significant promise in improving the accurate evaluation of FAST images while simultaneously alleviating the workload burden on expert physicians. LEVEL OF EVIDENCE: Diagnostic Test/Criteria; Level II.

Wound infection rate after skin closure of damage control laparotomy with wicks or incisional negative wound therapy: An EAST multi-center trial.

Damage control laparotomy (DCL) has a high risk of SSI and as an attempt to mitigate this, surgeons often leave the skin open to heal by secondary intention. A recent retrospective study showed that DCL wounds could be closed with the addition of wicks or incisional wound vacs with acceptable rates of wound infection. The aim of this prospective trial was to corroborate these results. This is a prospective multicenter observational trial performed by 7 institutions from July 2020 to April 2022. Adult patients who underwent DCL and fascia/skin closure with the addition of wicks or an incisional wound vac were included. Patients who died within seven days of DCL were excluded. Demographics, mechanism of initial presentation, wound classification, antibiotics given, surgical site infections, procedures performed, and mortality data was collected. Fisher's Exact test was used for categorical data and Wilcoxon Rank Sum test for continuous data. Mean days to closure was assessed using Student's t-test for independent groups. P-values <0.05 were considered indicative of statistical significance. Over the 21-month period, a total of 119 patients analyzed. Most patients were male (n = 66, 63 %), and the average age was 51 years. The average number of days the abdomen was kept open was 2.6. A majority of the DCLs were performed on acute care patients (n = 76, 63.8 %) and 92 patients (77.3 %) had a wound classification of contaminated or dirty. Most of the patients' skin was closed with wicks in place (68.9 %). There was a 9.8 % infection rate in patient's skin closed with wicks versus 16.2 % closed with an incisional wound vac (p = 0.361). Although the wick group had a higher proportion of class III and IV wound types, patients primarily treated with wicks had a lower risk of wound infection compared to those treated with incisional wound VACs; however, this difference was not statistically significant.

Failure rates of nonoperative management of low-grade splenic injuries with active extravasation: an Eastern Association for the Surgery of Trauma multicenter study.

Objectives: There is little evidence guiding the management of grade I-II traumatic splenic injuries with contrast blush (CB). We aimed to analyze the failure rate of nonoperative management (NOM) of grade I-II splenic injuries with CB in hemodynamically stable patients. Methods: A multicenter, retrospective cohort study examining all grade I-II splenic injuries with CB was performed at 21 institutions from January 1, 2014, to October 31, 2019. Patients >18 years old with grade I or II splenic injury due to blunt trauma with CB on CT were included. The primary outcome was the failure of NOM requiring angioembolization/operation. We determined the failure rate of NOM for grade I versus grade II splenic injuries. We then performed bivariate comparisons of patients who failed NOM with those who did not. Results: A total of 145 patients were included. Median Injury Severity Score was 17. The combined rate of failure for grade I-II injuries was 20.0%. There was no statistical difference in failure of NOM between grade I and II injuries with CB (18.2% vs 21.1%, p>0.05). Patients who failed NOM had an increased median hospital length of stay (p=0.024) and increased need for blood transfusion (p=0.004) and massive transfusion (p=0.030). Five patients (3.4%) died and 96 (66.2%) were discharged home, with no differences between those who failed and those who did not fail NOM (both p>0.05). Conclusion: NOM of grade I-II splenic injuries with CB fails in 20% of patients. Level of evidence: IV.

Propensity weighted analysis of chemical venous thromboembolism prophylaxis agents in isolated severe traumatic brain injury: An EAST sponsored multicenter study.

BACKGROUND: In patients with severe traumatic brain injury (TBI), clinicians must balance preventing venous thromboembolism (VTE) with the risk of intracranial hemorrhagic expansion (ICHE). We hypothesized that low molecular weight heparin (LMWH) would not increase risk of ICHE or VTE as compared to unfractionated heparin (UH) in patients with severe TBI. METHODS: Patients ≥ 18 years of age with isolated severe TBI (AIS ≥ 3), admitted to 24 level I and II trauma centers between January 1, 2014 to December 31, 2020 and who received subcutaneous UH and LMWH injections for chemical venous thromboembolism prophylaxis (VTEP) were included. Primary outcomes were VTE and ICHE after VTEP initiation. Secondary outcomes were mortality and neurosurgical interventions. Entropy balancing (EBAL) weighted competing risk or logistic regression models were estimated for all outcomes with chemical VTEP agent as the predictor of interest. RESULTS: 984 patients received chemical VTEP, 482 UH and 502 LMWH. Patients on LMWH more often had pre-existing conditions such as liver disease (UH vs LMWH 1.7 % vs. 4.4 %, p = 0.01), and coagulopathy (UH vs LMWH 0.4 % vs. 4.2 %, p < 0.001). There were no differences in VTE or ICHE after VTEP initiation. There were no differences in neurosurgical interventions performed. There were a total of 29 VTE events (3 %) in the cohort who received VTEP. A Cox proportional hazards model with a random effect for facility demonstrated no statistically significant differences in time to VTE across the two agents (p = 0.44). The LMWH group had a 43 % lower risk of overall ICHE compared to the UH group (HR = 0.57: 95 % CI = 0.32-1.03, p = 0.062), however was not statistically significant. CONCLUSION: In this multi-center analysis, patients who received LMWH had a decreased risk of ICHE, with no differences in VTE, ICHE after VTEP initiation and neurosurgical interventions compared to those who received UH. There were no safety concerns when using LMWH compared to UH. LEVEL OF EVIDENCE: Level III, Therapeutic Care Management.

Epic Sepsis Model Inpatient Predictive Analytic Tool: A Validation Study.

Earlier treatment of sepsis leads to decreased mortality. Epic is an electronic medical record providing a predictive alert system for sepsis, the Epic Sepsis Model (ESM) Inpatient Predictive Analytic Tool. External validation of this system is lacking. This study aims to evaluate the ESM as a sepsis screening tool and determine whether an association exists between ESM alert system implementation and subsequent sepsis-related mortality. DESIGN: Before-and-after study comparing baseline and intervention period. SETTING: Urban 746-bed academic level 1 trauma center. PATIENTS: Adult acute care inpatients discharged between January 12, 2018, and July 31, 2019. INTERVENTIONS: During the before period, ESM was turned on in the background, but nurses and providers were not alerted of results. The system was then activated to alert providers of scores greater than or equal to 5, a set point determined using receiver operating characteristic curve analysis (area under the curve, 0.834; p < 0.001). MEASUREMENTS AND MAIN RESULTS: Primary outcome was mortality during hospitalization; secondary outcomes were sepsis order set utilization, length of stay, and timing of administration of sepsis-appropriate antibiotics. Of the 11,512 inpatient encounters assessed by ESM, 10.2% (1,171) had sepsis based on diagnosis codes. As a screening test, the ESM had sensitivity, specificity, positive predictive value, and negative predictive value rates of 86.0%, 80.8%, 33.8%, and 98.11%, respectively. After ESM implementation, unadjusted mortality rates in patients with ESM score greater than or equal to 5 and who had not yet received sepsis-appropriate antibiotics declined from 24.3% to 15.9%; multivariable analysis yielded an odds ratio of sepsis-related mortality (95% CI) of 0.56 (0.39-0.80). CONCLUSIONS: In this single-center before-and-after study, utilization of the ESM score as a screening test was associated with a 44% reduction in the odds of sepsis-related mortality. Due to wide utilization of Epic, this is a potentially promising tool to improve sepsis mortality in the United States. This study is hypothesis generating, and further work with more rigorous study design is needed.

Damage Control Laparotomy Wound Incisions Can Be Closed With Low Rates of Superficial Surgical Site Infections.

Due to high rates of surgical site infections (SSIs) in damage control laparotomies (DCLs), many surgeons leave wounds to heal by secondary intention. We hypothesize that patients after DCL can have their wounds primarily closed with wicks/Penrose drains with low rates of superficial surgical site infections. A retrospective review of a prospectively maintained DCL database was performed for all patients who underwent DCL from January 2016 to June 2018. From January 2016 to June 2018, a total of 171 patients underwent DCL. After exclusions, 107 patients were reviewed to assess for SSI. 57 patients were closed with wicks/Penrose drains, 3 were closed with delayed primary closure, and 47 patients were closed completely at time of fascial closure. There were 4 (3.7%) superficial SSIs, 13 (12.1%) organ space infections, and 14 surgical site occurrences (3 of which required opening the skin). Primary closure of incisions after DCL has low superficial SSI rates.

Percutaneous Endoscopic Gastrostomy Placement in Trauma Patients: Early vs Delayed Initiation of Enteral Feeding.

In critically ill trauma patients, adequate nutrition is essential for the body's healing process. Currently, there is no clinical standard for initiating feeds after percutaneous endoscopic gastrostomy (PEG) tube placement. We aimed to demonstrate that early enteral nutrition (EN) is as safe as delayed EN in patients who have undergone PEG tube insertion. We conducted a multi-center, retrospective cohort study of 384 patients from the Prisma Health Trauma Registries who received PEGs. Feeding intolerance was defined as high gastric residuals, nausea, emesis, sustained diarrhea, or ileus. The probability that a patient would experience intolerance was 11.7% in those fed within 6 hours, 5.1% among patients fed between 6 and 12 hours, 6.0% among patients fed between 12 and 24 hours, and 7.6% among patients fed after 24 hours, for which no statistically significant difference was detected. These findings support that early EN after PEG placement is safe in critically ill, trauma patients.

Early VTE prophylaxis in severe traumatic brain injury: A propensity score weighted EAST multicenter study.

BACKGROUND: Patients with traumatic brain injury (TBI) are at high risk of venous thromboembolism events (VTE). We hypothesized that early chemical VTE prophylaxis initiation (≤24 hours of a stable head CT) in severe TBI would reduce VTE without increasing risk of intracranial hemorrhage expansion (ICHE). METHODS: A retrospective review of adult patients 18 years or older with isolated severe TBI (Abbreviated Injury Scale score, ≥ 3) who were admitted to 24 Level I and Level II trauma centers from January 1, 2014 to December 31 2020 was conducted. Patients were divided into those who did not receive any VTE prophylaxis (NO VTEP), who received VTE prophylaxis ≤24 hours after stable head CT (VTEP ≤24) and who received VTE prophylaxis >24 hours after stable head CT (VTEP>24). Primary outcomes were VTE and ICHE. Covariate balancing propensity score weighting was utilized to balance demographic and clinical characteristics across three groups. Weighted univariate logistic regression models were estimated for VTE and ICHE with patient group as predictor of interest. RESULTS: Of 3,936 patients, 1,784 met inclusion criteria. Incidences of VTE was significantly higher in the VTEP>24 group, with higher incidences of DVT in the group. Higher incidences of ICHE were observed in the VTEP≤24 and VTEP>24 groups. After propensity score weighting, there was a higher risk of VTE in patients in VTEP >24 compared with those in VTEP≤24 (odds ratio, 1.51; 95% confidence interval, 0.69-3.30; p = 0.307), however was not significant. Although, the No VTEP group had decreased odds of having ICHE compared with VTEP≤24 (odds ratio, 0.75; 95% confidence interval, 0.55-1.02, p = 0.070), the result was not statistically significant. CONCLUSION: In this large multi-center analysis, there were no significant differences in VTE based on timing of initiation of VTE prophylaxis. Patients who never received VTE prophylaxis had decreased odds of ICHE. Further evaluation of VTE prophylaxis in larger randomized studies will be necessary for definitive conclusions. LEVEL OF EVIDENCE: Therapeutic Care Management; Level III.

The Utility of Serial Hemoglobin Monitoring in Non-Operative Management of Blunt Splenic Injury.

BACKGROUND: The Eastern Association for the Surgery of Trauma (EAST) states there is not enough evidence to recommend a particular frequency of measuring Hgb values for non-operative management (NOM) of blunt splenic injury (BSI). This study was performed to compare the utility of serial Hgb (SHgb) to daily Hgb (DHgb) in this population. METHODS: We conducted a retrospective chart review of patients with BSI between 2013 and 2019. Demographics, comorbidities, lab values, clinical decisions, and outcomes were gathered through a trauma database. RESULTS: A total of 562 patients arrive in the trauma bay with BSI. In the NOM group, 297 were successful and 37 failed NOM. Of those that failed NOM, 8 (21.6%) changed to OM due to a drop in Hgb. 5 (62.5%) were hypotensive first, 2 (25%) were no longer receiving SHgb, and 1 (12.5%) had a repeat CT scan and was embolized. DHgb patients were not significantly different from SHgb patients in injury severity, length of stay, the largest drop in Hgb, and incidence of failing NOM. Patients taking aspirin were more likely to fall below 7 g/dl at 48 and 72 hours into admission. CONCLUSIONS: These results suggest that that trending SHgb may not influence clinical decision-making in NOM of BSI. Besides taking aspirin, risk factors for who would benefit from SHgb were not identified. Patients who received DHgb had similar injuries and outcomes than patients who received SHgb. Prospective studies are needed to evaluate the clinical utility of SHgb compared to DHgb.

Helicopter Transport of Trauma Patients Improves Survival Irrespective of Transport Time.

OBJECTIVES: Helicopter transport of trauma patients remains controversial. We examined the survival rates of patients undergoing helicopter versus ground transport to a Level 1 trauma center. METHODS: Retrospective analysis was performed on trauma patients treated between 2014 and 2017. Student's t-test was used to compare air versus ground transport times. A logistic regression was then used to examine the association of transportation type on survival controlling for demographics, mechanism of injury, transport time, field intubation, and injury severity. RESULTS: Of 3967 patients identified, 69.6% (2762) were male, and the average age was 40 years. Most patients suffered blunt injuries (86.8%, 3445), while the remaining had penetrating injuries (11.6%, 459) or burns (1.6%, 63). The majority of patients were transferred by ground (3449) with only 13% (518) transferred by air. Patients transported by air had increased Injury Severity Score (ISS) with a median of 17 (IQR 9-24) versus 9 (IQR 5-14), increased length of stay (LOS) at 6 days versus 3 (P < .001), and increased mortality at 12.6% vs 6.5% (P < .001). Patients transported by air arrived 16.6 ± 6.7 minutes faster compared with ground for the zip codes examined. When adjusting for the mechanism of injury, ISS, age, gender, intubation status, and transport time, air transport was associated with an increased likelihood of survival (odds ratio [OR] = 1.57, 95% CI = 1.06-2.40). CONCLUSION: In our analysis of 3967 patients, those transported by air had a significant improvement in the likelihood of survival compared with those transported by ground even when adjusting for both ISS and time.

Morbidity and Mortality in the Obese Trauma Intensive Care Unit Patient.

INTRODUCTION: Obesity is an epidemic in the United States, known to be associated with comorbidities. However, some data show that obesity may be a protective factor in some instances. The purpose of this study is to determine if there are differences in morbidity and mortality when comparing the obese and non-obese critically ill trauma patient populations. MATERIALS AND METHODS: This was a retrospective study conducted at Prisma Health Upstate in Greenville, South Carolina, an Adult Level 1 Trauma Center. Patients over the age of 18 years admitted due to trauma from February 6, 2016 to February 28, 2019 were included in this study. Burn patients were excluded. An online trauma database was used to obtain age, sex, body mass index, Glasgow coma score (GCS), injury severity score (ISS), revised trauma score (RTS), days on mechanical ventilation, hospital length of stay (LOS), and intensive care unit (ICU) LOS. RESULTS: There were 2365 critically ill trauma patients who met inclusion criteria for this study. 1570 patients were men (66.38%) and mean age was 53.2 ± 20.9. Of the patients, 2166 patients had blunt trauma (91.59%). Median GCS was 15 (interquartilerange [IQR]: 12, 15), median RTS was 12 (IQR: 11, 12), and median ISS was 17 (IQR: 9, 22). Obese critically ill trauma patients had significantly lower odds of mortality than nonobese (OR .686, CI 0.473-.977). Penetrating traumas (OR: 4.206, CI: 2.478, 6.990), increased ISS (OR: 1.095, CI: .473, 1.112), and increased age (OR: 1.036, CI: 1.038, 1.045) were associated with significantly increased odds of mortality. DISCUSSION: The obesity paradox is observed in the obese critically ill trauma patient population.

Avoiding Cribari gridlock 2: The standardized triage assessment tool outperforms the Cribari matrix method in 38 adult and pediatric trauma centers.

OBJECTIVES: The Cribari Matrix Method (CMM) is the current standard to identify over/undertriage but requires manual trauma triage reviews to address its inadequacies. The Standardized Triage Assessment Tool (STAT) partially emulates triage review by combining CMM with the Need For Trauma Intervention, an indicator of major trauma. This study aimed to validate STAT in a multicenter sample. METHODS: Thirty-eight adult and pediatric US trauma centers submitted data for 97,282 encounters. Mixed models estimated the effects of overtriage and undertriage versus appropriate triage on the odds of complication, odds of discharge to a continuing care facility, and differences in length of stay for both CMM and STAT. Significance was assessed at p <0.005. RESULTS: Overtriage (53.49% vs. 30.79%) and undertriage (17.19% vs. 3.55%) rates were notably lower with STAT than with CMM. CMM and STAT had significant associations with all outcomes, with overtriages demonstrating lower injury burdens and undertriages showing higher injury burdens than appropriately triaged patients. STAT indicated significantly stronger associations with outcomes than CMM, except in odds of discharge to continuing care facility among patients who received a full trauma team activation where STAT and CMM were similar. CONCLUSIONS: This multicenter study strongly indicates STAT safely and accurately flags fewer cases for triage reviews, thereby reducing the subjectivity introduced by manual triage determinations. This may enable better refinement of activation criteria and reduced workload.

Race Impacts Outcomes of Patients With Firearm Injuries.

BACKGROUND: To study the relationship between race and outcomes of patients with firearm injuries hospitalized in the United States. METHODS: The 2016 National Inpatient Sample was used. Patients were included if they had a principal diagnosis of firearm injury. Exclusion criteria were age <16 years and elective admissions. The primary outcome was in-hospital mortality. Secondary outcomes were morbidity (traumatic shock, prolonged mechanical ventilation, acute respiratory distress syndrome [ADRS], and ventilator-associated pneumonia [VAP]), and resource utilization (length of stay and total hospitalization charges and costs). RESULTS: The sample included 31 335 patients; 52% were Black and 29% were Caucasian. The mean age was 32 years and 88% were male. Black patients had lower odds of mortality (adjusted odds ratio (aOR): 0.41 (95% CI: 0.32-0.53), P < .01). However, compared with Caucasians, Blacks had higher mean total hospitalization charges (adjusted mean difference (aMD) : $14 052 (CI: $1469-$26 635), P = .03) and costs (aMD: $3248 (CI: $654-$5842), P = .01) despite similar mean length of stay (aMD: 0.70 (CI: -0.05-1.45), P = .07). Both racial groups had similar rates of traumatic shock (aOR: 0.91 (0.72-1.15), P = .44), prolonged mechanical ventilation (aOR: 0.82 (0.63-1.09), P = .17), ARDS (aOR: 1.18 (0.45-3.07), P = .74) and VAP (aOR: 1.27 (0.47-3.41), P = .63). DISCUSSION: Black patients with firearm injuries had a lower adjusted odds of in-hospital mortality compared with other races. However, despite having a similar hospital length of stay and in-hospital morbidity, -Black patients had higher total hospitalization costs and charges.

Development of Trauma Level Prediction Models Using Emergency Medical Service Vital Signs to Reduce Over- and Undertriage Rates in Penetrating Wounds and Falls of the Elderly.

Determining triage activation levels in geriatric patients who fall (GF), and patients with penetrating wounds can be difficult and inaccurate, resulting in excessive overtriage (OT) and undertriage (UT) rates. We developed trauma activation prediction models using field data to predict with greater accuracy trauma activation level and triage rates consistent with the ACS recommendations. Using data from the 2014 National Trauma Data Bank, we created binary regression equations for each type of injury (GF and penetrating wounds). The 2014 data were randomized and divided into two halves. The first half for each injury type was used to generate prediction models, whereas the second half of the 2014 data were combined with 2013 and 2015 National Trauma Data Bank data for model verification. Binary regression equations were generated from vital signs collected by EMS. A Cribari grid with ISS ≥ 15 was used to determine the appropriateness of activation level. Chi-square analysis was used to determine significant differences between OT, UT, and accuracy predictions. Using our triage models, we were able to obtain UT rates of less than 4 per cent for GF with OT rates of less than 40 per cent, UT rates less than 4.1 per cent and OT of less than 50 per cent for patients with gunshot wounds, and UT rates less than 4 per cent and OT rates less than 25 per cent for patients who had stab wounds. Our developed trauma level prediction models enable health providers to predict trauma activation levels that can result in OT and UT rates in the recommended ranges by the ACS.

Rethinking the definition of major trauma: The need for trauma intervention outperforms Injury Severity Score and Revised Trauma Score in 38 adult and pediatric trauma centers.

BACKGROUND: Patients' trauma burdens are a combination of anatomic damage, physiologic derangement, and the resultant depletion of reserve. Typically, Injury Severity Score (ISS) >15 defines major anatomic injury and Revised Trauma Score (RTS) <7.84 defines major physiologic derangement, but there is no standard definition for reserve. The Need For Trauma Intervention (NFTI) identifies severely depleted reserves (NFTI+) with emergent interventions and/or early mortality. We hypothesized NFTI would have stronger associations with outcomes and better model fit than ISS and RTS. METHODS: Thirty-eight adult and pediatric U.S. trauma centers submitted data for 88,488 encounters. Mixed models tested ISS greater than 15, RTS less than 7.84, and NFTI's associations with complications, survivors' discharge to continuing care, and survivors' length of stay (LOS). RESULTS: The NFTI had stronger associations with complications and LOS than ISS and RTS (odds ratios [99.5% confidence interval]: NFTI = 9.44 [8.46-10.53]; ISS = 5.94 [5.36-6.60], RTS = 4.79 [4.29-5.34]; LOS incidence rate ratios (99.5% confidence interval): NFTI = 3.15 [3.08-3.22], ISS = 2.87 [2.80-2.94], RTS = 2.37 [2.30-2.45]). NFTI was more strongly associated with continuing care discharge but not significantly more than ISS (relative risk [99.5% confidence interval]: NFTI = 2.59 [2.52-2.66], ISS = 2.51 [2.44-2.59], RTS = 2.37 [2.28-2.46]). Cross-validation revealed that in all cases NFTI's model provided a much better fit than ISS greater than 15 or RTS less than 7.84. CONCLUSION: In this multicenter study, NFTI had better model fit and stronger associations with the outcomes than ISS and RTS. By determining depletion of reserve via resource consumption, NFTI+ may be a better definition of major trauma than the standard definitions of ISS greater than 15 and RTS less than 7.84. Using NFTI may improve retrospective triage monitoring and statistical risk adjustments. LEVEL OF EVIDENCE: Prognostic, level IV.

Improved Intraoperative Communication after Utilization of Navigational Grids during Laparoscopic Cholecystectomies.

Efficient intraoperative communication (IC) between an attending and resident during surgery is highly valuable. Few tools, however, have been developed to improve IC. This study evaluates IC between residents and attendings after utilization of a navigational grid (NG) during laparoscopic cholecystectomies. Attendings and surgery residents completed a 10-question survey after performing a laparoscopic cholecystectomy. Surveys were collected for 12 weeks: six weeks before use of NGs and six weeks with use of NGs. The NGs were constructed to fit our 26-surgical monitors and allowed the monitors to be divided into a 7 × 4 grid. Hunderd and fifteen surveys were collected: 50 from attendings (pre-NG: 31 vs NG: 19) and 65 from residents (42 vs 23). Before NGs, attendings were less likely than residents to perceive attending instructions to be clear (64.5 vs 93.0%, P = 0.0001) and less likely to believe IC left little room for confusion during the procedure (64.5 vs 90.5%, P = 0.007). After NGs, attendings believed intraoperative directional guidance was more concise and clear (64.5 vs 89.5%, P = 0.062); they also reported that NGs left little room for IC confusion during the procedure (64.5 vs 94.7%, P = 0.039). Surveys showed the grid's utility to be inversely correlated with years of experience. Residents (