Timing of male urethral injury treatment in patients with polytrauma: A retrospective study.

INTRODUCTION: Male urethral injuries are uncommon, and the ideal timing of the definitive treatment remains controversial. This study aimed to compare the outcomes of early and delayed interventions (1 month or more after the injury) for male urethral injuries. PATIENT AND METHODS: We conducted a retrospective review of the medical records of 67 male patients with urethral injuries treated at our institution between 2011 and 2020. We examined patient age, injury severity score (ISS), abbreviated injury scale, mechanism, location and severity of injury, presence of pelvic fractures, surgical interventions, timing of treatment, and complications. We analysed factors associated with urinary complications based on the location of urethral injury. Additionally, we performed a subset analysis of patients with severe injuries (ISS≥16) to assess the impact of delayed surgery. RESULTS: Overall, 47 %, 37 %, and 27 % of patients in the delayed treatment group (N = 30) had urethral stricture (US), erectile dysfunction (ED), and/or urinary incontinence (UI). These rates were greater than the 22 % US, 3 % ED, and 11 % UI rates in the early treatment group (N = 37). The subgroup analysis revealed that patients with anterior urethral injury (AUI) who underwent delayed treatment (N = 18) tended to be more severely injured (ISS, 19 vs 9, p = 0.003) and exhibited higher rates of US (44% vs 21 %, p = 0.193) and ED (39% vs 0 %, p = 0.002) than those who received early treatment (N = 24). In the case of posterior urethral injury (PUI), the delayed treatment group (N = 13) had higher rates of US (50% vs 23 %, p = 0.326), ED (33% vs 8 %, p = 0.272), and UI (42% vs 0 %, p = 0.030) than the early treatment group. Regarding study limitations, more than 45 % of the enrolled patients were severely injured (ISS≥16), which may have potentially influenced the timing of urethral injury repair. CONCLUSIONS: The treatment of male urethral injuries may be delayed due to concurrent polytrauma and other associated injuries. However, delayed treatment is associated with higher rates of urinary complications. Early treatment of urethral injuries may be beneficial to male patients with urethral trauma, even in cases of severe injury.

Development and evaluation of a deep learning-based model for simultaneous detection and localization of rib and clavicle fractures in trauma patients' chest radiographs.

Purpose: To develop a rib and clavicle fracture detection model for chest radiographs in trauma patients using a deep learning (DL) algorithm. Materials and methods: We retrospectively collected 56 145 chest X-rays (CXRs) from trauma patients in a trauma center between August 2008 and December 2016. A rib/clavicle fracture detection DL algorithm was trained using this data set with 991 (1.8%) images labeled by experts with fracture site locations. The algorithm was tested on independently collected 300 CXRs in 2017. An external test set was also collected from hospitalized trauma patients in a regional hospital for evaluation. The receiver operating characteristic curve with area under the curve (AUC), accuracy, sensitivity, specificity, precision, and negative predictive value of the model on each test set was evaluated. The prediction probability on the images was visualized as heatmaps. Results: The trained DL model achieved an AUC of 0.912 (95% CI 87.8 to 94.7) on the independent test set. The accuracy, sensitivity, and specificity on the given cut-off value are 83.7, 86.8, and 80.4, respectively. On the external test set, the model had a sensitivity of 88.0 and an accuracy of 72.5. While the model exhibited a slight decrease in accuracy on the external test set, it maintained its sensitivity in detecting fractures. Conclusion: The algorithm detects rib and clavicle fractures concomitantly in the CXR of trauma patients with high accuracy in locating lesions through heatmap visualization.

Deep Learning for Automated Detection and Localization of Traumatic Abdominal Solid Organ Injuries on CT Scans.

Computed tomography (CT) is the most commonly used diagnostic modality for blunt abdominal trauma (BAT), significantly influencing management approaches. Deep learning models (DLMs) have shown great promise in enhancing various aspects of clinical practice. There is limited literature available on the use of DLMs specifically for trauma image evaluation. In this study, we developed a DLM aimed at detecting solid organ injuries to assist medical professionals in rapidly identifying life-threatening injuries. The study enrolled patients from a single trauma center who received abdominal CT scans between 2008 and 2017. Patients with spleen, liver, or kidney injury were categorized as the solid organ injury group, while others were considered negative cases. Only images acquired from the trauma center were enrolled. A subset of images acquired in the last year was designated as the test set, and the remaining images were utilized to train and validate the detection models. The performance of each model was assessed using metrics such as the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, positive predictive value, and negative predictive value based on the best Youden index operating point. The study developed the models using 1302 (87%) scans for training and tested them on 194 (13%) scans. The spleen injury model demonstrated an accuracy of 0.938 and a specificity of 0.952. The accuracy and specificity of the liver injury model were reported as 0.820 and 0.847, respectively. The kidney injury model showed an accuracy of 0.959 and a specificity of 0.989. We developed a DLM that can automate the detection of solid organ injuries by abdominal CT scans with acceptable diagnostic accuracy. It cannot replace the role of clinicians, but we can expect it to be a potential tool to accelerate the process of therapeutic decisions for trauma care.

Does a collecting system injury with grade iv renal blunt trauma always need urinary intervention?

PURPOSE: This study aimed to identify the characteristics associated with the need for urinary intervention for a blunt renal injury with collection system involvement using a computed tomography (CT) protocol for trauma. MATERIALS AND METHODS: Abdominal CT images of patients with blunt renal injuries from 2016 to 2020 were reviewed. Patients with low-grade renal trauma, non-collecting system involvement, American Association for the Surgery of Trauma grade V shattered kidney, and emergent nephrectomy were excluded. The largest perinephric mass thickness was measured in the axial view using CT, and a cutoff value was obtained using a receiver-operating characteristic curve analysis. Risk factors for further urinary intervention were analyzed. RESULTS: Among the 70 patients included in this study, those with perinephric mass thicknesses < 25 mm (n = 36) had a significantly lower rate of urinary intervention than those with perinephric mass thicknesses ≥ 25 mm (0 vs. 5; p = 0.023). There was no significant difference in the follow-up durations of the groups (19 days vs. 38 days; p = 0.198). More than 90% of the perinephric mass in the < 25 mm group resolved within a median follow-up duration of 38 days, whereas nearly half of the ≥ 25 mm group had a residual perinephric mass during a median follow-up duration of 19 days. CONCLUSION: The initial CT protocol for trauma was useful for predicting the need for further urinary interventions for collecting system injuries. A perinephric mass thickness < 25 mm is predictive of a low likelihood of requiring urinary intervention.

The effect of transarterial embolization and nephrectomy on acute kidney injury in patients with blunt renal trauma: An American College of Surgeons - Trauma Quality Improvement Program analysis.

INTRODUCTION: Transarterial embolization (TAE) or nephrectomy for patients with blunt renal trauma might result in acute kidney injury (AKI). Thus, we analyzed the American College of Surgeons - Trauma Quality Improvement Program (TQIP) to validate this. We hypothesized that nephrectomy, and not TAE, would be a risk factor for AKI in patients with blunt renal trauma. MATERIAL AND METHODS: Adult patients with blunt injuries from the TQIP between 2017 and 2019 were eligible for inclusion. The patients were divided into three treatment groups: conservative treatment, TAE, and nephrectomy. Multivariable logistic regression was used to clarify the AKI predictors. RESULTS: The study included 12,843 patients, wherein 12,373 (96.3 %), 229 (1.8 %), and 241 (1.9 %) patients were in the conservative, TAE, and nephrectomy groups, respectively. A total of 269 (2.2 %), 20 (8.7 %), and 29 (12.0 %) patients had AKI in the three groups, respectively. Both TAE (odds ratio [OR], 2.367; 95 % confidence interval [CI], 1.372-3.900; p = 0.001) and Nephrectomy (OR, 2.745; 95 % CI, 1.629-4.528; p < 0.001) were a statistically significant predictor for AKI in the multivariable logistic regression. CONCLUSIONS: TAE and nephrectomy were statistically associated with AKI in patients with blunt renal trauma. This result differs from our previous research findings that nephrectomy, but not TAE, was a risk factor for AKI in patients with blunt renal trauma. Further prospective and well-designed research may be needed.

Low-molecular-weight heparin is associated with lower venous thromboembolism events than factor Xa inhibitors in patients with severe blunt trauma: a cohort study from the Trauma Quality Improvement Program.

BACKGROUND: Venous thromboembolism (VTE), including deep venous thrombosis (DVT) and pulmonary embolism (PE), is a common complication of major trauma. Pharmacological VTE prophylactics are widely used, and low-molecular-weight heparin (LMWH) is recommended. Factor Xa inhibitors are increasingly being used for VTE prophylaxis in both medical and surgical patients. Evidence comparing LMWH and factor Xa inhibitors as VTE prophylactics for severe blunt trauma is lacking. This study aims to compare the efficacy and safety of factor Xa inhibitors and LMHW in VTE prophylaxis. MATERIALS AND METHODS: Patients with severe blunt trauma who received LMWH or a factor Xa inhibitor for VTE prophylaxis in the Trauma Quality Improvement Program between 2017 and 2019 were included. The comparison was performed after using propensity score matching. The outcomes included mortality and incidence of DVT, PE, post-prophylactics haemorrhage control procedures and length of stay. RESULTS: After 2:1 propensity score matching, 1128 patients ( n =752, LMHW group; n =376, factor Xa inhibitor group) were included in the analysis. Patients in the LMWH group had fewer VTE events than those in the factor Xa inhibitor group (DVT, 3.7% vs. 7.2%, P =0.013; PE, 0.4% vs. 3.2%, P <0.001). VTE risk was higher in the factor Xa group (DVT: odds ratio, 1.97; 95% CI, 1.12-3.44; P =0.018 and PE: odds ratio, 9.65; 95% CI, 2.91-44.12; P =0.001). The mortality rate was higher in the LMWH group; however, there was no significant difference (4.0% vs. 1.9%; P =0.075). The difference in the risk of undergoing haemorrhage control surgery after VTE prophylaxis between both groups was insignificant (0.3% vs. 0.0%; P =0.333). CONCLUSIONS: LMWH was associated with a lower risk of VTE than factor Xa inhibitors in patients with severe blunt trauma. The mortality rate was higher in the LMWH group; however, there was no statistically significant difference observed.

Feasibility of non-operative management for patients sustained blunt splenic traumas with concomitant aortic injuries.

PURPOSE: This study aimed to elucidate the treatment approach for blunt splenic injuries concurrently involving the aorta. We hypothesized that non-operative management failure rates would be higher in such cases, necessitating increased hemorrhage control surgeries. METHODS: Data from the Trauma Quality Improvement Program spanning 2017 to 2019 were utilized. All patients with blunt splenic trauma were considered for inclusion. We conducted comparisons between blunt splenic trauma patients with and without thoracic or abdominal aortic injuries to identify any potential disparities in treatment. RESULTS: Among the 32,051 patients with blunt splenic injuries during the study period, 752 (2.3%) sustained concurrent aortic injuries. Following 2:1 propensity score matching, it was determined that the presence of aortic injuries did not significantly affect the utilization of splenic transarterial angioembolization (TAE) (7.2% vs. 8.7%, p = 0.243) or the necessity for splenectomy or splenorrhaphy (15.3% vs. 15.7%, p = 0.853). Moreover, aortic injuries were not a significant factor contributing to TAE failure, regardless of the location or severity of the injury. Patients with simultaneous splenic and aortic injuries required more red blood cell transfusion within first 4 hours (0 ml [0, 900] vs. 0 ml [0, 650], p = 0.001) and exhibited a higher mortality rate (10.6% vs. 7.9%, p = 0.038). CONCLUSION: This study demonstrated that patients with concurrent aortic and splenic injuries presented with more severe conditions, higher mortality rates, and extended hospital stays. The presence of aortic injuries did not substantially influence the utilization of TAE or the necessity for splenectomy or splenorrhaphy. Patients of this type can be managed in accordance with current treatment guidelines. Nonetheless, given their less favorable prognosis, they necessitate prompt and proactive intervention.

Predicting the Risk of Total Hip Replacement by Using A Deep Learning Algorithm on Plain Pelvic Radiographs: Diagnostic Study.

BACKGROUND: Total hip replacement (THR) is considered the gold standard of treatment for refractory degenerative hip disorders. Identifying patients who should receive THR in the short term is important. Some conservative treatments, such as intra-articular injection administered a few months before THR, may result in higher odds of arthroplasty infection. Delayed THR after functional deterioration may result in poorer outcomes and longer waiting times for those who have been flagged as needing THR. Deep learning (DL) in medical imaging applications has recently obtained significant breakthroughs. However, the use of DL in practical wayfinding, such as short-term THR prediction, is still lacking. OBJECTIVE: In this study, we will propose a DL-based assistant system for patients with pelvic radiographs to identify the need for THR within 3 months. METHODS: We developed a convolutional neural network-based DL algorithm to analyze pelvic radiographs, predict the hip region of interest (ROI), and determine whether or not THR is required. The data set was collected from August 2008 to December 2017. The images included 3013 surgical hip ROIs that had undergone THR and 1630 nonsurgical hip ROIs. The images were split, using split-sample validation, into training (n=3903, 80%), validation (n=476, 10%), and testing (n=475, 10%) sets to evaluate the algorithm performance. RESULTS: The algorithm, called SurgHipNet, yielded an area under the receiver operating characteristic curve of 0.994 (95% CI 0.990-0.998). The accuracy, sensitivity, specificity, and F1-score of the model were 0.977, 0.920, 0932, and 0.944, respectively. CONCLUSIONS: The proposed approach has demonstrated that SurgHipNet shows the ability and potential to provide efficient support in clinical decision-making; it can assist physicians in promptly determining the optimal timing for THR.

Accelerated fixation is associated with fewer pulmonary embolism in patients with isolated and unilateral femoral shaft fractures: A TQIP analysis.

INTRODUCTION: The optimal time to intramedullary internal fixation for patients with isolated and unilateral femoral shaft fractures was investigated by the American College of Surgeons-Trauma Quality Improvement Program (ACS-TQIP). MATERIAL AND METHODS: Adult patients from the TQIP between 2017 and 2019 were eligible for inclusion. The associations between time to fixation and target outcomes (pulmonary embolism [PE], deep vein thrombosis [DVT], acute respiratory distress syndrome [ARDS], and mortality) were assessed. Maximization of the sum of sensitivity and specificity was used to determine the optimal cut point. The patients were divided into three groups according to different time to fixation intervals. A multinomial propensity scores weighting using generalized boosted models was performed for all unbiased pre-treatment factors between the groups. Multivariate logistic regression was used to clarify the outcomes predictors. RESULTS: The univariate and multivariate analysis before weighting showed that only PE was significantly associated with time to fixation (odds ratio [OR], 1.05; 95% confidence interval [CI], 1.01-1.09; p = 0.012). The optimal cut point of time to fixation was 13.88 h. The patients were divided into accelerated (≤ 13.88 h), standard (> 13.88 and ≤ 24 h), and delayed (> 24 and < 36 h) groups. After weighting for all unbiased pre-treatment factors between the groups, multivariate logistic regression showed that standard group significantly increased the risk of PE compared with accelerated group (OR, 4.436; 95% CI, 1.844-10.672; p = 0.001). CONCLUSION: Accelerated intramedullary internal fixation within 13.88 h was associated with a significantly decreased risk of PE compared with standard fixation in patients with isolated and unilateral femoral shaft fractures without additional mortality risks.

Speeding pelvic fracture fixation: CT scan with simultaneous 3-D pelvic reconstruction in the emergency department.

INTRODUCTION: Contrast-enhanced computed tomography (CT) scans are usually needed in the emergency department (ED) to evaluate intra-abdominal injuries associated with pelvic fractures. Three-dimensional (3-D) images for pelvis reconstruction are also needed for planning surgical fixation after admission. This study investigates the advantages integrating a one-stage computed tomography (CT) scan with these two diagnostic modalities simultaneously to reduce the time to surgery and improve the outcomes of pelvic fracture fixation. METHODS: A retrospective cohort study (2018-2021) of patients with pelvic fractures was performed. Patients were categorized into the one-stage CT group or the two-stage CT group, and propensity score matching was used to address biases. The outcome measures included time to surgical fixation, time to CT scan for 3-D pelvis reconstruction, and overall length of hospital stay. RESULTS: Four hundred forty-four pelvic fracture patients who underwent definite surgical fixation were identified. Of those, 320 underwent a one-stage CT scan, while the remaining 124 underwent a two-stage CT scan. After well-balanced matching, those in the one-stage CT group had a significantly shorter time to surgical fixation than those in the two-stage CT group (4.6 vs. 6.8 days, p < 0.001). Even among critically ill patients necessitating intensive care unit (ICU) admission, the one-stage CT scan group had a shorter time to definitive surgical fixation (5.5 vs. 7.2 days, p = 0.002) and a shorter hospital stay (19.0 vs. 32.7 days, p = 0.006). CONCLUSION: A one-stage contrast-enhanced CT scan combined with simultaneous 3-D pelvis reconstruction is promising for expediting surgical fixation in pelvic fracture patients. This innovative strategy may improve patient outcomes by facilitating timely surgical interventions and minimizing delays associated with additional CT scans.

The Role of Diagnostic Laparoscopy in the Evaluation of Abdominal Trauma Patients: A Trauma Quality Improvement Program Study.

PURPOSE: We aimed to identify factors related to delayed intervention in abdominal trauma patients who underwent diagnostic laparoscopy using a nationwide databank. METHODS: From 2017 to 2019, abdominal trauma patients who underwent diagnostic laparoscopy were retrospectively evaluated using the Trauma Quality Improvement Program. Patients who underwent delayed interventions after a primary diagnostic laparoscopy were compared with those who did not. Factors associated with poor outcomes that are usually correlated with overlooked injuries and delayed interventions were also analyzed. RESULTS: Of the 5221 studied patients, 4682 (89.7%) underwent inspection without any intervention. Only 48 (0.9%) patients underwent delayed interventions after primary laparoscopy. Compared with patients receiving immediate interventions during primary diagnostic laparoscopy, patients receiving delayed interventions were more likely to have small intestine injuries (58.3% vs. 28.3%, p < 0.001). Among patients with hollow viscus injuries, a significantly higher probability of overlooked injuries that required delayed intervention was observed in patients with small intestine injuries (small intestine injury: 16.8%; gastric injury: 2.5%; large intestine injury: 5.2%). However, delayed small intestine repair did not significantly affect the risk of surgical site infection (SSI) (p = 0.249), acute kidney injury (AKI) (p = 0.998), or hospital length of stay (LOS) (p = 0.053). In contrast, significantly positive relationships between delayed large intestine repair and poor outcomes were observed (SSI, odds ratio = 19.544, p = 0.021; AKI, odds ratio = 27.368, p < 0.001; LOS, ß = 13.541, p < 0.001). CONCLUSIONS: Most examinations and interventions (near 90%) were successful during primary laparoscopy for abdominal trauma patients. Small intestine injuries were easily overlooked. Delayed small intestine repair-related poor outcomes were not observed.

The Application of Design Thinking in Developing a Deep Learning Algorithm for Hip Fracture Detection.

(1) Background: Design thinking is a problem-solving approach that has been applied in various sectors, including healthcare and medical education. While deep learning (DL) algorithms can assist in clinical practice, integrating them into clinical scenarios can be challenging. This study aimed to use design thinking steps to develop a DL algorithm that accelerates deployment in clinical practice and improves its performance to meet clinical requirements. (2) Methods: We applied the design thinking process to interview clinical doctors and gain insights to develop and modify the DL algorithm to meet clinical scenarios. We also compared the DL performance of the algorithm before and after the integration of design thinking. (3) Results: After empathizing with clinical doctors and defining their needs, we identified the unmet need of five trauma surgeons as "how to reduce the misdiagnosis of femoral fracture by pelvic plain film (PXR) at initial emergency visiting". We collected 4235 PXRs from our hospital, of which 2146 had a hip fracture (51%) from 2008 to 2016. We developed hip fracture DL detection models based on the Xception convolutional neural network by using these images. By incorporating design thinking, we improved the diagnostic accuracy from 0.91 (0.84-0.96) to 0.95 (0.93-0.97), the sensitivity from 0.97 (0.89-1.00) to 0.97 (0.94-0.99), and the specificity from 0.84 (0.71-0.93) to 0.93(0.990-0.97). (4) Conclusions: In summary, this study demonstrates that design thinking can ensure that DL solutions developed for trauma care are user-centered and meet the needs of patients and healthcare providers.

Sarcopenia provides extra value outside the PULP score for predicting mortality in older patients with perforated peptic ulcers.

BACKGROUND: Perforated peptic ulcer (PPU) remains challenging surgically due to its high mortality, especially in older individuals. Computed tomography (CT)-measured skeletal muscle mass is a effective predictor of the surgical outcomes in older patients with abdominal emergencies. The purpose of this study is to assess whether a low CT-measured skeletal muscle mass can provide extra value in predicting PPU mortality. METHODS: This retrospective study enrolled older (aged ≥ 65 years) patients who underwent PPU surgery. Cross-sectional skeletal muscle areas and densities were measured by CT at L3 and patient-height adjusted to obtain the L3 skeletal muscle gauge (SMG). Thirty-day mortality was determined with univariate, multivariate and Kaplan-Meier analysis. RESULTS: From 2011 to 2016, 141 older patients were included; 54.8% had sarcopenia. They were further categorized into the PULP score ≤ 7 (n=64) or PULP score > 7 group (n=82). In the former, there was no significant difference in 30-day mortality between sarcopenic (2.9%) and nonsarcopenic patients (0%; p=1.000). However, in the PULP score > 7 group, sarcopenic patients had a significantly higher 30-day mortality (25.5% vs. 3.2%, p=0.009) and serious complication rate (37.3% vs. 12.9%, p=0.017) than nonsarcopenic patients. Multivariate analysis showed that sarcopenia was an independent risk factor for 30-day mortality in patients in the PULP score > 7 group (OR: 11.05, CI: 1.03-118.7). CONCLUSION: CT scans can diagnose PPU and provide physiological measurements. Sarcopenia, defined as a low CT-measured SMG, provides extra value in predicting mortality in older PPU patients.

The Feasibility and Performance of Total Hip Replacement Prediction Deep Learning Algorithm with Real World Data.

(1) Background: Hip degenerative disorder is a common geriatric disease is the main causes to lead to total hip replacement (THR). The surgical timing of THR is crucial for post-operative recovery. Deep learning (DL) algorithms can be used to detect anomalies in medical images and predict the need for THR. The real world data (RWD) were used to validate the artificial intelligence and DL algorithm in medicine but there was no previous study to prove its function in THR prediction. (2) Methods: We designed a sequential two-stage hip replacement prediction deep learning algorithm to identify the possibility of THR in three months of hip joints by plain pelvic radiography (PXR). We also collected RWD to validate the performance of this algorithm. (3) Results: The RWD totally included 3766 PXRs from 2018 to 2019. The overall accuracy of the algorithm was 0.9633; sensitivity was 0.9450; specificity was 1.000 and the precision was 1.000. The negative predictive value was 0.9009, the false negative rate was 0.0550, and the F1 score was 0.9717. The area under curve was 0.972 with 95% confidence interval from 0.953 to 0.987. (4) Conclusions: In summary, this DL algorithm can provide an accurate and reliable method for detecting hip degeneration and predicting the need for further THR. RWD offered an alternative support of the algorithm and validated its function to save time and cost.

Timely surgical fixation confers beneficial outcomes in patients' concomitant flail chest with mild-to-moderate traumatic brain injury: a trauma quality improvement project analysis - a cohort study.

BACKGROUND: Traumatic flail chest results in respiratory distress and prolonged hospital stay. Timely surgical fixation of the flail chest reduces respiratory complications, decreases ventilator dependence, and shortens hospital stays. Concomitant head injury is not unusual in these patients and can postpone surgical timing due to the need to monitor the status of intracranial injuries. Reducing pulmonary sequelae also assists in the recovery from traumatic brain injury and improves outcomes. No previous evidence supports that early rib fixation can improve the outcome of patients with concomitant flail chest and traumatic brain injury. RESEARCH QUESTION: Can early rib fixation improve the outcome of patients with concomitant flail chest and traumatic brain injury? STUDY DESIGN AND METHODS: Adult patients with blunt injuries from the Trauma Quality Improvement Project between 2017 and 2019 were eligible for inclusion. Patients were divided into two treatment groups: operative and nonoperative. Inverse probability treatment weighting was used to identify the predictors of mortality and adverse hospital events. RESULTS: Patients in the operative group had a higher intubation rate [odds ratio (OR), 2.336; 95% CI, 1.644-3.318; p <0.001), a longer length of stay (coefficient ß , 4.664; SE, 0.789; p <0.001), longer ventilator days (coefficient ß , 2.020; SE, 0.528; p <0.001), and lower mortality rate (OR], 0.247; 95% CI, 0.135-0.454; p <0.001). INTERPRETATION: Timely rib fixation can improve the mortality rate of patients with flail chest and a concomitant mild-to-moderate head injury.

Evaluation of ensemble strategy on the development of multiple view ankle fracture detection algorithm.

OBJECTIVE: To identify the feasibility and efficiency of deep convolutional neural networks (DCNNs) in the detection of ankle fractures and to explore ensemble strategies that applied multiple projections of radiographs.Ankle radiographs (AXRs) are the primary tool used to diagnose ankle fractures. Applying DCNN algorithms on AXRs can potentially improve the diagnostic accuracy and efficiency of detecting ankle fractures. METHODS: A DCNN was trained using a trauma image registry, including 3102 AXRs. We separately trained the DCNN on anteroposterior (AP) and lateral (Lat) AXRs. Different ensemble methods, such as "sum-up," "severance-OR," and "severance-Both," were evaluated to incorporate the results of the model using different projections of view. RESULTS: The AP/Lat model's individual sensitivity, specificity, positive-predictive value, accuracy, and F1 score were 79%/84%, 90%/86%, 88%/86%, 83%/85%, and 0.816/0.850, respectively. Furthermore, the area under the receiver operating characteristic curve (AUROC) of the AP/Lat model was 0.890/0.894 (95% CI: 0.826-0.954/0.831-0.953). The sum-up method generated balanced results by applying both models and obtained an AUROC of 0.917 (95% CI: 0.863-0.972) with 87% accuracy. The severance-OR method resulted in a better sensitivity of 90%, and the severance-Both method obtained a high specificity of 94%. CONCLUSION: Ankle fracture in the AXR could be identified by the trained DCNN algorithm. The selection of ensemble methods can depend on the clinical situation which might help clinicians detect ankle fractures efficiently without interrupting the current clinical pathway. ADVANCES IN KNOWLEDGE: This study demonstrated different ensemble strategies of AI algorithms on multiple view AXRs to optimize the performance in various clinical needs.

Effect of zinc supplement on patients with trauma: A systematic review and meta-analysis.

AIMS: This study aimed to assess the effect of zinc supplementation, with or without other antioxidants and trace elements, on clinical outcomes in patients with trauma. METHODS: A systematic review was conducted for adult patients with acute trauma who had been admitted to the hospital. Those who sustained burn injuries were excluded. Studies in PubMed, Web of Science, and Embase from 1990 to 2022 regarding the additional nutrition supplementation of zinc to patients, either in a single-agent or combined regimen, were included. Comparisons were made between the zinc supplement group and those who received a placebo or regular treatment. RESULTS: The primary outcomes of the study were mortality rate, length of hospital stay, and incidence of pneumonia. Seven studies qualified for the meta-analysis. Of the 594 patients eligible for analysis, 290 and 304 were in the zinc supplementation and control groups, respectively. The meta-analysis revealed that zinc supplementation was associated with a lower risk of pneumonia in patients with acute trauma than in the control group (odds ratio [OR], 0.506; 95% CI = 0.292-0.877; P = 0.015; heterogeneity, I2 = 12.7%). Zinc supplementation did not influence the mortality rate (OR, 0.755; 95% CI = 0.492-1.16; P = 0.612; heterogeneity, I2 = 0%) or the length of hospital stay (standard difference in means, -0.24; 95% CI = -0.544 to 0.063; P = 0.121; heterogeneity, I2 = 45.0%). CONCLUSION: Zinc supplementation, with or without other antioxidants and trace elements, in patients with trauma was associated with a lower incidence of pneumonia.

The three-dimensional weakly supervised deep learning algorithm for traumatic splenic injury detection and sequential localization: an experimental study.

BACKGROUND: Splenic injury is the most common solid visceral injury in blunt abdominal trauma, and high-resolution abdominal computed tomography (CT) can adequately detect the injury. However, these lethal injuries sometimes have been overlooked in current practice. Deep learning (DL) algorithms have proven their capabilities in detecting abnormal findings in medical images. The aim of this study is to develop a three-dimensional, weakly supervised DL algorithm for detecting splenic injury on abdominal CT using a sequential localization and classification approach. MATERIAL AND METHODS: The dataset was collected in a tertiary trauma center on 600 patients who underwent abdominal CT between 2008 and 2018, half of whom had splenic injuries. The images were split into development and test datasets at a 4 : 1 ratio. A two-step DL algorithm, including localization and classification models, was constructed to identify the splenic injury. Model performance was evaluated using the area under the receiver operating characteristic curve (AUROC), accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Grad-CAM (Gradient-weighted Class Activation Mapping) heatmaps from the test set were visually assessed. To validate the algorithm, we also collected images from another hospital to serve as external validation data. RESULTS: A total of 480 patients, 50% of whom had spleen injuries, were included in the development dataset, and the rest were included in the test dataset. All patients underwent contrast-enhanced abdominal CT in the emergency room. The automatic two-step EfficientNet model detected splenic injury with an AUROC of 0.901 (95% CI: 0.836-0.953). At the maximum Youden index, the accuracy, sensitivity, specificity, PPV, and NPV were 0.88, 0.81, 0.92, 0.91, and 0.83, respectively. The heatmap identified 96.3% of splenic injury sites in true positive cases. The algorithm achieved a sensitivity of 0.92 for detecting trauma in the external validation cohort, with an acceptable accuracy of 0.80. CONCLUSIONS: The DL model can identify splenic injury on CT, and further application in trauma scenarios is possible.

A Simple Scoring Model for Evaluation of Concomitant Vascular Injuries in Patients with Lower Extremity Open Fractures.

BACKGROUND: Timely diagnosis and management of concomitant vascular injuries is usually needed for the management of lower extremity open fractures. In the current study, a prediction model and simplified scoring system of vascular injuries were developed for the primary evaluation of patients with lower extremity open fractures. METHODS: Patients with lower extremity open fractures were retrospectively reviewed from 2017 to 2020. Multivariate logistic regression analysis was used to evaluate independent risk factors for concomitant vascular injuries in these patients using data collected from 2017 through 2019 and a prediction scoring model was created accordingly. Model performance was validated with data from 2020. RESULTS: In total, 949 patients with lower extremity open fractures (development cohort, 705 patients, 2017 through 2019; validation cohort, 244 patients, 2020) were enrolled. Concomitant vascular injuries occurred in 44 patients in the development cohort (6.2%). Three clinical variables were identified for a prediction scoring model with weighted points, including hard or soft vascular signs (3 points), segmental fractures (2 points), and degloving soft-tissue injury (1 point). The model showed good discrimination (area under the receiver operating characteristic curve, 0.928), calibration (Hosmer-Lemeshow test, P = 0.661), and precision (Brier score, 0.041). Subsequent management regarding different aspects (observation only, further imaging study, or direct surgical exploration) can thus be decided. The model also demonstrated good discrimination (area under the receiver operating characteristic curve, 0.949), good calibration (Hosmer-Lemeshow test, P = 0.174), and good precision (Brier score, 0.042) in the validation cohort. CONCLUSION: This model may guide the subsequent management of vascular injuries associated with lower extremity open fractures. CLINICAL QUESTION/LEVEL OF EVIDENCE: Risk, III.