RESUMO
PURPOSE: Intertrochanteric fracture (ITF) classification is crucial for surgical decision-making. However, orthopedic trauma surgeons have shown lower accuracy in ITF classification than expected. The objective of this study was to utilize an artificial intelligence (AI) method to improve the accuracy of ITF classification. METHODS: We trained a network called YOLOX-SwinT, which is based on the You Only Look Once X (YOLOX) object detection network with Swin Transformer (SwinT) as the backbone architecture, using 762 radiographic ITF examinations as the training set. Subsequently, we recruited 5 senior orthopedic trauma surgeons (SOTS) and 5 junior orthopedic trauma surgeons (JOTS) to classify the 85 original images in the test set, as well as the images with the prediction results of the network model in sequence. Statistical analysis was performed using the Statistical Package for the Social Sciences (SPSS) 20.0 (IBM Corp., Armonk, NY, USA) to compare the differences among the SOTS, JOTS, SOTS + AI, JOTS + AI, SOTS + JOTS, and SOTS + JOTS + AI groups. All images were classified according to the AO/OTA 2018 classification system by 2 experienced trauma surgeons and verified by another expert in this field. Based on the actual clinical needs, after discussion, we integrated 8 subgroups into 5 new subgroups, and the dataset was divided into training, validation, and test sets by the ratio of 8:1:1. RESULTS: The mean average precision at the intersection over union (IoU) of 0.5 (mAP50) for subgroup detection reached 90.29%. The classification accuracy values of SOTS, JOTS, SOTS + AI, and JOTS + AI groups were 56.24% ± 4.02%, 35.29% ± 18.07%, 79.53% ± 7.14%, and 71.53% ± 5.22%, respectively. The paired t-test results showed that the difference between the SOTS and SOTS + AI groups was statistically significant, as well as the difference between the JOTS and JOTS + AI groups, and the SOTS + JOTS and SOTS + JOTS + AI groups. Moreover, the difference between the SOTS + JOTS and SOTS + JOTS + AI groups in each subgroup was statistically significant, with all p < 0.05. The independent samples t-test results showed that the difference between the SOTS and JOTS groups was statistically significant, while the difference between the SOTS + AI and JOTS + AI groups was not statistically significant. With the assistance of AI, the subgroup classification accuracy of both SOTS and JOTS was significantly improved, and JOTS achieved the same level as SOTS. CONCLUSION: In conclusion, the YOLOX-SwinT network algorithm enhances the accuracy of AO/OTA subgroups classification of ITF by orthopedic trauma surgeons.
RESUMO
PURPOSE: SAM junctional tourniquet (SJT) has been applied to control junctional hemorrhage. However, there is limited information about its safety and efficacy when applied in the axilla. This study aims to investigate the effect of SJT on respiration when used in the axilla in a swine model. METHODS: Eighteen male Yorkshire swines, aged 6-month-old and weighing 55 - 72 kg, were randomized into 3 groups, with 6 in each. An axillary hemorrhage model was established by cutting a 2 mm transverse incision in the axillary artery. Hemorrhagic shock was induced by exsanguinating through the left carotid artery to achieve a controlled volume reduction of 30% of total blood volume. Vascular blocking bands were used to temporarily control axillary hemorrhage before SJT was applied. In Group I, the swine spontaneously breathed, while SJT was applied for 2 h with a pressure of 210 mmHg. In Group II, the swine were mechanically ventilated, and SJT was applied for the same duration and pressure as Group I. In Group III, the swine spontaneously breathed, but the axillary hemorrhage was controlled using vascular blocking bands without SJT compression. The amount of free blood loss was calculated in the axillary wound during the 2 h of hemostasis by SJT application or vascular blocking bands. After then, a temporary vascular shunt was performed in the 3 groups to achieve resuscitation. Pathophysiologic state of each swine was monitored for 1 h with an infusion of 400 mL of autologous whole blood and 500 mL of lactated ringer solution. Tb and T0 represent the time points before and immediate after the 30% volume-controlled hemorrhagic shock, respectively. T30, T60, T90 and T120, denote 30, 60, 90, and 120 min after T0 (hemostasis period), while T150, and T180 denote 150 and 180 min after T0 (resuscitation period). The mean arterial pressure and heart rate were monitored through the right carotid artery catheter. Blood samples were collected at each time point for the analysis of blood gas, complete cell count, serum chemistry, standard coagulation tests, etc., and thromboelastography was conducted subsequently. Movement of the left hemidiaphragm was measured by ultrasonography at Tb and T0 to assess respiration. Data were presented as mean ± standard deviation and analyzed using repeated measures of two-way analysis of variance with pairwise comparisons adjusted using the Bonferroni method. All statistical analyses were processed using GraphPad Prism software. RESULTS: Compared to Tb, a statistically significant increase in the left hemidiaphragm movement at T0 was observed in Groups I and II (both p < 0.001). In Group III, the left hemidiaphragm movement remained unchanged (p = 0.660). Compared to Group I, mechanical ventilation in Group II significantly alleviated the effect of SJT application on the left hemidiaphragm movement (p < 0.001). Blood pressure and heart rate rapidly increased at T0 in all three groups. Respiratory arrest suddenly occurred in Group I after T120, which required immediate manual respiratory assistance. PaO2 in Group I decreased significantly at T120, accompanied by an increase in PaCO2 (both p < 0.001 vs. Groups II and III). Other biochemical metabolic changes were similar among groups. However, in all 3 groups, lactate and potassium increased immediately after 1 min of resuscitation concurrent with a drop in pH. The swine in Group I exhibited the most severe hyperkalemia and metabolic acidosis. The coagulation function test did not show statistically significant differences among three groups at any time point. However, D-dimer levels showed a more than 16-fold increase from T120 to T180 in all groups. CONCLUSION: In the swine model, SJT is effective in controlling axillary hemorrhage during both spontaneous breathing and mechanical ventilation. Mechanical ventilation is found to alleviate the restrictive effect of SJT on thoracic movement without affecting hemostatic efficiency. Therefore, mechanical ventilation could be necessary before SJT removal.
Assuntos
Choque Hemorrágico , Doenças Vasculares , Masculino , Animais , Suínos , Choque Hemorrágico/terapia , Torniquetes , Axila , Hemorragia/terapia , RespiraçãoRESUMO
Background & aim: The association of perioperative blood transfusion (PBT) with long-term survival in perihilar cholangiocarcinoma (pCCA) patients after surgical resection with curative intent is controversial and may differ among different stages of the disease. This study aimed to investigate the impact of PBT on long-term survival of patients with different stages of pCCA. Methods: Consecutive pCCA patients from three hospitals treated with curative resection from 2012 to 2019 were enrolled and divided into the PBT and non-PBT groups. Propensity score matching (PSM) was used to balance differences in baseline characteristics between the PBT and non-PBT groups. Kaplan-Meier curves and log-rank test were used to compare overall survival (OS) and recurrence-free survival (RFS) between patients with all tumor stages, early stage (8th AJCC stage I), and non-early stage (8th AJCC stage II-IV) pCCA in the PBT and non-PBT groups. Cox regression analysis was used to determine the impact of PBT on OS and RFS of these patients. Results: 302 pCCA patients treated with curative resection were enrolled into this study. Before PSM, 68 patients (22 patients in the PBT group) were in the early stage and 234 patients (108 patients in the PBT group) were in the non-early stage. Patients with early stage pCCA in the PBT group had significantly lower OS and RFS rates than those in the non-PBT group. However, there were with no significant differences between the 2 groups with all tumor stages and non-early stage pCCA. After PSM, there were 18 matched pairs of patients with early stage and 72 matched pairs of patients with non-early stage. Similar results were obtained in the pre- and post-PSM cohorts: patients with early stage pCCA in the PBT group showed significantly lower OS and RFS rates than those in the non-PBT group, but there were no significant differences between the 2 groups for patients with all tumor stages and non-early stage pCCA. Cox regression analysis demonstrated that PBT was independently associated with worse OS and RFS for patients with early stage pCCA. Conclusions: PBT had a negative impact on long-term survival in patients with early stage pCCA after curative resection, but not in patients with non-early stage pCCA.