Journal of Trauma and Acute Care Surgery最新文献

Background: Appendicitis has traditionally been managed with urgent surgery. Because of operative room and logistic constraints, surgery may be delayed. Evidence on the impact of this delay remains equivocal. The aim of this systematic review and meta-analysis is to assess postoperative outcomes of delayed appendectomy amongst both pediatric and adult populations.

Methods: A systematic review with meta-analysis was performed including studies comparing cohorts of urgent versus delayed appendectomy, excluding those with interval appendectomy. Studies were included as long as one relevant postoperative complication was mentioned; for adults, only prospective studies were included. A comprehensive search of six databases was performed including studies from January 1, 2000, to January 15, 2024. A meta-analysis with a random effects model and restricted maximum likelihood was used.

Results: Of 11,227 citations, 20 pediatric and 5 adult studies were included, with 827,019 and 4250 patients, respectively. Definitions of early surgery cohorts were usually surgery within 4 to 12 hours or overnight, and delayed surgery >4 to 12 hours or next day. The pediatric meta-analysis revealed no increased risk of intraoperative perforation in delayed versus emergent cohorts (odds ratio [OR], 1.16; 95% confidence interval [CI], 0.78-1.72), length of stay (mean difference, 1.2 days; 95% CI, -0.3 to -2.8), abscess (OR, 0.80; 95% CI, 0.29-2.25), surgical site infection (OR, 1.11; 95% CI, 0.93-1.30), or readmission (OR, 0.82; 95% CI, 0.55-1.21). The adult meta-analysis results revealed no difference between the delayed and emergent appendectomy groups for intraoperative perforation (OR, 1.29; 95% CI, 1.00-1.67), abscess (OR, 1.54; 95% CI, 0.58-4.10), surgical site infection (OR, 1.35; 95% CI, 0.71-2.56), or conversion to open (OR, 0.81; 95% CI, 0.64-1.03). Subgroup analyses showed increased length of stay in pediatric population (mean difference, 0.42 days; 95% CI, 0.10-0.74).

Conclusion: These findings suggest that a modest delay in appendectomy may be permissible in pediatric and adult settings and adult patients presenting with acute appendicitis. While this does not replace surgeon clinical acumen, it may help guide decision making in resource-constrained settings.

Level of evidence: Systematic Review and Meta-analysis; Level II.

Introduction: REBOA is a recognized rescue strategy in exsanguinating shock. Nevertheless, effects of partial- versus total-REBOA inflation on both macrocirculatory and microcirculatory splanchnic flows are not fully understood.

Methods: Controlled hemorrhagic shock was induced in 18 landrace pigs. After 30 minutes of shock, animals were randomly allocated to receiving partial-REBOA (n = 6), total-REBOA (n = 6), or no-REBOA (n = 6). Resuscitation with whole blood was initiated 25 minutes after balloon inflation (in both REBOA groups) or attaining shock (in no-REBOA group). Thereafter, the balloon was progressively deflated according to hemodynamic tolerance. Aortic root, femoral, and end-diastolic left ventricular pressures were monitored throughout the experiment. Simultaneous carotid, supra-celiac abdominal aorta and superior mesenteric artery flows were recorded, while microvascular flows at jejunal-serosa and mucosa were assessed by laser Doppler flowmetry (LDF) and sidestream dark-field video-microscopy. Mesenteric-venous blood samples were drawn to measure blood gases and lactate levels. All macrohemodynamic and microhemodynamic parameters were followed up to 4 hours of completing REBOA deflation (or its equivalent-time in no-REBOA group).

Results: Total-REBOA group showed the highest increase in aortic-root and coronary perfusion pressures during inflation, but these decreased significantly during reperfusion period, compared with partial- and no-REBOA (p < 0.001). Partial- and total-REBOA groups showed significant decreases in superior mesenteric artery flow during reperfusion period compared with no-REBOA (p < 0.001). However, partial-REBOA allowed some flow during inflation while enabling significantly better jejunal-microvascular flow assessed by LDF during reperfusion period, when compared with total-REBOA (p = 0.048). The proportion of jejunal-villi with predominant continuous flow was significantly higher in partial- than total- or no-REBOA groups (p < 0.01). The total-REBOA group had higher arterial and mesenteric-venous lactate levels both during occlusion and reperfusion periods (p = <0.001; p = <0.001, respectively) when compared with partial-REBOA and no-REBOA groups.

Conclusion: Partial-REBOA preserved regional-mesenteric and intestinal microcirculatory blood flow during both balloon occlusion and the early reperfusion period compared with total-REBOA. Partial-REBOA was also related with more favorable mesenteric venous pH and lactate values during balloon occlusion and reperfusion phases.

Level of evidence: Animal experiment; Level III.

Background: No universal definition of trauma-induced coagulopathy exists, and no validated scoring system accurately evaluates coagulopathy in patients with severe trauma. Although developed for sepsis and other critical illnesses, the conventional Japanese Association for Acute Medicine (JAAM) disseminated intravascular coagulation (DIC) score's applicability to trauma populations remains controversial. This study aimed to evaluate the JAAM DIC score's diagnostic performance in patients with severe trauma and develop a trauma-specific coagulopathy scoring system based on its original components.

Methods: This retrospective study analyzed data from a multicenter trauma cohort between April 1, 2018, and March 31, 2019. Patients 18 years or older with severe trauma and an Injury Severity Score of ≥16 were included. The primary outcome was defined as all-cause in-hospital mortality. The secondary outcomes were 24-hour mortality, cause-specific in-hospital mortality (exsanguination, traumatic brain injury [TBI], and others), and massive transfusion incidence. The conventional JAAM DIC variables' predictive performance for all-cause in-hospital mortality, 24-hour mortality, death due to exsanguination, TBI-induced death, and massive transfusion occurrence were assessed using receiver operating characteristic curves. A new trauma DIC score was developed using the optimal cutoff values and compared with the conventional scores.

Results: Among 719 patients analyzed, optimal cutoff values for predicting in-hospital mortality were a prothrombin time-international normalized ratio of 1.080, fibrin degradation products of 116.0 μg/mL, and platelet count of 17.0 (×104/μL). The DIC group had higher all-cause mortality rates than the non-DIC group. The new trauma DIC score outperformed the conventional score for predicting all-cause mortality, 24-hour mortality, TBI-induced death, and massive transfusion requirements; however, both performed similarly for exsanguination-induced death.

Conclusion: We developed a trauma-specific DIC score that outperformed the conventional score for predicting clinical outcomes in patients with severe trauma. While promising for trauma populations, external validation in various clinical settings is warranted.

Level of evidence: Multicenter Retrospective Cohort Study; Level IV.

Background: Thrombocytosis in major trauma patients has been reported with equivocal clinical relevance. We aimed to describe the incidence and natural history of thrombocytosis in intensive care unit (ICU)-admitted polytrauma patients at risk of multiple-organ failure (MOF).

Methods: A 19-year retrospective study ending in December 2023 was performed on a Level 1 center's prospective institutional MOF database. All adults with an Injury Severity Score (ISS) of >15 and ICU patients who survived >48 hours were included. All adults with nonmechanical trauma, isolated traumatic brain injury (TBI), or spinal cord injury or those without sequential platelet monitoring were excluded. Platelet counts were collected until death, discharge, or 28 days. Thrombocytosis and extreme thrombocytosis (ET) were defined as >450,000/μL and >1,000,000/μL. Descriptive statistics were calculated, and mortality, MOF, and venous thromboembolic outcomes were compared between groups. For mortality, multivariable logistic regression was performed adjusting for age, ISS, TBI, and systolic blood pressure.

Results: A total of 797 patients were included (age, 48.8 years; 75% male; 96% blunt; median ISS, 29). Incidence of thrombocytosis was 63% (503 of 797 patients) with ET of 16.5% (83 of 797 patients). Thrombocytosis patients had higher admission counts peaking at 14 to 17 days. Groups did not differ in sex and TBI severity. Thrombocytosis patients were younger and had lower systolic blood pressure on admission and longer median ICU and hospital length of stay (8 vs. 6 and 27 vs. 12, p < 0.005). Incidence of MOF and venous thromboembolism did not differ. Mortality was lower in the thrombocytosis group (7.6% vs. 18%, p < 0.001). One ET patient died. Multiple-organ failure developed in 176 (22%) with incidence of thrombocytosis of 5% (103 of 176 patients). The mean Denver scores between thrombocytosis and no thrombocytosis did not differ, but mortality was lower in the thrombocytosis group (adjusted odds ratio, 0.05; 95% confidence interval, 0.01-0.15).

Conclusion: Thrombocytosis (63%) is frequent in polytrauma patients and is associated with favorable outcomes without higher risk for complications. Early thrombocytosis and rising platelet trajectories may act as a surrogate marker for better chance to survive, and its therapeutic potential warrants detailed exploration.

Level of evidence: Retrospective and observational study, level III.

Background: Balanced transfusion with (1:1:1) ratios of fresh frozen plasma (FFP), packed red blood cells (PRBCs), and platelets is a core tenet of management in traumatic hemorrhagic shock. However, preclinical data suggest that PRBCs transfusion may contribute to endothelial dysfunction, a complication that may be mitigated by FFP. We examined the impact of high FFP:PRBC ratios on mortality and major complications using the American College of Surgeons Trauma Quality Improvement Program database.

Methods: Trauma patients (18 years or older) in Trauma Quality Improvement Program (2017-2022) who received ≥1 U FFP and 1 U PRBC within 4 hours were included. Those transfused with whole blood or FFP:PRBC <0.9 were excluded. Patients were classified as balanced (FFP:PRBC 0.9-1.1) or excess FFP (FFP:PRBC >1.1), propensity matched based on odds of mortality, and compared by multivariate regression.

Results: Of 50,594 patients analyzed (75% male; median age, 38 years; Injury Severity Score, 25), 31,960 (63%) were classified as balanced, and 18,634 (37%) as excess FFP. Propensity matching generated 16,939 pairs (mean standard difference, <0.1). After multivariable adjustment, excess FFP was associated with 40% decreased odds of acute respiratory distress syndrome (p < 0.001) and 32% decreased odds of acute kidney injury (p = 0.027) compared with balanced transfusion, with no difference in hospital mortality (odds ratio, 1.01; 95% confidence interval, 0.92-1.11; p = 0.87).

Conclusion: We found that early excess FFP (FFP:PRBC ratio >1.1) was associated with a significantly decreased odds of key complications compared with a strictly defined balanced transfusion (FFP:PRBC ratio 0.9-1.1), supporting further investigation of the potential benefits of plasma.

Level of evidence: Therapeutic/Care Management; Level III.

Background: In this study, we develop a standardized porcine model of distant injury plus lung bacterial inoculation to allow translational investigations of the effects of tissue injury on susceptibility to infection. This generalizable model will allow testing of immune interventions on the evolution of infection.

Methods: A standardized liver crush (5 cm × 2.5 cm/3 kg) plus hemoperitoneum (6 mL/kg) or sham procedure was performed in 30-kg Yorkshire pigs, followed by intratracheal inoculation of bacteria ( Actinobacillus pleuropneumoniae ). We then compared gross pathology, histology, lung bacterial counts, danger-associated molecular pattern molecules, and serum cytokines between the two groups.

Results: The lungs of injured pigs demonstrated significantly enhanced responses to infection compared with sham injured pigs, both on the macroscopic and microscopic levels. Lung bacterial clearance was significantly impaired after trauma, with increased infiltration of neutrophils and differential location of myeloid cells on immunostaining. In lung parenchyma expression of the stress response genes, Hmox1 and Nrf2 were increased in both trauma alone and trauma plus infection. Plasma from pigs subjected to trauma showed increased levels of the danger-associated molecular patters heme and mitochondrial DNA and promoted bacterial growth in vitro compared with plasma from uninjured pigs.

Conclusion: We have developed a novel, clinically relevant, reproducible porcine model of abdominal injury with subsequent A. pleuropneumoniae pneumonia for the study and development of therapeutics against immune dysregulation induced by trauma. Additionally, a novel finding is that plasma from traumatized pigs provides a permissive environment for bacterial growth.

Background: The benefit of transfusion of fresh frozen plasma (FFP) and platelets in a 1:1 ratio with packed red blood cells (PRBCs) is well established; however, the benefit of a particular ratio of cryoprecipitate to PRBC is not. The Joint Trauma System updated its 2019 Damage Control Resuscitation Guideline by recommending empiric 1:1 cryoprecipitate/PRBCs. We hypothesized that patients receiving product within the cryoprecipitate/PRBC guideline range (high ratio) would have an associated reduction in mortality.

Methods: We included adult patients in the Trauma Quality Improvement Program data registry (2013-2021) who received at least 5 U of PRBCs and 1 U of FFP within 4 hours. Death within 30 minutes, nonsurvivable injury patterns, preexisting coagulopathy, advanced directives, transfers, and burns were excluded. Patients were partitioned into high (≥1:1), medium (≥1:2 to <1:1), and low (<1:2) cryoprecipitate/PRBC ratios. Treatment effects were estimated with propensity score-weighted risk adjustment models, clustering by center. The primary outcome was 6-hour mortality. Secondary outcomes included 24-hour and inpatient mortality. Adjusting for FFP, platelets, and whole blood was included as a sensitivity analysis.

Results: A total of 49,301 patients (high, 5,284; medium, 3,630; low, 40,387) were included. The mean age was 39, 79% were male, 58% suffered blunt trauma, and the mean Injury Severity Score was 29. Unadjusted 6-hour mortality was 11.8%, 18.8%, and 21.3% for high, medium, and low ratios. High ratio was protective as compared with low at 6 hour (adjusted odds ratio [aOR], 0.52; 95% confidence interval [CI], 0.45-0.58) and 24 hours (aOR, 0.74; 95% CI, 0.67-0.82), and medium ratio was protective as compared with low ratio at 6 hours (aOR, 0.78; 95% CI, 0.70-0.87). Blood product sensitivity analysis demonstrated that high and medium ratios were protective of 6-hour, 24-hour, and inpatient mortality.

Conclusion: High cryoprecipitate ratios were independently associated with decreased mortality in massively transfused civilian trauma patients during the first 24 hours. Future prospective multicenter randomized trials are warranted.

Level of evidence: Prognostic and Epidemiological; Level III.

Background: Acute care surgery (ACS) involves rapid, high-stakes decisions with limited opportunity for preoperative planning. While machine learning (ML) may improve risk prediction and decision making in this setting, its development, validation, and implementation in ACS remain understudied. We therefore evaluated the techniques, predictor features, and outcomes used in ML-driven risk prediction models in ACS and generated recommendations to inform future research and support clinically meaningful implementation.

Methods: A systematic review of ML-driven predictive models in ACS (emergency general surgery, surgical critical care, trauma) was conducted. Models were analyzed by predictor features, outcomes, algorithms, and performance. The best-performing models for the most commonly predicted outcome were identified.

Results: Of 52 studies, 57.7% focused on trauma populations. Most models used registry data (76.8%), fewer used electronic health records (28.8%), and only five studies performed external validation after model development. Common algorithms included logistic regression (44.2%), random forest (34.6%), and decision trees (26.9%). Mortality (59.6%), complications (30.8%), and triage/severity (15.4%) were the most frequent outcomes; patient-centered/reported outcomes were absent. Features commonly included demographics, physiologic scores, and vital signs, while imaging and intraoperative data were underused. Natural language processing was used in four studies. Model performance was typically assessed using area under the receiver operating characteristic curve (88.5%), with support vector machines demonstrating the highest performance. Machine learning models generally outperformed conventional risk scores among 11 comparative studies.

Conclusion: Machine learning-driven predictive models in ACS show promising performance but are constrained by limited methodological rigor, real-world validation, and substantial heterogeneity in features, outcomes, and algorithms, challenging systematic adoption and oversight. A grounded understanding of ACS decision making workflows and their postimplementation impact may ensure clinically relevant, seamless, and safe integration of ML-based risk prediction.

Level of evidence: Systematic Review Without Meta-analysis; Level IV.