儿童烧伤液体复苏循证知识图谱与多智能体决策研究

… • Calculate fluid resuscitation requirements in pediatric burn victims based on burn size, depth, … clinical indicators that warrant referral of a child with burns to a specialized burn center …

… Pediatric dehydration is a common condition requiring timely recognition and appropriate fluid resuscitation … The approach to fluid resuscitation in children with burns has evolved from …

… All arrived within 8 hours of injury, but prehospital fluid resuscitation was inadequate in all. Six patients required grafting; no escharotomies. All patients hospitalized >5 days required …

… pSOFA needs validation in other subsets of critically ill children, including burn patients. The … ill pediatric burn patients. METHODS: This is a retrospective cohort study with pediatric burn …

… Among the factors that can explain these conditions, the increase in vasoactive and cytokine secretion following the burn and the need for fluid resuscitation that could lead to non-…

Clinical Practice Guidelines (CPGs) encode evidence-based clinical knowledge but are primarily distributed as unstructured PDF documents, making them inaccessible to automated clinical decision support (CDS) systems. This paper proposes a service-oriented architecture that formalizes the IMSS Clinical Practice Guideline for Major Burn Management (IMSS-375) as a versioned REST/JSON microservice. Seven clinical decision endpoints are defined, each encapsulating a specific GPC recommendation: burn classification, initial assessment, fluid resuscitation, pain management, infection prevention, nutritional support, and transfer criteria, following HL7 FHIR R4 interoperability standards. A mapping between GPC clinical rules (including the Parkland formula, Benaim scale, Curreri formula, and Baux prognostic index) and structured JSON request/response schemas is presented and evaluated against related formalization approaches and verified through structured schema-level invocations against a representative clinical scenario, including a detailed comparison of Mexico IMSS-375 standard properties against HL7 FHIR and OpenEHR. A deployment architecture is described covering hospital-level integration, a centralized service layer, and a non-relational persistence tier based on document-oriented storage for unstructured clinical data.

Background: Burn injuries require accurate assessment for effective management, and artificial intelligence (AI) is gaining attention in burn care for diagnosis, treatment planning, and decision support. This study compares the effectiveness of AI-driven models with experienced plastic surgeons in burn assessment and management. Methods: Ten anonymized burn images of varying severity and anatomical location were selected from publicly available databases. Three AI systems (ChatGPT-4o, Claude, and Kimi AI) analyzed these images, generating clinical descriptions and management plans. Three experienced plastic surgeons reviewed the same images to establish a clinical reference standard and evaluated AI-generated recommendations using a five-point Likert scale for accuracy, relevance, and appropriateness. Statistical analyses, including Cohen’s kappa coefficient, assessed inter-rater reliability and comparative accuracy. Results: AI models showed high diagnostic agreement with clinicians, with ChatGPT-4o achieving the highest Likert ratings. However, treatment recommendations varied in specificity, occasionally lacking individualized considerations. Readability scores indicated that AI-generated outputs were more comprehensible than the traditional medical literature, though some recommendations were overly simplistic. Cohen’s kappa coefficient suggested moderate to high inter-rater agreement among human evaluators. Conclusions: While AI-driven models demonstrate strong diagnostic accuracy and readability, further refinements are needed to improve treatment specificity and personalization. This study highlights AI’s potential as a supplementary tool in burn management while emphasizing the need for clinical oversight to ensure safe and individualized patient care.

… Standards for Provision of Adult and Pediatric Burn Care developed by BBA were selected … Referral and transfer protocols should cover airway management, fluid resuscitation, urgent …

Burns are a significant public health concern, with thousands in the UK requiring treatment annually. Burn assessment and management are complex and require a systematic approach. This study aims to provide an in-depth review of how to evaluate and treat burns to enhance clinical decision-making and ultimately improve patient outcomes. This study explores key aspects of a burn assessment, including key points in the history, examination findings and the classification of the burn depth and total body surface area. It also highlights the Emergency Management of Severe Burn (EMSB) approach and its significance in managing burns, as well as different fluid resuscitation formulas such as Parkland and biological engineering technology (BET). Wound care strategies, indications for surgical and specialist management and additional measurements needed for special burns are also discussed.

… In conclusion, the anesthetic management of a child with … anesthetic management of a burned child is a crucial part of burn … Recent developments in the perioperative fluid management …

… Burn trauma typically induces in children a systemic inflammatory reaction and a robust … of the burn is equal to or greater than 20% of the TBSA and maximize when the size of the burn …

… than usual and burn dressings and … fluid management can be determined based on the patient’s condition, ensuring adequate urine output (0.5 mL/kg/h for adults, 1 mL/kg/h for children, …

OBJECTIVES Guiding fluid therapy in severely burned patients presents complex challenges, combining elements of both distributive and hypovolemic shock, accompanied by cardiovascular dysfunction. Whereas parts of hemodynamic assessment in septic shock such as central venous oxygen saturation (ScvO2) and the arterial-venous carbon dioxide (CO2) partial pressure difference (ΔpCO2 = pcvCO2 - paCO2) can both be used as indicators of hypoperfusion, they are not part of the standard clinical assessment of severely burned patients. The aim of this retrospective study was thus to investigate whether the ΔpCO2 and ScvO2 in severely burned patients correlates with fluid administration and mortality. METHODS Retrospective analysis of severely burned patients with burns larger than 20 % total body surface area (TBSA) admitted between 01/2017 and 06/2021 to the Burns Center of the University Hospital Zurich, Switzerland. Patients were treated according to international guidelines and in-house standards and ΔpCO2 and ScvO2 were assessed at least once within the first 24 h of intensive care unit admission. RESULTS In total 69 patients were included in this analysis. The median ΔpCO2 and ScvO2 were 1.16 kPa [inter-quartile range IQR, 0.82 - 1.50] and 76 % [IQR, 71 - 81]. This translated to an abnormally elevated ΔpCO2 (>0.8 kPa) in 53 (77 %) and an abnormally reduced ScvO2 (<70 %) in 13 (19 %) patients. Initial ΔpCO2, but not ScvO2, was non-linearly associated with intravenous fluid administration in the following 24 h (estimatelog 9.6 L [95 % confidence interval CI, 4.1 - 15.2], p = 0.001) and 48 h (estimatelog 6.0 L [95 % CI, 0.3 - 11.8], p = 0.039). Both initial ΔpCO2 (logarithmic odds ratio ORlog 4.3 [95 % CI, 1.1 - 20.9], p = 0.048) as well as ScvO2 (ORlog 0.1 [95 % CI, 0 - 0.42], p = 0.034) were non-linearly associated with in-hospital mortality. However, the best association with mortality was achieved by combining ΔpCO2 and ScvO2, with patients presenting with abnormalities in both ΔpCO2 and ScvO2 having a predicted mortality of 46 % [95 % CI, 23-71]. CONCLUSION In severely burned patients, both ΔpCO2 and ScvO2 are associated with the extent and severity of burns as well as with in-hospital mortality. However, only ΔpCO2 seems to possess potential as a predictor of fluid administration. A multimodal approach to fluid resuscitation including both parameters may show promise in severely burned patients; however, further prospective studies are required to define optimal thresholds and validate its integration into clinical fluid resuscitation practice.

… In burn resuscitation, they attenuate capillary leak and … , evaluates traditional resuscitation strategies, examines … current evidence, and discusses future directions for resuscitation in …

Introduction: Sepsis remains a serious and life-threatening condition with high morbidity and mortality. It is marked by imbalance between increased production of free radicals and insufficient neutralization by endogenous antioxidants, including vitamin C. This causes a spread of oxidation reactions and contributes to organ failure. Although this imbalance has been proven to be more marked in burn patients, few studies have focused on the intake of vitamin C in this population. Objective: Assess hemodynamic impact of high dose vitamin C (100 mg/weight/day) in septic burn patients. Methods: Case-control study conducted in intensive burn care department in Tunisia during 28 months (July 2019-October 2021). Were included adult burns presenting sepsis or septic shock (as defined in Sepsis 3). Were excluded pregnant woman, patients with history of lithiasis or vesico-renal calculus, of hemochromatosis or G6PD deficiency, and those on long-term vitamin C or glucocorticoids. Only one septic episode was considered per patient during his hospitalization. After inclusion, ascorbic acid was prescribed at a dose of 100 mg/weight/day in 3 doses over 4 days. The vitamin C group was compared with a retrospective group (non-vitamin C) from the same center matched in terms of age, sex, total body surface area (TBSA) and burn severity. Therapeutic management of sepsis was similar for 2 groups in terms of fluid resuscitation with same hemodynamic objectives (hourly output at 0.5cc to 1cc/weight and MAP > 65 mmHg). The primary endpoint was the impact of Vit C on hemodynamic state evaluated by fluid resuscitation, dose, and duration of catecholamines. Results: During study period, 100 patients were included divided into 2 groups: G1 (Vit C+ ; n= 50) and G2 (Vit C- : n= 50). Patients of two groups were comparable in terms of sex, age and severity of burns. Administration of vitamin C reduced fluid balance at day 3 (2 ml/weight/day for G1 vs 13 ml/weight/ day for G2; p= 0,008), reduced significantly doses of noradrenaline at day 3 (1,8 mg/h vs 3,5 mg/h ; p= 0,01), and shortened duration of noradrenaline dependence (4 days for G1 vs 4,84 days for G2 ; p= 0,28) in septic shock burns. No ascorbic acid-related adverse events were identified in the treatment group during the study. Conclusion: High-dose vitamin C therapy was associated with reduced fluid balance, doses of noradrenaline and duration of dependency in burn patients with septic shock during the first 3 days of sepsis.

We introduce MedGraphRAG, a novel graph-based Retrieval-Augmented Generation (RAG) framework designed to enhance LLMs in generating evidence-based medical responses, improving safety and reliability with private medical data. We introduce Triple Graph Construction and U-Retrieval to enhance GraphRAG, enabling holistic insights and evidence-based response generation for medical applications. Specifically, we connect user documents to credible medical sources and integrate Top-down Precise Retrieval with Bottom-up Response Refinement for balanced context awareness and precise indexing. Validated on 9 medical Q&A benchmarks, 2 health fact-checking datasets, and a long-form generation test set, MedGraphRAG outperforms state-of-the-art models while ensuring credible sourcing. Our code is publicly available.

Mental health is a major global concern, yet findings remain fragmented across studies and databases, hindering integrative understanding and clinical translation. To address this gap, we present the Mental Disorders Knowledge Graph (MDKG)—a large-scale, contextualized knowledge graph built using large language models to unify evidence from biomedical literature and curated databases. MDKG comprises over 10 million relations, including nearly 1 million novel associations absent from existing resources. By structurally encoding contextual features such as conditionality, demographic factors, and co-occurring clinical attributes, the graph enables more nuanced interpretation and rapid expert validation, reducing evaluation time by up to 70%. Applied to predictive modeling in the UK Biobank, MDKG-enhanced representations yielded significant gains in predictive performance across multiple mental disorders. As a scalable and semantically enriched resource, MDKG offers a powerful foundation for accelerating psychiatric research and enabling interpretable, data-driven clinical insights. Understanding the pathophysiological pathways of mental disorders and identifying reliable biomarkers remain challenging. This study introduces a large-scale knowledge graph tailored to mental disorders to improve knowledge discovery, disease prediction, and clinical validation

Recurrent spontaneous abortion (RSA), affecting 1%–5% of reproductive‐aged women, presents diagnostic challenges due to its complex multifactorial causes. Conventional guidelines are inadequate for idiopathic cases and emerging biomarkers, whereas artificial intelligence (AI) models struggle to integrate multimodal data. To address these issues, we developed RSA‐KG, a graph‐based AI knowledge graph that synthesizes multimodal clinical data and adapts to evolving guidelines. RSA‐KG integrates 5 international RSA guidelines, utilizing natural language processing (NLP) and multimodal models for data processing. Evaluation of RSA‐KG showed that LLMs enhanced by RSA‐KG outperformed naive retrieval‐augmented generation (RAG) and raw models in diagnostic accuracy. Reproductive specialists also rated the output of the RSA‐KG system more favorably than raw models and medical large language models (LLM). RSA‐KG represents a novel approach to RSA management, overcoming limitations of traditional AI by modeling systemic interactions and integrating real‐time evidence. Further validation through multicenter trials is required for broader clinical adoption.

Background Electronic health records (EHRs) and routine documentation practices play a vital role in patients’ daily care, providing a holistic record of health, diagnoses, and treatment. However, complex and verbose EHR narratives can overwhelm health care providers, increasing the risk of diagnostic inaccuracies. While large language models (LLMs) have showcased their potential in diverse language tasks, their application in health care must prioritize the minimization of diagnostic errors and the prevention of patient harm. Integrating knowledge graphs (KGs) into LLMs offers a promising approach because structured knowledge from KGs could enhance LLMs’ diagnostic reasoning by providing contextually relevant medical information. Objective This study introduces DR.KNOWS (Diagnostic Reasoning Knowledge Graph System), a model that integrates Unified Medical Language System–based KGs with LLMs to improve diagnostic predictions from EHR data by retrieving contextually relevant paths aligned with patient-specific information. Methods DR.KNOWS combines a stack graph isomorphism network for node embedding with an attention-based path ranker to identify and rank knowledge paths relevant to a patient’s clinical context. We evaluated DR.KNOWS on 2 real-world EHR datasets from different geographic locations, comparing its performance to baseline models, including QuickUMLS and standard LLMs (Text-to-Text Transfer Transformer and ChatGPT). To assess diagnostic reasoning quality, we designed and implemented a human evaluation framework grounded in clinical safety metrics. Results DR.KNOWS demonstrated notable improvements over baseline models, showing higher accuracy in extracting diagnostic concepts and enhanced diagnostic prediction metrics. Prompt-based fine-tuning of Text-to-Text Transfer Transformer with DR.KNOWS knowledge paths achieved the highest ROUGE-L (Recall-Oriented Understudy for Gisting Evaluation–Longest Common Subsequence) and concept unique identifier F1-scores, highlighting the benefits of KG integration. Human evaluators found the diagnostic rationales of DR.KNOWS to be aligned strongly with correct clinical reasoning, indicating improved abstraction and reasoning. Recognized limitations include potential biases within the KG data, which we addressed by emphasizing case-specific path selection and proposing future bias-mitigation strategies. Conclusions DR.KNOWS offers a robust approach for enhancing diagnostic accuracy and reasoning by integrating structured KG knowledge into LLM-based clinical workflows. Although further work is required to address KG biases and extend generalizability, DR.KNOWS represents progress toward trustworthy artificial intelligence–driven clinical decision support, with a human evaluation framework focused on diagnostic safety and alignment with clinical standards.

… This is due to challenges such as multiple modalities in patient clinical information and LLM’s … framework maps multimodal patient clinical information to a patient Knowledge Graph (KG) …

Background Sepsis is a complex, life-threatening condition characterized by significant heterogeneity and vast amounts of unstructured data, posing substantial challenges for traditional knowledge graph construction methods. The integration of large language models (LLMs) with real-world data offers a promising avenue to address these challenges and enhance the understanding and management of sepsis. Objective This study aims to develop a comprehensive sepsis knowledge graph by leveraging the capabilities of LLMs, specifically GPT-4.0, in conjunction with multicenter clinical databases. The goal is to improve the understanding of sepsis and provide actionable insights for clinical decision-making. We also established a multicenter sepsis database (MSD) to support this effort. Methods We collected clinical guidelines, public databases, and real-world data from 3 major hospitals in Western China, encompassing 10,544 patients diagnosed with sepsis. Using GPT-4.0, we used advanced prompt engineering techniques for entity recognition and relationship extraction, which facilitated the construction of a nuanced sepsis knowledge graph. Results We established a sepsis database with 10,544 patient records, including 8497 from West China Hospital, 690 from Shangjin Hospital, and 357 from Tianfu Hospital. The sepsis knowledge graph comprises of 1894 nodes and 2021 distinct relationships, encompassing nine entity concepts (diseases, symptoms, biomarkers, imaging examinations, etc) and 8 semantic relationships (complications, recommended medications, laboratory tests, etc). GPT-4.0 demonstrated superior performance in entity recognition and relationship extraction, achieving an F1-score of 76.76 on a sepsis-specific dataset, outperforming other models such as Qwen2 (43.77) and Llama3 (48.39). On the CMeEE dataset, GPT-4.0 achieved an F1-score of 65.42 using few-shot learning, surpassing traditional models such as BERT-CRF (62.11) and Med-BERT (60.66). Building upon this, we compiled a comprehensive sepsis knowledge graph, comprising of 1894 nodes and 2021 distinct relationships. Conclusions This study represents a pioneering effort in using LLMs, particularly GPT-4.0, to construct a comprehensive sepsis knowledge graph. The innovative application of prompt engineering, combined with the integration of multicenter real-world data, has significantly enhanced the efficiency and accuracy of knowledge graph construction. The resulting knowledge graph provides a robust framework for understanding sepsis, supporting clinical decision-making, and facilitating further research. The success of this approach underscores the potential of LLMs in medical research and sets a new benchmark for future studies in sepsis and other complex medical conditions.

Large language models (LLMs) have emerged as important tools in healthcare, showing growing potential for clinical reasoning and patient care. This survey examines recent progress in medical LLMs, focusing on reasoning applications and requirements. We present a dual-view approach that connects clinical practice with computational methods. On the clinical side, we establish a five-level competency scheme following Miller’s Pyramid, progressing from knowledge recall to dynamic case management. On the computational side, we link deductive, inductive, and abductive reasoning patterns to common medical goals and tasks. We also introduce a benchmark dataset spanning five levels of medical reasoning capability and report results on 18 state-of-the-art models, revealing that medical specialist models excel in diagnosis-centric tasks while general models lead in decision support and dialogue. We conclude by discussing current progress and open challenges, including data limitations, hallucination, and grounding issues, and outline directions toward safer, more reliable, and more workflow-ready systems.

Retrieval-Augmented Generation (RAG) improves large language models (LLMs) by retrieving relevant information from external sources and has been widely adopted for text-based tasks. For structured data, such as knowledge graphs, Graph Retrieval-Augmented Generation (GraphRAG) retrieves and aggregates information along graph structures. More recently, GraphRAG has been extended to general text settings by organizing unstructured text into graph representations, showing promise for reasoning and grounding. Despite these advances, existing GraphRAG systems for text data are often tailored to specific tasks, datasets, and system designs, resulting in heterogeneous evaluation protocols. Consequently, a systematic understanding of the relative strengths, limitations, and trade-offs between RAG and GraphRAG on widely used text benchmarks remains limited. In this paper, we present a comprehensive benchmark study comparing RAG and GraphRAG on established text-based tasks, including question answering and query-based summarization. We introduce a unified evaluation protocol that standardizes data preprocessing, retrieval configurations, and generation settings, enabling fair and reproducible comparisons. Our results highlight the distinct strengths of RAG and GraphRAG across different tasks and evaluation perspectives. Building on these findings, we explore selection and integration strategies that combine the strengths of both paradigms, leading to consistent performance improvements. We further analyze failure modes, efficiency trade-offs, and evaluation biases, and highlight key considerations for designing and evaluating retrieval-augmented generation systems. Our code is available at https://github.com/haoyuhan1/RAGvsGraphRAG.

Large language models (LLMs) augmented with knowledge graphs (KGs) offer a promising approach for knowledge-intensive reasoning. Central to this approach is the selection of appropriate reasoning paths in the KG. Yet, existing methods face a common limitation: reasoning path selection is often performed by separate modules using criteria that are only weakly connected to the reasoning requirements. This often results in selecting incorrect relations or premature pruning of relevant paths. We propose Search-on-Graph (SoG), a method that strengthens the connection between path selection and reasoning by having the LLM itself select which relations to follow, informed by both the available KG structure and the complete reasoning history. SoG follows an observe-think-navigate paradigm: at each step, the LLM observes the relational connections available at the current entity, reasons about which path best advances toward answering the question, and navigates accordingly. This context-aware navigation fully exploits the LLM's reasoning capabilities rather than relying on independent selection modules with surrogate criteria. Experiments on six knowledge graph question answering (KGQA) benchmarks demonstrate that SoG outperforms state-of-the-art methods while requiring no task-specific fine-tuning and generalizing across different KG schemas.

… Consensus rate is used for all adults except for very deep burns or small children. Small … occur despite closely monitoring urine output. While typically seen during burn resuscitation, …

BACKGROUND Accurate assessment of intravascular volume status is crucial for managing fluid resuscitation in pediatric burn patients. Conventional methods, such as clinical signs and urine output, may be insufficient. This study aimed to evaluate the effectiveness of transthoracic echocardiography (TTE) in guiding fluid therapy for severely burned children. MATERIALS AND METHODS This retrospective study included 40 pediatric patients with burns covering ≥ 20 % of total body surface area (TBSA), admitted within 24 h of injury to the pediatric burn center between 2019 and 2022. Patients were divided into two groups: TTE group (n = 20) and control group (n = 20). The TTE group received fluid therapy guided by TTE parameters, particularly the inferior vena cava to aorta (IVC/Ao) ratio. The control group was managed using conventional methods. Demographic data, clinical characteristics, fluid management, and outcomes were compared between the groups. RESULTS There were no significant differences in age, gender, TBSA burned, or burn type between groups. In the TTE group, fluid therapy was modified in 80 % of patients based on echocardiographic findings. The TTE group required less diuretic treatment (10 % vs 60 %, p < 0.05) compared to the control group. Although the TTE group had higher rates of initial intubation (55 % vs 20 %, p < 0.05) and inotropic support (35 % vs 5 %, p < 0.05), there was no significant difference in mortality between groups. A negative, moderate linear relationship was found between urine output and descending aorta diameter in the TTE group (correlation coefficient: -0.47, p = 0.04). CONCLUSION The addition of TTE to conventional methods for managing fluid resuscitation in pediatric burn patients may help reduce unnecessary fluid administration and diuretic use. TTE, particularly the IVC/Ao ratio, appears to be a valuable tool for assessing intravascular volume status in this population. While further prospective studies are needed to establish standardized reference values for TTE parameters in pediatric burn patients, this non-invasive method shows promise in improving fluid management and potentially reducing complications associated with over- or under-resuscitation.

INCIDENCE/EPIDEMIOLOGY In the United States, 136,453 children were injured from a fire or burn and treated in emergency rooms in 2012. This number includes more than 67,000 children 4 years of age and younger.1,2 In 2011, 325 children 19 years of age or younger died from fires or burns, 277 (85%) of which occurred in residential fires; 47% of children who died from fires or burns were 4 years of age or younger. The death rate for children this age (0.77 per 100,000) is almost ■ Children have thin skin; therefore their injuries tend to be deep, complicated by contracture and hypertrophic scarring.

Two-thirds of reported burn cases admitted to burn centers affect less than 10% of total body surface area (TBSA) and require a relatively short hospitalization. The majority of patients with major burn injury or significant physiological derangements survive. Improvement in mortality rates is related to advances in resuscitation, infection control, modulation of the hypermetabolic response to traumatic injury and early excision and grafting. Traditional classification of burns of first, second, third and fourth degree have been replaced by a more descriptive explanation of injury depth:

Burns represent one of the leading causes of morbidity and mortality in children. Aim. Aim of this review is to gain better understand of the pathophysiological changes and assessment of the severity of burn injuries in different ages of pediatric patients, which may help in early implementation of appropriate therapeutic procedures and improvement of the outcome of these patients. Children are more likely to develop wider and deeper burns, greater fluid and heat loss in comparison to adults. Therefore, the initial assessment of the TBSA and the depth of the burns in children are crucial for their further treatment. The most important approach in the treatment of children with burn injuries includes the management of airways, effective fluid resuscitation, pain control, and prevention of infection. In the current review we sought to provide recommendations that might help improve the assessment of the severity of burns in children, which may be important for improving their recovery and reducing mortality rate. DOI of the original article: 10.5937/afmnai42-53854 Erratum DOI: 10.5937/afmnai41-62369

… in burn patients despite fluid resuscitation volumes in excess of that recommended by the Parkland formula and despite normal average urine output (… intuitive monitor of renal function. …

Pediatric patients with larger total body surface area (TBSA) burns have a high surface area to volume ratio and are at risk of over-resuscitation. In 2015, our burn center revised our resuscitation algorithm for "difficult to resuscitate" patients with >15% TBSA burn to substitute albumin for a portion of crystalloid volume, hoping to reduce negative effects of volume overload while preserving resuscitation goals. We retrospectively reviewed patients <18 years of age treated between 2008-2024 who required burn resuscitation. Patients who had ≥15% TBSA burn and required >40% of baseline fluids in the first 24 hours were defined as "difficult to resuscitate." Patients who died in <48 hours were excluded. Patient demographics, burn characteristics, treatment details, and patient outcomes were collected. Patients were compared using Fisher's exact and Wilcoxon rank sum tests. Thirty-four patients were "difficult to resuscitate", with 8 patients admitted prior to substitution of albumin and 26 patients treated after. Demographic characteristics did not vary between groups. Patients in the albumin group received less total intravenous fluid volume within 48 hours (12.4 [IQR: 8.6-13.8] vs. 7.8 [IQR 6.3-9.3] mL/kg/TBSA p= 0.037) and had lower serum lactate at 48 hours (1.7 [IQR 1.7-2.2] vs. 1.0 [IQR 0.8-1.3] mmol/L p=0.018). Length-of-hospital-stay normalized to TBSA burn was shorter among those who received albumin (1.2 [IQR:0.8-1.6] vs 1.9 [IQR:1.3-2.4] days p= 0.027). The substitution of albumin for pediatric burn patients who are difficult to resuscitate reduced total intravenous volume and length-of-stay per TBSA, while preserving chemical markers of adequate resuscitation.

… burns received vitamin C 50 mg/kg every 6 h for 96 h after admission [1]. In the VITaCCA trial, resus… ending, this should not obscure the ongoing need for adequate nutritional support. …

Background Burn shock is a life-threatening, capillary-leak-driven hypovolemic state in which standard oral rehydration solution (ORS) fails to address the underlying endothelial dysfunction and oxidative stress. Vitamin C is an essential cofactor for catecholamine synthesis and endothelial barrier stabilization, yet the optimal oral concentration for burn shock resuscitation remains undefined, hindering clinical translation in mass casualty and resource-limited settings where intravenous access is unavailable. Methods Male Wistar rats (n = 350) with 50% total body surface area full-thickness scald burns were randomized to seven groups: sham, burn control, standard WHO-ORS, or one of four vitamin C-fortified ORS groups (1.16, 3.5, 5.98, and 11.6 g/L). Resuscitation was administered via oral gavage following a modified Parkland formula (half-volume regimen to avoid aspiration). Primary outcome was 72-hour survival (n = 20/group). Secondary assessments included intestinal microcirculation (laser speckle contrast imaging), hemodynamic parameters (hematocrit, lactate), oxidative stress markers (malondialdehyde, total antioxidant capacity), systemic inflammation (interleukin-17), and intestinal barrier integrity (serum diamine oxidase activity and histopathology). Results Gastrointestinal absorption of vitamin C was confirmed in the setting of burn shock, with a dose-dependent but nonlinear pharmacokinetic profile suggestive of saturable transport. The 3.5 g/L concentration achieved the highest 72-hour survival (85% versus 65% with standard ORS; P &lt; 0.05). This concentration optimally restored intestinal microcirculatory perfusion, attenuated hemoconcentration (lower hematocrit and hemoglobin; P &lt; 0.05), and accelerated lactate clearance. Mechanistically, the 3.5 g/L group demonstrated superior suppression of malondialdehyde and interleukin-17, with correspondingly enhanced total antioxidant capacity. Intestinal barrier preservation was most pronounced in this group, as evidenced by the lowest serum diamine oxidase activity and histopathological injury scores ( P &lt; 0.001 versus ORS). Conclusion The 3.5 g/L vitamin C concentration represents the optimal oral formulation for burn shock resuscitation in this preclinical model. Its efficacy is mechanistically linked to preserved microvascular perfusion, endothelial barrier stabilization, and attenuation of oxidative-inflammatory injury. This low-cost, field-deployable intervention warrants further clinical investigation as a bridge therapy when intravenous resuscitation is unavailable.

Acute or chronic burn injuries require hormonal responses that significantly influence patient prognosis. Elevated cortisol, catecholamines, and glucagon levels, lead to important metabolic changes, such as hyperglycaemia, insulin resistance, protein catabolism, free fatty acids oxidation, and secondary metabolic acidosis. These alterations impair immune function and wound healing and trigger a systemic inflammatory response. A multidisciplinary approach is needed in order to correctly manage the aforementioned endocrine and metabolic changes. During the acute phase, glucose monitoring, corticosteroid administration for transient or iatrogenic adrenal insufficiency and electrolyte balance maintenance are critical. Chronic phase requires hormonal replacement, nutritional optimization, and anabolic agents administration to counteract catabolic states. Despite continuous advances in burn care, understanding the complex interplay between hormonal changes and immune dysfunction remains challenging. Managing burn-associated endocrine responses could lead to the development of new therapeutic strategies, including personalised and stage- adapted treatment.

Nutrition is a key aspect of growth and development. Nutrition is especially important in times of increased physiological stress such as injury, infection, and surgery. In low- and middle-income countries (LMICs) malnutrition is a significant challenge that may uniquely impact management of surgical patients. In surgical patients, nutrition is important for wound healing, infection resistance, and recovery. Pediatric surgical patients are a special population as they have different nutritional requirements than their adult counterparts. Pediatric surgical patients are also a vulnerable population in LMICs where they have limited access to safe and affordable care. Subpopulations, such as burn, neonatal, and oncology patients, have unique nutritional needs that also impact surgical outcomes. This review explores the intersection of nutrition and infection outcomes in three key pediatric surgical populations in LMICs: burn patients, surgical neonates, and pediatric oncology patients.

BACKGROUND Biodegradable temporising matrix (BTM) is a dermal substitute developed to reconstruct full thickness burns, yet the subsequent metabolic effects are unknown. This pilot study aimed to 1) measure variation in energy expenditure using indirect calorimetry (IC) in response to BTM application across the continuum of acute burns management and 2) assess accuracy of predictive energy equations commonly used in Australia in adult patients with severe burns. METHODS Energy expenditure was measured (MEE) using IC during distinct time-periods: 'acute surgery'; 'BTM integration'; following 'skin grafting'; and 'acute recovery' and compared to predictive equations (Toronto, Schofield plus injury factor (IF), minimum ratio 35kcal/kg/day and maximum ratio 40kcal/kg/day). Agreement was assessed using Lin's concordance correlation coefficient (CCC) and Bland-Altman methods. RESULTS Eighteen patients were included (median [Interquartile range] 44 [29-70] years; 39% [25-56%] total body surface area burns). MEE reported as estimated marginal means (95% confidence intervals) for each time-period were: 'acute surgery' 2048 (1847, 2248) kcal; 'BTM integration' 2244 (2071, 2416) kcal; 'skin grafting' 2297 (2123, 2471) kcal; 'acute recovery' 2102 (1918, 2287) kcal, equating to 25%, 37%, 41% and 29% above predicted basal metabolic rate (Schofield, no injury factor), respectively (all p<0.001). During 'acute surgery' all equations (Schofield x injury factory, minimum and maximum ratio method), overestimated energy requirements by 24-42% (all p<0.001), except Toronto (-12%, p=0.071). Similarly, all equations overestimated energy requirements by 11-27% throughout 'BTM integration' (all p≤0.01), except Toronto (-5%, p=0.12). Following 'skin grafting' Schofield and maximum ratio equations overpredicted, while Toronto underpredicted requirements (+16%, +21% and -11%, respectively, p≤0.001). Maximum ratio overestimated and Toronto underestimated requirements during 'acute recovery' (+19%, p=0.04 and -9%, p=0.014, respectively). Average CCC (all time periods) was highest for Toronto at 0.77, with Bland-Altman plots also showing highest accuracy and reliability. CONCLUSIONS A substantial hypermetabolic response was not observed following BTM application. While the Toronto equation most closely predicted energy requirements considerable variability was observed, highlighting the value of IC to guide nutrition support in severe burns where nutritional needs change over time. A larger multicentre study is required to substantiate the effect of BTM application on energy expenditure.

… We hypothesized that burn patients with inhalation injury exhibit distinct … Nutritional support was progressively initiated within the first 12-24 h, once the initial phase of fluid resuscitation …

Background/Objectives: Severe burn injury is associated with high postoperative morbidity and mortality due to profound metabolic and nutritional stress. The blood glucose-to-serum albumin ratio (GAR) may reflect both metabolic derangement and nutritional status, but its prognostic significance in burn intensive care unit (ICU) patients remains unclear. This study evaluated the association between preoperative GAR and postoperative outcomes in adult burn ICU patients undergoing surgery. Methods: We retrospectively analyzed adult burn ICU patients who underwent surgery between 2014 and 2024. GAR was calculated using blood glucose and serum albumin levels measured within one day before surgery. The primary outcome was 90-day postoperative mortality. Secondary outcomes included 90-day hospital-free days and ICU-free days. Multivariable Cox regression and restricted cubic spline analyses were performed to assess the relationship between GAR and mortality risk. Receiver operating characteristic curve analysis was used to evaluate the discriminatory performance of GAR and determine the optimal cutoff value. Results: Among 1119 patients, the 90-day mortality rate was 25.6%. Higher preoperative GAR was independently associated with increased 90-day mortality in multivariable Cox regression analysis. Restricted cubic spline analysis demonstrated a significant overall association, with progressively increasing mortality risk at higher GAR levels. Receiver operating characteristic curve analysis showed moderate discriminatory performance (area under the curve = 0.789), and the optimal cutoff value based on the highest Youden index was 62.5. Patients with GAR ≥ 62.5 had significantly lower 90-day survival rates and fewer 90-day hospital-free days and ICU-free days than those with lower GAR values (all p < 0.001). Conclusions: Higher preoperative GAR was significantly associated with adverse postoperative outcomes, including increased 90-day mortality and fewer hospital-free and ICU-free days in adult burn ICU patients.

… nutrition risk or determine the adequacy of nutrition support in adult trauma patients, as they reflect inflammation and resuscitation … ) of nutritional support with n-3 PUFA in burn patients …

… The limitations identified in this review highlight several strategic research directions to advance clinical decision support systems that integrate multi-agent reinforcement learning, …

… impacts of multi-agent AI systems in enhancing clinical decision-making processes within … of how multi-agent AI systems can improve clinical decision-making and provide a foundation …

While Large Language Models (LLMs) have demonstrated potential in healthcare, they often struggle with the complex, non-linear reasoning required for accurate clinical diagnosis. Existing methods typically rely on static, linear mappings from symptoms to diagnoses, failing to capture the iterative, hypothesis-driven reasoning inherent in human clinicians. To bridge this gap, we introduce ClinicalAgents, a novel multi-agent framework designed to simulate the cognitive workflow of expert clinicians. Unlike rigid sequential chains, ClinicalAgents employs a dynamic orchestration mechanism modeled as a Monte Carlo Tree Search (MCTS) process. This allows an orchestrator to iteratively generate hypotheses, actively verify evidence, and trigger backtracking when critical information is missing. The foundation of this framework is a Dual-Memory architecture: a mutable working memory that maintains the evolving patient state for context-aware reasoning, and a static experience memory that retrieves clinical guidelines and historical cases via an active feedback loop. Extensive experiments demonstrate that ClinicalAgents achieves the best performance among evaluated baselines, significantly enhancing both diagnostic accuracy and explainability compared to strong single-agent and multi-agent baselines. Our code is released at https://github.com/ZhuohanGe/ClinicalAgents-Code.

In recent years, the advances of large language models and autonomous agents have revolutionized the healthcare field, facilitating diagnosis and improving treatment results. However, most existing AI systems rely on pre-trained knowledge and predefined pipelines, which struggle to learn dynamically from the interactive chat session history that contains patient outcomes and past failures. To address this limitation, we propose VIBEMed, a multi-agent framework with a built-in self-evolution mechanism and architecture-level safety sandbox for robust clinical decision support. The system integrates three specialized agents, including a Clinical Diagnostic Agent (CDA) for hypothesis generation, a Therapeutic Execution Agent (TEA) for treatment planning, and a Clinical Evolution Manager Agent (CEMA) that distills longitudinal clinical feedback into reusable knowledge, transforming multimodal patient information into personalized medical decisions. Through self-evolution mechanism, the framework enables iterative updates across memory, model behavior, and decision strategies, allowing the system to improve over time. Experimental results show that VIBEMed demonstrates superior performance through its evolving mechanism in complex clinical cases, particularly in tasks that require integrated decision-making and longitudinal planning. The framework also supports reliable end-to-end decisions in challenging scenarios such as oncology treatment planning, highlighting its feasibility in real-world clinical contexts. Overall, VIBEMed provides a practical path beyond static AI systems toward adaptive, experience-driven clinical decision support, demonstrating the value of combining multi-agent collaboration with continuous evolution for advancing precision medicine.

Diagnosing diseases through histopathology whole slide images (WSIs) is fundamental in modern pathology but is challenged by the gigapixel scale and complexity of WSIs. Trained histopathologists overcome this challenge by navigating the WSI, looking for relevant patches, taking notes, and compiling them to produce a final holistic diagnostic. Traditional AI approaches, such as multiple instance learning and transformer-based models, fail short of such a holistic, iterative, multi-scale diagnostic procedure, limiting their adoption in the real-world. We introduce PathFinder, a multi-modal, multi-agent framework that emulates the decision-making process of expert pathologists. PathFinder integrates four AI agents—the Triage Agent, Navigation Agent, Description Agent, and Diagnosis Agent—that collaboratively navigate WSIs, gather evidence, and provide comprehensive diagnoses with natural language explanations. The Triage Agent classifies the WSI as benign or risky; if risky, the Navigation and Description Agents iteratively focus on significant regions, generating importance maps and descriptive insights of sampled patches. Finally, the Diagnosis Agent synthesizes the findings to determine the patient's diagnostic classification. Our Experiments show that PathFinder outperforms state-of-the-art methods in skin melanoma diagnosis by 8% while offering inherent explainability through natural language descriptions of diagnostically relevant patches. Qualitative analysis by pathologists shows that the Description Agent's outputs are of high quality and comparable to GPT4o. PathFinder is also the first AI-based system to surpass the average performance of pathologists in this challenging melanoma classification task by 9%, setting a new record for efficient, accurate, and interpretable AI-assisted diagnostics in pathology. Data, code and models are available at https://pathfinder-dx.github.io/.

Clinical decision-making often involves navigating complex data, rapidly changing patient conditions, and the need for precise, context-aware recommendations. Traditional clinical decision support systems (CDSS) provide rule-based assistance but frequently fall short in dynamic environments due to their lack of contextual understanding, adaptability, and real-time responsiveness. In this paper, we introduce an agentic AI framework designed specifically for clinical environments. By integrating autonomous AI agents with medical reasoning capabilities, seamless access to electronic health records (EHRs), and structured inter-agent communication, our system enables real-time diagnosis, triage prioritization, and personalized treatment planning. Using synthetic yet medically realistic patient datasets, we evaluate system performance in terms of diagnostic accuracy, triage precision, interpretability, system latency, and physician satisfaction. The results demonstrate substantial performance gains over traditional systems and lay the groundwork for a new era in clinical AI. The complexity of clinical decision-making continues to grow due to rapidly changing patient data and the demand for timely, context-aware responses. Traditional clinical decision support systems (CDSS) often fall short in dynamic environments. In this paper, we introduce an agentic AI framework that integrates autonomous agents with structured memory and EHR interoperability. The system supports real-time diagnosis, triage, and treatment planning. In comparative evaluations using synthetic patient datasets, our framework achieved 92.4% diagnostic accuracy, 95.2% triage precision, and reduced average response latency to 3.7 seconds—a 40% improvement over rule-based CDSS. The results highlight the transformative potential of agentic AI in augmenting clinical workflows.

… Additionally, to better simulate real-world clinical decision processes, some studies have introduced multi-agent collaboration in sequential decision-making tasks, allowing agents to …

… Multi-agent conflict resolution in clinical settings relies on established decision support frameworks while addressing the unique challenges of healthcare delivery (Shortliffe and …

Although the efficiency of clinical decision support system has been greatly increased by integrating health care analytics technologies, typical AI models remain as black-box with limited interpretability and lacking integration of heterogeneous clinical data sources. We propose a new Hierarchical Agentic Explainable Multimodal Reasoning Network (HAEMR-Net), which is a set of intelligent agents-based AI architecture to achieve both explainability and accuracy with the integration of multimodal Electronic Health Records (EHRs) and medical imaging. In the proposed architecture, four collaborative intelligent agents are developed: (i) Data Harmonization Agent for data preprocessing, (ii) Clinical Knowledge Reasoning Agent for reasoning based on clinical knowledge graphs and context information, (iii) Explainability Agent for visualising the feature-attribution and attention maps via SHAP value and attention mechanisms, and (iv) Decision Validation Agent for verifying model decisions using confidence score estimation and rule-based analysis. We conducted extensive experimental evaluations on a multimodal EHRs dataset containing 15,240 patient records along with corresponding diagnostic images, retrieved from publicly accessible healthcare repositories. The experiments compared HAEMR-Net with existing multimodal models in terms of accuracy, precision, recall, F1-score, Area Under the Receiver Operating Characteristic Curve (AUC) and Explanation Fidelity. The results show that HAEMR-Net achieve promising results by reaching 97.4% accuracy, 96.9% precision, 97.1% recall, 97.0% F1-score, and 98.2% AUC, outperforming other mainstream deep learning and transformer-based multimodal models in major evaluation indicators by 3.8-6.5%. Additionally, the explanation component improves trust of clinical diagnosis via the production of feature-level explanation and region-based justification, with an Explanation Fidelity score of 95.6%. Our agentic model successfully tackles multimodal challenges by seamlessly integrating heterogeneous data sources and enhances explainability of AI-based healthcare system, representing a novel, effective and interpretable solution for next-generation intelligent health care.

Introduction Multi-agent AI systems are believed to bring significant improvements in digital health, but it also brings new and more serious ethical issues. Such systems distribute the decision-making process among multiple interacting agents, and this decentralized decision-making system has raised ethical concerns in the medical field. On the one hand, it continues the ethical issues of traditional AI tools; on the other hand, the interaction processes within complex systems have also brought about new dilemmas. This narrative review aims to synthesize the ethical issues related to multi-agent AI systems in healthcare presented and explore the corresponding mitigation strategies. Methods The study outcomes were synthesized using a narrative approach. Relevant records were gathered through Boolean searches in databases such as PubMed, Scopus, and Web of Science. A total of 21 articles related to multi-agent AI, healthcare, and ethical issues are included in this review. Results Seven key ethical challenges were identified: (1) compound opacity, where interacting AI agents create layers of inscrutable decision-making; (2) error propagation and attribution difficulties, complicating accountability for clinical harm; (3) increased clinician dependence and automation bias, leading to potential deskilling and overreliance; (4) erosion of human oversight, as multi-agent AI systems operate beyond effective human control; (5) privacy and data security risks, stemming from complex data flows among agents; (6) threats to patient autonomy and informed consent, due to opaque or paternalistic AI recommendations; and (7) contextual blindness, reflecting a loss of individualized patient understanding in modular AI workflows. Furthermore, this review also summarized solutions proposed in the existing literature for these ethical issues. Conclusions Multi-agent AI systems intensify existing ethical concerns in healthcare by distributing decision-making and blurring responsibility. To mitigate these issues, recent research advocates for the development of adaptive governance models, clear accountability frameworks, human–AI collaboration structures that preserve clinician authority, enhanced systems for explainability, and privacy-centered designs. In order to successfully incorporate agentic AI into healthcare, it is essential to maintain transparency, protect patient rights, and ensure that human-centered values continue to guide clinical decision-making in an era dominated by autonomous, interacting AI systems.

… decision-making during the exploration phase of multi-agent systems. By leveraging LLMs to assist in decision-making… enhancing clinical data analysis and supporting medical decision-…

Introduction This work presents an AI reasoning framework for burn depth diagnosis that emphasizes transparency and interpretability. The system addresses a critical need in burn care for consistent, explainable assessments, particularly in remote and resource-limited environments where access to burn specialists is limited. Materials and Methods The proposed framework employs a multimodal structural reasoning mechanism that integrates digital photographs with ultrasound imaging, including B-mode and Tissue Doppler Imaging (TDI). These complementary modalities capture both surface features and subsurface tissue dynamics. A chain-of-thought reasoning process links visual and acoustic cues to clinically meaningful indicators of burn severity, providing step-by-step explanations that mirror human diagnostic logic. Results The multimodal reasoning framework demonstrates high diagnostic accuracy in classifying burn depth across three clinically relevant categories. The structured reasoning output provides interpretable explanations for each diagnostic decision, enabling medics to validate the model’s conclusions. Experimental findings show that the system demonstrates strong performance relative to retrospective human evaluation. Conclusions This multimodal AI reasoning framework offers a practical, deployable solution for modern burn assessment. Its transparent, step-wise reasoning enhances clinical trust, improves decision confidence, and supports care delivery in settings lacking expert supervision. The approach establishes a foundation for explainable, high-performance AI in both military and civilian burn care.

Accurate assessment of burn depth and total body surface area (TBSA) is critical for clinical decision-making; however, it remains subjective and prone to interobserver variability. Multimodal large language models (MLLMs) are increasingly encountered in clinical contexts, but whether these systems can reliably assess burn images remains unclear. We evaluated four MLLMs (GPT-5.4 Pro, Grok 4.1, Gemini 3.1 Pro, and Claude Opus 4.6) on 50 clinical burn photographs using a repeated-inference design with five independent runs per model. Burn depth classification was assessed in numeric and text-based formats, alongside ordinal TBSA estimation. Performance varied across the models, with burn depth accuracy ranging from 34.0 ± 6.5% to 76.4 ± 6.8% and TBSA accuracy from 32.8 ± 9.4% to 68.4 ± 3.3%. Inter-run reliability (Fleiss' κ) ranged from slight (κ = 0.171) to almost perfect (κ = 0.916), demonstrating response variability not captured by single-query evaluations. Notably, no model combined high accuracy and high reliability, indicating a dissociation between performance and consistency. All models showed a tendency toward overestimation of burn depth, including assignment of fourth-degree burns despite their absence in the dataset. Error direction analysis revealed model-specific and task-dependent biases, including opposing patterns within the same model. Internal consistency between numeric and text classifications was near-perfect (99.6-100%), indicating format-invariant but systematically biased outputs. These findings demonstrate that MLLM performance is characterized by stochastic response instability invisible to single-query evaluations. Such inconsistency for identical inputs represents a fundamental limitation for workflows requiring consistent outputs across repeated evaluations.

INTRODUCTION Burn patients are a group highly prone to sepsis and bloodstream infections (BSIs) due to immune dysregulation, skin barrier loss, and complex inflammatory responses. Traditional diagnostic criteria for sepsis are often unreliable in this population. While conventional regression models and laboratory biomarkers have been widely studied, they have limitations. Machine learning (ML) techniques have been used for this purpose. However, no systematic review has yet summarized these applications specifically for burn care. This review aims to synthesize and critically appraise current evidence on ML models for predicting sepsis and BSI in burn patients. METHODS We systematically searched three databases: PubMed, Scopus, and Web of Science. Eligible studies included original research applying modern ML algorithms to predict sepsis or BSI in burn patients. Two independent reviewers screened and extracted data. Risk of bias was assessed using the PROBAST+AI tool. RESULTS Following the search and screening process, seven studies were selected for final inclusion, comprising four studies focused on developing predictive models for sepsis and three aimed at predicting BSI in burn patients. One out of seven studies used multi-center data. Included studies had sample sizes ranging from 82 to 302. The data types were heterogeneous. The majority of the studies used clinical and laboratory data. Notably, only one study performed temporal validation beyond internal resampling. CONCLUSION Despite the promising advancements identified in this review, the field of using AI techniques to predict sepsis and bloodstream infections in burn population remains underexplored. The limited number of studies, coupled with small sample sizes, indicates that this field is still an emerging area of research.

Total Burned Surface Area (%TBSA) estimation is essential in guiding effective burns resuscitation. Traditional methods of %TBSA estimation remain limited by poor inter-rater reliability and operator-dependent accuracy. Several digital applications have been developed to assist with %TBSA estimation; however, there is a paucity of evidence directly comparing their accuracy, inter-rater reliability, and clinical utility. We aimed to synthesise the performance metrics of digital tools for %TBSA estimation as compared with traditional methods and established reference standards, to assess their suitability for clinical integration. A PRISMA-guided search of Embase and MEDLINE (January 1, 1990-August 20, 2023) was conducted to identify studies reporting accuracy or inter-rater reliability for computer- or smartphone-based digital %TBSA estimation tools. 20 studies (2006-2023) evaluating 16 unique digital tools met the inclusion criteria. Against an established standard, digital tools demonstrated superior accuracy over traditional methods (mean error: digital -5.47% to +4%; traditional -0.47% to +19.7%), with EasyTBSA achieving the closest agreement to the established reference (-0.01 ± 3.59%) compared with Lund and Browder (4.42 ± 5.52%), Rule of Palms (3.92 ± 10.71%), and Rule of Nines (5.05 ± 6.87%). Digital tools demonstrated higher inter-rater reliability (ICC: digital 0.986-0.998; traditional 0.886-0.910). Time-to-estimation varied across digital platforms without a consistent advantage over traditional methods. While direct cross-application comparisons were precluded due to study heterogeneity, these findings suggest that digital tools for %TBSA estimation offer measurable improvements in accuracy and reliability over conventional methods, warranting consideration for broader clinical integration as digital health technologies continue to advance.

… When we resuscitate an adolescent with traumatic burns from a motor vehicle crash in a resource-intensive level-one trauma emergency department in the mountains of Utah, we see …

BACKGROUND Fluid resuscitation is a critical treatment in the shock phase for burn patients, but excessive fluid infusion has adverse effects. METHODS We conducted a multicenter retrospective cohort study involving 12 burn centers across mainland China. Adult patients (age >18 years) with severe burns were included. The primary exposure was the 24-hour fluid volume (ml/kg/%TBSA). The primary outcome was 28-day mortality. Cox proportional hazards models were used to evaluate the association between fluid volume and mortality. Secondary outcomes included changes in coagulation parameters (prothrombin time, activated partial thromboplastin time [APTT], INR, D-dimer, fibrinogen, thrombin time, and platelet count) and incidence of coagulopathy. Mediation analysis was performed to assess whether the effect of fluid rate on mortality was mediated through alterations in coagulation markers. RESULTS A total of 475 patients were included in the initial cohort for mortality analysis, and 278 patients with complete coagulation data were analyzed in the secondary cohort for mediation. After adjusting for multiple confounders, higher fluid volume was significantly associated with increased 28-day mortality (adjusted HR = 1.74, 95% CI: 1.30-2.32, P < 0.001). Elevated fluid volume was also associated with prolonged APTT, reduced platelet count, and higher incidence of coagulopathy. Mediation analysis revealed that the effect of fluid rate on mortality was partially mediated by decline in platelet count (P = 0.002), whereas other coagulation parameters did not show significant mediating effects. CONCLUSIONS Higher initial fluid volume not only has a direct effect on coagulation but also leads to an increased risk of 28-day mortality. The difference in platelet count statistically mediates the effect of fluid volume on mortality outcomes.

… The variability observed in burn assessment, resuscitation, and early wound care in this study is therefore better understood as a system-level gap rather than a series of isolated deficits…

Health and racial disparities can limit access to preventative, trauma, and chronic disease care but have not been addressed in burn resuscitation. Over- and under-resuscitation contribute to increased overall hospital costs, and morbidity and mortality rates. The primary objective of this study was to identify potential racial disparities that may exist during the initial fluid resuscitation after burn injury. This was a retrospective review of all burn patients > 14 years of age admitted between January 1, 2020 and December 31, 2022 to a county safety net hospital. Patients were excluded if they transitioned to comfort care within 24 hours of admission. Data collected included baseline demographics, relevant burn injury information, and laboratory parameters. Outcomes included hospital and ICU length of stay, duration of mechanical ventilation, payor status, and mortality. Patients were divided into white (59 %) vs. African American-Hispanic (AA-HIS) (41 %) and included 105 patients. The median age (IQR) was 44.5(30) for whites vs 34(36) for AA-HIS. There were no statistically significant differences in severity of burn injury, cause of burn injury, rates of inhalation injury, or ICU or hospital lengths of stay. In both groups 55 % of the patients required mechanical ventilation while 18 % required renal replacement therapy. Overall mortality was not higher in the AA-HIS group at 32.6 % vs 17.7 % (p = 0.081). There were no differences in amount of fluid administered, urine output, laboratory values during resuscitation, or patient outcomes between the groups. The use of protocols for burn resuscitation can be instrumental in protecting against racial and ethnic disparities.

In addition to aiding decision making, AI-based clinical decision-support systems may need to consider and support provider agency and collaboration between medical providers. We analyzed collaboration and agency within fast-response teams, identifying implications for designing decision-support systems that not only facilitate decision making, but also collaboration and agency. Using an Actor-Network Theory approach, we reviewed videos of 12 pediatric trauma resuscitations and conducted a secondary analysis of 27 interviews with trauma team members. We identified actants in trauma resuscitation, shifts in agency that can occur within fast-response teams during medical emergencies, and factors considered by providers when envisioning the design of decision-support systems. From our analysis, we propose implications for existing human-AI interaction guidelines when designing AI systems for fast-response medical teams. We also highlight parallels between the introduction of clinical practice guidelines and the introduction of AI-based decision-support systems, suggesting that these systems may influence the training and ''clinical gaze'' of providers.

Large-scale combat operations (LSCO) will challenge the current U.S. military trauma system with high casualty volumes, prolonged evacuation timelines, and degraded logistics. The Joint Trauma System demonstrated the effectiveness of data-driven medical performance optimization during the Wars in Iraq and Afghanistan, reducing battlefield mortality. However, the current system relies on delayed, manual documentation processes that are inadequate for the operational tempo and complexity of future conflicts. This article advocates for the development of a real-time, automated combat casualty care data ecosystem that supports decision-making, resource allocation, and command and control across echelons. Current modernization efforts, including digital tools such as the Battlefield-assisted Trauma Distributed Observation Kit (BATDOK) and integration platforms like the Operational Medicine Data Service (OMDS) and the System for Injury Monitoring and Outcomes Nexus (SIMON), may improve data capture but are still heavily dependent on human input. We propose a future system centered on passive data collection, scalable edge computing, artificial intelligence-enabled triage and decision support, and seamless integration with tactical networks and operational planning tools. This system must distinguish between data needed for real-time care and that required for archival documentation of injuries and care provided. It must also enable both clinical optimization and trauma system learning without adding a burden to providers. Investments in real-time data infrastructure, machine learning, and automated sensing are necessary to maximize survivability in LSCO. Without this transformation, the military trauma system risks delays in care, degradation in outcomes, and reduced operational effectiveness. Real-time data are essential for modern combat casualty care and future mission success.

Background/Objectives: Surgical debridement and early excision of burned areas followed by skin autograft is the gold standard of treatment for partial and full-thickness pediatric burns. However, skin autografting might be unfeasible or unlikely to succeed due limited availability of skin donor areas or inadequate conditions. In these circumstances, alternative treatment is required, and Skin Substitutes (SS) cold play a role. Recently, Biological Skin Substitutes (BSS) and Tissue Engineered Skin Substitutes (TESS) are emerging as alternative treatment options, but strong evidence is missing. This review investigates the current literature focusing on BSS and TESS, aiming to improve the medical and surgical management of pediatric patients. Methods: A systematic review was performed in accordance with the PRISMA 2020 guidelines and registered in the PROSPERO database (CRD42024627569). A comprehensive search was conducted in PubMed (MEDLINE) from 2000 to 2024 using Boolean logic and PICO-based inclusion criteria. Study quality was assessed using the Joanna Briggs Institute (JBI) critical appraisal checklists according to study design. Results: Twenty-nine articles and 2676 pediatric patients undergoing surgical reconstruction by BSS or TESS for burns were included. The methodological quality was generally moderate, with most studies being observational or case series. Several strategies were critically analyzed and possibly discussed. Conclusions: While BSS and TESS are safe and effective reconstructive options, the overall level of evidence remains low to moderate. A schematic classification of SS for pediatric burns is presented. Further prospective trials are needed to define standardized algorithms for pediatric burn reconstruction.

Background Burns are a significant cause of accidental injuries in children worldwide. This study investigated the epidemiology and burden of pediatric burns to identify the associated factors with these injuries and to formulate prevention strategies in underdeveloped minority areas in Guangxi, China. Methods A retrospective study was conducted on 660 pediatric burns admitted to two tertiary hospitals in Baise, Guangxi. Demographic data, burn characteristics, clinical factors, and burden information were analyzed. Multiple linear regression was performed to identify the associated factors associated with the length of hospital stay (LOHS) and costs. Results 660 pediatric burn cases were recorded from January 2014 to December2020. The median age of pediatric burn patients was 2 years. Winter is the peak period of pediatric burns. Scalding was the most frequent cause followed by flame burns. The median total body surface area (TBSA) affected being 6%. The median length of hospital stay (LOHS) was 9 days. The median cost was 7,558 CNY. The major associated factors for LOHS were the response rate [standardized beta coefficient (SBC) = 0.391], TBSA (SBC = 0.357), operations (SBC = 0.350). The major associated factors for cost were LOHS (SBC = 0.468), TBSA (SBC = 0.306), operations (SBC = 0.215), response rate (SBC = 0.120) and region (SBC = 0.081). The p value of all associated factors were p < 0.05. Conclusion Scalds and flame burns are key targets for prevention among children under 3 years in underdeveloped minority areas of Guangxi, China. Winter is identified as the peak season for pediatric burns. The burden of pediatric burns is associated with multiple associated factors; these findings provide a foundation for assessing the epidemiology and burden of pediatric burns in this areas. Given the central role of caregivers in daily child supervision, addressing their knowledge, caregiving practices, and environmental awareness—alongside consideration of the broader contextual factors—can help foster a safer and more sustainable caregiving environment for children.

Burns are challenging injuries requiring effective wound management, and while silver nanoparticle dressings show promise in promoting healing, there is a lack of clinical research on their use in pediatric burn patients. Thus, we aimed to investigate their effects on accelerating wound healing in children. We performed a retrospective study on 60 pediatric burn patients. The patients were grouped into control (30 cases, conventional wound management) and observation (30 cases, silver nanoparticle dressings + conventional management) groups. The clinical efficacy, wound healing time, impediments, Vancouver Scar Scale (VSS) score, scar hyperplasia degree, and satisfaction of patients with different burn sites and degrees were analyzed. Wound healing time and treatment effective rate of patients with different burn sites and degrees in the observation group was significantly higher than control (P < .05). The complication incidence, VSS scores for scar color, vascularity, pliability and thickness, and the proportion of severe hyperplasia were lower in the observation group, while the proportion of mild hyperplasia was higher (P < .05). Family satisfaction was also higher than control (P < .05). Correlation analysis revealed that shorter healing times were associated with improved scar outcomes, such that each 1-day reduction in healing time was linked to lower VSS thickness scores and a decreased likelihood of severe hyperplasia. This retrospective study suggests that silver nanoparticle dressings may be associated with faster wound healing, fewer complications, improved cosmetic outcomes, and higher family satisfaction in pediatric burn patients. Prospective randomized studies are warranted to confirm these findings.

Abstract Introduction Pediatric burn injuries are associated with a range of physical, social, and psychological consequences, including disruption to school adjustment. However, little is known about how smaller burns may interfere with school functioning. The current study evaluates the impact of acute burn injury on school functioning and relationships with demographic and burn-related factors in youth receiving outpatient care in a pediatric burn clinic. Methods Data were extracted from retrospective chart review as part of IRB-approved study at a children’s hospital with a pediatric burn center. Participants (N = 102) were school-aged burn patients (M age=9.9 years, 60% female) who attended an outpatient burn clinic < 31 days post-injury and completed the Children’s Dermatology Life Quality Index (CDLQI), which assesses impairments across several quality of life (QOL) domains (e.g., physical discomfort, negative emotions, sleep disturbance, interpersonal problems). The school interference item (scored 0-3) was used to measure potential negative impact on a child’s functioning at school. Spearman’s rho correlations and between-group analyses also included burn total body surface area (TBSA), age, gender, and ethnicity. Results Mild to severe levels of school interference were reported by 60% of participants (22.4% mild, 9.2% moderate, 28.6% severe), while 40% reported having no school interference. School interference was correlated with TBSA (ρ =.26, p <.05), age (ρ =.27, p <.05), difficulties coping with burn treatment (ρ =.25, p <.001, and interference with hobbies/playing (ρ =.44, p <.001). School interference was rated similarly by patients with an admission for their burn injury (M =.92, SD =.99) and patients without a hospital admission (M =.98, SD =.88; t(99)=.22, p =.83). Although youth with full thickness burns reported higher levels of total QoL impairments (M = 13.40, SD = 7.02) compared to those with partial thickness burns (M = 8.38, SD = 5.14, t(99)= -2.06, p =.04), there were no group differences on the school interference item in this small sample. Conclusions In this small retrospective study, larger burn TBSA and older child age were associated with worse school interference, which in turn was correlated with quality of life impairments (e.g., disruption to playing/hobbies, difficulties coping with burn treatment). This pattern did not vary by ethnic minority status, burn depth severity, or burn-related hospitalization, suggesting that the acute phase of burn injuries can disrupt school for many children. Applicability of Research to Practice Understanding the specific effects of burns during childhood is crucial for researchers and practitioners in tailoring interventions and supporting these children during their transition back to daily activities, including school. Funding for the Study N/A

… prevention efforts, especially in children aged less than 5 years, and to improve first-aid training. We identified deficiencies in pain assessment and management, as well as significant …

BACKGROUND Superficial second-degree burns pose major clinical challenges, including extended healing time duration, infection risk, and patient discomfort. Although advanced synthetic dressings such as self-adhesive silicone foam (Mepilex®XT) improve outcomes, novel interventions require evaluation. This study compares efficacy and safety between a novel subvacuum dressing and Mepilex®XT in managing superficial second-degree burns in children. The aim of this study was to investigate whether the novel subvacuum dressing for superficial second-degree burns is non-inferior to Mepilex®XT. METHODS This prospective, randomized, controlled multicenter non-inferiority trial enrolled 82 participants aged 2-14 years with superficial second-degree burns. Patients were randomly assigned to the experimental (novel subvacuum dressing) or control (Mepilex®XT) groups. A total of 78 participants completed the study, comprising 41 in the experimental group and 37 in the control group. The primary outcome was time to complete wound closure via standardized assessment. Secondary outcomes included infection rates and adverse reactions. Baseline comparability was assessed for patient demographics, burn size, and anatomical location. RESULTS No significant intergroup differences were observed in baseline characteristics (all P > 0.05). Time to complete wound closure was comparable between groups (P > 0.05). Safety profiles were similar, with no significant differences in adverse event frequency or severity (P > 0.05). Experimental dressing and standard care results demonstrated no significant differences across all metrics. CONCLUSIONS The novel subvacuum dressing demonstrated comparable efficacy and safety to Mepilex®XT, proving beneficial in promoting healing of superficial second-degree burns in pediatric patients.

Superficial partial thickness burns typically receive nonoperative treatment, whereas deep partial thickness burns, which are prone to hypertrophic scarring, are usually managed through debridement followed by autologous split-thickness skin grafting. Various therapies have been developed to prevent wound infection and to enhance wound healing in pediatric partial-thickness burns. However, the choice of dressing by the surgeon can be influenced by various factors. It is worth noting that there is no standardized approach across all burn centers, leading to variations in care practices. To optimize pediatric patient care, a systematic review was conducted following PRISMA guidelines to review existing treatment options for partial thickness burns in children. Outcomes of interest were wound healing time, dressing changes, length of hospital stay, wound infections, need for grafting despite treatment, and hypertrophic scarring. A total of 68 studies with 8199 patients were included. The mean age of the included patients was 3.1 years, and the mean total body surface area of the burns was 15.6 %. Treatment groups included topical agents, bandages, skin analogues, or unclassified. Considering all treatment outcomes evaluated in this systematic review of the literature, non-silver dressings and skin analogues may have some benefit over topical agents in terms of wound healing time, length of hospital stay, hypertrophic scarring, pain management, and cost saving. Dressing changes, wound infections, and need for grafting did not significantly change between various treatments.

Burn injuries in children often impair joint mobility and motor function, with concomitant pain often compromising rehabilitation compliance. Active video games (AVGs) may offer a motivating, fear-reducing alternative, yet their effects on burn-related outcomes remain understudied. This study aimed to evaluate AVGs’ impact on pain intensity, range of motion (ROM), and motor function in pediatric burn patients. This two-phase study was conducted in 2024 at Imam Reza Hospital, Mashhad, Iran. First, an active video game was developed; then, a single-blind randomized controlled trial was conducted with 69 children aged 5–15 years with second-degree upper limb burns. Participants were block-randomized (1:1) using Stata 17. The intervention group received Kinect Xbox-based game therapy; the control group used traditional physiotherapy. Each session included a 15-minute warm-up followed by gameplay/ physiotherapy, held 10 times per week for two weeks. Pain, range of motion (ROM), and motor function were assessed using the Wong-Baker FACES® scale, goniometry, and the Activities Scale for Kids (ASK), respectively. Data were analyzed using chi-square, t-test, Mann-Whitney U, repeated measures ANOVA, and Friedman tests in SPSS 28 (p < 0.05). The mean age was 7.94 ± 2.22 years in the intervention group and 8.29 ± 2.29 years in the control group (p = 0.890). The findings showed a significant reduction in pain intensity after the intervention in both groups compared to pre-intervention levels (p < 0.001, 95% CI: 0.87–1.30). Moreover, the increase in wrist joint range of motion after the intervention was significantly different between the two groups (p < 0.001, 95% CI: 2.61–10.62). However, no significant difference was observed between the groups regarding shoulder and elbow joint range of motion (p = 0.490, 95% CI: -39.91–26.86; p = 0.060, 95% CI: -1.70–19.14, respectively). Additionally, no significant difference was found in motor function between the two groups after the intervention (p = 0.058, 95% CI: -0.16–9.68). Our results demonstrated that active video game intervention significantly reduced pain intensity and improved wrist range of motion in children with burns. These findings suggest that active video games could serve as a complementary approach for wrist rehabilitation in pediatric burn patients. Future studies should validate these effects on other joints (shoulder/elbow) and functional outcomes. This RCT was registered in the Iran Registry of Clinical Trials (IRCT20230723058891N1, https//irct.behdasht.gov.ir/trial/72575) on 11/04/2023.

Background: Burns in children represent a significant public health issue, as there is no single targeted dressing for the treatment of burn wounds in children. The alloplastic epidermal skin substitute is the dressing of choice for treating burns in children in our burn center. However, it sometimes occurs that the dressing separates from the wound too early, before the process of full re-epithelialization. The inflammatory phase of wound healing seems to be crucial for maintaining the adhesion of the dressing, and thus, changes in parameters such as leukocyte levels and protein changes are of clinical significance. The aim of our study is to find laboratory factors that could contribute to premature dressing separation. Methods: The documentation of 182 children treated for acute burns at a major Polish burn center in the years 2009–2023 was analyzed. A demographic analysis was performed to collect information. The group was split into the following two categories based on the condition of the dressing: “attached to the wound” and “detached from the wound”. Laboratory tests were collected on admission and with control tests 3–5 days after injury. Results: The results indicate that only a few of the parameters studied showed a statistically significant difference between the groups of patients in whom the dressing did or did not attach. The most pronounced relationship was found for the pre-treatment leukocyte level (leuk1). Statistical significance was also demonstrated for hemoglobin levels and changes in protein (protein_diff) and also glucose levels (glucose_diff). Conclusions: Our study shows that there are blood parameters (leukocyte, protein, and glucose levels) that influence the adhesion of the dressing. Unfortunately, there are no other studies on this topic in the literature, so it seems very important to expand research in this direction.

Tension-free closure of a clean wound enables good healing and optimal scarring. Tense wounds are more likely to dehisce and lead to scar stretching, producing a poor cosmetic result.

Abstract Conservative management for hypertrophic scars (HSc) and scar contractures is of utmost importance to optimally reintegrate burn survivors into society. Many conservative treatment interventions have been described in the literature for the management of HSc. Recent advancements in the literature pertaining to postburn scarring and HSc formation have advanced our understanding of the mechanisms that support or refute the use of common rehabilitation treatment modalities after burn injury. This is particularly relevant for recent advancements in the fields of mechanotransduction and neurogenic inflammation, resulting in the need for rehabilitation clinicians to reflect upon commonly employed treatment interventions. The aim of this review article is to summarize and clinically apply the evidence that supports or refutes the use of common conservative treatment interventions for scar management employed after burn injury. The following treatments are discussed, and mechanotransduction and neurogenic inflammation concepts are highlighted: (1) edema management (compression, positioning/elevation, pumping exercises, retrograde massage, and manual edema mobilization); (2) pressure therapy (including custom fabricated pressure garments, inserts, face masks, and other low-load long-duration orthotic devices); (3) gels or gel sheets; (4) combined pressure therapy and gels; (5) serial casting; (6) scar massage; and (7) passive stretching. This review supports the following statements: (1) Compression for edema reduction should be initiated 48–72 hours postinjury and continued for wounds that require longer than 21 days to heal until scar maturation; (2) Elevation, pumping exercises, and retrograde massage/MEM should be used in combination with other edema management techniques; (3) Custom-fabricated pressure garments should be applied once the edema is stabilized and adequate healing has occurred. Garments should be monitored on a regular basis to ensure that optional pressure, >15 mm Hg, is maintained, adding inserts when necessary. The wearing time should be >16 hours/day; (4) Gels for postburn scar management should extend beyond the scar; (5) Serial casting should be applied when contractures interfere with function; (6) Forceful scar massage should be avoided early in the wound healing process or when the scar is inflamed or breaks down; and (7) Other treatment modalities should be prioritized over passive stretching for scar management.

… and thoracoabdominal constrictive burns. Discussion The available evidence suggests that … selected circumferential or constrictive burns at risk of burn-induced compartment syndrome. …

Effective management of burn mass casualty incidents (BMCIs) requires coordinated systems that include early stabilization, definitive burn care, surge capacity, and rehabilitation, supported by adequate infrastructure and workforce capacity. Although different countries have national guidelines, the full synthesis of the components of the health system needed by BMCIs response remains limited. This study aimed to systematically identify and classify the key components of the response to BMCIs. A systematic literature review was conducted to identify and synthesize the key components required for effective response to BMCIs. A systematic review was conducted in accordance with PRISMA guidelines and registered in PROSPERO (CRD42024625737). This research was conducted through a systematic review and searches of the databases PubMed, Scopus, Web of Science, and Google Scholar. A Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) was used for systematic search, and a Mixed Methods Appraisal Tool (MMAT) was used for quality assessment of final extracted articles. From an initial set of 7,975 articles, 27 studies were included for analysis. The key components for responding to BMCIs were analyzed in 5 categories and 36 subcategories. The primary themes and related main categories included: (1) Preparedness and Prevention; (2) Command, Control, Coordination, and Communication; (3) Prehospital and Transport Management; (4) Hospital and Surge Capacity Management; and (5) Clinical and Supportive Care. This review provides a comprehensive, evidence-informed framework outlining essential health system components for effective BMCIs response. The identified domains may guide the development of national and regional response models, strengthen disaster preparedness, and enhance health system resilience to BMCIs scenarios.

ABSTRACT Objective To present evidence‐ and consensus‐based guidelines for prehospital first aid in dogs and cats. Design Prioritized clinical questions pertaining to first aid posed in the Population–Intervention–Comparator–Outcome (PICO) format were used to inform systematic literature searches by information specialists, which facilitated extraction of research findings from relevant publications and these findings’ synthesis into evidence. This evidence was assessed for quality and used to draft treatment recommendations. These steps were followed by an international commenting period for veterinary professionals prior to finalization. These 2026 RECOVER First Aid Guidelines summarize treatment recommendations enabling veterinary professionals to provide clear, actionable instructions to animal owners, handlers, and paraprofessionals in the prehospital setting. Setting Transdisciplinary, international collaboration in university, specialty, emergency, and general practice. Results A total of 10 PICO questions pertaining to first aid in dogs and cats were addressed in the main portion of this project. This resulted in 38 treatment recommendations covering a variety of conditions for which timely aid in the prehospital setting may improve morbidity and mortality outcomes in dogs and cats. Conditions included heat stroke, choking, burns, noncaustic toxin ingestion, hypoglycemia, and severe extremity bleeding, among others. One main and four condition‐specific RECOVER First Aid Recognize and Act Algorithms display the flow of assessments and actions bystanders may use to help distressed animals outside the clinical setting. Six additional instructional figures can be used to teach small animal owners, handlers, and paraprofessionals to provide first aid. Conclusions These 2026 RECOVER First Aid Guidelines present a concise and comprehensive framework for first aid provision for dogs and cats in the prehospital setting. These works serve to support veterinary professionals, educational systems, and research initiatives in conducting, implementing, and advancing the provision of first aid in dogs and cats.

ABSTRACT Objective To systematically review the evidence on, to devise clinical recommendations for, and to identify critical knowledge gaps regarding the provision of first aid in dogs and cats. Design Standardized, systematic evaluation of literature pertinent to first aid following Grading of Recommendations, Assessment, Development, and Evaluation methodology. Identified literature regarding prioritized questions in the Population–Intervention–Comparator–Outcome (PICO) format was reviewed independently by Evidence Evaluators, and findings were reconciled by Domain Chairs and Reassessment Campaign on Veterinary Resuscitation Co‐Chairs to arrive at treatment recommendations commensurate with the quality of evidence and feasibility. This process was implemented using an evidence profile worksheet for each PICO question, including an introduction, consensus on science, treatment recommendations, justification of recommendations, and knowledge gaps. Treatment recommendations underwent a 4‐week comment period among international veterinary stakeholders prior to finalization. Setting Transdisciplinary, international collaboration in university, specialty, emergency, and general practice. Results Sixteen PICO questions pertaining to first aid were addressed in this project. We separated 10 general first aid scenario PICO questions from those pertaining to allergy and anaphylaxis, reported elsewhere. These main first aid PICO questions resulted in 38 treatment recommendations regarding a variety of conditions for which prehospital care may improve morbidity and mortality in dogs and cats. Many recommendations are supported by low or very low quality of evidence (15) or based on expert opinion (18). Conclusions Uncertainty remains regarding optimal ways for bystanders to recognize specific conditions and provide first aid in the prehospital setting to dogs and cats. However, through comprehensive evidence evaluation and a consensus process that included considerations of safety and feasibility, the resulting treatment recommendations lay the foundation for clear, actionable guidance in small animal first aid. Additionally, a list of high‐priority knowledge gaps was identified to guide collaborative clinical research to overcome a significant deficit in veterinary literature.

Autoimmune diseases pose a persistent challenge to clinical decision-making due to their heterogeneous presentations, episodic flare-ups, and complex pathophysiological mechanisms. Existing deep learning models frequently rely on correlation-driven patterns, which often lack robustness and interpretability in real-world clinical settings. This study proposes a novel causal deep learning framework designed to model and predict the progression of autoimmune diseases using longitudinal multimodal clinical data. The proposed approach integrates a temporal transformer architecture with a structural causal model (SCM)-guided counterfactual inference module to capture latent disease mechanisms while mitigating spurious correlations introduced by observational confounders. The model is trained and evaluated on a real-world cohort consisting of patients diagnosed with systemic lupus erythematosus (SLE) and rheumatoid arthritis (RA), incorporating electronic health records (EHRs), laboratory biomarkers, and immunological profiles. Empirical results demonstrate that the proposed framework outperforms conventional deep learning baselines in predicting disease transitions and flare-up events. In addition, interventional and counterfactual evaluation protocols are employed to validate the model’s causal consistency and interpretability. This work contributes to the advancement of clinically trustworthy AI systems by enabling robust, explainable, and data-efficient disease trajectory modeling in the autoimmune domain.

Maintaining regulatory compliance while detecting ever more complex fraud patterns via conventional rules-based systems presents unmatched difficulties for the financial services sector. The incorporation of understandable artificial intelligence approaches with hybrid architectures integrating knowledge graphs and natural language processing to automate compliance and fraud detection in banking is discussed in this article. Machine learning models show superior performance to conventional detection methods, but their black-box character goes against transparency and explainability regulations. Using transformer-based language models and heterogeneous graph neural networks, the hybrid design extracts semantic patterns from textual transaction data while encoding domain knowledge via structured knowledge representations. Using SHAP and attentional mechanisms, human-interpretable explanations that satisfy legislative obligations can be created while keeping identification accuracy. Regulatory compliance frameworks, including the GDPR and Basel Committee guidelines, provide openness requirements, yet execution issues with regard to clarity, specificity, adversarial robustness, and computational overhead persist. Deploying reliable artificial intelligence systems for financial compliance calls for balancing the conflicting needs of stakeholder trust, traceability performance, and operational efficiency by means of well-thought-out governance systems and multi-modal explainability strategies.

Severe burn injuries result in a massive systemic inflammatory and hypermetabolic response, often disrupting multiple organ systems, including the cardiovascular system. Cardiac troponins are frequently elevated in burn patients, but the interpretation of these elevations is hardly straightforward. Sometimes, elevated troponin means type 1 myocardial infarction (MI) due to an acute coronary event, but it might also reflect type 2 MI due to an imbalance between body oxygen demand and supply, or even non-ischemic myocardial injury from things like overwhelming inflammation, sepsis, or direct thermal effects on the heart itself. This narrative review explores the prevalence, underlying pathophysiological mechanisms, diagnostic challenges, and prognostic implications of troponin elevation in burn patients. We discussed the limitations of applying conventional MI diagnostic criteria in this unique population. The role of electrocardiographic and echocardiographic assessment, as well as the potential utility of high-sensitivity troponin assays, is also discussed. Understanding of how burn pathophysiology relates to myocardial injury is essential for accurate diagnosis, improved management, and better outcomes in this highly vulnerable group of patients.

The aim of this study is to share our experience regarding the diagnosis and therapeutic management of primary cutaneous aspergillosis (PCA) in the burn patient, an uncommon infection associated with increased mortality, morbidity, and treatment costs. The uniqueness of this article is the presence of PCA in pediatric patients where the Meek micrografting technique was used. We performed a retrospective study from June 2020 to November 2024. The inclusion criteria were the concomitant presence of burn injuries and confirmed PCA. We identified six patients, aged between 12 and 17 years, admitted with deep burns ranging from 55% to 90% of the total body surface area (TBSA). They required complex ICU treatment and underwent extensive excision–grafting surgeries. The suspicion of infection was raised by changes in the appearance of wounds. Systemic and topical antifungal treatment was established in patients after a PCA diagnosis. Five out of the six cases had a favorable outcome. The use of the Meek micrografting technique in burn treatment represents a new challenge in the treatment of PCA due to the polyamide gauze that covers the micrografts. Early detection and appropriate topical antifungal agents combined with systemic treatment may save the infected grafts and limit the infection spread without necessarily removing the polyamide gauze.

The Systemic Immune-Inflammation Index is a measurement of the systemic immune-inflammatory response (SII), which is used as both a diagnostic and predictive index for many diseases. Burns are a major public health problem among children. Infection caused by burns is the most important cause of mortality in children. In this study aims to investigate the predictive and diagnostic performance of SII for infection for pediatrics at the burn center and the causes of burns and responsible microorganisms and possible risk factors on infection. Data were collected retrospectively from 42 pediatric patients between 2013 and 2023 and analyzed in the burn center. Infected and uninfected burn patients were compared. Scalds were the most common cause of burns in both groups, (91.3%; 87%, respectively). The most frequently isolated microorganism was Pseudomonas aeruginosa (52,6%). Central venous catheter use was the biggest risk factor for infection (OR = 8,077; 95% CI 1,523 to 42,834). The AUC value demonstrated an acceptable diagnostic performance (AUC = 0,605; 95% CI 0,450 to 0,746) Similarly, the odds ratio suggested a potential relationship between SII and infection (OR = 2,057; 95% CI 0,489 to 8,657), but both failed to reach statistical significance. The results of this investigation indicate limited predictive and diagnostic utility for SII. CRP performed better diagnostically than SII (AUC = 0,877; 95% CI 0,747 to 0,955), suggesting that traditional inflammatory markers may still be a better way to predict infection in pediatric burns. Moreover, substantial disparities in hemoglobin levels, lymphocyte counts, CRP, and procalcitonin between infected and uninfected groups indicate that a multi marker strategy may prove more efficacious than dependence on a solitary index. While the SII showed a tendency to predict infection in pediatric burn patients, it did not achieve statistical significance in our research. These findings highlight the need for larger-scale studies to clarify the role of SII in infection prediction among pediatric burn patients. Further research with larger cohorts or multicenter studies could help determine whether SII has clinical utility in this population. Also, accurate identification of infectious agents, development of effective treatment strategies, avoidance of prophylactic antibiotic use, and strict adherence to isolation precautions will significantly reduce the risk of infection in centers where burn patients are followed up.

Boric acid (BA) is found in various household products. Paediatric intoxication via unusual routes remains a concern. This case report describes an 8-year-old boy who ingested BA through contaminated pasta water, with documented blood and urine boron levels. An 8-year-old boy consumed homemade pasta prepared with water inadvertently contaminated with a 3% BA solution (estimated intake: 2.7–8.1 g). He presented with nausea and abdominal pain. On admission, routine blood and urine analyses revealed normal creatinine and urea levels. He was admitted for observation, intravenous fluid therapy, and monitoring of boron levels. The patient experienced one episode of vomiting and headache, which resolved with acetaminophen. Serial blood and urine boron levels were measured. The patient’s initial plasma boron concentration was 16.5 mg/L, decreasing significantly within 24 h, consistent with BA’s pharmacokinetics. Despite the estimated BA intake, the patient remained largely asymptomatic after the initial episode. Follow-up tests one-month post-discharge were normal. This case highlights the potential for accidental BA ingestion in children and emphasizes the importance of prompt recognition and supportive treatment with intravenous fluids. While blood boron levels can confirm exposure and elimination, correlation with clinical manifestations requires further study. Ingestion of BA doses exceeding 3 g warrants immediate hyperhydration and monitoring. BA is readily absorbed, with renal excretion being the primary route of elimination. Prompt hydration is crucial. A one-month follow-up with clinical evaluation and renal function tests appears sufficient for asymptomatic patients. Further research is needed to better define the toxic dose of BA and its clinical correlation.

Background: Multidrug-resistant organism (MDRO) infections pose a growing threat to pediatric burn patients, contributing to prolonged hospitalization, increased health care burden, and limited therapeutic options. Despite their clinical significance, predictive tools specific to MDRO infections in this population remain scarce. The authors aimed to identify independent clinical predictors of MDRO infection among pediatric burn patients with health care–associated infections (HAIs) and to construct a validated nomogram for individualized risk estimation. Methods: A retrospective cohort study was conducted at a tertiary-care pediatric burn center in China, including patients 1 month to 18 years of age who developed HAIs between January of 2020 and December of 2023. Multivariable logistic regression was used to identify independent risk factors, and a nomogram was constructed accordingly. Model performance was evaluated using the area under the receiver operating characteristic curve, calibration plot, decision curve analysis, and clinical impact curve. Results: Among 763 pediatric burn patients with HAIs, 93 (12.2%) developed MDRO infections. Six independent predictors were identified: total burn surface area, hospital length of stay, intensive care unit admission, mechanical ventilation, central venous catheter use, and acute inhalation injury. The nomogram demonstrated excellent discrimination (area under the receiver operating characteristic curve 0.962), good calibration, and favorable clinical utility across a range of threshold probabilities. Conclusions: This study presents a clinically applicable nomogram for early prediction of MDRO infection in pediatric burn patients with HAIs. The tool may support individualized risk stratification, guide targeted preventive interventions, and inform empirical antibiotic decision-making in burn care settings. Prospective validation is warranted.

Acute kidney injury (AKI) is prevalent among children with severe malaria, contributing to considerable morbidity and mortality. Oxidative stress has been implicated in the pathophysiology of malaria-induced AKI, and paracetamol, with its antioxidant properties, has been proposed as a solution. This phase I/II randomized trial evaluated paracetamol as a potential renoprotective adjunct in children with severe malaria and acute kidney injury. We conducted a phase I/II open label parallel randomized controlled trial of 40 hospitalized children aged > 6 months to < 12 years with malaria-induced AKI in eastern Uganda. Participants were randomized by a sealed envelope 1:1 to receive either oral paracetamol 20 mg/kg 6 hourly for 48 h or tepid sponging every 30 min until fever subsided. Only the assessors of the primary outcome were masked to the intervention. The primary outcome was renal recovery at 48 h assessed using restricted mean survival time (RMST) in intention to treat population of children according to their randomization groups. Between 19 September 2021 and 25 August 2023, 250 children with hemoglobinuric severe malaria were screened and the 40 enrolled were randomly assigned paracetamol (n = 20) or tepid sponging (n = 20). The mean age was 6.54 (2.61) years. The mean time to renal recovery in the paracetamol group was 0.491 h (95% CI, −9.265 to 10.248; p = 0.921) longer than the control group within 48 h, but this difference was not statistically significant even after adjusting for age and weight: 1.04 h (95% CI, −8.61 to 10.70; p = 0.832). The safety assessment indicated no significant differences in adverse events and hepatotoxicity in either group. Although paracetamol was safe, it did not significantly improve renal recovery in children with malaria-induced AKI. Further larger studies are needed to explore this role of paracetamol. ISRCTN84974248. Registered on August 12, 2020.

Background Arginine growth hormone (GH) testing is a commonly used and generally safe pediatric procedure for assessing GH deficiency. The occurrence of gross hematuria following this testing is a rare adverse drug reaction that had not been previously encountered at our institution, raising significant concerns among both patients and clinicians. Case presentation An 11-year-and-3-month-old boy presented with short stature. Approximately 7 h after undergoing a combined arginine and levodopa GH stimulation testing, he developed painful gross hematuria accompanied by blood clots. A subsequent comprehensive clinical evaluation ruled out other potential etiologies, leading to the diagnosis of drug-induced hematuria. The patient was treated with urine alkalinization, and the gross hematuria substantially resolved within 8 days. Furthermore, a review of the literature identified 9 case reports involving 15 patients who developed hematuria following arginine GH stimulation testing. The majority of these patients were male, with a broad age distribution. Most individuals presented solely with a change in urine color; hematuria typically manifested within 1–3 days post-administration and resolved spontaneously within approximately 1 week. Standard management strategies primarily include urine alkalinization and fluid supplementation. Current studies hypothesize that the underlying mechanism may involve drug-induced alterations in the permeability of the glomerular filtration membrane or a triggered immune response. Conclusions Hematuria induced by GH stimulation testing is a rare but noteworthy adverse drug reaction. Therefore, prior to initiating the test, clinicians must thoroughly inform patients and their families of this potential risk to alleviate unnecessary anxiety, and closely monitor for the possible onset of hematuria post-testing.

Objectives: This study aimed to develop a population pharmacokinetic model for vancomycin in pediatric patients with sepsis or septic shock and to propose tailored initial dose recommendations based on the model’s findings. Methods: We conducted a retrospective analysis of pediatric patients with sepsis or septic shock who received vancomycin treatment from January 2017 to December 2019. A population pharmacokinetic model was developed using NONMEM. Monte Carlo Simulations were performed using Crystal Ball software to determine the probability of target attainment (PTA) for different vancomycin dosage regimens based on the target PK/PD index of AUC24hr/MIC &gt; 400 Results: Analyzing 580 vancomycin concentrations from 228 patients (median age 4.1 years; 33.8% with septic shock), we determined that a one-compartment model best described the pharmacokinetics, estimating clearance (CL) at 1.8 L/hr and volume of distribution (Vd) at 15.8 L. Body weight (BW) significantly influenced both CL and Vd, while serum creatinine (SCr) and blood urea nitrogen (BUN) affected only CL. The optimal doses adjusted by SCr and BW to achieved PTA range from 60 mg/kg/day (SCr 0.1–0.3 mg/dL) decreasing to 20–30 mg/kg/day (SCr 0.7–1.0 mg/dL) when the MIC ≤ 0.5 mg/L. For MIC = 1 mg/L, higher doses up to 90 mg/kg/day are needed in patients with low SCr. A loading dose of 25–35 mg/kg was necessary to reach PTA within 24 hours. No regimen achieved target PTA at MIC &gt; 1 mg/L. Conclusions: Customized vancomycin dosing based on BW and renal function may improve target AUC attainment for S. aureus in sepsis or septic shock children. Caution is warranted for MIC &gt;1 mg/L or SCr &gt;1 mg/dL due to limited data.

BACKGROUND Paediatric burn injuries pose a critical global health challenge, affecting millions of children annually. Beyond the immediate physical harm, burns disrupt developmental trajectories, leaving enduring psychological, physical, and relational impacts as children transition into adulthood. While advancements in acute care have improved survival rates, understanding the dyadic recovery process between parent and children remains underexplored. The prolonged and complex nature of post-burn recovery necessitates sustained caregiver support, emphasising the need to examine how parent-child dyads experience and collectively navigate this journey. OBJECTIVE To generate a theory explaining how parent-child dyads experience and navigate through the post-burn recovery process. DESIGN Constructivist grounded theory. PARTICIPANTS Paediatric burn survivors with a discharge status of up to 1 month or more and their informal caregivers who were present throughout the period of hospitalisation. METHODS Initial purposive and subsequent theoretical sampling approaches were employed. Data were collected through face-to-face dyadic interviews at mutually agreed time and locations, supplemented by field notes. Concurrent data analysis utilised constant comparative approach to iteratively refine emerging codes and categories. To ensure congruence with the dyadic interviewing approach, the constant comparative approach employed did not only focus on uncovering categories, but also the content (emerging narratives) and interaction (how parents and children communicated noting areas of convergence and divergence in their narratives). RESULTS Twenty-three (23) parent-child dyads participated in the study. The children and parents (19 mothers and 4 fathers) are aged 10-16 years and 33-49 years respectively. The core category, Coming to Terms, conceptualises the dyadic recovery journey from injury occurrence to acceptance and adaptation. Three interrelated categories characterised this process: (1) Being with Each Other in a Trauma Bubble (mutual emotional entrapment in acute distress), (2) Living with the Scars (physical and psychological marks shaping identity), and (3) Navigating the Recovery Maze (collaborative adaptation to post-burn realities). The Theory of Coming to Terms underscores post-burn recovery as an emotionally fraught negotiation marked by asymmetrical burdens for dyads, requiring reconciliation with lasting physical, psychological, and relational consequences. Although most aspects of the recovery journey showed commonalities across dyads, a notable divergence emerged in attitudes toward scars: while parents and older children tended to conceal these marks, younger children often chose to display them openly. CONCLUSION The findings highlight the dyadic interconnectedness in post-burn recovery, advocating for rehabilitative frameworks that prioritise both objective clinical outcomes and subjective experiences. A family-centered approach is critical to integrate ongoing support for children and caregivers.

… Clinical decision support: Diagnostic agents collaborate to suggest treatment … , making manual coordination both error-prone and resource-intensive. The multi-agent decision support …

… less familiar with burns, to guide resuscitation and ongoing decision making. A checklist can add structure for these teams, highlight differing approaches and support even experienced …

Background Vancomycin is used to treat severe infections caused by methicillin-resistant gram-positive bacteria such as methicillin-resistant Staphylococcus aureus (MRSA). Although several pharmacokinetic studies on vancomycin in burn patients have been reported, a robust population pharmacokinetic model based on routine therapeutic drug monitoring data from a large cohort is still lacking. This study aimed to develop a PopPK model for vancomycin in adult burn patients and evaluate its predictive performance. Methods Serum vancomycin concentrations from 93 burn patients were analyzed using non-linear mixed-effects modeling. Covariates including age, gender, creatinine clearance (CrCL), BMI, burn cause, and burn area were evaluated using stepwise forward selection and backward elimination. Model stability was assessed by bootstrap. Predictive performance was compared with Clincalc in 9 external burn patients. Results A one-compartment model best described vancomycin PK in burn patients. The final model identified CrCL as a significant covariate on clearance: CL (L/h) = 6.73 × (CrCL/111.67)0.708, with a volume of distribution of 49.1 L. Bootstrap analysis confirmed good model stability. Our model demonstrated superior predictive accuracy compared with Clincalc. Conclusion Vancomycin clearance is increased in burn patients and significantly correlated with CrCL. The developed PopPK model provides a reliable basis for individualized vancomycin dosing in this specific population.

Abstract Introduction Healthcare-acquired pressure injuries (HAPI) present a significant challenge in the care of burn patients, adding complexity and increasing morbidity. Within the setting of already inherently compromised skin integrity, factors such as prolonged immobility, critical illness, and utilization of medical devices have been shown to increase the incidence of HAPI. Nursing-driven protocols for patient repositioning and invasive device monitoring as well as attentiveness to thorough documentation may help mitigate risk. This quality improvement study investigates the impact of initiation of new nursing-driven protocols on the prevalence, characteristics, and clinical outcomes of HAPI in patients admitted to a high-volume regional Burn Center. Methods Concern for elevated HAPI rates, underutilization of Wound, Ostomy, and Continence Nursing (WOCN) consultation services, and inconsistent documentation spurred the introduction in September, 2023 of the Turn Your Burn program as well as in-service training on Heel Medics boots. Patients with HAPI were identified from the Burn Center registry for 8 months pre-implementation and 6 months post-implementation. Retrospective chart review collected data on patient demographics, burn injury characteristics, patient acuity, HAPI characteristics and timing of diagnosis, utilization of WOCN consultation, and associated clinical outcomes. Descriptive statistics were reported, and univariate analysis compared pre- and post-implementation groups. Results Throughout the 14-month study period (1/2023-3/2024), 45 HAPI were identified in 23 patients admitted to the Burn Unit. Patients were predominantly male (78.3%) with a mean age of 54.4 ± 20.1 years. Most were critically ill at the time of HAPI identification (78.3%), with mean Total Body Surface Area % (TBSA) 34.7% ± 24.1%. Heels and other locations on the feet accounted for 55.6% of cases, while the sacrum was involved in 13.3%. Medical devices were implicated in 26.7%. No HAPI was present at the time of admission, and the median time of diagnosis was hospital day 22.0 (IQR 38.0). Surgical intervention was required in 8.9%. Rates of WOCN involvement significantly increased following protocol implementation (68.4% vs. 15.4%, p < 0.001). Documentation of staging as well as addition of the injury to the Lines-Drains-Airways avatar (LDA) has been steadily increasing over time. Conclusions HAPI remains prevalent in burn patients. Implementing targeted interventions, such as Turn Your Burn, in collaboration with Wound, Ostomy, and Continence Nurses (WOCN), can help mitigate HAPI risk in this vulnerable patient population through increased awareness and improved documentation. Applicability of Research to Practice Nursing-driven quality improvement efforts through protocol development, implementation, and optimization allow for improved patient care as well as staff engagement. Funding for the Study N/A

… accurate weights for objective functions in the decision-making process [37]. Finally, the development of dedicated decision support software based on the proposed integrated model …

… and validation of an AI-driven decision support architecture aimed at improving cardiac arrest … Reporting guideline for the early stage clinical evaluation of decision support systems …