Search Results (1,322)

Background: Increasing demand for emergency medical services (EMS) challenges prehospital resource availability, particularly for time-critical emergencies requiring advanced life support (ALS). Although tiered BLS and ALS response models are established in many non-physician-led EMS systems, evidence regarding qualification-based dispatch stratification within physician-led EMS systems remains limited. This study evaluated the clinical and operational characteristics of an expanded BLS dispatch segment in Berlin EMS. Methods: A retrospective observational study analysed 2,132,246 EMS missions in Berlin, Germany, between 2020 and 2024. Missions were retrospectively classified according to dispatch codes included in the expanded BLS segment, designed to allocate incidents with lower expected requirements for ALS-level interventions to appropriately qualified EMS personnel. Analyses included dispatch characteristics, clinical findings from electronic patient care records, observed safety-related indicators, and ALS response intervals. Results: Overall, 28.7% of EMS missions were classified within the expanded BLS segment. Traumatic and psychiatric presentations represented the most frequent diagnostic groups. Based on predefined clinical indicators, no immediately life-threatening condition was documented in more than 95% of missions. Cardiopulmonary resuscitation occurred in 0.03% of cases, and emergency physician involvement was documented in approximately 3% of cases. Conclusions: A substantial proportion of EMS missions represented a population with predominantly lower expected prehospital treatment complexity within a qualification-based dispatch framework. Structured emergency call interrogation combined with dispatch classification and linked clinical data enabled retrospective evaluation of this approach. Further validation against independent clinical reference standards and linkage with downstream outcomes are required to determine broader applicability. Full article

Background: Medical artificial intelligence (AI), machine learning (ML), and deep learning (DL) studies frequently begin with datasets collected for routine care rather than for computational modeling. Such datasets may contain inconsistent variables, heterogeneous measurement time points, unexplained NaN values, poorly defined outcomes, missing metadata, and insufficient documentation, which can compromise model development before any algorithm is selected. Methods: This Technical Note proposes a physician-facing Clinical AI-Readiness Guide for preparing medical datasets before AI-based analysis. The guide was developed as a practical framework organized around pre-modeling decisions, including the clinical task, cohort, minimum common dataset, outcome definition, predictor variables, measurement timing, missing-data logic, standardization, non-tabular data linkage, data dictionary, and validation readiness. Results: The proposed guide translates AI-readiness principles into concrete data-collection rules for clinical, laboratory, imaging, physiological-signal, textual, follow-up, and multimodal data. It emphasizes clinically consistent data acquisition, reliable target labeling, explicit missing-data logic, patient-level linkage, structured metadata, and validation feasibility. A structured checklist and scoring approach are also proposed as practical pre-modeling assessment tools to classify datasets as not ready, exploratory only, ML-ready with limitations, or AI-ready for model development. Conclusions: Medical AI-readiness should be established before model development begins. By helping physicians collect, structure, and document data more consistently, the proposed guide may improve collaboration between clinical and technical teams and reduce preventable dataset-related failures in medical AI research. Full article

Background/Objectives: Proactively informing the patient prior to initiating medication and implementing preventive strategies are essential for mitigating the risk of developing medication-related osteonecrosis of the jaw (MRONJ). This study evaluated the awareness of MRONJ among a group of dental faculty patients who were scheduled to use, were using, or had previously utilized bisphosphonate and/or denosumab. Methods: A structured questionnaire-based survey was used as the survey tool to evaluate the patients’ awareness of MRONJ. A non-probability convenience sampling method was employed throughout a one-year period. Results: Throughout the study period, 202 participants were included, with a response rate of 97.58%. Only 35 (17.3%) of the patients had awareness of the risk of MRONJ, and 23 (11.4%) reported that their prescribing physicians had informed them about the potential side effects of their medication before the commencement of treatment. Univariate logistic regression, which was performed to elucidate the elements affecting awareness, revealed a significant odds ratio for healthcare professionals as a source of knowledge (29.229, p < 0.001), information of side effects given by a prescribing physician (25.343, p < 0.001), dental consultation (2.592, p = 0.018), and regular dental visits (3.205, p = 0.004). In the multivariate logistic regression analysis, “Healthcare professional as source of knowledge” was significantly associated with awareness (OR = 21.956, p < 0.001). “Information of side effects given by a prescribing physician” was also a significant predictor (OR = 11.450, p < 0.001), indicating that participants informed by prescribing physicians about the side effects were more likely to have awareness. Conclusions: This study revealed that despite patients’ limited awareness of MRONJ, the dissemination of information by healthcare professionals positively influenced their awareness and understanding of the risk factors associated with MRONJ. In particular, receiving information from healthcare professionals and prescribing physician-provided information about side effects appeared to be the strongest predictors of awareness. Full article

Gout is one of the most common inflammatory arthritides worldwide, and its burden is rising in Southeast Asia alongside rapid demographic and socioeconomic transitions. This region, characterized by diverse healthcare systems, cultural practices, and levels of economic development, presents a unique context in which the epidemiology and management of gout are evolving. Despite the availability of effective urate-lowering therapy (ULT) and well-established treat-to-target (T2T) strategies, real-world outcomes remain suboptimal. Barriers to optimal gout care operate at multiple levels. Physician-related factors include clinical inertia and limited awareness of evidence-based recommendations. Patient-related factors include limited understanding of gout and its treatment, compounded by socioeconomic disadvantages. System-level challenges include inequitable access to healthcare services, limited health insurance coverage, and resource-strained health systems. The current evidence base in Southeast Asia remains limited and heterogeneous, raising uncertainty regarding the applicability of global data to local populations. Nevertheless, emerging studies suggest that multidisciplinary, context-oriented approaches can improve gout outcomes. The objective of this review is to explore the evolving epidemiological landscape and barriers to gout management in Southeast Asia. Full article

Background/Objectives: Pelvic floor dysfunctions (PFDs) are highly prevalent chronic conditions associated with substantial functional and psychosocial burden. Although multidisciplinary pelvic floor rehabilitation is recommended as first-line management, its effectiveness depends not only on intervention efficacy but also on care delivery, communication, patient engagement, and long-term adherence. This study applied the Delphi-Dialogue Patients–Professionals (DDPP) framework to quantify patient–professional alignment and translate identified patient–professional perceptual differences into candidate service-improvement strategies. Methods: A two-phase observational study incorporating parallel stakeholder surveys and a modified Real-Time Delphi feasibility-assessment process was conducted under the auspices of the Spanish Society of Physical Medicine and Rehabilitation (SERMEF). In Phase 1, parallel online questionnaires were completed by 225 patients with PFD and 164 rehabilitation professionals. Shared items were rated on a six-point Likert scale, and patient–professional divergence was quantified using an operational composite index integrating standardized differences in mean ratings and high-agreement responses. In Phase 2, the implementation feasibility of 10 improvement recommendations was evaluated by a Delphi panel comprising 35 PM&R physicians and two representatives of the Spanish Incontinence Association (ASIA). Results: Patient–professional agreement was domain-specific. The greatest divergences concerned digital competence, timing of patient education, bladder and bowel diary use, and adherence to prescribed rehabilitation programmes. Professionals attributed greater value to bladder and bowel diaries and selected group-based interventions, whereas patients reported higher digital competence, more timely information delivery, and better adherence. Comparatively similar response patterns were observed regarding hybrid rehabilitation models, educational resources, app-supported follow-up, and overall satisfaction with rehabilitation care. The Delphi process identified structured home exercise programmes, adherence-monitoring strategies, and improved visibility of rehabilitation pathways as the recommendations with the highest perceived implementation feasibility. Conclusions: The findings suggest that important challenges in pelvic floor rehabilitation may arise from implementation-related factors in addition to the effectiveness of available interventions. The DDPP framework provides a structured, stakeholder-informed methodology for translating patient–professional divergence into candidate organizational recommendations with comparatively higher perceived implementation feasibility. Full article

Background: Procedural sedation and analgesia (PSA) is commonly used for shoulder dislocation reduction in emergency departments (EDs), but it requires monitoring resources and may be associated with sedation-related adverse events. Ultrasound-guided regional anesthesia (UGRA), particularly low-volume superior trunk (ST) block, has emerged as a potential alternative. However, little is known about how this technically demanding technique can be implemented and adopted in routine ED practice. This study aimed to describe the implementation and clinical adoption of low-volume ultrasound-guided ST block for shoulder reduction in a specialized ED over a six-year period. Methods: This retrospective single-center observational cohort study included all consecutive patients undergoing shoulder reduction in a specialized orthopedic ED between February 2018 and February 2024. Patients were managed according to routine clinical practice using either UGRA or PSA, with treatment choice determined by the treating physician. The primary objective was to describe the implementation and clinical adoption of low-volume ST block in routine practice. Additional analyses assessed temporal trends in technique utilization, local anesthetic volume, provider distribution, and ED length of stay. Results: A total of 206 patients were included (124 UGRA; 82 PSA). The use of UGRA increased progressively during the study period, whereas PSA decreased accordingly (p < 0.001). Most UGRA procedures were performed by a small group of experienced physicians, reflecting the gradual adoption of the technique in routine clinical practice. Local anesthetic volumes decreased significantly over time (p < 0.001), with most blocks ultimately performed using approximately 4–5 mL. Successful shoulder reduction was achieved in all patients. No conversion from UGRA to PSA was necessary. No clinically relevant complications such as respiratory or hemodynamic distress were observed in either group, although diaphragmatic function was not systematically assessed. Conclusions: This retrospective implementation analysis demonstrates that low-volume ultrasound-guided ST block was increasingly integrated into routine ED practice. The findings primarily describe the implementation and progressive adoption of a technically demanding regional anesthesia technique in a real-world setting. Prospective studies are warranted to further evaluate clinical effectiveness, patient-reported outcomes, and respiratory effects using standardized study protocols. Full article

Background: Human papillomavirus (HPV) is associated with multiple head and neck cancers, many of which are preventable with the 9-valent HPV vaccine. We developed an interactive HPV vaccination education program (IHVEP) to equip patients with HPV-associated cancers to serve as health educators. Objective: To pilot test the adapted IHVEP, evaluate its content using the Information-Motivation-Behavioral Skills (IMB) model and the Suitability Assessment of Materials (SAM) instrument, and identify factors relevant to optimizing digital health interventions. Methods: A previously developed cervical cancer–focused IHVEP was adapted for patients with oropharyngeal cancer. Participants completed the program and participated in semi-structured interviews. Feedback was analyzed and organized into themes aligned with the IMB and SAM frameworks. Results: Participants generally perceived the adapted IHVEP as informative, motivating, and empowering. Key motivators included physician recommendation and a desire to protect others. The program appeared to enhance participants’ perceived ability and self-efficacy to promote HPV vaccination. Modules were described as clear, well-organized, linguistically accessible, and culturally appropriate. Feedback identified areas for improvement related to technology, content, and functionality. Conclusions: Findings suggest that effective digital health interventions may benefit from incorporating training in evaluating health information, ensuring intuitive design and navigation, and including content featuring real individuals. These results will inform further refinement of IHVEP and the development of future digital health interventions. Full article

Background/Objectives: Medication safety is a critical concern in older adults due to age-related vulnerability, multimorbidity, and polypharmacy. Pro Re Nata (PRN, “as needed”) prescribing is common in geriatric care but may be interpreted inconsistently across healthcare professionals, potentially increasing the risk of medication errors. This study aimed to evaluate PRN prescribing patterns in hospitalized older adults and assess interprofessional perceptions of the clarity and appropriateness of PRN prescribing for conditions. Methods: We conducted a prospective, single-day observational study in a French geriatric university hospital, including all patients aged ≥65 years with at least one medication prescription. PRN prescriptions were extracted from electronic medical records and independently assessed by a pharmacist, nurse, and physician for clarity and appropriateness. Descriptive statistics summarized PRN use, chi-square tests assessed interprofessional differences, and multivariate logistic regression identified predictors of appropriate PRN wording. Results: Among 316 patients, 1035 PRN prescriptions were analyzed, representing 24.5% of all medication entries. Only 51.3% (n = 531) were deemed appropriate by all evaluators. Significant differences in interpretation were observed across professional groups (p < 0.001 for most pairwise comparisons). PRN prescriptions were most frequent in long-term care units and primarily involved gastrointestinal agents, analgesics, and psychotropic medications. Multivariate analysis showed that hospitalization in long-term care (OR = 4.17; 95% CI 2.48–7.03) or rehabilitation units (OR = 2.07; 95% CI 1.22–3.53) with a higher number of prescriptions administered under a therapeutic protocol PRN (OR = 2.27; 95% CI 1.39–3.63) were independently associated with appropriate wording, while a higher total number of PRN prescriptions reduced appropriateness (OR = 0.87; 95% CI 0.82–0.94). Conclusions: PRN prescribing is frequent in hospitalized older adults and often involves high-risk or potentially inappropriate medications. Substantial interprofessional variability in the interpretation of PRN medication wording requires appropriate, standardized, and clinical PRN guidelines to improve medication safety and reduce potential adverse events in geriatric practice. Full article

Professional and elite sport represents a high-pressure healthcare environment in which standard principles of patient-centered care, confidentiality, informed consent, shared decision-making, and clinical independence are tested by organizational and performance-related pressures. This narrative review examines reliance management in athlete–physician relationships. Reliance is conceptually distinguished from trust: whereas trust is an interpersonal attitude that can be betrayed, reliance denotes the broader dependence of the athlete-patient on the physician, the care process, and the surrounding governance system—a dependence that may persist even where trust is absent. Reliance management is conceptualized as the deliberate, multi-level design and maintenance of communicative, ethical, and governance conditions under which such reliance is well-placed. A structured, non-systematic literature search of PubMed, Scopus, Web of Science, SPORTDiscus, and Google Scholar, supplemented by backward citation tracking, informed a thematic synthesis organized around three dimensions: clinical ethics, patient-centered care, and governance. The synthesis distinguishes empirical findings, professional consensus guidance, ethical and legal analysis, and illustrative cases, and identifies domains in which evidence remains weak. An integrative conceptual model is proposed that links governance antecedents at the dyadic, clinical, organizational, and system levels to mechanisms of well-placed reliance and to outcomes that require empirical evaluation. Proposed safeguards—independent medical oversight, documented return-to-play pathways, formal data-sharing rules, second-opinion access, and contractual protection of physician autonomy—are intended to protect patient-centered care and patient safety; their effectiveness has, however, rarely been evaluated and should be the object of future intervention research. Full article

Background: Substantial inter-individual variability in imatinib systemic exposure may influence both efficacy and toxicity. Based on recommendations of the International Association for Therapeutic Drug Monitoring and Clinical Toxicology (IATDMCT), the IRCCS-CRO of Aviano started in 2018 offering physicians the opportunity to monitor plasma concentrations of imatinib and its active metabolite, norimatinib, as part of a diagnostic service for patients with Gastrointestinal Stromal Tumor (GIST). This study aims to evaluate the potential clinical utility of imatinib Therapeutic Drug Monitoring (TDM) in predicting response and toxicity in a real-world cohort of 63 patients with GIST. Methods: Imatinib and its active metabolite norimatinib trough concentrations (C_min) were quantified using a validated LC–MS/MS method. Associations between drug exposure and clinical–demographic characteristics, treatment-related toxicity, and progression-free survival (PFS) were evaluated. Results: A total of 437 plasma samples from 63 patients were collected. Marked inter- and intra-patient variability in imatinib exposure was observed (43% and 31% respectively); 67% of patients receiving 400 mg/day had a C_min below the recommended target concentration (1100 ng/mL). Female sex was significantly associated with higher imatinib exposure (median C_min 1114 vs. 786 ng/mL in males; p = 0.0018). Combined age–sex analysis showed a significant difference between women and men < 60 years (1041 vs. 668 ng/mL; p = 0.0068, Bonferroni-adjusted). A strong exposure–toxicity relationship emerged, with higher imatinib levels in samples collected at toxicity occurrence vs. the others (1728 vs. 927 ng/mL; p < 0.0001), without a direct association between sex and toxicity. No significant association between C_min and PFS was observed; however, disease progression was more frequent in patients with C_min < 500 ng/mL and absent in those with C_min >1500 ng/mL (60% vs. 0%, respectively). Conclusions: TDM was found to be a potentially useful tool for predicting clinically relevant toxicity and outcome. Sex and age significantly influence exposure, supporting tailored dosing strategies. Full article

Background: Artificial intelligence and large language models (LLMs) are emerging as transformative technologies in medicine. However, their ability to develop physical exercise rehabilitation programs and provide insights into musculoskeletal (MSK) disorders remains underexplored. This study aimed to evaluate the quality and readability of LLM-generated responses to consultation questions addressing various stages of the clinical process encountered by patients with MSK disorders. Methods: This study recruited 50 patients with musculoskeletal disorders and extracted disease-related frequently asked questions from Google search. We developed three clinical scenario-based question types simulating real consultations, which were processed by four LLMs (GPT-3.5-turbo, GPT-4-turbo, GPT-4o, and Claude-3-haiku-20240307). Response quality was assessed by orthopedic specialists and therapists using the DISCERN instrument, and GPT-4o-assisted evaluation was used to extend the assessment after validation with expert ratings. Readability was systematically assessed via six validated indices. Results: Among the 1476 LLM-generated responses, the generated rehabilitation programs demonstrated consistent adherence to the specified query requirements. The DISCERN scores ranged from 26 (poor) to 68 (excellent), with a mean score of 55.60 ± 8.40. The intraclass correlation coefficient (0.68) indicated moderate interrater agreement, and Cronbach’s α showed good internal consistency. The readability scores across six indices indicated that most responses exceeded the recommended reading levels (p < 0.05). Conclusions: LLMs generated moderate-to-high-quality PE rehabilitation recommendations for patients with MSK disorders across simulated consultation scenarios. However, limited supporting materials and suboptimal readability may restrict their effectiveness. With physician oversight, improved readability, and enhanced supplementary resources, LLMs demonstrate considerable potential as supportive tools in orthopedics. Full article

Background/Objectives: Burnout among healthcare professionals has emerged as a major occupational and organizational challenge with important implications for workforce sustainability, patient safety, and healthcare system performance. Although the scientific literature on burnout has expanded substantially, an integrated understanding of its global development, research structure, and scientific impact remains limited. This study aimed to examine the evolution, thematic development, and citation dynamics of burnout research among healthcare professionals through an integrated bibliometric approach combining science mapping, multilevel citation modelling, and critical literature synthesis. Methods: A bibliometric analysis was conducted using publications indexed in the Web of Science Core Collection between 1987 and 2024. Science mapping techniques were applied using Bibliometrix and VOSviewer to evaluate publication trends, collaboration networks, thematic development, and citation patterns. In addition, a multilevel negative binomial regression model was used to identify publication characteristics associated with citation impact. Results: A total of 1232 publications were included in the analysis. Scientific production increased markedly after 2020, temporally coinciding with the COVID-19 pandemic and growing scientific interest in healthcare workforce wellbeing. Research output was concentrated in high-income countries, particularly the United States, China, the United Kingdom, Canada, and Australia. Physicians and nurses dominated the literature, while non-clinical healthcare workers remained underrepresented. The thematic analysis indicated an increasing emphasis on organizational and system-level determinants of burnout alongside the continued presence of individual-centered perspectives. The multilevel negative binomial model identified longer title length and a greater number of author-provided keywords as being associated with lower expected citation counts, whereas a greater number of Keywords Plus terms was associated with higher expected citation counts. Conclusions: Burnout research has evolved into a rapidly expanding and increasingly interdisciplinary field. However, important gaps persist regarding geographic representation, workforce diversity, methodological standardization, and organizational intervention research. Future studies should move beyond descriptive approaches and further evaluate organizational interventions and workforce-related strategies to strengthen the evidence base that may inform healthcare policy and organizational practice. Full article

Background: Hereditary angioedema (HAE) is a rare, potentially life-threatening disease in which delayed recognition and inappropriate medication use may result in preventable morbidity and mortality. We aimed to assess physicians’ knowledge regarding HAE-related triggers, clinical features, and management strategies across multiple medical specialties. Methods: This single-center, cross-sectional survey was conducted among 350 physicians at a tertiary training and research hospital. A structured electronic questionnaire assessed knowledge of HAE pathophysiology, diagnosis, medication-related triggers, and management. A predefined composite knowledge score (0–14) was calculated. Results: Although self-reported familiarity with HAE was high, overall disease-specific knowledge was limited. Awareness of critical medication-related triggers—including angiotensin-converting enzyme inhibitors, dipeptidyl peptidase-4 inhibitors, and estrogen-containing therapies—was low across all specialties, with no significant between-group differences. Substantial knowledge gaps were identified in the recognition of clinical features, diagnostic evaluation, and acute management. In multivariable analysis, prior clinical exposure to HAE patients was the only independent predictor of higher knowledge scores (B = 1.097, p = 0.001), whereas specialty group, gender, and years of professional experience were not independently associated with knowledge scores. However, the regression model explained only a small proportion of the variance in knowledge scores. Conclusions: Significant gaps in clinically relevant HAE knowledge persist among physicians from multiple medical specialties. Prior clinical exposure was associated with higher knowledge scores, whereas specialty group, gender, and years of professional experience were not independently associated with physician knowledge. Targeted, practice-oriented educational interventions focusing on medication-related triggers and acute management may help bridge these knowledge gaps, complement experiential learning, and ultimately enhance patient safety. Full article

Background: This study aimed to characterize real-world patterns of in-hospital statin initiation and identify clinical characteristics associated with this prescribing decision among previously statin-naïve patients hospitalized for acute decompensated heart failure (ADHF) in the Polish HEROES registry. Methods: We analyzed 374 statin-naïve patients hospitalized for ADHF, including 168 patients in whom statin therapy was initiated during the index hospitalization and 206 in whom it was not initiated. Demographic, clinical, laboratory, echocardiographic, functional and treatment-related variables were compared between the groups. A prespecified forced-entry multivariable logistic regression model including 10 clinically relevant and nonredundant covariates was used to evaluate factors associated with statin initiation. Results: After multivariable adjustment, higher odds of statin initiation were associated with male sex (adjusted odds ratio [aOR], 2.01; 95% confidence interval [CI], 1.12–3.61), arterial hypertension (aOR, 1.65; 95% CI, 1.06–2.56), previous myocardial infarction (aOR, 1.82; 95% CI, 1.03–3.22), higher systolic blood pressure (aOR per 10 mmHg increase, 1.12; 95% CI, 1.01–1.25), higher LDL-C concentration (aOR per 10 mg/dL increase, 1.15; 95% CI, 1.07–1.22), and anticoagulant therapy (aOR, 1.88; 95% CI, 1.20–2.96). Increasing age (aOR per 1-year increase, 0.97; 95% CI, 0.95–0.99) and a higher KCCQ-12 score (aOR per 10-point increase, 0.88; 95% CI, 0.81–0.96) were associated with lower odds of treatment initiation. LVEF and the GDMT score were not significantly associated with statin initiation. Conclusions: In previously statin-naïve patients hospitalized for ADHF, statin initiation reflected a range of clinical and treatment-related characteristics. These variables should be interpreted as correlates of real-world physician-guided prescribing decisions rather than as causal determinants of statin initiation or predictors of clinical benefit. The study does not provide evidence supporting routine statin initiation solely because of ADHF. Full article

Background: Acute upper gastrointestinal bleeding (UGIB) remains a leading cause of emergency department (ED) admission and carries a mortality of 5–10%. Static hemodynamic parameters perform poorly as predictors of fluid responsiveness, exposing patients either to under-resuscitation or to the harms of fluid overload. The mini fluid challenge (MFC), in which a 100 mL bolus is used to elicit a measurable change in the left ventricular outflow tract velocity–time integral (ΔVTI), is a validated dynamic test in intensive care but has not been evaluated in undifferentiated ED haemorrhagic shock. We aimed to determine whether ΔVTI-guided resuscitation is feasible at the bedside in UGIB and to explore how ΔVTI relates to in-hospital mortality. Methods: Prospective, single-centre, randomised quasi-experimental comparative study conducted between February 2024 and February 2025. Adults presenting with UGIB were allocated by treatment period, and credentialed emergency physicians to an ultrasound-guided protocol group (UGPG) or a standard care protocol group (SCPG). ΔVTI was measured in both groups but was disclosed to the treating team only in UGPG; in SCPG, it was recorded offline and analysed post hoc. The primary outcome was the feasibility of ΔVTI acquisition; the association of ΔVTI with in-hospital mortality was a secondary, exploratory outcome. Results: Eighty-five patients were enrolled (UGPG n = 47; SCPG n = 38). ΔVTI acquisition was completed within the intended time window in all enrolled patients, establishing bedside feasibility. Across the whole cohort, ΔVTI discriminated non-survivors poorly (AUC 0.55, 95% CI 0.44–0.66), and its performance differed markedly by group: AUC 0.83 (95% CI 0.67–0.93) in SCPG versus 0.69 (95% CI 0.54–0.82) in UGPG. In SCPG, non-survivors had substantially higher ΔVTI than survivors (58.6 ± 30.5% vs. 28.4 ± 30.9%; p = 0.030), a pattern consistent with persistent, uncorrected fluid responsiveness. In UGPG, the association ran in the opposite direction, mortality being concentrated among ΔVTI < 10% non-responders (p = 0.020). Conclusions: ΔVTI-guided resuscitation is feasible in the ED management of UGIB. The direction of the ΔVTI–mortality association differed according to whether ΔVTI was visible to clinicians, generating the testable hypothesis that unrecognised fluid responsiveness contributes to death in standard care. These findings are hypothesis-generating only: the small sample and the low number of events preclude any causal inference and require confirmation in an adequately powered, randomised, etiology-stratified trial. Full article