Publications

BACKGROUND: Both whole blood (WB) and component therapy (CT) are used for hemostatic resuscitation in injured children. We hypothesize that early WB transfusion compared to CT alone is associated with decreased post-traumatic organ dysfunction.

STUDY DESIGN AND METHODS: This single-center observational study included children ages 0-17 years between January 2021 and March 2024 with trauma mechanism and intensive care unit admission. The primary outcome was Pediatric Logistic Organ Dysfunction 2 (PELOD-2) score on post-trauma days 1-7. Data were analyzed using linear regression adjusting for age, sex, race, year, injury mechanism, injury severity score (ISS), shock index pediatric age-adjusted, and total 4-h transfusion volume (mL/kg).

RESULTS: In total, 540 subjects met eligibility criteria; of the 52/540 (10%) who received blood transfusion within 4 h, 11/52 (21%) received RBC alone, 12/52 (23%) received WB alone, 9/52 (17%) of subjects received RBC plus other, and 20/52 (38%) received WB plus other. The cohort was 60% (326/540) male, 83% (449/540) blunt injury mechanism, median (interquartile range [IQR]) age 3 years (0-11), and median (IQR) ISS of 11 (8-18). In adjusted analysis, transfusion of WB + other was an independent predictor of lower PELOD-2 score through post-trauma day 7 in comparison to subjects receiving RBC + other.

DISCUSSION: In subjects who were transfused multiple blood products, receipt of any WB versus CT alone for hemostatic resuscitation after injury was associated with reduced organ dysfunction. Further investigation is needed in large cohorts to fully elucidate clinical benefit and improve mechanistic understanding.

BACKGROUND: Whole blood (WB) is a preferred resuscitation strategy, yet how storage affects hemostasis remains unclear. Prior studies have shown decreased fibrinogen function during storage; however, these were limited by incomplete removal of platelet contributions. Additionally, leukoreduced (LR) versus non-leukoreduced (NLR) status in preserving fibrinogen remains unknown.

STUDY METHODS: WB units (n = 15, 7 NLR, 8 nLR) were evaluated at serial time points through day 35. ROTEM EXTEM was performed. Platelet-poor plasma (PPP) from WB at day 0, day 21, and day 35 underwent western blot for fibrinogen. Functional fibrinogen was measured using the Clauss assay after freeze-thaw cycles of WB PPP to minimize platelet contribution.

RESULTS: As WB aged, there was a progressive decrease in alpha angle and maximum clot firmness. Platelet count dropped sharply by day 7. Western blot demonstrated no evidence of fibrinogen degradation in either LR or NLR WB. Clauss fibrinogen concentrations were unchanged from day 0 to day 35, with a modest non-significant decline by day 35 driven primarily by NLR units.

DISCUSSION: Fibrinogen did not undergo detectable proteolysis during storage of WB and had preserved function through day 35 after freeze-thaw cycles to minimize platelet contributions. NLR appeared to impact fibrinogen function with medians below 200 for day 21 and 35 but was not statistically significant. This suggests that previously reported reductions in fibrinogen function may be based on leukoreduction status, but that early changes in fibrinolysis sensitivity are more likely due to residual platelet contributions to the assays rather than loss of fibrinogen function itself.

INTRODUCTION: Accurate, reliable means to obtain fresh viable clinically localised prostate cancer tissue do not exist. We developed a method in which bespoke 3D-printed moulds can be created for any prostate and allow for ex vivo sampling of magnetic resonance imaging (MRI)-visible, biopsy-proven cancer lesions. We sought to demonstrate the ability of our platform to obtain fresh viable prostate cancer tissue after robot-assisted radical prostatectomy (RARP).

METHODS: Inclusion criteria were a patient that underwent MR-fusion biopsy (UroNav, Philips) with a resulting biopsy proven Gleason Grade (GG) ≥ 2 target. STL files for prostate boundary and target regions of interest (ROI), created as part of fusion biopsy, were exported from the UroNav and imported into SolidWorks (Dassault), a solid modelling computer-aided design and engineering application. A macro within SolidWorks was then applied to create a material-optimised mould around the prostate with needle guides to allow for targeted sampling. The 3D mould was exported as an STL file and then 3D-printed on a Stratasys Fortus 250 MC 3D Printer. During RARP, the specimen is extracted, the seminal vesicles detached and the prostate is placed in the mould for biopsy. The biopsy cores are sent to pathology for analysis and compared to specimens from the initial biopsy.

RESULTS: Twelve patients with MRI-visible lesions and biopsy proven GGG ≥ 2 cancer underwent RARP. In 12 out of 12 patients, ex vivo biopsies performed with the 3D-printed mould yielded prostate tissue with cancer.

CONCLUSIONS: Our 3D-printed mould platform allows for ex vivo sampling of MRI identified and previously biopsied prostate cancer at the time of RARP. The native, cancerous tissue may then be used to advance further research. The potential applications for a platform that can reliably sample living prostate cancer tissue are numerous, including the ability to advance future cancer research as well as other solid-organ malignancies with targetable lesions.

BACKGROUND: Critical illness dramatically increases the demand for and consumption of leukocytes. This demand induces emergency hematopoiesis (EH) whereby the hematopoietic stem and progenitor cells in the bone marrow become activated and proliferate to increase leukocyte production. Recent studies in murine models of trauma demonstrate that trauma induces EH through IL-1-dependent production of G-CSF. We hypothesized that non-septic critical illness (NSCI) would similarly drive IL-1 and G-CSF- mediated EH in critically ill humans.

METHODS: Two independent prospective observational trials were conducted. Plasma cytokine levels were measured in a cohort of 55 critically injured trauma patients. G-CSF levels were measured on arrival and at 1, 6, 24, and 72 hours after arrival. Correlation between G-CSF levels and injury severity, length of stay, ICU length of stay and inflammatory cytokine levels were measured. Bone marrow was isolated from an independent cohort 6 non-septic critically ill organ donation candidates and 5 healthy donors. Hematopoietic stem and progenitor cells were measured by flow cytometry.

RESULTS: Critical injury induced a rapid and significant increase in plasma G-CSF levels, detectable within 1 hour of hospital arrival, with peak levels measured 6 hours after injury. G-CSF levels 1 hour after arrival correlated with injury severity, length of stay and ICU length of stay. 6 hour G-CSF levels correlated with ICU LOS. 6 hour G-CSF levels correlated strongly with 6-hour levels of IL-1β, IL-6 and IL-8. NSCI caused a significant increase in hematopoietic stem cells (HSC) and common myeloid progenitors and a significant decrease in multipotent progenitors and common lymphoid progenitors as compared to healthy controls.

CONCLUSIONS: Trauma induces a rapid increase in plasma G-CSF levels that is strongly correlated with plasma IL-1β, suggesting that as in murine models, IL-1 may be driving G-CSF production. As in mice, NSCI induced EH, characterized by expansion of the HSC cohort and a shift toward myelopoiesis. In critically ill humans these data suggest that sterile injury induces IL-1 and G-CSF mediated EH, recapitulating the murine finding.

Anorexia nervosa is a severe psychiatric disorder characterized by persistent food restriction and often excessive physical activity, implicating dysfunction in neural circuits governing motivation, reward, and behavioral persistence. The nucleus accumbens (NAc) is a central component of these circuits, yet synaptic and cellular adaptations within this region during anorexia-like states remain poorly defined. Using the activity-based anorexia (ABA) paradigm in adult female mice, we examined glutamatergic signaling and intrinsic neuronal properties in the NAc shell. ABA exposure produced rapid weight loss, reduced food intake, and progressively increased running-wheel activity. Biochemical analyses of NAc shell tissue revealed elevated membrane-associated GluA2 AMPA receptor protein. Consistent with this finding, whole-cell patch-clamp recordings from medium spiny neurons showed increased amplitude of spontaneous excitatory postsynaptic currents. ABA also enhanced intrinsic neuronal excitability, reflected by greater firing in response to depolarizing current injections. Together, these convergent biochemical and electrophysiological results demonstrate that ABA induces coordinated postsynaptic strengthening and increased intrinsic excitability in NAc shell medium spiny neurons. These adaptations suggest a sustained increase in accumbal output that may bias motivational circuit function and contribute to excessive activity and suppressed feeding during anorexia-like conditions, paralleling glutamatergic plasticity observed in other compulsive disorders, including substance use disorder.

OBJECTIVES: Air medical transport (AMT) improves survival for selected trauma patients. Improving AMT triage is limited by a lack of evidence and a standardized definition for which patients and circumstances may warrant AMT. Our objective was to develop a consensus-based definition of AMT need.

METHODS: We recruited a multidisciplinary, nationally representative panel with expertise in AMT from trauma surgery, anesthesiology, critical care, emergency medicine, and emergency medical services (EMS). Panelists were presented with criteria from the literature representing the potential for AMT need that included patient injuries, time-sensitive interventions, and system factors. Panelists voted over 4 rounds to refine and select (≥70% agreement) a final set of criteria using a web-based Delphi methodology, including potential criteria combinations.

RESULTS: A total of 32 of 45 (71.1%) invited panelists agreed to participate. From 66 initial criteria, panelists reached consensus on 18 patient factors, 6 time-sensitive interventions, 3 system factors, and 7 combinations of criteria. Two key themes emerged: the need for specialized care from air medical crews that may not be available from ground ambulance clinicians, as well as overall prehospital time-savings. After narrative feedback and refinement to eliminate redundant and overlapping criteria, an algorithm for AMT need was developed along with a decision flow diagram suitable for educational dissemination.

CONCLUSIONS: We developed a consensus-based definition of AMT need for trauma patients that can be operationalized for AMT triage. Further validation of this concept with patient outcomes and identifying implementation barriers will contribute to field deployment of a useful AMT triage tool for EMS clinicians.

Understanding the decision-making process of black-box neural networks is crucial for safe use of AI in high-stakes medical tasks such as histopathology. We present Adaptive Example Selection (AES), a prototype-based explainable AI framework that improves interpretability of deep learning models for mitosis detection. AES retrieves a sparse set of supporting and contradicting real-world prototype images to locally approximate the model's confidence surface with high fidelity ([Formula: see text]). The framework is integrated with a robust Faster R-CNN detector that demonstrates strong cross-tumor performance, for example achieving an F1-score of 0.84 on the Canine Cutaneous Mast Cell Tumor dataset. AES generates concise, case-specific explanations that faithfully capture local decision boundaries while linking predictions to interpretable exemplars. This enables clinicians to visualize model reasoning, assess uncertainty, and conduct contrastive analyses. Unlike prior methods focused on discrete class predictions, AES shows how similarity to mitotic and non-mitotic prototypes shapes graded confidence, enhancing transparency, trust, and practical adoption of AI-assisted mitosis detection in cancer diagnostics.