Weekly Sepsis Research Analysis
This week’s sepsis literature converged on three high-impact domains: advanced diagnostics/phenotyping with geometric deep learning (InfEHR) enabling label-efficient detection of low-prevalence sepsis phenotypes; mechanistic targets linking innate immunity and organ dysfunction (FPR1–NETosis axis in septic cardiomyopathy and gut/translocation-driven sepsis physiology); and clinically actionable practice changes from implementation studies and meta-analyses (continuous β‑lactam infusion evidence,
Summary
This week’s sepsis literature converged on three high-impact domains: advanced diagnostics/phenotyping with geometric deep learning (InfEHR) enabling label-efficient detection of low-prevalence sepsis phenotypes; mechanistic targets linking innate immunity and organ dysfunction (FPR1–NETosis axis in septic cardiomyopathy and gut/translocation-driven sepsis physiology); and clinically actionable practice changes from implementation studies and meta-analyses (continuous β‑lactam infusion evidence, NICU IPC and stewardship, and rapid multiplex diagnostics improving outcomes). Together these studies push translational targets toward trials while offering pragmatic tools for risk stratification and antibiotic stewardship.
Selected Articles
1. InfEHR: Clinical phenotype resolution through deep geometric learning on electronic health records.
InfEHR converts whole EHRs into temporal graphs and applies deep geometric learning to infer clinical likelihoods with very few labeled examples. Tested across two health systems, it markedly improved sensitivity for culture-negative neonatal sepsis and postoperative AKI while preserving high specificity, demonstrating scalable, label-efficient sepsis phenotyping.
Impact: Introduces a novel, high-impact AI paradigm that addresses a key barrier in clinical ML—scarcity of labeled data—enabling earlier and more accurate detection of low-prevalence sepsis phenotypes directly from routine EHRs.
Clinical Implications: Hospitals can consider piloting InfEHR-like tools to flag culture-negative neonatal sepsis and postoperative AKI, improving early recognition and guiding diagnostic or therapeutic escalation without large manual labeling efforts.
Key Findings
- Temporal-graph EHR representation with deep geometric learning enables high-performance probabilistic inference with few labels.
- Sensitivity for culture-negative neonatal sepsis improved from 0.04 (physician heuristics) to 0.60 while specificity remained high.
- Outperformed physician heuristics across two independent health systems, supporting generalizability.
2. N-formyl methionine mediates NETosis of neutrophil to promote sepsis-induced cardiomyopathy via the FPR1 pathway.
This translational study links elevated serum N-formyl methionine (fMet) in sepsis-induced cardiomyopathy (SIC) to NETosis through FPR1/HIF-1α activation. Genetic FPR1 knockout and pharmacologic inhibition reduced NETosis, inflammation, and improved survival and cardiac function in CLP sepsis models, positioning FPR1 and NET modulation as actionable therapeutic targets.
Impact: Combines patient biomarker correlations with mechanistic human-cell and murine data to identify FPR1 as a tractable host target to mitigate sepsis-induced cardiac dysfunction via NETosis modulation.
Clinical Implications: Supports development of FPR1 inhibitors and monitoring of fMet as prognostic/risk-stratification biomarkers for SIC; suggests NET-modulating therapies warrant translational testing in early-phase trials for septic cardiomyopathy.
Key Findings
- Serum fMet is elevated in SIC patients and correlates with NET markers MPO and dsDNA.
- fMet plus LPS increased NET formation and upregulated FPR1 and HIF-1α in human neutrophils.
- FPR1 knockout or inhibition suppressed NETosis and improved survival, cardiac function, and mitochondrial function in CLP sepsis mice.
3. Mesenteric ischemia and bacterial translocation precipitate the intoxication phase of yellow fever.
In a reverse-translational study (hamster model corroborated by human fatal-case pathology), lethal ‘intoxication’ in yellow fever is shown to result from mesenteric ischemia causing intestinal epithelial erosion and portal-system bacterial translocation, producing a sepsis-like syndrome. Human fatal cases displayed portal/hepatic bacteria and plasma markers of intestinal injury, reframing late YF lethality as secondary sepsis-driven physiology.
Impact: Reframes a historically poorly understood lethal phase of a major viral hemorrhagic disease as secondary bacterial translocation–driven sepsis, suggesting immediate, testable clinical interventions (gut-directed management and early antimicrobial strategies) and changing management paradigms.
Clinical Implications: Clinicians managing suspected yellow fever intoxication should consider early assessment for mesenteric ischemia and intestinal barrier failure, and the potential role for prompt broad-spectrum antimicrobials and gut-directed supportive strategies while confirming viral dynamics—warranting updates to treatment algorithms and trials.
Key Findings
- Mesenteric ischemia and subsequent epithelial erosion preceded intoxication in an animal YF model.
- Portal-system bacterial translocation produced a sepsis-like syndrome; human fatal cases showed portal/hepatic bacteria and elevated intestinal-injury/bacteremia markers.
- Provides a unified mechanism explaining clinical features of late fatal YF (e.g., AST/ALT pattern, GI hemorrhage, pancreatitis, neutrophilia).