Age-Dependent Remodeling of Splenic Immune Responses in Murine Sepsis Revealed by Single-Cell Transcriptomics

Authors:
Christine Rodhouse, Dayuan Wang, Hongru Tang, Miguel Hernandez-Rios, Whitman Wiggins, Xuanxuan Yu, Leandro Balzano-Nogueira, Angel Charles, Larissa Langhi Prata, Tyler Loftus, Letitia Bible, Alicia Mohr, Robert Maile, Guoshuai Cai, Shawn Larson, Philip Efron, Jaimar Rincon

Body of Abstract:
Background: Decades of sepsis research have yet to produce an effective immune-therapeutic to improve patient outcomes. This is in part due to a failure to apply precision medicine to sepsis research, which includes accounting for the age of the host as the disease tends to affect the very young and older adults the most. Our goal is to delineate the transcriptomic patterns of leukocytes early after sepsis in key age (neonate, young adult, older adult) groups using a polymicrobial model of murine abdominal sepsis, allowing for comparisons of similarities and differences.

Methods: Neonatal (7-day-old), young adult (3-4 months), and older adult (18-24 months) C57BL/6 (B6) mice were challenged with 1.1 – 1.3 mg/g BW of cecal slurry (CS) to induce polymicrobial sepsis. Spleens were harvested prior to challenge (naive) and 18 hours post-sepsis (CS18h). Single-cell gene expression libraries were generated using the 10X Genomics platform and sequenced on an Illumina HiSeq® system (scRNAseq).  Data was processed with CellRanger and analyzed in Seurat.  Differentially expressed genes (DEGs) were identified using an adjusted Wilcoxon’s rank-sum test. IPA was used for functional enrichment, with significance assessed by Fisher’s exact test. Pseudotime trajectory analysis was inferred based on transcriptional dynamics. 

Results: Following sepsis, neonates showed expansion of HSPCs, PMNs, and mast cells; young adults exhibited reductions in HSPCs and mast cells, while older adults displayed increased DCs with otherwise stable myeloid proportions. Monocyte transcriptional responses differed markedly by age. Neonatal monocytes showed dysregulated activation and impaired maturation with induction of IL-10, Th2, TLR, LPS/IL-1, iNOS, IL-17, and IL-1 pathways and suppression of LXR/RXR and MAPK. Young adults mounted a balanced response, upregulating IL-10, IL-17, and PD-1/PD-L1 while downregulating Th1, HGF, and IL-4. Aged monocytes exhibited profound dysregulation with activation of cytokine-storm, IL-10, ID3, and sirtuin pathways and loss of PPAR/RXR, TRIM21, and communication signaling, reflecting a hyperinflammatory, metabolically compromised state. Pseudotime trajectory analysis revealed pronounced alterations suggesting perturbations in hematopoietic differentiation dynamics across the ages. In young adult mice, myeloid differentiation skewed towards inflammatory effector cells (e.g. monocytes, DCs). In contrast, in neonates and aged mice there were disrupted trajectories and reduced progression to the terminal macrophage branches, indicative of impaired myeloid maturation. 

Conclusions: scRNAseq analysis of splenic leukocytes revealed that host age results in tremendous differences in cell proportions, DEGs, and cell differentiation early after sepsis. Translation efforts in sepsis can be improved by applying precision medicine to research, including preclinical rodent sepsis models needed for FDA approval of therapies.

Antibiotic Duration Beyond 24 Hours Does Not Reduce Infection in Grade 3 Open Extremity Fractures

Authors:
Andrea Gochi, Henry Olivera Perez, Anna Huang Perez, Hyunjee Kwak, Lucas Thornblade, Naveen Balan, Adam Gutierrez, Genna Beattie, Gregory Victorino, April Mendoza

Body of Abstract:
Background: Grade 3 Gustilo Anderson open extremity fractures carry a high risk of infection, yet wide variation in antibiotic duration persists. Many patients receive a prolonged (> 48 hour) antibiotic course despite limited evidence of benefit. Prior studies have demonstrated no reduction in surgical site infection (SSI) with broader antibiotic coverage in severe open fractures, and in some cases increased nephrotoxicity with longer regimens. We hypothesized that antibiotic administration beyond 24 hours  from time of injury, provides no additional protection against development of SSI in Grade 3 open extremity fractures.

Methods: A retrospective cohort study of all adult Grade 3 Gustilo Anderson open extremity fractures treated at a Level I trauma center from October 2019 to September 2025 was conducted. Clinical characteristics, antibiotic treatment data, and infectious outcomes were collected. The primary outcome was development of a SSI (superficial or deep) at the site of Grade 3 open extremity injury. The secondary outcome was need for hardware removal. Multivariable logistic regression was used to identify independent predictors of SSI, and a separate model evaluated factors associated with hardware removal.

Results: Out of 1507 Grade 1-3 open extremity fractures, a total of 216 Grade 3 open extremity fractures, met inclusion criteria. Patients with incomplete data, those who died on arrival or died within 2 weeks were excluded from analysis. The overall SSI incidence was 18.5% (40/216), including 15 superficial SSI and 25 deep SSI infections, with 43% of the cohort receiving broad coverage (Gram positive and Gram negative and/or anaerobic coverage). Infection rates did not differ between patients receiving ≤24 hours versus >24 hours of total antibiotics (20.9% vs 17.9%, p=0.649). Postoperative antibiotic duration was similarly not associated with infection and neither was coverage type (all p>0.05). No significant relationship was found between antibiotic duration and hardware removal (p>0.05).  In multivariable analysis, active smoking (OR 4.47, p=0.004), diabetes mellitus (OR 5.12, p=0.018) and need for multiple washouts (OR 3.89, p=0.013), were independently associated with SSI. In the hardware removal model, superficial SSI (OR 6.60, p=0.022) was the strongest contributor.

Conclusions: Antibiotic administration beyond 24 hours for Grade 3 Gustilo Anderson open extremity fractures provided no protective benefit against SSI or need for hardware removal. These findings support emerging evidence that prolonged antibiotic courses may not improve outcomes in severe open fractures and reinforce the need for antibiotic stewardship in the management of Grade 3 injuries.

Antibiotic-induced gut dysbiosis enhances pathogen-mediated coagulopathy via shifts in quorum sensing and microbial metabolites

Authors:
Jessica Cao, Rebecca Meltzer, Andrew Benjamin, John Alverdy, Robert Keskey

Body of Abstract:
Background: Critically ill trauma patients are exposed to selective pressures including antibiotics that disrupt the gut microbiota, favoring expansion of Proteobacteria including pathogens like Pseudomonas aeruginosa (Pa). Intestinal dysbiosis in this setting has been linked to adverse outcomes including bleeding complications. We recently demonstrated that gut-derived Pa can induce clinically significant coagulopathy via secreted exoproducts in humans and mice. However, the extent to which Pa virulence expression–and thus its ability to drive coagulopathy–is modulated by the gut environment following antibiotic exposure remains unknown. We hypothesize that antibiotic-induced changes in the intestinal milieu (i.e. higher levels of the interspecies quorum-sensing molecule autoinducer-2 [AI-2], depletion of gut-derived metabolites) increases Pa virulence and enhances its ability to induce coagulopathy.

Methods: Pa isolated from an ICU patient was grown in tryptic soy broth (TSB) and TSB supplemented with the following: cecal metabolites from antibiotic-decontaminated mice (AbxCec); supernatant of E. Coli that constitutively expresses AI-2 (Ec-AI2), supernatant of a control E. Coli strain (Ec-Ctrl); 100 mM acetate (Ace); 10 mM butyrate (But); 10 mM propionate (Prop); 1 mM kynurenic acid (Kyn); 1 uM indole-3-propionate (IPA); and 50 mM tryptophan (Trp). Supernatant from Pa grown in each of these conditions (Pa-Sup) was collected at stationary phase at a similar optical density, filtered, and added to citrate-anticoagulated whole blood from healthy volunteers. Coagulation was analyzed using TEG6S. 

Results: Growth in TSB supplemented with AbxCec increased Pa-induced coagulopathy in a dose-dependent manner. Compared to TSB Pa-Sup, AbxCec Pa-Sup prolonged R time when added to blood at 1:15 (54.18s vs 4.64s; p<0.001) and decreased maximum amplitude (MA) at 1:20 (18.74 vs 57.76; p<0.001) and 1:15 (2.1 vs 43.46; p<0.01). Ec-AI2 Pa-Sup decreased MA to a greater extent than Ec-Ctrl Pa-sup and TSB Pa-Sup at 1:10 (7.52 vs 13.92 vs 29.8; Ec-AI2 vs Ec-Ctrl vs TSB Pa-sups; p<0.05). Ec-AI2 Pa-Sup also prolonged R time to a greater extent (31.1s vs 18.6s vs 3.88s; NS), but this was not statistically significant. Growing Pa in the presence of Ace, Kyn, IPA, and Trp suppressed Pa-Sup’s ability to decrease MA when added to blood at 1:10 (18.53 vs 59.53 vs 43.23 vs 53.2 vs 52.6; TSB vs Ace vs Kyn vs IPA vs Trp Pa-Sups; p<0.05), while growing Pa with But and Prop did not have this effect. Conclusion: By shifting the balance of quorum-sensing signals (i.e. AI-2) and gut-derived metabolites (i.e. SCFAs, Trp derivatives), antibiotics create a metabolic niche that favors Pa predominance and potentiates its virulence expression in a way that amplifies pathogen-mediated coagulopathy. This underscores the concept that antibiotic-mediated perturbations of the gut microbiota can transform commensal-pathogen dynamics and ultimately hinder recovery in critically ill patients.

Appendicitis in the Transplant Patient: A Common Problem in a Less Common Patient Population

Authors:
Michael Megaly, Todd Costantini, Qi Wang

Body of Abstract:
Background:
Acute appendicitis is an uncommon surgical emergency among solid organ transplant (SOT) recipients and may present atypically due to immunosuppression. Comparative data with non-transplant patients remain limited. We hypothesized that SOT recipients would demonstrate blunted inflammatory responses and atypical clinical presentations.

Methods:
We performed a retrospective review of adult SOT recipients diagnosed with appendicitis within a multi-center integrated health system from January 2011 to July 2025. SOT patients were compared with a contemporaneous cohort of non-transplant patients with appendicitis. Demographics, presenting vital signs, laboratory values, perforation rates, operative versus non-operative management, and antibiotic use were assessed.

Results:
Among 4,049 SOT recipients, 58 (1.4%) developed appendicitis. Median time from transplant to diagnosis was 889 days (IQR 116–3,399). SOT recipients were older (51.9 ± 16.6 vs 41.8 ± 18.2 years, p < 0.001) and had higher rates of hypertension (74% vs 20%) and diabetes (36% vs 8%) (p < 0.001 for both). Fever (>100.4 °F) occurred in 9.4% of SOT recipients versus 17.2% of non-transplant patients (p = 0.18), and leukocytosis (>11 × 10⁹/L) in 44.8% of SOT patients vs 68.6% non-transplant patients (p < 0.01). Non-operative management was more common in SOT recipients (14% vs 3%, p = 0.0028). Perforated appendicitis occurred more frequently (17% vs 8%, p = 0.04) in SOT patients. SOT patients had longer courses of antibiotics for all appendicitis management. Median length of stay was longer (2.9 vs 1.0 days, p = 0.02) in SOT recipients while no 30-day mortality occurred in either group. Conclusions: Appendicitis in SOT recipients is rare and frequently presents with attenuated inflammatory signs and increased risk of perforation.This underscores the need for a high index of suspicion for acute appendicitis in SOT recipients even in the absence of traditional diagnostic findings.

Association of Palliative Care Consultation with Resource Utilization and Infectious Outcomes in Advanced Age Surgical ICU Patients with Sepsis

Authors:
Francesca Bragg, Ryan Desrochers, Daithi Heffernan, Andrew Stephen

Body of Abstract:
Background: Surgical ICU patients with sepsis often face rapidly progressive deterioration, multisystem failure requiring a high degree of supportive care, and ultimately face a high mortality. Surgeons are trained to view sepsis as a reversible condition if timely operative source control is achieved which is often in contrast to other medical conditions which are more commonly accepted as displaying a terminal, or irreversible trajectory. This mindset may delay important goals-of-care discussions when frailty, advanced age, and organ failure limit the physiologic benefit of aggressive interventions. Earlier palliative care (PC) integration may align treatment decisions with prognosis and reduce non-beneficial invasive care, but its role in septic surgical patients remains poorly defined.

Methods: We conducted a retrospective analysis of critically ill surgical patients ≥65 years with sepsis requiring a procedural intervention. Patients receiving PC consultation were compared with those without PC involvement on mortality, infection-related complications, ventilator exposure, invasive interventions, and health care utilization. Continuous data is presented as mean and standard deviation

Results: A total of 11,090 patients were included; 5,520 (50%) received PC consultation and 5,570 (50%) did not. PC patients were older (76.4+/-0.9 vs 74.6+/-0.9; p<0.001) and had higher rates of heart failure (53% vs 39%; p<0.001), COPD (35% vs 26%; p<0.001) and chronic kidney disease (45% vs 34%; p<0.001). PC patients exhibited higher mortality (60% vs 16%; RR 3.8, 95% CI 3.548-4.036; p<0.001). PC involvement was associated with significantly greater downstream infection-related complications, including pneumonia (17% vs 12%; RR 1.3, p<0.001) and bacteremia (11% vs 9%; RR 1.3, p<0.001), and significantly fewer UTI (12% vs 15%; RR 0.83, p=<0.001). PC patients underwent fewer invasive procedures (56% vs 71%; RR 0.79, p<0.001), fewer specialist consults (66% vs 68%; RR 0.97, p=0.03), fewer CT scans (42% vs 47%; RR 0.88, p<0.001), and reduced antimicrobial exposure (83% vs 88%; RR 0.95, p<0.001). Conversely, ventilator (28% vs 22%; RR 1.3, p<0.001) and vasopressor (18% vs 12%; RR 1.5, p<0.001) utilization was higher in the PC group. Conclusions: In older septic surgical ICU patients, palliative care consultation is associated with higher rates of pneumonia and bacteremia but lower rates of urinary tract infection, underscoring the complexity of downstream infection-related outcomes in this population. Palliative care involvement is associated with decreased procedural intervention while appropriately identifying those with terminal illness trajectories. These findings support normalizing and earlier integration of palliative care in sepsis management, particularly when physiologic reserve is limited, to ensure patient-centered decision making and avoid non-beneficial escalation of care.