Dose-Dependent Effects of Whole Blood Transfusion on Associated Infection Risk in Trauma

/in ,

Dose-Dependent Effects of Whole Blood Transfusion on Associated Infection Risk in Trauma

Authors:
Hadi Hamdan, Ahmad El Nouiri, Camden Gardner, Drishti Lall, Kurt Kralovich, Jeffrey Johnson

Body of Abstract:
Background: Whole blood (WB) transfusion has emerged in trauma resuscitation as a method associated with improved survival over component therapy (CT), largely attributed to reduced hemorrhagic deaths. However, WB is a biologically complex fluid with distinct immunologic and volume-expanding properties that may affect infection risk differently than CT. Although infections have often been secondary outcomes in transfusion studies, their direct relationship with WB use remains unclear. We hypothesized that a higher WB-to-total-transfusion ratio would be associated with lower rates of nosocomial infection in a dose-dependent manner.

Methods: Adult trauma patients from the 2020–2023 national TQIP database who received transfusion were included. The WB ratio was defined as the proportion of total transfused volume given as WB. Outcomes were examined across the full ratio range (0–1), with extremes (CT-only vs WB-only) analyzed separately to confirm consistency. Variables significant on univariate analysis were entered into multivariable logistic regression models evaluating associations between WB ratio and infection outcomes: severe sepsis, surgical site infections, catheter-associated urinary tract infection (CAUTI), and central-line associated bloodstream infection (CLABSI). Models adjusted for demographics, physiology, injury severity, comorbidities, transfusion volume, and hospital characteristics. A subanalysis of ventilated ICU patients assessed ventilator-associated pneumonia (VAP), additionally adjusting for ICU stay and ventilator days.

Results: Among 221,142 transfused patients, 79.8% received CT only, 8.7% WB only, and 11.5% both. Increasing WB ratio was associated with lower odds of severe sepsis (aOR 0.82, p = 0.006), superficial SSI (aOR 0.79, p = 0.020), and CAUTI (aOR 0.73, p = 0.012). At the extremes, WB-only versus CT-only showed similar reductions in severe sepsis (aOR 0.74, p = 0.001), superficial SSI (aOR 0.73, p = 0.008), deep SSI (aOR 0.79, p = 0.028), organ/space SSI (aOR 0.65, p < 0.001), and CAUTI (aOR 0.73, p = 0.021), with no difference in CLABSI. In the ventilated ICU subanalysis (n = 94,206), higher WB ratio was associated with increased odds of VAP (aOR 1.21, p < 0.001), and this persisted when comparing WB-only to CT-only. Across models, older age, higher ISS, diabetes, steroid use, chemotherapy, and longer hospitalization remained independent infection predictors. Conclusion: Our findings demonstrate a consistent, dose-dependent protective association between increasing WB ratio and several nosocomial infections in trauma patients, an effect that remained stable at the extremes of transfusion practice (WB-only vs CT-only). However, this trend did not extend to ventilator-associated pneumonia, which increased with higher WB exposure. These results highlight a potentially heterogeneous effect of WB transfusion across infection types and support further study into the mechanisms and optimal transfusion balance.

Early Bronchoscopy for the Prevention of Pneumonia in Trauma Patients Undergoing Emergent Intubation: An Interim Analysis

/in ,

Early Bronchoscopy for the Prevention of Pneumonia in Trauma Patients Undergoing Emergent Intubation: An Interim Analysis

Authors:
Michelle Lippincott, Louis Perkins, Laura Haines, Jessica Weaver

Body of Abstract:
Background: Patients who are intubated emergently are at increased risk for developing ventilator associated pneumonia. This risk is theoretically further increased in trauma patients who often undergo prolonged noninvasive ventilation with altered mental status in the prehospital setting. We hypothesized that early bronchoscopy in these patients reduces the risk of pneumonia by decreasing airway bacterial burden and removing inflammatory gastric acid and digestive enzymes. Methods: After implementation of an early bronchoscopy protocol at a Level 1 Trauma center, a retrospective cohort study was performed on all patients intubated in the prehospital setting or within 1 hour of arrival to the hospital. Patients were excluded if they died or were extubated within 24 hours of arrival or they were transferred to another ICU service. Patients who underwent bronchoscopy within 24 hours of admission (early bronchoscopy) were compared to those who did not receive a bronchoscopy within 24 hours (control group). The primary outcome was diagnosis of pneumonia during the same hospitalization. Secondary outcomes included ICU length of stay (LOS), ventilator days, hospital LOS, and mortality. Results: Among the 59 patients included in our interim analysis, 7 (11.9%) underwent bronchoscopy within 24 hours of admission. No patients in the early bronchoscopy group developed pneumonia, compared to 28.8% in the control group. There was no difference in median ventilator days (3 vs 4 days, p=0.41), median ICU LOS (4 vs 6.5 days, p=0.49), hospital LOS (median 10 vs 11 days, p=0.93), or mortality (28.6% vs 21.2%, p=0.64). In the early bronchoscopy group, a bronchial alveolar lavage culture was sent in three patients and 100% were positive for organisms, though none of these patients developed a clinical pneumonia. One patient that received early bronchoscopy received antibiotics for prophylaxis against aspiration pneumonia, while 71% of patients received antibiotics for any indication. Conclusions: In this preliminary analysis, for trauma patients intubated in the pre-hospital setting or within 1 hour of arrival, early bronchoscopy was not associated with a statistically-significant difference in pneumonia rates, hospital LOS, ICU LOS, ventilator days or mortality. However, no patients developed pneumonia in the early bronchoscopy group. Continued data collection for a larger sample size is necessary to demonstrate if there is any benefit in these patients.

Ehrlichiosis and Anaplasmosis Infections in Solid Organ Transplant Recipients: A Single-Center Series

/in ,

Ehrlichiosis and Anaplasmosis Infections in Solid Organ Transplant Recipients: A Single-Center Series

Authors:
Benjamin Fisher, Pravin Meshram, Michael Megaly, Rubeena Naaz, Anmol Nigam, Anna Sachdeva, Jo-Anne Young, Michael Park, Raja Kandaswamy, James Harmon

Body of Abstract:
Introduction

Ehrlichiosis and anaplasmosis are emerging tick-borne rickettsial infections in the bacterial family of Anaplasmataceae. This series describes three cases of ehrlichiosis and four cases of anaplasmosis after organ transplantation at a single center. We report the time between transplant and infection diagnosis, laboratory values, and comorbidities.

Methods

We conducted a 13-year (2011-2024) retrospective cohort study using data from our institutional database. Demographic characteristics, transplant history, comorbidities, medications, and laboratory data were analyzed using R. Relevant laboratory measures and antibiotic usage were identified using pattern-based searches chronologically to characterize trends for each recipient. 

Results

We identified a total of 727 patients who were diagnosed with ehrlichiosis or anaplasmosis at our center. The median age was 66 years and 65% of these patients were male. Seven infections occurred in solid organ transplant (SOT) recipients. The median age of the 7 recipients was 65 years, and all were white. Six of 7 were male, and outdoor hobbies and occupations will be assessed with a later chart review. The dates of organ transplantation ranged from 1982 to 2022. Ehrlichiosis was diagnosed in 3 patients following kidney transplantation. Anaplasmosis was diagnosed in 4 recipients, 3 of whom were heart transplant recipients, and 1 of whom was a liver transplant recipient. One patient was diagnosed using cell-free DNA testing. Three of 7 were diagnosed during a hospital stay. Infections were diagnosed between June and November, consistent with the typical exposure season. The median time between transplant and diagnosis of ehrlichiosis was 4 years (range, 2-5), and anaplasmosis was 8 years (range, 2-31). The liver transplant recipient died 5 months after anaplasmosis infection. All recipients were chronically immunosuppressed with agents including tacrolimus, mycophenolate mofetil, everolimus, cyclosporine, and/or azathioprine. Recipient comorbidities included systemic hypertension in 7, chronic kidney disease in 6, coronary artery disease in 5, heart failure in 2, and diabetes in 1. CRP was elevated in all recipients in whom it was measured (4 out of 7, ehrlichiosis 3, anaplasmosis 1). Leukopenia, anemia, thrombocytopenia (5 out of 7, all less than 100,000), hyponatremia, and elevated creatinine were noted in 5 out of 7 transplant recipients. Aspartate aminotransferase was elevated in 5 of 7 recipients (range, 46-920; upper limit normal 33 U/L), and total bilirubin was elevated in 1 recipient. All recipients were treated with 100mg Doxycycline BID for 9-15 days.

Conclusion

We report 7 transplant recipients who were diagnosed with ehrlichiosis or anaplasmosis at a single center. Immune-suppressed transplant recipients are typically at increased risk of poor outcomes following infection. Our report highlights the features of ehrlichiosis and anaplasmosis infection in SOT recipients.

Evaluation of ASEPSIS Score as a Secondary Endpoint in the Phase 3 SHIELD II Trial of D-PLEX100 in Colorectal Surgery

/in ,

Evaluation of ASEPSIS Score as a Secondary Endpoint in the Phase 3 SHIELD II Trial of D-PLEX100 in Colorectal Surgery

Authors:
Robert Sawyer, Livnat Levy, Elena Zafirovikj

Body of Abstract:
Background: D-PLEX100 is a novel extended-release doxycycline delivery system applied as a single dose to the surgical site prior to incision closure. D-PLEX100 was evaluated in SHIELD II, a randomized, double-blind, controlled trial in patients undergoing abdominal colorectal surgery. The study met its primary endpoint, demonstrating a significantly lower rate of treatment failure (defined as occurrence of any of the following: surgical site infection (SSI) in the target incision, re-intervention, or death) in the D-PLEX100 arm compared with standard of care (SoC). ASEPSIS score was used as a post-operative wound surveillance tool to standardize the assessment and grading of surgical site infections based on observable clinical criteria. The score is determined based on the following characteristics: Additional treatment, Serous discharge, Erythema, Purulent exudate, Separation of deep tissue, Isolation of bacteria, Stay duration as inpatient. It provides an objective and standardized approach to wound assessment, offering a quantitative measure of SSI severity derived from specific, predefined clinical findings. Herein, we report on key secondary endpoints related to ASEPSIS scoring.     

Methods: Abdominal colorectal surgery patients were randomized to receive D-PLEX100 plus SoC systemic antibiotics or SoC alone. The primary outcome was treatment failure defined as a composite of any one of the following: adjudicated incision SSI, re-intervention at the target incision site, and mortality. One of the three key secondary endpoints was determining ASEPSIS scores at each study visit as part of the surgical site assessment. Number of subjects with at least one ASEPSIS score of >20 within 30 days post abdominal (index) surgery were compared between both study arms in the intention-to-treat (ITT) population. The averages of cumulative ASEPSIS assessment scores were also reported for those who experienced adjudicated SSI within 30 days post index surgery. 

Results: SHIELD II had 798 subjects [n = 405 (D-PLEX arm) n = 393 (SOC arm)] in the ITT population. There was a 38% relative risk reduction in the primary outcome in the D-PLEX100 arm compared to SoC [44 (10.9%) vs 71 (18.1%); difference -7.2 (95% CI, -12.1 to -2.3), p = 0.0039]. The proportion of subjects with at least one ASEPSIS score >20 was lower in the D-PLEX100 arm [8/391 (2.0%)] compared to the SoC arm [21/377 (5.6%)]. The stratified risk difference was -3.5 (95% CI -6.2 to ‑0.8; p = 0.0103). The median cumulative ASEPSIS score was 175 in the D-PLEX arm and 216.25 in the SoC arm. The median cumulative ASEPSIS score within the SSI period was 110 in the D-PLEX arm and 160 in the SoC arm. 

Conclusions: The use of D-PLEX100 resulted in improved outcomes for abdominal colorectal surgery patients, specifically demonstrating reduction in treatment failures compared to SoC antibiotics. D-PLEX100 was also associated with lower rates of wound infections as determined by ASEPSIS scoring.

Experience with infections involving the rare non-fermenter Alcaligines faecalis

/in ,

Experience with infections involving the rare non-fermenter Alcaligines faecalis

Authors:
Pooja Ajith, Saron Araya, Sridha Gona, Aaron George, Hugo Bonatti

Body of Abstract:
Background: Alcaligines faecalis is a rare non-fermentative Gram-negative rod found in water and soil and decaying material. Other previous Alcaligines spp recently underwent reclassification within the Burkholderiales order such as Achromobacter xyloxidans. Up to 25% of humans are colonized with the organism and only a limited number of infections have been published with the majority diagnosed in immunocompromised individuals.

Methods: Our institutional database was searched for all infections caused by Alcaligines faecalis during a 4-year period. Two isolates initially reported as Alcaligines xyloxidans were excluded from the study.

Results: In total 38 isolates in 33 patients were identified. Median age was 60 (range 35.3-95.6) years; 63.6% were male. Rates of comorbid conditions were DM 46%, hypertension 399%, hyperlipidemia 36%, COPD 21%, CAD 18%, and malignancies 15%. 46% of individuals were obese and 61% were active smokers. Demographic, clinical, and microbiology data are shown in table 1. Bacterial growth pattern based on streak appearance for surgical specimens was reported light in 30%, moderate in 13% and heavy in 57%; 70% of infections were polymicrobial with staphylococci in 43%, streptococci in 13%, Gram-negative rods in 28%, and anaerobes in 18% as co-pathogens. Blood cultures accounted for 6%, drainage fluids/tissue specimens for 42% and wound cultures for 42% of specimens, 9% came from drained abscesses. Only 6% of isolates came from blood cultures. Alcaligines faecalis was predominantly isolated in lower extremity soft tissue infections (88%), upper extremities were involved in 3% and another 3% were intraabdominal infections. Surgical services submitted 33% of specimens, medical services including infectious diseases 27% and the emergency department 27%; 12% of specimens came from primary care physicians. Treatment for surgical infection included incision and drainage, debridement and amputation as indicated together with antibiotics according to sensitivity testing considering the high rate of mixed infections. 

Conclusion: Alcaligines faecalis in this series was predominantly isolated in patients with chronic lower extremity infections such as diabetic foot syndrome. Whereas the majority of these infections were treated successfully, patients with Alcaligines faecalis infections had a survival of only 73% after a 2-year follow-up reflecting the high rates of comorbidities in these individuals.

Female Sex is an Independent Risk Factor for Mortality Following Post-Burn Sepsis

/in ,

Female Sex is an Independent Risk Factor for Mortality Following Post-Burn Sepsis

Authors:
Diana Julia Tedesco, Maria Fernanda Hutter, Fadi Khalaf, Marc Jeschke

Body of Abstract:
Background: Sepsis is the leading cause of morbidity and mortality among burn patients, affecting approximately 1 in 5 adults. Previous research has shown that adult female burn patients face an increased risk of mortality after injury compared to their male counterparts. However, the underlying reasons for this disparity remain unclear. It is uncertain whether the increased mortality in females is due to differences in sepsis risk, distinct physiological responses to sepsis, or variations in the timing of sepsis onset. Thus, in this study, we aim to clarify whether sex-based differences in post-burn mortality emerge primarily after sepsis develops, rather than through differential incidence of sepsis.

Methods: We conducted a cohort study at two provincial burn centres between 2006 and 2025. Patients ≥18 years with acute burn injuries covering ≥5% of total body surface area were included and stratified based on biological sex recorded on admission and sepsis diagnosis using the Sepsis-3 and ABA guidelines. Multivariable logistic regression was used to evaluate (1) the association between sex and mortality, (2) the association between sex and sepsis incidence, and (3) sex differences in 30-day mortality among patients with sepsis, adjusting for age, burn size, inhalation injury, and sepsis onset timing. 

Results: A total of 1483 burn patients were included, including 223 males diagnosed with sepsis, 844 male controls, 82 females diagnosed with sepsis, and 334 female controls. At admission, female patients were older (median (IQR) 48 (36-62) vs. 46 (22-59) years, p=0.020), had similar burn size (median (IQR) 11 (7-19) vs 12 (8-21) %, p=0.068), and similar incidence of inhalation injury (26% vs 20%, p=0.462) compared to all male patients. Overall 30-day mortality was higher in females than males (8% vs 4%, p=0.002), an association that persisted after adjustment (adjusted OR 2.20, 95% CI 1.29–3.74). Sepsis incidence (20% vs 21%, p=0.668) and the timing of sepsis onset (median (IQR) 9.5 (5-15) vs 10 (6-15) days post-injury, p=0.503) did not differ between sexes. However, among sepsis patients, females had higher 30-day mortality compared to males (20% vs 10%, p=0.031) and female sex remained independently associated with mortality after adjustment for age, burn size, inhalation injury, and sepsis onset (adjusted OR 2.97, 95% CI 1.27–6.95).

Conclusion: Female burn patients experienced higher mortality overall, and this disparity became more pronounced following sepsis. These differences were not attributable to differential sepsis incidence or timing, suggesting sex-specific biological responses to infection rather than differential exposure risk.

Hospital Trauma Volume and the Risk of Post-Injury Sepsis in Blunt Intestinal Injury: A Nationwide Analysis

/in ,

Hospital Trauma Volume and the Risk of Post-Injury Sepsis in Blunt Intestinal Injury: A Nationwide Analysis

Authors:
Yasmin Arda, Ioannis Karikis, John Hwabejire, Michael DeWane, Charudutt Paranjape, Joshua Ng-Kamstra, Lydia Maurer, Matthew Bartek, Jonathan Parks, Ali Salim, George Velmahos, Haytham Kaafarani

Body of Abstract:
Background: Blunt intestinal injury (BInI) is rare and often difficult to diagnose, resulting in delay in intervention and worse outcomes. This study aimed to evaluate whether hospital BInI volume influences the risk of post-injury sepsis in patients with BInI.

Methods: The 2017-2020 ACS-TQIP database was used to identify patients ≥18 years of age with full-thickness ileal, jejunal, or colonic perforation secondary to blunt trauma. Hospitals were stratified into tertiles by annual BInI volume. Multivariable logistic regression adjusting for demographics, comorbidities, and injury characteristics/severity was used to assess the impact of hospital volume on the risk of post-injury sepsis. To examine the potential role of delayed recognition, sensitivity analyses were conducted by stratifying patients undergoing early versus delayed (>24 hours) surgical intervention.

Results: Of 4,005,762 trauma patients, 3,954 were included: 1,397 (35.3%) in low BInI volume, 1,373 (34.7%) in medium BInI volume, and 1,184 (30%) in high BInI volume hospitals. The mean age was 41±18 years, 37% were females, the mean injury severity score was 19±10, and the most common injury was jejunal or ileal perforation (66%). On multivariable analyses, high BInI volume was independently associated with a 45% lower risk of post-injury sepsis (aOR 0.55, 95% CI 0.36-0.86) compared to low BInI volume. This association was not statistically observed in sensitivity analyses stratified by timing of surgery.

Conclusions: High trauma hospital volume of BInI is independently associated with decreased risk of post-injury sepsis. The attenuation of this effect after stratifying by operative timing may partially be related to earlier surgical intervention at high volume hospitals.

Immune cell temporal specific signatures clock track recovery status of critical injury patients

/in ,

Immune cell temporal specific signatures clock track recovery status of critical injury patients

Authors:
TeDing Chang

Body of Abstract:
BACKGROUND

Despite major advances in resuscitation and supportive care, trauma remains a leading cause of death and disability worldwide, particularly among young adults. Critical illness following severe trauma represents one of the most complex and dynamic immunological conditions encountered in modern intensive care. Survivors frequently experience prolonged recovery characterized by immune dysregulation, secondary infections, and MODS. These adverse outcomes highlight the need to understand not only the magnitude but also the temporal dynamics of immune responses during recovery from critical injury.

METHODS

We conducted a large, prospective, multicenter study enrolling 243 critically injured trauma patients and 57 healthy volunteers. Using sorted immune cell populations including T cells, monocytes, and PMNs, we constructed a comprehensive, time-resolved transcriptomic cohort of immune responses from injury onset through clinical recovery or death.

We applied three complementary analytical approaches: DEGs, WGCNA, and SLIDE to select the recovery related gene signature. From these analyses, we identified nine gene signatures associated with complicated recovery and used them to train an immune recovery clock.

We developed a novel framework, termed Temporal Gap (TempoGap), to quantify the deviation between the predicted and actual post-injury time, thereby providing a metric of immune recovery delay or acceleration. To model temporal progression, we employed the LASSO regression, optimized through 5-fold cross-validation. Model stability and performance variability were estimated via 500 bootstrapped iterations, aggregated into an ensemble of LASSO-based temporal predictors.

RESULTS

All TempoGap models showed significant associations with recovery duration. Notably, the T cell latent factor–based TempoGap demonstrated the strongest predictive effect (HR = 1.5, p < 0.001), indicating that higher expression of this temporal module was associated with faster recovery. Given the importance of early prognostication, we further evaluated model performance within the first week post-injury. The models retained substantial predictive power during this period, with the monocyte complicated recovery gene TempoGap model showing the best performance on day six (HR = 2.0, p < 0.05). Importantly, TempoGap scores exhibited only weak correlations with conventional clinical indicators, suggesting that TempoGap captures unique biological dimensions of immune recovery not reflected by traditional scoring systems. CONCLUSIONS In summary, we established an immune recovery clock model and introduced TempoGap, a novel temporal deviation metric that predicts recovery status in critically injured patients. This approach offers a conceptual and analytical foundation for precision monitoring of immune recovery, enabling early outcome prediction and guiding targeted interventions to promote recovery after critical injury.

Immunomodulatory Effect of Phospholipid Nanoparticle, VBI-S for Treatment of Sepsis.

/in ,

Immunomodulatory Effect of Phospholipid Nanoparticle, VBI-S for Treatment of Sepsis.

Authors:
Gracy Rosario, Gelilla Daniel, Benjamin Edwards, Cuthbert Simpkins

Body of Abstract:
Background: 

Sepsis, a leading cause of morbidity and mortality in humans, is an inflammatory disease caused by a dysregulated host response to an infection. Increased antibiotic resistance and lack of FDA approved drugs significantly limits treatment options for sepsis. Recently, VBI-S, a phospholipid nanoparticle colloid, has proven effective in Phase 2a clinical trial for septic shock1. Efficacy of VBI-S for septic shock is currently being evaluated in Phase III open-label randomized controlled clinical trial. 

In sepsis, macrophages play an important role in inflammation, which is one of the causes of multi-organ damage. As preliminary evidence indicates that VBI-S is effective in sepsis, we hypothesized that VBI-S is immunomodulatory in nature, and capable of reducing inflammation in septic patients. Hence, the present study initially evaluated the immunomodulatory potential of VBI-S on pro-inflammatory macrophage functions. The study analyzed the effect of VBI-S on pro-inflammatory gene expression by M1 macrophages in an in-vitro culture model. 

Methods: 

Human THP-1 monocytes were differentiated to M0 macrophages by treatment with 25nM phorbol-myristate-acetate (PMA) for 48h and then to M1 macrophages by treatment with 100 ng/ml lipopolysaccharide (LPS) and 20 ng/ml interferon gamma (IFN-γ) for 48h. Subsequently, the M1 cells were treated with different concentrations of VBI-S, (1%, 0.1%, 0.01%, 0.01%) for 24h in presence of LPS and IFN-γ for 24h. Control included M1 cells cultured in media with or without the aqueous phase. Expression of pro-inflammatory genes (CXCL10, CCR7 and IL-1b) were analyzed by quantitative SYBR Green RT-PCR. Viability of the VBI-S treated adherent M1 cells were studied by MTT assay and non-adherent floating M1 cells evaluated by Trypan blue staining. Statistical analysis was performed by Brown-Forsythe and Welsch ANOVA, nonparametric t-test and Mann Whitney U test (Graph Prism Software). 

Results: 

VBI-S significantly decreased the levels of candidate pro-inflammatory CXCL10, CCR7 and IL-1b genes in M1 cells at 1%, 0.1%, and 0.01% concentrations (P<0.05). Furthermore, the number of viable adherent or non-adherent dead cells were unaltered at 1%, 0.1%, 0.01% and 0.001% VBI-S as compared to the respective controls.  Conclusion:  VBI-S has immunomodulatory properties as it reduces pro-inflammatory cytokine/chemokine gene expression by M1 macrophages. This study points towards immunomodulation being a key mechanistic function of VBI-S in treatment of sepsis. Futuristic studies are aimed at assessing VBI-S uptake by M1 macrophages, and subsequent mechanistic actions on pro-inflammatory signaling events. 1 Simpkins C, et al., 2024: Efficacy and safety of phospholipid nanoparticles (VBI-S) in reversing intractable hypotension in patients with septic shock: a multicentre, open-label, repeated measures, phase 2a clinical pilot trial. EClinicalMedicine 68:102430.

Impact of Thoracic Irrigation on Empyema Patterns in Trauma Patients

/in ,

Impact of Thoracic Irrigation on Empyema Patterns in Trauma Patients

Authors:
Katherine Russo, Joshua Preston, Wen Yang, Randi Smith, Jonathan Nguyen, Jason Sciarretta

Body of Abstract:
Background:

Empyema is a significant complication following traumatic hemothorax, contributing to prolonged hospitalization, increased morbidity, and the need for invasive interventions. Thoracic irrigation (TI) has emerged as a minimally invasive strategy for managing retained hemothorax, yet its influence on the development and microbiologic profile of subsequent empyema remains unclear. 

Methods:

A retrospective review was performed of trauma patients who underwent TI for retained hemothorax at an urban Level I trauma center between 8/2023 and 8/2025, identifying those who subsequently developed empyema. A historical comparison cohort included all traumatic empyema cases over a four-year period (1/2019 – 7/2023) preceding TI protocol implementation. Empyema was defined by the presence of positive pleural cultures. Data collected included demographics, injury patterns, pleural microbiology, ICU length of stay (LOS), and mortality. Continuous and ordinal variables were summarized as median [IQR] and compared using the Wilcoxon rank-sum test. Categorical variables were summarized as n (%) and compared using Fisher’s exact test.

Results:

Among 124 patients who underwent TI during the study period, the overall empyema rate was 2.4%. A total of 35 culture-proven empyemas were identified: 3 post-TI patients and 32 historical controls without TI. Patients who developed empyema after TI had a significantly longer time to diagnosis compared with those without TI (30 days [29.5-33] vs. 13 days [9.8-19.5], p=0.031). Polymicrobial empyema occurred less frequently in the TI group (33% vs. 72%), though this difference did not reach statistical significance. Rates of anaerobic infection were similar between cohorts (33% vs. 41%). No significant differences were observed in demographics, injury mechanism, ISS, thoracic AIS, ICU LOS, or mortality between groups (Table 1).

Conclusion:

In trauma patients who developed empyema, those who underwent prior thoracic irrigation demonstrated delayed onset of empyema and a lower observed rate of polymicrobial infection compared with patients who did not undergo TI. Although limited by the small number of post-TI empyemas, these findings suggest that TI may influence the microbiologic profile or temporal development of pleural space infection. Larger studies are needed to clarify the impact of TI on empyema pathophysiology and clinical outcomes following traumatic hemothorax.