The Relationship Between Length of Stay and Increased Infection Rates in Trauma Patients with a Delay in Hospital Discharge.

Authors:
Fabiana C Sanchez, Grant V Bochicchio, Leonardo J Diaz Gandica, Marco J Henriquez, Melissa Canas, Michael W Alchaer, Ricardo A Fonseca

Body of Abstract:
Background:
A substantial body of literature explores the impact of length of stay (LOS) on complications, infections, morbidity, and mortality. However, detailed analyses of specific intervals within LOS and their contributions to patient outcomes, particularly in the context of delayed discharge, remain limited.
Methods:
We conducted an analysis of our multidisciplinary, prospectively maintained trauma database, including all trauma patients with a delay in discharge from 2018 to 2023 at our Level 1 trauma center. Patients with discharge delays related to non-medical reasons and those with delays related to COVID-19 were excluded.
LOS was defined as the total hospital stay. LOS minus delay in discharge time (LOS – DDT) was defined as the duration from admission to when patients were medically ready for discharge ([Phase I]), and DDT was defined as the time from when patients were medically ready for discharge to their actual discharge ([Phase II]). Patients were stratified into two cohorts: those with delays due to infections (DDI) and those with delays due to non-infectious medical reasons (DDNI).
Results:
Among 160 patients with a delay in discharge, 15 (9.4%) were in the DDI group, and 145 (90.6%) were in the DDNI group. Patients in the DDNI group were more likely to be African American (31.7% vs. 0%, p=0.035). Compared to DDNI patients, those in the DDI group had significantly higher total LOS (17.4 days vs. 9.58 days, p<0.001), higher LOS - DDT (11 days vs. 6.18 days, p=0.008), and higher DDT (6.4 days vs. 3.4 days, p<0.001). In a multivariate analysis controlling for Injury Severity Score (ISS), female sex, Glasgow Coma Scale (GCS), and age, LOS - DDT was significantly associated with DDI patients, with an odds ratio (OR) of 1.133 (p=0.017). Conclusion: A higher LOS - DDT ([Phase I]) was associated with increased infection rates among trauma patients experiencing delayed discharge for medical reasons. Additionally, the presence of infection correlated with longer DDT ([Phase II]). These findings suggest the need for strategies to optimize hospital discharge processes and minimize prolonged hospital stays to reduce the risk of infections.

Corticosteroids in Trauma Patients: Impact on Outcomes and Infection Risk

Authors:
Ameera Mazraany, George C Velmahos, Haytham M A Kaafarani, Ikemsinachi C Nzenwa, John O Hwabejire, Josh Ng, Michael DeWane, Vahe S Panossian, Veerle van Zon

Body of Abstract:
Background: Corticosteroids impair healing and increase infection risk in surgical patients, but their impact on injured patients is less understood. This study aimed to examine the relationship between corticosteroid use and outcomes, including infection, in trauma patients.

Methods: Using the ACS-TQIP 2017-2020 database, all trauma patients 18 years or older were included. Corticosteroid use is defined in the database as the routine use of corticosteroid medications within 30 days before the injury for the management of a chronic medical condition. Multivariable logistic regression adjusting for demographics (e.g. age, sex), injury characteristics (e.g. mechanism, severity), and comorbidities (e.g. COPD, cirrhosis) was used to study the correlation between corticosteroid use and 1) mortality, 2) overall morbidity (e.g. acute kidney injury, stroke, pressure ulcers), and 3) infectious complications (e.g. sepsis, pneumonia, surgical site infection- SSI). SSI was assessed only in patients who underwent open surgical procedures.

Results: A total of 3,278,447 patients were included, 33,977 (1.0%) of which were corticosteroid users. The median age was 54 years, 90.3% had blunt trauma, mortality was 3.5%, overall morbidity was 7.3%, and 0.9% of patients developed an infectious complication. Among those who underwent open surgery, 0.6% developed a SSI. On multivariable analyses, corticosteroid use was independently associated with increased odds of mortality (OR 1.49, 95% CI 1.38–1.60, p<0.001), overall morbidity (OR 1.29, 95% CI 1.24–1.34, p<0.001) and infectious complications (OR 1.45, 95% CI 1.29–1.63, p<0.001), but not SSI (OR 0.81, 95% CI 0.55–1.18, p=0.270) [Figure 1]. Conclusion: Corticosteroid use in trauma patients is independently associated with increased risk of mortality, morbidity and infectious complications. These findings are important for patient counseling, benchmarking the quality of care of trauma patients, and for creating targeted interventions to decrease the risk of complications and infection in this patient population.

NPWT Management of Burst Abdomen in Pancreatic Juice Leak Following Distal Pancreatectomy Complicated by Tuberculosis

Authors:
Adianto Nugroho, Aditomo Widarso, Indah Jamtani, Rofi Y Saunat

Body of Abstract:
Introduction

Pancreatic fluid collection after distal pancreatectomy is common and can be managed conservatively. However, in some cases, it can lead to a devastating sequel, such as a burst abdomen.

Case Illustration

A 55-year-old male was consulted to the digestive surgery service with acute abdomen 24 hours after endoscopic drainage of pancreatic pseudocyst. The patient was diagnosed with cystadenoma of the pancreas tail, and the endoscopist internist successfully did endoscopic internal drainage of the cyst. However, one day after the procedure, the patient complained of severe abdominal pain accompanied by nausea, vomiting, and fever. His laboratory showed a marked leukocytosis, and a repeat abdominal CT scan showed a large amount of intraperitoneal air, which was a deliberate process of the endoscopic procedure. The decision to take the patient to surgery was made after 12 hours of conservative management failed. Upon exploration of the abdomen, we discovered pancreatic juice collection at the left paracolic gutter and pelvic floor. We performed a formal distal pancreatectomy, using a linear cutter, along with the spleen, and primary repair of the stomach. A large peritoneal drain was left at the peripancreatic space and another at the pelvic floor. The abdomen was cleansed with sterile saline and sutured closed. In the postoperative follow-up, pancreatic juice leakage through the peripancreatic drain was confirmed by amylase workup of the fluid. The pancreatic juice leaked through the abdominal incision as well, causing a burst abdomen by postoperative day 6. The patient had also developed an overwhelming post-splenectomy infection in the form of hospital-acquired pneumonia and was admitted to the ICU.
Meanwhile, pancreatic juice leakage was treated conservatively with optimal drainage through the peripancreatic drain, and a negative pressure wound therapy (NPWT) device was applied to treat the burst abdomen. The pathology examination of the specimen revealed a pancreatic pseudocyst with no signs of malignancy. We screened the patient for tuberculosis as another factor in his slow healing, and oral anti-tuberculosis (OAT) medicine was administered. Soon after the administration of OAT, the pancreatic juice drainage slowly reduced, and his abdominal wall defect showed progress. The patient recovered from OPSI, and his abdominal wall defect was healing well with VAC. He was discharged POD-30 for homecare with the VAC device still applied and was scheduled for abdominal wall closure within the next month.

Conclusion
Conservative management of pancreatic juice collection can be successful owing to detailed and comprehensive patient management. In endemic regions, tuberculosis should always be suspected in slow-recovering patients.

Understanding the Risk of Infection after Ultra Massive Blood Product Transfusion

Authors:
Courtney Meyer, Jason Sciarretta, Jonathan Nguyen, Lucy Hart, Randi Smith, Will Dunne, Zachary Grady

Body of Abstract:
Background
Ultra massive transfusion (UMT), defined as the transfusion of > 20 units of red blood cell product, is utilized for trauma patients in severe hemorrhagic shock. Patients requiring UMT have high rates of mortality and survivors are at risk for infectious complications. Yet, the exact risk remains unknown. The aim of this study was to quantify the risk of infections after UMT by evaluating the incidence, type, causative organisms, treatment and outcomes in trauma survivors.
Methods
A retrospective review was conducted at a Level 1 trauma center from 05/2018- 11/2021 of all trauma patients who received UMT. Data were collected from prospectively maintained registries & chart review including demographics, comorbidities, infection type (bacteremia, fungemia, surgical site infection (SSI), osteomyelitis, pneumonia (PNA), urinary tract infection, empyema, & Clostridioides difficile infection), culture data, and antibiotic administration & duration. Outcomes of hospital & intensive care unit (ICU) length of stay and mortality were also evaluated.
Results
A total of 193 patients were included in the study period with an overall mortality rate of 54.4%. Infection occurred in 25.4% of patients and at least one antibiotic dose was administered for 93.8% of patients. For those that survived at least 48 hours, 50.9% had an infection. There was no difference in age, sex, comorbidities, injury mechanism, injury severity, initial vital signs & laboratory values, and operative interventions required between those with and without an infection. Patients without an infection had more bowel injuries (49.0% vs. 20.7%, p<0.05). In these patients, the most frequent infections were PNA (33%), bacteremia (19.8%), and deep SSI (9.4%). There were 55 unique organisms with the most frequent being Enterobacter cloacae (8.9%), Pseudomonas aeruginosa (8.1%), Enterobacter aerogenes (7.0%), and Enterococcus faecalis (7.0%). In the entire cohort, there were 1,182 unique antibiotic courses with an average antibiotic duration of 3.5 days. The most frequently administered antibiotics were cefazolin (27.6%), vancomycin (17.0%), and metronidazole (11.3%). Infectious disease consultation occurred for 22.6% of patients who survived at least 48 hours. Patients with infection had more ventilator days (14.5 vs. 4 days, p<0.001) and longer ICU (24.5 vs. 6 days, p<0.001) and hospital (41 vs. 13 days, p<0.001) length of stay. Patients without an infection had higher rates of mortality (32.7% vs. 12.1%, p=0.01). Conclusion Patients who require UMT following traumatic injury have high rates of infection, most frequently pneumonia, bacteremia, & deep surgical site infections caused by a variety of microorganisms. Though this high rate exists, it may be similar to other critically injured patients who do not require UMT such that once a patient is stabilized from a resuscitation standpoint, their infection rates may be similar to their non-UMT counterparts.

Race impacts the incidence of complicated appendicitis with abscess in urban populations but not in rural populations

Authors:
Manuel Castillo-Angeles, Reza Askari, Saba Ilkhani, Tandis Soltani

Body of Abstract:
Introduction:
Complicated appendicitis, characterized by generalized peritonitis and abscess formation, serves as a valuable “surgical access proxy,” with outcomes independently influenced by the timing of presentation, irrespective of care quality. Studies using nationally representative data suggest that Black patients are more likely to present with complicated appendicitis compared to White patients. Urbanicity also shapes these outcomes, with higher rates of complicated appendicitis presentations in rural areas. This study examines how racial disparities are influenced by rural versus urban settings.
Methods:
We analyzed data from the Nationwide Inpatient Sample (NIS) from 1990 to 2021, focusing on appendicitis cases. Among 1,218,245 cases of appendicitis, 14.5% were classified as complicated, involving generalized peritonitis and abscess formation. Multivariate logistic regression models were used to assess the impact of rural versus urban settings and racial disparities on complicated appendicitis, adjusting for age, gender, comorbidities, and insurance status. Secondary analyses examined urbanicity by transfer status and geographic region.
Results:
As expected, complicated appendicitis was more prevalent in rural areas compared to urban settings (14.7% vs. 13.9%, p< 0.001), and Black patients had a higher probability of complications compared to White patients (White: 14.8% [95% CI: 14.7–14.9], Black: 17.3% [95% CI: 17.1–17.6]). Urban areas exhibited significant racial disparities, with Black patients facing higher odds of complications compared to White patients (aOR = 1.05, 95% CI: 1.02–1.08, p = 0.001), while no significant racial disparities were observed in rural areas (aOR = 1.08, 95% CI: 0.96–1.22, p = 0.188). Stratification by transfer status revealed that among non-transferred patients, complication rates were similar between rural and urban settings (aOR = 0.97, p = 0.067). However, transferred patients in rural areas had significantly higher odds of complications (aOR = 1.21, 95% CI: 1.08–1.37, p = 0.005). Regional analysis showed higher complication rates in the rural Midwest and South but no significant differences in the West and Northeast. Conclusions: Our findings indicate that rural areas experience higher rates of complicated appendicitis, but racial disparities in these settings are absent, likely due to shared geographic barriers that affect all populations equally. In contrast, urban areas exhibit significant racial disparities. The lack of significant differences in complication rates among non-transferred rural patients underscores the critical role of timely access over the quality of care in determining outcomes. Addressing system-level inequities in urban settings and expanding healthcare infrastructure, particularly in vulnerable regions, is essential for achieving equitable patient outcomes.

Quick and Short: The Impact of Time to Surgery and Operative Duration on Infection Risk in Emergency Surgery

Authors:
George C Velmahos, Haytham M A Kaafarani, Ikemsinachi C Nzenwa, John O Hwabejire, Michael DeWane, Vahe S Panossian, Yasmin Arda

Body of Abstract:
Background: Emergency surgery (ES) is associated with up to 8 times higher risk of perioperative morbidity and mortality compared to elective surgery. This study aimed to identify the impact of time to surgery and operative duration on infectious complications after ES.

Methods: Using the ACS-NSQIP 2013-2017 database, all ES patients 18 years or older were identified using the variable ‘Emergency Case’. Patients with preoperative sepsis were excluded. The primary outcome was postoperative infectious complications, defined as the presence of any of the following: sepsis, septic shock, surgical site infection (i.e. superficial, deep, and organ space), pneumonia, and urinary tract infection. Multivariable logistic regression analyses that adjusted for age, body mass index, comorbidities, surgical approach, and type of surgery were used to investigate the impact of time to surgery (more than 12 hours) and operative duration (more than 2 hours) on postoperative infection. Sensitivity analyses were performed to examine the same relationship in three subsets of patients: 1) exploratory laparotomy, 2) cholecystectomy, and 3) appendectomy.

Results: Out of a total of 4,299,148 patients, 264,213 ES patients were included, and 24,921 (9.43%) had infectious complications. On univariate analysis, patients with infectious complications were more likely to have preoperative comorbidities (e.g. obesity, diabetes, heart failure), an American Society of Anesthesiologists (ASA) grade of 3 or higher, an open surgical approach, longer time to surgery (50.4% vs. 39.4%, p < 0.001), and a longer operative duration (31.6% vs. 14.3%, p < 0.001). On multivariable analyses, time to surgery longer than 12 hours was associated with a 14% higher risk of postoperative infectious complications (OR 1.14; 95% CI 1.1-1.18), and operative duration greater than 2 hours was associated with twice the risk of infection (OR 1.99; 95% CI 1.91-2.08). Sensitivity analyses showed that prolonged surgery was associated with infectious complications in all cohorts, while delayed surgery was significantly associated with infectious complications only in exploratory laparotomy (Figure 1). Conclusions: Prolonged time to surgery and operative duration are independently associated with increased risk of infectious complications in ES patients. These findings highlight the importance of early and abbreviated surgical interventions in ES.

Utility of Serum (1,3)-β-D-glucan in Diagnosing and Treating Fungal Infections in Patients with Burn Injury

Authors:
Jeffrey Shupp, Lauren Moffatt, Melissa McLawhorn, Salma Hassan, Shawn Tejiram, Taryn Travis

Body of Abstract:
Introduction
Invasive fungal infections are life-threatening and occur in immunocompromised or critically ill patients. Patients with severe burn injuries are at increased risk for developing invasive fungal infection. Fungal infections are independently associated with increased mortality in burn patients, making early detection in this population paramount. Conventional methods of fungal culture may be unable to detect invasive fungal infection or result in delayed diagnosis. Fungitell® is a diagnostic test that detects (1, 3)-β-D-glucan (BDG) in serum to help diagnose invasive fungal infections. There is a paucity of literature examining the use of BDG assays in burn injured patients to diagnose invasive fungal infection. The aim of this work was to examine the use of BDG assays in a population of severe burn injured patients to determine its utility in guiding antifungal use as well as patient outcomes at a single institution.
Methods
Patients admitted at a single burn center with severe burn injuries estimated to be 15% total body surface area (TBSA) of burn or greater from July 2019 to June 2024 were retrospectively examined. Medical records were reviewed for demographics, injury characteristics, BDG levels, use of antifungals, presence of concomitant bacterial infections, antibiotic therapy, resolution of infection, and mortality. Friedman’s test was used to compare BDG levels across multiple measurements following the initiation of antifungal treatment. A Spearman’s correlation test to analyze the correlation between initial BDG levels and duration of antifungal treatment.
Results
During the study period, 44 patients had BDG levels reported. Of these patients, the most common etiology of burn injury was flame (84.4%). The median (IQR) total body surface area of burn injury was 33.4% (24.8-47.5). Overall mortality rate was 52.3%. Of the patients that had BDG levels measured, 37 were started on antifungal treatment after the initial level was collected (84.1%). Of these patients, only 12 (32.4%) had culture data positive for fungal infection. Discontinuation of treatment was guided by decreasing or negative BDG levels in 73.3% of patients. Among patients with three or more BDG levels obtained, there were increased BDG levels from the first to second measurements [123 (51.75-201.0) vs 159 (113.5-238.8); p=0.007]. However, levels decreased between the second and third measurements [159 (113.5-238.8) vs 126 (85.0-252.0); p=0.004] (Figure 1). There is a weak correlation between initial BDG level and the duration of antifungal treatment (Spearman ρ =0.3194, 95% CI 0.01579-0.5690; p=0.0173).
Conclusion
This data demonstrates that patients were initiated on antifungal treatment after collection of initial BDG level and continued despite negative culture data in many patients, suggesting the utility of BDG assay to guide antifungal use in patients with burn injury. Further study is necessary to elucidate best practices in BDG level use.

Diverticulitis in Pregnancy: A National Readmissions Database Study

Authors:
Alan Smith, Dayle Colpitts, Jessica Weaver

Body of Abstract:
Background
Current guidelines recommend employing standard procedural indications for the management of appendicitis and biliary disease in pregnant patients. Withholding the standard of care for these conditions based on pregnancy status alone is associated with an increase in maternal and fetal complications. There is a paucity of data however, regarding diverticular disease in obstetric patients, and there remains no formal management guidelines for this population. The incidence of diverticulitis in patients younger than 50 years old increased by 132% from 1980-2007. Furthermore, CDC data continues to demonstrate an upward trend in the mean age of pregnant individuals, with 19% of all pregnancies in the US occurring in women 35 years and older. National trends toward earlier onset diverticulitis combined with more advanced maternal age may converge, making diverticulitis an increasingly important etiology of obstetric abdominal infection. The purpose of this study was to identify the prevalence of diverticulitis is pregnant patients as well as the common complications associated with this disease process.

Methods
The National Readmissions Database (NRD) was queried for all visits in the first 3 quarters of each year 2016 to 2019, that included a patient with an ICD-10 code for diverticulitis and pregnancy. These entries were then further investigated for demographics, gestation, severity of disease, rates of intervention, readmissions, and pregnancy-related complications.

Results
470 patients were identified with a diagnosis of diverticulitis and pregnancy. Age at admission ranged from 17 to 52 years old with most patients presenting in the third decade (n=296, 62.9%). Most patients (n=340, 72.3%) were in the third trimester of pregnancy. 160/470 (34.0%) of admissions led to fetal delivery within 48 hours. 3.8% of patients (n=18) had complicated diverticulitis on index admission, with 8 patients undergoing diagnostic laparoscopy (1.7%), 2 exploratory laparotomy (0.4%), and ultimately 5 had colonic resection (1.1%). The 90-day readmission rate for perforation or abscess was 1.3%. There were no reported maternal deaths. The overall rate of fetal loss was 1.9%, and fetal complications was 2.3%. The most frequent complication events were abnormal fetal heart rate 20.8%, gestational diabetes 19.1%, hypertension 17.6%, fetal distress 5.3%, preterm labor 5.1%, hemorrhage 4.5%, and amniotic space infection 2.8%.

Conclusions
Diverticulitis can affect a broad range of pregnant patients and is not a benign condition. It should not be neglected in the differential workup for obstetric intraabdominal infections, especially given the population-level trends towards earlier onset of diverticular disease and advancing maternal age. Further investigation is warranted to better define the outcomes and guide best practices in this population.

Severe Group A Streptococcal Toxic Shock Syndrome after Total Thyroidectomy

Authors:
Christine Cocanour, Michael Campbell, Sarah Lucente

Body of Abstract:
Background: Group A streptococcal (GAS) infection with toxic shock syndrome (TSS) is a rare complication of elective surgery with an incidence of 0.8 to 3.4 per 100,000 and a mortality rate of 1.8-12%; mortality may exceed 50% with misdiagnosis or delay in treatment. High index of suspicion is required when patients do not follow anticipated post-operative trajectory to initiate early treatment and supportive care.

Methods: Case report and review of the current literature.

Results: Rapid onset of hypotension, renal impairment, acute respiratory distress syndrome and a generalized macular rash across the chest and abdomen were observed 24 hours following elective total thyroidectomy. The patient returned to the operating room the evening of post-operative day one where a punctate tracheal injury was observed and wound cultures obtained, the final isolate being streptococcus pyogenes. This was successfully managed with early washout of the operative field, aggressive volume resuscitation, broad spectrum antibiotics with protein synthesis inhibition, and hemodynamic support with four vasopressors including Angiotensin-2 for refractory toxic shock syndrome.

Conclusion: High index of suspicion for TSS as the cause of refractory shock is essential to prevent high rates of mortality. Early re-exploration and source control are essential; the addition of broad-spectrum antibiotics should include protein synthesis inhibition. Angiotensin-2 for refractory shock may be useful in decreasing vasopressor requirements and improve mortality in severe septic shock.

Impact of Whole Blood Resuscitation on Infection Risk in Trauma Patients Undergoing Surgery for Hemorrhage Control

Authors:
Amy McDonald, Chris Guidry, Jonathan Wisler, Joseph Cuschieri, Justin Dvorak, Laura Kreiner, Nimitt Patel, Rondi Gelbard, Todd Costantini

Body of Abstract:
Background: Blood transfusion is a cornerstone of trauma resuscitation. While early transfusion is associated with improved survival, blood transfusion is also associated with increased risk of infection, likely due to immunomodulatory mediators in blood products. Whole blood use is becoming increasingly common in trauma resuscitation, with a growing body of literature showing improved survival in patients that receive whole blood. However, less is known about the impact of whole blood on other outcomes, including infection. The aim of this study is to evaluate the association between whole blood and infection in trauma patients undergoing hemorrhage control laparotomy.

Methods: The American College of Surgeons Trauma Quality Improvement (TQIP) registry was queried for adult trauma patients between the years 2020 – 2022. Patients were included if they underwent early laparotomy for hemorrhage control, were undergoing resuscitation with PRBCs, and survived at least 30 days, or were discharged alive. Patient demographics, arrival vital signs, Injury Severity Score (ISS), units of packed red blood cells (PRBC), Fresh Frozen Plasma (FFP), Platelets (Plt), and Whole Blood (WB) transfused, length of stay, and surgical site infection (SSI) were collected. Primary outcome was the association between receipt of WB with SSI (defined in TQIP as a Deep SSI, Organ/Space SSI, or superficial SSI). For the primary outcome, receipt of WB was a binary variable (received or did not receive). Secondary outcome was the association between the total number of units of WB received and SSI. Subgroup analysis of patients who underwent a Massive Transfusion (MT) was also performed. Logistic regression was used to evaluate the association between WB and infection, with multivariate regression used to control for potential confounders, including age, race, ISS, shock index, and transfusion requirement.

Results: 19,144 patients were included in the study. Patients who received WB in the first 4 hours of arrival had higher odds for SSI on multivariate analysis compared to patients who did not (OR 1.27, 95% CI 1.11 – 1.46, p < 0.01). Increase risk of infection with WB was also seen in a dose dependent manner, with each additional unit of WB conferring a 6% increase in odds of infection (OR 1.06, 95% CI 1.03 – 1.09, p < 0.01). On subgroup analysis of patients who underwent MT, the association between receiving WB with and infection persisted on multivariate analysis (OR 1.31, 95% CI 1.11 – 1.55, p < 0.01) (Table). This was also seen in a dose-dependent manner for patients undergoing MT (OR 1.07, 95% CI 1.02 – 1.12, p < 0.01). Conclusions: Trauma patients undergoing hemorrhage control laparotomy who receive WB had increased odds of SSI compared to patients who did not. This was also seen in a dose-dependent manner, with increased SSI odds with each unit of WB. As WB use becomes more common, further research is needed to evaluate outcomes beyond survival.