Title of Project: Did we cause that colon surgical site infection (SSI) or was it PATOS (present at time of surgery)?

/in ,

Authors:
Anne Macy, Bonita Allen, David Gemmel, Jennifer Yockey, Lillianne Stanitsas

Body of Abstract:
Background: Complex PATOS SSIs are distinct from complex non-PATOS SSIs and are excluded from the SIR (Standard infection ratio) to ensure accurate hospital comparisons. For PATOS, intraoperative infection evidence must be documented within the operative body or findings. Per NHSN guidelines pathology, imaging and cultures alone cannot confirm PATOS; however, these criteria are used to diagnose an SSI.

Methods: Retrospective cohort chart audit analyzed 1379 colon-coded procedures from 27 hospitals from May 2023 to Jan 2024. Charts were audited from three days pre- surgery to 30 days post-surgery following NHSN surveillance guidelines. Data abstraction was performed by a surgical critical care provider and system wide infection preventionists. Inclusion/exclusion adhered to NHSN criteria. Variables included demographics, clinical data and CT imaging done within three days pre-op, pathology, PATOS documentation, plan to re-operate and intra-operative cultures. Data Sources included hospital-wide surveillance data and electronic medical record.

Results: In this series rate of colon SSIs was 13.4%: with PATOS infection excluded the rate was 9.7%. Of note, overall mortality was 6.2%. Multiple factors were associated with higher SSI rates: fever, hypotension and lactate >2 (<0.001). CT imaging criteria illustrating free air, abscess, dehiscence, anastomotic leak, and inflammation as well as the number of CT findings were also significant (< 0.001). While CT findings of ischemic bowel (0.014) was significant, phlegmon was not (0.335). Pathological findings of perforation, necrosis, and number of pathological findings excluding cancer were associated with SSIs (<0.001). In addition, cancer (0.012), purulence (0.013), Inflammation (0.040) and bowel ischemia (0.004) were also associated with SSIs. Although the rate of intra-op positive cultures was higher among SSIs, 10.3% vs 7.0%, it was not significant (0.119). Plan to re-operate (<0.001) was associated with higher rates of SSI. Although PATOS documentation (p <0.001) was significant several patients had either a CT finding or pathological finding (excluding cancer alone) without documentation of PATOS, which if included as PATOS, decreased the SSI rate to 5.3%. Multivariate logistic regression was used to control for NHSN factors that increase SSIs for colon operations: diabetes, ASA, Male, Age, BMI >/=30, non-primary closure, and oncology hospitals. Only closure technique, BMI and Age were associated with SSIs.

Conclusions: Currently accepted NHSN PATOS documentation underestimates the number of patients that have features that significantly increase risk that infection was PATOS.

Utility of Abdominal MRI in Pediatric Patients with Suspected Appendicitis Who Undergo Non Diagnostic Ultrasound in a Low Resource Setting

/in ,

Authors:
Anna Willingham, Jason McKee, Lena Perger, Noah Lucero, Rachel Sidebottom

Body of Abstract:
Authors: A.Willingham1, N. Lucero1, R. Sidebottom1, L.Perger2, J. McKee2
1University of New Mexico, School of Medicine, Albuquerque, NM, USA
2University of New Mexico, Department Of Surgery, Albuquerque, NM, USA

Background: Pediatric patients with suspected appendicitis often undergo abdominal ultrasound as the initial diagnostic modality. However, in pediatric surgery centers without access to high-quality, dedicated ultra-sonographers, the appendix is frequently not visualized, leading to non-diagnostic results. This poses a clinical challenge for pediatric surgery and emergency department providers in determining who should undergo additional diagnostic imaging, such as abdominal MRI. To address this issue, we evaluated 5 years of pediatric patients at the University of New Mexico Hospital who underwent abdominal MRI following non-diagnostic ultrasounds to identify predictive factors of MRI-confirmed appendicitis.

Methods: A total of 411 pediatric patients who underwent abdominal MRI following non-diagnostic abdominal ultrasounds were identified. A comprehensive chart review of each patient was performed to collect data on patient age, year evaluated, maximum temperature reported, symptom duration, white blood cell count (WBC), and C-reactive protein (CRP) levels. Radiologic findings, including specification descriptions from ultrasound, CT, and MRI studies, were also recorded. Data analysis involved calculating relative risk ratios, sensitivity, and specificity for each variable to determine its contribution to predicting a positive MRI result.

Results: Of the 411 patients included in the study, 9 (2.1%) had positive MRI findings consistent with appendicitis. All patients with appendicitis met at least one diagnostic criterion, such as elevated WBC or CRP. Among these patients, 8 out of 9 had elevated WBC counts (>11.0) and elevated CRP levels (>0.5), while 4 out of 9 had a temperature >38°C. The sensitivity and specificity for WBC were 0.89 and 0.63, respectively, while CRP had a sensitivity of 0.89 and specificity of 0.51. In contrast, temperature >38°C demonstrated a lower sensitivity of 0.44 but a higher specificity of 0.81. These findings highlight the diagnostic value of combining clinical and laboratory parameters to identify appendicitis.

Conclusion: Abdominal MRI following a non-diagnostic abdominal ultrasound for appendicitis demonstrated a high true negative rate, with elevated WBC and CRP levels serving as highly sensitive predictors. Given the high healthcare utilization cost of obtaining an abdominal MRI, particularly in a lower resource hospital setting, a higher threshold for obtaining an abdominal MRI should be adopted for pediatric patients.

Necrotizing Fasciitis Associate with Small Bowel Perforation During Liposuction and Brazilian Butt Lift

/in ,

Authors:
micharl cooper

Body of Abstract:
57 year old female presented to the emergency department by ambulance in septic shock 48. hours after undergoing liposuction and a Brazilian Butt Lift in a doctor’s office. Physical examination was consistent with an acute abdomen. A CT Scan of the abdomen showed a pneumoperitonium and free air under the diaphragm. The patient was resuscitated and taken emergently to the operating room. Exploratory laparotomy was performed, and 3 perforations were noted in the jejunum. Necrotic jejunum was resected and the viable ends were stapled separately. Necrotic subcutaneous tissue and fascia were debrided. Negative pressure wound therapy was place over the bowel, an the skin was closed. The patient was taken back to the operating room the following day, and necrotic abdominal wall was excision ally derided. Wound and blood cultures grew Psedommonas aeruginosa an Candida. The jejunum was reanastomosed, mesh was placed over the bowel, a tracheostomy was performed, and the patient was transferred to a burn unit for further wound care. Further debridements of the chest and abdominal wall were performed, and split thickness autografts were harvested and placed for coverage. The was discharged to rehablitation medicine.

The Effect of Ketone Bodies and Lipid Nanoparticles on Lethal Infection

/in ,

Authors:
Cuthbert Simpkins, Gelilla Daniel, Gelilla Daniel, Vidit Minda, William Gutheil

Body of Abstract:
Introduction: With the rise of multidrug resistant organisms, sepsis has become increasingly difficult to treat. The discovery of a treatment that is not solely reliant on standard antibiotics is vital. We decided on an approach that optimized macrophage activity. Studies have shown that the ketone, beta-hydroxybutyrate, suppresses differentiation of pro-inflammatory M1 macrophages and augments anti-inflammatory M2 macrophage lineages. Our previous studies demonstrated that lipid nanoparticles (VBI-1) reduce the expression of M1 macrophage genes. Hence, we tested the efficacy of ketones [Ethyl Acetoacetate (EAA), Beta-Hydroxybutyrate (BHB), and Ethyl Hydroxybutyrate (EHB)] and VBI-1 on the survival of rats subjected to lethal intra-abdominal infection.

Methods: We utilized a cecal ligation and laceration (CLL) model to induce lethal sepsis in adult male and female Sprague-Dawley rats. The cecum was ligated below the ileocecal junction with silk suture; then, a 0.5 cm laceration was made in the distal cecum. The cecum was then returned to the peritoneal cavity and the laparotomy was closed. The control rats received a intraperitoneal (IP) injection of isotonic aqueous solution in a volume equal to 8% of the rat’s body weight. The treatment group received an IP injection of the same solution with 125 mM of the ketones (EAA, BHB, EHB), VBI-1, or a combination of VBI-1 and ketones. The injection was delivered thirty minutes or five hours after the procedure. No source control, antibiotics, or intravascular fluids were given. Survival of rats beyond 24 hours was considered a success. Rats were observed for a maximum of thirty days. Statistical analysis was carried out using a one-way ANOVA and LSD post-hoc test.

Results: All control rats died within 24 hours of the CLL. Within thirty minutes post-CLL, there was no statistical significance between rats that received EAA and control (p=0.072). 60% survival was noted with BHB (p=0.009). Survival of 80% was shown with EHB (p=0.001). We further assessed the efficacy of EHB by performing the injection five hours post-CLL with a survival rate of 67% (p = 0.004). 40% survival was noted with VBI-1 alone (p=0.029). We then injected a combination of VBI-1 and EHB thirty minutes post-CLL which showed a survival of 100% of rats (p <0.0001). We further analyzed the combination at five hours post-CLL with a survival of 50% (p=0.015). To better understand the mechanism, we performed an in vitro study to evaluate the minimum inhibitory concentrations of EHB and BHB against E. coli and MRSA. The EHB exhibited the greatest level of bacterial suppression, most effectively against E. coli. Conclusion: EHB is a ketone with minimal research on its role in the mitigation of sepsis. It exhibits an improvement on survival with and without VBI-1 in a lethal model of sepsis without reliance on standard antibiotics. Continued research is being conducted to further understand its mechanism of action in macrophage activity.

OUTCOMES BY RACE AND ETHNICITY IN TWO DIFFERENT CARE MODELS FOR ACUTE APPENDICITIS

/in ,

Authors:
Leo Andrew Benedict

Body of Abstract:
Background: Despite the use of standardized clinical pathways for acute surgical diseases, disparities in population-specific outcomes persist. Our tertiary care center developed an Acute Care Surgery (ACS) perioperative clinical pathway for patients diagnosed with acute appendicitis to standardize care delivery and enhance efficiency. The impact of the pathway on reducing or exacerbating racial and ethnic disparities in surgical outcomes is unknown.

Methods: This retrospective cohort study involved consecutive patients admitted to our tertiary care facility with acute appendicitis who underwent appendectomy. Demographic and clinical data were abstracted from patient medical records. White, Black, and Hispanic patients were classified by pre- (January 1, 2016-July 31, 2018) and post-initiation (August 1, 2018-December 31, 2023) of the ACS clinical pathway. Outcomes, including post-operative LOS (primary), recovery room discharge rates and time from computed tomography (CT) to operating room (OR), were analyzed using Negative Binomial regression with an interaction between race and clinical pathway, adjusted for age and perforation. Statistical analysis was performed using SAS with a p-value <0.05 determined as significant. Results: Of 705 consecutive patients, 220 (31%) were treated prior to implementing the ACS pathway. Overall, 532 (75%) were White, 88 (12%) were Hispanic, and 85 (12%) were Black, with relatively similar distributions before and after the ACS pathway was introduced. In adjusted analyses, there was an interaction for LOS, with White, but not Black or Hispanic, patients experiencing a shorter LOS with the new pathway (p=.03, Figure 1). Improvements in discharge from the recovery room (p=.40) and CT to OR time (p=.39) was similar across all demographic groups after introducing the ACS pathway (Figure 1). Conclusion: While a new ACS clinical pathway improved a range of outcomes, White patients had a greater improvement in shorter post-operative LOS than Black or Hispanic patients. Further prospective investigation into the factors associated with disparate post-operative LOS improvements among non-White race groups and solutions to mitigate barriers to timely discharge is warranted.

Assessing the Gap: SIS Guidelines vs. Real-World Antibiotic Use in Facial Fractures

/in ,

Authors:
Ahmad El Nouiri, Hadi Hamdan, Jeffrey Johnson

Body of Abstract:
Introduction:
Facial fractures account for over 400,000 ED visits annually in the United States. They are managed operatively (open reduction and internal fixation, ORIF), non-operatively (closed reduction), or via observation with diet and activity modification. Regardless of management, antibiotics are commonly prescribed. The Surgical Infection Society (SIS) published 2020 guidelines advocating against antibiotic use in all pre- and post-operative as well as non-operative cases (upper face, mid-face, or mandibular).
Methodology:
This retrospective chart review evaluates antibiotic prescribing practices at a quaternary regional academic level one trauma center for patients with isolated facial fractures. It assesses potential antibiotic days saved by adhering to SIS guidelines. Data for all admitted patients above 18 presenting to Henry Ford Hospital’s trauma service with a facial fracture from January 2019 to August 2024 were acquired from the trauma registry. Patients were manually screened to ensure eligibility. Patients with clear antibiotic indications (e.g., open fractures, complex lacerations) were excluded. Antibiotic timing classification followed SIS guidelines: pre-op (>1 hour before surgery), peri-op (within 1 hour before surgery to 24 hours after), and post-op (>24 hours after surgery). Statistical analysis included descriptive methods for quantitative variables and Chi-Square for qualitative variables.
Results:
A total of 119 patients fit the criteria. 89.1% received antibiotics, with 57.6% receiving at least two. 80.2% of the antibiotics given were against the recommendations of the SIS. Total antibiotic days were 584. Mandibular (49.6%) and midface (41.2%) fractures were most common. The antibiotic administration rate for mandibular fractures was 98.3%. Operative management accounted for 93.2% of cases, mostly open. Antibiotic use did not significantly differ between operative and non-operative management (p=0.18). The rate of antibiotic use between open and closed operative management did not differ significantly either (p=0.99). Cephalosporins and Penicillins were most prescribed. Peri-op antibiotics were used in 89.9% of cases, but non-guideline operative use (pre and post-op) rates were 47.5% and 63.6%, respectively. No significant association existed between guideline publishing and non-guideline use (p=0.60). Surgical site infection and Clostridium Difficile complication rates were 0%.
Conclusion:
There is a clear discordance between antibiotic prescribing practices at our institution and the SIS guideline recommendations for facial fractures. Implementing guidelines could yield substantial cost savings and is vital in combating antibiotic resistance.

Are Fewer Damage Control Laparotomies Associated with a Decrease in the Rate and Severity of Surgical Site Infections?

/in ,

Authors:
Chelsea Guy Frank, John A. Harvin, Lillian S Kao, Renee W, Green, Stephanie Martinez Ugarte

Body of Abstract:
Background: Local damage control laparotomy (DCL) rates have decreased since the implementation of a multi-faceted quality improvement intervention during the first half of the last decade. Given the association of DCL with both organ space and incisional surgical site infections (SSIs), we hypothesize that SSI rates and severity based on Clavien-Dindo (CD) classification have decreased over the same time period.

Methods: A retrospective study was performed of all patients (>16 years) who underwent trauma laparotomy at a large metropolitan Level 1 Trauma Center from 01/2011-12/2020. Patients who did not survive >48 hours were excluded. The data was divided into period 1 (1/2011-12/2015) and period 2 (1/2016-12/2020). The Center for Disease Control and Prevention definitions of organ space (OS-SSI) and superficial and deep incisional (SD-SSIs) were used. CD classification was further grouped into minor and major, with CD grade 1 (opening the wound) and 2 (antibiotics) considered minor and CD grade ≥ 3a (procedure requiring sedation, single and multiorgan failure, and death) considered major. Univariate and multivariable analyses were performed (p<0.05). Results: Of 1975 patients that met inclusion criteria, the median age was 33 (23, 46) years old. A total of 18% (355) of patients developed an SSI, 18.5% (179) in period 1 and 17.4% (176) in period 2 (P= 0.5). Between periods 1 and 2, there was an increase in injury severity score (ISS, 19 [ IQR 10, 29] vs 21 [IQR 12, 34], P<0.001) and penetrating mechanism of injury (426 (44%) vs. 513 (51%), p=0.002), while there was a decrease in the use of DCL (283 [29%] vs. 153 [15%], P=0.001). Despite the decrease in use of DCL, the rates and severity of SSIs remained stable between the two time periods (Table). Length of hospital stay decreased (LOS, 10 [IQR 6,21] vs 9.5 [IQR 5, 20] P= 0.04). On multivariable analysis, DCL, penetrating mechanism of injury, lower emergency room systolic blood pressure, higher wound classification, large bowel resection, and splenectomy were predictors of CD grade ≥3a SSIs. Conclusions: Contrary to our hypothesis, despite decreases in DCL use, the rates and severity of SSIs after trauma laparotomy in our patient population have not decreased but remained stable over the recent decade. The decreased use of DCLs may have been offset by the increase in penetrating injuries. Other potentially modifiable risk factors need to be identified to be able to impact the frequency and severity of SSIs after trauma laparotomy.

Retained Hemothorax and Empyema After Trauma: Predictive Factors and Cost

/in ,

Authors:
Benjamin Stocker, Briana Britton, Daniel VanDerPloeg, Dante Yeh, Eric Campion, Ernest Moore, Fredric M Pieracci, Kassandra Samuel, Mandi Roberts, Renaldo Williams

Body of Abstract:
Background: Thoracostomy tubes (TT) placed for traumatic hemothorax can be complicated by retained hemothorax (rHTX) with a recorded incidence ~20%, and rHTX may lead to complications including empyema. Secondary interventions to treat rHTX include additional TT, instillation of lytic agents (tPA), and operative intervention (thoracoscopy or thoracotomy). The purpose of this quality improvement project was to review our institution’s incidence and consequences of rHTX and identify opportunities for improvement.
Methods: This retrospective, single-center study from Jan 2021 to Dec 2023 included adult trauma patients receiving TT for hemothorax. Patients were excluded if: non-survivable injuries on arrival, died or transitioned to comfort care within 48 hours, bilateral TT, TT placed at an outside institution or in the operating room during thoracic procedure, TT for pneumothorax only, video-assisted thoracoscopic surgery (VATS) for rib fixation, and VATS or thoracotomy within 6 hours of arrival. rHTX was defined as residual pleural blood on subsequent imaging after TT or dictated as an intraoperative finding. Data collected included demographics, TT-specific details, hospital charges, and clinical outcomes. Subjects were divided into two groups based on the presence/absence of rHTX and group characteristics and outcomes were compared. We additionally divided patients into thoracic irrigation vs. no thoracic irrigation and compared rHTX incidence.
Results: A total 113 patients were included, of whom 30 (27%) developed rHTX (Table. The majority of TT were placed for hemopneumothorax (68%) and placed in the Emergency Department (67%). Univariate analysis did not identify any predictive factors for rHTX. All rHTX patients required additional intervention, most commonly VATS, which was performed in two-thirds of rHTX patients. Three patients received tPA and two required additional TT. Thoracotomy (3%) and empyema (2%) were overall rare. However, patients with a rHTX had a trend towards more commonly developing an empyema (7% vs. 0%, p=.069). rHTX patients had significantly longer hospital length of stay (13 [6-24] vs. 8 [4-15] days, p=.044), and total hospital charges were higher for the rHTX group ($231K vs. $167K USD, p=.056), though not statistically significant due to sample size. Thoracic irrigation was uncommon (n=19; 17% overall) and the difference between rHTX and no rHTX groups was not statistically significant (10% vs. 19%, p=.393). When comparing patients receiving thoracic irrigation (n=19) vs. no thoracic irrigation (n=94), the rate of rHTX was 16% vs. 29%, respectively (p=.393).
Conclusions: At our institution, the rate of retained hemothorax is higher (27%) than published rates in the literature (~20%), and the incremental cost of retained hemothorax is approximately $64,500 USD. Thoracic irrigation utilization is low and routine implementation may be considered to decrease the rate of retained hemothorax in a cost-effective manner.

Readability of Infection-Related Patient Education Materials

/in ,

Authors:
Krislynn Mueck, Lillian S Kao, Parker Towns, Stephanie Martinez Ugarte, William Rieger

Body of Abstract:
Background: Patient education materials (PEMs) are intended to increase patient understanding of their condition and to reduce readmissions. To maximize accessibility, PEMs should be readable at a grade level of 6 or below. Little information is available on the variation of PEMs across hospital types or infectious conditions. We hypothesize that the mean readability of infection-related PEMs does not meet the goal grade level of 6 or below.
Methods: PEMs were obtained from a tertiary referral center and a safety net hospital by searching existing electronic medical record-linked repositories with key words related to nosocomial infections: “infection,” “surgical site infection,” “urinary tract infection,” “pneumonia,” and “central line associate blood stream infection.” PEMs that did not explain the definition, prevention, or treatment of infections were excluded. Each PEM’s readability was assessed with 3 well-validated measures: Flesch Kincaid Grade Level (FKGL), Simple Measure of Gobbledygook (SMOG), and FORd-CAylor-STicht (FORCAST); each is based on a separate syntactical formula with lower grades indicating a text is easier to read. These measures were compared across hospitals systems, readability measure, and infection type. Descriptive analysis, univariate, and multivariable analysis were performed.
Results: Of 91 PEMs, 49 (53%) were from the safety net hospital; 13 (14%) focused on SSIs and 11 (12%) were designated as “easy to read” from the safety net hospital. The mean readability grade level for infection-related PEMS across all measures was 8.3 (SD 2.12) though using the FKGL measure alone demonstrated a lower median readability level of 5.8 (IQR 5.64-5.96). The different readability measures differed significantly in their scoring (X2 of 197, p<0.001), with the FKGL estimating the lowest grade level in 2-way and 3-way comparisons (p<0.001). “Easy-to-read” PEMs had a lower mean readability score of 7.2 (SD 2.05), compared to others at 8.46 (SD 2.05). PEMs from the tertiary referral center and safety net hospital demonstrated similar mean readability scores with a median of 8.35 (SD 1.99) and 8.26 (SD 2.23), respectively without a significant difference (t= -0.36, 95% CI -0.6 – 0.33, p=0.361). Conclusion: The readability of infection-related PEMs was similar across hospital settings and specific infections but differed based on assessment tool used. Although "easy-to-read" PEMs had lower readability grade level scores, consistent with their label, opportunity still exists across all infection-related PEMs for improvement. Further study is needed to determine if improving the readability of infection-related PEMs is associated with better understanding among patients with low health literacy.

Is mobility after trauma laparotomy associated with surgical site infections?

/in ,

Authors:
Lillian S Kao, Mokunfayo Fajemisin, Renee W, Green, Stephanie Martinez Ugarte, William Rieger

Body of Abstract:
Background:
Traumatic injury can drastically affect mobility in patients, regardless of surgical intervention. Enhanced Recovery After Surgery (ERAS) protocols that include early mobilization after elective and emergency surgery have shown a decrease in surgical site infections (SSIs). Although only certain ERAS components are applicable to emergency surgery, early mobilization may be feasible. However, few studies have explored the existing prevalence of early mobilization after emergency trauma laparotomy or explored the association between physical mobility and SSIs. We hypothesized that degree of post-laparotomy, as denoted by the Activity Measure for Post-Acute Care (AMPAC) score, and time to mobilization, are associated with lower SSI rates in trauma patients.

Methods:
A retrospective analysis of adults (>16 years old) who underwent trauma laparotomy from 6/2022 – 6/2023 was performed. Patients who died without or never received physical therapy (PT) were excluded. Patient demographics, hospitalization details, SSI status per Centers for Disease Control guidelines, and inpatient PT details of days to mobilization, as denoted by PT initiation, and AMPAC scores at first post-operative PT session were collected from the medical record. AMPAC scores are derived from 6 activities of daily living scored 1-4 for a max of 24 meaning full function. Descriptive statistics, univariate, and multivariable analysis were performed using known SSI-associated factors.

Results:
Of 160 patients, 120 were male (75.5%), their median age was 34 (IQR 23-45) with a median injury severity score of 22 (IQR 12-32). Their median AMPAC score was 16 (IQR 9-24) and median time to mobility was 2 days (IQR 0-4) (Table). Seventeen (11%) patients had an SSI. Univariate regression showed no significant association between initial AMPAC scores and SSIs (OR 1.0, 95% CI 0.91-1.09, p=0.93), but time to mobilization was significantly associated (OR 0.93, 95% CI 0.91-1.09, p=0.01). In separate models, each adjusted for injury severity score, AMPAC scores were not associated with SSIs (OR 1.0, 95% 0.88-1.10, p=0.85), while days to mobilization remained significantly associated (OR 1.04, 95% CI 0.94-1.02, p<0.01). Conclusions: This single center analysis showed that time to initial mobilization but not initial physical function per AMPAC scores was associated with SSI. While multiple confounders may impact these metrics, the data suggest an opportunity for earlier patient mobilization after emergency trauma laparotomy, potentially as a component of an ERAS protocol.