Citations

Abstract Pancreatic fungal infection (PFI) in patients with necrotizing pancreatitis can lead to significant morbidity and mortality. The incidence of PFI has increased during the past decade. Our study aimed to provide contemporary observations on the clinical characteristics and outcomes of PFI in comparison to pancreatic bacterial infection and necrotizing pancreatitis without infection. We conducted a retrospective study of patients with necrotizing pancreatitis (acute necrotic collection or walled-off necrosis), who underwent pancreatic intervention (necrosectomy and/or drainage) and had tissue/fluid culture between 2005 and 2021. We excluded patients with pancreatic procedures prior to hospitalization. Multivariable logistic and Cox regression models were fitted for in-hospital and 1-year survival outcomes. A total of 225 patients with necrotizing pancreatitis were included. Pancreatic fluid and/or tissue was obtained from endoscopic necrosectomy and/or drainage (76.0%), CT-guided percutaneous aspiration (20.9%), or surgical necrosectomy (3.1%). Nearly half of the patients had PFI with or without concomitant bacterial infection (48.0%), while the remaining patients had either bacterial infection alone (31.1%) or no infection (20.9%). In multivariable analysis to assess the risk of PFI or bacterial infection alone, only previous pancreatitis was associated with an increased odds of PFI vs. no infection (OR 4.07, 95% CI 1.13-14.69, p = .032). Multivariable regression analyses revealed no significant differences in in-hospital outcomes or one-year survival between the 3 groups. Pancreatic fungal infection occurred in nearly half of necrotizing pancreatitis. Contrary to many of the previous reports, there was no significant difference in important clinical outcomes between the PFI group and each of the other two groups.

Background. Surgical-site infection (SSI) is the most common early infectious complication after pancreas transplantation (PT). Although SSI has been shown to worsen outcomes, little data exist to guide optimal choices in perioperative prophylaxis. Methods. We performed a retrospective cohort study of PT recipients from 2010–2020 to examine the effect of perioperative antibiotic prophylaxis with Enterococcus coverage. Enterococcus coverage included antibiotics that would be active for penicillin-susceptible Enterococcus isolates. The primary outcome was SSI within 30 d of transplantation, and secondary outcomes were Clostridioides difficile infection (CDI) and a composite of pancreas allograft failure or death. Outcomes were analyzed by multivariable Cox regression. Results. Of 477 PT recipients, 217 (45.5%) received perioperative prophylaxis with Enterococcus coverage. Eighty-seven recipients (18.2%) developed an SSI after a median of 15 d from transplantation. In multivariable Cox regression analysis, perioperative Enterococcus prophylaxis was associated with reduced risk of SSI (hazard ratio [HR] 0.58; 95% confidence interval [CI], 0.35-0.96; P = 0.034). Anastomotic leak was also significantly associated with elevated risk of SSI (HR 13.95; 95% CI, 8.72-22.32; P < 0.001). Overall, 90-d CDI was 7.4%, with no difference between prophylaxis groups (P = 0.680). SSI was associated with pancreas allograft failure or death, even after adjusting for clinical factors (HR 1.94; 95% CI, 1.16-3.23; P = 0.011). Conclusions. Perioperative prophylaxis with Enterococcus coverage was associated with reduced risk of 30-d SSI but did not seem to influence risk of 90-d CDI after PT. This difference may be because of the use of beta-lactam/beta-lactamase inhibitor combinations, which provide better activity against enteric organisms such as Enterococcus and anaerobes compared with cephalosporin. Risk of SSI was also related to anastomotic leak from surgery, and SSI itself was associated with subsequent risk of a poor outcome. Measures to mitigate or prevent early complications are warranted.

AbstractBackgroundIn the management of Gram‐negative bloodstream infection (GN‐BSI), short antimicrobial courses have been increasingly demonstrated to be non‐inferior to prolonged therapy, with lower risk of Clostridioides difficile infection (CDI) and emergence of multi‐drug resistant (MDR) organisms. However, immunocompromised hosts were excluded from these studies. We investigated outcomes of short (≤10 days), intermediate (11–14 days), and prolonged (≥15 days) antimicrobial durations for GN‐BSI in neutropenic patients.MethodsA retrospective cohort study was conducted on neutropenic patients with monomicrobial GN‐BSI between 2018 and 2022. The primary outcome was a composite of all‐cause mortality and microbiologic relapse within 90 days after therapy completion. The secondary outcome was a composite of 90‐day CDI and development of MDR‐GN bacteria. Cox regression analysis with propensity score (PS) adjustment was used to compare outcomes between the three groups.ResultsA total of 206 patients were classified into short (n = 67), intermediate (n = 81), or prolonged (n = 58) duration. Neutropenia was predominantly secondary to hematopoietic stem cell transplantation (48%) or hematologic malignancy (35%). The primary sources of infection included intra‐abdominal (51%), vascular catheter (27%), and urinary (8%). Most patients received definitive therapy with cefepime or carbapenem. No significant difference in the primary composite endpoint was observed for intermediate versus short (PS‐adjusted hazard ratio [aHR] 0.89; 95% confidence interval [95% CI] 0.39–2.03) or prolonged versus short therapy (PS‐aHR 1.20; 95% CI 0.52–2.74). There was no significant difference in the secondary composite endpoint of CDI or MDR‐GN emergence.ConclusionOur data suggest that short antimicrobial courses had comparable 90‐day outcomes as intermediate and prolonged regimens for GN‐BSI among immunocompromised patients with neutropenia. image

Background Pediatric cardiac surgery is associated with abnormal coagulation, bleeding, and nearly ubiquitous transfusions. With the popularization of patient blood management, attempts are being made to decrease liberal transfusions by administering prothrombin complex concentrates (PCCs). The safety and efficacy of PCCs in adult cardiac surgery has been studied extensively, but only few reports address this in children. We performed an observational study focused on transfusion requirements after off-label use of activated PCC Factor Eight Inhibitor Bypassing Activity (FEIBA) as an adjunct to post-cardiopulmonary bypass (CPB) hemostatic protocol. Methods We reviewed the medical records of children ≤15 kg undergoing cardiac operations with CPB between May 2018 and March 2022. A propensity score (PS) analysis was performed to identify matched pairs of patients who did and did not receive FEIBA. Results Out of 210 patients who met the inclusion criteria, 44 patients received FEIBA. Propensity score-based analysis identified 40 matched pairs of patients with similar baseline characteristics. There was no statistically significant difference in the primary outcome—the volume of transfusion after CPB, which included all allogeneic blood products and salvaged washed red cells administered after protamine. Specifically, FEIBA patients were transfused 28 (22-34) mL/kg and controls were transfused 22 (11-49) mL/kg, P = .989. Upon arrival to ICU, the FEIBA group averaged an 8% lower international normalized ratio, compared with the controls ( P = .009) and a 1.08 g/dL higher hemoglobin ( P = .050). Neither difference remained significant on POD 1. Conclusions In this exploratory study, we found no change in transfusion requirements after CPB despite FEIBA administration.

Beta-hemolytic streptococci are common causes of bloodstream infection (BSI). There is emerging data regarding oral antibiotics for BSI but limited for beta-hemolytic streptococcal BSI.

AbstractBackgroundObesity has been associated with an increased risk of vascular complication during percutaneous coronary intervention, but there are no data on the risk of vascular complication during percutaneous transfemoral transcatheter aortic valve insertion (TAVI).ObjectivesWe hypothesized there would be a similar increased risk associated with TAVI.MethodsWe reviewed the records of 1176 patients who received percutaneous transfemoral transcatheter aortic valve insertion from September 2015 to September 2020. All patients received 1) preoperative computed tomoraphy angiography assessment of the abdomen and pelvis to delineate iliofemoral artery anatomy, 2) ultrasound‐guided percutaneous femoral arterial access, and 3) pre‐closure of the delivery sheath femoral access site. Vascular complication was recorded based on definitions set forth by Valve Academic Research Consortium 3.ResultsThe median age of patients was 81 years, and 60% were men. The median body mass index (BMI) was 29 kg/m2 (range, 11–67), and 91 (8%) patients had a value ≥40 kg/m2 (i.e., morbid obesity). Delivery sheath size was 14‐French in 859 (73%) patients, 16‐French in 311 (26%), and 18‐French in 6 (1%). Vascular complication occurred in 53 (5%) patients, including 39 (7%) among the first half of procedures and 14 (2%) among the second half (p < 0.001). When stratified by obesity status (BMI < or ≥30 kg/m2, p < 0.001), the complication rate was 4% in nonobese patients and 5% in obese patients. Multivariable analysis showed no overall association between risk of vascular complication and BMI categories (p = 0.583)BMI continuous values (p = 0.529), or sheath size (p = 0.217).ConclusionsObesity is not associated with a vascular complication during percutaneous transfemoral transcatheter aortic valve insertion. The operation should not be denied in obese patients.

Abstract OBJECTIVES Delayed sternal closure (DSC) after cardiac surgery is a temporizing measure to address coagulopathy or haemodynamic instability after cardiac surgery. We sought to study: (i) indications and temporal trends for DSC, (ii) factors associated with time to chest closure and (iii) its impact on short-term and long-term outcomes. METHODS From January 2007 to December 2017, 494 patients (median age 67 years, 66% males) required DSC after cardiac surgery. Medical records were reviewed for indications, risk factors, time to DSC and outcomes. Multivariable Cox regression via landmark analysis of 486 5-day survivors was used to investigate the impact of time to chest closure on early and late survival. RESULTS Coagulopathy and haemodynamic instability were the most common indications. Median time to chest closure was 2 days. Pre-/intraoperative extracorporeal membranous oxygenation, severe right ventricular dysfunction and diabetes mellitus were associated with longer time to chest closure. Longer time to closure was associated with increased risk of operative complications and operative mortality, but did not have a statistically significant association with late mortality. Increasing age, pulmonary hypertension and a greater number of prior sternotomies were also found to be associated with overall mortality. CONCLUSIONS While longer time to chest closure was associated with increased rates of operative complications and operative mortality, it did not reveal a statistically significant association with long-term survival.

Abstract Background Staphylococcus aureus has high mortality rates, which is influenced by several factors related to the host-pathogen interaction. The current study aims to provide a contemporary evaluation of predictors of 6-month mortality rate in a population-based cohort of incident Staphylococcus aureus bacteremia (SAB) cases. Methods Retrospective population-based study of 541 adult residents of Olmsted County, MN with monomicrobial SAB from January 1, 2006 through December 31, 2020. The study’s primary outcome was 6-month mortality rate, and a survival analysis was prepared by the Kaplan-Meier method. Multivariable Cox regression was used to investigate risk factors associated with 6-month mortality. Results The median (IQR) age of 541 patients with SAB was 66.8 (54.4-78.5) years and 39.6% were female. The median Charlson Comorbidity Index (CCI) was 2 (1-4). SAB cases were healthcare-associated in 49.2% and 56.2% were methicillin-susceptible. Twenty percent of SAB cases had an unknown infection source and 38.3% had complicated bacteremia. There were 7 relapses within the first 12 weeks of follow-up, corresponding to a cumulative incidence rate of 1.3%. There were no re-infections through the remainder of the 6-month follow-up, although there were 2 recurrences after 7 months. Eighty-nine (16.5%) patients died within 1 month of SAB diagnosis and 144 (26.7%) patients died during the 6-month follow-up period. In a multivariable analysis, older age, elevated CCI, unknown source of SAB, and intensive care unit (ICU) admission were significant predictors of an increased 6-month mortality rate. In contrast, presence of diabetes mellitus was significantly associated with a decreased 6-month mortality rate; also, undergoing infectious diseases consultation was associated with reduced mortality in the first 2 weeks, but not for the remainder of the 6-month follow-up period. Conclusion To our knowledge, the current investigation represents the only contemporary population-based study determining significant predictors of mortality in SAB patients, namely, older age, higher CCI, unknown source of SAB, and ICU admission. This study also described one of the lowest 30-day mortality rates reported over the past 2 decades. Figure 1.Survival and reinfection rates of patients with SAB over 6 months. Abbreviations: SAB, Staphylococcus aureus bacteremia. Foot notes: The overall survival curve was derived by the Kaplan-Meier estimator, while the curve for reinfection was estimated by the cumulative incidence function taking into account the competing risk of death. Disclosures Larry M. Baddour, M.D., Boston Scientific: Advisor/Consultant|Botanix Pharmaceuticals: Advisor/Consultant|Roivant Sciences: Advisor/Consultant|UpToDate, Inc.: Royalty payments - authorship duties.

Abstract Background There are anatomical, pharmacokinetic, and joint biomechanical differences between males and females. Whether these differences result in differing presentation, treatment tolerability and outcomes in periprosthetic joint infection (PJI) has not been widely evaluated. Methods We undertook a retrospective case matched study of patients with staphylococcal PJI managed with two-stage exchange arthroplasty. In order to control for differences other than sex which may influence outcome or presentation, males and females were case-matched for age, causative organism (Coagulase-negative staphylococci vs Staphylococcus aureus) and joint involved. The two matched groups were analyzed for differences in duration of symptoms, inflammatory markers, duration of hospital stay, antimicrobial side effects, post-operative complications and recurrence rate. Recurrence of infection was defined as per Musculoskeletal Infection Society criteria for PJI and included both relapse with original microorganism or with an additional organism. Results 78 pairs of males and females with staphylococcal PJI were successfully matched. There were no significant baseline differences by sex, except for greater use of chronic immunosuppression among females (16.7% vs. 4.1%; p=0.012). Clinical characteristics and outcomes of the patients are summarized in table 1. We did not detect a statistically significant between the outcomes of the two groups. Overall, 16 recurrent infections occurred during a median follow-up time of 2.9 (IQR: 1.5 - 5.3) years. The 3-year cumulative incidence of treatment failure depicted in figure 1 was 16.1% for females, compared with 8.8% for males (p=0.434). Conclusion Success rates for PJI treated with two-stage exchange arthroplasty are high consistent with previously reported literature. A larger cohort of patients may be required to detect differences in relapse rates. This study included only staphylococcal PJI and was thus unable to evaluate the differences in microbiology of PJI between males and females, which could also impact outcomes. This retrospective case-matched study did not detect any significant differences in outcomes between males and females with staphylococcal PJI who underwent two stage exchange arthroplasty. Disclosures Gina A. Suh, M.D., Adaptive Phage Therapeutics (APT): Grant/Research Support|Adaptive Phage Therapeutics (APT): Dr. Suh received grants and has an equity and royalty-bearing know-how agreement with Adaptive Phage Therapeutics but is unrelated to project Aaron J. Tande, M.D., Uptodate: Dr Tande has received an honorarium from Uptodate for medical writing unrelated to this project.

Abstract Background Bloodstream infection (BSI) following arterial aneurysm repair may signify vascular graft infection (VGI). The incidence of BSI and its implications in patients with arterial grafts have received scant attention with limited contemporary study. Therefore, the aim of the current investigation is to describe the incidence and epidemiology of BSI following arterial aneurysm repair in a population-based cohort. Methods Retrospective case-cohort study using the expanded Rochester Epidemiology Project (e-REP) to identify all arterial aneurysm repairs performed in adults (&gt;18 years) residing in eight counties in southern Minnesota between January 1, 2010 and December 31, 2020. All repairs were performed at Mayo Clinic Rochester. Electronic health records were screened for initial BSI episode following aneurysm repair and complicating VGI. Results Overall, 643 patients underwent 708 aneurysm repairs: 417 endovascular and 291 open surgical repairs. Forty-two patients developed BSI during 2,992 cumulative person-years of follow-up. The median time from repair to BSI was 3.1 [interquartile range 0.5, 6.4] years. The BSI incidence rate, in cases per 1000 person-years, was 14.0 (95% CI 10.1-19.0) overall, including 18.2 (95% CI 11.9-26.7) following open surgical repair and 10.2 (95% CI 5.8-16.6) following endovascular repair (p=0.148). Thirty-nine (92.9%) of the BSI cases were monomicrobial, with 18 (46.2%) due to Gram-positive organisms and 21 (53.8%) due to Gram-negative organisms. Of the 42 patients with BSI, 10 (23.8%) had VGI. VGI occurred in 8/18 (44.4%) monomicrobial Gram-positive BSI as compared to 2/21 (9.5%) monomicrobial Gram-negative BSI (p=0.013). The 3-year mortality rate following BSI was 52.3%. Conclusion There was no statistically significant difference in incidence rates of BSI following open surgical repair and endovascular repair. Higher rates of VGI were seen with Gram-positive BSI as compared to Gram-negative BSI. Disclosures Larry M. Baddour, M.D., Boston Scientific: Advisor/Consultant|Botanix Pharmaceuticals: Advisor/Consultant|Roivant Sciences: Advisor/Consultant|UpToDate, Inc.: Royalty payments - authorship duties.