An ex vivo study of infections of vascular grafts and endografts with scanning electron microscopy.

OBJECTIVE: Vascular graft and endograft infections (VGEIs) are complicated by high morbidity, mortality, and recurrence rates, notably due to biofilm formation on the graft surface, hardly dislodgeable by the sole anti-infectious treatment. The characteristics of this biofilm are still poorly documented. The aim of this study was to evaluate ex vivo biofilm on removed infected vascular grafts and endografts (VGEs). METHODS: Explanted VGEs were prospectively collected from 2019 to 2022 at Bordeaux University Hospital, France. Two samples per graft were used for scanning electron microscopy imaging; one was sonicated, and both grafts' sides were imaged. RESULTS: A total of 26 patients were included, 18 with VGEI, eight without any infection (endoleak and/or thrombosis), and 29 VGEs were collected. Microbial documentation was obtained in 83% of VGEIs. A thick layer of fibrin was visible on almost all grafts, mixed with a dense biofilm matrix on infected grafts visible as early as 1 month after the onset of infection. Bacteria were not always visualized on infected grafts' surface (80% on outer side and 85% on luminal side) but were surprisingly present on one-third of non-infected grafts. There was no significant difference between biofilm, fibrin, and microorganisms' distribution between the two grafts' sides. However, there were clear differences between infected and non-infected grafts, since immune cells, bacteria and biofilm were more frequently visualized on both sides of infected grafts (P < .05). Bacteria and immune cells although still visible, were significantly less present after sonication; the number of other elements including biofilm was not significantly different. CONCLUSIONS: The persistence of a thick layer of fibrin and biofilm embedding microorganisms on both sides of infected VGE even after 1 month of infection could be the explanation for the low success rates of conservative management and the usual need for graft removal to treat VGEIs.

Staphylococcus aureus Adhesion and Biofilm Formation on Vascular Polyester Grafts are Inhibited In Vitro by Triclosan.

OBJECTIVE: This study evaluated Staphylococcus aureus adhesion and biofilm formation on vascular grafts, which has seldom been investigated. METHODS: Adhesion and biofilm formation capabilities of three methicillin susceptible S. aureus strains (one biofilm forming reference strain and two clinical isolates) on five different vascular biomaterials were evaluated in vitro, including polyester (P), P + gelatin (PG), P + collagen (PC), PC + silver (PCS), and PCS + triclosan (PCST). Staphylococcus aureus adhesion on grafts was evaluated after one hour of culture and biofilm formation after 24 hours of culture by four different methods: spectrophotometry after crystal violet staining; sonicate fluid culture; metabolic assay; and scanning electron microscopy (SEM). Optical density was compared using Mann-Whitney pairwise test, and bacterial counts using Wilcoxon pairwise test. RESULTS: PCST grafts were most efficient in preventing S. aureus adhesion and biofilm formation, regardless of the method used. Bacterial counts and metabolic activity were significantly lower on PCST grafts after 24 hours (5.65 vs. 9.24 [PCS], 8.99 [PC], 8.82 [PG], and 10.44 log10 CFU/mL [P]; p < .015), and only PCST grafts were bactericidal. Biofilm formation was significantly diminished on PCST grafts compared with all other grafts (p < .001). Bacterial viability and metabolic activity after 24 hours were more impaired on PG compared with PC graft, and were surprisingly higher on PCS compared with PC grafts. Biofilm biomass formed after exposure to P, PG, PC, and PCS grafts was also reduced after 24 hours of incubation with PCST grafts (p < .001). After 24 hours, few bacteria were visible by SEM on PCST grafts, whereas bacterial biofilm colonies were clearly identified on other graft surfaces. CONCLUSION: Triclosan impregnated PCST grafts appeared to interfere with S. aureus adhesion from early stages of biofilm formation in vitro. Silver impregnation was not efficient in preventing biofilm formation, and collagen coating promoted S. aureus biofilm formation more than gelatin coating.

Essential Oil Composition of Phagnalon sordidum (L.) from Corsica, Chemical Variability and Antimicrobial Activity.

The chemical composition of Phagnalon sordidum (L.) essential oil was investigated for the first time using gas chromatography and chromatography/mass spectrometry. Seventy-six compounds, which accounted for 87.9% of the total amount, were identified in a collective essential oil of P. sordidum from Corsica. The main essential oil components were (E)-ß-caryophyllene (14.4%), ß-pinene (11.0%), thymol (9.0%), and hexadecanoic acid (5.3%). The chemical compositions of essential oils from 19 Corsican locations were investigated. The study of the chemical variability using statistical analysis allowed identifying direct correlation between the three populations of P. sordidum widespread in Corsica and the essential oil compositions they produce. The in vitro antimicrobial activity of P. sordidum essential oil was evaluated and it exhibited a notable activity on a large panel of clinically significant microorganisms.

Direct testing of bronchoalveolar lavages from ventilator-associated pneumonia patients.

In line with a rapid de-escalation of empirical antimicrobial therapy, this study assessed the validity of an E-test-based direct specimen testing method on bronchoalveolar lavage (BAL) samples from ventilator-associated pneumonia (VAP) patients. E-test strips were directly applied onto Mueller-Hinton agar plates seeded with BAL samples and read after 24 h of incubation. In parallel, the BAL samples were analyzed by the routine diagnostic laboratory. The microbroth dilution approach was used as a control method. In a cohort of 20 patients, 135 microorganism-antibiotic combinations were studied. Total agreement between the 2 methods was achieved for 88.9% combinations, with 1.5% very major errors (isolates susceptible by E-test and reported resistant by the diagnostic laboratory) and 9.6% major errors (isolates resistant by E-test and reported susceptible by the diagnostic laboratory). These results indicate that applying E-test directly on BAL samples is a promising method for obtaining susceptibility data after 24 h in critical patients with VAP.