Multicenter performance evaluation of the Unyvero IAI cartridge for detection of intra-abdominal infections

Abstract

Intra-abdominal infections (IAIs) are one of the most common type of infections in patients with sepsis and an important cause of death in intensive care units. Early detection and treatment are necessary to reduce patient complications and improve outcomes. The Unyvero IAI Application (Curetis GmbH) is the first automated assay to rapidly and simultaneously identify a large panel of bacteria, fungi, toxins, and antibiotic resistance markers directly from IAI-related samples. The assay was evaluated in four European clinical laboratories in comparison to routine microbiological practices. A total of 300 clinical samples were tested with an overall sensitivity of 89.3% and specificity of 99.5%, while time to results was reduced by an average of about 17 h compared to identification (ID) results and 41 h compared to full antibiotic susceptibility testing (AST) results. The Unyvero IAI was able to detect additional microorganisms compared with culture, in particular anaerobes, with most detections confirmed by sequencing. The most frequent resistance markers detected were mecA/mecC (n = 25), aacA4 (n = 20), and bla_CTX-M (n = 17) and carbapenemase genes were identified in nine specimens. Further studies are now required to determine the clinical impact of this new rapid test which could play a role in the successful treatment of IAI.

Appendix

Microbiology culture testing. The specimens were tested for pathogens using the standard procedure of each laboratory as described below.

Barts Health NHS Trust, London: Gram stain was performed on all sample types except drain fluids. Cell counts were performed on all ascitic fluids and peritoneal fluids and both specimens were concentrated prior to subculture on solid agar and incubation in BacT Alert (BioMerieux) blood culture bottles for 4 days. All sample types were cultured microaerophilically on blood agar at 37 °C for 48 h and also anaerobically on blood agar with a metronidazole disc (5 μg) for 48 h. Bile samples were also cultured on XLD agar (Xylose-Lysine-Desoxycholate) and incubated at 37 °C in O₂ for 24 h. If Actinomyces or Nocardia species were suspected, pus samples were cultured on FBH agar and incubated anaerobically for 10 days. Tissues were transferred (1 cm³) into 5 ml saline containing glass ballotini beads and vortexed at high speed for 10 s. The bead mill was then subcultured as described above, with the addition of a chocolate agar plate if indicated by the Gram stain result. All isolates were identified by MALDI-TOF MS (Bruker) and antibiotic susceptibility testing was performed on the MicroScan (Beckman Coulter), with the exception of streptococci, enterococci and anaerobes which underwent disc diffusion according to BSAC guidelines.

Toulouse University Hospital Center: Gram stain was performed on all sample types. Peritoneal fluids and bile were inoculated onto blood agar, chocolate agar (incubated microaerophilically), Colombia CAN agar, BromoCresol Purple agar, Sabouraud agar (incubated aerobically), and Wilkins-Chalgren agar (incubated anaerobically). Peritoneal fluids were also inoculated into blood culture bottles and subcultured to blood agar and BromCresol Purple agar when they became positive. For ascitic fluids, pancreatic fluids, and abscesses, samples were inoculated onto blood agar, and chocolate agar, (incubated microaerophilically), Sabouraud agar (incubated aerobically), blood agar (incubated anaerobically), and thioglycolate medium. Agar plates were read daily for up to 5 days and colonies was identified using MALDI-TOF MS (Bruker). Susceptibility testing was performed using the Vitek 2 system for Gram negative rods, Enterococcus species and staphylococci, with streptococci and anaerobes tested by disk diffusion according to the recommendations of the CA-SFM/EUCAST (Comité de l’Antibiogramme de la Société Française de Microbiologie/European Committee on Antimicrobial Susceptibility Testing).

Amiens University Hospital Center: Gram staining was performed on pus samples only. IAI samples were cultured onto blood agar under aerobic and anaerobic conditions; Schaedler Neo Vanco + 5% sheep blood (SNVS) agar and colistin-nalidixic acid (CNA) agar under anaerobic conditions; chocolate culture medium under microaerophilic conditions; and BHI broth under aerobic conditions. Plates were read at 24 h, 48 h, 72 h, and 96 h and colonies identified using MALDI-TOF MS (Bruker). Susceptibility testing was performed by agar diffusion following the CASFM 2013 recommendations.

The Great Romagna Hub Laboratory, Pievesestina: sterile fluids (ascitic, peritoneal, and biliary fluids and abdominal drainages) were collected in 4-mL heparinized tubes and routinely subcultured onto Columbia CNA agar + 5% sheep blood, chocolate agar PolyViteX, mannitol salt 2 agar, Mac Conkey agar, and Candida agar plus brain-heart infusion broth (BHI) at 37 °C. Gram staining was performed on all sample types. If anaerobic bacteria were suspected, the sample was cultured on three additional plates: Columbia agar + 5% sheep blood, Schaedler agar + 5% sheep blood, SNVS (BioMerieux). Plates were read at 24 h and 48 h (if cultured under aerobic conditions) or at 48 h and 72 h if incubated anaerobically. Turbid BHI tubes were subcultured only when solid plates showed no growth. Identification and susceptibility testing were performed using a Vitek2 instrument (BioMérieux). Additionally, a set of blood culture bottles were inoculated for each specimen and incubated at 37 °C for up to 5 days, with appropriate sub-culture depending on whether the aerobic or anaerobic bottle became positive.