Abstract
Objectives
A multicentre study evaluating NG-Test DetecTool OXA-23 for the detection of OXA-23 carbapenemase directly from positive blood cultures (PBCs).
Methods
The NG-Test DetecTool OXA-23 is an immunoassay that integrates a sample preparation device. We evaluated NG-Test DetecTool OXA-23 on 189 spiked and 126 clinical PBCs. The clinical samples’ standard-of-care procedure consisted of bacterial identification from the first day of positivity by MALDI-TOF MS, conventional culture and antimicrobial susceptibility testing. The immunoassay results were verified molecularly. The strains used for the spiked samples consisted of well-characterized Acinetobacter baumannii and Proteus mirabilis strains.
Results
The NG-Test DetecTool OXA-23 was evaluated on 315 PBCs and revealed sensitivity of 100% (95% CI: 98.21%–100.00%) and specificity of 100% (95% CI: 96.73%–100.00%). It provided 204 true-positive results for OXA-23 in 196 bottles with carbapenem-resistant A. baumannii (CRAB) and 8 bottles with carbapenem-resistant P. mirabilis and also provided 111 true-negative results. There were no false-positive and no false-negative results. Among the 315 PBCs studied, 83 clinical blood cultures collected in the ICU of a Greek university hospital, which were tested prospectively, all yielded CRAB, and OXA-23 was correctly detected in all samples from the first day of positivity using the NG-Test DetecTool OXA-23.
Conclusions
The NG-Test DetecTool OXA-23 has exhibited excellent sensitivity and specificity for OXA-23 detection in PBCs and can provide valuable information for appropriate selection of antibiotic therapy and early implementation of infection control measures.
Introduction
The emergence and spread of carbapenemase-producing Gram-negative bacteria (CPGNB) pose a significant global health concern. In addition to carbapenems, CPGNB are usually resistant to many other antibiotics.1 Early diagnosis of CPGNB infections is crucial for appropriate treatment and infection control. OXA-23 is the main carbapenemase identified in the Acinetobacter genus.2 Moreover there have been reports of OXA-23-producing Proteus mirabilis from France, Singapore, Belgium, Germany, Finland and other countries.3–7 The blaOXA-23 gene in Acinetobacter is most often found in several transposons associated with ISAba1 (e.g. Tn2006 or Tn2008) or ISAba4 (Tn2007).8 In P. mirabilis, the blaOXA-23 gene is mainly carried on the chromosome, except for some isolates from France and Singapore that carried blaOXA-23 on an AbaR4-like structure on an untypeable plasmid.9 Due to the association of blaOXA-23 with mobile genetic elements, it can spread through horizontal gene transfer among microorganisms, and its early detection is of great importance. New diagnostic tests are being developed for the rapid phenotypic or molecular detection of OXA-23-producing Gram-negative bacteria. We conducted a multicentre study that included eight European hospitals for the assessment of an immunochromatographic assay coupled with a filtration/incubation device (NG-Test DetecTool OXA-23) in order to rapidly detect OXA-23 directly from positive blood cultures.
Materials and methods
Participating centres
The study took place in eight European microbiology laboratories: Hospital Clinic of Barcelona (Barcelona, Spain), Attikon University Hospital (Athens, Greece), Assistance Publique Hôpitaux de Paris (AP-HP), Paris-Saclay, Bicêtre Hospital Le Kremlin-Bicêtre, (Paris, France), Semmelweis University (Budapest, Hungary), Bellvitge University Hospital (Barcelona, Spain), Florence University Hospital (Florence, Italy), Saint Laszlo-South Pest Hospital (Budapest, Hungary) and Centre Hospitalier Amiens (Amiens, France).
Study period and samples
The study was conducted from March 2022 till October 2022. We tested 315 positive blood cultures. Eighty-three positive clinical blood cultures with carbapenem-resistant Acinetobacter baumannii (CRAB), from different ICU patients, were analysed prospectively in Attikon University Hospital, Athens, Greece. Another 35 positive clinical blood cultures with A. baumannii from 35 patients were tested in AP-HP Paris-Saclay, Bicêtre Hospital, Florence University Hospital, Semmelweis University and Amiens University Hospital. One hundred and eighty-one blood cultures spiked with A. baumannii were also tested in the participating labs, 62 of which were spiked with A. baumannii susceptible to carbapenems, 103 were spiked with OXA-23-producing A. baumannii, eight were spiked with OXA-40 carbapenemase-producing A. baumannii, 5 were spiked with OXA-58-producing A. baumannii and 3 were spiked with NDM-producing A. baumannii. All the spiked isolates belonged to the bacterial collections of the Hospital Clinic of Barcelona, AP-HP Paris-Saclay, Bicêtre Hospital and Attikon University Hospital and had been tested molecularly for the presence of blaOXA-23, blaOXA-40, blaOXA-58 and blaNDM as described before.10,11
Moreover, 16 blood cultures positive for P. mirabilis were tested. Among these samples, eight were clinical samples prospectively tested in Florence University Hospital and Saint Laszlo-South Pest Hospital, Budapest (from eight patients) while eight blood cultures were spiked with OXA-23-producing P. mirabilis molecularly characterized from the bacterial collection in AP-HP Paris-Saclay, Bicêtre Hospital.6,10 For spiking experiments, blood culture bottles considered negative after 5 days of incubation were inoculated with 100 μL of a bacterial suspension containing 105 cfu/mL. Bottles were incubated and tested by NG-Test DetecTool OXA-23 once they were detected as positive by the automatic incubator.
Identification and susceptibility testing in blood cultures
Blood cultures were incubated using the automated systems BACTEC FX (Becton Dickinson, Hunt Valley, MD, USA) or BacT/ALERT Virtuo (bioMérieux, France). After positivity, Gram staining and subculture to appropriate solid culture media were performed. Identification was accomplished by MALDI-TOF MS (Microflex LT, Bruker Daltonik, Germany or Vitek MS, bioMérieux, France) on the first day of positivity from the broth’s sediment and was verified by MALDI-TOF MS from a colony the next day. Susceptibility testing was performed by disc diffusion directly from the positive blood culture and the semi-automated systems Vitek 2 Compact (bioMérieux, France) or Phoenix 50 (Becton Dickinson). Results were interpreted according to the EUCAST guidelines.
Sample processing and OXA-23 carbapenemase detection
The OXA-23 detection was performed using the NG-Test DetecTool OXA-23, which is a filtration/incubation device coupled to a lateral flow immunoassay strip (NG-Biotech, Guipry, France). For the clinical samples, the test was performed after the identification of Gram-negative bacteria by MALDI-TOF MS. OXA-23 detection was performed on the first day of blood culture positivity, the processing time was 3–5 min, and the result was read within 15 min. The process, in brief, was: 0.5 mL of the positive blood culture broth was collected from the bottle and mixed with 250μL of cell lysis buffer. After incubation for 2–5 min, 500μL of the lysed sample was transferred to the filtration membrane, and 300 μL of extraction buffer was added. The extract that potentially contained the resistance enzyme then passed through the membrane and migrated through the test device. Two red lines were interpreted as a positive result and one red line in the control region signified a valid negative result. The sample processing device called the BL-DetecTool, has been used previously with other lateral flow immunoassay tests.12 The NG-Test DetecTool OXA-23 is a prototype for research use and hasn’t yet been approved for diagnostic use. The positive blood culture broth that was used for the test from the clinical samples was taken from the remaining broth after performing all standard-of-care blood culture procedures on the first day of positivity.
Molecular methods
Verification of the OXA-23-positive results was done by molecular detection of blaOXA-23 by PCR and sequencing of selected amplicons.10
Statistical analysis
The calculation and statistical analysis of sensitivity and specificity (with 95% CI) were determined using MedCalc software.13
Results
Overall, the NG-Test DetecTool OXA-23 tested as true positive for the 196 OXA-23-producing A. baumannii and the 8 OXA-23-producing P. mirabilis and tested as true negative for the 111 OXA-23-non-producing A. baumannii and P. mirabilis (Tables 1 and 2). The results were verified as true positives and true negatives after molecular testing for blaOXA-23. In all samples tested, the immunochromatography’s positivity was revealed within 15 min and in most of them in less than 5 min, with easy-to-read strong bands. All immunochromatography results were valid since the control line was present. The sensitivity of the test was 100% (95% CI: 98.21%–100.00%) and the specificity was 100% (95% CI: 96.73%–100.00%).
Table 1.
Hospital | True positive | True negative | False positive | False negative |
---|---|---|---|---|
Attikon Athens, clinical | 83 | 0 | 0 | 0 |
Attikon Athens, spiked | 0 | 17 | 0 | 0 |
HC Barcelona, clinical |
— | — | — | — |
HC Barcelona, spiked |
5 | 19 | 0 | 0 |
Firenze, clinical | 5 | 0 | 0 | 0 |
Firenze, spiked | — | — | — | — |
Paris Saclay, clinical | 0 | 18 | 0 | 0 |
Paris Saclay, spiked | 13 | 19 | 0 | 0 |
Bellvitge, clinical | — | — | — | — |
Bellvitge, spiked | 27 | 23 | 0 | 0 |
Semmelweis Budapest, clinical | 4 | 7 | 0 | 0 |
Semmelweis Budapest, spiked | 8 | 0 | 0 | 0 |
Amiens, clinical | 1 | 0 | 0 | 0 |
Amiens, spiked | 50 | 0 | 0 | 0 |
Total | 196 | 103 | 0 | 0 |
Table 2.
Hospital | True positive | True negative | False positive | False negative |
---|---|---|---|---|
Firenze, clinical | 0 | 4 | 0 | 0 |
Firenze, spiked | — | — | — | — |
Paris Saclay, clinical | — | — | — | — |
Paris Saclay, spiked | 8 | 0 | 0 | 0 |
South Pest, clinical | 0 | 4 | 0 | 0 |
South Pest, spiked | — | — | — | — |
Total | 8 | 8 | 0 | 0 |
Discussion
This is the first study assessing the sensitivity and specificity of NG-Test DetecTool OXA-23, with the majority of positive results coming from clinical bacteraemic cases. We have shown that this test can serve as a valuable tool in the rapid detection of OXA-23- producing A. baumannii and other Gram-negative bacteria directly from positive blood cultures in less than 20 min. The sample processing device (BL-DetecTool), which is coupled to the immunoassay, simplifies the direct testing from the blood culture since there is no need for centrifugation of the sample and the lysed blood culture broth after the extraction step is filtered through the BL-DetecTool. Because of this simple innovative device included in the test, no equipment is needed and the immunochromatographic assay provides a fast and simple-to-read result.
The NG-Test DetecTool OXA-23 test exhibited sensitivity and specificity of 100% in the detection of OXA-23-producing bacteria.
Since OXA-23 is the most prevalent carbapenemase in Acinetobacter,2 we suggest using this rapid test on the positive blood culture on the first day of positivity after identification of Acinetobacter by MALDI-TOF MS on the sediment. Identification by MALDI-TOF MS from the sediment is easy to perform and provides pathogen identification a day earlier. In this study the identification on the first day of blood culture positivity from the sediment was verified by MALDI-TOF MS from the colony on the second day (the results were the same). This is a low-cost method that can provide important information for the adequate treatment of bacteraemic patients a day earlier than the usual conventional blood culture susceptibility testing. The test can be useful also in early detection of bacteraemia by Proteus producing OXA-23. Moreover, the rapid detection of patients carrying OXA-23-producing bacteria is important for the timely implementation of infection prevention and control measures. Our results are similar to those of previous studies that had evaluated the use of immunochromatography in carbapenemase detection (KPC, VIM, NDM, IMP, OXA-48) directly from blood cultures.12,14 The NG-Test DetecTool OXA-23 can be used for the rapid detection of OXA-23 directly from the positive blood culture bottle after identification of A. baumannii or P. mirabilis. It can provide early diagnosis, especially in severely ill bacteraemic patients, such as those hospitalized in ICUs, in hospitals where there is a high prevalence of OXA-23-producing CRAB or in patients who had previous known carriage of OXA-23-producing CRAB in surveillance cultures.
There are studies evaluating the use of other immunochromatographic tests for OXA-23 detection using bacterial colonies, namely the OXA-23 K-SeT and RESIST ACINETO (Coris BioConcept, Belgium). Both tests have exhibited 100% sensitivity and specificity for OXA-23 testing from colonies.15,16 Nevertheless, our study differs because it evaluates rapid OXA-23 detection directly from positive blood cultures of bacteraemic patients.
A limitation of this study is that the assay was tested in a few A. baumannii carrying carbapenemases other than OXA-23. Another limitation is that a clonality investigation of the isolates was not performed. Nevertheless, it is known that most of the CRAB recovered in recent years from highly endemic regions belong to international clone (IC)2 and produce OXA-23 and this was the case also in Greek hospitals17,18 and possibly in our setting. In that respect, we believe that the results of the study would be of clinical value for the numerous international regions that are highly endemic for OXA-23-producing CRAB. Of note, the BL-DetecTool can also be used for testing directly for β-lactamases in other samples such as rectal swabs and urine.13 At the moment, the BL-DetecTool and the NG-Test DetecTool OXA-23 are ‘research use only’ assays.
In conclusion, the NG-Test DetecTool OXA-23 assay seems to be a reliable, rapid and inexpensive test that could be incorporated in the routine workflow of clinical microbiology laboratories in areas with high prevalence of OXA-23-producing CRAB infections.
Acknowledgements
This study was part of the AMR DetecTool project. We want to acknowledge the contribution of the AMR DetecTool team: Gian Maria Rossolini, Chiara Chilleri, Alberto Antonelli, M. Angeles Domínguez-Luzón, María Bonilla-Moreno, Eszter Ostorhazy, Katalin Kristof, Jordi Bosch, Mariana Fernandez, Styliani Louka and Clara Ballesté-Delpierre.
Contributor Information
Alexandra Vasilakopoulou, Laboratory of Clinical Microbiology, Attikon University General Hospital, Medical School, National and Kapodistrian University, Athens, Greece.
Thierry Naas, Team ‘Resist’, INSERM Unit 1184, Faculty of Medicine, Université Paris-Saclay, Bacteriology-Hygiene Unit, Assistance Publique-Hôpitaux de Paris, AP-HP Paris-Saclay, Bicêtre Hospital Le Kremlin-Bicêtre, Paris, France.
Camille Gonzalez, Team ‘Resist’, INSERM Unit 1184, Faculty of Medicine, Université Paris-Saclay, Bacteriology-Hygiene Unit, Assistance Publique-Hôpitaux de Paris, AP-HP Paris-Saclay, Bicêtre Hospital Le Kremlin-Bicêtre, Paris, France.
Jordi Vila, Department of Clinical Microbiology, Hospital Clinic of Barcelona, Barcelona, Spain; Institute for Global Health (ISGlobal), University of Barcelona, Barcelona, Spain; Department of Clinical Microbiology/CIBER de Enfermedades Infecciosas, University of Barcelona, Madrid, Spain.
Dóra Szabo, Institute of Medical Microbiology, Semmelweis University, Budapest, Hungary.
Eleonora Riccobono, Department of Experimental and Clinical Medicine, University of Florence, Florence, Italy; Department of Medical Biotechnologies, University of Siena, Siena, Italy.
Katalin Kamotsay, Central Microbiology Laboratory, Central Hospital of Southern Pest National Institute of Hematology and Infectious Diseases, Budapest, Hungary.
Sophie Reissier, Department of Bacteriology, Amiens University Hospital, Amiens, France.
Dàmaris Berbel, Microbiology Department, Hospital de Bellvitge, IDIBELL, UB, CIBERES, Barcelona, Spain.
Albert Zoltan Aszalos, Health Services Management Training Centre, Semmelweis University, Budapest, Hungary.
Magda Rosenmoller, Department of Operations, Information and Technology, IESE Business School, Barcelona, Spain.
Milovan Stankov-Puges, R&D Department, NG Biotech, Guipry, France.
Panagiota-Christina Georgiou, Laboratory of Clinical Microbiology, Attikon University General Hospital, Medical School, National and Kapodistrian University, Athens, Greece.
Sophia Vourli, Laboratory of Clinical Microbiology, Attikon University General Hospital, Medical School, National and Kapodistrian University, Athens, Greece.
Hervé Volland, Département Médicaments et Technologies Pour la Santé (DMTS), Université Paris Saclay, CEA, INRAE, SPI, 91191 Gif-sur-Yvette, Paris, France.
Spyros Pournaras, Laboratory of Clinical Microbiology, Attikon University General Hospital, Medical School, National and Kapodistrian University, Athens, Greece.
Funding
This study was part of the AMR DetecTool project, which was funded by the European Institute of Innocation and Technology (EIT-Health, Reference 1.5.4; Project ID: 211010 AMR-DetecTool) (European Union).
Transparency declarations
M.S.P. is an employee of NG Biotech, R&D Department—Guipry (France). He was involved in the NG-Test DetecTool OXA-23 manufacturing but not in the data analysis. All other authors: none to declare.
References
- 1. WHO Regional Office for Europe, ECDC . Antimicrobial Resistance Surveillance in Europe 2022. 2022. https://www.ecdc.europa.eu/sites/default/files/documents/Joint-WHO-ECDC-AMR-report-2022.pdf.
- 2. Bonnin RA, Nordmann P, Poirel L. Screening and deciphering antibiotic resistance in Acinetobacter baumannii: a state of the art. Expert Rev Anti Infect Ther 2013; 11: 571–83. 10.1586/eri.13.38 [DOI] [PubMed] [Google Scholar]
- 3. Potron A, Hocquet D, Triponney Pet al. . Carbapenem-susceptible OXA-23-producing Proteus mirabilis in the French community. Antimicrob Agents Chemother 2019; 63: e00191-19. 10.1128/AAC.00191-19 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Österblad M, Karah N, Halkilahti Jet al. . Rare detection of the Acinetobacter class D carbapenemase blaOXA-23 gene in Proteus mirabilis. Antimicrob Agents Chemother 2016; 60: 3243–5. 10.1128/AAC.03119-15 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Octavia S, Xu W, Ng OTet al. . Identification of AbaR4 Acinetobacter baumannii resistance island in clinical isolates of blaOXA-23-positive Proteus mirabilis. J Antimicrob Chemother 2020; 75: 521–5. 10.1093/jac/dkz472 [DOI] [PubMed] [Google Scholar]
- 6. Bonnin RA, Girlich D, Jousset ABet al. . A single Proteus mirabilis lineage from human and animal sources: a hidden reservoir of OXA-23 or OXA-58 carbapenemases in Enterobacterales. Sci Rep 2020; 10: 9160. 10.1038/s41598-020-66161-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Hamprecht A, Sattler J, Noster Jet al. . Proteus mirabilis—analysis of a concealed source of carbapenemases and development of a diagnostic algorithm for detection. Clin Microbiol Infect 2023; 29: 1198.e1–e6. 10.1016/j.cmi.2023.05.032 [DOI] [PubMed] [Google Scholar]
- 8. Corvec S, Poirel L, Naas Tet al. . Genetics and expression of the carbapenem-hydrolyzing oxacillinase gene blaOXA-23 in Acinetobacter baumannii. Antimicrob Agents Chemother 2007; 51: 1530–3. 10.1128/AAC.01132-06 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Bonnin RA, Jousset AB, Emeraud Cet al. . Genetic diversity, biochemical properties, and detection methods of minor carbapenemases in Enterobacterales. Front Med (Lausanne) 2021; 7: 616490. 10.3389/fmed.2020.616490 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Pournaras S, Dafopoulou K, Del Franco Met al. . Predominance of international clone 2 OXA-23-producing-Acinetobacter baumannii clinical isolates in Greece, 2015: results of a nationwide study. Int J Antimicrob Agents 2017; 49: 749–53. 10.1016/j.ijantimicag.2017.01.028 [DOI] [PubMed] [Google Scholar]
- 11. Woodford N, Ellington MJ, Coelho JMet al. . Multiplex PCR for genes encoding prevalent OXA carbapenemases in Acinetobacter spp. Int J Antimicrob Agents 2006; 27: 351–3. 10.1016/j.ijantimicag.2006.01.004 [DOI] [PubMed] [Google Scholar]
- 12. Volland H, Ballesté-Delpierre C, Szabó Det al. . Rapid detection of CTX-M-type ESBLs and carbapenemases directly from biological samples using the BL-DetecTool. J Antimicrob Chemother 2022; 77: 2867–75. 10.1093/jac/dkac264 [DOI] [PubMed] [Google Scholar]
- 13. MedCalc Software Ltd . Diagnostic test evaluation calculator, Version 22.007. https://www.medcalc.org/calc/diagnostic_test.php.
- 14. Riccobono E, Antonelli A, Pecile Pet al. . Evaluation of the KPC K-SeTTM immunochromatographic assay for the rapid detection of KPC carbapenemase producers from positive blood cultures. J Antimicrob Chemother 2018; 73: 539–40. 10.1093/jac/dkx406 [DOI] [PubMed] [Google Scholar]
- 15. Riccobono E, Bogaerts P, Antonelli Aet al. . Evaluation of the OXA-23 K-SeT® immunochromatographic assay for the rapid detection of OXA-23-like carbapenemase- producing Acinetobacter spp. J Antimicrob Chemother 2019; 74: 1455–7. 10.1093/jac/dkz001 [DOI] [PubMed] [Google Scholar]
- 16. Mancini S, Seth-Smith HB, Kolesnik-Goldmann Net al. . Evaluation of the RESIST ACINETO multiplex immunochromatographic assay for detection of OXA-23-like, OXA-40/58-like and NDM carbapenemase production in Acinetobacter baumannii. J Antimicrob Chemother 2023; 78: 2771–4. 10.1093/jac/dkad253 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Nowak J, Zander E, Stefanik Det al. . High incidence of pandrug-resistant Acinetobacter baumannii isolates collected from patients with ventilator-associated pneumonia in Greece, Italy and Spain as part of the MagicBullet clinical trial. J Antimicrob Chemother 2017; 72: 3277–82. 10.1093/jac/dkx322 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Palmieri M, D’Andrea MM, Pelegrin ACet al. . Abundance of colistin-resistant, OXA-23- and ArmA-producing Acinetobacter baumannii belonging to international clone 2 in Greece. Front Microbiol 2020; 11: 668. 10.3389/fmicb.2020.00668 [DOI] [PMC free article] [PubMed] [Google Scholar]