Proof of Concept Evaluation of a New Diagnostic POCT Device for Culture- Independent Microbiological Diagnosis of Infective Endocarditis

Introduction: Accurate and fast microbiological diagnosis of infective endocarditis (IE) is of vital importance for patient outcome. Material & Methods: Forty culture-negative heart valves were evaluated with a new POCT multiplex-PCR cartridge (UnyveroTM, Curetis AG, Holzgerlingen, Germany), advertised to detect several Gram positive/negative bacteria and fungi, together with several antibiotic resistance genes. Those POCT results were compared to conventional 16S rDNA PCR/sequencing results. Results: POCT multiplex-PCR was positive in 13 cases [Staphylococcus aureus (n=5), Enterococcus spp. / E. faecalis (n=5), ConS (n=1), Granulicatella adjacens (n=1), Abiotrophia adjacens (n=1)]. Antibiotic resistances were found in 44 specimens, from which 2 specimens were without any pathogen identification. 16S rDNA PCR was positive in 20 cases. Consecutive sequencing identified those as Staphylococcus spp. (n=6), Enterococcus faecalis (n=4), Streptococcus spp. (n=4), Leifsonia shinshuensis (n=1), Granulicatella elegans / G. adjacens (n=2), Abiotrophia adjacens (n=1). One case was positive in 16S PCR without any reliable signal in sequencing. When comparing both methods, identification was consistent in 9 cases and divergent in other 9 cases. Discussion: This POCT cartridge is easy to integrate into the daily microbiology laboratory work flow, and is less laborious than 16S sequencing PCR. For the application in routine IE diagnosis, the system needs to be optimized to include targets for viridans streptococci and HACEK group. In addition, problems with invalid resistance and pathogen target detection need to be fixed by the producer. Conclusion: The analyzed POCT system might be a future diagnostic tool for IE detection following assay optimization.


Introduction
Infective endocarditis (IE) is a serious disease with a high mortality rate [1] and accurate and fast microbiological diagnosis of IE is of vital importance for patient outcome. IE is difficult to diagnose as blood cultures and culture from heart valve tissue samples often remain negative, mainly due to previous microbial treatment or fastidious IE causing microorganisms (e.g. Bartonella sp., Coxiella burnetti, Tropheryma whipplei). For the culture identification of IE-causing HACEK (Aggregatibacter aphrophilus, Cardiobacterium hominis, Eikenella corrodens, Kingella kingae) bacteria and streptococci, a prolonged incubation for 7-14 days is often needed for identification [2]. Although the modified Duke criteria (DC) have been established for the diagnosis of IE, they have limitations in the diagnosis of IE in prosthetic valves and intra cardiac devices [3]. The yearly rate of IE in patients with a prosthetic valve is approximately 3 cases per 1,000 patients [4]. PCR may offer significant advantages in diagnosis due to the generation of fast results [5]. The correct identification of IE-causing pathogens is important for immediate targeted antibiotic therapy and the avoidance of unnecessary antibiotics, in the context of antibiotic stewardship (ABS) programs. Overall, there is a high need for optimized and faster molecular detection assays for early and targeted treatment of IE-causing pathogens [6][7][8].

POCT multiplex-PCR assay and conventional diagnosis
We performed an evaluation that examined the sensitivity and specificity of a new point of care testing (POCT) multiplex PCR assay (Unyvero ™ , Curetis, Holzgerlingen, Germany) for swift detection of causative microbes in culture-negative cases of IE. The Unyvero ™ i60 ITI cartridge is advertised to detect several Grampositive/-negative bacteria and fungi in implant and tissue, together with some of the most important antibiotic resistance genes. Specimens from 40 fresh frozen heart valves were cultured on Columbia/chocolate agar plates and incubated in brain-heart broth for 7 days. In cases of no bacterial growth after 24 h, tissue specimens were evaluated with the new POCT multiplex-PCR assay for the detection of heart valve infections and compared to 16S rDNA PCR results. In cases of pathogen growth after >24 hours, identification of colonies was performed using MALDI-TOF (VITEK MS) or VITEK 2 technology. Heart valves were frozen after culture inoculation and thawed after 24 h for lysis in the POCT lysator which is able to process a wide range of clinical sample types using a standardized protocol. The Unyvero ™ sample tube prepares the patient sample containing glass beads and buffers for bacterial lysis and sample liquefaction. The Unyvero ™ i60 ITI cartridge is equipped with integrated reagent containers, a DNA purification column, eight separate PCR chambers and a corresponding number of arrays. The cartridge contains buffers for DNA purification, reagents and fluorescence-marked primers for PCR amplification, as well as probes for array hybridization and is assembled by inserting the Unyvero ™ sample tube with the lysed sample and the master mix tube. Once assembled, the cartridge is physically closed, minimizing the risk of contamination. An internal control is also included in the cartridge in order to verify the DNA purification, PCR and array hybridization for each measurement. This gene is amplified in each of the eight PCR chambers and hybridized on each array The Unyvero ™ analyzer processes up to two Unyvero ™ Cartridges in random access mode and automatically performs DNA purification, specific amplification and detection. The Unyvero ™ cockpit is equipped with a touchscreen and connects the Unyvero ™ analyzer to the lysator. Resistance genes that can be measured included mec A/C (resistance to methicillin, and other ß-lactams), van A/B (resistance to glycopeptides), erm A/C (erythromycin-resistance genes), vim/imp/kpc/ndm (carbapenemases, metallo-β-lactamases), aacA4 (aminoglycoside 6'-N-acetyltransferase, resistance-modifying enzyme gene), ctx-M (most prevalent extended-spectrum betalactamases), rpoB oxa-23/-24/-48/-58(carbapenemases), gyrA (quinolone-resistance), aac(6')/aph(2'') (aminoglycoside-6′-Nacetyltransferase/2′′-O-phosphoryltransferase).

16S PCR/sequencing
Bacterial DNA was automatically extracted by the Maxwell Tissue DNA Purification Kit on a Maxwell 15 machine (Promega, Mannheim, Germany). Amplification was performed using the Multiplex PCR Kit (Qiagen, Hilden, Germany) and the ligonucleotides 16S-F (5' TGGTAGTCCACGCCGTAACC 3') and 16S-GRP-R (5' TCATAAGGGGCATGATGAT 3') to detect Gram positive pathogens or 16S-GRN-R (5' CGTAAGGGCCATGATGACT 3') to detect Gram negative pathogens, respectively. The amplification was carried out on a Master cycler epi gradient S (Eppendorf, Hamburg, Germany) with the following conditions: initial denaturation at 95°C for 15 min, 35 cycles of 94°C for 15 sec and 53°C for 15 sec and 72°C for 1 min, followed by terminal elongation at 72°C for 5 min. The quality of each PCR run was ensured by negative control, positive control (DNA of M.luteus for Gram-positive PCR and DNA of M.catarrhalis for Gramnegative PCR) and inhibition control.
All PCR amplicons obtained from patient samples were subject to sequencing in both sense and antisense directions performed with a Big Dye Terminator v1.1 cycle sequencing kit and an ABI Prism 310 sequencer (Applied Bio systems, Foster City, CA, USA) according to standard protocols. Sequencing was performed by using forward and reverse primers (10 pmol each). PCR amplicons were purified from the gel with an Invisorb spin DNA extraction kit (Invitek, Berlin, Germany), and further treated with the sequencing kit material. The following program was executed: 1 cycle at 96°C for 1 min and 25 cycles of 96°C for 10 s, 50°C for 5 s, and 60°C for 4 min. The electropherog rams obtained were analyzed using sequencing analysis software (version 3.7; Applied Biosystems). The results were aligned and examined by Gen Bank NCBI genetic sequence database searching.

Results
Fresh frozen heart valves were examined in routine bacteriology laboratory and with the Unyvero ™ i60 ITI Cartridge.

Discussion
PCR testing of explanted heart valves is recommended in addition to culture techniques to increase diagnostic yield [9]. In case of negative culture result, current guidelines recommend that tissues from excised heart valves or vegetations from patients with suspected IE should be referred for broad-range bacterial PCR and sequencing [10]. Unfortunately, this 16S PCR is prone to contamination of reagents with bacterial DNA, which represents a major problem exacerbated by the highly sensitive nature of 16S PCR. These methods are also laborious and time-consuming, thus fully-automated assays which can be easily integrated in the routine work flow are needed. This evaluation highlights the importance of molecular analysis in diagnostically challenging culture-negative IE as time to result is of critical importance in the diagnosis of IE. For the majority of pathogens detected in this study, the information provided by the new POCT seems to be sufficient in the first place, given the high incidence of staphylococci, streptococci and enterococci in this disease. Although bacteria from the HACEK group were not discovered with 16S PCR IE, they should be covered by a POCT system, despite their rareness ( Table 2). The detection of 5 staphylococcal cases using multiplex PCR assay is not surprising, as Staphylococcus aureus is described as the most common cause of IE in the developed world [11]. There was only a slight difference in the absolute numbers of staphylococcal cases using 16S rDNA PCR/sequencing (6 cases, although Staphylococcus spp in 6 cases. Streptococcal species are common causative IE pathogens [7]. This is in line with our finding of 4 streptococcal cases in this evaluation. Accurate identification within some streptococcal groups was limited with both techniques used. In case of divergent results, such results should be interpreted with caution. Biochemical streptococcal species identification were shown to result in false identifications in more of half of the patients when compared to genetic discrimination methods ( Table 3). Nutritionally variant streptococci (Abiotrophia defective or Granulicatella spp.), are thought to account for 2% of all infective endocarditis cases [12]. Due to difficulties in obtaining positive microbiology cultures, Granulicatella adjacens is known to be responsible for culture-negative infective endocarditis [13]. Both methods used in this evaluation were convergent for Abiotrophia spp, and one Granulicatella case, but divergent for 2 other cases of Granulicatella.
Thus these two results should be interpreted with caution. Infections caused by multi-resistant enterococci (E. faecium, E. faecalis) have increased over the recent years to a point that they now represent the 3rd most common cause of IE worldwide [14].  This is also reflected in our evaluation. Unlike streptococci and taphylococci, most enterococci do not produce a set of potent pro-inflammatory toxins, but they are equipped with many genes encoding adhesion proteins that may mediate adherence to host tissues, consistent with their pathogenic role in infective endocarditis. Vancomycin-resistnat E. faecium (VRE) is difficult to treat [15].

Conclusion
The Unyvero ITI cartridge could represent a useful tool for IE diagnosis. It can be easily integrated into the lab work flow and is less laborious. However, for its application in routine IE diagnosis the multiplex system needs to be optimized and extended to include targets for viridans streptococci and the HACEK group. Therefore, a new specific IE cartridge needs to be developed. In addition, problems with invalid resistance and pathogen target detection need to be fixed before routine testing.