Diagnostic Microbiology and Infectious Disease 79 (2014) 98–101
Contents lists available at ScienceDirect
Diagnostic Microbiology and Infectious Disease journal homepage: www.elsevier.com/locate/diagmicrobio
A commercially available multiplex real-time PCR for detection of pathogens in cardiac valves from patients with infective endocarditis Christian Leli a, Amedeo Moretti a, Maria Bruna Pasticci b, Elio Cenci a, Francesco Bistoni a, Antonella Mencacci a,⁎ a b
Department of Experimental Medicine and Biochemical Sciences, Microbiology Section, University of Perugia, Perugia, Italy Department of Experimental Medicine and Biochemical Sciences, Infectious Disease Section, University of Perugia, Perugia, Italy
a r t i c l e
i n f o
Article history: Received 19 August 2013 Received in revised form 13 December 2013 Accepted 13 December 2013 Available online 21 December 2013 Keywords: Infective endocarditis Real-time PCR SeptiFast Heart valve
a b s t r a c t Infective endocarditis (IE) is a life-threatening condition, burdened by high mortality. Current guidelines recommend that, in case of negative culture result, tissues from excised heart valves or vegetations from patients with suspected IE should be referred for broad-range bacterial PCR and sequencing. In this proof-ofconcept study, the diagnostic utility of the commercially available multiplex real-time PCR system SeptiFast (SF), performed on cardiac valves, was evaluated in a selected population of 20 patients with definite IE of known origin, in comparison with culture. A significant difference was found between SF and culture in the rate of pathogen detection (19 versus 3 respectively; chi-square 14.06; P = 0.0002). SF sensitivity was 95%; specificity, 100%; positive predictive value (PPV), 100%; and negative predictive value (NPV), 83.3%. Culture sensitivity was 15%; specificity, 100%; PPV, 100%; and NPV, 22.7%. SF assay, performed on culture-negative excised heart valves, can be useful for the etiological diagnosis of IE. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Infective endocarditis (IE) is a life-threatening condition, burdened by high mortality (Prendergast, 2006). Gram-positive cocci account for the majority of the cases (Prendergast, 2006). Recently, it has been reported that 31% of the cases are due to Staphylococcus aureus; 29%, to streptococci; 11%, to coagulase-negative staphylococci (CoNS); and 10%, to enterococci (Murdoch et al., 2009). Other microorganisms, such as Gram-negative bacteria, fastidious organisms, and fungi, sum up to a prevalence of around 14% in IE (Berbari et al., 1997; Gould 2012; Watkin et al., 2003). The diagnosis of IE is established with certainty only when vegetations obtained from cardiac valves or emboli are examined histologically or microbiologically (Horstkotte et al., 2004). Nevertheless, IE is usually diagnosed by clinical, laboratory, and echocardiographic findings, according to the modified Duke criteria (Durack et al., 1994). Among these, blood culture (BC) represents a cornerstone for the microbiological diagnosis of IE (Gould, 2012), but it can be negative in a substantial number of patients (Lamas and Eykyn, 2003; Naber and Erbel, 2007). Similarly, culture of excised cardiac valves can be burdened by a considerable number of false negative results, mostly in patients under antibiotic therapy, or by false-positive results, due to contamination (Muñoz et al., 2008). Current guidelines recommend that tissues from excised heart
⁎ Corresponding author. Tel.: +39-075-578-4285; fax: +39-075-578-4298. E-mail address: [email protected] (A. Mencacci). 0732-8893/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.diagmicrobio.2013.12.011
valves or vegetations from patients with suspected IE should be referred for broad-range bacterial PCR and sequencing in cases of culture-negative result (Gould, 2012). To date, different molecular techniques proved to be endowed with considerable sensitivity and specificity in IE diagnosis, overcoming the problems of poor sensitivity of culture (Breitkopf et al., 2005; Gauduchon et al., 2003; Vondracek et al., 2011; Wallet et al., 2013). Unfortunately, these techniques are not easily available in clinical microbiology laboratories. The commercially available SeptiFast (SF) real-time PCR system (Roche Molecular Systems, Mannheim, Germany), used for diagnosis of sepsis, can be successfully used on blood samples for diagnosis of IE (Casalta et al., 2009; Mencacci et al., 2012) and on specimens different from blood for the diagnosis of other infections (Mencacci et al., 2011). The aim of the present study was to evaluate the utility of SF, performed on excised cardiac valves, in the etiological diagnosis of IE. To this purpose, the diagnostic accuracy of SF was evaluated in a proof-of-concept study in a selected population of 20 patients with definite IE of known origin, in comparison with culture. 2. Materials and methods 2.1. Patients and specimens Between June 2011 and January 2013, 20 cardiac valves (11 mitral and 9 aortic) excised from an equal number of patients with definite IE diagnosis of known etiology were tested by conventional culture and SF test. Five aortic valves excised from non-IE patients, affected by non-infective cardiac complications, were included in the study as
C. Leli et al. / Diagnostic Microbiology and Infectious Disease 79 (2014) 98–101
controls. Mean age of the patients was 64.2 years (SD ±12.6). All patients were on antimicrobial therapy during sampling. Inclusion criteria were cardiac valve tissues excised from patients with definite IE, diagnosed before surgery according to modified Duke's criteria (Durack et al., 1994), and known infectious agent, because of previous positive BC. Exclusion criteria were specimens from patients with no definite IE diagnosis or with IE of unknown etiology. 2.2. Cultures All the excised valves were processed in a laminar flow cabinet prior to examination using Gram staining and for inoculation of solid and liquid media. The valves were disrupted by homogenization in PCR grade water (Roche Applied Science, Penzberg, Germany) and cultured on: chocolate agar supplemented with IsoVitaleX and Columbia sheep blood agar, incubated at 35–37 °C, with 5% CO2 for up to 4–5 days; Sabouraud agar, incubated under aerobic conditions at 35–37 °C for up to 2–3 days; Schaedler agar and vancomycinkanamycin Schaedler agar, incubated in an anaerobic atmosphere at 35–37 °C for up to 4–5 days (all media from Becton Dickinson, Sparks, MD, USA). Specimens were also inoculated into brain heart infusion broth (bioMerieux, Marcy-l'Etoile, France) and in aerobic and anaerobic BACTEC FX Plus bottles (Becton Dickinson) and incubated at 35 °C for 7 days. Isolated colonies were identified according to standard procedures and by means of the automatic Phoenix system (Becton Dickinson). 2.3. SF real-time PCR Explanted valve tissues were homogenized in 3 mL PCR grade water (Roche Applied Science). The entire cardiac tissue explanted from each patient was used for homogenization. Concomitantly with cultures, 1.5 mL sample was processed by SF test (Roche Diagnostics GmbH, Mannheim, Germany), as previously described (Mencacci et al., 2011). Briefly, mechanical lysis using the SF Lys Kit MGRADE and the MagNALyser® was performed. Using the SF Prep Kit MGRADE, DNA was extracted as described by the manufacturer. Hybridization probes were used. An internal control was introduced into each specimen, and a negative control supplied by the manufacturer was included in each extraction series. Using the LightCycler® SF kit MGRADE, real-time PCR was performed in a LightCycler 2.0 instrument (Roche Diagnostics). Three different primer mixes were used to amplify Gram-positive and Gram-negative bacteria and fungi. The internal transcribed spacer region was the specific target for the detection of bacterial and fungal pathogens. Species identification (melting temperature analysis of specimens and controls in each channel) and report generation were obtained using the SF identification software (Roche Diagnostics). The microorganisms identified by SF have been listed elsewhere (Lehmann et al., 2008). 2.4. Definition of pathogen An organism detected by SF test was considered true pathogen if it coincided with the infectious agent previously isolated from BC, collected from the same IE patient. 2.5. Statistical analysis SPSS statistical package, release 13.0 (SPSS, Chicago, IL, USA), was used for all statistical analyses. Sensitivities, specificities, and predictive values were compared by means of the chi-square test, using the 2-by-2 contingency table. The McNemar test was used for testing the differences between paired proportions. Agreement beyond chance among tests was assessed by Cohen's kappa statistics (b0, no agreement; 0–0.19, poor; 0.20–0.39, fair; 0.40–0.59, moder-
99
ate; 0.60–0.79, substantial; and 0.80–1.00, almost perfect agreement) (Landis and Koch, 1977). The significance level was 0.05. 3. Results During the study period, 20 cardiac valves were excised from an equal number of patients with definite IE, diagnosed according to modified Duke criteria (Durack et al., 1994), in which the etiological agents, recovered from BC, were as follows: 3 S. aureus, 3 Staphylococcus epidermidis, 3 Streptococcus mitis, 2 Streptococcus bovis, 2 Streptococcus constellatus, 2 Streptococcus agalactiae, 2 Enterococcus faecalis, 1 Streptococcus gordonii, 1 Enterococcus faecium, and 1 Kytococcus sedentarius. All patients were on antimicrobial therapy during sampling. Cardiac valves (11 mitral and 9 aortic) from IE patients, together with 5 aortic valves from non-IE patients, were examined by culture and SF test. Culture was positive in 3 out of 20 valves from IE patients, for microorganisms matching those previously isolated from BC, and was negative in all valves from non-IE patients (Table 1). Culture sensitivity was 15.8%; specificity, 100%; positive predictive value (PPV), 100%; and negative predictive value (NPV), 22.7%. Gram staining was positive in 16 valves from IE patients and negative in all non-IE patients (Table 1). Gram staining sensitivity was 80%; specificity, 100%; PPV, 100%; and NPV, 55.5%. SF test gave a positive result in 19 out of 20 valves from patients with IE and a negative result in all the 5 valves from non-IE patients. In all cases, the organism detected by the test was concordant with the known etiologic agent and Gram staining. In K. sedentarius endocarditis, SF test was negative (Table 1). SF assay sensitivity was 95%; specificity, 100%; PPV, 100%; and NPV, 83.3%. A significant difference in the rate of pathogen detection was found between SF and culture (19 versus 3, respectively; chi-square 14.06; P = 0.0002), with poor degree of agreement (kappa value = 0.083; 95% confidence interval, −0.025–0.190). SF sensitivity and NPV were significantly higher than sensitivity (chi-square 25.08; P b 0.0001) and NPV (chi-square 7.54; P = 0.01) of culture. No significant differences were found between SF and Gram staining. 4. Discussion The present study found that SF assay, performed on heart valves from patients with IE, showed a greater pathogen detection rate than conventional culture. SF sensitivity and NPV were significantly higher than those of culture. Differently from other studies, the diagnostic accuracy of SF was evaluated by means of a proof-of-concept study, in which only selected cases with both definite IE diagnosis and known infectious agent were included. This approach was chosen to rule out the possibility that false-positive or false-negative IE patients could be included in the study and to avoid possible misinterpretation of the clinical significance of the microorganisms detected by the test. In addition, this study was carried out using a commercial real-time PCR test, easily available in clinical microbiology laboratories. Nevertheless, the results are in line with studies using other broad-range PCR and different methods to evaluate the diagnostic accuracy of PCR (Breitkopf et al., 2005; Marín et al., 2007; Voldstedlund et al., 2008; Vollmer et al., 2010). Voldstedlund et al. (2008), by means of 16S rDNA PCR and sequencing, found for the molecular method a sensitivity (72%) and specificity (100%) better than culture in a sample of 74 heart valves from 74 patients, 57 of which with definite IE, 54 with the infectious agent identified from BC. Similar results were obtained by Marín et al. (2007) with 96% PCR sensitivity, 95.3% specificity, and 98.4% NPV, in a total of 177 heart valves of which 35 from patients with definite IE. A high sensitivity (80.6%) for the molecular method was also observed by means of a 23 rDNA PCR, which can improve the diagnosis of culture-negative IE (Vollmer et al., 2010). Likewise, in a sample of 52 heart valves from 51 patients with
100
C. Leli et al. / Diagnostic Microbiology and Infectious Disease 79 (2014) 98–101
Table 1 Gram staining, cultures, and SF real-time PCR in heart valves excised from patients with and without definite IE. Organism isolated from blood culture Patients with definite IE S. aureus S. aureus S. aureus S. epidermidis S. epidermidis S. epidermidis S. mitis S. mitis S. mitis S. bovis S. bovis S. constellatus S. constellatus S. agalactiae S. agalactiae E. faecalis E. faecalis S. gordonii E. faecium K. sedentarius Patients without IE None None None None None
Heart valve tissue Valve
Gram staining
Culture
SF real-time PCR
Mitral Aortic Mitral Mitral Aortic Aortic Mitral Mitral Mitral Aortic Mitral Aortic Aortic Mitral Aortic Mitral Aortic Aortic Mitral Mitral
Gram-positive cocci Gram-positive cocci Negative for bacteria and fungi Gram-positive cocci Gram-positive cocci Negative for bacteria and fungi Negative for bacteria and fungi Gram-positive cocci in chains Negative for bacteria and fungi Gram-positive cocci in chains Gram-positive cocci Gram-positive cocci in pairs Gram-positive cocci in pairs Gram-positive cocci in chains Gram-positive cocci in pairs Gram-positive cocci Gram-positive cocci Gram-positive cocci in chains Gram-positive cocci in pairs Gram-positive cocci
Negative Negative Negative Negative S. epidermidis Negative Negative S. mitis Negative Negative Negative Negative Negative Negative Negative Negative Negative Negative E. faecium Negative
S. aureus S. aureus S. aureus CoNSa CoNSa CoNSa Streptococcus spp.b Streptococcus spp.b Streptococcus spp.b Streptococcus spp.b Streptococcus spp.b Streptococcus spp.b Streptococcus spp.b Streptococcus spp.b Streptococcus spp.b E. faecalis E. faecalis Streptococcus spp.a E. faecium Negative
Aortic Aortic Aortic Aortic Aortic
Negative Negative Negative Negative Negative
Negative Negative Negative Negative Negative
Negative Negative Negative Negative Negative
for for for for for
bacteria bacteria bacteria bacteria bacteria
and and and and and
fungi fungi fungi fungi fungi
a Coagulase-negative Staphylococcus from SML: Staphylococcus capitis, Staphylococcus caprae, Staphylococcus cohnii, S. epidermidis, Staphylococcus haemolyticus, Staphylococcus hominis, Staphylococcus lugdunensis, Staphylococcus pasteuri, Staphylococcus saprophyticus, Staphylococcus warneri, and Staphylococcus xylosus. b Streptococcus from SML: S. agalactiae, Streptococcus anginosus, S. bovis, S. constellatus, Streptococcus cristatus, S. gordonii, Streptococcus intermedius, Streptococcus milleri group, S. mitis, Streptococcus mutans, Streptococcus oralis, Streptococcus parasanguinis, Streptococcus pyogenes, Streptococcus salivarius, Streptococcus sanguinis, Streptococcus thermophilus, Streptococcus vestibularis, and Streptococcus viridans group.
suspected IE, PCR showed a sensitivity and NPV higher than culture (Breitkopf et al., 2005), although the PCR sensitivity (41.2%) observed does not match that of the present study, probably due to different inclusion criteria. To our knowledge, only few studies evaluated commercially available real-time PCR tests for diagnosis of IE on heart valves. In a study using 2 universal rRNA gene PCR plus sequencing tests, sensitivity of PCR (85%) was nearly twice as high as that of culture (45%) (Kühn et al., 2011), and a previous study on various clinical specimens suggested that SF assay performed on cardiac valves could be a useful tool for molecular diagnosis in patients with IE (Mencacci et al., 2011). Fernández et al. (2010) found a sensitivity of 100% for SF test in a sample of 15 heart valves from patients with definite IE in comparison with culture (30.7%). Indeed, also in the present study, excluding the case caused by K. sedentarius, an organism not included in the SF master list (SML), SF showed a sensitivity of 100%, mirroring the results of the study reported above. However, it should be stressed that, as all the pathogens were isolated from blood culture and all but one were detectable by SF, the sensitivity, specificity, NPV, and PPV data may not perfectly reflect those expected in a true study population or in a routine diagnostic setting, in which it is possible to find patients with clinical diagnosis of IE, but negative blood cultures or IE from pathogens not included in SML. Moreover, as no case included in the study was due to Gram negative bacteria or fungi, the diagnostic accuracy of the test for IE caused by these pathogens remains to be defined. Similarly to other studies employing other commercially available molecular tests (Kühn et al., 2011; Haag et al. 2013), real-time PCR sensitivity was superior to that of culture. This finding can be due to the fact that all cardiac valves were from patients under antibiotic treatment, which hampers the sensitivity of culture. Thus, SF could
have detected DNA of intact non-cultivable microorganisms, as suggested by the positivity on Gram staining of culture-negative valve tissues. However, differently from studies using molecular systems in which free pathogen DNA is eliminated during the extraction protocol (Kühn et al., 2011; Haag et al., 2013), it is also possible that DNA remnants from infectious agents have been detected by SF, in which DNA depletion is not done prior to DNA extraction. Indeed, the finding that DNA can persist on cardiac valves of IE patients for several months after the starting of antimicrobial therapy (Gould, 2012) points for the notion that a positive real-time PCR result can be reliably used to identify the cause of endocarditis but cannot be used to infer ongoing presence of infection and should not, therefore, be used alone to judge the duration of post-operative antimicrobial therapy (Gould, 2012). On the other hand, it is unlikely that the DNA detected by SF was due to sample contamination, due to different findings: i) microorganisms detected matched the known causative pathogens, previously isolated from BC; ii) microorganisms detected were consistent with those observed on Gram staining of valve tissues, in 16/20 cases; iii) SF gave negative results with valves from non-IE patients and K. sedentarius IE patient; and iv) absence of contamination in the real-time PCR experiments, suggested by adequate internal and negative controls. The impossibility to identify pathogens not included in SML is 1 limitation of the SF assay, given that fastidious or slow-growing bacteria, like HACEK (Haemophilus spp., Aggregatibacter actinomycetemcomitans, Cardiobacterium hominis, Eikenella spp., and Kingella spp.), Coxiella burnetii, Bartonella spp., or others can be responsible of IE. All together these pathogens sum up to a prevalence of around 14% in IE (Berbari et al., 1997; Gould, 2012; Watkin et al., 2003). Thus, in case of negative SF result, it would be advisable to test cardiac valves
C. Leli et al. / Diagnostic Microbiology and Infectious Disease 79 (2014) 98–101
by a more sensitive method, such as a 2-step broad-range PCR method or real-time PCR procedures specific for uncommon infectious agents, which can improve molecular diagnosis of IE (Boussier et al., 2013), with additional costs. Nevertheless, since the majority of IE cases are due to microorganisms detectable by SF (Murdoch et al., 2009), laboratories performing the molecular diagnosis of sepsis can use this test for culture-negative heart valves to overcome the problem of employing in-house molecular procedures, requiring a high-quality validation and expert dedicated staff, not easily available in all the clinical microbiology laboratories. However, it must be underlined that not every clinical microbiology laboratory has the equipment and staff for this multiplex real-time PCR analysis as well. Another limitation of SF is that it does not identify at the species level streptococci other than S. pneumoniae and CoNS (Lehmann et al., 2008), although in case of IE from staphylococci, methicillin resistance can be assessed by detection of the mecA gene (Wallet et al., 2011). Even if etiological diagnosis of streptococcal or staphylococcal IE at the genus level could be sufficient enough to guide appropriate antimicrobial therapy (Gould, 2012), nevertheless, the need for DNA sequencing of the amplicons could be required for species identification, with additional costs. Although the strict criteria used ruled out the possibility of erroneous evaluation of the diagnostic accuracy of SF, further studies with a greater number of patients, including cases with IE of unknown origin or suspected, are needed to better define the usefulness of this test in the diagnostic practice. In conclusion, SF assay can be used on culture-negative heart valve tissues, excised from patients with IE, showing sensitivity and NPV higher than culture, with the same high specificity and PPV. In cases of SF positive for CoNS or streptococci, a DNA sequencing of the amplicons would be required to identify the pathogen at the species level. In case of negative SF result, it would be advisable to test cardiac valves by PCR procedures specific for uncommon infectious agents of IE.
References Berbari EF, Cockerill III FR, Steckelberg JM. Infective endocarditis due to unusual or fastidious microorganisms. Mayo Clin Proc 1997;72:532–42. Boussier R, Rogez S, François B, Denes E, Ploy MC, Garnier F. Two-step bacterial broadrange polymerase chain reaction analysis of heart valve tissue improves bacteriological diagnosis of infective endocarditis. Diagn Microbiol Infect Dis 2013;75:240–4. Breitkopf C, Hammel D, Scheld HH, Peters G, Becker K. Impact of a molecular approach to improve the microbiological diagnosis of infective heart valve endocarditis. Circulation 2005;111:1415–21. Casalta JP, Gouriet F, Roux V, Thuny F, Habib G, Raoult D. Evaluation of the LightCycler SeptiFast test in the rapid etiologic diagnostic of infectious endocarditis. Eur J Clin Microbiol Infect Dis 2009;28:569–73. Durack DT, Lukes AS, Bright DK. New criteria for diagnosis of infective endocarditis: utilization of specific echocardiographic findings. Duke Endocarditis Service. Am J Med 1994;96:200–9.
101
Fernández AL, Varela E, Martínez L, Martínez A, Sierra J, González-Juanatey JR, et al. Evaluation of a multiplex real-time PCR assay for detecting pathogens in cardiac valve tissue in patients with endocarditis. Rev Esp Cardiol 2010;63:1205–8. Gauduchon V, Chalabreysse L, Etienne J, Célard M, Benito Y, Lepidi H, et al. Molecular diagnosis of infective endocarditis by PCR amplification and direct sequencing of DNA from valve tissue. J Clin Microbiol 2003;41:763–6. Gould FK. Guidelines for the diagnosis and antibiotic treatment of endocarditis in adults: a report of the Working Party of the British Society for Antimicrobial Chemotherapy. J Antimicrob Chemother 2012;67:269–89. Haag H, Locher F, Nolte O. Molecular diagnosis of microbial aetiologies using SepsiTest™ in the daily routine of a diagnostic laboratory. Diagn Microbiol Infect Dis 2013;76:413–8. Horstkotte D, Follath F, Gutschik E, Lengyel M, Oto A, Pavie A, et al. Task Force Members on Infective Endocarditis of the European Society of Cardiology; ESC Committee for Practice Guidelines (CPG); Document Reviewers. Guidelines on prevention, diagnosis and treatment of infective endocarditis executive summary; the task force on infective endocarditis of the European society of cardiology. Eur Heart J 2004;25:267–76. Kühn C, Disqué C, Mühl H, Orszag P, Stiesch M, Haverich A. Evaluation of commercial universal rRNA gene PCR plus sequencing tests for identification of bacteria and fungi associated with infectious endocarditis. J Clin Microbiol 2011;49:2919–23. Lamas CC, Eykyn SJ. Blood culture negative endocarditis: analysis of 63 cases presenting over 25 years. Heart 2003;89:258–62. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33:159–74. Lehmann LE, Hunfeld KP, Emrich T, Haberhausen G, Wissing H, Hoeft A, et al. A multiplex real-time PCR assay for rapid detection and differentiation of 25 bacterial and fungal pathogens from whole blood samples. Med Microbiol Immunol 2008;197:313–24. Marín M, Muñoz P, Sánchez M, del Rosal M, Alcalá L, Rodríguez-Créixems M, et al, Group for the Management of Infective Endocarditis of the Gregorio Marañón Hospital. Molecular diagnosis of infective endocarditis by real-time broad-range polymerase chain reaction (PCR) and sequencing directly from heart valve tissue. Medicine (Baltimore) 2007;86:195–202. Mencacci A, Leli C, Cardaccia A, Montagna P, Moretti A, Bietolini C, et al. Comparison of conventional culture with SeptiFast real-time PCR for microbial pathogen detection in clinical specimens other than blood. J Med Microbiol 2011;60:1774–8. Mencacci A, Leli C, Montagna P, Cardaccia A, Meucci M, Bietolini C, et al. Diagnosis of infective endocarditis: comparison of the LightCycler SeptiFast real-time PCR with blood culture. J Med Microbiol 2012;61:81–3. Muñoz P, Bouza E, Marín M, Alcalá L, Rodríguez Créixems M, et al. Heart valves should not be routinely cultured. J Clin Microbiol 2008;46:2897–901. Murdoch DR, Corey GR, Hoen B, Miró JM, Fowler Jr VG, Bayer AS, et al. Clinical presentation, etiology, and outcome of infective endocarditis in the 21st century: the International Collaboration on Endocarditis-Prospective Cohort Study. Arch Intern Med 2009;169:463–73. Naber CK, Erbel R. Infective endocarditis with negative blood cultures. Int J Antimicrob Agents 2007;30:S32–6. Prendergast BD. The changing face of infective endocarditis. Heart 2006;92:879–85. Voldstedlund M, Nørum Pedersen L, Baandrup U, Klaaborg KE, Fuursted K. Broad-range PCR and sequencing in routine diagnosis of infective endocarditis. APMIS 2008;116:190–8. Vollmer T, Piper C, Horstkotte D, Körfer R, Kleesiek K, Dreier J. 23S rDNA real-time polymerase chain reaction of heart valves: a decisive tool in the diagnosis of infective endocarditis. Eur Heart J 2010;31:1105–13. Vondracek M, Sartipy U, Aufwerber E, Julander I, Lindblom D, Westling K. 16S rDNA sequencing of valve tissue improves microbiological diagnosis in surgically treated patients with infective endocarditis. J Infect 2011;62:472–8. Wallet F, Herwegh S, Decoene C, Courcol RJ. PCR-electrospray ionization time-of-flight mass spectrometry: a new tool for the diagnosis of infective endocarditis from heart valves. Diagn Microbiol Infect Dis 2013;76:125–8. Wallet F, Loïez C, Herwegh S, Courcol RJ. Usefulness of real-time PCR for the diagnosis of sepsis in ICU-acquired infections. Infect Disord Drug Targets 2011;11:348–53. Watkin RW, Lang S, Lambert PA, Littler WA, Elliott TSJ. The microbial diagnosis of infective endocarditis. J Infect 2003;47:1–11.
Contents lists available at ScienceDirect
Diagnostic Microbiology and Infectious Disease journal homepage: www.elsevier.com/locate/diagmicrobio
A commercially available multiplex real-time PCR for detection of pathogens in cardiac valves from patients with infective endocarditis Christian Leli a, Amedeo Moretti a, Maria Bruna Pasticci b, Elio Cenci a, Francesco Bistoni a, Antonella Mencacci a,⁎ a b
Department of Experimental Medicine and Biochemical Sciences, Microbiology Section, University of Perugia, Perugia, Italy Department of Experimental Medicine and Biochemical Sciences, Infectious Disease Section, University of Perugia, Perugia, Italy
a r t i c l e
i n f o
Article history: Received 19 August 2013 Received in revised form 13 December 2013 Accepted 13 December 2013 Available online 21 December 2013 Keywords: Infective endocarditis Real-time PCR SeptiFast Heart valve
a b s t r a c t Infective endocarditis (IE) is a life-threatening condition, burdened by high mortality. Current guidelines recommend that, in case of negative culture result, tissues from excised heart valves or vegetations from patients with suspected IE should be referred for broad-range bacterial PCR and sequencing. In this proof-ofconcept study, the diagnostic utility of the commercially available multiplex real-time PCR system SeptiFast (SF), performed on cardiac valves, was evaluated in a selected population of 20 patients with definite IE of known origin, in comparison with culture. A significant difference was found between SF and culture in the rate of pathogen detection (19 versus 3 respectively; chi-square 14.06; P = 0.0002). SF sensitivity was 95%; specificity, 100%; positive predictive value (PPV), 100%; and negative predictive value (NPV), 83.3%. Culture sensitivity was 15%; specificity, 100%; PPV, 100%; and NPV, 22.7%. SF assay, performed on culture-negative excised heart valves, can be useful for the etiological diagnosis of IE. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Infective endocarditis (IE) is a life-threatening condition, burdened by high mortality (Prendergast, 2006). Gram-positive cocci account for the majority of the cases (Prendergast, 2006). Recently, it has been reported that 31% of the cases are due to Staphylococcus aureus; 29%, to streptococci; 11%, to coagulase-negative staphylococci (CoNS); and 10%, to enterococci (Murdoch et al., 2009). Other microorganisms, such as Gram-negative bacteria, fastidious organisms, and fungi, sum up to a prevalence of around 14% in IE (Berbari et al., 1997; Gould 2012; Watkin et al., 2003). The diagnosis of IE is established with certainty only when vegetations obtained from cardiac valves or emboli are examined histologically or microbiologically (Horstkotte et al., 2004). Nevertheless, IE is usually diagnosed by clinical, laboratory, and echocardiographic findings, according to the modified Duke criteria (Durack et al., 1994). Among these, blood culture (BC) represents a cornerstone for the microbiological diagnosis of IE (Gould, 2012), but it can be negative in a substantial number of patients (Lamas and Eykyn, 2003; Naber and Erbel, 2007). Similarly, culture of excised cardiac valves can be burdened by a considerable number of false negative results, mostly in patients under antibiotic therapy, or by false-positive results, due to contamination (Muñoz et al., 2008). Current guidelines recommend that tissues from excised heart
⁎ Corresponding author. Tel.: +39-075-578-4285; fax: +39-075-578-4298. E-mail address: [email protected] (A. Mencacci). 0732-8893/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.diagmicrobio.2013.12.011
valves or vegetations from patients with suspected IE should be referred for broad-range bacterial PCR and sequencing in cases of culture-negative result (Gould, 2012). To date, different molecular techniques proved to be endowed with considerable sensitivity and specificity in IE diagnosis, overcoming the problems of poor sensitivity of culture (Breitkopf et al., 2005; Gauduchon et al., 2003; Vondracek et al., 2011; Wallet et al., 2013). Unfortunately, these techniques are not easily available in clinical microbiology laboratories. The commercially available SeptiFast (SF) real-time PCR system (Roche Molecular Systems, Mannheim, Germany), used for diagnosis of sepsis, can be successfully used on blood samples for diagnosis of IE (Casalta et al., 2009; Mencacci et al., 2012) and on specimens different from blood for the diagnosis of other infections (Mencacci et al., 2011). The aim of the present study was to evaluate the utility of SF, performed on excised cardiac valves, in the etiological diagnosis of IE. To this purpose, the diagnostic accuracy of SF was evaluated in a proof-of-concept study in a selected population of 20 patients with definite IE of known origin, in comparison with culture. 2. Materials and methods 2.1. Patients and specimens Between June 2011 and January 2013, 20 cardiac valves (11 mitral and 9 aortic) excised from an equal number of patients with definite IE diagnosis of known etiology were tested by conventional culture and SF test. Five aortic valves excised from non-IE patients, affected by non-infective cardiac complications, were included in the study as
C. Leli et al. / Diagnostic Microbiology and Infectious Disease 79 (2014) 98–101
controls. Mean age of the patients was 64.2 years (SD ±12.6). All patients were on antimicrobial therapy during sampling. Inclusion criteria were cardiac valve tissues excised from patients with definite IE, diagnosed before surgery according to modified Duke's criteria (Durack et al., 1994), and known infectious agent, because of previous positive BC. Exclusion criteria were specimens from patients with no definite IE diagnosis or with IE of unknown etiology. 2.2. Cultures All the excised valves were processed in a laminar flow cabinet prior to examination using Gram staining and for inoculation of solid and liquid media. The valves were disrupted by homogenization in PCR grade water (Roche Applied Science, Penzberg, Germany) and cultured on: chocolate agar supplemented with IsoVitaleX and Columbia sheep blood agar, incubated at 35–37 °C, with 5% CO2 for up to 4–5 days; Sabouraud agar, incubated under aerobic conditions at 35–37 °C for up to 2–3 days; Schaedler agar and vancomycinkanamycin Schaedler agar, incubated in an anaerobic atmosphere at 35–37 °C for up to 4–5 days (all media from Becton Dickinson, Sparks, MD, USA). Specimens were also inoculated into brain heart infusion broth (bioMerieux, Marcy-l'Etoile, France) and in aerobic and anaerobic BACTEC FX Plus bottles (Becton Dickinson) and incubated at 35 °C for 7 days. Isolated colonies were identified according to standard procedures and by means of the automatic Phoenix system (Becton Dickinson). 2.3. SF real-time PCR Explanted valve tissues were homogenized in 3 mL PCR grade water (Roche Applied Science). The entire cardiac tissue explanted from each patient was used for homogenization. Concomitantly with cultures, 1.5 mL sample was processed by SF test (Roche Diagnostics GmbH, Mannheim, Germany), as previously described (Mencacci et al., 2011). Briefly, mechanical lysis using the SF Lys Kit MGRADE and the MagNALyser® was performed. Using the SF Prep Kit MGRADE, DNA was extracted as described by the manufacturer. Hybridization probes were used. An internal control was introduced into each specimen, and a negative control supplied by the manufacturer was included in each extraction series. Using the LightCycler® SF kit MGRADE, real-time PCR was performed in a LightCycler 2.0 instrument (Roche Diagnostics). Three different primer mixes were used to amplify Gram-positive and Gram-negative bacteria and fungi. The internal transcribed spacer region was the specific target for the detection of bacterial and fungal pathogens. Species identification (melting temperature analysis of specimens and controls in each channel) and report generation were obtained using the SF identification software (Roche Diagnostics). The microorganisms identified by SF have been listed elsewhere (Lehmann et al., 2008). 2.4. Definition of pathogen An organism detected by SF test was considered true pathogen if it coincided with the infectious agent previously isolated from BC, collected from the same IE patient. 2.5. Statistical analysis SPSS statistical package, release 13.0 (SPSS, Chicago, IL, USA), was used for all statistical analyses. Sensitivities, specificities, and predictive values were compared by means of the chi-square test, using the 2-by-2 contingency table. The McNemar test was used for testing the differences between paired proportions. Agreement beyond chance among tests was assessed by Cohen's kappa statistics (b0, no agreement; 0–0.19, poor; 0.20–0.39, fair; 0.40–0.59, moder-
99
ate; 0.60–0.79, substantial; and 0.80–1.00, almost perfect agreement) (Landis and Koch, 1977). The significance level was 0.05. 3. Results During the study period, 20 cardiac valves were excised from an equal number of patients with definite IE, diagnosed according to modified Duke criteria (Durack et al., 1994), in which the etiological agents, recovered from BC, were as follows: 3 S. aureus, 3 Staphylococcus epidermidis, 3 Streptococcus mitis, 2 Streptococcus bovis, 2 Streptococcus constellatus, 2 Streptococcus agalactiae, 2 Enterococcus faecalis, 1 Streptococcus gordonii, 1 Enterococcus faecium, and 1 Kytococcus sedentarius. All patients were on antimicrobial therapy during sampling. Cardiac valves (11 mitral and 9 aortic) from IE patients, together with 5 aortic valves from non-IE patients, were examined by culture and SF test. Culture was positive in 3 out of 20 valves from IE patients, for microorganisms matching those previously isolated from BC, and was negative in all valves from non-IE patients (Table 1). Culture sensitivity was 15.8%; specificity, 100%; positive predictive value (PPV), 100%; and negative predictive value (NPV), 22.7%. Gram staining was positive in 16 valves from IE patients and negative in all non-IE patients (Table 1). Gram staining sensitivity was 80%; specificity, 100%; PPV, 100%; and NPV, 55.5%. SF test gave a positive result in 19 out of 20 valves from patients with IE and a negative result in all the 5 valves from non-IE patients. In all cases, the organism detected by the test was concordant with the known etiologic agent and Gram staining. In K. sedentarius endocarditis, SF test was negative (Table 1). SF assay sensitivity was 95%; specificity, 100%; PPV, 100%; and NPV, 83.3%. A significant difference in the rate of pathogen detection was found between SF and culture (19 versus 3, respectively; chi-square 14.06; P = 0.0002), with poor degree of agreement (kappa value = 0.083; 95% confidence interval, −0.025–0.190). SF sensitivity and NPV were significantly higher than sensitivity (chi-square 25.08; P b 0.0001) and NPV (chi-square 7.54; P = 0.01) of culture. No significant differences were found between SF and Gram staining. 4. Discussion The present study found that SF assay, performed on heart valves from patients with IE, showed a greater pathogen detection rate than conventional culture. SF sensitivity and NPV were significantly higher than those of culture. Differently from other studies, the diagnostic accuracy of SF was evaluated by means of a proof-of-concept study, in which only selected cases with both definite IE diagnosis and known infectious agent were included. This approach was chosen to rule out the possibility that false-positive or false-negative IE patients could be included in the study and to avoid possible misinterpretation of the clinical significance of the microorganisms detected by the test. In addition, this study was carried out using a commercial real-time PCR test, easily available in clinical microbiology laboratories. Nevertheless, the results are in line with studies using other broad-range PCR and different methods to evaluate the diagnostic accuracy of PCR (Breitkopf et al., 2005; Marín et al., 2007; Voldstedlund et al., 2008; Vollmer et al., 2010). Voldstedlund et al. (2008), by means of 16S rDNA PCR and sequencing, found for the molecular method a sensitivity (72%) and specificity (100%) better than culture in a sample of 74 heart valves from 74 patients, 57 of which with definite IE, 54 with the infectious agent identified from BC. Similar results were obtained by Marín et al. (2007) with 96% PCR sensitivity, 95.3% specificity, and 98.4% NPV, in a total of 177 heart valves of which 35 from patients with definite IE. A high sensitivity (80.6%) for the molecular method was also observed by means of a 23 rDNA PCR, which can improve the diagnosis of culture-negative IE (Vollmer et al., 2010). Likewise, in a sample of 52 heart valves from 51 patients with
100
C. Leli et al. / Diagnostic Microbiology and Infectious Disease 79 (2014) 98–101
Table 1 Gram staining, cultures, and SF real-time PCR in heart valves excised from patients with and without definite IE. Organism isolated from blood culture Patients with definite IE S. aureus S. aureus S. aureus S. epidermidis S. epidermidis S. epidermidis S. mitis S. mitis S. mitis S. bovis S. bovis S. constellatus S. constellatus S. agalactiae S. agalactiae E. faecalis E. faecalis S. gordonii E. faecium K. sedentarius Patients without IE None None None None None
Heart valve tissue Valve
Gram staining
Culture
SF real-time PCR
Mitral Aortic Mitral Mitral Aortic Aortic Mitral Mitral Mitral Aortic Mitral Aortic Aortic Mitral Aortic Mitral Aortic Aortic Mitral Mitral
Gram-positive cocci Gram-positive cocci Negative for bacteria and fungi Gram-positive cocci Gram-positive cocci Negative for bacteria and fungi Negative for bacteria and fungi Gram-positive cocci in chains Negative for bacteria and fungi Gram-positive cocci in chains Gram-positive cocci Gram-positive cocci in pairs Gram-positive cocci in pairs Gram-positive cocci in chains Gram-positive cocci in pairs Gram-positive cocci Gram-positive cocci Gram-positive cocci in chains Gram-positive cocci in pairs Gram-positive cocci
Negative Negative Negative Negative S. epidermidis Negative Negative S. mitis Negative Negative Negative Negative Negative Negative Negative Negative Negative Negative E. faecium Negative
S. aureus S. aureus S. aureus CoNSa CoNSa CoNSa Streptococcus spp.b Streptococcus spp.b Streptococcus spp.b Streptococcus spp.b Streptococcus spp.b Streptococcus spp.b Streptococcus spp.b Streptococcus spp.b Streptococcus spp.b E. faecalis E. faecalis Streptococcus spp.a E. faecium Negative
Aortic Aortic Aortic Aortic Aortic
Negative Negative Negative Negative Negative
Negative Negative Negative Negative Negative
Negative Negative Negative Negative Negative
for for for for for
bacteria bacteria bacteria bacteria bacteria
and and and and and
fungi fungi fungi fungi fungi
a Coagulase-negative Staphylococcus from SML: Staphylococcus capitis, Staphylococcus caprae, Staphylococcus cohnii, S. epidermidis, Staphylococcus haemolyticus, Staphylococcus hominis, Staphylococcus lugdunensis, Staphylococcus pasteuri, Staphylococcus saprophyticus, Staphylococcus warneri, and Staphylococcus xylosus. b Streptococcus from SML: S. agalactiae, Streptococcus anginosus, S. bovis, S. constellatus, Streptococcus cristatus, S. gordonii, Streptococcus intermedius, Streptococcus milleri group, S. mitis, Streptococcus mutans, Streptococcus oralis, Streptococcus parasanguinis, Streptococcus pyogenes, Streptococcus salivarius, Streptococcus sanguinis, Streptococcus thermophilus, Streptococcus vestibularis, and Streptococcus viridans group.
suspected IE, PCR showed a sensitivity and NPV higher than culture (Breitkopf et al., 2005), although the PCR sensitivity (41.2%) observed does not match that of the present study, probably due to different inclusion criteria. To our knowledge, only few studies evaluated commercially available real-time PCR tests for diagnosis of IE on heart valves. In a study using 2 universal rRNA gene PCR plus sequencing tests, sensitivity of PCR (85%) was nearly twice as high as that of culture (45%) (Kühn et al., 2011), and a previous study on various clinical specimens suggested that SF assay performed on cardiac valves could be a useful tool for molecular diagnosis in patients with IE (Mencacci et al., 2011). Fernández et al. (2010) found a sensitivity of 100% for SF test in a sample of 15 heart valves from patients with definite IE in comparison with culture (30.7%). Indeed, also in the present study, excluding the case caused by K. sedentarius, an organism not included in the SF master list (SML), SF showed a sensitivity of 100%, mirroring the results of the study reported above. However, it should be stressed that, as all the pathogens were isolated from blood culture and all but one were detectable by SF, the sensitivity, specificity, NPV, and PPV data may not perfectly reflect those expected in a true study population or in a routine diagnostic setting, in which it is possible to find patients with clinical diagnosis of IE, but negative blood cultures or IE from pathogens not included in SML. Moreover, as no case included in the study was due to Gram negative bacteria or fungi, the diagnostic accuracy of the test for IE caused by these pathogens remains to be defined. Similarly to other studies employing other commercially available molecular tests (Kühn et al., 2011; Haag et al. 2013), real-time PCR sensitivity was superior to that of culture. This finding can be due to the fact that all cardiac valves were from patients under antibiotic treatment, which hampers the sensitivity of culture. Thus, SF could
have detected DNA of intact non-cultivable microorganisms, as suggested by the positivity on Gram staining of culture-negative valve tissues. However, differently from studies using molecular systems in which free pathogen DNA is eliminated during the extraction protocol (Kühn et al., 2011; Haag et al., 2013), it is also possible that DNA remnants from infectious agents have been detected by SF, in which DNA depletion is not done prior to DNA extraction. Indeed, the finding that DNA can persist on cardiac valves of IE patients for several months after the starting of antimicrobial therapy (Gould, 2012) points for the notion that a positive real-time PCR result can be reliably used to identify the cause of endocarditis but cannot be used to infer ongoing presence of infection and should not, therefore, be used alone to judge the duration of post-operative antimicrobial therapy (Gould, 2012). On the other hand, it is unlikely that the DNA detected by SF was due to sample contamination, due to different findings: i) microorganisms detected matched the known causative pathogens, previously isolated from BC; ii) microorganisms detected were consistent with those observed on Gram staining of valve tissues, in 16/20 cases; iii) SF gave negative results with valves from non-IE patients and K. sedentarius IE patient; and iv) absence of contamination in the real-time PCR experiments, suggested by adequate internal and negative controls. The impossibility to identify pathogens not included in SML is 1 limitation of the SF assay, given that fastidious or slow-growing bacteria, like HACEK (Haemophilus spp., Aggregatibacter actinomycetemcomitans, Cardiobacterium hominis, Eikenella spp., and Kingella spp.), Coxiella burnetii, Bartonella spp., or others can be responsible of IE. All together these pathogens sum up to a prevalence of around 14% in IE (Berbari et al., 1997; Gould, 2012; Watkin et al., 2003). Thus, in case of negative SF result, it would be advisable to test cardiac valves
C. Leli et al. / Diagnostic Microbiology and Infectious Disease 79 (2014) 98–101
by a more sensitive method, such as a 2-step broad-range PCR method or real-time PCR procedures specific for uncommon infectious agents, which can improve molecular diagnosis of IE (Boussier et al., 2013), with additional costs. Nevertheless, since the majority of IE cases are due to microorganisms detectable by SF (Murdoch et al., 2009), laboratories performing the molecular diagnosis of sepsis can use this test for culture-negative heart valves to overcome the problem of employing in-house molecular procedures, requiring a high-quality validation and expert dedicated staff, not easily available in all the clinical microbiology laboratories. However, it must be underlined that not every clinical microbiology laboratory has the equipment and staff for this multiplex real-time PCR analysis as well. Another limitation of SF is that it does not identify at the species level streptococci other than S. pneumoniae and CoNS (Lehmann et al., 2008), although in case of IE from staphylococci, methicillin resistance can be assessed by detection of the mecA gene (Wallet et al., 2011). Even if etiological diagnosis of streptococcal or staphylococcal IE at the genus level could be sufficient enough to guide appropriate antimicrobial therapy (Gould, 2012), nevertheless, the need for DNA sequencing of the amplicons could be required for species identification, with additional costs. Although the strict criteria used ruled out the possibility of erroneous evaluation of the diagnostic accuracy of SF, further studies with a greater number of patients, including cases with IE of unknown origin or suspected, are needed to better define the usefulness of this test in the diagnostic practice. In conclusion, SF assay can be used on culture-negative heart valve tissues, excised from patients with IE, showing sensitivity and NPV higher than culture, with the same high specificity and PPV. In cases of SF positive for CoNS or streptococci, a DNA sequencing of the amplicons would be required to identify the pathogen at the species level. In case of negative SF result, it would be advisable to test cardiac valves by PCR procedures specific for uncommon infectious agents of IE.
References Berbari EF, Cockerill III FR, Steckelberg JM. Infective endocarditis due to unusual or fastidious microorganisms. Mayo Clin Proc 1997;72:532–42. Boussier R, Rogez S, François B, Denes E, Ploy MC, Garnier F. Two-step bacterial broadrange polymerase chain reaction analysis of heart valve tissue improves bacteriological diagnosis of infective endocarditis. Diagn Microbiol Infect Dis 2013;75:240–4. Breitkopf C, Hammel D, Scheld HH, Peters G, Becker K. Impact of a molecular approach to improve the microbiological diagnosis of infective heart valve endocarditis. Circulation 2005;111:1415–21. Casalta JP, Gouriet F, Roux V, Thuny F, Habib G, Raoult D. Evaluation of the LightCycler SeptiFast test in the rapid etiologic diagnostic of infectious endocarditis. Eur J Clin Microbiol Infect Dis 2009;28:569–73. Durack DT, Lukes AS, Bright DK. New criteria for diagnosis of infective endocarditis: utilization of specific echocardiographic findings. Duke Endocarditis Service. Am J Med 1994;96:200–9.
101
Fernández AL, Varela E, Martínez L, Martínez A, Sierra J, González-Juanatey JR, et al. Evaluation of a multiplex real-time PCR assay for detecting pathogens in cardiac valve tissue in patients with endocarditis. Rev Esp Cardiol 2010;63:1205–8. Gauduchon V, Chalabreysse L, Etienne J, Célard M, Benito Y, Lepidi H, et al. Molecular diagnosis of infective endocarditis by PCR amplification and direct sequencing of DNA from valve tissue. J Clin Microbiol 2003;41:763–6. Gould FK. Guidelines for the diagnosis and antibiotic treatment of endocarditis in adults: a report of the Working Party of the British Society for Antimicrobial Chemotherapy. J Antimicrob Chemother 2012;67:269–89. Haag H, Locher F, Nolte O. Molecular diagnosis of microbial aetiologies using SepsiTest™ in the daily routine of a diagnostic laboratory. Diagn Microbiol Infect Dis 2013;76:413–8. Horstkotte D, Follath F, Gutschik E, Lengyel M, Oto A, Pavie A, et al. Task Force Members on Infective Endocarditis of the European Society of Cardiology; ESC Committee for Practice Guidelines (CPG); Document Reviewers. Guidelines on prevention, diagnosis and treatment of infective endocarditis executive summary; the task force on infective endocarditis of the European society of cardiology. Eur Heart J 2004;25:267–76. Kühn C, Disqué C, Mühl H, Orszag P, Stiesch M, Haverich A. Evaluation of commercial universal rRNA gene PCR plus sequencing tests for identification of bacteria and fungi associated with infectious endocarditis. J Clin Microbiol 2011;49:2919–23. Lamas CC, Eykyn SJ. Blood culture negative endocarditis: analysis of 63 cases presenting over 25 years. Heart 2003;89:258–62. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33:159–74. Lehmann LE, Hunfeld KP, Emrich T, Haberhausen G, Wissing H, Hoeft A, et al. A multiplex real-time PCR assay for rapid detection and differentiation of 25 bacterial and fungal pathogens from whole blood samples. Med Microbiol Immunol 2008;197:313–24. Marín M, Muñoz P, Sánchez M, del Rosal M, Alcalá L, Rodríguez-Créixems M, et al, Group for the Management of Infective Endocarditis of the Gregorio Marañón Hospital. Molecular diagnosis of infective endocarditis by real-time broad-range polymerase chain reaction (PCR) and sequencing directly from heart valve tissue. Medicine (Baltimore) 2007;86:195–202. Mencacci A, Leli C, Cardaccia A, Montagna P, Moretti A, Bietolini C, et al. Comparison of conventional culture with SeptiFast real-time PCR for microbial pathogen detection in clinical specimens other than blood. J Med Microbiol 2011;60:1774–8. Mencacci A, Leli C, Montagna P, Cardaccia A, Meucci M, Bietolini C, et al. Diagnosis of infective endocarditis: comparison of the LightCycler SeptiFast real-time PCR with blood culture. J Med Microbiol 2012;61:81–3. Muñoz P, Bouza E, Marín M, Alcalá L, Rodríguez Créixems M, et al. Heart valves should not be routinely cultured. J Clin Microbiol 2008;46:2897–901. Murdoch DR, Corey GR, Hoen B, Miró JM, Fowler Jr VG, Bayer AS, et al. Clinical presentation, etiology, and outcome of infective endocarditis in the 21st century: the International Collaboration on Endocarditis-Prospective Cohort Study. Arch Intern Med 2009;169:463–73. Naber CK, Erbel R. Infective endocarditis with negative blood cultures. Int J Antimicrob Agents 2007;30:S32–6. Prendergast BD. The changing face of infective endocarditis. Heart 2006;92:879–85. Voldstedlund M, Nørum Pedersen L, Baandrup U, Klaaborg KE, Fuursted K. Broad-range PCR and sequencing in routine diagnosis of infective endocarditis. APMIS 2008;116:190–8. Vollmer T, Piper C, Horstkotte D, Körfer R, Kleesiek K, Dreier J. 23S rDNA real-time polymerase chain reaction of heart valves: a decisive tool in the diagnosis of infective endocarditis. Eur Heart J 2010;31:1105–13. Vondracek M, Sartipy U, Aufwerber E, Julander I, Lindblom D, Westling K. 16S rDNA sequencing of valve tissue improves microbiological diagnosis in surgically treated patients with infective endocarditis. J Infect 2011;62:472–8. Wallet F, Herwegh S, Decoene C, Courcol RJ. PCR-electrospray ionization time-of-flight mass spectrometry: a new tool for the diagnosis of infective endocarditis from heart valves. Diagn Microbiol Infect Dis 2013;76:125–8. Wallet F, Loïez C, Herwegh S, Courcol RJ. Usefulness of real-time PCR for the diagnosis of sepsis in ICU-acquired infections. Infect Disord Drug Targets 2011;11:348–53. Watkin RW, Lang S, Lambert PA, Littler WA, Elliott TSJ. The microbial diagnosis of infective endocarditis. J Infect 2003;47:1–11.
