Impact of serology and molecular methods on improving the microbiologic diagnosis of infective endocarditis in Egypt

Transcription

1 Infection (2015) 43: DOI /s ORIGINAL PAPER Impact of serology and molecular methods on improving the microbiologic diagnosis of infective endocarditis in Egypt Amany Aly El Kholy 1 Nevine Gamal El din El Rachidi 1 Mervat Gaber El Enany 1 Eiman Mohammed AbdulRahman 1 Reem Mostafa Mohamed 1 Hussien Hasan Rizk 2 Received: 29 December 2014 / Accepted: 12 March 2015 / Published online: 26 March 2015 Springer-Verlag Berlin Heidelberg 2015 Abstract Background Conventional diagnosis of infective endocarditis (IE) is based mainly on culture-dependent methods that may fail because of antibiotic therapy or fastidious microorganisms. Objectives We aimed to evaluate the added values of serological and molecular methods for diagnosis of infective endocarditis. Patients and methods One hundred and fifty-six cases of suspected endocarditis were enrolled in the study. For each patient, three sets of blood culture were withdrawn and serum sample was collected for Brucella, Bartonella and Coxiella burnetii antibody testing. Galactomannan antigen was added if fungal endocarditis was suspected. Broad range PCR targeting bacterial and fungal pathogens were done on blood culture bottles followed by sequencing. Culture and molecular studies were done on excised valve tissue when available. Results One hundred and thirty-two cases were diagnosed as definite IE. Causative organisms were detected by blood cultures in 40 (30.3 %) of cases. Blood culture-negative endocarditis (BCNE) represented 69.7 %. Of these cases, PCR followed by sequencing on blood and valvular tissue could diagnose five cases of Aspergillus flavus. Eleven patients with BCNE (8.3 %) were diagnosed as zoonotic endocarditis by serology and PCR including five cases of Brucella spp, four cases of Bartonella spp and two cases of * Eiman Mohammed AbdulRahman eimanmousa@yahoo.com 1 2 Clinical Pathology Department, Faculty of Medicine, Cairo University, Giza, Egypt Cardiology Department, Faculty of Medicine, Cairo University, Giza, Egypt Coxiella burnetii. PCR detected three cases of Brucella spp and two cases of Bartonella spp, while cases of Coxiella burnetii were PCR negative. The results of all diagnostic tools decreased the percentage of non-identified cases of BCNE from 69.7 to 49.2 %. Conclusion Our data underline the role of serologic and molecular tools for the diagnosis of blood culture-negative endocarditis. Keywords Blood culture-negative endocarditis Brucella spp. Broad range PCR Introduction Infective endocarditis (IE) is a life-threatening disease with variable clinical signs making it a diagnostic challenge. Modified Duke criteria propose diagnostic algorithms that help to standardize the diagnosis of IE [1]. Besides clinical and laboratory parameters, the main pillars of the Duke criteria are vegetations on the heart valves detected by echocardiography and positive blood cultures. Unfortunately, in a substantial number of IE cases ( %), cultures are negative and the infectious agent remains undiagnosed [1]. Blood culture-negative endocarditis (BCNE) may be due to the inability of fastidious microorganisms to grow by routine methods and growth inhibition of pathogens due to prior antibiotic administration [2]. Such blood culturenegative infections delay both the diagnosis and the initiation of adequate treatment, thereby profoundly impacting clinical outcome. The complement of culture-dependent methods by serologic techniques was shown to be useful for diagnosis of IE in cases of BCNE and for selected organisms such as

2 524 A. A. El-Kholy et al. Coxiella spp, Bartonella spp, Brucella spp, Legionella spp and Aspergillus spp. [3]. PCR methods followed by DNA sequence analysis are highly sensitive and specific tools for detecting and identifying the etiological agents of IE [4]. Hypothesis Serological and molecular methods are of added values of for diagnosis of infective endocarditis. Patients and methods Study design The study included all patients with suspected infective endocarditis (IE) referred to the Endocarditis Service, affiliated to the Cardiology Department, Cairo University Hospitals from May 2010 to September Modified Duke criteria were used for case definition [1]. All patients were subjected to history taking, including demographic data, presenting symptoms, underlying heart disease, co-morbid conditions and contact with animals. Informed consent was taken from each patient before enrollment. Laboratory methods Blood cultures At least three sets of blood culture were obtained as soon as possible after enrollment. We used the BACTEC Plus aerobic/f and BACTEC Plus anaerobic/f blood culture Vials (Becton, Dickinson Spain, MD). Microbiologic examination of the surgically excised specimens For patients who were candidates for surgical intervention, the surgically removed valves were divided into three portions: The first was aseptically ground and subjected to KOH wet mount, Gram s stain, and bacterial culture and sensitivity, and fungal culture [5]. a. Smear Examination: Gram staining and Potassium hydroxide (KOH) wet mount. b. Culture: on blood and MacConkey s agars and on chocolate agar incubated in 5 % CO 2 at 35 C for 72 h and on Sabouraud s dextrose agar incubated at 25 C for up to 21 days and examined every 48 h for any growth. Part of the specimen was inoculated in thioglycolate broth (Oxoid, UK) and incubated at 37 C for 21 days and subcultured if turbidity was noticed. Microorganisms were identified by their growth characteristics, Gram-stained smears and routine biochemical reactions [5]. The second part was put in 4 % formalin to be subjected to histopathological examination. The third part was immediately frozen at 70 C for PCR. Molecular techniques For detection of broad range bacterial 16S rrna and the panfungal 18S and 28S rrna from blood culture bottles and surgical materials when available, followed by sequencing for identification of positive cases. DNA extraction was done from blood culture bottles using the wash/alkali/heat lysis method previously described by Millar et al. [6] and from surgical materials using GeneJET genomic DNA purification kit (#K0721, Fermentas, Vilnius, Lithuania). DNA amplification and cycling conditions for detection of broad range 16SrRNA bacterial gene were done using eubacterial broad range 16SrRNA primer set 536f 5 CAGCAGCCGCGGTAATAC and RP2 5 eacggcacc TTGTTACGACTT (Bioneer, AccuOligo ) [7], and two oligonucleotide pan-fungal primers to amplify the conserved regions of the 18S (ITS-1) and 28S (ITS-4) rrna. We used primer sequences: ITS-1 5 TCCGTAGGTGAACCTG CGGG and ITS-4 5 GCTGCGTTCTTCATCGATGC as previously described [8]. A positive control was included in each run using Staph aureus (ATCC 25923), E.coli (ATCC 25922) and Candida spp. clinical isolate in addition to a negative control. PCR products any band >200 bp was considered positive [9]. Purification of PCR products was done using GenJET PCR purification kit (#K0701, Fermentas, Vilnius, Lithuania) according to manufacturer instructions. Cycle sequencing The nucleic acid sequence of PCR amplified gene fragments was determined using automatic DNA sequencing (ABI Prism 310 Genetic Analyzer, Applied Biosystems). Samples with a minimum length of 200 bp were sequenced, while those <200 bp were considered inadequate for sequencing. Sequencing was done using Applied Biosystem PRISM BigDye terminator cycle sequencing ready reaction kit (V3.1) according to manufacturer instructions using the 536f 5 CAGCAGCCGCGGTAATAC primer for bacterial pathogens [10] and the ITS-1 5 TCCGTAGGTGAACC TGCGGG primer for fungal pathogen as previously described [11].

3 Impact of serology and molecular Sequence analysis Sequences obtained were compared to those available in GenBank database using Mega BLAST ( nlm.nih.gov/blast). PCR was regarded as positive if the results matched those of blood culture. If blood culture was negative, the PCR result was regarded as positive if the agent was typical for IE [12]. Serologic testing Detection of agglutinating anti-brucella antibodies was done by tube agglutination test (linear chemicals, Montgat- Barcelona, Spain). Testing for IgG antibodies against Bartonella henselae and Bartonella quintana, IgG, IgM and IgA antibodies against Coxiella burnetii was done by indirect Immunofluorescence assay (Vircell S.L. microbiologist, Granada, Spain). Interpretation of the serology results The patient was considered to have Brucella when antibody titers 1/320 [13], Bartonella endocarditis when IgG titers 1:800 [14], Coxiella endocarditis when IgG phase I titer 1:800 was observed [1]. Aspergillus galactomannan antigen detection was done using the Platelia EIA (Bio-Rad, Marnes-La-Coquette, France). Patients with an index >0.5 were considered positive for Aspergillus antigen [15]. Results One hundred and fifty-six patients with suspected infective endocarditis were enrolled, based on modified Duke criteria [1], and 132 cases were diagnosed as definite endocarditis. The patients age was years (Mean 31.07, SD 9.43). Native valve IE occurred in 70 % of cases, compared to prosthetic valve IE in 30 %. The mitral valve was the most commonly involved (31.3 %), followed by the aortic valve (25.4 %) then combined aortic and mitral affection (20.9 %) and Tricuspid valve (14.9 %). One patient (2.6 %) had pacemaker; 25 % of cases gave history of rheumatic heart disease, 6 % had CVL-related blood stream infection and 5.3 % had history of intravenous drug abuse. History of antibiotic intake prior to blood culture withdrawal was positive in 98 patients (74.2 %). Blood culture was positive in 40 (30.3 %) patients, so blood culture-negative infective endocarditis (BCNE) represented 69.7 % of patients S.aureus Enterococcus P.aeruginosa Acinetobacter spp. 525 Staphylococcus aureus was the mostly encountered microorganism represented (10.6 %), followed by viridans Streptococci (7.6 %), Enterococcus spp. (5.3 %), coagulase-negative staphylococci (CoNS) (3 %), Pseudomonas aeruginosa (1.5 %), Propionobacterium acne, Acinetobacter spp., Candida albicans each represented (0.75 %) and mixed growth of CoNS and viridans Streptococci in (0.75 %) of cases (Fig. 1). MRSA was identified in nine (64.3 %) isolates of S.aureus; no vancomycin resistance was detected in staphylococcus or streptococcus strains isolated. Time to blood culture positivity was 1 6 days (Mean 3.33, SD 1.459). Broad range PCR for amplification of bacterial 16SrRNA and fungal 18-28SrRNA on blood culture samples was positive in 44 cases (33.3 %) and 8 cases (6.1 %), respectively. Subsequent sequencing of the amplified product was performed on 45 positive PCR cases, double-band or shortband length were inadequate for sequencing. In our study, microorganisms identified by blood culture compared to those identified by blood broad range PCR followed by sequencing were concordant in 67.5 % of cases, and PCR showed a sensitivity of 88.3 %, specificity 92 %, positive predictive value 79.1 %, negative predictive value 95.8 % and accuracy 91 %. The discordance between blood culture and PCR results is shown in Table 1. Geobacillus, Bordatella holmesii, Staph hemolyticus, Aspergillus oryzae and Saccharomyces cerviciae are considered as contaminants as they are of questionable clinical significance. PCR could additionally diagnose three cases of Aspergillus flavus, all were BCNE. PCR corrected the identification of microorganisms isolated from blood culture in three cases (Table 1). PCR failed to diagnose eight blood culture-positive cases (Table 1). Surgical material Mixed growth Fig. 1 Distribution of microorganisms isolated from blood culture Of the 156 patients, 57 (36.5 %) underwent surgical intervention for valve replacement, embolectomy or abscess

4 526 A. A. El-Kholy et al. Table 1 Discordant results between organisms identified by blood culture and those identified by blood PCR-sequencing Organisms isolated from blood culture (number) Results of blood PCR-sequencing (number) Staph. aureus (3) PCR was negative Enterococcus spp (3) PCR was negative (1), Strept anginosus (1) and Achromobacter dentrificans (1) Achromobacter dentrificans (1) Propionobacterium acne (1) Bacillus safensis Strept. viridans (5) PCR was negative (1), inadequate band (3) and Brevibacillus brevi (1) Blood culture negative (9) Geobacillus (1), Bordatella holmesii (1), Achromobacter xylosoxidans (1), Staph.hemolyticus (1), Aspergillus flavus (3), Aspergillus oryzae (1) and Saccharomyces cerviciae (1) Table 2 Distribution of identified organisms by valve culture Organisms identified by valve culture (number) Result of blood culture CoNS (6) No growth (5) and CoNS (1) Acinetobacter spp. (1) No growth Pseudomonas aeruginosa (1) Pseudomonas aeruginosa Staph. aureus (1) Staph.aureus Citrobacter spp (1) Acinetobacter junii (1) Aspergillus flavus (1) No growth Candida albicans (1) Candida albicans excision, 55 of which were available for culture and histopathological examination and 45 to molecular studies. Histopathological examination of surgically excised tissue sections showed the presence of typical IE cases in 34 (61.8 %) of cases, classified according to Voldstedlund et al. [12]. The presence of fungal hyphae was demonstrated in valve tissue sections stained with hematoxylin and eosin in four cases. Valve culture was positive in 12 out of 55 (21.8 %) of cases, and valve Gram stain was positive for the presence of an organism in 6 (10.9 %) of cases, all were valve culture positive. KOH examination of the excised valves revealed only two positive cases (3.6 %): one was culture positive for Candida albicans and the other one was positive for Aspergillus flavus. The distribution of identified organisms by valve culture compared to the results of blood culture is shown in Table 2. PCR done on valve tissue using primers for 16SrRNA was positive in 23 out of 45 (51.1 %) of cases, while broad range fungal PCR using primers for 18-28SrRNA was positive in 3 cases (6.7 %). Agreement between the results of valve culture and PCR was 62.2 % for the 16S primer and 80 % for the 18-28S primers. Results of sequencing of amplified PCR products from valve tissue are shown in Table 3. PCR done on valvular tissue could additionally diagnose four cases of BCNE (Table 3). PCR on valve tissue corrected the identification of microorganisms isolated from valve culture in one case (Table 3). PCR failed to detect two-valve culture-positive cases (Table 3). Using serology and PCR, zoonotic infective endocarditis could be diagnosed in 11 (8.3 %) of definite IE cases; five cases (3.8 %) with Brucella spp., four (3 %) Bartonella spp. and two (1.5 %) Coxiella burnetii, all was BCNE. PCR on blood and excised valves correctly detected 3/5 cases of Brucella, 2/4 cases of Bartonella spp., while cases of Coxiella burnetii were PCR negative. PCR followed by sequencing on blood and valvular tissue could diagnose five cases of Aspergillus flavus which were BCNE. Aspergillus galactomannan was positive in 19 (14.4 %) of cases, all were BCNE. Aspergillus galactomannan results showed an agreement of 89.8 and 84.4 % with 18-28S rrna (ITS) PCR performed on blood and valves, respectively. One case that was diagnosed as Aspergillus flavus by PCR, and histopathology was galactomannan negative. Ten patients (7.6 %) died in spite of proper treatment guided by the lab results, two cases with S. aureus, two with S. viridans, two with achromobacter spp., one with aspergillus spp. and three were BCNE. Relapse was recorded in five patients (3.8 %), two cases of Brucella spp., two with CoNS and one with S. aureus. Discussion Our study represented the cooperation of the Endocarditis Service, affiliated to the Cardiology Department, Cairo University Medical School with the Clinical Microbiology laboratory in an attempt to improve the diagnosis of cases of infective endocarditis (IE). The Endocarditis Service is a referral center for the Greater Cairo regions as well as

5 Impact of serology and molecular 527 Table 3 Organisms identified by sequencing of PCR positive valve tissue compared to results of valve culture, blood culture and blood PCR positivity Organisms identified by valve sequencing Result of valve culture Result of blood culture Result of blood PCR (number) Brucella spp. (4) Negative Negative Negative Strept. mutans (3) Negative 1 was positive 1 was positive, 1 was negative and 1 yielded inadequate band Staph. aureus (3) 1 was positive All was positive 2 cases were positive Strept. oralis (2) Negative All was positive All was positive Aspergillus flavus (2) 1 was positive Negative Negative Bartonella quintana (2) 1 was positive for Negative Negative Pseudomonas aeruginosa Strept. mitis (2) 1 was positive for CoNS and 1 was negative 1 was positive and the other was negative 1 was positive and the other was negative Acinetobacter baumannii (1) Negative Negative Negative Acinetobacter junii (1) It was positive for Citrobacter spp. Negative Negative Staph. epidermidis (1) Positive Negative Negative Kytococcus schroeteri (1) It was positive for CoNS Negative Negative Candida albicans (1) Positive Positive Positive Pseudomonas stutzeri (1) Negative Negative Negative Staph. lugdunensis/epidermidis (1) It was positive for CoNS Negative Negative Staph. spp./micro-coccus (1) It was positive for CoNS Negative Negative Governorates from Delta and Upper Egypt where the IE medical care is inadequate. In our study, blood culture was positive in 40 patients, Staphylococcus aureus represented (10.6 %), the most encountered organism from blood culture, followed by viridans Streptococci (6.8 %) then Enterococcus spp. (5.3 %). Similarly, in Sengal, Staphylococcus aureus represented (15.3 %) [16], in Saudi Arabia Staphylococcus aureus was isolated in 22.7 % and Streptococci in 18.2 % [17]. In a study conducted in Japan, Streptococcus isolated in 58 % of IE cases with positive blood culture [18]. In our study, blood culture was negative in 69.7 % of definite IE cases; similarly, BCNE in Senegal represented 69 % [16], in Algeria 61 % [19] and in Saudi Arabia 45.5 % [17]. On the other hand, in Western countries, BCNE is relatively rare and was reported as 9 % in France [21], 10 % in Sweden [20] and 28 % in Denmark [12]. This difference may be related to the unrestricted antibiotic intake circumstances in developing countries and the high prevalence of zoonotic infections with fastidious organisms. Using serologic and molecular techniques, 8.3 % of cases were diagnosed as zoonotic infective endocarditis; all of them were blood culture negative. Higher rates were reported; in a study conducted in Algeria where zoonotic IE represented 15.6 % [19]. Our cases of zoonotic IE include five (3.8 %) cases of Brucella spp., four (3 %) cases of Bartonella spp. and two (1.5 %) cases of Coxiella burnetii. Previous studies stated that brucellosis is endemic in Egypt with high seropositivity titers [13]; reinforcement of control measures of brucellosis in animals should be a priority. Higher rate of Bartonella (11.4 %) was reported in Algeria, followed by Brucella (3.3 %) and Coxiella (3.3 %) [19] and in Tunisia, Bartonella accounted for (9.8 %) [22]. On the other hand, in UK, the most common organism was Coxiella burnetii (12.7 %) followed by Bartonella (1.1 %) [23], and in France, Coxiella burnetii accounted for 57.3 % of BCNE cases and Bartonella (19.2 %) [24]; these variations may be due to differences in socioeconomic standards, type of animal exposure, standards of veterinary care and laboratory methods. When we used broad range PCR followed by sequencing, concordance with blood culture was 67.5 %. Other studies reported 53.4 [25] and 48 % [26] agreements between the two methodologies. In our study, PCR showed a sensitivity of 88.3 %, specificity 92 %, positive predictive value 79.1 %, negative predictive value 95.8 % and accuracy 91 %, another study reported lower PCR sensitivity (61 %) but higher specificity (100 %) [27], while other authors reported higher sensitivity (96 %) and comparable specificity 95.3 % [28]. In our study, there were eight samples positive by blood culture but negative for 16SrRNA PCR; all was Gram-positive cocci. PCR-negative results in culture-positive patients especially with Gram-positive pathogens were previously

6 528 A. A. El-Kholy et al. reported and were explained by the effect some of technical factors related to some Gram-positive bacteria, such as, difficulties in breaking the cell walls of Gram-positive organisms during sample preparation that resulted in failure in the DNA extraction process [29, 30]. PCR was positive for Achromobacter xylosoxidans in a BCNE patient on hemodialysis; it may be present in low count that was not detected by the blood culture and needed confirmation by valve culture, histopathology or valve PCR-sequencing to demonstrate its role in IE [31]; unfortunately, this patient was not candidate for surgery. In our study, the sensitivity of histological examination of excised valves was 61.8 %, and the specificity was 100 %; similar results were reported in previous study which reported 61 % sensitivity and specificity 100 % [27]. Valve culture in our study was positive in 21.8 % of cases; previous studies reported 23 and 24 % positivity of valve culture [19, 20]. The performance of 16SrRNA PCR on valvular tissue followed by sequencing showed an agreement of 53.3 % with blood culture results. Our results were more favorable than an agreement of 30 % found in a previous study [32]. Of the 11 cases of zoonotic IE, PCR correctly detected three cases of Brucella (one of them was negative by serology) and four cases of Bartonella spp. (one of them was negative by serology), while cases of Coxiella burnetii were PCR negative. All cases belonged to BCNE patients. Those organisms are present mainly intracellular in vegetations on affected cardiac valves, so the positivity of each test may be affected by the disease stage [33]. This raises the importance of serological diagnosis and the confirmation by valve PCR after surgical resection. PCR followed by sequencing on blood and valvular tissue could diagnose five cases of Aspergillus flavus, all belonged to BCNE patients; fungal IE was documented as a cause for BCNE in previous studies [34, 35]. Aspergillus galactomannan results showed an agreement of 89.8 and 84.4 % with 18-28S rrna (ITS) PCR performed on blood and valves, respectively; as one of the five cases of Aspergillus flavus had a negative result for galactomannan antigen but was confirmed by histopathological examination identifying fungal hyphae, a finding that was previously reported [36]. Many variables contribute to the release of circulating galactomannan, some of them related to the fungus in the form of fungal strain, growth rate and rate of release of fungal antigen, other factors are host-related such as underlying diseases, extent of fungal disease, antifungal treatment and age [37]. Aspergillus galactomannan antigen assay can be a useful adjunctive test to establish the diagnosis of invasive aspergillosis, but its role has not been studied well in the setting of endocarditis. The cost of PCR method performed in our study is affordable and easily applicable to the clinical microbiology laboratory. Its cost-effectiveness does not only pertain to materials and labor but also to the add-on value of the early administration of appropriate treatment when a patient is bacteremic, thus reducing the mortality and morbidity rates, and the reduction in empiric therapy in patients who have no systemic infections. However, DNA amplification techniques do not establish antibacterial susceptibilities, which must still be determined by conventional methods. From our study, we could establish the add-on value of molecular techniques in the settings of infective endocarditis especially BCNE; however, PCR-based techniques cannot replace conventional culture techniques. It must be interpreted with caution and hand in hand with traditional histopathology and culture. In addition, introduction of serological techniques is crucial as a supplement for diagnosis especially in patients with zoonotic infections. Conflict of interest None. References 1. Li JS, Sexton DJ, Mick N, et al. Proposed modifications to the Duke Criteria for the diagnosis of infective endocarditis. Clin Infect Dis. 2000;30: Habib G, Hoen B, Tornos P, et al. Guidelines on the prevention, diagnosis, and treatment of infective endocarditis. Eur Heart J. 2009;30: Raoult D, Casalta JP, Richet H, et al. Contribution of systematic serological testing in diagnosis of infective endocarditis. J Clin Microbio. 2005;43: Moter A, Musci M, Schmiedel D. Molecular methods for diagnosis of infective endocarditis. Curr Infect Dis Rep. 2010;12: Manual of clinical Microbiology 9th ed. Murray PR, Baron EJ, Jorgensen JH et al (eds), American Society for Microbiology, 2009, Washington DC. 6. Millar BC, Jiru XU, Moore JE, et al. A simple and sensitive method to extract bacterial, yeast and fungal DNA from blood culture material. J Microbiol Methods. 2000;42: Goldenberger D, Kunzli A, Vogt P, et al. Molecular diagnosis of bacterial endocarditis by broad-range PCR amplification and direct sequencing. J Clin Microbiol. 1997;35: White T, Burns T, Lee S et al. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In PCR protocols. A guide to methods and applications. Innis MA, Gelfand DH, Sninsky JJ and White TJ (ed.) 1990, p Grijalva M, Horváth R, Dendis M, et al. Molecular diagnosis of culture negative infective endocarditis: clinical validation in a group of surgically treated patients. Heart. 2003;89: Bousbia S, Papazian L, Auffray JP, et al. Tropheryma whipplei in patients with pneumonia. Emerg Infect Dis. 2010;16: Henry TP, Iwen C, Hinrichs SH. Identification of Aspergillus species using internal transcribed spacer regions 1 and 2. J Clin Microbiol. 2000;38:

7 Impact of serology and molecular 12. Voldstedlund M, Pedersen NL, Baadrup U, et al. Broad range PCR and sequencing in routine diagnosis of infective endocarditis. APMIS. 2008;116: El-Sherbini A, Kabbash I, Schelling E, et al. Seroprevalence and local variation of human and livestock brucellosis in two villages in Gharbia governorate Egypt. Trans R Soc Trop Med Hyg. 2007;101: Fournier PE, Mainardi JL, Raoult D. Value of micro-immunoflourescence for diagnosis and follow up of Bartonella endocarditis. Clin Diag Lab Immunol. 2002;9: Kalkhe AS, Rouphael N, El Chami MF, et al. Aspergillus endocarditis: a review of literature. Internat J Infect Dis. 2010;14:e Ndiaye MB, Diao M, Kane A, et al. Infective endocarditis in the middle cardiac Dakar: descriptive study about 39 cases The Pan African Medical Journal ISSN 2010; Assiri AS. Clinical and microbiological profiles of infective endocarditis in a tertiary hospital in Aseer region, Saudi Arabia. J Saudi Heart Assoc. 2011;23: Miyazato A, Ohkusu K, Tabata M, et al. Comparative molecular and microbiological diagnosis of 19 infective endocarditis cases in which causative microbes were identified by PCR-based DNA sequencing from the excised heart valves. J Infect Chemother. 2011s. doi: /s Benslimani A, Fenollar F, Lepidi H, et al. Bacterial zoonoses and infective endocarditis, Algeria. Emerg Infect Dis. 2005;11: Vondracek M, Sartipy U, Aufwerber E, et al. 16S rdna sequencing of valve tissue improves microbiological diagnosis in surgically treated patients with infective endocarditis. J Infect. 2011;62: Hoen B, Alla F, Setton-Suty C, et al. Changing profile of infective endocarditis: results of a one year survey in France. JAMA. 2002;288: Znazen A, Rolain JM, Hammami N, et al. High prevalence of Bartonella quintana endocarditis in Sfax Tunisia. Am J Trop Med Hyg. 2005;72: Lamas CC, Eykyn SJ. Blood culture negative endocarditis: analysis of 63 cases presenting over 25 years. Heart. 2003;89: Fournier PE, Thuny F, Richet H, et al. Comprehensive diagnostic strategy for blood culture-negative endocarditis: a prospective study of 819 new cases. Clin Infect Dis. 2010;51: Wellinghausen N, Kochem AJ, Disqué C, et al. Diagnosis of Bacteremia in Whole-Blood Samples by Use of a Commercial 529 Universal 16S rrna Gene-Based PCR and Sequence Analysis. J Clin Microbiol. 2009;47: Ruppenthal RD, de Pereira FS, Cantarelli VV, et al. Application of broad-range bacterial PCR amplification and direct sequencing on the diagnosis of neonatal sepsis. Braz J Microbiol. 2005;36: Greub G, Lepidi H, Rovery C, et al. Diagnosis of infectious endocarditis in patients undergoing valve surgery. Am J Med. 2005;118: Marín M, Muñoz P, Sánchez M, et al. 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: Rovery C, Greub G, Lepidi H, et al. PCR detection of bacteria on cardiac valves of patients with treated bacterial endocarditis. J Clin Microbiol. 2005;43: Rothman RE, Majmudar MD, Kelen GD, et al. Detection of bacteremia in emergency department patients at risk for infective endocarditis using universal 16S rrna primers in a decontaminated polymerase chain reaction assay. J Infect Dis. 2002;186: Ahmed MS, Nistal C, Jayan R, et al. Achromobacter xylosoxidans, an emerging pathogen in catheter-related infection in dialysis population causing prosthetic valve endocarditis: a case report and review of literature. Clin Nephrol. 2009;71: Podglajen I, Bellery F, Poyart C, et al. Comparative molecular and microbiologic diagnosis of bacterial endocarditis. Emerg Infect Dis. 2003;9: Brouqui P, Raoult D. Endocarditis due to rare and fastidious bacteria. Clin Microbiol Rev. 2001;14: Gumbo T, Taege AJ, Mawhorter S, et al. Aspergillus valve endocarditis in patients without prior cardiac surgery. Medicine. 2000;79: Kalokhe AS, Rouphael N, El Chamia MF, et al. Aspergillus endocarditis: a review of the literature. Internat J Inf Dis. 2010;14:e Pemán J, Ortiz R, Osseyran F, et al. Native valve Aspergillus fumigatus endocarditis with blood culture positive and negative for galactomannan antigen. Case report and literature review. Rev Iberoam Micol. 2007;24: Mennink-Kersten MA, Donnelly JP, Verweij PE. Detection of circulating galactomannan for the diagnosis and management of invasive aspergillosis. Lancet Infect Dis. 2004;4: