Infective Endocarditis in Northeastern Thailand

George Watt

, Orathai Pachirat, Henry C. Baggett, Susan A. Maloney, Viraphong Lulitanond, Didier Raoult, Saithip Bhengsri, Somsak Thamthitiwat, Anucha Paupairoj, Michael Kosoy, Nongrak Ud-Ai, Wichuda Sukwicha, Toni Whistler, and Pierre-Edouard Fournier

Author affiliations: Thailand Ministry of Public Health–US Centers for Disease Control and Prevention Collaboration, Nonthaburi, Thailand (G. Watt, H.C. Baggett, S.A. Maloney, S. Bhengsri, S. Thamthitiwat, W. Sukwicha, T. Whistler); Khon Kaen University Faculty of Medicine, Khon Kaen, Thailand (O. Pachirat, V. Lulitanond, A. Paupairoj, N. Ud-Ai); Aix Marseille Université, Marseille, France (D. Raoult, P.-E. Fournier); Centers for Disease Control and Prevention, Atlanta, Georgia, USA (H.C. Baggett, S.A. Maloney, T. Whistler); Centers for Disease Control and Prevention, Fort Collins, Colorado, USA (M. Kosoy)

Abstract

Despite rigorous diagnostic testing, the cause of infective endocarditis was identified for just 60 (45.5%) of 132 patients admitted to hospitals in Khon Kaen, Thailand during January 2010–July 2012. Most pathogens identified were Viridans streptococci and zoonotic bacteria species, as found in other resource-limited countries where underlying rheumatic heart disease is common.

Serologic testing of patients with blood culture–negative endocarditis has identified Coxiella burnetii, the causative agent of Q fever, and Bartonella spp. as noteworthy causes of infective endocarditits (IE) in resource–limited countries (1–4). Many cases of IE that were not diagnosed by standard blood culture were caused by zoonotic bacteria (5). Prospective, systematic descriptions of the etiology and characteristics of IE in Southeast Asia are lacking. We therefore collected detailed clinical, laboratory, and epidemiologic information for patients with confirmed IE in Khon Kaen, Thailand, and conducted specialized testing methods in addition to standard blood cultures to facilitate assessment for zoonotic and nonzoonotic bacteria as the cause of IE.

The Study

During January 25, 2010–July 19, 2012, patients were prospectively enrolled in this study at 2 tertiary care referral hospitals located on the campus of the medical school of Khon Kaen University. Srinagarind Hospital is a 777-bed general hospital, and the Queen Sirikit Heart Center of the Northeast is a 200-bed specialized cardiac center in which ≈10 heart valve replacement surgeries are performed each month. Patients with suspected IE are referred from much of northeastern Thailand, a region of ≈21 million persons.

Transthoracic echocardiography was performed for patients suspected of having IE; consenting patients >16 years of age who met modified Duke criteria for endocarditis were enrolled in this study. Underlying cardiac conditions were assessed by cardiologists on the basis of patients’ medical records, history, physical examination, and echocardiographic findings. At admission to a hospital, 3 separate blood samples for culture were obtained in <90 minutes. Blood was inoculated into aerobic medium (BD BACTEC Plus Aerobic/F Medium; Becton Dickinson, Franklin Lakes, NJ, USA), and cultures were processed by using an automated system (BD Bactec Fx series; Becton Dickinson). Pathogens were identified to species level whenever possible, but some blood culture isolates were defined only to the genus level (e.g., viridans group streptococci). One month after admission, a convalescent-phase serum specimen was obtained from each study patient and these patients were evaluated by a cardiologist.

Acute- and convalescent-phase serum specimens were tested for C. burnetii and Legionella pneumophila by indirect immunofluorescence assay (IFA) as described (5). Phase 1 IgG reciprocal titers >800 for C. burnetii and total antibody reciprocal titers ≥256 for L. pneumophila on either serum specimen were defined as positive (5). IFA IgG reciprocal titers of ≥800 to Bartonella quintana and Bartonella henselae were deemed positive. Specific antibodies to Brucella melitensisand Mycoplasma pneumoniae were detected with a commercial immunoenzymatic antibody test (Brucella antibody and Platellia M. pneumoniae IgM kits, respectively; Bio-Rad, Marnes-la-Coquette, France). Reciprocal titers ≥200 were considered positive. DNA was extracted from surgically excised heart valves by using the QIAamp DNA FFPE Tissue Kit (QIAGEN, Courtaboeuf, France) as described by the manufacturer. Previously described broad spectrum PCR primers and amplification and sequencing conditions (4) were used to detect all bacteria (16S rRNA); all fungi (18S rRNA); Staphylococcus aureus, mitis and gallolyticus group streptococci, Enterococcus faecalis and E. faecium, Mycoplasma hominis, C. burnetii;, Bartonella spp., and Tropheryma whipplei.