Patient population

The study was performed in the Paediatric cardiology unit, Yuzuncu Yil University, Faculty of Medicine, Van, Turkey. The study group included children diagnosed with brucellosis in the Paediatric infectious unit. Sixty-seven children with brucellosis and 40 age- and sex-matched healthy patients with an innocent murmur were included as a control group. A full history was recorded, and a complete physical examination was performed by the same physician. Bodyweight, heart rate, blood pressure and electrocardiogram of all children were recorded.

Children with the structural and functional cardiac abnormality, pulmonary and systemic hypertension, anaemia, moderate or severe malnutrition, any other systemic disorders, and comorbid diseases, such as upper or lower respiratory infection, were excluded. The diagnosis was established by a positive Brucella Wright and Coombs tube agglutination tests, and/or isolation Brucella species from the blood. Echocardiographic examinations were performed before given the antibiotic therapy. This study was approved by the local ethics committee. All parents gave informed consent for participation in the study.

Echocardiography

Transthoracic echocardiography was performed by a single experienced Paediatric cardiologist. An electrocardiogram was simultaneously recorded for all patients. The patients were studied without sedation while they were lying quietly in the supine and left lateral positions. Assessments were performed using the General Electric Medical Systems ViVid 6 dimension echocardiography device, a 6-MHz probe (GE Vingmed Ultrasound AS, Horten, Norway). A complete echocardiographic study was performed using standard views and techniques. All measurements followed the recommendations of the American Society of Echocardiography. ^{Reference Pearlman, Gardin and Martin4} Imaging windows obtained from the parasternal long axis and the apical four-chamber were used. The measurements of at least three cardiac consecutive cycles were averaged in sinus rhythm. All echocardiographic assessments were digitally recorded to enable later investigation. The analyses were performed using a commercially available computer software program (Echopac 2008; GE Vingmed).

End-diastolic and end-systolic dimensions of the left ventricle and anterior wall thickness were measured by M-mode at the parasternal long axis. Fractional shortening (%) and ejection fraction (%) measurements were obtained on the echocardiography. Mitral annular plane systolic excursion and tricuspid annular plane systolic excursion were measured in millimetres as previously described. ^{Reference Pai, Bodenheimer and Pai5,Reference Karaye, Habib and Mohammed6} Peak systolic velocities were evaluated by pulsed-wave Doppler. Peak systolic velocity waveforms from the ascending aorta were obtained from the five-chamber view. Peak systolic velocity waveforms from the pulmonary artery were obtained from the short axis view. Pulmonary artery pressures were obtained by the systolic transtricuspid gradient. ^{Reference Garson, Bricker, Fisher and Geva7} All pulmonary artery pressures were within normal limits.

Mitral and tricuspid filling velocities were recorded from the apical four-chamber view with the pulsed-wave Doppler sample volume placed at the tip of its leaflets during diastole. The peak velocity during early diastole (E), peak velocity during late diastole (A) and deceleration time of the early diastolic velocity were used as both ventricular diastolic function parameters. The E/A ratios were calculated for mitral and tricuspid valves. ^{Reference Lang, Bierig and Devereux8}

Right and left ventricle functions were also evaluated using tissue Doppler imagination. For optimal records, Nyquist limits were adjusted to 15–20 cm/s, sample volume to 2– 5 mm and monitor velocity to 100 mm/s. Tissue Doppler imagination of mitral and tricuspid valves were performed at the level of the valve annuli in apical four-chamber planes. Isovolumic relaxation time from the end of the peak systolic velocity (S-wave) to the beginning of Ea-wave and isovolumic contraction time from the end of the Aa-wave to the beginning of the S-wave were measured. The ejection time was measured from the beginning to the end of the S-wave. Left and right ventricular myocardial performance index combining systolic and diastolic time intervals were calculated as the sum of isovolumic contraction time and isovolumic relaxation time divided by ejection time. ^{Reference Tei, Dujardin and Hodge9,Reference Rybczynski, Koschyk and Aydin10} The resulting velocities were recorded for three cardiac cycles. The mean values of the measurements were used for statistical analysis.