Reproducibility of transthoracic echocardiography in small animals using clinical equipment

Abstract

Objective Transthoracic echocardiography has been employed to assess left ventricular dimensions and function in small animals. The aim of this study was to identify the limits of transthoracic echocardiography in a commonly used Wistar rat model by assessing intraobserver variability, interobserver variability, and day-to-day variability of examinations implying registrations and measurements. Methods Twenty male adult Wistar rats (body weight 496 52 g) were examined under volatile isoflurane anesthesia (heart rate 302 26 bpm) by transthoracic echocardiography (Sonos 7500; Philips) with a 15 MHz-transducer. For calculation of intraobserver variability, examinations were repeated by the same examiner and for interobserver variability, examinations were performed independently by two investigators. For day-to-day variability, examinations were repeated 14 days later. Left ventricular diameters and areas were analyzed in parasternal short axis and in a modified parasternal long axis. Fractional shortening, area shortening, ejection fraction, stroke volume, and cardiac output were calculated. Results Left ventricular end-diastolic diameter was 8.9 +/- 0.6 mm, fractional shortening 39.0 +/- 5.3%, area shortening 59.6 +/- 6.1%, ejection fraction 83.3 +/- 5.1%, stroke volume 0.24 +/- 0.06 ml, and cardiac output 72.9 +/- 20.6 ml/ min. Intraobserver variability of left ventricular end-diastolic diameter, fractional shortening, area shortening, and ejection fraction was less than 10%, increasing to 19% for stroke volume and cardiac output. Interobserver variability of left ventricular end-diastolic diameter, fractional shortening, area shortening, ejection fraction was less than 13%, increasing to 23% for stroke volume and 25% for cardiac output. Day-to-day variability of left ventricular end-diastolic diameter, area shortening, ejection fraction was less than 11% whereas for stroke volume it was 21% and for cardiac output it was 22%. F-ratio test comparing investigated variabilities did not reveal significant differences. Conclusions M-mode and two-dimensional echocardiography in large rats by clinically common high-end ultrasound systems can be assessed reliably. Parameters of global left ventricular performance like stroke volume and cardiac output could not be assessed with similar reliability.