Studies were undertaken using transoesophageal Doppler echocardiography to monitor left ventricular systolic function in anaesthetised horses. A 3.5 MHz transoesophageal probe was specifically developed in collaboration with Vingmed Sound for equine use. The indices of systolic function investigated were maximum acceleration of aortic blood flow (dv/dt_max), maximum blood flow velocity, (V_max), cardiac output (CO), left ventricular pre-ejection period (PEP) and left ventricular ejection time (ET). The feasibility of the technique was demonstrated in a group of 8 healthy Thoroughbred horses anaesthetised using a standard protocol. It was established that two dimensional transoesophageal echocardiography provided a reference view of the left ventricular outflow tract and aorta that consistently allowed high quality Doppler echocardiographic measurement of aortic blood flow velocity. The flow envelopes obtained were suitable for measurement of indices of left ventricular systolic function. The repeatability of the measured indices was similar to that of the maximum rate of rise of left ventricular pressure (LVdp/dt_max), obtained simultaneously by cardiac catheterisation. Cardiac output estimations made using transoesophageal Doppler echocardiography were compared with those obtained by thermodilution in the same group of horses under general anaesthesia. Cardiac output was altered by infusions of the sympathomimetic amine, dobutamine. Aortic velocity spectra obtained both by high pulse repetition frequency and continuous wave insonation modes were used to obtain the velocity time integral for calculation of cardiac output. The measurements derived from transoesophageal echocardiography agreed well with those obtained by thermodilution. Both correlation coefficients and limits of agreement between the two techniques were better than those obtained from similar studies in standing horses using transthoracic echocardiography. The sensitivity of the Doppler derived indices of left ventricular function on inotropic intervention was assessed in the final sequence of studies. As these indices are derived during the ejection period they are load dependent, so their response to changes in ventricular loading was also assessed and compared with the most commonly used index of myocardial contractility in horses, LVdp/dt_max.