Jain U. University of California at San Francisco.J Clin Anesth 1997 Sep;9(6):487-92. STUDY OBJECTIVES: To evaluate the automated determination of onset and offset times and amplitudes of all the PQRST waves from simultaneously recorded surface electrocardiogram (SECG) and unipolar esophageal ECG (EsECG). The occurrence of ST segment deviation is also examined. DESIGN: Prospective, observational study. SETTING: University hospital. PATIENTS: 30 patients undergoing coronary artery bypass graft (CABG) surgery. INTERVENTIONS: SECG and two-lead unipolar EsECG were recorded after induction of anesthesia and before cardiopulmonary bypass (CPB). MEASUREMENTS AND MAIN RESULTS: The amplitudes of the P and T waves and the ST segment deviation were measured. EsECG had more noise than SECG. Slight movement of the esophageal electrodes occasionally caused substantial changes in the wave amplitudes and ST segment deviation in the unipolar EsECG. The maximum P wave amplitude in EsECG was, on average, 97% greater than the maximum P wave amplitude in SECG, ST segment deviation in EsECG was observed in the absence of ST segment deviation in SECG and vice versa. CONCLUSIONS: The recognition and measurement of all the PQRST waves can be improved and automated by simultaneous use of EsECG and SECG. The P wave amplitude is greater in EsECG than in SECG, which may faciliate the identification of supraventricular versus ventricular arrhythmias. ST segment deviation in the unipolar EsECG may not be suitable for the routine detection of ischemia.