Transoesophageal echocardiography in cardiac and vascular surgery: implications and observer variability

Schmidlin, D. ; Bettex, D. ; Bernard, E. ; Germann, R. ; Tornic, M. ; Jenni, R. ; Schmid, E. R.

In: British Journal of Anaesthesia, 2001, vol. 86, no. 4, p. 497-505

Ajouter à la liste personnelle
    Summary
    Transoesophageal echocardiography (TOE) has gained widespread acceptance among cardiac anaesthetists as a tool to facilitate peri‐operative decision‐making. This observational study analyses the impact of TOE and its inter‐observer variability on intra‐operative patient management during cardiac and major vascular surgery. From June 1996 to December 1998, standardized reports were obtained from 11 anaesthetists in 1891 adult cardiac and vascular surgery patients undergoing routine biplane or multiplane TOE. Inter‐observer variability and the difference between variables of interest were tested using the chi‐squared test or factorial analysis of variance as appropriate. TOE examinations were performed before and after the operation; 1673 (88.5%) patients underwent cardiopulmonary bypass (CPB), and 218 (11.5%) patients had surgery without CPB, including 42 (2.2%) coronary revascularizations. In 923 patients (49%), TOE provided additional information that influenced the patient's therapy. In 968 patients (51%), TOE had only minor or no impact on clinical decision‐making. In two patients (0.10%) the scheduled operation was not performed, and in another two patients the TOE examination led to major complications. Observer‐dependent variables were: implications of TOE for intra‐operative decision‐making (P<0.0001), estimation of image quality (P<0.0001), pre‐operative left ventricular fractional area change (FAC) (P=0.0026), difference between pre‐operative FAC and post‐operative FAC (P=0.033), and requests for supervision (P<0.0001). There was no significant difference in the case mix between observers. TOE had an important impact on intra‐operative patient management. Inter‐observer variability was significant for several variables but not for the frequency of additional surgical procedures. Br J Anaesth 2001; 86: 497-505