Transesophageal echocardiography is a diagnostic and monitoring modality. The objectives of our study were to compare the diagnoses obtained with continuous transesophageal echocardiography and hemodynamic monitoring in the intensive care unit, to determine interobserver variability of diagnosis obtained with both modalities, and to evaluate its impact.Design
Prospective cohort study.Setting
Surgical intensive care unit.Patients
Consecutive hemodynamically unstable patients after cardiac surgery.Interventions
At admission, unstable patients were monitored during 4 hrs with transesophageal echocardiography and standard hemodynamic monitoring. The critical care physician evaluated the patients based on all information except the transesophageal echocardiography at 0, 2, and 4 hrs and formulated a hypothesis on the most likely cause of hemodynamic instability. Transesophageal echocardiography information was provided after each evaluation. To evaluate interobserver variability, all the hemodynamic and echocardiographic information was gathered, randomized, and evaluated by five clinicians for the hemodynamic data and five echocardiographers for the transesophageal echocardiography data. The evaluators were blinded to all other information. Kappa statistics were used to evaluate agreement. Impact of transesophageal echocardiography was assessed retrospectively by using the Deutsch scale.Results
Twenty patients qualified for the study. The agreement between the hemodynamic and echocardiographic diagnosis showed a kappa at admission, 2 hrs, and 4 hrs of 0.33, 0.47, and 0.28. The interobserver agreement for the initial diagnosis (p = .014) and between all evaluators (p < .001) was significantly higher in the echocardiographic compared with the hemodynamic group. The transesophageal echocardiographic information was considered retrospectively to be essential in 34% and valuable in 34% of cases.Conclusions
These observations support the belief that transesophageal echocardiographic monitoring in the intensive care unit is associated with higher interobserver agreement in diagnosing and excluding significant causes of hemodynamic instability for postoperative cardiac surgical patients.