Protocol B: All TEE studies were reviewed by the same investigators, who again were unaware of the final diagnoses. In each patient who exhibited an intra-aortic linear image, both investigators used the TEE diagnostic criteria established in protocol A to determine whether the linear image was an artifact or a true flap. The diagnostic accuracy of TEE for the identification of intra-aortic flaps was then determined and compared to that obtained in protocol A.
Protocol A: In order to determine the independent TEE parameters predictive of the presence of a linear artifact in patients exhibiting an intra-aortic linear image, a logistic regression model was developed for both the ascending and descending thoracic aorta. Initially, continuous variables were compared between patients with linear artifacts and patients with true flaps, using the nonparametric Mann-Whitney test. Qualitative variables were compared using the x2 test or the Fisher’s Exact Test, when appropriate. Parameters with a probability value < 0.25 in the univariate analysis were then included in the regression model. Odd ratios and 95% confidence intervals (CIs) were calculated for each parameter. canadian-familypharmacy.com
To confirm the origin of multiple-paths linear artifacts, the diameters of both the ascending aorta and the anatomic structure located posteriorly (ie, the left atrium or the right pulmonary artery), as well as the ratio between these two measurements were compared between patients with linear artifacts and patients with normal ascending aortas, using the Student’s t test. Results were expressed as mean ± SD, and p < 0.05 was considered statistically significant. All tests of significance were two tailed.
Finally, the sensitivity and specificity, as well as the positive and negative predictive values, were determined by comparing the initial interpretation of TEE studies with the corresponding results of the reference imaging techniques, or with anatomic findings. Ranges were obtained by considering initially inconclusive TEE studies as either positive or negative results.
Protocol B: The frequency of intra-aortic linear artifacts and true flaps in the ascending and descending segments of the thoracic aorta was compared between protocols A and B, using the Z test. The diagnostic accuracy of TEE for the identification of aortic flaps was determined using the diagnostic criteria established in protocol A. Diagnostic accuracy was obtained using either a single or a combination of TEE criteria. The combination providing the highest diagnostic accuracy was then established.