OBJECTIVE: This study evaluated widening of the tracheal bifurcation (subcarinal and interbronchial) angle on posteroanterior chest radiographs as a sign of left atrial enlargement. The purpose was to determine the diagnostic sensitivity and specificity of the sign. MATERIALS AND METHODS: The chest radiographs and echocardiograms of 122 clinically stable patients were retrospectively reviewed. The mean interval between examinations was 10 days (range, 0-30 days). The left atrial dimension was measured on the two-dimensional echocardiograms. Sixty-five patients had a normal left atrial dimension (mean, 34 mm), and 45 patients had an enlarged atrium (mean, 47 mm) on echocardiography. Subcarinal and interbronchial angles were jointly measured by two observers who were unaware of the echocardiographic findings. Correlation analysis was used to determine which angle measurement best predicted the left atrial size. Discriminant analysis was used to derive a threshold angle that predicted left atrial enlargement. RESULTS: The carina was inadequately seen on 14 radiographs. Left atrial size correlated poorly with both the interbronchial (r = .33) and the subcarinal (r = .25) angel values. An interbronchial angle of 76.4 degrees and a subcarinal angle of 65.4 degrees were the best discriminators between patients with normal and those with enlarged left atrial dimensions (sensitivities, 61% and 51%, respectively; specificities, 63% and 66%). CONCLUSION: Our findings show that widening of the tracheal bifurcation angle on chest radiographs is an insensitive and nonspecific sign of left atrial enlargement. This sign is of little value in diagnosing left atrial enlargement.
OBJECTIVE: This study evaluated widening of the tracheal bifurcation (subcarinal and interbronchial) angle on posteroanterior chest radiographs as a sign of left atrial enlargement. The purpose was to determine the diagnostic sensitivity and specificity of the sign. MATERIALS AND METHODS: The chest radiographs and echocardiograms of 122 clinically stable patients were retrospectively reviewed. The mean interval between examinations was 10 days (range, 0-30 days). The left atrial dimension was measured on the two-dimensional echocardiograms. Sixty-five patients had a normal left atrial dimension (mean, 34 mm), and 45 patients had an enlarged atrium (mean, 47 mm) on echocardiography. Subcarinal and interbronchial angles were jointly measured by two observers who were unaware of the echocardiographic findings. Correlation analysis was used to determine which angle measurement best predicted the left atrial size. Discriminant analysis was used to derive a threshold angle that predicted left atrial enlargement. RESULTS: The carina was inadequately seen on 14 radiographs. Left atrial size correlated poorly with both the interbronchial (r = .33) and the subcarinal (r = .25) angel values. An interbronchial angle of 76.4 degrees and a subcarinal angle of 65.4 degrees were the best discriminators between patients with normal and those with enlarged left atrial dimensions (sensitivities, 61% and 51%, respectively; specificities, 63% and 66%). CONCLUSION: Our findings show that widening of the tracheal bifurcation angle on chest radiographs is an insensitive and nonspecific sign of left atrial enlargement. This sign is of little value in diagnosing left atrial enlargement.