Abstracts

RETURN TO ABSTRACT LISTING


1698. Diagnostic Performance of Chest Radiography Measurements Compared to Cardiac MRI in the Assessment of Cardiac Chamber Enlargement
Authors * Denotes Presenting Author
  1. Felipe Torres *; University of Toronto
  2. Diego Eifer; University of Toronto
  3. Felipe Sanchez Tijmes; University of Toronto
  4. Elsie Nguyen; University of Toronto
  5. Kate Hanneman; University of Toronto
Objective:
The cardiothoracic ratio (CTR) is commonly assessed on chest radiography (CXR) for detection of cardiac chamber enlargement. However, a traditional cut-point of 0.5 has low specificity and accuracy. The purpose of this study was to evaluate optimal measurement techniques and cut-off values for CTR in the detection of cardiac chamber enlargement on CXR in comparison to cardiac magnetic resonance imaging (MRI).

Materials and Methods:
Consecutive patients with CXR and cardiac MRI performed within a 14-day interval were included in this retrospective study (n=152, 101 posterior-anterior [PA] CXR and 51 anterior-posterior [AP] CXR; 53% men, mean age 52+/-17 years). The presence of cardiac chamber enlargement was established using cardiac MRI as the reference standard. Heart size on frontal CXR was assessed using two measurements, 1) as the sum of the maximum distance from midline to right and left heart borders and 2) maximum transverse diameter assessed as a single measurement. Thoracic size was measured between the inner margins of the ribs at three levels, 1) at the dome of the right hemi-diaphragm, 2) at the mid heart, and 3) maximum diameter at any level. Separate CTRs were calculated by dividing each heart measurement by each thoracic measurement. The diagnostic performance of CTRs assessed using different measurement techniques and cut-points were evaluated using receiver operating characteristic curve analysis. Inter-observer agreement was evaluated with intra-class correlation (ICC).

Results:
The prevalence of cardiac chamber enlargement was 61%. CTR assessed as the ratio of maximum heart diameter (single measurement)/maximum thoracic diameter (at any level) had the highest reproducibility (ICC 0.986, 95%CI[0.972, 0.992]) and area under the curve (0.788, 95%CI[0.715, 0.861]) for detection of chamber enlargement. In the subgroup of PA studies, a cut-point of 0.50 had only moderate sensitivity (72%) and specificity (72%). However, a cut-point of 0.45 had very high sensitivity (88%) with low specificity (40%) and a cut-point of 0.60 had very high specificity (95%) with low sensitivity (24%). CTR was significantly higher in the sub-group of AP studies (0.57+/-0.08 vs. 0.52+/-0.08, p=0.002). However, even in the AP sub-group, a cut-point of 0.45 had very high sensitivity (97%) and a cut-point of 0.60 had very high specificity (94%) for detection of chamber enlargement.

Conclusion:
CTR should be assessed as the ratio of maximum heart diameter to maximum thoracic diameter given that this technique has the best diagnostic performance and reproducibility. A CTR cut-point of 0.45 has very high sensitivity on both PA and AP studies, and is therefore useful in ruling out chamber enlargement when negative (CTR <0.45). A CTR cut-point of 0.60 has very high specificity on both PA and AP studies, and is therefore useful in ruling in chamber enlargement when positive (CTR>=0.60). If CTR>=0.60, further cardiac investigation should be considered, for example with echocardiography. CTR values between 0.45 and 0.60 are indeterminate and may not be helpful in ruling in or out cardiac chamber enlargement.