Cardiac CT & MRI Literature

Double aortic arch and tracheal compression

This is an article by Chan MSM et al from Hong Kong, in this month's AJR, that shows how with a single dataset, we can get exquisite angiogram and tracheo-bronchial tree images of a double aortic arch patient. When the patient is under anesthesia, "inspiratory expiratory" images can be obtained to look for concomitant tracheomalacia.

We recently saw a similar patient, where however the tracheal compression was mild to moderate. We could not however perform a dynamic study for tracheomalacia, since all our parients are scanned under deep sedation only.

JAMA -- Abstract: Noninvasive Coronary Angiography With Multislice Computed Tomography, May 25, 2005, Hoffmann et al. 293 (20): 2471

This is finally the "big" one. Following in the vein of all the previous articles by authors from Germany and Netherlands, this one by Hoffman et al shows how well CT angiography on a 16-slice CT does, as compared to conventional catheter angiography. The negative and positive predictive values are above 90%. As with previous articles, almost all false positive cases are related to the presence of extensive calcification.

Link: JAMA -- Abstract: Noninvasive Coronary Angiography With Multislice Computed Tomography, May 25, 2005, Hoffmann et al. 293 (20): 2471.

Cardiac CT in assessing chronic total occlusions for stent placement

Continuing from yesterday with understanding possible indications of cardiac CT, is an article from Mollet NR et al from Erasmus University, Rotterdam, published in the January issue of the American Journal of Cardiology.

This article is an initial study on 15 patients, studying the factors that affect the outcome of stent placement in chronic total occlusions. "Multivariate analysis identified a blunt stump (by conventional angiography), occlusion length >15 mm, and severe calcification (by multislice computed tomographic coronary angiography) as independent predictors of procedural failure."

This should be one more potential indication for cardiac CT. Patients with chronic total occlusions identified on coronary angiography, prior to stent placement could get a CT done to assess the length of the stenosis and the extent of calcification.

Noninvasive evaluation of type 2 diabetes patients with MSCT

Schuijf et al from Leiden have another article similar to the one published yesterday on this blog, in Diabetes Care, describing the use of 16-slice CT in patients with type 2 diabetes. The results obtained are a shade inferior to those in the patients with hypertension.

They had a sensitivity and specificity of 95% but with 86% interpretable segments. With the inclusion of non-interpretable segments, the sensitivity and specificity drop to the low 80s.

Noninvasive evaluation of hypertensive patients with MSCT

This is an interesting article by Schuijf et al from Leiden, in Hypertension, describing a study performed on a 16-slice Toshiba, without beta-blockers, assessing the usefulness of cardiac CT in triaging patients with hypertension.

31 patients were studied and 243 coronary artery segments could be evaluated. The sensitivity and specificity was 93% and 96% respectively, with an accuracy of 90% (24/31) on a per patient basis. The evaluation of ejection fraction and wall motion abnormality as compared to 2D-echocardiography was also accurate.

The authors suggest that in hypertensives, who have some form of atherosclerotic disease, using cardiac CT as an initial modality to evaluate function and the coronary arteries may be feasible proposition.

Accuracy of 16-slice CT with a 375ms rotation tube

Two new articles on the accuracy of 16-slice CT with a 375ms rotation tube have been published in JACC in the early part of this year. The first is by Mollet Nico et al from Erasmus University, Rotterdam and the other by Kuettner Axel et al from Tubingen. At both these centres, the scanners on which these studies were performed, no longer exist. They have now been superseded by 64-slice scanners with 330ms rotation tubes and the results from both these centres on the new scanners are better than the ones published using the older scanners. These new results were presented at the recent ECR held last month.

Mollet et al's study is of 51 patients where catheter angiography was the gold standard and the negative predictive value for detection of significant lesions on a segment-based analysis was 99%. The positive predictive value was 87%.

In the second study by Kuettner et al, 72 patients were studied and their negative predictive value for the whole study group was also 97%.

This is again in line with our own experience. A study published at our centre (Bajaj et al, unpublished) on 50 consecutive patients, showed a negative predictive value of 97% as well.

16-slice CT for assessment of post-bypass graft analysis

This is one more article on the accuracy of 16-slice CT in the evaluation of venous and arterial conduit patency, by Martuscelli et al in a recent issue of Circulation. Using a 500ms GE scanner, the authors could achieve a sensitivity of 96% in diagnosing significant stenoses.

This is in line with the previous articles that we have discussed in this blog as well as our experience.

Use of beta-blockers improving image quality in 16-slice CT

This article from Korea, published in the February issue of AJR, by Shim SS, shows how the use of beta-blockers on a 16-slice CT (12 x 0.75mm) improved the image quality, especially in the RCA. The study was not performed on the same patient population, but on two separate groups.

This is something that is now an accepted fact, though there are very few papers showing a head-to-head comparison with and without beta-blockers. It would be difficult and probably unethical to perform such a study.

Having said this, on the new 64-slice CT, though we insist on heart rates below 65 for excellent image quality, if the heart rate is in the 80s, using 2-segment reconstruction, we still get excellent images.

Cardiac function analysis with an 8-slice CT as compared to SPECT and MRI

This is one  more article from Japan, by Yamamuro M, in the February issue of Radiology, on an 8-slice CT using segment reconstruction techniques, comparing cardiac CT to SPECT and CMR (the gold standard) in assessing ejection fraction and volumes.

The found excellent correlation with CMR with results better than with SPECT. The authors did not find the overestimation of systolic volumes and consequent underestimation of the ejection fraction, that has been the general problem in practice and with other authors. They ascribe this to the use of segmental reconstruction techniques, but as shown in the earlier post on the article in European Radiology, that does not seem to make much of a difference.

The use of segmental reconstruction techniques would help only with high heart rates. With low heart rates, in any case only single or at the most two segment reconstruction can be performed and most of the earlier studies have controlled heart rate using beta-blockers. This argument by the authors does not seem to make sense.

For the time being, it would be better to assume that there is some underestimation of the ejection fraction on cardiac CT as compared to CMR.

Measurement of LV volumes as compared to CMR, with and without multi-segment reconstructions

This is a technical paper from Juergens KW et al from Munster, in the January issue of European Radiology, trying to examine whether increased temporal resolution using multi-segment reconstructions from 3, 5 and 7 rotations, improves our ability to assess ejection fraction and volumes as compared to 2-segment reconstruction, using CMR as the gold standard.

They found that it makes no difference and that there is a consistent underestimation of the ejection fraction by 2-5% as compared to CMR. This has been our observation as well and we have found no difference using one-segment or two-segment reconstruction on the 64-slice scanner. We have had two patients so far who have had CMR and cardiac CT performed on the same day and we have found 5% underestimation of the ejection fraction and volumes on CT.

Having said that, both techniques as we know are far superior to echocardiography.