Canadian Health&Care Mall: Results of Coronary Angiography Utilizing Computer-Based Digital Subtraction Methods
In the first 27 patients studied, a total of 95 arteries (41 left anterior descending, 41 circumflex, and 13 right) were visualized in the identical spatial orientation using both conventional cineangiography and digital angiography. In total, 66 lesions were identified in 46 arteries by cineangiography; 49 arteries were free of lesions. The digitally subtracted angiograms identified within one grade of severity 47 of the 66 stenoses (71 percent), and identified as normal or <50 percent stenotic 38 of the 49 normal arteries (71 percent), seen on the single view cineangiogram (Fig 1); the digital angiograms were 91 percent sensitive in detecting lesions of < 50 percent severity, and were 78 percent specific in detecting arteries without obstructions.
In the next 50 patients studied, 144 major arteries were visualized in multiple views by both conventional cineangiography and digital angiography. (Damped pressure tracings during selective intubation of six right coronary arteries precluded digital angiography of these arteries.) In total, 141 lesions were identified by conventional cineangiography, and 48 arteries were considered normal. As shown in Figure 2, there was complete agreement or agreement within one grade of severity in 167 of the 189 (90 percent) of lesions or normal arteries. When conventional angiography was used as the standard for comparison, the digital method was 92 percent sensitive in detecting lesions of <50 percent severity, and was 88 percent specific in identifying arteries which were normal. Visit to Canadian Health Care – news online and will know such facts about which you haven’t never heard at all.
Forty-two arteries were visualized by collateral flow when conventional methods were employed. Digital angiography detected 40 of 42 (95 percent) of collateralized vessels that were seen on cineangiography. The two collateral vessels that were not seen in digitized images were graded poor by conventional cineangiography. Digital subtraction angiography did not permit visualization of vessels that were not seen by conventional angiography. There was agreement in the grade of collateral vessels in 32 of 42 (81 percent) collateral vessels (Fig 3).
In the last 50 patients, three digital angiograms were found to be technically inadequate for review, invariably due to pixel misregistration between mask and post-contrast image. In one case, this artifact was caused by a paroxysm of coughing; the other two occurred as a consequence of obtaining the mask during inadvertent partial, rather than full, inspiration; during subsequent imaging, more complete inspiration resulted in more caudad displacement of the diaphragm than that recorded on the mask. Although a release of a deep inspiration in three other instances resulted in partial obscuration of the image, in all instances an evaluable image was obtained.
Figure 1. Concordance in interpretation between single view selective coronary cineangiograms and single view digital subtracted angiograms.
Figure 2. Concordance in interpretation between multiple view selective coronary cineangiograms and multiple view digital subtracted angiograms.
Figure 3. Concordance in interpretation of collateral vessels between multiple view selective coronary cineangiograms and multiple view digital subtracted angiograms.