Jonathan L. Elion

Left heart opacification with peripheral venous injection of a new saccharide echo contrast agent in dogs

Opacification of the left heart chambers after venous injection of echo contrast agents with transpulmonary capabilities has been difficult to achieve because of a lack of availability of a biodegradable nontoxic agent that produces uniformly small microbubbles. SHU-508 is a new saccharide echo contrast agent that produces bubble sizes from 2 to 8 microns in diameter, capable of traversing the pulmonary capillary bed and resulting in left heart contrast. The echo intensity produced by this agent was compared with that of agitated saline solution, indocyanine green and SHU-454 (another experimental saccharide agent for right-sided contrast) during 136 injections in eight dogs. Videotaped two-dimensional echographic images were digitized and analyzed with the use of videodensitometry for peak right and left ventricular intensity, pulmonary transit times and time of persistence of contrast. The highest right ventricular intensity value (3,594 +/- 1,393) was achieved with SHU-508 (p less than 0.05 compared with the other agents). The right ventricular contrast seen with SHU-508 also persisted for a longer period (22.8 +/- 12 s) than with the standard agents (p less than 0.001). Left ventricular contrast with SHU-508 was visually evident in all 42 injections, whereas the peak left ventricular intensity was 35% as bright as that produced in the right ventricle by the same agent. Peak left ventricular intensity values from SHU-508 were compared with those from agitated saline solution injected from the pulmonary capillary wedge position in four dogs. SHU-508 produced brighter left ventricular intensity (1,281 +/- 607) compared with that obtained with the saline-wedge technique (p les than 0.002).(ABSTRACT TRUNCATED AT 250 WORDS)

Value and limitations of computer analysis of digital subtraction angiography in the assessment of coronary flow reserve.

Conventional coronary angiography has significant limitations in quantifying the severity and functional significance of coronary stenoses. However, coronary reactive hyperemia is an excellent physiologic indicator of coronary reserve. Digital subtraction angiography offers the potential to analyze coronary blood flow dynamics quantitatively. Therefore we assessed the accuracy of digital angiographic methods to detect and quantify reductions in coronary flow reserve secondary to stenoses of varying magnitude in an experimental canine preparation. Studies were performed in nine anesthetized open-chest dogs with an electromagnetic flow (EMF) probe and two pneumatic occluders positioned on the left circumflex coronary artery. One occluder served to induce reactive hyperemia by temporary total occlusion, while the other served to produce variable gradations of stenosis. Digital angiography was performed after the subselective injection of contrast under basal conditions and during reactive hyperemia. Time-intensity curves were obtained from digital angiograms for both a coronary and a myocardial region of interest. Measurements included area under the curve, time to peak contrast, and contrast disappearance rate. An index of coronary reserve was computed as the ratio of hyperemic to basal measurements for each of these methods. Coronary blood flow ranged from 6.5 to 142 ml/min, with hyperemic to basal EMF flow ratios of 0.80 to 4.2:1. The index derived from contrast decay rate showed a poor correlation with EMF (r = .34). The correlation between measurements of time to peak myocardial contrast and coronary blood flow was r = .68 (y = 0.16 x + 0.97). The area under the time-intensity curve from a coronary region of interest showed a close correlation with coronary blood flow (y = 0.91 x + 0.1, r = .86). Thus estimates of coronary reserve by computer analysis of digital subtraction angiograms can yield information regarding the physiologic consequences of coronary stenoses.

Correlation of continuous wave doppler velocities with cardiac catheterization gradients: An experimental model of aortic stenosis

The purpose of this study was to use a canine preparation of experimental aortic stenosis to compare estimates of pressure gradient derived from continuous wave Doppler ultrasound with gradients measured directly by catheterization. Aortic stenosis was created in six mongrel dogs by placing an elastic band around the aorta. Eighty-eight different pressure gradients, ranging from 5 to 160 mm Hg, were produced by variable tightening of the aortic band. Pressure gradients were measured by micromanometer-tipped catheters placed in the left ventricle and aorta. Doppler spectral signals were simultaneously obtained using a 2.0 MHz nonimaging transducer placed directly on the surface of the ascending aorta. Doppler and pressure recordings were analyzed using a custom-designed software program to measure maximal instantaneous, mean and peak to peak gradients, as well as ejection and acceleration times. Maximal instantaneous Doppler gradient showed an excellent linear correlation with maximal instantaneous catheterization gradient (r = 0.98, SEE = 5.3 mm Hg). The correlation of Doppler-estimated maximal gradient to peak to peak catheterization gradient was also linear (r = 0.97, SEE = 6.2 mm Hg) but resulted in a systematic overestimation of pressure drop (mean overestimation = 9.0 mm Hg). Measurement of the Doppler gradient at mid-systole resulted in a more accurate correlation with the peak to peak catheterization gradient (r = 0.98, SEE = 6.1 mm Hg) and eliminated the problem of overestimation.(ABSTRACT TRUNCATED AT 250 WORDS)

Contrast echo washout curves from the left ventricle: application of basic principles of indicator-dilution theory and calculation of ejection fraction

Time-intensity curves can be obtained from contrast echocardiography of the left ventricle. The purposes of this study were: 1) to verify whether these curves conform to the basic principles of indicator-dilution theory; and 2) to derive indexes of left ventricular ejection fraction from curve analysis. In seven closed chest dogs, 31 doses of the polysaccharide agent SHU-454 were injected into the left ventricular cavity during apical four chamber two-dimensional echocardiography. Data were obtained at different levels of ejection fraction, which were induced by changes in preload, afterload and contractility, and measured by single plane Simpsons rule analysis of digital subtraction left ventriculograms. In a subset of two dogs, eight incremental doses (from 1 to 8 ml) of SHU were injected in the basal state. Contrast echocardiograms were digitized off-line, the mean gray level/pixel of a region of interest inside the left ventricular cavity was measured, and the average value for three systolic frames of each beat was used to obtain time-intensity curves. A good correlation was observed between the peak of the time-intensity curve and the quantity of contrast injected (correlation coefficient r = 0.91 by a logarithmic fit). The echo intensities observed in each animal were subsequently transformed in quantity of contrast according to these functions and their natural logarithm was calculated both with and without background subtraction. All curves relating time and the natural logarithm of the corrected intensity exhibited a descending rectilinear portion (washout) in which the correlation was very good (r = 0.97 +/- 0.02 = mean +/- SD) and which was not significantly affected by background subtraction. The validity of this fit was also unaffected by heart rate (55 to 158 beats/min) and angiographic ejection fraction (22 to 74%), and only minimally influenced by duration of contrast washout (3.3 to 14.6 seconds). Ejection fraction was calculated by an algorithm derived from indicator-dilution theory: ejection fraction = [1 - e(-bd)] X 100, where b = slope of the curve and d = cardiac cycle duration. Linear regression analysis between values of ejection fraction derived by angiography and contrast echo yielded r = 0.73. A second index, based on b and d, was derived by multiple regression analysis. Linear regression analysis of this index and angiographic ejection fraction yielded a correlation of r = 0.87.(ABSTRACT TRUNCATED AT 400 WORDS)

Recommended guidelines for training in adult clinical cardiac electrophysiology

Abstract Training in clinical cardiac electrophysiology should take place in an Accreditation Council for Graduate Medical Education accredited cardiology program, and the electrophysiology training program itself should be accredited by the Council. Each trainee must be eligible for board certification in Internal Medicine and either eligible for certification in Cardiovascular Diseases or in a program leading to eligibility. Training faculty should be certified in clinical cardiac electrophysiology or demonstrate equivalent credentials. At least two training faculty members are preferred. The faculty must be dedicated to teaching, active in performing or promoting research and must spend a substantial portion of their time in research, teaching and practice of clinical electrophysiology. A curriculum of training should be established. Faculty experts in the related basic sciences should be available and involved in teaching. The institution should have a fully equipped clinical electrophysiology laboratory and complete noninvasive capabilities. A close working relation with a cardiac surgery faculty member skilled in surgical treatment of arrhythmias is required. Training in application of pharmacologic and all current nonpharmacologic therapies, in the outpatient and inpatient setting, is necessary. The clinical exposure must include all facets of arrhythmia diagnosis and treatment and must be quantitatively sufficient to allow the trainee to develop proficiency. The period of training should not be less than one year in addition to the period of cardiology fellowship required by the ABIM for board eligibility. A continuous period of training is preferred.

Comparison of simultaneously performed digital and film-based angiography in assessment of coronary artery disease.

This study compared digital angiography (Digital) to conventional cineangiography (Cine) for the diagnosis and quantification of coronary artery disease. Digital and Cine were obtained simultaneously under identical radiographic conditions during routine coronary arteriography. Using visual inspection and manual calipers, four independent observers identified 131 stenoses in 18 patients with multivessel coronary disease. There was no difference in interobserver variability between Digital and Cine during multiple subgroup analyses. Overall, Digital yielded significantly greater estimates of stenosis severity than did either of two separate Cine observations (p less than 0.0001; average difference, 6.25%), but the differences fell below the level of statistical significance when only the group of stenoses 50% or greater were considered. Digital and Cine correlated well for the assessment of stenosis severity (r = 0.88), but linear regression comparisons of multiple subgroups consistently indicated modest overestimation of Cine by Digital. Smaller vessels, branch vessels, and mild lesions increased the likelihood of overestimation by Digital. Digital was highly sensitive for identification of clinically relevant stenoses, but less specific and less predictive than a second observation of Cine. Our results indicate that Digital and Cine are not interchangeable imaging techniques and that potential differences must be considered when Digital is used for clinical decision making.

Determination of left ventricular ejection fraction by computer densitometric analysis of digital subtraction angiography: Experimental validation and correlation with area-length methods

Conventional methods for calculating left ventricular (LV) ejection fraction (EF) require accurate edge definition and geometric assumptions, which may be compromised in the presence of dyssynergy. Computer densitometric analysis (CDA) of digital subtraction angiography offers the potential for calculation of EF, independent of LV shape, by comparing summated brightness for regions of interest at end diastole and end systole. Therefore, the accuracy of CDA was validated for 2 mechanical heart models of differing geometry, spherical and rectangular. Both models confirmed the close correlation between calculated and measured EF (r = 0.98 and r = 0.99, respectively). Subsequently, the CDA was compared with single and biplane area-length EF calculations in 72 patients, half with a previous myocardial infarction. In patients without previous myocardial infarction, CDA correlated closely with both single-plane and biplane EF (r = 0.91 and 0.93, respectively). The close correlation was maintained regardless of whether CDA was applied to direct LV injection or intravenous digital subtraction angiography. However, in 36 patients with previous myocardial infarction, CDA correlated less closely with single-plane (r = 0.74) than with biplane (r = 0.86) area-length EF. Thus, CDA permits calculation of EF without geometric assumptions, and may be superior to the area-length method in patients with LV dyssynergy after myocardial infarction.

Determination of coronary flow reserve by digital angiography: Validation of a practical method not requiring power injection or electrocardiographic gating

Although coronary flow reserve is a well established measure of the physiologic significance of atherosclerotic stenosis, cumbersome methodology has prevented its widespread clinical application. This study evaluated a new simplified method of measuring coronary flow reserve based on indicator-dilution analysis of hand-injected digital coronary arteriograms. In five dogs, the circumflex artery was instrumented with an angiographic catheter, an electromagnetic flow probe and a pneumatic occluder. For each of 18 stenoses of varying severity, arteriograms were obtained under basal conditions and during papaverine-induced hyperemia. A pair of background-corrected arterial time-density curves was generated for each stenosis by off-line computer analysis of the circumflex artery arteriograms. Coronary flow reserve was calculated from the measured areas of the time-density curves and the known volume of contrast medium used to produce each curve. Angiographic flow reserve ranged from 0.9 to 6.1 (mean 2.99), whereas electromagnetic flow reserve ranged from 0.7 to 6.9 (mean 3.02). Angiographic and electromagnetic measurements of coronary flow reserve correlated well (r = 0.86). This study establishes that indicator-dilution analysis of 30 frames/s digital coronary arteriography permits the accurate determination of coronary flow reserve. The technique described employs hand injection of small doses of radiographic contrast medium using conventional catheters, and should be readily applicable to the study of human coronary artery disease.

A new method for structure recognition in unsubtracted digital angiograms

Existing methods for automatically finding arteries in coronary angiograms rely on preprocessing. The authors develop a structure recognition system which is not sensitive to variations in image quality. This system utilizes a probabilistic contextual segmentation technique which performs image segmentation without preprocessing. This approach, the deformable template matcher, combines prior knowledge of the arterial tree, encoded as mathematical templates, with a stochastic deformation process described by a hidden Markov model. An introduction to the technique is presented along with recent enhancements of the structure recognition system.<<ETX>>

Clinical use of lossy image compression in digital angiography

I n this issue of The American Journal of Cardiology, Rigolin et al present data to suggest that the assessment of coronary stenosis in digital angiograms is not adversely affected by the use of mathematic compression.’ This should not be taken to mean that the technique can be used clinically with no fear of losing diagnostic information, because there are more broad-reaching implications of image compression technology that must be considered. Since many digital angiographic review systems that use compression are now showing up in the marketplace, this is a good time to look at the technology, and to reach a better understanding of some of the pertinent issues. What is meant by image compression? Many computer methods have been developed that can reduce the storage requirements for computerized data. “Reversible” or “lossless” compression methods (such as those in common use to distribute computer software) can allow the original data to be completely restored by the corresponding decompression procedures. The maximum degree of reversible compression that can be achieved is typically in the range of 3:l for radiographs. Irreversible or “lossy” compression techniques recover only an approximation of the original image, but allow for much greater degrees of compression. This technique is particularly well suited to nuclear medicine2 and echocardiographic image data,3 where the artifact introduced by the compression is not as apparent as it is in higher resolution imaging like radiographs. The lossy compression techniques developed by the Joint Photographic Expert Group (JPEG)’ have become widely available, and have the advantages of being very standardized, and with specialized computer chips that have been developed to increase the processing speed. The JPEG standard also defines a form of compression that is lossless, but this has been less well known until its use in the DICOM standard for digital angiography.‘.6 Why is compression desired? A few simple calculations show the large amount of data that is generated by digital angiography. In its simplest form in cardiac imaging, each frame is represented by a matrix of 5 12 X 5 12 picture elements ( “pixels” ) . Each has a numeric value between 0 and 255, and takes up 1 “byte” of computer storage. At 30 frames per second (the standard frame rate in the United