D.C., HOFSTAMER R., KERNOFF R.S,, OTIS R.S., THOMPSON A.C. Nu~l. ... RUBENSTEIN E., HOFSTADTER R., Z M H.D., THaMPSON A.C., OTIS J.N., B R m.
JOURNAL DE PHYSIQUE Colloque C9, supplkment au n012, Tome 48, decembre 1987
PROGRESS REPORT ON SYNCHROTRON RADIATION ANGIOGRAPHY ON HUMAN SUBJECTS
E. RUBENSTEIN, G.S. BROWN*, J.C. GIACOMINI, H.J. GORDON, R. HOFSTADTER** , R.S. KERNOFF, J.N. OTIS** , W, THOMLINSON** * , A.C. THOMPSON*'** and H.D. ZEMAN** Department of Medicine, Stanford University School of Medicine, Stanford, CA 94305, U.S.A. "stanford Synchrotron Radiation Laboratory, Stanford, CA 94305, U.S.A. t * Hansen Laboratories of Physics, Department of Physics, Stanford University, Stanford, CA 94305, U.S.A. **"~ational Synchrotron Light Source, Brookhaven National Laboratory, Long Island, NY 11973, U.S.A. *st* Lawrence Berkeley Laboratory, University of California, Berkeley, CA 95720, U.S.A.
Synchrotron radiation has been used as,theilluminating source for imaging the coronary artery circulation of human subjects. Iodine-containing contrast agent is introduced into the venous side of the circulation. The principle involved in the imaging procedure is that of iodine K-edge dichranography, in which two monochromatic X-ray beams, closely bracketing the K-edge of iodine (33 keV), are used to acquire a pair of line images, recorded within 0.4 msec of each other. The logarithmic subtraction of the images provides an image which enhances attenuation signals arising f r m iodine and virtually eliminates signals arising fran attenuation by body structures. The images are recorded in line scan fashion, usually at a rate of 12 an/sec. Photons are detected by a siliconlithium linear detector with sensitive elements which provide spatial resolution of 0.5 mn X 0.5 mn. Typical operating parameters have been: E=3.0 GeV, I=50-80 mA, S1.8-1.9 T. Six transvenous coronary angiograms have been done on human subjects during two sets of experiments, the first in April 1986 and the second in May 1987. These experiments confirm that the left anterior descending and right coronary arteries can be visualized by this technique. Further experience is needed to determine whether the left main coronary artery and the circumflex artery can be visualized free of overlapping contrast. Greater X-ray flux is required to achieve a signal-to-noise ratio adequate to provide clinically useful images.
Coronary atherosclerobis is the leading cause of death in Europe and the United States 111. Because of the gradual nature of the process and because myocardial perfusion remains adequate until the regional coronary artery is approximately 80 percent occluded, the disease is silent for decades before it becanes symptanatic. Unfortunately, in nearly half of its victims, the first clinical apression is sudden death. The insensitivity of other diagnostic m e t h c d s has lead physicians to cane to rely on coronary angiography as the standard means of establishing the presence and extent of the process. This procedure requires the injection of undiluted
Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1987901
JOURNAL DE PHYSIQUE
iodine-containing contrast agent directly into the coronary arteries. This is achieved by arterial invasion through a puncture into a peripheral artery and the introduction of a catheter, which is advanced through the entire length of the aorta, until its tip is engaged inside the ostium of each of the two coronary arteries. The high dollar costs (approximately $3,500 U.S.) and the risks (morbidity approximately 1:100, mortality approximately 1:500-1:1000) restrict the application of the procedure to patients in w h m there is suspicion of far-advanced disease. A less invasive method is needed.
The principal hurdle to overcane is the requirement for arterial invasion, which creates the threat of heart attack and stroke, awing to the dislodganent by the catheter of loosely adherent atheranatous plaques on the arterial wall. Since pervenous injections of contrast agent are not associated with these acmplications, a transvenous methcd of coronary angiography would be preferable. Attempts to use conventional X-ray sources for pervenous digital timesubtraction coronary angicgraphy have failed because this method involves the acquisition of two images, the first before the arrival of the contrast agent, and the second after the heart has been apacified several seconds later. Misregistration invariably occurs as a consequence of the rapid and canplex motion of the coronary arteries.
The vessels dilate and elongate with each
pulse and ride on the surface of the heart, whose intrathoracic position changes continuously with the sequential phases of the cardiac and respiratory cycles. Misregistration is eliminated by the use of synchrotron radiation as the illuminating source, since there is an insignificant time interval between the recording of the two images. Moreover, the intensity of synchrotron radiation allaws for very brief, motion-freezing exposures. Furthermore, the monochromaticity of synchrotron radiation greatly increases sensitivity to iodine; this is important in the transvenous apprcach because the contrast agent undergoes marked dilution as it passes f r m the venous to the arterial side of the circulation. The principle of the methcd is K-edge iodine dichramgraphy, in which two images are recorded with monochranatic X-ray hems, whose energies closely bracket the K-edge of iodine, 33.17 keV [ 2 ] . The subject is translated vertically through these stationary beams, and the image is recorded in line-scan fashion, using a solid-state lithium-drifted silicon detector which provides a spatial resolution suitable for the application. In the present version of the imaging system, the the exposure of each beam is 1.7 spatial resolution is 0.5 m x 0.5 mn, msec, and the interval between the two exposures is 0.4 msec [3,4]. T h e contrast agent (Rencgrafin-76)is p e r injected in bolus doses of 35-50 m l
into a central vein, and the imaging procedure is cxmnenced approximately six sec later, when the contrast agent has reached the left side of the circulation. Images of the human coronary circulation have been recorded in two sets of v i m e n t s at the Stanford Synchrotron Radiation Laboratory ( S m ) [5,6]. These have confirmed the feasibility of the technique. lPne left anterior descending coronary artery, the right coronary artery, the internal m a n n ~ ~ ~ artery and vein bypass grafts have been visualized (Fig. 1). problm remain. The first is that of overlapping iodine-containing structures. This may prove to be particularly challenging for the visualization of the circum£lex artery. A systematic evaluation of projection angles that provide clear views of this vessel and of image-enhancment methods will be undertaken in future studies. ?tuo
The second problem is an inadequate signal-to-noise ratio which degrades the image quality. This will require an increase in X-ray fluence, probably by a factor of 5-10. An w a d e d imaging systen is now being fabricated, which w ~ l l allow for the s i m u l ~ u s recording of the two monochranatic images and will increase fluence by a factor of 2.5. w a t i n g under conditions of increased beam energy, such as 3.5 GeV at SSRL, or increased wiggler field strength, such as 5T at the National Synchrotron Light Source at the Brookhaven National Laboratory, should provide for the remaining needed increase in X-ray intensity.
Figure 1 !Cransvenous coronary angiogram of a 48-year-old man who had previously undergone bypass swgery in which the left internal marmiry artery (LIM7i) was inplanted into the left anterior descending coronary artery (LAD) and vein bypass grafts (VBG) were anastanosed at several sites including the distal right coronary artery (RCA). Also identified are the aorta (A01 and the left ventricle (LV). The electron beam energy was 3.0 GeV, the beam current was 58.1 mA, the X-ray dose was 0.16 rad, and the contrast agent (Renografin-76)volme was 40 ml (0.64 m l per kg), given at a rate of 15 ml per sec. Courtesy of Proc. Natl. Acad. Sci. U.S.A., Medical Sciences.
Vol. 83, pp. 9724-9728, December 1986
JOURNAL DE PHYSIQUE
The specifications and designs of synchrotron radiation facilities dedicated to angiography have been set forth by Winick and Wiedemann [ 7,8I. Acknowledgment: 'Ihiswork was supported by National Institutes of Health Grant 1 R01 HL29024-01-A1, National Institutes of Health Contract 1 IN-38039, and Dqarbwnt of Energy Contract DE-A5003-84ER60200. The work reported herein was done at stanford Synchrotron Radiation Laboratory, which is supported by the Department of Energy, Office of Basic Energy Sciences, and the National Institutes of Health, Biotechnology Resource Program, Division of Research Resources. References GOIDBEET R.J,. GORE J.M., ALPERT J.S., DALEN J.E. (1986) 2774.
J. Am. Med. Assoc.
RUBENSTEIN E., HUGHES E.B.,
al.
CAMP= L.E., HOFR., KIRK R.L., Proc. Soc. Photo-Cpt. Instrum. Eng. 314 (1981) 42.
255
et
HUGHFS E.B., RUBENSTEIN E., Z M H.D., BRCWN G.S., BUCHBINDER M., HARRISON D.C., HOFSTAMER R., KERNOFF R.S,, OTIS R.S., THOMPSON A.C. Nu~l. Instrum. &thods Phys. Res. Sect A 246: 719 (1986) 725.
RUBENSTEIN E., HOFSTADTER R., Z M H.D., THaMPSON A.C., OTIS J.N., B R m G.S., GIA-I J.C., GORDON H.J. KD3KOOFF R.S. HARRISON D.C., THCMJNSON W. proc. N a t l . Acad. S c i . USA 83 (1986) 9724-9728. THaMPSON A., HOFS'IRUlXR R., OTIS J.N., Z M H.D., KEF@JOFF R.S., RUBENSTFZN E., GIACaMINI J.C., GORDON H.J., B R m G.S., THCMLINSON W. Nuc. Instru and mthcds (1987) in press. WINICK H.
Societa Italiana di Fisica (1987) in press.
WIH)lWWN, H.
Societa Italiana di Fisica (1987) in press.
