Contrast Echocardiography

4MEDICAL REVIEW Contrast Echocardiography

Mark A. Friedman Echocardiography Laboratory Barnes-Jewish Hospital Washington University School of Medicine St. Louis, MO 53110

ABSTRACT CONTRAST AGENTS

Ultrasound contrast agents are widely used in clinical The ability to opacify vascular structures on echocardio- practice for left ventricle opacification in sub-optimal grams by injecting agitated saline solution has been well echocardiograms. Recently, significant research has recognized for more than thirty years (Gramiak and Shah, focused on the use of contrast echocardiography as a 1968). Early contrast agents were produced by manually non-invasive means to evaluate myocardial perfusion. agitating solutions such as saline, yielding bubbles that Advances in contrast agents as well as ultrasound tech- provided brief ultrasound contrast by scattering the nology have enabled investigations into myocardial con- incoming ultrasound energy before dissolving rapidly in trast echocardiography as a possible alternative to blood. These hand-agitated intravenous solutions have nuclear imaging studies. This review will focus on the been shown to be safe and are still used clinically to pro- development and current uses of contrast echocardiog- vide brief contrast during echocardiograms for evaluating raphy, as well as future indications, including myocardial cardiac shunts and patent foramen ovale (Bommer et al., perfusion and risk stratification following myocardial 1984). Hand-agitated intravenous solutions, however, infarction. cannot be used to evaluate left-sided cardiac chambers because the bubbles have neither a small enough size nor the stability to cross the pulmonary vasculature. (Table 1) INTRODUCTION

Echocardiography has become a standard tool in the TABLE 1 PROPERTIES OF AN IDEAL ECHOCARDIOGRAPHIC CONTRAST AGENT evaluation and diagnosis of a wide range of cardiac conditions. In addition to the determination of cardiac size • High echogenicity (scattering) and function, echocardiography is widely used for the • Minimal diffusion of gas outside the microbubble assessment of left ventricular wall motion, systolic and diastolic function, valvular structure and function, as well • Small size to transverse pulmonary circulation as imaging of the great vessels. The use of agitated saline • Stability during pulmonary passage as an intravenous ultrasound contrast agent to enhance images has been well described and is currently in use for • Confinement to the intravascular compartment evaluating right to left shunts. However, the large size and short half-life of saline bubbles prevent their passage through the pulmonary circulation and thus pre- Sonication, or the exposure of a solution to ultrasound cludes their use as an intravenous agent for enhance- energy leading to the formation of stable microbubbles, ment of the left ventricle. Recent advances in microbub- was originally described by Feinstein (Feinstein et al., ble technology, along with improved ultrasound trans- 1984). The first commercially available contrast agent, ducer technology, have enabled the wide-spread use of Albunex, consisted of sonicated albumin microbubbles intravenous contrast for left ventricular opacification. containing air, and was approved by the Food and Drug However, it has not historically been possible to assess Administration (FDA) in 1992 as an intravenous injection myocardial perfusion, that is, the functional state of the for left ventricular opacification in sub-optimal echocar- coronary vasculature, using echocardiography. Expand- diograms. The air within Albunex bubbles diffused rap- ing on the technology of ventricular opacification, much idly into the circulation following peripheral injection, research has focused on the use of echocardiography causing the bubbles to become progressively smaller. with intravenous contrast to evaluate myocardial perfu- Because the ability to detect a microbubble by ultra- sion. Numerous phase III clinical trials are presently sound is a function of the sixth power of the radius of underway investigating transthoracic ultrasound as an the bubble, these first-generation microbubbles such as alternative to nuclear imaging studies currently used in Albunex began to lose their echogenicity in a relatively assessing myocardial perfusion. This review of contrast short time as air diffused out into the circulation (Cheng, echocardiography will discuss the development and cur- et al., 1998). In low output states, with increased contact rent indications for contrast echocardiography, as well as time between the microbubbles and blood, Albunex did possible future uses of the technology. not reliably lead to cavity opacification. Gandhok et

2 Einstein J. Biol. Med. (2004) 21:2-8. 4MEDICAL REVIEW Contrast Echocardiography

TABLE 2 INTRAVENOUS CONTRAST AGENTS (partial list) *approved by FDA

Name ManufacturerShell Composition Gas

Albunex * Molecular BiosystemsAlbumin Air (no longer available)

Optison * Mallinckrodt MedicalAlbumin Octafluoropropane

Definity * Dupont PharmaceuticalsLipid Octafluoropropane Levovist ScheringGalactose/Palmitate Air (approved in Europe) Imagent ScheringSurfactant/powder Perflexane SonoVue Bracco ImagingPhospholipid Sulfur-hexafluoride PB-127 Point-BiomedicalAlbumin/Polylactide Nitrogen al.,(1997) found that only 44% of selected patients with signal generates a maximal signal from the microbubble ≥1 of the following conditions (moderate-to-severe sys- (Kaul, 2001). The goal in using contrast agents is to tolic dysfunction, moderate-to-severe mitral or tricuspid increase the signal from the tissue or cavity that contains regurgitation, pulmonary hypertension, or atrial fibrilla- the contrast while simultaneously minimizing the signal tion) achieved significant left ventricular opacification from surrounding tissues. In opacifying the left ventricle, with Albunex, compared to 88% of patients without blood within the cavity contains maximum amount of these conditions. In view of the large percentage of contrast, which generates a more intense signal com- patients referred for echocardiography with these condi- pared to the myocardium which contains comparatively tions, Albunex achieved less than ideal opacification in a few microbubbles. Soft tissues scatter the incoming ultra- significant number of patients. sound wave in a linear fashion, while microbubbles have the ability to scatter a given frequency at multiples of Recently, second-generation contrast agents have been the original frequency, termed harmonics (Porter et al., developed using poorly soluble gases such as high-molec- 1996). By filtering out the fundamental frequency and ular-weight perfluorcarbons. The resulting microbubbles processing only the harmonics, the technology described have an increased longevity compared to first-genera- as harmonic imaging has enhanced the differentiation tion agents, and maintain their echogenicity for several between contrast and surrounding tissue. minutes following peripheral injection, providing sufficient time to obtain the desired opacification (Skyba and In numerous clinical trials, harmonic imaging echocardio- Kaul, 2001). Several second-generation agents have been graphy with intravenous contrast has improved the qual- granted FDA approval for ventricular opacification in ity and diagnostic yield compared to echocardiograms sub-optimal studies and are now widely used in clinical without contrast. The impaired transmission of ultra- practice for left ventricular opacification (Table 2). sound energy caused by body habitus, lung disease, hyperventilation, or tachycardia can result in non-diagnostic images in up to 30% of patients referred for ULTRASOUND TECHNOLOGY echocardiogram (Marwick et al., 1992). Contrast offers the potential to convert many of these non-diagnostic Along with advances in microbubble technology, signifi- studies into diagnostic images by improving the ability to cant progress has been made in ultrasound imaging. The visualize the endocardial border and myocardial thicken- ability of an ultrasound wave to detect microbubbles ing (Figures 1 & 2). Cohen et al. (1998) studied 203 depends on the degree of physical change that the bub- patients with initial sub-optimal non-contrast echocar- ble undergoes. At progressively higher diagnostic ultra- diograms and found that the use of Optison, a second- sound intensity, which can be monitored by the mechan- generation contrast agent consisting of an albumen shell ical index, the bubbles oscillate with increasing intensity, filled with octafluoropropane, achieved 100% left ven- eventually breaking apart and yielding a maximum sig- tricular opacification in 89 out of 102 (87%) patients. In nal. Mechanical index is directly proportional to the addition, Optison increased the amount of well-visual- power output of the ultrasound transducer and inverse- ized left ventricular endocardium by as much as 7.6 cm at ly proportional to the square root of the transducer fre- the highest dose used. This study by Cohen et al. (1998) quency. This means that a high intensity, low frequency compared Optison to Albunex, a first-generation, air-

The Einstein Journal of Biology and Medicine 3 4MEDICAL REVIEW Contrast Echocardiography

FIGURE 1 Sub-Optimal Echocardiogram (without contrast) In this apical four-chamber view, the apex of the left ventricle is not visualized adequately, which prevents an accurate assessment of wall motion or wall thickening and making the evaluation of ejection fraction impossible.

filled contrast agent and found that in all clinical end- MYOCARDIAL PERFUSION points, including the degree of left ventricular opacification, the length of visualized endocardium, and the A large area of research has focused on the potential use duration of contrast enhancement, the second genera- of transthoracic echocardiography with intravenous con- tion agent yielded significantly superior results. Specifi- trast to assess myocardial perfusion. At least in theory, cally, 63 out of 85 patients (74%) with non-diagnostic echocardiography has several potential advantages over baseline studies were converted to diagnostic studies nuclear imaging studies currently used to evaluate with Optison, compared to only 22 out of 85 patients myocardial perfusion. In addition to its wide-spread (26%) with Albunex. In a follow-up study by Shaw et al. availability and relatively low cost, echocardiography has (1998), the improvement in image quality using Optison very good spatial resolution (<1mm in the axial direc- resulted in a greater ability to answer the primary refer- tion), which is superior to that offered by single-photon ral question in as many as 50% of patients. In terms of emission computed tomography (SPECT) or positron cost-savings, Thanigaraj et al. (2001) estimate that the emission tomography (PET) (Kaul, 2001). In addition, con- regular use of contrast during stress echocardiography in trast echocardiography offers the potential to image the patients with sub-optimal images would yield a savings coronary microcirculation because, unlike other imaging of $238 per patient, primarily by reducing the number of modalities, the microbubbles reside entirely in the patients that require a follow-up nuclear stress study microvasculature without entering the extravascular because the initial non-enhanced stress echocardiograph space (Keller et al., 1989). Myocardial contrast echocar- was sub-optimal in quality. diography (MCE) describes the use of transthoracic

FIGURE 2 Sub-Optimal Echocardiogram (with contrast) This second image from the same patient using Optison, a second generation ultrasound contrast agent, to facilitate left ventricular opacification. A clear delineation between the myocardium and the ventricular cavity is observed. The apex (arrow), which was not visualized without contrast, is now clearly seen, and wall thickening can be assessed.

echocardiography with intravenous contrast to evaluate during the preceding year received both a nuclear imag- perfusion of the myocardium. The underlying concept ing study and MCE, performed with the use of a second behind MCE is that the relative concentration of generation intravenous contrast. Segmental analysis of microbubbles detected by ultrasound in different por- the ultrasound images obtained from MCE resulted in a tions of the microvasculature reflects the myocardial specificity of 93%, although with a poor sensitivity of blood flow in those regions and thus the status of the 31% at the highest contrast dose. This study, one of the coronary circulation. first direct comparisons of intravenous contrast MCE with nuclear imaging found that MCE had limited sensitivity Initial studies of MCE during experimental coronary for detecting moderate to severe perfusion defects. artery occlusion reported a clear delineation between perfused and non-perfused myocardium, suggesting that MCE may be useful for determining specific regions of FUTURE DEVELOPMENTS AND APPLICATIONS coronary stenosis (Wei et al., 1998; Cheng et al., 1998). Early clinical trials of MCE on patients with known coro- Following the study by Marwick et al. (1998), significant nary artery disease, however, yielded disappointing advances in ultrasound technology have been realized, results. Marwick et al. (1998) compared MCE to SPECT including the development of harmonic power Doppler nuclear imaging in evaluating coronary artery disease. In imaging (HPDI). In power Doppler, the imaging system cal- the study, 203 patients with a prior myocardial infarction culates the differences in amplitude between the initial

The Einstein Journal of Biology and Medicine 5 4MEDICAL REVIEW Contrast Echocardiography ultrasound pulse and the returning signal, in contrast to sue have been destroyed by the first pulse. The purpose of traditional color Doppler which analyzes only the change the destruction frame is to confirm that the mechanical in frequency of the signal (Villanueva et al., 2001). When index of the initial pulse was high enough to destroy a microbubble is exposed to ultrasound energy, the return microbubbles present in the myocardium. With a continu- signal is scattered, or non-linear. Power Doppler technolo- ous intravenous infusion of microbubbles, myocardial gy averages the scattering and filters out the linear, or sta- blood flow can be quantified by measuring the rate that tionary components of the signal which originate from new microbubbles enter the tissue between each dual- the myocardium. A second advance in ultrasound technol- triggered pulse, usually separated by 3-5 seconds. During ogy is dual-frame triggering, in which two ultrasound this pause, microbubbles which are in high concentration pulses are transmitted a fraction of a second apart. The within the left ventricular cavity and have not been first of the two pulses destroys the microbubbles present destroyed by the preceding pulses, enter the coronary in the tissue and assigns a color based on the amplitude of microcirculation. By repeating the triggered pulses at a set the returning signal. Regions with more microbubbles will interval, a pattern develops: the first frame demonstrates generate a stronger return signal because more microbub- the high-amplitude signal from microbubble destruction bles are destroyed by the initial ultrasound pulse. The sec- and the second frame appears black, confirming that ond pulse, called the destruction frame, occurs 40 to 50 microbubbles were indeed destroyed. Stenosed vessels milliseconds later and yields a significantly less intense causing a defect in perfusion in a specific region of the return signal, often appearing as black on the ultrasound myocardium will appear as a decreased signal, or more color display since most of the microbubbles within the tis- black, on the initial pulse (Figure 3).

FIGURE 3 Myocardial Contrast Echocardiography Harmonic power Doppler with trigger mode imaging. The perfusion image on the left demonstrates the presence of microbubbles in the ventricular septum. The destruction image on the right, a fraction of a second later, demonstrates that microbubbles within the septum have been destroyed, appearing black. A perfusion defect in the septal base is noted in the first image (arrow).

Numerous studies have shown harmonic power Doppler circulation vessels which are smaller than the 100um imaging to yield superior opacification of the myocardi- limit of angiography (Kaul, 1997). Sabia et al. (1992) um in detecting coronary stenosis compared to tradi- studied patients with recent myocardial infarctions and tional color Doppler ultrasound (Masugata et al., 2000). an occluded infarct-related artery by injecting microbub- Heinle et al. (2000) compared harmonic power Doppler bles during catheterization directly into the left main MCE to SPECT nuclear imaging during rest and adeno- and right coronary arteries just prior to intervention. sine stress and found an overall concordance of 81%. After successful angioplasty, microbubbles were again For specific regions of stenosis, concordance was 81% directly injected to define the vascular bed and search for the LAD, 76% for the RCA, and 72% for the LCX. The for collaterals. Nearly 80% of the patients that exhibited Heinle et al. (2000) study demonstrates the ability of adequate collateral perfusion by MCE within the infarct harmonic power Doppler MCE to detect myocardial per- bed prior to intervention had a demonstrated improve- fusion with reasonably high concordance to nuclear ment in regional systolic function one month after imaging. restoring blood flow. One significant finding in this study was that as long as collateral perfusion was pres- Utilizing these advances in ultrasound technologies, a ent within the infarct zone, function improved after large number of clinical trials have sought FDA approval angioplasty even when anterograde flow was not estab- for the use of MCE to assess myocardial perfusion. Stud- lished for days to weeks after the acute myocardial ies have attempted to determine whether the quantifi- infarction. This study demonstrates that a direct intra- cation of myocardial blood flow through MCE corre- coronary injection of microbubbles provides significant lates to the degree of coronary stenosis. Wei et al. prognostic data and may one day be used to predict via- (2003) compared MCE to SPECT nuclear imaging in a bility pre-coronary intervention. phase II FDA trial using a continuous intravenous infusion of a second-generation microbubble contrast Janardhanan et al. (2003) studied the use of MCE to dif- agent. In the trial, 54 patients with known or suspected ferentiate necrotic from viable myocardium after reper- coronary artery disease, who were referred for stress fusion therapy in acute myocardial infarction. In this nuclear studies, received both the nuclear study and study, baseline echocardiography followed by MCE was MCE with harmonic power Doppler. The study demon- performed 8 days after the infarction in 50 patients, all strated concordance between the two methods in 84% of whom had undergone coronary angiography. of the patients, with MCE having a sensitivity and speci- Echocardiography was repeated 12 weeks later to evalu- ficity of 96% and 63%. In 15 of the patients in the Wei ate recovery. The study showed that functional recovery et al. (2003) trial, cardiac catheterization was performed could be predicted based on the presence of myocardial following the imaging studies, demonstrating severe perfusion as detected by MCE. While 85% of segments stenosis in at least one coronary artery in all 15 patients. that achieved perfusion demonstrated functional In this subset of patients that eventually required car- improvement at 12 weeks, 150 segments (93%) with no diac catheterization, the prior contrast echocardiogra- demonstrated myocardial perfusion showed an absence phy study had detected a perfusion defect in 15/15 cases of functional improvement at 3 months. One possible (100%), whereas only 11/15 (73%) of the patients had explanation for the results is that the detection of lesions detected with nuclear imaging. Redefining the microbubbles at the capillary level acts as a marker of gold-standard as results on either SPECT or cardiac capillary integrity, and thus the degree of collateral cir- catheterization, the sensitivity and specificity of MCE culation (Swinburn et al., 2001). These studies demon- improve to 97% and 83%. This study by Wei et al. (2003) strate that MCE offers significant prognostic data, which demonstrates a significant improvement in the sensitiv- may be used in the future to assess viability following ity and specificity compared to that of Marwick et al. myocardial infarction. This technology may have the (1998) several years earlier. One difference between the capability to risk-stratify patients following coronary two trials is the use of harmonic power Doppler which catheterization to predict the potential for recovery of was not available at the time of Marwick’s study. Cur- function. rently, phase III trials of MCE are underway. FDA approval for the use of transthoracic echocardiography with intravenous contrast to assess myocardial perfu- CONCLUSION sion is being sought and results are expected within the next 12-18 months. Ultrasound contrast agents are widely used for ventricular opacification in patients with a technically sub-opti- Future applications of MCE may include predicting mal echocardiogram, currently the only FDA approved recovery following acute myocardial infarction. The use. Recent improvements in microbubble technology accurate visualization of collateral perfusion is one area and harmonic power Doppler imaging have given impe- where MCE may offer superior results compared to tra- tus for the use of MCE to evaluate myocardial perfusion. ditional percutaneous coronary angiography. Because Only when current phase III FDA trials are complete, of the small size and intravascular location of the which compare myocardial contrast echocardiography microbubbles, MCE allows the visualization of collateral to nuclear imaging, will there be sufficient data to pre-

The Einstein Journal of Biology and Medicine 7 4MEDICAL REVIEW Contrast Echocardiography

dict the future role of MCE in clinical cardiology. Future Marwick, T.H., Nemic, J.J., Pashkow, F.J., Stewart, W.J., Salcedo, E.E. (1992) Accu- racy and limitations of exercise echocardiography in a routine clinical setting. use of MCE includes providing prognostic information J. Am. Coll. Cardiol. 19:74-81. following coronary catheterization and is an area which is currently being actively pursued in clinical research. Marwick, T.H., Brunken, R., Meland, N., Brochet, E., Baer, F.M., Binder, T., Flach- skampf, F., Kamp, O., Nienaber, C., Nihoyannopoulos, P., Pierard, L., Vanover- schelde, J.L., Wouw, P., Lindvall, K. (1998) Accuracy and feasibility of contrast echocardiography for detection of perfusion defects in routine practice. J. Am. Coll. Cardiol. 32:1260-1269. ACKNOWLEDGEMENT Masugata, H., Cotter B., Peters, B., Ohmori, K., Mizushige, K., Demaria, A.N. (2000) Assessment of coronary stenosis severity and transmural perfusion I would like to thank Dr. Julio E. Perez for his support and gradient by myocardial contrast echocardiography: Comparison of gray-scale guidance in writing this review. B-mode with power doppler imaging. Circulation 102:1427-1433.

Porter, T.R., Xie, F., Kricsfeld, D., Armbruster, R.W. (1996) Improved myocardial contrast with second harmonic transient ultrasouns response imaging in REFERENCES humans using intravenous perfluorocarbon-exposed sonicated dextrose albumin. J. Am. Coll. Cardiol. 27:1497-1501. Bommer, W.J., Shah, P.M., Allen H., Meltzer R., Kisslo J. (1984) The safety of contrast echocardiography: Report of the committee on contrast echocardiogra- Sabia, P.J., Powers, E.R., Ragosta, M., Sarembock, I.J., Burwell, L.R., Kaul, S. phy for the American Society of Echocardiography. J. Am. Coll. Cardiol. 3:6-13. (1992) An association between collateral blood flow and myocardial viability in patients with recent myocardial infarction. N. Engl. J. Med. 372:1825-1831. Cheng, S.C., Dy, T.C., Feinstein, S.B. (1998) Contrast echocardiography: Review and future directions. Am. J. Cardiol. 81:41G-48G. Shaw, L.J., Gillam. L., Feinstein, S., Dent, J., Plotnick, G. (1998) Technology assessment in the managed care era: use of an intravenous contrast agent Cohen, J.L., Cheirif, J., Segar, D.S., Gillam, L.D., Gottdiener, J.S., Hausnerova, E., (FS069-Optison) to enhance cardiac diagnostic testing. Am. J. Manag. Care Bruns, D.E. (1998) improved left ventricular endocardial border delineation and 4:SP169-179. opacification with OPTISON (FS069), a new echocardiographic contrast agent, results of a phase III multicenter trial. J. Am. Coll. Cardiol. 32:746-752. Skyba, D.M., Kaul, S. (2001) Advances in microbubble technology. Coron. Artery Dis. 11:211-219. Feinstein, S.B., Tencate, F., Werner, Z., Ong, K., Maurer, G., Tei, C., Shah, P.M., Meerbaum, S., Corday, E. (1984) Two-dimensional contrast echocardiography. I. Swinburn, J.M., Lahiri, A., Senior, R. (2001) Intravenous myocardial contrast In vitro development and quantitative analysis of echo contrast agents. J. Am. echocardiography predicts recovery of dysnergic myocardium early after acute Coll. Cardiol. 3:14-20 myocardial infacrction. J. Am. Coll. Cardiol. 38:19-25. Gandhok, N., Block, R., Ostoic, T., Rawal, M., Hickle, P., Devries, S., Feinstein, S.B. Thaniaraj, S., Nease, J.R., Schechtman, K.B., Wade, R.L., Loslo, S., Perez, J.E. (1997) Reduced forward output states affect the left ventricular opacification of (2001) Use of contrast for image enhancement during stress echocardiography intravenously administered Albunex. J. Am. Soc. Echocardiogr. 10:25-30. Is cost-effective and reduces additional diagnostic testing. Am. J. Cardiol. 87:1430-1432. Gramiak, R., Shah P.M. (1968) Echocardiography of the aortic root. Invest. Radi- ol. 5:356-66. Villanueva, F.S., Gertz E.W., Csikari, M., Pulido, G., Fisher, D., Sklenar, J. (2001) Heinle, S.K., Noblin J., Goree-Best, P., Mello, A., Ravad, G., Mull, S., Mammen, Detection of coronary artery stenosis with power doppler imaging. Circulation P., Grayburn, P.A. (2000) Assessment of myocardial perfusion by harmonic 103:2624-2630. power doppler imaging at rest and during adenosine stress. Comparison with Tc-Sestamibi SPECT Imaging. Circulation 102:55-60. Wei, K., Jayaweear, A.R., Firoozan, S., Linka, A., Skyba, D.M., Kaul, S. (1998) Quantification of myocardial blood flow with ultrasound-induced destruction Janardhanan, R., Swinburn, J., Greaves, K., Senior, R. (2003) Usefulness of of microbubbles administered as a constant venous infusion. Circulation myocardial contrast echocardiography using low-power continuous imaging 97:473-483. early after acute myocardial infarction to predict late functional left ventricular recovery. J. Am. Coll. Cardiol. 92:493-497. Wei, K., Jayaweera, A.R., Firoozan, S., Linka, A., Skyba, D.M., Kaul, S (1998) The basis for stenosis detection using venous administration of microbubbles dur- Kaul, S. (1997) Myocardial contrast echocardiography. 15 years of research and ing myocardial contrast echocardiography. Bolus of continuous infusion? J. Am. development. Circulation 96:3745-3760. Coll. Cardiol. 32:252-260.

Kaul, S. (2001) Myoardial contrast echocardiography: Basic principles. Prog. Wei, K., Crouse. L., Weiss, J., Villanueva, F., Schiller, N.B., Naqvi, T.Z., Siegel, R., Cardiovasc. Dis. 44:1-11. Monaghan, M., Goldman, J., Aggarwal, P., Feigenbaum, H., DeMaria, A. (2003) Comparison of usefulness of dipyridamole stress myocardial contrast echocar- Keller M.W., Segal, S.S., Kaul, S., Duling, B. (1989) The behavior of sonicated diography to technetium-99m sestamibi single-photon emission computed albumin microbubbles within the microcirculation: A basis for their use during tomography for detection of coronary artery disease (PB127 multicenter phase myocardial contrast echocardiography. Circ. Res. 65:458-467. 2 trial results). Am. J. Cardiol. 91:1293-1298.

MedicalEJBM Review A Medical Review summarizes the current published literature and offers perspective based on that literature. Medical Reviews can be submitted by electronic mail ([email protected]) or regular mail (1300 Morris Park Avenue; Forchheimer Building, Room 306; Bronx, New York 10461). Receipt of a Medical Review is acknowledged. Medical Reviews are peer reviewed and edited for space and clarity.

EINSTEIN JOURNAL