Angioscopic Visualization of Pulmonary Emboli* Kenneth M. Moser, M.D., F.C.C.P.; Deborah Shure, M.D.; ]ames H. Harrell, M.D., F.C.C.P.; and Joseph Tulumello, M.D.
A fteDble, 8beroptic device, fitted with an inftatable distal baBoon, bas been used to directly "risulize the peat veins, right cardiac chambers, pulmonary arteries, and experimental emboU iD 'Vivo iD the dog. Insertion
A safe method for direct visualization of pulmo-
nary emboli and venous thrombi in vivo would have broad clinical and investigative applications. For example, the diagnosis of embolism still is difficult in some instances, despite the availability of such useful procedures as lung scintiphotography and pulmonary angiography. The literature discloses no previous attempts to develop a method for in vivo visualization of thromboemboli. However, there have been periodic spurts of investigative interest in the development of techniques for in vivo visualization of the heart and great vessels as an aid to the diagnosis and management of congenital and acquired cardiac and arterial diseases. Indeed, the first such effort occurred more than 60 years ago when Rhea and Walker, in 1913,1 reported attempts to use a rigid, illuminated tube, inserted into the heart at thoracotomy, to achieve direct intracardiac visualization. In 1922, Allen and Graham, 2 using a similar device with a lens system capping the distal end, reported that (in dogs) "endocardium can be seen through [the device] as clearly as we can see the mucosa of the bladder through the cystoscope if the lens of the instrument is in contact with the structure to be seen." Over the subsequent 30 years, several additional rigid "cardioscopes" were developed, all of them designed to be introduced directly into the heart at thoracotomy. A persistent problem was the inteder•From the Pulmonary Division, Deparbnent of Medicine, University of California, School of Medicine1 San Diego. Supp
198 MOSER ET AL
and manipulation of the device have been well-tolerated. The angioscope bas significant potential for investigative and cUDical application.
ence of blood with visualization. u Pedusion with saline solution to "clarify" the heart was limited by the volumes required. In 1958, Sakakibara et al6 attempted to solve this problem by placing an inflatable balloon on the distal end of the device. Touching the balloon to structures resulted in the clear images described by Allen and Graham; 2 and these workers used this device to guide aortic valve surgery in several patients. All of these instruments were introduced directly through the atria or ventricles at thoracotomy; but in 1961, Carlens and Silander7 directed a rigid, balloon-tipped device into the right atria of dogs via the jugular vein. In 1964, Silander6 extended this experience in dogs and used the device to visualize the right atrium in seven patients. The next development was the use of a flexible fiberoptic device. Greenstone et al 9 used such an instrument to visualize the major arteries in dogs, using saline solution to flush the vessels. In 1967, Gamble and Innis 10 employed a flexible fiberoptic instrument, covered at its distal end by an inflatable latex balloon, to examine the aorta, aortic valve, and left ventricle of dogs via femoral and carotid artery insertion; and the right atrium, ventricle, tricuspid and pulmonary valves, pulmonary artery, and superior and inferior vena cava by femoral and jugular vein insertions. In more than 40 dogs, they reported the procedure was well-tolerated even when the balloon was inflated in the right and left ventricular outflow tracts for several minutes. Our efforts began some three years ago. This report describes the technique and instrumentation we have devised and the results achieved, including the first report of successful in vivo visualization of experimentally induced pulmonary emboli in the dog. CHEST, 77: 2, FEBRUARY, 1980
FIGURE 3 (left). Looking down superior vena cava toward tricuspid valve through angioscope. Distal "bright spot" is reflection from balloon which is displacing blood. (right). FiGURE 4 (right). In right :ventricle, chordae tendineae are clearly visualized.
FIGURE 1. Latex balloon inflated over tips of fiberoptic device. Balloon is affixed with silk suture in this early version. Somewhat eccentric inflation is noted.
FIGURE 5 (left). View from left main pulmonary artery into lobar branches. Note whitish color of normal pulmonary arterial intima (right). FIGURE 6 (right). Swan-Ganz catheter (upper left) visualized entering main pulmonary artery.
FIGVRE 2. Metal yarn reinforcement of latex balloon; this leads to inflation in a less eccentric manner. METHODS
Our initial work was done with a modified 4 mm diameter fiberoptic bronchoscope (FOB). A balloon is fixed to the distal end of the FOB (Fig 1). Substantial effort bas been devoted to development of the balloon. The initial latex balloon was replaced by latex balloons incorporating IDetal yarn (Fig 2) because such balloons inflate more reliably as direct extensions of the FOB. Most recently, we have constructed balloons of a transparent polyurethane. The initial FOB has given way to an "angioscope." This device is 4 mm in diameter, bas a central channel 0.8 mm in diameter, and is 80 em long. It bas 70• of flexion and a 70•
CHEST, 77: 2, FEBRUARY, 1980
FIGURE 7. Pulmonary arterial branch (lobar) visualized ODe hour after embolization. There is intense hyperemia of the intima. The dark region below is the embolus previously released and identified by the angioscope in this lobar branch. angle of view. A cold light source (Olympus CLX) is used. The angioscope is introduced into the jugular vein of the dog via a longitudinal incision; a suture is placed around vein and angioscope to prevent air embolism. When visualization is desired. the balloon is inflated with carbon dioxide, air;or saline solution. Gentle touching of the structure by the balloon results in a clear image. This image can be recorded, employing appropriate optic systems, on motion picture film, videotape, or still photographs.
ANGIOSCOPIC YISUAUZAnON OF PULMONARY EMBOU 199
fu:suLTS To date, the device has been introduced into 30 dogs. It is guided with relative ease, after some experience, into right atrium, right ventricle, pulmonary arteries, inferior vena cava, and iliofemoral veins. All procedures have been done under fluoroscopic control as a "secondary" guidance system. However, as familiarity is gained with direct visualization, this becomes unnecessary. Structures visualized and photographed include the following: the superior vena cava (Fig 3), right atrium, tricuspid valve, right ventricle including papillary muscles and chordae tendineae (Fig 4), the pulmonary valve, the main pulmonary artery, two orders of branching of the right and left pulmonary arteries (Fig 5), the inferior vena cava (and entering branches such as the renal veins) and the iliac and femoral veins. Valvular visualization is difficult because of the rapid canine heart rate; it is improved when motion pictures taken are run at slow speeds. We have also visualized Swan-Ganz catheters during passage to, and after entry into, the pulmonary arteries (Fig 6). Finally, we have visualized emboli induced in the femoral veins and released to the pulmonary circulation. Initially, methylene blue was introduced into these thrombi during formation to facilitate visualization. Subsequently, nonstained emboli have been visualized without difficulty. We also have been impressed by the intense intimal hyperemia that appears proximal to emboli within minutes after they lodge (Fig 7). Thus far, no untoward hemodynamic events have occurred in the animals studied except for transient arrhythmias induced as the angioscope traverses the tricuspid valve and right ventricular outflow tract. Extended periods of partial obstruction of the tricuspid and pulmonary valves (with the angioscope lying in the pulmonary artery) have not induced arrhythmias or alterations in cardiac output The balloons and angioscope have been siliconized Significant fibrin formation has not occurred on them to date during in vivo residence exceeding one hour. DISCUSSION
While we were not aware of the fact when we commenced development of the angioscope more than three years ago, efforts had been made sporadically by various investigators to develop a device for direct visualization of cardiac and vascular structures. Most of these prior attempts were catalyzed by the desire to enhance the diagnosis of congenital or acquired heart disease, or to allow cardiac surgery 200 MOSER ET AL
under direct visualization. With the introduction of open heart surgery, the latter objective essentially disappeared and led to a long hiatus of interest in such devices that had persisted, almost uninterrupted, for 20 years. Our major objective has been direct visualization of the pulmonary arterial and systemic venous circulation, as a means for enhancing clinical and investigative progress in the field of thromboembolism. Those concerned with improving the diagnostic and management approach to venous thromboembolism will recognize the advantages that would attach to direct visualization of thromboemboli in veins and pulmonary arteries. In the experimental animal, this would allow unequivocal confirmation of the results of existing techniques ( eg, lung scan, angiogram) and new approaches to diagnosis. Similar diagnostic application could be envisioned in man. Our focus on this problem (thromboembolism) has made us less concerned with certain difficulties encountered by others: the motion inside cardiac chambers, particularly the rapid motion of valves; and potential chamber obstruction during balloon inflation during valvular approach and passage. There are, of course, numerous other potential applications which would require chamber or valve inspection. Such inspection is certainly possible with the device described, but we have regarded this as a fascinating "dividend" to our inspection of the more tranquil areas of primary interest: the pulmonary arteries and systemic veins. Passage of the device in the dog is not difficult; but, as with all procedures, becomes easier with experience. The same is true of recognizing structures during the procedure. The current device is adequate, but a number of modifications and adaptations are in progress, planned, or could be proposed. Among these are the following: further alterations in balloon materials, configuration and fixation; angioscopes of narrower diameter, with different flexion/ extension characteristics; channels within or external to the device for such purposes as infusion, blood withdrawal, passage of instruments, pressure measurements, and thrombus manipulation. Finally, as all experienced endoscopists recognize, it is difficult to reproduce faithfully what is seen by the operator to a photographic form that can be shared by others. Our figures suffer in this translation from visual image to reproduction. But, as indicated by Allen and Graham in 1922,2 using their transcardiac rigid device, the directly visualized images are. remarkably clear. It has taken more than 50 years to proceed from their description to this first direct visualization of pulmonary emboli in vivo in CHEST, 77: 2, FEBRUARY, 1980
the intact animal. Our expectation is that the next several years will witness further extensions of this technique including, after expansion of our current animal studies, application to selected patients.
1 Rhea L, Walker IC, cited by Cutler et al: Arch Surg 9:689, 1913 2 Allen DS Graham EA: Intracardiac surgery: A new method. JAMA 79:1028, 1922 3 Harker DE, Glidden EM: Experiments in intracardiac surgery: II. Intracardiac visualization: Description of instruments. J Thorac Surg 12:566, 1943 4 Butterworth RF: A new operating cardioscope. J Thorac
Surg 22:319, 1951 5 Bolton HE, Bailey CP, Costas-Durieux J, et al: Cardioscopy-simple and practical. J Thorac Surg 27:323, 1954 6 Sakakibara S, Okawa T, Hattori J, et al: Direct visual operation for aortic stenosis: Cardioscopic studies. J Int Coli Surg 29:548, 1958 7 Carlens E, Silander T: Method for direct inspection of the right atrium: Experimental investigation in the dog. Surgery 49:622, 1961 8 Silarider T: Cardioscopy without thoracotomy. Acta Chir Scand 127:67, 1964 9 Greenstone SM, Shore JM, Heringman GC, et al: Arterial endoscopy (arterioscopy). Arch Surg 93:811, 1966 10 Gamble WJ, Innis RE: Experimental intracardiac visualization. N Engl J Med 276:1397, 1967
THIRD WORLD CONGRESS ON BRONCHOESOPHAGOLOGY April 9-11 , 1980 The Breakers Palm Beach, Florida Scientific Program Chairmen: Arthur M. Olsen, M.D., FCCP David R. Sanderson, M.D., FCCP For further information, please contact Department of Education, AMERICAN COLLEGE OF CHEST PHYSICIANS, 911 Busse Highway, Park Ridge, IL 60068 (312) 698-2200
CHEST, 77: 2, FEBRUARY, 1980
ANGIOSCOPIC VISUAUZATION OF PULMONARY EMBOU 201