- Email: [email protected]
Percutaneous Transmyocardial Intracardiac Retroperfusion Shunts: Technical Feasibility in a Canine Model
Percutaneous Transmyocardial Intracardiac Retroperfusion Shunts: Technical Feasibility in a Canine Model1 Nilesh H. Patel, MD Kenneth P. Moresco, MD Gordon McLennan, MD R. Gerald Dreesen, AS, RT Index terms: Coronary vessels, diseases ● Heart, perfusion ● Myocardium, ischemia ● Shunts, cardiac JVIR 2000; 11:382–390 Abbreviations: CABG ⫽ coronary artery bypass graft, LAD ⫽ left anterior descending, PTCA ⫽ percutaneous transluminal coronary angioplasty, TICRS ⫽ transmyocardial intracardiac retroperfusion shunt, TMR ⫽ transmyocardial revascularization
PURPOSE: To test the technical feasibility of creating a left ventricle to coronary sinus shunt using endovascular techniques. MATERIALS AND METHODS: By means of a right jugular vein approach, a needle puncture was made from the coronary sinus to the left ventricle in 10 dogs. The tracts were balloon dilated and lined with 6-mm Wallstents. Shunt patencies, immediate and 4-hours later, were fluoroscopically assessed by contrast material injection into the left ventricle. Blood pressure, pulse, oxygen saturation, and cardiac rhythm were monitored. The dogs were then euthanized. Thoracic cavities and hearts were dissected and inspected. RESULTS: Technical success and immediate shunt patency were 100%. No cardiac dysrhythmias, electrocardiographic changes, or reduction in voltage potential were seen. Eight (80%) of the shunts were patent at 4 hours, one (10%) had thrombosed, and one dog died. Nine (90%) dogs had no pericardial hematoma and one (10%) had minimal pericardial blood from needle passes into the pericardial sac. The coronary sinuses were intact and no injuries to the valve leaflets or chordae tendineae were seen. The puncture sites were from the coronary sinus, 1–2 mm (mean, 1.3) from its auricular orifice, into the left ventricle, just below the inferior margin of the posterior leaflet of the mitral valve. One dog died at 3 hours with no preceding electrocardiographic evidence of impending demise. Autopsy showed no pericardial hematoma and the heart findings were no different from the other nine dogs. CONCLUSIONS: Creation of a left ventricle to coronary sinus shunt with use of endovascular techniques is technically feasible. Study of a transmyocardial intracardiac coronary retroperfusion shunt to deliver oxygenated blood to the ischemic myocardium is warranted.
1
From the Department of Radiology, Indiana University Medical Center, Indiana University School of Medicine, Indianapolis, Indiana. Received May 21, 1999; revision requested June 9; revision received and accepted September 14. From the SCVIR 1999 annual meeting. Address correspondence to N.H.P., Department of Radiology, Room 0279, Indiana University Hospital, 550 N. University Blvd., Indianapolis, IN 46202; E-mail: [email protected] © SCVIR, 2000
382
IN the United States, coronary artery disease affects 12 million people, 6.2 million of whom have angina pectoris and an additional 7 million of whom will incur myocardial infarction (1). Coronary artery disease is the single largest killer of Americans, accounting for almost a half million deaths in 1996 (1). Rapid restoration of blood flow is the mainstay of current therapy.
Effective methods of re-establishing flow to the ischemic myocardium are percutaneous endovascular interventions (ie, transluminal coronary angioplasty, laser angioplasty, atherectomy, thrombolytic therapy, stent placement) and coronary artery bypass graft surgery. In 1996, an estimated 666,000 percutaneous transluminal coronary angioplasty/ stenting (PTCA) and 598,000 coro-
Patel et al Volume 11
nary artery bypass graft (CABG) procedures were performed (1). Each year, an estimated 80,000 people are diagnosed with severe stable angina whose coronary arterial atherosclerotic disease cannot be treated by conventional revascularization methods (2). Their symptoms of angina pectoris are treated with drugs that improve the efficiency of the heart muscle, thereby reducing myocardial oxygen demand. However, they may still experience debilitating angina because the medications do not sufficiently increase the supply of oxygenated blood to the ischemic myocardium. In these more severely affected patients, perfusion of the remaining viable myocardium may not be adequate to maintain pump function. Heart transplantation may be the only available treatment option. There are currently 158 heart transplant programs in the United States, where, in 1997, 2,290 heart transplantations were performed (1). These numbers fall far short of the patients at risk for coronary artery disease. Alternative revascularization methods are needed to treat these patients, some of whom may be awaiting a heart transplant. Alternative means of delivering oxygenated blood to the ischemic myocardium in these patients are transmyocardial revascularization and coronary sinus retroperfusion (3,4). The latter entails the surgical interposition of a vein graft between the aorta and the coronary sinus, or one of its major tributaries (5,6). In 1997, Nelson et al proposed the concept of delivering oxygenated blood to the ischemic myocardium by placing a retroperfusion shunt from the left ventricle to the coronary sinus (US patent no. 5,655,548). The present study was performed to test the technical feasibility of percutaneously creating a left ventricular to coronary sinus shunt using endovascular techniques.
MATERIALS AND METHODS The animal protocol was approved by the Animal Care and Use
Committee. All aspects of handling and caring for the animals in this study including anesthesia, analgesia, and euthanasia adhered to the recommendations of the NIH “Guide for the Care and Use of Laboratory Animals,” as well as state and institutional guidelines. After development and refinement of an access set, creation of a left ventricle to coronary sinus shunt was performed in 10 adult female mongrel dogs (25–30 kg). Induction anesthesia was achieved with sodium thiopental (20 –30 mg/kg intravenous). An endotracheal tube was inserted. Isoflurane inhalation anesthesia (2.5–3%) was used for maintenance, with the dose titrated to affect. Monitoring of all vital signs was performed every 10 minutes using Laboratory Animal Resource Center forms in accordance with institutional guidelines. Oxygen saturation (pulse oximeter) and cardiac rhythm were continuously monitored. No antiarrhythmic drugs were administered. The right neck and right groin were prepared with surgical scrub and covered with a sterile drape. The right femoral artery and right jugular vein were exposed and secured with “O” silk suture. A 7-F vascular sheath (Cook, Bloomington, IN) was placed into the right femoral artery. A 6-F, angled pigtail catheter (Cordis, Warren, NJ) was introduced through the femoral arterial sheath and advanced under fluoroscopic guidance into the left ventricle. A 10-F vascular sheath (Cook) was placed into the right internal jugular vein. A 5-F catheter (Judkins JR5; Cook) was introduced through the jugular venous sheath. Under fluoroscopic guidance, the catheter was advanced into the coronary sinus. Hand-injection of contrast material was performed to confirm catheter position and to determine the exact location of the auricular orifice of the coronary sinus. The catheter was exchanged for a 16-gauge custom directional cannula (Cook). A 21-gauge, curved Chiba needle (Cook) was coaxially introduced through the directional cannula. The cannula tip was then rotated anteriorly. With use of the
●
383
Number 3
pigtail catheter as a fluoroscopic “target,” the Chiba needle was advanced from the coronary sinus into the left ventricular cavity (Fig 1). The stylet of the needle was removed and a syringe was attached to its hub. The needle was slowly withdrawn until blood was freely aspirated, and then contrast material was injected through the needle to confirm cannulation of the left ventricle. A torque controlled 0.018inch SST guide wire (Cook) was advanced through the Chiba needle into the left ventricular chamber and manipulated out the aortic outflow tract into the descending thoracic aorta. The directional cannula and Chiba needle were exchanged for a hydrophilic-coated, 4-mm balloon small vessel PTA catheter (Sub-4; Boston Scientific Vascular, Natick, MA). The transmyocardial tract was balloon dilated, followed by the deployment of a 6 ⫻ 20 mm Wallstent endoprosthesis (Boston Scientific Vascular). In six of the 10 dogs, the tract was postdilated after stent deployment with a 6-mm PTA balloon (Ultrathin; Boston Scientific Vascular). Immediate shunt patency was fluoroscopically determined by hand-injection of contrast material through the 6-F left ventricular pigtail catheter. After approximately 4 hours, shunt patency was again assessed angiographically (Fig 2). The dogs were then euthanized by intravenous injection of Fatal Plus (1 mL/10 lb, pentobarbital sodium 390 mg/mL; Vortech Pharmaceuticals, Dearborn, MI). Autopsy of each dog included systematic inspection of the mediastinum and thoracic cavities, followed by en-bloc removal of the heart, aorta, superior vena cava, and a portion of the inferior vena cava. Gross inspection of the pericardial sac and surface of the heart was followed by anatomical dissection of the right atrium, coronary sinus, great cardiac vein, and left ventricle.
RESULTS One observation readily apparent was the difficulty in precise positioning of the Wallstent in the beat-
384
●
Percutaneous TICRS
March 2000
JVIR
Figure 1. Schematic diagrams depicting the steps for creating the left ventricle to coronary sinus shunt. Step 1: Introduction of a directional cannula through a vascular sheath introduced from a right internal jugular venous approach. The cannula is rotated such that its tip is pointed anteriorly and inferiorly. Step 2: Shows the path of the coaxial Chiba puncture needle from the coronary sinus into the left ventricular chamber in the (a) frontal and (b) lateral projections. Step 3: Predilation of the transmyocardial tract with an angioplasty balloon catheter. Step 4: Shows the position and course of the Wallstent lining the shunt tract in the (a) frontal projection and (b) lateral projection. (Ao ⫽ aorta; SVC ⫽ superior vena cava; RA ⫽ right atrium; LA ⫽ left atrium; LV ⫽ left ventricle; CS ⫽ coronary sinus; LCx ⫽ left circumflex coronary artery).
ing heart. Other confounding factors were the quality of the fluoroscopy image and the use of a nonreconstrainable Wallstent delivery system. The image intensifier was an old radiography and fluoroscopy room and magnification required changing the source-to-image dis-
tance, which resulted in degradation and distortion of the fluoroscopic image. A left ventricle to coronary sinus shunt was successfully created in all 10 dogs. All dogs maintained a normal sinus rhythm during the peri- and postprocedural time pe-
riod. Occasional premature ventricular contractions were detected during manipulation of the guide wire through the aortic outflow tract. No cardiac dysrhythmias, ECG changes, or reduction of electric voltage potential were seen. The animals maintained normal oxygen
Patel et al Volume 11
●
385
Number 3
Figure 2. Contrast material injection through the left ventricular catheter at 4 hours after shunt creation in the (a) frontal and (b) lateral projections. There is flow through the Wallstent lined shunt (arrowheads) into the coronary sinus (arrow) and right atrium (asterisk).
saturation by pulse oximetry. There was no clinical or gross pathologic evidence of acute pulmonary edema after shunt creation. All shunts were patent angiographically immediately after creation. Intermittent fluoroscopy of the shunt during the observation period showed no evidence of stent collapse with the cardiac cycle, especially at peak systole. A focal waist remained in the initial four stents that were not dilated after deployment, corresponding to the location and length of the intracardiac tract. The subsequent six stents, postdilated, showed no such waist. A single acute mortality occurred. One (10%) dog died at 3 hours after shunt creation. There was no preceding electrocardiographic evidence of impending demise. Postmortem, the shunt was patent angiographically. Subsequent pathologic examination showed the stent to be in the appropriate location. The heart findings were not different from the other dogs with no pericardial hematoma. Eight (80%) of the 10 shunts were shown to be angiographically patent at 4 hours. In the single stent angiographically occluded, which was postdilated to 6 mm,
thrombus was found at autopsy occluding the intracardiac tract. On autopsy, the stent was found lying in the ventricular cavity, behind the posterior leaflet of the mitral valve. Inspection of the en-bloc specimen revealed no blood in the pericardial sac in nine of 10 dogs. In one dog, a scant amount of blood was present in the pericardium. In this dog, the first two attempted needle passes were into the pericardial sac. Autopsy revealed the integument of the pericardial sac to be intact and no gross hematoma was present in the mediastinum. Gross examination of the surface of the heart showed localized bruising of the myocardium at the site of needle passes through the posterior left ventricular wall (Fig 3). In addition, there was bruising of the wall of the coronary sinus at and near its auricular orifice, most likely from the manipulation of the guiding cannula and vascular sheath (Fig 3). The coronary sinus and great cardiac vein were grossly intact and free of thrombus. In one dog, there was bruising of the posterior vein of the left ventricle due to cannulation with the tip of the directional cannula and needle passes directly into the vein. In nine of the 10 cases, the Wallstent
Figure 3. Posterior view of the heart showing localized bruising of the left ventricular myocardium (arrow) and at the auricular orifice (arrowhead) of the coronary sinus.
Figure 4. Dissection of the right atrium and the coronary sinus shows the entrance of the Wallstent into the left ventricular transmyocardial tract. The arrows denote the auricular orifice of the coronary sinus.
remained in its postdeployment location and, on autopsy, lined the myocardial tract. The metallic free ends of the Wallstent did not protrude through the wall of the coronary sinus. The initial puncture site from the coronary sinus into the myocardium was 1–2 mm (mean, 1.3 mm) from its auricular orifice (Fig 4). The
386
●
Percutaneous TICRS
March 2000
JVIR
Figure 5. Shows the course of the Wallstent. (a) Dissection of the right atrium and the coronary sinus shows the Wallstent extending to the auricular orifice of the coronary sinus. (b) Dissection of the left ventricle and aortic outflow tract shows the Wallstent entering the ventricular chamber just below the inferior margin of the posterior leaflet of the mitral valve.
entrance into the left ventricle was just below the inferior margin of the posterior leaflet of the mitral valve. Although nine of the 10 shunts were lined with the Wallstent, there was variability as to the length of the stent extending into either the ventricular cavity or the coronary sinus (Fig 5). The metallic free ends of the stents extended onto the ventricular cavity abutting against the chordae tendineae, however, no disruption of the chordae tendineae or tear of the valve leaflet was seen. DISCUSSION In August 1998, the Food and Drug Administration approved a new surgical myocardial revascularization procedure, transmyocardial revascularization (TMR). TMR is used to treat patients with severe, stable angina (class III and intravenous) who are not responsive to medical therapy and have diffuse coronary artery disease not amenable to conventional revascularization methods (3). TMR uses an electrocardiogram-synchronized, highenergy laser to create small channels in the ischemic heart muscle. Experimental studies have shown these channels provide oxygenated blood flow to the ischemic myocardium (7). Early results of clinical
studies have shown the procedure to be safe and effective in decreasing angina pectoris by two or more classes in more than 75% of the patients (3,7). Animals and human PET studies of the heart after TMR suggest that the primary means by which TMR provides symptomatic improvement of the patients angina appears to be by sympathetic denervation of the heart rather than by an increase in myocardial perfusion (3,8). It is well known that the coronary venous system can serve as a pathway by which oxygenated blood and cardioplegic solutions may be delivered to the ischemic myocardium. In humans, the venous drainage of the heart is via the coronary sinus, anterior cardiac veins, and smallest cardiac veins (9 –12). Sixty to 70% of the venous drainage is via the coronary sinus and anterior cardiac veins into the right atrium (9 – 12). The smallest cardiac veins (thebesian veins, arterioluminal vessels, and arteriosinusoidal vessels) drain directly into the chambers of the heart (9 –12). Tributaries draining into the coronary sinus are the great cardiac vein, small cardiac vein, middle cardiac vein, posterior vein of left ventricle, and oblique vein of the left atrium (Fig 6) (13). The coronary sinus lies in the atrio-
ventricular groove and is 25 mm in length, with a diameter of 9 –10 mm (13,14). Less than 1% of people have a coronary sinus too small to permit cannulation (14). The pressure in the coronary sinus is 2–10 mm Hg and runs parallel to the right atrial pressure (4). In 1706, Raymond Vieussens ligated the inferior vena cava, superior vena cava, and pulmonary veins of human and animal hearts (15). He then injected the coronary arteries with a solution of safranine and noted that the solution ran out of the coronary sinus and small ducts (termed “ducti carnosi”) into the cavities of the heart. Adam Christian Thebesius, unaware of this prior discovery, injected the coronary sinus with water, air, and/or colored solution and found channels (termed “Thebesian veins”) in the wall of the heart directly communicating with the heart chambers (16). Further anatomical studies of the human heart by several investigators demonstrated that a rich network of venovenous and arteriovenous anastomoses exist (Fig 7) (9 –12). With its numerous interconnections (arteriovenous, venovenous, and intramuscular sinusoidal plexus), the venous system is a remarkably dense meshwork of normally unused volume capacity
Patel et al Volume 11
Figure 6. Posteriorlateral view of the heart shows the coronary venous drainage.
(11,12). Approximately two thirds of the total coronary blood volume belongs to the venous compartment. The coronary venous drainage is dependent on coronary arterial flow and resistance of other draining channels (4,6,9 –12,14). The rich network of arteriovenous and venovenous anastomoses allows retroperfusion through the coronary sinus to increase blood flow beyond a coronary occlusion (Fig 8) (4,6,9 – 12,14). Moreover, in patients with coronary arterial atherosclerotic disease, there is an increase in arteriovenous anastomoses within the myocardium (12). These interconnected channels create a redundant system that becomes effective when there is a change in venous outflow impedance. An important finding of anatomic and pathologic studies of the human heart is that the coronary venous system is not affected by atherosclerotic disease (9 –12). In 1893,
Pratt et al proposed the concept that oxygenated blood could be delivered to the ischemic myocardium via the coronary sinus (10). It was shown in experiments that an isolated dog heart could be kept beating continuously by using only the coronary veins for inflow of oxygenated blood. In 1943, Roberts et al showed that delivery of oxygenated blood from a great vessel (arterial) branch to the coronary sinus via a glass cannula conduit was possible in dogs (17). The dogs survived even after ligation of the coronary sinus at its auricular orifice. He injected dye into the coronary sinus, performed histopathologic sections of the heart, and showed complete staining of the capillaries. Beck et al performed an acute experiment in which he ligated (complete or partial) the coronary sinus followed by a surgical anastomosis between the carotid artery and the coronary sinus (18). He later ligated the left
●
387
Number 3
anterior descending (LAD) coronary artery. In the control group, undergoing LAD ligation only, there was a 70% mortality rate. This was compared to a 61% mortality rate in the group that had coronary sinus ligation prior to LAD ligation and 0% mortality in the group that had CS ligation and patent carotid arteryto-CS shunt prior to the LAD ligation (18). In the latter group, coronary sinus injection identified flow into the cardiac chambers via intramyocardial channels, as well as into in the LAD distal to the site of ligation. Both Roberts and Beck observed that, with shunting of arterial blood, the coronary sinus became arterial in color, distended, and pulsatile in nature. The Beck operation was fraught with complications, including myocardial edema and hemorrhage, intimal thickening of the veins and subsequent venous thrombosis, and thrombosis of the graft (19,20). These complications were solved by making the operation a two-staged procedure, limiting the flow in the graft and partial occlusion of the coronary sinus (20). Moreover, if the flow in the coronary retroperfusion shunt averaged approximately 50 mL/min, intimal hyperplasia of the veins was not seen (21,22). Subsequently, Beck and colleagues clinically implemented a two-stage surgical technique of creating a coronary sinus retroperfusion shunt to deliver oxygenated blood to the ischemic myocardium of patients with severe angina due to diffuse coronary artery. In stage one of the operation, a venous bypass graft was interposed between the aorta and the coronary sinus (23). This was followed 1–2 weeks later with partial occlusion of the coronary sinus at its auricular orifice. In patients who underwent the operation, 89% reported a dramatic clinical symptomatic relief in angina pectoris (complete resolution or improved) (20,23). The Beck II surgical retroperfusion procedure never gained widespread acceptance because of the technical difficulty of the operative procedure, the need for two operations (staged proce-
388
●
Percutaneous TICRS
March 2000
JVIR
Figure 7. Myocardial circulation and its interconnections.
dure), poor patient selection, and a high mortality rate (26%) (20,23). An alternative to “long-term” retroperfusion exists in the form of “short-term,” catheter-directed coronary sinus retroperfusion (4,5,14). This technique has been used to perfuse the acutely ischemic myocardium when rapid revascularization by conventional coronary revascularization techniques is not possible. It also has been effective in protecting the myocardium during PTCA or CABG in high-risk patients. Experimental and clinical studies have shown coronary sinus retroperfusion to be effective in reducing infarct size, improving myocardial contractility, and increasing the myocardial energy metabolism (4,5,14,24,25). Our rationale for choosing a 6-mm-diameter stent was the fact that the coronary sinus in dogs has a diameter of 6 mm. In our study, we created a 6-mm-diameter shunt (n ⫽ 6) and a 4-mm-diameter shunt (n ⫽ 4). We saw no thrombosis of the 4-mm shunt group and one
thrombosis in the 6-mm shunt group. However, the latter was due to a mal-deployment of the stent and recoil of the tract to 4-mm diameter, as noted on autopsy. We lined the shunt tract with Wallstent endoprosthesis for anatomical localization of the shunt, as well as to help facilitate and guide anatomic dissection. We chose the Wallstent endoprosthesis because of its flexibility and self-expanding radial force, with the fear that a balloonexpandable stent may be crushed by the myocardial contractions. However, by intermittent fluoroscopy, we did not see any collapse of the stent with the cardiac cycle during the 4-hour observation period. The puncture sites into the left ventricular myocardium from the coronary sinus were approximately 1.3 cm from its auricular orifice. Five additional dogs were utilized to development and refine the access set and technique. When the directional cannula was introduced further into the CS, all passes with the coaxial puncture needle were
into the pericardial sac. The heart is relatively fixed in the superior middle mediastinum by the great vessels and relatively mobile at its base on the diaphragmatic surface. We hypothesize that rotation of the directional cannula when its tip was beyond 2 cm into the CS resulted in rotation of the entire heart in a clockwise direction, thereby directing all needle passes into the pericardial sac. The punctures were from the CS into the left ventricle, just below the inferior margin of the posterior leaflet of the mitral valve. Because of this location, the Wallstent is not ideal because its free metallic ends may damage the chordae tendineae or mitral valve leaflet. There are a few possible solutions. One is to modify the access set such that the needle entry into the left ventricular chamber is toward the apex of the heart, away from the papillary muscle bundles and chordae tendineae. This may be technically feasible, especially in patients who have dilated left ventricles. Another solution would be to cauterize the transmyocardial intracardiac tract by thermal (laser) or electrical energy rather then to line the shunt with a stent. A third solution is to develop a stent with a buttressed end, like that of a rivet, which would sit flush with the left ventricular endocardium. The Beck II surgical retroperfusion procedure entailed reduction of the CS orifice to a luminal diameter of 2 mm (23). For the creation of a percutaneous transmyocardial intracardiac retroperfusion shunt (TICRS), we propose the use of a “T” or “Y” stent, with one limb directed into the left ventricular chamber and the other two opposing limbs located within the CS. The limb directed toward the auricular orifice of the CS will have a smaller luminal diameter, thereby effectively limiting the shunting of oxygenated blood from the left ventricle into the right atrium. Pathologic examination of the heart postmortem showed evidence of limited acute injury to the traversed myocardium with no apparent untoward effect to the normal functioning
Patel et al Volume 11
●
389
Number 3
Figure 8. Basis for coronary sinus retroperfusion. (a) Normal myocardial blood flow. (b) Myocardial ischemia due to coronary artery occlusion and (c) blood flow to the ischemic myocardium and the coronary artery distal to the occlusion with coronary sinus retroperfusion. For simplicity, only the coronary sinus limb of the retroperfusion “end-to-side” shunt is depicted (arrow).
of the heart. Importantly, there was no electrophysiologic disturbance in cardiac conduction as shown by maintenance of a normal cardiac rate and rhythm both during and after creation of the shunt. The single mortality acutely after the procedure occurred suddenly with no apparent change in cardiac function. The postmortem examination showed no significant difference in the appearance to the heart. We speculate that the death may have been related to the anesthesia. We tested the technical feasibility of creating a left ventricle to coronary sinus shunt using endovascular techniques for the future purpose of developing a retroperfusion shunt to provide oxygenated blood to the ischemic myocardium. We have shown that percutaneous cre-
ation of the shunt is technically feasible. Studies are needed to show that TICRS can deliver oxygenated blood to the ischemic myocardium distal to a coronary arterial occlusion(s). Experiments will need to be performed to study parameters, such as the flow and hemodynamic pressures, in the coronary sinus and right atrium both before and after the creation of the shunt of various diameters and with varying degrees of occlusion of the orifice of the coronary sinus. Also, the efflux of blood from the coronary arteries after TICRS and the oxygen saturation of the blood in the coronary arteries distal to an occlusion will need to be measured. The human population to benefit from TICRS are severe stable angina (class III and intravenous) patients whose
coronary artery disease is not amenable to PTCA and CABG, and in a subset of heart transplant recipients who develop atherosclerosis in the graft. For physiologic and hemodynamic testing of TICRS, a chronic myocardial ischemia pig model may best suite and reflect human cardiac physiology. We hope that TICRS, like transjugular intrahepatic portosystemic shunt, may one day be promising as a “bridge” to transplantation. References 1. American Heart Association. Heart and Stroke Statistical Update. Dallas: American Heart Association, 1999. 2. Ballard JC, Wood LL, Lansing AM. Transmyocardial revascularization:
390
●
Percutaneous TICRS
March 2000
3.
4. 5.
6. 7.
8.
9.
JVIR
criteria for selecting patients, treatment, and nursing care. Crit Care Nurs 1997; 17:42– 49. Kantor B, McKenna CJ, Caccitolo JA, et al. Transmyocardial and percutaneous myocardial revascularization: current and future role in the treatment of coronary artery disease. Mayo Clin Proc 1999; 74: 585–592. Mohl W. Coronary sinus interventions: from concept to clinics. J Card Surg 1987; 2:467– 493. Moll JW, Dziatkowiak AJ, Edelman M, Iljin W, Ratajczyk-Pakalska E, Stengert K. Arterialization of the coronary veins in diffuse coronary arteriosclerosis. J Cardiovasc Surg 1975; 16:520 –525. Hochberg MS. Selective retrograde coronary venous perfusion. Ann Thorac Surg 1980; 29:578 –588. Okada M, Nakamura M. Experimental and clinical studies on transmyocardial laser revascularization (TMLR). J Clin Laser Med Surg 1998; 16:197–201. Al-Sheikh T, Allen KB, Straka SP, et al. Cardiac sympathetic denervation after transmyocardial laser revascularization. Circulation 1999; 100:135–140. Wearn JT. The role of the thebesian vessels in the circulation of the heart. J Exp Med 1928; 47:293–316.
10. Pratt FH. The nutrition of the heart through the vessels of thebesius and the coronary veins. Am J Physiol 1893; 1:86 –103. 11. Printzmetal M, Simkin B, Bergman HC, Kruger HE. Studies on the coronary circulation II: the collateral circulation of the normal human heart by coronary perfusion with radioactive erythrocytes and glass spheres. Am Heart J 1947; 33: 420 – 442. 12. Pakalska E, Kolff WJ. Anatomical basis for retrograde coronary vein perfusion: venous anatomy and veno-venous anastomoses in the hearts of humans and some animals. Minn Med 1980; 63:795– 801. 13. Tschabitscher M. Anatomy of the coronary veins. In: Mohl W, Wolner E, Glogar D, eds. The coronary sinus: Proceedings of the 1st International Symposium of Myocardial Protection via the Coronary Sinus. New York: Springer-Verlag, 1984; 8 –10. 14. Kar S, Nordlander R. Coronary veins: an alternate route to ischemic myocardium. Heart Lung 1992; 21: 148 –157. 15. Vieussens R. Nouvelle decourvertes sur le coeur. Toulouse, 1706. 16. Thebesius AC. Dissertatio medica de circulo saguinis in corde. Lugduni Batavorum, 1708. 17. Roberts JT, Browne RS, Roberts G.
18.
19. 20.
21.
22.
23. 24. 25.
Nourishment of the myocardium by way of the coronary veins. Fed Proceed 1943; 2:90. Beck CS, Stanton E, Batiuchok W, Leiter E. Revascularization of heart by graft of systemic artery into coronary sinus. JAMA 1948; 137:436 – 442. Beck CS. Revascularization of the heart. Ann Surg 1948; 128:854 – 864. Beck CS, Hahn RS, Leighninger DS, McAllister FF. Operation for coronary artery disease. JAMA 1951; 147:1726 –1731. Hochberg MS, Roberts WC, Morrow AG, Austen WG. Selective arterialization of the coronary venous system: encouraging long-term flow evaluation utilizing radioactive microspheres. J Thorac Cardiovasc Surg 1979; 77:1–12. Beck CS, Leighninger DS. Scientific basis for the surgical treatment of coronary artery disease. JAMA 1955; 159:1264 –1271. Beck CS, Leighninger DS. Operation for coronary artery disease. JAMA 1954; 156:1226 –1233. Lazar H. Coronary sinus interventions during cardiac surgery. Ann Thorac Surg 1988; 46:475– 482. Mohl W. The relevance of coronary sinus interventions in cardiac surgery. Thorac Cardiovasc Surg 1991; 39:245–250.
PURPOSE: To test the technical feasibility of creating a left ventricle to coronary sinus shunt using endovascular techniques. MATERIALS AND METHODS: By means of a right jugular vein approach, a needle puncture was made from the coronary sinus to the left ventricle in 10 dogs. The tracts were balloon dilated and lined with 6-mm Wallstents. Shunt patencies, immediate and 4-hours later, were fluoroscopically assessed by contrast material injection into the left ventricle. Blood pressure, pulse, oxygen saturation, and cardiac rhythm were monitored. The dogs were then euthanized. Thoracic cavities and hearts were dissected and inspected. RESULTS: Technical success and immediate shunt patency were 100%. No cardiac dysrhythmias, electrocardiographic changes, or reduction in voltage potential were seen. Eight (80%) of the shunts were patent at 4 hours, one (10%) had thrombosed, and one dog died. Nine (90%) dogs had no pericardial hematoma and one (10%) had minimal pericardial blood from needle passes into the pericardial sac. The coronary sinuses were intact and no injuries to the valve leaflets or chordae tendineae were seen. The puncture sites were from the coronary sinus, 1–2 mm (mean, 1.3) from its auricular orifice, into the left ventricle, just below the inferior margin of the posterior leaflet of the mitral valve. One dog died at 3 hours with no preceding electrocardiographic evidence of impending demise. Autopsy showed no pericardial hematoma and the heart findings were no different from the other nine dogs. CONCLUSIONS: Creation of a left ventricle to coronary sinus shunt with use of endovascular techniques is technically feasible. Study of a transmyocardial intracardiac coronary retroperfusion shunt to deliver oxygenated blood to the ischemic myocardium is warranted.
1
From the Department of Radiology, Indiana University Medical Center, Indiana University School of Medicine, Indianapolis, Indiana. Received May 21, 1999; revision requested June 9; revision received and accepted September 14. From the SCVIR 1999 annual meeting. Address correspondence to N.H.P., Department of Radiology, Room 0279, Indiana University Hospital, 550 N. University Blvd., Indianapolis, IN 46202; E-mail: [email protected] © SCVIR, 2000
382
IN the United States, coronary artery disease affects 12 million people, 6.2 million of whom have angina pectoris and an additional 7 million of whom will incur myocardial infarction (1). Coronary artery disease is the single largest killer of Americans, accounting for almost a half million deaths in 1996 (1). Rapid restoration of blood flow is the mainstay of current therapy.
Effective methods of re-establishing flow to the ischemic myocardium are percutaneous endovascular interventions (ie, transluminal coronary angioplasty, laser angioplasty, atherectomy, thrombolytic therapy, stent placement) and coronary artery bypass graft surgery. In 1996, an estimated 666,000 percutaneous transluminal coronary angioplasty/ stenting (PTCA) and 598,000 coro-
Patel et al Volume 11
nary artery bypass graft (CABG) procedures were performed (1). Each year, an estimated 80,000 people are diagnosed with severe stable angina whose coronary arterial atherosclerotic disease cannot be treated by conventional revascularization methods (2). Their symptoms of angina pectoris are treated with drugs that improve the efficiency of the heart muscle, thereby reducing myocardial oxygen demand. However, they may still experience debilitating angina because the medications do not sufficiently increase the supply of oxygenated blood to the ischemic myocardium. In these more severely affected patients, perfusion of the remaining viable myocardium may not be adequate to maintain pump function. Heart transplantation may be the only available treatment option. There are currently 158 heart transplant programs in the United States, where, in 1997, 2,290 heart transplantations were performed (1). These numbers fall far short of the patients at risk for coronary artery disease. Alternative revascularization methods are needed to treat these patients, some of whom may be awaiting a heart transplant. Alternative means of delivering oxygenated blood to the ischemic myocardium in these patients are transmyocardial revascularization and coronary sinus retroperfusion (3,4). The latter entails the surgical interposition of a vein graft between the aorta and the coronary sinus, or one of its major tributaries (5,6). In 1997, Nelson et al proposed the concept of delivering oxygenated blood to the ischemic myocardium by placing a retroperfusion shunt from the left ventricle to the coronary sinus (US patent no. 5,655,548). The present study was performed to test the technical feasibility of percutaneously creating a left ventricular to coronary sinus shunt using endovascular techniques.
MATERIALS AND METHODS The animal protocol was approved by the Animal Care and Use
Committee. All aspects of handling and caring for the animals in this study including anesthesia, analgesia, and euthanasia adhered to the recommendations of the NIH “Guide for the Care and Use of Laboratory Animals,” as well as state and institutional guidelines. After development and refinement of an access set, creation of a left ventricle to coronary sinus shunt was performed in 10 adult female mongrel dogs (25–30 kg). Induction anesthesia was achieved with sodium thiopental (20 –30 mg/kg intravenous). An endotracheal tube was inserted. Isoflurane inhalation anesthesia (2.5–3%) was used for maintenance, with the dose titrated to affect. Monitoring of all vital signs was performed every 10 minutes using Laboratory Animal Resource Center forms in accordance with institutional guidelines. Oxygen saturation (pulse oximeter) and cardiac rhythm were continuously monitored. No antiarrhythmic drugs were administered. The right neck and right groin were prepared with surgical scrub and covered with a sterile drape. The right femoral artery and right jugular vein were exposed and secured with “O” silk suture. A 7-F vascular sheath (Cook, Bloomington, IN) was placed into the right femoral artery. A 6-F, angled pigtail catheter (Cordis, Warren, NJ) was introduced through the femoral arterial sheath and advanced under fluoroscopic guidance into the left ventricle. A 10-F vascular sheath (Cook) was placed into the right internal jugular vein. A 5-F catheter (Judkins JR5; Cook) was introduced through the jugular venous sheath. Under fluoroscopic guidance, the catheter was advanced into the coronary sinus. Hand-injection of contrast material was performed to confirm catheter position and to determine the exact location of the auricular orifice of the coronary sinus. The catheter was exchanged for a 16-gauge custom directional cannula (Cook). A 21-gauge, curved Chiba needle (Cook) was coaxially introduced through the directional cannula. The cannula tip was then rotated anteriorly. With use of the
●
383
Number 3
pigtail catheter as a fluoroscopic “target,” the Chiba needle was advanced from the coronary sinus into the left ventricular cavity (Fig 1). The stylet of the needle was removed and a syringe was attached to its hub. The needle was slowly withdrawn until blood was freely aspirated, and then contrast material was injected through the needle to confirm cannulation of the left ventricle. A torque controlled 0.018inch SST guide wire (Cook) was advanced through the Chiba needle into the left ventricular chamber and manipulated out the aortic outflow tract into the descending thoracic aorta. The directional cannula and Chiba needle were exchanged for a hydrophilic-coated, 4-mm balloon small vessel PTA catheter (Sub-4; Boston Scientific Vascular, Natick, MA). The transmyocardial tract was balloon dilated, followed by the deployment of a 6 ⫻ 20 mm Wallstent endoprosthesis (Boston Scientific Vascular). In six of the 10 dogs, the tract was postdilated after stent deployment with a 6-mm PTA balloon (Ultrathin; Boston Scientific Vascular). Immediate shunt patency was fluoroscopically determined by hand-injection of contrast material through the 6-F left ventricular pigtail catheter. After approximately 4 hours, shunt patency was again assessed angiographically (Fig 2). The dogs were then euthanized by intravenous injection of Fatal Plus (1 mL/10 lb, pentobarbital sodium 390 mg/mL; Vortech Pharmaceuticals, Dearborn, MI). Autopsy of each dog included systematic inspection of the mediastinum and thoracic cavities, followed by en-bloc removal of the heart, aorta, superior vena cava, and a portion of the inferior vena cava. Gross inspection of the pericardial sac and surface of the heart was followed by anatomical dissection of the right atrium, coronary sinus, great cardiac vein, and left ventricle.
RESULTS One observation readily apparent was the difficulty in precise positioning of the Wallstent in the beat-
384
●
Percutaneous TICRS
March 2000
JVIR
Figure 1. Schematic diagrams depicting the steps for creating the left ventricle to coronary sinus shunt. Step 1: Introduction of a directional cannula through a vascular sheath introduced from a right internal jugular venous approach. The cannula is rotated such that its tip is pointed anteriorly and inferiorly. Step 2: Shows the path of the coaxial Chiba puncture needle from the coronary sinus into the left ventricular chamber in the (a) frontal and (b) lateral projections. Step 3: Predilation of the transmyocardial tract with an angioplasty balloon catheter. Step 4: Shows the position and course of the Wallstent lining the shunt tract in the (a) frontal projection and (b) lateral projection. (Ao ⫽ aorta; SVC ⫽ superior vena cava; RA ⫽ right atrium; LA ⫽ left atrium; LV ⫽ left ventricle; CS ⫽ coronary sinus; LCx ⫽ left circumflex coronary artery).
ing heart. Other confounding factors were the quality of the fluoroscopy image and the use of a nonreconstrainable Wallstent delivery system. The image intensifier was an old radiography and fluoroscopy room and magnification required changing the source-to-image dis-
tance, which resulted in degradation and distortion of the fluoroscopic image. A left ventricle to coronary sinus shunt was successfully created in all 10 dogs. All dogs maintained a normal sinus rhythm during the peri- and postprocedural time pe-
riod. Occasional premature ventricular contractions were detected during manipulation of the guide wire through the aortic outflow tract. No cardiac dysrhythmias, ECG changes, or reduction of electric voltage potential were seen. The animals maintained normal oxygen
Patel et al Volume 11
●
385
Number 3
Figure 2. Contrast material injection through the left ventricular catheter at 4 hours after shunt creation in the (a) frontal and (b) lateral projections. There is flow through the Wallstent lined shunt (arrowheads) into the coronary sinus (arrow) and right atrium (asterisk).
saturation by pulse oximetry. There was no clinical or gross pathologic evidence of acute pulmonary edema after shunt creation. All shunts were patent angiographically immediately after creation. Intermittent fluoroscopy of the shunt during the observation period showed no evidence of stent collapse with the cardiac cycle, especially at peak systole. A focal waist remained in the initial four stents that were not dilated after deployment, corresponding to the location and length of the intracardiac tract. The subsequent six stents, postdilated, showed no such waist. A single acute mortality occurred. One (10%) dog died at 3 hours after shunt creation. There was no preceding electrocardiographic evidence of impending demise. Postmortem, the shunt was patent angiographically. Subsequent pathologic examination showed the stent to be in the appropriate location. The heart findings were not different from the other dogs with no pericardial hematoma. Eight (80%) of the 10 shunts were shown to be angiographically patent at 4 hours. In the single stent angiographically occluded, which was postdilated to 6 mm,
thrombus was found at autopsy occluding the intracardiac tract. On autopsy, the stent was found lying in the ventricular cavity, behind the posterior leaflet of the mitral valve. Inspection of the en-bloc specimen revealed no blood in the pericardial sac in nine of 10 dogs. In one dog, a scant amount of blood was present in the pericardium. In this dog, the first two attempted needle passes were into the pericardial sac. Autopsy revealed the integument of the pericardial sac to be intact and no gross hematoma was present in the mediastinum. Gross examination of the surface of the heart showed localized bruising of the myocardium at the site of needle passes through the posterior left ventricular wall (Fig 3). In addition, there was bruising of the wall of the coronary sinus at and near its auricular orifice, most likely from the manipulation of the guiding cannula and vascular sheath (Fig 3). The coronary sinus and great cardiac vein were grossly intact and free of thrombus. In one dog, there was bruising of the posterior vein of the left ventricle due to cannulation with the tip of the directional cannula and needle passes directly into the vein. In nine of the 10 cases, the Wallstent
Figure 3. Posterior view of the heart showing localized bruising of the left ventricular myocardium (arrow) and at the auricular orifice (arrowhead) of the coronary sinus.
Figure 4. Dissection of the right atrium and the coronary sinus shows the entrance of the Wallstent into the left ventricular transmyocardial tract. The arrows denote the auricular orifice of the coronary sinus.
remained in its postdeployment location and, on autopsy, lined the myocardial tract. The metallic free ends of the Wallstent did not protrude through the wall of the coronary sinus. The initial puncture site from the coronary sinus into the myocardium was 1–2 mm (mean, 1.3 mm) from its auricular orifice (Fig 4). The
386
●
Percutaneous TICRS
March 2000
JVIR
Figure 5. Shows the course of the Wallstent. (a) Dissection of the right atrium and the coronary sinus shows the Wallstent extending to the auricular orifice of the coronary sinus. (b) Dissection of the left ventricle and aortic outflow tract shows the Wallstent entering the ventricular chamber just below the inferior margin of the posterior leaflet of the mitral valve.
entrance into the left ventricle was just below the inferior margin of the posterior leaflet of the mitral valve. Although nine of the 10 shunts were lined with the Wallstent, there was variability as to the length of the stent extending into either the ventricular cavity or the coronary sinus (Fig 5). The metallic free ends of the stents extended onto the ventricular cavity abutting against the chordae tendineae, however, no disruption of the chordae tendineae or tear of the valve leaflet was seen. DISCUSSION In August 1998, the Food and Drug Administration approved a new surgical myocardial revascularization procedure, transmyocardial revascularization (TMR). TMR is used to treat patients with severe, stable angina (class III and intravenous) who are not responsive to medical therapy and have diffuse coronary artery disease not amenable to conventional revascularization methods (3). TMR uses an electrocardiogram-synchronized, highenergy laser to create small channels in the ischemic heart muscle. Experimental studies have shown these channels provide oxygenated blood flow to the ischemic myocardium (7). Early results of clinical
studies have shown the procedure to be safe and effective in decreasing angina pectoris by two or more classes in more than 75% of the patients (3,7). Animals and human PET studies of the heart after TMR suggest that the primary means by which TMR provides symptomatic improvement of the patients angina appears to be by sympathetic denervation of the heart rather than by an increase in myocardial perfusion (3,8). It is well known that the coronary venous system can serve as a pathway by which oxygenated blood and cardioplegic solutions may be delivered to the ischemic myocardium. In humans, the venous drainage of the heart is via the coronary sinus, anterior cardiac veins, and smallest cardiac veins (9 –12). Sixty to 70% of the venous drainage is via the coronary sinus and anterior cardiac veins into the right atrium (9 – 12). The smallest cardiac veins (thebesian veins, arterioluminal vessels, and arteriosinusoidal vessels) drain directly into the chambers of the heart (9 –12). Tributaries draining into the coronary sinus are the great cardiac vein, small cardiac vein, middle cardiac vein, posterior vein of left ventricle, and oblique vein of the left atrium (Fig 6) (13). The coronary sinus lies in the atrio-
ventricular groove and is 25 mm in length, with a diameter of 9 –10 mm (13,14). Less than 1% of people have a coronary sinus too small to permit cannulation (14). The pressure in the coronary sinus is 2–10 mm Hg and runs parallel to the right atrial pressure (4). In 1706, Raymond Vieussens ligated the inferior vena cava, superior vena cava, and pulmonary veins of human and animal hearts (15). He then injected the coronary arteries with a solution of safranine and noted that the solution ran out of the coronary sinus and small ducts (termed “ducti carnosi”) into the cavities of the heart. Adam Christian Thebesius, unaware of this prior discovery, injected the coronary sinus with water, air, and/or colored solution and found channels (termed “Thebesian veins”) in the wall of the heart directly communicating with the heart chambers (16). Further anatomical studies of the human heart by several investigators demonstrated that a rich network of venovenous and arteriovenous anastomoses exist (Fig 7) (9 –12). With its numerous interconnections (arteriovenous, venovenous, and intramuscular sinusoidal plexus), the venous system is a remarkably dense meshwork of normally unused volume capacity
Patel et al Volume 11
Figure 6. Posteriorlateral view of the heart shows the coronary venous drainage.
(11,12). Approximately two thirds of the total coronary blood volume belongs to the venous compartment. The coronary venous drainage is dependent on coronary arterial flow and resistance of other draining channels (4,6,9 –12,14). The rich network of arteriovenous and venovenous anastomoses allows retroperfusion through the coronary sinus to increase blood flow beyond a coronary occlusion (Fig 8) (4,6,9 – 12,14). Moreover, in patients with coronary arterial atherosclerotic disease, there is an increase in arteriovenous anastomoses within the myocardium (12). These interconnected channels create a redundant system that becomes effective when there is a change in venous outflow impedance. An important finding of anatomic and pathologic studies of the human heart is that the coronary venous system is not affected by atherosclerotic disease (9 –12). In 1893,
Pratt et al proposed the concept that oxygenated blood could be delivered to the ischemic myocardium via the coronary sinus (10). It was shown in experiments that an isolated dog heart could be kept beating continuously by using only the coronary veins for inflow of oxygenated blood. In 1943, Roberts et al showed that delivery of oxygenated blood from a great vessel (arterial) branch to the coronary sinus via a glass cannula conduit was possible in dogs (17). The dogs survived even after ligation of the coronary sinus at its auricular orifice. He injected dye into the coronary sinus, performed histopathologic sections of the heart, and showed complete staining of the capillaries. Beck et al performed an acute experiment in which he ligated (complete or partial) the coronary sinus followed by a surgical anastomosis between the carotid artery and the coronary sinus (18). He later ligated the left
●
387
Number 3
anterior descending (LAD) coronary artery. In the control group, undergoing LAD ligation only, there was a 70% mortality rate. This was compared to a 61% mortality rate in the group that had coronary sinus ligation prior to LAD ligation and 0% mortality in the group that had CS ligation and patent carotid arteryto-CS shunt prior to the LAD ligation (18). In the latter group, coronary sinus injection identified flow into the cardiac chambers via intramyocardial channels, as well as into in the LAD distal to the site of ligation. Both Roberts and Beck observed that, with shunting of arterial blood, the coronary sinus became arterial in color, distended, and pulsatile in nature. The Beck operation was fraught with complications, including myocardial edema and hemorrhage, intimal thickening of the veins and subsequent venous thrombosis, and thrombosis of the graft (19,20). These complications were solved by making the operation a two-staged procedure, limiting the flow in the graft and partial occlusion of the coronary sinus (20). Moreover, if the flow in the coronary retroperfusion shunt averaged approximately 50 mL/min, intimal hyperplasia of the veins was not seen (21,22). Subsequently, Beck and colleagues clinically implemented a two-stage surgical technique of creating a coronary sinus retroperfusion shunt to deliver oxygenated blood to the ischemic myocardium of patients with severe angina due to diffuse coronary artery. In stage one of the operation, a venous bypass graft was interposed between the aorta and the coronary sinus (23). This was followed 1–2 weeks later with partial occlusion of the coronary sinus at its auricular orifice. In patients who underwent the operation, 89% reported a dramatic clinical symptomatic relief in angina pectoris (complete resolution or improved) (20,23). The Beck II surgical retroperfusion procedure never gained widespread acceptance because of the technical difficulty of the operative procedure, the need for two operations (staged proce-
388
●
Percutaneous TICRS
March 2000
JVIR
Figure 7. Myocardial circulation and its interconnections.
dure), poor patient selection, and a high mortality rate (26%) (20,23). An alternative to “long-term” retroperfusion exists in the form of “short-term,” catheter-directed coronary sinus retroperfusion (4,5,14). This technique has been used to perfuse the acutely ischemic myocardium when rapid revascularization by conventional coronary revascularization techniques is not possible. It also has been effective in protecting the myocardium during PTCA or CABG in high-risk patients. Experimental and clinical studies have shown coronary sinus retroperfusion to be effective in reducing infarct size, improving myocardial contractility, and increasing the myocardial energy metabolism (4,5,14,24,25). Our rationale for choosing a 6-mm-diameter stent was the fact that the coronary sinus in dogs has a diameter of 6 mm. In our study, we created a 6-mm-diameter shunt (n ⫽ 6) and a 4-mm-diameter shunt (n ⫽ 4). We saw no thrombosis of the 4-mm shunt group and one
thrombosis in the 6-mm shunt group. However, the latter was due to a mal-deployment of the stent and recoil of the tract to 4-mm diameter, as noted on autopsy. We lined the shunt tract with Wallstent endoprosthesis for anatomical localization of the shunt, as well as to help facilitate and guide anatomic dissection. We chose the Wallstent endoprosthesis because of its flexibility and self-expanding radial force, with the fear that a balloonexpandable stent may be crushed by the myocardial contractions. However, by intermittent fluoroscopy, we did not see any collapse of the stent with the cardiac cycle during the 4-hour observation period. The puncture sites into the left ventricular myocardium from the coronary sinus were approximately 1.3 cm from its auricular orifice. Five additional dogs were utilized to development and refine the access set and technique. When the directional cannula was introduced further into the CS, all passes with the coaxial puncture needle were
into the pericardial sac. The heart is relatively fixed in the superior middle mediastinum by the great vessels and relatively mobile at its base on the diaphragmatic surface. We hypothesize that rotation of the directional cannula when its tip was beyond 2 cm into the CS resulted in rotation of the entire heart in a clockwise direction, thereby directing all needle passes into the pericardial sac. The punctures were from the CS into the left ventricle, just below the inferior margin of the posterior leaflet of the mitral valve. Because of this location, the Wallstent is not ideal because its free metallic ends may damage the chordae tendineae or mitral valve leaflet. There are a few possible solutions. One is to modify the access set such that the needle entry into the left ventricular chamber is toward the apex of the heart, away from the papillary muscle bundles and chordae tendineae. This may be technically feasible, especially in patients who have dilated left ventricles. Another solution would be to cauterize the transmyocardial intracardiac tract by thermal (laser) or electrical energy rather then to line the shunt with a stent. A third solution is to develop a stent with a buttressed end, like that of a rivet, which would sit flush with the left ventricular endocardium. The Beck II surgical retroperfusion procedure entailed reduction of the CS orifice to a luminal diameter of 2 mm (23). For the creation of a percutaneous transmyocardial intracardiac retroperfusion shunt (TICRS), we propose the use of a “T” or “Y” stent, with one limb directed into the left ventricular chamber and the other two opposing limbs located within the CS. The limb directed toward the auricular orifice of the CS will have a smaller luminal diameter, thereby effectively limiting the shunting of oxygenated blood from the left ventricle into the right atrium. Pathologic examination of the heart postmortem showed evidence of limited acute injury to the traversed myocardium with no apparent untoward effect to the normal functioning
Patel et al Volume 11
●
389
Number 3
Figure 8. Basis for coronary sinus retroperfusion. (a) Normal myocardial blood flow. (b) Myocardial ischemia due to coronary artery occlusion and (c) blood flow to the ischemic myocardium and the coronary artery distal to the occlusion with coronary sinus retroperfusion. For simplicity, only the coronary sinus limb of the retroperfusion “end-to-side” shunt is depicted (arrow).
of the heart. Importantly, there was no electrophysiologic disturbance in cardiac conduction as shown by maintenance of a normal cardiac rate and rhythm both during and after creation of the shunt. The single mortality acutely after the procedure occurred suddenly with no apparent change in cardiac function. The postmortem examination showed no significant difference in the appearance to the heart. We speculate that the death may have been related to the anesthesia. We tested the technical feasibility of creating a left ventricle to coronary sinus shunt using endovascular techniques for the future purpose of developing a retroperfusion shunt to provide oxygenated blood to the ischemic myocardium. We have shown that percutaneous cre-
ation of the shunt is technically feasible. Studies are needed to show that TICRS can deliver oxygenated blood to the ischemic myocardium distal to a coronary arterial occlusion(s). Experiments will need to be performed to study parameters, such as the flow and hemodynamic pressures, in the coronary sinus and right atrium both before and after the creation of the shunt of various diameters and with varying degrees of occlusion of the orifice of the coronary sinus. Also, the efflux of blood from the coronary arteries after TICRS and the oxygen saturation of the blood in the coronary arteries distal to an occlusion will need to be measured. The human population to benefit from TICRS are severe stable angina (class III and intravenous) patients whose
coronary artery disease is not amenable to PTCA and CABG, and in a subset of heart transplant recipients who develop atherosclerosis in the graft. For physiologic and hemodynamic testing of TICRS, a chronic myocardial ischemia pig model may best suite and reflect human cardiac physiology. We hope that TICRS, like transjugular intrahepatic portosystemic shunt, may one day be promising as a “bridge” to transplantation. References 1. American Heart Association. Heart and Stroke Statistical Update. Dallas: American Heart Association, 1999. 2. Ballard JC, Wood LL, Lansing AM. Transmyocardial revascularization:
390
●
Percutaneous TICRS
March 2000
3.
4. 5.
6. 7.
8.
9.
JVIR
criteria for selecting patients, treatment, and nursing care. Crit Care Nurs 1997; 17:42– 49. Kantor B, McKenna CJ, Caccitolo JA, et al. Transmyocardial and percutaneous myocardial revascularization: current and future role in the treatment of coronary artery disease. Mayo Clin Proc 1999; 74: 585–592. Mohl W. Coronary sinus interventions: from concept to clinics. J Card Surg 1987; 2:467– 493. Moll JW, Dziatkowiak AJ, Edelman M, Iljin W, Ratajczyk-Pakalska E, Stengert K. Arterialization of the coronary veins in diffuse coronary arteriosclerosis. J Cardiovasc Surg 1975; 16:520 –525. Hochberg MS. Selective retrograde coronary venous perfusion. Ann Thorac Surg 1980; 29:578 –588. Okada M, Nakamura M. Experimental and clinical studies on transmyocardial laser revascularization (TMLR). J Clin Laser Med Surg 1998; 16:197–201. Al-Sheikh T, Allen KB, Straka SP, et al. Cardiac sympathetic denervation after transmyocardial laser revascularization. Circulation 1999; 100:135–140. Wearn JT. The role of the thebesian vessels in the circulation of the heart. J Exp Med 1928; 47:293–316.
10. Pratt FH. The nutrition of the heart through the vessels of thebesius and the coronary veins. Am J Physiol 1893; 1:86 –103. 11. Printzmetal M, Simkin B, Bergman HC, Kruger HE. Studies on the coronary circulation II: the collateral circulation of the normal human heart by coronary perfusion with radioactive erythrocytes and glass spheres. Am Heart J 1947; 33: 420 – 442. 12. Pakalska E, Kolff WJ. Anatomical basis for retrograde coronary vein perfusion: venous anatomy and veno-venous anastomoses in the hearts of humans and some animals. Minn Med 1980; 63:795– 801. 13. Tschabitscher M. Anatomy of the coronary veins. In: Mohl W, Wolner E, Glogar D, eds. The coronary sinus: Proceedings of the 1st International Symposium of Myocardial Protection via the Coronary Sinus. New York: Springer-Verlag, 1984; 8 –10. 14. Kar S, Nordlander R. Coronary veins: an alternate route to ischemic myocardium. Heart Lung 1992; 21: 148 –157. 15. Vieussens R. Nouvelle decourvertes sur le coeur. Toulouse, 1706. 16. Thebesius AC. Dissertatio medica de circulo saguinis in corde. Lugduni Batavorum, 1708. 17. Roberts JT, Browne RS, Roberts G.
18.
19. 20.
21.
22.
23. 24. 25.
Nourishment of the myocardium by way of the coronary veins. Fed Proceed 1943; 2:90. Beck CS, Stanton E, Batiuchok W, Leiter E. Revascularization of heart by graft of systemic artery into coronary sinus. JAMA 1948; 137:436 – 442. Beck CS. Revascularization of the heart. Ann Surg 1948; 128:854 – 864. Beck CS, Hahn RS, Leighninger DS, McAllister FF. Operation for coronary artery disease. JAMA 1951; 147:1726 –1731. Hochberg MS, Roberts WC, Morrow AG, Austen WG. Selective arterialization of the coronary venous system: encouraging long-term flow evaluation utilizing radioactive microspheres. J Thorac Cardiovasc Surg 1979; 77:1–12. Beck CS, Leighninger DS. Scientific basis for the surgical treatment of coronary artery disease. JAMA 1955; 159:1264 –1271. Beck CS, Leighninger DS. Operation for coronary artery disease. JAMA 1954; 156:1226 –1233. Lazar H. Coronary sinus interventions during cardiac surgery. Ann Thorac Surg 1988; 46:475– 482. Mohl W. The relevance of coronary sinus interventions in cardiac surgery. Thorac Cardiovasc Surg 1991; 39:245–250.