Patent ductus arteriosus (PDA) is the most common congenital

J Vet Intern Med 2004;18:325 329 Transarterial Coil Embolization of Patent Ductus Arteriosus in Small Dogs with 0.025-Inch Vascular Occlusion Coils: 10 Cases Daniel F. Hogan, Henry W. Green III, Sonya Gordon, and Matthew W. Miller Patent ductus arteriosus (PDA) is the most common congenital cardiac disease in the dog and generally leads to severe clinical signs, including left-sided congestive heart failure. Historically, definitive treatment consisted of surgical ligation; however, the use of vascular occlusion devices by minimally invasive techniques has gained popularity in veterinary medicine during the past decade. Adequate vascular access is a major limiting factor for these minimally invasive techniques, precluding their use in very small dogs. The clinical management of PDA with 0.025-in vascular occlusion coils in a minimally invasive transarterial technique in 10 dogs is described. The dogs were small (1.38 0.22 kg), were generally young (6.70 5.74 months), and had small minimal ductal diameters (1.72 0.81 mm from angiography). Vascular access was achieved, and coil deployment was attempted in all dogs with a 3F catheter uncontrolled release system. Successful occlusion, defined as no angiographic residual flow, was accomplished in 8 of 10 (80%) dogs. Successful occlusion was not achieved in 2 dogs (20%), and both dogs experienced embolization of coils into the pulmonary arterial tree. One of these dogs died during the procedure, whereas the other dog underwent a successful surgical correction. We conclude that transarterial PDA occlusion in very small dogs is possible with 0.025-in vascular occlusion coils by means of a 3F catheter system and that it represents a viable alternative to surgical ligation. The risk of pulmonary arterial embolization is higher with this uncontrolled release system, but this risk may decrease with experience. Key words: Congenital heart disease; Interventional devices; Pulmonary thromboembolism; Vascular access. Patent ductus arteriosus (PDA) is the most common congenital cardiac disease in dogs, and it accounts for almost 30% of all congenital defects. 1 Any breed of dog can be affected, but the highest risk is seen in small or toy breeds, including the Maltese, Poodle (toy and miniature), Bichon Frise, Pomeranian, and Yorkshire Terrier. 1 Uncorrected, PDA often results in left-sided congestive heart failure with a mortality rate of greater than 60% within the 1st year. 2 For these reasons, correction of the defect should be performed as early as possible. Correction of PDA by surgical ligation has been performed for more than 40 years. Success rates generally are very high (65 to 95%) with a relatively low and acceptable perioperative mortality rate ( 10%) in uncomplicated cases. 2 7 However, mortality rates rise dramatically (40 100%) when substantial hemorrhage is associated with the procedure, which occurs more frequently in dogs of older age at the time of surgery and is most likely to be fatal in very small dogs because of rapid exsanguination. 2,3,6,8 During the past decade, the use of catheter-delivered intravascular occlusion devices by minimally invasive techniques has gained popularity. 7,9 18 Correction with vascular From the School of Veterinary Medicine, Department of Veterinary Clinical Sciences-Lynn Hall, Purdue University, West Lafayette, IN (Hogan, Green); and the Department of Small Animal Medicine and Surgery, College of Veterinary Medicine, Texas A&M University, College Station, TX (Gordon, Miller). The results of this study were presented at the 21st Annual American College of Veterinary Internal Medicine Forum, 2003, Charlotte, NC. The work was performed at Purdue University and Texas A&M University. The author represents and warrants that his or her part of the work as submitted will in no way violate any copyright, or any other right. Reprint requests: Daniel F. Hogan, DVM, DACVIM-Cardiology, School of Veterinary Medicine, Department of Veterinary Clinical Sciences-Lynn Hall, 625 Harrison Street, West Lafayette, IN 47907-2026; e-mail: hogandf@purdue.edu. Submitted September 26, 2003; Revised November 20, 2003; Accepted December 22, 2003. Copyright 2004 by the American College of Veterinary Internal Medicine 0891-6640/04/1803-0011/$3.00/0 occlusion coils by transarterial access is the most common technique. This procedure most frequently is performed by veterinary cardiologists and is the most commonly used technique for PDA correction at some institutions. Success rates are generally high (90%), and perioperative mortality rates are very low ( 1%). 9,12 Limitations of these techniques include inadvertent pulmonary arterial or aortic embolization, incomplete occlusion, and inadequate vascular access in very small dogs. Originally, occlusion coils were deployed by catheter manipulation and extrusion of the coil with a wire guide, giving the operator limited control over coil release. To decrease the risk of pulmonary or aortic embolization, techniques have been modified or equipment developed that allows operator-controlled release of the occlusion coil. a d,12 14,16 18 Unfortunately, these modifications have often resulted in the requirement for larger delivery devices, further limiting their use in very small dogs and simultaneously increasing cost. To address inadequate arterial vascular access, transvenous (retrograde) coil deployment techniques have been used successfully in small dogs and cats. 10,13,16 18 Access to the PDA is accomplished from the pulmonary arterial side, which can be difficult in some cases. Additionally, this approach represents a more circuitous route to the PDA and may limit multiple coil deployment, resulting in inadequate occlusion and residual ductal flow. Our goal was to demonstrate the ability to successfully occlude PDA in very small dogs by the standard and more direct transarterial technique along with standard, easily acquired, and cost-effective equipment. We explored the use of 0.025-in vascular occlusion coils e deployed through a 3F diagnostic catheter. Materials and Methods Ten consecutive dogs, precluded from standard transarterial PDA embolization techniques because of inadequate vascular access, were entered in this study between January 2002 and October 2003. Eight dogs presented to the Purdue University (West Lafayette, IN) Veterinary Teaching Hospital (PUVTH), and 2 dogs presented to the Texas A&M University (College Station, TX) Veterinary Medical Teaching

326 Hogan et al Hospital (TAMU VMTH). Dogs ranged in weight from 0.9 to 1.7 kg (1.38 0.22 kg) and in age from 2 to 19 months (6.70 5.74 months). There were 4 males (40%) (1 neutered and 3 intact) and 6 females (60%) (6 intact). Breeds represented included 3 Maltese (30%), 2 Pomeranians (20%), and 1 (10%) each of Shetland Sheepdog, Yorkshire Terrier, Papillon, Bichon Frise, and mixed breed. The diagnosis of PDA was confirmed by physical examination, thoracic radiography, echocardiography, and angiography in all dogs. General anesthesia was induced by standard techniques according to the anesthesiologist on duty. The dogs were placed in right lateral recumbency, and the right femoral artery was accessed by surgical exposure. The femoral artery was punctured by a 21-gauge, thinwalled vascular access needle f through which a 0.021-in wire guide was passed. The puncture site was expanded by passing the dilator portion of a 3F vascular access sheath. g Leaving the wire guide in place, the dilator was removed, and a 3F diagnostic catheter h was advanced into the femoral artery to the proximal descending thoracic aorta. Angiography was performed by injecting an iodinated i (PUVTH) or a noniodinated j (TAMU VMTH) contrast agent of approximately 1 ml/kg, and the minimal ductal diameter (MDD) was measured. An appropriately sized embolization coil was chosen for which the unrestricted coil diameter was at least twice that of the MDD. The diagnostic catheter was moved caudally to engage the origin of the ductus and was then advanced to the midportion of the ductal ampulla. The embolization coil was advanced to the tip of the catheter with a 0.021-in wire guide and was slowly extruded into the ampulla. To avoid pulmonary arterial embolization, 1 1.5 coil loops were extruded from the catheter within the midportion of the ductal ampulla and then advanced cranially to try to avoid distortion of the coil that might result in passage across the internal diaphragm of the PDA. Angiography was repeated 5 15 minutes after coil placement to determine if there was residual ductal flow. If there was residual flow, additional coils were placed within the ampulla to achieve complete closure. When additional coils were deployed, every attempt was made to engage the previously deployed coil to limit the risk of inadvertent arterial embolization. After catheter removal, the femoral artery was double-ligated both proximally and distally, and the surgical wound was routinely closed. Results MDDs were relatively small (1.72 0.81 mm) with angiographic ductal anatomy 20 as follows: IIA (7 dogs) and IIB (3 dogs). Successful occlusion, defined as no residual angiographic ductal flow, was achieved in 8 of 10 (80%) dogs (Fig 1). In 5 of 8 (63%) dogs, only 1 embolization coil was required (for 3 dogs, a 3-mm-diameter coil, and for 2 dogs, a 5-mm-diameter coil). Of the remaining 3 dogs, 2 required 2 coils for complete closure (two 5-mm coils for one dog and, for the other dog, one 5-mm coil and one 3- mm coil), and 1 required 3 coils (two 5-mm coils and one 3-mm coil). In one dog that required 2 coils, the 1st 5-mmdiameter coil embolized to the pulmonary arterial tree. Because we felt that the original coil was not well positioned, a 2nd 5-mm-diameter coil was deployed that achieved stable fixation within the ductal ampulla. Other than the pulmonary arterial embolization, no complications occurred in any of the successful occlusions. Successful occlusion could not be obtained in 2 of 10 (20%) dogs. Each of these dogs had small MDDs (1 mm), but 5-mm-diameter coils passed through the internal diaphragms with little or no apparent distortion, resulting in pulmonary arterial embolization. Occlusion was attempted with two 5-mm coils in each dog. One dog subsequently was corrected by surgical ligation. The other dog died during the procedure. Attempts to retrieve the embolized coils with a vascular snare a in this dog were unsuccessful. Retrieval of embolized coils was not attempted in the other 2 dogs with pulmonary arterial embolization because the dogs were stable and the risk for hemodynamic compromise was judged to be low. In the 9 dogs that survived the procedure, all recovered and experienced no short- or longterm complications. Discussion Transcatheter correction of PDA with vascular occlusion coils was first explored as a viable alternative to surgical ligation that could provide adequate clinical success rates and reduce morbidity and mortality rates while reducing cost. Originally, 0.038-in coils were deployed through a 5F diagnostic catheter by means of a transarterial technique with limited operator-controlled release. 7,9 11 Limitations with this technique included inadvertent pulmonary arterial and aortic embolization, along with incomplete PDA occlusion with residual shunting. Larger-gauge coils (0.052 in) were used to provide better occlusion, but these required a 6F guiding catheter for deployment and necessitated larger vascular access. In addition, the catheter minimal internal diameter was larger than the coil wire gauge. This difference sometimes resulted in the coil being sucked out of the distal end of the catheter before the operator s decision to deploy the device. Different techniques were developed to facilitate controlled coil deployment, providing better coil positioning and a reduction in the risk of inadvertent pulmonary arterial and aortic embolization. These techniques included the use of a vascular snare, a interlocking release wire, b threaded delivery wire, c and myocardial bioptome. d,12 14,16,18 Although all of these techniques provided substantial improvements to transcatheter correction of PDA, they have potential limitations. Minimal catheter size for the original or interlocking release wire technique is 4F, which often precludes vascular access in very small dogs. Larger delivery devices (a 6F guiding catheter or a 4F vascular introducer) are required for the threaded delivery wire and myocardial bioptome. Venous and arterial access is required with the vascular snare technique in which the snare engages the pulmonary arterial aspect of the coil that is deployed through a transarterial approach. 14 Additionally, the coils occasionally can become entangled in the snare, making release difficult. Smaller wire gauge occlusion coils are available (0.018 in) but are composed of platinum, which has less radial strength than stainless steel and generally is not acceptable for high flow rate applications such as PDA. The wire gauge of vascular occlusion coils reported in product literature is larger than the actual wire gauge. For example, a 0.038-in coil actually is a 0.035-in wire with thrombogenic fibers extending from the wire. For this reason, 0.038-in coils can be deployed through 4.1F catheters that have a 0.035-in internal diameter. We determined that such also was the case for 0.025-in coils that could be deployed through a 3F catheter with a 0.021-in internal diameter. The 0.025-in coils are constructed of stainless steel and therefore have increased radial strength compared to the 0.018-in platinum coils. A 3F catheter allows arterial

Small Dog PDA 327 Fig 1. Descending aortography performed 5 minutes after the placement of a 0.025-in, 4-cm-long, 3-mm-diameter (25 4 3) occlusion coil (approximately 4 magnification). vascular access in all but the smallest of dogs, and we were able to gain vascular access in all of the dogs of this study (0.90 1.70 kg). In fact, we found that vascular access was not difficult, but the femoral artery is small and must be gently manipulated. Careful loading of the coil into the catheter was necessary, as the coils are very small and can get entangled in the catheter hub (Fig 2). Equipment costs were minimal ($62 total; $27 1 coil, $16 wire guide, and $19 catheter) and were thought to be cost-effective from a clinical perspective. Eight of 10 dogs (80%) were occluded successfully; however, the other 2 dogs experienced pulmonary arterial embolization, and their PDA could not be occluded. A coil embolized to the pulmonary arterial tree in 1 other dog, but this dog s PDA was successfully occluded (dog 5). However, all of these dogs were from the PUVTH, where most procedures were performed, and the author (DFH) has had less experience with uncontrolled coil release. Data from studies in human patients suggest that embolization rates decrease with operator experience when uncontrolled release techniques are used. 21 One of these dogs was successfully corrected with surgical ligation. Unfortunately, the other dog died unexpectedly from cardiac arrest during the procedure. This dog showed more signs of severe cardiac disease ie, moderate mitral insufficiency and evidence of systolic dysfunction and these abnormalities have been identified as markers for increased mortality with surgical correction. 2 The most plausible explanation for the cause of

328 Hogan et al Fig 2. Photograph of a 0.025-in, 4-cm-long, 3-mm-diameter coil (25 4 3) demonstrating the very small size of the coil. death was acute pulmonary hypertension from the embolized coils. Postmortem examination identified a slitlike internal diaphragm in which the length was much larger than the height of the slit. This finding could explain why the 5-mm-diameter coil passed through the internal diaphragm of the PDA without apparent distortion when the MDD measured only 1 mm. The length of the slit was parallel to the fluoroscopic beam, and so the height of the slit was measured. Although it is not known for certain, it is possible that the other dog with unsuccessful occlusion had a similar internal diaphragm. Vascular snares a,k are available and can be used to retrieve coils that have embolized into the pulmonary or systemic arterial tree. These include microsnares k that can be deployed through 2.3F to 3F catheters and thus are compatible with the small femoral arteries in dogs of this size. It is possible to have a controlled release system for deployment of the 0.025-in coils. Currently, a commercially available detachable system that uses 0.011- to 0.018-in coils can be deployed through 2.6F to 3F microcatheters. l This system uses a threaded delivery wire similar to that of larger systems. However, the veterinary medical market is too small and the human medical market is not large enough to justify production expenses. An improvement in success rate and a lowering of perioperative mortality rate is desirable and expected with increased experience with this technique. Similar trends were seen with early results of surgical ligation of PDA in dogs. 4 However, most dogs (5 of 8 [63%]) were occluded successfully with only 1 occlusion coil, most likely because of the small MDD. Although influenced by the experience of the operator, anesthetic, procedural, and fluoroscopic times were relatively short (86.0 30.1 minutes, 48.0 17.0 minutes, and 5.9 4.7 minutes, respectively) and were comparable to anesthetic and procedural times for surgical ligation (166.0 34.6 minutes and 96.5 26.0 minutes, respectively; Hogands, unpublished data). We conclude that 0.025-in vascular occlusion coils deployed through a 3F diagnostic catheter with an uncontrolled release technique can be used successfully to correct PDA in very small dogs and that they represent a viable alternative to surgical ligation. We believe that the rate of successful occlusion will increase and that the risk of inadvertent embolization will decrease when more experience has been acquired with this technique. Footnotes a Amplatz Goose Neck Snare Kit, Microvena Corp, White Bear Lake, WI b Flipper detachable embolization coil delivery system, Cook, Inc, Bloomington, IN c Detachable embolization coils, Cook, Inc, Bloomington, IN d Myocardial biotome, FBF-3.0-120, Cook, Inc, Bloomington, IN e Occlusion coils, MWCE-25-4-3, MWCE-25-5-5, Cook, Inc, Bloomington, IN f Needle, SDN-21UT-2.5, Cook, Inc, Bloomington, IN g Sheath, VCF-3.0-21-5-J, Cook, Inc, Bloomington, IN h Diagnostic catheter, N3.0-21-75-P-NS-JR2.5, Cook, Inc, Bloomington, IN i Hypaque 60%, Nycomed Amersham, Princeton, NJ j Oxilan-300, Cook, Inc, Bloomington, IN k Amplatz Goose Neck Microsnare Kit, Microvena Corp, White Bear Lake, WI l Detach, detachable coil system, Cook, Inc, Bloomington, IN References 1. Buchanan JW. Prevalence of cardiovascular disorders. In: Fox PR, Sisson DD, Moïse NS, eds. Textbook of Canine and Feline Cardiology: Principles and Clinical Practice, 2nd ed. Philadelphia, PA: WB Saunders; 1999:457 470. 2. Eyster GE, Eyster JT, Cords GB, Johnston J. Patent ductus ar-

Small Dog PDA 329 teriosus in the dog: Characteristics of occurrence and results of surgery in one hundred consecutive cases. J Am Vet Med Assoc 1976;168: 435 438. 3. Bellenger CR, Hunt GB, Goldsmid SE, Pearson MRB. Outcomes of thoracic surgery in dogs and cats. Aust Vet J 1996;74:25 30. 4. Buchanan JW. Patent ductus arteriosus. Semin Vet Med Surg Small Anim 1994;9:168 176. 5. Ackerman N, Burk R, Hahn AW, Hayes HM. Patent ductus arteriosus in the dog: A retrospective study of radiographic, epidemiologic, and clinical findings. Am J Vet Res 1978;39:1805 1810. 6. Birchard SJ, Bonagura JD, Fingland RB. Results of ligation of patent ductus arteriosus in dogs: 201 cases (1969 1988). J Am Vet Med Assoc 1990;196:2011 2013. 7. Van Israel N, French AT, Dukes-McEwan J, Corcoran BM. Review of left-to-right shunting patent ductus arteriosus and short term outcome in 98 dogs. J Small Anim Pract 2002;43:395 400. 8. Hunt GB, Simpson DJ, Beck JA, et al. Intraoperative hemorrhage during patent ductus arteriosus ligation in dogs. Vet Surg 2001;30:58 63. 9. Miller MW, Meurs KM, Gordon SG. Transarterial ductal occlusion using Gianturco vascular occlusion coils: 43 cases 1994 1998 (abstract). Proceedings of the 17th Annual Veterinary Forum, American College of Veterinary Internal Medicine, Chicago, IL, 1999. 10. Fellows CG, Lerche P, King G, Tometzki A. Treatment of patent ductus arteriosus by placement of two intravascular embolisation coils in a puppy. J Small Anim Pract 1998;39:196 199. 11. Snaps FR, Mc Entee K, Saunders JH, Dondelinger RF. Treatment of patent ductus arteriosus by placement of intravascular coils in a pup. J Vet Med Assoc 1995;207:724 725. 12. Stokhof AA, Sreeram N, Wolvekamp WTC. Transcatheter closure of patent ductus arteriosus using occluding spring coils. J Vet Intern Med 2000;14:452 455. 13. Schneider M, Hildebrandt N, Schweigel T, et al. Transvenous embolization of small patent ductus arteriosus with single detachable coils in dogs. J Vet Intern Med 2001;15:222 228. 14. Fox PR, Bond BR, Sommer RJ. Nonsurgical transcatheter coil occlusion of patent ductus arteriosus in two dogs using a preformed nitinol snare delivery technique. J Vet Intern Med 1998;12:182 185. 15. Glaus TM, Berger F, Ammann FW, et al. Closure of large patent ductus arteriosus with a self-expanding duct occluder in two dogs. J Small Anim Pract 2002;43:547 550. 16. Tanaka R, Hoshi K, Nagashima Y, et al. Detachable coils for occlusion of patent ductus arteriosus in 2 dogs. Vet Surg 2001;30:580 584. 17. Schneider M, Hildebrandt N. Transvenous embolization of the patent ductus arteriosus with detachable coils in 2 cats. J Vet Intern Med 2003;17:349 353. 18. Miller MW. Transarterial coil occlusion of patent ductus arteriosus: Outcome in 120 cases. Proceedings of the 20th Annual Veterinary Forum, American College of Veterinary Internal Medicine, Dallas, TX, 2002. 19. Schneider M, Fischer C, Hildebrandt N, Bierent N. Embolization of small- and medium-sized PDA ( 4.0 mm) with a single elongated coil. J Vet Intern Med 2003;17:442 (abstract). 20. Miller MW, Meurs KM, Lehmkuhl L, et al. Angiographic classification of patent ductus arteriosus in the dog (abstract). Proceedings of the 16th Annual Veterinary Forum, American College of Veterinary Internal Medicine, San Diego, CA, 1998. 21. Rosenthal E, Qureshi SA, Reidy J, et al. Evolving use of embolisation coils for occlusion of the arterial duct. Heart 1996;76:525 530.