Patent ductus arteriosus (PDA), accounting for nearly 30% of all congenital defects, is the most common congenital cardiac disease in dogs.1 It can occur in any breed, but small breeds appear to be overrepresented, with Maltese, Pomeranian, Shetland Sheepdog, English Springer Spaniel, toy Poodle, miniature Poodle, and Yorkshire Terrier being most at risk.2 Clinically, this defect is seen more often in females than in males.1 PDA leads to overcirculation of the lungs, volume overload of the left heart, and in some patients, left-sided congestive heart failure. If uncorrected, PDA results in a 1-year mortality that approached 65%.3 Surgical ligation of PDA has been performed in veterinary medicine for many years, with generally high success rates and acceptable (<8%) mortality rates.3 The most important complication of surgical ligation is hemorrhage, which reportedly occurs in <10% of dogs, but is frequently fatal (79%).4 The clinical utility of PDA occlusion with vascular occlusion coils is well established.5,a Benefits of the procedure include minimal invasiveness, high success rate, and low morbidity and mortality. Potential complications include pulmonary embolization of coils (3%), arterial embolization (<5%), recannalization of the PDA (5–20%) and hemolysis secondary to mechanical damage to erythrocytes.6–10 Other devices, such as the Amplatzer Duct Occluder and Gianturco-Grifka vascular occlusion device, also have been used successfully for PDA occlusion in dogs.11–14 Recently, the use of Amplatzer vascular plugs15,16,b,c and the Amplatz Canine Duct Occluder17,18,d,e for transvascular occlusion of PDA have been reported. The potential benefits of these devices compared with thrombogenic coils include ease of deployment, ability to reposition the device before release, and cost effectiveness. Our goal was to evaluate the utility of the Amplatzer vascular plug for occlusion of PDA in a large series of dogs.
Background: Transarterial ductal occlusion with the Amplatzer vascular plug was first reported in dogs by Hogan et al in 2005.
Hypothesis: Use of the Amplatzer vascular plug is a safe, efficacious method of patent ductus arteriosus (PDA) occlusion.
Animals: Thirty-one client-owned dogs with PDA.
Methods: Records of 31 dogs in which transarterial occlusion of PDA with an Amplatzer vascular plug was attempted were reviewed.
Results: All dogs had a type II PDA, with 27 dogs having type IIA morphology and 4 dogs having type IIB morphology. Appropriate device deployment was achieved in 29 of 31 dogs. Postdeployment angiography in 21 dogs documented complete occlusion in 10 dogs, trivial residual flow in 5 dogs, mild residual flow in 2 dogs, moderate residual flow in 3 dogs, and severe residual flow in 1 dog. Transthoracic color Doppler echocardiography documented complete occlusion in 22 dogs, whereas 2 dogs had trivial residual flow, 2 dogs had mild residual flow, 2 dogs had mild to moderate residual flow, and 1 dog had severe residual flow. Of the 7 dogs with residual flow, 2 had complete occlusion 2–4 months postoperatively, 1 had moderate residual flow 1 month postoperatively, and 4 were lost to follow-up. One dog required a larger device than was able to be deployed through the largest sheath placed in the femoral artery. Pulmonary embolization of the device occurred in 1 dog.
Conclusion: We conclude that ductal occlusion with an Amplatzer vascular plug is a safe and efficacious therapy for PDA in dogs.
Materials and Methods
The medical records of 31 dogs referred for definitive therapy of PDA via transcatheter occlusion from June 2005 to May 2007 were evaluated. Data regarding signalment, physical examination, thoracic radiographs, echocardiograms, angiograms, and device deployment were reviewed and analyzed. The diagnosis of PDA was confirmed in all dogs by physical examination, thoracic radiography, echocardiography, and angiography. Transesophageal echocardiography (TEE) was performed before and during the interventional catheterization procedure in 16 dogs to further characterize ductal size and morphology, help guide device deployment, and assess PDA occlusion postdeployment.
General anesthesia was induced by a standardized protocol which included premedication with hydromorphonef or butoraphanol,g and glycopyrrolateh; induction with diazepami and etomidatej; and balanced maintenance anesthesia with fentanylk and midazolaml constant rate infusion and sevoflurane inhalant.m The dogs were placed in right lateral recumbency, and the right femoral artery was isolated. Once isolated, access to the femoral artery was obtained by a modified Seldinger technique. A 4- or 6-French long sheathn was advanced over a guide wireo either directly into the artery or through an appropriately sized introducerp,q to the descending thoracic aorta immediately distal to the ductus. This position was easily identified, because the guide wire almost invariably entered the ductus during advancement, serving as a reliable landmark. Angiography was performed by injecting approximately 1 mL/kg of a nonionic contrast agentr with either an automated injectors or more frequently, vigorous hand injection (Fig 1A). PDA morphology, minimal ductal diameter, and both the width and length of the ductal ampulla were measured as described previously.19,20 An Amplatzer vascular plug was chosen such that the diameter of the plug was approximately 30% larger than the width of the maximal diameter of the ductal ampulla. The sheath and appropriate dilator then were advanced over the guide wire until the end of the sheath was located at the midportion of the ductal ampulla. The guide wire and dilator were removed, and the Amplatzer vascular plug was advanced to the tip of the sheath and slowly extruded (Fig 1B). The plug subsequently was advanced to the most distal aspect of the ductus (Fig 1C). In 22 dogs, angiography was performed and recorded 5–15 minutes after device deployment, but before release from the delivery cable, to determine, based on standardized criteria, if there was residual ductal flow (Table 1). In 9 dogs, angiograms were performed, but not recorded. Of those 9 dogs, 3 dogs also were evaluated for procedural success by TEE. The device then was disengaged from the delivery cable and repeat angiography performed (Fig 1D). After sheath removal, the femoral artery was ligated both proximally and distally, and the surgical wound closed routinely. Within 24 hours postoperatively, complete echocardiograms were performed on all dogs. Special emphasis was placed on evaluation of cardiac size, indices of systolic function, and presence or absence of residual flow (Table 1).
|Residual Flow||Angiographic Description||Echocardiographic Description|
|Trivial||Small puff of contrast which appears to float across orifice of PDA with no discrete jet||Random color pixels appearing at orifice of PDA with no discrete jet of color flow|
|Mild||Discrete jet of contrast which does not opacify the |
|Discrete jet of color flow which does not fill the pulmonary artery|
|Moderate||Discrete jet of contrast which partially opacifies the |
pulmonary artery, but with substantially less opacity
than the ductus
|Discrete jet of color flow which partially fills the pulmonary artery|
|Severe||Discrete jet of contrast which opacifies the pulmonary artery, but with similar opacity to the ductus||Discrete jet of color flow which fills the pulmonary artery|
Data are reported as mean±standard deviation (SD) when normally distributed, or as median and range when not normally distributed.
Transarterial ductal occlusion with the Amplatzer vascular plug was attempted in 31 dogs over the 2-year period. The study population consisted of 22 females and 9 males, ranging in age from 2.5 to 91 months (median, 6 months) and weighing between 2.4 and 22 kg (median, 6.4 kg). Breeds represented included 6 Bichon Frise (19%), 4 Shetland Sheepdogs (13%), 3 mixed breeds (10%), 2 (6.5%) each of Australian Shepherds, miniature Poodles, Weimaraners, German Shepherds, West Highland White Terriers, and 1 (3%) each of Australian Heeler, Newfoundland, Petit Basset Griffon Vendeen, Cavalier King Charles Spaniel, Pomeranian, Border Collie, Boston Terrier, and Affenpinscher.
On physical examination, all dogs had a left basilar continuous murmur. Heart murmur intensity ranged from grade III to VI/VI (median, grade V/VI). Six dogs (19%) also had a left apical systolic murmurs and 1 dog (3%) also had a right basilar systolic murmur. Femoral arterial pulse strength was determined to be hyperdynamic in 20/31 dogs (68%).
Mitral regurgitation secondary to annular dilatation was identified by color flow Doppler on transthoracic echocardiography in 16 dogs. Concurrent cardiac abnormalities were identified by transthoracic echocardiography in 5 dogs (16%). In addition to PDA, 1 dog each had moderate pulmonic stenosis (70 mmHg gradient), chronic valvular disease (CVD), severe subaortic stenosis (122 mmHg gradient), suspected pulmonary hypertension (PH) based on severe pulmonic insufficiency and an enlarged pulmonary artery (PA), and severe PH (estimated systolic PA pressure 93 mmHg) with bidirectional, but preferential left to right shunting. PH was assumed to be a sequela of the PDA in both dogs. Both dogs with PH had PA pressures measured intraoperatively, and were found to have systolic PA pressures of 70 and 74 mmHg, respectively. The difference between direct systolic PA pressure and those estimated by Doppler was assumed to be secondary to the effects of anesthesia.
Thoracic radiographs were available for evaluation in 28 dogs. Vertebral heart scores ranged from 9.8 to 13 (mean, 11.6 ± 2.4). Pulmonary overcirculation was present in 19/28 dogs (68%), with evidence of congestive heart failure in 2/28 dogs (7%).
PDA size and morphology were classified by angiography from right lateral thoracic radiographs in all dogs. All dogs had type II PDA, with 27 dogs having type IIA morphology and 4 dogs having type IIB morphology.20 Angiographic ductal length was 12.3 ± 2.8 mm (mean ± 1 SD; range, 6.8–17.9). The mean diameter of the ductal ampulla was 5.9 ± 2.1 mm (range, 2.3–14.3 mm). Mean minimal ductal diameter was 2.0 ± 0.9 mm (range, 0.5–5.0 mm). Mean plug diameter to angiographic ductal ampulla diameter ratio was 1.3 ± 0.3. Median fluoroscopy time was 8.7 minutes (range, 3.4–41.5 minutes).
At the time of deployment, devices were appropriately positioned in 31 dogs and the position was maintained in 29 dogs. In 1 dog, appropriate deployment was achieved, but angiography before release revealed severe residual flow through the PDA, and the device was removed. Occlusion with a larger device was not performed owing to inability to place an appropriately sized sheath, and the dog was subsequently sent to surgery for surgical ligation. In another dog, the device rotated in place after release from the delivery cable, which resulted in pulmonary embolization of the device. Angiography was performed and recorded 5–15 minutes after device deployment in 21 dogs and documented complete occlusion in 10 dogs, trivial residual flow in 5 dogs, mild residual flow in 2 dogs, moderate residual flow in 3 dogs and severe residual flow in 1 dog. In 3 of the 9 dogs that did not have postdeployment angiograms recorded, postdeployment procedural success was evaluated by TEE and documented complete occlusion in 1 dog, trivial residual flow in 1 dog and mild residual flow in 1 dog. Based on transthoracic color flow Doppler echocardiography performed on all dogs the day after the procedure, complete occlusion was achieved in 22 dogs (76%), 2 dogs had trivial residual flow, 2 dogs had mild residual flow, 2 dogs had mild to moderate residual flow, and 1 dog had severe residual flow. Of the 22 dogs with complete occlusion, 3 dogs with trivial residual flow, 1 dog with mild residual flow, 1 dog had mild to moderate residual flow, 1 dog had moderate residual flow, and 1 dog had severe residual flow on postdeployment angiogram. Of the 3 dogs with trivial flow on transthoracic echocardiography, 1 had trivial residual flow and 1 had moderate residual flow on postdeployment angiogram.
Transthoracic echocardiography was performed the day before and the day after the procedure. In addition to ductal flow, echocardiographic evaluation included left ventricular internal dimensions in systole and diastole (LVIDs and LVIDd) (M-mode), fractional shortening (M-mode), and area shortening (2-dimensional) measured from right parasternal short axis view at the level of the papillary muscles; left atrium to aorta ratio (LA : Ao) via M-mode measured from the right parasternal short axis view at the level of the aortic valve; and transaortic peak velocities obtained from a subcostal imaging plane. Post-PDA occlusion, median LVIDd decreased by 10.3 ± 16.3% whereas median LVIDs decreased by only 1.3 ± 17.5%. The median decrease in LVIDs does not truly reflect the changes seen in all dogs, as 4 out of 29 dogs experienced an increase in LVIDs postoperatively suggestive of residual systolic dysfunction. Mean fractional shortening decreased 6.1 ± 8.2%, and mean area shortening decreased 8.5 ± 9.2% postoperatively. Preoperatively, mean LA : Ao was slightly increased at 1.40 ± 0.38, and decreased to 1.18 ± 0.29 postoperatively. Mean transaortic velocities were mildly increased preoperatively at 2.11 ± 0.85 m/s, and decreased to 1.54 ± 0.85 m/s postoperatively.
Of the 7 dogs with residual flow, 2 dogs with trivial and mild flow based on transthoracic echocardiography the day after the procedure, respectively, had complete occlusion 2–4 months postoperatively. The dog with preexisting CVD, that had severe residual flow based on transthoracic echocardiography the day after the procedure, had moderate residual flow and moderate CVD with systolic dysfunction 1 month postoperatively. Four of the 7 dogs with residual ductal flow were lost to follow-up.
The only complication was pulmonary embolization in 1 dog. The affected dog's PDA was later (during the same anesthetic procedure) occluded with thrombogenic coils, with mild residual flow 24 hours postoperatively. A pulmonary perfusion scan performed 48 hours after plug embolization and subsequent coil occlusion revealed mildly decreased perfusion of the right caudal lung lobe. The dog recovered without complication and was discharged from the hospital.
Transcatheter interventional procedures developed for occlusion of PDA have gained wide acceptance in veterinary medicine.5,11–21,a,c Despite the advantage of a noninvasive method for PDA occlusion, the presently available occlusion devices still have potential disadvantages including a requirement for large vascular sheaths for deployment, complicated delivery mechanisms, residual leakage, and device embolization. The Amplatzer vascular plug requires a large delivery sheath, limiting its use to dogs with adequately sized femoral arteries. The sheaths required for some of the larger Amplatzer vascular plugs are larger than those required for controlled deployment of 0.052 vascular occlusion coils. The use of a long delivery sheath without an introducer allows for larger devices to be delivered. Even so, the use of an Amplatzer vascular plug in dogs <2.4 kg was not possible in this study owing to limitations of vascular access. Despite this relatively minor limitation, we found the Amplatzer vascular plug to have many benefits over methods used previously. The device comes preloaded on a delivery cable, which allows for controlled deployment, or withdrawal of the device if it is not positioned optimally. The presence and severity of residual flow then can be accessed before release of the device. The Amplatzer vascular plug is available in 7 diameter sizes (4, 6, 8, 10, 12, and 14 mm), which is especially useful for relatively large PDAs. Multiple vascular occlusion coils typically are required to occlude most PDAs, leading to increased cost, increased risk of embolization, as well as increased anesthetic, surgical, and fluoroscopy time.
The percentage of dogs with initial residual flow after PDA occlusion with the Amplater vascular plug in our experience is similar to previous experience with vascular occlusion coils.5,a Of the dogs with residual flow after the procedure, all but 1 had clinically unimportant residual flow and did not require a second procedure to further occlude the PDA. The dog with severe residual flow based on transthoracic echocardiography after the procedure had a decrease in transaortic flow velocities, an increase in diastolic blood pressure, loss of bounding femoral pulses, and a decrease in LV diameter after the procedure, all suggesting a substantial reduction in the flow across the PDA and a second procedure was not attempted. Three months later, further reduction in flow across the PDA was noted in that dog.
Complications associated with the use of the Amplatzer vascular plug for PDA occlusion are uncommon.15,16 The manufacturer recommends that the device be 130–150% the size of the vessel to be occluded, which correlates well with the upsize factor (device relative to ductal diameter) used in this series of dogs. In the 1 dog in which pulmonary embolization occurred, the plug diameter to angiographic ductal ampulla diameter ratio was 1.4, suggesting adequate sizing of the device. Embolization was not immediate. Once deployed, the plug was documented to be in the desired position. However, after occlusion of the ductus, dramatic dilatation of the ductal ampulla occurred. Once the ampulla dilated, the device began to slowly rotate within the ductal lumen. After approximately 45° of rotation, the edge of the device apparently became engaged in the ostium of the ductus, which halted further rotation. While in that position, the plug progressively compressed, and ultimately was forced through the ductus into the PA with subsequent embolization. Rotation of vascular plugs within the ductus, but without subsequent pulmonary embolization, after deployment has been reported.c No other clinically relevant complications were noted in this series of dogs.
The main limitation of this study was a lack of long-term follow-up on the majority of cases. Although recannalization has been reported as a potential complication with this device experimentally in peripheral veins in dogs, we have not recognized this complication when with the device for PDA occlusion.22 More long-term follow-up is needed to evaluate the importance of this potential complication when using the device for this application.
Results of the present study suggest that the Amplatzer vascular plug can be successfully used for occlusion of PDAs in dogs. Potential advantages including multiple device sizes, ease of delivery, and ability to provide complete closure make the device a good cost effective option in the transcatheter occlusion of PDAs in dogs.
The authors thank Kathy Glaze, RVT, and Katy Waddell, RVT, for their technical assistance.
aMiller MW, Meurs KM, Gordon SG, et al. Transarterial ductal occlusion using Gianturco vascular occlusion coils: 43 cases (1994–1998). J Vet Intern Med 1999;13:247 (abstract)
bAmpltazer Vascular Plug, AGA Medical Corp, Golden Valley, MN
cBussadori C, Carminati M, Domenech O, et al. A new employment of Amplatzer vascular plug: Implantation in two dogs with patent ductus arteriosus. 16th ECVIM-CA Congress, 2006 (abstract)
dAmplatz Canine Duct Occluder, LLC, Infiniti Medical, West Hollywood, CA
eNguyenba TP, Tobias AH. Patent ductus arteriosus occlusion with an investigational Amplatzer canine ductal occluder. J Vet Intern Med 2006;20:730 (abstract)
fHydromorphone HCl, Baxter Healthcare Corp, Deerfield, IL
gDolorex (Butorphanol tartrate), Intervet, Millsboro, DE
hGlycopyrollate, Fort Dodge Animal Health, Fort Dodge, IA
iDiazepam, Hospira Inc, Lake Forest, IL
jEtomidate, BenVenue Labs Inc, Bedford, OH
kFentanyl citrate, Hospira Inc
lMidazolam HCl, Baxter Healthcare Corp
mSevoflurane, Abbott Laboratories, North Chicago, IL
n4F, 6F Check-Flo Performer Mullins Introducer Set, Cook Inc, Bloomington, IN
oGlidewire Stiff type, Terumo Medical Corp, Somerset, NJ
p4F, 6F, 11F pediatric introducer, Infiniti Medical, Haverford, PA
q6F, 8F introducer, Boston Scientific Corp, Natick, MA
rOxilan, Cook Inc
sMark V Plus Injection System KMP 900 E, MedRad Inc, Indianola, PA