Successful interventional occlusion of muscular ventricular septal defect in a dog

Abstract

Ventricular septal defect (VSD) is a rare congenital heart disease in dogs. Hemodynamically important interventricular defects must be closed to improve the prognosis. This case report describes successful interventional transcatheter closure of a muscular VSD in a young Maltese and poodle mixed-breed dog with a large muscular interventricular defect (~5 mm in diameter) with a high rate of left-to-right shunt flow. The VSD was closed with a customized Amplatzer-type VSD occluder via a percutaneous transvenous (jugular) approach. We concluded that interventional occlusion of a muscular VSD with an Amplatzer-type occluder is a viable treatment option for dogs. A regular follow-up study for this dog is ongoing and has not detected complications.

Key clinical message:

Interventional occlusion of a muscular VSD with an Amplatzer-type occluder is a viable treatment option for dogs.

Ventricular septal defect (VSD), which is a hole(s) in the interventricular septum of the heart separating the right and left ventricles, is a congenital heart disease often progressing to congestive heart failure and pulmonary hypertension. Depending on the location and anatomic features of the defect, it is broadly classified as a membranous, muscular, or perimembranous type (1). Although several studies reported VSD as the 4th- to 6th-most common canine congenital heart disease (1–4), the prevalence rate of VSD in dogs is not well-documented, especially for muscular-type VSD, as these are rarer than other types (5–7). Most VSD are predominantly incidental findings and, as isolated, small VSD may not cause obvious clinical signs, even in adulthood (5,7), the actual prevalence rate is likely higher than reported in the literature.

Small defects may not cause a hemodynamically significant shunt or any clinical signs, even if untreated. However, large defects causing a hemodynamically significant shunt should be treated medically and/or surgically before the shunt causes severe pulmonary hypertension (i.e., right-to-left shunt). Ventricular septal defect can be closed surgically or by an intervention. The first successful interventional closure of VSD in humans was reported in 1988 (8,9). Since then, technology for interventional closure of VSD has continued to evolve, with improvements in devices used for closure and techniques used to access the heart (10–18). The first successful interventional closures of VSD in dogs were reported in the 2000s (19–21). Although interventional closure of VSD is now a well-established treatment option for dogs with VSD, a large defect, especially of a muscular type, is still challenging to successfully close. This case report describes a successful closure of a large VSD with a customized Amplatzer-type VSD occluder in a dog.

Case description

A 5-month-old male Maltese and poodle mixed-breed dog weighing 2.7 kg was presented for VSD closure. The chief complaint was a nonproductive cough. The dog was being given cardiac medications including sildenafil (Viagra; Pfizer, Seoul, Korea), 2 mg/kg, PO, q8h and pimobendan (Vetmedin; Boehringer Ingelheim, Ingelheim, Germany), 0.3 mg/kg, PO, q12h, in addition to drugs for a lower respiratory tract infection. On physical examination, a mild precordial thrill and a grade V/VI systolic murmur on the right heart base were present. Jugular vein distension was noted during blood collection but the hepatojugular reflex was negative. The remainder of the physical examination was unremarkable, including a normal systolic blood pressure of 126 mmHg. Red blood cell concentration was within the normal range (5.79 M/μL, reference range: 5.65 to 8.87 M/μL) and a remarkable increase in nucleated RBCs was not detected on the blood smear, indicating that the dog was not suffering from severe hypoxemia. There were no remarkable findings from the laboratory examinations.

The electrocardiogram indicated sinus tachycardia (180 to 200 beats per min) with deep S-wave, suggesting right ventricular enlargement. There was no heart block during the ECG. Thoracic radiographs revealed generalized cardiomegaly (vertebral heart scale: 11.6 V, intercostal space: 3.5; Figure 1 A). The caudal vena cava (CVC) was distended [CVC to T7 vertebral length (VL) ratio: 1.8]. In addition to enlargement of the right cardiac silhouette, the main pulmonary artery (MPA) and pulmonary vessels were dilated and over-circulated (Figure 1 B). However, there was no pulmonary edema. On abdominal radiographs, there was no evidence of ascites despite mild hepatomegaly.

Figure 1.

Thoracic radiography at presentation. A — Right lateral projection revealed marked generalized cardiomegaly (vertebral heart scale: 11.6 V, intercostal space: 3.5) and distended caudal vena cava. B — Ventrodorsal projection showed “reverse D”-shaped cardiac silhouette along with prominent dilation of main pulmonary artery, indicating enlargement of right cardiac chamber.

On echocardiography, right ventricular enlargement with interventricular septal flattening was observed (Figure 2 A), but there were no detectable anatomical abnormalities on cardiac valves, including the tricuspid valve. The left ventricular internal dimension at diastole (LVIDd) to right ventricular internal dimension at diastole (RVIDd) ratio was < 1, whereas the MPA to aorta (AO) ratio and the right pulmonary artery (RPA) to AO ratio were 1.41 and 0.72, respectively, indicating significant pulmonary hypertension (LVIDd/RVIDd < 1, MPA/AO > 1.1, and RPA/AO > 0.6 are suggestive of pulmonary hypertension; Figure 2 B). The pulmonary regurgitation peak velocity was measured as 4.3 m/s, suggesting that the systolic pressure of the pulmonary artery was > 74 mmHg (Figure 2 C). A large muscular interventricular defect (~5 mm in diameter) with a large volume of left-to-right shunt flow was visualized. The calculated pulmonary blood flow (Qp) to systemic blood flow (Qs) ratio was 2.75, suggesting that the interventricular defect was hemodynamically important (Figure 2 D). No additional congenital heart diseases (i.e., patent ductus arteriosus, atrial septal defects) were observed.

Figure 2.

Echocardiography at presentation. A — Two-dimensional image of right parasternal short axis view at papillary muscle level. A large muscular interventricular defect (measuring ~5 mm in diameter) was clearly visualized. B — Two-dimensional image of right parasternal short axis view at pulmonary artery level. Note the marked dilation of main and right pulmonary arteries, indicating pulmonary hypertension. The main pulmonary artery to aorta ratio was 1.41, whereas the aorta to right pulmonary artery ratio was 0.72. C — Pulse Doppler interrogation of the pulmonary artery. The peak velocity of pulmonary regurgitation was 4.3 m/s (pressure gradient: 74 mmHg), indicating severe pulmonary hypertension. D — Continuous Doppler interrogation at the interventricular defect detected a marked left-to-right shunt at systole but minimal right-to-left shunt at diastole. The calculated pulmonary blood flow (Qp) to systemic blood flow (Qs) ratio was 2.75, indicating left-to-right shunt direction.

The presence of cardiac remodeling and shunt flow ratio (Qp/Qs) indicated that the muscular VSD was hemodynamically significant and required repair to improve the prognosis. Therefore, transcatheter VSD occlusion was chosen and was performed in this dog 5 wk after the first visit. The diameter of the defect, the thickness of the interventricular septal wall, and the distance to the valves or chordae tendineae were taken into consideration when choosing the size and type of occlusion device. To cover this defect well, the occlusion device was customized by a commercial company (S&G Biotech, Seoul, Korea) based on the echocardiographic measurement (waist length: 5 mm, waist diameter: 6 mm, and disc diameter: 10 mm; Figure 3).

Figure 3.

Schematic diagram of the occluder used in this dog.

The body weight of the dog had increased to 3.7 kg. For the transcatheter VSD occlusion, the dog was premedicated with butorphanol (0.2 mg/kg, IV) and midazolam (0.2 mg/kg, IV). Anesthesia was induced with alfaxalone (2 mg/kg, IV) and maintained with 1 to 5% sevoflurane in oxygen. In addition, a preventive antibiotic (cefazolin; 30 mg/kg, IV) was administered before the intervention. The dog was positioned in left lateral recumbency to expose a right jugular vein.

A 5 Fr introducer-dilator set (Flexor Check-Flo Introducer Set; Cook Medical, Bloomington, Indiana, USA) was placed in the right jugular vein. Then, a guide wire (Roadrunner UniGlide Hydrophilic Wire Guide, 0.035″ × 150 cm; Cook Medical) was advanced through it. The angiocatheter (Renal Access Cobra Catheter, 5 Fr × 90 cm; Cook Medical) was then located at the VSD through the guide wire. After confirming the location of the angiocatheter across the VSD hole, the delivery catheter (Flexor Tuohy-Borst Side-Arm Introducer, 5 Fr × 0.038″ × 90 cm; Cook Medical) replaced the angiocatheter. A self-expandable VSD occluder (customized Amplatzer-type VSD occluder; S&G Biotech, Seoul, Korea) was then inserted into the delivery catheter until the first disc was exposed fully. Then, the occluder was pulled back to stick to the left-sided interventricular septal wall. After confirming the first disc was not moving, the second disc was exposed by gently pulling back the delivery catheter. After confirming that the occluder was positioned appropriately, it was detached from the delivery wire by rotating it counterclockwise.

The dog recovered uneventfully. Post-procedural thoracic radiography confirmed that the device was positioned properly (Figure 4). After the procedure, heart size was not dramatically reduced, although the CVC/VL ratio was reduced from 1.8 to 1.4 (Figure 4). The dog was discharged with oral antimicrobial medication (cephalexin; 30 mg/kg, PO, q12h) and a pulmonary vasodilator (sildenafil; 1 mg/kg, PO, q12h). The dog was in good condition 1 mo after the procedure, with considerably improved activity, and the device remained in place despite mild residual shunt flow.

Figure 4.

Thoracic radiography at 1 mo after intervention. Right lateral (A) and ventrodorsal (B) projections indicated the occluder (arrow) remained in place and in shape.

The LVIDd to RVIDd ratio was 2.1, whereas the AO/MPA ratio and AO/RPA ratios were 0.81 and 0.6, respectively, indicating a dramatic reduction of pulmonary hypertension. The pulmonary regurgitation peak velocity was measured as 1.84 m/s (pulmonary arterial pressure gradient was reduced from 74 to 13.5 mmHg; Figure 5 A), whereas tricuspid regurgitation peak velocity was measured as trivial. The calculated Qp/Qs ratio was 1.35 (right ventricular outflow tract diameter and velocity time integral were 13.0 mm and 6.7 cm, respectively; left ventricular outflow tract diameter and velocity time integral were 12.0 mm and 5.8 cm, respectively), indicating successful reduction of left-to-right shunt flow. The occluder was placed at the defect and successfully closed the shunt flow, although there was minimal residual flow at the occlusion site (Figure 5 B, C).

Figure 5.

Echocardiography at 1 mo after intervention. A — Severity of pulmonary hypertension was remarkably reduced (pulmonary arterial pressure gradient was reduced from 74 to 13.5 mmHg). B and C — The occluder (arrow) was placed at the defect and successfully closed the shunt flow, with minimal residual flow at the occlusion site.

There were no detectable short-term complications associated with transcatheter occlusion (e.g., aortic thromboembolism). Follow-up studies (at 1, 2, 4, and 6 mo after the procedure) confirmed that the device was well-maintained and the clinical condition had gradually normalized after a successful VSD closure. The dog was medicated with a pulmonary vasodilator (sildenafil; 0.5 mg/kg, PO, q12h) for 3 mo after surgery.

Approximately 9 mo after the procedure, the dog was returned to this clinic due to foreign body ingestion. At that time, there was no detectible murmur and were no clinical signs related to heart disease. The owner also confirmed the absence of clinical signs. Thoracic radiographs revealed the device was in the same position without changes in its morphology. However, the owner declined additional echocardiographic evaluation.

Discussion

Since left ventricular pressure is typically higher than right ventricular pressure, the shunt direction in most VSD is left to right. However, as right ventricular diastolic pressure increases over left ventricular diastolic pressure with time, the shunt may be reversed and cause severe pulmonary hypertension and secondary polycythemia (5). However, 1 human study reported that 40% of the VSD closed spontaneously and an additional 25 to 30% of defects became small enough to not cause a hemodynamically significant shunt (17). Therefore, not all VSD need to be closed.

Most isolated VSD in dogs and cats are associated with a good long-term prognosis and may not affect either quality or duration of life (5). An important criterion for surgical or interventional closure of VSD is Qp/Qs (pulmonary to systemic flow ratio) > 2.5. The Qp/Qs ratio is an index reflecting the direction and magnitude of shunt flow (22–24). A Qp/Qs < 1.0 indicates a right-to-left shunt, whereas Qp/Qs > 1.0 indicates a left-to-right shunt. In this case, the Qp/Qs at the first presentation was 2.75, along with evidence of severe pulmonary hypertension, indicating the requirement for either surgical or interventional closure of VSD to improve the prognosis. Surgical (open-heart) closure of muscular VSD is a limited therapeutic option in dogs because it requires cardiopulmonary bypass and has a high risk of peri- and post-period complications associated with surgery (25–27). Therefore, transcatheter occlusion of the muscular VSD was a more attractive option for this dog, although this procedure has rarely been done in dogs.

Only 2 case reports have described successful transcatheter occlusion of muscular VSD in dogs (6,28). Those 2 studies used Amplatzer muscular VSD occluders (Amplatzer Muscular VSD Occluder; Abbott Medical, Abbott Park, Illinois, USA) designed for human patients. Applying the ready-made occlusion devices from Abbott was problematic because the desired diameter of occluder for this case (matched to the diameter of the defect and thickness of surrounded septum) was not available. Moreover, the diameter of the delivery system was too large to use in this dog since it requires a 6 to 9 Fr vascular sheath. Therefore, we decided to order a customized occluder from a commercial company (S&G Biotech, Seoul, Korea).

The occluder was designed with dimensions of 6 mm (waist diameter), 5 mm (waist length), and 10 mm (disc diameter), based on echocardiographic measurements at systole and diastole. Furthermore, the device was designed to be accommodated in a 5 Fr delivery system. Since the maximal diameter of the defect at diastole in this case was ~5 mm, the diameter of the disc was designed as 10 mm, to cover the defect more completely. In addition, the length of the waist was designed as 5 mm, to optimally occlude the defect, since the septal thickness around the defect was 5.8 mm at systole.

Suboptimal occlusion with persistent residual shunt around the occluder has been reported in a dog with a ready-made Amplatzer muscular VSD occluder (28). For ready-made occluders, the shortest waist length was 7 mm, making it too loose to fit and possibly resulting in suboptimal occlusion even though the diameter of the disc covered the defect. To prevent suboptimal occlusion in this case, the waist length was designed 0.8 mm smaller than the actual septal thickness around the VSD. With this design, residual shunt activity was minimal at 1 mo after intervention.

The interval from design to delivery of the occluder was only 2 wk, which was relatively shorter than the interval when purchasing an off-the-shelf occluder from the local distributor (usually takes 2 to 4 wk to import). However, delivery may take longer if the occluder must be imported, and this could be a limitation. In this case, the clinical condition and severity of pulmonary hypertension were dramatically improved after the intervention, although normalization of cardiac remodeling took longer.

No short- or long-term complications associated with transcatheter occlusion of VSD have been detected to date in this dog (the last full examination was 6 mo after the procedure). Furthermore, no procedure- or device-related complications were observed. Complications (i.e., hemolysis, conduction disorders, and valve dysfunction) have been associated with transcatheter occlusion of VSD in humans (8–10). According to the human literature, the most important risk with device closure of membranous VSD is complete heart block (14,15), which was also reported in dogs during and after the procedure (29). In the present case, noticeable heart blocks were not observed during or after the procedure, although catheter-triggered premature ventricular contractions were often noticed during the procedure.

The second-most important risks with device occlusion are device-related valvular dysfunction (especially mitral and tricuspid valves) and hemolysis (15–18). These complications were observed in an experimental canine study (29). The complications described above might be closely related to the type of VSD and occlusion devices, because membranous VSD have higher risks of complications in human studies. In this case, the absence of procedure- or device-related complications was attributed to the device being very well-fitted to the VSD and the fact that the VSD was muscular (located at the lower part of the interventricular septum).

In conclusion, this report described a minimally invasive interventional occlusion of a muscular VSD with a customized Amplatzer-type occluder in a dog. The VSD was almost completely closed with this customized occluder, which was designed based on echocardiographic measurements. This was apparently the first case of successful occlusion of a muscular VSD with a customized occluder in a dog.