Asian Cardiovasc Thorac Ann 2002;10:124-128
© 2002 Asia Publishing EXchange Pte Ltd
Arterial Duct Closure With Detachable Coils: Application in the Small Child
Ümrah Aydo
an, MD
|
Department of Pediatric Cardiology Istanbul Medical Faculty Istanbul University Istanbul, Turkey
|
|
|
Ümrah Aydog an, MD Tel: 90 532 612 4719 Fax: 90 212 621 1643 email: uaydogan{at}turk.net Department of Pediatric Cardiology, Istanbul Medical Faculty, Istanbul University, Millet Caddesi, Çapa, Istanbul 34390, Turkey.
|
 |
ABSTRACT
|
|---|
Transcatheter closure of patent ductus arteriosus using controlled-release coils was performed in 16 patients weighing < 10 kg. No embolization occurred. Procedure-related complications occurred in 3 patients (18.8%): massive femoral hemorrhage in 1 and femoral artery thrombosis in 2. The ductus recanalized in 1 of them because of mechanical hemolysis caused by streptokinase treatment. This was the only patient who underwent another occlusion procedure. Complete occlusion was achieved in 7 patients (43.8%) immediately, in 13 (81.2%) the following day, and in all 15 patients who had completed the 6-month follow-up. During follow-up, flow velocities between the left and the main pulmonary arteries and between the descending and the ascending aortae did not differ significantly. Flow velocity was > 2 m•sec-1 in 3 patients in the left pulmonary artery and in 1 in the descending aorta. Protrusion of the coil was seen in 3 of these patients. Flow velocity was also high in the main pulmonary artery in the 4th patient. In conclusion, coil occlusion of ductus arteriosus is feasible in the small child, but no more than half a loop of the coil should be left at the pulmonary site. High flow velocity does not always mean obstruction.
|
INTRODUCTION
|
|---|
Transcatheter ductal occlusion has become an accepted nonsurgical treatment option for patients with clinically evident patent ductus arteriosus (PDA). The Rashkind umbrella has been used extensively for this purpose, but it can potentially cause significant obstruction of the left pulmonary artery (LPA) and the descending aorta (DAo) as well as residual shunting despite proper device delivery.1–3 Subsequently, embolization techniques using Gianturco coils were employed, producing excellent rates of complete ductal occlusion.4,5 However, there is a high rate of embolization of the coil to the peripheral pulmonary arteries with this device, which is not release-controlled.6 Recently, this problem has been resolved with controlled-release Jackson coils, and they have become widely used by pediatric cardiology centers.7
Although anatomical closure of the PDA is typically completed on the first days of life after birth in most babies, rarely there may be a delay in this phenomenon until the age of 1 year.8 Hence, most institutions rec-ommend intervention after 12 months of age if there is any evidence of heart failure.9 However, small infants with congestive heart failure that is refractory or poorly responsive to medical management need urgent inter-vention, regardless of the age or weight of the patient. This report presents our experience with the implantation of controlled-release coils in children weighing under 10 kg and assesses the immediate and midterm hemodynamic influence of the coils on periductal vessels.
|
PATIENTS AND METHODS
|
|---|
From July 1997 to August 2000, the parents of patients who needed early intervention for PDA were given information about surgical division/ligation or trans-catheter coil occlusion. Patients whose parents preferred surgical therapy or whose echocardiographic and angio-graphic findings revealed PDA morphology and diameter unsuitable for coil occlusion were excluded from the study. Transcatheter occlusion using controlled-release Jackson coils (Cook Inc., Bloomington, IN, USA) was performed in 16 patients.
The coil was implanted under local anesthesia, and ketamine was used for sedation in all patients. A 5F NIH venous catheter was advanced to the juxtaductal region of the DAo, followed by a 5F right Judkins arterial catheter delivered retrogradely to the pulmonary site whenever possible, passing through the PDA. The morphology and the narrowest diameter of the duct were evaluated on lateral aortography. Jackson coils with a diameter 1.5 to 2 times the measured minimum ductal diameter and with 3 to 5 loops were selected for the procedure (Figure 1
). Occlusion was performed either anterogradely or retrogradely as described elsewhere.7 No more than 1 loop of the coil was left at the pulmonary site in anterograde applications, while up to 1.5 loops were inadvertently released in the pulmonary artery in retrograde technique. In patients whose post-occlusion control aortograms, taken 15 minutes after implantation, revealed manifest residual shunting, a second coil was implanted retrogradely in the same procedure.
View larger version (19K):
[in this window]
[in a new window]
|
Figure 1. Delivery of the Jackson coil. (A) The mandril of the delivery system was pushed into the coil. (B) The mandril was withdrawn with the delivery catheter placed across the arterial duct, protruding half to 1 loop of the coil into the pulmonary artery.
|
|
Echocardiographic studies and physical examination were conducted on the day after the occlusion procedure and, subsequently, every 6 months in all patients. Transthoracic echocardiography was performed using commercially available Vingmed CFM750 ultrasound instruments (Vingmed Sounds, Horten, Norway). Standard images included the parasternal short-axis and high left parasternal view to evaluate the pulmonary arteries and the PDA, as well as the suprasternal notch view to assess the PDA and the proximal DAo. Coil position was evaluated by 2-dimensional echocardiography. Blood flow velocities across the LPA and the DAo were measured and the presence of residual shunting investigated by color and continuous Doppler imaging. At the latest follow-up (at 1 day to 3 years after occlusion), blood flow velocities in the main pulmonary artery (MPA) and the ascending aorta (AAo) were also measured. Flow velocities in the MPA and the LPA were measured by pulsed Doppler imaging at that investigation.
Peak flow velocities are expressed as mean ± standard deviation. Comparison of flow velocities between the MPA and the LPA and between the AAo and the DAo measured at the latest follow-up was performed by Student's t test. Statistical significance was defined as p < 0.05.
|
RESULTS
|
|---|
The 16 patients, 12 (75%) of whom were girls, were aged 3.1 to 29.8 months (median, 11.4 months) and weighed 4.5 to 10 kg (median, 7.6 kg) on the day of the procedure. Five patients had additional cardiac abnormalities: 2 had bicuspid aortic valve without stenosis, while the other 3 had, respectively, discrete coarctation of the aorta, discrete coarctation of the aorta with a moderate-size ventricular septal defect, and a moderate-size ventricular septal defect. One patient had ulnar agenesis and another had torticollis as noncardiac abnormalities.
Minimal ductal diameter measured on aortograms was 1.5 to 4.5 mm (median, 4 mm). The coil was implanted via the arterial route in 4 patients, transvenously in 10, and via both routes simultaneously in the other 2. A single coil was used in 10 patients (62.5%). Including the simultaneous implantations, 6 patients (37.5%) needed a second coil, implanted via the arterial route, because of manifest residual shunting.
Complete angiographic occlusion was achieved imme-diately in 7 patients (43.8%). However, femoral artery thrombosis developed in one of them. Recanalization of the PDA and mechanical hemolysis occurred during streptokinase treatment in this patient. A second coil was implanted 38 hours after the initial one. Only trivial leakage was observed on post-occlusion angiography in the other 9 patients (56.2%); it was decided that they would be followed up by echocardiography.
Successful balloon angioplasty was performed in the 2 patients with coarctation of the aorta in the same procedure with PDA occlusion, after determining angiographically that the coils in the ampulla had not protruded into the DAo. Post-procedure aortography revealed an angioplasty-related microaneurysm in the DAo in one of the patients. Its size had remained the same during follow-up.
Procedure-related complications were encountered in 3 cases (18.8%). Massive femoral hemorrhage occurred in 1 patient, but he did not need blood transfusion. Femoral artery thrombosis developed in the other 2. Initial therapy with heparin infusion was not successful after 24 hours. The problem was resolved with streptokinase treatment over a few hours; but, as mentioned earlier, mechanical hemolysis occurred in one of them, who needed blood transfusion.
The number of complete occlusion increased to 13 patients (81.2%) with the addition of 6 more patients after 24 hours, and to 15 patients after 6 months. The last patient had not completed the 6-month follow-up.
Echocardiography performed on the day after occlusion showed that flow velocity in the LPA exceeded 2 m•sec-1 (2.26 m•sec-1) in 1 patient. This patient had a single 5-mm coil and only half a loop left at the pulmonary site. No protrusion of the device was detected on 2-dimensional echocardiography. However, pulsed Doppler echocar-diography at her latest examination revealed that flow velocity in the MPA was also high (1.86 m•sec-1), while it was 2.19 m•sec-1 in the LPA.
Two patients had elevated LPA peak flow velocity of 2.01 (at 12 months) and 2.23 m•sec-1 (at 24 months), respectively, during follow-up, whereas it was within normal limits on the day following occlusion. Two coils had been implanted in both patients. Their post-occlusion angiograms revealed that a complete loop of a 5-mm coil in 1 patient and 1.5 loops of an 8-mm coil in the other had been left at the pulmonary site. Two-dimensional echocardiography revealed protrusion of the devices into the pulmonary artery in both patients. Pulsed Doppler flow velocities in the MPA at the latest follow-up were 0.95 and 0.90 m•sec-1, respectively.
Peak flow velocity in the DAo was high (3.46 m•sec-1) with diastolic runoff in a patient on the day after occlusion (Figure 2). This patient had a wide PDA with narrow ampulla, and 2 coils (8 mm and 5 mm, respectively, with 5 loops each) had been implanted to achieve complete occlusion. Her post-occlusion aortography and 2-dimen-sional echocardiography showed protrusion of the devices into the DAo. Peak flow velocity in the DAo decreased to 2.14 m•sec-1 at 6 months and 1.88 m•sec-1 at 18 months without diastolic runoff, while it was 1.34 m•sec-1 in the AAo.
View larger version (205K):
[in this window]
[in a new window]
|
Figure 2. Continuous wave Doppler echocardiography (A) on the day after ductal occlusion, showing coarctation of the aorta with diastolic runoff in which the ductal ampulla is too narrow, and (B) at 18-month follow-up, showing spontaneous resolution of the coarctation.
|
|
At the latest follow-up, peak flow velocities in this series of patients were in the range of 0.87 to 1.86 m•sec-1 (mean, 1.25 ± 0.28; median, 1.26 m•sec-1) in the MPA and 0.81 to 2.23 m•sec-1 (mean, 1.51 ± 0.44; median, 1.36 m•sec-1) in the LPA. There was no statistical difference between the MPA and LPA values. Peak flow velocities ranged from 0.89 to 2.08 m•sec-1 (mean, 1.45 ± 0.33; median, 1.44 m•sec-1) in the AAo and 1.05 to 1.88 m•sec-1 (mean, 1.52 ± 0.36; median, 1.44 m•sec-1) in the DAo. The AAo and DAo values did not differ significantly.
|
DISCUSSION
|
|---|
The incidence of catheterization-related femoral artery thrombosis in pediatric patients is as high as 7% when prophylactic heparin is used.10 It should be even higher if only small babies are considered. Hence, we think that the rate of 12.5% (2 patients) found in our study group is not higher than that reported in the literature. Further-more, prophylactic heparin was not used in our patients in order to speed up thrombus formation on the coil.
The systemic use of thrombolytic/fibrinolytic agents is widely accepted in the treatment of catheterization-related thrombosis.11 Recanalization of the completely occluded PDA after fibrinolytic therapy has not been reported before. It is well known that the biological half-life of streptokinase is 82 to 184 minutes and the return of serum plasminogen to normal levels takes 24 hours.12 Thus, in the patient with recanalization it was hoped that complete occlusion of the PDA would be reestablished spontaneously with thrombus reformation on the coil several hours after the cessation of streptokinase treatment. Unfortunately, recanalization and mechanical hemolysis persisted, and the patient needed blood transfusion and another occlusion procedure.
Nykanen and colleagues3 found residual shunting in 33% of their patients 6 months after PDA occlusion with the Rashkind umbrella in children weighing under 10 kg. It may have been higher in the first days of implantation. Implanting another device to stop residual shunting is not possible for several days after the initial placement of a Rashkind umbrella. In contrast, coils have the advantage of multiple-coil implantation even in the same procedure if residual shunting is detected after the first implantation. Therefore, it was possible to achieve early complete occlusion in a large proportion of our patients.
Infective endocarditis prophylaxis should be continued for at least as long as residual shunting persists after transcatheter occlusion. However, the main problem en-countered in residual shunting is mechanical hemolysis.13 It is obvious that the risk of mechanical hemolysis would increase when the Rashkind umbrella is used since the incidence of residual shunting is high with this device. Since a second device cannot be implanted immediately when mechanical hemolysis occurs after the implantation of the Rashkind umbrella, surgical intervention or retrieval of the device by a complicated technique is needed.14 In this situation, the coil has the advantage of a second implantation to stop mechanical hemolysis, as was the case in our study. It is clear that treating mechanical hemolysis in small infants is much more difficult than in bigger children.
The incidence of LPA stenosis after Rashkind umbrella implantation is quite high according to the literature.1,2 Nykanen's group reported 9.9% in patients weighing under 10 kg.3 Gianturco coils have also been reported to cause LPA stenosis: Stromberg's group found LPA flow velocity above 2 m•sec-1 in 3% of their patients on Doppler echocardiography in the first days after occlusion.15 It was found high in a patient in our study, but detailed echocardiographic study revealed that it was not a true stenosis since the flow velocity was also high in the MPA. LPA flow velocity gradually increased during follow-up in a patient in Stromberg's study.15 We obtained similar findings in 2 patients. Implantation of 2 coils with at least a complete loop at the pulmonary site may be the reason for this phenomenon, although it was explained by incomplete endothelialization of the coil by Shim and associates.5
Stenosis of the DAo is another problem encountered in transcatheter PDA occlusion. Ottenkamp's group found protrusion of the Rashkind umbrella into the DAo in 76.5% of their patients, and turbulent flow on color Doppler echocardiography was seen in 41.5%.2 But none of them had flow velocity above 2 m•sec-1. Coil occlusion of the PDA carries a greater risk of DAo obstruction. Flow velocity in the DAo after catheterization was 2.1 m•sec-1 in a patient (4%) in Stromberg's study,15 but it normalized at intermediate follow-up with the endothelialization of the coil. On the other hand, in another of their patients and in the patient reported by Moore and coworkers,16 DAo flow velocity rose beyond 2 m•sec-1 at intermediate follow-up, although it was within normal limits in the postcatheterization period. This phenomenon was also explained by incomplete endothelialization of the coil. The changes in flow velocity when there is diastolic runoff on Doppler echocardiography have not been documented up to now. It can regress, as seen in a patient in our study, which can be explained by either endothelialization of the coil and/or the widening of the DAo with rapid weight gain following the elimination of heart failure after PDA occlusion. Post-occlusion angiography in our patient revealed that the ductal ampulla was too narrow and all the loops of the coils at the aortic site were protruding into the DAo.
In conclusion, our study shows that coil occlusion is feasible in the small child when the diameter of the PDA is less than 4.5 mm. It has advantages over the Rashkind umbrella when there is residual shunting or mechanical hemolysis. However, it is not completely without complications, as indicated by Stromberg's and Dalvi's groups, who found LPA obstruction in some of their patients after occlusion.15,17 It would be erroneous to assume that there is DAo or LPA stenosis based only on the flow velocities in these vessels. They should be compared with AAo and MPA flow velocities, especially when there is an additional heart defect. Furthermore, no more than half a loop of the coil should be left at the pulmonary site in order to prevent stenosis, especially when coils with large diameters are used. This can be more easily achieved with controlled-release coils and by using the venous route for implantation.
Our study shows that coil occlusion may cause DAo obstruction when there is no wide ductal ampulla for the aortic loops to fit into. Coils with fewer loops or other occlusion techniques should be used in such cases. Detailed 2-dimensional and color Doppler echocardiography should be performed in such cases even if peak flow velocity is within normal limits because if the device protrudes into the DAo (Figure 3), infective endocarditis prophylaxis is necessary.
View larger version (210K):
[in this window]
[in a new window]
|
Figure 3. (A) Doppler study showing normal flow velocity in the descending aorta. (B) Two-dimensional echocardiography showing protrusion of the coil into the descending aorta in the same patient.
|
|
|
REFERENCES
|
|---|
-
Fadley F, Al-Halees Z, Galal O, Kumar N, Wilson N. Left pulmonary artery stenosis after transcatheter occlusion of persistent arterial duct. Lancet
1993;341:559–60.[Medline]
-
Ottenkamp J, Hess J, Talsma MD, Buis-Liem TN. Protrusion of the device: a complication of catheter closure of patent ductus arteriosus. Br Heart J
1992;68:301–3.[Abstract/Free Full Text]
-
Nykanen DG, Hayes AM, Benson LN, Freedom RM. Transcatheter patent ductus arteriosus occlusion: application in the small child. J Am Coll Cardiol
1994; 23:1666–70.[Abstract]
-
Hijazi ZM, Geggel RL. Results of anterograde transcatheter closure of patent ductus arteriosus using single or multiple Gianturco coils. Am J Cardiol
1994;74:925–9.[Medline]
-
Shim D, Fedderly RT, Beekman RH III, Ludomirsky A, Young ML, Schork A, et al. Follow-up of coil occlusion of patent ductus arteriosus. J Am Coll Cardiol
1996;28: 207–11.[Abstract]
-
Lloyd TR, Fedderly R, Mendelsohn AM, Sandhu SK, Beekman RH III. Transcatheter occlusion of patent ductus arteriosus with Gianturco coils. Circulation
1993;88: 1412–20.[Abstract/Free Full Text]
-
Rosenthal E, Qureshi SA, Reidy J, Baker EJ, Tynan M. Evolving use of embolisation coils for occlusion of the arterial duct. Heart
1996;76:525–30.[Abstract/Free Full Text]
-
Gaynor JW, Sabiston DC. Patent ductus arteriosus, coarctation of the aorta, aortopulmonic window, anomalies of the aortic arch. In: Sabiston DC, Spencer FC, editors. Surgery of the chest. Philadelphia: WB Saunders, 1990: 1128–72.
-
Mullins CE, Pagotto L. Patent ductus arteriosus. In: Garson A, Bricker JT, Fisher DJ, Neish SR, editors. The science and practice of pediatric cardiology. 2nd ed. Baltimore: Williams & Wilkins, 1997:1181–98.
-
Anderson RH, Macartney FJ, Shinebourne EA, Tynan M. Cardiac catheterization and angiocardiography. In: Paediatric cardiology. Edinburgh: Churchill-Livingstone, 1987:363–93.
-
Aydog an Ü, Cantez T, Dindar A, Tanman B, Ertugrul T, Omeroglu R. Fibrinolytic therapy for femoral arterial thrombosis after cardiac catheterization in infants and children. J Invasive Cardiol
1992;4:445–7.
-
Chintagumpala MM, Steuber CP. Anticoagulant and thrombolytic agents. In: Garson A Jr, Bricker JT, Fisher DJ, Neish SR, editors. The science and practice of pediatric cardiology. 2nd ed. Baltimore: Williams & Wilkins, 1998:2541–52.
-
Ladusans EJ, Murdoch I, Franciosi J. Severe haemolysis after percutaneous closure of a ductus arteriosus (arterial duct). Br Heart J
1989;61:548–50.[Abstract/Free Full Text]
-
Grifka RG, O'Laughlin MP, Mullins CE. Late transcatheter removal of a Rashkind PDA occlusion device for persistent hemolysis using a modified transseptal sheath. Cathet Cardiovasc Diagn
1992;25:140–3.[Medline]
-
Stromberg D, Pignatelli R, Rosenthal GL, Ing FF. Does ductal occlusion with the Gianturco coil cause left pulmonary artery and/or descending aorta obstruction? Am J Cardiol
1999;83:1229–35.[Medline]
-
Moore JD, Shim D, Mendelsohn AM, Kimball TR. Coarctation of the aorta following coil occlusion of a patent ductus arteriosus. Cathet Cardiovasc Diagn
1998;43:60–2.[Medline]
-
Dalvi B, Nabar A, Goyal V, Naik A, Kulkarni H, Ramakanthan R. Transcatheter closure of patent ductus arteriosus in children weighing < 10 kg with Gianturco coils using the balloon occlusion technique. Cathet Cardiovasc Diagn
1998;44:303–8.[Medline]
TOP
ABSTRACT
INTRODUCTION
