Follow-up after coil closure of patent ductus arteriosus

T occlusion of the patent ductus arteriosus (PDA) with Gianturco coils has emerged as a promising technique because of the small size of the delivery catheter and lower cost without compromising safety and efficacy.1–3 Modification in coil design, route of delivery, and technique have been described.4,5 Whereas several reports describe the acute results of PDA closure with Gianturco coils, there is limited follow-up data on the intermediateand longterm prevalence and course of residual shunts after coil placement.6,7 We report our intermediate-term follow-up of 84 patients who were subjected to transcatheter closure of the PDA with Gianturco coils using the temporary balloon occlusion technique.8 • • • All patients seen in the pediatric cardiology clinic after April 1995 with PDA were eligible for transcatheter closure. The diagnosis was confirmed on color Doppler echocardiography and duct size estimated in the standard ductal view on color flow mapping. The Gianturco coils were deployed using standard retrograde transarterial technique assisted by Swan-Ganz catheter inflation to prevent embolization as has been reported earlier.8 In the initial 15 patients, only 1 coil was deployed because of concern of coil dislodgment during the delivery of the second coil. However, subsequently, 1 or multiple coils were used to achieve complete angiographic closure. All patients underwent clinical examination, electrocardiography, and color Doppler echocardiographic evaluation at 24 hours and at 3, 6, 12, 18, and 24 months after the procedure. Left atrial, aortic, and left ventricular dimensions and flow velocities across branch pulmonary arteries and aortic arch were serially assessed at each follow-up. Clinical and laboratory evidence of hemolysis (hemoglobin, reticulocyte count, serum bilirubin, and peripheral smear for fragmented red cells) was looked for immediately after the procedure and at days 1 and 3 and after 1 and 3 months following the procedure. Residual shunts on color Doppler were graded as grade 0 5 no shunt, grade I 5 single jet with noncontinuous flow, grade II 5 single jet with continuous flow, and grade III 5 multiple jets with continuous flow.8 Cine angiography grading of residual shunts was done as per the classification used by the International PDA Registry. Complete closure of the PDA was considered if no residual shunting was noted on aortogram or at color Doppler. Spontaneous closure of the residual shunt was diagnosed as absence of the shunt on color Doppler in a patient (without further interventions) with previously documented residual shunt. Recanalization was diagnosed when persistent reappearance of the shunt across the PDA was detected in patients with previously documented complete occlusion. Transient reopening was defined as recurrence of ductal flow after complete closure, which was followed by spontaneous reclosure. The procedure was repeated at 18 months in patients with a grade II residual shunt or earlier in patients with clinically significant residual shunt. The paired t, chi-square, and Fischer’s exact test were used for comparing before and after procedure variables. The cumulative residual ductal shunting was evaluated using a Kaplan-Meier product-limit analysis technique. • • • Eighty-four patients (58 females and 26 males) with median age 4 years (range 1 month to 45 years) and median weight 14 kg (range 2.3 to 38) underwent Gianturco coil embolization of the PDA. Seventy-five of 84 patients were symptomatic with repeated lower respiratory tract infection, shortness of breath, or failure to thrive. Two patients had recanalization after surgical ligation of PDA and 1 patient had residual shunt after attempted buttoned device closure.9 There were 72 patients with type A, 6 with type B, 4 with type C and 1 with type D and E duct.10 The mean angiographic duct diameter was 2.8 (SD 0.87 mm), and 9 had duct diameters .4 mm. Semiquantitative evaluation of the shunt on the aortogram before the procedure revealed 5 patients with small shunts, 32 with moderate, and 47 with large shunts. Sixty patients had some degree of pulmonary hypertension (median pulmonary artery diastolic pressure, 19 mm Hg, range 9 to 50 mm Hg). Baseline left atrium to aorta ratio was 1.5 6 0.3, which was reduced to 1.2 6 0.3 (p 5 0.03) after occlusion of the ductus. The differences in blood flow velocities in left (before, 1.3 6 0.86 m/s; after, 1.22 6 0.7 m/s) and right (before, 1.04 6 0.34 m/s; after, 1.1 6 0.3 m/s) pulmonary arteries, and descending thoracic aorta (before, 1.4 6 1.15 m/s; after, 1.2 6 1.2 m/s) were statistically insignificant (p .0.05). Occluding spring coils (Cook, Bloomington, Indiana) of various sizes were used (MWCE-38-8-8, MCWE-38-5-5, and occasionally MCWE-38-12-10). The first coil was selected to have a helical diameter at least twice the minimum angiographic diameter, and its length was such that it formed at least 3 loops. Subsequent coil size was based on the degree of residual shunt. Most of our patients required .1 coil From the Department of Cardiology, King Edward VII Memorial Hospital, Parel, Mumbai, India. Dr. Dalvi’s address is: Department of Cardiology, King Edward VII Memorial Hospital, Parel, Mumbai 400 012, India. Manuscript received March 24, 1998; revised manuscript received and accepted August 26, 1998.