This study examines whether transcatheter closure of a patent ductus arteriosus (PDA) using a Rashkind PDA occluder device is safe and effective in adults, or if adults have complications not sited in children owing to prolonged aorticopulmonary communication, high surgical risks or calcified PDAs. Fifteen patients aged 22 to 76 years (mean 42 [+ or -] 14) were referred for transcatheter PDA occlusion. Exercise intolerance was the most frequent clinical manifestation. Eleven of 1S patients had surgical risk factors that included left ventricular failure (n = 10), biventricular failure (n = 1), elevated pulmonary pressures (n = 1), and a calcified PDA (n = 5). Twelve millimeter devices were placed in 4 PDAs [less than or equal to]3 mm in diameter; 17 mm devices were placed in 11 PDAs 3 to 6 mm in diameter. Seven (47%) were occluded angiographically shortly after device placement; another 5 PDAs (33%) were occluded echocardiographically within 24 hours of the procedure. Completed occlusion in this time interval was more likely to occur in PDAs Although surgical correction of a patent ductus arteriosus (PDA) is a low risk procedure in children,[1-3] the risks of this procedure theoretically may be increased for certain adults who have prolonged shunting through a PDA resulting in left ventricular failure or pulmonary vascular disease. In particular, calcification of the PDA will make it too friable for standard clamping, necessitating the occasional use of a more invasive bypass procedure.[4] Anesthesia risks may also be increased significantly because of other diseases, in adults not related to the presence of the PDA. An alternative to surgical closure of a PDA is transcatheter occlusion using a Rashkind PDA occlusion system (USCI, Inc., Billeria, Massachusetts). Multiple studies have reported the safety and efficacy of this procedure in children.[5-9] Only 1 study to date elaborates on the use of this technique in an elderly man.[10] This investigation examines whether transcatheter occlusion of a PDA in adults is as safe and effective as in children, or if adults have complications not found in children. METHODS Patients: From July 1981 through May 1991, 15 of 233 patients (11 women and 4 men) referred to Texas Children's Hospital for transcatheter closure of a PDA were aged >21 years (range 22 to 76, mean 42 [+ or -] 14). Clinical history, physical examinations and investigational studies were reviewed retrospectively. Patients were placed into New York Heart Association classification on the basis of symptoms. Catheterization and occlusion: During each cardiac catheterization, the presence of a PDA was confirmed by the prograde passage of a 'marker' angiocardiographic catheter through the PDA. This was followed by a selective cineangiogram in the descending aorta just distal to the PDA. Pulmonary to systemic flow ratios were determined for each patient before PDA closure using saturation data obtained during catheterization and an assumed basal oxygen consumption. Pulmonary arteriolar resistance was calculated in each case by subtracting the value of mean left atrial pressure from the mean pulmonary artery pressure, dividing by the pulmonary blood flow, and then correcting for body surface area. Delivery of each device was accomplished under mild sedation following the methods described by Rashkind et al.[5] All patients received a double-disk occluder with the knuckle-and-eye attach-release mechanism. An aortogram with injection into the descending aorta was performed approximately 20 minutes after occluder device implantation to check for residual PDA flow. All patients with residual shunts in the catheterization laboratory were reassessed for PDA flow within 24 hours of device placement by echo/doppler studies. Two of the 7 patients with PDA occlusion angiographically did not have echocardiographic assessment, but did have complete ductal occlusion by physical examination. Complications: Complications of transcatheter PDA occluder placement reported in the past and defined as such in our study included postrelease embolization of a device, occluder-related endarteritis or hemolysis, and encroachment of the occluder arms into the left pulmonary artery leading to stenosis.[6] Other potential hazards for PDA closure specifically in adults were evaluated and included rupture of calcified ductus, hemodynamic compromise in patients with documented left ventricular failure or elevated pulmonary resistance, and development of malignant systolic hypertension.[11,12] Follow-up: Late follow-up data were obtained by mailed questionnaire or telephone communication with the referring physician. All patients with residual shunts within 24 hours of PDA occluder device placement had echocardiograms to assess flow in late follow-up. The 2 patients with PDAs occluded angiographically who did not have echocardiograms within 24 hours of device placement did have color/doppler investigation in late follow-up. Any change in the patient's symptoms including exercise tolerance, angina and arrhythmias were also assessed. Data analysis: PDA diameter versus successful PDA closure within 24 hours of device placement was evaluated by 1 factor analysis of variance. Chi-square analysis was used to determine significance between PDA calcification and complete PDA occlusion within 24 hours of the procedure as well as in late follow-up. A p value RESULTS Patients description at time of referral for patent ductus arteriosus occlusion (Table I): A continuous murmur located over the second left intercostal space was present in every patient except for the patient in New York Heart Association functional class IV, whose murmur was isolated to systole. Intensity of the murmurs generally ranged from grade II/IV to grade III/VI. Eleven of 15 patients exhibited some degree of exercise intolerance. The 2 patients with angina had previous cardiac catheterizations. One was found to have significant obstruction of the left anterior descending coronary artery. The other patient had complete occlusion of the first diagonal coronary artery and minimal diffuse disease distally. Two patients with a history of palpitations had atrial arrhythmias confirmed by Holter monitoring. Reasons for referral for transcatheter closure: Three patients (nos. 10, 12 and 15 [Table I]) had contraindications to general anesthesia that prompted their referring physicians to choose transcatheter PDA closure over surgery. The remaining 12 patients were referred by their physician based on personal patient preference. Most patients had surgical risks as well, having ventricular failure by symptoms or calcified PDAs. Surgical risk factors included left ventricular failure (n = 10), biventricular failure (n = 1), elevated pulmonary resistance (n = 1), a calcified PDA (n = 5), chronic obstructive pulmonary disease (n = 1), atrial arrhythmias (n = 2) and renal disease (n = 1). Catheterization and occlusion data (Table I): Five of 15 PDAs were noted to be calcified at the time of catheterization. All patients had left to right shunting through their PDAs during their catheterizations. Patients in New York Heart Association functional class 1, II, III and IV had mean pulmonary to systemic flow ratios of 1.65 [+ or -] 0.66, 1.51 [+ or -] 0.23, 2.32 [+ or -] 1.17, and 1.5, respectively. The mean pulmonary arteriolar resistance in U/[m.sup.2]) Was 1.5 [+ or -] 0.77 for class 1, 1.24 [+ or -] 0.39 for class II, 1.7 [+ or -] 0.55 for class III, and 6.6 for class IV. Seven of the 15 PDAs, or 47%, were occluded angiographically shortly after device implantation. Five additional PDAs, or 33%, were noted to be occluded within 24 hours by 2-dimensional echo/doppler imaging. Once there was documentation that these PDAs were occluded, none reopened by clinical or echocardiographic examination. Three of the 15 PDAs, or 20%, had residual leaks throughout a mean follow-up time of 23 months. All of these PDAs were 5 to 6 mm in diameter before device placement. Two of these 3 patients had short systolic murmurs immediately after device placement. Both patients had no murmur when trivial shunting was noted by color/doppler evaluation in late follow-up. The third patient had persistence of his continuous murmur immediately after device implant. He underwent placement of a second device 2 years later, with complete occlusion immediately. Successful occlusion was obtained within 24 hours of the delivery of a Rashkind PDA occluder in PDAs, with mean internal diameters of 3.6 [+ or -] 0.6 mm; the 3 PDAs not completely closed had a mean diameter of 5.3 [+ or -] 0.3 mm. A significant inverse relation was found between ductal diameter and successful occlusion within 24 hours of device placement (p = 0.0009). The presence of PDA calcification was not significantly correlated with complete PDA occlusion within 24 hours of the procedure (p = 0.09) or in late follow-up (p = 0.04). Figure 1 illustrates angiographically, successful occlusion of a calcified 4 mm PDA 20 minutes after device placement. Complication data: No complications related to device implantation or PDA closure were noted in any patient. The 3 patients with residual PDAs did not have hemolysis or endarteritis throughout their total follow-up interval of 5.9 patient years. Follow-up data by symptoms: The time interval between device implant and late follow-up ranged from 1 month to 38 months (mean 11.5 [+ or -] 9.6 months). Figure 2 shows the outcome in New York Heart Association functional class by late follow-up for 14 of the 15 patients who could be evaluated. All patients with exercise intolerance improved, including those with residual PDAs. One patient was lost to follow-up by her physician. This patient was in New York Heart Association functional class IV before device placement and demonstrated improved exercise tolerance in the hospital before her discharge 48 hours after her procedure. Both patients with documented coronary artery disease and angina had relief of their exertional chest pain after complete ductal occlusion. Only 1 of the 2 patients had resolutions of his arrhythmias by symptoms, with documentation by Holter monitoring. The other patient with persistent atrial fibrillation/flutter was reported to have significant improvement of her left ventricular failure. In fact, she was feeling so well that she began swimming again and became noncompliant with her antiarrhythmic medication. This patient died suddenly of unknown causes 3 years after device implantation at 79 years of age. DISCUSSION Closure of an audible PDA during childhood and adolescence is recommended because closure reduces the risk of infective endarteritis, left ventricular failure, pulmonary vascular disease and thus, ultimately, premature death.[11] However, a PDA may escape clinical detection and persist into adulthood, with rare spontaneous closure after age 40 years.[13] If adults do not develop significant pulmonary vascular disease, they tend to have an unpredictable progression of left ventricular failure.[12] Current publications mitigate against complacency regarding PDA closure, even when an asymptomatic PDA is discovered, because patients with cardiomegaly at the time of surgical closure have a worse prognosis than patients with a normal heart size.[14] Most of our patients were referred for transcatheter rather than surgical closure of their PDAs on the basis of patient preference, although both procedures were offered to them by their physicians. The lack of 100% PDA occlusion by transcatheter technique is probably one reason for hesitancy in referral.[7-9] The major limitation of obtaining successful PDA occlusion in this study and others using a Rashkind PDA occluder device is the PDA diameter. All PDA with residual leaks in this investigation were 5 to 6 mm at their narrowest internal diameter. Only 1 patient in our study had significant residual shunting with a continuous murmur after attempted transcatheter occlusion and needed placement of another device. The other 2 patients with residual flow had enough reduction of shunting immediately after attempted occlusion to decrease their murmurs to systole only and relieve them of their left ventricular dysfunction. Both of these patients have demonstrated gradual closure of their residual leaks in follow-up, with loss of their ductal murmurs and trivial leaks detected by color/doppler flow alone. Further evaluation is necessary to determine if complete occlusion is eventually attained. A question raised with the presence of residual PDA flow after occluder device placement is whether there is an increased risk for bacterial endarteritis. No studies to date have found any endarteritis in PDAs with trivial leaks detected only by color/doppler now.[15,16] It is prudent, however, to continue antibiotic prophylaxis with these patients until more is known about the possible risks. Another speculated complication of the transcatheter procedure is PDA rupture during catheter and occluder manipulation. This was, in particular, a concern for the 5 patients in our study with calcified ductus. None of the 15 patients had any evidence of vessel leak by angiograms in the catheterization laboratory or by echo/doppler studies on the day after their procedure. There was no unexplained loss of blood or development of a pleural effusion in any patients. In fact, the rim of PDA calcification often simplified a catheterization procedure by obviating repeated contrast injections to localize the PDA. The use of transcatheter rather than surgical PDA closure eliminated the need for a long convalescence as seen after an open thoracotomy. Additionally, all patients in this study who had exercise intolerance exhibited subjective clinical improvement, usually within 1 day of their occluder device placement. Anginal relief occurred in the 2 patients with documented coronary artery disease. This relief may have been related to improved coronary artery filling with increased diastolic pressures after PDA closure or a decreased left ventricular work load. Only 1 of the 2 patients with atrial arrhythmias had complete resolution. It is known, however, that atrial arrhythmias can continue after surgical PDA closure and are believed to be related to long-standing left atrial enlargement.[17] The patient in our study who had protracted atrial arrhythmias, nevertheless, was reported to have significant improvement in left ventricular function. Currently, the Rashkind PDA occluder device lacks approval of the Food and Drug Administration and has limited availability until the Food and Drug Administration review is completed. With the safety and efficacy of transcatheter occlusion demonstrated in this study, it is hoped that the device will soon be unrestricted to allow its use for PDA occlusion in patients with personal or physician opposition to surgical intervention. [1.] Panagopoulos PH, Tatooles CJ, Aberdeen E, Waterston DJ, Carter RE. Patent ductus arteriosus in infants and children. Thorax 1971;26:137-143. [2.] Cotten RB, Stahlman MT, Bender MD, Graham TP, Catterton WZ, Kovar I. Randomized trial of early closure of symptomatic patent ductus arteriosus in small preterm infants. J Pediatr 1978;98:647-651. [3.] Wagner HR, Ellison C, Zierler S, Lang P, Purohit DM, Behrendt D, Waldhausen JA. Surgical closure of patent ductus arteriosus in 268 preterm infants. J Thorac Cardiovasc Surg 1984;87:870-875. [4.] Johnson AM, Kron IL. Closure of the calcified patent ductus in the elderly: avoidance of ductal clamps and shunts. Ann Thorac Surg 1988;45:572-573. [5.] Rashkind WJ, Mullins CE, Hellenbrand WE, Tait MA. Nonsurgical closure of patent ductus arteriosus: clinical application of the Rashkind PDA occluder system. Circulation 1987;75:583-592. [6.] Dyck JD, Benson LN, Smallhorn JF, McLaughlin PR, Freedom RM, Rowe RD. Catheter occlusion of persistently patent ductus arteriosus. Am J Cardiol 1988;62: 1089-1092. [7.] Latson LA, Hofshire PJ, Kugler JD, Cheatham JP, Gumbiner CH, Danford DA. Transcatheter closure of patent ductus arteriosus in pediatric patients. J Pediatr 1989;115:549-553. [8.] Hosking MCK, Benson LN, Musewe N, Dyck JD, Freedom RM. Transcatheter occlusion of the persistently patent ductus arteriosus: forty-month follow-up and prevalence of residual shunting. Circulation 1991;84:2313-2317. [9.] Ali Khan MA, Mullins CE, Nihill MR, Al Yousef S, Al Oufy S, Abdullah M, Al Fagih MR, Sawyer W. Percutaneous catheter closure of the ductus arteriosus in children and young adults. Am J Cardiol 1989;64:218-221. [10.] Vita JA, Bittl JA, Selwyn AP, Lock JE. Transcatheter closure of a calcified patent ductus arteriosus in an elderly man. J Am Coll Cardiol 1988;12:1382-1385. [11.] Morgan JM, Gray H, Miller GAH, Oldershaw PJ. The clinical features, management and outcome of persistence of the arterial duct presenting in adult life. Int J Cardiol 1990;27:193-199. [12.] Marquis RM, Miller HC, McCormack RJM, Matthews MB, Kitchin AH. Persistence of ductus arteriosus with left to right shunt in the older patient. Br Heart J 1982;48:469-484. [13.] Campbell M. Natural history of persistent ductus arteriosus. Br Heart J 1968; 30:4-13. [14.] Fisher RG, Moodie DS, Sterba R, Gill CC. Patent ductus arteriosus in adults -- long-term follow-up: nonsurgical versus surgical treatment. Jam Coll Cardiol 1986; 8:280-284. [15.] Tunick PA, Kronzon I. Diagnosis of patent ductus arteriosus by serendipity in the adult. J Am Soc Echo 1988;1:446 449. [16.] Sorensen KE, Kristensen BO, Hansen OK. Frequency of occurrence of residual ductal flow after surgical ligation by color-flow mapping. Am J Cardiol 1991; 67:653-654. [17.] Trippestad A, Efskind L. Patent ductus arteriosus. Surgical treatment of 686 patients. Scand J Cardiovasc Surg 1972;6:38-42. From the Lillie Frank Abercrombie Section of Pediatric Cardiology, Texas Children's Hospital and Department of Internal Medicine, Section of Cardiology, Baylor College of Medicine, Houston, Texas. Manuscript received December 14, 1992; revised manuscript received April 27, 1993, and accepted April 28. Address for reprints: Margaret H. Schenck, MD, Pediatric Cardiology, Texas Children's Hospital, 6621 Fannin, Houston, Texas 77030.