Echocardiographic and Radionuclide Pulmonary Blood Flow Patterns After Transcatheter Closure of Patent Ductus Arteriosus
Background Transcatheter occlusion of patent ductus arteriosus (PDA) has been associated with protrusion of the occluder device into the left pulmonary artery (LPA). This study was conducted to evaluate the significance of occluder protrusion and its implications for potential obstruction of the LPA and associated decrease of left lung perfusion.
Methods and Results Fifty-two patients underwent successful transcatheter PDA occlusion over a period of 5 years. In this study, 49 were reexamined between March and June 1995. In addition to clinical and echocardiographic examination, lung scintigraphy was performed. Protrusion into the LPA was present in 5 of 49 patients (10%). In these patients, maximal flow velocity in the LPA was significantly (P<.01) increased. Decreased left lung perfusion, defined as <40% of total pulmonary blood flow, was found in 7 of 49 patients (14%). Although mean left pulmonary perfusion was significantly (P=.02) decreased in patients with protrusion, there was considerable overlap with patients without protrusion, and only a weak correlation was found (r=−.35, P=.01) between flow-velocity and left lung perfusion.
Conclusions Our results demonstrate that protrusion of the device in the LPA is an infrequent finding. If present, it is associated with increased maximal flow velocity in the LPA and diminished left lung perfusion. However, echocardiography and lung scintigraphy are weakly correlated: Increased maximal blood flow velocities in the proximal LPA proved to be a poor indicator of impaired left lung perfusion. Also, decreased perfusion occurs in the absence of echocardiographic evidence of device protrusion.
Nowadays, transcatheter occlusion of a patent ductus arteriosus (PDA) with the use of a Rashkind PDA occluder device is recognized as a safe and effective procedure for patients of any age, including smaller children.1 2 3 Although the current results are promising,4 5 6 7 an increasing number of potential sequelae have been reported since the technique was first used; among these sequelae are persistent residual shunting, device embolization, severe hemolysis,7 injury of the tricuspid and/or pulmonary valves,8 and impaired pulmonary blood flow distribution.9 10
To assess the importance of device-related effects on pulmonary blood flow distribution, we reviewed our first consecutive series of patients who underwent transcatheter duct occlusion and correlated right and left pulmonary blood flow velocities measured by Doppler echocardiography with the relative flow distributions measured by radionuclide lung scintigraphy.
Between November 1988 and December 1993, 58 elective PDA closure attempts were conducted at the Sophia Children's Hospital Rotterdam with the Rashkind PDA occluder device in a group of 56 patients (20 boys, 36 girls) who presented with a PDA (diagnosed by clinical, two-dimensional echocardiographic and Doppler flow examinations). Patient characteristics are shown in Table 1⇓.
Initial procedures were technically successful in 52 patients (93%). Two underwent a second procedure (without any problems) because of significant residual shunting (7 months after the first procedure in one patient and 26 months after the first procedure in the other).
All patients had isolated PDA. Three patients had trisomy 21, 1 patient had congenital rubella syndrome but without peripheral pulmonary artery stenosis, and 1 patient had Hirschsprung's disease. Except for one patient who had severe congenital kyphoscoliosis, none had thoracic cage abnormalities.
None of the patients had differences between the flow velocities in the pulmonary arteries of >0.2 m/s before implantation of the device.
The procedure was performed with the patient under general anesthesia and endotracheal intubation. The Rashkind PDA occluder device and the technique of occlusion have been described previously.1 2 6 7
Posteroanterior and lateral chest radiographs and complete multimodal echocardiographic studies were performed 1 day after the procedure.
All 52 patients were periodically followed after the procedure on an outpatient basis and underwent clinical and echocardiographic examinations at each visit.
All echocardiographic examinations were performed from apical, subcostal, parasternal, and suprasternal views (multiple axis), including the high parasternal short-axis view (so-called ductal cut), with particular attention to the descending aorta, pulmonary arteries, ductus arteriosus anatomy, and eventual residual shunting and position of the occluder device.
In all cases, adequate visualization of the occluder device for evaluation of residual shunting was achieved. To evaluate protrusion of the device into the left pulmonary artery (LPA), the position of the legs of the device in relation to the arterial wall was assessed by two-dimensional echocardiography. If one or more legs were not “attached” to the wall and protruding into the vessel lumen, the device position was judged to be protruding. However, quantification of this phenomenon was not possible.
Between March 1995 and June 1995, 49 of 52 patients (94%) were reexamined in a follow-up visit, including clinical examination and complete transthoracic echocardiographic examination. All examinations were performed by the same investigator without knowledge of preliminary data nor the outcome of radionuclide lung scintigraphy. Using pulsed- and continuous-wave Doppler flow mapping, in combination with Doppler color flow mapping, we determined the maximal flow velocities in the proximal right and left pulmonary arteries. For this follow-up study, informed parental consent was obtained.
Methodology of Lung Scintigraphy
Lung scintigraphy was performed with the use of 99mTc-labeled human albumin macroaggregates (SOLEOrMAA) and a gamma camera (Siemens [FOV]). After intravenous injection the 99mTc-labeled macroaggregate particles (5 to 20 μm) are homogeneously distributed in the lungs through trapping of particles in the precapillary pulmonary arterioles during their first pass, enabling the study of pulmonary radioactivity distributions that reflect the pulmonary blood flow.
In children (ranging in age from young infants to 4 years), approximately 30 000 particles labeled with 18.5 MB2 99mTc were administered.
The adult dose (100 000 to 150 000 particles labeled with 75 MB2 99mTc) was used in patients ≥18 years of age.
The number of particles and the amount of radioactivity administered were adjusted to the age of the patient. The lung scintigrams were obtained from anterior and posterior directions by collecting 400 000 counts per view. The percentage of the perfusion of the left lung was calculated by dividing the mean radioactivity (counts) of the left lung measured from anterior and posterior views (by the mean counts of both views) divided by 2.
Data are expressed as mean±SD and range. Comparison of continuous data between groups was performed with the Student's t test or Mann-Whitney U test when population distribution was not normal. Categorical data were compared with the use of Fisher's exact test. Correlations between continuous variables were evaluated by means of the Pearson product-moment correlation coefficient. The prevalence of residual ductal shunting was evaluated with the use of Kaplan-Meier actuarial analysis. A value of P<.05 was considered significant.
Of the 3 patients who did not participate in the follow-up study, the last available data were collected after 5 months for 1 patient, 2.4 years for 1, and 2.5 years of follow-up for 1.
Mean follow-up duration time was 3.6±1.5 years (range, 1.6 to 6.5 years), the mean age was 8.1±5.5 years (range, 1.3 to 24.5 years), the mean weight was 30.3±16.9 kg (range, 14.0 to 80.0 kg), and the mean length was 126.0±22.1 cm (range, 95.0 to 180.0 cm) at the last visit.
Prevalence of Residual Shunting
With the use of actuarial analysis, the prevalence of residual shunting in our population at different follow-up moments was estimated; this showed a decrease from 65% directly after the procedure to 37% at 1 year, 18% at 2 years, and 10% at 3 years.
The characteristics of our population were compared in relation to the size of the occluder device. Table 1⇑ shows that the two populations do not differ except for the size of the shunt, which was greater, as expected, in patients in whom a 17-mm device was implanted (patients with two occluders were excluded).
In addition, residual shunting directly after the procedure was observed more frequently in patients who received a 17-mm device, although this difference did not reach statistical significance. However, at the last follow-up, the prevalence of residual shunting was equal in both patient groups irrespective of the size of the occluder device used. On the basis of the length and velocity of the Doppler flow signal in the pulmonary artery, all residual shunts were judged to be trivial at the time of the follow-up measurements.
Protrusion of the Device Into the LPA and Maximal Flow Velocities
At the last follow-up visit, protrusion of the device into the LPA lumen was present in 5 of the 49 (10%) patients who received one occluder, of whom 2 had a 12-mm occluder (n=29 patients, 7%) and 3 a 17-mm occluder (n=20 patients, 15%).
One patient who had a second occluder had protrusion of the device into the LPA. The maximal flow velocity in the LPA was significantly increased in the presence of protrusion of the device (Table 2⇓). The characteristics of the patients with and without a protrusion into the LPA did not differ.
Correlation Between Protrusion in the LPA and the Results of Lung Scintigraphy
Except for the patient with severe kyphoscoliosis, all underwent lung scintigraphy. Seven patients (14%) proved to have decreased perfusion of the left lung (<40% of total pulmonary blood flow), of whom 2 had protrusion of the device into the LPA. We observed a relative decrease of the left lung perfusion in association with a protrusion of the device into the LPA as assessed by flow velocity measurement (Table 2⇑). However, in terms of absolute values of lung perfusion, there was an important overlap between the two populations (patients with and without a protrusion). Of the 5 patients who had a residual shunt, none had decreased left lung perfusion.
The flow velocities and the perfusion of the left lung were only weakly correlated (r=−.35, P=.01). There was no difference between the patients with a 12-mm device and those with a 17-mm device.
In our population, the rate of successful ductal occlusion with the use of the Rashkind PDA occluder device was ≈90%, which is similar to that reported by others.4 7 We also observed a substantial decrease of residual shunting during the first years after implantation. Although residual shunting directly after the implantation was more frequent in patients with a 17-mm device than in those with a 12-mm device, this difference was not present in the longer term.7 This might be related to the initial size and shape of the ductus arteriosus.
Protrusion of a device into the LPA was first reported by Dyck et al2 in 1988 and was associated with an estimated 10-mm Hg Doppler gradient. In 1989, Musewe et al11 suggested a tendency of the proximal arms of the device to straddle the LPA orifice, inducing a potential stenosis of the proximal LPA. Later, other investigators9 10 reported on this complication, both in the LPA and descending aorta. We conducted this study to evaluate the significance of occluder protrusion and its implications for potential obstruction of the proximal LPA with particular reference to impaired pulmonary blood flow distribution.
We observed a rate of device protrusion (10%) into the proximal LPA comparable to that reported by others.9 Protrusion was more frequently observed in patients with a 17-mm device (15%) than in those with a 12-mm device (7%), but this difference did not reach statistical significance. Moreover, despite a relatively limited number of patients, we observed that an increase in maximal blood flow velocity in the proximal LPA (assessed by Doppler echocardiography) was associated with device protrusion. However, two considerations should be made: Assessment of protrusion on two-dimensional echocardiography is difficult because of the occurrence of reverberations by the device. Furthermore, the degree of protrusion is a continuum rather than a “yes or no” phenomenon.
In our study, we found a relative decrease of the left lung perfusion in association with protrusion of the device. However, there are no hard data on radionuclide lung perfusion ratios regarding normal ranges for children. In our own laboratory, <40% flow to the left lung is judged to be abnormal on the basis of findings in children with ostensibly normal hearts. Therefore, when an abnormal left lung perfusion scan is defined to be <40%, a significant discrepancy between the presence of protrusion of the device and decreased left lung perfusion was noted. We also observed a poor correlation between the maximal LPA blood flow velocities and the left lung perfusion scan values. This means that protrusion of the device, resulting in a higher flow velocity in the LPA than that in the right pulmonary artery, is not necessarily associated with impaired perfusion of the left lung.
It could be argued that residual shunting may have diluted the radioactivity in the left lung by nontagged blood selectively streaming into the left lung. However, since all residual shunts in our study were trivial and none of the patients with a residual shunt had decreased left lung perfusion, this is not significant.
We did not identify specific morphological ductal patterns that predisposed for these sequelae. However, it has been put forward that both the shape and the geographical positioning of the device might play a role in the occurrence of both protrusion and residual shunting.12
Protrusion of the device into the LPA is an infrequent finding in the longer term. However, when present, it is frequently associated with an increase of the maximal flow velocity in the LPA (assessed by Doppler echocardiography). Yet increased maximal blood flow velocities in the proximal LPA proved to be a poor indicator of stenosis leading to impaired left lung perfusion. Furthermore, impaired perfusion of the left lung also may occur in the absence of device protrusion, at least as can be assessed by echocardiography.
The association between transcatheter occlusion of PDA by means of the Rashkind occluder device and substantial impaired left lung perfusion is not of great importance in the shorter term. However, longer follow-up is indicated to assess the final clinical relevance and the final position of this device in the management of PDA. In addition, studies on pulmonary blood flow patterns in patients who underwent surgical ductal closure also should be initiated for realistic comparisons. Since there are no reliable echocardiographic predictors to assess impaired left lung perfusion, routine assessment of radionuclide lung perfusion ratios in patients after transcatheter closure of PDA should be considered.
- Received March 27, 1996.
- Revision received May 15, 1996.
- Accepted May 17, 1996.
- Copyright © 1996 by American Heart Association
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.
Gatzoulis MA, Rigby ML, Redington AN. Umbrella occlusion of persistent arterial duct in children under two years. Br Heart J.. 1994;72:364-367.
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.
Latson LA. Residual shunts after transplantation closure of patient ductus arteriosus: a major concern or benign ‘technomalady’? Circulation.. 1991;84:2591-2593.
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;68:301-303.
Musewe NN, Benson LN, Smallhorn JF, Freedom RM. Two-dimensional echocardiographic and color flow Doppler evaluation of ductal occlusion with the Rashkind prosthesis. Circulation.. 1989;80:1706-1710.
Talsma M, Benitez O, Witsenburg M, van Oort A, Hess J, Rohmer J. Anatomical characteristics of arterial ducts with regard to successful transcatheter occlusion. Cardiology in the Young. 1994;4(suppl I):13. Abstract.