E60
Occult Progressive Pulmonary Arterial Occlusion Associated with Right Ventricular Hypertension in Patients with Systemic Arteriopathy Brian H. Morray, MD, Lisa Bergersen, MD, MPH, James E. Lock, MD, and Audrey C. Marshall, MD Department of Cardiology, Boston Children’s Hospital, Boston, Mass, USA ABSTRACT
Objectives. Peripheral pulmonary artery stenosis (PPS) is a heterogeneous disease associated with numerous clinical or syndromic disorders including Williams syndrome. Previous studies focusing on patients with Williams syndrome have documented spontaneous improvement in branch pulmonary artery (PA) stenosis and right ventricular pressure. The goal of this study is to identify angiographic evidence of spontaneous occlusion or loss of segmental PAs in patients with PPS. Methods. Patients undergoing serial catheterization for PPS at Boston Children’s Hospital between 1999 and 2014 were selected. Patients with structural heart disease such as tetralogy of Fallot, truncus arteriosus, transposition of the great arteries, or single ventricle were excluded. Hemodynamic data were extracted from catheterization reports, all available angiograms were reviewed, and evidence of segmental PA loss was documented based on defined angiographic criteria. Results. Forty-two patients were included in the study. Each patient underwent an average of 5 catheterizations over the 15-year study period. The average RV pressure at initial catheterization was near systemic levels. Evidence of segmental PA loss was present in 40% of patients. Almost 30% of patients with vessel loss demonstrated loss at their initial catheterization, and 65% of patients experienced loss of multiple vessels over time. Patients with vessel loss tended to be older at initial catheterization and have higher average RV pressures at their most recent catheterization than those without vessel loss. Conclusions. Progressive segmental PA loss in patients with systemic arteriopathy and PPS has not been well described previously. In this cohort, segmental PA loss occurred in 40% of patients, resulting in persistently elevated PA pressures relative to patients who did not develop vessel loss. These findings highlight the importance of surveillance in this population of patients with systemic arteriopathy and severe PPS. Key Words. Peripheral Pulmonary Artery Stenosis; Balloon Pulmonary Artery Angioplasty; Systemic Arteriopathy
Introduction
P
eripheral pulmonary artery stenosis (PPS) occurs in approximately 2–3% of patients with congenital heart disease.1 It is a heterogeneous disease that can extend from the mediastinal pulmonary arteries out into more distal, posthilar vessels and can occur in association with congenital heart disease, several well-described syndromes (Williams, Alagille, Noonan, and congenital rubella syndromes) or in isolation.2 PPS is the
Congenit Heart Dis. 2015;10:E60–E67
second most common cardiovascular disorder in Williams syndrome and the most common cardiovascular finding in Alagille syndrome. The literature describing the natural history of PPS has largely focused on Williams syndrome. These data suggest that there is spontaneous improvement in right ventricular (RV) pressure and angiographic appearance of PPS over time without intervention.3–9 However, some of these series document a small subset of patients with worsening RV hypertension but provide little © 2014 Wiley Periodicals, Inc.
E61
Segmental Occlusion in PPS angiographic evidence of progressive PPS. We believe there is a subset of patients with PPS who develop progressive, segmental pulmonary artery (PA) vascular occlusions. Loss of PA crosssectional area can lead to persistent RV hypertension and hypertensive vascular changes in patent segments. The goal of catheter-based interventions is to prevent vessel loss and distribute blood flow throughout the vascular tree to prevent chronic pulmonary arterial hypertension. Maintaining patency of segmental PA branches may also improve ventilation–perfusion matching and gas exchange, as has been described in patients with tetralogy of Fallot and PPS following balloon pulmonary angioplasty.10,11 The goal of this study is to describe the previously undocumented angiographic characteristics of progressive segmental PA occlusion in patients with syndromic and nonsyndromic arteriopathy and PPS. Methods
Patients and Study Protocol Patients were selected by searching the clinical databases in the Department of Cardiology at Boston Children’s Hospital (BCH). The databases were queried for patients coded with a diagnosis of PA stenosis or PPS. Additionally, patients with syndromes with known association with PPS were selected. These included the following: (1) Williams syndrome diagnosed genetically (based on fluorescence in situ hybridization) or based on clinical characteristics; (2) non-Williams familial arteriopathy (at least 1 first-degree relative with supravalvar aortic stenosis or PPS and no syndromic diagnosis); (3) Alagille syndrome; (4) nonsyndromic, nonfamilial arteriopathy (presence of PPS with no family history or syndromic diagnosis); and (5) moyamoya disease (based on the presence of characteristic clinical history and neuroimaging findings). Patients were included in the study if they had PPS, underwent catheterization at BCH between January 1999 and January 2014, and subsequently underwent ≥3 catheterizations for evaluation and dilation of the branch pulmonary arteries. Patients with structural heart disease associated with PPS, such as tetralogy of Fallot, truncus arteriosus, transposition of the great arteries, single-ventricle disease, or discontinuous pulmonary arteries, were excluded. Patients with other congenital heart disease, such as isolated atrial or ventricular septal defects, as well as systemic arterial defects such as
supravalvar aortic stenosis or coarctation of the aorta, were included. Patients were included even if they underwent initial intervention at another institution or underwent surgical repair of an associated minor defect (i.e., ventricular septal defect, patent ductus arteriosus). The Children’s Hospital Committee on Clinical Investigation approved the retrospective study.
Catheterization Data Serial catheterization data were reviewed for each patient by a single investigator (BM). RV and descending aortic pressures were extracted from each catheterization report. All available angiograms from each study were thoroughly reviewed for the presence of segmental vessel occlusion. According to our own definitions, vessel loss was diagnosed by (1) the inability to opacify with contrast a segment of lung to which flow was visible on previous catheterizations, despite adequate angiographic technique (i.e., appropriate catheter position, camera position, and contrast delivery) or (2) the presence of collateral vessels from an adjacent segment that filled the segment in question in a retrograde fashion, resulting in a beaked appearance to the vessel where the occlusion occurred. Statistical Analysis The primary outcome variable was the angiographic documentation of segmental PA loss, as described above. Secondary variables included age at catheterization, number of catheterizations performed, RV pressures, RV-to-aorta (RV : Ao) pressure ratio, presence of associated lesions (i.e., supravalvar aortic stenosis), and the segmental distribution of PA interventions. Predictor variables were compared between those patients with angiographic evidence of vessel loss and those without vessel loss. Continuous variables were expressed as mean ± SD, and categorical variables were expressed as median and range. Results
The initial search identified 172 patients. Of this group, 82 patients were excluded because they underwent initial catheterization at BCH prior to 1999, making image retrieval from nondigital media incomplete. An additional 47 patients were excluded because they underwent fewer than 3 catheterizations and thus lacked serial data. One patient with Williams syndrome and isolated coarctation, but no PA disease, was excluded. A Congenit Heart Dis. 2015;10:E60–E67
E62 Table 1.
Morray et al. Baseline Demographics (n = 42)
Age at initial catheterization (years), mean ± SD Male, n (%) Genetic syndromes, n (%) Williams syndrome Familial arteriopathy Alagille syndrome Nonsyndromic, nonfamilial arteriopathy Moyamoya disease Number of catheterizations, median (range) Patients with prior catheterizations at outside hospitals, n (%) Preintervention RV pressure (mm Hg), mean ± SD Preintervention RV-to-aorta pressure ratio, mean ± SD Associated cardiovascular anomalies, n (%) Supravalvar aortic stenosis Mild Moderate Severe Coarctation Ventricular septal defect Patients with pulmonary artery occlusion, n (%) Occluded vessels, n Distribution of occluded vessels, n Right lung Upper lobe Middle lobe Lower lobe Left lung Upper lobe Lingula Lower lobe
3.96 ± 5.29 19 (45%) 15 (36%) 13 (31%) 6 (14%) 5 (12%) 4 (9%) 5 (3–15) 14 (33%) 90 ± 29 0.99 ± 0.30
21 (50%) 10 (24%) 2 (5%) 9 (21%) 9 (21%) 4 (10%) 17 (40%) 35 13 5 3 5 22 5 3 14
total of 42 patients undergoing 228 catheterizations met all criteria and were included in the study. Baseline demographics are listed in Table 1. With regard to associated diagnosis, 15 patients (36%) had Williams syndrome, 13 patients (31%) had non-Williams familial arteriopathy, 6 patients (14%) had Alagille syndrome, 5 patients (12%) had nonsyndromic, nonfamilial arteriopathy, and 4 patients had moyamoya disease. One patient had concurrent diagnoses of both Alagille syndrome and moyamoya disease. The mean age at the initial catheterization reviewed for this study was 3.9 years. Each patient underwent a median of 5 (range 3–15) catheterizations. Fourteen patients (33%) had undergone catheterization at another institution prior to receiving care at our institution. Associated cardiac defects were common: 21 patients had supravalvar aortic stenosis, 4 patients had a ventricular septal defect, 9 patients had coarctation, and 1 patient had partial anomalous venous return of the right upper pulmonary vein. The mean RV pressure was 90 mm Hg and the mean RV : Ao pressure ratio 0.99 (Table 1) at the time of the initial catheterization. Congenit Heart Dis. 2015;10:E60–E67
Figure 1. PA angiograms in anteroposterior (left column) and lateral (right column) projections. (A) Infant with Williams syndrome. There is proximal PA hypoplasia with discrete branch point stenoses and mild poststenotic dilation of the segmental vessels. (B) Female with Alagille syndrome. There is proximal PA hypoplasia that extends out to the segmental vessels with few branch-point stenoses. (C) Male with moyamoya disease. The proximal PAs are normal-caliber. The segmental and subsegmental vessels are hypoplastic and tortuous with a beaded appearance.
Angiographic Appearance On focused angiographic review, subtle differences were observed in the appearance of the PA vessels depending on the severity of disease or underlying syndrome or diagnosis (Figure 1). Patients with Williams syndrome, non-Williams arteriopathy, and nonsyndromic arteriopathy often had diffusely small pulmonary arteries with discrete stenoses at vessel branch points, followed by mild poststenotic dilation of the vessel immediately beyond the branch points. Patients with Alagille syndrome frequently had diffusely small vessels with marked luminal hypoplasia that extended down to the subsegmental branches with fewer discrete stenoses. The patients with moyamoya disease were often the most distinct and tended to have normal-caliber proximal pulmonary arteries with severe vessel hypoplasia that
E63
Segmental Occlusion in PPS Table 2.
Procedural Data
Patients, n Age at initial catheterization (years), mean ± SD Total number of catheterizations, median (range) Williams syndrome, n Alagille syndrome, n Familial arteriopathy, n Nonsyndromic, nonfamilial arteriopathy, n Moyamoya disease, n Initial RV pressure (mm Hg), mean ± SD Initial RV-to-aorta ratio, mean ± SD RV pressure at most recent catheterization, mean ± SD RV-to-aorta ratio at most recent catheterization, mean ± SD Associated cardiac lesions, n (%) Moderate to severe supravalvar aortic stenosis Coarctation
Vessel Occluded
No Vessel Occluded
17 6.21 ± 6.2
25 2.44 ± 4
5 (3–15) 5 3 3 3
4 (3–8) 10 3 10 2
4 93 ± 33
0 87 ± 25
0.99 ± 0.32 74.3 ± 28.5
0.99 ± 0.29 54 ± 15.7
0.75 ± 0.28
0.59 ± 0.2
2 (12%)
9 (36%)
4 (24%)
5 (20%)
Figure 2. Scatter plot of RV-to-aorta (RV : Ao) pressure ratio over time for all interventions. Patients with segmental pulmonary artery occlusion and patients without pulmonary artery occlusion are shown separately.
began at the lobar branch points. The segmental vessels were tortuous and had a beaded appearance. There was frequently pruning of the smallcaliber subsegmental vessels.
Pulmonary Artery Outcomes In our series of 42 patients, 17 (40%) demonstrated angiographic evidence of segmental or subsegmental PA loss over the series of observed catheterizations (Table 2). Patients with vessel loss tended to be older than those patients without vessel loss at the time of their initial catheterization at our institution (6.2 years vs. 2.4 years). Evidence of vessel loss was present in 5 patients with Williams syndrome, 3 patients with Alagille syndrome, 3 patients with familial arteriopathy, 3 patients with nonfamilial arteriopathy, and 4 patients with moyamoya disease. The initial RV pressure and RV : Ao pressure ratio were similar between those patients with vessel loss and those without. Over time, the RV : Ao pressure ratio decreased for the entire cohort, but there was an increase in the ratio in those patients who underwent catheterizations into their teenage years (Figure 2). At the most recent catheterization, the RV pressure and RV : Ao pressure ratio were higher in patients with evidence of vessel loss (Figure 3). The patients with evidence of vessel loss underwent a median of 5 catheterizations (range 3–15), while those patients without vessel loss underwent a median of 4 catheterizations
Figure 3. Boxplot of RV-to-aorta (RV : Ao) pressure ratio at the first catheterization and most recent catheterization for patients with vessel loss and those without vessel loss. Patients with vessel loss did not experience as great a decline in RV : Ao pressure ratio at most recent catheterization as those patients without vessel loss.
(range 3–8). Patients with Williams syndrome, familial arteriopathy, and Alagille syndrome tended to undergo initial interventions at earlier ages relative to patients with nonsyndromic arteriopathy and those with moyamoya disease (Table 3). Among the 17 patients with evidence of vessel loss, a total of 35 vessels were documented as occluded: 22 vessels were occluded in the left lung, while 13 were occluded in the right lung, 19 in the lower lobes, 6 in the middle lobe/lingual, and 10 in the upper lobes (Figure 4). Seven patients with Congenit Heart Dis. 2015;10:E60–E67
E64 Table 3.
Morray et al. Data by Diagnosis
Williams syndrome Familial arteriopathy Alagille syndrome Nonsyndromic, nonfamilial arteriopathy Moyamoya disease
Patients, n
Age at First Catheterization (Years), Mean ± SD
Vessel Loss, n
Vessel Loss at Initial Catheterization, n
15 13 6 5 4
2.8 ± 3.9 3.2 ± 5.3 2.9 ± 2.4 5.9 ± 8.7 12.3 ± 1.9
5 3 3 3 4
0 1 1 1 3
Figure 4. Male with Williams syndrome who underwent initial catheterization at 1 year of age. (A) Left PA angiogram (anteroposterior and lateral projections) at the initial catheterization in 2004. Lung scan demonstrates a slight reduction in left upper lobe flow. (B) Left PA angiogram (anteroposterior and lateral projections) at the most recent catheterization in 2014. The lung scan demonstrates diminishing left upper lobe flow. The angiograms demonstrate occlusion of the superior left lower lobe and the lateral left upper lobe vessels (indicated by the arrows).
documented vessel loss had undergone an intervention (balloon dilation or stent placement) on that same vessel at a prior catheterization. Two patients had undergone PA stent placement at another institution prior to referral. Both of those patients presented to our institution with complete occlusion of the stented segment. Five of the 17 patients (29%) with vessel loss had evidence of vessel loss at their initial catheterizaCongenit Heart Dis. 2015;10:E60–E67
tion. Those 5 patients tended to be older at initial catheterization (9.6 years vs. 4.8 years) and with higher RV pressures (119 ± 48 mm Hg vs. 85 ± 25 mm Hg) relative to patients who developed vessel loss at later catheterizations. The median age at which a vessel was first documented as occluded was 7 years (range 11 months to 23 years). The earliest evidence of vessel loss was in a patient with Williams syndrome at 11 months of age. This was at the fourth catheterization, 8 months after the initial intervention. There was a median 2.5 years and 4 (range 2–6) catheterizations between initial intervention and documentation of segmental vessel loss in those patients who did not have vessel loss at their initial catheterization. Eleven patients (26%) lost different segmental vessels over the course of multiple catheterizations. One patient with nonsyndromic, nonfamilial arteriopathy underwent 15 catheterizations with evidence of progressive distal occlusions despite intervention and ultimately underwent bilateral orthotopic lung transplantation. All 4 patients with moyamoya disease had evidence of PA vessel loss. Three of those 4 patients had evidence of vessel loss at their first catheterization at our institution and progressive loss of different segmental vessels over multiple catheterizations. Discussion
PPS has been described in a number of small cohort studies and case reports, largely in conjunction with Williams syndrome.4–7,9 The natural history of PPS documented in these studies suggests spontaneous improvement in RV pressure and PA stenosis over time. However, in some of these series there was a subset of patients with persistent, severe PA stenosis and RV hypertension.6 There are limited data in patients with Alagille syndrome or moyamoya disease documenting spontaneous regression of PPS. This study documents a group of patients with PPS encompassing a wide array of syndromic diagnoses presenting to a single institution over a 15-year period. In this cohort, 40% of patients had
E65
Segmental Occlusion in PPS evidence at some point in their treatment course of loss of a segmental or subsegmental PA. The average RV pressures at first catheterization in this cohort were near systemic levels. This group may be different than some of the previously reported cohorts in which the initially reported RV pressures were less than systemic pressures.4 The initial RV pressures and the presence of additional cardiovascular anomalies (i.e., coarctation) were similar between those patients who lost PA vessels over the course of their treatment and those who did not. Each patient underwent an average of 5 catheterizations over the 15-year study period. One patient underwent 15 catheterizations for progressive PPS and RV hypertension with continued vessel loss, but ultimately survived to bilateral lung transplantation. Similarly to previous studies, there was an overall reduction in absolute RV pressure and RV : Ao pressure ratio during the course of the study. Even those patients who lost vessels during the course of the study demonstrated a slight reduction in RV pressure and RV : Ao pressure ratio, although not as great a reduction as those patients without vessel loss. The reduction in RV pressure in those patients with vessel loss should be interpreted in the context of multisegment PPS. The goal of an intervention is to prevent vessel loss and distribute blood flow more evenly throughout the vascular tree to prevent hypertensive vascular changes in the patent vessels. In the setting of diffuse PPS, loss of a single segmental branch can further shift blood flow and produce persistent hypertension in the patent vessels. As demonstrated in this cohort, those patients with vessel occlusion experienced a smaller reduction in RV pressure and maintained average RV pressures approximately three-fourths of systemic pressures at the most recent catheterization. Eleven patients lost multiple vessels over the study period, with approximately 50% of those vessels lost in the lower lobes. Almost one-third of the patients with vessel loss had loss at their initial catheterization. Those patients tended to be older at initial intervention than patients who did not demonstrate vessel loss or who lost vessels later in their treatment course. In this cohort there was no strong association of vessel loss with a particular diagnosis. However, patients with moyamoya disease and PPS seemed to have particularly aggressive disease, and all lost vessels during the study. Three of the 4 patients with moyamoya disease lost multiple vessels over the course of the study, in some cases well into adolescence. Patients
with moyamoya disease and nonsyndromic arteriopathy tended to present for initial intervention later relative to the rest of the cohort, which may reflect the difficulty in making the diagnosis early and may have some bearing on the severity of their PA disease at initial presentation. The association of moyamoya disease with congenital heart disease has rarely been reported, but case reports have documented PPS in patients with moyamoya disease.12,13 The angiographic findings in moyamoya disease are distinct, and serial angiograms demonstrate the progressive nature of the distal vessel loss, highlighting the importance of ongoing surveillance in this patient population. Surgical or catheter-based therapies have been advocated in those patients judged to have more severe disease based on absolute RV pressure or RV : Ao pressure ratio.3,14,15 However, as demonstrated in this cohort, there is often little difference in these basic hemodynamic parameters at the initial interventions, and it can be very difficult to predict which patients will go on to develop vascular occlusions. The presence of additional cardiovascular defects or the severity of those defects also does not seem to help differentiate those patients who will go on to lose segmental branches. Despite a strategy of aggressive intervention in this cohort, 7 of 17 patients (41%) lost vessels that had previously been intervened upon. The average balloon size used to dilate these vessels prior to loss was 4 mm (3–4.5 mm), suggesting these were small-caliber vessels. It may be that these vessels had more severe disease and would have progressed to occlusion regardless of intervention. The mechanism of vessel loss in this group was difficult to determine. There was no evidence of filling defects within the lumen of the vessel that might suggest thrombus or an obstructive intimal tear, although these mechanisms cannot be completely ruled out based on the angiographic findings. Prior case series have demonstrated the difficulty in differentiating thromboembolic pulmonary hypertension from other mechanisms.16,17 Limitations
This study is limited by its small size and retrospective design. As described in a previous study, our institution serves as referral center for patients with severe PPS, which likely impacts the severity of disease observed here.18 There is no control group to track the natural history of PPS in patients with severe underlying disease and Congenit Heart Dis. 2015;10:E60–E67
E66 compare outcomes with those patients who receive aggressive catheter-based therapies. As with previous retrospective studies of this condition, it is difficult to control for intervention bias. There is no systematic approach to vessel dilation in this cohort. Typically, the vessels with the greatest degree of hypoplasia or stenosis are intervened upon, so we anticipate that a greater incidence of vessel loss in those vessels that have previously been intervened upon may not reflect the effect of the intervention itself but more likely the fact that those vessels started with more severe disease. This question could really only be answered in a prospective fashion in which vessels met criteria for dilation and were randomized to undergo intervention. An additional important limitation to this study is inconsistency in angiographic technique. The ability to detect the presence or absence of a vessel is dependent on appropriate contrast opacification, catheter position, and fluoroscopic angle. This is not uniform between studies. Review of serial studies reduces this inconsistency by offering multiple data points and views of the vessels in question but does not completely eliminate this concern. With attention to the issue of vessel loss and awareness of its diagnosis using standard angiographic or other axial imaging techniques, additional knowledge may be gained about natural history and timing and effect of intervention on at-risk vessels.
Conclusions
The findings of this study suggest that there is a subset of patients with Williams syndrome and other conditions associated with PPS that are at risk for progressive segmental PA loss, warranting careful surveillance and possible intervention to prevent persistent RV hypertension and pulmonary vascular disease in the patent segments. However, there seem to be few early risk factors that characterize the group of patients that will go on to develop vessel loss. Our review suggests that those rare patients with moyamoya disease and evidence of RV hypertension should be evaluated and monitored for progression of vascular occlusions that may accelerate cardiac consequences. Early recognition of these angiographic findings may aid in devising a strategy of serial interventions aimed at addressing progressive distal PA stenoses. Once we have identified and characterized the features of this subset of patients, we can Congenit Heart Dis. 2015;10:E60–E67
Morray et al. begin to assess risk factors for disease progression and response to therapy.
Author Contributions All the authors were directly involved in the study design, data analysis, interpretation of findings and manuscript completion. All the authors reviewed the manuscript in detail and have approved it in its submitted form.
Corresponding Author: Lisa Bergersen, MD, MPH, Department of Cardiology, 300 Longwood Avenue, Boston, MA 02115, USA. Tel: (617) 355-6529; Fax: (617) 730-0710; E-mail: [email protected] Conflict of interest: None. Accepted in final form: July 3, 2014.
References
1 Gay BB Jr, French RH, Shuford WH, Rogers JV Jr. The roentgenologic features of single and multiple coarctations of the pulmonary artery and branches. Am J Roentgenol Radium Ther Nucl Med. 1963;90:599–613. 2 Franch RH, Gay BB Jr. Congenital stenosis of the pulmonary artery branches. A classification, with postmortem findings in two cases. Am J Med. 1963;35:512–529. 3 Collins RT II, Kaplan P, Somes GW, Rome JJ. Cardiovascular abnormalities, interventions, and long-term outcomes in infantile Williams syndrome. J Pediatr. 2010;156:253–8 e1. 4 Giddins NG, Finley JP, Nanton MA, Roy DL. The natural course of supravalvar aortic stenosis and peripheral pulmonary artery stenosis in Williams’s syndrome. Br Heart J. 1989;62:315–319. 5 Wessel A, Pankau R, Kececioglu D, Ruschewski W, Bursch JH. Three decades of follow-up of aortic and pulmonary vascular lesions in the Williams–Beuren syndrome. Am J Med Genet. 1994;52:297–301. 6 Wren C, Oslizlok P, Bull C. Natural history of supravalvular aortic stenosis and pulmonary artery stenosis. J Am Coll Cardiol. 1990;15:1625–1630. 7 Zalzstein E, Moes CA, Musewe NN, Freedom RM. Spectrum of cardiovascular anomalies in Williams– Beuren syndrome. Pediatr Cardiol. 1991;12:219– 223. 8 Kim YM, Yoo SJ, Choi JY, Kim SH, Bae EJ, Lee YT. Natural course of supravalvar aortic stenosis and peripheral pulmonary arterial stenosis in Williams’ syndrome. Cardiol Young. 1999;9:37–41. 9 Miyamura H, Watanabe H, Tatebe S, Eguchi S. Spontaneous regression of peripheral pulmonary artery stenosis in Williams syndrome. Jpn Circ J. 1996;60:311–314.
Segmental Occlusion in PPS 10 Rhodes J, Dave A, Pulling MC, et al. Effect of pulmonary artery stenoses on the cardiopulmonary response to exercise following repair of tetralogy of Fallot. Am J Cardiol. 1998;81:1217–1219. 11 Sutton NJ, Peng L, Lock JE, et al. Effect of pulmonary artery angioplasty on exercise function after repair of tetralogy of Fallot. Am Heart J. 2008;155:182–186. 12 Lutterman J, Scott M, Nass R, Geva T. Moyamoya syndrome associated with congenital heart disease. Pediatrics. 1998;101:57–60. 13 Reardon L, Maree AO, de Moor M. Moyamoya disease with peripheral pulmonary artery stenoses and coronary artery fistulae. Case Rep Med. 2009;840–904. 14 Monge MC, Mainwaring RD, Sheikh AY, Punn R, Reddy VM, Hanley FL. Surgical reconstruction of peripheral pulmonary artery stenosis in Williams and Alagille syndromes. J Thorac Cardiovasc Surg. 2013;145:476–481.
E67 15 Gandy KL, Tweddell JS, Pelech AN. How we approach peripheral pulmonary stenosis in Williams–Beuren syndrome. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu. 2009;12:118–121. 16 Ulrich S, Fischler M, Pfammatter T, et al. Multiple peripheral pulmonary artery stenoses in adults: a rare cause of severe pulmonary hypertension necessitating lung transplantation. J Heart Lung Transplant. 2005;24:1984–1987. 17 Kreutzer J, Landzberg MJ, Preminger TJ, et al. Isolated peripheral pulmonary artery stenoses in the adult. Circulation. 1996;93:1417–1423. 18 Cunningham JW, McElhinney DB, Gauvreau K, et al. Outcomes after primary transcatheter therapy in infants and young children with severe bilateral peripheral pulmonary artery stenosis. Circ Cardiovasc Interv. 2013;6:460–467.
Congenit Heart Dis. 2015;10:E60–E67
Copyright of Congenital Heart Disease is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
Occult Progressive Pulmonary Arterial Occlusion Associated with Right Ventricular Hypertension in Patients with Systemic Arteriopathy Brian H. Morray, MD, Lisa Bergersen, MD, MPH, James E. Lock, MD, and Audrey C. Marshall, MD Department of Cardiology, Boston Children’s Hospital, Boston, Mass, USA ABSTRACT
Objectives. Peripheral pulmonary artery stenosis (PPS) is a heterogeneous disease associated with numerous clinical or syndromic disorders including Williams syndrome. Previous studies focusing on patients with Williams syndrome have documented spontaneous improvement in branch pulmonary artery (PA) stenosis and right ventricular pressure. The goal of this study is to identify angiographic evidence of spontaneous occlusion or loss of segmental PAs in patients with PPS. Methods. Patients undergoing serial catheterization for PPS at Boston Children’s Hospital between 1999 and 2014 were selected. Patients with structural heart disease such as tetralogy of Fallot, truncus arteriosus, transposition of the great arteries, or single ventricle were excluded. Hemodynamic data were extracted from catheterization reports, all available angiograms were reviewed, and evidence of segmental PA loss was documented based on defined angiographic criteria. Results. Forty-two patients were included in the study. Each patient underwent an average of 5 catheterizations over the 15-year study period. The average RV pressure at initial catheterization was near systemic levels. Evidence of segmental PA loss was present in 40% of patients. Almost 30% of patients with vessel loss demonstrated loss at their initial catheterization, and 65% of patients experienced loss of multiple vessels over time. Patients with vessel loss tended to be older at initial catheterization and have higher average RV pressures at their most recent catheterization than those without vessel loss. Conclusions. Progressive segmental PA loss in patients with systemic arteriopathy and PPS has not been well described previously. In this cohort, segmental PA loss occurred in 40% of patients, resulting in persistently elevated PA pressures relative to patients who did not develop vessel loss. These findings highlight the importance of surveillance in this population of patients with systemic arteriopathy and severe PPS. Key Words. Peripheral Pulmonary Artery Stenosis; Balloon Pulmonary Artery Angioplasty; Systemic Arteriopathy
Introduction
P
eripheral pulmonary artery stenosis (PPS) occurs in approximately 2–3% of patients with congenital heart disease.1 It is a heterogeneous disease that can extend from the mediastinal pulmonary arteries out into more distal, posthilar vessels and can occur in association with congenital heart disease, several well-described syndromes (Williams, Alagille, Noonan, and congenital rubella syndromes) or in isolation.2 PPS is the
Congenit Heart Dis. 2015;10:E60–E67
second most common cardiovascular disorder in Williams syndrome and the most common cardiovascular finding in Alagille syndrome. The literature describing the natural history of PPS has largely focused on Williams syndrome. These data suggest that there is spontaneous improvement in right ventricular (RV) pressure and angiographic appearance of PPS over time without intervention.3–9 However, some of these series document a small subset of patients with worsening RV hypertension but provide little © 2014 Wiley Periodicals, Inc.
E61
Segmental Occlusion in PPS angiographic evidence of progressive PPS. We believe there is a subset of patients with PPS who develop progressive, segmental pulmonary artery (PA) vascular occlusions. Loss of PA crosssectional area can lead to persistent RV hypertension and hypertensive vascular changes in patent segments. The goal of catheter-based interventions is to prevent vessel loss and distribute blood flow throughout the vascular tree to prevent chronic pulmonary arterial hypertension. Maintaining patency of segmental PA branches may also improve ventilation–perfusion matching and gas exchange, as has been described in patients with tetralogy of Fallot and PPS following balloon pulmonary angioplasty.10,11 The goal of this study is to describe the previously undocumented angiographic characteristics of progressive segmental PA occlusion in patients with syndromic and nonsyndromic arteriopathy and PPS. Methods
Patients and Study Protocol Patients were selected by searching the clinical databases in the Department of Cardiology at Boston Children’s Hospital (BCH). The databases were queried for patients coded with a diagnosis of PA stenosis or PPS. Additionally, patients with syndromes with known association with PPS were selected. These included the following: (1) Williams syndrome diagnosed genetically (based on fluorescence in situ hybridization) or based on clinical characteristics; (2) non-Williams familial arteriopathy (at least 1 first-degree relative with supravalvar aortic stenosis or PPS and no syndromic diagnosis); (3) Alagille syndrome; (4) nonsyndromic, nonfamilial arteriopathy (presence of PPS with no family history or syndromic diagnosis); and (5) moyamoya disease (based on the presence of characteristic clinical history and neuroimaging findings). Patients were included in the study if they had PPS, underwent catheterization at BCH between January 1999 and January 2014, and subsequently underwent ≥3 catheterizations for evaluation and dilation of the branch pulmonary arteries. Patients with structural heart disease associated with PPS, such as tetralogy of Fallot, truncus arteriosus, transposition of the great arteries, single-ventricle disease, or discontinuous pulmonary arteries, were excluded. Patients with other congenital heart disease, such as isolated atrial or ventricular septal defects, as well as systemic arterial defects such as
supravalvar aortic stenosis or coarctation of the aorta, were included. Patients were included even if they underwent initial intervention at another institution or underwent surgical repair of an associated minor defect (i.e., ventricular septal defect, patent ductus arteriosus). The Children’s Hospital Committee on Clinical Investigation approved the retrospective study.
Catheterization Data Serial catheterization data were reviewed for each patient by a single investigator (BM). RV and descending aortic pressures were extracted from each catheterization report. All available angiograms from each study were thoroughly reviewed for the presence of segmental vessel occlusion. According to our own definitions, vessel loss was diagnosed by (1) the inability to opacify with contrast a segment of lung to which flow was visible on previous catheterizations, despite adequate angiographic technique (i.e., appropriate catheter position, camera position, and contrast delivery) or (2) the presence of collateral vessels from an adjacent segment that filled the segment in question in a retrograde fashion, resulting in a beaked appearance to the vessel where the occlusion occurred. Statistical Analysis The primary outcome variable was the angiographic documentation of segmental PA loss, as described above. Secondary variables included age at catheterization, number of catheterizations performed, RV pressures, RV-to-aorta (RV : Ao) pressure ratio, presence of associated lesions (i.e., supravalvar aortic stenosis), and the segmental distribution of PA interventions. Predictor variables were compared between those patients with angiographic evidence of vessel loss and those without vessel loss. Continuous variables were expressed as mean ± SD, and categorical variables were expressed as median and range. Results
The initial search identified 172 patients. Of this group, 82 patients were excluded because they underwent initial catheterization at BCH prior to 1999, making image retrieval from nondigital media incomplete. An additional 47 patients were excluded because they underwent fewer than 3 catheterizations and thus lacked serial data. One patient with Williams syndrome and isolated coarctation, but no PA disease, was excluded. A Congenit Heart Dis. 2015;10:E60–E67
E62 Table 1.
Morray et al. Baseline Demographics (n = 42)
Age at initial catheterization (years), mean ± SD Male, n (%) Genetic syndromes, n (%) Williams syndrome Familial arteriopathy Alagille syndrome Nonsyndromic, nonfamilial arteriopathy Moyamoya disease Number of catheterizations, median (range) Patients with prior catheterizations at outside hospitals, n (%) Preintervention RV pressure (mm Hg), mean ± SD Preintervention RV-to-aorta pressure ratio, mean ± SD Associated cardiovascular anomalies, n (%) Supravalvar aortic stenosis Mild Moderate Severe Coarctation Ventricular septal defect Patients with pulmonary artery occlusion, n (%) Occluded vessels, n Distribution of occluded vessels, n Right lung Upper lobe Middle lobe Lower lobe Left lung Upper lobe Lingula Lower lobe
3.96 ± 5.29 19 (45%) 15 (36%) 13 (31%) 6 (14%) 5 (12%) 4 (9%) 5 (3–15) 14 (33%) 90 ± 29 0.99 ± 0.30
21 (50%) 10 (24%) 2 (5%) 9 (21%) 9 (21%) 4 (10%) 17 (40%) 35 13 5 3 5 22 5 3 14
total of 42 patients undergoing 228 catheterizations met all criteria and were included in the study. Baseline demographics are listed in Table 1. With regard to associated diagnosis, 15 patients (36%) had Williams syndrome, 13 patients (31%) had non-Williams familial arteriopathy, 6 patients (14%) had Alagille syndrome, 5 patients (12%) had nonsyndromic, nonfamilial arteriopathy, and 4 patients had moyamoya disease. One patient had concurrent diagnoses of both Alagille syndrome and moyamoya disease. The mean age at the initial catheterization reviewed for this study was 3.9 years. Each patient underwent a median of 5 (range 3–15) catheterizations. Fourteen patients (33%) had undergone catheterization at another institution prior to receiving care at our institution. Associated cardiac defects were common: 21 patients had supravalvar aortic stenosis, 4 patients had a ventricular septal defect, 9 patients had coarctation, and 1 patient had partial anomalous venous return of the right upper pulmonary vein. The mean RV pressure was 90 mm Hg and the mean RV : Ao pressure ratio 0.99 (Table 1) at the time of the initial catheterization. Congenit Heart Dis. 2015;10:E60–E67
Figure 1. PA angiograms in anteroposterior (left column) and lateral (right column) projections. (A) Infant with Williams syndrome. There is proximal PA hypoplasia with discrete branch point stenoses and mild poststenotic dilation of the segmental vessels. (B) Female with Alagille syndrome. There is proximal PA hypoplasia that extends out to the segmental vessels with few branch-point stenoses. (C) Male with moyamoya disease. The proximal PAs are normal-caliber. The segmental and subsegmental vessels are hypoplastic and tortuous with a beaded appearance.
Angiographic Appearance On focused angiographic review, subtle differences were observed in the appearance of the PA vessels depending on the severity of disease or underlying syndrome or diagnosis (Figure 1). Patients with Williams syndrome, non-Williams arteriopathy, and nonsyndromic arteriopathy often had diffusely small pulmonary arteries with discrete stenoses at vessel branch points, followed by mild poststenotic dilation of the vessel immediately beyond the branch points. Patients with Alagille syndrome frequently had diffusely small vessels with marked luminal hypoplasia that extended down to the subsegmental branches with fewer discrete stenoses. The patients with moyamoya disease were often the most distinct and tended to have normal-caliber proximal pulmonary arteries with severe vessel hypoplasia that
E63
Segmental Occlusion in PPS Table 2.
Procedural Data
Patients, n Age at initial catheterization (years), mean ± SD Total number of catheterizations, median (range) Williams syndrome, n Alagille syndrome, n Familial arteriopathy, n Nonsyndromic, nonfamilial arteriopathy, n Moyamoya disease, n Initial RV pressure (mm Hg), mean ± SD Initial RV-to-aorta ratio, mean ± SD RV pressure at most recent catheterization, mean ± SD RV-to-aorta ratio at most recent catheterization, mean ± SD Associated cardiac lesions, n (%) Moderate to severe supravalvar aortic stenosis Coarctation
Vessel Occluded
No Vessel Occluded
17 6.21 ± 6.2
25 2.44 ± 4
5 (3–15) 5 3 3 3
4 (3–8) 10 3 10 2
4 93 ± 33
0 87 ± 25
0.99 ± 0.32 74.3 ± 28.5
0.99 ± 0.29 54 ± 15.7
0.75 ± 0.28
0.59 ± 0.2
2 (12%)
9 (36%)
4 (24%)
5 (20%)
Figure 2. Scatter plot of RV-to-aorta (RV : Ao) pressure ratio over time for all interventions. Patients with segmental pulmonary artery occlusion and patients without pulmonary artery occlusion are shown separately.
began at the lobar branch points. The segmental vessels were tortuous and had a beaded appearance. There was frequently pruning of the smallcaliber subsegmental vessels.
Pulmonary Artery Outcomes In our series of 42 patients, 17 (40%) demonstrated angiographic evidence of segmental or subsegmental PA loss over the series of observed catheterizations (Table 2). Patients with vessel loss tended to be older than those patients without vessel loss at the time of their initial catheterization at our institution (6.2 years vs. 2.4 years). Evidence of vessel loss was present in 5 patients with Williams syndrome, 3 patients with Alagille syndrome, 3 patients with familial arteriopathy, 3 patients with nonfamilial arteriopathy, and 4 patients with moyamoya disease. The initial RV pressure and RV : Ao pressure ratio were similar between those patients with vessel loss and those without. Over time, the RV : Ao pressure ratio decreased for the entire cohort, but there was an increase in the ratio in those patients who underwent catheterizations into their teenage years (Figure 2). At the most recent catheterization, the RV pressure and RV : Ao pressure ratio were higher in patients with evidence of vessel loss (Figure 3). The patients with evidence of vessel loss underwent a median of 5 catheterizations (range 3–15), while those patients without vessel loss underwent a median of 4 catheterizations
Figure 3. Boxplot of RV-to-aorta (RV : Ao) pressure ratio at the first catheterization and most recent catheterization for patients with vessel loss and those without vessel loss. Patients with vessel loss did not experience as great a decline in RV : Ao pressure ratio at most recent catheterization as those patients without vessel loss.
(range 3–8). Patients with Williams syndrome, familial arteriopathy, and Alagille syndrome tended to undergo initial interventions at earlier ages relative to patients with nonsyndromic arteriopathy and those with moyamoya disease (Table 3). Among the 17 patients with evidence of vessel loss, a total of 35 vessels were documented as occluded: 22 vessels were occluded in the left lung, while 13 were occluded in the right lung, 19 in the lower lobes, 6 in the middle lobe/lingual, and 10 in the upper lobes (Figure 4). Seven patients with Congenit Heart Dis. 2015;10:E60–E67
E64 Table 3.
Morray et al. Data by Diagnosis
Williams syndrome Familial arteriopathy Alagille syndrome Nonsyndromic, nonfamilial arteriopathy Moyamoya disease
Patients, n
Age at First Catheterization (Years), Mean ± SD
Vessel Loss, n
Vessel Loss at Initial Catheterization, n
15 13 6 5 4
2.8 ± 3.9 3.2 ± 5.3 2.9 ± 2.4 5.9 ± 8.7 12.3 ± 1.9
5 3 3 3 4
0 1 1 1 3
Figure 4. Male with Williams syndrome who underwent initial catheterization at 1 year of age. (A) Left PA angiogram (anteroposterior and lateral projections) at the initial catheterization in 2004. Lung scan demonstrates a slight reduction in left upper lobe flow. (B) Left PA angiogram (anteroposterior and lateral projections) at the most recent catheterization in 2014. The lung scan demonstrates diminishing left upper lobe flow. The angiograms demonstrate occlusion of the superior left lower lobe and the lateral left upper lobe vessels (indicated by the arrows).
documented vessel loss had undergone an intervention (balloon dilation or stent placement) on that same vessel at a prior catheterization. Two patients had undergone PA stent placement at another institution prior to referral. Both of those patients presented to our institution with complete occlusion of the stented segment. Five of the 17 patients (29%) with vessel loss had evidence of vessel loss at their initial catheterizaCongenit Heart Dis. 2015;10:E60–E67
tion. Those 5 patients tended to be older at initial catheterization (9.6 years vs. 4.8 years) and with higher RV pressures (119 ± 48 mm Hg vs. 85 ± 25 mm Hg) relative to patients who developed vessel loss at later catheterizations. The median age at which a vessel was first documented as occluded was 7 years (range 11 months to 23 years). The earliest evidence of vessel loss was in a patient with Williams syndrome at 11 months of age. This was at the fourth catheterization, 8 months after the initial intervention. There was a median 2.5 years and 4 (range 2–6) catheterizations between initial intervention and documentation of segmental vessel loss in those patients who did not have vessel loss at their initial catheterization. Eleven patients (26%) lost different segmental vessels over the course of multiple catheterizations. One patient with nonsyndromic, nonfamilial arteriopathy underwent 15 catheterizations with evidence of progressive distal occlusions despite intervention and ultimately underwent bilateral orthotopic lung transplantation. All 4 patients with moyamoya disease had evidence of PA vessel loss. Three of those 4 patients had evidence of vessel loss at their first catheterization at our institution and progressive loss of different segmental vessels over multiple catheterizations. Discussion
PPS has been described in a number of small cohort studies and case reports, largely in conjunction with Williams syndrome.4–7,9 The natural history of PPS documented in these studies suggests spontaneous improvement in RV pressure and PA stenosis over time. However, in some of these series there was a subset of patients with persistent, severe PA stenosis and RV hypertension.6 There are limited data in patients with Alagille syndrome or moyamoya disease documenting spontaneous regression of PPS. This study documents a group of patients with PPS encompassing a wide array of syndromic diagnoses presenting to a single institution over a 15-year period. In this cohort, 40% of patients had
E65
Segmental Occlusion in PPS evidence at some point in their treatment course of loss of a segmental or subsegmental PA. The average RV pressures at first catheterization in this cohort were near systemic levels. This group may be different than some of the previously reported cohorts in which the initially reported RV pressures were less than systemic pressures.4 The initial RV pressures and the presence of additional cardiovascular anomalies (i.e., coarctation) were similar between those patients who lost PA vessels over the course of their treatment and those who did not. Each patient underwent an average of 5 catheterizations over the 15-year study period. One patient underwent 15 catheterizations for progressive PPS and RV hypertension with continued vessel loss, but ultimately survived to bilateral lung transplantation. Similarly to previous studies, there was an overall reduction in absolute RV pressure and RV : Ao pressure ratio during the course of the study. Even those patients who lost vessels during the course of the study demonstrated a slight reduction in RV pressure and RV : Ao pressure ratio, although not as great a reduction as those patients without vessel loss. The reduction in RV pressure in those patients with vessel loss should be interpreted in the context of multisegment PPS. The goal of an intervention is to prevent vessel loss and distribute blood flow more evenly throughout the vascular tree to prevent hypertensive vascular changes in the patent vessels. In the setting of diffuse PPS, loss of a single segmental branch can further shift blood flow and produce persistent hypertension in the patent vessels. As demonstrated in this cohort, those patients with vessel occlusion experienced a smaller reduction in RV pressure and maintained average RV pressures approximately three-fourths of systemic pressures at the most recent catheterization. Eleven patients lost multiple vessels over the study period, with approximately 50% of those vessels lost in the lower lobes. Almost one-third of the patients with vessel loss had loss at their initial catheterization. Those patients tended to be older at initial intervention than patients who did not demonstrate vessel loss or who lost vessels later in their treatment course. In this cohort there was no strong association of vessel loss with a particular diagnosis. However, patients with moyamoya disease and PPS seemed to have particularly aggressive disease, and all lost vessels during the study. Three of the 4 patients with moyamoya disease lost multiple vessels over the course of the study, in some cases well into adolescence. Patients
with moyamoya disease and nonsyndromic arteriopathy tended to present for initial intervention later relative to the rest of the cohort, which may reflect the difficulty in making the diagnosis early and may have some bearing on the severity of their PA disease at initial presentation. The association of moyamoya disease with congenital heart disease has rarely been reported, but case reports have documented PPS in patients with moyamoya disease.12,13 The angiographic findings in moyamoya disease are distinct, and serial angiograms demonstrate the progressive nature of the distal vessel loss, highlighting the importance of ongoing surveillance in this patient population. Surgical or catheter-based therapies have been advocated in those patients judged to have more severe disease based on absolute RV pressure or RV : Ao pressure ratio.3,14,15 However, as demonstrated in this cohort, there is often little difference in these basic hemodynamic parameters at the initial interventions, and it can be very difficult to predict which patients will go on to develop vascular occlusions. The presence of additional cardiovascular defects or the severity of those defects also does not seem to help differentiate those patients who will go on to lose segmental branches. Despite a strategy of aggressive intervention in this cohort, 7 of 17 patients (41%) lost vessels that had previously been intervened upon. The average balloon size used to dilate these vessels prior to loss was 4 mm (3–4.5 mm), suggesting these were small-caliber vessels. It may be that these vessels had more severe disease and would have progressed to occlusion regardless of intervention. The mechanism of vessel loss in this group was difficult to determine. There was no evidence of filling defects within the lumen of the vessel that might suggest thrombus or an obstructive intimal tear, although these mechanisms cannot be completely ruled out based on the angiographic findings. Prior case series have demonstrated the difficulty in differentiating thromboembolic pulmonary hypertension from other mechanisms.16,17 Limitations
This study is limited by its small size and retrospective design. As described in a previous study, our institution serves as referral center for patients with severe PPS, which likely impacts the severity of disease observed here.18 There is no control group to track the natural history of PPS in patients with severe underlying disease and Congenit Heart Dis. 2015;10:E60–E67
E66 compare outcomes with those patients who receive aggressive catheter-based therapies. As with previous retrospective studies of this condition, it is difficult to control for intervention bias. There is no systematic approach to vessel dilation in this cohort. Typically, the vessels with the greatest degree of hypoplasia or stenosis are intervened upon, so we anticipate that a greater incidence of vessel loss in those vessels that have previously been intervened upon may not reflect the effect of the intervention itself but more likely the fact that those vessels started with more severe disease. This question could really only be answered in a prospective fashion in which vessels met criteria for dilation and were randomized to undergo intervention. An additional important limitation to this study is inconsistency in angiographic technique. The ability to detect the presence or absence of a vessel is dependent on appropriate contrast opacification, catheter position, and fluoroscopic angle. This is not uniform between studies. Review of serial studies reduces this inconsistency by offering multiple data points and views of the vessels in question but does not completely eliminate this concern. With attention to the issue of vessel loss and awareness of its diagnosis using standard angiographic or other axial imaging techniques, additional knowledge may be gained about natural history and timing and effect of intervention on at-risk vessels.
Conclusions
The findings of this study suggest that there is a subset of patients with Williams syndrome and other conditions associated with PPS that are at risk for progressive segmental PA loss, warranting careful surveillance and possible intervention to prevent persistent RV hypertension and pulmonary vascular disease in the patent segments. However, there seem to be few early risk factors that characterize the group of patients that will go on to develop vessel loss. Our review suggests that those rare patients with moyamoya disease and evidence of RV hypertension should be evaluated and monitored for progression of vascular occlusions that may accelerate cardiac consequences. Early recognition of these angiographic findings may aid in devising a strategy of serial interventions aimed at addressing progressive distal PA stenoses. Once we have identified and characterized the features of this subset of patients, we can Congenit Heart Dis. 2015;10:E60–E67
Morray et al. begin to assess risk factors for disease progression and response to therapy.
Author Contributions All the authors were directly involved in the study design, data analysis, interpretation of findings and manuscript completion. All the authors reviewed the manuscript in detail and have approved it in its submitted form.
Corresponding Author: Lisa Bergersen, MD, MPH, Department of Cardiology, 300 Longwood Avenue, Boston, MA 02115, USA. Tel: (617) 355-6529; Fax: (617) 730-0710; E-mail: [email protected] Conflict of interest: None. Accepted in final form: July 3, 2014.
References
1 Gay BB Jr, French RH, Shuford WH, Rogers JV Jr. The roentgenologic features of single and multiple coarctations of the pulmonary artery and branches. Am J Roentgenol Radium Ther Nucl Med. 1963;90:599–613. 2 Franch RH, Gay BB Jr. Congenital stenosis of the pulmonary artery branches. A classification, with postmortem findings in two cases. Am J Med. 1963;35:512–529. 3 Collins RT II, Kaplan P, Somes GW, Rome JJ. Cardiovascular abnormalities, interventions, and long-term outcomes in infantile Williams syndrome. J Pediatr. 2010;156:253–8 e1. 4 Giddins NG, Finley JP, Nanton MA, Roy DL. The natural course of supravalvar aortic stenosis and peripheral pulmonary artery stenosis in Williams’s syndrome. Br Heart J. 1989;62:315–319. 5 Wessel A, Pankau R, Kececioglu D, Ruschewski W, Bursch JH. Three decades of follow-up of aortic and pulmonary vascular lesions in the Williams–Beuren syndrome. Am J Med Genet. 1994;52:297–301. 6 Wren C, Oslizlok P, Bull C. Natural history of supravalvular aortic stenosis and pulmonary artery stenosis. J Am Coll Cardiol. 1990;15:1625–1630. 7 Zalzstein E, Moes CA, Musewe NN, Freedom RM. Spectrum of cardiovascular anomalies in Williams– Beuren syndrome. Pediatr Cardiol. 1991;12:219– 223. 8 Kim YM, Yoo SJ, Choi JY, Kim SH, Bae EJ, Lee YT. Natural course of supravalvar aortic stenosis and peripheral pulmonary arterial stenosis in Williams’ syndrome. Cardiol Young. 1999;9:37–41. 9 Miyamura H, Watanabe H, Tatebe S, Eguchi S. Spontaneous regression of peripheral pulmonary artery stenosis in Williams syndrome. Jpn Circ J. 1996;60:311–314.
Segmental Occlusion in PPS 10 Rhodes J, Dave A, Pulling MC, et al. Effect of pulmonary artery stenoses on the cardiopulmonary response to exercise following repair of tetralogy of Fallot. Am J Cardiol. 1998;81:1217–1219. 11 Sutton NJ, Peng L, Lock JE, et al. Effect of pulmonary artery angioplasty on exercise function after repair of tetralogy of Fallot. Am Heart J. 2008;155:182–186. 12 Lutterman J, Scott M, Nass R, Geva T. Moyamoya syndrome associated with congenital heart disease. Pediatrics. 1998;101:57–60. 13 Reardon L, Maree AO, de Moor M. Moyamoya disease with peripheral pulmonary artery stenoses and coronary artery fistulae. Case Rep Med. 2009;840–904. 14 Monge MC, Mainwaring RD, Sheikh AY, Punn R, Reddy VM, Hanley FL. Surgical reconstruction of peripheral pulmonary artery stenosis in Williams and Alagille syndromes. J Thorac Cardiovasc Surg. 2013;145:476–481.
E67 15 Gandy KL, Tweddell JS, Pelech AN. How we approach peripheral pulmonary stenosis in Williams–Beuren syndrome. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu. 2009;12:118–121. 16 Ulrich S, Fischler M, Pfammatter T, et al. Multiple peripheral pulmonary artery stenoses in adults: a rare cause of severe pulmonary hypertension necessitating lung transplantation. J Heart Lung Transplant. 2005;24:1984–1987. 17 Kreutzer J, Landzberg MJ, Preminger TJ, et al. Isolated peripheral pulmonary artery stenoses in the adult. Circulation. 1996;93:1417–1423. 18 Cunningham JW, McElhinney DB, Gauvreau K, et al. Outcomes after primary transcatheter therapy in infants and young children with severe bilateral peripheral pulmonary artery stenosis. Circ Cardiovasc Interv. 2013;6:460–467.
Congenit Heart Dis. 2015;10:E60–E67
Copyright of Congenital Heart Disease is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
