Endovascular Stents in the Pulmonary Circulation
Clinical Impact on Management and Medium-term Follow-up
Background The use of endovascular stents to relieve obstructions in the setting of non–balloon dilatable pulmonary artery stenosis has been encouraging. The benefits in management and the potential for restenosis, however, have not been defined. This study attempts to assess the impact of such implants on clinical outcomes and the pattern of stent incorporation within the vessel wall.
Methods and Results Fifty-five balloon-expandable stents were implanted in 42 patients 6.1±4.7 years of age. Patients were followed prospectively (median, 15 months) and recatheterized 1 year after implantation. Thirty-eight patients had the implants positioned percutaneously (49 implants), while 4 patients (6 implants) had intraoperative implantations. There was a diameter increase in the stenotic area of 109±79% (P<.0001) and a gradient reduction of 74±26% (P<.0001). Twelve stents straddled the orifice of side-branch pulmonary arteries and reduced flow to the branch vessel acutely in 7 patients. Twenty-nine patients underwent recatheterization, and various degrees and locations of acquired intraluminal narrowing were observed in all cases, particularly in areas of diameter mismatch between the stented and nonstented vessels. Eleven patients had further dilation with diameter improvement. Of the 38 patients who underwent percutaneous implantation, planned surgery for pulmonary artery stenosis was avoided in 33 and deferred in 4 patients. One patient who was considered inoperable had stent implantation as a palliative procedure. Symptomatic improvement was reported in 27 patients, and 15 patients remained asymptomatic.
Conclusions Endovascular stents have a role in the treatment of pulmonary artery stenoses and positively affect clinical care. The stenosis relief, however, may be tempered by the development of intraluminal stent obstruction, which may require redilation (15 of 55 stents) and mandates long-term follow-up.
Congenital or acquired pulmonary artery stenosis remains a clinically challenging lesion, particularly when distal to the lung hilum. To manage lesions that experience suboptimal relief from balloon angioplasty, endovascular stents have been used to provide a rigid framework for vessel support.1 2 Early studies confirmed that such implants could provide substantial vessel enlargement in this setting2 and have proved similarly effective in a variety of other vascular stenoses observed in congenital heart disorders.3 4 In particular, clinical observations regarding the safety and efficacy of stent implantation in branch pulmonary arteries and their influence on clinical management have been encouraging.2 3 4 The objectives of this study were to (1) review acute efficacy and safety of pulmonary artery stent implantation; (2) assess pulmonary artery flow characteristics; (3) note the presence, extent, and degree of luminal ingrowth; and (4) define the impact on management algorithms that device implantation had on clinical outcomes.
Between October 1990 and July 1993, 42 patients (17 girls, 25 boys), underwent placement of balloon-expandable endovascular stents (Johnson & Johnson Interventional System) in branch pulmonary arteries under a protocol approved by the Human Subjects Protection Committee of The Hospital for Sick Children, Faculty of Medicine, University of Toronto. Informed parental consent was obtained in all cases. Patient records and angiograms were reviewed for demographic data, characteristics of the initial lesions, and previous surgery. All patients were found at the time of catheterization to have unilateral or bilateral pulmonary artery stenosis with average peak systolic gradients across the narrowings of 28±21 mm Hg. Balloon angioplasty was attempted in all patients with previously published methods,5 6 and all lesions were found dilatable. Failure to achieve adequate obstruction relief, defined as vessel size enlargement to within 75% of the adjacent normal vessel, a reduction in right ventricular pressure to less than two thirds the systemic pressure, and the elimination of the gradient across the lesion, provided the substrate for stent implantation. Fifty-five balloon-expandable stents were implanted into 42 patients: 22 to the left pulmonary artery, 15 to the right pulmonary artery, and 9 bilaterally. Percutaneous stent implantation was performed in 38 patients, whereas stents were placed at the time of surgery for additional cardiac defects in 4 patients. Stents were of three sizes: large stents (30×3.4-mm design, n=29); medium stents (20×2.5 mm, n=16); and small, articulated stents (with a 1-mm separation, 15×1.8 mm, n=10). Acute anatomic success was judged by use of the criteria of Rothman et al,5 ie, a >50% increase in vessel diameter, improvement or normalization of flow to the ipsilateral lung by perfusion scan, or a reduction in the ratio of right ventricular to systemic pressure to >33% (in the absence of a residual ventricular septal defect or an obstruction proximal to the branch pulmonary arteries). Clinical success6 was defined as symptomatic improvement (increased exercise tolerance), avoidance of anticipated pulmonary artery surgery, or sufficient improvement in vessel diameter to render a previously inoperable patient operable, thus altering treatment options.
Method of Implantation
Cardiac catheterization was performed under general anesthesia. Selective pulmonary artery angiography was obtained in axial projections to maximize stenosis profile and pulmonary artery topography. Care was taken to note the position of branching vessels and their relation to the stenosis. The largest stent that could be delivered safely was chosen to allow the potential for future dilation. The technique for implantation was described in detail previously.3 In brief, stents were mounted on 8-, 10-, or 12-mm-diameter low-profile balloons (Mansfield Scientific) and directed to straddle the lesions with an 8F, 10F, or 11F long sheath guide (with built-in backbleed taps, Cook Inc) over a 0.035-in extrastiff exchange interventional wire (Amplatz type, Cook Inc). After initial placement, further stent enlargement with a larger-diameter balloon was performed to eliminate any residual stenosis. Dilations were attempted to conform to the configuration of the pulmonary vessel both proximally and distally, with attempts made to avoid overdistention of the pulmonary artery segment. During the course of the procedure, the activated clotting time was adjusted to 250 seconds after an intravenous dose of 150 U/kg heparan sulfate. After catheterization, heparin sulfate was administered intravenously at 10 U · kg−1 · h−1 for 24 hours, and acetylsalicylic acid 3 to 5 mg · kg−1 · d−1 was prescribed for 6 months.
Before stent implantation, patients underwent routine precatheterization noninvasive assessment and lung perfusion scans. After implantation, patients were assessed clinically, with chest radiography and pulmonary perfusion scans at 6 months and with repeated cardiac catheterization at 1 year. Follow-up catheterizations were performed with patients under sedation (narcoleptic technique), and attempts were made to match angiographic projections used at the time of implantation. Dilation was repeated when there was an observed intraluminal obstruction, growth of a proximal or distal vessel, or a persistent narrowing within the stent caused by residual vessel stenosis. Depending on the mechanism and degree of the noted obstructions, balloon diameters were of the same size or 2 to 3 mm larger than that used at implantation. Pulmonary artery and stent measurements were determined from the cineangiograms with electronic calipers, with the known angiographic catheter diameter used to calculate the magnification correction factor.
Results are expressed as mean±SD or median with range. Because changes in dimensions, hemodynamic parameters, and lung perfusion scans were not uniformly collected over all time periods, repeated-measures ANOVA could not be applied. Therefore, selected comparisons were made with paired Student’s t tests with α increased to P<.01 to minimize the probability of a type I error with multiple comparisons. The number of subjects for each statistical comparison is also given.
Mean age at the time of implantation was 6.1±4.7 years (range, 0.2 to 16.9 years), and the mean weight was 21.9±17.4 kg (range, 3.5 to 69.5 kg). Forty-one patients had had prior surgery for congenital heart lesions, and 1 patient developed left pulmonary artery stenosis after device occlusion for a patent ductus arteriosus (the Table⇓).
Optimal stent positioning, defined as straddling the stenotic lesion in the center of the implant and avoiding protrusion of the stent into the main pulmonary artery or the orifice of a side branch, was achieved in 53 of 55 individual implants. Mean stenosis diameter was 5±2 mm (n=46), increasing to 10±3 mm (n=48) after implantation (mean increase, 109±79%; P<.0001; Fig 1A⇓). Peak systolic pressure gradients fell from 28±21 to 7±9 mm Hg (n=45; P<.0001; mean decrease, 74±26%; Fig 1B⇓), and right ventricular systolic pressure fell from 48±18 to 41±14 mm Hg (n=24; P<.001; Fig 1C⇓). The ratios of right ventricular to systemic pressure decreased from 0.56±0.22 (n=21) to 0.48±0.16 (n=13) (mean decrease, 25±18.9%; n=11; P=.0007). Twelve stents straddled the orifice of side-branch pulmonary arteries. In 5 implants, flow into these side branches was unaffected; in 7 implants, reduced flow was noted after implantation.
One patient developed distal left pulmonary artery thrombosis and was treated successfully with intravenous thrombolytic therapy. This event was thought to be due to prolonged balloon inflation (malfunction in balloon deflation) and low flow to the involved branch. One stent that was placed into the proximal left pulmonary artery and protruded into the main pulmonary artery (by >50%) was removed and replaced at a subsequent catheterization. A further stent migrated to a left lower lobe branch and was dilated in place, leaving no residual obstruction to other side-branch pulmonary arteries.
Median follow-up was 15 months (range, 1 to 36 months). Symptomatic improvement was noted in 27 patients (improved exercise tolerance by history), whereas 15 patients remained asymptomatic. Surgery for pulmonary artery stenosis was avoided in 33 of the 38 patients undergoing percutaneous stent implantation; 4 patients had planned cardiac surgery for additional cardiac lesions simplified by stent implantation. The remaining of these 38 patients, considered inoperable because of multiple peripheral pulmonary stenoses, had stent implantation as a palliative procedure, which improved perfusion to the stented lung.
Twenty-nine patients underwent recatheterization 10±4 months after implantation. The narrowest (minimal) endoluminal diameter was 8±2 mm (32 implants, Fig 1A⇑), systolic pressure gradient was 18±17 mm Hg (35 implants, Fig 1B⇑), and right ventricular systolic pressure was 41±14 mm Hg (from 19 patients, Fig 1C⇑). Right ventricular end-diastolic pressures had no significant change from before stent implantation (8.5±2.8 mm Hg, n=17) to follow-up catheterization (8.8±2.9 mm Hg, n=17) in these patients. Ten patients with unilateral obstructions had lung perfusion scans before and after (5.2±2.6 months) implantation. The mean percentage of flow to the ipsilateral lung increased from 28±13% to 40±16% (n=6; P<.005) for left pulmonary artery implants, whereas no significant change was observed (32±25% to 42±11%, n=4) for right pulmonary artery stents. Comparison of the angiographic and hemodynamic findings at follow-up to those immediately after stent implantation, available in 26 implants, showed a 20±21% decrease in the lumen diameter (from 10±2 to 8±2 mm; P<.0001), a nonsignificant increase in mean systolic pressure gradient of 198±240% (from 6±9 to 15±14) in 27 implants, and no significant increase in right ventricular pressure (14 patients) or in ratios of right ventricular to systemic pressure (9 patients). Restenosis was significantly higher in small stents, with a diameter decrease of 40±7% (n=4) for 15×1.8-mm stents and 37±20% (n=5) for 20×2.5-mm stents compared with 9±20% (n=17) for 30×3.4-mm stents (P=.001).
Selective angiography identified endoluminal tissue ingrowth of various degrees in all stents. Various patterns of intimal growth were noted (Fig 2⇓) that involved the proximal or distal stent, where the implant was larger than the adjacent vessel diameter (n=18); a focal ingrowth at the site of the previous artery stenosis (n=3); or a diffuse pattern of growth (n=2). Of the 7 branch pulmonary arteries that were traversed by a stent with an apparent reduction in flow, persistent reduction was noted in 6 arteries, and 1 appeared fully patent. The 5 remaining branch arteries that were crossed by a stent with no change in flow remained fully patent (see Fig 3⇓).
Redilation of 16 stents was attempted in 11 patients as a result of significant restenosis using standard pressure balloons, questioning whether the stents could be enlarged reducing the gradients, and noting the appearance of the intimal ingrowth after expansion. The diameter of the narrowest area within the stent was increased from 7±2 to 10±2 mm (52±47% enlargement; n=16; P<.0001; Fig 1A⇑), systolic pressure gradient decreased from 23±13 to 14.1±12.3 mm Hg (52±26% reduction; n=15; P<.0001; Fig 1B⇑), and right ventricular systolic pressure decreased from 41±14 to 37±9 mm Hg (4±35% reduction; n=8; P=NS; Fig 1C⇑). No stent demonstrated thrombosis formation, migration, aneurysm formation, or calcification.
Balloon angioplasty for pulmonary artery stenosis is not uniformly successful, with several recent reports suggesting an acute success rate in the range of only 50%, with clinical management favorably influenced in only 35% of patients.5 6 Furthermore, the noted incidence of recurrent narrowing after balloon angioplasty (about 17%) may be related to the elastic recoil properties of the pulmonary artery wall. To overcome this, frequent redilation and overdistention of the stenotic areas may be required and can contribute to procedural complications.4
The objective of intravascular stenting is to provide a framework to avoid vessel recoil and support the vessel wall, eliminating the stenosis. Anticipated full surface neoendothelialization, a process that develops rapidly within the first 3 weeks after implantation, then protects against thrombosis.7 Initial surface endothelialization evolves through various stages. Within days to weeks of implantation, a thrombotic layer covering the stent struts is progressively replaced by a fibromuscular tissue layer, forming a surface for neoendothelial ingrowth. The later progression of events is thought to decline with time, becoming quiescent years after placement.7 8
The efficacy of stent relief of such obstructions and the safety of implantation were discussed in a number of recent publications.2 3 4 One issue that remains unanswered, however, is the pattern of endoluminal growth in stents within the pulmonary circulation. Early experimental studies identified a uniform neoendothelial cell layer covering stents placed in piglet pulmonary arteries,1 but examinations were performed only 3 months after implantation. From one large previously reported series,4 only 1 of 25 recatheterized patients was found to have hemodynamic or angiographic evidence of restenosis (at the junction of two stents implanted in tandem), while the majority of stents showed very mild (1- to 2-mm) wall thickening; in addition, the site of maximal ingrowth was not detailed. From studies of stented adult coronary arteries, restenosis tissue consists of proliferating smooth muscle cells that grow over the stent wires and subsequently reendothelialize.9
This report supports previous early observations after stent implantation.3 4 In this setting, the acute success rate was 96%, and clinical management was favorably influenced in 95% of patients. Our observations suggest that such implants do positively affect patient treatment algorithms by alleviating symptoms, improving pulmonary hemodynamics, and enlarging branch vessels to the degree that surgical enlargement can be avoided. Furthermore, those patients with only mildly elevated right ventricular pressures (eg, unilateral obstruction) and pulmonary insufficiency might benefit from vessel enlargement, reducing the degree of regurgitation by increasing the perfused pulmonary vascular bed; in this small series, however, right ventricular end-diastolic pressure remained unchanged at follow-up. This could improve hemodynamics in the right side of the heart by decreasing ventricular volume and could improve exercise performance. Whether this is a reality requires further investigation in this population. In addition, in a majority of patients, anticipated surgery was avoided and symptoms were improved as a result of the procedure. Although small stents have limited dilation potential and increased risk of restenosis, they appeared useful in delaying surgery in younger patients for which additional lesions would require future surgical intervention. Considerable care must be exercised in the application of these smaller implants in that expansion, although beneficial in the short-term, has long-term consequences. In such situations, however, the stent could be either removed or opened lengthwise at surgery and patch-enlarged at a later date. Acute complications were few and could be managed medically; no emergency surgery resulting from the implant was required.
Medium-term angiographic follow-up demonstrated wall thickening within the stent lumen of various degrees, occurring primarily in areas where there was an abrupt variation in vessel-to-lumen diameter after implantation. With overdilation, vessel endothelium denuded and separated; cells within this injured area multiplied and migrated to replace cells within the denuded zone. Such stationary cell replication creates a hyperplastic regeneration zone.10 Finally, the difference in diameters between the stented and the nonstented vessels creates heightened shear forces that vary with the compliance of these two areas. High shear forces activate platelets, which adhere to the vessel wall and release mitogenic factors that stimulate subendothelial cells in the vicinity of the adjacent vascular wall to proliferate and migrate.10 In two cases in which no diameter differences between the stented and nonstented vessels were observed, a diffuse pattern of mild intimal growth was noted. This may suggest that when there is no flow acceleration (as when there are no diameter differences and no residual focal narrowing), the lumen narrowing will be minimal.
Stent implantation into the central right or left pulmonary artery may compromise flow to branch vessels arising at angles from the stented segments. Although flow reduction to such branches might be the result of the underlying disease (stenosis of the origin of the crossed side vessel before stent implantation), it more likely results from mechanical distortion created by the stent implantation. Experimental studies in stented pulmonary arteries from pigs and clinical observations would suggest that this is not related to endothelial growth over stent strut members or a strut thrombus obstructing flow1 (Fig 4⇓). Such findings may be related to entrapment of the side-branch orifices by the strut, acutely occluding or compromising the orifice by compression. In one case, flow to the branch pulmonary artery improved at follow-up, which suggested the possibility of vasospasm contributing to flow reduction.
A limitation of this study was the diversity of congenital heart lesions addressed by the implant that may affect the pulmonary vascular bed differently. In addition, the follow-up time (median, 15 months) may be too short, and changes in the intimal growth patterns or the hemodynamics may evolve further. Although objective data were gathered prospectively, clinical decisions regarding patient management in this population were examined retrospectively, hence introducing bias into the study.
- Received December 21, 1994.
- Revision received February 2, 1995.
- Accepted February 19, 1995.
- Copyright © 1995 by American Heart Association
O’Laughlin MP, Perry SB, Lock JE, Mullins CE. Use of endovascular stents in congenital heart disease. Circulation. 1991;83:1923-1939.
O’Laughlin MP, Slock MC, Grifka RG, Perry SB, Lock JE, Mullins CE. Implantation and intermediate-term follow-up of stents in congenital heart disease. Circulation. 1993;88:605-614.
Schatz RA, Baim DS, Leon M, Ellis SG, Goldberg S, Hirshfeld JW, Cleman MW, Cabin HS, Walker C, Stagg J, Buchbinder M, Teirstein PS, Topol EJ, Savage M, Perez JA, Curry RC, Whitworth H, Sousa JE, Tio F, Almagor Y, Ponder R, Penn IM, Leonard B, Levine SL, Fish RD, Palmaz JC. Clinical experience with the Palmaz-Schatz coronary stent: initial results of a multicenter study. Circulation. 1991;83:148-161.
Herring BM. Endothelial seeding of blood surfaces. In: Wright CB, ed. Vascular Grafting. Littleton, Mass: PSG Publishing; 1983:275-314.