Author + information
- Received November 11, 2015
- Revision received January 25, 2016
- Accepted February 11, 2016
- Published online June 13, 2016.
- Shyam K. Sathanandam, MDa,∗ (, )
- T.K. Susheel Kumar, MDb,
- Deepthi Hoskoppal, MDc,
- Lauren M. Haddad, MDa,
- Saradha Subramanian, MDa,
- Ryan D. Sullivan, DVMd,
- David Zurakowski, PhDe,
- Christopher Knott-Craig, MDb and
- B. Rush Waller III, MDa
- aDepartment of Pediatrics, Division of Pediatric Cardiology, University of Tennessee Health Science Center, Memphis, Tennessee
- bDepartment of Surgery, University of Tennessee Health Science Center, Memphis, Tennessee
- cDepartment of Pathology, University of Tennessee Health Science Center, Memphis, Tennessee
- dDepartment of Veterinary Medicine, University of Tennessee Health Science Center, Memphis, Tennessee
- eDepartment of Biostatistics, Harvard Medical School, Boston, Massachusetts
- ↵∗Reprint requests correspondence:
Dr. Shyam Sathanandam, University of Tennessee Health Science Center, 848 Adams Avenue, Memphis, Tennessee 38103.
Objectives This study sought to determine the feasibility and safety of unzipping small-diameter stents (SDS) in a growing animal model.
Background SDS implanted to relieve stenosis of blood vessels in infants may result in refractory stenosis as the child grows. If stents can be longitudinally fractured—unzipped—then the target vessel can potentially be redilated to the eventual adult vessel diameter.
Methods Fifty stents (diameter 4 to 7 mm) were implanted in 5 neonatal piglets (mean age and weight = 1.5 weeks and 3.4 kg). Pre-mounted coronary (CS) (n = 24), biliary (BS) (n = 14), nitinol (NS) (n = 3), and renal stents (RS) (n = 9) were implanted in pulmonary arteries (n = 13), systemic arteries (n = 25), and systemic veins (n = 12). Three months later (median weight = 32 kg), unzipping was attempted by dilating the stents.
Results All CS and RS unzipped at twice their nominal diameter with <20% shortening. None of the NS unzipped. The BS shortened the most (∼40%), with only 69% of the stents unzipping. Stainless steel CS and RS with an open cell design were significant predictors (p ≤ 0.01) for unzipping. On histopathology, unzipping of the BS caused the most medial dissection and vessel wall injury, while unzipping of the CS caused the least.
Conclusions Unzipping of small-diameter CS and RS implanted in systemic and pulmonary vessels is more feasible than the BS and NS. This study may encourage the implantation of small stents in infant blood vessels and aid in selection of appropriate stent type.
Endovascular stent implantation can be used to treat stenosis of blood vessels in infants and young children (1–9). The circumferential size of the implanted small-diameter stent (SDS) does not adapt to the growth of the vessel, resulting in refractory stenosis (RSS) (10) and therefore limiting the use of stents in infants with stenotic vessels (11). If an implanted SDS can be fractured along its length, then the stenotic segment can potentially be restented with a larger stent, which should be amenable to further dilation. The deliberate fracturing of a stent, such that the fracture points on each strut are aligned, leading to a linear fracture along the length of the stent is termed unzipping (12). Unzipping of SDS (Figure 1, Online Videos 1 and 2) may prove valuable in treating stenosis of growing blood vessels. We have described this technique by an in vitro experiment on several commercially available SDS (12). The technique of unzipping involves serial dilations of stents using slow inflation with ultra high pressure (UHP) angioplasty balloons, with sequential small increments in balloon size to prevent shortening of the stent. Stainless steel CS were found to be best suited for unzipping (12). However, the effect of unzipping endothelialized stents using UHP balloons on blood vessels of growing children is unknown. The primary objective of this study was to determine the feasibility and safety of unzipping SDS in a growing animal model. The secondary objective was to determine stent properties that favor unzipping among various stent types.
Approval for this study was obtained from the University of Tennessee Health Science Center Institutional Animal Care and Use Committee of University of Tennessee Health Science Center and the Medical Education and Research Institute in Memphis, Tennessee. The domestic farm pig, which has a rapid growth rate, was chosen as the model for this chronic animal study. The study design included implantation of 10 stents per piglet and redilation of the stents when the piglet grew to 10 times the weight at original implantation.
Fifty SDS (median diameter 4.5 mm, range 4 to 7 mm) were implanted in 5 neonatal piglets (mean age 1.5 weeks, range 1 to 2 weeks; mean weight 3.4 kg, range 3 to 3.6 kg). Aspirin (40.5 mg/day, oral) was started 3 days prior to the procedure. The piglets were pre-medicated with 6 mg/kg telazol, intubated, maintained on isoflurane with neuromuscular blockade, mechanically ventilated with peak inspiratory pressure of 20 cm H2O, an inspired oxygen fraction of 1 and a rate of 25 to 30 breaths/min to achieve a normal pH and arterial CO2 tension. Intravenous cefazolin (25 mg/kg) was administered. Universal sterile precautions were applied. The right internal jugular vein and the right carotid artery were accessed via cut-down. A 7-F sheath was introduced into the vein, and a 5-F sheath was introduced into the artery. Continuous electrocardiography, pulse oximetry, and invasive blood pressure monitoring through the side arm of the arterial sheath were performed. Arterial blood gasses and activated clotting times were measured every 30 min. An intravenous heparin bolus of 200 U/kg was administered with additional boluses given as needed to maintain activated clotting times between 200 to 250 s. Ten stents were implanted under fluoroscopic guidance in various locations in each piglet (Figure 1, Online Video 3). Angiograms were obtained before and after stent implantation in an Artiz Zee Zeego suite (Siemens, Munich, Germany). Diameters of the blood vessel prior to stent implantation and of the stent post-implantation were measured on the basis of the angiograms. Following the procedure, the piglets were extubated and transferred to a warming mat in a pen located in a temperature-controlled area. They were maintained on oral cephalexin for 2 days post-procedure.
Six types of pre-mounted stents were implanted, the properties of which are described in Table 1. They were grouped into 4 categories: CS, biliary (BS), renal (RS), and nitinol stents. Only 1 type of CS, the VeriFLEX stent (n = 24, median diameter 4 mm; Boston Scientific, Natick, Massachusetts) was implanted, as this stent has been best suited for unzipping in vitro, predictably fracturing at twice its nominal diameter with minimal shortening. Two types of BS were implanted; Palmaz Genesis and Blue (n = 14, median diameter 6 mm; Cordis, Johnson & Johnson, Piscataway, New Jersey). In vitro, these stents had unzipped at 3 times their nominal diameter, but underwent significant shortening. Three nitinol Protégé self-expanding stents (SES) (ev3, Plymouth, Minnesota) were implanted (median diameter 6 mm). These stents did not unzip in vitro, but developed disorganized fractures at twice nominal diameter. Two types of RS were implanted; Express SD (Boston Scientific) and Formula 414 (n = 9, median diameter 6 mm; Cook Medical, Bloomington, Indiana). These stents had unzipped at twice nominal diameter in vitro. The uneven distribution of stent types that were implanted was secondary to match the vessel size to the nominal stent diameter.
The blood vessels in which the stents were implanted were grouped as pulmonary arteries (PA), systemic arteries, and systemic veins. A total of 13 stents were implanted in the PA (CS = 9, BS = 3, and RS = 1), 25 stents in systemic artery (aorta and common iliac arteries; CS = 10, BS = 7, SES = 1, and RS = 7), and 12 stents in systemic veins (inferior vena cava and common iliac veins; CS = 5, BS = 4, SES = 2, and RS = 1). Figure 1 demonstrates implantation of stents in 3 individual piglets. Complete physical exams with assessment for weight gain were performed weekly. Sedated echocardiograms and fluoroscopy were performed once every 3 weeks. To assess for blood vessel growth, the diameters of the blood vessels adjacent to the stents were measured and Doppler gradients across the stents were recorded. The piglets were maintained on aspirin (81 mg/day, oral), throughout the growth phase.
Unzipping of stents
Three months after stent implantation, a second catheterization was performed with anesthesia, medications, and monitoring as previously described. At the time of the second catheterization, the piglets had achieved a median weight of 32 kg (range 28 to 35 kg). An 8-F sheath and a 7-F sheath were introduced via cut-down into the left internal jugular vein and the left carotid artery, respectively. The percentage in-stent stenosis (ISS) was calculated from angiograms. UHP balloons (Conquest or Dorado up to 10 mm, and Atlas for >10 mm; Bard Peripheral Vascular, Tempe, Arizona) were used to dilate the stents serially, first with 1 mm increments in balloon diameter until 6 mm, followed by 2 mm increments until the stents either unzipped (Figure 2, Online Video 4) or shortened without unzipping (napkin-ringed). Pressure gradients were measured by catheter pullback before and after the stents were dilated. A hemodynamically relevant stenosis was judged by a systolic pressure gradient >20 mm Hg in arteries and a mean gradient >5 mm Hg in veins. Angiograms were performed after each dilation to evaluate for complications. The diameter and length of each stent after each dilation was measured by fluoroscopy until the outcome was achieved. The percentage decrease in length (Δlength) to the percentage increase in diameter (Δdiameter) was calculated (dL/dD ratio) for each stent following unzipping or napkin-ringing. The ultimate fracture diameter was represented as a multiple of the nominal diameter of the stent (X-nominal). The hoop stress (σθ) for each dilation was calculated. As stent fracturing is a dynamic process and the σθ at the exact moment of stent fracture could not be calculated, the mean σθ of the penultimate and ultimate dilations from a mathematical model was calculated as the yield stress (σy) for the stent. RSS was defined as the ratio of the final diameter of the stented segment of the vessel compared to adjacent, normal vessel diameter. If over-dilation of the stent resulted in the stented segment being larger than the adjacent, nonsegmented vessel, then RSS was considered to be 0% and not expressed as a negative percentage. RSS and ISS ≥20% were considered significant.
At the end of the procedure, the piglets were euthanized with IV pentobarbital and phenytoin. Necropsy was performed immediately to evaluate for vessel contusion, rupture, aneurysm or bleeding. The stented vessels were excised, flushed, and fixed in 10% neutral buffered formalin for 30 min at a pressure equal to the former intraluminal pressure by mounting on a balloon the size of the ultimate unzipping diameter in vivo. In vitro macroscopic examination under fluoroscopy was performed to confirm stent unzipping (Figure 2).
One set of vessels was analyzed with the stent in situ and another set was analyzed following removal of the stent. The in situ stent specimens were processed in plastic, dehydrated in alcohol, cleared with Xylene, infiltrated with Epon spur, cut at 5 μm and stained with Hematoxilin and Eosin, Trichrome and Von Geison stain (Elastin). The specimens with the stents removed were processed similarly, embedded with paraffin and the same stains were used. Medial dissection was defined as dissection of the tunica media circumferentially along the external elastic lamina. No dissection was scored as 0, localized dissection as 1, 25% to 50% dissection as 2, 50% to 75% as 3 and >75% as 4. Vessel wall injury was scored 1 for rupture of the internal elastic lamina, 2 for injury to the media, 3 for injury to the external elastic lamina and 4 for rupture extending up to the adventitia. Two sets of controls were used for comparison. Normal blood vessels from piglets of the same age and size were used to compare the degree of neo-endothelial proliferation. To differentiate injuries caused by unzipping of stents compared to regular balloon angioplasty, nonstented controls of similar blood vessel types that were subjected to UHP balloon angioplasty and were used as the second set of controls.
A hierarchical approach using generalized estimating equations in analyzing the 50 stents in the 5 piglets was applied, as the stents are not independent but represent repeated or multiple measurements. This statistical approach was applied to account for the correlated data with the piglet ID number used as the subject and the stent itself as the repeated measures term in the model. A random effects linear model was used when analyzing the continuous outcome variables such as end diameter and length and % change in diameter and length, and yield stress; we fit a logistic regression model with a binomial distribution when assessing the binary outcome of “unzipping” and again treated the piglet as the subject factor with stent as the replicate or nested term within animal. Continuous data are reported as median (interquartile range) and categorical data as numbers with percentages. Chi-square tests were performed for categorical data and Mann-Whitney U-tests were performed for continuous data. Sensitivity and specificity were calculated for dL/dD ratio, X-nominal, stent alloy density, and strut thickness. Receiver-operating characteristic (ROC) curve analysis was performed to determine area under the curve (AUC) to assess utility of variables in differentiating between stents that unzipped and those that failed to unzip. Optimal cut-off values were determined by maximizing Youden’s J-index for dL/dD ratio, X Nominal, stent alloy density, strut thickness, and σy. Groups on the basis of stent types and blood vessel types were compared for outcomes. An analysis of variance was used to analyze the differences between group means and their associated variables and outcomes. In addition, the Bonferroni method was used for correcting for multiple comparisons. Histology grade, Δlength, dL/dD ratio, X-nominal, ISS, RSS, and σy were compared between stent types and vessel types using an analysis of variance. A p value <0.05 was considered statistically significant. Statistical analysis was performed using the IBM SPSS software package (version 21.0, IBM, Armonk, New York).
The type of blood vessel in which each of the 50 stents was implanted and the eventual outcome are as listed in Table 1. At the time of recatheterization, there were only minimal pressure gradients across the stents (mean systolic gradient <20 mm Hg for all stent types; p = 0.76). Baseline angiograms demonstrated mean ISS <10% for all stent types (p = 0.22) and all vessel types (p = 0.38) as seen in Tables 2, 3, and 4. Unzipping was attempted on 45 of the 50 stents; 38 were successfully unzipped (Table 2). Unzipping was not attempted in 5 stents; 2 were crushed during implantation (1 CS in a PA branch and another CS in an iliac vein while stenting an adjacent PA branch and iliac artery, respectively) and 3 stents were inaccessible (a CS and BS in PA branches, and 1 CS in an iliac artery) due to the operator’s technical difficulties. Attempted unzipping of all CS and RS was successful (p = 0.004, in comparison to other stent types). Of the BS, all Palmaz Blue stents unzipped; 6 of the 10 Genesis stents unzipped. Four of the Genesis stents shortened without fracturing. None of the Protégé stents unzipped, with 1 developing disorganized fractures. There was no difference in the outcome on the basis of the type of vessel (p = 0.59) with 80% of the PA stents, 88% of the systemic artery stents, and 72% of the systemic vein stents being successfully unzipped.
Open cell stents were more easily unzipped compared to the closed cell design stents (91% vs. 69%; p = 0.01). Stainless steel stents were more amenable to unzipping than nickel titanium stents (90% vs. 0%; p = 0.012). The nominal diameter and length of the stents, the density of the stent alloy, the stent strut thickness, the yield stress, the degree of ISS, the vessel type, or the presence of a gradient through the stent prior to dilation did not influence the ability to unzip (Table 3). Stents that unzipped did so with fewer number of balloon dilations (median: 2 vs. 4; p = 0.005), with a smaller ultimate fracture diameter (median X-nominal: 2 vs. 2.8; p < 0.001). Stents that failed to unzip shortened significantly more than those that unzipped (median dL/dD ratio: 0.36 vs. 0.09; p = 0.03). The CS shortened the least (% change in length: 7 ± 3%), while the BS shortened the most even when they unzipped (% change in length: 42 ± 24% for Genesis and 34 ± 16% for Blue; p < 0.001) as demonstrated in Table 2. Therefore, the VeriFLEX stents had the smallest dL/dD ratios, although the Genesis stents had the greatest (mean dL/dD ratio: 0.09 vs. 0.65; p < 0.001). Figure 3 shows the shortening of the various stents as they were dilated beyond their nominal diameters. Although the Protégé stents shortened the least, it did not unzip even when dilated with a balloon to 3 times the nominal diameter of the stent. The CS and RS unzipped at twice their nominal diameters, although the BS did so at about 2.5 times their nominal diameter. The Genesis and the Protégé stents, although larger than the CS, had more RSS (mean RSS: 6% and 8% vs. 0%; p = 0.01). However, there was no significant RSS of any blood vessel or stent after unzipping the stent (Tables 2, 3, and 4). The mean post-intervention systolic pressure gradient was <10 mm Hg for all stent types and vessel types.
The predictive accuracy of various stent properties for unzipping are listed in Table 5. The stent alloy density had an inverse correlation (ρ = –0.43; p = 0.01) to the ultimate fracture diameter of the stent (Figure 4A). The number of dilations required to unzip a stent had a strong positive correlation to the ultimate fracture diameter (ρ = 0.62; p < 0.001), as shown in Table 6. The σy was not different for stents that unzipped versus those that did not (median σy 72 vs. 90 MPa; p = 0.29). A cut-off of σy of 80 MPa had the highest predictive accuracy for unzipping (sensitivity 60%, specificity of 67%; 95% CI as listed in Table 5; receiver-operating characteristic curves shown in Figure 4B).
There were no complications during stent implantation. However, there were 2 major and 2 minor complications while attempting to unzip stents. All complications resulted from oversizing of balloons with respect to the adjacent, normal vessel size. The minor events included contained dissections in 2 iliac arteries. The major complications included extravasation of contrast seen on follow-up angiograms from vessel injury secondary to overdilation of 2 CS to 8 mm when the vessel diameter was 5 mm.
Fluoroscopy, macroscopic, and histopathologic examination
In vitro fluoroscopic examination of all stented vessel segments confirmed unzipping of 35 stents. This was seen as a linear fracture along the entire length of the stent with diverging stent struts. On macroscopic exam, the stent network was irregular in one-half of all CS and a third of all RS suggestive of loss of radial strength of the stents (Figure 2). The BS and Protégé stents maintained their structural integrity. On microscopic exam, only minimal neointimal proliferation was observed for all stent types and for all vessel types. The mean thickness of the neointima was ≤100 μm for all CS and approximately 150 μm for the RS, BS, and nitinol SES with accentuation at the site of the struts in contrast to nonstented controls. Microscopic hemosiderin in the media was observed in 2 iliac arteries reflecting old hemorrhage, possibly secondary to localized vessel trauma during stent implantation. Unzipping or attempted unzipping of the Genesis stents had the highest medial dissection scores, although the VeriFLEX stents had the least (2 ± 1 vs. 1 ± 0.5; p = 0.03) (Figure 5). Among the RS, unzipping of the Express stents created more medial dissection than the Formula stents (mean score: 1.5 vs. 1). Similarly, dilation of the Genesis and the Express stents created greater vessel wall injury than the VeriFLEX and the Formula stents (mean score: 3 vs. 2; p = 0.01) (Figure 5). The medial dissection and vessel wall injury scores for vessels with unzipped stents (Figure 6) compared to nonstented controls of similar blood vessel types subjected to UHP balloon angioplasty alone were similar (mean medial dissection score: 1.5 vs. 1.5; p = 0.64; mean vessel was injury score: 2 vs. 2; p = 0.38). No inflammatory cells were identified in the vessel wall for any stent by immunohistochemical staining for CD-3 (T-type lymphocytes), CD-79 (B-type lymphocytes), CD-68 (macrophages), or MAC (granulocytes, monocytes, tissue macrophages).
Transcatheter treatment of both PA stenosis and coarctation of the aorta in neonates and infants remains a challenging clinical problem primarily because of the discrepancy between the size of the target blood vessel and the outer diameter of the balloons and sheaths necessary to deliver a stent that can ultimately reach large diameters (3–6,8–11). In addition, SDS that can be implanted easily in stenotic blood vessels of infants and small children cannot be redilated to adult vessel diameter leading to fixed or RSS (10). The use of resorbable stents offers an attractive option for children (13,14). Several investigators have conceptualized breakable stents and open-ring stents for implantation in small children (15–18), which would allow for dilation and possibly restenting of these vessels to larger sizes at a later stage when the child has grown. In an in vitro study describing the technique to unzip stents with a longitudinal fracture, we determined that CS, especially the VeriFLEX stent, was best suited for unzipping. The current study was therefore aimed to address the feasibility and safety of unzipping SDS in vivo using a growing piglet model. The neonatal piglet is similar in size to human neonates, with blood vessels that measure similar to human neonatal blood vessels. At 3 to 4 months of age, the piglet weighs about 30 kg comparable to an 8- to 10-year-old human. The blood vessels of a pre-adolescent human are 2 to 3 times the size of the neonatal vessels. In vitro testing established that most SDS unzip at 2 to 3 times their nominal diameter. Therefore, we conjectured that a 30 kg pig’s blood vessels would have doubled or tripled in size, making the implanted stents amenable to unzipping.
A thorough determination of successful unzipping was performed using fluoroscopic, gross, and microscopic inspection. Unzipping was successful in 38 of the 45 stents in which it was attempted (85% feasibility). All of the VeriFLEX, Express, Formula, and Blue stents unzipped. Only 60% of the Genesis stents and none of the Protégé stents were feasible for unzipping, as in the in vitro study. The CS and RS unzipped predictably at twice the nominal diameter, while the BS did so when dilated to 2.5 times their nominal diameter. The CS and RS also shortened less than the BS when they were dilated. On the basis of these results, optimal cut-off points for stents to unzip included, a strut thickness of 108 μm, alloy density of 7.7 g/cm3, dL/dD ratio of 0.12, and yield stress of 80 MPa.
There were no procedural complications in this experiment during stent implantation despite an average of 10 stents being implanted in each piglet in multiple vessels. As per the study protocol, we intended to unzip the stents when the pigs were approximately 30 kg, expecting the blood vessels to have at least doubled in size. However, the segments of the blood vessels adjacent to the stented segments had not doubled or tripled the size by the time of the second procedure for stent dilation as expected. The implantation of several stents along the length of each blood vessel in close proximity to each other likely contributed to restricting the growth of the blood vessels. Hence, most stents ended up being dilated to diameters greater than the size of the adjacent, nonstented blood vessel to determine feasibility of unzipping. The described complications were all due to this unplanned overdilation of the adjacent normal vessel. In a clinical setting, where stents are only dilated to match the adjacent, normal blood vessel diameter, this would not be expected to happen. Therefore, these complications can possibly be avoided by attempting to unzip a stent only when the adjacent vessel is larger than the original diameter of the stent. No stents in this study developed significant ISS. We hypothesize the lack of ISS was because the stented vessels were normal, nonstenotic vessels. Typically, ISS is proportional to the degree of stenosis and the degree of stent-related inflammatory reaction. By immunohistochemistry, no inflammatory response was noted either.
Serial dilation with slow inflations using small increments in balloon size was the best technique for both unzipping stents and preventing rapid stent shortening or napkin ringing. The use of noncompliant, UHP balloons layered with woven ultra high molecular weight polyethylene such as the Conquest, Dorado, and Atlas (BARD peripheral vascular) were most effective. The pressure exerted by the balloon is less important than the diameter of the balloon. It was observed that, as stents were dilated, there was minimal recoil with each dilation. The recoil was exaggerated when the stent was unzipped. We hypothesize that unzipping makes the stent lose its radial strength. However, the stent does not collapse, possibly secondary to endothelization. The recoil was prominent with CS and in systemic arteries. Although this experiment was not designed to implant a second stent within the unzipped stent, it is our conjecture that a large-diameter stent could easily be implanted to the diameter of the larger vessel, thus providing radial support and allowing for subsequent dilations to the eventual adult vessel diameter as the child grows. Larger-diameter stent implantation within the unzipped stent may be important for its continuing intentional enlargement, as the vessel will be unsupported and further dilation of the vessel could lead to that segment of the vessel to get weaker and leading to aneurysm formation.
There is always a chance for later development of pseudoaneurysms at these localized dissection sites. Microscopy only looks at a thin section of the vessel and not the entire length of the vessel and therefore other injuries could have been missed. Longer-term survival studies are therefore necessary to determine other potential complications that could develop in the future. Although medial dissection and vessel wall injury have been well described secondary to balloon angioplasty of stenosis in PA and coarctation of aorta (19,20), the histopathologic impact of using UHP balloons to dilate normal blood vessels has not been described. Therefore, we subjected a second control set of normal blood vessels to UHP angioplasty to compare with the stented blood vessels. We found that the degree of medial dissection and vessel wall injury with stent unzipping was minimal and not excessive compared to balloon dilation with UHP balloon alone. No vessel wall avulsion was noted. However, the mere presence of vessel wall injury, albeit mild, makes it difficult to conclude that unzipping a stent in vivo, although feasible, is entirely safe. As almost all stented vessels were overdilated in relation to the adjacent, nonstented segment, it is possible that the injury to the vessels was exaggerated and could be minimized if overdilation is avoided. Manufacture of the VeriFLEX stent has recently been discontinued. However, any other open cell, stainless steel CS may be amenable for unzipping and can be considered for stenting of stenosis in blood vessels of infants that are smaller than 5 mm in diameter. The lower density of the stainless steel alloy, the open cell design and the small strut thickness allows for unzipping of CS. The Express stent can be used for treating larger vessels. These stents can potentially be redilated and unzipped in the future and a larger diameter stent can be implanted within it and serially dilated to achieve the desired adult vessel diameter. We are subsequently conducting further long-term survival experiments aimed to determine the feasibility of such a strategy.
Unzipping SDS in vivo is most feasible with open-cell stainless steel CS and RS such as the VeriFLEX, Express, and Formula stents. The technique of serial dilations with slow inflations using small increments in balloon diameter is most effective in unzipping stents. Further experiments are required to establish the safety of this technique. If proven safe, the technique of unzipping SDS may be useful in treating stenosis of infant blood vessels.
WHAT IS KNOWN? SDS used to treat stenosis blood vessels of infants and small children cannot be redilated to adult vessel diameter leading to fixed or refractory stenosis.
WHAT IS NEW? If an implanted SDS can be fractured along its length—“unzipped”—then the stenotic segment can potentially be restented with a larger stent, which should be amenable to further dilation. This study determines the feasibility and safety of unzipping commercially available SDS in the blood vessels of a growing piglet model and establishes the stent properties that are ideal for unzipping.
WHAT IS NEXT? This study could encourage the implantation of small stents in infant blood vessels and aid in selection of appropriate stent type. The study could also help in designing an “unzippable” stent in the future for use in neonates, infants, and young children.
For supplemental videos and their legends, please see the online version of this article.
All authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- biliary stent(s)
- coronary stent
- in-stent stenosis
- pulmonary artery
- renal stent(s)
- refractory stenosis
- systemic artery
- small-diameter stent(s)
- self-expanding stent(s)
- ultra high pressure
- Received November 11, 2015.
- Revision received January 25, 2016.
- Accepted February 11, 2016.
- American College of Cardiology Foundation
- O'Laughlin M.P.,
- Perry S.B.,
- Lock J.E.,
- Mullins C.E.
- Mullins C.E.,
- O'Laughlin M.P.,
- Vick G.W.,
- et al.
- O'Laughlin M.P.,
- Slack M.C.,
- Grifka R.G.,
- Perry S.B.,
- Lock J.E.,
- Mullins C.E.
- Shaffer K.M.,
- Mullins C.E.,
- Grifka R.G.,
- et al.
- McMahon C.J.,
- El-Said H.G.,
- Grifka R.G.,
- Fraley J.K.,
- Nihill M.R.,
- Mullins C.E.
- Ebeid M.R.,
- Prieto L.R.,
- Latson L.A.
- Ward C.J.,
- Mullins C.E.,
- Nihill M.R.,
- Grifka R.G.,
- Vick G.W.
- Redington A.N.,
- Hayes A.M.,
- Ho S.Y.
- Hatai Y.,
- Nykanen D.G.,
- Williams W.G.,
- Freedom R.M.,
- Benson L.N.
- Rosenthal E.,
- Qureshi S.A.
- Sigler M.,
- Schneider K.,
- Meissler M.,
- Koenig K.,
- Schneider M.B.
- Ing F.F.,
- Fagan T.E.,
- Kearney D.L.
- McElhinney D.,
- Marshall A.,
- Schievano S.