Long-term endothelial dysfunction is more pronounced after stenting than after balloon angioplasty in porcine coronary arteries
Experimental Studies
Abstract
Objectives. To compare percutaneous transluminal coronary angioplasty (PTCA) and stent implantation with respect to the long-term changes they induce in the newly formed endothelium in porcine coronary arteries by studying both morphological and functional parameters of the endothelium at 2 weeks and 3 months after intervention.
Background. Problems affecting PTCA or stent implantation have been overcome to a large extent by means of better techniques and the availability of new drugs. Late problems, however, still exist in that restenosis affects a large number of patients. With an increasing number of patients being treated with stents, the problem of in-stent restenosis is of even greater concern, as this seems difficult to treat. A functional endothelial lining is thought to be important in controlling the growth of the underlying vascular tissue. We hypothesized that the enhanced neointimal hyperplasia observed after stenting is associated with a more pronounced and prolonged endothelial dysfunction.
Methods. Arteries were analyzed using a dye-exclusion test and planimetry of permeable areas. Thereafter, the arteries were processed for light and scanning electron microscopy for assessment of morphology and proliferative response.
Results. Leakage of the endothelium for molecules such as Evans blue-albumin as well as prolonged endothelial proliferation is observed as late as 3 months after the intervention, and is more pronounced after stenting. Permeability is associated with distinct morphologic characteristics: endothelial retraction, the expression of surface folds, and the adhesion of leukocytes.
Conclusions. Stenting especially decreases long-term vascular integrity with respect to permeability and endothelial proliferation, and is associated with distinct morphologic characteristics.
Introduction
Problems affecting percutaneous transluminal coronary angioplasty (PTCA) or stent implantation such as acute closure, stent thrombosis, or bleeding complications due to the stringent anticoagulation protocols from the early days of stent use have been overcome to a large extent by means of better techniques and the availability of new drugs (1–4). Late problems, however, still exist in that restenosis affects approximately 15–20% of patients after primary stenting and 30–50% after PTCA alone. With an increasing number of patients being treated with stents (up to 50%), the problem of in-stent restenosis is of even greater concern as this seems difficult to treat.
Every intervention aimed at increasing lumen size inevitably leads to damage of the vessel wall. The subsequent healing response, necessary to pacify the inflicted wound, triggers the growth of a neointimal thickening (NI). Excessive growth of this NI is only one of the contributors to restenosis after PTCA, but is likely the sole responsible factor when dealing with in-stent restenosis. PTCA mechanically damages the endothelial cells and induces endothelial dysfunction that persists for several weeks both in the laboratory animal and in patients (5–7). A functional endothelial lining is important in controlling the growth of the underlying vascular tissue (8), and it may well be that the enhanced neointimal hyperplasia observed after stenting is associated with a more pronounced and prolonged period of endothelial dysfunction.
The objective of the present study was, therefore, to compare PTCA and stent implantation with respect to the long-term changes they induce in the newly formed endothelium in porcine coronary arteries. We therefore studied both morphological (as assessed by general pathology, morphometry, histochemistry, electron microscopy [EM]) and functional parameters of the endothelium (barrier function and proliferative status).
Methods
Animal care
Animal care
Experiments were performed under the regulations of the animal care committee of the Erasmus University Rotterdam and in accordance with the “Guide for the Care and Use of Laboratory Animals” (9).
Animal preparation
Experiments were performed in Yorkshire pigs (25–30 kg; HVC). After an overnight fast, the animals were sedated with 20 mg/kg ketamine hydrochloride. After induction of anesthesia with thiopental (12 mg/kg) and after endotracheal intubation, the pigs were connected to a ventilator that administered a mixture of oxygen and nitrous oxide (1:2 [vol/vol]). Anesthesia was maintained with 0.5– 2.5 vol% isoflurane. Antibiotic prophylaxis was administered by an intramuscular injection of 1,000 mg of a mixture of procaine penicillin-G and benzathine penicillin-G.
Under sterile conditions, an arteriotomy of the left carotid artery was performed and a 9-F introduction sheath was placed. Then 10,000 IU heparin sodium were administered followed by left coronary angiography using the nonionic contrast agent iopamidol (Iopamiro 370) after intracoronary administration of 1 mg isosorbide dinitrate.
Coronary interventions
From the angiograms (analyzed on-line using a quantitative coronary angiography analysis system), arterial segments of 2.5–3.5 mm in diameter were selected in the left anterior descending and/or left circumflex coronary arteries. Typically, balloon sizes were chosen 0.2–0.5 mm larger than the recipient artery. The stents (PS 153, Palmaz-Schatz Coronary Stent; JJIS, and Wiktor stent; Medtronic) were placed as described before (10,11). PTCA was performed in a similar way, using identical inflation parameters. After repeat angiography of the treated coronary arteries, the guiding catheter and the introducer sheath were removed, the arteriotomy was repaired, and the skin was closed in two layers. The animals were then allowed to recover from anesthesia.
Experimental groups and follow-up
Interventions were performed in four groups of animals, as shown in Table 1. In groups 1 and 2 (a subset of animals from a previously published study [10]) the animals received a Palmaz-Schatz stent only and were followed for 4 and 12 weeks, respectively, to assess: 1) the “molecular window,” i.e., to which extent the barrier function of the endothelial lining was impaired, and 2) whether this “window” of permeability changed in time. In groups 3 and 4, both PTCA and stent implantation were performed in each animal. These animals were followed for 2 and 12 weeks to assess the morphologic determinants correlating with the impaired barrier both in morphologically immature and mature endothelium, and to study differences between the two types of intervention.
| Group | Intervention | Follow-up | n | Molecular Weight | Study Objective |
|---|---|---|---|---|---|
| 1 | Palmaz-Schatz stent | 4 weeks | 10 | 1 and 70 kD | Window of permeability |
| 2 | Palmaz-Schatz stent | 12 weeks | 10 | 1 and 70 kD | Window of permeability |
| 3A | Wiktor stent | 2 weeks | 9 | 70 kD | Morphologic determinants |
| 3B | PTCA | 2 weeks | 9 | 70 kD | Morphologic determinants |
| 4A | Wiktor stent | 12 weeks | 5 | 70 kD | Morphologic determinants |
| 4B | PTCA | 12 weeks | 5 | 70 kD | Morphologic determinants |
Assessment of cell proliferation
To assess the proliferative response to stent implantation and PTCA in comparison with control coronary arteries, five animals each in groups 3A, 3B, 4A, and 4B were given three intramuscular injections of BrdU (Sigma Chemical Co.) at 100, 50, and 50 mg/kg at 8-h intervals, starting 24 h before sacrifice. Using light microscopy (LM), the total number and number of BrdU-positive cells were counted for each section in several high-power fields both proximal and distal in the treated arteries. The right coronary artery served as a control.
Assessment of intimal permeability at follow-up
In this test (Fig. 1, dye-exclusion test) Evans blue (EB) (Sigma Chemical Co.) was used in two configurations (12,13).
EB can be administered both intravenously and intracoronarily. Intravenous administration results in the spontaneous binding of EB to albumin, and subjection of the arterial wall to the 70-kD large complex. Intracoronary administration after a saline flush to remove serum proteins results in subjection of the arterial wall to the smaller 1-kD molecule. Blue staining of the arterial wall indicates a breach in the luminal barrier.
EB-albumin
To subject the arteries to the large molecular marker (70 kD), 300 mL of EB in saline (0.3% [wt/vol]) was administered intravenously, to allow for EB binding to albumin. The infusion was given for 30 min, and then 1 h was allowed for recirculation of the EB-albumin complex.
Binding control
During and after the EB infusion, arterial blood samples were taken, proteins precipitated with tri-chloric acid (final concentration 20%), and then spun down to check the supernatant for unbound dye.
EB-saline
To subject the arteries to the small molecular marker (1 kD), 300 mL of EB in saline (0.3% [wt/vol]) was administered directly into the coronary circulation after a saline flush. After completion of the EB infusions, the coronary arteries were flushed with approximately 300 mL saline before pressure fixation in situ (approximately 100 mm Hg) with 500 mL 4% buffered formaldehyde. Then the heart was excised, and the treated and control coronary arteries (not treated with balloon or stent) were dissected from the epicardial surface.
Macroscopic assessment
The excised treated and control coronary arteries were opened longitudinally and checked under a dissection microscope for penetration of the blue dye. The arteries were documented on film and used for planimetric analysis of the permeable areas. Thereafter, both proximal and distal areas of the specimen were divided for EM and LM.
Routine histology
To check for abnormal vascular reactions to the interventions and for a general assessment of the histological appearance, all specimens were processed for routine histology as described before (10). Sections were stained with hematoxylin-eosin as a routine stain and resorcin-fuchsin as a collagen and elastin stain.
Histochemistry
Lectin- and immunocytochemistry
Lectin- and immunocytochemistry
This was performed to confirm the identity of the endothelium and smooth muscle cells as described before (14).
Detection of BrdU incorporation
After acid DNA denaturation and elimination of endogenous tissue peroxidase activity, rehydrated paraffin sections were exposed to mouse anti–BrdU antibody (Becton Dickinson & Co.), dilution 1:80, to detect BrdU-positive cells. As a second antibody, HRP-labeled rabbit anti–mouse antibody (Dakopatts) was used, with 670 μg/ml Di Amino Benzidine (Sigma Chemical Co.) in phosphate-buffered saline as a detecting reagent.
Morphometry
Intimal and medial thickness was determined along the length of the treated arterial segments. A distinction was made between intimal and medial thickness within and outside the PTCA lesion area and the media underneath or between the stent struts. Data were analyzed using elastin-stained sections and assessed on a microscopy image analysis system (Impak C, Clemex vision Image analysis system; Clemex Technologies Inc.) as described before (11). In addition, lesion length was determined in the arteries treated with PTCA, which was defined as the percentage of the internal elastic lamina containing discontinuities or associated with an intimal and/or medial thickening (15).
Scanning and transmission EM
To study endothelial morphology (scanning EM) and to assess endothelial cell-cell contact (transmission EM), selected tissues were fixed with 2.5% glutaraldehyde in 0.15 M cacodylate buffer, postfixed with 0.1 M cacodylate buffer containing 1% OsO4and 50 mM ferricyanide (K3[Fe{CN}6]), and further processed as described before (18). Specimens were examined in a JSM25 scanning electron microscope (Jeol Ltd.) and a CM100 transmission microscope (Philips).
Statistical analysis
Analysis was performed using Sigmastat (versions 1.0 and 2.0, Jandel Scientific). Data are given as mean ± standard deviation. Morphometry was analyzed with a one-way ANOVA, the planimetry and angiography with a one-way repeated measures ANOVA, and followed by an all-pairwise comparison in case of statistical significance using a Student Neuman Keuls or Tukey test. A p value of <0.05 was considered statistically significant.
Results
Procedural outcome
Procedural outcome
A total of 39 animals were enrolled in the study. In group 3, one animal died suddenly within 1 h after the procedure due to stent thrombosis. In group 4, four animals died: one animal died during the procedure due to ventricular fibrillation, one died within a few hours after the procedure due to stent migration and subsequent stent thrombosis, one animal due to respiratory problems during recovery from anaesthesia, and one animal died at 3 weeks after the procedure ex causa ignota (not stent related). The remaining 34 animals were used for analysis, as summarized in Table 1. Quantitative angiographic measurements are shown in Table 2. Quantitative coronary angiography (QCA) confirmed that all stent and balloon sizes closely matched coronary artery vessel size with a balloon-artery ratio between 0.95 and 1.1.
| Group | Pre | Balloon | Ratio | Post | Fu |
|---|---|---|---|---|---|
| 1 (n = 10) | 3.05 ± 0.28 | 2.98 ± 0.38 | 0.95 ± 0.07 | 3.06 ± 0.33 | 2.37 ± 0.39 |
| 2 (n = 10) | 3.11 ± 0.21 | 3.03 ± 0.30 | 0.97 ± 0.06 | 3.02 ± 0.17 | 3.12 ± 0.27 |
| 3A (n = 9) | 2.87 ± 0.27 | 3.02 ± 0.41 | 1.02 ± 0.09 | 2.74 ± 0.29 | 2.65 ± 0.28 |
| 3B (n = 9) | 2.95 ± 0.31 | 2.85 ± 0.32 | 0.95 ± 0.05 | 2.66 ± 0.34 | 2.79 ± 0.48 |
| 4A (n = 5) | 2.61 ± 0.46 | 2.92 ± 0.39 | 1.08 ± 0.06 | 2.67 ± 0.30 | 2.80 ± 0.25 |
| 4B (n = 5) | 2.58 ± 0.25 | 2.65 ± 0.42 | 1.04 ± 0.06 | 2.28 ± 0.28 | 3.08 ± 0.42 |
Intimal permeability
Binding control confirmed that complete binding of EB to the albumin was achieved.
The “window” of permeability (groups 1 and 2)
Macroscopy (Fig. 2A and B)revealed that both the 1- and 70-kD markers were able to stain the vessel wall at 4 as well as at 12 weeks post stenting. This indicates that there is a wide window of permeability that does not change during the first 3 months.
Macroscopy of the dye-exclusion test. A,Control right coronary artery, shows a clean surface with occasional small blue areas. B,Group 2, Palmaz-Schatz stent, 1 kD. For both molecular weights, blue staining is found mainly at the stent ends and in the area of the coupler (arrows). C,Group 3A Wiktor stent. Blue staining is seen especially in the region of the intimal tissue covering the stent struts (arrows). D,Group 3B, PTCA. Randomly stained areas (arrows)are found in the arterial segment treated with PTCA and are more apparent and intense in group 3B than 4B.
The “extent” of permeability (all groups)
Except for occasional small areas distal to side branches, where the endothelium is often subject to hemodynamic stress, the control arteries did not reveal staining (Fig. 2A). Both stent- and balloon-treated arteries, however, did reveal a distinct staining of the vessel wall (Fig. 2B–D). In the stented arteries staining of the intima was generally observed over the stent wires, while between the stent struts a much lower level of staining was seen (Fig. 2B and C). Both the Wiktor stent and the Palmaz-Schatz stent, however, revealed a specific staining pattern. In the Palmaz Schatz stent staining was seen over the stent struts in the area of the stent ends and in the area of the coupler (Fig. 2B), while in the Wiktor stents staining was observed over the wire along the whole length of the stent (Fig. 2C). These patterns did not change during the observation period. In group 3B (2 weeks after PTCA alone) there were also areas that revealed staining albeit less prominently (Fig. 2D). In group 4B (12 weeks after PTCA) the pattern was the same but the staining intensity was less.
Planimetry
Planimetry of the area permeable to EB at 2 weeks showed that the percentage was 35.4 ± 19.7% for the stented arteries (p < 0.05 vs balloon and control), 10.1 ± 6.8% for the PTCA treated arteries, and 2.5 ± 2.2% for the control arteries.
Electron microscopy and the endothelial barrier function
Both scanning and transmission EM were performed on groups 3 and 4. It confirmed that, in contrast to normal endothelium (Fig. 3A), at 2 weeks after either stent implantation or PTCA the endothelial covering was still incomplete. The areas covering the stent struts especially showed missing cells, often in association with adhesion of leukocytes and platelets. These areas were highly permeable to the EB dye. The more diffusely permeable areas were characterized by an endothelial layer where the cells appeared “retracted” (Fig. 3B). Transmission EM showed small or nonexistent intercellular junctional complexes (Fig. 3C)in the areas with retracted cells. Twelve weeks after the interventions, scanning EM showed that the endothelial covering was complete in all groups, and transmission EM now showed more extensive junctional complexes even with tight junctions (Fig. 3D). Permeability was now associated with a different phenomenon, namely an endothelial layer with adhesion of leukocytes with a rounded morphology that were also often seen penetrating the endothelium as well as surface folds (Fig. 3E), which indicates an increased endocytotic activity, although endothelial retraction was still observed occasionally.
EM of groups 3 and 4. A,Scanning EM of a nonblue area showing normal endothelium. B,Endothelial cell retraction at 2 weeks after stenting as illustrated by scanning EM, shows fingerlike projections between adjacent endothelial cells (arrowhead).Some have broken during critical point drying (asterisk). C,Transmission EM at 2 weeks after stenting shows loose junctions (arrow)between adjacent endothelial cells. D,At 12 weeks after stenting, transmission EM shows tight junctions (arrow). E,The permeable areas in the endothelium at 12 weeks after stenting are characterized by surface folds as observed with transmission EM (arrow). N = nucleus.
Morphometry and assessment of cell proliferation
Morphometric analysis is summarized in Table 3. At 2 weeks after stenting (group 3A) there is a trend (p > 0.05) toward a larger NI over the stent struts than after PTCA at the site of the lesion (group 3B). At 12 weeks both the Wiktor stent (group 4A) and the Palmaz-Schatz stent (group 2) do induce a statistically significant larger NI than after PTCA. Also, the Wiktor stent induces a significantly more pronounced NI as compared with the Palmaz-Schatz stent. Morphometric analysis of the medial layers show that the stents significantly impress the media, whereas at the site of the PTCA lesion the media has thickened. At 4 and 12 weeks (groups 1, 2, and 4A) the media has thickened between the stent struts and is now similar to the lesion area in the balloon-treated arteries at 2 and 12 weeks.
| Group | NI | Media | |
|---|---|---|---|
| Stent or PTCA Lesion | Underneath PTCA Lesion or Stent Struts | Between Stent Struts or PTCA Lesion/Nonlesion | |
| 1 (stent) | 259 ± 104∗3a,3b,4b | 99 ± 22 | 187 ± 28∗3a,3b,4a |
| 2 (stent) | 192.1 ± 63.3∗3b,4b | 109.8 ± 21.6 | 184.3 ± 16.3∗3a,3b,4b |
| 3A (stent) | 114 ± 60 | 71 ± 40 | 122 ± 26 |
| 3B (PTCA) | 33 ± 11 | 221 ± 77∗1,2,3a,4a | 119 ± 15 |
| 4A (stent) | 305 ± 155∗3a,3b,4b | 73 ± 29 | 182 ± 29∗3a,3b,4b |
| 4B (PTCA) | 19 ± 12 | 181 ± 53∗1,3a,4a | 128 ± 33 |
For assessment of cell proliferation, the BrdU-positive and total number of cells were counted in several sections. For each artery this amounted to 200–400 endothelial cells; 200–400 (balloon group) and 1500–5500 (stent group) intimal cells; 1000–4000 medial and adventitial cells. The percentage of BrdU incorporation is summarized in Table 4, and shows that at 2 weeks after intervention the stent (group 3A) induced a higher percentage of BrdU incorporation as compared with PTCA in all tissue layers (p < 0.05). In the PTCA vessels (both lesion area and nonlesion area) there was a trend (p > 0.05) toward a higher percent proliferating cells as compared with control (not injured) values. Twelve weeks after intervention the levels of BrdU incorporation had decreased in all groups. Only the endothelium overlying the stent struts was still significantly higher as compared with control values (4.8 ± 2%; p < 0.05).
| Group (n) | EC | NI | M | ADV |
|---|---|---|---|---|
| Control (n = 6) | 0.4 ± 0.1 | Nonexistent | 0.9 ± 0.1 | 1.07 ± 0.1 |
| 3A (n = 6) | 22.2 ± 7.5∗ | 18.8 ± 2.7∗ | 6.5 ± 1.2∗ | 7.9 ± 2.1∗ |
| 3B: lesion area (n = 5) | 5 ± 3.8 | 6.14 ± 8.2 | 2.66 ± 2.62 | 3.98 ± 4.41 |
| 3B: non-lesion area (n = 6) | 2.64 ± 2.65 | Nonexistent | 0.86 ± 0.43 | 0.98 ± 0.78 |
| 4A (n = 5) | 3.5 ± 1.1 | 2.7 ± 1.1 | 1.6 ± 1 | 2.3 ± 1.1 |
| 4B: lesion area (n = 4) | 1.3 ± 1.5 | 1.2 ± 0.91 | 1.5 ± 0.67 | 0.75 ± 0.76 |
| 4B: non-lesion area (n = 5) | 0.3 ± 0.6 | Nonexistent | 0.99 ± 0.91 | 0.64 ± 0.54 |
Microscopy
There were no adverse or unexpected vascular reactions to the interventions as performed in groups 1–4, and in general the tissue response was as described before (10,11). In short, the stented arteries in groups 1, 2, and 4 were covered by a variable intimal thickness consisting of smooth muscle cells in a collagenous matrix and covered by endothelium. Inflammatory reactions were limited for groups 1 and 2. In group 4A, there were areas where the stent strut lacerated the media, a phenomenon associated with a diffuse inflammatory response. At 2 weeks (group 3A), the stent was embedded in a mass of organizing thrombus, containing leukocytes and macrophage giant cells, and was covered by an incomplete layer of endothelial cells. There was medial hyperplasia (medial thickening and longitudinal orientation of smooth muscle cells) in association with BrdU incorporation between the stent struts.
PTCA
At 2 weeks after PTCA (group 3B), there were focal areas with a limited amount of neointima, which consisted of smooth muscle cells in a collagenous matrix. In these areas we observed fragmentation of the Lamina Elastica Interna (LEI) and significant medial hyperplasia in association with incorporation of BrdU (Fig. 4A). These lesions encompassed approximately 30% of the measured circumference. The endothelial covering (as confirmed by lectin histochemistry) was incomplete and had a variable morphology. At 12 weeks after PTCA (group 4B), there was still a limited amount of intimal hyperplasia. Although the media was still thickened in these areas and the lesions still encompassed approximately 30% of the measured circumference, there was no incorporation of BrdU, nor a clear fragmentation of the LEI. On the contrary, we often observed an additional internal elastic membrane under the newly formed endothelium (Fig. 4B). The adventitia was unremarkable.
LM. A,Group 3B. At 2 weeks after PTCA, focal lesions can be found with BrdU incorporation (arrow),especially underneath the endothelial lining but also elsewhere in the intima (i)and media (m).HRP-DAB with hematoxylin counterstain. Bar = 50 μm. B,Group 4B. At 12 weeks after PTCA, there is still a limited NI, sometimes with an additional elastic membrane underneath the endothelium (arrow).Resorcin-Fuchsin, i = intima, m = media. Bar = 50 μm.
Discussion
In contrast to restenosis after PTCA, which is dictated both by constrictive remodeling and tissue growth (16,17), restenosis after stenting is the result of tissue growth alone. With increasing numbers of patients being treated with stents (up to 50%), the problem of in-stent restenosis is becoming more and more of a problem, as this seems more resistant to effective treatment.
Vascular dysfunction and in particular endothelial dysfunction has been described after PTCA both in humans and animals (5–8). As one of the first changes in the etiology of atherosclerosis, endothelial dysfunction might also be involved in the ongoing tissue growth after angioplasty procedures (8,18). The aim of our study, therefore, was to investigate endothelial function after stenting and compare this with PTCA alone by assessing both morphologic and functional parameters.
The main finding in our study is that both PTCA and stent implantation result in an impairment of the vascular barrier function at least up to 3 months after the procedure as evidenced by the uptake of the EB dye. This loss of barrier function is more pronounced after stenting than after PTCA, and showed a stent-specific pattern. This breach was characterized by specific endothelial morphologic correlates: in the early phase by incomplete endothelialization and endothelial retraction or loose intercellular connections. The late phase was characterized by the expression of surface folds and the adhesion of leukocytes. Both phenomena were also observed in stented human vein grafts (19).
Transport routes across the endothelial lining
The extravasation of (macro)molecules, such as EB bound to albumin, proceeds mainly by two routes. One is through diffusion via the cellular junctions, i.e., paracellular exchange (20,21), and the other is by vesicle-mediated transport, i.e., transcellular transport (22).
Paracellular exchange is morphologically associated with small interendothelial gaps caused by contractile forces in the cell and by disintegration of cell–cell junctions (23,24). This process is regulated by actin fibers, which are connected to other proteins anchoring the cells to their neighbors and to the extracellular matrix (25,26). Vasoactive agents and thrombin can affect the integrity of the endothelium through phosphorylation of specific target proteins. As a consequence, actin reorganization may occur through RhoA– and protein kinase C–activated pathways. The interaction between actin and nonmuscle myosin, activated by phosphorylation of the myosin light chain, subsequently causes contraction and gap formation (24,27). Interaction of leukocytes with the endothelium, which is enhanced by inflammatory mediators, can enhance this response (28). In addition to the effects of vasoactive agents, paracellular permeability can also be enhanced by the lingering proliferative response of the endothelium itself. Cell retraction associated with cell mitosis causes paracellular gaps in arterial endothelial cells in vivo (29). Because cell division is found until at least 3 months after intervention, particularly in areas overlying stent struts, it may contribute to the observed leakage of EB-albumin complex. Both endothelial barrier function and proliferation are under control of the extracellular matrix. Remodeling of the extracellular matrix by proteases can reduce the firm interaction between cell and matrix. This reduction may enhance the efficacy of endothelial permeability-increasing agents (25,30).
Transcellular (vesicle-mediated) transport is the second major route for macromolecular exchange across the endothelial lining. Both paracellular and transcellular exchange of albumin are increased by VEGF (31), a growth factor that is induced in injured arteries (32,33). Further studies have to elucidate whether and for how long VEGF is induced in stented coronary arteries.
In vivo (34,23)and in vitro studies (35,36)have shown that an increase in endothelial permeability in the microcirculation can be reversed by exposing the endothelium to cAMP-elevating agents. In preliminary experiments we found that intracoronary administration of 1 mM dibutyryl-cAMP within 10 min indeed partly normalizes the barrier function in areas of increased permeability covering and adjacent to the implanted stents (Fig. 5). The areas that were permeable because of missing endothelial cells were not affected by this treatment. A reduced cellular cAMP concentration/content has been reported in arterial cells in intimal tissue affected by atherosclerosis (37). However, it remains to be established whether this also occurs in the new layer of endothelial cells in stented arteries. The reduction in endothelial permeability by elevation of cellular cAMP levels indicates that a major part of the leakage of EB-albumin occurs via impaired junctional complexes between endothelial cells. However, the current state of knowledge does not permit exclusion of a minor contribution by vesicular transport.
Macroscopy c-AMP–treated arteries. Macroscopy of the coronary arteries at 2 weeks after implantation of a Wiktor stent. While a “control” artery (A)was not exposed to db-c-AMP, but only to EB, the c-AMP-treated artery was first exposed to 1 mM db-c-AMP and then to EB with a molecular weight of 1 kD (B),showing an improvement especially in the areas between the stent struts.
Permeability after PTCA takes place preferentially in the area designated as the lesion area containing intimal hyperplasia. Clearly these are areas where the endothelium is attached to a changed basement membrane or extracellular matrix. In the stented segments, the intima covering the stent struts is also constantly changing during the process of scar maturation, which might explain permeability over the stent struts in general but not the differences between stent designs. Whereas the Wiktor stent shows preferential dye-uptake over all of the stent wires throughout the whole stent, the Palmaz-Schatz stent shows dye-uptake preferentially over the stent struts at both stent extremities and over the area of the coupler. It would seem that these specific patterns of dye-uptake are a reflection of the design of the stent. The Wiktor stent has a more open design than the Palmaz-Schatz stent, and as permeability seems to occur in areas where theoretically movement between the tissue and the stent struts may occur, this may be a factor influencing the chronic vascular irritation by the stents. From the literature it is known that mechanical instability of healing bone fractures, for instance, affects the composition of the extracellular matrix with respect to the level of sulfate incorporation in the glycosaminoglycans (38). If these processes also take place in the vessel wall it would certainly influence endothelial function and permeability.
Conclusions
This study indicates that especially stenting decreases long-term vascular integrity with respect to permeability. Leakage is observed with molecules such as EB and EB-albumin complex, and is associated with prolonged endothelial proliferation and distinct morphologic characteristics such as endothelial retraction, the expression of surface folds, and the adhesion of leukocytes.
1. Colombo A, Ferraro M, Itoh A, et al. Results of coronary stenting for restenosis. J Am Coll Cardiol 1996 Oct;28:830–6.
2. : "Intracoronary stent implantation without ultrasound guidance and with replacement of conventional anticoagulation by antiplatelet therapy. 30-day clinical outcome of the French Multicenter Registry". Circulation 1996; 94: 1519.
3. : "Heparin-coated Palmaz-Schatz stents in human coronary arteries. Early outcome of the Benestent-II Pilot Study". Circulation 1996; 93: 412.
4. : "Evidence for prevention of death and myocardial infarction with platelet membrane glycoprotein IIb/IIIa receptor blockade by abciximab (c7E3 Fab) among patients with unstable angina undergoing percutaneous coronary revascularization. EPIC Investigators. Evaluation of 7E3 in preventing ischemic complications". J Am Coll Cardiol 1997; 30: 149.
5. : "Endothelium dependent inhibition of ergonovine-induced contraction is impaired in porcine coronary arteries with regenerated endothelium". Circulation 1989; 80: 643.
6. : "Persistent dysfunction of regenerated endothelium after PTCA of rabbit iliac artery". Circulation 1990; 81: 1667.
7. : "Early evidence of endothelial vasodilator dysfunction at coronary branchpoints". Circulation 1990; 82: 1169.
8. : "Endothelium and atherosclerosis". J Hyperten 1992; 10: 43.
9. NIH publication 85-23, revised 1985.
10. : "Reduction of thrombotic events with heparin-coated Palmaz-Schatz stents in pigs". Circulation 1996; 93: 423.
11. : "Coronary stenting with a new, radiopaque, balloon expandable endoprosthesis in pigs". Circulation 1991; 83: 1788.
12. : "Endothelial integrity and viability in the aorta of the normal rabbit and rat as evaluated with dye exclusion tests and interference contrast microscopy". Atherosclerosis 1972; 15: 285.
13. : "Increased endothelial cell turnover in areas of in vivo evans blue uptake in pig aorta". Atherosclerosis 1973; 17: 401.
14. : "Histological features of a polymer endovascular prosthesis after transcatheter implantation in porcine arteries". Cardiovascular Pathology 1993; 2: 41.
15. : "Increasing arterial wall injury after long term implantation of two stent types in a porcine coronary model". Eur Heart J 1998; 19: 601.
16. : "The relative importance of arterial remodeling compared with intimal hyperplasia in lumen renarrowing after PTCA: a study in the normal rabbit and the hypercholesterolemic Yucatan micropig". Circulation 1994; 89: 2816.
17. : "Intravascular ultrasound comparison of restenotic and de novo coronary artery narrowings". Am J Cardiol 1994; 74: 1278.
18. : "The pathogenesis of atherosclerosis. An update". N Engl J Med 1986; 314: 488.
19. : "Histology after stenting of human saphenous vein bypass grafts: Observations from surgically excised grafts 3 to 320 days after stent implantation". J Am Coll Cardiol 1993; 21: 45.
20. : "Transport of macromolecules across microvascular walls: the two pore theory". Physiol Rev 1994; 74: 163.
21. : "Transport of macromolecules through microvascular walls". Cardiovasc Res 1996; 32: 644.
22. :
The microvascular endothelium revisited . In: Endothelial Cell Biology in Health and Disease . Edited by . New York: Plenum Publishing Corp1988: 3.23. : "The B2-adrenergic receptor agonist formoterol reduces microvascular leakage by inhibiting endothelial gap formation". Am J Physiol 1994; : L461.
24. : "Endothelial permeability for macromolecules. Mechanistic aspects of pathophysiological modulation". Arterioscl Thromb Vasc Biol 1997; 17: 1018.
25. : "Regulation of vascular endothelial barrier function". Lung Cell Mol Physiol 1994; 11: L223.
26. :
The endothelial contractile cytoskeleton . In: Vascular Endothelium. Physiology, Pathology and Therapeutic Opportunities . Edited by . Stuttgart: Schattauer1997: 1.27. : "Regulation of endothelial cell gap formation and barrier dysfunction: role of myosin light chain phosphorylation". J Cell Physiol 1995; 163: 510.
28. : "Physiological mechanisms of postischemic tissue injury". Ann Rev Physiol 1995; 57: 311.
29. : "Transendothelial transport of low density lipoprotein in association with cell mitosis in rat aorta". Arteriosclerosis 1989; 9: 230.
30. : "The extracellular matrix as a cell survival factor". Mol Biol Cell 1993; 4: 953.
31. : "Increase microvascular permeability and endothelial fenestrations induced by vascular endothelial growth factor". J Cell Sci 1995; 108: 2369.
32. : "Expression of VEGF receptors in arteries after endothelial injury and lack of increased endothelial regrowth in response to VEGF". Arterioscler Throm Vasc Biol 1996; 16: 1399.
33. : "Reciprocal relation between VEGF and NO in the regulation of endothelial integrity". Nature Medicine 1997; 3: 879.
34. : "Effects of isoprenaline and cooling on histamine induced changes of capillary permeability in the rat hindquarter vascular bed". Acta Physiol Scand 1978; 103: 252.
35. : "Role of cyclic adenosine monophosphate in the induction of endothelial barrier properties". J Cell Physiol 1989; 139: 157.
36. : "Norepinephrine and iloprost improve barrier function of human endothelial cell monolayers: role of cAMP". Am J Physiol 1991; 260: C1052.
37. : "Disorders in the system of cyclic nucleotides in atherosclerosis: cyclic AMP and cyclic GMP content and activity of related enzymes in human aorta". Tissue Cell 1987; 19: 21.
38. : "The effects of mechanical stability on the macromolecules of the connective tissue matrices produced during fracture healing. II. The glycosaminoglycans". Histochem J 1987; 19: 39.
Abbreviations
| BrdU | bromodeoxy uridine |
| EB | Evans blue |
| EM | electron microscopy |
| LM | light microscopy |
| NI | neointimal thickening |
| PTCA | percutaneous transluminal coronary angioplasty |
Footnotes
☆ This study was supported by the Netherlands Heart Foundation grant 93-158, and the Interuniversity Cardiology Institute of the Netherlands (ICIN) project 18.
