Enhanced Cellular Proliferation in Intact Stenotic Lesions Derived From Human Arteriovenous Fistulas and Peripheral Bypass Grafts
Does It Correlate With Flow Parameters?
Background Vascular interventions are often complicated by the development of intimal thickening, leading to stenosis. Cellular proliferation is a key event in stenosis formation in animals, but the role of cell proliferation in intimal thickening in humans is still unclear. Furthermore, the relation between proliferation in human stenotic lesions and flow parameters has not been established.
Methods and Results We studied the proliferation patterns of 35 anatomically intact human stenotic lesions derived from either peripheral bypasses (normal flow) or hemodialysis AV fistulas (high flow) with the use of Ki-67, a cell proliferation marker. Local flow parameters were assessed with ultrasound. Proliferation patterns were similar in AV fistula and bypass stenoses. In the intima, proliferation was highest in the area just below the endothelium (AV fistulas, 3.6%; bypasses, 3.5%; P=NS). In adjacent nonstenotic vessel segments that were used as controls, proliferation rate in the intima was 0.3%. Double-labeling studies revealed that subendothelial-intimal proliferation consisted mainly (90%) of vascular smooth muscle cells, whereas proliferation in the other layers of the vessel wall also consisted of endothelial cells and macrophages. Blood flow velocity was negatively correlated with subendothelial-intimal proliferation (r=−.61, P<.05). The endothelial cell coverage of the lumen was positively correlated with proliferation (r=.85, P<.01).
Conclusions These data suggest enhanced cellular proliferation in human stenotic tissue derived from AV fistulas and peripheral bypass grafts. Furthermore, high proliferation rates seem to be associated with endothelial cell coverage of the lumen and low local flow velocities.
Failures of vascular reconstructions are caused mainly by the development of stenotic lesions, often located in or adjacent to the anastomo-ses.1 2 3 4 5 The incidence of stenosis in the first postoperative year varies from 20% to 30% in femorodistal bypass grafts to 50% to 60% in hemodialysis AV fistulas when constructed with the use of interposition grafts.6 7 8 Proliferation of intimal VSMCs is thought to be a dominant cellular event in the development of an intima.5 9 10 The kinetics of this proliferative response are well described in animal studies.11 However, data on proliferation in human restenotic coronary tissue,12 13 peripheral artery restenotic lesions,14 and stenoses derived from AV fistulas10 have been obtained only recently. The studies so far have produced rather conflicting results; very high (up to 20%) and very low (0% to 0.5%) proliferation indexes have been reported. Consequently, the exact contribution of VSMC proliferation to restenosis in humans remains unclear. A difficulty of the reported studies may be the use of PCNA as the marker for proliferation. Overestimation of the proliferation index with this marker has been demonstrated and may explain the reported high proliferation indexes.15 A difficulty of evaluating atherectomy-derived stenotic tissue may be the possible influence of sampling error, resulting in underestimation of the contribution of proliferation.16 The first goal of the present study was to obtain reliable data on proliferation in human stenotic vascular grafts by circumventing the possible disadvantages of earlier studies. Instead of PCNA, we used Ki-67, which has been shown to provide an easy and reliable method to determine proliferative activity in tissue samples.15 17 18 19 Moreover, our studies were not performed in small samples of stenotic lesions but on complete and intact stenotic lesions derived from AV fistulas and peripheral bypass grafts, which enabled us to exactly localize the Ki-67-positive cells in the lesion.
One factor that may influence the growth of an intimal lesion is local flow, and decreased flow is associated with increased intimal thickening.20 21 22 23 24 25 26 In humans, accelerated atherosclerosis has been reported in bypassed coronary vessels, which have almost zero flow.27 However, information on the possible correlation between flow and intimal growth in humans lesions is lacking. Therefore, we studied the correlation between initial and late local flow characteristics and proliferation in the stenotic segments. Because AV fistulas are an example of a high-flow system, and a peripheral bypass is an example of a low-flow system, we hypothesized that the extent of proliferation in stenotic lesions derived from these two types of vascular reconstructions might be different, ie, high proliferation in the peripheral bypass stenoses and low proliferation in the AV fistula stenoses.
We also measured the endothelial cell coverage of the lumen of the stenotic segments because data on the relationship between the presence of endothelial cells and proliferation in human stenotic lesions have not been reported so far. However, numerous animal studies showed an inhibiting effect of endothelial cells on intimal growth,28 29 30 whereas some investigators observed a positive correlation between the presence of endothelial cells and subendothelial proliferation.31 32
Two weeks after construction, flow characteristics were determined with the use of ultrasound techniques at several sites in the graft and in the inflow and outflow vessel segments in a consecutive series of patients receiving a hemodialysis AV fistula or a femorodistal bypass. During follow-up, the local flow characteristics were assessed again at the site of the stenosis in patients developing stenoses in the graft or at anastomoses necessitating surgical intervention. After the complete stenotic vessel segments were harvested, proliferation was assessed in different layers of the stenosis with Ki-67. Local flow characteristics, assessed both initially and just before intervention, were correlated with the development of intimal thickening and the extent of proliferation in the stenotic vessel segments.
During the study period of January 1992 through June 1994, 48 reinterventions were performed after construction of either an AV fistula or a peripheral bypass in the surgery department of the University Hospital of Maastricht or the Catharina Hospital of Eindhoven. In 13 patients, the vascular conduit was corrected by patchplasty at the site of the stenosis. Because collection of stenotic tissue in these cases was impossible without additional surgical procedures, which might jeopardize the success of the reintervention, it was decided not to obtain tissue specimens from this group of patients. In the remaining 35 patients, surgical repair was performed by placement of an interposition graft, allowing collection of the stenotic process and adjacent vessel segments. In all patients, the stenotic lesion was focal and narrowing or occluding the lumen. The age of the graft was defined as the time between construction of the vascular conduit and surgical intervention to repair the stenotic lesion.
Patient Clinical Profile
Patients With AV Fistulas
This group consisted of 9 men and 6 women whose mean age was 56 years (range, 37 to 76 years). The AV fistula was repaired after an average time of 10.4 months (range, 3 to 27 months). The diagnosis of the stenosis was based on duplex ultrasound and confirmed by angiography.8 In all patients, the occurrence of a stenosis was associated with an increase in venous pressure and/or low flow of the fistula during hemodialysis that necessitated reintervention. In 1 patient, a second reintervention was performed during the study period, and in 1 patient, two stenotic lesions were derived from different sites in the vascular conduit, resulting in a total of 17 stenotic samples (n=17). In 12 patients, the stenotic samples were derived from the efferent vein of a Brescia-Cimino fistula or a fistula constructed with a PTFE interposition graft. In 5 patients, the stenotic sample was derived from the venous anastomosis of a PTFE interposition graft.
Patients With a Peripheral Bypass
This group of patients consisted of 12 men and 4 women whose mean age was 64 years (range, 54 to 77 years). Reintervention was performed after an average time of 4.5 months (range, 1 to 24 months). The diagnosis of a stenosis was based on generally accepted duplex criteria and confirmed by angiography.33 Patients with a graft at risk underwent repair of the stenotic segment. In most of these patients, occurrence of stenosis was associated with a recurrence of symptoms. In 2 patients, stenotic tissue was derived from two different sites in the graft, giving 18 total stenotic vessel specimens in this group (n=18). In 10 patients, the stenotic tissue was derived from reversed saphenous vein grafts; in 6 patients, from in situ saphenous vein grafts; and in 2 patients, from a PTFE graft. Stenotic samples were derived from the proximal anastomosis (n=11), midgraft (n=5), and distal anastomosis (n=2).
Tissue Staining with Ki-67
Cell proliferation was determined with Ki-67, an antibody to an antigen expressed in all phases of the cell cycle except G0 and the early part of G1. The Ki-67 antigen is localized at the periphery of the chromosome scaffold and cellular cortex.15 The exact function of this antigen in replication has not been determined yet, but a reaction of the Ki-67 antibody with nuclear proteins has been established.15 Furthermore, the proliferation indexes determined with Ki-67 and 3H-thymidine are closely correlated.17 19
Immediately after harvesting, the tissue specimens were processed carefully to preserve the sections intact and to prevent loss of endothelial cells covering the lumen. After deparaffinizing of the sections and blocking of endogenous peroxidase activity with 0.3% H2O2, the slides were placed in a 0.01-mol/L citrate buffer, pH 6.0, and boiled for 5 minutes in a microwave oven. After subsequent washes in H2O and TBS, the slides were incubated with a mouse anti-Ki-67 monoclonal antibody (MIB, Immunotech) for 60 minutes at room temperature at a dilution of 1:100, followed by incubation with biotinylated sheep anti-mouse IgG (1:250, DAKO) at room temperature for 30 minutes. Sections were then labeled for 30 minutes with an avidin-biotin peroxidase complex (Vectastain, ABC Kit, Vector Labs) by use of 3,3′-diaminobenzidine as the chromogen. The sections were counterstained with hematoxylin, dehydrated through graded series of alcohols, and mounted. A piece of human small intestine was used as a positive control in all staining sessions. As an internal positive control, we used one stenotic AV fistula segment that showed consistent positive staining during all sessions. As a negative control, the staining procedure was performed without addition of the primary antibody. A spare saphenous vein segment also was used as a negative control, showing no proliferation throughout the different staining sessions.
To assess the phenotype of proliferating cells, double-labeling studies were performed. For double immunocytochemistry, paraffin sections were treated and incubated similarly to those used for single Ki-67 immunocytochemistry. After the incubation with diaminobenzidine to stain Ki-67-positive nuclei and washes in H2O and TBS, the sections were incubated with one of the second primary antibodies (smooth muscle actin for VSMC or CD68 for macrophages) for 45 minutes at room temperature. The antibodies used were mouse anti-α-smooth muscle actin (DAKO, 1:4000 in TBS/1% BSA/0.1% Tween, pH 7.4 to 7.6) and mouse anti-CD68 (DAKO, 1:500). Subsequently, the sections were washed three times with TBS at room temperature and incubated for 30 minutes with a biotin-labeled sheep anti-mouse IgG (DAKO, 1:250) at room temperature followed by 30 minutes of incubation with an avidin-biotin alkaline phosphatase complex (DAKO), with fast blue as the chromogen (Sigma Chemical Co). The sections were mounted with Immunomount (Shandon). No counterstaining was performed. This staining protocol results in dark-brown Ki-67-positive nuclei and blue cytoplasm in cells positive for one of the other two antibodies.
Because double-staining protocols of Ki-67 and lectin were unsuccessful, another endothelial cell marker (CD34) was used. The protocol for the CD34-Ki-67 (endothelial cells) double staining was comparable to the above-described protocol. However, CD34 immunostaining (mouse anti-CD34 [Becton Dickinson, 1:400]) preceded the Ki-67 staining. Diaminobenzidine was used as the chromogen for Ki-67, and fast blue was used as the chromogen for the CD34 antibody. This staining protocol also results in brown nuclei of Ki-67-positive cells and blue cytoplasm of cells positive for CD34. Sections without the respective primary antibodies were used as negative controls (data not shown).
Cell Counting and Proliferation Rates
Tissue sections were investigated by light microscopy at ×400 magnification with a standard field size. Cells containing dark nuclear staining with Ki-67 were considered positive17 18 (Figs 1 and 2⇓⇓). To obtain insight into the distribution of the proliferative response in the anatomically intact lesions, cells were counted separately in subendothelial-intimal and medial-intimal areas. Furthermore, medial and adventitial areas were counted. All cell counting was performed with an ocular grid by two investigators (Drs Hofstra and Daemen). In each of the vessel layers, four randomly selected fields were counted. For each specimen, assessment of proliferation in the different cell layers was performed in three parallel sections. The values of the four fields in the three parallel sections were averaged. This average value was used in this study. The proliferation index, defined as the number of positive cells divided by the total number of cells, was calculated for each layer. In the stenotic lesions derived from PTFE grafts, no medial proliferation index was calculated.
Double-labeling studies in parallel sections were used to assess the fraction of the Ki-67-positive VSMCs, macrophages, or endothelial cells in the different areas (subendothelial-intimal, medial-intimal, medial, and adventitial).
The number of proliferating cells, also positive for one of the phenotype-specific antibodies, was divided by the total number of proliferating cells to calculate the fraction. Fractions for the three different cell types were calculated for each area of the vessel wall (subendothelial-intimal, medial-intimal, medial, and adventitial).
Endothelial Cell Coverage of the Stenotic Lumen
The degree of endothelial cell coverage of the lumen was estimated by use of the Ulex Europaeus-stained parallel sections. After blocking of endogenous peroxidase activity with 0.3% H2O2, sections were incubated with the lectin antibody (DAKO, 1:250, 45 minutes) after predigestion with pepsin (1 mg/mL, 30 minutes). Subsequently, the sections were incubated for 45 minutes with an anti-Ulex Europaeus lectin peroxidase complex (DAKO, 1:100), followed by staining with the chromogen 3,3′-diaminobenzidine. In two parallel sections of each specimen, the cell coverage of the lumen was determined by morphometry (Quantimet 570, Image Analyzer, Leica). The extent of endothelial cell coverage was defined as the circumference of the lumen covered by Ulex Europaeus-positive cells divided by the total luminal circumference. Tissue processing and all cell counting were performed with investigators blinded to the clinical information.
In 15 tissue samples, proliferation was assessed in vessel segments adjacent to the stenosis showing no intimal thickening. Proliferation was assessed in the intima, media, and adventitia of 8 bypass tissue specimens and 7 AV fistula tissue specimens.
Measurement of Flow Characteristics
To correlate the initial and final local flow characteristics with proliferation, PSV (in centimeters per second) and nPSV (in s−1) were assessed in vivo at predefined sites in the graft and inflow and outflow segments, with special attention given to sites predisposed to developing intimal lesions, like the perianastomotic regions. PSV was assessed with the use of duplex ultrasound (ATL, Ultramark IV, Advanced Technology Laboratories) with the Doppler sample placed midstream at defined sites in the vascular conduit. The diameter was determined at the same defined sites in the graft and inflow and outflow arterial segments with vessel wall Doppler tracking. This system was described in detail elsewhere.34 35 Validation studies showed a coefficient of variation of 2% to 4% in the determination of diameter of a medium-sized muscular artery with this system.36
From PSV and diameter, nPSV was calculated from the following equation:nPSV|<|=|>|PSV/diameter
The local flow characteristics, determined 2 weeks after construction of the vascular conduit (initial flow) and 1 week before surgical repair of the stenosis (final flow), were used for analysis and correlation with proliferative patterns.
Data are presented as mean±SD. To compare results between the bypass group and the AV fistula group, the Mann-Whitney U test was used. To compare proliferation between the different cell layers within the stenoses, the Wilcoxon test was performed (paired test). A multiple regression model was used to assess the correlation between proliferation and parameters, such as endothelial cell coverage of the lumen, nPSV, and age of the stenosis. Linear regression analysis (Pearson's correlation) was used to define correlations between different parameters. When the data suggested a nonlinear relation, nonlinear regression analysis was used to define the relationship. A value of P<.05 was considered statistically significant.
Patient and Flow Characteristics
Table 1⇓ summarizes the patient characteristics. No differences in mean age were observed between the groups (Mann-Whitney). Surgical repair of the vascular conduit in patients with AV fistulas was performed after an average time of 10.4 months and in bypass patients after an average time of 4.5 months (P=.002). Two weeks after operation, PSV and nPSV were higher at the site of the later stenosis in the AV fistula group compared with the bypass group (P=.002 and P=.05). However, no differences in local PSV and nPSV were observed between the two groups shortly before correction (final) of the stenotic segment. In the bypass group, the final PSV and nPSV values at the site of the stenosis were higher compared with the initial PSV and nPSV values (P<.01), but in the AV fistula group, no differences between the initial and final PSV and nPSV values were observed.
In all stenotic samples, Ki-67-positive cells were observed, indicating proper expression of this proliferation marker in the vessel specimens (Fig 1⇑). Table 2⇓ provides a summary of the results. No differences in proliferation index were observed between the AV fistula and bypass groups in the subendothelial-intimal areas (7234 cells counted), medial-intimal areas (8371 cells counted), medial areas (3893 cells counted), and adventitial areas (6124 cells counted; Mann-Whitney). In both the AV fistula and bypass groups, subendothelial-intimal proliferation was higher compared with medial-intimal proliferation (3.6% versus 1.1% and 3.5% versus 2.0%, respectively; P<.05; Wilcoxon). In the adventitia, relatively high proliferation indexes were found (7.5% and 4.2% for AV fistula and bypass stenoses, respectively; P=.26). In the nonstenotic vessel segments, proliferation in the intima was 0.3%; in the media, 0.2%; and in the adventitia, 3.2%.
Double labeling showed that 90% of Ki-67-positive cells in the subendothelial-intimal area were VSMCs (Fig 1⇑). The contribution of macrophages and endothelial cells to total proliferation in the subendothelial-intimal area was low (4% and 6%, respectively; Table 2⇑). However, in the intima close to the media (intimal-medial), the fraction of proliferating VSMCs decreased to 27%, whereas the contribution of macrophages and endothelial cells to total proliferation increased to 18% and 55%, respectively (Table 2⇑ and Fig 1⇑). In the media, the contribution of proliferating macrophages to total proliferation was low (2%), whereas the contribution of endothelial cells was very high (76%). In the adventitia, both the macrophages and the endothelial cells contributed substantially to total proliferation (23% and 35%, respectively).
Endothelial Cell Coverage of the Lumen
The degree of endothelial cell coverage of the lumen of the stenotic segments was 48% in the AV fistula group and 50% in the bypass group. Endothelial cell coverage of the adjacent nonstenotic vessel segments was 98%, indicating that processing of the segments did not result in a loss of endothelial cells. The proliferation index in the endothelium covering the lumen of the stenotic segment was 2.9% in the AV fistula group and 3.5% in the bypass group (P=.31).
Because no differences in proliferation patterns were found between stenotic segments derived from AV fistulas and peripheral bypasses, they were treated as one group in the correlations of different parameters with proliferation. A multiple regression model was used to identify factors that independently correlated with subendothelial-intimal proliferation, including the age of the lesion, nPSV at the site of the stenosis, graft type (vein or PTFE), and endothelial cell coverage of the lumen of the stenosis. The only factor that independently correlated with subendothelial-intimal cell proliferation was endothelial cell coverage (Table 3⇓). A positive correlation was found between endothelial coverage of the lumen and subendothelial-intimal proliferation (r=.85, P<.001; Fig 2a⇑). The final nPSV could not be identified as an independent factor. However, when nPSV was correlated only with subendothelial-intimal cell proliferation, a negative correlation was observed (nonlinear regression analysis, r=−.67, P<.05; Fig 2b⇑). Therefore, a relation between endothelial cell coverage of the lumen and nPSV was sought (PTFE stenoses were excluded because no endothelial cells were present in these specimens). Linear regression analysis demonstrated a negative correlation between these parameters (y=100−0.4x; r=−.60, P<.05), suggesting that high flow velocity results in a loss of endothelial cell coverage of the lumen. A positive correlation was observed between proliferation of endothelial cells covering the lumen of the stenosis and subendothelial-intimal cell proliferation (y=1.4+0.47x; r=.54, P<.01).
The present study has several interesting findings. First, it shows increased proliferation in human vascular stenotic lesions derived from AV fistulas and peripheral bypass grafts. Moreover, the data indicate that ≈90% of the proliferating cells in the subendothelial area of the intima are VSMCs. In the intimal-medial area, media, and adventitia, a substantial fraction of proliferating cells in the stenotic lesion are macrophages and endothelial cells covering microvessels. Our data further suggest an inhibitory effect of high shear rate on the development of stenosis in human vascular grafts because an increased nPSV, measured 2 weeks before surgical repair of the stenosis, was associated with a decreased proliferation in the stenosis. Finally, the data suggest that luminal endothelial cells may be involved in the regulation of intimal proliferation in human vascular grafts.
Proliferation in Human Stenotic Tissue
Numerous pharmacological agents have been reported to inhibit VSMC proliferation in animal models (for a review, see Reference 37). However, administration of these agents to patients undergoing vascular intervention failed to show beneficial effects,9 38 39 indicating that extrapolation of findings in animal models to humans may be restricted and that proliferation patterns may differ between human and animal stenotic lesions.9 16 39 Therefore, more information on cell proliferation in human stenotic lesions is required to address this question.
The recent reports on proliferation in human vascular tissue may be criticized because of the small tissue samples and/or the proliferation marker used. The advantage of studying surgically harvested fresh stenotic tissue from anastomoses of AV fistulas and peripheral bypasses, as in the present study, is that the tissue samples are well defined and complete. This allows us to exactly localize the site of proliferation within the specimen and excludes sample errors, as may be encountered in studies evaluating atherectomy-derived tissue.14 16 The replication rates in the different vessel layers found in the present study (1.1% to 3.6% for the intima) are, however, also considerably lower compared with those found in another study evaluating proliferation in AV fistula stenoses.10 In that study, the PCNA marker was used, and proliferation rates as high as 15% to 20% were observed in the intima. High proliferation rates also were observed in a study evaluating proliferation in restenotic tissue after PTCA with PCNA.14 The use of PCNA may lead to overestimation of proliferation, as has been observed in validation studies, and may explain the reported high replication rates.15 In the present study, we studied replication in stenotic tissue with the use of Ki-67. This proliferation marker has been validated with 3H-thymidine labeling and has gained acceptance as a prognostic marker in cancer research.15 17 18 It may give a better representation of proliferation than PCNA.17 19 In our opinion, the use of Ki-67 on complete stenotic specimens as performed in the present study provides a good basis for reliable data on cell proliferation in human stenotic vascular grafts.
Initial Flow Conditions and Graft Stenosis
We hypothesized that the extent and distribution of proliferation may differ between stenotic lesions derived from AV fistulas (a relatively high-flow system) and peripheral bypasses (a relatively low-flow system).20 21 22 23 24 25 26 However, no apparent differences between proliferation in stenotic tissues derived from these two groups were found, although a large difference in the age of the graft was observed between the AV fistulas and the bypasses (4.5 versus 10.4 months, P=.002). Although it is true that the clinical profiles of patients with AV fistulas and peripheral bypasses differ in many aspects, our data indicate that the initial flow in the graft is not related to late graft proliferation. However, in contrast to the initial values, PSV and nPSV values just before repair of the stenosis (Table 1⇑) were of the same magnitude in the AV fistula and bypass groups, indicating that at the time of harvesting the local flow conditions at the site of the stenosis were similar in the two groups. This may explain why proliferation rates did not differ in the stenotic tissues derived from these groups. We cannot, however, exclude that proliferation in the two groups may have been different at an earlier point in the development of an intimal lesion.
Endothelial Cells and Proliferation
Both the higher age of the graft in the high-flow AV fistulas compared with the peripheral bypasses with normal flow and the negative correlation between end-point proliferation and prestenotic normalized PSV suggest that the concept of inhibiting effects of high shear rate on the development of intimal thickening as found in animal models is also valid for human stenotic lesions. However, nPSV was not an independent factor influencing proliferation, and endothelial cell coverage of the lumen appeared to be intermediate between nPSV and subendothelial-intimal proliferation. This suggests that an increased nPSV may lead to a loss of endothelial cells covering the stenotic segments, thereby decreasing proliferation.
Increased proliferation in areas covered with endothelial cells compared with denuded areas was described previously in organ cultures of porcine aorta.31 In a recent study using a human saphenous vein organ culture model, higher proliferation indexes also were found in the intima with an intact endothelial cell layer compared with denuded vein cultures.32 The results of these in vitro studies are in concordance with our study. Increased proliferation of endothelial cells covering the stenosis, as found in the present study, suggests that these cells are not in their normal quiescent state and that this proliferative state of endothelial cells may be related to a loss of their growth inhibitory function.
A difficulty in correlating endothelial cell coverage of the lumen and proliferation is that during processing of tissue samples, loss of endothelial cells may occur. However, the nonstenotic adjacent vessel segments, which were processed in the same fashion, showed an almost complete coverage of the lumen with endothelial cells. This argues against a substantial loss of endothelial cells of the stenotic segments during processing. Furthermore, a loss of endothelial cells would tend to increase the proliferation index in nonendothelialized regions. Hence, our results probably underestimate rather than overestimate the actual proliferation in endothelialized regions.
From the lumen to the adventitia, increasing proportions of proliferating cells were identified as endothelial cells. A possible explanation may be endothelial regrowth within the stenotic segment. Another explanation may be the formation of newly formed microvessels, which has been suggested to occur in human stenotic lesions derived from AV fistulas.10 The close correlation of subendothelial-intimal proliferation with the degree of endothelial cell coverage and the large percentage of proliferating endothelial cells covering microvessels in the vascular wall suggest that endothelial cells may be a major source of growth factors influencing proliferation patterns in human stenotic lesions. Interestingly, a substantial number of proliferating cells in the intimal-medial areas were identified as macrophages, which was observed previously in atherectomy-derived restenotic tissue.16 Also, restenosis after PTCA has been associated with activation and accumulation of macrophages.40 41 42 Extensive research on the role of macrophages in the development of atherosclerosis has suggested that these cells play a pivotal role in atherosclerosis progression. Furthermore, numerous studies have shown that macrophages can produce various growth factors and chemoattractants.28 29 43 44 45 So far, it is unclear how and to what extent these cells contribute to the initiation and perpetuation of intimal thickening in peripheral bypasses and AV fistulas.
The use of flow parameters like PSV and nPSV is the main limitation of the present study. Although nPSV is expressed in the same unit as shear rate, nPSV can only give a rough estimation of the actual shear stresses (defined as shear rate times local blood viscosity) exerted on the endothelial cell layer. In the systolic phase, the local velocity profile probably is blunted, especially in narrowed segments, as studied here. A blunted velocity profile leads to a very narrow angle between the velocity profile and the vessel wall, resulting in very high shear rates. However, the negative correlation between nPSV and endothelial cell coverage of the lumen indicates that nPSV may be of physiological importance. nPSV possibly gives a better characterization of the local pulsatile shear conditions than mean shear rate, which has been used by other investigators in similar flow models in animals25 26 and was calculated by use of the Hagen-Poiseuille formula. In those models, no correlation was observed between mean shear stress and proliferation.25
At this time, ultrasound systems to assess shear rate in humans are being developed. In a pilot study using such a system to assess shear rate in human carotid arteries of healthy volunteers, we observed an almost linear relation between midstream velocity and wall shear, which supports the use of nPSV.46
In summary, a proliferation rate of 0.9% to 3.7% was observed in anatomically intact human stenotic tissue derived from severely stenosed peripheral bypasses and AV fistulas. The data suggest an inhibitory effect of high shear rate on the development of intimal thickening in human vascular grafts because local normalized flow velocity was associated with a decreased stenotic proliferation. Moreover, the data suggest that the presence of endothelial cells covering the stenosis is associated with increased cellular proliferation.
Selected Abbreviations and Acronyms
|nPSV||=||peak systolic velocity normalized for diameter|
|PCNA||=||proliferating cell nuclear antigen|
|PSV||=||peak systolic velocity|
|VSMC||=||vascular smooth muscle cell|
We are indebted to Jean Willigers for her technical assistance. We thank Petra Aarts, Anique Janssen, and Monique Verluyten for their excellent technical support. We thank Drs J. Buth and M. Idu for supplying the samples from the Catharina Hospital Eindhoven. This study was supported by grants from the Netherlands Heart Foundation (NHS grants 90.055 and 93.165).
Preliminary results of this study were presented at the VIIIth International Symposium on the Biology of Vascular Cells, Heidelberg, Germany, August 30-September 4, 1994.
- Received January 17, 1996.
- Revision received March 21, 1996.
- Accepted March 26, 1996.
- Copyright © 1996 by American Heart Association
Swedberg SH, Brown BG, Sigley R, Wight TN, Gordon D, Nicholls SC. Intimal fibromuscular hyperplasia at the venous anastomosis of PTFE grafts in hemodialysis patients. Circulation. 1989;80:1726-1736.
Liu MW, Roubin GS, King SB III. Restenosis after coronary angioplasty: potential biologic determinants and role of intimal hyperplasia. Circulation. 1989;79:1374-1387.
Rekhter M, Nicholls S, Ferguson M, Gordon D. Cell proliferation in human arteriovenous fistulas used for hemodialysis. Arterioscler Thromb. 1993;13:609-617.
Clowes AW, Reidy MA. Prevention of stenosis after vascular reconstruction: pharmacologic control of intimal hyperplasia: a review. J Vasc Surg. 1991;6:885-891.
Dzau VJ, Safar ME. Large conduit arteries in hypertension: role of the vascular renin-angiotensin system. Circulation. 1988;77:947-954.
Pickering JG, Weir L, Jekanowski J, Kearny MA, Isner JM. Proliferative activity in peripheral and coronary atherosclerotic plaque among patients undergoing percutaneous revascularisation. J Clin Invest. 1993;91:1469-1480.
Verheyen R, Kuypers HJH, Schlingeman RO, Boehmer ALM, van Driel R, Brakenhof GJ, Ramakers FCS. Ki-67 detects a nuclear matrix-associated proliferation-related antigen. J Cell Sci. 1989;92:123-130.
O'Brien ER, Alpers CE, Stewart DK, Ferguson M, Tran N, Gordon D, Benditt EP, Hinchara T, Simpson JB, Schwartz SM. Proliferation in primary restenotic coronary atherectomy tissue: implications for antiproliferative therapy. Circ Res. 1993;73:223-231.
Zarins CK, Bomberger RA, Glagov S. Local effects of stenoses: increased flow velocity inhibits atherogenesis. Circulation. 1981;64(suppl II):II-221-II-227.
Zarins CK, Giddens DP, Bharadvaj BK, Sottiurai VS, Mabon RF, Glagov S. Carotid bifurcation atherosclerosis: quantitative correlation of plaque localization with flow velocity profiles and wall shear stress. Circ Res. 1983;53:502-514.
Asakura T, Karino T. Flow patterns and spatial distribution of atherosclerotic lesions in human coronary arteries. Circ Res. 1990;66:1045-1066.
Kohler TR, Jawien A. Flow affects development of intimal hyperplasia after arterial injury in rats. Arterioscler Thromb. 1992;12:963-971.
Kohler TR, Kirkman TR, Kraiss LW, Zierler BK, Clowes AW. Increased blood flow inhibits neointimal hyperplasia in endothelialized vascular grafts. Circ Res. 1991;69:1557-1565.
Hofstra L, Willigers JM, Huvers FC, Schaper NC, Kester ADM, Kitslaar PJEHM, Hoeks APG. Short-term variations in elastic properties of a muscular artery in humans. Clin Sci. 1993;50:472-480.
Ferrell M, Fuster V, Gold HK, Chesebro JH. A dilemma for the 1990s: choosing appropriate experimental animal model for the prevention of restenosis. Circulation. 1992;85:1630-1631.
Ferns GAA, Forster L, Stewart LA, Konneh M, Nourooz-Zadeh J, Anggard EE. Probucol inhibits neoinitimal thickening and macrophage accumulation after balloon injury in the cholesterol fed rabbit. Proc Natl Acad Sci U S A. 1992;89:11312-11316.
Moreno PR, Falk E, Palacios IF, Newell JB, Fuster V, Fallon JT. Macrophage infiltration in acute coronary syndromes: implication for plaque rupture. Circulation. 1994;90:775-778.
Hanke H, Hassenstein S, Ulmer A, Kamenz J, Oberhoff M, Haasse K, Baumbach A, Gaun AM, Karsch KR. Accumulation of macrophages in the arterial vessel wall following experimental balloon angioplasty. Eur Heart J. 1994;15:691-698.
Gordon D, Schwartz SM. Cell proliferation in human atherosclerosis. Trends Cardiovasc Med. 1991;1:24-28.
Nelken NA, Coughlin SR, Gordon D, Wilcox JN. Monocyte chemoattractant protein-1 in human atheromatous plaques. J Clin Invest. 1991;88:1121-1127.
Brands PJ, Hoeks APG, Hofstra L, Reneman RS. A non-invasive method to estimate wall shear rate using ultrasound. Ultrasound Med Biol. 1995;12:171-185.