View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

Eur J Vasc Endovasc Surg 23, 226–233 (2002) doi:10.1053/ejvs.2001.1556, available online at http://www.idealibrary.com on

Local Delivery of NO-Donor Molsidomine Post-PTA Improves Haemodynamics, Wall Mechanics and Histomorphometry in Atherosclerotic Porcine SFA

P. H. Rolland∗1, J. M. Bartoli1,2, Ph. Piquet1,4, C. Mekkaoui1, S. H. Nott5, G. Moulin1,2, P. Amabile1,4 and T. Mesana1,3

1Hemodynamics and Cardiovascular Mechanics Laboratory, School of Medicine, and Departments of 2Radiology and 3Cardiac Surgery, Hoˆpital La Timone, and 4Vascular Surgery, Hoˆpital Sainte-Marguerite, 27 Bd Jean-Moulin, 13385 Marseille Cedex 5, F; and 5Boston Scientific Corporation, Natick, Massachusetts, U.S.A.

Objectives: we investigated the therapeutic effect of angioplasty with local drug delivery (LDD) of the wall-accumulating NO-donor molsidomine (M) in the superficial femoral arteries (SFA) of atherosclerotic swine. Materials and methods: atherosclerotic Pietrin swines (n=14) underwent PTA-LDD-M (4 mg/2 ml) vs contralateral PTA-LDD-Placebo in the SFA using a channelled balloon angioplasty catheter. Invasive and colour Doppler energy (CDE) assessments of haemodynamics and wall mechanics were performed at 24 h (n=4) and 5 months (n=10). Immuno- histolabelling of cell proliferation and histomorphometry were serially performed in perfusion fixed SFA samples. Results: at 24 h, PCNA-positive nuclei revealed 33±14 and 12±3 proliferating cells/mm2 at placebo and molsidomine PTA-LDD sites, respectively (p<0.001). At 5 months, PTA-LDD-M vessels, compared with PTA-LDD-P, had increased compliance (66±9vs11±4 ml/mmHg) and lowered impedance (0.11±0.05 vs 0.45±0.14 mmHg/ml min−1)(p<0.05). CDE revealed low, middle and high velocity peaks at 7.5, 20 and 35, and 8, 15 and 22 cm s−1 in systolic and diastolic flows, respectively; and PTA-LDD-M prevented emergence of restenosis-associated increases in low blood velocities (p<0.01). PTA-LDD-M inhibited restenotic intimal thickening and medial thinning which decreased mean lumenal diameter in placebo-treated (2.6±0.3) as compared to molsidomine-treated (3.4±0.3 mm) vessels (p<0.05). Conclusions: in the atherosclerotic porcine SFA model, PTA-LDD with molsidomine consistently improved haemodynamic wall mechanics, lowered cell proliferation and prevented late lumen loss observed with PTA-LDD with placebo.

Key Words: Post-angioplasty restenosis; Molsidomine; Atherosclerotic pigs; Local drug delivery; NO-donors.

Introduction inhibit thrombosis,12,13 inhibit extra-cellular matrix deposition7,14–16 and promote the maintenance of nor- Following angioplasty in the superficial femoral artery mal, vascular smooth muscle function, and inhibit (SFA), only 50% of patients show long-term symp- inflammation.11,14–18 tomatic improvement.1–3 Specifically, the clinical out- The sydnonimines, especially molsidomine, liberate come appears to be dependent upon (i) haemodynamic NO leading to development of tolerance.19,20 They have success and restoration of wall mechanics post-inter- no specific intramural binding properties,11 but when vention, (ii) reactive and adaptive remodelling pro- coupled with the channelled balloon angioplasty cath- cesses,2,4,5 and (iii) thrombo-embolic complications.2,4 eter, there appears21–24 capable of efficient, transmural Atherosclerotic swine provide a good model of the drug delivery. Furthermore, this agent has a unique human situation.6–9 Local drug delivery (LDD) is a property of accumulating in significant amounts logical approach to post-angioplasty restenosis and within the heart muscle and arterial walls.25 In a por- occlusion in the SFA.8–10 Specifically, nitric oxide (NO)- cine model we have previously reported that these donors11 affect vascular tone and wall dynamics,11,12 compounds have a beneficial impact on restenosis through effects on NO and vascular smooth muscle cells.7,26–30 ∗ Please address all correspondence to: P. H. Rolland, Laboratoire d’He´modynamique et de Me´canique Cardiovasculaire, Faculte´ de However, early attempts to demonstrate ex- Me´decine, 27 Bd Jean-Moulin, 13385 Marseille Cedex 5. perimental and clinical benefits of administration of

1078–5884/02/030226+08 $35.00/0  2002 Elsevier Science Ltd. All rights reserved. NO-Donor Molsidomine Post-PTA 227 sydnonimines at the time of PTA,31,32 continued or not by oral administration of molsidomine, have yielded promising results with the conclusions that both chem- ical nature of the NO-donor and drug delivery device must be optimised in order to yield the expected anti-restenotic effect.26,32,33 The present study therefrom investigate the haemodynamic and morphologic con- sequences of molsidomine delivery with the chan- nelled angioplasty balloon in the SFA in the atherosclerotic swines.

Materials and Methods

Experimental protocol

The animals were handled in accordance with the Guide for the Care and Use of Laboratory Animals (National Research Council, Natl Acad Press, Wash- ington, DC, U.S.A., 1996). For 6 months, 14 male Pietrin pigs (40.9±4.2 kg at 6 months of age; Blossin SA, Aubagne, France) were fed the Marseille atherogenic diet which accelerates development of athero- sclerosis.6,30,31 At 6 months, angioplasty with local drug delivery (AP-LDD) was performed with molsidomine and placebo in bilateral superficial femoral arteries (SFA); and the animals were placed back on control diet.30,31 Twenty-four hours (n=4) and 5 months (n= 10) after AP-LDD, animals were submitted to an- giographic, haemodynamic, vascular wall rheologic and colour Doppler energy investigations. Vessels Fig. 1. Graph shows that cholesterol-related serum parameters were were pressure fixed at physiologic condition and increased in swines at time of AP-LDD (6 months, mean±s.d., n= 29 14). Microphotograph illustrates that atherosclerotic lesions in the sampled for histomorphological assessment. SFA contained a fibrous cap (FibC), a lipidic core (LpC) with focal necrosis (nec) and media intimalisation (MedInt) with elastic laminae disruption (red arrow) and reorientation and migration of smooth muscle cells (black arrow). Radiologic procedures and angioplasties saphenous artery (ISA) branch and the popliteal-pos- All procedures were performed on anaesthetised terior tibial arteries (PopA-postTA) bifurcation (Fig. pigs28,29 in dorsal position with the lower legs folded 1). The pre-procedure lumenal diameter by using the down thereby defining the radiologic posture. A digit- dimensional characteristics of the balloon. A 3 min ised subtraction angiographic Stenoscop system (Gen- inflation time at 10 atm was performed using a mixture eral Electric Medical System, Minneapolis, U.S.A.) was of saline and contrast medium (50:50, v/v) during used for radiologic procedures with access through which molsidomine delivery was conducted by a 2 min surgically isolated, right carotid artery (7F vascular perfusion time at 1 ml/min, at 6 atm (60 and 10 ml, sheath, Cordis/J&J, Miami, FL, U.S.A.). No heparin Leveen Inflators, BSC-MediTech). Molsidomine (4 mg/ was used. Routine angiograms were obtained from a 2 ml) and control saline solution (2 ml isotonic sodium NightHawk angiographic catheter (5F, 100 cm, BSC- chloride solution containing dextran 60 – 6%, p/v and MediTech, Watertown, MA, U.S.A.) advanced over glucose, 10%, p/v) in ampules were a kind gift from a glidex-coated hydrophilic 0.035 in guidewire (BSC- Dr Grewe (Hoechst, Taunus, Germany). The lumen MediTech). The channelled balloon angioplasty cath- diameter of the SFAs was 5.4±0.3 mm (n=14), with a eter (BSC-CB, 6×40 mm; Boston Scientific Corp, Na- 1.10–1.15:1 balloon/artery diameter ratio. The tick, MA, U.S.A.) was advanced in between the internal sequence for local drug delivery involved AP-LDD

Eur J Vasc Endovasc Surg Vol 23, March 2002 228 P. H. Rolland et al. using control saline in one SFA, randomly chosen; and repeated AP-LDD using molsidomine in the opposite SFA. Post-procedural angiograms and lumenal dia- meter measurements were obtained as described above. After surgical closure of puncture site and wound, the pig was transferred to the postoperative suite for recovery.

Haemodynamics and wall rheology in SFAs, echodoppler

From a local incision, SFA was carefully freed (without apparent disruption of blood supply to the adventitia) from the profunda femoralis branch down to the PopA- postTA arteries bifurcation. Haemodynamics and wall mechanics measurements29,34 were centred on AP-LDD site and consisted of intravascular measurements of pulsatile blood pressure with a 2F Mikro-Tip Catheter pressure transducer SPR-249 (Millar Instruments, Inc., Houston, TX, U.S.A.). Pulsatile flow measurements were carried out by using the T206 blood flow meter (Transonic Systems, Inc., Ithaca, NY, U.S.A.) and ap- propriate perivascular RB flowprobes. Pulsatile ex- ternal diameter was assessed by using a facing pair of ultrasonic probes (10 MHz) inserted within a silicone clip-probe, and connected to 4-channels sono- micrometer (Triton Technology Inc., San Diego, CA, U.S.A.). Systolic and diastolic blood flow velocities at the PTA site (as a frame fixed in depth and centred length in between ISA branch and PopA-PostTA bifurcation) were further non-invasively investigated by Colour Fig. 2. Angiograms show that luminal narrowing was observed in Doppler Energy (CDE) measurements of the lon- the native distal SFA (a). The angioplasty site was delimited by the gitudinal density profile of velocities (i.e. a range from internal saphenous artery (ISA) embranchment, proximally, and by the popliteal artery (PopA) – posterior tibial artery (postTA) 0 to 50 cm/s) along a middle-vessel line (i.e. to estimate bifurcation, distally. Low-pressure channel balloon inflation dis- velocities found in the middle of the vessel). Having played the stenotic lesion (b, arrow). Lipiodol infusion shows the a colour signal independent of the angular Doppler circumferential infusion pattern (c, arrow). Postprocedure an- giogram (d) of the native SFA in (a) illustrate the efficacy of the shots, a lower signal/noise ratio than velocity Doppler angioplasty with the channel balloon. At 5 months in the same and being devoid of spectral foldings (i.e. which led animal, control angiograms show the beneficial angiographic results to a high accuracy and sensitivity for the blood flows in the molsidomine-treated SFA (e), as compared to the placebo- treated SFA (f). in the low range), CDE signal intensity was proved to be observer independent and influenced by fluid flow velocity, only.35,36 At the same depth and instrument probe) displaying the varying pulsatile flow velocities settings (as is the case in the study), the CDE signal (see Fig. 3). From the video recordings, Adobe Premi- was found to be proportional to the returning signal ere software (Adobe Systems Inc.) was used to digitise, strength and to provide high-resolution cross-sectional extract and convert (into a 256 grey levels scale) the display of the vessel which can be stored on video tape individual systolic and diastolic images along the ori- or numerised and subsequently subjected to computer ginal cardiac cycles triggered by the R-wave. The analysis for extracting flow velocities from the scale SAMBA 2005 automatic scanning image analyser given by the instrument.35,36 With standardised ap- (TITN-Alcatel, Grenoble, France)6,7,27,30 enables drawing plication, CDE was recorded in SFAs under constant a mid-vessel line along the entire segment to maintain amplifying magnitude for 15 cardiac cycles with an identical length in systolic and diastolic images. The HDI 5000 ATL echodoppler instrument (7.5 MHz optical density (OD) profiles were recorded pixel by

Eur J Vasc Endovasc Surg Vol 23, March 2002 NO-Donor Molsidomine Post-PTA 229

Fig. 3. Graphs display the density count analysis at the location of data values (n=10, confidence intervals from 8 to 16% are not shown) of mid-vessel distribution of blood velocities measured by colour doppler energy (illustrated in insert) in the middle of the SFA (positioned as a white line), 5 months after angioplasty with local delivery of placebo (dashed line) or molsidomine (dark line), both in systolic (B, D) and diastolic (A,C) situations, respectively. Note that the drug reduced the angioplasty-induced enlargement in the range of CDE velocities by shifting the lowest velocities to the highest range values. pixel along the line and further converted to blood elastic fibres, or processed for immunohistological velocities according to the instrument-dependent col- PCNA labelling.28,32 Morphometric analysis was per- our map, treated accordingly. Finally the blood velocity formed under ×15–400 magnification light micro- values in individual cycles were averaged in the sys- scopy, using an automatic videoanalyser, SAMBA 2005 tolic and diastolic phases for both placebo or mol- (TITN-Alcatel, Grenoble, France).6,7,27,30 Densitometry sidomine-treated legs; and profiles of velocities in the analysis of PCNA staining was computerised as the averaged cycles were established using density count number of labelled nuclei per unit surface. In the technique in Systat 9.0 statistical software (SPSS Inc., lesional sites, morphometric analysis of the arterial Chicago, IL, U.S.A.). section geometry consisted of (i) delineating on the computer screen, the boundaries of the lumen, intima- media and media-adventitia junctions, (ii) centring 24 radial pairs of 2 orthogonal diameters on the arterial Morphometry and immunocytochemistry section, and (iii) measurement of 96 triplets of radial lumen, intima and media thickness. In each sample at Immediately after haemodynamics measurements, the the lesional site, the distribution of the morphometric animals were euthanised (midazolam 15 mg and chlor- data were further submitted to density count analysis, promazine 25 mg in KCl 15% (p/v), 20 ml, i.v. bolus) and further averaged to yield the final density count and serial pathologic slides, orthogonally sectioned profile per leg and per animal. (6
m) were obtained after in situ fixation with 10% formalin under a constant 120 mmHg perfusing pres- Statistical analysis sure.28,32 Deparaffined slides were stained with Mas- son’s trichrome for general observation or with the Data treatment and statistical analyses were performed Lillie and Fullmer’s orceine for specific staining of by using Systat 9.0 software (SPSS Inc., Chicago, IL,

Eur J Vasc Endovasc Surg Vol 23, March 2002 230 P. H. Rolland et al.

U.S.A.). Results are expressed as mean±standard de- mm2, at the AP-LDD with placebo and molsidomine, viation (SD). In constructing and analysing the blood respectively (p<0.001), which indicate that, 24 h after velocities and vessel dimensions, multicates in each angioplasty, molsidomine delivery maintained cell leg and in each subject were averaged to yield an proliferation state to baseline level of 9±2 stained individual value or profile subjected to dot density nuclei/mm2.27 Finally, local delivery of molsidomine count analysis along the horizontal data scale at the as compared to placebo delivery at the time of an- location of data values. Statistical analysis of the meas- gioplasty induced parallel magnification of both sys- ures for molsidomine-treated arterial sites vs controls tolic (374±54 vs 135±34 ml/min) and diastolic was performed by using the Kruskal–Wallis test (i.e. (176±51 vs 78±13 ml/min) flows with increased the non-parametric analogue of a one-way analysis of mean blood flow (374±54 vs 135±35 ml/min) variance), which reduced to the Mann–Whitney test (p<0.05), due to increased compliance in the treated (i.e. the non-parametric analogue of the two-sample t- sites (90±10 vs 27±5 ml/mmHg) (p<0.05), re- test) when there are only two groups. spectively.

Results Angioplasty with LDD, mid-term results at 5 months

Atherosclerosis model As they were in pre-procedure angiograms, luminal diameter narrowing was observed in some of the The animals remained healthy throughout the pro- placebo-treated sites (Fig. 2). The early vasodilatory gram period. At 6 months, the atherogenic diet in- changes in the haemodynamics and wall mechanics creased cholesterol-related serum parameters in the induced by molsidomine delivery during angioplasty swines (Fig. 1) without significant additional changes persisted at 5 months, as compared to the placebo in the serum biochemistry.6,27 the type IIa hy- sites: both maximum (270±98 vs 171±82 ml/min) percholesterolemia-induced atherosclerotic lesions and minimum (85±57 vs 57±14 ml/min) flows had were detectable under gross pathologic examination parallel magnification due to increased compliance of the SFA proximal to the angioplasty site at 24 h (66±9vs11±4 ml/mmHg) and lowered entry im- − (Fig. 1). These lesions had the common pathological pedance (0.11±0.05 vs 0.45±0.10 mmHg/ml min 1) appearance relevant to the reported assessment of (p<0.05), which also indicate the worsening situation human lesions.6,7,27 Upon examination of the pre- at the placebo-treated site. angioplasty arteriograms, angiographic reduction in In dot histograms (Fig. 3) showing the density (with luminal diameter was observed (Fig. 2a) in some a unique symbol as the relative concentration of data instances but not generally influencing the average points in intervals) of every velocity across the range diameter. of the distribution, non-invasive haemodynamic in- vestigations in the SFA at 5 months demonstrated that, in SFAs following angioplasty with placebo delivery, blood flow elicited a 3 peak-shaped distribution of Angioplasty with LDD velocities in both the systolic and diastolic situations due to the emergence of a lowest velocity group. By The BSC-CB device was placed and inflated in the SFA definition, low, middle and high velocities peaked at in all cases (Fig. 2a). No acute complication occurred 7.5, 20 and 35 cm s−1 and 8, 15 and 22 cm s−1 in the during intraSFA insertion, deployment, liquid solution systolic and diastolic flows, respectively. The results delivery or retrieval of the balloon along the guidewire. show that both diastolic and systolic flows were in- The average SFA lumen diameter at the dilatation site creased at the molsidomine delivery sites, due to a was similar in both SFAs before angioplasty: shift of the lowest velocities to the higher range of the (5.5±0.2 mm; n=28 vessels). At 24 h (n=4), the SFA spectrum. Specifically, the density counts in the systolic luminal diameter was 5.7±0.2 mm (n=4) and peaks of velocity were 9.4±1.1 vs 0.7±0.1∗, 40.2±3.9 5.9±0.3 mm (n=4) at the AP-LDD with placebo and vs 28.8±3.1, and 48.9±5.4 vs 103.2±12.1 counts for molsidomine, respectively, which shows that equi- the placebo vs molsidomine-treated angioplasty sites valent angioplasty was performed at both sites. In the (∗: p<0.01), respectively; and similarly the density arterial wall at 24 h, the number of PCNA-positive counts for the diastolic flows, the density counts were nuclei, i.e. the proliferating nuclei, was found to be 19.1±3.0 vs 8.7±1.6∗, 93.8±9.6 vs 99.2±7.3, and 33±14 stained nuclei/mm2 and 12±3 stained nuclei/ 37.5±5.2 vs 139.4±11.2 counts for the placebo vs

Eur J Vasc Endovasc Surg Vol 23, March 2002 NO-Donor Molsidomine Post-PTA 231

1.10±0.12 mm in the molsidomine-treated site, whereas the placebo-treated site had a dramatically enlarged peak centred at 0.85±0.24 mm (p<0.001). Finally, the mean lumenal radius in the placebo-treated and molsidomine-treated PTA sites were 2.6±0.34 and 3.4±0.31 mm, respectively (p<0.05). The results indicate that PTA-LDD-M prevented the increases in both the intima thickness and intima-media ratio, as well as the media thinning. As a consequence, the lumenal radius was lowered by 24% in the placebo- treated arm as compared to the treated arm (p<0.05).

Discussion

The main result of the present study is that local delivery of the aqueous perfusate of NO-donor parent compound molsidomine at the time of angioplasty Fig. 4. Graphs display the morphometry of the lesional site in the with the channelled balloon catheter led to late lumen SFA 5 months after angioplasty with local delivery of placebo preservation superior to those of PTA-LDD with (dashed line) (insert A and B) or molsidomine (dark line) (insert C), as the density count analysis at the location of data values placebo, as well as to beneficial post-angioplasty (n=10, confidence intervals from 7 to 14% are not shown) of changes in haemodynamics and wall mechanics, and to computerised thickness of the media (a) and intima (b) and the inhibition of proliferative response in the SFA arterial radius length (c) along 96 radiant radii per slide. Inserts are orceine- stained slides (×250). walls. The data further demonstrated that local de- livery of molsidomine in the porcine SFA minimised occurrence of (i) a low-velocity pattern in blood flow, molsidomine-treated angioplasty sites (∗: p<0.01), re- known to favour development of arterial lesions,37,38 spectively. Therefrom, AP-LDD with molsidomine cre- and (ii) a focalised intimal thickening and converse ated a homogeneous blood flow distribution and the media thinning with an increased intima/media ratio, treatment did not increase the blood flow velocities thereby leading to a larger lumenal diameter in treated outside of the spectrum range in the placebo-treated vessels. Finally, the safety of PTA procedure is sup- SFAs. ported by early angiographic results which did not Pathologic examinations of the PTA sites in the show any significant arterial damages or differences explanted SFA segments at 5 months revealed non- between LDD with placebo or molsidomine. lipidic fibrosclerotic and cell proliferations and al- Haemodynamics and wall mechanics at endo- terations of the elastic structures, histologic char- vascular intervention play a significant role in the acteristics of restenotic processes with (Fig. 4). location, development, progression of atherosclerosis, Morphometric analysis of the lesions identified the early and late restenosis, as well as the propensity restenosis-associated intimal thickening and converse for subacute thrombosis.37,38 Consequently, in the SFA medial thinning; and increased intima-media ratio and it is established that return to normal ankle-brachial lowered lumenal diameter (Fig. 4). Specifically, intimal pressure index (ABPI >0.9 at 24 h) is a strong thickness in the placebo-treated site was characterised indicator of a favourable long-term prognosis after by the emergence of a density peak at 0.75±0.12 mm angioplasty.1 By investigating the vessel compliance and an increased density peak at 0.48±0.14 mm which estimates how the elastic wall reacts in (p<0.05 as compared with the molsidomine-treated response to the haemodynamic forces, as well as site); whereas the molsidomine-treated sites had a how the vessel recoils to further propel blood homogeneous peak (at 0.21 mm) in intimal thickness. flow along the vasculature,39 we have shown that Similarly, the intima-media ratio evidenced two dens- molsidomine delivery led to a persistent increase in ity peaks (centred at 0.28 (80%) and 0.60 (20%)) in the the SFA vascular strain (i.e. changes in volume due placebo-treated sites, whereas a single density peak at to wall circumferential elongation) in response to 0.16 was characteristic of the intima/media ratio in the stress (i.e. distending blood pressure wave). This the molsidomine-treated site. The medial thickness in turn resulted in decreased entry impedance and peak was uniformly shaped and centred at increased blood flow with constant pulse flow. The

Eur J Vasc Endovasc Surg Vol 23, March 2002 232 P. H. Rolland et al. latter is further demonstrated by colour Doppler Future directions/next steps energy investigations revealing that PTA-LDD-M prevented emergence of restenosis-associated low We report here the mid-term therapeutic efficacy of blood velocities. As mechanical forces play an im- local molsidomine delivery at the time of angioplasty portant role in the modification of the vascular under the experimental conditions of pre-existing ath- structure, it is presently accepted that the vessel erosclerotic lesions in the porcine SFA. The ex- remodels its geometry, structure and composition in perimental protocol was based upon technically response to angioplasty in a way to produce a more challenging solutions for simultaneous local drug de- homogeneous stress distribution from the inner wall livery and angioplasty derived to address the key to the outer wall.40 It is also acknowledged that determinants in SFA restenosis. The design of the these remodelling targets may be reached at the channelled angioplasty balloon enables delivery of expense of pathological changes in wall metabolism adequate quantities of drug to the vessel wall without and mechanical properties.41 The residual strains, either injuring the wall or compromising flow. The which determine the stress distribution through the commercially available, aqueous i.v. perfusate allows artery wall following geometrical changes,40 are penetration, retention and convection of the 4 mg ad- concentrated in the inner, intimal layers, which ministered drug by using the increased hydraulic con- contain more elastin than the collagenous outer ductance of atherosclerosis. The optimisation of the layers which contribute little to the residual strains. NO-donor molecule, which is retained for several From these, the inner layers remodel themselves in days and continuously released, avoids the risk of cell response to stress concentration by developing more toxicity due to a concentrated localised presence of compressive residual strains than the outer layers.41 NO within the vessel wall. Additionally, NO, as the By preserving the integrity of the internal elastic active agent, possesses the “multi-purpose” phar- laminae, molsidomine delivery obviously favours the macologic properties addressing the “multi-com- latter processes. On the other hand, changes in ponent” compensatory responses of the restenotic arterial compliance, detectable as early as at 24 h after development in the atherosclerotic SFA. molsidomine delivery, may be due to a synergistic Finally, caution has to be taken in extrapolating the combination of adaptive structural changes and of present results because of the reported disparity laying relaxation in arterial smooth muscle tone. This would between animal model data and clinical trial ex- in turn alter the relative loading of collagen and perience.16 The efficacy for the present protocol es- elastin fibres by transferring stress from the less tablished for angioplasty with local molsidomine extensive collagen fibres to elastin.42 The net result delivery will be examined in a controlled clinical trial of these structural adaptations is that intimal layers as the next step. and endothelial cells are in a new dynamic equi- librium with their ambient fluid and solid mechanical environment.43 The observation of beneficial, thera- peutic effects for molsidomine delivery implies that References cellular activation in the vessel wall at time of 1Golledge J, Ferguson K, Ellis M et al. Outcome of femoro- angioplasty are attainable by the NO-donor delivery popliteal angioplasty. Ann Surg 1999; 229: 146–153. within the early hours after angioplasty. At the same 2Johnston KW. Femoral and popliteal arteries: reanalysis of time and later on, arterial wall remodels itself to results of balloon angioplasty. Radiology 1992; 183: 767–771. 3Krepel VM, van Andel GJ, van Erp WF, Breslau PJ. Per- reach a new distribution for the residual stress under cutaneous transluminal angioplasty of the femoropopliteal ar- conditions of vasorelaxation, as the production of tery: initial and long-term results. Radiology 1985; 156: 325–328. NO from molsidomine promotes blood flow by 4Sapoval MR, Long AL, Raynaud AC et al. Femoropopliteal stent placement: long-term results. Radiology 1992; 184: 833–839. inhibiting smooth muscle contraction and platelet 5Glagov S. Intimal hyperplasia, vascular remodeling, and the aggregation and adhesion.44 Therefore, it is con- restenosis problem. Circulation 1994; 2888–2891. ceivable that harmonious and moderate haemo- 6Charpiot P, B escond A, Augier T et al. Hyperhomocysteinemia induces elastolysis in minipig arteries: structural consequences, dynamic and mechanical forces induced by arterial site specificity and effect of captopril-hydro- angioplasty plus molsidomine, in conjunction with chlorothiazide. Matrix Biol 1998; 17: 559–574. the cellular therapeutic effects of the low, continuous 7Augier T, Charpiot P, C hareyre C et al. Medial elastic structure alterations in atherosclerotic arteries in minipigs: plaque prox- release of NO from 4 mg molsidomine for the 4 imity and arterial site specificity. Matrix Biol 1997; 15: 455–467. days the drug was last in the arterial walls, may 8Geary RL, Adams MR, Benjamin ME, Williams JK. Conjugated exert a coordinated influence on intimal structures equine estrogens inhibit progression of atherosclerosis but have no effect on intimal hyperplasia or arterial remodeling induced and endothelial functions, the net effect of which is by balloon catheter injury in monkeys. J Am Coll Cardiol 1998; anti-restenotic. 31: 1158–1164.

Eur J Vasc Endovasc Surg Vol 23, March 2002 NO-Donor Molsidomine Post-PTA 233

9Mondy JS, Williams JK, Adams MR, Dean RH, Geary RL. cells but not adventitial contribution. Circulation 1997; 96:I- Structural determinants of lumen narrowing after angioplasty 84–88. in atherosclerotic nonhuman primates. J Vasc Surg 1997; 26:27Latrille V, Ghiringhelli O, Jourdheuil-Rahmani D et al. 875–883. Long-term treatment of atherosclerotic minipigs with isosorbide 10 Faxon DP. Effect of high dose angiotensin-converting enzyme dinitrate restores nitric oxide release from endothelial cells, and inhibition on restenosis: final results of the MARCATOR Study, inhibits vascular smooth muscle cell proliferation. Endothelium a multicenter, double-blind, placebo-controlled trial of cilazapril. 1996; 4: 235–245. The Multicenter American Research Trial With Cilazapril After 28 Rolland PH, Friggi A, Barlatier A et al. Hyper- Angioplasty to Prevent Transluminal Coronary Obstruction and homocysteinemia-induced vascular damage in the minipig. Cap- Restenosis (MARCATOR) Study Group. J Am Coll Cardiol 1995; topril-hydrochlorothiazide combination prevents elastic 25: 362–369. alterations. Circulation 1995; 91: 1161–1174. 11 Ignarro LJ, Lippton H, Edwards JC et al. Mechanism of vascular 29 Rolland PH, Charifi AB, Verrier C et al. Hemodynamics and smooth muscle relaxation by organic nitrates, nitrites, nitro- wall mechanics after stent placement in swine iliac arteries: prusside and nitric oxide: evidence for the involvement of S- comparative results from six stent designs. Radiology 1999; 213: nitrosothiols as active intermediates. J Pharmacol Exp Ther 1981; 229–246. 218: 739–749. 30 Rolland PH, Charpiot P, F riggi A et al. Effects of angiotensin- 12 Rolland PH, Sampol J, Lacarelle B et al. Differential inhibitory converting enzyme inhibition with perindopril on hemo- effects of isosorbide dinitrate and its mononitrate metabolites dynamics, arterial structure, and wall rheology in the hind- on platelet aggregation and thromboxane formation. Pharmacol quarters of atherosclerotic mini-pigs. Am J Cardiol 1993; 71: Res 1985; 2: 87–90. 22E–27E. 13 Berenger FP, Cano JP, Rolland PH. Antithrombogenic endo- 31 Lablanche JM, Grollier G, Lusson JR et al. Effect of the thelial cell defense. Basal characteristics in cultured endothelial direct nitric oxide donors linsidomine and molsidomine on cells and modulation by short-term and long-term exposure to angiographic restenosis after coronary balloon angioplasty. The isosorbide nitrates. Circ Res 1987; 60: 612–620. ACCORD Study. Angioplastic Coronaire Corvasal Diltiazem. 14 Garg UC, Hassid A. Nitric oxide-generating vasodilators and Circulation 1997; 95: 83–89. 8-bromo-cyclic guanosine monophosphate inhibit mitogenesis 32 Kalinowski M, Alfke H, Bergen S et al. Comparative trial of and proliferation of cultured rat vascular smooth muscle cells. J local pharmacotherapy with L-arginine, r-Hirudin, and mol- Clin Invest 1989; 83: 1774–1777. sidomine to reduce restenosis after balloon angioplasty of steno- 15 Sarkar R, Meinberg EG, Stanley JC, Gordon D, Webb RC. tic rabbit iliac arteries. Radiology 2001; 219: 716–723. Nitric oxide reversibly inhibits the migration of cultured vascular 33 Groves PH, Banning AP, Penny WJ et al. The effects of exo- smooth muscle cells. Circ Res 1996; 78: 225–230. genous nitric oxide on smooth muscle cell proliferation following 16 Kolpakov V, Gordon D, Kulik TJ. Nitric oxide-generating porcine carotid angioplasty. Cardiovasc Res 1995; 30: 87–96. compounds inhibit total protein and collagen synthesis in cul- 34 Mekkaoui C, Friggi A, Rolland PH et al. Simultaneous meas- tured vascular smooth muscle cells. Circ Res 1995; 76: 305–309. urements of arterial diameter and blood pressure to determine 17 Wever RMF, Luscher TF, Cosentino F, Rabelink TJ. Athero- the arterial compliance, wall mechanics and stresses in vivo. Eur sclerosis and the two faces of endothelial nitric oxide synthase. J Vasc Endovasc Surg 2001; 21: 208–213. Circulation 1998; 97: 108–112. 35 Hoyer PF, Schmid R, Wunsch L, Vester U. Color doppler 18 Polte T, Abate A, Dennery PA, Schroder H. Heme oxygenase- energy – a new technique to study tissue perfusion in renal 1 is a cGMP-inducible endothelial protein and mediates the transplants. Pediatr Nephrol 1999; 13: 559–563. cytoprotective action of nitric oxide. Arterioscler Thromb Vasc Biol 36 Gudmundsson S, Valentin L, Pirhonen J et al. Factors affecting 2000; 20: 1209–1215. color doppler energy ultrasound recordings in an in-vitro model. 19 Messin R, Karpov Y, B aikova N et al. Short- and long-term Ultrasound Med Biol 1998; 24: 899–902. effects of molsidomine retard and molsidomine nonretard on 37 Davies PF. Spatial hemodynamics, the endothelium, and focal exercise capacity and clinical status in patients with stable angina atherogenesis: a cell cycle link? Circ Res 2000; 86: 114–116. – a multicenter randomized double-blind crossover placebo- 38 Giddens DP, Zarins CK, Glagov S. The role of fluid mechanics controlled trial. J Cardiovasc Pharm 1998; 31: 271–276. in the localization and detection of atherosclerosis. J Biomech Eng 20 Rosenkranz B, Winkelmann BR, Parnham MJ. Clinical phar- 1993; 115: 588–594. macokinetics of molsidomine. Clin Pharmacokinet 1996; 30: 372– 39 Drexler H. Nitric oxide and coronary endothelial dysfunction 384. in humans [review]. Cardiovasc Res 1999; 43: 572–579. 21 Baldwin AL, Wilson LM, Gradus-Pizlo I, Wilensky R, March 40 Delfino A, Stergiopulos N, Moore JE, Meister JJ. Residual K. Effect of atherosclerosis on transmural convection an arterial strain effects on the stress field in a thick wall finite element ultrastructure. Implications for local intravascular drug delivery. model of the human carotid bifurcation. J Biomech 1997; 30: Arterioscler Thromb Vasc Biol 1997; 17: 3365–3375. 777–786. 22 Brieger D, Topol E. Local drug delivery systems and prevention 41 Greenwald SE, Moore JE, Rachev A, Kane TP, Meister JJ. of restenosis. Cardiovasc Res 1997; 35: 405–413. Experimental investigation of the distribution of residual strains 23 Hong MK, Barry JJ, Leon MB. Multichannel balloon catheter. in the artery wall. J Biomech Eng 1997; 119: 438–444. Semin Interv Cardiol 1996; 1: 34–35. 42 Belz GG. Elastic properties and Windkessel function of the 24 Hong MK, Wong SC, Farb A et al. Localized drug delivery in human aorta. [Review] [87 refs]. Cardiovasc Drugs Ther 1995; 9: atherosclerotic arteries via a new balloon angioplasty catheter 73–83. with intramural channels for simultaneous local drug delivery. 43 Gimbrone MA, Topper JN, Nagel T, Anderson KR, Gar- Cathet Cardiovasc Diagn 1995; 34: 263–270. cia-Cardena G. Endothelial dysfunction, hemodynamic forces, 25 Fromson JM, Illing HP, Ings RM et al. Absorption and dis- and atherogenesis. Ann NY Acad Sci 2000; 902: 230–239. position of [14C]-molsidomine in laboratory animals. Arz- 44 Hironaka K, Yano M, Kohno M et al. In vivo aortic wall neimittelforschung 1981; 31: 337–345. characteristics at the early stage of atherosclerosis in rabbits. Am 26 Rolland PH, Charifi AB, Jourdheuil-Rahmani D, Piquet P, J Physiol 1997; 273: H1142–H1147. Bartoli JM. Nitric oxyde-donor (Molsidomine) local delivery during angioplasty prevents restenosis due to smooth muscle Accepted 7 November 2001

Eur J Vasc Endovasc Surg Vol 23, March 2002