Association of Primary Varicose Veins with Dysregulated Vein Wall Apoptosis

Eur J Vasc Endovasc Surg 35, 224e229 (2008)

doi:10.1016/j.ejvs.2007.08.015, available online at http://www.sciencedirect.com on

Association of Primary Varicose Veins withDysregulated Vein Wall Apoptosis

E. Ducasse,1,2,3* K. Giannakakis,3 F. Speziale,2 D. Midy,1

E. Sbarigia,2 J.C. Baste1 and T. Faraggiana3

1Unit of Vascular Surgery, Hospital Tripode-Pellegrin,Universite de Bordeaux 2, CHU de Bordeaux, Bordeaux, France,

2Unit of Vascular Surgery, Clinic Umberto I, University ‘‘La Sapienza’’, Rome, Italy, and3Department Experimental Medicine and Pathology, Section of Ultrastructural Pathology,

University ‘‘La Sapienza’’, Rome, Italy

*CorrehospitE-mail

1078–5

Background. Disordered programmed cell death may play a role in the development of superficial venous incompetence.We have determined the number of cells in apoptosis, and the mediators regulating the intrinsic and extrinsic pathways inspecimens of varicose vein.Methods. Venous segments were obtained from 46 patients undergoing surgical treatment for primary varicose veins.Controls samples were obtained from 20 patients undergoing distal arterial bypass grafting surgery. Segments of the distaland proximal saphenous trunk as well as tributaries were studied. Cell apoptoses and mediators of the mitochondrial andtrans membrane pathway were evaluated with peroxidase in situ apoptosis detection, Bax and Fas detection, caspase-9 and8 detection in the medial layer.Results. Disorganised histological architecture was observed in varicose veins. Primary varicose veins also contained fewerperoxidase in situ-positive cells than control veins (2.6% S.D. 0.2% versus 12% S.D. 0.93%, P¼ .0001, Mann-Whitney utest), fewer Bax positive cells (2.1.% S.D. 0.3% versus 13% S.D. 0.9%, P¼ .0001) and fewer Caspase 9 positive cells(3.2% S.D. 1% versus 12% S.D. 1.3%, P¼ .0001). Similar findings were observed in saphenous trunk, main tributariesand accessory veins. In patients with recurrent varicose veins in whom the saphenous trunk had been preserved showedsimilar findings to primary varicose veins. Residual varicose veins contained fewer peroxidase in situ-positive cells thanhealthy veins (3.2% S.D. 0.6% versus 11% S.D. 2%, P¼ .0001), fewer Bax positive cells (2.2% S.D. 0.3% versus12% S.D. 0.7%, P¼ .0001) and fewer Caspase 9 positive cells (2.6% S.D. 0.6% versus 12% S.D. 1%, P¼.0001). Immu-nohistochemical detection for Fas and caspase 8 remained equal was the same in the varicose vein and control groups.Conclusion. Apoptosis is down regulated in the medial layer of varicose veins. This dysregulation is attributable to a dis-order of the intrinsic pathway and involves the great saphenous vein trunk, major tributaries and accessory veins. Thisprocess may be among the causes of primary varicose veins.� 2007 Published by Elsevier Ltd on behalf of European Society for Vascular Surgery.

Keywords: Apoptosis; Varicose vein; Primary varicose veins; Vascular apoptosis; Mitochondrial pathway.

Introduction

Current research indicates that dysregulation in thenumber of parietal cells undergoing apoptosis in thevenous wall is important in the onset and progressionof human primary varicose veins.1e5 Studies in hu-man renal, coronary and carotid arteries have shownincreased apoptosis and increased numbers of intrin-sic pathway promotors including the apoptotic

sponding author. Dr. E. Ducasse, Unit of vascular surgery,al tripode-pellegrin, tripode 5eme, 33076 Bordeaux, France.address: [email protected]

884/000224 + 06 $34.00/0 � 2007 Published by Elsevier Ltd on

promoter Bax.6 The apoptotic process is regulated byspecific activation of a family of intracellular cysteineproteases known as caspases (cysteine aspartyl-specificproteases) inducible by various cellular and externalstimuli.7 Two regulatory pathways are recognised.The intrinsic pathway (mitochondria pathway or typeI) is regulated by apoptosis promoter proteins [suchas Bax or Bcl-2] and involves specific caspases [espe-cially caspase 9] along with the participation of mito-chondria. The extrinsic pathway (trans-membranepathway or type II) is regulated by proteins [such asfas, the tumour-necrosis factor receptor (TNF-R) familyproteins] and involves specific caspases [especially

behalf of European Society for Vascular Surgery.

225Primary Varicose Veins

caspase 8] connecting ligand binding at the cell surfaceto apoptosis induction.8e11 Apoptotic dysregulation inthe superficial vein wall of varicose veins is mediatedvia the intrinsic pathway. This involves the whole sa-phenous vein independently of whether a particularsegment is incompetent.12

The aim of the present study was to correlate theextent of histological changes in varicose veins withmediators regulating the intrinsic and the extrinsicpathways of apoptosis by detecting specific proe-tins.11e16 We wished to evaluate these phenomena inthe saphenous trunks, major tributaries and accessoryveins. To find out whether apoptotic dysregulationwas a reversible process, we compared apoptotic cel-lular regulation in the medial layer in the proximalgreat saphenous veins in limbs previously strippedof the distal saphenous trunk and tributaries as partof varicose vein treatment.

Material and Methods

The protocol for this study was considered by the lo-cal committee on medical ethics and all patients in-cluded in this work gave their informed writtenconsent.

Cohort

In protocol 1, a total of 26 varicose veins were obtainedfrom 26 patients who were undergoing lower-extremityvaricose vein surgery for primary varicose veins. Thecharacteristics of patients are given in Table 1. The oper-ative indications classified according to the Clinical Eti-ological Anatomical and Physiopathological (CEAP)

Table 1. Characteristics of the 46 patients with healthy and pri-mary varicose veins

Varicose veins(n¼ 30)

Healthy veins(n¼ 20)

Number of patients 26 20GenderMen: women 7: 19 6: 14Mean age (years) 51 S.D. 9 58 S.D. 7Height (cm) 167 S.D. 5 168 S.D. 6Weight (kg) 76 S.D. 12 68 S.D. 13Body mass index (BMI) 27 S.D. 6 24 S.D. 2< 20 4 320e26 9 13> 26 13 4

Clinical EtiologicalAnatomical and Physiopathological classification:

Type 2 2 —Type 3 21 —Type 4 3 —

All data are expressed as means and SD.

system as class 2 (n¼ 4), class 3 (n¼ 15) and class 4(n¼ 7). Pre-operative duplex ultrasound confirmedthat the GSV was incompetent and had a diameter ofat least 5 mm. A total 20 healthy veins were obtainedfrom 20 patients who were undergoing distal arterial by-pass grafting surgery. Duplex ultrasonography con-firmed that these veins were competent. For all thevenous specimens, the segment analyzed was the greatsaphenous vein one centimetre distal to the sapheno-femoral junction. In addition, the termination of the 3principal tributaries of the GSV (anterior accessory sa-phenous vein, superficial epigastric and external puden-dal veins) were removed for analysis. Three accessoryveins were resected to analyse segments along the greatsaphenous vein trunk in varicose and healthy veins.

In protocol 2, a total of 20 segments from varicoseveins were obtained from 20 patients previously oper-ated upon for varicose veins. Operations had beenperformed a mean of 4.6 S.D. 1.6 years previouslysparing the proximal part of the great saphenousvein. Segments of residual GSV were obtained 1 cmdistally to the SFJ and were immuno-histochemicallyanalysed and compared with equivalent venous seg-ments obtained from healthy patient group. A total20 healthy veins were obtained from 20 patientswho were undergoing distal arterial bypass graftingsurgery.

Tissues specimens

All venous segments were fixed in 10% neutralbuffered formalin solution and paraffin embedded.Vein specimens were sectioned (3 mm) (LEICAJUNG-RM-2035) perpendicular to the axis of the ves-sel, when the original venous segment was longenough the whole block was sectioned and sectionswere fixed each 5 mm (two to five sections perblock/slide).

Histology

Specimens were stained with haematoxylin and eosinfor routine histological evaluation. Trichrome stainingwas used to assess vein wall collagen fibres. Victoriablue was used to study elastin fibres. Results wereevaluated by two independent investigators whowere blinded to the patients’ clinical findings.

Immunohistochemical detection

Cell apoptoses, mediators of the intrinsic mitochon-drial and extrinsic trans membrane pathways weredetected with five primary antibodies: Apoptag� per-oxidase in situ Apoptosis Detection Kit e S7101

Eur J Vasc Endovasc Surg Vol 35, February 2008

226 E. Ducasse et al.

(BIOTECH, Andover, MA, USA) (diluted: 33/110);rabbit polyclonal bax P-19 e SC526 (SANTA CRUZBiotechnology, INC) (diluted 1/40); rabbit polyclonalantibody Caspase 9 Ab-4 (Cat RB-1205 e P1ABX)(NEOMARKERS, Fremont, CA, USA) (diluted 1/40);rabbit polyclonal antibody Fas Ab-2 (ONCOGENE,code N� AM01 L) (dilution 1/100); and polyclonal an-tibody Caspase 8 Ab-4 (category RB-1200-P1ABX,NEOMARKERS, Fremont, CA, USA) (dilution 1/40).

The immunohistochemical techniques used havebeen detailed elsewhere.5

For the histological analysis four fields per sectionwere examined for staining at 400X magnification.The immunohistochemical expression of apoptosiswas determined according to the intensity of nuclearstaining and the density and morphology of chroma-tin. Results were evaluated by two independent inves-tigators who were blinded to the patients’ clinicalfindings. Peroxidase in situ, Bax, fas, caspase 8 and9 -immunoreactive cells considered as positive for ap-optosis in the medial layer were counted manually at400X magnification and 5 random fields per sectionwere counted. Two to five sections per specimenwere evaluated. Immunohistochemical positive cellstaining was not investigated in the intima andvasa-vasorum because previous studies found no dif-ferences in intimal apoptosis in varicose and healthyveins.3 Nor was positive cell staining investigated inthe adventitia because this layer is more severelyand diffusely disorganised and cell counts were thusmore approximate in varicose veins than in healthyveins. The apoptotic index was calculated as the num-ber of positive cells in each field divided by numbertotal of cells in the field multiplied by 100.

Table 2. Immunohistochemical quantification of the specimensfrom healthy and primary varicose veins

Number of positive Varicose veins Healthy veins *P value

Statistical analysis

All data are expressed as means and standard devia-tion. Immunohistochemical results are expressed asthe apoptotic index for the five antibodies tested.Data obtained were analyzed by the Mann-Whitneytest comparing the results obtained in the two groups.A P value less than .05 was considered to be statisti-cally significant. Statistical analyses were performedwith StatView (SAS Institute, Cart, NC, USA)software.

stained cells (in %)

Peroxidase in situapoptosis detection

2.6 S.D. 0.2 12 S.D. 0.9 .0001

Bax detection 2.1 S.D. 0.3 13 S.D. 0.9 .0001Caspase 9 detection 3.2 S.D. 1 12 S.D. 1 .0001Fas detection 11 S.D. 2 13 S.D. 2 .65Caspase 8 detection 11 S.D. 1 10 S.D. 1 .61

All data are expressed as mean and SD. * Mann-Whitney test.

Results

Histology standard

Varicose veins from the saphenous trunk, tributariesand accessory veins exhibited severely disorganised

Eur J Vasc Endovasc Surg Vol 35, February 2008

parietal architecture. In trichrome-stained specimens,the medial layer showed an increased number of col-lagen fibres with a decrease in smooth muscular cellsin the sub-intimal layer. The total number of parietalcells has not been compared between the two groups.The adventitial layer was thinner in all varicose veins(distal trunk, main branches and collaterals) than incomparable healthy veins and contained more colla-gen fibres. In Victoria blue stained specimens, the me-dial layer consisted of elastin fibres constantlyfragmented with parcelled interruption of the internalelastic lamina. The adventitial layer showed de-creased density and size of elastin fibres and in-creased fibre degradation.

Immunohistochemical staining

PROTOCOL 1 (Table 2)Peroxidase in situ apoptosis detection. Fig. 1(a and b)Varicose veins arising in the trunk, main tributariesand accessory veins contained fewer peroxidase insitu-positive cells than comparable healthy veins. Nodifference was found in the apoptotic index betweenthe saphenous trunk, main tributaries and accessoryveins (data not shown).

Bax detection. The apoptosis promoter index ofpositive cell staining differed significantly in the twogroups: varicose veins from the saphenous trunk,main tributaries and accessory veins contained fewerBax positive cells than healthy veins. No differencewas found in the Bax detection index between the sa-phenous trunk, main tributaries and accessory veins(data not shown).

Caspase 9 detection. Again, the apoptosis promoterindex of positive cell staining differed significantly inthe two groups: varicose veins with distal trunk,main branches and collaterals mixed together con-tained fewer caspase 9 positive cells than healthy veinsNo difference was found in Caspase 9 detection indexbetween the saphenous distal trunk, main branchesand collaterals (data not shown).

Fig. 1. (a, b): Peroxidase in situ apoptosis detection showingfewer peroxidase in situ-positive cells in varicose veins(Fig. 1a) than in healthy veins (Fig. 1b) (original magnifica-tion, X 400).

Table 3. Immunohistochemical quantification of the specimensfrom healthy and previously treated varicose veins

Number of positivestained cells (in %)

Varicose veins Healthy veins *P value

Peroxidase in situapoptosis detection

3.2 S.D. 0.61 11 S.D. 1.9 .0001

Bax detection 2.2 S.D. 0.3 12 S.D. 0.7 .0001Caspase 9 detection 2.6 S.D. 0.6 12 S.D. 1.3 .0001Fas detection 9.8 S.D. 2 10 S.D. 1 .4Caspase 8 detection 10 S.D. 1 10 S.D. 1 .52

All data are expressed as mean and SD. * Mann-Whitney u test.

227Primary Varicose Veins

Fas detection. There was no difference in the apo-ptosis promoter index of positive cell staining forFas detection in the two groups: Fas immunoreactivecells were found in similar numbers in varicose veinswith distal trunk, main branches and collateralsmixed together and in healthy veins. No differencewas found in the Fas detection index for the saphe-nous distal trunk, main branches and collaterals(data not shown).

Caspase 8 detection. Again, there was no differ-ence in the apoptosis promoter index of immunoreac-tive cell staining for caspase 8 detection in the twogroups: similar numbers of caspase 8 immunoreactivecells were detected in varicose veins with distal trunk,main branches and collaterals mixed together and inhealthy veins. The Caspase 8 detection index was sim-ilar in the saphenous distal trunk, main branches andcollaterals (data not shown).

PROTOCOL 2 (Table 3)Peroxidase in situ apoptosis detection. The apoptoticindex of immunoreactive cell staining differed

significantly in the two groups: the residual proximalpart of varicose veins contained fewer peroxidase insitu-immunoreactive cells than comparable healthyveins.

Bax detection. The apoptosis promoter index of im-munoreactive cell staining also differed significantlyin the two groups: the residual proximal part of vari-cose veins contained fewer Bax-immunoreactive cellsthan healthy.

Caspase 9 detection. Again, the apoptosis pro-moter index of positive cell staining differed signifi-cantly in the two groups: residual proximal part ofvaricose veins contained fewer caspase 9 positive cellsthan healthy.

Fas detection. There was no difference in the apo-ptosis promoter index of immunoreactive cell stainingfor Fas detection in the two groups: the residual prox-imal part of varicose veins and healthy veins con-tained similar numbers of Fas immunoreactive cells.

Caspase 8 detection. Again, there was no differ-ence in the apoptosis promoter index of immunoreac-tive cell staining for caspase 8 detection in the twogroups: the residual proximal part of varicose veinsand healthy veins contained similar numbers of cas-pase 8 immunoreactive.

Discussion

We detected marked histological changes in the archi-tectural organisation and distribution of collagen andelastin fibres in varicose veins compared to controlveins, in agreement with other published work. Wealso confirmed that immunoreactivity to apoptotic de-tection was less prominent in the medial layer of var-icose veins than in healthy veins. In varicose veins, wefound fewer specific mediators regulating the intrinsicapoptosis pathway. These cellular findings corrobo-rate our previous clinical studies suggesting deregu-lated programmed cell death, acting through theintrinsic mitochondrial apoptosis pathway promoter,as a possible cause of primary varicose veins in hu-mans.5,12 We found that the changes involved not

Eur J Vasc Endovasc Surg Vol 35, February 2008

228 E. Ducasse et al.

only the saphenous trunk, but the accessory veins andtributaries as well. Similar phenomena were presentin patients with primary varices and in those patientswho had previously undergone varicose vein surgery.

The diffusely disorganised architecture and distri-bution of collagen and elastin fibres corresponds tofibrotic degradation of the parietal wall previouslyobserved and a loss of mechanical properties in vari-cose veins.3,17 Further studies are required to preciselydetermine whether apoptotic dysregulation is relatedto the global number of parietal cells observed in thehistological changes.

Apoptosis plays a major role in tissue homeostasisand maintenance so the decreased cell turnover weobserved in all parts of the superficial venous systemcould be a major determinant of the parietal weaknessin primary varicose veins. Deregulated apoptosis isa fundamental mechanism in a variety of diseasesprincipally cancer, but also, autoimmune disorders,immunodeficiency, inflammation, ischemic heart dis-ease, stroke and neurodegenerative diseases.18e20 Al-though less is known about cell turnover in thevascular wall, recent evidence shows that apoptosisis inversely correlated with re-stenosis related to arte-rial intimal hyperplasia after angioplasty and venousgrafting.21,22 Our findings in varicose veins contrastwith the increased apoptosis and increased numbersof intrinsic pathway promotors including the apopto-tic promoter Bax reported by others in renal, coronaryand carotid artery specimens from patients with ad-vanced atherosclerotic lesions.6

Our findings confirm that the process of pro-grammed cell death is down-regulated in primaryvaricose veins as suggested by other authors 13e15 Aprevious study showed a decrease in the apoptotic in-dex in the adventitia of varicose veins compared withcontrol veins and no difference in the media layer (asin our study).13 These discrepancies presumably re-flect the different tissues examined. Cell culturesfrom varicose veins show that the main componentof the venous wall, smooth-muscle cells, principallylocalised in the medial layer, transform from normallycontractile to secretory cell types thus leading to thefibrotic nature of in the varicose vein wall.23 Cell cyclechanges in varicose veins mediate the smooth-musclecell dedifferentiation from a contractile to a secretoryphenotype.24 Fibrotic degradation seen in the varicosevenous wall is associated with dysregulation of apo-ptosis owing to decreased turnover of smooth-musclecells in the media layer. The intrinsic pathway can beinduced by mitochondrial release of cytochrome c, inresponse to various stimuli, including elevated levelsof the pore-forming pro-apoptotic Bcl-2 family, in-cluding Bax. Within the cytosol, cytochrome c binds

Eur J Vasc Endovasc Surg Vol 35, February 2008

and activates Apaf-1, thus binding and activatingpro-caspase-9.25e27 Active caspase-9 serves as the api-cal caspase of the intrinsic (mitochondrial) pathway,and directly cleaves and activates the apoptotic effec-tor protease, caspase-3.28,29

We studied a group of patients who had previouslyundergone varicose vein surgery which preserved thesaphenous trunk. Findings in this group were identi-cal to those in patients with primary varices. This sug-gests that the abnormal apoptosis is irreversible onceit affects the venous system. However, our previouslyoperated group also had recurrent varices whichmight have caused a return of the apoptotic malfunc-tion. Our findings indicate that diseased veins mayextend beyond the clinically detectable limits of ve-nous disease.

In conclusion, our results provide evidence of de-creased programmed cell death in the medial layerof human primary varicose veins. Assay of specificproteins indicates that the physiological decrease innormal apoptosis is modulated through the intrinsicmitochondrial-dependent pathway. This observa-tional study implicates apoptotic down-regulation inthe veins wall as a cause of varicose veins but doesnot exclude the possibility that other mechanismsare also responsible.

References

1 ASHER E, THERESA J, HINGORANI A, GUNDUZ Y, MAZZARIOL F,KALLAKURI S. Programmed cell death (Apoptosis) and its rolein the pathogenesis of lower extremity varicose veins. AnnChir Vasc 2000;14:24e30.

2 ASCHER E, JACOB T, HINGORANI A, TSEMEKHIN B, GUNDUNZ Y.Expression of molecular mediators of apoptosis and their rolein the pathogenesis of lower-extremity varicose veins. J VascSurg 2001;33:1080e1086.

3 BUJAN J, JIMENEZ-COSSIO JA, JURADO F, GIMENO MJ, PASCUAL G,GARCIA-HONDUVILLA B et al. Evaluation of the smooth musclecell component and apoptosis in the vein wall. Histol Histopathol2000;15:745e752.

4 URBANEK T, SKOP B, WIADERKIEWICZ R, WILCZOK T, ZIAJA K, LEBDA-WYBORNY T et al. Smooth muscle cell apoptosis in primary vari-cose veins. Eur J Vasc Endovasc Surg 2004;28:600e611.

5 DUCASSE E, GIANNAKAKIS K, CHEVALIER J, DASNOY D, PUPPINCK P,SPEZIALE F et al. Dysregulated apoptosis in primary varicoseveins. Eur J Vasc Endovasc Surg 2005;29:316e323.

6 SAXENA A, MCMEEKIN JD, THOMSON DJ. Expression of Bcl-x, Bcl-2,Bax and Bak in endarterectomy and atherectomy specimens.J Pathol 2002;196:335e342.

7 ALNEMRI ES, LIVINGSTON DJ, NICHOLSON DW, SALVESEN G,THORNBERRY NA, WONG WW et al. Human ICE/CED-3 proteasenomenclature. Cell 1996;87:171.

8 WALLACH D, VARFOLOMEEV EE, MALININ NL, GOLTSEV YV,KOVALENKO AV, BOLDIN MP. Tumor necrosis factor receptor andFas signaling mechanisms. Annu Rev Immunol 1999;17:331e367.

9 YUAN J. Transducing signals of life and death. Curr Opin Cell Biol1997;9:247e251.

10 REED JC. Cytochrome c: can’t live with it; can’t live without it.Cell 1997;91:559e562.

229Primary Varicose Veins

11 GREEN DR, REED JC. Mitochondria and apoptosis. Science 1998;281:1309e1312.

12 DUCASSE E, GIANNAKAKIS K, SPEZIALE F, FIORANI P, SASSOUST G,MIDY D et al. Superficial venous insufficiency: new data. AnnVasc Surg in press.

13 KORSMEYER SJ, SHUTTER JR, VEIS DJ, MERRY DE, OLTVAI ZN. Bcl-2/Bax: a rheostat that regulates an anti-oxidant pathway and celldeath. Semin Cancer Biol 1993;4:327e332.

14 OLTVAI ZN, MILLIMAN CL, KORSMEYER SJ. Bcl-2 heterodimerizes invivo with a conserved homolog, Bax, that accelerates pro-grammed cell death. Cell 1993;74:609e619.

15 DIAZ JL, OLTERSDORF T, HORNE W, MCCONNEL M, WILSON G,WEEKS S et al. A common binding site mediates heterodimeriza-tion and homodimerization of Bcl-2 family members. J Biol Chem1997;272:11350e11355.

16 VILLA P, KAUFMANN SH, EARNSHAW WC. Caspases and caspase in-hibitors. Trends Biochem Sci 1997;22:388e393.

17 KOCKX MM, KNAAPEN MW, BORTIER HE, CROMHEEKE KM,BOUTHERIN T, FALSON O et al. Vascular remodeling in varicoseveins. Angiology 1998;49:871e877.

18 ROY N, MAHADEVAN MS, MCLEAN M, SHUTLER G, YARAGHI Z,FARAHANI R et al. The gene for neuronal apoptosis inhibitoryprotein is partially deleted in individuals with spinal muscularatrophy. Cell 1995;80:167e178.

19 FISHER GH, ROSENBERG FJ, STRAUSS SE, DALE JK, MIDDELTON LA,LIN AY et al. Dominant interfering fas gene mutations impair ap-optosis in a human autoimmune lymphoproliferative syndrome.Cell 1995;81:935e946.

20 REED JC, MIYASHITA T, TAKAYAMA S, WANG HG, SATO T, KRAJEWSKI Set al. Bcl-2 family proteins: regulators of cell death involved inthe pathogenesis of cancer and resistance to therapy. J Cell Bio-chem 1996;60:23e32.

21 DURAND E, MALLAT Z, ADDAD F, VILDE F, DESNOS M, GUEROT C et al.Time courses of apoptosis and cell proliferation and their relation-ship to arterial remodeling and restenosis after angioplasty in anatherosclerotic rabbit model. J Am Coll Cardiol 2002;39:1680e1685.

22 MAYR U, MAYR M, LI C, WERNIG F, DIETRICH H, HU Y et al. Loss ofp53 accelerates neointimal lesions of vein bypass grafts in mice.Circ Res 2002;90:197e204.

23 JUROKOVA Z, MILENKOV C. Ultrastructural evidence for collagendegradation in the walls of varicose veins. Exp Mol Pathol1982;37:37e47.

24 PAPPAS PJ, GWERTZMAN GA, DEFOUW DO, PADBERG Jr FT, SILVA Jr MB,DURAN WN et al. Retinoblastoma protein: a molecular regulator ofchronic venous insufficiency. J Surg Res 1998;76:149e153.

25 ZOU H, HENZEL WJ, LIU X, LUTSCHG A, WANG X. Apaf-1, a humanprotein homologous to C. Elegans CED-4, participates in cyto-chrome c-dependent activation of caspase 3. Cell 1997;90:405e413.

26 ZOU H, LI Y, LIU X, WANG X. An Apaf-1 cytochrome c multimericcomplex is a functional apoptosome that activates procaspase 9.J Biol Chem 1999;274:11549e11556.

27 SALEH A, SRINIVASULA S, ACHARYA S, FISHEL R, ALNEMRI E. Cyto-chrome c and dATP-mediated oligomerization of Apaf-1is a pre-requisite for procaspase-9 activation. J Biol Chem 1999;274:17941e17945.

28 KUIDA K, HAYDAR TF, KUAN CY, GU Y, TAYA C, KARASUYAMA H et al.Reduced apoptosis and cytochrome c-mediated caspase activa-tion in mice lacking caspase-9. Cell 1998;94:325e337.

29 HAKEM R, HAKEM A, DUNCAN GS, HENDERSON JT, WOO M,SOENGAS MS et al. Differential requirement for caspase-9 in apo-ptotic pathways in vivo. Cell 1998;94:339e352.

Accepted 30 August 2007

Available online 23 October 2007

Eur J Vasc Endovasc Surg Vol 35, February 2008