1018

In cardiac and peripheral arterial surgery, there isa need for small caliber grafts for arterial reconstruc-tion. Although in coronary artery bypass graftingthe internal mammary artery is considered the graftof choice, the widespread use of the saphenous veincontinues because of the limitations in the use of the

internal mammary artery, because of the lack of suf-ficient arterial conduit for grafting in patients withmultiple-vessel disease.1,2 However, saphenous veinsunder arterial pressure experience an accelerated ath-erosclerotic process.3 For this reason, different typesof grafts, either natural or synthetic, are being inves-tigated as substitutes for saphenous vein and as acomplement of the internal mammary artery.

The preservation of functional endothelium isessential to maintain the vascular tone and antithrom-botic properties of a vessel. Therefore an adequatepreservation of the arterial wall function may play arole in medium and long-term patency of small caliberarterial grafts.

Cryopreservation is a well-developed techniqueto preserve both valvular grafts4 and large-caliberarterial allografts.5 The clinical need for arterialgrafts in coronary surgery has increased the interestin this technique of banking small-caliber arterialgrafts. Cryopreservation of small arterial vessels has

Changes in the cooling rate and mediumimprove the vascular function incryopreserved porcine femoral arteriesMontserrat Rigol, PhD, Magda Heras, MD, Anna Martínez, MD, María JesúsZurbano, BS, Elba Agustí, BS, Eulàlia Roig, MD, José Luís Pomar, MD, andGinés Sanz, MD, Barcelona, Spain

Purpose: The purpose of this study was to design an adequate technique with which tocryopreserve pig femoral arteries and to assess the influence of storage times in vascularfunction.Methods: Fifty-two femoral arteries were distributed in seven groups. In group A (con-trol), 10 arteries were studied after harvest; in groups B1 and B2, 19 arteries were sus-pended in RPMI 1640 plus fetal calf serum plus dimethylsulfoxide and were cryopre-served at 1˚C per minute or 0.3˚C per minute, respectively. In groups C1 to C4, 23arteries were suspended in modified Krebs-Henseleit plus dimethylsulfoxide plussucrose, cryopreserved at 0.7˚C per minute, and kept frozen for 1, 15, 60, or 180 days,respectively. After being thawed, arteries were examined for contraction and endothelial-dependent vasodilation (organ bath studies), antithrombotic properties of the endothe-lial layer(perfusion studies), and vessel structure (electron microscopy). Results: Endothelial cells were present in both cryopreserved and control arteries. Thecontrol vessels showed a mean contraction to norepinephrine (10–7 mol/L) of 13010 ±3181 mg. Arteries in groups B1 and B2 did not respond to norepinephrine. Contractionin groups C1 to C4 was as follows: C1, 5354 ± 1222 mg; C2, 5187 ± 2672 mg; C3, 6867± 2292 mg; C4, 7000 ± 2858 mg, which represent 50% of the control values (P < .001).Vasodilation was similar in control (99% ± 3%) and cryopreserved arteries (C1, 90% ±13%; C2, 93% ± 12%; C3, 89% ± 15%; C4, 88% ± 22%). Storage time did not influencevascular function. Platelet interaction was almost absent and similar in all groups.Conclusion: A modified cryopreservation technique preserves endothelial function inde-pendently of the storage time up to 6 months. (J Vasc Surg 2000;31:1018-25.)

From the Cardiovascular Institute, Hospital Clinic, and Institutd’Investigacions Biomèdiques August Pi Sunyer (IDIBAPS),University of Barcelona.

Competition of interest: nil.Supported in part by a grant from Dirección General de

Investigación Científica y Técnica-DGICYT 1992.Presented in part at the Seventy-first Scientific Sessions of the

American Heart Association, Dallas, Texas, November 1998.Reprint requests: Montserrat Rigol, PhD, Institut de Malalties

Cardiovasculars, Hospital Clínic de Barcelona, Villarroel, 170,08036 Barcelona, Spain.

Copyright © 2000 by The Society for Vascular Surgery andInternational Society for Cardiovascular Surgery, NorthAmerican Chapter.

0741-5214/2000/$12.00 + 0 24/1/103793doi:10.1067/mva.2000.103793

JOURNAL OF VASCULAR SURGERYVolume 31, Number 5 Rigol et al 1019

already been shown to be a useful technique for thestorage of animal and human arteries. Nevertheless,most cryopreservation studies have used othermethods and short arterial segments, which are onlysuitable for pharmacologic testing but not for graft-ing purposes.6 The relaxant and contractile respons-es of cryopreserved arteries in these studies vary con-siderably, mainly because of the differences in thesusceptibility to cryoinjury of the endothelial andsmooth muscle cells from different arteries and tothe cryopreservation procedures used.7-12

The aim of the present study was (1) to design acomputerized program to cryopreserve pig femoralarteries, (2) to assess whether different storage timesmay affect the endothelial function of these cryopre-served arteries, and (3) to study the antithromboticand ultrastructural endothelium properties of cryo-preserved banked arteries.

METHODSThirty-six common pigs, a mixture of Landrace

and Largewhite breeds, were used in this study. Theywere purchased from a local farmer and were fed aregular diet. They were 4 months old, a mixture ofmale and female with a mean body weight of 29 ± 5kg. All procedures were performed according to theGuide for the Care and Use of Laboratory Animalspublished by the US National Institutes of Health(NIH Publication No 85-23, revised 1996) and theguidelines of the local government.13

The animals were sedated with an intramuscularinjection of azaperone (2 mg/kg) and were anes-thetized with thiopental 30 mg/kg; thereafter, anes-thesia was maintained with intravenous thiopental(0,2 mg/kg/min). They were intubated and con-nected to a volumetric ventilator with a mixture ofoxygen and air room. The carotid artery was dis-sected and catheterized with 8F sheaths (Input 8F;BARD, CR Bard Ireland Ltd, Galway, Ireland) towithdraw blood for perfusion studies.

Blood samples were mixed with citrate-phos-phate-dextrose as anticoagulant, with a final concen-tration of 13 mmol/L. After both femoral arterieswere harvested, the animals were killed with a largedose of intravenous thiopental.

Tissue preparation and storage methods. Thefemoral arteries were dissected, carefully removed,and divided into seven groups. Group A (n = 10arteries) consisted of fresh vessels, which were usedimmediately after removal for organ bath, perfusionstudies, and electronic microscopy and served as acontrol group.

In groups B1 (n = 9 arteries) and B2 (n = 10

arteries), the femoral arteries were cut into smallersegments of about 1.5 cm and placed in 2 mL liquidnitrogen storage ampoules (Cryotubes Nunc; LabClinics, Barcelona, Spain) filled with RPMI 1640 tis-sue culture media (BioWhittaker, Walkersville, Md)containing 10% fetal calf serum (FCS) and 10% ofdimethylsulfoxide (DMSO). The ampoules werethen introduced in a programmable freezer (CryosonBV-10; Carburos Metálicos SA, Barcelona, Spain)and slowly frozen at a cooling rate of 1˚C per minute(group B1) or 0.3˚C per minute (group B2). Afterthe cryopreservation process, the ampoules weretransferred to liquid nitrogen (–196˚C) and storedfor 90 days.

Because contractile response of arteries that werecryopreserved with this technique was abolished, thefollowing modifications were introduced in the cryo-preservation technique for groups C1 to C4. Thearterial segments (approximately 4 cm) were placed inplastic bags (Cryocyte Freezing container, Pl 269plastic; Baxter Healthcare Corporation, Irvine, Calif)filled with 100 mL of modified Krebs-Henseleit(NaCl 118 mmol/L, glucose 11 mmol/L, NaHCO325 mmol/L, KCl 4.7 mmol/L, MgSO4 1.2mmol/L, KH2PO4 1.2 mmol/L, CaCl2 1.2mmol/L, EDTA 0.03 mmol/L) solution containing12% DMSO and 0.1 mol/L sucrose.14 The bags werethen introduced in the same programmable freezerand slowly frozen at a cooling rate of 0.7˚C perminute. When –75˚C was achieved, the bags weretransferred to liquid nitrogen (–196˚C) and stored for1 day (group C1; n = 6 arteries), 15 days (group C2;n = 7 arteries), 60 days (group C3; n = 7 arteries), or180 days (group C4; n = 3 arteries). Table I shows thedifferent techniques of programmed cryopreservationand the study length for each study group.

At removal from the liquid nitrogen, the femoralarteries were thawed by placing the ampoules andbags in a 37˚C water bath for approximately 3 min-utes. Thereafter, the vessels were rinsed in a dish con-taining Krebs-Henseleit solution. The femoral arter-ies were then examined immediately for their con-traction and endothelial-dependent vasodilation inthe organ bath. Perfusion and electronic microscopystudies were also performed only in groups A and Cto assess the antithrombotic properties and the struc-ture of the endothelial layer, respectively.

Organ bath. One section of each artery wascleaned of adherent connective tissue and cut intorings of approximately 5 mm in length. Special carewas taken to avoid contact with the luminal surfaceof the blood vessel to preserve the endothelial layer.The rings were mounted horizontally between two

JOURNAL OF VASCULAR SURGERY1020 Rigol et al May 2000

stainless steel stirrups in individual organ bath cham-bers filled with 10 mL of Krebs-Henseleit solution.The saline solution was constantly bubbled with amixture of 95% oxygen-5% carbon dioxide and main-tained at 37˚C with an outer jacket and circulatingheat pump. Changes in tension were measured withan isometric transducer (TRI 201; Leica SA,L’Hospitalet, Spain) and recorded in a multichannelpolygraph Uni-Graph 2000-506 (Leica SA). Therings were allowed to equilibrate for 120 minuteswith several adjustments until a baseline force of 12g was obtained. In previous experiments, we foundthat a resting tension of 12 g is optimal for theexpression of 10–7 mol/L norepinephrine-inducedcontraction of femoral rings obtained from controlpigs. All the experiments were performed in the pres-ence of 10–5 mol/L indomethacin to prevent therelease of endogenous prostanoids.

After the equilibration period, the rings werecontracted with 10–7 mol/L norepinephrine. Thisdose induced approximately 40% (6320 ± 1225 mg)of maximal contraction (16240 ± 2456 mg) in thecontraction-response curves of previous experimentsfrom this laboratory (Fig 1; unpublished results).When the contractile response had reached aplateau, the preparations were exposed to cumula-tive concentrations (10–9 to 10–4 mol/L) of acetyl-choline. The relaxant activity of the agonist wasexpressed as a percentage of reduction of the tensiondeveloped by norepinephrine.

Perfusion studies. The endothelial antithrom-botic properties were evaluated under blood flowconditions. Perfusions were performed at 37˚C inperfusion chambers as developed by Baumgartner15

with an effective annular width of 2.2 mm and a rodlength of 7.2 cm. Blood flow was obtained bypumping blood through a peristaltic pump. Pigblood was perfused at 140 mL/min (wall shear rate800/sec). One section of each artery was everted,mounted on a plastic rod, and rinsed with phosphatebuffered saline solution (PBS; BioMérieux, Marcy,l’étoile, France) for 10 minutes. After 10 minutes of

perfusion with the blood samples, the segmentswere rinsed with the same phosphate buffer andfixed with 3% glutaraldehyde.

After 1 hour, the segments were separated fromthe rods, washed with PBS, dehydrated through agraded series of ethanol, embedded in JB-4 embed-ding material (Polysciences, Warrington, Pa), thincross-sectioned for light microscopy (3 µm), andstained with toluidine blue. Platelet interactions withthe arterial wall were morphometrically evaluatedwith the use of a computerized image analysis sys-tem.16 The morphometric parameters were deter-mined according to the criteria previously establishedby Baumgartner.17 Briefly, platelets or groups ofplatelets were classified as (1) contact (platelets thatwere attached to but not spread on the endotheli-um); (2) adhesion (platelets that had spread on theendothelial surface, including small aggregates of lessthan 5 µm in height); and (3) thrombi (plateletaggregates of more than 5 µm in height). All thesebasic parameters were expressed as a percentage ofthe total surface examined. Another parameter relat-ed to those previously defined was the total coveredsurface, which was determined by adding contact +adhesion + thrombi.

Electron microscopy. One section of eachartery was placed in 2.5% glutaraldehyde for scan-ning electron microscopy. The section was postfixedin 1% osmium tetraoxide in PBS, dehydrated ingraded ethanols, critical point-dried, and stored in adessicator. The sections were then gold coated(Polaron E500; Polaron Watford Hertfordshire,England) and viewed with a Leica LC-360 scanningelectron microscope (Leica, Cambridge, England)to assess the integrity of the endothelial layer.

Drugs. All drugs (azaperone 2 mg/kg, intra-muscularly [Stresnil; Esteve, Barcelona, Spain]; sodi-um thiopental is 30 mg/kg, intravenous bolus, toinduce anesthesia and intravenous perfusion 0.25mg/kg/min [Pentotal; Laboratorio Abbot, Madrid,Spain]; indomethacine 10–5 mol/L [Sigma Quimica,Alcobendas, Madrid, Spain]; norepinephrine 10–7

Table I. Different techniques and storage times of programmed cryopreservation for each study group

Group B Group C

Group B1 Group B2 Group C1 Group C2 Group C3 Group C4

Arteries (n) 9 10 6 7 7 3Cryomedium RPMI 1640, 10% DMSO, 10% FCS Modified Krebs-Henseleit, 12% DMSO, 0.1 mol/L sucroseCooling rate (˚C/min) 1 0.3 0.7 0.7 0.7 0.7Storage at 90 90 1 15 60 180

–196˚C (day)

JOURNAL OF VASCULAR SURGERYVolume 31, Number 5 Rigol et al 1021

mol/L [Sigma Quimica]; acetylcholine 10–9 to 10–4

mol/L [Sigma Quimica]; and sucrose 0.1 mol/L[Sigma Quimica]) were freshly prepared daily in dis-tilled water (except for norepinephrine, which wasdissolved in HCl 1%). All drugs were properly dilutedand added to the organ bath in volumes of 100 µL.Concentrations are expressed as final molar concen-tration in the organ chambers.

Statistical analysis. All values are expressed asmean ± SD of the mean. A one-way analysis of vari-ance was used to compare means between differentgroups. Significant levels were adjusted for multiplecomparisons by the Scheffe method. A probabilityvalue of less than .05 indicated a significant differ-ence. The statistical calculations were performed withthe SPSS/PC package (SPSS, Inc, Chicago, Ill).

RESULTSContractile responses. Table II shows the con-

tractile responses of femoral arteries induced by 10–7

mol/L norepinephrine for groups A and C. Thevasoconstrictor response to the norepinephrine in thecontrol arteries (group A) was 13010 ± 3181 mg; ingroups B1 and B2, it was zero. However, in groupsC1 to C4, the contractile responses to the norepi-nephrine were approximately 50% with respect to thefresh arteries (group A; P < .01). There were no sig-nificant differences between storage times.

Relaxant responses. Fig 2 illustrates the effectinduced by an increase in the concentrations ofacetylcholine in fresh and cryopreserved arteries.

Table II also shows the maximal endothelium-dependent relaxation, expressed as a percentage ofthe induced contraction. Endothelium-dependentrelaxation could not be investigated in groups B1and B2 because they had lost contractility. No sig-nificant changes in the response to acetylcholinewere observed between fresh (group A) and cryo-preserved arteries (groups C1 to C4); it was inde-pendent of the storage time.

Perfusion studies. Platelet interaction with theendothelium was almost absent in control arteries(group A) and in all cryopreserved arteries studied(groups C1 to C4). Moreover, there were no differ-ences between different storage times. Fig 3 showsperfusion studies of fresh and cryopreserved arteriesof groups C3 and C4.

Electron microscopic examination. All cryo-preserved arteries (groups C1 to C4) exhibited con-fluent endothelial surfaces. No morphologic differ-ences could be made between fresh (group A) andcryopreserved arteries. There were no differencesbetween different storage times. Representative scan-ning electron microscopic images of arterial endothe-lium from groups A and C2 are shown in Fig 4.

DISCUSSIONThe results of this study demonstrate that

endothelial morphologic features and function aremaintained in porcine femoral arteries, when cryo-preserved with a technique consisting of a coolingrate of 0.7˚C per minute and a modified Krebs-

Fig 2. Relaxations in response to increasing concentra-tions of acetylcholine in fresh and cryopreserved arteriesstored for 1, 15, 60, and 180 days. Values are expressed aspercentage of the induced contraction by 10–7 mol/Lnorepinephrine.

Fig 1. Dose-response curve to norepinephrine in ringspreparation of femoral arteries (n = 3 arteries; 12 rings).Values are expressed in milligrams.

JOURNAL OF VASCULAR SURGERY1022 Rigol et al May 2000

Henseleit solution that contains 12% of DMSO and0.1 mol/L sucrose as cryomedium. In contrast, thearteries cryopreserved with the standard techniquefor cardiac valve cryopreservation lost both smoothmuscle and endothelial cells function. We have alsoshown that this preservation is not negatively affect-ed by the storage in liquid nitrogen up to 6 months.

During the cryopreservation process, the cryo-medium and the cooling rate are determinants of thecell damage by the formation of intracellular andextracellular ice crystals and by the osmotic disequi-librium because of concentrated electrolytes duringfreezing.9 Therefore an adequate cryopreservationprocedure is essential for the postthawed recovery ofvascular function.

In this study, we used modified Krebs-Henseleitsolution containing 12% of DMSO and 0.1 mol/Lsucrose, without amino acids and vitamins, and acooling rate of 0.7˚C per minute as an effectivemethod to preserve pig femoral arteries. Theincrease in the viability of the arteries could beascribed to both, the cryopreservation solution andthe freezing rate. However, we think that, becauseboth solutions used in this study have been used suc-cessfully by other authors, the major changes in thefreezing rate may be the most important factor inthe differences found in our study. Therefore arter-ies cryopreserved in such a way showed an approxi-mate 50% reduction in the contractility whenexposed to norepinephrine, as compared with fresh

Table II. Maximal contractile and relaxant responses to norepinephrine and acetylcholine, respectively, offresh (group A) and cryopreserved arteries (groups C1 to C4)

Group A

Group C

Group C1 Group C2 Group C3 Group C4

Arteries (n) 10 6 7 7 3Contraction to

norepinephrine (mg) 13010 ± 3181 5354 ± 1222* 5187 ± 2672* 6867 ± 2292* 7000 ± 2858*

Vasodilation (%) toacetylcholine 99 ± 3 90 ± 13 92 ± 12 89 ± 15 88 ± 22

All values are expressed as mean ± SD. *P < .001, vs group A, by one-way analysis of variance.

Fig 3. Light micrographs obtained from cross-sections of everted segments of fresh (A) andcryopreserved arteries stored for 60 (B) and 180 days (C). The arrows show endothelial cells.In all cases, the interaction of platelets with the endothelium is almost absent. (Toluidine bluestain; original magnification, ×850.)

A

B

C

JOURNAL OF VASCULAR SURGERYVolume 31, Number 5 Rigol et al 1023

studied arteries. These results compare favorablywith those observed in other vascular segments, suchas human coronary arteries, in which contractionresponses have been reported to be only 20% to 30%that of fresh ones.10

Our results are similar to the findings of Müller-Schweinitzer et al14,18 in canine femoral, humanmesenteric, and left circumflex coronary arteries.These authors observed contractile responses similarto those of the present investigation. However, inthe present investigation, the maximal endothelial-dependent relaxation was similar between cryopre-served and fresh arteries (approximately 90% of theinduced contraction) and higher than the 30% to50% reported by Müller-Schweinitzer et al. The twomain reasons for these differences can be ascribed toa variation in susceptibility to the cryopreservationprocedure, which may vary from tissue to tis-sue,6,10,18,19 and to methodologic differences in thecooling process. We used a computerized programthat was designed and is used in our institution toensure a constant slow rate of 0.7˚C per minute, incontrast to the manual cooling rate. Moreover, thegreater detrimental effect of the cryopreservation oncontraction as compared with vasodilation may berelated to the density and packing of the smoothmuscle cells in large arteries, which supports the ideathat different cell types have a different susceptibili-ty to the freezing procedure.9,20

Pompilio et al21 demonstrated a full preservationof the relaxant activity and vasoconstrictor responsein human mammary arteries after cryopreservationusing RPMI, FCS, and DMSO as cryomedium anda computerized cooling rate of 1˚C per minute.However, these results have not been reproduced byany other author. In fact, with the methods ofPompilio et al, Nataf et al12 and Vischjager et al22

found (as we did in our groups that were preservedby the standard technique) a severe damage of thevascular segments with extremely poor contractionin human mammary arteries and canine carotidarteries, respectively.

It has been previously demonstrated that storageat –196˚C yields a better preservation of vascularfunction than storage at –70˚C.23 However, it is notwell known whether continued storage of small cal-iber arteries at –196˚C could affect arterial function.

The present investigation has addressed thisquestion and has demonstrated that storage for anextended period of time in liquid nitrogen at–196˚C does not negatively affect the vascular func-tion. Our results are different from those of Ku et al8who reported that cryopreserved canine coronary

arteries stored for 1 day at –75˚C showed contractileand relaxant responses not significantly differentfrom the fresh control, although longer cryostorageat –75˚C resulted in a significant decrease in theseresponses. This difference can be probably related tothe different banking temperature. In that sense,Brockbank et al24 found that the contractile respons-es generated by long-term storing (more than 2years at less than –150˚C) of cryopreserved caninesaphenous veins were the same as in the short-termstorage of veins (1-28 days at less than –150˚C).These authors demonstrated that vein viability ismaintained in cryopreserved banked veins, as wehave done with pig femoral arteries.

Our study also investigated the antithromboticproperties and endothelial morphologic features of thecryopreserved arteries. These two aspects have neverbeen looked at previously in studies with the presentcryopreservation methods. The fact that a cryopre-served artery retains its antithrombotic properties (noplatelet adhesion) is a very relevant finding to predict

Fig 4. Scanning electron micrographs of fresh (A) andcryopreserved arteries stored for 15 days (B). In bothcases, the endothelium is confluent, and the cell morpho-logic condition is normal. (Original magnification, ×200.)

A

B

JOURNAL OF VASCULAR SURGERY1024 Rigol et al May 2000

the future patency when used as a grafting segment. Inthis sense, the preservation of the endothelial layermorphologic features further confirms that thismethod fully respects the vascular structure.

Clinical relevance. In spite of antiplatelet treat-ment, saphenous venous grafts still have a high attri-tion rate, as compared with arterial conduits.Furthermore, the number of patients who require asecond intervention is increasing as the populationages. Therefore cryopreserved arteries may represent auseful alternative to saphenous vein. We can speculatethat preservation of vascular function, including theendothelial antithrombotic properties, may help topreserve the graft patency. In this sense, the cryop-reservation technique developed in the present investi-gation may offer better vascular grafts for cardiovascu-lar surgery and peripheral vascular surgery. However,the performance of cryopreserved grafts whenimplanted in the circulation and the effect of rejectionshould be addressed in further studies. At thismoment, our group has started to implant cryopre-served femoral arteries in a porcine model, and histo-logic analysis to observe tissue viability and immuno-logic rejection is underway. Other authors25-27 havealready published studies that deal with cryopreservedallografts used for arterial bypass grafting. In thatsense, Pacholewicz et al25 cryopreserved internalmammary canine arteries and obtained overall patencyrates of 75% at 90 days when they were used as graft-ing segments, which suggests that these cryopreservedarteries may have potential as a substitute graft.

In conclusion, the results from the present studyindicate that computerized cryopreservation, with acooling rate of 0.7˚C per minute and modifiedKrebs-Henseleit solution that contains 12% DMSOand 0.1 mol/L sucrose as cryomedium, preservesthe morphologic condition and function of porcinefemoral arteries. These findings are not negativelyaffected by the storage time in liquid nitrogen,which supports the feasibility of small caliber arteri-al banking. Moreover, the preservation of theantithrombotic and ultrastructural endothelial prop-erties of cryopreserved banked arteries may ensuretheir patency when they are used as arterial grafts.

REFERENCES

1. Lytle BW, Loop FD, Cosgrove DM, Ratliff NB, Easly K,Taylor PC. Long term serial studies of internal mammaryartery and saphenous vein coronary bypass grafts. J ThoracCardiovasc Surg 1985;89:248-58.

2. Okies JE, Page US, Bigelow JC. The left internal mammaryartery: the graft of choice. Circulation 1984;70(suppl):1213-21.

3. Chesebro JH, Clements IP, Fuster V, Elveback LR, SmithHC, Bardsley WT, et al. A platelet-inhibitor drug trial in

coronary-artery bypass operations: benefit of perioperativedipyridamole and aspirin therapy on early postoperative vein-graft patency. N Engl J Med 1982;307:73-8.

4. Loose R, Markant S, Sievers H-H, Bernhard A. Fate ofendothelial cells during transport, cryopreservation, andthawing of heart valve allografts. Transplant Proc1993;25:3247-50.

5. Mestres CA, Mulet J, Pomar JL. Large-caliber cryopreservedarterial allografts in vascular reconstructive operations: earlyexperience. Ann Thorac Surg 1995;60:S105-7.

6. Müller-Schweinitzer E. Applications of cryopreserved bloodvessels in pharmacological research. Cryobiology 1994;31:57-62.

7. Schoeffter P, Müller-Schweinitzer E. The preservation of fun-tional activity fo smooth muscle and endothelium in pigcoronary arteries after storage at –190˚C. J Pharm Pharmacol1990;42:646-51.

8. Ku DD, Willis WL, Caulfield JB. Retention of endothelium-dependent vasodilatory responses in canine coronary arteriesfollowing cryopreservation. Cryobiology 1990;27:511-20.

9. Müller-Schweinitzer E, Ellis P. Sucrose promotes the func-tional activity of blood vessels after cryopreservation inDMSO-containing fetal calf serum. Naunyn-Schmiedeberg’sArch Pharmacol 1992;345:594-7.

10. Ku DD, Winn MJ, Grigsby T, Caulfield JB. Human coronaryvascular smooth muscle and endothelium-dependent respons-es after storage at –75˚C. Cryobiology 1992;29:199-209.

11. Song YC, Hunt CJ, Pegg DE. Cryopreservation of the com-mon carotid artery of the rabbit. Cryobiology 1994;31:317-29.

12. Nataf P, Hadjisky P, Lechat PH, Mougenot N, PeuchmaurdM, Gouezo R, et al. Effect of cold anoxia and cryopreserva-tion and contractile functions of human mammary artery.Cryobiology 1995;32:327-33.

13. Llei 5/1995 del 21 de juny de protecció dels animals per aexperimentació i per a altres finalitats científiques. DOGC1995;2073.

14. Müller-Schweinitzer E, Mihatsch MJ, Schilling M, HaefeliWE. Functional recovery of human mesenteric and coronaryarteries after cryopreservation at –196˚C in a serum-freemedium. J Vasc Surg 1997;25:743-50.

15. Baumgartner HR. The role of blood flow in platelet adhe-sion, fibrin deposition and formation of mural thrombi.Microvasc Res 1973;5:167-79.

16. Escolar G, Bastida E, Castillo R, Ordinas A. Development ofcomputer program to analyze the parameters of platelet-ves-sels wall interaction. Haemostasis 1986;16:8-14.

17. Baumgartner HR. Platelet interaction with collagen fibrils inflowing blood: reaction of human platelet with α-chy-motrypsin digested subendothelium. Thromb Haemost1977;37:1-15.

18. Müller-Schweinitzer E. Arterial smooth muscle function afterprolonged exposure to a medium containing dimethyl sul-foxide (Me2SO) and storage at –196˚C. Cryobiology 1994;31:330-5.

19. Ellis P, Müller-Schweinitzer E. Maintenance of functionalactivity of human pulmonary arteries after cryopreservation.Br J Pharmacol 1991;103:1377-80.

20. Ku DD, Liu Q, Norton P, Caulfield JB. Cryopreservation ofcoronary endothelium and endothelial-mediated responses.Cryobiology 1994;31:82-9.

21. Pompilio G, Polvani GL, Antona C, Rossoni G, Guarino A,Porqueddu M, et al. Retention of endothelium-dependentproperties in human mammary arteries after cryopreserva-tion. Ann Thorac Surg 1996;61:667-73.

JOURNAL OF VASCULAR SURGERYVolume 31, Number 5 Rigol et al 1025

22. Vischjager M, Van Gulik TM, Van Marle J, Pfaffendorf M,Jacobs MJ. Function of cryopreserved arterial allograftsunder immunosuppressive protection with cyclosporine A. JVasc Surg 1996;24:876-82.

23. Müller-Schweinitzer E, Tapparelli C. Pharmacological studieson frozen stored canine saphenous veins and basilar arteries.Naunyn-Schmiedeberg’s Arch Pharmacol 1986;332:74-8.

24. Brockbank KGM, Carpenter JF, Schwartz LB, Nardiello C,Hagen PO. Smooth muscle function in short- and long-termstored cryopreserved veins. Vasc Surg 1992;26:116-22.

25. Pacholewicz JK, Adkins MS, Boris WJ, Gu J, Xenachis C,Klabunde RE, et al. Evaluation of cryopreserved internal thoracic artery as an alternative coronary graft: evidence for

preserved functional, metabolic and structural integrity. EurJ Cardiothorac Surg 1996;10:20-5.

26. Laffaldano RA, Lewis BE, Johnson SA, Piffare R, McKiernanTL. Patency of cryopreserved saphenous vein grafts as conduitsfor coronary artery bypass surgery. Chest 1995;108:725-9.

27. Korb H, Borowski A, Esbati A, Schlesinger U, Dhein S.Cryopreservation of veins as an alternative to autografts incoronary by-pass grafting. J Cardiovasc Surg (Torino)1994;35:223-6.

Submitted Jul 9, 1999; accepted Sep 21, 1999.