Eur J Cardiothorac Surg 1998;14:503-507
© 1998 Elsevier Science NL
Endothelial prostacyclin (PGI-2) production of human and porcine valve allografts related to ischemic history1
Peter Biberthaler,
Nikolaus Mendler,
Ursula Ettner,
Hans Meisner
German Heart Center Munich, Lazarettstrasse 36, 80636 Munich, Germany
Received 5 April 1998;
received in revised form 11 August 1998;
accepted 11 August 1998.
Corresponding author. Deutsches Herzzentrum München, Lazarettstrasse 36, 80636 München, Germany. Tel.: +49 89 12183500; fax: +49 89 1218 3513; e-mail: mendler@dhm.mhn.de
 |
Abstract
|
|---|
Background: The significance of cellular viability in human valve allografts for functional clinical longevity continues to be debated. Meaningful tests for this biological entity are therefore in demand to quantify the relative merits of graft origin and procurement techniques. The valve leaflet endothelium is recognized as a particularly sensitive target to noxes and its continued ability to produce prostacyclin (PGI-2) after explantation has been suggested as indicating viability. Objective: Graft ischemic history and species differences were therefore studied in human and porcine valve leaflets by the measurement of endothelial prostacyclin production, post-explantational, basal and after stimulation with bradykinin. Methods: Four groups of aortic valve donors were established. Fresh human heart-beating donors (h-HBD), cadaveric human donors (h-NHBD) processed within 24 h, fresh porcine donors (p-HBD) and cadaveric porcine donors (p-NHBD) also processed within 24 h. Leaflets were separately incubated at 37°C for successive periods of 30 min up to 5 h in Earle's Medium 199. After 240 min PGI-2 production was stimulated by 10 µM bradykinin. Postincubational release was stopped with indomethacin 10 µg/ml. Prostacyclin production was measured as 6-kPGF1a using an ELISA. Results: Initial PGI-2 production is significantly higher in porcine than in human grafts and in both species enhanced by previous warm ischemia. While baseline species differences disappear during progressive incubation, differences resulting from graft history are maintained. After PGI-2 stimulation species differences dominate again while ischemic history has no effect. Conclusion: Ischemia and surgical manipulation are stimulators of endothelial PGI-2 production in both human and porcine allografts and, therefore, a correlation of this metabolic activity with cellular integrity may be misleading. Valid data are obtained only if the natural time-course and reaction to stimulation of PGI-2 production are duely recognized and species differences in the response to mechanical and ischemic stress are considered.
Key Words: Heart valve allograft • Viability • Prostacyclin
|
Introduction
|
|---|
Since their introduction 1962 human heart valve allografts have been successfully used as cardiac valve substitutes providing superior hemodynamic performance, freedom from anticoagulation, minimal thromboembolic complications and excellent long term durability
[1]
[2]
[3].
More than three decades of research into the long-term clinical performance suggested the correlation with post-harvesting procurement of the graft regarding different sterilization and preservation techniques
[4]
[5]. Although the final proof of the contribution of living cellular components in the graft to its longevity is still missing, the experience with preservation and sterilization techniques supports the existence of a positive influence. A significant improvement of clinical function was shown
[5] with cell-protecting procedures as compared with the toxic procurements of the early seventies. The quality of preservation and sterilization techniques is therefore judged by the post-procurement viability of the graft today
[6]. Various viability tests have been established setting fibrocytes as a main target of viability signs
[7]. The H3 incorporation test into fibrocyte-produced proline has become the gold standard
[6], but unfortunately the setting of the test is destructive for the valves and thus allows no study of the time course of biological activity. The endothelial cell (EC) layer covering the valve leaflets became a focus of interest when its metabolic activity with production of prostacyclin (PGI-2) was recognized as a particularly sensitive target to noxes
[9]
[10]. The continued ability to produce PGI-2 was established as a quantitative marker of damage to the cellular components of the graft. Although the ultimate fate of the allograft EC is uncertain
[18] it is known as the most vulnerable component of the valve and may serve as an indicator for the overall cell survival after exposure to various steps of the procurement process
[9]
[10]
[11]
[25]. The test allows the study of time-dependent phenomena while not being destructive to the examined material. This metabolic test system was established in a porcine model. The purpose of our study was to determine the validity of the data comparing human with porcine models by measuring the differences in endothelial cell metabolic activity related to ischemic history.
The obtained data will facilitate the comparison of previous published information from endothelial metabolic testing systems.
|
Material and methods
|
|---|
Experimental groups
Four donor groups were studied: Human heart-beating donors (h-HBD) from organ recipients of our heart transplantation program (n=7). Five males and two females between 44 and 64 years all being transplanted because of dilatative cardiomyopathy. The explanted hearts were processed within 30 min. All h-HBD donors suffered from chronic disease. Therefore, stimulation of prostacyclin production by acute stretching of the endothelial cell layer was considered negligible. Human non-heart-beating donors (h-NHBD) explanted under the standard protocol of our allograft valve bank. The exclusion criteria have been described elsewhere
[6]. This group contained 12 valves, 10 male and two female, between 6.5 months and 58 years donor age. The warm ischemic time was always less than 24 h with a mean of 15.75 h. The bodies of the h-NHBD group were collected at common human environmental temperatures. Spontaneous cooling of the cadavers was assumed to result in an ischemic temperature of 20°C at harvesting. The porcine heart-beating donor (p-HBD) group (n=11) was harvested under operating room conditions from anaesthetized animals between 27 and 54 kg and processed within 30 min. In the non-heart-beating donor (p-NHBD) group the hearts of the animals were stored in approximately 500 ml of their own blood and allowed to cool to room temperature to simulate warm ischemic time of cadaveric human harvesting conditions. Storage time was always 24 h in this group modelling closely the warm ischemic time of the h-NHBD group accepted up to a maximum of 24 h. Eleven hearts of animals between 31.5 and 52 kg were examined in this group. All animals received humane care in compliance with the European Convention on Animal Care and the study was approved by the institutional ethics committee.
Valve procurement
Valve procurement was similar in all groups: The hearts were excised in toto from the donors, stored in 4°C nutrition solution, transferred to our laboratory and processed within 30 min after explantation in the beating-heart groups, and within 24 h in the non-beating hearts. The valves were dissected with a no-touch technique in a 0.9% NaCl containing vessel to prevent drying damage of the endothelial layer. All three leaflets were incubated for successive periods of 30 min up to 5 h in 2.5 ml Earle's Medium 199 (20 mM HEPES). Following each incubation step, further PGI-2 release by eventually abraded endothelial cells in the incubation solution was stopped by addition
[8] of indomethacin (10 µg/ml) and samples were frozen at -20°C. Previous published studies
[10] revealed a characteristic PGI2 production time course with a peak at 150 min after starting incubation, levelling to a steady state at 240 min. The peak was interpreted as a sign of activation due to surgical valve dissection, the steady state reflecting a non-activated basic production rate. To obtain discriminating data about stimulation we incubated the valves until a non-activated steady state was reached at 240 min. After 240 min PGI-2 production was stimulated by 10 µM bradykinin (Sigma, St. Louis, MO)
[8]. The ability to react to stimulation by bradykinin is considered to quantitatively reflect viability (9), and PGI-2 production was therefore measured 30 and 60 min after stimulation (270 and 300 min) (
Fig. 1
).
View larger version (31K):
[in this window]
[in a new window]
|
Fig. 1. Endothelial PGI-2 production of human and porcine aortic valve leaflets from cadaveric (NHBD) and heart-beating (HBD) donors incubated in Earle's Medium 199 before (30, 240 min), and after (270, 300 min) stimulation by 10 µM bradykinin. Values given in pg/ml per 30 min normalized to 1 cm2 of leaflet surface as mean (standard deviation (sd). Significance (*) indicated between groups for each incubation step shown. All stimulated values are significantly elevated against corresponding base-line values. Values given in pg/ml per 30 min normalized to 1 cm2 of leaflet surface as mean±SEM.
|
|
Prostacyclin measurement
PGI-2 is known as an instabile compound with a half-life time between 6 min
[8] and 10 min
[12], it transforms spontaneously in a non-enzymatic reaction into the stabile metabolite 6-k-PGF1a
[12]. To measure this stabile product an ELISA (Perseptive Diagnostics, Cambridge, MA) was used. As the metabolic activity of the endothelial layer is dependent on the absolute number of cells
[9], the 6-k-PGF1a production was normalized to leaflet area by photo-planimetry. Results are given in pg/ml per cm2±SEM. Statistical analyses were performed using Student's t-test for independent populations and ANOVA.
|
Results
|
|---|
Initial PGI-2 production at t=30 min with 304 (±84) pg/ml per cm2 in the porcine HBD group and 622 (±157) pg/ml per cm2 in the porcine NHBD group is significantly (P<0.05) higher than in their human reference groups (see Table 1). The previous warm ischemia in both species induced a tending increase in PGI-2 production with 50 (±11) pg/ml per cm2 in the human HBD versus 226 (±69) pg/ml per cm2 in the human NHBD group (P=0.06) and 304 (±84) pg/ml per cm2 in the porcine HBD versus 622 (±157) pg/ml per cm2 in the porcine NHBD group (P=0.09). After 150 min the PGI-2 production levels fell to a baseline production with the disappearance of species as well as ischemic history differences. During progressive incubation at 240 min differences between graft history are maintained with 234 (±34) pg/ml per cm2 in h-HBD versus 434 (±138) pg/ml per cm2 in h-NHBD and 177 (±40) pg/ml per cm2 in p-HBD versus 577 (±121) in p-NHBD group (P=0.005). Thirty minutes after stimulation at 240 min, PGI-2 production differs between species with 516 (±60) pg/ml per cm2 in h-HBD and 590 (±143) pg/ml per cm2 in h-NHBD versus 1495 (±211) pg/ml per cm2 in p-HBD and 1692 (±187) in p-NHBD group (P=0.0001, P=0.003), while ischemic history has no longer any detectable effect (270 min). Further incubation showed a continued increase of production in the human hearts, whereas in both the porcine groups production decreased again, still remaining significantly elevated above baseline levels (see Table 1).
View this table:
[in this window]
[in a new window]
|
Table 1. The time course of PGI-2 production of human and porcine aortic value leaflets from cadaveric (NHBD) and heart-beating donors (HBD). Values given in pg/mg/30 min normalized to 1 cm2 of leaflet surface as mean±sem.
|
|
|
Discussion
|
|---|
Improved techniques for post-explantation procurement of allografts have resulted in prolonged hemodynamic function with less deterioration of the valve
[1]
[2]
[4]
[5]. A judgement on the procurement quality may be obtained retrospectively from the reoperation rate caused by primary valve malfunction
[6]. The time between implantation and loss of hemodynamic function often spans several years so that the decision process about possible improvements takes too long resulting in an increase in reoperation rates and patient morbidity. This developement was obvious during the late sixties and early seventies when first trials of sterilization and preservation with very toxic techniques like radiation or shock-freezing led to worse clinical performance than experienced with the previous fresh-inserted allografts
[6]. The results with cell protecting sterilization and a gentle freezing technique suggested the importance of the graft's cellular viability for long term durability
[4]
[5]. This theory is supported by many later observations
[1]
[7]
[13]
[14]
[15]
[16]
[17]. The final proof of the significance of cellular viability for longevity of allografts is still missing. The London group analyzed results from a 20 years follow-up of cryopreserved (-196°C) viable and fridge-preserved (+4°C) non-viable grafts and found no significant differences between both groups
[23].
However, as there was the need for a testing system to accelerate the decision process on post-explantation procurement alternatives, the evidence of cellular viability being a determining factor for longevity calls for a standard viability test to judge new techniques. During the past 30 years of research different testing systems were established mainly setting fibrocytes as the target. Autoradiographic proline incorporation as a marker of collagen production became the widely accepted standard
[6]. The destructive setting of the H3 proline method does not allow to follow the time course of biological activity during the procurement process
[24]. For viability testing at different time points, different material has to be used. As the declining viability of allografts is known as a very time sensitive process
[25], a non-destructive testing device was searched for. In the early nineties a metabolic test system was established aiming at the ability of endothelial cells to produce prostacyclin (PGI-2)
[9]. PGI-2 was identified in 1976 as an arachidonic-acid metabolite produced by EC. It was recognized as an unstable compound with a short half-life time transforming by a non-enzymatic reaction to its stable and biologically inert metabolite 6-k-PGF1a. An immuno-assay for 6-k-PGF1a was established in 1978
[19]. The role of the endothelium for the allograft fate is not clear yet.
The cryopreservation of the graft leads to a nearly complete disappearance of the EC
[20]
[21]. This is judged by some authors as an advantage for long-term performance as the EC is recognized as presenting antigens initiating an immune reaction by the host
[18]
[22]. However, as the EC is not only sensitive to cryopreservation but also to toxic substances according to its anatomical exposition its viability is regarded as an indicator for the overall cell survival after exposure to sterilization solutions
[9]
[10]
[11]. The purpose of our study was to compare endothelial PGI-2 production rates of human with porcine valve leaflets related to ischemic history and thereby obtain information on the validity of the frequently used porcine animal model. An attempt was made to model human allograft harvesting and procurement from heart-beating and cadaveric donors by matched animal groups. This certainly has limitations from chronic disease and the use of cardiovascular acting drugs in the transplant recipients and from the variety of death courses and the agonal period in cadaveric donors. In the latter group, however, there was no history of chronic drug treatment, and the absence of differences in stimulated PGI-2 production between the human groups would indicate unimpaired metabolic activity in the domino donors.
Our results show species differences in the early phase after surgical manipulation and ischemic stress. The latter is identified as the causal factor for raised PGI-2 production in both species groups. As the differences between species disappear at a steady-state the testing system must allow the EC to recover from initial mechanical stimulation to obtain valid data. It has been reported that the first steady-state conditions of porcine material are reached after 30–45 min followed by a peak at 150 min and decreasing to a second steady state at 240 min
[10]. The peak was identified as a response to mechanical stress during valve dissection, while the steady state was interpreted as a non-activated basal production rate. We therefore choose the stimulation point at 240 min to transform the endothelial cell layer from a non-activated status into an activated one.
However, our results revealed the first steady state at 90 min in the human HBD as well as in the porcine HBD group. The NHBD groups reached the first steady state at 120 min. In neither group a peak could be detected from our data before stimulation, although the natural production rates were still increasing at 240 min. Therefore, the reservation must be made that stimulation was induced too early in the spontaneous time course.
As we derived our stimulation point from previous published data
[10] further studies should consider the obvious delay of natural endothelial production rates due to ischemic history. The ability of defined stimulated PGI-2 production by bradykinin awaits determination to distinguish it precisely from natural oscillating production waves.
Summarizing from our results we conclude that endothelial PGI-2 production in the porcine model for the study of preservation and procurement of human valve allografts must be regarded with high care. Valid data are obtained only if the natural time-course and reaction to stimulation of PGI-2 production are duely recognized and species differences in the response to mechanical stress and ischemia are considered.
|
Footnotes
|
|---|
Presented at the 11th Annual Meeting of the European Association for Cardio-thoracic Surgery, Copenhagen, Denmark, September 28 – October 1, 1997. 
|
References
|
|---|
- Angell W.W. Long-term follow-up of viable frozen aortic homografts. J Thorac Cardiovasc Surg 1987;93:815-822.[Abstract]
- O'Brien M.F. Allograft aortic valve replacement: long term comparative clinical analysis of he viable cryopreserved and antibiotic 4°C stored valves. J Cardiac Surg 1991;6(Suppl. II):65-70.
- Scheld H.H. The pathology of bioprostetic heart valves and allografts. Curr Top Pathol 1994;86:127-157.[Medline]
- Barrat-Boyes B.G., Roche A.H.G., Subramanyan R. Long-term follow-up on patients with the antibiotic-sterilized aortic homograft valve inserted freehand in the aortic position. Circulation 1987;75:768-777.[Abstract/Free Full Text]
- O'Brien M.F., Stafford E.G., Gardner M.A.H. The viable cryopreserved allograft aortic valve. J Cardiac Surg 1987;2(Suppl. I):153-167.[Medline]
- Hopkins RA. Cardiac reconstruction with allograft valves. New York: Springer, 1989: p. 22.
- Brockbank K.G.M., Bank H.L. Measurement of postcryopreservation viability. J Cardiac Surg 1987;1(Suppl. I):145-151.
- Jaffe E.A. Cell biology of endothelial cells. Hum Pathol 1987;18:234-239.[Medline]
- Feng X.J., Van Hove C.E.J., Mohan R., Andries L., Rampart M., Herman A.G., Walter P.J. Improved endothelial viability of heart valves cryopreserved by a new technique. Eur J Cardio-thorac Surg 1992;6:251-255.[Abstract]
- Feng X.J., Van Hove C.E., Mohan R., Walter P.J., Herman A.G. Effects of different antibiotics on the endothelium of the porcine aortic valve. J Heart Valve Desease 1993;2:694-704.[Medline]
- Fischlein T., Schütz A., Uhlig A., Frey R., Reichart B. Integrity and viability of homograft valves. Eur J Cardio-thorac Surg 1994;8:425-430.[Abstract]
- Needleman P., Turk J., Jakschik B.A. Arachidonic acid metabolism. Ann Rev Biochem 1986;55:69-74.[Medline]
- Yankah A.C., Wottke H.U., Muller-Ruchholtz W. Prognostic importance of viability and a study of a second set allograft valve: an experimental study. J Cardiac Surg 1988;3:263-270.[Medline]
- McGiffin D.C., O'Brien M.F., Stafford E.G., Gardner M.A., Pohlner P.G. Long-term results of the viable cryopreserved allograft aortic valve: continuing evidence for superior valve durability. J Cardiac Surg 1988;3:289-296.[Medline]
- Schütz A., Fischlein T., Hammer C., Reichart B. Cytoimmunological monitoring after homograft valve replacement. Eur J Cardio-thorac Surg 1994;8:609-612.[Abstract]
- Motomura N., Imanita M. Cell viability of aortic allografts after long-term cryopreservation and clinical application to aortic root replacement in patients with aortitis. Cardiovasc Surg 1995;2:231-234.
- Yacoub M., Rasmi N.R., Sundt T.M. Fourteen-years experience with homovital homografts for aortic valve replacement. J Thorac Cardiovasc Sur 1995;110:186-194.[Abstract/Free Full Text]
- Yankah A.C., Wottge H.U. Allograft conduit wall calcification in a model of chronic arterial graft rejection. J Card Surg 1997;12:86-92.[Medline]
- Salmon J.A. A radioimmunoassay for 6-keto-prostaglandin F-1a. Prostaglandins 1978;15:382-397.
- Solberg S., Larsen T., Jorgensen L., Sorlie D. Cold induced endothelial cell detachment in human saphenous vein grafts. J Cardiovasc Surg 1987;28:571-575.[Medline]
- Lupinetti F.M., Tsai T.T., Kneebone J.M., Bove E.L. Effect of cryopreservation on the presence of endothelial cells on human valve allografts. J Thorac Cardiovasc Surg 1993;106:912-917.[Abstract]
- Yankah A.C., Wottge H.U., Müller-Hermelink H.K. Transplantation of aortic and pulmonary allografts, enhanced viability of endothelial cells by cryopreservation: importance of histocompatibility. J Card Surg 1987;1(Suppl. I):209-220.
- Bodnar E., Matsuki O., Parker R., Ross D.N. Viable and nonviable aortic homografts in the subcoronary position: a comparative study. Ann Thorac Surg 1989;47:799-805.[Abstract]
- Song Y.C., Yao L.Y., Kneebone J.M., Lupinetti F.M. Effect of cryopreservation and histocompatibility on type I procollagen gene expression in aortic valve grafts. J Thorac Cardiovasc Surg 1997;114:421-427.[Abstract/Free Full Text]
- Pompilio G., Polvani G.L., Rossoni G., Porqueddu M., Berti F., Barajon I., Petruccioli M.G., Guarino A., Aguggini G., Biglioli P., Sala A. Effects of warm ischemia on valve endothelium. Ann Thorac Surg 1997;63:656-662.[Abstract/Free Full Text]
Top
Abstract
Introduction
