| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Eur J Cardiothorac Surg 2004;25:98-104
© 2004 Elsevier Science NL
a Institute of Experimental Medicine, University of Cologne, Robert-Koch-Str. 10, 50931 Cologne, Germany
b Second Department of Surgery, Kagoshima University, Kagoshima, Japan
c Department of Cardiothoracic Surgery, University of Cologne, Cologne, Germany
Received 9 July 2003; received in revised form 29 September 2003; accepted 20 October 2003.
* Corresponding author. Tel.: +49-221-478-4129; fax: +49-221-478-6264
e-mail: jh.fischer{at}uni-koeln.de
 |
Abstract |
|---|
 Top Abstract 1. Introduction2. Methods3. Results4. DiscussionReferences |
|---|
Key Words: Endothelium derived relaxation • Coronary oxygen persufflation • Heart preservation • Non-heart-beating donor • NHBD • Heart transplantation
Abbreviations: BDM, 2,3-butanedione monoxime • CPB, cardiopulmonary bypass • COP, coronary oxygen persufflation • EDHF, endothelium derived hyperpolarizing factor • EDRF, endothelium derived relaxing factor • HBD, heart-beating donor • HLM, heart–lung machine • HTK, histidine-tryptophan-ketoglutarate solution Custodiol® • Indo, indomethacin • mBHTK, modified BDM containing HTK solution • NHBD, non-heart-beating donor
|
1. Introduction |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. DiscussionReferences |
|---|
Experiments on several species have shown that organs of NHBD should be oxygenated immediately following the normothermic ischemic phase to minimize irreversible damage. The possible means to achieve this are: (1) the use of a heart–lung-machine for normothermic oxygenated continuous perfusion; (2) an oxygenated hypothermic perfusion system; or (3) gaseous oxygen persufflation during storage preservation. Methods 1 and 2 are very costly and technically demanding, but method 3 is simple yet equally effective. This technique, which was developed as retrograde oxygen persufflation (ROP) for kidney and liver transplantation [4,5], has recently been adapted as a coronary oxygen persufflation (COP) for the heart [8,9].
Retrograde oxygen persufflation has already been performed in several studies for preserving livers and kidneys after normothermic ischemic predamage. Organs preserved in this way were clearly superior to organs merely stored hypothermically using clinically accepted preservation solutions [4–7]. It has been shown in several studies that the oxygen persufflation in hypothermia maintains the pool of energy-rich phosphates and decreases the structural cell damage, thus resulting in a better restitution of organ function.
COP has also been shown to be effective for 14 h of pig heart preservation [8,9]. In our previous paper the functional and metabolic recovery of pig hearts preserved for 3.3 h following a NHBD phase of 16 min was tested [11]. In contrast to hearts stored without COP for the same period, the persufflated hearts were able to support the recipient's circulation within 2 h after orthotopic transplantation, with a cardiac output reaching 68% of the normal values before transplantation, and could easily be weaned from the heart–lung-machine. Without oxygenation during the preservation period and use of HTK solution, the cellular energy stores of these predamaged hearts decreased further, resulting in myocardial contracture and a cardiac output below 5% of the pretransplant control values.
Coronary oxygen persufflation (COP) differs from the previous retrograde gaseous persufflation techniques (ROP) of kidney and liver, as it is an orthograde persufflation technique with the oxygen gas coming in contact with all regions of the vascular bed.
In previous experiments, coronary arteries of pigs, which had been COP preserved for 14 h or even 18 h in modified HTK solution without a preceding NHB phase, have been shown to be unchanged in their ability for endothelium-derived relaxation [11–13]. Particularly in our experiments with heterotopic transplantation after 14 h COP-preservation of pig hearts, including light and electron microscopy at the end of a 7-day recovery period, no structural defects were found related to the COP technique.
However, it is not known whether COP combined with preceding warm ischemia affects the coronary arteries and especially the vascular endothelium, which in vivo regulates the vascular tone by producing different vasoactive substances.
In our present study, we tested the endothelium-derived relaxation of coronary rings at the end of experiments on pigs including a NHB phase of 16 min, 3.3 h preservation with modified HTK solution including simultaneous COP (mBHTK+COP), orthotopic transplantation, and 3 h in situ recovery. Sixteen-minute normothermic ischemia was used because this period allows retrieval of NHBD organs in several countries (for further discussion see [10]); 3.3 h cold preservation was used because this is within the clinically accepted range of 3–4 h for the preservation period for cold stored hearts. All these hearts had reached an excellent functional recovery, and the recipients were weaned from cardiopulmonary bypass 1 h before the end of the experiment (see Ref. [10]). For controls, similar experiments were performed without COP, using commercial HTK-solution or modified HTK-solution (mBHTK).
|
2. Methods |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. DiscussionReferences |
|---|
Sixteen pigs (37 kg average body weight) received premedication of azaperon 4 mg/kg and atropine 0.02 mg/kg intramuscularly, followed by propofol 1–2 mg/kg and ketamine hydrochloride 5–10 mg/kg intravenously. Pigs were placed in a supine position and endotracheally intubated. Controlled ventilation with room air was started using a volume-cycled respirator (Engström Respirator System 300, LKB Medical AB, Bromma, Sweden) and continuous expiratory CO2 measurement. Anesthesia was maintained by continuous infusion of 1 mg/kg per min ketamine hydrochloride. When required, 0.2 mg/kg pancuronium was infused intravenously for muscle relaxation. The common carotid artery and the internal jugular vein were cannulated for continuous measurement of arterial blood pressure (AP), and central venous blood pressure (CVP). Heart rate (HR) was also measured continuously.
Hemorrhagic shock was induced by exsanguination from the abdominal aorta, following the administration of 500 IU/kg of heparin. Blood pressure rapidly dropped to values below 10 mmHg within 1 min. The heart was then left ischemic for 16.7±1.2 min until cardioplegic perfusion was begun. After the normothermic ischemic period, during which a sternotomy and a pericardiotomy were performed, the ascending aorta was cross-clamped and the coronary system was anterogradely perfused with cold (0–1 °C) preservation solution, cooled in an ice-water mixture. The coronary perfusion pressure was measured via a separate catheter positioned in the ascending aorta, and was adjusted to 75 mmHg for the first minute and lowered to 40 mmHg for the following 9 min as recommended for the use of Bretschneider‘s cardioplegic histidine-tryptophan-ketoglutarate (HTK) solution (Custodiol®, Dr. F. Köhler Chemie, Alsbach-Hähnlein, Germany). This solution contains (in mmol per liter): 15 NaCl, 9 KCl, 4 MgCl2, 18 histidinechloride, 180 histidine, 2 tryptophan, 30 mannitol, 1 potassiumhydrogen-2-ketoglutarate, and 0.015 CaCl2.
In four pigs, cardiac arrest was induced by pressure-controlled anterograde coronary perfusion with original HTK solution (HTK, n=4). In 12 pigs, a modified HTK solution was used (mBHTK). The modification consisted in an addition of: 40 mg/l of hyaluronidase (Boehringer-Mannheim/Roche, Germany), 30 mmol/l 2,3-butanedione monoxime (BDM), 15 µmol/l adenosine (Sigma-Aldrich, Taufkirchen, Germany), and 50 µmol/l CaCl2.
The hearts were then removed and stored in 200 ml of the respective preservation solution at 0–1 °C (four HTK storage controls and six mBHTK storage controls) in a container surrounded by iced water for at least 3 h, including the period of transplantation with continued cooling on a cooling jacket until the start of reperfusion. During the storage period, six of the pig hearts preserved in mBHTK solution were additionally oxygenated by continuous anterograde coronary oxygen persufflation (COP). In these hearts, gaseous oxygen was administered into the aortic root via a plastic cylinder with a central opening fixed into the ascending aorta. To avoid leakage of gaseous oxygen into the left ventricle, the aortic valve was securely closed gas-tight using a self-made silicone rubber valve guard [9] (a self-made silicone rubber device, stabilizing the aortic valve). The valve guard was fastened to the tip of the aortic valve cusps with a single 8-0 polypropylene stitch, which could be easily removed later on before starting the reperfusion. COP pressure was 45 mmHg, and oxygen gas flow was 80±10 ml/min.
Orthotopic transplantation was performed using the technique of Lower and Shumway [14]. The transplantation procedure has been described in detail in our previous paper [10], but is briefly described here. Recipient pigs with a body weight similar to the donors were anaesthetized as described for the donor pigs; 500 IU/kg heparin and 500 mg methylprednisolone were given before the start of the cardio-pulmonary bypass with a heart–lung machine (HLM-CAPS, Stöckert Instruments, Munich, Germany). During the entire procedure, except for the final suture of the aorta, oxygen persufflation of the coronaries was continued in the hearts of the mBHTK+COP group.
Heart reperfusion was started with warm (37 °C) modified Krebs–Henseleit solution including 15 µmol/l adenosine, 1 mmol/l uric acid, and 1 IU/l insulin added, but containing only 50 µmol/l calcium. The Ca2+ content of the solution was gradually elevated to 1 mmol/l between the 5th and 10th minute. After 10 min, coronary blood reperfusion started. For the first hour, adenosine (13.5 µmol/min) was continuously infused into the aortic root. In cases of ventricular fibrillation, 100–200 mg lidocaine hydrochloride (Xylocain 2%® Astra, Wedel, Germany) was injected into the aortic root. Measurements of ventricular pressure and aortic flow were performed continuously; after 180 min of blood reperfusion, transmural myocardial specimens were taken from ventricular and atrial regions, frozen immediately in liquid nitrogen and myocardial metabolic state tested after perchloric acid extraction using HPLC (hemodynamic and metabolic results were published separately [10]).
Measurements of coronary function were carried out by means of a lever transducer system (Fig. 1) , which records contractions and or dilations of the coronary rings depending on the substance applied in the organ bath. Any alteration of vascular tone was recorded (Multi-pen Recorder; Rikadenki Kogyo Co. Tokyo, Japan) via a transducer (Lever Transducer B 40 Type 373; Hugo Sachs Elektronik, March, Germany) and an amplifier (Transducer- Amplifier Module Type 705/1; Hugo Sachs Elektronik, March, Germany).
|
After 18 h of incubation, up to four rings from each coronary were tested in a standardized procedure of four different runs. The coronary rings were fixed between two triangular steel-wire holders under a load of 2 g, placed in a container with 10 ml Krebs–Henseleit solution (KH, Table 1), and administered with carbogen at 37 °C (see Fig. 2)
. After reaching a steady state of vascular tension in a first run, KCl (60 mmol/l) was applied to trigger a contraction. All concentration values are the concentrations in the organ bath of 10 ml KH (see Table 2). Following each run, the KH was exchanged at least three times to remove all traces of the test substances. Each of the following runs was initiated by giving indomethacin (10 µmol/l) to block the cyclooxygenase pathway. The second run was then continued by adding PGF2
(10 µmol/l; Dinolytic, Pharmacia Upjohn GmbH, Erlangen, Germany), which induced contraction of the rings, followed by application of Substance P (10 nmol/l, SP, Fluka Chemie GmbH, Neu-Ulm, Germany), which caused endothelium-dependent dilation. The second run was repeated with an additional incubation with L-NIL (10 µmol/l, L-N6-[-1-iminoethyl]-lysine; Sigma-Aldrich GmbH, Steinheim, Germany) to block NO production by iNOS. The last run differed from the preceding ones by blocking all NO production with the addition of L-NNA (300 µmol/l N-nitro-L-arginine, Sigma-Aldrich GmbH Steinheim Germany). At the end of the test, Papaverin (200 nmol/l; Knoll, Ludwigshafen am Rhein, Germany) was added to achieve maximal dilation. If no contraction was achieved by the first addition of PGF2 in the tests immediately after the end of the reperfusion, contraction was induced by 0.125 µmol/l of the thromboxan-agonist U46619 and in a parallel ring PGF2 induced contraction was tested after addition of L-NNA.
|
|
|
All data are expressed as mean values±standard deviation (S.D.). Significance of differences between groups was tested using analysis of variance, followed by Bonferroni corrected t-test for multiple comparison. Statistical significance was considered for P<0.05.
|
3. Results |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. DiscussionReferences |
|---|
, the endothelium-derived relaxation (EDR) caused by SP was 78±7% in the mBHTK+COP group; this is not significantly different from the physiologic values measured in coronaries from the slaughterhouse, which were 75±9%. The EDR in the control groups was also similar with 77±20% in the mBHTK group and 72±7% in the HTK group without significant differences versus the mBHTK+COP group or physiologic values (P>0.05 in all comparisons, see Fig. 3)
.
|
After application of L-NNA for blockage of all NO production, a significant reduction of SP-induced relaxation (P<0.05) could be observed in all groups versus initial values or versus the values after L-NIL incubation. EDR was significantly reduced to 58±8% in the mBHTK+COP group (P<0.05 vs. normal EDR or L-NIL test) and reduced significantly to 48±8% in the mBHTK control group (P<0.05 vs. normal EDR or L-NIL test). The reduction of 55±13% in the HTK control group was significant only versus L-NIL test. No significant differences were found between the mBHTK+COP group and the control groups mBHTK or HTK (see Fig. 3).
Immediately at the end of transplantation and reperfusion of the hearts and isolation of the coronary rings, PGF2 induced contraction was hardly measurable, while following the 18-h immersion in oxygenated cell culture medium at room temperature (22 °C), a regular contractile response similar to physiologic values was restored.
A contractile response of rings to PGF2 immediately after the end of reperfusion was found only after incubation with 300 µmol/l L-NNA. However, it could be induced by using 125 pmol/l U 46619 instead of PGF2. The SP dilation values were similar in coronary rings tested immediately using U 46619 contraction and in those from the same coronary artery with PGF2 precontraction after restoring their contractile response by an 18-h cell culture storage.
|
4. Discussion |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. DiscussionReferences |
|---|
The hearts of the control groups were unable to restore an acceptable cardiovascular function [10], due to the lack of oxygenation. Nonetheless, they were still capable of producing an EDR similar to physiologic values, without any difference to the mBHTK+COP group. In these groups, only the extent of contraction induced by PGF2 was smaller compared to the mBHTK+COP group, and the lowest values were in the HTK group.
We used Substance P for the investigation of endothelium-derived relaxations. It stimulates the release of the EDRF nitric oxide (NO), as well as the release of endothelium-derived hyperpolarizing factor (EDHF), in a large number of vessels including large and small coronary arteries of the pig [15]. Damage to the coronary endothelium, which results in less vascular relaxation in situ and thus leads to reduced myocardial perfusion, can easily be demonstrated by measuring the capability for Substance P induced relaxation of vessel segments in vitro. The use of acetylcholine, the substance for which Furchgott and Zawadzki first described the existence of EDRF [16], was not possible, because, in contrast to the coronary resistance vessels and many other arteries [13], the large coronary arteries of the pig are not relaxed by any concentration of this substance (unpublished results from our laboratory).
It has been shown in previous studies that the use of University of Wisconsin (UW) solution [17] or of St. Thomas hospital solution [18] for heart preservation damages the endothelial function. While UW solution seems to damage the endothelium by its high potassium concentration or by the pronounced development of edema during the reperfusion [19,20], St. Thomas hospital solution may cause this damage by its high calcium content of 1.2 mmol/l, resulting in cellular calcium gain during hypothermia.
The modification of HTK solution to mBHTK results in less edema formation by hyaluronidase and in contracture inhibition by BDM. This modification had no negative influence on the EDR.
The test sequence of coronary rings in our experiments started with 60 mmol/l potassium chloride to test the ability for maximal contraction and to exclude rings with damaged muscle layers from the study. The concentration of PGF2 used in the following tests resulted in about 50% of the potassium chloride contraction of normal coronary rings from the slaughterhouse.
Using Substance P for EDR tests, we incubated the rings in all tests with indomethacin, in order to block possibly interfering prostaglandin effects, which can also induce vasodilatation, for instance by prostacyclin (PGI2) [22].
The purpose of the second test including L-NIL was to differentiate the NO production by activation of endothelial NOS (eNOS=NOS III) from the NO production by activation of inducible NOS (iNOS=NOS II). This was performed by incubation with L-NIL, a specific iNOS inhibitor [23]. While the presence of cerebral NOS type NOS I can be excluded from coronary arteries, the inducible enzyme can be synthesized by a variety of cell types in response to inflammatory stimuli, which are a typical effect of HLM perfusion. Although coronary arteries of pigs without endothelial damage show eNOS activity but no iNOS activity, an increase of iNOS activity and a decrease of eNOS activity caused by endothelial damage has been demonstrated for pig carotid arteries [24]. But as our results show, there was no difference between EDR with or without L-NIL. Therefore, it is quite unlikely that there was any noticeable effect of iNOS in the coronary arteries of our study.
The purpose of the last test including L-NNA was to differentiate between NO and EDHF, which are both parts of the total EDR response. Previous studies have shown that total NO production can be almost completely eliminated by 300 µmol/l of L-NNA and that the remaining dilation response to substance P is the effect only of EDHF [25]. EDR was reduced similarly in all groups after L-NNA incubation. This reflects the similar capacity of relaxation of the coronary vessels from the EDHF mechanisms, irrespective of the preservation technique. In a previous experiment, isolated pig coronaries from the slaughterhouse were preserved for 3 or 18 h and reperfused in vitro with Kreb–Henseleit solution. This resulted in somewhat lower EDHF relaxations of 41% or 39% on average, while the total EDR without L-NNA, similar to the present results, was at physiologic values of 75% or 76% on average [13].
As described above, the standard test of the coronary rings was done after an 18-h incubation of the rings in oxygenated cell culture solution (without fetal albumin) at room temperature. Without this incubation immediately at the end of the reperfusion period, no or only minimal contractions could be induced by PGF2. But following the incubation with L-NNA, the contraction response to PGF2 recovered. Since L-NNA only blocks NO production, the most likely interpretation for the initial lack of contractile reaction to PGF2 seems to be a very high level of NO which was inducing maximal dilation.
The thromboxan-agonist U46619 is a stable PGH2 analogon [21] and a stronger vasoconstrictor than PGF2. U46619 was able to induce contraction of the rings immediately at the end of our experiment. This fact supports the interpretation that the contractile response was still present but weakened by a simultaneous dilation. We compared rings from the same coronary which either: (1) had undergone U46619 contraction at the end of the experiment; or (2) had undergone 18 h incubation and then PGF2 contraction. These two sets of rings demonstrated a similar percentage of endothelium-derived relaxation by Substance P. Of course there might be a difference between the endothelium of the large or small coronary arteries with respect to the availability of oxygen during the normothermic ischemic phase from the contained blood. The first aim of our study was to evaluate the endothelial function after the consequent exposure to normothermic ischemia and gaseous oxygen persufflation. In a second approach we currently test the survival of vascular reactivity of myocardial resistance vessels in rabbit hearts, reperfused after various periods of normothermic ischemia followed by storage preservation with or without COP. Preliminary results show a good recovery of the vascular function measured by flow increase during stimulation of endothelium derived relaxation with various substances.
In conclusion, the results of our study demonstrate that the use of coronary oxygen persufflation with gaseous oxygen during a 3.3-h storage period (after 16 min of warm ischemia) does not damage the endothelium-dependent function of the large coronary arteries. Pig hearts treated in this way are able to guarantee sufficient blood supply for the recipient. They also show endothelium-derived relaxations similar to those of controls preserved without COP and even similar to the physiologic values of coronaries from the slaughterhouse. iNOS seems to not be involved in the EDR of these coronaries, and the relaxation by EDHF which remains after L-NNA blockage of NO production also does not differ from the controls.
The use of coronary oxygen persufflation is therefore a promising means to reduce the shortage of available hearts for transplantation. It allows excellent recovery of preserved NHBD hearts with full functional integrity of the coronary endothelium.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. DiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. H. Fischer and M. Steinhoff Effects of aprotinin on endothelium-dependent relaxation of large coronary arteries Eur. J. Cardiothorac. Surg., December 1, 2005; 28(6): 801 - 804. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
