| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 1999;118:665-673
© 1999 Mosby, Inc.
SURGERY FOR CONGENITAL HEART DISEASE |
From the Division of Cardiovascular Surgery, The Heart Institute for Children, Hope Children’s Hospital, Oak Lawn, Ill.
M.T.K. is supported in part by the Pillsbury Fellowship.
Address for reprints: Bradley S. Allen, MD, The Heart Institute for Children, Hope Children’s Hospital, 4440 West 95th St, Oak Lawn, IL 60453.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionAppendix: DiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAppendix: DiscussionReferences |
|---|
1-5 However, at higher concentrations, nitric oxide is detrimental because of overproduction of the toxic oxygen radical peroxynitrite. 2,5-8 It would therefore be advantageous if the beneficial effects of L-arginine (nitric oxide) could be achieved by another method since the optimum dose of L-arginine, as well as whether nitric oxide production or inhibition is desired, varies depending on several factors. White blood cell (WBC) filters have been shown to reduce the reperfusion injury, and because one of the actions of nitric oxide is to prevent WBC adherence, leukodepleting the cardioplegic solution should have a similar effect. 9-11 Nitric oxide, however, also promotes vasodilation, as well as decreasing platelet adherence; thus a WBC filter does not completely mimic its actions. Prostaglandins are secreted by vascular endothelial cells and may be one of the mediators by which nitric oxide works. 7,12-14 Prostaglandins promote vasorelaxation, as well as preventing WBC and platelet adherence, and therefore more closely mimic the actions of nitric oxide. By investigating cardioplegic solutions with and without L-arginine, leukodepleted, and supplemented with prostaglandin E1 (PGE 1, alprostadil), this study aims (1) to elucidate whether the primary beneficial effect of nitric oxide is vasodilatation, prevention of a WBC-mediated injury, or both, and (2) to provide an alternative approach to L-arginine for improving myocardial protection. |
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAppendix: DiscussionReferences |
|---|
15,16
Experimental protocols
Hypoxic-ischemic stress.
Twenty piglets underwent a hypoxic-ischemic stress as previously described by lowering the fraction of inspired oxygen to 8% to 10% for 60 minutes (ventilator hypoxia), followed by cardiopulmonary bypass at an inspired oxygen fraction of 100% for 5 minutes (reoxygenation), and finally clamping the aorta for 20 minutes at 37°C (normothermic ischemia). 5 Piglets then underwent 70 minutes of cardioplegic arrest according to the protocol described below.
Administration of cardioplegic solution.
Cardioplegic solutions (CAPS Service, Research Medical Inc, Salt Lake City, Utah) are shown in Tables I and II. After the 20-minute normothermic-ischemic insult (see above), piglets underwent 70 minutes of cardioplegic arrest. The protocol, previously described, consisted of 5 minutes of warm (37°C) induction (Table I) followed by 4 minutes of cold multidose cardioplegia (Table II), a 2-minute cold multidose infusion every 20 minutes, and a 4-minute warm (37°C) cardioplegic reperfusate ("hot shot") before aortic unclamping. 5 Cardioplegic solution was always infused at a continuously measured aortic root pressure of 40 to 50 mm Hg. All piglets were weaned from cardiopulmonary bypass with no inotropic support 30 minutes after aortic unclamping. Final functional and biochemical measurements were made 30 minutes after arterial blood gas, Ca2+, and K + levels were normalized.
|
|
Unmodified blood cardioplegic solution (group 1).
In 5 piglets, the blood cardioplegic solution was not enriched with L-arginine or PGE1 or passed through a leukocyte filter.
L-Arginine (group 2).
In 5 piglets, the cardioplegic solution was enriched with L-arginine (4 mmol/L).
Prostaglandin (group 3).
In 5 piglets, the cardioplegic solution was supplemented with PGE 1 (alprostadil, 4 µg/L).
WBC filter (group 4).
In the final 5 piglets, the cardioplegic solution was passed through a Pall BC-1 leukocyte depleting filter (Pall Corp, G1en Cove, NY) before administration.
Myocardial oxygen consumption.
After cardioplegic arrest, blood was obtained at 1-minute intervals from the cardioplegia line and coronary sinus over the 5 minutes of warm cardioplegic induction, and myocardial oxygen consumption was calculated as previously described. 17 The cumulative 5-minute myocardial oxygen consumption was determined by the sum of the individual 1-minute values and expressed per 100 g of heart tissue.
Myocardial performance.
Left ventricular (LV) pressure and conductance catheter signals were amplified and digitized to inscribe LV pressure volume loops, the end-systolic and end-diastolic pressure volume relationship, and preload recruitable stroke work, analyzed as previously described. 5,15-18 Measurements were made before hypoxia (baseline) and 30 minutes after cardiopulmonary bypass was discontinued. Functional measurements are expressed as absolute values as well as percent recovery of baseline values, with each piglet acting as its own control. After final hemodynamic measurements, all piglets were placed back on bypass and hearts were arrested with cold (4°C) blood cardioplegic solution. Transmural LV biopsy specimens were obtained, as previously described, for biochemical analysis and myocardial water. 5
Physiologic measurements.
Coronary vascular resistance (CVR) was determined during each cardioplegic infusion by measuring coronary sinus pressure and cardioplegic flow once a constant infusion rate with an aortic root pressure between 40 and 50 mm Hg was achieved. CVR was calculated as previously described and expressed as dynes · sec · cm–5. 15,16
Biochemical analysis
Production of conjugated dienes.
Blood samples were obtained from the cardioplegia line and coronary sinus 1 and 5 minutes after the start of warm cardioplegia induction, and myocardial conjugated dienes were assessed as previously described. 5 Production of conjugated diene was expressed per 100 gm heart tissue by weighing the left ventricle at the conclusion of the experiment.
Adenosine pool and myeloperoxidase activity.
The adenosine pool and quantitative myeloperoxidase activity were determined as described previously. 5,11,15 Adenosine triphosphate (ATP) levels are expressed as micrograms per gram of dry tissue and myeloperoxidase activity as the change in optical density per minute per milligram of tissue protein (
OD/min/mg protein).
Antioxidant reserve capacity.
Myocardial antioxidant reserve capacity was determined as described previously by exposing the tissue to a 4 mmol/L concentration of t-butylhydroperoxide and measuring the production of malondialdehyde. 5,19 Antioxidant reserve capacity is expressed as malondialdehyde production in nanomoles per gram protein of heart tissue. The greater the production of malondialdehyde to the oxidant stress (t-butylhydroperoxide), the lower the levels of endogenous tissue antioxidants (lower antioxidant reserve capacity), indicating prior exposure of the myocardial tissue to oxygen-derived free radicals during reperfusion.
Myocardial water.
The percent myocardial water was calculated as previously described by placing ventricular samples in preweighed vials and drying to a constant weight at a temperature of 85°C. 5
Statistics.
Data were analyzed with JMP V2.0 software (SAS Institute, Inc, Cary, NC) on a Macintosh IIVX computer (Apple Inc, Cupertino, Calif). The paired Student t test and 1-way analysis of variance were used for comparison of variables among experimental groups. If the analysis of variance revealed a significant interaction, pair-wise tests of individual group means were compared by means of multiple comparisons (Tukey’s test) using a level of significance of P < .05, P < .01, and P < .001. Group data are expressed as mean ± standard error of the mean.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAppendix: DiscussionReferences |
|---|
Hemodynamic and physiologic measurements.
Results are depicted in Table III and Figs 1 to 4. There was no significant change or difference (P > .2) in the X-axis intercept for end-systolic elastance or preload recruitable stoke work between prebypass (baseline) and postbypass values in any experimental group. Therefore, the change in the slope of end-systolic elastance and preload recruitable stroke work can be interpreted to express variability in the contractile state of the myocardium compared with baseline values. This allows the change in slope to be expressed as a percentage of baseline with each piglet acting as its own control. Unmodified blood cardioplegic solution (group 1) was unable to resuscitate the severely stressed (hypoxic-ischemic) myocardium, resulting in decreased postbypass systolic contractility, markedly increased diastolic stiffness, and reduced preload recruitable stroke work. In contrast, cardioplegic solution supplemented with L-arginine (group 2), PGE1 (group 3), or passed through a WBC filter (group 4) resuscitated and protected the severely stressed myocardium, resulting in complete return of systolic function and preloaded recruitable stroke work, and minimally increased diastolic stiffness. These values (groups 2-4) were not statistically different (P > .2) from each other, indicating similar beneficial effects for all 3 interventions with regard to myocardial functional recovery. The Pall BC-1 leukodepleting filter was extremely efficient at lowering the concentrations of WBCs (13.2 ± 0.4 prefilter vs 0.3 ± 0.1 postfilter) and neutrophils (7.5 ± 0.3 prefilter vs 0.2 ± 0.1 postfilter), but only mildly reduced the platelet count (234 ± 12 prefilter vs 139 ± 10 postfilter).
|
|
|
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAppendix: DiscussionReferences |
|---|
1-4 However, L-arginine also produces the toxic oxygen-derived free radical peroxynitrite and thus under certain conditions it can be detrimental. 5,6,20 Because it may be difficult to determine under what conditions and at what dose L-arginine may be harmful, an alternative approach that achieves the same beneficial effects would be advantageous. Prostaglandins are released by endothelial cells and, like nitric oxide, prevent WBC adherence and vasodilate coronary vessels. 7,12-14,21 Whereas L-arginine needs to be converted to nitric oxide by the intact endothelial cell, prostaglandins act directly. Prostaglandins could therefore be superior in the presence of damaged endothelium. PGE1 does not produce oxygen radicals at higher doses, and this study used a concentration of PGE1 that results in a low systemic dose (0.03-0.04 µg · kg–1 · min–1). Systemic doses up to 5 to 10 times this dose are routinely used clinically without problems. 21 A WBC filter does not act directly on the vascular endothelium or cause vasodilatation. It was therefore used in 1 group of piglets to allow quantification of the contribution of WBCs in mediating the reperfusion injury independent of other factors (ie, vasodilation).
This study used both an ischemic and hypoxic stress because each increases WBC adherence. This protocol thus allowed the methods of protection to be tested in vulnerable hearts. Furthermore, compared with hypoxia alone, the combination of hypoxia and ischemia may more closely resemble the condition of the cyanotic child. 5,15,16,18,22-24 After a hypoxic-ischemic stress, hearts protected with unmodified blood cardioplegic solution sustained a significant oxygen-derived free radical–mediated reperfusion injury resulting in vascular dysfunction (elevated CVR), mitochondrial damage, and depressed myocardial function. In contrast, oxygen-derived free radical production was equally reduced if the cardioplegic solution was enriched with L-arginine or PGE1 or was passed through a WBC filter. By avoiding the reperfusion injury, mitochondrial function was preserved (ATP/ADP ratio), which allowed energy levels to be replenished, resulting in increased oxygen uptake during cardioplegic induction, as well as complete functional recovery. Compared with L-arginine and PGE 1, use of leukocyte filtration resulted in similar levels of oxygen-derived free radical production, myeloperoxidase activity, and metabolic and functional recovery. This implies that WBCs are responsible for the generation of the majority of oxygen-derived free radicals during reperfusion and that most of the beneficial effects of L-arginine and prostaglandins are due to inhibition of WBC adherence. We cannot prove, however, that all these modalities work via an identical pathway, because we did not specifically examine the mechanism of action with each intervention. It is possible that L-arginine and PGE1 limit the reperfusion injury via a non-WBC-dependent mechanism or by a combination of effects, since these agents also prevent platelet adherence and induce vasodilatation. However, because one of the primary effects of L-arginine (nitric oxide) and PGE1 is to prevent adherence of activated WBCs, this mechanism is probably responsible for a substantial portion of their beneficial effects.
CVR was measured with each cardioplegic infusion to determine whether any of the beneficial effects of L-arginine or PGE1 were due to their ability to promote vasodilatation. 1-3,7,12,21,25 After an ischemic stress, the coronary vasculature should be vasodilated. The higher CVR with unmodified blood cardioplegic solution (group 1) therefore implies vascular dysfunction resulting from endothelial cell injury. Conversely, the lower CVR with L-arginine, prostaglandins, or WBC filtration implies avoidance of this injury with preservation of vascular function. Because a WBC filter does not cause vasodilatation, the similar CVR in these 3 groups suggests that L-arginine and prostaglandins are acting primarily to preserve vascular function through inhibition of WBCs, and not as vasodilators. This hypothesis remains speculative because specific tests of vascular endothelial cell function were not performed after bypass. However, elevated CVR most likely indicates vascular damage since it correlates with increased production of oxygen-derived free radicals, mitochondrial damage, and poor functional recovery.
We cannot determine whether combining a WBC filter with a pharmacologic agent (L-arginine and prostaglandins) would further improve results, because hearts receiving L-arginine, PGE1, or leukodepletion recovered completely. However, because of differences between these modalities, it is possible that they should be used together. For instance, L-arginine is not as effective when given cold, whereas WBC filters are relatively unaffected by temperature. The optimal dosages of L-arginine and prostaglandins are unknown in human beings under various pathologic conditions; therefore, if a low dose is used for safety, these agents may not be as effective as they were in this study. Some WBCs escape filtration, and leukocyte filters become less efficient with greater volumes or higher flow rates. A chemical (pharmacologic) blocker may help prevent adhesion of these unfiltered leukocytes. L-Arginine and prostaglandins also promote vasodilation, as well as limiting platelet adherence. 1,3,7,13,14,21,25 Increased vascular resistance may limit cardioplegic distribution, and platelet deposition results in capillary plugging with release of potent vasocontrictors. In contrast, the Pall BC-1 leukocyte-depleting filter only slightly reduces platelet counts. The aforementioned factors probably explain why Hiramatsu and associates 2 demonstrated improvement when WBC filtration and L-arginine were combined. Using a pharmaceutical agent along with a WBC filter may therefore improve myocardial protection in cyanotic children. However, because of potential hazards with L-arginine, we would recommend using prostaglandins, especially since they have a similar effect.
In conclusion, L-arginine, PGE 1, and leukocyte filtration all substantially and equally improve myocardial protection by limiting a WBC-mediated injury, which reduces oxygen-derived free radical production and helps to restore endothelial, metabolic, and functional recovery. Because pediatric patients are more sensitive to surgical ischemia, as well as being prone to postbypass myocardial dysfunction, surgeons should consider incorporating these modalities into their protective strategies.
|
Appendix: Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAppendix: DiscussionReferences |
|---|
You used a model of combined subacute hypoxia followed by ischemic stress in piglets. Do you believe these results are applicable to adult mature myocardium and endothelium?
Dr Allen. To make the heart more sensitive to differences in myocardial protection, we subjected piglets to an ischemic and hypoxic stress. We used the combination of ischemia and hypoxia because cyanotic infants often have depressed myocardial ATP levels owing to episodes of ischemia that occur during times of stress. Although our model might not completely simulate the condition of the cyanotic infant, it allows us to test different cardioplegia strategies in vulnerable hearts. This is no different from simulating an ischemic stress in adult studies by crossclamping the aorta. This may not exactly simulate chronic angina, but it subjects the adult heart to a clinically relevant stress. Investigating stressed hearts is extremely important in both adult and pediatric studies, because they are the most difficult to protect. It is my belief that the principles of this study are applicable to both pediatric and adult patients. However, without a definitive investigation, I cannot be certain.
Dr Hofer. I was impressed that the effects on both the CVR and myocardial oxygen consumption are immediate; 1 minute after your induction dose you see an amelioration of these in all 3 groups. This is understandable in the WBC filter groups, since they theoretically have been treated the entire course, before their induction therapy. However, how can you explain this happening so rapidly in the PGE1 and the nitric oxide groups? Does this imply that a very small dose at the beginning of induction is sufficient, as opposed to a longer infusion?
Dr Allen. The initial reintroduction of blood caused a reperfusion injury in the unmodified group, resulting in an increase in both oxygen-derived free radical production and CVR. Conversely, this injury was avoided (low oxygen radical formation) if the cardioplegic solution was enriched with PGE1, L-arginine, or leukodepleted, improving functional recovery. Because oxygen-derived free radical formation and CVR are decreased during cardioplegic induction with L-arginine, PGE1, and leukodepletion, it is possible these agents are not required after the initial period of reperfusion, or during subsequent cardioplegic infusions. However, previous studies suggest they need to be delivered for several minutes to obtain maximal benefit. Furthermore, because these modalities help prevent the reperfusion injury, and hearts are ischemic between cardioplegic infusions, they are probably beneficial during each cardioplegic infusion. None of the groups were pretreated. A WBC filter was used only to filter leukocytes in the cardioplegic solution. Therefore, the only difference between groups was the cardioplegic solution. The PGE 1 concentration was relatively low and equivalent to a systemic dose of 0.2 to 0.4 µg · kg–1 · min–1 , which is one fifth of what is often used to keep a patent ductus open.
Dr Hofer. Since you are concerned that a higher dose of L-arginine is injurious, would an experiment in which you apply only one dose, as opposed to continuing through the whole infusion, theoretically remove those injurious side effects?
Dr Allen. No, I do not believe so. In the previous study (see page 655) we found a marked increase in oxygen radical formation and CVR during cardioplegic induction if the concentration of L-arginine was increased. Therefore the detrimental effects of L-arginine seem to be dependent on the concentration, not the total dose.
Dr Hofer. Would you expect the exact same results if you were to give adenosine as a vasodilator or if you were to block adhesion by another mechanism?
Dr Allen. Yes, I believe the same results could be expected. There are dozens of prostaglandins as well as prostacyclin and adenosine, all of which seem to have the similar actions of limiting WBC and platelet adherence and inducing vasodilatation. We chose PGE1 because it is routinely used in infants, it may be one of the mediators by which nitric oxide works, and several studies have documented its effectiveness in reducing reperfusion injury after ischemia. However, we probably would have had similar results had we used adenosine, prostacyclin, or a variety of other prostaglandins.
Dr Edward Verrier (Seattle, Wash). I am a bit concerned about the conclusions. A number of other cells are involved with hypoxic injuries, including endothelial cells and platelets. Therefore, to say your observations are only mediated by WBCs may be true but unrelated. One simple experiment that could be done would be to add the different components into the same solution and see whether the results change. For instance, one could use either prostaglandin or the WBC filter and see whether the factors are additive in some fashion. The fact that you get the same level with each intervention does not necessarily mean that they are mediated by exactly the same mechanism.
Dr Allen. You are probably right. In addition to inhibiting WBCs, prostaglandins block platelet adherence and induce vasodilatation. These effects are also important in preventing cellular injury after ischemia. Dr Mayer and associates recently demonstrated that using L-arginine and WBC filters together improved myocardial protection much more effectively than using either modality separately. It may be that our model was not severe enough to allow us to demonstrate the added benefits of vasodilatation or decreased platelet adherence. However, the present study does suggest that a major beneficial effect of L-arginine and prostaglandin is due to inhibition of WBC-mediated injury. Platelets and vascular dysfunction definitely play a role in the pathogenesis of the reperfusion injury, and modalities that also block these adverse effects may be superior. We therefore believe that WBC filters and chemical blockers, such as PGE1, should probably be used together.
Dr Verrier. This type of hypoxic stress induces endothelial cells, for instance, to up-regulate a huge number of inflammatory mediators. They do not even get expressed out until 4 hours. Thus the results that you report here are very early-phase results and may be significantly different from what would be seen at 6 hours (in terms of the return of function).
Dr Allen. There may be an up-regulation of inflammatory mediators that will increase the inflammatory response several hours later. Using L-arginine, PGE1, or leukodepletion, however, prevented the initial production of free radicals during reperfusion. By limiting the reperfusion injury, inflammatory up-regulation should at least be partially suppressed. Furthermore, limiting the reperfusion injury may also allow the cell to be less susceptible to any inflammatory mediators several hours later.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAppendix: DiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  B. S. Allen, M. Castella, G. D. Buckberg, and Z. Tan Conditioned blood reperfusion markedly enhances neurologic recovery after prolonged cerebral ischemia J. Thorac. Cardiovasc. Surg., December 1, 2003; 126(6): 1851 - 1858. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. J. Kelly Distant Effects of Experimental Renal Ischemia/Reperfusion Injury J. Am. Soc. Nephrol., June 1, 2003; 14(6): 1549 - 1558. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. S. Allen, J. S. Veluz, G. D. Buckberg, E. Aeberhard, and L. J. Ignarro Deep hypothermic circulatory arrest and global reperfusion injury: Avoidance by making a pump prime reperfusate--A new concept J. Thorac. Cardiovasc. Surg., March 1, 2003; 125(3): 625 - 632. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. S Allen The role of leukodepletion in limiting ischemia/reperfusion damage in the heart, lung and lower extremity Perfusion, March 1, 2002; 17(2_suppl): 11 - 22. [Abstract] [PDF]
|
|
|
|
 |
|
 B. S. Allen Controlled pulmonary reperfusion: what is the optimal method of delivery? Ann. Thorac. Surg., October 1, 2000; 70(4): 1449 - 1450. [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 |
