Ann Thorac Surg 1997;63:1003-1011
© 1997 The Society of Thoracic Surgeons
Original Article: Cardiovascular
Effect of Desferrioxamine Cardioplegia on Ischemia-Reperfusion Injury in Isolated Rat Heart
Sonia C. Nicholson, MA,
Marian Squier, MBBS,
David J. P. Ferguson, DSc,
Zsuzsanna Nagy, DPhil,
Stephen Westaby, MS,
Rhys D. Evans, DPhil
Nuffield Department of Anaesthetics and Department of Neuropathology, University of Oxford, Radcliffe Infirmary, and Departments of Electron Microscopy and Cardiothoracic Surgery, John Radcliffe Hospital, Oxford, England
Accepted for publication October 24, 1996.
 |
Abstract
|
|---|
Background. The protective effect of desferrioxamine against myocardial ischemia-reperfusion injury remains uncertain. In this study we examined a broad range of ischemia-reperfusion indices to determine the effect of desferrioxamine cardioplegia in a model that reflects surgical practice.
Methods. Isolated rat hearts were subjected to 90 minutes of ischemia with cold cardioplegia, with or without 0.5-mmol/L desferrioxamine. Left ventricular mechanical function and the levels of thiobarbituric acid–reactive substances and nonprotein thiol compounds were measured after reperfusion. Electron microscopic analysis of mitochondria was performed using diaminobenzidine staining, together with histochemical staining for glycogen and marker enzymes in left ventricular muscle and the atrioventricular node.
Results. The desferrioxamine group showed better preservation of diastolic function (chamber stiffness coefficient at 15 minutes and maximum rate of decrease of left ventricular pressure at 45 minutes of reperfusion). Histochemical analysis showed that mitochondria-specific succinate dehydrogenase and the nonspecific esterase of the atrioventricular node were better preserved in the desferrioxamine group.
Conclusions. The findings from this study indicate that there is added protection against ischemia-reperfusion injury when desferrioxamine is added to the cardioplegic solution; however, the study also highlighted that, in this clinically applicable model, desferrioxamine is not universally protective against all aspects of ischemia-reperfusion injury.
|
Introduction
|
|---|
There is substantial evidence for the generation of oxygen-derived free radicals during reperfusion after ischemia, as shown both by direct measurement [1, 2] and indirectly by the prevention of damage by free radical scavengers [2–7]. However, the precise role and clinical relevance of these free radicals in myocardial ischemia-reperfusion injury remain uncertain. The hydroxyl radical is one of the most cytotoxic of the oxygen-derived free radicals generated during reperfusion. It causes a chain reaction, starting with lipid peroxidation, which in turn oxidizes membrane lipids and thereby damages the cell wall. This leads to changes in membrane permeability, produces electrolyte and cell volume disturbances, and alters the function of membrane-bound proteins and mitochondria. It is probable that oxygen-derived free radicals are produced in the mitochondria. Ischemia decreases the ability of the mitochondria to detoxify them [8], and mitochondria are consequently particularly sensitive to ischemia-reperfusion injury.
Iron chelation prevents the generation of hydroxyl radicals from superoxide radicals by means of the Fe(II) to Fe(III)–driven Fenton reaction. In addition, the iron chelator desferrioxamine has a direct role in scavenging superoxide radicals, thereby decreasing the generation of hydroxyl radicals from superoxide radicals and hydrogen peroxide by means of the Haber-Weiss reaction. Desferrioxamine has been used to prevent reperfusion injury, but the findings have been variable. Some studies have shown that desferrioxamine brings about an improvement in systolic function [4, 9], but others have shown no such change [7]. Some have shown it produces less damage to mitochondrial structure [1, 5, 10] and function [3]. A decrease in the incidence of arrhythmias in association with desferrioxamine use was reported by Bernier and associates [6], but it was observed to produce no change in infarct size in rabbits [11]. Although these experiments were conducted in a variety of species with different timing of the administration of desferrioxamine, there is no correlation between either of these factors.
The value of desferrioxamine in clinical practice remains unclear. Because of advances in cardiac surgery that have led to shorter myocardial ischemic times, the present investigation was conducted to assess a much wider range of outcomes in greater depth than previous studies have, including the more sensitive indices of diastolic function (maximum rate of decrease of left ventricular pressure [-dP/dtmax], chamber stiffness coefficient, and time constant of relaxation) not previously studied in this context. This was done using a protocol of mild ischemia-reperfusion injury applicable to the clinical setting.
|
Material and Methods
|
|---|
The investigation was performed in accordance with the UK Home Office Guidance on the Operation of the Animals (Scientific Procedures) Act 1986, published by HMSO, London UK.
Chemicals
Lactate dehydrogenase was obtained from Boehringer Corps, Lewes, Sussex, UK; 5,5'-dithiobis(2-nitrobenzoic acid) and sulfosalicylic acid were obtained from Aldrich Chemical Co Ltd, Gillingham, Dorset, UK. Other chemicals were obtained from Sigma Chemical Company, Poole, Dorset, UK.
Isolated Heart Model
Male Wistar rats weighing 300 to 450 g were anesthetized with intraperitoneal sodium pentobarbitone (60 mg/kg body weight). The hearts were rapidly excised, briefly cooled in ice-cold Krebs-Henseleit buffer, and then perfused through the aorta using a modified Langendorff apparatus. The perfusion solution consisted of a modified Krebs-Henseleit bicarbonate solution at 37°C containing calcium chloride (1.3 mmol/L) and glucose (10 mmol/L) and oxygenated with 95% oxygen and 5% carbon dioxide. Perfusion pressure was maintained at 100 cm H2O. The first 50 mL of 250 mL of perfusate was discarded to free the circuit of red blood cells, and the remaining 200 mL was recirculated through a 5-µm filter. A right ventriculotomy was performed to allow free drainage of the coronary sinus, and the left ventricle was vented by apical puncture. A latex balloon, which was slightly larger than the left ventricle, was inserted through the mitral valve and connected by means of a polyethylene catheter and three-way tap to a saline-filled high-precision syringe (Hamilton syringe, model 1001 TLL; Hamilton Co, Reno, NV) and to a pressure transducer (Elcomatic EM751; Elcomatic Ltd, Neilston, Glasgow, UK). Epicardial pacing wires were placed in the right ventricle and the heart paced at 240 beats/min. The balloon volume was adjusted to achieve an end-diastolic pressure of between 5 and 10 mm Hg (V10). The heart was surrounded by a water jacket maintained at 37°C.
The ventricular pressure signal was digitalized using an analog-digital converter (AT-MIO-16; National Corporation, Austin, TX) and continuously displayed on the screen of an IBM AT personal computer by means of a real-time mode of software designed in this department. The software permitted electronic differentiation of the pressure signal. For each recording, data were downloaded at a sampling frequency of 500 Hz and stored on hard-disk for off-line analysis.
Experimental Protocol
Only those hearts were studied that developed systolic pressures of greater than 80 mm Hg after a stabilization period of 30 minutes (total, 28 hearts). At the end of this period, baseline measurements were made of the coronary flow rate at volume V10 and the left ventricular pressure trace downloaded for analysis of variables. Samples were collected for measurement of coronary venous lactate concentrations. The left ventricular balloon was then emptied until the end-diastolic pressure was 0 mm Hg and increased in 20-µL increments until the end-diastolic pressure was 35 mm Hg. The left ventricular pressure trace was recorded 10 seconds after each volume change to prevent hysteresis and the effect of stress relaxation. Pacing was then stopped, the balloon emptied, the temperature of the heart chamber water jacket decreased to 14°C, and the hearts perfused via the aortic cannula with 20 mL of St. Thomas' Hospital cardioplegic solution (containing 2-mmol/L calcium, 1-mmol/L procaine, and 20-mmol/L potassium chloride) at 4°C, at an infusion pressure of 80 mm Hg. The cardioplegic solution in the test group contained 0.5-mmol/L desferrioxamine. After a 90-minute period of cold global ischemia (no aortic perfusion), the water jacket temperature was increased to 37°C, the aortic line opened, and the pacing restarted. Measurements were repeated at 15, 30, and 45 minutes of reperfusion.
The apex of the heart was then excised and placed in ice-cold phosphate buffer for tissue assays, and the remaining heart tissue snap-frozen in isopentane cooled in liquid nitrogen for histochemical staining. Hearts for electron microscopy were perfused with 4% gluteraldehyde in 0.1-mol/L phosphate buffer at 4°C, removed from the cannula, and placed under fixative. The left ventricular apex was removed, cut into 1-mm cubes, and returned to the 4% gluteraldehyde solution at 4°C.
Mechanical Data Analysis
The left ventricular pressure wave and its first differential were analyzed for the peak systolic pressure, left ventricular end-diastolic pressure, the maximum rate of increase of left ventricular pressure (+dP/dtmax), and the -dP/dtmax, and the time constant of isovolumic relaxation, 
, calculated. The latter was calculated using the method of Weiss and associates [12]. The slope of the natural logarithm of the left ventricular pressure (lnP) versus time (t) relation (from the peak -dP/dt to the point when pressure declined to the level of the end-diastolic pressure) was obtained by the least-squares linear regression: lnP = At + c, where A is the slope and c the intercept of the lnP versus t relation; was defined as the negative inverse of that slope, = -1/A.
The chamber stiffness constant (Kp) was calculated from the relation between the left ventricular end-diastolic pressure (EDP) versus volume (V): lnEDP = KpV + B, where B is the intercept of the slope.
The average value of at least three beats at each stage were used for analysis and was calculated in duplicate.
Lactate Assay
Samples of coronary effluent were deproteinized in 6% (weight per volume) perchloric acid. After neutralization the lactate concentration was assayed according to the method of Hohorst [13] using lactate dehydrogenase.
Tissue Assays
The levels of thiobarbituric acid–reactive substances (TBARS) were measured by the method of Ohkawa [14]. Absorbance values were measured at 515-nm excitation and 553-nm emission by fluorometric spectrophotometry (LS-3 Fluorescence Spectrometer; Perkin-Elmer Ltd, Beaconsfield, Buckinghamshire, UK) and the TBARS levels calculated from a fresh 1,1,3,3-tetraethoxypropane standard. The levels of nonprotein thiol compounds were measured using a modification of the method of Ellman [14]. The absorbance was measured at 412 nm (Lambda 3 UV/VIS Spectrophotometer; Perkin-Elmer Ltd) and the levels of nonprotein thiol compounds calculated using 13.6 x 10-3•mol-1•cm-1 as the molar extinction coefficient.
Histochemistry
Longitudinal serial sections were cut at 5 µm parallel to the interventricular septum and stained with hematoxylin-eosin (to identify the atrioventricular [AV] node), periodic acid–Schiff (for glycogen), and with histochemical methods to identify the activities of nonspecific esterase, cytochrome oxidase, and succinate dehydrogenase. Automated densitometric analysis was performed using the VIDAS 21 computer-assisted image analysis system (Kontron Electronics, GmbH, Munich, Germany).
The optical density of the AV node and ventricular muscle on the periodic acid–Schiff–stained and histochemical preparations was measured. To avoid the problem of a differing depth of stain on each slide, the density was expressed as the ratio of the optical density of the region of interest to the optical density of the connective tissue on the same slide. The characteristic color (red-green-blue) distribution and intensity were also defined for each stain. The extent to which these specified variables were present in the node and in the ventricular muscle was taken as the measure of the amount of the biochemical marker studied and was expressed as the proportion of the area examined. By using the two techniques, the problem of increased optical density posed by the staining of areas other than cytoplasm (such as nuclei) was minimized.
Electron Microscopy
The blocks of tissue initially fixed in 4% gluteraldehyde were postfixed in 2% (weight per volume) osmium tetroxide, dehydrated in ethanol, treated with propylene oxide, and embedded in Spurr's epoxy resin. Thin sections were stained with uranyl acetate and lead citrate and examined with a Jeol 1200EX electron microscope (Jeol Corp, Tokyo, Japan).
3,3'-Diaminobenzidine (DAB) staining was performed as described by Mack and colleagues [15]. The gluteraldehyde-fixed tissue blocks were cut under fixative into 100-µm slices with a vibrotome. The first 0.3 mm of tissue from each cube was discarded and random slices incubated in imidazole buffer with 1 mg/mL of DAB. Subsequent processing was done as already described.
Statistics
Results are expressed as the mean ± the standard error of the mean. Mechanical function, biochemical assay results, and histochemistry findings were all analyzed using unpaired t tests. Statistical significance was set at a p value of less than 0.05.
|
Results
|
|---|
Cardiac Mechanical Function
There was no significant difference between the baseline values of the two groups. There was also no significant difference in the coronary flow rates between the two groups at any time point during the reperfusion period, and both groups showed the expected reactive coronary dilatation during early reperfusion (Fig 1
). Left ventricular diastolic function (Fig 2) showed an initial deterioration after reperfusion, with a subsequent return toward the baseline value; this effect was partly modified by the presence of desferrioxamine in the cardioplegic solution, as shown by the fact that the chamber stiffness constant in early (15 minutes) reperfusion and the -dP/dtmax in late (45 minutes) reperfusion were normalized in the hearts exposed to desferrioxamine (Fig 2). When systolic function was examined (Fig 3), both the systolic pressure and contractility (+dP/dtmax) were found to be higher at each time point in the presence of desferrioxamine than in controls, although the difference was not statistically significant.
View larger version (37K):
[in this window]
[in a new window]
|
Fig 1. . Effect of desferrioxamine on coronary flow rate after 15, 30, and 45 minutes of reperfusion, expressed as a percentage of the baseline value (B). Baseline values are given in milliliters per minute. Results are expressed as the mean ± the standard error of the mean (n = 11 to 15).
|
|
View larger version (24K):
[in this window]
[in a new window]
|
Fig 2. . Effect of desferrioxamine on left ventricular diastolic function after 15, 30, and 45 minutes of reperfusion. Values are expressed as the percentage of the baseline value (B). Baseline values are given in millimeters mercury per second for the maximum rate of decrease of left ventricular pressure (maximum negative dP/dt), milliseconds for the time constant of relaxation, and milliliters-1 for the chamber stiffness coefficient. Results are expressed as the mean ± the standard error of the mean (n = 7 for each group). (*p < 0.05, as tested by Student's t test.)
|
|
View larger version (32K):
[in this window]
[in a new window]
|
Fig 3. . Effect of desferrioxamine on systolic function after 15, 30, and 45 minutes of reperfusion. Values are expressed as the percentage of the baseline value (B). Baseline values are given in millimeters mercury (systolic pressure) and millimeters mercury per second (maximum rate of increase of left ventricular pressure [maximum positive dP/dt]). Results are expressed as the mean ± the standard error of the mean (n = 7 for each group). No significant difference was observed between groups.
|
|
Coronary Venous Lactate Production
The coronary venous lactate production increased from baseline levels in both groups during reperfusion. This was significant in the desferrioxamine-treated group at 15 minutes and the control group at 30 minutes. However, there was no significant difference between the two groups at any time point examined (Fig 4).
View larger version (28K):
[in this window]
[in a new window]
|
Fig 4. . Effect of desferrioxamine on coronary venous lactate production per gram of wet weight at baseline (B) and after 15, 30, and 45 minutes of reperfusion. Results are expressed as the mean ± the standard error of the mean (n = 7 for each group). ( p < 0.05 compared with baseline values, as shown by Student's t test; no significant differences between the two groups were observed.)
|
|
Biochemical Assays of Heart Tissue
There were no significant differences in the tissue levels of non–protein-associated reducing equivalents, as estimated by the levels of nonprotein thiol compounds (control group [n = 7], 1,472 ± 196 µmol/g heart; desferrioxamine-treated group, 1,522 ± 218 µmol/g heart), and although there was a tendency for the tissue lipid peroxide levels, (as estimated by the TBARS levels: control group, 55.6 ± 6.0 nmol/g heart; desferrioxamine-treated group, 39.9 ± 3.9 nmol/g heart) to be decreased, this did not reach statistical significance (p = 0.06).
Histochemistry
Initial periodic acid–Schiff staining showed good delineation of the AV node and ventricular muscle (Fig 5), permitting assessment of the glycogen content and orientation for further staining.
View larger version (157K):
[in this window]
[in a new window]
|
Fig 5. . Photomicrographs of rat heart showing atrioventricular node (arrow). Staining with periodic acid–Schiff was done after cold cardioplegic ischemia with desferrioxamine and reperfusion. Well-preserved glycogen stores are seen. (A, x46; B, x230; both before 33% reduction.)
|
|
Optical density measurements indicated a tendency for the activity of all enzymes studied in the group treated with desferrioxamine to be increased compared with the levels in the control group in both the conducting pathway (AV node) and the ventricular muscle. This was statistically significant for nonspecific esterase (Table 1).
View this table:
[in this window]
[in a new window]
|
Table 1. . Ratio of Optical Density of Atrioventricular Node or Ventricular Muscle to Connective Tissue for Each Histochemical Staina
|
|
The sections were further examined using a technique that assessed the proportion of the region occupied by the characteristic color of the reaction product of a particular enzyme. As expected, this showed less succinate dehydrogenase in the AV node than in the myocardium. The activity of succinate dehydrogenase in ventricular muscle, as determined by this method, was better preserved in the desferrioxamine-treated group than in the control group; however, the cytochrome oxidase level in the AV node was decreased in the desferrioxamine-treated group compared with that in control hearts (Table 2).
View this table:
[in this window]
[in a new window]
|
Table 2. . Proportion of Region Staining for Chosen Color in Connective Tissue, Atrioventricular Node, and Ventricular Musclea
|
|
Electron Microscopy
The ultrastructural appearance of several muscle fibers from each sample were examined (D.J.P.F., who was blinded to the treatment received). The myocytes in certain biopsy specimens appeared normal and had well-preserved mitochondria. In these cases all the cells appeared similar (Fig 6A). The mitochondria of the cardiac muscle cells in other biopsy specimens, however, appeared abnormal, with varying degrees of swelling and disruption of the cristae. The swollen mitochondria appeared more electron lucent (Fig 6B). In these samples all the mitochondria of all cells appeared damaged but the degree of swelling varied between adjacent cells. Because of this variability it was not possible to grade the degree of damage between samples. Therefore they were rated as either damaged or undamaged. There was no evidence of tissue edema in the sections except where the mitochondria were severely damaged. One of six hearts treated with desferrioxamine had poorly preserved mitochondria compared with two of the six in the control group (Table 3).
View larger version (0K):
[in this window]
[in a new window]
|
Fig 6. . Transmission electron micrographs of rat cardiac myocytes after routine processing (A, B) or diaminobenzidine staining (C, D, E). (A) A well-preserved myocyte is seen adjacent to a capillary showing well-preserved mitochondria (Mi) with an electron-dense matrix and normal cristae. (Bar = 0.5 µm.) (En = endothelial cell.) (B) Part of a poorly preserved myocyte showing swollen electron-lucent mitochondria (Mi) with disrupted cristae. (Bar = 1 µm.) (C) Low-power view of part of myocyte with well-preserved mitochondria (Mi) showing lipid droplets (arrows) with electron-dense peroxidase product around their periphery. (Bar = 1 µm.) (D) Detail of two lipid droplets in which the electron-dense peroxidase product can be seen around the periphery (arrowheads). (Bar = 0.5 µm.) (E) Detail of the myocyte cytoplasm showing two densely stained peroxisomes (arrows) surrounded by mitochondria. (Bar = 0.5 µm.)
|
|
There was strong staining of the few small peroxisomes in sections treated with DAB (Fig 6E
). The myocytes in certain hearts contained numerous small lipid droplets, and in these cases, reaction product was found around the periphery of the lipid droplet (Figs 6C, 6D). In no case was there any evidence of increased mitochondrial density resulting from DAB staining. Using a semiquantitative examination of DAB staining (+, some; ++, moderate; +++, most), it was found that there was a tendency for the control hearts to show more DAB staining than those treated with desferrioxamine (see Table 3). Hearts perfused for a similar period but not subjected to any cardioplegia or ischemia showed some (+) DAB staining and morphologically well preserved mitochondria (data not shown).
|
Comment
|
|---|
This study showed that desferrioxamine confers added protection on the rat heart against ischemia-reperfusion injury when added to cold cardioplegic solution. This is seen in the better preservation of diastolic function and enzyme activity.
To mimic the surgical setting, we used cold ischemia for 90 minutes, whereas others have used warm or normothermic ischemia [3, 7, 9, 10] or longer periods of cold ischemia [4]. Injury was clearly demonstrated during early reperfusion, indicating the adequacy of this protocol to cause ischemic damage.
The desferrioxamine concentration was chosen after an assessment of the concentrations used in previous studies, including that of DeBoer and Clark [7], who showed a maximum protective effect in rat heart at a concentration of 0.5 mmol/L and toxicity at a concentration of more than 0.76 mmol/L.
The diastolic function of the heart, studied as the -dP/dtmax, chamber stiffness coefficient, and time constant of relaxation, has been shown to be a sensitive marker of myocardial ischemia-reperfusion damage [16]. Our results showed better preservation of both early (-dP/dtmax) and late (chamber stiffness coefficient) diastolic function in the desferrioxamine group. The changes in systolic function during reperfusion noted in the present study were less marked than those in diastolic function, and a significant difference was not seen in this experimental protocol. However, in general agreement with our findings of better-preserved mechanical function, Menasché and colleagues [4] showed some improved preservation of +dP/dtmax as well as compliance in rat hearts given desferrioxamine in the cardioplegic solution, and Ambrosio and associates [9] showed a greater recovery of developed pressure in rabbit hearts given desferrioxamine at reperfusion (70% of baseline) compared with controls (35% of baseline). The developed pressure in the control group of the present study decreased to 80% of baseline after 45 minutes of reperfusion indicating that the damage to systolic function produced by ischemia-reperfusion in this model may not have been sufficiently severe to be significantly improved by desferrioxamine; other groups have been unable to show an improvement in mechanical function in rat [7] or rabbit [9, 17] hearts exposed to desferrioxamine either before ischemia or at the time of reperfusion.
The TBARS assay is a standard technique for estimating peroxide activity and was used to assess lipid peroxidation despite its known limitations, thus permitting comparison with the DAB staining of tissue peroxidation seen on electron microscopy studies. Conditions for the TBARS assay were chosen to achieve optimal sensitivity [11]. The assay has been criticized mostly on the grounds that malondialdehyde is generated in the presence of acid and heat, pigmented samples are overread, and other aldehydes are included in the assay. Our samples were all subjected to identical conditions; standards were used throughout and showed a high degree of reproducibility. The level of TBARS in the treatment group was less than that in the control group, and this approached significance (p = 0.06). We previously showed a 50% decrease in the TBARS level in human atrial tissue when desferrioxamine was added to the cardioplegic solution [2].
The DAB studies, with a tendency to show less overall staining in the desferrioxamine-treated group than in the control group, suggest support for these findings. Peroxidation was seen in the form of a dense stain in peroxisomes and in some of the lipid droplets present. Evaluation of lipid peroxidation was difficult, however, because fewer lipid droplets are seen in rat myocardium (present study) than in rabbit myocardium [15]. No increase in the mitochondrial density was seen in the DAB-stained tissue, in contrast to the findings in rabbit [15] and dog [18] heart after ischemia-reperfusion. Even nonischemic controls showed some DAB staining, possibly as a result of ischemia at the time of cardiac excision; the desferrioxamine-treated group would therefore be likely to demonstrate at least an equivalent amount of staining.
Coronary venous lactate production indicates the occurrence of cytosolic lactate production. Although the levels increased after ischemia, this was only transient, indicating a washout effect and the presence of sufficient cellular integrity to allow a return to aerobic metabolism.
The preservation of reducing equivalents, as estimated by the level of nonprotein thiol compounds (predominantly reduced glutathione), in both groups is further evidence for the mild degree and clinically applicable nature of this model and emphasizes the importance of the improved diastolic function in the desferrioxamine-treated group. However, it raises a question regarding the cause of the poor diastolic function in the control group.
Succinate dehydrogenase, a sensitive mitochondrial marker enzyme, was better preserved in the myocardium of hearts treated with desferrioxamine than in controls. This was not seen for cytochrome oxidase, the terminal component of the mitochondrial respiratory chain and a good indicator of mitochondrial function. However, Yanagiya [19] suggested that decreases in cytochrome oxidase activity lag behind the ultrastructural changes resulting from the ischemia-reperfusion damage to mitochondria and that therefore the precise timing may be critical in the assessment of injury.
Mitochondrial morphology was generally less well preserved in the control group than in the desferrioxamine-treated group, although one heart in the desferrioxamine-treated group showed mitochondrial damage. We were unable to grade the degree of damage between poorly preserved and well-preserved mitochondria, as has been described by other groups [1, 5] studying tissue and mitochondrial damage. It has been suggested that the morphologic damage to mitochondria seen immediately after reperfusion recovers with an increasing duration of reperfusion, and it may be that this change was missed in the samples obtained at 45 minutes. However, the clinical importance of the transient changes in morphology that resolve after 60 minutes of reperfusion is questionable. In contrast, Badylak and associates [10] showed better preservation of mitochondria after 60 minutes of reperfusion in hearts exposed to desferrioxamine, although only one heart from each group was examined in this study. Better preservation of mitochondrial morphology was also shown by Ferreira and colleagues [1] in human hearts and Karwatowska-Prokopczuk and co-workers [5] in rat hearts. Similarly, mitochondrial respiration and phosphorylation have been noted to be improved when desferrioxamine (0.76 mmol/L) was given before ischemia or at reperfusion [5].
We elected to study the AV node, because it is more resistant than the myocardium to damage caused by ischemia-reperfusion and direct exposure to reactive oxygen species [20]. The AV node and His-Purkinje system have a relatively high glycogen content and fewer mitochondria than ventricular myocardium, indicating both a high metabolic rate and an ability to maintain anaerobic glycolysis for limited periods. The nodal glycogen levels, as estimated by periodic acid–Schiff staining, were not better preserved in the desferrioxamine-treated group, although the nonspecific esterase activity was. However, computerized analysis of the periodic acid–Schiff–stained tissue is technically difficult because the nuclei stain darkly, making the optical density measurement difficult to interpret because the nodal cells are smaller and hence have a greater number of nuclei for a given area. It is also likely that cytosolic glycolytic pathways are less sensitive to lipid peroxidation damage than is mitochondrial oxidation-respiration, which is highly membrane dependent. Thus desferrioxamine may be expected to have less effect on cytosolic anaerobic metabolism and glycogen status.
The better preservation of nodal nonspecific esterase found in the present study could explain the decrease in reperfusion arrhythmias reported by Bernier and associates [6], although the mechanism suggested by that group was decreased damage to membrane pumps and hence less electrolyte disturbance. Lurie and colleagues [21] suggested that electrophysiologic changes in the subendocardial Purkinje fibers may be responsible for postreperfusion arrhythmias, and the current study supports this in its finding of better-preserved conducting tissue.
In summary, this study highlights the variability in the effects of desferrioxamine when a wide variety of parameters of ischemia-reperfusion injury are examined. This variability may in part be due to different sensitivities of the biochemical processes to ischemia-reperfusion injury and their modification by desferrioxamine. Desferrioxamine has been shown to decrease some aspects of ischemia-reperfusion injury, but this study did not examine the mechanism of this action (iron chelation, nonspecific inhibition of lipid peroxidation, or superoxide radical scavenging). It also emphasizes the importance of the experimental design.
|
Acknowledgments
|
|---|
Sonia C. Nicholson, MA, was supported financially by Cardiac Surgical Sciences Trust Fund, University of Oxford.
|
Footnotes
|
|---|
Address reprint requests to Dr Evans, Nuffield Department of Anaesthetics, University of Oxford, Radcliffe Infirmary, Woodstock Rd, Oxford, OX2 6HE England.
|
References
|
|---|
- Ferreira R, Burgos M, Milei J, et al. Effect of supplementing cardioplegic solution with deferoxamine on reperfused human myocardium. J Thorac Cardiovasc Surg 1990;100:708–14.[Abstract]
- Drossos G, Lazou A, Panagopoulos P, Westaby S. Deferoxamine cardioplegia reduces superoxide radical production in human myocardium. Ann Thorac Surg 1995;59:169–72.[Abstract/Free Full Text]
- Van Jaarsveld H, Potgieter GM, Kuyl JM, Barnard HC, Barnard SP. The effect of desferal on rat heart mitochondrial function, iron content, and xanthine dehydrogenase/oxidase conversion during ischemia-reperfusion. Clin Biochem 1990;23:509–13.[Medline]
- Menasché P, Grousset C, Gauduel Y, Mouas C, Piwnica A. Prevention of hydroxyl radical formation: a critical concept for improving cardioplegia. Circulation 1987;76(Suppl 5):180–5.
- Karwatowska-Prokopczuk E, Czarnowska E, Beresewicz A. Iron availability and free radical induced injury in the isolated ischaemic/reperfused rat heart. Cardiovasc Res 1992;26:58–66.[Abstract/Free Full Text]
- Bernier M, Hearse DJ, Manning AS. Reperfusion-induced arrhythmias and oxygen-derived free radicals. Circ Res 1986;58:331–40.[Abstract/Free Full Text]
- DeBoer DA, Clark RE. Iron chelation in myocardial preservation after ischemia-reperfusion injury: the importance of pretreatment and toxicity. Ann Thorac Surg 1992;53:412–8.[Abstract]
- Ferrari R, Ceconi C, Curello S, Cargnoni A, Medici D. Oxygen free radicals and reperfusion injury; the effect of ischemia and reperfusion on the cellular ability to neutralise oxygen toxicity. J Mol Cell Cardiol 1986;18(Suppl 4):67–9.[Medline]
- Ambrosio G, Zweier JL, Jacobus WE, Weisfeldt ML, Flaherty JT. Improvement of postischemic myocardial function and metabolism induced by administration of deferoxamine at the time of reflow: the role of iron in the pathogenesis of reperfusion injury. Circulation 1987;76:906–15.[Abstract/Free Full Text]
- Badylak SF, Simmons A, Turek J, Babbs CF. Protection from reperfusion injury in the isolated rat heart by postischaemic deferoxamine and oxypurinol administration. Cardiovasc Res 1987;21:500–6.[Medline]
- Kikugawa K, Kojima T, Kosugi H. Major thiobarbituric acid–reactive substances of liver homogenate are alkadienals. Free Radic Res Commun 1990;8:107–13.[Medline]
- Weiss JL, Frederiksen JW, Weisfeldt ML. Hemodynamic determinants of the time-course of fall in canine left ventricular pressure. J Clin Invest 1976;58:751–60.
- Hohorst HJ. L-(+)-lactate. In: Bergmeyer HU, ed. Methods of enzymatic analysis. New York: Academic, 1963:215–9.
- Lapenna D, Porreca E, Del Boccio G, et al. Regional distribution of glutathione-related antioxidant differences in the normal rabbit aorta. Life Sci 1991;48:2225–8.[Medline]
- Mack CP, Brosamer KM, Shlafer M. Ultrastructural demonstration of peroxidative activity and peroxidation in ischaemic and ischaemic-reperfused rabbit hearts. Cardiovasc Res 1993;27:371–6.[Abstract/Free Full Text]
- Bonow RO, Bacharach SL, Green MV, et al. Impaired left ventricular diastolic filling in patients with coronary artery disease, assessment with radionuclide angiography. Circulation 1981;64:315–23.[Abstract/Free Full Text]
- Maxwell MP, Hearse DJ, Yellon DM. Inability of desferrioxamine to limit tissue injury in the ischaemic and reperfused rabbit heart. J Cardiovasc Pharmacol 1989;13:608–15.[Medline]
- Vaughan DM, Koke JR, Bittar N. Ultrastructure, peroxisomes and lipid peroxidation in reperfused myocardium. Cytobios 1988;55:71–80.[Medline]
- Yanagiya N. Ultrastructural and histochemical changes of mitochondria in global ischemic cardiac muscle of rat. Cell Mol Biol 1994;40:1151–64.
- Tribulova N, Slezak J, Ravingerova T, Styk J, Gabauer I. Vulnerability of specialized conductive tissue to ischemia and reperfusion related injury. Bratisl Lek Listy 1991;92:81–90.[Medline]
- Lurie KG, Argentieri TM, Sheldon J, Frame LH, Matschinsky FM. Metabolism and electrophysiology in subendocardial Purkinje fibers after infarction. Am J Physiol 1987;253:H662–70.[Abstract/Free Full Text]
Top
Abstract
Introduction
