| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2004;77:226-232
© 2004 The Society of Thoracic Surgeons
a Second Department of Cardiac Surgery, Katowice, Poland
b Department of Cardiology, Medical University of Silesia, Katowice, Poland
* Address reprint requests to Dr Deja, Second Department of Cardiac Surgery, Medical University of Silesia, Ul. Ziolowa 47, 40-635 Katowice, Poland.
e-mail: narizol{at}slam.katowice.pl
Presented at the Thirty-ninth Annual Meeting of The Society of Thoracic Surgeons, San Diego, CA, Jan 31–Feb 2, 2003.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
METHODS: The isolated electrically driven human right atrial trabeculae, obtained during cardiac surgery, were studied. The tissue bath was oxygenated with 95% oxygen and 5% carbon dioxide, hypoxia being obtained by replacing oxygen with argon. The influence of diazoxide on atrial contractility was studied first. Next, the two trabeculae from one atrial appendage were studied simultaneously, adding diazoxide to the tissue bath 10 minutes before hypoxia in one, with another serving as a control. We tested 10-4.5 mol/L and 10-4 mol/L diazoxide in three sets of experiments testing 30, 60, and 90 minutes of hypoxia. We continued reoxygenation for 120 minutes (in 60-minute and 90-minute hypoxia experiments) and subsequently tested reaction to 10-4 mol/L norepinephrine. Apart from continuous recording of the contraction force, we measured the troponin I release into the tissue bath after ischemia and reoxygenation.
RESULTS: Diazoxide exerted a negative inotropic effect in human atrial muscle (pD2=3.96 ± 0.18). Both concentrations of diazoxide studied resulted in better functional recovery of atrial trabeculae subjected to 30 minutes of hypoxia. With longer hypoxia, only the higher diazoxide concentration provided significant protection as assessed by contractility. After 120 minutes of reoxygenation, only diazoxide-treated muscle was viable enough to respond to norepinephrine. Only 10-4 mol/L diazoxide resulted in lower troponin I release during hypoxia and reoxygenation.
CONCLUSIONS: This study shows that diazoxide provides significant concentration-dependent protection against hypoxia/reoxygenation injury to human myocardium in vitro.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Whether the mitochondrial KATP channel is the effector or mediator of preconditioning, it plays an important role in cardioprotection [11]. Importantly, it may be an attractive target for eliciting the protection against ischemia in human myocardium. The basic, selective mitochondrial KATP channel activator, used in most laboratory studies examining the role of mitochondrial KATP channels in preconditioning [3–13] (diazoxide), had been used for years in humans for the treatment of hypertensive crisis and hyperinsulinemia.
It has been shown to confer significant myocardial protection in many experiments [3–13], and it has been proposed as a possible adjunct to myocardial protection during cardiac surgical procedures [15–19]. Nevertheless, diazoxide use for myocardial protection during cardiac surgery in real patients has not been reported yet. Before it can be used clinically, its influence on the contractility of human myocardium has to be established. It is important to find what level of protection can be afforded by diazoxide in human myocardium and what concentration is necessary to provide significant myocardial protection in human tissue. These are the questions addressed by the present study.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
The tissue was transferred in ice-cold Krebs-Henseleit solution to the laboratory. The single trabecula less than 1 mm in diameter was mounted in the organ chamber (Schuler Organbath, Hugo Sachs Elektronik, March-Hugstetten, Germany [HSE]) containing Krebs-Henseleit solution of the following composition [mol/L]: NaCl, 118.0; KCl, 4.70; CaCl2, 2.52; MgSO4, 1.64; NaHCO3, 24.88; KH2PO4, 1.18; glucose, 5.55; sodium pyruvate, 2.0 (pH 7.4). It was oxygenated with carbogen (95% oxygen, 5% carbon dioxide) and maintained at 37°C. The trabecula was driven with 1 Hz, 50 ms square stimuli using platinum field electrodes and the potential of 150% of the threshold for given preparation. The stimulator type 215 (HSE) was used. The contraction force was measured with F30 isometric force transducer type 372 (HSE). The signal was enhanced with bridge amplifier type 336 (HSE) and recorded with Graphtec Linearcorder WR 3320 (HSE).
The trabecula was gradually stretched to seek for the optimal preload according to the Frank-Starling relationship. Once the maximal contraction force was achieved, the preparation was relaxed to 90% of optimal tension and left for 30 minutes of stabilization. During this period, it was thoroughly washed.
In the first set of experiments (n = 20), the influence of diazoxide (Sigma Chemical Co, St. Louis, MO) on myocardial contractility was assessed. Diazoxide was dissolved in dimethyl sulfoxide (DMSO) (Sigma Chemical Co). The concentration of DMSO in tissue bath was always kept below 0.4%. We checked that this concentration of DMSO did not influence the myocardial contractility. Diazoxide was applied in increasing concentrations, starting with 10-8 mol/L and increasing in negative logarithm half-molar cumulative steps up to 10-3 mol/L to establish the concentration-effect relationship. The contractility after diazoxide was expressed as a percentage of the initial contraction force, and presented at each concentration level as a mean ± SEM. The concentration-effect relationship was obtained from a regression analysis to the general logistic equation of Michaelis and Menten (1913):
 |
During subsequent experiments, the hypoxia was simulated by substituting oxygen in carbogen with argon (95% argon, 5% carbon dioxide). This resulted in the drop of tissue bath oxygen partial pressure from 475 ± 52 mm Hg to 51 ± 1.8 mm Hg (p < 0.001) and caused significant and rapid impairment of muscle inotropism. On reoxygenation, the carbogen (95% oxygen, 5% carbon dioxide) was added again and the tissue was washed several times.
The two trabeculae from one atrial appendage were always studied simultaneously, adding diazoxide to the tissue bath 10 minutes before simulated hypoxia in one, with another serving as a control. We tested 10-4.5 mol/L (31.6 µmol/L) and 10-4 mol/L (100 µmol/L) diazoxide.
Six separate sets of experiments (10 patients each) were performed testing two different diazoxide concentrations added 10 minutes before 30, 60, and 90 minutes of hypoxia respectively. In the experiments with a longer (60 or 90 minutes) hypoxia period, we kept the reoxygenation period for 120 minutes and subsequently added 10-4 mol/L norepinephrine [(-)-arterenol bitartrate (Sigma Chemical Co)] to test for stunning. We recorded the force of contraction continuously. We compared the recovery of contraction force after 40 minutes of reoxygenation. This was the time after which contractility achieved plateau and before it started declining again. In experiments with 60 and 90 minutes of hypoxia, we additionally analyzed the contraction force after 120 minutes of reoxygenation and after adding norepinephrine.
For each preparation, the contraction force recorded was compared with its initial contraction force (after stabilization period) and expressed in percent, the initial contraction force being 100%.
To allow for troponin I accumulation, the tissue bath was not replaced throughout the hypoxia period, and apart from the initial washout, it was not replenished throughout the reoxygenation as well. The sample of tissue bath was obtained for troponin I concentration measurement after the hypoxia period and after 120 minutes of reoxygenation (not in the 30-minute hypoxia experiments). The troponin I concentration was measured using the immunoenzymatic method (Diagnostic Resort Group International Co, Mountainside, NJ) and expressed in nanograms per milligram wet tissue weight.
Statistical analysis
We first used two-way repeated measures (RM) analysis of variance (ANOVA) with the time of hypoxia as one factor and the concentration of diazoxide (10-4 vs 10-4.5 vs 0 mol/L [control]) as another. Because two-way RM ANOVA showed influence of the hypoxia time on contractility, we analyzed separately the influence of diazoxide concentration on the outcomes (both contractility and troponin I release) in experiments with different hypoxia times using one-way RM ANOVA. We used the Holm-Sidak method for the post hoc multiple comparison procedure. We considered p less than 0.05 significant. SigmaStat for Windows Version 3.0 (SPSS Inc, Cary, NC) was used for statistical analysis.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
|
The contraction force recovered after 40 minutes of reoxygenation depended not only on the length of hypoxia (p < 0.001) but also on the concentration of diazoxide (p < 0.001), as shown by two-way RM ANOVA. We analyzed separately the recovery of contractility in experiments with different hypoxia times using one-way RM ANOVA with post hoc testing (Fig 2). Both concentrations of diazoxide allowed significantly better recovery of contractility after 30 minutes of hypoxia and 40 minutes of reoxygenation (65% ± 7% vs 47% ± 5%; p = 0.002; and 62% ± 4% vs 45% ± 6%; p = 0.013). With 60 minutes of hypoxia, the recovery was significantly improved by 10-4 mol/L diazoxide. In experiments with 90 minutes of hypoxia, the contractility of atrial muscle treated with 10-4 mol/L diazoxide was not only higher than the control (56% ± 9% vs 24% ± 3%, p < 0.001) but also on average by 30% higher (p < 0.001) than in the muscle treated with a lower (10-4.5 mol/L) diazoxide concentration.
|
|
|
|
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Having established the diazoxide influence on human atrium contractility, we chose to study the myocardioprotective effects of its two different concentrations. We elected to study the influence of 10-4.5 mol/L (approximately 30 µmol/L) and 10-4 mol/L (100 µmol/L) of diazoxide for several reasons. First, these very concentrations were used in various laboratory and animal studies before. Second, the authors generally used either the concentration of 30 µmol/L [14, 15, 19, 20] or 100 µmol/L [9, 12–13], or concentrations in between the two [8, 16–18, 25], and there is little comparison between the cardioprotective action of them. In the work of Garlid and associates [3], both doses possessed similar cardioprotective effect in isolated rat hearts, and the authors elected to use 30 µmol/L diazoxide in their further experiments. On the other hand, Liu and associates [4] studied the cardioprotection and opening of mitochondrial KATP channels mediated by diazoxide in rabbit ventricular myocytes. They showed that at concentrations up to 100 µmol/L, diazoxide dose-dependently opened the mitochondrial without activating the sarcolemmal KATP channel. K1/2 value was 27 µmol/L, suggesting marked cardioprotective benefit of increasing the diazoxide concentration above this value. Last but not the least, 100 µmol/L diazoxide caused 50% contractility drop in our experiments (EC50), whereas 30 µmol/L caused only little (approximately 20%) decrease in contractility. Therefore, we considered it important, especially in view of potential clinical use, to find whether increasing diazoxide concentration would confer significant cardioprotective benefit.
Our experimental model of myocardial hypoxia and reoxygenation is based on always studying a pair of atrial trabeculae from the same patient. This is necessary, as we observed a different level of myocardial contractility suppression with hypoxia and a different level of recovery with reoxygenation in different patients. It was not surprising, as the patients differed from the point of view of not only exact cardiac pathology but also from the point of view of their medication, demographics, and so on. We have reported previously, for instance, that the atrial trabeculae of patients with different disease states differ considerably in their inotropic responses [24]. To study cardioprotection with diazoxide, we elected not to use tissue from diabetic patients, but we decided that further exclusion criteria would not be practically feasible. After all, one could never achieve the situation close enough to the "clean" laboratory animal experiment. We confirmed, nevertheless, in earlier sham experiments that the results analyzed pairwise were consistently showing functional (contractility) and structural (troponin I release) deterioration in preparations subjected to ischemia compared with preparations oxygenated throughout the experiment.
To be able to assess the troponin I release to the tissue bath, time was needed to allow it to leak from the tissue to the Krebs-Henselait solution and to accumulate. Therefore, we did not analyze troponin I release in the 30-minute hypoxia experiments. For the same reason, we extended the reoxygenation period to 120 minutes and we did not replace the tissue bath throughout this time. This is in contrast to atrial trabecula being continually superfused during the "reperfusion" period in the very similar experimental model by Yellon and associates [20]. Yellon's group concentrated, however, on functional recovery only, and no attempt to collect the indicator enzymes was made.
Our study showed significant myocardial protection against hypoxia and reperfusion conferred by diazoxide. The recovery of contractility after 40 minutes of reoxygenation did not depend on diazoxide concentration if the trabeculae were subjected to the short period of hypoxia. However, a clear advantage of using higher diazoxide concentration was visible when assessing the functional recovery after longer, particularly 90-minute, hypoxia. The longer reoxygenation period resulted in a gradual loss of contractile function, probably related to reoxygenation injury. However, 10-4 mol/L diazoxide protected, to some extent, against this injury.
The functional recovery of human atrial trabeculae from hypoxia/reoxygenation insult in our study should be compared with the results of the already cited Yellon's group study, performed in a very similar model [20]. They studied the classic ischemic preconditioning and the role of protein kinase C and KATP channels. The functional recovery of the human atrial trabeculae from 90 minutes of simulated ischemia in their study was of the same level as the recovery we observed with 10-4 mol/L of diazoxide. However, they did not observe the decline of myocardial function over the 120-minute "reperfusion" period. As mentioned earlier, their preparation was continually superfused during the "reperfusion," whereas we did not replace the bath solution to allow for troponin I accumulation. This might have enhanced the reoxygenation injury in our experiments.
Second, the tissue bath solution used in Yellon's laboratory contained a high concentration (10 mmol/L) of pyruvate. Pyruvate has been shown to possess cardioprotective properties [21–22], particularly against the reperfusion injury. And, indeed, when we used a high pyruvate level Krebs-Henseleit solution in our laboratory, it enhanced recovery (data not published). Therefore, we resigned from using it so as not to confound the results.
To test whether the deteriorating contractility during the reoxygenation period depended on the loss of cellular integrity or on the increasing stunning, we added norepinephrine to the tissue bath. The reversibility of contractility loss in human atrial trabeculae under positive inotropic challenge fulfils the definition of stunned myocardium [23]. We found that all diazoxide-protected trabeculae responded with an increase of contraction force, whereas no response in control muscle was observed. This finding would suggest that the function decline with reoxygenation depended on structural damage in control group, and on growing stunning in diazoxide-treated group. In other words, the diazoxide did not prevent stunning development, but it preserved structural integrity necessary for inotropic response to norepinephrine. We draw no conclusion from the magnitude of the response to norepinephrine, as we have shown previously in our laboratory that this may vary from patient to patient [24]. However, the rise in contractility of muscle treated with 10-4.5 mol/L diazoxide and subjected to 90 minutes of hypoxia failed to reach statistical significance compared with the control. This confirms once more the benefit of using the higher diazoxide concentration.
Although we observed the tendency to lower troponin I release after diazoxide treatment in both hypoxia and reoxygenation in all experiments, the 90-minute ischemia period was necessary to make differences statistically significant. Perhaps 60-minute ischemia was too short to allow for significant troponin I accumulation in tissue bath. Still, in 90-minute hypoxia experiments, the differences were significant only with higher diazoxide concentration. This parallels the results of functional recovery in 90-minute hypoxia experiments.
Based on the above, we conclude that diazoxide protected human atrial myocardium from hypoxia/reoxygenation injury in a concentration-dependent fashion. It both improved functional recovery and maintained cellular integrity. This is opposite to earlier reports showing that mitochondrial KATP channel opening reduces infarct size but has little, if any, effect on functional recovery [16–17, 26], unless we speculate that diazoxide might have other points of action in human myocardium, different from mitochondrial KATP channel [26].
The last issue to address is the difference between our results and the results obtained in study of Pomerantz and associates [14]. They found that subjecting the myocardium to diazoxide for a short period before ischemia provided significant protection, whereas leaving the diazoxide in the tissue bath throughout the ischemic period resulted in loss of protective effect. However, the authors studied 30 µmol/L diazoxide only. Perhaps with 10-4 mol/L diazoxide, the authors would observe the protective effect similar to our results.
We chose to leave the diazoxide in the tissue bath for the period of ischemia to mimic the situation in which the diazoxide is one of the components of cardioplegic solution. In this case, it would be infused at the beginning of (or hopefully with blood cardioplegia before) the ischemic period. It would, with a bit of luck, stay in coronaries in between the infusions, and would be replenished with maintenance cardioplegia doses. There are data from other authors suggesting that diazoxide given in cardioplegia, and even reaching the myocardium only at the beginning of reperfusion, may be beneficial for myocardial preservation [16–19]. We suggest that to achieve maximal cardioprotection, the concentration of diazoxide reaching the myocardium should be rather high (ie, at the level of 10-4 mol/L [100 µmol/L]. It may cause a significant negative inotropic effect, but at least in vitro, this effect easily disappears with washout. Even with this high diazoxide concentration, the total diazoxide dose given to the patient with one initial and two maintenance doses of cardioplegia would be 60 mg, five times less than the bolus intravenous dose given systemically for the treatment of hypertension.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
DR DEJA: We actually hoped to show that diazoxide prevented stunning to some extent. There were physiological studies done previously using this kind of tool (inotropic stimulation for finding or proving the stunning). To our surprise, we found a significant response to norepinephrine in diazoxide-treated muscle, which means that the muscle was simply stunned when losing the contractile force over 120 minutes of reoxygenation, and no response at all or no significant response in the muscle not treated with diazoxide, which, in our interpretation, together with the troponin level, means that probably the structural damage to the muscle was big enough not to allow for any significant inotropic response. Yes, we think it is a very interesting finding. Unfortunately, we have not found a group where there was a smaller response to norepinephrine in myocardium not treated with diazoxide and a bigger response in treated, which probably would look even more elegant.
DR TOMASZ TIMEK (Redlands, CA): Thank you for the presentation. It was beautiful. The question I had was, could you comment on the mechanism of the negative effect of diazoxide before hypoxia, and is that effect also dose-dependent?
DR DEJA: Yes, it is dose-dependent. As you have probably noticed, there is around 10% to 15% drop of contractile force with the lower concentration (10-4.5 mol/L) and almost 30% to 40% contractile force drop with the higher concentration (10-4 mol/L) of diazoxide. Now, it is very difficult to comment on it, and what we are planning to do is to try and exclude the possibility that diazoxide was actually acting via the sarcolemmal channel at this high concentration. To do this, we need a sarcolemmal ATP-dependent potassium channel blocker. In the experiments I know, there is no proof that diazoxide in such a concentration can work on sarcolemmal channels. At least Gross and Garlid's group studies show that with this concentration and with the relative affinity to sarcolemmal channels being 2,000 times less than to the mitochondrial, one should not expect it. So, the only thing I can think of is the sarcolemmal channel being involved, and we obviously need to test for it.
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
activation in the mechanism of preconditioning. Am J Physiol Heart Circ Physiol 2002;283:H440-447.This article has been cited by other articles:
|
|
 |
|
  M. A. Deja, M. Malinowski, K. S. Golba, M. Kajor, T. Lebda-Wyborny, D. Hudziak, W. Domaradzki, D. Szurlej, A. Bonczyk, J. Biernat, et al. Diazoxide protects myocardial mitochondria, metabolism, and function during cardiac surgery: A double-blind randomized feasibility study of diazoxide-supplemented cardioplegia J. Thorac. Cardiovasc. Surg., April 1, 2009; 137(4): 997 - 1004. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Deja, K. S. Golba, M. Malinowski, K. Widenka, J. Biernat, D. Szurlej, and S. Wos Diazoxide Provides Maximal KATP Channels Independent Protection if Present Throughout Hypoxia Ann. Thorac. Surg., April 1, 2006; 81(4): 1408 - 1416. [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 |
= EC50 ± SEM (n = 20).
