The importance of conducting relevant animal studies to advance cardioprotection for cardiac surgery

In this issue of the European Journal of Cardio-Thoracic Surgery, Veitinger et al. [1] have conducted cardioplegic arrest studies that extend small animal studies into a more clinically relevant direction. Since the start of cardiac surgery in the 1950s, the induction of elective cardiac arrest has been used by surgeons alongside the development of cardiopulmonary bypass. The early clinical success with elevated potassium citrate (hyperkalemia) in blood (after limited experimental studies by Melrose et al. [2] and termed ‘cardioplegia’ by Lam (as cited in [3])) was, unfortunately, a false dawn. High incidence of cardiac dysfunction and areas of myocardial necrosis lead to increased patient mortality. Cardioplegia was replaced by either normothermic or hypothermic ischaemic arrest or direct coronary perfusion; however, in many patients this resulted in the ‘stone heart’, a fatal irreversible contracture! After Katz and Tada [4] challenged biochemists to solve the problem of the stone heart, the early 1970s saw a revival of interest in hyperkalemic cardioplegic arrest. Collaborative studies by David Hearse (a biochemist) and Mark Braimbridge (a cardiac surgeon) resulted in the development [5] and ultimate clinical introduction [6] of the St Thomas’ Hospital cardioplegic solution, based on an extracellular-type formulation with moderately elevated potassium chloride as the primary arresting agent. Most experimental studies leading to this clinical cardioplegic solution were conducted using isolated perfused healthy rat hearts that mimicked the cardiopulmonary bypass situation of aortic occlusion with ischaemic cardiac arrest using cardioplegia. Since then, there have been hundreds of studies (again, mostly in isolated healthy rat hearts) conducted into cardioprotection using various cardioplegic solutions [3, 7]. The predominant common factor was that these cardioplegic solutions were hyperkalemic (inducing a ‘depolarized’ arrest at a membrane potential of around −50 mV) representing the most effective way to induce rapid arrest of the heart and produce the best cardioprotection. Increasingly, however, recent studies have explored the concept that hyperkalemic depolarized arrest does not induce an optimal arrest and that arrest at a more ‘polarized’ membrane potential (closer to the resting membrane potential of between −85 and −70 mV) will avoid many of the potential damaging effects of depolarization, such as ionic inhomogeneity and consequent maintained energy utilization, and lead to improved protection from a cellular perspective [3, 7]. Agents that are capable of inducing such a polarized arrest include sodium-channel blockers (tetrodotoxin, lidocaine), potassium channel openers (adenosine, pinacidil) and calcium-channel blockers (diltiazem, magnesium); many of these compounds, however, are toxic at the concentrations required to induce cardiac arrest. Recent studies from my group [7–9] have demonstrated the potential of esmolol, an ultra-short-acting β-blocker, as an effective and safe arresting agent that acts to block both the fast sodium channel and the L-type calcium channel and hence to induce a polarized arrest.

The present study [1] describes experiments in isolated perfused rat hearts that confirm the efficacy of esmolol cardioplegia in both crystalloid-based and blood-based solutions in comparison to Calafiore solution, a blood-based hyperkalemic cardioplegia. However, these studies have broadened the clinical relevance of these solutions by also examining their protective efficacy in damaged rat hearts (infarction generated by left anterior descending coronary artery ligation for 30 min prior to arrest and release at the end of the 90-min ischaemic arrest period) to simulate clinical conditions. The study goes some way to fulfil the ‘refinement’ part of the 3 Rs (replacement, reduction and refinement) of animal model studies [10], as it mimics the specific clinical condition of coronary artery bypass surgery. It demonstrates that, in non-infarcted rat hearts, both crystalloid esmolol-based and blood esmolol-based cardioplegic solutions improve cardioprotection compared to the conventional hyperkalemic Calafiore blood-based cardioplegia. However, in the infarcted rat hearts, similar protection was seen with all 3 cardioplegic solutions.

There are some areas of the study that are worthy of comment. One of these relates to the temperature of the hearts during ischaemic cardioplegic arrest. Hearts in the blood-based solution groups (esmolol and hyperkalemic Calafiore) were subjected to a constant temperature of 36°C for the cardioplegia infusion and global ischaemia throughout; in contrast, the crystalloid-based esmolol solution was infused at 32°C for each of the 5 infusions but global ischaemia was maintained at 36°C. This could influence the outcome in 2 ways; either it could lead to improved protection due to the moderate hypothermia of the infusions, or it could reduce the protection due to the constantly changing myocardial temperature throughout the 90 min of ischaemic arrest! It would have been more scientifically correct to maintain the same temperature throughout the ischaemic arrest in all groups, but the fact that both esmolol groups recover to very similar levels may indicate that this is only of minor importance.

Also of interest is the considerably higher levels of left ventricular end-diastolic pressure (contracture) during the ischaemic period in the Calafiore blood cardioplegia groups (non-infarcted and infarcted hearts), compared to the esmolol groups. This may be interpreted as preventing, or reducing, an elevation in intracellular calcium that can be caused by the non-inactivating ‘window currents’ at the membrane potential generated by hyperkalemia [3, 7]. In addition, the higher levels of post-ischaemic coronary flow in the esmolol cardioplegia groups might suggest that esmolol cardioplegia has a more protective effect on coronary endothelium than the hyperkalemic cardioplegia.

To conclude, the study by Veitinger et al. [1] is well-conducted and clinically relevant and presents as an interesting direction for the next generation of cardioprotective research. Like all good studies, many more questions are posed than are answered and hence additional work will be required. It is, however, encouraging that studies such as this continue to demonstrate that ‘polarizing’ arrest has the potential for safe and advantageous myocardial protection in the context of cardiac surgery.

REFERENCES