To compare two different methods of protecting heart muscle from damage caused by a decreased blood supply. Exposure to the anaesthetic agent sevoflurane can allow the heart muscle to resist longer periods of low blood or oxygen supply without sustaining the amount of damage that it would otherwise expect to. The use of thoracic epidural analgesia improves the blood flow to the heart muscle and has also been shown to reduce the amount of damage the heart muscle may otherwise sustain. The aim of this study is to compare these two methods.

Coronary artery bypass grafting (CABG) is a common procedure performed to improve blood flow to the heart in patients with severe ischaemic heart disease. Commonly, the heart has to be stopped to allow this procedure to be performed and this is often achieved with cooling the heart and perfusing it with a solution that stops the activity of the heart muscle. These techniques stop the heart from beating which allows the surgery to be performed and also reduce the oxygen requirements of the heart. This in turn reduces the damage the heart suffers from the reduced blood flow to it which occurs while the arteries are being operated on. Unfortunately, however, these techniques do not completely eliminate the risk of heart muscle damage and so new methods of further reducing damage to the heart are continually being investigated. Two methods relating to the anaesthetic techniques used have recently been identified as potentially of benefit in this regard - the use of volatile anaesthetic agents and the use of high thoracic epidural analgesia.

Volatile anaesthetic agents have been extensively investigated in the past few years with regard to their apparent ability to mimic ischaemic preconditioning. Ischaemic preconditioning refers to the phenomenon that if heart tissue is exposed to frequent, short episodes of reduced blood or oxygen supply, followed by a longer spell, the heart is likely to suffer a smaller area of damage than if it had never been exposed to the brief ischaemic spells. This can be related clinically to the observation that patients with angina, who subsequently suffer a heart attack, have a better prognosis than those patients who suffer a heart attack without ever experiencing angina prior to the event. There have been numerous studies demonstrating that the volatile anaesthetic agents (isoflurane, sevoflurane, desflurane) appear to mimic this phenomenon, both in animal and human models. Numerous studies have demonstrated favourable postoperative blood concentrations of cardiac troponin I (a sensitive marker of heart damage), CK-MB (another marker of heart damage), atrial and brain natriueretic peptides (markers of heart function) compared to those who did not receive preconditioning. It has therefore been suggested that this may improve outcome following cardiac surgery.

High thoracic epidural analgesia (HTEA) has also been shown to be beneficial following cardiac surgery. This involves placing a small catheter near the nerves as they leave the spinal cord. Local anaesthetic ccan be administered down this catheter to numb the areas of the body supplied by these nerves. This provides very good pain relief and is widely used in our hospital for this operation. It has been shown to be associated with a shorter time of required artificial ventilation compared to standard pain relief with drugs such as morphine. It has also been suggested that it may influence outcome with improved heart function following the operation compared to those without HTEA. It appears to do this by improving the blood flow to the heart. It has also been associated with a lower postoperative concentration of cardiac troponin , CK-MB, atrial and brain natriuretic peptides. There has not been any direct comparison of the two techniques, however, to assess if one is superior to the other, or if the benefits of the two techniques are additive.

Comparison:

This study aims to assess if one technique confers more benefit than the other and if the benefits are additive. This is important as many centres do not use thoracic epidural analgesia for cardiac surgery and not all clinicians currently use volatile anaesthetics for cardiac surgery.

Inclusion Criteria:

• Patients presenting for elective CABG operations, with >40 % left ventricular ejection fraction, aged 40-80.

Exclusion Criteria:

• Patients at risk of epidural haematomas (i.e. platelets <100, coagulopathies, on anticoagulant therapies, INR>1.4) or unsuited to HTEA (patient refusal, spinal deformities).

Patients on drugs shown to manipulate the preconditioning phenomenon will also be excluded. This includes those on diazoxide, nicorandil, oral sulfonylureas and theophyllines.

Patients with preoperative ECG morphologies likely to make interpretation difficult or impossible including: left bundle branch block, cardiac pacemaker dependence.

Haemodynamically compromised patients requiring inotropic or balloon pump support preoperatively.

Significant valvular disease. Preoperative elevated levels of troponin I or CKMB. Unstable angina or angina in 24 hours preop