Peri-operative silent myocardial ischaemia

Introduction Silent or symptomatic myocardial ischaemia has been regarded as a risk factor for cardiovascular complications of anaesthesia and surgery for many years. As a result, the prevention of myocardial ischaemia is considered an important goal. Studies carried out more than two decades ago have shown that beta-adrenoceptor blockers reduce the incidence of myocardial ischaemia, probably because they make the circulation more stable [1]. However, data correlating intra-operative myocardial ischaemia and adverse cardiovascular outcome, i.e. myocardial infarction, were lacking until Slogoff and Keats [2] reported a clear association between the severity of myocardial ischaemia and acute myocardial infarction in patients who had undergone coronary bypass surgery. In this study, the authors demonstrated that myocardial ischaemia occurred essentially at times of haemodynamic instability (tachycardia, hypo- or hypertension). Silent myocardial ischaemia was first described by Holter in 1961 [3]. It can occur in patients suffering from angina, previous myocardial infarction or in patients who are completely asymptomatic. In all groups of patients with silent myocardial ischaemia, the risk of cardiovascular events is greater than in patients with coronary artery disease, including those with previous myocardial infarction, who do not suffer from episodes of silent ischaemia. A study by Raby and colleagues [4] has shown that the prognosis in patients with silent ischaemia is worse than in those without: nine out of 10 major cardiac events occurred in 29 patients with silent myocardial ischaemia and only one cardiac event in 71 patients without silent myocardial ischaemia. Why ischaemia should be silent is unclear. In some patients, ischaemia is associated with the increased release of beta-endorphins. This may explain the absence of pain. Recent evidence suggests that autonomic dysfunction may play a role [5]. Predictors for silent ischaemia include a history of previous myocardial infarction, ECG evidence of previous infarction and non-specific alterations in the ECG [6], hypertension and left ventricular hypertrophy [7]. There is also an association between low ejection fraction (less than 40%) and silent ischaemia [6]. Episodes of silent myocardial ischaemia do not appear to differ from symptomatic episodes in terms of their effects on myocardial metabolism and myocardial function. The metabolic effects include reductions in high-energy phosphates, leakage of enzyme (CK-MB) and troponins. Myocardial ischaemia causes a reduction in systolic function (decreased systolic shortening and thickening), an increase in paradoxical wall motion (early systolic lengthening or thinning; post-systolic shortening or thickening) (Fig. 1)[8] and increases in the time constant of isovolumic relaxation and myocardial stiffness. Both the reduction in systolic function and the diastolic dysfunction contribute to the development of acute, ischaemic, left ventricular failure. In addition, both contribute to the worsening of ischaemia caused by a reduction in coronary blood flow (Fig. 2). Acute ischaemia is also responsible for acute dysrhythmias because of re-entry phenomena in the myocardium. Finally, prolonged ischaemia can cause myocardial infarction.Fig. 1: Effects of progressive reductions in coronary blood flow on systolic shortening and post-systolic shortening in a canine model (adapted from Marsch et al. [8]).Fig. 2: Schematic representation of the consequences of myocardial ischaemia in terms of systolic function, diastolic function and myocardial perfusion.Peri-operative silent myocardial ischaemia Ambulatory ECG monitoring has revealed that preoperative silent myocardial ischaemia is common [6], especially in patients presenting for coronary bypass or vascular surgery [9]. Several authors have shown an association between pre-operative silent myocardial ischaemia and postoperative cardiovascular complications of anaesthesia and surgery [10,11]. Raby and colleagues [10] reported that 12 out of 13 major cardiac events occurred in patients with pre-operative silent ischaemia and only one in those without silent ischaemia. As the group with silent ischaemia represented only 20% of the patients studied, the association between silent ischaemia and adverse events was striking. Even more impressive is the association between postoperative silent ischaemia and adverse outcome. Raby and colleagues [12] showed that the odds ratio for adverse cardiovascular events rose from 2.7 (pre-operative ischaemia) to 16 in patients with postoperative silent ischaemia. This is not surprising because the number of episodes of ischaemia, as well as their duration, are greater postoperatively than pre-operatively [13,14]. This increased ischaemic burden is an important feature as Landesberg and colleagues [15] have shown that complications occur most frequently in patients who have suffered from ischaemia for more than 2 h. Even more disquieting is the observation by Mangano and colleagues [16] of the long-term effects of postoperative ischaemia: over the 2 years after surgery, mortality was substantially greater in patients with than in patients without post-operative silent ischaemia. Long-term adverse outcome was worse when the patient had suffered myocardial infarction after their operation. Whether postoperative silent ischaemia is a marker of more severe coronary disease (hence the worse long-term prognosis) or represents a further insult to the myocardium is not known. Intuitively, both factors may contribute to the long-term adverse outcome, and it is tempting to postulate that prevention and/or treatment of perioperative silent ischaemia may improve outcome. This hypothesis is supported by the known temporal association between duration of ischaemia and cardiac events [15], as well as the temporal association between silent ischaemia and adverse outcome [12]. It is well established that cardiovascular complications of anaesthesia and surgery occur predominantly during the second, third and fourth post-operative days [17,18]. This is also the time at which silent myocardial ischaemia is more frequently observed, and episodes of ischaemia are of longer duration [9]. Causes of peri-operative myocardial ischaemia The increased frequency and duration of ischaemia may have multiple causes. The association between cardiac aberrations such as tachycardia, hypertension, hypotension, left ventricular dilatation and myocardial ischaemia is well established [2]. It is also well documented that, during the postoperative period, heart rate is substantially faster than pre-operatively [9], and that episodes of silent ischaemia are frequently, but not always, associated with tachycardia [2]. The extent of the adverse role of an increase in heart rate can also be inferred from studies of outcome after myocardial infarction. Heart rate above 120 beats per minute increases the mortality of infarction by a factor of 7.5 [19]. An imbalance in autonomic activity has been reported during the early postoperative period, attested by reduced heart rate variability [20]. This imbalance may alter coronary vascular tone and allow vasoconstriction to occur if dynamic stenoses are present. After surgery, changes in both coagulation and fibrinolysis are known to occur. The increase in coagulation factors peaks after 7-14 days, while the nadir of fibrinolysis occurs at 1-5 days after surgery. Not surprisingly, this is associated with an increased risk of deep venous thrombosis and arterial occlusion. These changes in coagulation may be responsible for postoperative myocardial ischaemia and/or infarction. Indeed, it is increasingly recognized that transient occlusion followed by lysis of small clots may be responsible for myocardial ischaemia. Role of postoperative hypoxaemia Another important factor is hypoxaemia. The advent of pulse oximetry and computerized data collection has made it possible to obtain continuous records of arterial oxygen saturation after surgery and in patients with sleep disorders. Severe episodic desaturation can occur in the early postoperative period. Hypoxaemia is most likely to appear during sleep, sometimes because of central or obstructive apnoea. Nocturnal hypoxaemia occurs frequently in patients recovering from major surgery. Indeed, in the absence of administration of supplementary oxygen, post-operative hypoxaemia is very common, and the lowest level of hypoxaemia is noted during the second night, while episodic hypoxaemia is more common during the third night [21]. When supplementary oxygen is given according to perceived clinical needs, very low saturations may occur during the night after surgery of the abdominal aorta. However, the most dramatic episodes of desaturation are likely to occur during the third or fourth postoperative nights as, at that time, supplementary oxygen is unlikely to be given routinely. In 1991, we reported a temporal association between hypoxaemia and silent myocardial ischaemia in three patients who had undergone surgery of their abdominal aorta [22]. In each patient, administration of supplementary oxygen corrected their saturation and suppressed the signs of ischaemia. In these patients, postoperative tachycardia was a persistent phenomenon. This study was extended to 34 patients who had undergone major vascular surgery and revealed the high incidence and considerable severity of postoperative hypoxaemia, especially during the third and fourth postoperative nights [23]. Most patients were receiving supplementary oxygen for the first two nights, but oxygen use was much more variable on the following three nights. Not surprisingly, the proportion of the night during which saturation was below 90% was highest during the third and fourth rather than first and second nights. However, episodes of hypoxaemia were still observed until the fifth postoperative night. It is clear that, without supplementary oxygen, patients undergoing major abdominal vascular surgery are at risk of severe hypoxaemia until the fifth night after their operation. The pathogenesis of postoperative hypoxaemia is unlikely to be uniform. In the early postoperative phase, it is likely to be caused by the reduction in functional residual capacity (FRC), consequent upon anaesthesia and surgery, and the development of atelectasis in the dependent lung regions [24]. The respiratory depressant effect of anaesthesia and opioids contributes to the early postoperative hypoxaemia. Later, hypoxaemia is more likely to relate to persistent reductions in FRC, impaired diaphragmatic function and additional splinting of the diaphragm by abdominal distension and pain. The site of the surgical incision is an important factor, as well as the extent of the transection of the abdominal muscles and the severity of the postoperative pain [25]. In our patients, the pattern of periods of relatively stable baseline hypoxaemia and repetitive cyclic desaturation of short duration was identical to that reported in patients with sleep apnoea. While we did not record the EEG, others have investigated the association between REM sleep and hypoxaemia. The highest incidence of hypoxaemia may coincide with the rebound of REM sleep, as an association between REM sleep, sleep-disordered breathing and episodic hypoxaemia has been observed during the second and third post-operative nights [26]. Not surprisingly, disordered patterns of breathing have also been reported after thoracotomy, ranging from absence of hypoxaemia, hypoxaemia stable and improving, stable and constant, and unstable and deteriorating [27]. Clearly, the development of laparoscopic surgery may reduce the incidence of postoperative hypoxaemia, as it can be shown that laparoscopic cholecystectomy, for example, is associated with smaller reductions in vital capacity and forced expiratory volume than conventional cholecystectomy [28]. That nocturnal respiratory disorders and hypoxaemia are associated with adverse outcome is strongly suggested by the circadian variation in unexpected postoperative deaths. The study of Rosenberg and colleagues [29] showed that 13 out of 18 unexpected deaths occurred during the night. Postoperative hypoxaemia has been shown to be associated not only with myocardial ischaemia and cardiac arrhythmias but also with mental confusion, surgical delirium and poor wound healing. That post-operative delirium is caused, at least in part, by hypoxaemia can be inferred from the efficacy of supplementary oxygen reported by Aakerlund and Rosenberg [30]. Hypoxaemia may also exert an adverse effect on wound healing and resistance to bacterial infection [10]. The association between nocturnal hypoxaemia and cardiovascular accidents is not unique to the perioperative period. There is abundant evidence for an association between disordered breathing and cardiac events. Disordered breathing is common during sleep in patients suffering from coronary artery disease, even in those who do not have associated symptomatic pulmonary disease. The disorder consists of obstructive apnoea or Cheyne-Stokes breathing. The episodes of disordered breathing can be as frequent as 20 times per hour and are associated with marked reductions in saturation [31]. In patients with chronic obstructive airway disease, the low basal mean saturation is made worse by episodes of more severe hypoxaemia. During such episodes, atrial or ventricular ectopic activity, tachycardia, prolongation of Q-T interval and S-T depression have been reported and are corrected by oxygen therapy [32]. In addition to hypoxaemia, sleep apnoea caused by upper airway collapse or, less commonly, by absent central respiratory drive is associated with severe arterial hypoxaemia, modest hypercarbia, systemic hyper- or hypotension, and cyclic acceleration and slowing of heart rate. These effects in themselves may precipitate myocardial ischaemia in patients with coronary artery disease and, in the longer term, together with hypoxaemia, may accelerate the progression of atherosclerosis. Indeed, sleep apnoea has been found to be more severe in male survivors of acute myocardial infarction than in male subjects without evidence of coronary artery disease. Both average and minimum saturations were lower in patients with acute myocardial infarction, and these patients spent longer periods of time with saturation below 90% [33]. Similarly, a case-control study has revealed that snoring is a strong risk factor for acute myocardial infarction [34]. At the time of acute myocardial infarction, sleep apnoea is common and is particularly prominent in patients with a low cardiac index. The episodes of sleep apnoea are associated with desaturation in 40% of patients. In these patients, rhythm disturbances are frequently observed, confirming the association between hypoxaemia and adverse cardiac events [35]. In the case of obstructive apnoea, the adverse effects on the heart result from the association between hypoxaemia and periodic variations in both heart rate and blood pressure Management of peri-operative hypoxaemia As postoperative hypoxaemia and the associated myocardial ischaemia increase the risk of the complications of anaesthesia and surgery, the question of its prevention is a major issue. It can be argued that better methods of pain relief should minimize the risk of hypoxaemia. The epidural route of administration of opioids is popular, effective, and its effects on oxygenation have been studied. Even in patients with normal pre-operative saturation, hypoxaemia may occur postoperatively at a time when patients are receiving opioids intramuscularly or epidurally [36]. While epidural opioids usually provide better pain relief, as assessed by pain scores, than other modalities of opioid administration, they are usually associated with more hypoxaemia, particularly when given as continuous infusions [37]. Because the extent of postoperative nocturnal hypoxaemia is variable, it would be useful to be able to predict which patients are most at risk for severe desaturation. In this respect, the mean overnight saturation is a useful predictor, while the awake saturation is not [38]. As cardiovascular complications (myocardial infarction, life-threatening arrhythmias, acute left ventricular failure and cerebrovascular accidents) often occur after episodes of desaturation associated with prolonged episodes of ischaemia [14], the role of monitoring saturation and myocardial ischaemia during the postoperative period must be considered. The increasing availability of pulse oximeters and of ECG monitors with which ST segment trends can be detected may offer new avenues for the detection of both arterial hypoxaemia and myocardial ischaemia. If the risk of complications relates to the severity and duration of both hypoxaemia and silent ischaemia, detection of hypoxaemia and ischaemia within a short period of their development may be sufficient, as 'instantaneous' detection is logistically impossible outside the intensive therapy or high-dependency unit. If periods of hypoxaemia and ischaemia of less than 15 min are much less likely to cause major cardiovascular events than more prolonged episodes, then monitors with the appropriate alarms could be used on ordinary wards provided protocols were available for the management of hypoxaemia and ischaemia. Drug prevention of postoperative myocardial ischaemia As a study carried out in vascular surgical patients showed that peri-operative silent myocardial ischaemia was associated with a long-term reduction in survival [16], intense research activity has been directed at the prophylaxis of peri-operative myocardial ischaemia with a view to reducing the incidence of immediate postoperative cardiovascular complications and of improving the long-term outcome of patients with coronary artery disease, especially those undergoing coronary revascularization or peripheral vascular surgery. Beta-blockers have been shown to be effective in decreasing the incidence of myocardial ischaemia in the immediate peri-operative period, whether given intravenously at the time of induction of anaesthesia or for 3 days before anaesthesia and surgery [1]. Similarly, the administration of a single dose of a beta-blocker given with the premedication has been shown to decrease the risk of peri-operative ischaemia, arguably by preventing tachycardia [39]. When the administration of a beta-blocker is maintained for 1 week, the reduction in the incidence of myocardial ischaemia is also maintained [40]. More importantly, this mode of administration of a beta-blocker has been shown to improve the 2-year survival of the patients so treated [41]. As alpha2-adrenoceptor agonists are known to make the circulation more stable and to decrease the release of catecholamines, it is not surprising that the alpha2-adrenoceptor agonist mivazerol has been shown to minimize the risk of intra-operative and emergence ischaemia [42]. This may be associated with reduced peri-operative morbidity. Another approach is the use of drugs that act on the coronary circulation itself. Adenosine modulators enhance the release of adenosine by the ischaemic but not by the normal myocardium. The adenosine modulator acadesine has been shown to reduce the risk of cardiac death and of postoperative myocardial infarction in a recent meta-analysis of several studies of this adenosine modulator [43]. Thus, some classes of drugs may be able to reduce both the incidence of silent myocardial ischaemia and the risk of both early and late cardiovascular complications (including cardiac death) of anaesthesia and surgery in patients with coronary heart disease or with factors of risk for coronary artery disease. In conclusion, silent myocardial ischaemia is a common, multifactorial complication of anaesthesia and surgery. To minimize its incidence and the associated cardiovascular risks, it is essential to ensure haemodynamic stability throughout the peri-operative period. It is also important to prevent the development of hypoxaemia, particularly during the first few postoperative nights. Of the cardiovascular drugs used in the prevention of peri-operative ischaemia, alpha2-adrenoceptor agonists and adenosine modulators have shown promise. Beta-blockers have been shown to be effective, and their peri-operative administration deserves to be considered in patients with coronary heart disease or with risk factors for this condition. This view has been clearly stated in recent guidelines issued by the American Heart Association [44].