Myocardial conditioning techniques in off-pump coronary artery bypass grafting

Off-pump coronary artery bypass surgery by avoiding cardioplegic arrest seems to reduce the risk of ischemic myocardial injury. However, even short-term regional ischemic periods, hemodynamic instability and arrhythmias associated with the procedure can be responsible for myocardial damage. Conditioning, a potential cardio-protective tool during on-pump cardiac surgery, has hardly been investigated in the context of off-pump surgery. There are virtually no large trials on remote ischemic preconditioning and the majority of reports have focused on central ischemic conditioning. Similarly, volatile anesthetic agents with conditioning effect like ischemic preconditioning have been shown to reduce cardiac injury during on-pump procedures but have not been validated in the off-pump scenario. Here, we review the available evidence on myocardial conditioning, either with ischemia/reperfusion or volatile anesthetic agents in patients undergoing off-pump coronary artery surgery.


Review
Conditioning is an umbrella definition that consists of pre-conditioning, per-conditioning and post-conditioning [1]. Conditioning can be elicited remotely (e.g. at the level of a limb) [2] or centrally (e.g. at the level of the heart) [3]. Stimulus normally consists of an ischemic period followed by reperfusion [4], but other triggers like pain-stimulus [5], hyperbaric oxygen [6] and most importantly methods such as volatile anesthetics [7] have been advocated as being equally effective. Conditioning has been experimented in cardiac surgery mainly in the form of remote ischemic preconditioning (RIPC) and the majority of the largest trials have been in on-pump CABG patients (ONCAB) [8] or during valve surgery [9]. RIPC seems to lowers troponin release in patients undergoing ONCAB ('proof of concept'), but still unclear is if this leads to any better clinical outcome. RIP Heart-Study will recruit over 2000 on-pump patients and will include as primary outcomes all-cause mortality, non-fatal MI, any new stroke and/or acute renal failure [10] and the ERICCA [11] trial will establish if there would be better clinical outcome in on-pump high-risk patients. However, it is still an open question whether conditioning has potential benefits in patients undergoing OPCAB in terms of lowering troponin release and improving clinical outcomes. Ischemia during OPCAB can happen in 10% of cases [12] with ST segment elevation in up to 40% of patients [13]. Although the use of intracoronary shunts can reduce the ischemic time, even shortterm regional ischemic periods or heart manipulations can result in myocardial injury [14], and subsequent arrhythmias and/or hemodynamic instability which can lead even to conversion to ONCAB [15].
Several reports on the use of volatile anesthetics cardio-protective agents in ONCAB surgery have suggested a similar ischemic preconditioning effect. However, only small trials have investigated the effect of anaesthetic preconditioning in OPCAB. The use of alternative methods such as adenosine, hyperbaric oxygen or temperature in OPCAB is also yet to be validated. Optimizing protection of the heart from ischemia/reperfusion injury (I/R) during OPCAB is, therefore, a worthwhile goal.

Ischemic conditioning in OPCAB
Although there are evidences supporting the benefits of ischemic preconditioning (IP) in terms of cardio-protection, its adoption in OPCAB has been relatively limited (Table 1) with just few studies published so far.
Malkowsky et al. [13] induced IP with 5 minutes of local coronary artery occlusion and 5 minutes of reperfusion in a series of 17 single vessel OPCAB patients. Left anterior descending artery (LAD) occlusion/reperfusion increased left ventricle wall motion score (LVWMS) and pulmonary pressure (PA). However, subsequent ischemia during the construction of the anastomosis did not alter the regional LV systolic function. van Aarnhem et al. [12] retrospectively reviewed a cohort of 200 OPCAB patients in whom ischemic preconditioning was used before anastomosis by occluding the vessel for 5 minutes and then allowing 5 minutes of reperfusion. They reported a 10% incidence of-intra operative ischemia (defined as 1 mm ST segment elevation) but no perioperative myocardial injury. Intra-coronary shunts were used in case of critical ischemia and no conversion to ONCAB was observed. Laurikka et al. [14] showed that IP, induced with two cycles of 2 minutes left anterior descending artery occlusion (LAD) followed by 3 minutes of reperfusion before the first coronary artery anastomosis, decreased the immediate myocardial enzyme release, reduced the post-operative increase in HR, and enhanced the recovery of volume index after surgery. Doy et al. demonstrated that a single cycle of 5 minutes of central I/R attenuated ischemia-induced electrophysiological changes in patients undergoing MIDCAB [23]. Wu et al. [22] reported that IP, induced by to cycles of 2 minutes occlusion of LAD followed by reperfusion, led to a positive suppression of HR and reduced the incidence of supra-ventricular and ventricular arrhythmias, although the incidence of post-operative atrial fibrillation (AF) remained similar between preconditioned and non-preconditioned groups. Drengen et al. [21] compared prospectively a cohort of patients preconditioned with a single cycle of 5 minutes of LAD occlusion followed by 5 minutes of reperfusion and a cohort preconditioned with 1.6% enflurane with a non-preconditioned group. They reported a significant reduction of metabolic deficit in both the IP and volatile groups compared to the non-preconditioned group [21]. Hong et al. [19] carried out a prospective controlled randomized trial on RIPC in patients undergoing OPCAB. Remote ischemic preconditioning was elicited with 4 cycles of 5 min upper limb ischemia and 5 min of reperfusion after anesthesia. Troponin release was lower in the RIPC group but did not reach statistical significance (p = 0.172) [19]. Succi et al. [20] reported that IP, induced only by 1 minutes of LAD occlusion followed by 2 minutes of reperfusion before anastomosis prevented decrease in left ventricular contractility Hong et al. [18] used both RIPC and postconditioning (PostC) in a cohort of 35 OPCAB patients. Ischemic remote preconditioning and PostC were elicited with 4 cycles of 5 minutes of ischemia and 5 minutes of reperfusion of the lower limb before and after anastomosis. They reported a significant decrease in post-operative myocardial enzymes. Forouzinna et al. [17] prospectively compared patients preconditioned either with adenosine or with 2 cycles of 2 minutes of LAD occlusion followed by reperfusion with a non preconditioned control group. They reported no differences in terms of post-operative EF and enzyme release but the incidence of post-operative AF was higher in the IP group although it did not reach statistical significance. A recent pilot study by Jung et al. focusing on neurological outcome did not show any benefits of RIPC in patients undergoing OPCAB in terms of cognitive outcomes [16].

Volatile anesthetic agents conditioning in OPCAB
Several reports investigating the potential use of volatile anesthetics as preconditioning agents in OPCAB (Table 2) have reported rather different results. Sevoflurane was found to reduce troponin release in three trials [25][26][27], although Orriach et al. used sevoflurane as well as postconditioning agent, extending its administration during the first post-operative hours [25]. No differences in troponin release were reported by other authors when sevoflurane was compared to desflurane and propofol [28], to propofol in a remifentanil-based anesthesia regime [29,30] and to isoflurane alone [31,32]. Sevoflurane was also showed to have several other benefits: better antioxidative properties than propofol [33], reduced incidence of arrhythmias compared to desflurane [34], better preservation of cardiac function compared to control and to propofol [35,36] and to reduced NBP release and plasma protein A release when compared to propofol [37]. Isoflurane was reported to reduce troponin release when compared to propofol [38] but no differences were found when compared to sevoflurane [31,32]. Other authors on the contrary did not find differences in terms of troponin release when it was compared to a propofol group of patients [39]. Desflurane was reported to improve LSWI when compared to a propofol anesthetic regime, although no differences in troponin release were observed [40], and moreover was found to be less effective in reducing incidence of arrhythmias than sevoflurane [34]. Guarracino et al. observed a significant reduction in troponin release in a cohort of 57 patients anesthetized with desflurane [41]. Finally, remifentanil was found to have a preconditioning effect by lowering troponin release [42], however no differences were reported when was associated with sevoflurane and compared to propofol [30].

Alternative way to standard conditioning in OPCAB
There have been few reports on the use of adenosine as preconditioning agent in OPCAB. Forouzinna et al. [17] reported no differences between preconditioned and non Thirty two OPCAB IP induced with twice 2 min LAD occlusion followed 3 min reperfusion before the first anastomosis Incidence of post-operative arrhythmias IP suppressed the HR elevation during the time of myocardial ischemia and reperfusion and significantly reduced the incidence of VT after surgery. Incidence of SVT during 2 to 24 hours after surgery was lower in the IP patients but incidence of SVES, VES, and AF were similar between the 2 groups Arrhythmia was a common phenomenon during and after OPCAB procedure; IP protocol significantly suppressed HR elevation, episodes of SVT, and incidence of VT after surgery but incidence of post-op AF was similar in between groups Temporary segmental occlusion was safe before anastomosis in OPCAB; shunts were used in critical ischemiaIschemic dysfunction was precipitated by the 5-min LAD occlusion, as shown by the increase in LVWMS and PA pressure. However, a 5-min coronary occlusion and the resulting ischemia did not alter regional LV systolic function during subsequent ischemia There were no perioperative MI/no conversion to ONCAB LVWMS decreased significantly after first cycle but improved after IP No significant differences in pulmonary artery pressures were after IP and during anastomosis Malkowski (1998)      preconditioned groups in terms of EF and troponin release. Li et al. [6] randomized a small number of patients to on and off-pump to be preconditioned with the use of hyperbaric oxygen (HBO). Patients in the preconditioning group underwent HBO for 70 min/daily for 5 consecutive days before surgery. Preconditioning with HBO resulted in both cerebral and cardiac protective effects as determined by biochemical markers of neuronal and myocardial injury and clinical outcomes in patients undergoing ONCAB while no benefits were observed in the OPCAB group. Matsumoto et al. [24] reviewed a cohort of 48 OPCAB patients. Among them, the subgroup treated with IP plus allopurinol and nicorandil had an improved post-ischemic functional recovery after surgery.

IP
Myocardial conditioning in cardiac surgery can be achieved in different ways. The most used are central or remote preconditioning and volatile anesthetics, while adenosine or other pharmacological agents less frequently used. Ischemic conditioning was first reported in 1986 [4] and conditioning by volatile anesthetic with halothane in 1976 [44]. Although different, both techniques share probably some common mechanisms of action and, most importantly, can be used simultaneously. Both have been frequently utilized in ONCAB [45], but with regard of OPCAB surgery no large trials have been published so far evaluating their effect on troponin [19] or clinical outcomes. Moreover, according to clinical trial.gov [46], while there are different trials in different phases, registered and ongoing, investigating the effect of RIPC in ONCABG (isolated or plus/minus valve surgery) the only trial aiming to compare the effect of RIPC on ON and OFF-pump CABG is the RIP-CON study, expected to report in June 2015 [47]. Small trials are present in the literature, focusing on the effect of RIPC on different subsystem outcomes. Joung et al. [16] randomised 70 OPCAB patients to RIPC and control with cognitive function as primary end points. Hong et al. reported no differences in terms of troponin release in OPCAB after RIPC [19]. Same Authors [18], however, studied the effect of RIPC and PostC together and they observed, this time, a reduction of troponin release in a series of OPCAB. All the other trials [13,14,17,[20][21][22]48] carried out in OPCAB with conditioning used 'central preconditioning' , hence eliciting protection by cycles of occlusion/reperfusion at the level of targeted vessel. All of them led to different and non-homogeneous results. Central remote ischemic preconditioning has been, however, fairly dismissed in on pump cardiac surgery and during interventional cardiology procedure for its impracticality and its high level of invasiveness [2,3,49]. It is important to stress that some of the RIPC and central IP trials did not state the type of anesthesia used [13,23], or deliberately used propofol as anesthetic maintenance [12,[16][17][18]20], while very few did not use it [19,21,22].
Recently it has been reported that propofol may inhibit preconditioning effect, and large trials have intentionally avoided it use [50].
Since the first report in 1976 [44] volatile anaesthetics have been frequently used as conditioning techniques in ONCAB. Two large meta-analyses from 2006 [51,52] have suggested that there was a significant difference in in patients who received volatile agents in terms of reduced troponin release, improved CI, less use of inotropes and reduced need of mechanical ventilation. However they did not point out any differences with regards of mortality, LOS and myocardial infarction rate. Moreover a relatively recent review [53] has reported that sevoflurane had beneficial effects only in naïve patients with no previous exposures to episode of angina. Although there is sufficient evidence to support the use of volatile agents during cardiac surgery and taking into account their feasibility to be used as pre, per and post conditioning with the potential to prolong the protective effect [25], no large trials have been set in place in order to identify potential benefits in patients undergoing OPCAB. Up till now there are 17 trials comparing different volatile anesthetics vs propofol or vs control or volatiles among themself but the vast majority of them were underpowered and led to conflictive results ( Table 2). Sevoflurane is proven to reduce troponin level after on-pump surgery but its effect is still unclear in the off-pump scenario. With regard to OPCAB, sevoflurane was found to preserve cardiac function in three studies [35,37,54], to reduce troponin release in two studies [26,27], to have antioxidant properties in one [33] and to prevent incidence of arrhythmias in one [34]. On the contrary sevoflurane did not show any better troponin reduction when compared to other agents in four studies [28][29][30][31]. Only Orriach et al. used sevoflurane during and after the operation as post-conditioning and reported a significant reduction in troponin release [25].

Conclusions
Although OPCAB is thought to reduce the extent of general ischemia, ischemic cardiac insult can be found in up to 10% of the patients [12]. Data coming from the largest pooled analysis on mainly on-pump experience reported a trend toward a consistent reduction of almost a half of MI in RIPC groups [55]; hence there may be the theoretical potential to translate the same advantage to off-pump patients. However, up till now there are no strong evidences supporting the use of ischemic preconditioning, either central or remote, in OPCAB patients.
In terms of volatile anaesthetics, taking into account all the trials, we can conclude here that has not yet been demonstrated if they can reduce troponin level after off-pump surgery. Other forms of conditioning used in OPCAB such as adenosine/HBO/pharmacological have been used and some times in association with IP, but led to different conclusions. Conditioning, either elicited by ischemia/reperfusion or by volatile agents, may theoretically be a valid method to increase cardiac protection in off-pump surgery, but further trials are definitely needed.