Ischemic Preconditioning Suppresses Ventricular Tachyarrhythmias After Myocardial Revascularization
Background— Ventricular fibrillation (VF) and tachycardia (VT) are the common and potential life-threatening complications after CABG. Ischemic preconditioning (IP) has been proved effective in reducing ischemia reperfusion arrhythmia in animals and humans. Whether IP is effective in suppressing postoperative VF/VT in patients with CABG has not been studied.
Methods and Results— Eighty-six patients with CABG with stable and unstable 3-vessel disease were equally randomly assigned into an IP and a control group. The patients who received IP received 2 periods of 2-minute ischemia followed by 3-minute reperfusion. Twenty-four-hour electrocardiographic data were collected. IP resulted in fewer cases of VF after declamping (48.8% versus 79.1% in IP and control, P=0.004) and a shorter VF period (2.28±0.44 versus 4.41±0.51 minutes, P=0.002). The episodes of VT were significantly reduced in patients in the IP group during early reperfusion and 24 hours after reperfusion (0.65±0.16 versus 3.71±0.46, P=0.000 and 0.07±0.04 versus 2.12±1.41, P=0.002, respectively). De novo sustained VT occurred in 3 control patients as against none in the IP group after surgery. As a result, IP significantly curtailed the mechanical ventilation period and reduced the need for inotropes.
Conclusions— IP significantly reduced postoperative VF/VT in patients with CABG with 3-vessel disease. Suppression of VT during early reperfusion and 24 hours after reperfusion suggests early and delayed IP phenomena in patients undergoing CABG surgery.
Received July 31, 2002; revision received September 19, 2002; accepted September 22, 2002.
Ischemic heart disease carries an increased risk of malignant ventricular fibrillation (VF) and tachycardia (VT) and sudden cardiac death.1–6⇓⇓⇓⇓⇓ VT in patients with coronary artery disease reflects a complex interaction between reversible ischemia and chronic fixed electrophysiological abnormalities. Hemodynamic instability, transient metabolic abnormalities, and a defined anatomic electrical substrate have an established arrhythmogenic potential in the postoperative setting.1–6⇓⇓⇓⇓⇓ Relief of underlying ischemia by CABG eliminates susceptibility to VF and VT, whereas CABG also appears to exert effects on the arrhythmogenic substrates.1,2⇓ Postcardioplegic reperfusion-induced nonsustained VT is generally considered to be a benign transient event, although it may increase the risk of future life-threatening arrhythmic events in some patients.1,3–6⇓⇓⇓⇓ It has been considered to be a manifestation of ischemia-reperfusion injury and used as a variable to compare strategies for myocardial protection during cardiac operations.3,4⇓ De novo sustained VT after CABG is an uncommon but serious complication with an unsatisfactory prognosis.1,2,5,6⇓⇓⇓
See p 2999
Abundant evidence shows that ischemic preconditioning (IP) is the most powerful protective mechanism against ischemia-reperfusion injury in animals.7,8⇓ There are reports of the IP phenomenon during coronary balloon angioplasty.9 Protection afforded by previous unstable angina episodes against subsequent myocardial ischemia also supports the conception of IP effects in clinical patients.10,11⇓ IP has also been proved effective in preserving high-energy phosphate,12 improving heart performance,13,14⇓ and reducing cardiac troponin T release15 during open heart surgery. However, the precise mechanism involved in this is as yet unclear; as derived from an animal model, it might also not totally fit clinical settings.8 Controversy exists as to the effect and safety of IP in open heart surgery.16 Differences in IP protocol, end points, patients selected, and myocardial protective methods may contribute to the divergence of conclusions.12–17⇓⇓⇓⇓⇓
IP has been reported to be associated with a reduction in ischemia-reperfusion arrhythmias in experimental animals. These antiarrhythmic effects of preconditioning appear to be as potent as those induced by standard antiarrhythmic drugs.18–24⇓⇓⇓⇓⇓⇓ There is also evidence to demonstrate that IP may also induce antiarrhythmic protection in humans.9,10,25,26⇓⇓⇓ Whether IP protects the heart from postoperative VF/VT in open heart surgery has not been studied.
The study design was accepted by the Ethics Committee of Tampere University Hospital, Finland, and informed consent was obtained from all patients.
Ninety consecutive stable and unstable patients undergoing CABG were included in the study. All patients were randomly assigned into 2 groups: a control group and a study (IP) group receiving an IP protocol. All unstable patients had unstable angina pectoris within 3 days before the operation. These were diagnosed by ECG and treated with nitroglycerin infusion. Patients with severe calcified ascending aorta, recent myocardial infarction (<3 months), the presence of a preoperative basic rhythm of atrial fibrillation, additional cardiac diseases, and severe noncardiac diseases, cardiac reoperation, emergency operation, and intraoperative or postoperative death were excluded from the study in both groups.
Two patients in IP and 2 control subjects were excluded after unnoticed accidental disconnection of the wires or malfunction of the recording apparatus in 24-hour ECG. This left a final study cohort of 86 patients (43 cases in IP and 43 in control subjects). The preoperative characteristics of the patients in the respective groups were similar (Table 1).
After the cardiopulmonary bypass (CPB) was created and the pump run to vent the heart, the ascending aorta was occluded by cross-clamping for 2 minutes, followed by 3 minutes of reperfusion, the procedure being repeated once. The control group also had the pump running for 10 minutes before the routine operation. The temperature was kept normothermic during this period. The protocol was selected because it causes myocardial ischemia as manifested by higher cardiac troponin I release during the IP period and resulted in better hemodynamic outcome in our previous study.14
Anesthesia, CPB, and Surgical Technique
A standardized anesthetic technique was used with sufentanil, midazolam, and pancuronium. CPB with nonpulsatile perfusion flow (2.2 to 2.4 L/min per meter squared) was conducted with the use of membrane oxygenators with arterial line filtration. Mild hypothermia (32°C) was maintained without topical cooling. Surgical techniques were the same in all cases. Aortic root and 2-stage single venous cannula were used for CPB. A retrograde, self-inflating coronary sinus cardioplegia cannula (RC014, Research Medical Inc) with a pressure monitoring port was guided into place. A 9-gauge cannula was placed in the aortic root for antegrade cardioplegia or for venting. Distal anastomoses were made in the order of right coronary artery–circumflex artery–left anterior descending coronary artery. The proximal anastomoses were constructed during cross-clamping. Left internal mammary artery to left anterior descending coronary artery was used in all patients.
Blood from the pump reservoir was mixed with crystalloid in a ratio of 4:1, yielding a cardioplegia solution with a 0.21 hematocrit value and 21 mmol/L potassium concentration in the initial and 9 mmol/L in subsequent doses. In antegrade delivery, cardioplegia was administered at a pressure of 80 mm Hg and in retrograde 30 to 50 mm Hg, with a flow of at least 200 mL/min. The initial high-potassium cardioplegia was given 1.5 minutes antegrade, then 2.5 minutes retrograde, at a temperature of 6° to 9°C. One minute was given retrograde to the right coronary artery and left circumflex area grafts after each distal anastomosis. Warm cardioplegia (37°C) was given retrograde for 3 minutes before release of cross-clamping.
Measurements of VT/VF
Data from 24-hour ECG were collected 1 day before the operation, during the operation, and up to the 2nd postoperative morning. Holter monitoring was performed with 2-channel tape recorders (Oxford Medilog 4500). The recordings were analyzed with the Oxford Medilog ECG Replay system (Rel 8.5 version). An analysis was conducted blind to the results of other investigations. It was performed by visual review of the recordings of any potential episodes of rhythm disturbances. Details of VT/VF were noted by review of a hard-copy ECG printout. The data were preserved and analyzed in the same period with the same standards. The time periods of recording were set before the study and illustrated in Figure 1. The recording periods were similar between the groups, which lasted 13.6±0.5 hours in the IP group and 14.1±0.4 hours in the control group in T1, 18.1±0.5 hours and 18.4±0.3 hours in T5, and 22.6±0.6 hours and 23.4±0.3 hours in T6.
Nonsustained VT was defined as ≥3 successive ventricular extrasystoles (VESs) at a rate of >120 beats/min. Sustained VT was defined as VT lasting >30 seconds.
Volume infusion was aimed to maintain filling pressure at least at preoperative level. Pharmacological therapy with inotropes was used to maintain the cardiac index of >2.0 L/min per meter squared. β-Blocker was continued after weaning from the respirator. The cardiologist on duty was responsible for antiarrhythmic interventions, including medication, pacing, and electric cardioversion when needed. Perioperative infarction was diagnosed if any new Q wave appeared, with one-third QRS height and for >0.04 second or CK-MB passed beyond 100 μg/L. The ICU team and cardiologists were blinded in the treatment of postoperative care.
The unpaired Student’s t test was used for continuous data (2-tailed) and χ2 test or Fisher’s exact test for categoric data when comparing variables between two groups. The Mann-Whitney U test was used for skewed distributions. One-way ANOVA with Bonferroni correction was used for continuous data when comparing variables for more than 2 groups. The level of significance was set at 0.05. Data are presented as mean±SEM. Statistical analyses were performed with the SPSS/Win (version 10.0) statistical package program.
Outcome of Surgery
The period of mechanical ventilation was significantly shorter in the IP group than in the control group. The length of stay in ICU was similar in both groups. More patients in the control group needed adrenaline support. The period of inotropic medication was also marginally shorter in the patients who received IP than in the control subjects (Table 2). Adrenaline administration did not influence the incidence of postoperative VT/VF. There was no perioperative myocardial infarction or early postoperative death in either group. Intra-aortic balloon pumping was not required in any patient.
Data From 24-Hour ECG
The baseline levels of heart rate (HR) and preoperative VT episodes were similar between the groups. HR was significantly increased after the operation compared with baseline (P<0.001). HR was higher during the IP protocol in patients who received IP, albeit without statistically significant difference compared with the control subjects (P=0.125).
There was no VF before surgery. Most VF episodes appeared at T3; only 2 in controls lasted longer. In the control patients, 79.1% of the cases (77.3% in the stable and 81.0% in the unstable patients) had VF after declamping. IP significantly suppressed cardioplegic ischemia-reperfusion VF, with VF developing in only 48.8% of the cases (47.8% in the stable patients and 50.0% in the unstable patients). The period of VF was also statistically significantly shorter in the patients who received IP (Table 3).
There were 20 cases (23.3%) with preoperative VT, 12 IP (27.9%) with 34 episodes, and 8 control subjects (18.6%, P=0.307) with 14 episodes (P=0.397). Preoperative VT did not correlate with postoperative VT episodes (r=−0.125, −0.055, −0.055, and −0.096 at T4, T5, and T6 and total episodes after surgery, respectively).
The IP protocol resulted in more VT episodes, with mean VT episodes during the IP period (T2) being significantly more numerous than in the control group (1.91±0.41 versus 1.05±0.20, P=0.033; Figure 2).
Nearly all control patients (97.7%) had VT episodes after the operation, compared with 55.8% in the IP group. The occurrence of VT after the operation was markedly less in the patients who received IP during T4 and T6 but not during T5 (Table 3). Mean VT episodes were significantly fewer in the IP group than in the control group at T4 (P<0.001) and T6 (P=0.002) but not at T5 (P=0.137, Figure 2).
Sustained VT occurred in 2 patients before surgery (1 in each group). There was no sustained VT after CABG in these two cases. Sustained VT after surgery was noted in 3 control subjects (2 stable and 1 unstable) and none in patients who received IP. One case was encountered with 8 repeated sustained VT during T5 but none later. The other two had a single episode, one during T5 and the other during T6. All patients were successfully resuscitated. There were no VT episodes before surgery in these 3 patients. Patients with post-CABG VT had more nonsustained VT episodes after the operation (Table 4).
This study showed that cardioplegic ischemia-reperfusion resulted in a high incidence of VT/VF in patients undergoing CABG. The findings demonstrated that 2 periods of 2-minute ischemia followed by 3-minute reperfusion was effective in suppressing the incidence and severity of postoperative VT/VF in patients with CABG. We have previously reported that the same IP protocol benefits hemodynamic recovery.14 As the result of the protective IP effect on left ventricular function and antiarrhythmic IP effect, the patients who received IP required shorter mechanical ventilation support period. To our knowledge, this is the first report of an IP effect in suppressing ventricular tachyarrhythmia in cardiac surgery.
The electrical instability seen after heart surgery has been explained by a preexisting structural arrhythmia substrate, ischemia, and perioperative myocardial infarction, catecholamine stimulation, hemodynamic instability, abnormal calcium modulation and free radical generation during reperfusion, antiarrhythmic and adrenergic drugs, and metabolic disturbance.1–6,22,27⇓⇓⇓⇓⇓⇓⇓ Adrenergic stimulation causes VF only in the presence of ischemia.27 Restoration of the aerobic metabolism or recovery of the membrane ionic pump function, rapid washout of potassium, restoration of temperature, and an increase in ionized calcium concentration and restoration of pH value during reperfusion are responsible for the return of a nonfibrillating rhythm during reperfusion.3 Because myocardial ischemia may be an important factor in the genesis of VF/VT, postcardioplegic VF/VT has been considered as a variable in comparing strategies for myocardial protection.3,5⇓ The high incidence of postcardioplegic VF/VT in the present study would suggest the existence of myocardial ischemia-reperfusion injury in the patient concerned. The IP protocol could thus be used as an additional myocardial protective strategy in CABG surgery.
Two phases of the IP effect have been investigated. Early IP effect occurs within minutes and lasts for 1 to 3 hours in different species.7,8,11,23,24⇓⇓⇓⇓ Delayed IP effect occurs 24 hours after IP stimuli and may last up to 72 hours.8,11,20⇓⇓ Studies of the time course of early and delayed IP in humans are still few in numbers and not unambiguous.11 The mechanism of IP is not yet clear. Receptor (adenosine A1, α-adrenoceptor, bradykinin B2) activation by triggers (adenosine, norepinephrine, free radicals, bradykinin, and nitric oxide), translocation and activation of protein kinase C, and opening of ATP-dependent potassium channels are known to be involved. The delayed IP involves gene transcription and the subsequent production of new protective proteins, including stress proteins and antioxidant enzyme systems.8 The antiarrhythmic IP mechanism might be different from the antinecrosis IP mechanism. The protective antiarrhythmic role of endothelium-derived protective mediators such as bradykinin, prostacyclin, nitric oxide, and cGMP, which do not appear to be involved in the anti-ischemic effect, is important in the antiarrhythmic effects of IP.10,19,20⇓⇓
In the present study, VT was significantly reduced by IP within 2 hours and over the period of 24 to 48 hours after reperfusion but not 2 to 24 hours after reperfusion. This pattern fits well with the time course of early and delayed IP. VT episodes have been proved to be related to recent myocardial ischemia or infarction.2,5,6⇓⇓ It has also emerged that myocardial ischemia remains common even after successful revascularization, and current techniques cannot rule out ischemia of small foci in the myocardium.2,5,6⇓⇓ VT episodes in the patients that we studied suggested the existence of myocardial ischemia even after successful revascularization. This would have room for delayed IP protective effects. The present findings supported those in previous studies that IP effects were transient but could reoccur in the delayed phase after 24 hours.19,20⇓ However, the time periods were arbitrarily set before the study. The present results only indicate the approximate rather than the exact time course of early and delayed IP phenomena.
Nonsustained VT has been reported in 17% to 58% of patients after CABG.5 The incidence of sustained VT is definitely lower, with an occurrence of 0.41% to 3.1%.1,2,5⇓⇓ The incidence of VT (98%) and sustained VT (7%) in our control patients was higher than previously reported.1,2,5⇓⇓ Sustained VT is reported to be related to the number of vessels diseased and bypassed.1 All patients selected for the present study had 3-vessel coronary artery disease and hence showed higher VT. The use of 24-hour ECG monitoring was another reason for the high VT incidence in our patients. It has been reported that sustained VT is not seen during the first postoperative day.2 In the present study, 2 cases with sustained VT were found within 24 hours after surgery, which would indicate that VT may occur unexpectedly any time from immediately after the operation.5
Nonsustained VT has been regarded as a benign phenomenon after CABG, whereas sustained VT carries a high mortality risk both during the acute phase and after hospital discharge.1,2⇓ The VT event is unstable and requires resuscitation.2 In the present study, successful CABG corrected the preoperative sustained VT. Postoperative sustained VT was not related to the preoperative episodes, as demonstrated by previous reports.1,5⇓ However, patients with sustained VT appeared to have postoperative nonsustained VT episodes, which is in contrast to previous findings.1 Therefore, even nonsustained VT does not in general require treatment but may increase the risk of future life-threatening arrhythmic events.5
Ventricular arrhythmias are often detected in patients during unstable angina episodes.10 The cases selected in the present study were patients with recent unstable angina episodes; emergency cases were not included. When patients remained unstable with nitroglycerin infusion, an emergency operation was performed. Hence, we could not record more VT events before surgery in unstable patients. In most cases, the preoperative onset of angina episodes in our recent unstable patients with CABG fits the time course of delayed IP protection. Previous studies have suggested that antecedent angina before a prolonged ischemic event may act as a preconditioning stimulus.11,12⇓ There was a trend of less postoperative VT in unstable control subjects compared with stable control subjects in the present findings. However, there was no statistically significant difference between the groups. The antiarrhythmic IP effect is a dose-dependent phenomenon and has an independent mechanism from antinecrosis effect.10,18–21,27⇓⇓⇓⇓⇓ In a clinical study setting, it is difficult to estimate the duration and frequency of unstable ischemic episodes, since some may be asymptomatic or difficult to monitor.5 Part of the unstable ischemic episodes in our patients might thus be too weak to constitute antiarrhythmic IP stimuli or too strong to cause cumulative deleterious effects.
In summary, IP with 2 periods of 2-minute ischemia followed by 3-minute reperfusion significantly reduced postoperative VF/VT in patients with CABG with 3-vessel disease. The decreased incidence of VT in the IP group during early reperfusion and 24 hours after reperfusion suggests early and delayed IP phenomena in patients with CABG. Preceding unstable angina in the patients that we studied may not act as an effective IP stimulus in suppressing VF/VT.
The study was supported by the Research Foundation of Tampere University Hospital.
- ↵Steinberg JS, Gaur A, Sciacca R, et al. New-onset sustained ventricular tachycardia after cardiac surgery. Circulation. 1999; 99: 903–908.
- ↵Murry CE, Jennings RB, Reimer KA. Preconditioning with ischemia: a delay of lethal cell injury in ischemic myocardium. Circulation. 1986; 74: 1124–1136.
- ↵Pasceri V, Lanza GA, Patti G, et al. Preconditioning by transient myocardial ischemia confers protection against ischemia-induced ventricular arrhythmias in variant angina. Circulation. 1996; 94: 1850–1856.
- ↵Jenkins DP, Pugsley WB, Alkhulaifi AM, et al. Ischemic preconditioning reduces troponin T release in patients undergoing coronary artery bypass surgery. Heart. 1997; 77: 314–318.
- ↵Parratt JR, Vegh A. Delayed protection against ventricular arrhythmias by cardiac pacing. Heart. 1997; 78: 423–425.