- Email: [email protected]
Beneficial effects of ischemic preconditioning on right ventricular function after coronary artery bypass grafting
Beneficial Effects of Ischemic Preconditioning on Right Ventricular Function After Coronary Artery Bypass Grafting Zhong-Kai Wu, MD, Matti R. Tarkka, MD, Erkki Pehkonen, MD, Liisa Kaukinen, MD, Eva L. Honkonen, MD, and Seppo Kaukinen, MD Division of Cardiac Surgery, Department of Anesthesiology and Intensive Care, Tampere University Hospital, Tampere, Finland
Background. Preservation of right ventricular myocardium is unsatisfactory in patients with critical stenosis or occlusion of the right coronary artery. The aim of this study was to investigate whether ischemic preconditioning (IP) improved the recovery of right ventricular function after coronary artery bypass grafting. Methods. Forty patients with three-vessel disease who had coronary artery bypass grafting were randomly assigned to the IP group (n ⴝ 20) or control group (n ⴝ 20). In the IP group, two cycles of two minutes of ischemia after three minutes of reperfusion were given before cross-clamping. Hemodynamic data were collected. Right ventricular ejection fraction was measured by thermodilution. Results. Right ventricular ejection fraction and right
ventricular systolic volume index were decreased postoperatively (lowest value at 6 hours postoperatively). The changes in right ventricular ejection fraction were significantly milder in the IP group postoperatively (p ⴝ 0.012). The decrease in right ventricular systolic volume index postoperatively was also less in IP patients (p ⴝ 0.002). Fewer inotropic drugs were used in the IP group compared with controls. Conclusions. Ischemic preconditioning had a myocardial protective effect on recovery of right ventricular contractility in patients who had coronary artery bypass grafting.
S
especially severe stenosis of the right coronary artery (RCA).
ingle or repeated brief periods of myocardial ischemia after reperfusion increases myocardial tolerance to subsequent long-term ischemic insult. This paradoxical phenomenon was first described and termed ischemic preconditioning (IP) by Murry and colleagues [1]. A similar IP effect has since been found in almost all mammals tested [2], including human myocardial tissue [3] and in patients [4]. This mode of myocardial protection also has been studied in clinical practice, in procedures such as coronary angioplasty [5] and open heart operations [6 –9]. There are nevertheless controversial reports about its effect in cardiac operations [10, 11]. Recent studies have shown that the current myocardial protective method does not provide adequate protection to the ischemia-reperfusion damaged myocardium in patients with severe coronary stenosis, especially right ventricular (RV) function [12, 13]. Thus, IP might afford additional protection in patients with severe coronary stenosis [14]. This study was devised to determine whether ischemic preconditioning protects RV function during coronary artery bypass grafting (CABG) with combined delivery of antegrade and retrograde cold blood cardioplegia in patients with three-vessel disease,
(Ann Thorac Surg 2000;70:1551–7) © 2000 by The Society of Thoracic Surgeons
Material and Methods The study design was approved by the Ethics Committee of Tampere University Hospital, Finland, and informed consent was obtained from all patients. Forty patients with stable angina and three main coronary artery stenosis admitted for CABG operation were randomly assigned into the control group, in which routine myocardial protection methods with cold blood cardioplegia were used, or the study (IP) group, which received IP before cross-clamping. Patients with low ejection fraction (EF) (⬍ 40%), unstable angina, recent myocardial infarction (⬍ 3 months), additional cardiac diseases, severe noncardiac diseases, and calcified or dilated ascending aorta were excluded. The preoperative characteristics of the patients in the respective groups were similar; there were no statistically significant differences between the groups in patients’ age, sex, New York Heart Association class, diseased vessel, history of myocardial infarction, diabetes, risk factor, and preoperative medications (Table 1).
Accepted for publication May 4, 2000.
Preconditioning Protocol
Address reprint requests to Dr Tarkka, Clinic of Cardiothoracic Surgery, Tampere University Hospital, 33521 Tampere, Finland; e-mail: [email protected].
After cardiopulmonary bypass (CPB) was established with ventilation of the heart, the ascending aorta was occluded by cross-clamping for 2 minutes, followed by 3
© 2000 by The Society of Thoracic Surgeons Published by Elsevier Science Inc
0003-4975/00/$20.00 PII S0003-4975(00)01850-6
1552
WU ET AL PRECONDITIONING AND RIGHT VENTRICULAR FUNCTION
Ann Thorac Surg 2000;70:1551–7
Table 1. Preoperative Dataa Variable
Ischemic Preconditioning
Control
p Value
63.7 ⫾ 10.7 20
66.6 ⫾ 9.3 25
0.359 0.705
20 65 15 63.4 ⫾ 9.0 91.8 ⫾ 11.7
10 85 5 66.8 ⫾ 8.4 82.7 ⫾ 18.3
0.333
15 60 25 45 35 10
40 25 35 45 20 25
0.064
20 55 20 0 0 5
20 40 25 10 0 5
0.410
15 25 45
15 30 45
0.999 0.723 0.999
Age (y) Sex (% female) New York Heart Association class (%) II III IV LVEF (%) RCA (stenosis %) RCA 50%–75% 75%–99% 100% Left main coronary artery (⬎ 50% stenosis) History of myocardial infarction Diabetics Risk factor (Cleveland) 0 1 2 3 4 5 Preoperative medication ACE inhibitor Calcium-channel antagonist Antihyperlipidemia drug a
0.262 0.060
0.999 0.288 0.212
Numerical data are presented as mean ⫾ standard deviation. Categoric data are presented as percentage of positive finding.
ACE ⫽ angiotensin-converting enzyme;
NYHA ⫽ New York Heart Association;
minutes of reperfusion, and the procedure was repeated once. The control group also had cardiopulmonary bypass running for 10 minutes before the routine operation (Fig 1). The temperature was kept normal during this period.
Anesthesia, Cardiopulmonary Bypass, and Surgical Technique A standardized anesthetic technique was used with sufentanil, midazolam, and pancuronium. Cardiopulmonary bypass with nonpulsatile perfusion flow (2.2 to 2.4 L/m2 per minute) was conducted using 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 two-stage single venous cannulas were used for CPB. A retrograde, self-inflating coronary sinus cardioplegia cannula (RC014, Research Medical Inc, UT) with a pressure-monitoring port was used. A nine-gauge cannula was placed in the aortic root for antegrade cardioplegia or for venting. Distal anastomoses were made in the order of RCA-circumflex artery(CX)-left anterior descending artery. The proximal anastomoses were constructed during cross-clamping. Left internal mammary artery to left anterior descending artery was used in all patients.
RCA ⫽ right coronary artery.
Blood from the pump reservoir was mixed with crystalloid in a ratio of 4:1, yielding a cardioplegic solution with a 0.21 hematocrit value and 21 mmol/L potassium concentration in the initial dose and 9 mmol/L in subsequent doses. In antegrade delivery, cardioplegia was administered at a pressure of 80 mm Hg, and in retrograde delivery, 30 to 50 mm Hg, with a flow of at least 200 mL/minute. The initial high-potassium cardioplegia was given for 1.5 minutes antegrade then 2.5 minutes retrograde, at a temperature of 6°C to 9°C. One minute of retrograde cardioplegia was given to RCA and left CX area grafts after each distal anastomosis. Warm cardioplegia (37°C) was given retrograde for 3 minutes before release of cross-clamping.
Fig 1. Ischemic preconditioning (IP) protocol. (CPB ⫽ cardiopulmonary bypass; I ⫽ ischemia achieved by aortic cross clamping; min ⫽ minutes; R ⫽ reperfusion by releasing cross-clamping.)
Ann Thorac Surg 2000;70:1551–7
Measurements and Data Collections Heart rate, mean pulmonary artery pressure, pulmonary capillary wedge pressure, cardiac output, and right ventricular ejection fraction (RVEF) were monitored. Derived cardiovascular variables, including cardiac index (CI), right ventricular stroke work index (RVSWI), left ventricular stroke work index, pulmonary vascular resistance index, and right ventricular end-diastolic volume index were calculated using standard formulas. Right ventricular function was measured using a fast-response volumetric thermister-tipped pulmonary artery catheter (93A-434H-7.5F, Baxter Health Care Corp, Glendale, CA) and a microprocessor (Explorer; Baxter Health Care Corp, Edwards Division, Irvine, CA), which allowed measurement of the diastolic washout plateaus of a thermodilution cardiac output curve using exponential curve analysis. All measurements based on the thermodilution technique were made at end-expirium in triplicate using ice-cold saline. The mean value of three consecutive measurements at one time point was calculated. Before each measurement of RVEF, the correct positions of the catheter and right atrial delivery site were confirmed by analysis of the transduced pressure waveform. Hemodynamic data were collected at the following four time points [1]: baseline (before induction of anesthesia [2]), 1 hour after declamping [3], 6 hours after declamping [4], and on the first postoperative day.
WU ET AL PRECONDITIONING AND RIGHT VENTRICULAR FUNCTION
Table 2. Perioperative Data
HEMODYNAMIC DATA.
Blood samples were collected from peripheral vessels before CPB, after IP or 10 minutes of CPB, 5 minutes after declamping, 6 hours after declamping, and on the first and second postoperative days. Samples were collected in heparin-coated plastic tubes and centrifuged. Serum samples were measured with a Chiron ACS180 analyzer (ACS: 180R; Chiron/ Diagnostics, Emeryville, CA) using a direct chemiluminescence method. CREATINE KINASE-MB.
Volume infusion was intended to maintain filling pressure to at least the preoperative level. Pharmacologic therapy with inotropic agents was used to keep the CI greater than 2.0 L/m2 per minute; that therapy was not interrupted when hemodynamic data were measured. Dopoxamine was used as the first-choice inotropic agent, and amrinone with noradrenaline if necessary. Perioperative infarction was diagnosed if any new Q wave appeared with one third QRS height and for longer than 0.04 seconds or if creatine kinase-MB (CKMB) was greater than 100 g/L. The intensive care unit team was masked to the treatment group. POSTOPERATIVE CARE.
Statistics Unpaired Student’s t test was used for continuous data (two-tailed), and 2 test for categoric data was used to compare variables between the groups. Repeatedmeasures analysis of variance was used to test the repeated observation variables postoperatively. Baseline values were used as a covariate when appropriate in the analysis. Mann-Whitney U test was used for skewed
1553
Variable Vessels bypassed (n) Cross-clamping time (min) Cardiopulmonary bypass time (min) Ventricular fibrillation after declamping Cardioversion
Ischemic Preconditioning
Control
p Value
4.2 ⫾ 0.6 86.40 ⫾ 12.66
3.7 ⫾ 0.9 77.47 ⫾ 15.09
0.041 0.042
117.5 ⫾ 13.45
107.15 ⫾ 17.02
0.052
30%
55%
0.110
30%
60%
0.057
Data are presented as mean ⫾ standard deviation. Categoric data are presented as percentage of positive finding. a
distributions. Data are presented as mean ⫾ standard deviation (SD). Level of significance was set at 0.05. The statistical analyses were performed using SPSS for Windows (version 9.0; SPSS Inc, Chicago, IL).
Results Perioperative Course There were significantly more vessels bypassed in the IP group (p ⫽ 0.041), involving a significantly longer period of clamping (p ⫽ 0.042) and almost significant CPB time (p ⫽ 0.052). More patients in the control group needed defibrillation because of ventricular fibrillation after commencement of cross-clamping release, but the differences were not statistically significant (p ⫽ 0.057 and 0.110, respectively) (Table 2).
Hemodynamic Data There were no statistically significant differences in the variables, including heart rate, mean arterial pressure, central venous pressure, pulmonary capillary wedge pressure, mean pulmonary artery pressure, pulmonary vascular resistance index, and right ventricular enddiastolic volume index, between the groups (Table 3). The baseline data in the IP and control groups were similar for RVEF (40.1% ⫾ 4.3% compared with 40.9% ⫾ 6.2%, p ⫽ 0.636), CI (2.35 ⫾ 0.33 compared with 2.53 ⫾ 0.39 L/m2 per minute, p ⫽ 0.116), RVSWI (6.92 ⫾ 1.72 compared with 7.34 ⫾ 2.88 (g-m)/m2/b, p ⫽ 0.591) and left ventricular stroke work index (44.2 ⫾ 9.7 compared with 47.1 ⫾ 12.1 (g-m)/m2/b, p ⫽ 0.424). Right ventricular ejection fraction decreased postoperatively and was lowest at 6 hours after declamping in both groups. The changes were milder in the IP group but statistically significant (p ⫽ 0.012). Both RVSWI and left ventricular stroke work index decreased in both groups postoperatively but more severely in the control group (p ⫽ 0.002 and 0.027, respectively). Cardiac index decreased 1 hour and 6 hours postoperatively in the control group and recovered on the first postoperative day. In the IP group, on the other hand, CI increased at all three postoperative time points. The changes in CI postoperatively were statistically significant between the two groups (p ⫽ 0.013) (Fig 2).
1554
WU ET AL PRECONDITIONING AND RIGHT VENTRICULAR FUNCTION
Ann Thorac Surg 2000;70:1551–7
Table 3. Hemodynamic Data
Variable Heart rate (beats/min) Mean arterial pressure (mm Hg) Central venous pressure (mm Hg) Mean pulmonary artery pressure (mm Hg) Pulmonary capillary wedge pressure (mm Hg) Right ventricular end-diastolic volume index (mL/m2)
Pulmonary vascular resistance index (dyn/sec/cm5/m2)
6 h After Declamping
First Postoperative Day
Group
Baseline
1 h After Declamping
IP C IP C IP
58.7 ⫾ 11.8 61.2 ⫾ 10.6 89.3 ⫾ 12.1 94.1 ⫾ 10.4 7.6 ⫾ 3.1
80.7 ⫾ 18.9 80.0 ⫾ 16.0 76.3 ⫾ 10.3 77.4 ⫾ 11.2 9.8 ⫾ 2.0
92.7 ⫾ 13.9 85.7 ⫾ 24.0 79.8 ⫾ 11.8 77.9 ⫾ 11.6 10.3 ⫾ 2.9
80.4 ⫾ 8.4 82.6 ⫾ 11.7 77.4 ⫾ 10.6 82.1 ⫾ 12.7 9.5 ⫾ 2.4
C IP
8.5 ⫾ 2.0 20.7 ⫾ 4.3
9.9 ⫾ 2.2 19.1 ⫾ 2.8
10.1 ⫾ 2.7 23.8 ⫾ 5.4
8.6 ⫾ 3.4 21.5 ⫾ 3.3
C IP
23.0 ⫾ 7.2 11.8 ⫾ 2.8
21.4 ⫾ 5.3 10.5 ⫾ 2.3
24.3 ⫾ 5.1 10.5 ⫾ 3.4
21.9 ⫾ 5.9 12.0 ⫾ 1.8
C IP
14.1 ⫾ 5.7 99.8 ⫾ 14.2
11.7 ⫾ 2.6 79.4 ⫾ 15.6
11.6 ⫾ 3.1 88.7 ⫾ 17.8
11.6 ⫾ 2.9 95.7 ⫾ 14.7
C IP C
102.9 ⫾ 14.1 252.1 ⫾ 112.6 272.2 ⫾ 101.0
79.0 ⫾ 20.7 290.9 ⫾ 68.9 347.4 ⫾ 136.1
84.9 ⫾ 15.4 386.8 ⫾ 130.4 424.7 ⫾ 137.8
98.2 ⫾ 26.8 287.3 ⫾ 100.5 284.2 ⫾ 91.4
Data are presented as mean ⫾ SD. Categorical data were presented as percentage of positive finding (%). C ⫽ control group;
IP ⫽ ischemic preconditioning group.
Creatine Kinase-MB The base line level of CK-MB was similar (1.9 ⫾ 1.7 and 2.0 ⫾ 1.2 g/L in the IP and control groups, respectively, p ⫽ 0.898). Creatine kinase-MB increased significantly
after CPB, IP, and the operation (p ⬍ 0.001). Peak elevation of CK-MB occurred 6 hours postoperatively (30.5 ⫾ 17.2 and 25.9 ⫾ 12.3 g/L in the IP and control groups, respectively, p ⫽ 0.348). There was no statistically signif-
Fig 2. Right ventricular function and global hemodynamics in patients who had coronary artery bypass grafting (CABG). (RVEF ⫽ right ventricular ejection fraction; RVSWI ⫽ right ventricular stroke work index; CI ⫽ cardiac index; LVSWI ⫽ left ventricular stroke work index; T0 ⫽ Baseline data before induction of anesthesia; T1 ⫽ 1 hour after declamping; T2 ⫽ 6 hours after declamping; T3 ⫽ first postoperative morning.) Data are presented as mean ⫾ standard deviation. *p ⬍ 0.05, **p ⬍ 0.01.
Ann Thorac Surg 2000;70:1551–7
WU ET AL PRECONDITIONING AND RIGHT VENTRICULAR FUNCTION
1555
Comment
Fig 3. Creatine kinase-MB (CK-MB) in patients who had coronary artery bypass grafting (CABG). (T1 ⫽ before cardiopulmonary bypass; T2 ⫽ after ischemic preconditioning (IP) or 10 minutes of cardiopulmonary bypass; T3 ⫽ 5 minutes after declamping; T4 ⫽ 6 hours after declamping; T5 ⫽ first postoperative morning; T6 ⫽ second postoperative morning.) Data are presented as mean ⫾ standard deviation.
icant difference in CK-MB release on the first and second days postoperatively (p ⫽ 0.226) (Fig 3).
Postoperative Care One patient in the control group who was 79 years old and had systemic arteriosclerosis died of cerebral complications on the fourth postoperative day. There was no evidence of myocardial infarction or poor hemodynamic performance in this patient, and according to our standard, there was no apparent perioperative myocardial infarction. No intraaortic balloon pump was used. The duration of mechanical ventilation and stay in intensive care unit was nonsignificantly shorter in the IP group. Significantly more patients in IP group were free of inotropic agents. The duration of inotropic support in the IP group was also nonsignificantly shorter (Table 4).
Table 4. Postoperative Care Variable Mechanical ventilation (hours) Stay in intensive care unit (hours) Free of inotropic agents Duration of inotropic agents (hours) Dopexamine or adrenaline (%) Amrinone and noradrenaline (%)
Ischemic Preconditioning
Control
p Value
12.7 ⫾ 3.5
21.2 ⫾ 23.5
0.138
37.0 ⫾ 23.9
46.2 ⫾ 43.3
0.425
35
5
0.018
11.7 ⫾ 11.8
26.4 ⫾ 31.8
0.060
45
75
0.053
50
55
0.621
Data are presented as mean ⫾ standard deviation. Categoric data are presented as percentage of positive finding.
Right ventricular function is known to contribute essentially to the maintenance of global heart performance [15]. The RV is bonded anatomically to the left ventricle (LV) by subepicardial muscle fibers that run from the free wall of RV to the anterior wall of the LV. The ventricles also share the interventricular septum and an overlapping blood supply. Right ventricular pump function is important in preventing LV failure by ensuring delivery of the necessary preload required to subserve LV output. Abnormalities in the LV affect RV function (eg, diastolic volume, systolic function, and RV afterload) by pressure elevation in the pulmonary circulation [16]. Preservation of RV integrity during cardiac operations is more difficult than LV protection. Hines and Barash [17] reported that perioperative onset of ischemia can be associated with RV dysfunction, manifested by a decrease in RVEF. Patients with severe right coronary artery stenosis are at increased risk of developing perioperative RV ischemia, which likewise manifests itself in decreased RVEF [12]. Inadequate RV protection can lead to unexpected low postoperative cardiac output despite good preservation of left ventricular function [18]. Current myocardial protective methods do not always offer adequate protection of the RV; especially in patients with RCA stenosis, neither antegrade nor retrograde cardioplegia provides adequate protection of the right atrium and ventricle [12]. Boldt and colleagues [16] showed that acute volume loading after CPB in patients with severe RCA stenosis and prolonged aortic cross-clamping time led to reduced RVEF and cardiac output, whereas in patients without RCA stenosis RVEF and cardiac output increased. The decrease in RV systolic function was most severe at 4 to 6 hours after CPB, manifested by decreased stroke volume and RVEF [12, 19]. Our results in the control group were in concordance with such findings. Right ventricular ejection fraction, RVSWI, left ventricular stroke work index, and CI decreased postoperatively, with a nadir 6 hours postoperatively in the controls. Neither RVEF nor RVSWI recovered within 24 hours, but the CI recovered. This finding supports previous conclusions that combined antegrade and retrograde cold blood cardioplegia offers inadequate protection to the right ventricle. Right ventricular disorder results from the deleterious effect of ischemia and reperfusion damage to the heart and lung, as well as from systemic inflammatory response to CPB, which causes pulmonary vasoconstriction and congestion, resulting in increased RV afterload [20, 21]. The microvascular endothelium in the RV also might be more vulnerable to damage by cardioplegia and reperfusion than that in the LV [22]. Above all, however, RV disorder must be attributed to the limitation of RV myocardial protection. Protection of the RV myocardium is more difficult with cold blood cardioplegia than the LV because of the closer contact of the RV to the right atrium, with warmer
1556
WU ET AL PRECONDITIONING AND RIGHT VENTRICULAR FUNCTION
systemic circuit blood, and the anterior position of the right heart, which favors rewarming of that chamber by handling, contact with room air, and exposure to radiant energy from the surroundings [23]. It is well established that when the heart is diseased, antegrade cardioplegia can fail to give adequate protection to all its regions. Retrograde cardioplegia has been proposed as an alternative or additive to overcome that limitation [24]. Combined delivery of antegrade and retrograde cardioplegia protects the myocardium in jeopardy of inadequate cardioplegic protection [25]. If the balloon catheter does not obstruct the terminal tributaries of the coronary sinus, retrograde delivery of cardioplegia can ensure RV protection with adequate flow rate [26]. However, controversial reports suggest that RV perfusion is poor with retrograde delivery despite the absence of coronary stenosis [13], as the RV free wall drains directly into the lesser venous system (Thebesian veins) and the RV [25]. A more important reason is that the location of the inflated balloon of the coronary catheter might occlude the posterior interventricular vein into which the blood from the RV diaphragmatic wall and two thirds of the ventricular septum drains [26]. The leakage of about 22% of the cardioplegic solution to the right atrium delivered retrogradely by an autoinflatable balloon cannula also might be associated with inadequate myocardial perfusion [26]. The low nutritive retrograde flow (26% to 70% compared with 87% to 90% antegradely) to the right ventricle through the coronary sinus could further increase the difficulty of RV protection [12, 24]. Protection is more difficult in patients with RCA stenosis, where neither antegrade nor retrograde techniques ensure cardioplegic delivery to the RV [13, 23]. Obstructive lesion of the coronary artery results in inconsistent distribution of antegrade cardioplegia and thus leads to inadequate preservation of the myocardial area subserved by the stenotic vessels. Although the problem of maldistribution of cardioplegia in the presence of critical stenosis can be reduced by retrograde perfusion, there is evidence of less flow to the posterior LV septum and the RV free wall [25]. Retrograde cardioplegia also did not improve the ischemic or infarcted myocardium in the left anterior descending artery area [27]. Combined delivery of cardioplegia did not protect the RV better than the single technique in patients with RCA stenosis [12, 13]. Thus consideration of strategies that use IP in myocardial protection against ischemia and reperfusion injury is necessary. Clinical data show that IP effectively preserves highenergy phosphate, protects myocardium against ischemia and reperfusion injury, and improves postischemic functional recovery after cardiac operations [6 –9]. Controversy exists about the effect and safety of IP in open heart operations [10, 11]. As far as we know, however, the protection of IP with respect to RV function in patients who had open heart operations has not been studied. The present results show that, in patients with severe threevessel disease with stable angina who had CABG, IP protected right ventricular function from ischemia and
Ann Thorac Surg 2000;70:1551–7
reperfusion myocardial injury. It might also afford protection when combined with combined antegrade and retrograde delivery of cold blood cardioplegia. The decreased RV function and the need for inotropic support in most control patients indicated inadequate RV protection in patients with severely stenotic three-vessel disease. There is room for IP to improve suboptimal RV protection. The precise mechanism of IP remains unknown. Brief episodes of myocardial ischemia result in the production of adenosine, norepinephrine, free radicals, and bradykinin. These chemical factors act on one or more types of myocyte receptors, leading to translocation of protein kinase C to the cellular membrane, working with inhibitory G protein, subsequently phosphorate target proteins, ion channels, and myofilaments to achieve the IP effect [2, 14]. Our data suggest that IP effectively attenuated myocardial stunning but did not affect cellular necrosis as measured by CK-MB, possibly because peak release of CK-MB occurred 6 hours after ischemia. The release of CK-MB after the IP protocol was mainly found at the later time points. The second reason is that the longer clamping and CPB time in the IP patients also could cause higher CK-MB release, which would lead to observation bias in this variable. However, it is possible that the IP protocol did not improve the CK-MB release. We did not find that IP significantly decreased the severe ventricular arrhythmia during the early reperfusion period, but there was nevertheless a tendency toward less ventricular fibrillation and decreased requirement for cardioversion in the IP group. More accurate observations should be done to investigate the antiarrhythmic effect of IP. Right ventricular ejection fraction is the most frequently used factor to quantify RV function and has been reported to correlate with death in patients who had congestive heart failure associated with coronary artery disease [15]. The injection-to-injection reproducibility of the thermodilution method has been studied in an in vitro validation model, and the coefficient of variation for RVEF has been found to be as low as 4.7% [28]. Right ventricular ejection fraction is dependent on contractility, preload, and afterload. Stroke work index is a function of both contractility and preload and also has been considered one of the best measures of mechanical efficiency of the ventricles [29]. In our study, both preload (central venous pressure, right ventricular end-diastolic volume index ) and afterload (mean pulmonary artery pressure and pulmonary vascular resistance) remained stable, and changes were not different in the respective groups. However, RVEF and RVSWI were better in the IP group. Thus IP protects the RV function by preserving its contractility.
This study was supported by The Research Foundation of Tampere University Hospital. The authors thank Riina Metsa¨noja for statistical assistance.
Ann Thorac Surg 2000;70:1551–7
WU ET AL PRECONDITIONING AND RIGHT VENTRICULAR FUNCTION
References 1. Murry CE, Jennings RB, Reimer KA. Preconditioning with ischemia: a delay of lethal cell injury in ischemic myocardium. Circulation 1986;74:1124–36. 2. Schwarz ER, Whyte WS, Kloner RA. Ischemic preconditioning. Curr Opin Cardiol 1997;12:475– 81. 3. Morris SD, Yellon DM. Angiotensin-converting enzyme inhibitors potentiate preconditioning through bradykinin B2 receptor activation in human heart. J Am Coll Cardiol 1997;29:1599 – 606. 4. Kloner RA, Shook T, Przyklenk K, et al. Previous angina alters in-hospital outcome in TIMI 4. A clinical correlate to preconditioning? Circulation 1995;91:37– 45. 5. Airaksinen KE, Huikuri HV. Antiarrhythmic effect of repeated coronary occlusion during balloon angioplasty. J Am Coll Cardiol 1997;29:1035– 8. 6. Yellon DM, Alkhulaifi AM, Pugsley WB. Preconditioning the human myocardium. Lancet 1993;342:276–7. 7. Lu EX, Chen SX, Yuan MD, et al. Preconditioning improves myocardial preservation in patients undergoing open heart operations. Ann Thorac Surg 1997;64:1320– 4. 8. Illes RW, Swoyer KD. Prospective, randomized clinical study of ischemic preconditioning as an adjunct to intermittent cold blood cardioplegia. Ann Thorac Surg 1998;65:748–53. 9. Jenkins DP, Pugsley WB, Alkhulaifi AM, Kemp M, Hooper J, Yellon DM. Ischemic preconditioning reduces troponin T release in patients undergoing coronary artery bypass surgery. Heart 1997;77:314– 8. 10. Perrault LP, Menasche P, Bel A, et al. Ischemic preconditioning in cardiac surgery: a word of caution. J Thorac Cardiovasc Surg 1996;112:1378– 86. 11. Kaukoranta PK, Lepojarvi MP, Ylitalo KV, Kiviluoma KT, Pauhkurinen KJ. Normothermic retrograde blood cardioplegia with or without preceding ischemic preconditioning. Ann Thorac Surg 1997;63:1268–74. 12. Honkonen E, Kaukinen L, Pehkonen EJ, Kaukinen S. Combined antegrade-retrograde blood cardioplegia does not protect right ventricle better than either technique alone in patients with occluded right coronary artery. Scand Cardiovasc J 1997;31:289–95. 13. Allen BS, Winkelmann JW, Hanafy H, et al. Cardiopulmonary bypass, myocardial management, and support techniques. Retrograde cardioplegia does not adequately perfuse the right ventricle. J Thorac Cardiovasc Surg 1995;109: 1116–26. 14. Cleveland JC Jr, Meldrum DR, Rowland RT, Banerjee A, Harken AH. Optimal myocardial preservation: cooling, cardioplegia, and conditioning. Ann Thorac Surg 1996;61:760– 8. 15. Polak JF, Holman BL, Wynne J, Colucci WS. Right ventricu-
16. 17. 18. 19. 20.
21. 22. 23. 24. 25.
26. 27.
28. 29.
1557
lar ejection fraction: an indicator of increased mortality in patients with congestive heart failure associated with coronary artery disease. J Am Coll Cardiol 1983;2:217–24. Boldt J, Kling D, Moosdorf R, Hempelmann G. Influence of acute volume loading on right ventricular function after cardiopulmonary bypass. Crit Care Med 1989;17:518–22. Hines R, Barash PG. Intraoperative right ventricular dysfunction detected with a right ventricular ejection fraction catheter. J Clin Monitoring 1986;2:206– 8. Mullen JC, Fremes SE, Weisel RD, et al. Right ventricular function: a comparison between blood and crystalloid cardioplegia. Ann Thorac Surg 1987;43:17–24. Stein KL, Breisblatt W, Wolfe C, et al. Depression and recovery of right ventricular function after cardiopulmonary bypass. Crit Care Med 1990;18:1197–200. Kirklin JK, Westaby S, Blackstone EH, Kirklin JW, Chenoweth DE, Pacifico AD. Complement and the damaging effects of cardiopulmonary bypass. J Thorac Cardiovasc Surg 1983;86:845–57. Byrick RJ, Kay C, Noble WH. Extravascular lung water accumulation in patients following coronary artery surgery. Can Anaesth Soc J 1977;24:332– 45. Murphy CO, Pan-Chih, Gott JP, Guyton RA. Microvascular reactivity after crystalloid, cold blood, and warm blood cardioplegic arrest. Ann Thorac Surg 1995;60:1021–7. Christakis GT, Fremes SE, Weisel RD, et al. Right ventricular dysfunction following cold potassium cardioplegia. J Thorac Cardiovasc Surg 1985;90:243–50. Buckberg GD. Update on current techniques of myocardial protection. Ann Thorac Surg 1995;60:805–14. Partington MT, Acar C, Buckberg GD, Julia PL. Studies of retrograde cardioplegia. II. Advantages of antegrade/ retrograde cardioplegia to optimize distribution in jeopardized myocardium. J Thorac Cardiovasc Surg 1989;97:613–22. Menasche P. Experimental comparison of manually inflatable versus autoinflatable retrograde cardioplegia catheters. Ann Thorac Surg 1994;58:533–5. Carrier M, Gregoire J, Khalil A, et al. Myocardial distribution of cardioplegia administered by antegrade and retrograde routes to ischemic myocardium. Can J Surg 1997;40:108–13. Ferris SE, Konno M. In vitro validation of a thermodilution right ventricular ejection fraction method. J Clin Monitoring 1992;8:74– 80. Mangano DT. Biventricular function after myocardial revascularization in humans: deterioration and recovery patterns during the first 24 hours. Anaesthesiology 1985; 62:571–7.
Background. Preservation of right ventricular myocardium is unsatisfactory in patients with critical stenosis or occlusion of the right coronary artery. The aim of this study was to investigate whether ischemic preconditioning (IP) improved the recovery of right ventricular function after coronary artery bypass grafting. Methods. Forty patients with three-vessel disease who had coronary artery bypass grafting were randomly assigned to the IP group (n ⴝ 20) or control group (n ⴝ 20). In the IP group, two cycles of two minutes of ischemia after three minutes of reperfusion were given before cross-clamping. Hemodynamic data were collected. Right ventricular ejection fraction was measured by thermodilution. Results. Right ventricular ejection fraction and right
ventricular systolic volume index were decreased postoperatively (lowest value at 6 hours postoperatively). The changes in right ventricular ejection fraction were significantly milder in the IP group postoperatively (p ⴝ 0.012). The decrease in right ventricular systolic volume index postoperatively was also less in IP patients (p ⴝ 0.002). Fewer inotropic drugs were used in the IP group compared with controls. Conclusions. Ischemic preconditioning had a myocardial protective effect on recovery of right ventricular contractility in patients who had coronary artery bypass grafting.
S
especially severe stenosis of the right coronary artery (RCA).
ingle or repeated brief periods of myocardial ischemia after reperfusion increases myocardial tolerance to subsequent long-term ischemic insult. This paradoxical phenomenon was first described and termed ischemic preconditioning (IP) by Murry and colleagues [1]. A similar IP effect has since been found in almost all mammals tested [2], including human myocardial tissue [3] and in patients [4]. This mode of myocardial protection also has been studied in clinical practice, in procedures such as coronary angioplasty [5] and open heart operations [6 –9]. There are nevertheless controversial reports about its effect in cardiac operations [10, 11]. Recent studies have shown that the current myocardial protective method does not provide adequate protection to the ischemia-reperfusion damaged myocardium in patients with severe coronary stenosis, especially right ventricular (RV) function [12, 13]. Thus, IP might afford additional protection in patients with severe coronary stenosis [14]. This study was devised to determine whether ischemic preconditioning protects RV function during coronary artery bypass grafting (CABG) with combined delivery of antegrade and retrograde cold blood cardioplegia in patients with three-vessel disease,
(Ann Thorac Surg 2000;70:1551–7) © 2000 by The Society of Thoracic Surgeons
Material and Methods The study design was approved by the Ethics Committee of Tampere University Hospital, Finland, and informed consent was obtained from all patients. Forty patients with stable angina and three main coronary artery stenosis admitted for CABG operation were randomly assigned into the control group, in which routine myocardial protection methods with cold blood cardioplegia were used, or the study (IP) group, which received IP before cross-clamping. Patients with low ejection fraction (EF) (⬍ 40%), unstable angina, recent myocardial infarction (⬍ 3 months), additional cardiac diseases, severe noncardiac diseases, and calcified or dilated ascending aorta were excluded. The preoperative characteristics of the patients in the respective groups were similar; there were no statistically significant differences between the groups in patients’ age, sex, New York Heart Association class, diseased vessel, history of myocardial infarction, diabetes, risk factor, and preoperative medications (Table 1).
Accepted for publication May 4, 2000.
Preconditioning Protocol
Address reprint requests to Dr Tarkka, Clinic of Cardiothoracic Surgery, Tampere University Hospital, 33521 Tampere, Finland; e-mail: [email protected].
After cardiopulmonary bypass (CPB) was established with ventilation of the heart, the ascending aorta was occluded by cross-clamping for 2 minutes, followed by 3
© 2000 by The Society of Thoracic Surgeons Published by Elsevier Science Inc
0003-4975/00/$20.00 PII S0003-4975(00)01850-6
1552
WU ET AL PRECONDITIONING AND RIGHT VENTRICULAR FUNCTION
Ann Thorac Surg 2000;70:1551–7
Table 1. Preoperative Dataa Variable
Ischemic Preconditioning
Control
p Value
63.7 ⫾ 10.7 20
66.6 ⫾ 9.3 25
0.359 0.705
20 65 15 63.4 ⫾ 9.0 91.8 ⫾ 11.7
10 85 5 66.8 ⫾ 8.4 82.7 ⫾ 18.3
0.333
15 60 25 45 35 10
40 25 35 45 20 25
0.064
20 55 20 0 0 5
20 40 25 10 0 5
0.410
15 25 45
15 30 45
0.999 0.723 0.999
Age (y) Sex (% female) New York Heart Association class (%) II III IV LVEF (%) RCA (stenosis %) RCA 50%–75% 75%–99% 100% Left main coronary artery (⬎ 50% stenosis) History of myocardial infarction Diabetics Risk factor (Cleveland) 0 1 2 3 4 5 Preoperative medication ACE inhibitor Calcium-channel antagonist Antihyperlipidemia drug a
0.262 0.060
0.999 0.288 0.212
Numerical data are presented as mean ⫾ standard deviation. Categoric data are presented as percentage of positive finding.
ACE ⫽ angiotensin-converting enzyme;
NYHA ⫽ New York Heart Association;
minutes of reperfusion, and the procedure was repeated once. The control group also had cardiopulmonary bypass running for 10 minutes before the routine operation (Fig 1). The temperature was kept normal during this period.
Anesthesia, Cardiopulmonary Bypass, and Surgical Technique A standardized anesthetic technique was used with sufentanil, midazolam, and pancuronium. Cardiopulmonary bypass with nonpulsatile perfusion flow (2.2 to 2.4 L/m2 per minute) was conducted using 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 two-stage single venous cannulas were used for CPB. A retrograde, self-inflating coronary sinus cardioplegia cannula (RC014, Research Medical Inc, UT) with a pressure-monitoring port was used. A nine-gauge cannula was placed in the aortic root for antegrade cardioplegia or for venting. Distal anastomoses were made in the order of RCA-circumflex artery(CX)-left anterior descending artery. The proximal anastomoses were constructed during cross-clamping. Left internal mammary artery to left anterior descending artery was used in all patients.
RCA ⫽ right coronary artery.
Blood from the pump reservoir was mixed with crystalloid in a ratio of 4:1, yielding a cardioplegic solution with a 0.21 hematocrit value and 21 mmol/L potassium concentration in the initial dose and 9 mmol/L in subsequent doses. In antegrade delivery, cardioplegia was administered at a pressure of 80 mm Hg, and in retrograde delivery, 30 to 50 mm Hg, with a flow of at least 200 mL/minute. The initial high-potassium cardioplegia was given for 1.5 minutes antegrade then 2.5 minutes retrograde, at a temperature of 6°C to 9°C. One minute of retrograde cardioplegia was given to RCA and left CX area grafts after each distal anastomosis. Warm cardioplegia (37°C) was given retrograde for 3 minutes before release of cross-clamping.
Fig 1. Ischemic preconditioning (IP) protocol. (CPB ⫽ cardiopulmonary bypass; I ⫽ ischemia achieved by aortic cross clamping; min ⫽ minutes; R ⫽ reperfusion by releasing cross-clamping.)
Ann Thorac Surg 2000;70:1551–7
Measurements and Data Collections Heart rate, mean pulmonary artery pressure, pulmonary capillary wedge pressure, cardiac output, and right ventricular ejection fraction (RVEF) were monitored. Derived cardiovascular variables, including cardiac index (CI), right ventricular stroke work index (RVSWI), left ventricular stroke work index, pulmonary vascular resistance index, and right ventricular end-diastolic volume index were calculated using standard formulas. Right ventricular function was measured using a fast-response volumetric thermister-tipped pulmonary artery catheter (93A-434H-7.5F, Baxter Health Care Corp, Glendale, CA) and a microprocessor (Explorer; Baxter Health Care Corp, Edwards Division, Irvine, CA), which allowed measurement of the diastolic washout plateaus of a thermodilution cardiac output curve using exponential curve analysis. All measurements based on the thermodilution technique were made at end-expirium in triplicate using ice-cold saline. The mean value of three consecutive measurements at one time point was calculated. Before each measurement of RVEF, the correct positions of the catheter and right atrial delivery site were confirmed by analysis of the transduced pressure waveform. Hemodynamic data were collected at the following four time points [1]: baseline (before induction of anesthesia [2]), 1 hour after declamping [3], 6 hours after declamping [4], and on the first postoperative day.
WU ET AL PRECONDITIONING AND RIGHT VENTRICULAR FUNCTION
Table 2. Perioperative Data
HEMODYNAMIC DATA.
Blood samples were collected from peripheral vessels before CPB, after IP or 10 minutes of CPB, 5 minutes after declamping, 6 hours after declamping, and on the first and second postoperative days. Samples were collected in heparin-coated plastic tubes and centrifuged. Serum samples were measured with a Chiron ACS180 analyzer (ACS: 180R; Chiron/ Diagnostics, Emeryville, CA) using a direct chemiluminescence method. CREATINE KINASE-MB.
Volume infusion was intended to maintain filling pressure to at least the preoperative level. Pharmacologic therapy with inotropic agents was used to keep the CI greater than 2.0 L/m2 per minute; that therapy was not interrupted when hemodynamic data were measured. Dopoxamine was used as the first-choice inotropic agent, and amrinone with noradrenaline if necessary. Perioperative infarction was diagnosed if any new Q wave appeared with one third QRS height and for longer than 0.04 seconds or if creatine kinase-MB (CKMB) was greater than 100 g/L. The intensive care unit team was masked to the treatment group. POSTOPERATIVE CARE.
Statistics Unpaired Student’s t test was used for continuous data (two-tailed), and 2 test for categoric data was used to compare variables between the groups. Repeatedmeasures analysis of variance was used to test the repeated observation variables postoperatively. Baseline values were used as a covariate when appropriate in the analysis. Mann-Whitney U test was used for skewed
1553
Variable Vessels bypassed (n) Cross-clamping time (min) Cardiopulmonary bypass time (min) Ventricular fibrillation after declamping Cardioversion
Ischemic Preconditioning
Control
p Value
4.2 ⫾ 0.6 86.40 ⫾ 12.66
3.7 ⫾ 0.9 77.47 ⫾ 15.09
0.041 0.042
117.5 ⫾ 13.45
107.15 ⫾ 17.02
0.052
30%
55%
0.110
30%
60%
0.057
Data are presented as mean ⫾ standard deviation. Categoric data are presented as percentage of positive finding. a
distributions. Data are presented as mean ⫾ standard deviation (SD). Level of significance was set at 0.05. The statistical analyses were performed using SPSS for Windows (version 9.0; SPSS Inc, Chicago, IL).
Results Perioperative Course There were significantly more vessels bypassed in the IP group (p ⫽ 0.041), involving a significantly longer period of clamping (p ⫽ 0.042) and almost significant CPB time (p ⫽ 0.052). More patients in the control group needed defibrillation because of ventricular fibrillation after commencement of cross-clamping release, but the differences were not statistically significant (p ⫽ 0.057 and 0.110, respectively) (Table 2).
Hemodynamic Data There were no statistically significant differences in the variables, including heart rate, mean arterial pressure, central venous pressure, pulmonary capillary wedge pressure, mean pulmonary artery pressure, pulmonary vascular resistance index, and right ventricular enddiastolic volume index, between the groups (Table 3). The baseline data in the IP and control groups were similar for RVEF (40.1% ⫾ 4.3% compared with 40.9% ⫾ 6.2%, p ⫽ 0.636), CI (2.35 ⫾ 0.33 compared with 2.53 ⫾ 0.39 L/m2 per minute, p ⫽ 0.116), RVSWI (6.92 ⫾ 1.72 compared with 7.34 ⫾ 2.88 (g-m)/m2/b, p ⫽ 0.591) and left ventricular stroke work index (44.2 ⫾ 9.7 compared with 47.1 ⫾ 12.1 (g-m)/m2/b, p ⫽ 0.424). Right ventricular ejection fraction decreased postoperatively and was lowest at 6 hours after declamping in both groups. The changes were milder in the IP group but statistically significant (p ⫽ 0.012). Both RVSWI and left ventricular stroke work index decreased in both groups postoperatively but more severely in the control group (p ⫽ 0.002 and 0.027, respectively). Cardiac index decreased 1 hour and 6 hours postoperatively in the control group and recovered on the first postoperative day. In the IP group, on the other hand, CI increased at all three postoperative time points. The changes in CI postoperatively were statistically significant between the two groups (p ⫽ 0.013) (Fig 2).
1554
WU ET AL PRECONDITIONING AND RIGHT VENTRICULAR FUNCTION
Ann Thorac Surg 2000;70:1551–7
Table 3. Hemodynamic Data
Variable Heart rate (beats/min) Mean arterial pressure (mm Hg) Central venous pressure (mm Hg) Mean pulmonary artery pressure (mm Hg) Pulmonary capillary wedge pressure (mm Hg) Right ventricular end-diastolic volume index (mL/m2)
Pulmonary vascular resistance index (dyn/sec/cm5/m2)
6 h After Declamping
First Postoperative Day
Group
Baseline
1 h After Declamping
IP C IP C IP
58.7 ⫾ 11.8 61.2 ⫾ 10.6 89.3 ⫾ 12.1 94.1 ⫾ 10.4 7.6 ⫾ 3.1
80.7 ⫾ 18.9 80.0 ⫾ 16.0 76.3 ⫾ 10.3 77.4 ⫾ 11.2 9.8 ⫾ 2.0
92.7 ⫾ 13.9 85.7 ⫾ 24.0 79.8 ⫾ 11.8 77.9 ⫾ 11.6 10.3 ⫾ 2.9
80.4 ⫾ 8.4 82.6 ⫾ 11.7 77.4 ⫾ 10.6 82.1 ⫾ 12.7 9.5 ⫾ 2.4
C IP
8.5 ⫾ 2.0 20.7 ⫾ 4.3
9.9 ⫾ 2.2 19.1 ⫾ 2.8
10.1 ⫾ 2.7 23.8 ⫾ 5.4
8.6 ⫾ 3.4 21.5 ⫾ 3.3
C IP
23.0 ⫾ 7.2 11.8 ⫾ 2.8
21.4 ⫾ 5.3 10.5 ⫾ 2.3
24.3 ⫾ 5.1 10.5 ⫾ 3.4
21.9 ⫾ 5.9 12.0 ⫾ 1.8
C IP
14.1 ⫾ 5.7 99.8 ⫾ 14.2
11.7 ⫾ 2.6 79.4 ⫾ 15.6
11.6 ⫾ 3.1 88.7 ⫾ 17.8
11.6 ⫾ 2.9 95.7 ⫾ 14.7
C IP C
102.9 ⫾ 14.1 252.1 ⫾ 112.6 272.2 ⫾ 101.0
79.0 ⫾ 20.7 290.9 ⫾ 68.9 347.4 ⫾ 136.1
84.9 ⫾ 15.4 386.8 ⫾ 130.4 424.7 ⫾ 137.8
98.2 ⫾ 26.8 287.3 ⫾ 100.5 284.2 ⫾ 91.4
Data are presented as mean ⫾ SD. Categorical data were presented as percentage of positive finding (%). C ⫽ control group;
IP ⫽ ischemic preconditioning group.
Creatine Kinase-MB The base line level of CK-MB was similar (1.9 ⫾ 1.7 and 2.0 ⫾ 1.2 g/L in the IP and control groups, respectively, p ⫽ 0.898). Creatine kinase-MB increased significantly
after CPB, IP, and the operation (p ⬍ 0.001). Peak elevation of CK-MB occurred 6 hours postoperatively (30.5 ⫾ 17.2 and 25.9 ⫾ 12.3 g/L in the IP and control groups, respectively, p ⫽ 0.348). There was no statistically signif-
Fig 2. Right ventricular function and global hemodynamics in patients who had coronary artery bypass grafting (CABG). (RVEF ⫽ right ventricular ejection fraction; RVSWI ⫽ right ventricular stroke work index; CI ⫽ cardiac index; LVSWI ⫽ left ventricular stroke work index; T0 ⫽ Baseline data before induction of anesthesia; T1 ⫽ 1 hour after declamping; T2 ⫽ 6 hours after declamping; T3 ⫽ first postoperative morning.) Data are presented as mean ⫾ standard deviation. *p ⬍ 0.05, **p ⬍ 0.01.
Ann Thorac Surg 2000;70:1551–7
WU ET AL PRECONDITIONING AND RIGHT VENTRICULAR FUNCTION
1555
Comment
Fig 3. Creatine kinase-MB (CK-MB) in patients who had coronary artery bypass grafting (CABG). (T1 ⫽ before cardiopulmonary bypass; T2 ⫽ after ischemic preconditioning (IP) or 10 minutes of cardiopulmonary bypass; T3 ⫽ 5 minutes after declamping; T4 ⫽ 6 hours after declamping; T5 ⫽ first postoperative morning; T6 ⫽ second postoperative morning.) Data are presented as mean ⫾ standard deviation.
icant difference in CK-MB release on the first and second days postoperatively (p ⫽ 0.226) (Fig 3).
Postoperative Care One patient in the control group who was 79 years old and had systemic arteriosclerosis died of cerebral complications on the fourth postoperative day. There was no evidence of myocardial infarction or poor hemodynamic performance in this patient, and according to our standard, there was no apparent perioperative myocardial infarction. No intraaortic balloon pump was used. The duration of mechanical ventilation and stay in intensive care unit was nonsignificantly shorter in the IP group. Significantly more patients in IP group were free of inotropic agents. The duration of inotropic support in the IP group was also nonsignificantly shorter (Table 4).
Table 4. Postoperative Care Variable Mechanical ventilation (hours) Stay in intensive care unit (hours) Free of inotropic agents Duration of inotropic agents (hours) Dopexamine or adrenaline (%) Amrinone and noradrenaline (%)
Ischemic Preconditioning
Control
p Value
12.7 ⫾ 3.5
21.2 ⫾ 23.5
0.138
37.0 ⫾ 23.9
46.2 ⫾ 43.3
0.425
35
5
0.018
11.7 ⫾ 11.8
26.4 ⫾ 31.8
0.060
45
75
0.053
50
55
0.621
Data are presented as mean ⫾ standard deviation. Categoric data are presented as percentage of positive finding.
Right ventricular function is known to contribute essentially to the maintenance of global heart performance [15]. The RV is bonded anatomically to the left ventricle (LV) by subepicardial muscle fibers that run from the free wall of RV to the anterior wall of the LV. The ventricles also share the interventricular septum and an overlapping blood supply. Right ventricular pump function is important in preventing LV failure by ensuring delivery of the necessary preload required to subserve LV output. Abnormalities in the LV affect RV function (eg, diastolic volume, systolic function, and RV afterload) by pressure elevation in the pulmonary circulation [16]. Preservation of RV integrity during cardiac operations is more difficult than LV protection. Hines and Barash [17] reported that perioperative onset of ischemia can be associated with RV dysfunction, manifested by a decrease in RVEF. Patients with severe right coronary artery stenosis are at increased risk of developing perioperative RV ischemia, which likewise manifests itself in decreased RVEF [12]. Inadequate RV protection can lead to unexpected low postoperative cardiac output despite good preservation of left ventricular function [18]. Current myocardial protective methods do not always offer adequate protection of the RV; especially in patients with RCA stenosis, neither antegrade nor retrograde cardioplegia provides adequate protection of the right atrium and ventricle [12]. Boldt and colleagues [16] showed that acute volume loading after CPB in patients with severe RCA stenosis and prolonged aortic cross-clamping time led to reduced RVEF and cardiac output, whereas in patients without RCA stenosis RVEF and cardiac output increased. The decrease in RV systolic function was most severe at 4 to 6 hours after CPB, manifested by decreased stroke volume and RVEF [12, 19]. Our results in the control group were in concordance with such findings. Right ventricular ejection fraction, RVSWI, left ventricular stroke work index, and CI decreased postoperatively, with a nadir 6 hours postoperatively in the controls. Neither RVEF nor RVSWI recovered within 24 hours, but the CI recovered. This finding supports previous conclusions that combined antegrade and retrograde cold blood cardioplegia offers inadequate protection to the right ventricle. Right ventricular disorder results from the deleterious effect of ischemia and reperfusion damage to the heart and lung, as well as from systemic inflammatory response to CPB, which causes pulmonary vasoconstriction and congestion, resulting in increased RV afterload [20, 21]. The microvascular endothelium in the RV also might be more vulnerable to damage by cardioplegia and reperfusion than that in the LV [22]. Above all, however, RV disorder must be attributed to the limitation of RV myocardial protection. Protection of the RV myocardium is more difficult with cold blood cardioplegia than the LV because of the closer contact of the RV to the right atrium, with warmer
1556
WU ET AL PRECONDITIONING AND RIGHT VENTRICULAR FUNCTION
systemic circuit blood, and the anterior position of the right heart, which favors rewarming of that chamber by handling, contact with room air, and exposure to radiant energy from the surroundings [23]. It is well established that when the heart is diseased, antegrade cardioplegia can fail to give adequate protection to all its regions. Retrograde cardioplegia has been proposed as an alternative or additive to overcome that limitation [24]. Combined delivery of antegrade and retrograde cardioplegia protects the myocardium in jeopardy of inadequate cardioplegic protection [25]. If the balloon catheter does not obstruct the terminal tributaries of the coronary sinus, retrograde delivery of cardioplegia can ensure RV protection with adequate flow rate [26]. However, controversial reports suggest that RV perfusion is poor with retrograde delivery despite the absence of coronary stenosis [13], as the RV free wall drains directly into the lesser venous system (Thebesian veins) and the RV [25]. A more important reason is that the location of the inflated balloon of the coronary catheter might occlude the posterior interventricular vein into which the blood from the RV diaphragmatic wall and two thirds of the ventricular septum drains [26]. The leakage of about 22% of the cardioplegic solution to the right atrium delivered retrogradely by an autoinflatable balloon cannula also might be associated with inadequate myocardial perfusion [26]. The low nutritive retrograde flow (26% to 70% compared with 87% to 90% antegradely) to the right ventricle through the coronary sinus could further increase the difficulty of RV protection [12, 24]. Protection is more difficult in patients with RCA stenosis, where neither antegrade nor retrograde techniques ensure cardioplegic delivery to the RV [13, 23]. Obstructive lesion of the coronary artery results in inconsistent distribution of antegrade cardioplegia and thus leads to inadequate preservation of the myocardial area subserved by the stenotic vessels. Although the problem of maldistribution of cardioplegia in the presence of critical stenosis can be reduced by retrograde perfusion, there is evidence of less flow to the posterior LV septum and the RV free wall [25]. Retrograde cardioplegia also did not improve the ischemic or infarcted myocardium in the left anterior descending artery area [27]. Combined delivery of cardioplegia did not protect the RV better than the single technique in patients with RCA stenosis [12, 13]. Thus consideration of strategies that use IP in myocardial protection against ischemia and reperfusion injury is necessary. Clinical data show that IP effectively preserves highenergy phosphate, protects myocardium against ischemia and reperfusion injury, and improves postischemic functional recovery after cardiac operations [6 –9]. Controversy exists about the effect and safety of IP in open heart operations [10, 11]. As far as we know, however, the protection of IP with respect to RV function in patients who had open heart operations has not been studied. The present results show that, in patients with severe threevessel disease with stable angina who had CABG, IP protected right ventricular function from ischemia and
Ann Thorac Surg 2000;70:1551–7
reperfusion myocardial injury. It might also afford protection when combined with combined antegrade and retrograde delivery of cold blood cardioplegia. The decreased RV function and the need for inotropic support in most control patients indicated inadequate RV protection in patients with severely stenotic three-vessel disease. There is room for IP to improve suboptimal RV protection. The precise mechanism of IP remains unknown. Brief episodes of myocardial ischemia result in the production of adenosine, norepinephrine, free radicals, and bradykinin. These chemical factors act on one or more types of myocyte receptors, leading to translocation of protein kinase C to the cellular membrane, working with inhibitory G protein, subsequently phosphorate target proteins, ion channels, and myofilaments to achieve the IP effect [2, 14]. Our data suggest that IP effectively attenuated myocardial stunning but did not affect cellular necrosis as measured by CK-MB, possibly because peak release of CK-MB occurred 6 hours after ischemia. The release of CK-MB after the IP protocol was mainly found at the later time points. The second reason is that the longer clamping and CPB time in the IP patients also could cause higher CK-MB release, which would lead to observation bias in this variable. However, it is possible that the IP protocol did not improve the CK-MB release. We did not find that IP significantly decreased the severe ventricular arrhythmia during the early reperfusion period, but there was nevertheless a tendency toward less ventricular fibrillation and decreased requirement for cardioversion in the IP group. More accurate observations should be done to investigate the antiarrhythmic effect of IP. Right ventricular ejection fraction is the most frequently used factor to quantify RV function and has been reported to correlate with death in patients who had congestive heart failure associated with coronary artery disease [15]. The injection-to-injection reproducibility of the thermodilution method has been studied in an in vitro validation model, and the coefficient of variation for RVEF has been found to be as low as 4.7% [28]. Right ventricular ejection fraction is dependent on contractility, preload, and afterload. Stroke work index is a function of both contractility and preload and also has been considered one of the best measures of mechanical efficiency of the ventricles [29]. In our study, both preload (central venous pressure, right ventricular end-diastolic volume index ) and afterload (mean pulmonary artery pressure and pulmonary vascular resistance) remained stable, and changes were not different in the respective groups. However, RVEF and RVSWI were better in the IP group. Thus IP protects the RV function by preserving its contractility.
This study was supported by The Research Foundation of Tampere University Hospital. The authors thank Riina Metsa¨noja for statistical assistance.
Ann Thorac Surg 2000;70:1551–7
WU ET AL PRECONDITIONING AND RIGHT VENTRICULAR FUNCTION
References 1. Murry CE, Jennings RB, Reimer KA. Preconditioning with ischemia: a delay of lethal cell injury in ischemic myocardium. Circulation 1986;74:1124–36. 2. Schwarz ER, Whyte WS, Kloner RA. Ischemic preconditioning. Curr Opin Cardiol 1997;12:475– 81. 3. Morris SD, Yellon DM. Angiotensin-converting enzyme inhibitors potentiate preconditioning through bradykinin B2 receptor activation in human heart. J Am Coll Cardiol 1997;29:1599 – 606. 4. Kloner RA, Shook T, Przyklenk K, et al. Previous angina alters in-hospital outcome in TIMI 4. A clinical correlate to preconditioning? Circulation 1995;91:37– 45. 5. Airaksinen KE, Huikuri HV. Antiarrhythmic effect of repeated coronary occlusion during balloon angioplasty. J Am Coll Cardiol 1997;29:1035– 8. 6. Yellon DM, Alkhulaifi AM, Pugsley WB. Preconditioning the human myocardium. Lancet 1993;342:276–7. 7. Lu EX, Chen SX, Yuan MD, et al. Preconditioning improves myocardial preservation in patients undergoing open heart operations. Ann Thorac Surg 1997;64:1320– 4. 8. Illes RW, Swoyer KD. Prospective, randomized clinical study of ischemic preconditioning as an adjunct to intermittent cold blood cardioplegia. Ann Thorac Surg 1998;65:748–53. 9. Jenkins DP, Pugsley WB, Alkhulaifi AM, Kemp M, Hooper J, Yellon DM. Ischemic preconditioning reduces troponin T release in patients undergoing coronary artery bypass surgery. Heart 1997;77:314– 8. 10. Perrault LP, Menasche P, Bel A, et al. Ischemic preconditioning in cardiac surgery: a word of caution. J Thorac Cardiovasc Surg 1996;112:1378– 86. 11. Kaukoranta PK, Lepojarvi MP, Ylitalo KV, Kiviluoma KT, Pauhkurinen KJ. Normothermic retrograde blood cardioplegia with or without preceding ischemic preconditioning. Ann Thorac Surg 1997;63:1268–74. 12. Honkonen E, Kaukinen L, Pehkonen EJ, Kaukinen S. Combined antegrade-retrograde blood cardioplegia does not protect right ventricle better than either technique alone in patients with occluded right coronary artery. Scand Cardiovasc J 1997;31:289–95. 13. Allen BS, Winkelmann JW, Hanafy H, et al. Cardiopulmonary bypass, myocardial management, and support techniques. Retrograde cardioplegia does not adequately perfuse the right ventricle. J Thorac Cardiovasc Surg 1995;109: 1116–26. 14. Cleveland JC Jr, Meldrum DR, Rowland RT, Banerjee A, Harken AH. Optimal myocardial preservation: cooling, cardioplegia, and conditioning. Ann Thorac Surg 1996;61:760– 8. 15. Polak JF, Holman BL, Wynne J, Colucci WS. Right ventricu-
16. 17. 18. 19. 20.
21. 22. 23. 24. 25.
26. 27.
28. 29.
1557
lar ejection fraction: an indicator of increased mortality in patients with congestive heart failure associated with coronary artery disease. J Am Coll Cardiol 1983;2:217–24. Boldt J, Kling D, Moosdorf R, Hempelmann G. Influence of acute volume loading on right ventricular function after cardiopulmonary bypass. Crit Care Med 1989;17:518–22. Hines R, Barash PG. Intraoperative right ventricular dysfunction detected with a right ventricular ejection fraction catheter. J Clin Monitoring 1986;2:206– 8. Mullen JC, Fremes SE, Weisel RD, et al. Right ventricular function: a comparison between blood and crystalloid cardioplegia. Ann Thorac Surg 1987;43:17–24. Stein KL, Breisblatt W, Wolfe C, et al. Depression and recovery of right ventricular function after cardiopulmonary bypass. Crit Care Med 1990;18:1197–200. Kirklin JK, Westaby S, Blackstone EH, Kirklin JW, Chenoweth DE, Pacifico AD. Complement and the damaging effects of cardiopulmonary bypass. J Thorac Cardiovasc Surg 1983;86:845–57. Byrick RJ, Kay C, Noble WH. Extravascular lung water accumulation in patients following coronary artery surgery. Can Anaesth Soc J 1977;24:332– 45. Murphy CO, Pan-Chih, Gott JP, Guyton RA. Microvascular reactivity after crystalloid, cold blood, and warm blood cardioplegic arrest. Ann Thorac Surg 1995;60:1021–7. Christakis GT, Fremes SE, Weisel RD, et al. Right ventricular dysfunction following cold potassium cardioplegia. J Thorac Cardiovasc Surg 1985;90:243–50. Buckberg GD. Update on current techniques of myocardial protection. Ann Thorac Surg 1995;60:805–14. Partington MT, Acar C, Buckberg GD, Julia PL. Studies of retrograde cardioplegia. II. Advantages of antegrade/ retrograde cardioplegia to optimize distribution in jeopardized myocardium. J Thorac Cardiovasc Surg 1989;97:613–22. Menasche P. Experimental comparison of manually inflatable versus autoinflatable retrograde cardioplegia catheters. Ann Thorac Surg 1994;58:533–5. Carrier M, Gregoire J, Khalil A, et al. Myocardial distribution of cardioplegia administered by antegrade and retrograde routes to ischemic myocardium. Can J Surg 1997;40:108–13. Ferris SE, Konno M. In vitro validation of a thermodilution right ventricular ejection fraction method. J Clin Monitoring 1992;8:74– 80. Mangano DT. Biventricular function after myocardial revascularization in humans: deterioration and recovery patterns during the first 24 hours. Anaesthesiology 1985; 62:571–7.