On-Pump Beating-Heart Coronary Artery Bypass
Grafting for Acute Myocardial Infarction
Yuichi Izumi, MD, PhD, Katsuaki Magishi, MD, PhD, Noriyuki Ishikawa, MD, and
Fumiaki Kimura, MD
Department of Thoracic and Cardiovascular Surgery, Nayoro City General Hospital, Nayoro, Hokkaido, Japan
Background. The early results of emergent coronary
artery bypass grafting by conventional operative method
for acute myocardial infarction are reported to be poor.
The purpose of this study is to evaluate on-pump beat-
ing-heart coronary artery bypass grafting for acute myo-
cardial infarction.
Methods. Thirty-one patients with acute myocardial
infarction underwent emergent surgery between January
1998 and June 2004 at Nayoro City General Hospital. In
16 patients, on-pump surgery was performed on the
arrested heart, and in the other 15, on-pump surgery was
performed on the beating heart. Early results were com-
pared between the two groups.
Results. Preoperative and perioperative patient charac-
teristics revealed no significant differences between the
two groups. Although there was no statistically differ-
ence between the two groups, the early mortality rates of
on-pump arrested-heart coronary bypass grafting (31.3%)
was higher than that of on-pump beating-heart coronary
bypass grafting (13.3%). Postoperatively, the creatine
T he early results of emergent coronary artery bypass
grafting (CABG) for acute myocardial infarction
have been reported to be poor. Percutaneous catheter
intervention (PCI) is generally the first choice for treat-
ment of acute myocardial infarction to immediately
reperfuse the coronary flow. However, there are some
cases requiring emergent surgical treatment, and in these
cases the clinical results have been generally unaccept-
able because patients are often hemodynamically unsta-
ble owing to cardiogenic shock. Moreover, cardioplegic
arrest during cardiopulmonary bypass can induce myo-
cardial damage. Recently, in these situations, we have
performed on-pump CABG on the beating heart,
whereas in the past cases would have been treated with
on-pump CABG on the arrested heart. We have now
studied the results of on-pump beating-heart CABG for
acute myocardial infarction and have evaluated the use-
fulness of this technique.
Accepted for publication Aug 23, 2005.
Address correspondence to Dr Izumi, Nayoro City General Hospital,
West 7, South 8, Nayoro, Hokkaido 096-8511, Japan; e-mail: yi398ngh@
seagreen.ocn.ne.jp.
2006 by The Society of Thoracic Surgeons
Published by Elsevier Inc
CARDIOVASCULAR
kinase myocardial band value for the on-pump beating-
heart group was significantly lower than that for the
on-pump arrested-heart group (221 200 IU/L versus 666
540 IU/L, p 0.008). The incidence of postoperative
acute renal failure was significantly higher in the on-
pump arrested-heart group than in the on-pump beating-
heart group (p 0.034). The durations of ventilator use
and inotropic agent use were longer in the on-pump
arrested-heart group than in the on-pump beating-heart
group, though the differences were not statistically dif-
ferent (p 0.152, p 0.223).
Conclusions. On-pump beating-heart coronary artery
bypass grafting has the possibility to eliminate intra-
operative global myocardial ischemia and to be an
acceptable surgical option for acute myocardial infarc-
tion associated with lower postoperative mortality and
morbidity.
(Ann Thorac Surg 2006;81:573 6)
2006 by The Society of Thoracic Surgeons
Patients and Methods
Thirty-one patients with acute myocardial infarction, 23
men and 8 women with a mean age of 66 years (66.0
11.8; range, 37 to 82), underwent emergent CABG be-
tween January 1998 and June 2004 at Nayoro City General
Hospital. They all had already performed coronary an-
giograms by cardiovascular internist, and the decision for
emergent operation was made before transferring to our
department. The indications for emergent operation
were failed or unsuccessful precutaneous coronary inter-
vention in 16 patients, concomitance with a severe ste-
nosis or aneurysm of the left main coronary artery in 7
patients, and severe three-vessel disease in 8 patients.
Although the culprit vessel was the left anterior descend-
ing artery in all cases, there were 2 cases concomitant
with the infarction of right coronary artery area from the
findings of the electrocardiogram. The electrocardiogram
indicated ST-segment elevation in leads V1 to V4 in all
cases, and 2 cases were associated with ST-segment
elevation in leads II and III and aVf.
In 16 patients, between 1998 and 2001, on-pump sur-
gery was performed on the arrested heart; in 15 patients,
between 2002 and 2004, on-pump surgery was done on
the beating heart. The same surgeon performed the
CABG in all cases.
0003-4975/06/$32.00
doi:10.1016/j.athoracsur.2005.08.036
 574 IZUMI ET AL Ann Thorac Surg
ON-PUMP BEATING-HEART CABG 2006;81:573 6

Cardiopulmonary bypass was instituted by canulation Table 1. Preoperative Characteristics

of the ascending aorta and the right atrium, and a
Arrested Beating
standard circuit with a hollow fiber membrane oxygen- CABG CABG
ator and a centrifugal blood pump were used in both (n 16) (n 15) p Value
groups. In the on-pump arrested-heart CABG, after car-
CARDIOVASCULAR

Age 66.4 13.1 65.5 10.7 0.845

diopulmonary bypass was established, aortic cross clamp
Sex 11:5 12:3 0.685
and cardiac arrest were induced with Youngs solution.
DM 5 (31.3%) 7 (46.6%) 0.609
Myocardial protection was achieved by means of inter-
LMTD 8 (50.0%) 7 (46.6%) 0.862
mittent antegrade and retrograde cold blood cardioplegia
after every anastomosis. A terminal warm shot was IABP 11 (68.9%) 10 (66.7%) 1.000
administered before declamping of the aorta at finishing PCPS 1 (6.3%) 1 (6.7%) 1.000
of all distal anastomoses as a rule. In patients with CPR 3 (18.8%) 2 (13.3%) 1.000
aneurysm of the left main coronary trunk, declamping of CKMB (IU/L) 601116 60680
the aorta was performed after transaortic suture of the Median 154 105
left coronary ostium and proximal anastomosis on the Mean SD 260 307 199 184 0.536
ascending aorta. Patient body temperature was main- Duration to operation (hrs) 9.4 3.7 10.4 3.4 0.456
tained at 28C. CKMB creatine kinase myocardial band; CPR, cardiopulmonary
In the on-pump beating-heart CABG, total cardiopul- resuscitation; DM diabetes mellitus; IABP intra-aortic bal-
monary bypass was established, and body temperature loon pump; LMTD left main trunk disease; PCPS percutane-
ous cardiopulmonary support.
was maintained at 36C. Bypass grafting surgery was
performed on the beating heart using a heart stabilizer
and heart positioner. The left anterior descending artery
was the first target vessel, the right coronary artery status, creatine kinase myocardial band (CK-MB) values,
or duration between onset and operation.
system the next, and the circumflex system the last.
Three patients in the on-pump arrested-heart CABG
During anastomosing of the next graft, the bypass grafts
group and 2 in the on-pump beating-heart CABG group
that had already been sutured were unclamped and
required cardiopulmonary resuscitation or temporary
opened.
pacing because of cardiopulmonary arrest or ventricular
Urine output was measured every hour in the intensive
tachycardia before transferring. Before reaching our de-
care unit, and intravenous continuous infusion of the
partment, 11 and 10 patients in each group, respectively,
mannitol and furosemide cocktail was given to the pa-
were assisted by intra-aortic balloon pumping (IABP);
tient with low urine output below 0.5 mL/kg an hour
and a patient in the on-pump arrested-heart group was
lasting for 3 hours. Continuous hemofiltration started
supported by percutaneous cardiopulmonary support
when urine output did not increase or renal shutdown (PCPS). Cardiopulmonary internists made the decision of
occurred in spite of mannitol and furosemide. Hemodi- indication for them because of hypotension and low
alysis was introduced after the patients hemodynamic output heart. Only 1 patient was conducted on PCPS
state improved. The condition requiring hemofiltration after reaching our hospital, and was in the on-pump
or hemodialysis was defined as postoperative renal fail- beating-heart CABG group. The mean overall number of
ure in this study. distal anastomosis was 2.3 0.7 versus 2.5 0.5 in the
We analyzed the clinical results, mortality rate, mor- on-pump arrested-heart CABG and on-pump beating-
bidity rate, and blood chemistry data between the two heart CABG groups, respectively (p not significant).
groups retrospectively. Although there were no significant differences in the
target vessels between the two groups, arterial grafts
Informed Consent were utilized in the on-pump beating-heart CABG group
The Institutional Review Board of Nayoro City General more than in the on-pump arrested-heart CABG group (p
Hospital approved this study, and waived the individual 0.003; Table 2.).
consent because this study was retrospective. The cardiopulmonary bypass time in the on-pump
arrested-heart CABG group was longer, but the differ-
Statistical Analysis ence was not statistically significant. Blood loss volume
Results are expressed as mean SD. Statistical analysis was also similar between the two groups (Table 3). There
comparing the two groups was performed with an un- was a patient in each group required immediately intro-
paired Students t test for the means or with a 2 test for duction of cardiopulmonary bypass before graft take-
the variables. Probability values less than or equal to 0.05 down because of unstable hemodynamic state.
were considered significant. The postoperative clinical data for the on-pump beat-
ing-heart CABG group suggest better results than do
those for the on-pump arrested-heart CABG group. Post-
Results
operative maximum CK-MB values in the on-pump ar-
The preoperative patient data analysis (Table 1) between rested-heart CABG group were 666 540 IU/L, whereas
the two groups demonstrated no significant differences those in the on-pump beating-heart CABG group were
in mean age, sex, comorbidities, clinical preoperative statistically lower at 221 200 IU/L (p 0.008), despite
Ann Thorac Surg IZUMI ET AL 575
2006;81:573 6 ON-PUMP BEATING-HEART CABG

Table 2. Target Vessels and Bypass Grafts Table 4. Clinical Results

Arrested Beating Arrested Beating
CABG CABG p Value CABG CABG p

Anastomosis 2.3 0.7 2.5 0.5 0.202 Peck CKMB (IU/L)

CARDIOVASCULAR
Target vessel Range 1282,117 60701
LAD 16 15 1.000 Median 479 145
Diagonal 1 2 0.600 Mean SD 666 540 221 200 0.008
CX 9 12 0.303 Ventilator (hrs) 108 90 63 76 0.152
RCA 10 9 0.821 Inotropic use (days) 6.8 2.8 4.8 4.2 0.223
Bypass graft Renal failure 10 (64.3%) 2 (13.3%) 0.009
ITA 5 15 IABP support (hrs) 35.8 26.6 21.4 14.9 0.195
RA 2 8 0.003 PCPS support 3 (18.8%) 3 (20.0%) 0.654
SVG 30 14 PCPS weaning 0 (0.0%) 1 (33.3%) 0.273
Mortality 5 (31.3%) 2 (13.3%) 0.394
CX circumflex artery; ITA internal thoracic artery; LAD left
anterior descending artery; RA radial artery; RCA right CKMB creatine kinase myocardial band; IABP intra-aortic bal-
coronary artery; SVG saphenous vein graft. loon pump; PCPS percutaneous cardiopulmonary support.

the similar preoperative values between the two groups.
The durations of use for ventilator and inotropic agents
in the on-pump beating-heart CABG group were shorter
than for the on-pump arrested-heart CABG group (p
0.152, p 0.223). Postoperative renal failure (requiring
hemodialysis or hemofiltration) appeared to have oc-
curred frequently in the on-pump arrested-heart CABG
(p 0.009). There were 5 cardiac-related deaths in the
on-pump arrested-heart CABG group and 2 in the on-
pump beating-heart CABG group, making the mortality
rate lower in the on-pump beating-heart CABG group (p
0.394; Table 4).
Comment
The goals of treatment for acute myocardial infarction are
saving life and preserving myocardial function by
prompt revascularization of coronary artery. Percutane-
ous catheter intervention techniques have recently be-
come more popular and are now considered to be the
first-choice treatment for acute myocardial infarction.
However, there are some cases that require surgical
revascularization, especially in patients who have done
into in cardiogenic shock. The early results of emergent
conventional CABG for acute myocardial infarction have
generally been reported to be poor [13]. These poor
results can possibly be explained not only by the pa-
tients poor status, but also by operative procedures
associated with extracorporeal circulation and cardiople-
Table 3. Operation Data
Arrested Beating
CABG CABG
CPB (min) 164.7 69.8 130.7 59.8
Aortic cross clamp (min) 56.5 21.5 0 hypoperfusion. Although renal failure requiring dialysis
Blood loss (mL) 1,091 547 1,109 436
Graft takedown
after CPB introduction 1/16 (6.3%) 1/15 (6.7%)
CPB cardiopulmonary bypass.
gic arrest, which induce ischemic injury and reperfusion
injury to the myocardium. Although off-pump CABG has
some advantages [4 6], it is not always technically feasi-
ble. Beginning in 2001, as an alternative to conventional
CABG with cardioplegic arrest, we have conducted
CABG based on maintenance of a beating heart with
cardiopulmonary bypass but without aortic cross-
clamping for acute myocardial infarction.
It has been demonstrated that keeping the heart beat-
ing is associated with less myocardial edema and better
function [7]. As cardioplegic arrest can trigger intense
inflammatory responses, aortic cross-clamping should be
avoided in unstable patients in acute myocardial infarc-
tion. The avoidance of aortic cross-clamping and cardiac
arrest can contribute to myocardial protection and elim-
inate intraoperative global myocardial ischemia. Our
data show that CK-MB values in the on-pump beating-
heart CABG are significantly lower than those in the
on-pump arrested-heart CABG. It has previously been
demonstrated that in the on-pump beating-heart CABG
is a lower release of troponin Ic, which is a highly cardiac
specific marker of tissue damage [8]. In the present study,
the operative mortality was lower in the on-pump beat-
ing-heart CABG than the on-pump arrested-heart
CABG, although the difference was not significant.
Renal failure is a frequent complication in patients who
have experienced hemodynamic failure or undergone
conventional cardiac surgery. Ascione and associates [9]
have demonstrated that beating-heart CABG offers a low
risk of systemic hypoperfusion during surgery and as a
consequence superior renal protection, as demonstrated
by a low incidence of postoperative renal complications.
Prifti and colleagues [10] have also indicated that this
p Value technique offers better renal protection associated with
0.162 fewer postoperative complications due to intraoperative
0.918 or hemofiltration was identified postoperatively in both
groups in our series, the incidence was significantly
lower in the on-pump beating-heart CABG group.
1.000
Another point to consider is the technical features of
the on-pump beating-heart CABG technique. When
 576 IZUMI ET AL
ON-PUMP BEATING-HEART CABG
fully assisting the heart with cardiopulmonary bypass,
it is advantageous to place the graft on the circumflex
artery area easier owing to the reduction in heart
volume compared with off-pump CABG. Utilizing a
heart stabilizer and positioner, which are routinely
CARDIOVASCULAR
used in off-pump CABG, optimal exposure in the
posterior and lateral fields of the heart can be obtained
even in the beating heart. Borowski and colleagues [11]
have indicated that the on-pump beating-heart tech-
nique can help avoid extreme upward retraction of the
heart during revascularization of the circumflex artery
branch, thus contributing to better myocardial protec-
tion. Although the likelihood of incomplete revascular-
ization tends to increase in off-pump CABG [12], we
can achieve complete revascularization with the on-
pump beating-heart method.
Our results lead us to believe that on-pump beating-
heart CABG offers the possibility to eliminate intraoper-
ative global myocardial ischemia and is an acceptable
surgical option for acute myocardial infarction associated
with lower postoperative mortality and morbidity.
Limitations
Several limitations of this study need to be addressed.
First, our investigation was retrospective. Second, as the
patients undergoing conventional CABG were from a
former series, whereas the beating-heart CABG was
conducted during the most recent decade, a historical
difference exists; however, the operations were per-
formed by the same surgeon. Third, the number of
patients in each group was small. Further studies will be
necessary and are expected.
References
1. Curtis JJ, Walls JT, Salam NH, et al. Impact of unstable
angina on operative mortality with coronary revasculariza-
Ann Thorac Surg
2006;81:573 6
tion at varying time intervals after myocardial infarction.
J Thorac Cardiovasc Surg 1991;102:86773.
2. Applebaum R, House R, Rademaker A, et al. Coronary artery
bypass grafting within thirty days of acute myocardial in-
farction: early and late results in 406 patients. J Thorac
Cardiovasc Surg 1991;102:74552.
3. Quigley R, Milano CA, Smith LR, et al. Prognosis and
management of anterolateral myocardial infarction in pa-
tients with severe left main disease and cardiogenic shock:
the left main shock syndrome. Circulation 1993;88:6570.
4. Angelini GD, Taylor FC, Reeves BC, Ascione R. Early and
midterm outcome after off-pump and on-pump surgery in
Beating Heart Against Cardioplegic Arrest Studies (BHA-
CAS 1 and 2): a pooled analysis of two randomized con-
trolled trials. Lancet 2002;539:1194 9.
5. Hazama S, Eishi K, Yamachika S, et al. Inflammatory re-
sponse after revascularization: off-pump versus on-pump.
Ann Thorac Cardiovasc Surg 2004;10:90 6.
6. Wan IY, Arifi AA, Wan S, et al. Beating heart revasculariza-
tion with or without cardiopulmonary bypass: evaluation of
inflammatory response in a prospective randomized study.
J Thorac Cardiovasc Surg 2004;127:1624 31.
7. Mehlhorn U, Allen SJ, Adams DL, Davis KL, Gogola GR,
Warters RD. Cardiac surgical conditions induced by
-blockade: effect on myocardial fluid balance. Ann Thorac
Surg 1996;62:14350.
8. Perrault LP, Menasche P, Peynet J, et al. On-pump, beating-
heart coronary artery operations in high-risk patients: an
acceptable trade-off? Ann Thorac Surg 1997;64:1368 73.
9. Ascione R, Lloyd CT, Underwood MJ, Gomes WJ, Angelini
GD. On-pump versus off-pump coronary revasculariza-
tion: evaluation of renal function. Ann Thorac Surg 1999;
68:493 8.
10. Prifti E, Bonacchi M, Frati G, et al. Beating heart myocardial
revascularization on extracorporeal circulation in patients
with endstage coronary artery disease. Cardiovasc Surg
2001;9:608 14.
11. Borowski A, Korb H. Myocardial infarction in coronary
bypasss surgery using on-pump, beating heart technique
with pressure- and volume-controlled coronary perfusion.
J Card Surg 2002;17:272 8.
12. DiMauro M, Iaco AL, Contini M, et al. Reoperative coronary
artery bypass grafting: analysis of early and late outcomes.
Ann Thorac Surg 2005;79:817.