Propensity Case-Matched Analysis of Off-Pump
Versus On-Pump Coronary Artery Bypass Grafting
CARDIOVASCULAR
in Patients With Atheromatous Aorta
Manisha Mishra, MD, Rajneesh Malhotra, MCh, Anil Karlekar, MD, Yugal Mishra, MD,
and Naresh Trehan, MD
Departments of Cardiothoracic Surgery and Cardiac Anesthesiology, Escorts Heart Institute and Research Centre, New Delhi,
India
Background. Patients with severe atheromatous disease
of the aorta who underwent coronary artery bypass
grafting have an increased risk of stroke and death. We
hypothesize that in these high-risk patients off-pump
coronary artery bypass grafting is associated with lower
rates of stroke and mortality.
Methods. From January 1995 through June 2004, a total
of 24,107 patients underwent coronary artery bypass
grafting. Routine intraoperative transesophageal echo-
cardiography was performed in 18,501, of which 6,991
(29.0%) were found to have severe atheromatous disease
in the ascending aorta or aortic arch. Propensity matched-
pairs analysis was used to match patients undergoing
off-pump coronary artery bypass grafting (n 3,000)
with 3,000 patients undergoing conventional coronary
artery bypass grafting by age, sex, ejection fraction,
diabetes, preoperative intraaortic balloon pump, conges-
tive heart failure, chronic obstructive pulmonary disease,
acute myocardial infarction, peripheral vascular disease,
history of stroke or cerebrovascular disease, renal dis-
ease, carotid artery disease, atrial fibrillation, emergency
surgery, or previous cardiac surgery.
Results. Univariate analysis revealed decreased hospi-
N eurologic complications constitute a major cause of
morbidity and mortality after coronary artery by-
pass grafting (CABG) [1, 2]. Severe atheromatous disease
of the aorta has been identified as an independent risk
factor for both stroke and mortality [3 6]. Risk of stroke
attributable to aortic atheromatous disease is an impor-
tant consideration in the referral of patients with coro-
nary artery disease for surgical myocardial revasculariza-
tion. Several surgical techniques have been suggested to
minimize the risk of stroke in these patients [79]. The
significance of the aorta as a source of atheroembolism
and hence as a risk factor for perioperative stroke was
highlighted by Katz and colleagues [4], who found a
stroke incidence of 25% among patients with mobile
plaques of the aortic arch. Several subsequent studies
Accepted for publication March 24, 2006.
Address correspondence to Dr Mishra, Department of Cardiac Anesthe-
siology, Escorts Heart Institute & Research Centre, Okhla Rd, New Delhi
110025, India; e-mail: manishamishra@yahoo.com.
2006 by The Society of Thoracic Surgeons
Published by Elsevier Inc
tal mortality (1.4% versus 3.3%; p < 0.001) and stroke
prevalence (0.50% versus 0.97%; p 0.05) in off-pump
coronary artery bypass grafting compared with conven-
tional coronary artery bypass grafting. Multivariate anal-
ysis revealed that increased mortality was associated
with conventional coronary artery bypass grafting (odds
ratio, 2.6; p 0.001), age (odds ratio, 2.1; p 0.003), acute
myocardial infarction (odds ratio, 1.8; p 0.03), history of
stroke or cerebrovascular disease (odds ratio, 1.6; p
0.04), congestive heart failure (odds ratio, 2.1; p 0.04),
and diabetes mellitus (odds ratio, 1.9; p 0.03). Multi-
variate analysis showed that off-pump coronary artery
bypass grafting technique was the only independent
predictor of decreased stroke rate (odds ratio, 1.4; p
0.05).
Conclusions. Off-pump coronary artery bypass grafting
surgery in patients with atheromatous disease of the
aorta is associated with lower risk of stroke and death.
Routine intraoperative evaluation of the aorta is helpful
in identifying the disease and directs the appropriate
surgical technique.
(Ann Thorac Surg 2006;82:608 14)
2006 by The Society of Thoracic Surgeons
have identified severe atheromatous disease of the aorta
as an independent risk factor for both mortality and
stroke in patients undergoing coronary revascularization
[10 13].
Off-pump coronary artery bypass (OPCAB) grafting
has proven to be a feasible and safe alternative to
conventional myocardial revascularization, as it obviates
most of the perioperative and postoperative morbidity
related to on-pump CABG [14, 15]. The safety and effi-
cacy of OPCAB are well established [16, 17]. Decreased
morbidity, shorter length of hospital stay, and reduced
cost are often cited as some of the advantages of OPCAB
[18 20]. Nonrandomized comparative studies have
shown significant reduction in both mortality and major
neurologic events with OPCAB technique [13, 17, 21, 22].
A prior unmatched comparative study also demonstrated
an association between the OPCAB technique and im-
proved outcomes [5].
The routine use of intraoperative transesophageal
0003-4975/06/$32.00
doi:10.1016/j.athoracsur.2006.03.071
Ann Thorac Surg
2006;82:608 14
echocardiography (TEE) to evaluate all patients under-
going coronary revascularization for the presence of
atheromatous aortic disease has helped in identifying,
grading, and localizing the atheromatous lesions that are
associated with a high risk of perioperative neurologic
risk [2325]. We have been routinely using intraoperative
TEE at our institution for all cardiac surgical procedures
for the last 15 years. Each patient is evaluated and graded
for presence of atheromatous disease in the ascending
aorta, aortic arch, and descending aorta [2, 10, 11].
In the present study, we hypothesized that in these
high-risk patients, OPCAB is associated with lower rates
of stroke and mortality. We have used propensity case
matching to explore the relationship between surgical
revascularization technique and outcomes in the subset
of patients with aortic atheromatous disease.
Patients and Methods
From January 1995 through June 2004, a total of 24,107
patients underwent isolated CABG. Patients with associ-
ated other cardiac procedures (ie, valve surgery, left
ventricular aneurysm repair, ascending aorta replace-
ment, carotid endarterectomy) were excluded from the
study. We screened 18,501 patients intraoperatively by
TEE to identify aortic atheromatous disease, and 6,991
(29.0%) patients were found to have severe atheromatous
disease in the ascending aorta or aortic arch. The study
was approved by the institutional review board. Because
this is a retrospective study with a large number of
patients whose data had been collected over a long
duration, the institutional review board waived the re-
quirement for informed consent on the condition that the
subjects identities were hidden before any matching or
analytical procedures were performed.
The grading of atheromatous lesions of the aorta on
TEE were done according to our previously established
institutional criteria as follows [2]:
Grade I: Simple, smooth-surfaced plaques, focal increase
in echocardiographic density, and thickening of in-
tima extending less than 5 mm into the aortic lumen.
Grade II: Marked irregularity of intimal surface, focal
increase in echocardiographic density, and thicken-
ing of adjoining intima with overlying shaggy echo-
genic material extending more than 5 mm into the
aortic lumen.
Grade III: Plaques with a mobile element.
If atheromatous lesions of any grade were demon-
strated in the ascending aorta on TEE, then epiaortic scan
was routinely performed for better localization of the
lesions. Severe atheromatous disease of the aorta is
defined as grade II and III atheromatous disease of the
ascending aorta or the aortic arch, and the technique of
CABG was individualized to suit each patient for avoid-
ance of atheroembolism. Most surgical techniques that
we currently use concentrate on minimizing the direct
handling of the diseased aorta [8, 10, 11]. In the presence
of extensive atherosclerosis of the ascending aorta, the
technique of CABG was individualized to suit each
MISHRA ET AL 609
OPCAB VS CCAB IN ATHEROMATOUS AORTA
patient for avoidance of atheroembolism. In the conven-
tional coronary artery bypass grafting (CCAB) group, if
the disease was present in the proximal portion of as-
cending aorta, then a high aortic cannulation was done or
a long aortic cannula was passed beyond the left subcla-
CARDIOVASCULAR
vian artery, along with relocation of vein grafts and
cardioplegia needle to another site. Greater use was
made of arterial conduits, and a side-biting clamp was
avoided. If fewer proximal anastomoses were possible on
the diseased aorta, then more sequential arterial grafts
were done or conduits were taken piggyback on the
internal mammary arteries. If the disease was present in
the distal part of the ascending aorta or aortic arch
precluding aortic cannulation, cardiopulmonary bypass
(CPB) was established by femoral artery cannulation and
fibrillatory arrest without cross-clamping the aorta.
In the OPCAB group, we performed more total arterial
grafts with more sequential anastomoses or conduits
were taken piggyback on the internal mammary arteries.
When proximal anstomoses had to be performed on the
aorta, we used the side-biting clamp only once to mini-
mize the handling of the diseased aorta. We used special
proximal anastomotic devices to avoid application of the
side-biting clamp all together, ie, the Enclose device from
Novare Surgicals (Novare Surgical Systems Inc, Cuper-
tino, CA) was most commonly used. Hybrid procedures
were also performed in which CABG was combined with
percutaneous transluminal coronary angioplasty. The
other technique used was transmyocardial laser revascu-
larization combined with OPCAB depending on the cor-
onary anatomy and location of the atheromatous disease.
Propensity matched-pairs analysis was used to match
patients undergoing OPCAB with patients undergoing
CABG with CPB. The propensity matched-pairs analysis
is a balancing score method that attempts to correct bias
in patient selection by creating equivalent risk groups for
analysis. The propensity score is the predicted probabil-
ity of the dependent variable for each observation in the
data set. The single score (between 0 and 1) then repre-
sents the relationship between multiple characteristics
and dependent variables as a single characteristic. The
propensity score also provides the probability that pa-
tients received a particular treatment, in this case
OPCAB; patients from off-pump and on-pump groups
were matched by using this propensity score so that the
treatment outcomes could be compared.
We determined the independent factors associated
with group membership in OPCAB versus CCAB by
use of multivariate logistic regression. These factors
included age, acute myocardial infarction, history of
stroke or cerebrovascular disease, congestive heart
failure, hypertension, and diabetes mellitus. After this
parsimonious model was created, we established a
saturated model by adding other important clinical
variables. These factors included sex, renal disease,
carotid artery disease, atrial fibrillation, peripheral
vascular disease, chronic obstructive pulmonary dis-
ease, left ventricular ejection fraction less than 0.30,
urgent or emergency operation, previous cardiac sur-
gery, and patients with preoperative intraaortic bal-
 610 MISHRA ET AL Ann Thorac Surg
OPCAB VS CCAB IN ATHEROMATOUS AORTA 2006;82:608 14

Table 1. Comparison of Patient Characteristics

Characteristic OPCAB n (%) CCAB n (%) p Value

Number of patients 3,000 3,000

Mean age (y) SD 58.3 9.2 58.7 9.1 0.090
CARDIOVASCULAR

Age 70 y 384 (12.8%) 376 (12.5%) 0.786

Male sex 2,588 (86.3%) 2,623 (87.4%) 0.193
Diabetes 1,018 (33.9%) 993 (33.1%) 0.511
Hypertension 1,068 (35.6%) 1,105 (36.8%) 0.334
Hx of stroke or cerebrovascular disease 42 (1.4%) 51 (1.7%) 0.403
Carotid artery disease 70% 228 (7.6%) 234 (7.8%) 0.809
Hx of atrial fibrillation 108 (3.6%) 102 (3.4%) 0.725
Preoperative IABP 126 (4.2%) 141 (4.7%) 0.381
CHF 362 (12.1%) 345 (11.5%) 0.522
COPD 252 (8.4%) 244 (8.1%) 0.743
LVEF 0.30 852 (28.4%) 891 (29.7%) 0.280
Renal disease 43 (1.4%) 49 (1.6%) 0.599
Acute MI 69 (2.3%) 63 (2.1%) 0.660
PVD 32 (1.1%) 36 (1.2%) 0.714
Previous cardiac operation 35 (1.2%) 38 (1.3%) 0.814
Urgent or emergent operation 354 (11.8%) 375 (12.5%) 0.429

CCAB conventional coronary artery bypass grafting; CHF congestive heart failure; COPD chronic obstructive pulmonary disease; Hx
history; IABP intraaortic balloon pump; LVEF left ventricular ejection fraction; MI myocardial infarction; OPCAB off-pump
coronary artery bypass; PVD peripheral vascular disease; SD standard deviation.

loon pump for hemodynamic instability. Finally pa- Results

tients in the on-pump group were matched to patients
in the OPCAB group by use of greedy matching strat-
egy. The cohorts of 3,000 patients, who underwent
OPCAB, were matched with 3,000 CCAB patients.
Patient data was prospectively collected. Stroke was
defined as an episode of new focal or global loss of
cerebral function with symptoms lasting more than 24
hours, confirmed on computed tomography or mag-
netic resonance imaging.
Severe stenosis of carotid artery was defined as
narrowing of more than 70% of vessel cross-sectional
area. Patients with symptomatic or asymptomatic dis-
ease of more than 70% confirmed on carotid angiogra-
phy were treated with simultaneous carotid endarter-
ectomy along with CABG. These patients are not
included in the study.
Statistical analysis of categorical variables was carried
out using cross-tables with the Pearson 2 test. If the
expected values were small, Fishers exact test (two-
sided) was used. In all statistical tests, a probability value
of less than 0.05 was considered to be significant. Values
are expressed as the mean standard deviation unless
otherwise indicated. Multivariate logistic regression
analysis was performed on all independent variables
found significant by univariate analysis, including CPB,
for predicting the outcome. Stepwise regression analysis
was done separately for hospital mortality and neuro-
logic injury with background elimination method. A
variable is entered into the model if the probability is less
than 0.05, and is removed if the probability is greater than
0.1. Statistical analysis was performed with the statistical
software SPSS 13.0 (SPSS, Inc, Chicago, IL).
Of the 6,991 (29.0%) with severe atheromatous disease of
the aorta, 3,105 underwent OPCAB, whereas the remaining
3,886 had CCAB surgery. The cohorts of 3,000 patients who
underwent OPCAB were matched with 3,000 CCAB
patients.
The demographic profile of the patients and the risk
factors were fully matched and balanced, with a propen-
sity score difference of 0.005 between the two groups.
There were almost 87% males, one third of the patients
had diabetes mellitus, poor left ventricular ejection frac-
tion (0.30) was present in 28.4% in the OPCAB versus
29.7% in the CCAB group, and 12.6% were older than 70
years of age. Previous history of stroke or cerebrovascular
disease was present in 1.4% in the OPCAB group and
1.7% in the CCAB group, whereas 354 in the OPCAB and
375 in the CCAB group underwent an urgent or emergent
surgery. Carotid artery disease was present in nearly
7.7% of patients in both groups (Table 1).
The intraoperative and postoperative data are shown
in Table 2. The mean number of grafts was 3.05 0.82 in
the OPCAB versus 3.24 0.84 in the CCAB group (p
0.001). The data revealed decreased hospital mortality
(1.4% in the OPCAB versus 3.3% in the CCAB group; p
0.001) and stroke prevalence (0.50% in the OPCAB versus
0.97% in the CCAB group; p 0.05). The duration of
intensive care stay and the total length of hospital stay
were significantly shorter in the OPCAB group, whereas
freedom from any major complications was significantly
higher in the OPCAB group (p 0.001).
Tables 3 and 4 summarize the results of univariate
analysis of risk factors for mortality and stroke in the two
Ann Thorac Surg MISHRA ET AL 611
2006;82:608 14 OPCAB VS CCAB IN ATHEROMATOUS AORTA

Table 2. Comparison of Intraoperative and Postoperative Data Between Groups

Variable OPCAB (n 3,000) CCAB (n 3,000) p Value

Mean bypass time (min) NA 102.8

Mean cross-clamp time (min) NA 55.8

CARDIOVASCULAR
Hospital mortality (%) 42 (1.4%) 90 (3.3%) 0.001
Number of grafts 3.05 0.82 3.24 0.84 0.001
Reoperation for postoperative bleeding (%) 64 (2.1%) 129 (4.3%) 0.001
Deep wound infection 18 (0.6%) 38 (1.3%) 0.011
Renal failure (%) 39 (1.3%) 66 (2.2%) 0.010
Prolonged ventilation 24 hours (%) 132 (4.4%) 234 (7.8%) 0.001
GI bleeding 29 (0.97%) 33 (1.10%) 0.702
Stroke 15 (0.50%) 29 (0.97) 0.05
ICU stay in hours (mean SD) 20 7 32 8 0.001
Hospital stay in days (mean SD) 62 83 0.001
Complication free (%) 2,838 (94.6%) 2,646 (88.2%) 0.001

CCAB conventional coronary artery bypass grafting; GI gastrointestinal; ICU intensive care unit; OPCAB off-pump coronary artery
bypass grafting; SD standard deviation.

groups. The significant risk factors for hospital mortality
were revealed to be associated renal disease, previous
history of stroke or cerebrovascular disease, history of
atrial fibrillation, left ventricular ejection fraction less
than 0.30, congestive heart failure, acute myocardial
infarction, age 70 years and older, male sex, and diabetes
mellitus. The significant risk factors for neurologic injury
were renal disease, acute myocardial infarction, age 70
years and older, and history of previous cardiac surgery.
Multivariate analysis revealed that increased mortality
was associated with CCAB (odds ratio, 2.6; p 0.001), age
(odds ratio, 2.1; p 0.003), acute myocardial infarction
(odds ratio 1.8; p 0.03), history of stroke or cerebrovas-
cular disease (odds ratio, 1.6; p 0.04), congestive heart
failure (odds ratio, 2.1; p 0.04), and diabetes mellitus
(odds ratio, 1.9; p 0.03).
Multivariate analysis showed that OPCAB technique
was the only independent predictor of decreased stroke
rate (odds ratio, 1.4; p 0.05; Table 5).
Comment
The present study shows that the OPCAB technique in
patients with severe atheromatous disease is associated
with lower risk of hospital mortality, stroke, and other
complications as compared with the group of patients
undergoing CCAB. Previous studies have shown OPCAB to
be safe and effective with early outcome comparable to
CCAB. However, most studies have highlighted the use of
OPCAB technique in the high-risk group of patients (ath-
eromatous aorta, octogenarians, renal impairment, chronic
obstructive pulmonary disease) in which no significant
difference in perioperative mortality between the OPCAB
and CCAB groups could be demonstrated [26 29].

Table 3. Risk Factor Analysis for Hospital Mortality

Risk Factor Yes (n 132) No (n 5,868) p Value

Renal disease 5 (3.8%) 87 (1.5%) 0.051

History of stroke or cerebrovascular disease 6 (4.5%) 87 (1.5%) 0.016
Carotid artery disease 70% 17 (12.9%) 445 (7.6%) 0.364
History of atrial fibrillation 9 (6.8%) 201 (3.4%) 0.041
PVD 3 (2.3%) 63 (1.1%) 0.177
LVEF 0.30 49 (37.1%) 1,694 (28.9%) 0.049
CHF 35 (26.5%) 672 (11.4%) 0.001
Acute myocardial infarction 8 (6.1%) 124 (2.1%) 0.008
Urgent/emergency operation 19 (14.4%) 710 (12.1%) 0.788
Age 70 y 38 (28.8%) 722 (12.3%) 0.001
COPD 14 (10.6%) 482 (8.2%) 0.408
Preoperative IABP 5 (3.8%) 262 (4.5%) 0.067
Male sex 108 (81.8%) 5,103 (87.0%) 0.034
Diabetes 58 (43.9%) 1,893 (32.3%) 0.006
Previous cardiac surgery 4 (3.0%) 69 (1.2%) 0.076

CHF congestive heart failure; COPD chronic obstructive pulmonary disease; IABP intraaortic balloon pump; LVEF left ventricular
ejection fraction; PVD peripheral vascular disease.
 612 MISHRA ET AL Ann Thorac Surg
OPCAB VS CCAB IN ATHEROMATOUS AORTA 2006;82:608 14

Table 4. Risk Factor Analysis for Neurologic Injury

Risk Factor Yes (n 44) No (n 5,956) p Value

Renal disease 3 (6.8%) 89 (1.5%) 0.029

PVD 2 (4.5%) 66 (1.1%) 0.088
CARDIOVASCULAR

LVEF 0.30 18 (40.9%) 1,725 (29.0%) 0.116

CHF 7 (15.9%) 700 (11.8%) 0.537
Acute myocardial infarction 5 (11.4%) 127 (2.1%) 0.003
Urgent/emergency operation 6 (13.6%) 723 (12.1%) 0.851
Carotid artery disease 70% 5 (11.4%) 457 (7.8%) 0.248
History of atrial fibrillation 4 (9.1%) 206 (3.5%) 0.066
Age 70 y 11 (25.0%) 749 (12.6%) 0.025
COPD 4 (9.1%) 492 (8.3%) 0.782
Preoperative IABP 2 (4.5%) 265 (4.4%) 0.730
Male sex 38 (86.4%) 5,173 (86.8%) 0.898
Diabetes 19 (43.2%) 1,932 (32.4%) 0.175
Previous cardiac surgery 3 (6.8%) 70 (1.8%) 0.016

CHF congestive heart failure; COPD chronic obstructive pulmonary disease; IABP intraaortic balloon pump; LVEF left ventricular
ejection fraction; PVD peripheral vascular disease.

Magee and colleagues [14] reviewed two large data- stroke, and myocardial infarction) for patients who had
bases from two institutions to demonstrate that elimina- OPCAB (odds ratio, 0.48; 95%, confidence interval, 0.21 to
tion of CPB improves early survival in multivessel CABG. 1.09; p 0.08). There was no heterogeneity with respect
They had 6,466 patients who underwent CCAB and 1,983 to the primary end points among all trials [31]. Puskas
had OPCAB; the mortality was 3.5% versus 1.8% in the and associates [32] in a recent preoperative randomized
two groups, respectively. Another study compared the comparison of 200 unselected patients undergoing
two groups, off-pump and on-pump CABG surgery, by OPCAB versus CCAB found the 30-day mortality and
univariate analysis for risk factors and postoperative stroke rate to be similar in both groups.
complications, and predicted risk was determined by The A study by Cleveland and colleagues [18] suggests
Society of Thoracic Surgeons risk algorithm. There was a OPCAB reduces risk-adjusted postoperative mortality
significant difference in the observed mortality in the compared with CCAB across all risk groups. They stud-
OPCAB and CCAB groups, 1.9% versus 3.5% [30]. The ied a total of 126 experienced centers, which performed
results of our study are comparable for both groups, with 118,140 total CABG procedures. The use of OPCAB was
a mortality of 1.4% in our off-pump group versus 3.3% in associated with a decrease in risk-adjusted operative
the on-pump group. mortality from 2.9% with CCAB to 2.3% with OPCAB (p
A published meta-analysis of randomized trials com- 0.001). The use of an off-pump procedure decreased the
paring OPCAB and CCAB showed a trend toward a risk-adjusted major complication rate from 14.15% with
reduction in the risk of composite end points (death, conventional CABG to 10.62% in the off-pump group (p
0.001). Patients receiving OPCAB were less likely to die
Table 5. Multivariate Analysis of Hospital Mortality and (adjusted odds ratio, 0.81; 95% confidence interval, 0.70 to
Neurologic Injury 0.91) and less likely to have major complications (ad-
justed odds ratio, 0.77; 95% confidence interval, 0.72 to
Risk Factor Odds Ratio 95% CI p Value 0.82) [18].
Hospital mortality The importance of the aorta as a source of emboli has
CCAB 2.6 2.13.3 0.001 become apparent only since the advent of TEE. This
Age 70 y 2.1 1.72.7 0.003 technique has made possible high-resolution imaging of
Acute myocardial 1.8 1.42.2 0.03 the atherosclerotic aortic wall in great detail. We have
infarction developed elaborate screening techniques to detect pre-
History of stroke 1.6 1.22.3 0.04 operatively and intraoperatively some of the known
or cerebrovascular lesions and factors that may cause perioperative stroke.
disease We have demonstrated in our previous studies a drastic
Congestive heart 2.1 1.62.8 0.04 reduction in stroke rate with the routine use of intraop-
failure
erative TEE and with appropriate modification of surgical
Diabetes mellitus 1.9 1.42.7 0.03 technique [2, 8].
Neurologic injuries Although the proportion of strokes caused by aortic
CCAB 1.4 1.12.3 0.05 atheroemboli rather than concomitant cerebral athero-
CCAB conventional coronary artery bypass grafting; CI confi- sclerotic disease has not been clearly defined, it has been
dence interval. demonstrated in a small number of patients with severe
Ann Thorac Surg
2006;82:608 14
aortic atherosclerosis that a decrease in perioperative
stroke incidence can be affected by modifying cross-
clamping, cannulation, and graft anastomosis techniques
that specifically respect the embolic potential of aortic
plaques [10]. In a recent collective review of more than
35,000 patients, the stroke rate ranged from 0.9% to 3.9%
after isolated CABG, with a mean stroke rate of 2%. The
mortality from stroke in the review was 13% to 25% [33].
Stamou and coworkers [34] studied the stroke rate in
propensity-matched groups of off-pump and on-pump
patients undergoing CABG. Patients undergoing CCAB
were 1.8 (95% confidence interval, 1.0 to 3.1; p 0.3) times
more likely to suffer a stroke postoperatively than
OPCAB patients after controlling for preoperative risk
factors through matching.
Sharony and associates [12] demonstrated outcomes of
OPCAB and CCAB in patients with severe atheromatous
aortic disease by propensity case-match methods. The
study demonstrated a significantly lower prevalence of
hospital mortality, perioperative stroke, and overall com-
plications in the OPCAB group. Multivariable analysis of
preoperative risk factors showed that increased hospital
mortality was associated with CCAB (odds ratio, 2.7; p
0.01), fewer grafts (p 0.05), acute myocardial infarction
(odds ratio, 11.5; p 0.001), chronic obstructive pulmo-
nary disease (odds ratio, 2.4; p 0.03), previous cardiac
surgery (odds ratio, 10.2; p 0.05), and peripheral vas-
cular disease (odds ratio, 2.1; p 0.05). Cardiopulmonary
bypass was the only independent risk factor for stroke
(odds ratio, 3.6; p 0.03). In our study, the multivariate
analysis of risk factors showed that increased mortality
was associated with conventional on-pump CABG, age of
70 years or older, acute myocardial infarction, previous
history of stroke or cerebrovascular disease, peripheral
vascular disease, congestive heart failure, and diabetes,
whereas conventional on-pump CABG performed on
CPB was the only predictor of increased stroke rate in the
present study.
It is not clear whether the avoidance of CPB has an
independent advantage in reducing neurologic events,
whether it is more important just to avoid aortic manip-
ulation and proximal anastomosis, or whether elimina-
tion of CPB combined with complete avoidance of aortic
manipulation may further reduce the incidence of peri-
operative stroke [7, 9, 25]. The current study shows that
avoidance of CPB is independently associated with a
lower stroke rate when there is significant atheromatous
disease in the ascending aorta or arch. The separate effect
of a no-touch technique for the diseased ascending aorta
could not be ascertained from our data.
A recently published scientific statement from the
American Heart Association has compared the two sur-
gical procedures, OPCAB and CCAB, after reviewing
many clinical trials, including several large retrospective
analyses, meta-analyses, and randomized trials that ad-
dress the different aspects of both procedures. Although
it was difficult to reach a definitive conclusion, it showed
that patients may achieve an excellent outcome with
either type of procedure, and that individual outcomes
are likely to depend more on factors other than whether
MISHRA ET AL 613
OPCAB VS CCAB IN ATHEROMATOUS AORTA
patients underwent OPCAB or CCAB. Nonetheless, it
showed a trend in which length of hospital stay, mortality
rate, and long-term neurologic function and cardiac out-
come appear to be similar in the two groups [35].
An important limitation of the technique is the non-
CARDIOVASCULAR
randomization of patients into the two groups, although
the sample size is large. However, by using the propen-
sity score matching analysis, we attempted to minimize
bias between the two groups. Indeed, the CCAB and
OPCAB groups had an approximately equal distribution
of all significant preoperative variables [36]. Another
limitation of the study is that the OPCAB and CCAB
groups are not distributed uniformly on a yearwise basis,
as OPCAB surgery was performed more frequently in the
later years. Although we made every attempt to acknowl-
edge all clinically significant variables, this time period
also encompasses our transition from CCAB to OPCAB
along with the learning curve, and the surgical proce-
dures were performed by a group of surgeons.
We conclude that the off-pump technique of myocar-
dial revascularization in patients with atheromatous dis-
ease of the aorta is associated with a lower risk of stroke
and death. The routine use of intraoperative TEE in
evaluation of the aorta is helpful in identifying the
disease in patients who are at a higher risk for neurologic
events after CABG.
We thank Sudhir Shekhawat for providing assistance with
statistical analysis.
References
1. Gardner TJ, Horneffer PJ, Manolio TA, et al. Stroke following
coronary artery bypass grafting: a ten-year study. Ann
Thorac Surg 1985;40:574 81.
2. Trehan N, Mishra M, Dhole S, Mishra A, Karlekar A, Kohli
VM. Significantly reduced incidence of stroke during coro-
nary artery bypass grafting using transesophageal echocar-
diography. Eur J Cardiothorac Surg 1997;11:234 42.
3. Sharony R, Bizekis CS, Kanchuger M, et al. Off-pump
coronary artery bypass grafting reduces mortality and stroke
in patients with atheromatous aortas: a case control study.
Circulation 2003;108 (Suppl 2):II1520.
4. Katz ES, Tunick PA, Rusinek H, Ribakove G, Spencer FC,
Kronzon I. Protruding aortic atheromas predict stroke in
elderly patients undergoing cardiopulmonary bypass: expe-
rience with intraoperative transesophageal echocardiogra-
phy. J Am Coll Cardiol 1992;20:70 7.
5. Grossi EA, Bizekis CS, Sharony R, et al. Routine intraoper-
ative TEE identifies patients with atheromatous aortas: im-
pact on off-pump coronary artery bypass and periopera-
tive stroke. J Am Soc Echocardiogr 2003;16:7515.
6. Roach GW, Kanchuger M, Mangano CM, et al. Adverse
cerebral outcomes after coronary bypass surgery. Multi-
center Study of Perioperative Ischemia Research Group and
the Ischemia Research and Education Foundation Investiga-
tors. N Engl J Med 1996;335:1857 63.
7. Wareing TH, Davila-Roman VG, Daily BB, et al. Strategy for
the reduction of stroke incidence in cardiac surgical patients.
Ann Thorac Surg 1993;55:1400 8.
8. Trehan N, Mishra M, Kasliwal RR, Mishra A. Surgical
strategies in patients at high risk for stroke undergoing
coronary artery bypass grafting. Ann Thorac Surg 2000;70:
1037 45.
 614 MISHRA ET AL
OPCAB VS CCAB IN ATHEROMATOUS AORTA
9. Calafiore AM, Di Mauro M, Canosa C, Giammarco GD, Iaco
AL, Contini M. Early and late outcome of myocardial revas-
cularization with and without cardiopulmonary bypass in
high risk patients (EuroSCORE/6). Eur J Cardiothorac
Surg 2003;23:360 7.
10. Trehan N, Mishra M, Kasliwal RR, Mishra A. Reduced
CARDIOVASCULAR
neurological injury during CABG in patients with mobile
aortic atheromas: a five-year follow-up study. Ann Thorac
Surg 2000;70:1558 64.
11. Trehan N, Mishra M, Sharma OP, Mishra A, Kasliwal RR.
Further reduction in stroke after off-pump coronary artery
bypass grafting: a 10-year experience. Ann Thorac Surg
2001;72(Suppl):S1026 32.
12. Sharony R, Grossi EA, Saunders PC, et al. Propensity case-
matched analysis of off-pump coronary artery bypass graft-
ing in patients with atheromatous aortic disease. J Thorac
Cardiovasc Surg 2004;127:406 13.
13. Patel NC, Pullan DM, Fabri BM. Does off-pump total arterial
revascularization without aortic manipulation influence
neurological outcome? A study of 226 consecutive, uns-
elected cases. Heart Surg Forum 2002;5:28 32.
14. Magee MJ, Jablonski KA, Stamou SC, et al. Elimination of
cardiopulmonary bypass improves early survival for mul-
tivessel coronary artery bypass patients. Ann Thorac Surg
2002;73:1196 203.
15. Patel NC, Grayson AD, Jackson M, et al. The effect off-pump
coronary artery bypass surgery on in-hospital mortality and
morbidity. Eur J Cardiothorac Surg 2002;22:255 60.
16. Nathoe HM, van Dijk D, Jansen EW, et al. A comparison of
on-pump and off-pump coronary bypass surgery in low-risk
patients. N Engl J Med 2003;348:394 402.
17. Arom KV, Flavin TF, Emery RW, Kshetty VR, Janey PA,
Petersen RJ. Safety, and efficacy of off-pump coronary artery
bypass grafting. Ann Thorac Surg 2000;69:704 10.
18. Cleveland JC Jr, Shroyer ALW, Chen AY, Peterson E, Grover
FL. Off-pump coronary artery bypass grafting decreases
risk-adjusted mortality and morbidity. Ann Thorac Surg
2001;72:12829.
19. Lancey RA, Soller BR, Vander Salm TJ. Off-pump versus
on-pump coronary artery bypass surgery: a case-matched
comparison of clinical outcomes and costs. Heart Surg Fo-
rum 2000;3:20710.
20. Hernandez F, Cohn WE, Baribeau YR, et al. In-hospital
outcomes of off-pump versus on-pump coronary artery
bypass procedures: a multicenter experience. Ann Thorac
Surg 2001;72:1528 34.
21. Al-Ruzzeh S, George S, Yacoub M, Amarani M. The clinical
outcome of off-pump coronary artery bypass surgery in the
elderly patients. Eur J Cardiothorac Surg 2001;20:1152 6.
22. Al-Ruzzeh S, Nakamura K, Athanasiou T, et al. Does off-
pump coronary artery bypass (OPCAB) surgery improve the
outcome in high-risk patients?A comparative study of 1398
high-risk patients. Eur J Cardiothorac Surg 2003;23:50 5.
Ann Thorac Surg
2006;82:608 14
23. Tunick PA, Rosenzweig BP, Katz ES, Freeberg RS, Perez JL,
Kronzon I. High risk for vascular events in patients with
protruding aortic atheromas: a prospective study. J Am Coll
Cardiol 1994;23:108590.
24. Karalis DG, Chandrasekaran K, Victor MF, Ross JJ Jr, Mintz
GS. Recognition and embolic potential of intraaortic athero-
sclerotic debris. J Am Coll Cardiol 1991;17:73 8.
25. Wareing TH, Davila-Roman VG, Barzilai B, Murphy SF,
Kouchoukos NT. Management of the severely atheroscle-
rotic ascending aorta during cardiac operation. J Thorac
Cardiovasc Surg 1992;103:453 62.
26. Yokoyama T, Baumgartner FJ, Gheissari A, Capouya ER,
Panagiotides GP, Declusinet RJ. Off-pump verses on-pump
coronary bypass in high-risk subgroups. Ann Thorac Surg
2000;70:1546 50.
27. Ricci M, Karamanoukain HL, Abraham R, et al. Stroke in
octogenarians undergoing coronary artery surgery with and
without cardiopulmonary bypass. Ann Thorac Surg 2000;69:
14715.
28. Demers P, Cartier R. Multivessel off-pump coronary artery
bypass surgery in the elderly. Eur J Cardiothorac Surg
2001;20:908 12.
29. Sabik JF, Gillinov AM, Blackstone EH, et al. Does off-pump
coronary surgery reduce morbidity and morality? J Thorac
Cardiovasc Surg 2002;124:698 707.
30. Mack M, Bachand D, Acuff T, et al. Improved outcomes in
coronary artery bypass grafting with beating- hart tech-
niques. J Thorac Cardiovasc Surg 2002;124:598 607.
31. Parolari A, Alamanni F, Cannata A, et al. Off-pump versus
on-pump coronary artery bypass: meta-analysis of currently
available randomized trials. Ann Thorac Surg 2003;76:37 40.
32. Puskas JD, Williams WH, Duke PG, et al. Off-pump coronary
artery bypass grafting provides complete revascularisation
with reduced myocardial injury, transfusion requirements,
and length of stay: a prospective randomized comparison of
two hundred unselected patients undergoing off-pump ver-
sus conventional coronary artery bypass grafting. J Thorac
Cardiovasc Surg 2003;125:797 808.
33. Engleman DT, Cohn LH, Rizor RJ. Incidence of predictors of
TIAs and strokes following coronary artery bypass grafting:
report and collective review. Heart Surg Forum 1999;2:2425.
34. Stamou SC, Jablonski KA, Pfister AJ, et al. Stroke after
conventional versus minimally invasive coronary artery by-
pass. Ann Thorac Surg 2002;74:394 9.
35. Sellke FW, DiMaio JM, Caplan LR, et al. Comparing on
pump and off pump coronary artery bypass grafting. A
scientific statement from American Heart Association Coun-
cil on Cardiovascular Surgery and Anesthesia in collabora-
tion with the Interdisciplinary Working Group of Quality of
Care and Outcomes Research. Circulation 2005;111:2858 64.
36. Grunkemeier GL, Payne N, Jin R, Handy JR Jr. Propensity
score analysis of stroke after off-pump coronary artery
bypass grafting. Ann Thorac Surg 2002;74:3015.