Aggressive Preoperative Use of Intraaortic Balloon
Pump in Elderly Patients Undergoing Coronary
Artery Bypass Grafting
Dan E. Gutfinger, MD, PhD, Richard A. Ott, MD, Mark Miller, MD, Arthur Selvan, MD,
Michele A. Codini, MD, Hossein Alimadadian, MD, and Teresa M. Tanner
Division of Cardiothoracic Surgery, University of California Irvine Medical Center, Orange, California
Background. The use of the intraaortic balloon pump
(IABP) in patients undergoing coronary artery bypass
grafting has been traditionally associated with a high
complication rate and adverse outcomes. However, recent
reports show that many of these catastrophic outcomes
can be avoided by preoperatively placing the IABP in
high-risk patients. To further validate these reports, we
defined a set of liberal criteria for preoperative IABP
insertion and applied them to a series of elderly patients
(70 years or older) undergoing isolated coronary artery
bypass grafting.
Methods. Two hundred six consecutive patients who
underwent isolated coronary artery bypass grafting with
cardiopulmonary bypass were retrospectively reviewed.
A rapid recovery protocol emphasizing reduced cardio-
pulmonary bypass time, an anesthetic protocol for early
extubation, perioperative administration of corticoste-
roids and thyroid hormone, and aggressive diuresis was
applied to all patients. Patients who required an urgent
operation because of failed percutaneous transluminal cor-
onary angioplasty, a critical left main stenosis (70% or
greater), pronounced left ventricular dysfunction (left ven-
tricular ejection fraction 40% or less), or unstable angina
refractory to medical therapy or who required an emer-
gency reoperation received preoperative IABP support.
H istorically, use of the intraaortic balloon pump
(IABP) in patients undergoing coronary artery by-
pass grafting (CABG) has been associated with a high
complication rate and poor outcomes [1–7]. Many of the
catastrophic outcomes have been reported for patients in
whom the IABP was placed emergently during either
intraoperative or postoperative hemodynamic decom-
pensation. Several recent reports, however, have shown
that many of these complications and catastrophic out-
comes can be minimized by preoperative placement of
the IABP [6 –9].
Commonly accepted guidelines for preoperative inser-
tion of the IABP include cardiogenic shock after percu-
taneous transluminal coronary angioplasty (PTCA) and
Accepted for publication July 16, 1998.
Address reprint requests to Dr Ott, Division of Cardiothoracic Surgery,
University of California Irvine Medical Center, Bldg 53, Rte 81, 101 City
Drive South, Orange, CA 92668.
© 1999 by The Society of Thoracic Surgeons
Published by Elsevier Science Inc
Results. The 30-day mortality rate for the entire group
was 4.4%. There were 97 patients (47%) who received a
preoperative IABP (group II) in comparison with 109
patients (53%) who did not fulfill the preoperative inser-
tion criteria (group I). Patients in group II had a lower left
ventricular ejection fraction (mean, 46% versus 59%, p <
0.001) and a higher incidence of congestive heart failure
(35% versus 17%, p < 0.01) and acute myocardial infarc-
tion (37% versus 17%, p < 0.01) than patients in group I.
The average postoperative hospital length of stay for
patients in group II was slightly longer than for those in
group I (9.0 6 10.5 versus 6.0 6 3.7 days, p < 0.01).
However, there were no statistically significant differ-
ences in complication or mortality rates between the two
groups. Only 2 patients (2.2%) had complications related
to IABP insertion. Lower extremity ischemia occurred in
both patients, and both were treated successfully with
thromboembolectomy.
Conclusions. Liberal preoperative insertion of the
IABP can be performed safely in high-risk elderly pa-
tients undergoing coronary artery bypass grafting, with
results comparable to those in lower risk patients.
(Ann Thorac Surg 1999;67:610 –3)
© 1999 by The Society of Thoracic Surgeons
unstable angina refractory to medical therapy [1, 6].
Other less commonly accepted guidelines for preopera-
tive insertion of the IABP that have recently been re-
ported [7–9] include emergency cardiac reoperation, crit-
ical left main stenosis (70% or greater), and significant left
ventricular dysfunction (left ventricular ejection fraction
40% or less). These less commonly accepted guidelines
evolved from the clinical experience with intraoperative
and postoperative IABP insertion, which demonstrated
that many patients who ultimately required insertion of
an IABP were those undergoing repeat CABG or those
with critical left main stenosis or poor left ventricular
function.
In an effort to minimize the complications and cata-
strophic outcomes associated with either intraoperative
or postoperative IABP insertion, we defined a set of
liberal criteria for preoperative placement of the IABP
and applied them to a series of elderly patients (70 years
0003-4975/99/$20.00
PII S0003-4975(98)01201-6
Ann Thorac Surg GUTFINGER ET AL 611
1999;67:610 –3 PREOPERATIVE INTRAAORTIC BALLOON PUMP

or older) undergoing isolated CABG. Our intent was to Table 1. Preoperative Comorbidity and Operative Mortalitya
improve operative outcomes by avoiding delays in IABP
No IABP IABP
institution while minimizing the complications associ- Variable (n 5 109) (n 5 97) p Value
ated with IABP use. The results of this experience are
described here. Age (y) 75 6 4 76 6 5 NS
Female 52 (48) 39 (40) NS
Acute MI 18 (17) 36 (37) , 0.01
Material and Methods Congestive heart failure 18 (17) 34 (35) , 0.01
From January 1993 through September 1996, 423 patients LV ejection fraction (%) 59 6 10 46 6 14 , 0.001
underwent isolated CABG with cardiopulmonary bypass Obesity 15 (14) 11 (11) NS
(CPB) by a single surgeon (R.A.O.). Of these patients Diabetes 21 (19) 25 (26) NS
there were 206 consecutive elderly patients (70 years or Hypertension 66 (61) 51 (53) NS
older) who were retrospectively reviewed. All operations COPD 19 (17) 11 (11) NS
were performed within 24 hours of cardiac catheteriza- Symptomatic vascular disease 24 (22) 15 (16) NS
tion, unless additional time was needed to optimize the Ambulatory difficulties 8 (7) 7 (10) NS
clinical condition. After coronary catheterization, when Parsonnet score 15.0 6 6.4 21.4 6 7.3 , 0.001
the arterial sheath was still in place, a decision was made 30-day mortality 3 (2.8) 6 (6.2) NS
as to whether preoperative placement of the IABP was a
Data presented are mean value 6 standard deviation or number (%) of
necessary. Patients who required an urgent operation patients.
because of failed percutaneous transluminal coronary COPD 5 chronic obstructive pulmonary disease; IABP 5 intraaortic
angioplasty (PTCA), critical left main stenosis (70% or balloon pump; LV 5 left ventricular; MI 5 myocardial infarction;
NS 5 not statistically significant; Parsonnet score 5 risk assessment
greater), significant left ventricular dysfunction (left ven- score developed by Parsonnet and colleagues [12].
tricular ejection fraction 40% or less), or unstable angina
refractory to medical therapy or who required an emer-
gency cardiac reoperation received a preoperative IABP.
The intent was to identify a group of patients who were at bleeding, and after successful early extubation and re-
risk for perioperative cardiac decompensation and to moval of the IABP, patients were transferred to a moni-
provide additional hemodynamic stability, such that ep- tored floor and subsequently rehabilitated in the normal
isodes of low-output syndrome and subsequent end- manner.
organ dysfunction were minimized. In all instances the A database and risk assessment profile [12] were com-
IABP was inserted preoperatively using a percutaneous pleted retrospectively for each patient. Results are ex-
technique. No cutdown or transthoracic insertion meth- pressed as mean value 6 standard deviation. Compari-
ods were used. son of continuous variables was accomplished using the
Emergency cardiac reoperation refers to patients with t test, whereas categoric variables were compared with
a previous CABG requiring emergent operation second- the x2 test. Statistical significance was tested for p , 0.05.
ary to (1) failed PTCA; (2) significant lesions (70% stenosis
or greater) of the left main coronary artery or a dominant Results
coronary system that is unprotected with patent bypass
grafts; or (3) graft dependency to a dominant coronary Two hundred six consecutive elderly patients undergo-
vascular system with a significant stenosis (70% or great- ing isolated CABG with CPB were retrospectively re-
er); and (4) unstable anginal pattern. A cutoff of 70% viewed. The 30-day operative mortality rate for the entire
stenosis of the left main coronary artery was chosen series was 4.4% (9 patients). Ninety-seven patients (47%)
because more significant flow obstruction occurs with received a preoperative IABP (group II) versus 109 pa-
angiographic stenosis of 70% or greater. A left ventricular tients (53%) who did not fulfill the preoperative insertion
ejection fraction of 40% was chosen because it is the criteria (group I). Table 1 shows a comparison of the
accepted cutoff between moderate and severe left ven- preoperative comorbidity for groups I and II. Patients in
tricular dysfunction. group II had a lower left ventricular ejection fraction
A previously described rapid recovery protocol [10, 11] (mean, 46% versus 59%, p , 0.001) and a higher inci-
that emphasizes short CPB time, early extubation, peri- dence of congestive heart failure (35% versus 17%, p ,
operative steroid and thyroid administration, and aggres- 0.01) and acute myocardial infarction (37% versus 17%,
sive diuresis was applied to all patients. Attention was p , 0.01) than patients in group I. Patients in group II
directed toward reducing CPB time so that the deleteri- naturally represented a group of patients with a higher
ous effects of CPB on recovery would be minimized. A risk for perioperative cardiac decompensation. This is
combination of antegrade and retrograde cardioplegia also reflected by the higher Parsonnet risk assessment
was used, as previously described [10, 11]. Triiodothyro- score for group II compared with group I ( p , 0.001) [12].
nine was administered intraoperatively, whereas dexa- Of the 97 patients satisfying the insertion criteria, 17
methasone, thyroxine and early extubation techniques (18%) had an emergency repeat CABG, 20 (21%) had a
were used postoperatively [10, 11]. Every effort was made failed PTCA, 20 (21%) had critical left main disease with
to wean the patient from the IABP and to remove it 70% or greater stenosis, 42 (43%) had left ventricular
within 12 to 24 hours postoperatively. In the absence of dysfunction with a left ventricular ejection fraction 40%
612 GUTFINGER ET AL Ann Thorac Surg
PREOPERATIVE INTRAAORTIC BALLOON PUMP 1999;67:610 –3

Table 2. Intraoperative Variablesa extremity ischemia, which were treated successfully with
thromboembolectomy.
Variable No IABP IABP

No. of survivors 106 91

No. of grafts 3.1 6 1.2 2.8 6 0.9 Comment
Cross-clamp time (min) 33 6 12 33 6 13 Application of the IABP has traditionally been reserved
CPB time (min) 63 6 22 64 6 22 for patients who cannot be weaned from cardiopulmo-
Length of operation (min) 133 6 32 140 6 45 nary bypass and are in cardiogenic shock. Intraoperative
and postoperative IABP insertion has been disappointing
a
Data presented are mean value 6 standard deviation, unless otherwise
indicated. because of the high associated mortality rate (35% to
CPB 5 cardiopulmonary bypass; IABP 5 intraaortic balloon pump.
50%) [1, 2, 6, 7], as well as the high device-related
complication rate (10% to 30%) [3–5]. In an effort to
reduce the high morbidity related to intraoperative and
or less and 31 (32%) had isolated angina that was refrac- postoperative IABP insertion, there has been a trend
tory to medical therapy. Of the 97 patients satisfying the toward preoperative IABP insertion [6 –9]. However, it
insertion criteria, 30 (31%) satisfied more than one inser- has been extremely challenging to identify in advance
tion criterion simultaneously. No patient required either those patients who are at risk for perioperative cardiac
intraoperative or postoperative IABP insertion. decompensation and who may benefit the most from
Table 2 shows a comparison of the intraoperative preoperative IABP insertion and to minimize device-
variables. For the entire series the average number of related complications.
bypass grafts was 2.9 6 1.1, the aortic cross-clamp time In the present study we focused on elderly patients (70
was 33 6 12 minutes, and the CPB time was 64 6 27 years or older) and defined a set of liberal criteria for
minutes. There were no statistically significant differ- preoperative IABP use, so that the number of patients
ences in intraoperative variables between the two requiring intraoperative or postoperative IABP insertion
groups. would be minimized (in our study no such patients
Table 3 shows a comparison of the average postoper- required intraoperative or postoperative IABP insertion).
ative hospital length of stay (LOS), postoperative compli- The insertion criteria included patients who required an
cations and number of survivors for groups I and II. No urgent operation because of a failed PTCA or an emer-
patient in the series developed end-organ dysfunction gency repeat CABG and patients with a critical left main
characterized by renal failure requiring dialysis, adult stenosis (70% or greater), significant left ventricular dys-
respiratory distress syndrome, liver failure, or gastroin- function (left ventricular ejection fraction 40% or less), or
testinal complications. For the entire series the LOS was unstable angina refractory to medical therapy. These
7.4 6 7.8 days. The LOS for group I was 6.0 6 3.7 days insertion criteria are not unique to the present study but
versus 9.0 6 10.5 days for group II ( p , 0.01). There were have recently been used by others [8, 9]. Of 206 consec-
no statistically significant differences in the number of utive elderly patients undergoing isolated CABG with
postoperative complications and mortality between the CPB, 97 patients (47%) satisfied the insertion criteria,
two groups, which may be due to the relatively small with 30 (31%) of these 97 satisfying more than one
sample size. There were only two complications (2.2%) criterion simultaneously.
associated with IABP use. Both of these were lower Patients who received a preoperative IABP (group II)
represented higher risk patients, with a greater propor-
tion having congestive heart failure and acute myocardial
Table 3. Postoperative Complicationsa infarction than those not receiving a preoperative IABP
(group I). These patients also had a higher predicted
Variable No IABP IABP operative risk according to the Parsonnet scale (21.4 6 7.3
No. of survivors 106 91 versus 15.0 6 6.4, p , 0.001) [12]. Despite this higher
Sternal wound infection 1 (0.9) 0 predicted operative risk, there were no statistically sig-
Leg wound infection 1 (0.9) 2 (2.2) nificant differences (p . 0.05) in the operative mortality
Myocardial infarction 0 1 (1.1) (6.2% versus 2.8%) and postoperative complication rates,
Bleeding 6 (5.7) 6 (6.6) between patients who received a preoperative IABP and
New-onset atrial fibrillation 12 (11.3) 12 (13.2) those who did not. However, patients who received a
Pneumonia 3 (2.8) 4 (4.4) preoperative IABP had a longer postoperative hospital
Sepsis 0 2 (2.2) LOS than those who did not require an IABP (9.0 6 10.5
IABP complication ... 2 (2.2) versus 6.0 6 3.7 days, p , 0.01). This difference probably
Stroke 2 (1.9) 1 (1.1) does not reflect any consequence of IABP use; rather, it
LOS (d) 6.0 6 3.7 9.0 6 10.5 reflects the increased preoperative morbidity associated
with patients in group II.
a
There are no statistically significant differences between the two groups Although many of the patients may have done well
according to the chi-square test; data presented are mean value 6
standard deviation or number (%) of patients. without an IABP, four recent studies [6 –9] demonstrate
IABP 5 intraaortic balloon pump; LOS 5 postoperative hospital that patients who fulfill the insertion criteria do signifi-
length of stay. cantly worse when the IABP is not used preoperatively.
Ann Thorac Surg
1999;67:610 –3
Such studies show that intraoperative or postoperative
IABP insertion generally results in catastrophic outcomes
and is not sufficient to reverse the clinical outcome. By
avoiding delays in IABP use through preoperative inser-
tion, improved outcomes comparable to those attained in
lower risk patients can be achieved successfully.
Another important issue regarding preoperative IABP
is the low observed rate of device-related complications
compared with previous reports [4 – 6]. In our series there
were only 2 patients (2.2%) who had device-related
complications, both of which were lower extremity isch-
emia that was treated successfully with thromboembo-
lectomy. This result may further justify preoperative use
of the IABP in a large proportion of patients. Additionally
the studies by Dietl and colleagues [8] and Christenson
and associates [9] demonstrate that the extra cost in-
curred by preoperative IABP insertion is reclaimed
through avoidance of the complications and the pro-
longed postoperative hospital LOS in patients who re-
ceive the IABP intraoperatively or postoperatively.
Liberal use of the IABP preoperatively for support of
high-risk CABG remains controversial but, as demon-
strated by the present study, can be safely accomplished.
Our selection criteria were specifically designed to ag-
gressively use the IABP preoperatively in an effort to
avoid the complications and mortality associated with
intraoperative or postoperative insertion (delayed inser-
tion). It is clear from these results and others [6 –9] that
serious device-related complications can be avoided in
the preoperative period when pulsatile pressure, along
with the absence of vasoconstrictive agents, aids in un-
eventful IABP insertion. Similarly, the improved mortal-
ity in the present series compared with after intraopera-
tive and postoperative IABP placement [1, 2, 6, 7] most
likely reflects the avoidance of progressive cardiac dys-
function before insertion. Through early preoperative
IABP insertion, episodes of low-flow state with subse-
quent end-organ dysfunction are minimized [9]. There-
fore, because older and sicker patients constitute a
greater percentage of candidates for CABG, indications
GUTFINGER ET AL 613
PREOPERATIVE INTRAAORTIC BALLOON PUMP
for use and the timing of application of the IABP should
be reexamined.
We thank Pamela Dumas for her assistance.
In memory of our colleague Michele A. Codini, MD
References
1. Creswell LL, Rosenbloom M, Cox JL, et al. Intraaortic bal-
loon counterpulsation: patterns of usage and outcome in
cardiac surgery patients. Ann Thorac Surg 1992;54:11–20.
2. Christenson JT, Buswell L, Velebit V, et al. The intraaortic
balloon pump for postcardiotomy heart failure: experience
with 169 intraaortic balloon pumps. Thorac Cardiovasc Surg
1995;43:129–33.
3. Mackenzie DJ, Wagner WH, Kulber DA, et al. Vascular
complications of the intra-aortic balloon pump. Am J Surg
1992;164:517–21.
4. Makhoul RG, Cole CW, McCann RL. Vascular complications
of the intra-aortic balloon pump: an analysis of 436 patients.
Am Surg 1993;59:564– 8.
5. Busch T, Sirbu H, Zenker D, Dalichau H. Vascular compli-
cations related to intraaortic balloon counterpulsation: an
analysis of ten years experience. Thorac Cardiovasc Surg
1997;45:55–9.
6. Torchiana DF, Hirsch G, Buckley MJ, et al. Intraaortic
balloon pumping for cardiac support: trends in practice and
outcome, 1968 to 1995. J Thorac Cardiovasc Surg 1997;113:
758– 69.
7. Arafa OE, Pedersen TH, Svennevig JL, Fosse E, Geiran OR.
Intraaortic balloon pump in open heart operations: 10-year
follow-up with risk analysis. Ann Thorac Surg 1998;65:741–7.
8. Dietl CA, Berkheimer MD, Woods EL, et al. Efficacy and
cost-effectiveness of preoperative IABP in patients with
ejection fraction of 0.25 or less. Ann Thorac Surg 1996;62:
401–9.
9. Christenson JT, Badel P, Simonet F, Schmuziger M. Preop-
erative intraaortic balloon pump enhances cardiac perfor-
mance and improves the outcome of redo CABG. Ann
Thorac Surg 1997;64:1237– 44.
10. Ott RA, Gutfinger DE, Miller M, et al. Rapid recovery
following coronary artery bypass grafting: Is the elderly
patient eligible? Ann Thorac Surg 1997;63:634–9.
11. Ott RA, Gutfinger DE, Miller M, et al. Coronary artery
bypass grafting “on-pump”: Role of three-day discharge.
Ann Thorac Surg 1997;64:478– 81.
12. Parsonnet V, Dean D, Berstein A. A method of uniform
stratification of risk for evaluating the results of surgery in
acquired heart disease. Circulation 1989;79(Suppl I):I-3–12.