Interactive CardioVascular and Thoracic Surgery 23 (2016) 310–313 BEST EVIDENCE TOPIC – ADULT CARDIAC

doi:10.1093/icvts/ivw122 Advance Access publication 11 May 2016

Cite this article as: Rogers L, Cochrane E, Blundell D, Zakkar M. What is the optimum method of weaning intra-aortic balloon pumps? Interact CardioVasc Thorac
Surg 2016;23:310–3.

What is the optimum method of weaning intra-aortic balloon pumps?

Luke Rogersa, Elliott Cochraneb, David Blundellb and Mustafa Zakkara,*
a
Department of Cardiothoracic Surgery, Bristol Heart Institute, Bristol, UK
b
Department of Cardiac Surgery, Leeds General Infirmary, Leeds, UK

* Corresponding author. Bristol Heart Institute, Bristol Royal Infirmary, Bristol BS2 8HW, UK. Tel: +44-117-923-0000; e-mail: mustafazakkar@me.com (M. Zakkar).

Received 28 August 2015; received in revised form 29 March 2016; accepted 9 April 2016

Downloaded from https://academic.oup.com/icvts/article/23/2/310/2237833 by guest on 03 March 2024

Abstract
A best evidence topic in cardiac surgery was written according to a structured protocol. The question addressed was: is weaning an
intra-aortic balloon pump by volume superior to ratio reduction in terms of failure of weaning, inotropic support and haemodynamic
parameters? A total of 667 papers were identified as a result of the search described below. Six papers were relevant to the question asked.
The author, date and country of publication, patient group studied, study type, relevant outcomes, results and study weaknesses of the
papers are tabulated. Little published evidence exists, although weaning by ratio is more common particularly in high-volume centres. The
published data highlight the heterogeneity of weaning protocols not only between countries but also between hospitals in the same
country. Current evidence is unable to establish any difference in clinical outcomes including mortality, reinsertion of intra-aortic balloon
pumps and requirement for inotropic support between weaning by ratio, volume weaning and abrupt cessation. Despite this, the only ran-
domized trial demonstrates improved haemodynamic profiles in those weaned by volume weaning. In addition, given the difficulty in
obtaining clear clinical outcomes, N-terminal pro-brain natriuretic peptide and central venous oxygen saturation may be useful as surro-
gate markers for successful weaning.
Keywords: Intra-aortic balloon pump • Weaning • Inotropic support • Cardiac surgery

INTRODUCTION IABP OR counterpulsation] AND [weaning OR ratio OR volume]

AND [cardiac surgery OR valve surgery OR coronary bypass OR
A best evidence topic was constructed according to a structured coronary surgery OR CABG]. The search was limited to English
protocol. This is fully described in the ICVTS [1]. articles and human studies only. In addition, the reference lists of
each publication were reviewed.

THREE-PART QUESTION
SEARCH OUTCOME
In patients with an [intra-aortic balloon pump], is weaning an
intra-aortic balloon pump by [volume] superior to [ratio reduction] A total of 667 papers were identified. Of these, six papers provided
in terms of [failure of weaning, inotropic support and haemo- the best evidence to answer the question. These are summarized
dynamic parameters]? in Table 1.

CLINICAL SCENARIO RESULTS

You are asked to remove an intra-aortic balloon pump (IABP) in a
patient in intensive care unit after emergency coronary artery
bypass, prior to sedation wean. The consultant surgeon asks you
to wean the IABP by ratio reduction, and the consultant intensivist
advises you to wean by volume weaning. You resolve to check
available evidence.
SEARCH STRATEGY
Medline from 1948 to July 2015 using the PubMed interface
[intra-aortic balloon pumping OR intra-aortic balloon pump OR
Onorati et al. [2] presented the only randomized trial data (n = 30)
comparing weaning by ratio (Group R; 1 : 2 for 4 h, followed by
1 : 3 for an 1 h, n = 15) and progressive volume weaning (Group V;
10% of total volume every hour for 5 h and then removal; n = 15),
all in postoperative cardiac surgery patients. All but one patient
(who was in Group R) were successfully weaned without the need
for reinsertion of IABP; there was no significant difference in the
amount of inotropic support or difference in troponin I; 3 patients
(all from Group R) required post-withdrawal inotropic augmenta-
tion. Furthermore; patients weaned by volume weaning demon-
strated significantly better haemodynamic profiles (cardiac index,

© The Author 2016. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.
 L. Rogers et al. / Interactive CardioVascular and Thoracic Surgery
Table 1: Best evidence papers
Author, date, journal Patient group
and country
Study type
(level of evidence)
Onorati et al. (2013), N = 30
J Thorac Cardiovasc IABP inserted for
Surg, Italy [2] perioperative low-output
cardiac syndrome
Single-centre post-cardiac surgery, the
prospective majority had a concomitant
randomized trial procedure, AVR or MVR or
(level 1b) aortic surgery
Group R: n = 15
4 h 1 : 2 ratio followed by 1-h
1 : 3 and then removal
Group V: n = 15
10% of total volume
reduction every hour for 5 h
then removal
Manohar et al. (2012), N = 429
J Interv Cardiol, Surgical and non-surgical
USA [3] patients.
Single-centre Fast weaned: 1 : 2
retrospective cohort n = 164
study
(level 2b) Gradual wean: 1 : 2 and 1 : 3
n = 180
Abrupt cessation:
n = 85
Lewis et al. (2006), 192 intensive care units were
Crit Care Resusc, surveyed about the use of
Australia [4] IABP, weaning and need for
reinsertion or increase
Multi-centre inotrope use after removal
retrospective cohort
study Patients had different cardiac
(level 2b) pathologies
54 units reported the use of
IABP
Bignami et al. (2012), Survey of 66 cardiac centres.
Ann Card Anaesth,
Italy [5] Total 40 675 cardiac
procedures
Multicentre n = 40 675 procedures
retrospective cohort
study
(level 2b)
311
Outcomes Key results Comments
Weaning failure One wean failure (P = 1.0) Small sample size
Inotrope support No difference (P = 0.753) No objective data to precisely
quantify the progressive
Haemodynamics Greater CI, ISVR, CCE and lower CVP in volume weaning achieved at
Group V each cycle
(P = 0.0001)
Troponin I No difference (P = 0.715)
Downloaded from https://academic.oup.com/icvts/article/23/2/310/2237833 by guest on 03 March 2024
Lactate Lower in Group V
(P = 0.002)
ICU stay Group R 9.0 ± 1.7 vs Group V 7.8 ± 1.9 days
(P = 0.097)
Length of stay Group R 18.3 ± 1.9 vs Group V 17.4 ± 2.7
days; (P = 0.359)
HR and MAP Weaning led to higher postoperative HR Selection bias and limitations
and MAP both at removal (P = 0.002 and inherent to retrospective
0.03, respectively) and at 2 h post-removal cohort studies
(P = 0.002 and 0.03, respectively)
Notable difference in baseline
Inotrope support No significance (no exact P-value) medications between groups
BEST EVIDENCE TOPIC
Mortality No significance between no wean (14.1%)
and wean group (11.9) (no exact P-value)
IABP
No significant differences (none in the no
Reinsertion wean group and 2 patients (0.5%) in the
wean group) (no exact P-value)
Weaning was associated with a trend
Length of stay towards a longer length of stay
(13.4 vs 15.6)
(P = 0.06)
Reinsertion Weaning by ratio tended to less frequently The survey presented a set of
require reinsertion criteria that the contributing
(P = 0.07) groups felt important for
weaning and included: blood
Need for 90% of ICUs reported increasing inotropes pressure (92%); heart rate
post-removal after balloon removal only rarely (1 : 50 (76%); pulmonary artery wedge
inotropic support patients) or occasionally (1 : 10 patients), pressure (59%); noradrenaline
while 87% of ICUs reported never needing dose (78%); adrenaline dose
to reinsert the balloon or only rarely (57%) and dobutamine dose
(57%)
Day of removal Postoperative day 1 (43%), day 2 (34%), No comparison of weaning
day 3 (23%) techniques
57% centres weaned by ratio and 34% IABP was always used together
centres weaned by volume reduction, with pharmacological inotropic
whereas the rest (9%) had different support in the majority of
methods (using the two techniques centres (66%), whereas 39% of
simultaneously, turning the IABP off centres used one of the
abruptly and using modified techniques) following agents only if
clinically required: dobutamine,
dopamine, epinephrine,
norepinephrine, enoximone
and levosimendan
Continued
312 L. Rogers et al. / Interactive CardioVascular and Thoracic Surgery

Table 1: (Continued)

Author, date, journal Patient group Outcomes Key results Comments

and country
Study type
(level of evidence)

Hsin et al. (2013), N = 85
Int J Cardiol, This cohort included patients
Taiwan [6] from coronary intensive care
unit with different cardiac
Single-centre, pathologies
prospective,
observational study All patients were on inotropic
support including dopamine,
HR and MAP No significant changes in HR and MAP Small sample size 2-h weaning
(P = 0.57 and 0.8, respectively) strategy
Successful/failed Successful: n = 75
weans Failed: n = 10
The ScVO2 of the success group showed
minimal changes (68.4 ± 8.2 to 69.0 ± 7.2

Downloaded from https://academic.oup.com/icvts/article/23/2/310/2237833 by guest on 03 March 2024

noradrenaline and to 71.0 ± 7.7%, P = 0.326) as the assist ratio
adrenaline of IABP decreased

The failure group failed to sustain ScVO2

levels which showed a trend of decrement
(72.5 ± 3.4 to 64.3 ± 4.1 to 59.6 ± 3.8%,
P < 0.0001)

Multivariate analysis demonstrated that

ScVO2 is the only significant factor
predictable of successful wean (P < 0.01)

Tokita et al. (2014), N = 30 Successful/failed Successful: n = 20 Small sample size

Am J Cardiol, The study included a cohort weans Failed: n = 10
USA [7] of patients with different Clinical course only monitored
cardiac pathologies At initiation: for 24 h post-IABP cessation
Single-centre, No significance
retrospective, 4200 [1330–9015] vs 5620 [1700 to No comparison of NT-proBNP
observation study 14 000] pg/ml (P = 0.3) levels in patients weaned by
progressive volume weaning
At cessation: or abrupt cessation
Successful
No significance 4200 [1330–9015] vs 4175
[1690–8495] pg/ml (P = 0.97).
Failed
NT-proBNP rise 5620 [1700–14 000] vs
9995 [2920–15 100] pg/ml (P = 0.008)

CI: cardiac index; ISVR: indexed systemic vascular resistance; CCE: cardiac cycle efficiency; CVP: central venous pressure; HR: heart rate; MAP: mean arterial
pressure; ScVO2: mixed venous oxygen saturation; MACE: major adverse cardiovascular event; LVEDD: left ventricular end-diastolic dimension; NT-proBNP:
N-terminal pro-brain natriuretic peptide; AVR: aortic valve replacement; MVR: mitral valve replacement.

central venous pressure, indexed systemic vascular resistance and
cardiac cycle efficiency, P = 0.001). No difference was identified in
either ICU stay or hospitalization.
Manohar et al. [3] (n = 429) retrospectively analysed clinical
outcomes (all-cause mortality, cardiac mortality, haematoma
formation, transfusion rates and systemic embolization) stratified
by a weaning protocol either 1 : 2 alone (n = 164), 1 : 2/1 : 3 coun-
terpulsation (n = 180) or abrupt cessation (n = 85). No difference
in vasopressor use or mortality was demonstrated, although those
weaned tended towards a longer length of stay rather than abrupt
cessation (P = 0.06). It is important to note that patients weaned
by 1 : 2/1 : 3 represented a more acutely ill subpopulation as a
higher proportion of patients in this group required IABP pre-
operatively to support high-risk cardiac surgery with an increased
likelihood to be on both pressors and inotropes. These patients
also had a significantly longer length of stay (P = 0.05).
Lewis et al. [4] reported the results of a five-part questionnaire
to 192 ICUs in Australia and New Zealand. The group had returns
from 116 units with only 54 units reporting the use of IABPs. The
commonest method of weaning was by ratio alone in 33 units,
whereas 9 units weaned by ratio followed by volume weaning,
6 units weaned by volume then ratio and 4 units used volume
weaning alone. They demonstrated no statistically significant dif-
ferences between methods in terms of IABP reinsertion and ino-
tropic requirement following balloon removal.
Bignami et al. [5] surveyed the practice of IABP use in 66 cardiac
centres. A total of 40 675 procedures were analysed over 12
months. No comparison between weaning techniques and clinical
outcomes was made, but the heterogeneous nature of IABP man-
agement was clearly illustrated. The survey showed that 38 centres
weaned by ratio, 23 centres weaned by volume, 2 centres used
the two techniques simultaneously, 1 centre weaned by turning
the IABP off abruptly and 2 centres used modified techniques
which were not specified in paper.
Two studies, Hsin et al. [6] and Tokita et al. [7] did not directly
compare clinical outcomes between weaning by ratio and
volume, but presented potential surrogate markers with which the
success of weaning can be predicted. Hsin studied central venous
 L. Rogers et al. / Interactive CardioVascular and Thoracic Surgery
oxygen saturation (ScVO2) as a marker for global tissue oxygen-
ation in 85 patients commenced on IABP support for the treat-
ment of acute coronary syndrome. Patients were weaned by ratio
(1 : 2 then 1 : 3 for 1 h each). The group reported that 10 patients
(11.8%) failed to wean after withdrawal of IABP and although
mean arterial pressure and heart rate were comparable between
those that failed to be weaned and those that were successful,
ScVO2 declined. As the assist ratio decreased, ScVO2 significantly
decreased (72.5 ± 3.4 to 64.3 ± 7.2 to 59.6 ± 3.8%; P < 0.0001),
whereas in those that was weaned successfully minimal change
(68.4 ± 8.2 to 69.0 ± 7.2 to 71.0 ± 7.7%; P = 0.326) was demon-
strated. It is therefore hypothesized that a down slope of ScVO2
during the decrease of assist ratio indicates a requirement for
greater support.
Tokita et al. [7] measured serum N-terminal pro-brain natriuret-
ic peptide (NT-proBNP) levels in 30 consecutive patients sched-
uled for weaning from IABP to predict the success of weaning.
Attending physicians were blinded to NT-proBNP levels at base-
line and cessation of IABP. A standard weaning protocol (1 : 1 then
1 : 2 for 6 h and then 1 : 3 for 6 h) was used in all patients and
monitoring continued for 24 h after cessation of IABP. Weaning
was successful in 20 patients and unsuccessful in 10. There was no
difference between baseline and cessation NT-proBNP levels in
the successfully weaned group (4200 [1330–9015] vs 5620 [2920–
15 100] pg/ml; P = 0.3). However, in those that were unsuccessfully
weaned, there was a significant increase in NT-proBNP (5620
[1700–14 000] vs 9995 [2920–15 100] pg/ml; P = 0.008). The re-
ceiver operating characteristics curve analysis identified an in-
crease of 15.1% in NT-proBNP as the threshold for failure to wean.
CLINICAL BOTTOM LINE
There is little published evidence comparing different methods for
weaning of IABPs. The included studies are small and are widely
heterogeneous in terms of patients, indications for support and
weaning protocols (with weaning by ratio being the commonest
method). No differences in clinical outcomes are demonstrated
313
between weaning techniques; however, improved haemodynamic
profiles are apparent with volume deflation in one randomized
study. Furthermore, the measurement of NT-proBNP and ScVO2
may provide useful surrogates for successful weaning.
Funding
This research was supported by the National Institute for Health
Research Biomedical Research Unit in Cardiovascular Disease at
the University Hospitals Bristol NHS Foundation Trust and the
University of Bristol.
Conflict of interest: none declared.
Downloaded from https://academic.oup.com/icvts/article/23/2/310/2237833 by guest on 03 March 2024
REFERENCES
[1] Dunning J, Prendergast B, Mackway-Jones K. Towards evidence-based
medicine in cardiothoracic surgery: best BETS. Interact Cardiovasc Thorac
Surg 2003;2:405–9.
[2] Onorati F, Santini F, Amoncelli E, Campanella F, Chiominto B, Faggian G
et al. How should I wean my next intra-aortic balloon pump? Differences
between progressive volume weaning and rate weaning. J Thorac
Cardiovasc Surg 2013;145:1214–21.
[3] Manohar V, Levin R, Karadolian S, Usmani A, Timmis R, Dery M et al. The
impact of intra-aortic balloon pump weaning protocols on in-hospital
clinical outcomes. J Interven Cardiol 2012;25:145–6.
[4] Lewis P, Mullany D, Courntey M, Coyer F. Australasian trends in intra-aortic
balloon counterpulsation weaning: results of a postal survey. Crit Care
BEST EVIDENCE TOPIC
Resusc 2006;8:361–7.
[5] Bignami E, Tritapepe L, Pasin L, Meroni R, Corno L, Testa V et al. A survey on
the use of intra-aortic balloon pump in cardiac surgery. Ann Card Anaesth
2012;15:274–7.
[6] Hsin H, Chen L, Lin P, Shieh J, Ao C. Central venous oxygen saturation
(ScVO2) facilitates the weaning of intra-aortic balloon pump in acute heart
failure related to acute myocardial infarction. Int J Cardiol 2013;168:
4568–70.
[7] Tokita Y, Yamamoto T, Sato N, Hosokawa Y, Munakata R, Akutsu K et al.
Usefulness of N-terminal pro-brain natriuretic peptide levels to predict
success of weaning from intra-aortic balloon pumping. Am J Cardiol 2014;
114:942–5.