Interactive CardioVascular and Thoracic Surgery (2019) 1–7 ORIGINAL ARTICLE

doi:10.1093/icvts/ivz235

Cite this article as: Haraldsen P, Cunha-Goncalves D, Metzsch C, Algotsson L, Lindstedt S, Ingemansson R. Sevoflurane provides better haemodynamic stability than
propofol during right ventricular ischaemia–reperfusion. Interact CardioVasc Thorac Surg 2019; doi:10.1093/icvts/ivz235.

EXPERIMENTAL
Sevoflurane provides better haemodynamic stability than
propofol during right ventricular ischaemia–reperfusion

Downloaded from https://academic.oup.com/icvts/advance-article-abstract/doi/10.1093/icvts/ivz235/5580618 by guest on 09 October 2019

a
Pernille Haraldsen , Doris Cunha-Goncalves a,*, Carsten Metzscha, Lars Algotsson a
,
Sandra Lindstedt b and Richard Ingemanssonb
a
Department of Cardiothoracic Anaesthesia and Intensive Care, Lund University, Lund, Sweden
b
Department of Cardiothoracic Surgery, Lund University, Lund, Sweden

* Corresponding author. Thorax Department, plan 8, Skane University Hospital, SUS Lund, 22185 Lund, Sweden. Tel: +46-46-171000; e-mail: doris.cunha_goncalves@med.lu.se
(D. Cunha-Goncalves).

Received 22 April 2019; received in revised form 21 August 2019; accepted 29 August 2019

Abstract
OBJECTIVES: To assess whether sevoflurane provides better haemodynamic stability than propofol in acute right ventricular (RV) ischae-
mia–reperfusion.
METHODS: Open-chest pigs (mean ± standard deviation, 68.8 ± 4.2 kg) anaesthetized with sevoflurane (n = 6) or propofol (n = 6) underwent
60 min of RV free wall ischaemia and 150 min of reperfusion. Haemodynamic parameters and blood flow in the 3 major coronary arteries
were continuously monitored. Biomarkers of cardiac ischaemia were analysed.
RESULTS: Mean arterial pressure and stroke volume decreased, whereas pulmonary vascular resistance increased equally in both groups.
Heart rate increased 7.5% with propofol (P < 0.05) and 17% with sevoflurane (P < 0.05). At reperfusion, left atrial pressure and systemic vas-
cular resistance decreased with sevoflurane. While RV stroke work (mmHg
ml) and cardiac output (l
min-1) decreased in the propofol
group (4.2 ± 1.2 to 2.9 ± 1.7 and 2.65 ± 0.44 to 2.28 ± 0.56, respectively, P < 0.05 both), they recovered to baseline levels in the sevoflurane

C The Author(s) 2019. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.
V
2 P. Haraldsen et al. / Interactive CardioVascular and Thoracic Surgery

group (4.1 ± 1.5 to 4.0 ± 1.5 and 2.77 ± 0.6 to 2.6 ± 0.5, respectively, P > 0.05). Circumflex and left anterior descending coronary artery blood
flow decreased in both groups. Right coronary artery blood flow (ml
min-1) decreased with propofol (38 ± 9 to 28 ± 9, P < 0.05), but not
with sevoflurane (28 ± 11 to 28 ± 17, P > 0.05). Biomarkers of cardiac ischaemia increased in both groups.
CONCLUSIONS: Compared to propofol, sevoflurane-anaesthetized pigs showed higher RV stroke work, cardiac output and right coronary
artery blood flow during reperfusion. These findings warrant a clinical trial of sevoflurane in RV ischaemia in humans.
Keywords: Sevoflurane • Propofol • Right ventricular • Ischaemia–reperfusion • Cardioprotection

Downloaded from https://academic.oup.com/icvts/advance-article-abstract/doi/10.1093/icvts/ivz235/5580618 by guest on 09 October 2019

decreased transpulmonary blood flow with reduced cardiac out-
ABBREVIATIONS
put (CO) and decreased coronary blood flow [13, 14]. Recently,
we described an open-chest model suitable for the study of acute
CO Cardiac output
RV free wall ischaemia and showed that, compared to historical
CVP Central venous pressure
controls, sevoflurane improved cardiac function during pro-
Cx Circumflex artery
longed RV ischaemia without reperfusion [13, 15]. In a study by
HR Heart rate
another group, isoflurane reduced haemodynamic compromise
LAD Left anterior descending artery
and improved RV recovery during RV ischaemia–reperfusion [16].
LAP Left atrial pressure
As propofol and sevoflurane are probably the anaesthetic agents
LV Left ventricular
MAP Mean arterial pressure most commonly used worldwide, a prospective, randomized
mPAP Mean pulmonary artery pressure study comparing the cardioprotective effects of these 2 agents in
PVR Pulmonary vascular resistance acute RV ischaemia–reperfusion is warranted.
RCA Right coronary artery We hypothesized that, compared to propofol, sevoflurane
RV Right ventricular would provide greater cardioprotection during acute RV ischae-
RVSW RV stroke work mia–reperfusion, which would translate into superior haemody-
SV Stroke volume namic performance and improved transpulmonary blood flow.
SVR Systemic vascular resistance We prospectively assessed the effects of these 2 anaesthetic
TnT Troponin T agents on cardiovascular function and the release of cardiac in-
jury biomarkers in pigs during acute RV ischaemia–reperfusion.

INTRODUCTION MATERIALS AND METHODS

Right ventricular (RV) ischaemia–reperfusion injury is not uncom-
mon during complex cardiac surgery. Usually, it reflects pro-
longed aortic cross-clamp time, suboptimal cardioplegia or air
emboli, and severe cases are often associated with acute life-
threatening haemodynamic dysfunction and poor short-term
outcome [1, 2]. Myocardial viability is, however, generally pre-
served in acute RV ischaemic dysfunction, and RV function usu-
ally recovers with time [3].
In patients at risk of left ventricular (LV) myocardial ischaemic
injury, cardioprotective drugs, particularly volatile anaesthetic
agents, have been used perioperatively to improve ischaemic tol-
erance and reduce reperfusion injury [4, 5]. The greatest cardio-
protection is usually achieved when volatile agents are
administered throughout the surgical procedure [6], and even
though these drugs cause dose-dependent myocardial depres-
sion [7, 8], there is evidence from clinical and experimental stud-
ies showing that several volatile agents and some agents used for
total intravenous anaesthesia protect the heart in acute LV-
ischaemia. While sevoflurane anaesthesia decreased biochemical
markers of myocardial injury and resulted in better cardiac func-
tion 1 year after coronary artery bypass [9, 10], propofol im-
proved mechanical function in ischaemic rat hearts [11]. Besides,
in a Danish review of more than 10 000 cardiac surgical proce-
dures, propofol provided a superior outcome in patients with se-
vere preoperative ischaemia and cardiovascular instability [12].
Nevertheless, little is known about the cardioprotective effects of
anaesthetic drugs during RV ischaemia–reperfusion.
Acute postischaemic RV failure is characterized by severe hae-
modynamic compromise, including reduced RV contractility and
The Animal Research Ethical Committee at Lund University ap-
proved the experiments, and the animals received care in accor-
dance with the European Guidelines for the Use of Experimental
Animals (Directive 2010/63/EU). The primary end point was a
protective effect on RV stroke work (RVSW) with preservation
of CO.
Animal preparation
Twelve Swedish Landrace pigs (68.8 ± 4.2 kg) were randomized to
receive open-label anaesthesia with fentanyl supplemented by ei-
ther propofol (n = 6) or sevoflurane (n = 6).
All animals were premedicated with 15 mg
kg-1 ketamine,
0.07 mg
kg-1 midazolam and 2 mg
kg-1 xylazine given intramus-
cularly. Anaesthesia was maintained by an infusion of
1.2 lg
kg-1
h-1 fentanyl and 0.15 mg
kg-1
h-1 pancuronium and
supplemented throughout the experiment by propofol infusion
(6 mg
kg-1
h-1) or sevoflurane inhalation (end-tidal range 1.1–
1.3%). Fluid losses were replaced by Ringer Acetate
(3 ml
kg-1
h-1).
The airway was secured via tracheostomy with a 7.5-mm tra-
cheal tube (SIMS Portex Inc., Keene, NH, USA), and the animals
were mechanically ventilated (Servo-Ventilator 300, Siemens,
Solna, Sweden). Tidal volume was 8 ml
kg-1, PEEP 5 cmH2O and
FiO2 0.3. The respiratory rate was adjusted to achieve normoven-
tilation (end-tidal CO2 4.5–5.5 kPa). Sevoflurane was given via an
anaesthetic delivery system containing a vaporizer and an anaes-
thetic reflecting filter (AnaConDa, anaesthetic conserving device,
Sedana Medical AB, Sundbyberg, Sweden). Respiratory
 P. Haraldsen et al. / Interactive CardioVascular and Thoracic Surgery 3

EXPERIMENTAL
Figure 1: Protocol timeline.

parameters, oxygenation and end-tidal gases were analysed in a Spacelabs monitor (model 90309, Spacelabs Medical Inc.,
NICO CO2 monitor (Novametrix Medical System Inc., Redmond, WA, USA). Coronary artery and aortic blood flows

Downloaded from https://academic.oup.com/icvts/advance-article-abstract/doi/10.1093/icvts/ivz235/5580618 by guest on 09 October 2019

Wallingford, CT, USA) and a VAMOS gas monitor (Dräger
Medical, Lübeck, Germany).
The right carotid artery and internal jugular vein were cannu-
lated with polyurethane catheters for invasive pressure monitor-
ing, blood sampling and intravenous administration of
anaesthetics and fluids. After a sternotomy, perivascular ultra-
sonic probes (Transonic Systems Inc., Ithaca, NY, USA) were
placed around the ascending aorta (14 mm) and the 3 main cor-
onary arteries [3-mm probes around the left anterior descending
artery (LAD), the circumflex artery (Cx) and the right coronary ar-
tery (RCA)]. Aortic blood flow was used as a measure of CO.
Through purse-string sutures, the main pulmonary artery and the
left atrial appendage were punctured and cannulated with poly-
urethane catheters for pressure monitoring. An indwelling urinary
catheter was placed via a small abdominal incision. Care was
taken to minimize the risk of bleeding during surgery. Body tem-
perature was monitored with a rectal probe, and we used a
heated operation table and plastic drapes to avoid hypothermia.
Experimental protocol
were measured by transit time technique (HT207 Medical
Volume Flowmeter, Transonic Systems Inc., Ithaca, NY, USA).
Haemodynamic variables were calculated as: LV stroke volume
(SV, ml) = CO
heart rate-1 (HR); SVR (dynes
s
cm-5) =
(MAP - CVP)
80
CO-1; PVR (dynes
s
cm-5) = (mPAP - LAP)
80
CO-1;
and RVSW (mmHg
ml) = SV
(mPAP - CVP)
0.0136, where SVR is
the systemic vascular resistance and PVR the pulmonary vascular
resistance.
Analysis of biochemical markers of cardiac
ischaemia
Venous blood samples were centrifuged for 10 min at 2200  g
and stored at -20 C. Aspartate-aminotransferase, alanine-amino-
transferase and troponin T (TnT) were measured at baseline and
the end of the experiment in a Cobas analyser (Roche
Diagnostics AG, Rotkreuz, Switzerland). The normal TnT reference
range was 0.01–0.03 ng/ml. Blood gases were analysed in a
Radiometer ABL 700-series blood gas analyser (Radiometer
Medical A/S, Copenhagen, Denmark).

The experimental overview is shown in Fig. 1.

Surgical preparation was followed by 90 min of stabilization.
Statistical analysis
After baseline measurements (T0), RV ischaemia was induced by
Based on our earlier studies, the sample size was calculated to
snaring the RCA branches perfusing the RV free wall with vessel
detect a 15% difference in CO from baseline with 80% power at a
loops. Vessel occlusion was performed in a standardized manner,
significance level of a = 0.05.
i.e. starting with occlusion of the proximal branch and progress-
Haemodynamic data were analysed using the software pack-
ing with downstream branch ligation every 5 min until all RV free
age SPSS v.25. A mixed linear model was used with the fixed
wall branches were occluded. LAD branches supplying the RV
effects (Treatment and Time) and random effect (Subject) on all
free wall were also occluded, and septal perforating branches to
the haemodynamic variables. Post hoc testing of significant effects
the interventricular septum were avoided. The induction of is-
was performed using multiple comparison adjustment. In identi-
chaemia took approximately 20 min to be established, and it was
fying which covariance matrix was most appropriate, the differ-
designed to mitigate the development of malignant arrhythmias
ence in the 2 log-likelihood was determined to compare models
that often occur with complete acute occlusion of the coronary
with different covariance matrices. Residuals were also inspected
arterial supply [17].
to assess goodness of fit. The within-group differences relative to
Further, because many antiarrhythmic drugs exhibit cardiopro-
the baseline values are presented.
tective properties, we avoided their use in this model. Thirty min
For the cardiac ischaemia biomarkers, within-group compari-
after the start of ischaemia, new measurements were taken (T1).
sons were made with the paired t-test or the Mann–Whitney test
The third set of measurements (T2) were obtained after another
when appropriate.
30 min of stable RV ischaemia when the vessel loops were re-
All data are presented as mean, standard deviation and 95%
leased for myocardial reperfusion. All haemodynamic parameters
confidence interval. A P-value <_0.05 was considered statistically
were then monitored every 30 min for the following 150 min of
significant.
reperfusion (T3–T7) when the experiment was ended by euthana-
sia with a lethal intravenous injection of potassium chloride.
RESULTS
Haemodynamic measurements
All animals survived the experiment. Blood gases remained
Heart rhythm was monitored by 5-lead ECG. Mean arterial pres- within the normal range, and there was no hypoxaemia or acid-
sure (MAP), central venous pressure (CVP), mean pulmonary ar- aemia in any of the groups (data not shown). Two animals in the
tery pressure (mPAP) and left atrial pressure (LAP) were propofol group developed ventricular tachycardia/fibrillation
continuously monitored via fluid-filled pressure transducers during ischaemia. During reperfusion, 3 animals in each
(Becton Dickinson Critical Care Systems Pte. Ltd, Singapore) on a group had ventricular tachycardia/fibrillation. All ventricular
4 P. Haraldsen et al. / Interactive CardioVascular and Thoracic Surgery

Table 1: Systemic haemodynamics

Propofol (n = 6) Sevoflurane (n = 6)
Mean SD 95% CI Mean SD 95% CI

HR, beat
min-1
T0 baseline 67 16 51–83 78 11 66–90
T1 I 30 min 67 14 52–81 76 9 66–85
T2 I 60 min 69 11 57–80 77 9 67–87

Downloaded from https://academic.oup.com/icvts/advance-article-abstract/doi/10.1093/icvts/ivz235/5580618 by guest on 09 October 2019

T3 R 30 min 71* 14 57–86 81 8 73–88
T4 R 60 min 71 14 56–85 83* 8 75–91
T5 R 90 min 71 14 56–86 86* 7 79–94
T6 R 120 min 73* 15 57–88 88* 9 78–98
T7 R 150 min 72* 15 56–87 91* 7 83–98
MAP, mmHg
T0 baseline 82 20 61–104 87 7 79–95
T1 I 30 min 80 21 58–103 79* 6 72–85
T2 I 60 min 76* 21 53–98 74* 6 68–80
T3 R 30 min 73* 17 55–91 69* 8 60–77
T4 R 60 min 70* 16 54–86 69* 7 61–76
T5 R 90 min 68* 14 53–83 67* 9 58–76
T6 R 120 min 66* 14 52–81 66* 11 55–77
T7 R 150 min 63* 13 50–77 65* 10 54–76
CVP, mmHg
T0 baseline 5 2 3–7 6 3 2–9
T1 I 30 min 6 2 4–8 5 3 2–9
T2 I 60 min 6 1 4–7 6 3 2–10
T3 R 30 min 5 1 4–6 6 3 2–11
T4 R 60 min 5 1 4–7 6 3 2–10
T5 R 90 min 5 1 4–7 5 3 1–9
T6 R 120 min 6 1 5–6 5 3 1–9
T7 R 150 min 5 1 4–7 5 3 1–9
SVR, dynes
s
cm-5
T0 baseline 2350 516 1808–2892 2476 773 1644–3287
T1 I 30 min 2787* 850 1895–3678 2679 1096 1528–3829
T2 I 60 min 2572 557 1987–3157 2555 1022 1483–3628
T3 R 30 min 2334 589 1716–2952 2277 756 1484–3070
T4 R 60 min 2233 575 1630–2837 2131* 614 1487–2775
T5 R 90 min 2212 577 1607–2818 1994* 627 1336–2653
T6 R 120 min 2132 534 1572–2692 1981* 810 1130–2831
T7 R 150 min 2104 517 1562–2646 1939* 748 1154–2724
*P-value <0.05 compared to baseline (within-group differences).
CI: confidence interval; CVP: central venous pressure; HR: heart rate; I: is-
chaemia; MAP: mean arterial pressure; R: reperfusion; SD: standard devia-
tion; SVR: systemic vascular resistance.

arrhythmias were successfully treated with electric shocks. The

cumulative energy/total number of shocks applied in each animal
were: 50 J/1, 60 J/2, 60 J/2 and 170 J/4 shocks in the propofol
group and 100 J/2, 450 J/9 and 220 J/6 shocks in the sevoflurane
group. In all animals, ST segments normalized after a few minutes
of reperfusion.

Haemodynamics and cardiac function

HR was unchanged during ischaemia but increased in both
groups during reperfusion (Table 1). With ischaemia, SV de-
creased 15–20% in both groups and remained so throughout the
experiment (Fig. 2B). Similarly, CO decreased with ischaemia, but
in the sevoflurane group, it recovered to baseline levels during
the reperfusion period (Fig. 2A). Ischaemia also caused a persis-
tent fall in MAP in all animals paralleled by a lower SVR in the
sevoflurane group but not in the propofol group (Table 1). SVR
increased in both groups at the induction of ischaemia, but this
increase did not reach statistical significance in the sevoflurane
group. During reperfusion, SVR returned to normal with propofol
Figure 2: Cardiac function: (A) CO, (B) SV, (C) RVSW. Box plot with whiskers
showing median, 5th, 25th, 75th and 95th percentile. *P-value <0.05 changes
versus baseline for propofol, †P-value <0.05 changes versus baseline for sevo-
flurane. CO: cardiac output; RVSW: right ventricular stroke work; SV: stroke
volume.
and decreased below baseline with sevoflurane (Table 1). CVP
was unchanged in both groups, whereas LAP decreased in the
sevoflurane group only (Tables 1 and 2). The changes in RVSW
resembled those observed in CO: RVSW decreased in both
groups during ischaemia, but returned to baseline level in the
sevoflurane group as reperfusion started (Fig. 2C). PVR increased
 P. Haraldsen et al. / Interactive CardioVascular and Thoracic Surgery 5

Table 2: Pulmonary haemodynamics and LAP

EXPERIMENTAL
Propofol (n = 6) Sevoflurane (n = 6)
Mean SD 95% CI Mean SD 95% CI

mPAP, mmHg
T0 baseline 13 1 11–14 14 2 12–17
T1 I 30 min 12 1 11–13 14 2 11–16
T2 I 60 min 12 1 11–13 14 2 12–16

Downloaded from https://academic.oup.com/icvts/advance-article-abstract/doi/10.1093/icvts/ivz235/5580618 by guest on 09 October 2019

T3 R 30 min 12 2 10–14 15 2 13–17
T4 R 60 min 13 2 11–14 15 2 13–17
T5 R 90 min 13 3 10–16 15 2 13–18
T6 R 120 min 13 3 10–16 15* 2 13–18
T7 R 150 min 12 3 9–16 15* 2 13–18
PVR, dynes
s
cm-5
T0 baseline 255 74 178–333 284 132 147–423
T1 I 30 min 293 99 190–397 355* 145 202–508
T2 I 60 min 302* 90 208–396 371* 108 258–484
T3 R 30 min 298 97 197–400 393* 119 268–518
T4 R 60 min 310* 85 220–399 393* 129 258–528
T5 R 90 min 365* 60 291–440 375* 111 259–491
T6 R 120 min 372* 65 291–452 372* 148 217–527
T7 R 150 min 386* 86 279–492 395* 125 264–526
LAP, mmHg
T0 baseline 4 2 2–6 5 3 1–8
T1 I 30 min 4 2 1–6 4 2 1–7
T2 I 60 min 4 3 1–7 3* 2 1–5
T3 R 30 min 4 2 2–7 4 2 1–6
T4 R 60 min 4 2 2–6 4* 2 2–5
T5 R 90 min 4 2 2–7 3* 2 2–5
T6 R 120 min 5 3 1–8 3* 2 2–5
T7 R 150 min 4 2 2–6 3* 2 1–5
*P-value <0.05 compared to baseline (within-group differences).
CI: confidence interval; I: ischaemia; LAP: left atrial pressure; mPAP: mean
pulmonary artery pressure; PVR: pulmonary vascular resistance; R: reperfu-
sion; SD: standard deviation.

with ischaemia and remained high in both groups throughout

the experiment (Table 2). mPAP remained unchanged in the pro-
pofol group, but increased in the sevoflurane group during re-
perfusion (Table 2).

Coronary blood flow

RCA blood flow decreased in the propofol group during ischae-
mia, temporarily recovered during reperfusion and dropped
again below baseline during the last 90 min of observation. In
contrast, it was preserved with sevoflurane, showing even a brief
increase above baseline at 30 min reperfusion (Fig. 3A).
In the propofol group, LAD blood flow was maintained during
ischaemia, but permanently decreased from 60 min reperfusion Figure 3: Coronary artery blood flow: (A) RCA, (B) LAD, (C) Cx. Box plot with
whiskers showing median, 5th, 25th, 75th and 95th percentile. *P-value <0.05
to the end of the experiment. In the sevoflurane group, LAD
changes versus baseline for propofol, †P-value <0.05 changes versus baseline
blood flow showed significant oscillations during ischaemia and for sevoflurane. Cx: circumflex artery; LAD: left anterior descending artery; RCA:
reperfusion, but was also lower than baseline at the end of the right coronary artery.
experiment (Fig. 3B).
Blood flow in the Cx artery showed the same pattern in both
rises earlier and is more sensitive and specific to myocardial in-
groups: it decreased during ischaemia, showed temporary recov-
jury, increased sevenfold in both groups (Table 3).
ery at some points during reperfusion, but was lower than base-
line by the end of the experiment (Fig. 3C).
DISCUSSION
Biochemical markers of cardiac ischaemia
To our knowledge, this is the first prospective and randomized
While alanine-aminotransferase remained unchanged, aspartate- in vivo study evaluating sevoflurane’s cardioprotection during
aminotransferase increased in both groups (Table 3). TnT, which acute RV ischaemia–reperfusion. Sevoflurane administered
6 P. Haraldsen et al. / Interactive CardioVascular and Thoracic Surgery

RCA blood flow decreased 25–30% in the propofol group,

Table 3: Biochemical markers of cardiac ischaemia mainly due to the persistent decrease in perfusion pressure. In
contrast, RCA blood flow was unchanged during coronary occlu-
Propofol (n = 6) Sevoflurane (n = 6) sion in the sevoflurane group and showed even a brief flow over-
Mean SD 95% CI Mean SD 95% CI
shoot above baseline at the start of reperfusion. The opening of
KATP channels is involved in sevoflurane’s preconditioning effects
ASAT, mkat/l and its potent vasodilation properties (4). Flow preservation with
T0 baseline 0.51 0.11 0.39–0.62 0.57 0.12 0.44–0.69
sevoflurane might have played a role in avoiding arrhythmias dur-
T7 end 2.48* 3.53 0.01–6.19 1.45* 0.78 0.63–2.27
ALAT, mkat/l ing coronary occlusion. Sevoflurane is a coronary vasodilator that

Downloaded from https://academic.oup.com/icvts/advance-article-abstract/doi/10.1093/icvts/ivz235/5580618 by guest on 09 October 2019

T0 baseline 0.88 0.20 0.67–1.09 1.10 0.25
T7 end 1.02 0.38 0.63–1.42 1.12 0.27
TnT, ng/ml
T0 baseline 0.10 0.09 0.01–0.19 0.09 0.07
T7 end 0.71* 0.86 0.01–1.61 0.65* 0.55
*P-value <0.05 compared to baseline (within-group differences).
ALAT: alanine-aminotransferase; ASAT: aspartate-aminotransferase; CI:
confidence interval; SD: standard deviation; TnT: troponin T.
throughout an experimental surgical procedure involving RV is-
chaemia–reperfusion protected the heart against postischaemic
RV myocardial dysfunction. In contrast to propofol, sevoflurane
restored RVSW and maintained transpulmonary and RCA blood
flow during reperfusion.
Acute RV ischaemia can cause biventricular systolic and dia-
stolic dysfunction, with decreased SV and increased myocardial
stiffness [16]; by avoiding volume overload, as in our study, filling
pressures might remain unchanged even when myocardial stiff-
ness increases. During reperfusion, although transpulmonary
blood flow was restored in the sevoflurane group, LAP decreased.
The lower LAP possibly denotes reduced preload secondary to
vasodilation, but it may also indicate attenuation of myocardial
stiffness by sevoflurane at reperfusion, which has been demon-
strated in rabbits [18]. Nevertheless, despite decreased filling
pressures, this group generated the same SV as in the propofol
group, suggesting biventricular recovery of systolic function. The
increase in HR, probably a compensatory autonomic reflex in re-
sponse to sevoflurane’s systemic vasodilation [7], may partly ex-
plain the recovery of CO (7), but RV contractility must have
increased with sevoflurane, an assumption that is supported by
the restoration of RVSW during reperfusion. Similar observations
have been made in other studies with inhalational agents. In a
rodent model of RV ischaemia–reperfusion, sevoflurane im-
proved contractility in isolated RV trabeculae [19]. In an ischae-
mia–reperfusion study in pigs anaesthetized with xenon or
isoflurane, RV contractile performance indexes derived from
pressure-volume loops from a pressure-conductance catheter
improved as RV afterload increased [16]. Provided RCA blood
flow is adequate [20], a rise in RV afterload triggers an increase in
RV contractility through a compensatory mechanism known as
homeometric autoregulation [14, 21]. Although many anaesthetic
agents exert cardioprotection, most of them also depress myo-
cardial function [7, 8]. The balance between these 2 effects
remains unknown, but our results show that, unlike propofol,
sevoflurane improves RV haemodynamic performance during RV
ischaemia–reperfusion injury.
Despite increased PVR, the failing right ventricle may present
with normal or even low mPAP depending on the magnitude of
the decrease in transpulmonary blood flow, as we have shown in
pigs with RV ischaemia [13]. In the present study, however, be-
cause RV performance and CO were restored by sevoflurane,
mPAP increased.
0.84–1.36 preserves coronary flow reserve [22], and the flow overshoot at
0.84–1.40 early reperfusion undoubtedly reflects ischaemic reactive hyperae-
mia. The right ventricle is homogeneously perfused in systole and
0.02–0.16
0.07–1.23 diastole, and despite the reduced perfusion pressure, RCA blood
flow could be maintained during reperfusion due to a decrease in
coronary vascular resistance. What is surprising is that flow
remained unchanged even during coronary occlusion, but the
RCA is almost always dominant in pigs, and besides the RV free
wall, it perfuses 25% of the LV wall mass, 42% of the septum, the
sinoatrial and atrioventricular node and the right atrium [23].
Hence, even during coronary occlusion, total RCA blood flow
could be maintained probably as a result of the coronary vasodila-
tion and flow redistribution to other branches of the RCA.
In contrast, LAD and Cx blood flow decreased by 20% in both
groups. Perfusion pressures equally decreased in both groups,
but while LAD and Cx blood flow decreased, RCA blood flow was
maintained in the sevoflurane group, indicating that different
mechanisms were involved in the reduction of blood flow in
these vessels. As the left ventricle works under high pressure, the
LV coronary arteries are externally compressed during systole,
and LV coronary flow is predominantly diastolic [24]. Further, be-
cause tachycardia has an exponential effect on the shortening of
diastole [25], it might have slightly contributed to the reduction
of LV perfusion time in the sevoflurane group. These phasic char-
acteristics of LV coronary flow physiology explain why RCA blood
flow was maintained while left-sided coronary flow decreased.
None of the drugs prevented the increase in the biomarkers of
cardiac ischaemia, i.e. aspartate-aminotransferase and TnT, which
suggests some myocardial injury. These biomarkers could also
have increased due to myocardial injury from electric defibrilla-
tion [26], but these findings are not surprising, given the com-
plexity of our model with a total ischaemic time of 60 min. We
believe that RV free wall perfusion was extensively impaired for
enough time to cause myocardial injury and cardiac enzyme
release.
Limitations
There are some important limitations to be considered. First, the
swine model is not fully translatable to human physiology, and
the small sample size of this pilot study precludes comparisons
between groups. Second, as anaesthesia requires a combination
of different agents involving complex drug interactions, it is not
possible to have a control group without anaesthesia and surgical
confounders. To mitigate these effects, our study was prospective
and randomized, and all surgery was performed by the same se-
nior cardiac surgeon. Further, to minimize myocardial depres-
sion, we maintained a stable end-tidal concentration of
sevoflurane corresponding to 0.5 MAC for a grown-up pig [27]
and an equivalent low-dose of propofol [28]. Thus, care was
taken to reproduce the clinical settings observed during cardiac
surgery, and the distinct differences found in RV haemodynamic
 P. Haraldsen et al. / Interactive CardioVascular and Thoracic Surgery
performance’s outcome strongly suggest that our findings do re-
flect greater cardioprotection by sevoflurane. Third, this open-
chest, open pericardium model is comparable to open-heart sur-
gery, but not to situations where the pericardium is intact due to
the importance of the pericardium on the interaction between
RV and LV function. Finally, for technical reasons, we did not
monitor cardiac preload and function with echocardiography,
which would have been more appropriate, as diastolic dysfunc-
tion is an essential part of the injury mechanism in ischaemia–re-
perfusion. Nonetheless, fluid administration was moderate, and
CVP was stable, suggesting that there was no significant RV dila-
tion or septal shift in our study.
Sevoflurane administered to open-chest pigs throughout RV is-
chaemia–reperfusion preserved RV contractility, increasing trans-
pulmonary and RCA blood flow. Our findings indicate that
sevoflurane supports RV function during RV ischaemia–reperfu-
sion and might contribute to the preservation of haemodynamic
stability during procedures involving RV myocardial ischaemia.
ACKNOWLEDGEMENTS
All statistical calculations were performed in cooperation with
Mathias Grahn, Klagshamn, Sweden.
Funding
This work was supported by research funds from Region Skane and
donation funds from Skane University Hospital.
Conflict of interest: none declared.
Author contributions:
Pernille Haraldsen: Conceptualization; Data curation;
Investigation; Methodology; Writing - Original Draft; Writing -
Review & Editing; Doris Cunha-Goncalves: Formal analysis;
Supervision; Visualization; Writing - Original Draft; Writing - Review
& Editing; Carsten Metzsch: Conceptualization; Data curation;
Investigation; Methodology; Writing - Review & Editing; Lars
Algotsson: Conceptualization; Methodology; Supervision; Writing -
Review & Editing; Sandra Lindstedt: Conceptualization; Data cura-
tion; Investigation; Supervision; Writing - Review & Editing; Richard
Ingemansson: Conceptualization; Data curation; Funding acquisi-
tion; Methodology; Project administration; Resources; Supervision;
Writing - Review & Editing.
REFERENCES
[1] Haddad F, Couture P, Tousignant C, Denault AY. The right ventricle in
cardiac surgery, a perioperative perspective: II. Pathophysiology, clinical
importance, and management. Anesth Analg 2009;108:422–33.
[2] Goldstein JA. Acute right ventricular infarction: insights for the interven-
tional era. Curr Prob Cardiol 2012;37:533–57.
[3] Goldstein JA. Pathophysiology and management of right heart ischae-
mia. J Am Coll Cardiol 2002;40:841–53.
[4] Weber NC, Schlack W. The concept of anaesthetic-induced cardiopro-
tection: mechanisms of action. Best practice & research. Clin
Anaesthesiol 2005;19:429–43.
[5] Landoni G, Biondi-Zoccai GG, Zangrillo A, Bignami E, D’Avolio S, Marchetti C
et al. Desflurane and sevoflurane in cardiac surgery: a meta-analysis of ran-
domized clinical trials. J Cardiothorac Vasc Anesth 2007;21:502–11.
[6] De Hert SG, Van der Linden PJ, Cromheecke S, Meeus R, Nelis A, Van
Reeth V et al. Cardioprotective properties of sevoflurane in patients
7
undergoing coronary surgery with cardiopulmonary bypass are related
to the modalities of its administration. Anesthesiology 2004;101:
299–310.
EXPERIMENTAL
[7] Hettrick DA, Pagel PS, Warltier DC. Desflurane, sevoflurane, and isoflur-
ane impair canine left ventricular-arterial coupling and mechanical effi-
ciency. Anesthesiology 1996;85:403–13.
[8] Hettrick DA, Pagel PS, Warltier DC. Alterations in canine left ventricular-
arterial coupling and mechanical efficiency produced by propofol.
Anesthesiology 1997;86:1088–93.
[9] Garcia C, Julier K, Bestmann L, Zollinger A, von Segesser LK, Pasch T et al.
Preconditioning with sevoflurane decreases PECAM-1 expression and
Downloaded from https://academic.oup.com/icvts/advance-article-abstract/doi/10.1093/icvts/ivz235/5580618 by guest on 09 October 2019
improves one-year cardiovascular outcome in coronary artery bypass
graft surgery. Br J Anaesth 2005;94:159–65.
[10] Julier K, da Silva R, Garcia C, Bestmann L, Frascarolo P, Zollinger A et al.
Preconditioning by sevoflurane decreases biochemical markers for myo-
cardial and renal dysfunction in coronary artery bypass graft surgery: a
double-blinded, placebo-controlled, multicenter study. Anesthesiology
2003;98:1315–27.
[11] Ko SH, Yu CW, Lee SK, Choe H, Chung MJ, Kwak YG et al. Propofol
attenuates ischaemia-reperfusion injury in the isolated rat heart. Anesth
Analg 1997;85:719–24.
[12] Jakobsen CJ, Berg H, Hindsholm KB, Faddy N, Sloth E. The influence of
propofol versus sevoflurane anesthesia on outcome in 10, 535 cardiac
surgical procedures. J Cardiothorac Vasc Anesth 2007;21:664–71.
[13] Haraldsen P, Lindstedt S, Metzsch C, Algotsson L, Ingemansson R. A por-
cine model for acute ischaemic right ventricular dysfunction. Interact
CardioVasc Thorac Surg 2014;18:43–8.
[14] Vieillard-Baron A, Naeije R, Haddad F, Bogaard HJ, Bull TM, Fletcher N
et al. Diagnostic workup, etiologies and management of acute right ventri-
cle failure: a state-of-the-art paper. Intensive Care Med 2018;44:774–90.
[15] Haraldsen P, Metzsch C, Lindstedt S, Algotsson L, Ingemansson R.
Sevoflurane anesthesia during acute right ventricular ischaemia in pigs
preserves cardiac function better than propofol anesthesia. Perfusion
2016;31:495–502.
[16] Hein M, Roehl AB, Baumert JH, Bleilevens C, Fischer S, Steendijk P et al.
Xenon and isoflurane improved biventricular function during right ven-
tricular ischaemia and reperfusion. Acta Anaesthesiol Scand 2010;54:
470–8.
[17] Lumb GD, Hardy LB. Collateral circulation and survival related to gradual
occlusion of the right coronary artery in the pig. Circulation 1963;27:
717–21.
[18] Andrews DT, Royse AG, Royse CF. Functional comparison of anaesthetic
agents during myocardial ischaemia-reperfusion using pressure-volume
loops. Br J Anaesth 2009;103:654–64.
[19] Bouwman RA, Musters RJ, van Beek-Harmsen BJ, de Lange JJ, Lamberts
RR, Loer SA et al. Sevoflurane-induced cardioprotection depends on
PKC-alpha activation via production of reactive oxygen species. Br J
Anaesth 2007;99:639–45.
[20] Mechelinck M, Hein M, Bellen S, Rossaint R, Roehl AB. Adaptation to
acute pulmonary hypertension in pigs. Physiol Rep 2018;6:e13605.
[21] Sarnoff SJ, Mitchell JH, Gilmore JP, Remensnyder JP. Homeometric
autoregulation in the heart. Circ Res 1960;8:1077–91.
[22] Larach DR, Schuler HG. Direct vasodilation by sevoflurane, isoflurane,
and halothane alters coronary flow reserve in the isolated rat heart.
Anesthesiology 1991;75:268–78.
[23] Weaver ME, Pantely GA, Bristow JD, Ladley HD. A quantitative study of
the anatomy and distribution of coronary arteries in swine in compari-
son with other animals and man. Cardiovasc Res 1986;20:907–17.
[24] Ramanathan T, Skinner H. Coronary blood flow. Cont Educ Anaesth Crit
Care Pain 2005;5:61–4.
[25] Sarnari R, Kamal RY, Friedberg MK, Silverman NH. Doppler assessment
of the ratio of the systolic to diastolic duration in normal children: rela-
tion to heart rate, age and body surface area. J Am Soc Echocardiogr
2009;22:928–32.
[26] Guensch DP, Yu J, Nadeshalingam G, Fischer K, Shearer J, Friedrich MG.
Evidence for acute myocardial and skeletal muscle injury after serial
transthoracic shocks in healthy swine. PLoS One 2016;11:e0162245.
[27] Manohar M, Parks CM. Porcine systemic and regional organ blood flow
during 1.0 and 1.5 minimum alveolar concentrations of sevoflurane an-
esthesia without and with 50% nitrous oxide. J Pharmacol Exp Ther
1984;231:640–8.
[28] Swindle MM, Smith AC. Swine in the Laboratory: Surgery, Anesthesia,
Imaging & Experimental Techniques. 3rd edn. Boca Raton, FL: CRC Press,
2015.