687114

research-article2016
ACC0010.1177/2048872616687114European Heart Journal: Acute Cardiovascular CareHusebye et al.

EUROPEAN
SOCIETY OF
Original scientific paper CARDIOLOGY ®

European Heart Journal: Acute Cardiovascular Care

Systolic mitral annulus velocity is 1­–9

© The European Society of Cardiology 2017
Reprints and permissions:
a sensitive index for changes in left sagepub.co.uk/journalsPermissions.nav
https://doi.org/10.1177/2048872616687114
DOI: 10.1177/2048872616687114
ventricular systolic function during journals.sagepub.com/home/acc

inotropic therapy in patients with

acute heart failure

Trygve Husebye1,2,3, Jan Eritsland1,2, Reidar Bjørnerheim1,2

and Geir Ø Andersen1,2,4

Abstract
Background: Echocardiography is recommended for assessment of left ventricular systolic function in patients with
acute heart failure but few randomised trials have validated techniques like tissue Doppler (TDI) and speckle tracking
(STE) in patients with acute heart failure following ST-elevation myocardial infarction.
Methods: This was a substudy from the LEAF (LEvosimendan in Acute heart Failure following myocardial infarction)
trial (NCT00324766 ), which randomised 61 patients developing acute heart failure, including cardiogenic shock, within
48 hours after ST-elevation myocardial infarction, double-blind to a 25-hour infusion of levosimendan or placebo.
TDI-derived systolic mitral annulus velocity (S′), STE-derived global longitudinal strain (Sl) and strain rate (SRl) were
measured at baseline, day 1, day 5 and after 42 days.
Results: Datasets rejected for analyses were 2% (TDI) and 17% (STE). S′ increased by 23% in the levosimendan group
versus 8% in the placebo group from baseline to day 1 (p= 0.011) and by 30% vs. 3% from baseline to day 5 (p <0.0005).
Significant, but less pronounced, improvements in global Sl (p = 0.025 and p = 0.032) and in global SRl (p = 0.046 and p
= 0.001) in favour of levosimendan were also present.
Conclusion: S′ by TDI and STE-derived Sl and SRl were sensitive indices for changes in left ventricular systolic function
related to treatment with levosimendan. However, S′ by TDI was more feasible and sensitive and might be preferred
for assessment of changes in left ventricular systolic function in critically ill patients with acute heart failure receiving
inotropic therapy.

Keywords
Systolic mitral annulus velocity, global longitudinal strain, acute heart failure, levosimendan, myocardial contractility

Date received: 18 July 2016; accepted: 11 December 2016

Introduction
1Department of Cardiology, Oslo University Hospital Ullevål, Norway
Echocardiography is the image modality of choice in 2Centre for Heart Failure Research, University of Oslo, Norway
patients with heart failure due to its availability, safety and 3Faculty of Medicine, University of Oslo, Norway

low-cost.1 However, use of echocardiography in patients 4Centre for Clinical Heart Research, Oslo University Hospital Ullevål,

with acute heart failure (AHF) is less well defined, both in Norway
clinical practice and in clinical trials.2
Corresponding author:
Assessment of left ventricle (LV) systolic function by Trygve Husebye, Department of Cardiology, Oslo University Hospital
indices like wall motion score index (WMSI) and LV ejec- Ullevål Hospital, P.O. Box 4950, Nydalen N-0424, Oslo, Norway.
tion fraction are largely based on visual judgement and Email: tr-huse@online.no
2 European Heart Journal: Acute Cardiovascular Care
their user-dependency represents a major limitation, espe-
cially in clinical trials. Less user-dependent indices such as
peak systolic velocity of the mitral annulus (S′) by tissue
Doppler imaging (TDI), and global longitudinal strain (Sl)
and strain rate (SRl) by speckle tracking echocardiography
(STE), offer both objective, quantitative and validated
measurement of LV function.3,4 These indices have been
studied in patients with chronic heart failure (CHF).5,6
However, no data exist regarding the feasibility of these
indices in patients with advanced AHF complicating
ST-elevation myocardial infarction (STEMI).
We have previously reported results from the ran-
domised, placebo-controlled LEAF (LEvosimendan in
Acute heart Failure following myocardial infarction) trial
on patients with AHF following a percutaneous coronary
intervention (PCI)-treated STEMI.7 We were able to dem-
onstrate that levosimendan improved the primary endpoint
WMSI from baseline to day 5 compared with placebo.
The main objective of this substudy of the LEAF trial
was to elucidate whether repetitive measurements of S′ by
TDI and global Sl and SRl by STE were feasible and sensi-
tive indices for changes in LV systolic function during ino-
tropic stimulation by levosimendan in STEMI patients with
AHF including cardiogenic shock.
Methods
The LEAF trial was an investigator initiated, manufacturer
independent study conducted at Oslo University Hospital
Ullevål. The Regional Ethics Committee approved the
study, which was conducted in accordance with the princi-
ples of the Declaration of Helsinki. All patients provided
written informed consent. The study was registered at
www.clinicaltrial.gov; NCT00324766.
Study design and patient population
The LEAF trial was a randomised, double-blind, placebo-
controlled, single-centre, parallel-group study. Details on
study design and main results have been published previ-
ously.7 Patients were included in an intensive care (ICCU)
setting and assigned to an infusion of 0.2 µg/kg per min for
1 h followed by 0.1 µg/kg per min for 24 h of levosimendan
or placebo. Included in the study were patients with AHF
(including stratified randomisation of patients in cardio-
genic shock) complicating a PCI-treated STEMI. The
inclusion criteria were: (a) opening of an occluded or dila-
tion of a stenotic coronary artery presumed to be the
infarct-related artery, (b) signs of decreased wall-motion in
at least three of 16 segments of the LV assessed by echo-
cardiography, (c) clinical HF. Additional criteria in the car-
diogenic shock subgroup were: (a) systolic blood pressure
(SBP) < 90 mmHg after 60 min of adequate volume ther-
apy or SBP between 90 and 100 mmHg in spite of ino-
tropic support by catecholamine infusion and (b) signs of
organ hypoperfusion such as oliguria, cold and clammy
extremities or reduced consciousness.
All patients received standard medical therapy accord-
ing to current guidelines. The use of intravenous inotropic
drugs was restricted to patients with cardiogenic shock,
except noradrenaline in the setting of hypotension not
responding to volume therapy and reduction in study drug
infusion.
Echocardiography
Examinations were performed with two digital ultrasonic
device systems from the same vendor (GE Vingmed
Ultrasound, Horten, Norway), Vivid i at baseline and on
day 1 immediately after end of study drug infusion and
Vivid 7 on day 5 and after 42 days (four echocardiograms
in each patient). All recordings were analysed with the
same software (GE Echopac v 108.15). Two experienced
echocardiographers (TH, GØA) performed all the record-
ings and a single one (TH) performed all the analyses.
Interobserver reproducibility was tested by an independent
echocardiographer blinded to the results of the performed
analyses. Three consecutive cardiac cycles were recorded,
five if atrial fibrillation was present. Recordings included
the standard parasternal and apical four-chamber, two-
chamber and long-axis views.
TDI-recordings were obtained by colour tissue Doppler
with a mean frame rate of 146 (±30)/s. Care was taken to
avoid an angle of incidence >20° between the ultrasound
beam and the direction of the motion of the basal segment
of the LV walls. A sampling area of 6 mm × 6 mm was
positioned in the basal segment of the LV wall within 1
cm of the insertion of the mitral leaflets (Figure 1). S′
(TDI) was measured as the average of the velocities
acquired from the base of the septal, lateral, anterior and
inferior walls over three consecutive cycles (five if atrial
fibrillation).
Recordings for STE analyses were obtained from grey-
scale images in the three apical views with a mean frame
rate of 76 (±13)/s. Only the cycle with the best image qual-
ity was chosen for the analyses. Tracking of the myocardial
walls was automatically performed by the software (GE
Vingmed Ultrasound). Segments with suboptimal tracking
were discarded. If ≥3 segments were discarded or if atrial
fibrillation was present, the whole dataset was rejected for
analysis. End-systole was defined as closure of the aortic
valve determined from the apical long-axis view. Peak Sl
was defined as the maximal shortening before end-systole
and by the software given as a mean value in each of the
three apical views. Global Sl was then calculated as the
average value of the whole LV. Correspondingly, global SRl
was calculated as the average of the mean values of the
maximal systolic strain rate.
WMSI was measured using a 16-segments model.8 LV
ejection fraction was measured by Simpson’s biplane
Husebye et al. 3
Figure 1.  Measurements of left ventricular function by colour tissue Doppler imaging. Placement of sample volume for
measurement of mitral annulus velocities in apical four-chamber view (arrows to the left) and the corresponding maximal velocities
(arrows to the right) are shown.
method or by visual assessment.8 The LV outflow tract
(LVOT) peak velocity and stroke volume (SV) was meas-
ured according to current guidelines.9 Total afterload was
determined by the effective arterial elastance (Ea) (0.9 ×
SBP/SV).10,11 Mitral regurgitation was graded as mild,
moderate or severe.12
Outcome data
Major adverse events were defined as a composite of all-
cause death, heart transplantation and implantation of
implantable cardioverter-defibrillator (ICD) after success-
fully surviving cardiac arrest.
Feasibility and reproducibility
The acquired frame rates, both by TDI and STE, were not
significantly different between the levosimendan and pla-
cebo groups at any timepoint. Figure 2 shows the number
of patients analysed by TDI and STE according to treat-
ment groups. Less than 2% of the datasets were discarded
for analyses by TDI compared with 17% by STE (Figure 2).
Reproducibility was tested in 20 datasets randomly
selected from the LEAF-database. The intraobserver/
interobserver intraclass correlation coefficients (95% con-
fidence interval (CI)) for S′(TDI) were 0.97 (0.93–
0.99,)/0.94 (0.85–0.98), for global Sl 0.98 (0.95–0.99)/0.93
(0.84–0.97) and for global SRl 0.91 (0.79–0.96)/0.91
(0.78–0.96), respectively.
Statistical analyses
Values are reported as mean (SD) or median (25th, 75th
percentiles) according to the distribution of the data.
Analyses of covariance were used for changes from base-
line between treatment groups in S′(TDI) and global Sl
with the respective baseline values as covariates.
Analyses of other continuous data were performed by
two-sample t-test. If data were not normally distributed,
differences in changes from baseline between groups
were tested by Mann–Whitney U-test. Univariate Cox
proportional hazard analyses were performed in order to
explore a possible association between S′, Sl and ejection
fraction at baseline and time to a first major adverse
event. The relatively low number of events precluded a
complete multivariate analysis, and outcome results were
therefore only adjusted for age. Log rank test was used
for analysis of difference between the levosimendan and
placebo groups. A significance level of 5% with a two-
sided test was used. Adjustments for multiple compari-
sons were not performed due to the exploratory design of
the study. Statistical analyses were performed using IBM
SPSS version 20.0 (SPSS Inc., Chicago, IL, USA).
Results
A total of 61 patients were randomised double-blind to
levosimendan (n = 30) or placebo (n = 31). The two treat-
ment groups were not significantly different with respect
4 European Heart Journal: Acute Cardiovascular Care

Figure 2.  Echocardiograms analysed in the levosimendan and placebo group. Total number and the number rejected from analyses
by TDI and STE from baseline until day 42 are shown.
TDI: colour tissue Doppler imaging; STE: speckle tracking echocardiography.

to baseline characteristics (Table 1). Patients were charac-
terised by large infarct size (peak troponin T ~ 12,000
ng/l) and severe heart failure at inclusion (28% treated
with an intra-aortic balloon pump and 15% were in car-
diogenic shock).
increased significantly at day 1 (p = 0.046) and day 5 (p =
0.001) in favour of levosimendan (Figure 3(c)). Significant
improvements in LVOT peak velocity from baseline to day
1 (p = 0.030) and day 5 (p = 0.002) in favour of levosi-
mendan were also found (Figure 3(d)).

Echocardiographic indices Haemodynamic parameters

A total of four echocardiograms in one patient were dis-
carded for analysis by TDI compared with 40 in 14 patients
by STE (Figure 2). S′(TDI) at baseline was 4.67 ± 1.33
cm/s (levosimendan) and 4.73 ± 1.05 cm/s (placebo) and
increased significantly at day 1 (p = 0.011) and day 5 (p
<0.0005) in favour of levosimendan, relative increase 23%
vs. 8 % (day 1) and 30% vs. 3% (day 5) (Figure 3(a)).
Global Sl at baseline was −9.4% ± 1.75 (levosimendan) and
−8.8 % ± 2.48 (placebo). There was a significant improve-
ment at day 1 (p = 0.025) and day 5 (p = 0.032) in the levo-
simendan compared with the placebo group (Figure 3(b)).
However, the differences in the relative changes in global Sl
between the treatment groups were modest compared with
the corresponding differences in S′(TDI) (Figure 3(b)).
Baseline levels of global SRl were −0.59/s (−0.67, −0.53)
(levosimendan) and −0.57/s (−0.67, −0.50) (placebo) and
There was a significant decrease in heart rate (HR) from
baseline to day 5 in the placebo group compared with levo-
simendan. Cardiac output (CO) increased significantly
from baseline to day 5 in patients receiving levosimendan
compared with placebo. Otherwise no significant between-
group differences in HR, SBP, SV or Ea were found (Table
2). In one patient in each treatment group mitral regurgita-
tion was graded as >2 after 42 days, otherwise no changes
in mitral regurgitation appeared during the study period.
Outcome data
There were 21 major adverse events (deaths (n = 19),
heart transplant (n = 1), ICD after cardiac arrest (n = 1))
during follow-up (median 85 months, range 69–125);
however, no significant difference between the
Husebye et al. 5

Table 1.  Baseline characteristics of the study population according to treatment arms.

Levosimendan Placebo p-valuea

n = 30 n = 31
Age 66 (56, 74) 62 (56, 74) 0.61
Female, n (%) 12 (40) 6 (19) 0.10
Chronic heart failure, n (%) 0 (0) 1 (3) 1.0
Hypertension, n (%) 10 (33) 11 (36) 1.0
Dyslipidaemia, n (%) 3 (10) 10 (32) 0.06
Diabetes mellitus, n (%) 5 (17) 1 (3) 0.10
Current smoking, n (%) 12 (41) 10 (33) 0.68
Previous myocardial infarction, n (%) 7 (23) 4 (13) 0.34
Multivessel disease, n (%) 13 (43) 18 (58) 0.31
Infarct related artery, n (%)
  Left anterior descending artery 22 (73) 24 (77) 0.77
  Left main stem 2 (7) 2 (7) 1.0
  Circumflex artery 3 (10) 5 (16) 0.71
  Right coronary artery 3 (10) 0 (0) 0.11
Cardiogenic shock, n (%) 4 (13) 5 (16) 1.0
IABP at baseline, n (%) 8 (27) 9 (29) 1.0
Hours from start of symptoms to PCI 3 (2, 8) 3 (2, 6) 0.93
Hours from PCI to study infusion 24 (14, 33) 22 (14, 26) 0.63
TnT peak, ng/l 12195 (7990, 16,187) 11828 (5670, 18,640) 0.95
Creatinine, µmol/l 82 (66, 90) 82 (73, 110) 0.25
NT-proBNP, pmol/l 386 (297, 597) 474 (248, 922) 0.54
WMSI at baseline 1.94 (1.75, 2.13) 2.0 (1.88, 2.19) 0.17
LVEF at baseline, % 43 (38, 49) 40 (33, 47) 0.13
Mitral regurgitation at baseline, %
  Grade 1–2 2 (7) 3 (10) 1.0
  Grade >2 0 0  

Continuous data are presented as median (25th, 75th percentiles) unless indicated otherwise.
aDifferences between treatment groups.

IABP: intra-aortic balloon counter-pulsation; PCI: percutaneous coronary intervention; TnT: troponin T; NT-proBNP: N-terminal pro B-type natri-
uretic peptide; WMSI: wall motion score index; LVEF: left ventricular ejection fraction

levosimendan and placebo groups was found (p = 0.72).
Both S′ and Sl at baseline were associated with the com-
posite endpoint in univariate analyses, while LVEF only
reached borderline significance (Table 3). The HRs were
significant after adjusting for age (S′, HR 0.58 (0.37–0.91)
p = 0.018), (Sl, HR 1.68 (1.27–2.21) p < 0.0005) and
(LVEF (0.87–0.99) p = 0.016).
Use of beta-blockers and inotropic drugs
other than levosimendan
No significant differences in the use of beta blockers
between the levosimendan and placebo groups were found;
baseline: 6/30 vs. 6/31, day 1: 3/29 vs. 2/30, day 5: 18/29 vs.
18/29, day 42: 29/30 vs. 24/25, respectively. All patients in
the cardiogenic shock group (n = 9) received inotropic
drugs (dopamine or dobutamine) before inclusion and dur-
ing the first day of the study. After five days, one patient in
the levosimendan group and two patients in the placebo
group received dopamine. A total of four patients in the
levosimendan group and one patient in the placebo group
received noradrenaline due to transient hypotension during
the study drug infusion (first 25 h). This treatment was dis-
continued in all patients >30 min before the end of the drug
infusion and did not affect the subsequent echo-recordings
on day 1 or day 5.
Discussion
The principal finding of this prospective clinical study is
that S′ measured by TDI was a feasible and sensitive
marker of inotropic effect of levosimendan in patients
with advanced AHF. This is, to our best knowledge, the
first report of the use of this echocardiographic index to
quantify effects of inotropic stimulation in patients with
AHF complicating STEMI. It was more feasible and sen-
sitive than STE-derived Sl and SRl in an ICCU setting
with critically ill patients, including patients on a
mechanical ventilator. The randomised, placebo-con-
trolled study design ensured that changes in myocardial
function related to inotropic stimulation and not to effects
related to time per se were measured.
6 European Heart Journal: Acute Cardiovascular Care

Figure 3.  Changes from baseline in echocardiographic indices in patients treated with a 25 h infusion of levosimendan or placebo
S′ by colour tissue Doppler (a), Sl by speckle tracking (b), SRl by speckle tracking (c) and LVOT peak (d). Mean values ± SEM ((a),
(b) and (d)) or median values with 25th,75th percentiles (c). p-values (* p <0.05, **p <0.005, *** p <0.0005) are given for differences
between treatment groups.
S′: peak systolic mitral annulus velocity; Sl: global longitudinal strain; SRl: global longitudinal strain rate; LVOT peak: left ventricular outflow tract peak velocity.

Table 2.  Haemodynamic data according to treatments arms.

Levosimendan Placebo p-valuea

SV, ml
 Baseline 44 (37, 53) 45 (37, 53)  
  Day 1 49 (44, 56) 50 (44, 60) 0.48
  Day 5 53 (49, 60) 50 (47, 60) 0.66
  Day 42 60 (53, 69) 57 (49, 65) 0.33
CO, l/min
 Baseline 3.8 (3.4, 4.4) 3.8 (3.3, 4.3)  
  Day 1 4.9 (4.2, 5.5) 4.3 (3.9, 4.6) 0.057
  Day 5 4.6 (4.3, 5.2) 4.1 (3.7, 4.6) 0.022
  Day 42 3.9 (3.3, 4.4) 3.9 (3.6, 4.4) 0.66
HR, beats/min
 Baseline 86 (76, 95) 89 (75, 101)  
  Day 1 91 (79, 104) 87 (80, 94) 0.28
  Day 5 86 (81, 93) 80 (71, 87) 0.01
  Day 42 66 (59, 74) 71 (60, 83) 0.30
SBP, mmHg
 Baseline 102 (93, 114) 107 (93, 115)  
  Day 1 100 (90, 118) 105 (93, 111) 0.71
  Day 5 111 (100, 126) 110 (90, 120) 0.37
  Day 42 120 (105, 140) 116 (109, 130) 0.54
Ea, mmHg/ml
 Baseline 2.12 (1.72, 2.33) 1.97 (1.59, 2.47)  
  Day 1 1.81 (1.48, 2.19) 1.95 (1.50, 2.28) 0.40
  Day 5 1.92 (1.59, 2.23) 1.95 (1.57, 2.11) 0.50
  Day 42 1.77 (1.63, 2.18) 1.95 (1.50, 2.23) 0.69

Values are median (25th, 75th percentiles).

aDifferences in changes from baseline between treatment groups.

SV: stroke volume; CO: cardiac output; HR: heart rate; SBP: systolic blood pressure; Ea: effective arterial elastance
Husebye et al. 7
Table 3.  Cox regression analyses for the association between
baseline factors and future adverse events (all-cause death,
heart transplantation or implantable cardioverter-defibrillator
implantation after surviving cardiac arrest).
Univariate analysis  
  Hazard ratio (95% CI) p-value
Age 1.05 (1.01–1.09) 0.017
Female 0.60 (0.22–1.66) 0.327
S′ 0.61 (0.41–0.91) 0.016
Sl 1.57 (1.18–2.08) 0.002
LVEF 0.94 (0.89–0.999) 0.045
TnT peak, µg/l 1.07 (0.096–1.16) 0.061
CI: confidence interval; S′: systolic mitral annulus velocity by colour tis-
sue Doppler; Sl: global longitudinal strain by speckle tracking; LVEF: left
ventricular ejection fraction; TnT peak: peak level of troponin T
Feasibility of S′ by TDI and STE-derived Sl
and SRl
S′(TDI) is easy to obtain, as visualisation of the mitral
annulus is possible in nearly all patients. It has a satisfac-
tory reproducibility when measured as the average velocity
from the septal, lateral, anterior and inferior mitral annu-
lus.4,13 In our study, as demonstrated in Figure 2, S′(TDI)
was analysed in all patients except one, confirming the fea-
sibility of the index.
STE depends on good quality grey-scale images of the
whole LV.4 This may limit its application in different clini-
cal settings such as an ICCU. This was confirmed in this
study on STEMI patients with AHF where many recordings
were discarded due to image quality reasons. Our results
indicate that STE may not be an ideal method for serial
measurements in critically ill patients.
Assessment of inotropic alterations
The inotropic effects of levosimendan occur a few hours
after start of the drug infusion and, due to its active
metabolites, last for several days.14 Due to systemic
vasodilating properties, levosimendan may also improve
LV function by afterload reduction.15 However, the
changes in SBP and the calculated total afterload (Ea)
were not significantly different between the two groups
during the first five days after start of treatment in this
study. Therefore, the observed changes in S′(TDI)
between the levosimendan and placebo groups were
probably mostly explained by the inotropic effects of
levosimendan and not by its vasodilating properties.
The capability of systolic velocity by TDI to detect
minor alterations in LV contractility has been documented
in animal studies during dobutamine infusion.16-18 Although
this index is influenced by LV load and elastic properties of
the myocardium,11,19-21 S′ was reported to be relatively
independent of haemodynamic parameters like blood
pressure and systemic vascular resistance in 179 patients
during dobutamine stress echocardiography.22
Differences in the relative changes in global Sl between
the levosimendan and placebo group were modest, while
differences in global SRl and LVOT peak velocity were
comparable to the relative differences in S′(TDI). SRl and
LVOT peak velocity are, as S′, markers of events that occur
in the first half of the systole, as opposed to Sl, which is
measured at or near end-systole.23 Previously, animal stud-
ies have shown that SRl correlated strongly with contractil-
ity, while Sl correlated best with SV.16,24 Also, in a study on
healthy humans there was a significantly lower relative
increase in global Sl than in global SRl, LVOT peak veloc-
ity and S′ during low-dose dobutamin infusion.25 Our find-
ings indicate that these relationships also apply during
inotropic therapy in STEMI patients with AHF.
Levosimendan improved LV systolic function in post-
ischaemic myocardium measured as change in WMSI from
baseline to day 5 compared with placebo (p = 0.031, pri-
mary endpoint of the LEAF trial); however, no significant
change was present the first day after start of drug infusion.7
WMSI is based on visual assessment and the semi-quantita-
tive scoring with only four levels makes it less sensitive to
detect minor changes in LV systolic function over time. In
this substudy, the TDI and STE-derived indices were more
sensitive for such changes, probably because they measure
on a continuous scale and are less user-dependent.
Haemodynamic changes
Only a non-significant increase in SV related to levosi-
mendan treatment was found in the present study despite
the improvement in LV function. A dose–response relation-
ship between levosimendan treatment and changes in
haemodynamic indices after 23–24 h was reported in a pla-
cebo controlled study on patients with acute decompen-
sated HF.26 The main effect of levosimendan in the low-dose
range (<0.2 µg/kg per min) seemed to be reduction in ven-
tricular filling pressure and increase in cardiac output due
to increased HR, while a more modest increase in SV was
observed. AHF in STEMI patients commonly is precipi-
tated by high filling pressures due to reduced contractility
and stiffness of ischaemic myocardium, more than reduced
SV. Improvement of contractility of ischaemic myocardium
consistent with anti-stunning effect of levosimendan has
been reported.27 An improvement in contractility of post-
ischaemic myocardium resulting in reduced filling pres-
sures, in spite of a modest increase in SV, might be of
benefit in patients with AHF complicating PCI-treated
STEMI and explain the results of our study.
Outcome data
Although the key message of the study was that measure-
ments of S′ and Sl are sensitive indices for changes in LV
8 European Heart Journal: Acute Cardiovascular Care
systolic function in STEMI-patients with AHF, our study is
also the first to report the prognostic information of early
measurements of the indices. S′ has been reported to be a
predictor of death in patients with CHF6,28 and clinical out-
come in PCI treated STEMI patients,29 while Sl was shown
to be associated with future adverse events in patients with
CHF30 and in patients after AMI.31 Our results expand the
potential clinical importance of early measurements of S′
and Sl in STEMI patients complicated by AHF. However,
these results must be interpreted with caution as the sample
size and few endpoints generally restricted the use of mul-
tivariable analyses and should therefore be confirmed in
larger study populations.
Limitations of our study
The sample size was based on changes in WMSI, the pri-
mary endpoint of the LEAF trial, and not on changes in
the TDI and STE indices examined in this study. An inher-
ent weakness of TDI is the impact of global heart motion
and angle of incidence of the ultrasound beam on the
measurements. Both factors may have affected the serial
measurements of S′(TDI). Analysing only one cycle by
STE may have introduced more random variability in the
data compared with TDI. The low temporal resolution of
the STE recordings in combination with high HR might
have underestimated SRl in some patients.3 Measurement
of SRl by TDI might have avoided this. However, as both
TDI derived strain rate and strain generally are signifi-
cantly affected by noise,3 we decided not to use these indi-
ces in this study. Calculations of SV, CO and total afterload
were based on noninvasive measurements by pulsed wave
Doppler. Serial measurements by this method are sensi-
tive to errors in placement of the sample volume in LVOT.
This might have affected the ability to detect significant
changes over time.
Conclusion
We have for the first time demonstrated the feasibility and
usefulness of serial measurements of S′ measured by TDI
and STE-derived global Sl and SRl to assess changes in LV
systolic function in an ICCU setting with critically ill
patients with AHF. S′ was more feasible than STE-derived
indices and performed better than the established index,
WMSI. These results suggest that S′(TDI) might be the pre-
ferred echocardiographic index for assessment of changes
in LV systolic function during inotropic therapy in patients
with AHF.
Acknowledgements
We are grateful for skilful help from Dr Christian Shetelig and to
the Centre for Clinical Heart Research, Department of Cardiology,
Oslo University Hospital, with Professors Harald Arnesen and
Ingebjørg Seljeflot, for research support in the LEAF trial.
Conflict of interest
The Department of Cardiology, Oslo University Hospital Ullevål
received an unrestricted educational grant from the manufac-
turer of levosimendan, Orion Pharma in 2005. Orion Pharma did
not, however, provide study medication or participate in the
design, monitoring or analyses of the LEAF trial or the present
substudy. TH has received a lecture honorarium from Orion
Pharma in 2012.
Funding
This work was supported by the Centre for Heart Failure Research,
University of Oslo, South-Eastern Norway Regional Health
Authority and The Scientific Council at Oslo University Hospital
Ullevål.
References
1. McMurray JJ, Adamopoulos S, Anker SD, et al. ESC guide-
lines for the diagnosis and treatment of acute and chronic
heart failure 2012: The Task Force for the Diagnosis and
Treatment of Acute and Chronic Heart Failure 2012 of the
European Society of Cardiology. Developed in collaboration
with the Heart Failure Association (HFA) of the ESC. Eur J
Heart Fail 2012; 14: 803–869.
2. Papadimitriou L, Georgiopoulou VV, Kort S, et al.
Echocardiography in acute heart failure: Current perspec-
tives. J Card Fail 2016; 22: 82–94.
3. Marwick TH. Measurement of strain and strain rate by echo-
cardiography: Ready for prime time? J Am Coll Cardiol
2006; 47: 1313–1327.
4. Mor-Avi V, Lang RM, Badano LP, et al. Current and evolv-
ing echocardiographic techniques for the quantitative evalu-
ation of cardiac mechanics: ASE/EAE consensus statement
on methodology and indications endorsed by the Japanese
Society of Echocardiography. Eur J Echocardiogr 2011; 12:
167–205.
5. Ikonomidis I, Parissis JT, Paraskevaidis I, et al. Effects of
levosimendan on coronary artery flow and cardiac perfor-
mance in patients with advanced heart failure. Eur J Heart
Fail 2007; 9: 1172–1177.
6. Nikitin NP, Loh PH, Silva R, et al. Prognostic value of sys-
tolic mitral annular velocity measured with Doppler tissue
imaging in patients with chronic heart failure caused by left
ventricular systolic dysfunction. Heart 2006; 92: 775–779.
7. Husebye T, Eritsland J, Muller C, et al. Levosimendan in
acute heart failure following primary percutaneous coronary
intervention-treated acute ST-elevation myocardial infarc-
tion. Results from the LEAF trial: A randomized, placebo-
controlled study. Eur J Heart Fail 2013; 15: 565–572.
8. Lang RM, Badano LP, Mor-Avi V, et al. Recommendations for
cardiac chamber quantification by echocardiography in adults:
An update from the American Society of Echocardiography
and the European Association of Cardiovascular Imaging.
Eur Heart J Cardiovasc Imaging 2015; 16: 233–270.
9. Baumgartner H, Hung J, Bermejo J, et al. Echocardiographic
assessment of valve stenosis: EAE/ASE recommendations
for clinical practice. Eur J Echocardiogr 2009; 10: 1–25.
10. Kelly RP, Ting CT, Yang TM, et al. Effective arterial elastance
as index of arterial vascular load in humans. Circulation
1992; 86: 513–521.
Husebye et al. 9
11. Borlaug BA, Melenovsky V, Redfield MM, et al. Impact of
arterial load and loading sequence on left ventricular tissue
velocities in humans. J Am Coll Cardiol 2007; 50: 1570–
1577.
12. Lancellotti P, Moura L, Pierard LA, et al. European
Association of Echocardiography recommendations for the
assessment of valvular regurgitation. Part 2: Mitral and tricus-
pid regurgitation (native valve disease). Eur J Echocardiogr
2010; 11: 307–332.
13. Thorstensen A, Dalen H, Amundsen BH, et al. Reproducibility
in echocardiographic assessment of the left ventricular global
and regional function, the HUNT study. Eur J Echocardiogr
2010; 11: 149–156.
14. Parissis JT, Rafouli-Stergiou P, Paraskevaidis I, et al.
Levosimendan: From basic science to clinical practice.
Heart Fail Rev 2009; 14: 265–275.
15. Kass DA and Solaro RJ. Mechanisms and use of calcium-
sensitizing agents in the failing heart. Circulation 2006; 113:
305–315.
16. Greenberg NL, Firstenberg MS, Castro PL, et al. Doppler-
derived myocardial systolic strain rate is a strong index of
left ventricular contractility. Circulation 2002; 105: 99–105.
17. Seo JS, Kim DH, Kim WJ, et al. Peak systolic velocity of
mitral annular longitudinal movement measured by pulsed
tissue Doppler imaging as an index of global left ventricular
contractility. Am J Phys Heart Circ Phys 2010; 298: H1608–
H1615.
18. Gorcsan J 3rd, Strum DP, Mandarino WA, et al. Quantitative
assessment of alterations in regional left ventricular con-
tractility with color-coded tissue Doppler echocardiography.
Comparison with sonomicrometry and pressure-volume rela-
tions. Circulation 1997; 95: 2423–2433.
19. Bijnens BH, Cikes M, Claus P, et al. Velocity and deforma-
tion imaging for the assessment of myocardial dysfunction.
Eur J Echocardiogr 2009; 10: 216–226.
20. Kass DA, Maughan WL, Guo ZM, et al. Comparative
influence of load versus inotropic states on indexes of ven-
tricular contractility: Experimental and theoretical analysis
based on pressure-volume relationships. Circulation 1987;
76: 1422–1436.
21. Urheim S, Edvardsen T, Torp H, et al. Myocardial strain by
Doppler echocardiography. Validation of a new method to
quantify regional myocardial function. Circulation 2000;
102: 1158–1164.
22. Cain P, Napier S, Haluska B, et al. Influence of left ventricu-
lar size and hemodynamics on the systolic longitudinal myo-
cardial Doppler velocity response to stress. Am Heart J 2002;
143: 169–175.
23. Marsan NA, Tops LF, Westenberg JJ, et al. Usefulness of
multimodality imaging for detecting differences in temporal
occurrence of left ventricular systolic mechanical events in
healthy young adults. Am J Cardiol 2009; 104: 440–446.
24. Weidemann F, Jamal F, Sutherland GR, et al. Myocardial
function defined by strain rate and strain during alterations
in inotropic states and heart rate. Am J Phys Heart Circ Phys
2002; 283: H792–H799.
25. Thorstensen A, Dalen H, Amundsen BH, et al. Peak systolic
velocity indices are more sensitive than end-systolic indices
in detecting contraction changes assessed by echocardiogra-
phy in young healthy humans. Eur J Echocardiogr 2011; 12:
924–930.
26. Nieminen MS, Akkila J, Hasenfuss G, et al. Hemodynamic
and neurohumoral effects of continuous infusion of levosi-
mendan in patients with congestive heart failure. J Am Coll
Cardiol 2000; 36: 1903–1912.
27. Sonntag S, Sundberg S, Lehtonen LA, et al. The calcium
sensitizer levosimendan improves the function of stunned
myocardium after percutaneous transluminal coronary angi-
oplasty in acute myocardial ischemia. J Am Coll Cardiol
2004; 43: 2177–2182.
28. Wang M, Yip G, Yu CM, et al. Independent and incremental
prognostic value of early mitral annulus velocity in patients
with impaired left ventricular systolic function. J Am Coll
Cardiol 2005; 45: 272–277.
29. Biering-Sorensen T, Jensen JS, Pedersen S, et al. Doppler
tissue imaging is an independent predictor of outcome in
patients with ST-segment elevation myocardial infarction
treated with primary percutaneous coronary intervention. J
Am Soc Echocardiogr 2014; 27: 258–267.
30. Hasselberg NE, Haugaa KH, Bernard A, et al. Left ventricu-
lar markers of mortality and ventricular arrhythmias in heart
failure patients with cardiac resynchronization therapy. Eur
Heart J Cardiovasc Imaging 2016; 17: 343–350.
31. Ersbøll M, Valeur N, Mogensen UM, et al. Prediction of all-
cause mortality and heart failure admissions from global left
ventricular longitudinal strain in patients with acute myocar-
dial infarction and preserved left ventricular ejection frac-
tion. J Am Coll Cardiol 2013; 61: 2365–2373.