61 / 64
CARDIOVASCULAR JOURNAL OF AFRICA • Volume 31, No 6, November/December 2020
AFRICA
341
myocardial stunning was adequately improved; eventually, the
Bi-VAD was removed successfully.
Table 1 presents the biochemistry data, inotrope dosages
and echocardiography presentation during the VAD course.
The patient was weaned off the ventilator, and extubation
was performed three days after VAD removal. The day after
extubation, the patient was transferred to an ordinary ward
and discharged one week later. Out-patient follow up revealed
normal cardiac and renal function and cognition, and adequate
control of diabetes.
Discussion
Hypokalaemia is a common electrolyte imbalance present
in 20% of hospitalised patients,
5
and some of these patients
require immediate pharmacological treatment. Insulin-induced
hypokalaemia results in a decrease in serum potassium level due
to intracellular potassium shifts and, potentially, the aldosterone-
like effect of insulin on the renal tubule further increases urinary
potassium losses.
The goal of the treatment for insulin-induced hypokalaemia
(K
+
<
2.5 mmol/l) is to replenish potassium stores through slow
intravenous infusion of KCl,
6
with insulin therapy delayed until
serum potassium levels are corrected back to
>
2.5 mmol/l.
7
The
most severe complication of hypokalaemia is lethal arrhythmia,
such as VT/Vf. Potassium replenishment and cardioversion
defibrillation should be performed immediately.
In our case, the patient experienced in-hospital cardiac
arrest (IHCA) resulting from hypokalaemia-induced VT/Vf.
Extracorporeal CPR (ECPR) restored tissue and end-organ
perfusion to allow stabilisation and recovery of function. ECPR
can be defined as the implantation of VA-ECMO in a patient who
has experienced a sudden and unexpected pulseless condition
attributable to cessation of cardiac mechanical activity.
8
Many
prospective and retrospective studies have demonstrated the
superiority of ECPR over conventional CPR regarding the odds
of survival and neurological outcome.
9–11
ECPR can be viewed as
a late intervention in a moribund patient, possibly a candidate
for an earlier circulatory support system in case of IHCA.
Compared with ECMO, which provides both cardiac and
pulmonary support, a Bi-VAD usually provides cardiac support
only. However, a Bi-VAD can be implemented long term with
more cardiac support than ECMO, especially when the ECMO
is set up peripherally. Moreover, patients on ECMO support
usually require large doses of inotropes, which cause extreme
vasoconstriction and lead to malperfusion of the visceral organs.
In patients with refractory cardiogenic shock, a VAD has been
reported to provide a better survival rate than VA-ECMO.
12
In the current case, although VA-ECMO was instituted
for mechanical circulatory support and the potassium
level was corrected back to the normal range, the patient
experienced cardiogenic shock with multiple organ dysfunction
and exacerbations. Therefore, ECMO was substituted with
Bi-VAD implantation for optimal systemic perfusion. More
importantly, the Bi-VAD completely unloaded the bilateral
ventricle, maximising the likelihood of recovery from myocardial
stunning.
13
Based on our experience, the indications for VAD
intervention can be defined for these critical patients with
ECMO support (Table 2).
In our case, following Bi-VAD implantation, we were able
to immediately withdraw the inotropes and all the visceral
organs were preserved. Bedside echocardiography showed no
distention of the bilateral ventricle. Initially, the pulse pressure
was narrowed but returned three days later, which implied that
the myocardial stunning was completely resolved.
The CentriMag VAD (Levitronix LLC) was chosen for
several reasons. First, it has continuous flow, which is reported
to have better outcomes than pulsatile flow, especially for
Table 1. Biochemistry data, inotrope dosage and echocardiography presentation during the VAD course
Before
VAD POD1
POD2
POD3
POD4
POD5
POD7
POD11
Day 3
after
removal
K
+
(mmol/l)
1.6
4.2
4.7
3.7
3.5
3.5
3.3
3.5
3.1
BNP (pg/ml)
176
CK (U/l)
4862
>
10000
>
10000
>
10000
3292
Tro-I (ng/ml)
8.28
7.11
5.765
3.813
1.516
BUN (mg/dl)
60
26
28
31
61
78
Cr (mg/dl)
4.2
2.5
2.6
2.1
3.7
3.0
Urine output (ml/day)
170
995
(haemodyalysis)
1720
(haemodyalysis)
1620
(haemodyalysis)
3060
(haemodyalysis)
3420
(haemodyalysis)
4160
(haemodyalysis)
1480
2295
Norepinephrine (mcg/kg/min) 26.5
14.4
12.8
2.65
–
–
–
–
–
Dopamine (mcg/kg/min)
17.3
9.4
9.35
8.7
8.65
8.65
8.65
8.65
–
Epinephrine (mcg/kg/min)
16.7
13.3
13
2.7
–
–
–
–
–
L-VAD (rpm/flow)
3700/4.74
3700/5.07
3700/4.86
3600/4.5
3500/4.14
3400/3.81 2100/1.30 –
R-VAD (rpm/flow)
3000/4.87
3000/5.02
2700/4.4
2600/4.2
2400/3.75
2200/3.31 1200/0.91 –
MAP (mmHg)
65
65–75
65–75
80–90
78–86
88–100
97–105
72–82 95–100
Echocardiography
LVEF (%)
10–15
30–35
51
POD
=
post-operative day; L-VAD
=
left ventricular assist device; R-VAD
=
right ventricular assist device, BUN
=
blood urea nitrogen; CK
=
creatinine kinase.
Table 2. Indications of VAD intervention after ECMO support
1 ECMO flow insufficiency; ECMO complications
2 Any organ dysfunction with ECMO maximal flow
3 Three or more inotropes or large dose
4 Narrow pulse pressure,
≥
10 mmHg
5 Sustained VT resulted from LV distension
6 Echocardiography:
No opening of aortic valve
LV thrombus formation
Blood stasis in LV, presented as smoke swirl sign