ELECTRICALLY
STIMULATED BIOLOGICAL NEO-VENTRICLE FOR AORTIC COUNTERPULSATION: AN ANIMAL
EXPERIMENT IN SHEEP
M.Rab1, W.Girsch1, R.Koller1,
R.Seitelberger2, H.Lanmüller3, W. Haslik4,
L.-P. Kamolz4, U. Windberger5, H.Schima5,
H.G.Stöhr5 and M. Frey1
1 Department for Plastic and Reconstructive Surgery, Clinic
for Surgery
2 Department of Cardio-Thoracic Surgery
3 Department of Biomedical Engineering and Physics
4 Institute of Anatomy, Department III
5 Center for
Biomedical Research
Vienna Medical
School, University of Vienna, Austria
SUMMARY
14
adult sheep were used for acute experiments: an aorto-pericardial pouch of a
donor sheep was created. This biological conduit was anastomized in parallel to
the descending aorta of a recipient sheep, using the aortic root as an inflow
valve to the conduit. ECG-triggered nervous FES was applied during cardiac
diastole to simulate aortic counterpulsation. Stimulation was performed during
various hemodynamic conditions.
During
6 experiments a standardized surgical procedure suitable for long term studies
was established. A neoventricle with 70 to 80ml filling volume was found to be
optimal in size. In another 8 experiments hemodynamic measurements were
performed. Under stable hemodynamic conditions the stimulation of the biological
skeletal muscle ventricle induced a significant increase in mean diastolic
pressure of 58.8% (p<0.0004). During pharmacologically induced periods of
cardiac failure the stimulation of the APPC increased the mean diastolic
pressure significantly by 60.5% (p<0.002). Supra-systolic pressures were
obtained during all 8 experiments.
STATE OF THE ART
The
chronic shortage of donor organs for cardiac transplantation and the high costs
for mechanical assist devices demand the development of alternative cardiac
assist devices for the treatment of severe chronic heart failure. Therefore,
cardiac assistence by stimulated skeletal muscles is currently investigated as
a possible alternative. The goal of the presented study was to evaluate the
hemodynamic efficacy of a newly designed biological skeletal muscle ventricle
in an acute sheep model. The biological pump was used as an aortic
counterpulsation device.
MATERIAL AND METHODS
14
sheep weighing 57.5 +/- 6.2 kg were used. The animals were put under general anesthesia
and intubated. Anesthesia was maintained with halothan and nitrous oxide. At
the end of experiment the animals were sacrifized.
Surgical
procedure:
The
pericardium and the entire thoracic aorta including the aortic valve were
excised from fresh sheep cadavers prior to the operation. These excised
homografts were cryopreserved according to clinically approved techniques and
defrosted at the time of the experiment. Surgery was started by two teams, one
preparing the recipient animal and the other constructing the neo-ventricle
from the aortic homograft.
Construction of the biological
neo-ventricle:
The
aortic homograft was incised longitudinally twice in its middle section and
enlarged by two strips of pericardium to create an aorto-pericardial pouch
conduit (APPC).
Preparation of the recipient sheep:
The
left latissimus dorsi muscle (LDM) was detached from the thoracic wall under
careful preservation of its insertion to the humeral bone and the supplying
neurovascular pedicle. LDM was divided longitudinally into two branches with
respect to the intramuscular neurovascular supply. A segment of the third rib
was removed and LDM was placed in the left hemithorax. To continue the
procedure the fifth and sixth rib were
resected.
Connection of the APPC to the
circulation and positioning of LDM:
The APPC was
connected in parallel with the descending aorta of the recipient sheep. The
proximal part was anastomosed with the recipient aorta distal to the commune
brachiocephalic trunk, using the aortic root as an inflow valve to the conduit.
The distal end of the APPC was cut back to the appropriate length and connected
with the descending aorta above the diaphragm. The two branches of LDM were
wrapped around the APPC in counterrotating fashion and fixed to each other and
to the remaining parts of the fifth and sixth rib.
Activation
of the biological skeletal muscle ventricle (SMV) by FES:
Device:
Four
stimulation electrodes were applicated to the epineurium of the thoracodorsal
nerve and the electrode leads were led out percutaneously. Three ECG-sensing
electrodes were fixed to muscles of wall and both, pacing and sensing leads
were connected with an external stimulation device.
Stimulation
parameters:
Rectangular
pulses with 0.6 msec duration at a frequency of 28 Hz were used for burst
stimulation. Current was adjusted to achieve maximum tetanic contraction of
LDM. R-wave triggered stimulation at a rate of 1:2 or 1:3 with the native heart
rate was applied during diastole to simulate counterpulsation.
Hemodynamic Measurements:
A
flow-directed pulmonary artery catheter was introduced from the left jugular
vein. A aortic catheter was introduced into the left ventricle from the left
carotid artery. Saline-filled plastic catheters were placed either directly
into the left carotid artery close to the brachiocephalic trunk and introduced
into the abdominal aorta from the left femoral artery. These hydraulic pressure
catheters were connected to Van-den-Burg disposable pressure transducers. For
flow measurements three Flow Probes were placed around the proximal and distal
part of the homograft and around the descending aorta between the proximal and
distal anastomosis. All hemodynamic variables were recorded simultaneously by a
computerized registration unit. This unit includes an analog to digital
converter and systems for data analysis
Experimental sequence
(N=8):
Short periods
of stimulation, consisting of 10 to 20 contractions were performed repetitively
to avoid fatigue of the unconditioned, fast fatigable LDM. Heart failure was
induced by rapid intravenous infusion of a betablocker (Breviblockâ) and
stimulation was repeated.
RESULTS
Surgical
procedure:
During
six experiments a standardized surgical procedure, suitable for long term
studies was developed. Two pericardial patches, each sized 8 x 4 cm, created a
neo-ventricle of 70 to 80ml filling volume, which turned out to be optimal in
size and therefore was used in all further experiments. Macroscopically the
division of the LDM did not cause marked cyanosis of parts of the muscle or
denervation of parts of the LDM in any case (n = 14). At the end of each
experiment an investigation of the inner surface of the APPC was performed (n =
14). Visual inspection did not reveal any aggregates or thrombotic formations.
Hemodynamic
measurements:
During
eight experiments the hemodynamic efficiency of the neo-ventricle was
evaluated.
Stimulation
was performed under stable conditions and did affect left ventricular peak
pressure (LVP-max), mean arterial
pressure (p-mean) and mean diastolic
pressure (p-dia), which increased to
supra-systolic values in all experiments. Right ventricular (RVP) and pulmonary artery (PAP) blood pressure did not reveal
alterations due to stimulation.
Measurements
during normal heart function:
Stimulation of
the SMV caused a significant increase of pT-max
by 19% (p< 0,04) and pA-max by
27% (p< 0,02), while LVP-max decreased
not significantly by 5% (p<0.1). pT-min,
pA-min and LVPmin showed a
tendency to decrease, but were not altered significantly by FES. pT-mean and pA-mean increased significantly by 14% (p<0,02) and 17%
(p<0,02) and pT-dia showed a
significant increment of 26% (p<0,01). pA-mean
was not applicable, because the pressure curve derived from the abdominal aorta
did not allow differentiation between diastole and systole.
Measurements
during induced heart failure:
Under this
condition the stimulation of the SMV caused a significant increase of pT-max by 13% (p<0,04) and pA-max by 28% (p<0,01), while LVP-max decreased significantly by 8%
(p<0,04). pT-min, pA-min and LVPmin were not altered significantly by
FES. pT-mean and pA-mean increased significantly by 13% (p<0,002) and 11%
(p<0,002). pT-dia showed a
significant increment of 19% (p<0.01). Again pA-mean was not applicable, due to the reasons mentioned above.
Flow
measurements during normal and induced
heart failure:
Flow
measurements under stable hemodynamic conditions and also under induced periods
of cardiac failure revealed that the proximal as well as the distal part of SMV
was filled from the aorta during the systole. During the stimulation of the SMV
under both hemodynamic conditions blood was ejected from the proximal and
distal part of the homograft and led to an inversed cranial flow in the aorta
(Mean: Q-graft prox suff:-7,1l/min, Q-graft dist suff: 3,5l/min, Q-aort suff:
-0,5l/min;
Q-graft prox insuff: -6,4l/min, Q-graft dist insuff: 3,2l/min, Q-aort
insuff:-0,1l/min). Despite the existence of the aortic valve this flow
phenomenon was observed in each acute experiment .
DISCUSSION:
Various
approaches have been investigated to achieve chronic cardiac assistance using
skeletal muscles in the past, but only dynamic cardiomyolasty and recently
dynamic aortomyoplasty have found their way into clinical practice. Both
configurations are characterized by the presence of an uninterrupted
endothelium and a direct coupling of the skeletal muscle contraction to the
circulation. In our experimental setup we wanted to combine some basic aspects
of cardio- and aortomyoplasty, with new ideas concerning the positioning of the
LDM.
The
presented APPC is made of hemocompatible biological materials only. Although an
endothelium is not present at the time of operation, the preserved basal lamina
will provide reendothelialisation of the aortic homograft and the pericardial
patches according to the used preservation technique. The activation of the SMV
by its muscular envelopment means direct coupling of the skeletal muscle
contraction to the circulation, thus fulfilling another basic requirement for
efficient cardiac assist with skeletal muscles.
As a result a new type of skeletal muscle ventricle, different from already presented configurations was realized in sheep and a standardised surgical procedure, suitable for long term experiments was established. Activation of the APPC led to reduction of left ventricular peak pressure and induced marked increases of mean arterial and mean diastolic blood pressure. Referring to criterias for aortic counterpulsation the configuration did produce some of the required hemodynamic changes. In fact counterpulsation-efficacy was not to be expected in this series of acute experiments, in which an unconditioned LDM was used.
According
to our flow data we found aortic valves with different states of insufficiency
in the proximal part of the homograft in all animals. Taking into account that
the highest increase of mean diastolic pressure was produced in case of a
completely insufficient aortic valve of the SMV, no further use of this valve
should be considered for the next experiments
Summarizing
our experimental studies, it is too early for direct comparison with
Stephenson´s pericardium lined SMV, which worked in circulation up to 589 days
or aortomyoplasty, which already has been performed clinically. However, the
achieved results encourage us to continue the investigation of our newly
designed fully biological SMV. The presented data clearly demonstrate the
hemodynamic efficacy of this configuration as an aortic counterpulsation
device. Chronic animal experiments using a conditioned LDM will be performed in
order to investigate the long-term behaviour and reliability of the
configuration and its overall influence to the circulation.
REFERENCES
1) Chachques JC, Radermercker M, Tolan MJ,
Fischer EI, Grandjean PA, Carpentier AF (1996) Aortomyoplasty counterpulsation:
experimental results and early clinical experience. Ann Thorac Surg
61(1):420-425
2) El-Oakley RM, Jarvis JC, Barman D,
Greenhalgh DL, Currie J, Downham DY, Salmons S, Hooper TL (1995) Factors
affecting integrity of latissimus dorsi muscle grafts. J Heart Lung Transplant
14(2):359-365
3) Thomas GA, Lu HP, Isoda S, Hammond RL,
Lu H, Nakajima H, Nakajima HO, Colson M, Stephenson LW (1994) Pericardium-lined
skeletal muscle ventricles in circulation up to 589 days. Ann Thorac Surg
58(4):978-987
4) Gealow KK, Solien EE, Bianco RW, Chiu
RC, Shumway SJ (1993) Conformational adaption of muscle: implications in
cardiomyoplasty and skeletal muscle ventricles. Ann Thorac Surg 56(3):520-526
5) Grabenwöger M, Grimm M, Eybl E, Moritz
A, Müller MM, Bock P, Wolner E (1992) Endothelial cell lining of bioprosthetic
heart valve material. J Card Surg 7(1):79-84
6) Koller R, Girsch W, Huber L, Rab M,
Stöhr HG, Schima H, Losert U, Thoma H, Wolner E (1994) Experimental in situ
conditioning of the latissimus dorsi muscle for circulatory assist by
multichannel stimulation. Art Org 18/7:523-528
7) Koller R, Girsch W, Huber L, Rab M,
Stöhr HG, Schima H, Rokitansky AM, Losert U, Thoma H, Wolner E (1996) Influence
of different conditioning methods on force and fatigue resistance in
chronically stimulated skeletal muscles. PACE 19/2:222-230
8) Lazzara RR, Trumble DR, Magovern JA
(1994) Dynamic descending thoracic aortomyoplasty: comparison with intraaortic
balloon pump in a model of heart failure. Ann Thorac Surg 58(2):366-370
Matthias
Rab,M.D., resident at the Department of Plastic and Reconstructive Surgery,
e-mail:
matthias.rab@univie.ac.at