U.S. patent application number 14/251973 was filed with the patent office on 2014-08-14 for intracorporeal gas exchange devices, systems and methods.
This patent application is currently assigned to UNIVERSITY OF PITTSBURGH - OF THE COMMONWEALTH SYSTEM OF HIGHER EDUCATION. The applicant listed for this patent is UNIVERSITY OF PITTSBURGH - OF THE COMMONWEALTH SYSTEM OF HIGHER EDUCATION. Invention is credited to WILLIAM J. FEDERSPIEL, BRIAN J. FRANKOWSKI, SAMUEL C. LIEBER, KEVIN M. MIHELC.
Application Number | 20140228741 14/251973 |
Document ID | / |
Family ID | 40952467 |
Filed Date | 2014-08-14 |
United States Patent
Application |
20140228741 |
Kind Code |
A1 |
FRANKOWSKI; BRIAN J. ; et
al. |
August 14, 2014 |
INTRACORPOREAL GAS EXCHANGE DEVICES, SYSTEMS AND METHODS
Abstract
A system for intracorporeal gas exchange includes a flexible,
rotatable shaft; a plurality of axially spaced agitation mechanisms
positioned on the rotatable shaft, such that the rotatable shaft
can flex between the axially spaced agitation mechanisms; a
plurality of hollow gas permeable fibers adapted to permit
diffusion of gas between intracorporeal fluid and an interior of
the hollow fibers. The plurality of hollow fibers is positioned
radially outward from the agitation mechanisms. A blood contacting
medical system includes at least one seal, and a purge system via
which a flushing fluid is introduced under pressure over a
blood-side face of the seal and caused to flow through a space on
the blood side of the seal. The system can further include a
rotatable member. The space on the blood side of the seal can be in
fluid connection with an annular space adjacent to a rotating
element operatively connected to the rotatable member.
Inventors: |
FRANKOWSKI; BRIAN J.;
(IMPERIAL, PA) ; FEDERSPIEL; WILLIAM J.;
(PITTSBURGH, PA) ; MIHELC; KEVIN M.; (SOMERVILLE,
MA) ; LIEBER; SAMUEL C.; (JERSEY CITY, NJ) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
UNIVERSITY OF PITTSBURGH - OF THE COMMONWEALTH SYSTEM OF HIGHER
EDUCATION |
PITTSBURGH |
PA |
US |
|
|
Assignee: |
UNIVERSITY OF PITTSBURGH - OF THE
COMMONWEALTH SYSTEM OF HIGHER EDUCATION
PITTSBURGH
PA
|
Family ID: |
40952467 |
Appl. No.: |
14/251973 |
Filed: |
April 14, 2014 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
12865958 |
Aug 3, 2010 |
8734382 |
|
|
PCT/US2009/033399 |
Feb 6, 2009 |
|
|
|
14251973 |
|
|
|
|
61026735 |
Feb 7, 2008 |
|
|
|
Current U.S.
Class: |
604/26 |
Current CPC
Class: |
A61M 1/262 20140204;
B01D 2313/125 20130101; A61M 37/00 20130101; A61M 2005/1403
20130101; B01D 2313/105 20130101; A61M 1/26 20130101; A61M 2005/006
20130101; A61M 1/1678 20130101; B01D 63/02 20130101; B01D 2313/025
20130101; A61M 2025/0057 20130101; B01D 2313/20 20130101; A61M 5/14
20130101; A61M 2025/0031 20130101 |
Class at
Publication: |
604/26 |
International
Class: |
A61M 5/14 20060101
A61M005/14 |
Goverment Interests
GOVERNMENTAL INTEREST
[0002] This invention was made with government support under grant
HL070051 awarded by the National Institute of Health: National
Heart, Lung and Blood Institute. The government has certain rights
in this invention.
Claims
1-18. (canceled)
19. A system for intracorporeal gas exchange comprising: a
rotatable shaft; at least one agitation mechanism positioned on the
rotatable shaft; a plurality of hollow gas permeable fibers adapted
to permit diffusion of gas between intracorporeal fluid and an
interior of the hollow fibers, the plurality of hollow fibers being
positioned radially outward from the agitation mechanism; at least
a first seal, and a purge system via which a flushing fluid is
introduced under pressure over a blood-side face of the at least a
first seal and caused to flow through a space on a blood side of
the at least a first seal.
20. The system of claim 19 wherein the space on the blood side of
the at least a first seal is in fluid connection with an annular
space positioned radially outward from the rotatable shaft.
21. A blood contacting medical system, comprising: at least one
seal, and a purge system via which a flushing fluid is introduced
under pressure over a blood-side face of the at least one seal and
caused to flow through a space on a blood side of the at least one
seal.
22. The system of claim 21 further comprising a rotatable
member.
23. The system of claim 22 wherein the space on the blood side of
the at least one seal is in fluid connection with an annular space
positioned radially outward from the rotatable member.
24. The system of claim 20 wherein the purge system comprises a
first conduit to deliver the flushing fluid the annular space.
25. The system of claim 19 wherein the system further comprises a
medial manifold and a distal manifold, a first end of each of the
plurality of hollow fibers being in fluid connection with the
medial manifold, a second end of each of the plurality of hollow
fibers being in fluid connection with the distal manifold.
26. The system of claim 25 wherein the medial manifold comprising
the at least a first seal and the distal manifold comprises at
least a second seal, the purge system further introducing the
flushing fluid under pressure over a blood-side face of the at
least a second seal and causing the flushing fluid to flow through
a space on a blood side of the at least a second seal
27. The system of claim 26 wherein the space on the blood side of
the at least a first seal is in fluid connection with an annular
space adjacent to a first rotating element operatively connected to
the rotatable shaft and the space on the blood side of the at least
a second seal is in fluid connection with an annular space adjacent
to a second rotating element operatively connected to the rotatable
shaft.
28. The system of claim 27 wherein the purge system comprises a
first conduit to deliver the flushing fluid to the medial manifold
and a second conduit to deliver the flushing fluid to the distal
manifold.
Description
CROSS-REFERENCE TO RELATED APPLICATION
[0001] The present application claims benefit of U.S. Provisional
Patent Application Ser. No. 61/026,735, filed Feb. 7, 2008, the
disclosure of which is incorporated herein by reference.
BACKGROUND OF THE INVENTION
[0003] The present invention relates to intracorporeal gas exchange
devices, systems and methods and, particularly, to devices, systems
and method to effect gas exchange within a patient's vascular
structure.
[0004] The following information is provided to assist the reader
to understand the invention disclosed below and the environment in
which it will typically be used. The terms used herein are not
intended to be limited to any particular narrow interpretation
unless clearly stated otherwise in this document. References set
forth herein may facilitate understanding of the present invention
or the background of the present invention. The disclosure of all
references cited herein are incorporated by reference.
[0005] Lung disease remains one of the major healthcare problems
present in the United States today. Two significant contributors to
lung disease are acute respiratory distress syndrome (ARDS), and
chronic obstructive pulmonary disease (COPD).
[0006] Acute respiratory distress syndrome is defined as a severe
syndrome of inflammation and increased vascular permeability that
is associated with a constellation of clinical, radiologic, and
physiologic abnormalities. It is a non-cardiogenic, acute
respiratory complication characterized by a profound reduction in
systemic oxygenation or ventilation, with an in-hospital mortality
rate of 38.5%. ARDS develops as a result of injury to the
endothelium and epithelial layers of the alveolar membrane from
stimuli such as sepsis, infection or trauma. The breakdown of the
capillary-alveolar barrier leads to alveolar flooding and an
eventual deterioration of gas exchange capability.
[0007] In contrast to ARDS, chronic obstructive pulmonary disease
is a persistent, irreversible condition that slowly progresses over
time. COPD refers to the existence or co-existence of chronic
bronchitis and emphysema and is characterized by obstructed
airways, enlarged air spaces and destruction of lung parenchyma,
occlusion of small airways, and reduced lung elasticity. When
compared to normal, healthy lung function, patients with advanced
stages of COPD can experience 10 to 20 times the work necessary to
facilitate breathing.
[0008] Hospital treatment for ARDS, acute exacerbations of COPD,
and advanced COPD exist in three broad generalizations:
pharmacological, mechanical ventilation and application of an
extracorporeal membrane device (artificial lung). The treatment
modalities depend on the severity of the disease as well as patient
response to prior therapies. As indicated by the high mortality
rates for the lung conditions, however, each therapy has associated
limitations which can interrupt patient discharge from the hospital
and/or recovery.
[0009] The least invasive and first line of defense in treating
lung failure is to administer drugs that may improve the ailing
condition. Multiple pharmacologic strategies have been investigated
since the 1960's, but despite therapeutic benefits, none of the
investigated treatments have demonstrated an ability to improve
patient survival.
[0010] Mechanical ventilation is the most common therapy and serves
to maintain respiratory function by rhythmically inducing a
controlled flow of air into the lungs. In healthy persons, normal
breathing consists of contracting the diaphragm to distend the
lungs and create a negative pressure from atmosphere to lungs
forcing fresh air in. Following oxygen and carbon dioxide gas
exchange, the diaphragm relaxes compressing the lungs and forcing
expiratory air to the external environment. Mechanical ventilation
creates this effect but in an opposing manner; fresh air is driven
into the lungs by positive pressure and expiratory air is pumped
out of the lungs by a negative pressure.
[0011] Mechanical ventilation treatment is associated with multiple
shortcomings termed ventilator-induced lung injury, or VILI. VILI
covers a range of detrimental insults to the lung that can postpone
recovery, cause unfavorable outcomes, or even intensify preexisting
injury.
[0012] The third clinical therapy being administered utilizes a
membrane oxygenator and accompanying flow circuit. The treatment is
denoted as extracorporeal membrane life support (ECLS), often times
referred to as extracorporeal membrane oxygenation (ECMO). An ECLS
device processes patient blood by adding oxygen and removing carbon
dioxide through fiber membrane technology, replicating the natural
gas exchange function of the lungs.
[0013] ECLS is employed under circumstances of severe, reversible
respiratory failure or to patients responding adversely to all
advanced modes of mechanical ventilation. Operation of the circuit
relies on a pump to draw blood from the vena cava, transport it
through the membrane oxygenator, and return the blood either to the
right atrium (venovenous bypass) or aorta (venoarterial bypass).
Patients still receive mechanical ventilation while on ECLS,
however settings are reduced to minimize VILI as a result of the
ability of the oxygenator to exchange blood gases. With less
required work from the lungs, ECLS permits physiological
complications to abate and the therapy can be applied for weeks
barring complications.
[0014] Limitations of ECLS primarily arise from external circuitry
and artificial blood contacting surfaces. To avert thrombosis
within the circuit, patient blood is continuously anticoagulated
and bleeding is a major risk whether internal (intercranial) or
from cannula dislodgement. Patients are paralyzed and/or heavily
sedated to minimize movement causing dislodgement, which creates a
high risk scenario for decubitis ulcers. Also, the continuous
exposure of the blood to artificial surfaces causes platelets to
adhere and/or alter function (thrombocytopenia) requiring the
patient to receive multiple platelet transfusions. In addition, the
ECLS circuit must be constantly monitored for mechanical failures
such as tubing degradation, oxygenator or pump failure, and
presence of gaseous emboli or clot formation. Other noted
complications include sepsis and renal failure. Finally, ECLS
requires a multidisciplinary team to provide care. Staffing and
overall cost of the procedure, as well as restriction to major
medical centers are further limitations to providing this
therapy.
[0015] Recently, there have been attempts to develop intracorporeal
artificial lung devices. Percutaneous, intravascular respiratory
support therapy, which refers to treatments employing a gas
exchange device or an artificial lung device within the vasculature
of a patient to supplement lung function, can be used as an
additional and/or alternative approach to treating ARDS and acute
exacerbations of COPD. Such devices typically use hollow fiber
membrane or HFM technology to achieve gas exchange, oxygenating
blood and removing carbon dioxide. In most situations, carbon
dioxide removal is a primary goal of intravascular devices since
sufficient oxygenation levels can be attained in the clinical
setting through nasal oxygen or specific ventilation modes.
[0016] The ability of intravascular respiratory support devices to
facilitate carbon dioxide removal from the circulation provides an
advantage over sole mechanical ventilation strategies. By
regulating hypercapnia (or elevated carbon dioxide levels in the
blood), intravascular respiratory therapy can allow ventilation at
lower tidal volumes and pressures and thereby eliminate the
deleterious effects that often develop with mechanical ventilation.
Decrease in mechanical ventilation intensity to support patients
has been demonstrated to improve mortality rates. In addition, the
diseased lung tissue experiences a lower workload since the device
itself is performing partial respiratory function. The reduced
workload allows the injured tissue to rest and may improve tissue
recovery.
[0017] ECLS is able to regulate hypercapnia but is associated with
a number of complications resulting from blood circulating outside
the body. Utilizing an intracorporeal/intravascular device
eliminates external circuitry, thereby lessening the risks of
thrombocytopenia and activation of complement resulting from
artificial surfaces. Less artificial surface can also result in
lower anticoagulation levels thereby decreasing bleeding risks.
Overall implementation of the intravascular respiratory therapy is
easier than ECLS making it less demanding on hospital resources,
less expensive, and potentially available in more hospitals.
[0018] Factors regulating gas exchange in the intravascular devices
include gas partial pressures in both blood and fiber lumens, total
HFM surface area, fiber bundle geometry and relative velocity at
which the blood passes by the fiber surfaces. A concentration
boundary layer that forms near the external walls of the individual
fibers can result in significant resistance to diffusion that
limits gas exchange. The boundary layer forms as a result of blood
flow patterns around the fiber walls.
[0019] Existing intravascular devices can be categorized as either
passive or active by the means in which boundary layer reduction is
approached. Passive devices rely on blood flowrates and fiber
bundle geometry to mix flow patterns and disrupt layer formation
near stationary fibers. Active devices implement motion to the
fiber membranes and or to the blood subjacent to the fiber
membranes to disrupt boundary layer formation. In general, to
increase gas exchange, additional surface area can be provided to
boost overall diffusion and/or boundary layers can be reduced to
decrease mass transfer resistance. High relative blood velocity to
fibers facilitates boundary layer reduction.
[0020] As clear to those skilled in the art, a relatively small
size is beneficial for intracorporeal/intravascular respiratory
assist devices. Such devices can, for example, be inserted into a
patient's vascular structure in the manner of a catheter. Reducing
insertion size of the device is desirable to prevent tissue damage
and facilitate placement. However, limiting the size of the
intravascular device also results in limitations upon surface area
of the gas exchange membrane system, thereby limiting gas exchange.
Further, the amount of motion that can be imparted to blood and/or
fibers to increase gas exchange efficiency is limited by potential
damage to surrounding tissue and to the blood. Previously studied
intracorporeal/intravascular devices have met with only limited
success as a result of these limitations.
[0021] It thus remains desirable to develop improved intracorporeal
gas exchange devices, systems and methods that are safe, effective,
and/or amenable to easy insertion.
SUMMARY OF THE INVENTION
[0022] In one aspect, the present invention provides a system for
intracorporeal gas exchange including a flexible, rotatable shaft;
a plurality of axially spaced agitation mechanisms positioned on
the rotatable shaft, such that the rotatable shaft can flex (more
readily) between the axially spaced agitation mechanisms; a
plurality of hollow gas permeable fibers adapted to permit
diffusion of gas between intracorporeal fluid and an interior of
the hollow fibers. The plurality of hollow fibers is positioned
radially outward from the agitation mechanisms.
[0023] The system can further include a support member positioned
radially outward from the agitation mechanisms. The plurality of
hollow fibers can, for example, be positioned radially outward of
and adjacent to the support member.
[0024] Each of the plurality of agitation mechanisms can, for
example, include at least one radially outward extending vane.
[0025] The system can further include a medial manifold and a
distal manifold. A first end of each of the plurality of hollow
fibers can, for example, be in fluid connection with the medial
manifold. A second end of each of the plurality of hollow fibers
can be in fluid connection with the distal manifold. The device can
further include a first fluid path for flow of an inlet gas into
the medial manifold to flow into the first ends of the hollow
fibers and a second fluid path in fluid connection with the distal
manifold into which outlet gas exiting the second ends of the
hollow fibers can flow to exit the device.
[0026] The system can also include an extracorporeal proximal
manifold including an inlet in fluid connection with the first
fluid path and an outlet in fluid connection with the second fluid
path.
[0027] At least a portion of the second fluid path can be within
the flexible rotatable shaft.
[0028] The system can also include a drive shaft and a drive system
to impart rotation to the drive shaft. The drive shaft can, for
example, pass through the proximal manifold and through the second
fluid path to be in operative connection with the rotating
shaft.
[0029] The system can further include a vacuum in fluid connection
with the outlet of the proximal manifold.
[0030] In several embodiments, at least one of the medial manifold
and the distal manifold includes at least one seal, and the device
further includes a purge system via which a flushing fluid is
introduced under pressure over a blood-side face of the seal and
caused to flow through a space on the blood side of the seal. The
space on the blood side of the seal can, for example, be in fluid
connection with an annular space adjacent to a rotating element in
operative connection with the rotatable shaft.
[0031] The medial manifold can, for example, include at least a
first seal, and the distal manifold can include at least a second
seal. The purge system can introduce flushing fluid under pressure
over a blood-side face of the first seal and cause flushing to flow
through a space on the blood side of the first seal. The purge
system can also introduce flushing fluid under pressure over a
blood-side face of the second seal and cause flushing fluid to flow
through a space on the blood side of the second seal.
[0032] The space on the blood side of the first seal can be in
fluid connection with an annular space adjacent to a first rotating
element operatively connected to the rotatable shaft, and the space
on the blood side of the second seal is an annular space adjacent
to a second rotating element operatively connected to the rotatable
shaft.
[0033] The purge system can further include a first conduit to
deliver flushing fluid to the medial manifold and a second conduit
to delivery flushing fluid to the distal manifold.
[0034] In several embodiments, the drive system and drive shaft are
adapted to rotate the rotating shaft at a rate of at least 10,000
to 20,000 RPM.
[0035] The distal manifold, the hollow fibers and the medial
manifold can, for example, be adapted to be inserted
intravascularly and can have a maximum outer diameter of no more
than 33 French or no more than 25 French.
[0036] In several embodiments, the device comprises immobilized
carbonic anhydrase on or in the vicinity of at least a portion of a
first surface of the hollow fibers.
[0037] The system can further include immobilized carbonic
anhydrase on or in the vicinity of at least a portion of a first
surface of the hollow fibers such that the immobilized carbonic
anhydrase comes into contact with the fluid. The first surface can,
for example, exhibit carbonic anhydrase activity of at least 20% of
maximum theoretical activity of the first surface of the hollow
fibers based on monolayer surface coverage of carbonic anhydrase in
the case that the carbonic anhydrase is immobilized on the first
surface of the hollow fibers.
[0038] In another aspect, the present invention provides a system
for intracorporeal gas exchange including: a rotatable shaft; at
least one agitation mechanism positioned on the rotatable shaft; a
plurality of hollow gas permeable fibers adapted to permit
diffusion of gas between intracorporeal fluid and an interior of
the hollow fibers, the plurality of hollow fibers being positioned
radially outward from the agitation mechanism; at least one seal,
and a purge system via which a flushing fluid is introduced under
pressure over a blood-side face of the seal and caused to flow
through a space on the blood side of the seal. The space on the
blood side of the seal can, for example, be in fluid connection
with an annular space adjacent to a first rotating element
operatively connected to the rotatable shaft.
[0039] In a further aspect, the present invention provides a blood
contacting medical system including at least one seal, and a purge
system via which a flushing fluid is introduced under pressure over
a blood-side face of the seal and caused to flow through a space on
the blood side of the seal. The system can further include a
rotatable member. In several embodiments, the space on the blood
side of the seal is in fluid connection with an annular space
adjacent to a rotating element operatively connected to the
rotatable member.
[0040] The devices of the present invention provide significant
advantages over previously studied intracorporeal/intravascular gas
exchange devices. To be effective and obtain clinical acceptance,
such devices are preferably safe in operation, effective, and
amenable to easy insertion. In general, these criteria require an
intravascular device to be relatively small in size. In general,
increased gas exchange efficiencies (both CO.sub.2 removal and
O.sub.2 exchange) of the devices and systems of the present
invention, allow the devices and systems to be fabricated with
relatively small outer diameters. The devices of the present
invention can, for example, be used for respiratory support of
patients with acute or acute on chronic lung failure. In many
situations, the devices will be used for short periods of time (for
example, two weeks or less) in clinical setting. In most cases, a
primary performance goal will be CO.sub.2 removal as sufficient
oxygenation levels can usually be attained in a clinical setting
through nasal oxygen or specific lung protective ventilation modes.
The devices of the present invention can, for example, achieve
CO.sub.2 removal rates of at least 30 ml/min, at least 50 ml/min,
and even at least 75 ml/min under normocapnia conditions. The
devices of the present invention can facilitate ventilatory lung
protective strategies by regulating hypercapnia conditions. Removal
rates can increase under permissive hypercapnia.
[0041] The present invention, along with the attributes and
attendant advantages thereof, will best be appreciated and
understood in view of the following detailed description taken in
conjunction with the accompanying drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
[0042] FIG. 1 illustrates insertion of an embodiment of an
intravascular device of the present invention through the femoral
vein for placement in the vena cava.
[0043] FIG. 2A illustrates a photomicrograph of a fiber wall of a
hollow fiber membrane.
[0044] FIG. 2B illustrates a perspective view of fiber ends of a
hollow fiber membrane bundle.
[0045] FIG. 3 illustrates a model of gas exchange (oxygenation of
and carbon dioxide removal from) blood via a hollow fiber
membrane.
[0046] FIG. 4 illustrates a perspective view of an embodiment of an
intravascular device of the present invention.
[0047] FIG. 5A illustrates an enlarged view of an impeller unit of
the device of FIG. 4A positioned upon the impeller shaft of the
device of FIG. 4.
[0048] FIG. 5B illustrates a cross-sectional view of the device of
FIG. 4.
[0049] FIG. 5C illustrates an enlarged cross-sectional view of the
device of FIG. 4.
[0050] FIG. 5D illustrates an enlarged cross-sectional view of a
proximal manifold of the device of FIG. 4.
[0051] FIG. 6A illustrates a perspective view of the medial
manifold of the device of FIG. 4.
[0052] FIG. 6B illustrates an enlarged cross-sectional view of the
medial manifold.
[0053] FIG. 6C illustrates another enlarged cross-sectional view of
the medial manifold.
[0054] FIG. 6D illustrates another enlarged cross-sectional view of
the medial manifold.
[0055] FIG. 7A illustrates a perspective view of the distal
manifold of the device of FIG. 4.
[0056] FIG. 7B illustrates an enlarged cross-sectional view of the
distal manifold.
[0057] FIG. 8A(1) illustrates a "2-flat" impeller unit geometry
studied in the present invention.
[0058] FIG. 8A(2) illustrates a "4-flat" impeller unit geometry
studied in the present invention.
[0059] FIG. 8A(3) illustrates a "sawtooth" impeller unit geometry
studied in the present invention.
[0060] FIG. 8A(4) illustrates a "paddle" impeller unit geometry
studied in the present invention.
[0061] FIG. 8A(5) illustrates a "centrifugal" impeller unit
geometry studied in the present invention.
[0062] FIG. 8A(6) illustrates a "cone" impeller unit geometry
studied in the present invention.
[0063] FIG. 8A(7) illustrates a "RAX" impeller unit geometry
studied in the present invention.
[0064] FIG. 8A(8) illustrates an "axial" impeller unit geometry
studied in the present invention.
[0065] FIG. 8B illustrates an enlarged perspective view of a
sawtooth embodiment of an impeller unit of the present
invention.
[0066] FIG. 8C illustrates an enlarged front view of a sawtooth
embodiment of an impeller unit of the present invention.
[0067] FIG. 9A illustrates an experimental setup for studies of gas
exchange efficiency of devices of the present invention.
[0068] FIG. 9B illustrates an experimental setup for studies of
fatigue testing in bovine blood to assess performance of the
devices of the present invention.
[0069] FIG. 9C illustrates a study of carbon dioxide exchange rate
per unit area for four impeller geometries of the present invention
as a function of impeller rotation rate.
[0070] FIG. 9D illustrates a study of oxygen exchange rate per unit
area for four impeller geometries of the present invention as a
function of impeller rotation rate.
[0071] FIG. 9E(1) illustrates a graph of experimental results for
carbon dioxide removal as a function of impeller rotation speed in
deionized water and in a viscous fluid for the indicated impeller
unit geometry.
[0072] FIG. 9E(2) illustrates a graph of experimental results for
carbon dioxide removal as a function of impeller rotation speed in
deionized water and in a viscous fluid for the indicated impeller
unit geometry.
[0073] FIG. 9E(3) illustrates a graph of experimental results for
carbon dioxide removal as a function of impeller rotation speed in
deionized water and in a viscous fluid for the indicated impeller
unit geometry.
[0074] FIG. 9E(4) illustrates a graph of experimental results for
carbon dioxide removal as a function of impeller rotation speed in
deionized water and in a viscous fluid for the indicated impeller
unit geometry.
[0075] FIG. 9E(5) illustrates a graph of experimental results for
carbon dioxide removal as a function of impeller rotation speed in
deionized water and in a viscous fluid for the indicated impeller
unit geometry.
[0076] FIG. 9E(6) illustrates a graph of experimental results for
carbon dioxide removal as a function of impeller rotation speed in
deionized water and in a viscous fluid for the indicated impeller
unit geometry.
[0077] FIG. 9E(7) illustrates a graph of experimental results for
carbon dioxide removal as a function of impeller rotation speed in
deionized water and in a viscous fluid for the indicated impeller
unit geometry.
[0078] FIG. 9E(8) illustrates a graph of experimental results for
carbon dioxide removal as a function of impeller rotation speed in
deionized water and in a viscous fluid for the indicated impeller
unit geometry.
[0079] FIG. 9E(9) illustrates a graph of experimental results for
carbon dioxide removal as a function of impeller rotation speed in
deionized water and in a viscous fluid for the indicated impeller
unit geometry.
[0080] FIG. 9F(1) illustrates a graph of experimental results for
carbon dioxide removal as a function of the square root of impeller
rotation speed for the indicated impeller unit geometry.
[0081] FIG. 9F(2) illustrates a graph of experimental results for
carbon dioxide removal as a function of the square root of impeller
rotation speed for the indicated impeller unit geometry.
[0082] FIG. 9F(3) illustrates a graph of experimental results for
carbon dioxide removal as a function of the square root of impeller
rotation speed for the indicated impeller unit geometry.
[0083] FIG. 9F(4) illustrates a graph of experimental results for
carbon dioxide removal as a function of the square root of impeller
rotation speed for the indicated impeller unit geometry.
[0084] FIG. 9F(5) illustrates a graph of experimental results for
carbon dioxide removal as a function of the square root of impeller
rotation speed for the indicated impeller unit geometry.
[0085] FIG. 9F(6) illustrates a graph of experimental results for
carbon dioxide removal as a function of the square root of impeller
rotation speed for the indicated impeller unit geometry.
[0086] FIG. 9F(7) illustrates a graph of experimental results for
carbon dioxide removal as a function of the square root of impeller
rotation speed for the indicated impeller unit geometry.
[0087] FIG. 9F(8) illustrates a graph of experimental results for
carbon dioxide removal as a function of the square root of impeller
rotation speed for the indicated impeller unit geometry.
[0088] FIG. 9F(9) illustrates a graph of experimental results for
carbon dioxide removal as a function of the square root of impeller
rotation speed for the indicated impeller unit geometry.
[0089] FIG. 10 illustrates a graph of motor current versus elapsed
time for a control and for a device of the present invention
including saline purge.
[0090] FIG. 11 illustrates a graphical representation of carbon
dioxide removal rate per unit area in an animal study for three
impeller geometries of the present invention.
[0091] FIG. 12 illustrates carbon dioxide removal rate per unit
area in an animal study for three impeller geometries of the
present invention as a function of impeller rotation rate. For each
device results are shown for two different positions within the
animal vena cava, one of which has the device spanning the right
atrium
[0092] FIG. 13 illustrates blood plasma free hemoglobin (PFHB)
levels as a function of time in two animal studies for two impeller
geometries of the present invention.
[0093] FIG. 14A illustrates a comparison of the operation of a
standard hollow fiber membrane with a membrane of the present
invention in which diffusion is facilitated by carbonic anhydrase
immobilized on or in the vicinity of the membrane surface in
contact with the blood plasma (which occurs because the immobilized
enzyme generates diffusion of bicarbonate ion (an alternative form
of CO.sub.2), which is a substrate of the enzyme, towards the
membrane, at which the bicarbonate ion is converted to CO.sub.2,
which then diffuses into and across the membrane).
[0094] FIG. 14B illustrates a schematic representation of an RFGD
process for immobilization of carbonic anhydrase on a polymeric
material.
[0095] FIG. 15A illustrates a study of the effect of fluid
viscosity on required motor torque.
[0096] FIG. 15B illustrates a tortuous test section used in studies
of the present invention.
[0097] FIG. 15C illustrates a study of measured torque levels of
tubing samples in tortuous configurations.
[0098] FIG. 16A illustrates a schematic representation of
concentric rotating cylinders.
[0099] FIG. 16B illustrates schematically a side view an embodiment
of an impeller catheter external sleeve.
[0100] FIG. 16C illustrates schematically a front view an
embodiment of an impeller catheter external sleeve.
[0101] FIG. 17 illustrates a study of hemolysis generation in
several animal implant studies.
DETAILED DESCRIPTION OF THE INVENTION
[0102] In several embodiments, the present invention provides
intracorporeal respiratory assist devices that can, for example, be
inserted in a percutaneous manner similar to insertion of a
catheter which are operable to at least partially support native
lung function in patients with, for example, acute respiratory
distress syndrome and/or acute exacerbations of chronic obstructive
pulmonary disease. Intravascular devices of the present invention
are sometimes referred to herein as catheters. As described above,
primary current clinical therapies (including pharmacotherapy,
mechanical ventilation, and ECLS) are associated with patient
injury and high mortality rates. The devices of the present
invention can be used in combination with or as an alternative to
such clinical therapies to reduce patient injury and mortality.
[0103] Although the devices, systems and methods of the present
invention are discussed primarily herein in connection with
oxygenation of and removal of carbon dioxide from blood, one
skilled in the art appreciates that the devices, systems and
methods of the present invention can be used for many other types
of gas exchange (for example, for delivery of anesthesia).
[0104] As used herein and in the appended claims, the singular
forms "a," "an", and "the" include plural references unless the
content clearly dictates otherwise. Thus, for example, (unless
clearly indicated otherwise) reference to "an impeller unit"
includes a plurality of impeller units and equivalents thereof
known to those skilled in the art, and so forth, and reference to
"the impeller unit", is a reference to one or more such impeller
units and equivalents thereof known to those skilled in the art,
and so forth.
[0105] The devices of the present invention can, for example, be
inserted through the femoral vein for placement in the vena cava in
the manner of a catheter to actively process venous blood as
illustrated in FIG. 1. In several embodiments, the devices of the
present invention include a gas exchange membrane system including,
for example, a plurality of hollow fiber membranes (HFM), sometime
referred to herein as a HFM bundle, that supplement oxygenation and
carbon dioxide removal through diffusional processes as illustrated
in FIGS. 2A through 3.
[0106] In general, the intracorporeal devices and systems of the
present invention can be of reduced size when compared to previous
devices designed for intravascular respiratory assist, which
facilitates percutaneous insertion. Reduction in the size, however,
can limit available membrane (for example, HFM) surface area. In a
number of embodiments, efficiency in removal of CO.sub.2 per unit
membrane surface area is improved to offset limits in available
surface area. Furthermore, oxygen (O.sub.2) transfer rates per unit
surface are also improved as compared to currently available
intracorporeal and other devices. In several embodiments of the
devices of the present invention, agitation elements or mechanisms
such as rotating elements, impellers or rotors (sometimes referred
to herein collectively as impellers) are positioned within a
stationary HFM bundle to impel flow or create forced flow within
and through the HFM bundle to increase efficiency of gas exchange
per unit surface area. Internalization of rotational components
within the HFM bundle of the devices of the present invention
assists in protecting the surrounding tissue from damage.
[0107] Devices of the present invention were evaluated in vitro and
in vivo to assess design and performance Acceptable hemolysis
levels were observed with high-speed rotational impellers within
HFM bundles. Standard gas exchange characterization tests in, for
example, deionized or DI water showed an increase in CO.sub.2
removal efficiency and O.sub.2 transfer/exchange as compared to
previously disclosed percutaneous respiratory assist devices or
oxygenators. Several embodiments of devices of the present
invention were evaluated in a bovine model and an average
efficiency of 513.+-.20 ml CO.sub.2/min/m.sup.2 was attained at
20,000 RPM.
[0108] As described above, membrane oxygenators employ hollow fiber
membrane (HFM) technology to achieve gas transfer. The HFMs can,
for example, be made from polymers such as polymethylpentane or
polypropylene that are extruded or otherwise formed into
microporous tubes. The miniature tubes can further be arranged into
bundles that are used in a number of currently available
oxygenators. FIGS. 2A and 2B displays a magnified view of HFMs.
Wall porosity enables the membranes to exchange oxygen and carbon
dioxide gases while blood flows external to the fibers. The walls
of the fibers are hydrophobic and prevent fluid from leaking into
the lumens.
[0109] The process driving gas exchange in HFM is diffusion.
Oxygenators can, for example, be setup to process venous blood that
is high in carbon dioxide (CO.sub.2) concentration and low in
O.sub.2 concentration. Gas rich in oxygen or pure oxygen (O.sub.2)
gas is passed through the HFM lumens at a pressure (for example,
pO.sub.2.apprxeq.740 mmHg) that is greater than the gas
tension/partial pressure of oxygen in the blood. A transmembrane
concentration gradient is thus created between venous blood
(pO.sub.2=38 mmHg, pCO.sub.2=45 mmHg) and fiber lumen. The high
concentration of CO.sub.2 in the blood favors diffusion into the
low concentration side within the HFM lumen, and conversely the
high concentration of O.sub.2 in the HFM favors diffusion into the
low concentration blood-side. A schematic representing CO.sub.2
diffusion through a hollow fiber membrane is shown in FIG. 3. Blood
passing through the HFM bundle is thus supplemented with oxygen and
removed of carbon dioxide in a fashion similar to the natural
lung.
[0110] During insertion, the devices or catheters of the present
invention can, for example, extend through the femoral vein into
the vena cava to span the right atrium of the heart. FIG. 1
illustrates insertion of a device 10 of the present invention
within the vena cava prior to reaching a position spanning the
right atrium. A position spanning the right atrium allows the
device to process all venous blood returning from the major vessels
of the body. Furthermore, flow from the vena cava into the right
atrium creates an additional blood mixing component since
cross-flow will occur. This mixing regime may influence boundary
layer formation.
[0111] Pure oxygen gas can, for example, be shuttled through the
HFM bundle under a vacuum so that O.sub.2 transport occurs by
diffusion only. The vacuum assists in ensuring that oxygen sweep
gas is not bubbled through the pores of the HFM into the blood as a
result of a high transmembrane pressure gradient. In several
embodiments described in further detail below, manifolds at the
proximal and distal ends of the fiber bundle collectively organize
the fibers to permit sweep gas flow in parallel through the bundle.
Conduits in fluid connection with the manifolded bundle create a
single, continuous flow path for sweep gas. Flowrate through the
device can, for example, be maintained approximately 40-60 times
greater than the CO.sub.2 exchange rate to maximize efficiency.
See, for example, Federspiel W J, Hattler B G: Sweep gas flowrate
and CO.sub.2 exchange in artificial lungs. Artif Organs 20:
1050-1052, 1996, the disclosure of which is incorporated herein by
reference. Lower flowrates can lead to CO.sub.2 accumulation within
the HFMs which reduces the gas concentration driving gradient
responsible for diffusion. The flowrate range set forth above
assists in assuring that gas transfer is maximized and not
dependent upon sweep gas flowrate.
[0112] Preferably, the devices of the present invention are
operable without causing harmful effects to the blood. In that
regard, actively mixing the blood can cause hemolysis if
erythrocytes experience excessive shear stresses and exposure
times. In several embodiments, the devices of the present invention
were designed (but not optimized) to minimize hemolysis (for
example, as determined by maintaining acceptable blood plasma free
hemoglobin (PFHB) levels).
[0113] As described above, in reducing the insertion size of the
catheter, there is a resultant loss in HFM surface area. In the
devices of the present invention, a novel active mixing approach is
incorporated to reduce concentration boundary layers to
counterbalance the loss of HFM surface area and provide acceptable
overall gas exchange. In several embodiments, at least one rotating
impeller unit was positioned within the HFM bundle to distribute
fluid through the bundle. The impeller units were designed so that
the HFM bundle would remain stationary and enclose the rotating
components.
[0114] Gas exchange tests were first performed to evaluate gas
exchange efficiency. The tests were further subdivided to analyze
the effects of impeller geometry on gas exchange in both the
standard deionized (DI) water solution and a blood representative
viscous solution. Tortuous loop tests were implemented in early
stages to ensure mechanical operation in severe bending
environments. Fatigue tests were conducted in bovine blood to
assess hemolysis. The devices of the present invention were also
tested in four acute calf studies to examine gas exchange between
different impeller geometries.
[0115] One embodiment of a device 10 of the present invention is
illustrated in FIGS. 4 through 7B. Device 10 includes a plurality
of spaced internal impeller units or elements 100 capable of
rotating at, for example, 20,000 RPM to impart fluid motion and to
force fluid radially outward. In the embodiment as illustrated, for
example, in FIGS. 4 and 5A, impeller units 100 included four flat
vanes 110 extending radially outward from a common cylindrical base
104 which was positionable on flexible impeller shaft 120. The
impeller system can also be described as a single agitation or
impeller system including a plurality of groups of vanes that are
axially spaced on shaft 120. An HFM bundle 200, as described above,
enclosed rotating impeller units 100. In the illustrated
embodiment, HFM fabric 200 was formed around a support 300 in the
form of, for example, a stainless steel spring or helical coil. In
several devices 10 studied in the present invention, a 5 mm
diameter cavity existed within stent or spring 300 to create a
hollow bundle module for impeller units 100 to reside. Spring 300
served as both a backbone or support to support HFM fabric/HFM
bundle 300 and as a barrier to shield HFM bundle 300 from rotating
vanes 110 of impeller, even in tortuous configurations. Support 300
was secured to a medial manifold 400 and a distal manifold 500 and
stretched to provide ample spacing for fluid passage. As
illustrated in, for example, FIG. 5B, in several studied
embodiments, HFM bundle was approximately 30 cm in length and 12 to
14 (14 in the illustrated embodiment) impeller units 100 were
positioned along shaft 120 with a spacing of approximately 0.5
inches therebetween. In several studied embodiments of device 10,
CELGARD.RTM. ARX30-240 hollow fiber membrane fabric (Celgard, Inc.,
of Charlotte, N.C.) was used as fiber bundle 200. Characteristics
of the individual HFMs and the HFM bundle studied are set forth in
Table 1.
TABLE-US-00001 TABLE 1 ARX30-240 HFM Material Polypropylene Outer
Diameter (.mu.m) 300 Inner Diameter (.mu.m) 240 Average Pore Size
(.mu.m) 0.03 Porosity (%) 40 Fibers per Inch (FPI) 54 HFM Bundle
Bundle Length (cm) 30 Outer Diameter (mm) 8 # of Fibers 250 Fiber
Layers 4 Wefts none removed Surface Area (m.sup.2) 0.07
[0116] Hollow fiber membranes in artificial lungs (blood
oxygenators or respiratory assist devices) are typically made using
polyolefin polymers, with polypropylene, polyethylene and
polymethylpentene being commonly used materials. The hollow fiber
membranes are created by extrusion or other manufacturing processes
to create a microporous fiber wall with sub-micron sized pores
spanning the walls. These microporous fibers are adequate in
commercial blood oxygenators for short term blood contact. In
respiratory assist devices and artificial lungs required for
longer-term blood contact (>6 hours), the microporous fibers can
leak plasma into the fiber lumens and create a problem known as
plasma wetting. To prevent or retard plasma wetting, composite or
asymmetric microporous hollow fiber membranes can be used in the
devices and systems of the present invention. A composite HFM is a
standard microporous hollow fiber membrane on which a thin coating
of a nonporous, permeable polymer is applied. As the nonporous
coating should be very gas permeable, the coating is often made
from a highly gas permeable polymer such as siloxane polymers. An
asymmetric microporous hollow fiber membrane is one in which the
fiber wall is fabricated so that the porosity varies across the
wall and goes to zero at the fiber surface. The pores are then
essentially sealed by the same gas permeable material from which
the fiber is made.
[0117] Studied devices 10 were designed so that impeller units 100
could be easily interchanged to rapidly evaluate different impeller
geometries. Impeller units 100 were mounted directly to impeller
shaft 120 via adhesive. Fiber bundle 200 was constructed from 300
.mu.m OD fibers 210 (see FIG. 1) in fabric form, with fibers 210
potted into medial manifold 400 and distal manifold 500. In that
regard, epoxy was applied to fiber bundle 200 to fix manifolds 400
and 500 to bundle 200.
[0118] Distal manifold 500, medial manifold 400 and a proximal
manifold 600 operated, in part, to separate gas pathways in device
10. Proximal manifold 600 was used to introduce, for example,
oxygen sweep gas via an inlet port 610 and also included an outlet
port 620 for exhaust gas removal. Oxygen gas was introduced into
inlet port 610 of proximal manifold 600 and directed through HFM
bundle 200. Distal manifold 500 redirected exhaust gas back through
impeller tubing 120 towards proximal outlet port 620. Proximal
manifold 600 also operated to couple a device drive line 700 to a
drive system such as a servomotor drive system 800. As described in
more detail below, medial manifold 400 and distal manifold 500
included bearings and seals to stabilize rotation and prevent fluid
entry into device 10. Manifolds 400, 500 and 600 were stationary
during operation, thereby providing a base to attach support 300
and HFM bundle 200. Flow through the device was driven by a vacuum
pump 750 connected to the proximal manifold outlet port 620. Drive
system 800, including a DC servomotor and an amplifier, provided
the required torque to actuate and control drive line 700.
[0119] In the illustrated embodiment, the sweep gas flowed into
(via inlet port 610) and through a flexible pneumatic shaft or tube
350 (fabricated, for example, from a polymer such as polyurethane)
and into medial manifold 400. Drive line 700 passed through a
conduit 710 which was positioned within tube 350. As illustrated by
arrow S in FIG. 5D, inflowing sweep gas flowed in an annular
passage between conduit 710 and the inner wall of tube 350. At
medial manifold 400, the sweep gas (represented by arrows S in, for
example, FIGS. 6C and 6D) was distributed into a first end of
hollow fibers 210 of hollow fiber bundle 200. Upon exiting a second
end of hollow fibers 210, the sweep gas was collected within distal
manifold 500. In distal manifold 500, the returning sweep gas
(represented by arrows R) entered a fluid path within a connector
510 rotatably positioned within distal manifold 500. Returning
sweep gas passed from connector 510 into a central fluid path
within impeller shaft 120, with which connector 510 was in
operative and fluid connection.
[0120] The returning sweep gas passed through the center of device
10 (within the central fluid path within impeller shaft 120) back
to medial manifold 400 as, for example, illustrated in FIGS. 6C and
6D. In that regard, impeller shaft 120 was in operative and fluid
connection with a distal end of a connector 410 rotatably
positioned within medial manifold 400. Drive line 700 was in
operative connection with a proximal end of connector 410 and
imparted rotational motion to connector 710 and thereby to impeller
shaft 120. From medial manifold 400, returning sweep gas passed
through an annular space between drive line 700 and an interior
wall of conduit 710 to proximal manifold 600. The returning sweep
gas exited device 10 by flowing through conduit 710 to outlet 620
of proximal manifold 600 as, for example, illustrated in FIG.
5D.
[0121] Once again, a negative pressure gradient (created, for
example, via a sealed vacuum) was used to drive oxygen flow from
inlet port 610 to exhaust gas outlet 620. The sweep gas flowrate
requirement for a prototype exchanging 75 ml CO.sub.2/min was
determined to be at least 3 LPM using the 40-60 multiplier
discussed above. The main resistance to flow in device 10 was the
HFM bundle and the resistance was estimated according to the
following equation assuming oxygen to be a compressible gas:
P o 2 - P L 2 = 2 ( 128 .mu. L N .pi. d 4 ) P atm Q RTP ( 1 )
##EQU00001##
[0122] where P.sub.o is upstream pressure, P.sub.L is downstream
pressure, t is gas viscosity, L is fiber length, N is total number
of fibers, d is fiber inner diameter, P.sub.atm is atmospheric
pressure, and Q.sup.RTP is sweep gas flowrate (LPM). For a flowrate
of, for example, 3.5 LPM, the corresponding pressure drop over the
bundle was approximately 50 mmHg. See Eash H J, Frankowski B J,
Hattler B G, Federspiel W J: Evaluation of local gas exchange in a
pulsating respiratory support catheter. Asaio J 51: 152-157, 2005,
the disclosure of which is incorporated herein by reference. Vacuum
pump 750 was capable of producing 9 LPM flow at this pressure head,
and therefore the bundle design was adequate to attain sufficient
sweep flow.
[0123] High-speed electromechanical drive system 800 included a
brushless DC-servomotor (Series 2444-024B, MicroMo Electronics Inc.
of Clearwater, Fla.) and 2-quadrant PWM servo amplifier (Series BLD
3502, MicroMo Electronics Inc.) and provided rotation and control.
A conditional analog voltage of 0-5 volts set by a potentiometer
dictated motor speed. The unit was powered by a 24 volt AC/DC quad
output power supply (Series NFS110-7602P, Artesyn Technologies,
Inc. of Boca Raton, Fla.). The components were housed within a
single box for ease of use and protection from electrical failure.
Drive system 800 included a mini DC cooling fan to remove heat from
the power supply. Motor speed was measured with a digital rate
indicator (Model CUBS, Red Lion Controls of York, Pa.). A wire
connection was introduced to the amplifier to tap voltage pulses
returning from the servomotor hall-sensors. Connection from the
amplifier to the digital rate indicator enabled pulses to be
counted and displayed. A mechanism for measuring motor torque was
also included in the setup. The torque generated by the servomotor
was directly proportional to the current requirement for motor
operation. A wire connection was made to directly measure the DC
current output from power supply to amplifier. The measurement was
representative of the supplied current from amplifier to motor and
was used to calculate motor torque by the servomotor torque
constant. Finally, a fuse was placed in series from power supply to
amplifier. The fuse automated the system to shutdown upon motor
stall and protected the motor from damaging current levels.
[0124] As discussed above, each of medial manifold 400 and distal
manifold 500 included a seal and a bearing to, for example, prevent
blood intake and to support concentric impeller rotation. A diamond
like carbon coating (DLC) can, for example, be employed where the
rotating shaft and the seals make contact to, for example, reduce
the coefficient of friction and add lubricity. Medial manifold 400
is illustrated in, for example, FIGS. 6A through 6D. A rotary seal
404 manufactured from nitrile rubber (#711800, Chicago Rawhide/SKF
USA, Inc. of Elgin, Ill.) was used in the studied embodiments. Seal
404 accommodated shaft diameters of 0.110''-0.140'' and had an
outer diameter of 0.313 inches. Seal 404 was modified to an outer
diameter of 0.250 inches to meet size specifications. Seal 404 was
also configured with a shielded, mini ball bearing 406 (#35380815,
MSC Industrial Supply Co. of Melville, N.Y.) possessing a 0.125
inches shaft diameter and 0.250 inch outer diameter. In several
studied embodiments, a generally cylindrical sleeve of material 408
(such as VESPEL.RTM., a polyamide based polymeric material
available from DuPont) was positioned within the housing of medial
manifold 400 adjacent to the distal or blood-side face of seal 404.
Material 408 was 0.5 inches in length and had an inner diameters of
0.113 inches. That diameter created a 0.0005 inch clearance between
a connector section 122 of impeller shaft 120 and material 408.
[0125] Distal manifold 500 is illustrated in FIGS. 7A and 7B. As
with medial manifold 400, a rotary seal 504 fabricated from nitrile
rubber (#711800, Chicago Rawhide/SKF USA, Inc.) was used. Seal 504
accommodated shaft diameters of 0.110''-0.140'' and had an outer
diameter of 0.313''. Seal 504 was modified to an outer diameter of
0.250 inches to meet size specifications. Seal 504 was also
configured with a shielded, mini ball bearing 506 (#35380815, MSC
Industrial Supply Co.) having a 0.125 inch shaft diameter and 0.250
inches outer diameter. As described above in connection with
generally cylindrical sleeve of material 408, a generally
cylindrical sleeve of material 508 (VESPEL) was positioned within
the housing of distal manifold 500 adjacent to the proximal or
blood-side face of seal 504. As was also the case with material
408, material 508 was 0.5 inches in length and had an inner
diameters of 0.113 inches. The diameter created a 0.0005 inch
clearance between a connector section 124 of impeller shaft 120 and
material 508.
[0126] In several studied embodiments of device 10, the selected
material for flexible impeller shaft 120 was wire reinforced
PEBAX.RTM. (a polyether block amide available from Arkema France
Corporation of Colombes, France) catheter tubing (#N98-TOL-664,
available from New England Wire Technologies of Lisbon, N.H.). The
tubing was chosen from several samples based on tortuous loop test
results discussed in Appendix A. Impeller shaft 120 was operatively
connected to drive shaft or drive line 700 via connector 410 as
described above, which was produced from an internal component of
bi-directional 0.072 inch diameter shaft (#067N133C, S.S. White
Technologies of Piscataway, N.J.) that was coupled to servomotor
system 800 via proximal manifold 600 as also described above.
[0127] Tortuous loop tests described in the Experimental Section
hereof demonstrated that 0.5 inch length impeller units 100 with
0.5 inches spacing therebetween upon flexible impeller shaft 120
were suitable to attain desired impeller flexibility. Moreover, the
spacing between impeller units 100 can promote recirculating flow
from impeller segments into nonimpeller regions and thereby promote
gas exchange. Various impeller/vane units 100 evaluated in device
are shown in FIGS. 8A(1) through 8A(9). Enlarged illustrations of a
sawtooth impeller unit 100 of the present invention are set forth
in FIGS. 8B and 8C. A cylindrical blank was also made as a control
and is not displayed. Paddle and axial impeller units were
fabricated in-house. Axial units were produced by turning static
mixers (#04667-18, Cole Parmer Instrument Co. of Vernon Hills,
Ill.) on a lathe to a 0.152'' outer diameter. Other impeller units
100 were produced by SLA rapid prototyping. 2-flat, 4-flat,
sawtooth, paddle, centrifugal and cone were developed as radial
units. The RAX vanes included a combination of radial and axial
geometries to assist in carrying fluid to the radial blades of the
unit. All studied impeller units 100 had inner diameters of 0.084
inches, outer diameters of 0.156 inches, and lengths of 0.5
inches.
[0128] WATERSHED.TM. 11120 stereolithography resin available from
DSM Somos of Elgin, Ill. was used for impeller unit manufacture.
The material was compatible with SLA prototyping and also resistant
to water adsorption/swelling. Limited clearance within the cavity
within HFM bundle 200 made water resistance a desirable property.
Once again, materials/sleeves 408 and 508 were made from DuPont
VESPEL; which has demonstrated applicability as a low wear material
which absorbs very little water. Drive components, medial manifold
400 and distal manifold 500 were fabricated from 316 stainless
steel on a Hardinge lathe. Proximal manifold 600 was formed from
acrylic polymer. Inlet and exhaust tubings were made of
polyurethane plastic. Component mating was accomplished using the
2:3 polycin to vorite epoxy or PERMATEX.RTM. clear RTV silicone
adhesive sealant (available from Permatex, Inc. of Hartford,
Conn.). Mechanical drive parts were connected using silver
solder.
[0129] Initial evaluations of a number of devices 10 including
PEBAX tubing for impeller shaft 120 in an animal model revealed
mechanical failures of such an impeller shaft 120. Prototype
dissections also showed accumulations of paste-like blood
constituents in the external (blood side) materials/sleeves 408 and
508 that were believed to cause variations in friction, ultimately
resulting in the tubing failure. The annular space between
connector sections 122 and 124 and material/sleeves 408 and 508,
respectively, may, for example, have undesirably served as a
pathway through which blood traveled to contact seals 404 and 504.
In that regard, the blood may have penetrated the external housing
prior to contacting seals 404 and 504. Shear stress calculations
estimated a stress exposure of 5,000 dynes/cm.sup.2 at 20,000 RPM
(Appendix B.1). Paul et al. found the shear stress threshold for
hemolysis to be 3750 dynes/cm.sup.2 and 4250 dynes/cm.sup.2 at
exposure times of 1.4 and 0.425 seconds respectively. Paul R, Apel
J, Klaus S, Schugner F, Schwindke P, Reul H: Shear stress related
blood damage in laminar couette flow. Artif Organs 27: 517-529,
2003, the disclosure of which is incorporated herein by reference.
The shear stress level was thus possibly causing hemolysis and
contributing to the accumulation of blood constituents.
[0130] The PEBAX tubing also experienced non-uniform rotation
between medial manifold 400 and distal manifold 500 that created
torsional stresses along the tubing length. The combination of
high-speed and torque caused the PEBAX tubing to fail in two ways:
either the tubing completely ruptured leaving a helical pattern in
the direction of motor rotation, or the hollow tubing collapsed on
itself with the same observed pattern. The pattern indicated a
delay or complete seizure in impeller angular velocity in the
distal manifold causing the tubing to overturn.
[0131] Given the failure of PEBAX tubing for impeller shaft 120 in
several studies, a concurrent effort was undertaken to prevent
blood thrombosis and improve impeller tubing strength. A flushing
fluid (for example, heparinized saline) purge of seals 404 and 504
and exterior (blood-side) annular spaces between connector sections
122 and 124 and materials/sleeves 408 and 508, respectively, was
implemented within medial manifold 400 and distal manifold 500,
respectively, to prevent thrombus and blood damage. The purge
benefited device 10 three-fold by: (1) impeding blood penetration
into the annular spaces adjacent connector sections 122 and 124,
(2) preventing direct blood contact with seals 404 and 504, and (3)
removing friction-generated heat from seals 404 and 504.
[0132] To implement a saline purge, orifices were drilled into
medial manifold 400 and distal manifold 500 to deliver the saline
to the exterior face of the seals 404 and 504 and flush through the
annuli between materials/sleeves 408 and 505 and connector sections
122 and 124, respectively. As, for example, illustrated in FIGS. 5D
and 6D, two 304 stainless steel hypo tubing lines 900 and 910
(#HTX-25R-24, Small Parts Inc. of Logansport, Ind.) with IDs of
0.010'' and ODs of 0.020'' were passed through tubing 350 to enter
the annular path between conduit 710 and tubing 350. Alternatively,
saline (and/or other flushing fluid(s)) can enter device 10 via a
saline port 630 formed in proximal manifold 600 (see FIG. 4), which
can, for example. be in fluid connection with one or more delivery
conduits such as hypo tubing lines 900 and 910. In the studied
embodiments, hypo tubing line 900 connected to medial manifold 400
to deliver saline. Hypo tubing line 910 passed through medial
manifold 400 and along the exterior of HFM bundle 200 to connect to
proximal manifold 500. Sufficient saline pressure (for example, at
least 200 mmHg) was used to prevent blood seepage. Using a pressure
of 200 mmHg as a guideline, the flowrate of saline through a single
annular space to generate the pressure standard was calculated to
be 12 ml/hr (see Experimental Section (B.2.)). A minimal saline
flowrate of 12 ml/hr was used in all studies.
[0133] A controlled volume of flushing fluid or liquid
(saline/heparin) was thus introduced into an engineered space (the
annular space between materials/sleeves 408 and 508 and connector
sections 122 and 124, respectively) between the blood and seals 404
and 504. As discussed above, this space was approximately 0.5
Inches long, and there was an approximately 0.0005 inch gap between
the sleeves/seal housings and the rotating impeller shaft connector
sections. The flushing fluid entered into the bearing seal housing
of the manifolds on the external side (or blood side) of seals 404
and 504 and traveled through the engineered gap wherein it
purged/prevented inflow of blood. This space acted as a barrier
between the blood and the seals and rotating components. The
flushing fluid purge minimized or eliminated contact between blood
and rotating components that cause thrombus and blood damage by
heat or friction generated at the interface of a rotating shaft and
the seal. The flushing fluid/saline also provided lubrication and
cooling to the shaft-seal interface.
[0134] To improve impeller shaft strength, a stainless steel coil
(#HVT18642-01, available from Heraues Vadnais Inc. of Saint Paul,
Minn.) was used in several studies for impeller shaft/tubing 120.
The coil included 4-filar units that served to individually
distribute torque loads. The coil was made to dimensions of 0.060''
ID and 0.083'' OD from 0.010'' 304 stainless steel wire wound in a
clockwise direction. This embodiment of impeller tubing 120 was
formed as a hollow flexible drive shaft and could transmit higher
loads than plastic tubing. PEBAX heat shrink tubing
(#P2-140-006-CLR, available from Cobalt Polymers of Cloverdale,
Calif.) with a wall thickness of 0.006'' was applied to the coil to
produce the finalized, sealed tubing used in devices 10. Impeller
shaft 120 was tested in a torsion testing apparatus. Impeller
shafts 120 formed from stainless steel coil achieved much larger
torsional deflections prior to failure than the previously studied
PEBAX tubing and were integrated into device 10 for further
studies. Although the stainless steel coil embodiment of impeller
shaft 120 exhibited greater strength than the PEBAX tubing
embodiment, shear stresses were decreased sufficiently via use of a
flushing fluid purge as described above, that the PEBAX tubing
embodiment of impeller shaft 120 could have been implemented
without failure when used in connection with such a purge.
[0135] As described above, various impeller units 100 geometries
were evaluated in bench studies, which were followed by studies in
a bovine model. Gas exchange tests of devices 10 were conducted in
vitro to assess gas exchange efficiencies with the various impeller
units 100. Devices 10 were tested in the standard DI water and a
viscous solution similar to blood to provide insight regarding
impeller geometry performance.
[0136] The in vitro test was an idealized simulation of the vena
cava environment and provided rapid device characterizations
allowing device to device comparisons and indication of trends that
could be observed in vivo. Testing was performed by introducing the
device into separate fluid and gas circuits. The fluid loop
re-circulated de-ionized water past the device. Water was thus the
blood representative fluid and served as the gas carrying medium.
Device 10 was placed in the fluid circuit within a 7/8'' diameter
acrylic test section that simulated the vena cava. The fluid
circuit contained a reservoir, fluid pump, thermometer, heater,
flow meter, pressure transducer and de-oxygenator, and all were
connected with standard laboratory TYGON.RTM. tubing (plastic
tubing) available from Saint-Gobain Performance Plastics
Corporation of Aurora, Ohio. A schematic of the test setup is
illustrated in FIG. 9A.
[0137] The de-oxygenator was used to establish test section water
inlet conditions. Conditions set were a pO.sub.2 of 25-35 mmHg and
a pCO.sub.2 of 47-53 mmHg, which represented physiological
normocapnia venous blood. The water conditions were monitored on an
ABL 555 soluble gas analyzer machine. Water flowrate through the
loop was maintained at 3 LPM with a temperature of 37.degree. C.,
and the water gauge pressure within the test section was maintained
at approximately 10 mmHg. No micro-bubble formation was
observed.
[0138] The gas circuit established the pathway for oxygen sweep gas
introduction and exhaust gas analysis. Device inlet 610 was
connected to a pure oxygen source maintained at atmospheric
pressure. Upon exiting device 10 via outlet 620 the sweep gas was
passed through a cold moisture trap for dehumidification and then
directed through a flow meter and CO.sub.2 analyzer. The flowrate
was driven by a vacuum pressure station that pulled sweep gas
through the entire circuit. Gas pressure drop over device inlet 610
to outlet 620 was monitored using a pressure transducer.
[0139] In gathering gas exchange data, water conditions were first
measured to ensure that inlet pCO.sub.2 and pO.sub.2 were within
desired range. Acceptance of inlet conditions was followed by
recording of water temperature, flowrate, and pressure in the fluid
circuit. Also, the sweep gas flowrate, device pressure drop, motor
torque, and exhaust CO.sub.2 concentration were recorded. The RPM
rate was then changed and the procedure was repeated to obtain a
minimum of two random data points for each RPM to ensure
repeatability.
[0140] CO.sub.2 removal, V.sub.CO2 was calculated for the rates by
the following equation:
V.sub.CO2=Q.sup.STPF.sub.CO2 (2)
where Q.sup.STP is the flowrate of sweep gas at standard
temperature and pressure and F.sub.CO2 is the fraction of CO.sub.2
in the exhaust of the catheter, as measured per the CO.sub.2
analyzer. The exchange rate was then divided by the HFM surface
area and normalized to the target pCO.sub.2 inlet of 50 mmHg to
calculate efficiency (VCO.sub.2) and normalize variability
associated with small changes in inlet pCO.sub.2
concentrations:
VCO 2 = V CO 2 50 p CO 2 INLET SA ( 3 ) ##EQU00002##
[0141] Gas exchange testing evaluated the effect of impeller
geometry on CO.sub.2 removal efficiency and O.sub.2 exchange
efficiency. CO.sub.2 removal efficiency in deionized water for four
different impeller geometries as a function of impeller rotation
rate is illustrated in FIG. 9C. O.sub.2 exchange efficiency in
deionized water for four different impeller geometries as a
function of impeller rotation rate is illustrated in FIG. 9D. As
described above, in addition to testing in deionized water,
CO.sub.2 removal efficiency was also measured in a viscous
solution. The viscous solution test evaluated the interplay between
the impeller transmitting motion to the fluid versus the bundle
impeding fluid velocity. A higher viscosity solution, such as
blood, transmits fluid motion better thus increasing relative
velocity to fiber, whereas the bundle region impedes fluid motion
and decreases relative velocity. A 0.4% w/w concentration of
carboxymethylcellulose (CMC) sodium salt (#C5678-500G,
Sigma-Aldrich Inc.) and DI water was interchanged with homogenous
DI water to test prototypes in both solutions. The viscosity of the
CMC solution was measured at 37.degree. C. during each test using a
capillary viscometer (#150, Cannon Instrument Co) to ensure that
the viscosity remained in a range of 2.5-2.7 cP. CMC was used as
the solution thickener because a minute concentration could be
added to water without altering the gas solubility properties.
[0142] Maximum gas exchange in the studies occurred at the peak
rotational rate of 20,000 RPM. FIGS. 9E(1) through 9E(9) display
prototype characterization plots of VCO.sub.2 versus RPM in both
testing solutions for the nine impeller unit geometries testes.
Maximum CO.sub.2 removal efficiencies for all geometries are listed
in Table 2. The 4-flat, RAX, sawtooth, and 2-flat geometries
performed similarly. Without limitation to any mechanism, it was
hypothesized that the centrifugal geometry had a lower efficiency
compared to other radial geometries since the ends were enclosed
and fluid could not penetrate the vanes as well as open-ended
geometries. The cone geometry was the worst performing unit, even
when compared to the cylindrical control. Without limitation to any
mechanism, it was hypothesized that the tapered end caused the cone
unit to have a lower average linear velocity at the surface, which
transmitted less momentum to fluid particles, causing the cone
geometry to underperform even the control.
TABLE-US-00002 TABLE 2 VCO.sub.2 Impeller (ml CO.sub.2/min/m.sup.2)
control 307 .+-. 3 4-flat 529 .+-. 21 RAX 514 .+-. 11 sawtooth 482
.+-. 6 2-flat 482 .+-. 3 paddle 457 .+-. 1 axial 446 .+-. 3
centrifugal 341 .+-. 2 cone 264 .+-. 10
[0143] The effect of increased viscosity was consistent between
geometries. A fluid viscosity in the range of 2.5-2.7 cP decreased
gas exchange by approximately 20-30% at 20,000 RPM when compared to
water. Without limitation to any mechanism, it is believed that the
reduction was a result of increased drag forces on fluid particles
as they traversed the HFM bundle. Velocity of a fluid flowing
through a porous medium can be theorized according to Darcy's Law
which states:
V o = - k .mu. ( .gradient. P - .rho. g ) ( 4 ) ##EQU00003##
where {right arrow over (V)}o is the velocity field vector, k is
permeability, .mu. is viscosity, .gradient.P is the pressure
gradient, .rho. is fluid density, and {right arrow over (g)} is the
force of gravity. The fluid velocity through the medium is
indirectly proportional to the viscosity and fluids with higher
viscosities exhibit lower relative velocities to fibers at a given
RPM rate, thus inhibiting boundary layer reduction. This effect was
observed in comparing results obtained for the two solutions, per
the reduction in efficiency.
[0144] CO.sub.2 exchange in blood was approximately 20-40% greater
than in water when compared, for example, to the intravascular gas
exchange device described in Federspiel W J, Hewitt T, Hout M S, et
al: Recent progress in engineering the Pittsburgh intravenous
membrane oxygenator. Asaio J 42: M435-442, 1996, referred to as the
HATTLER.TM. Catheter. Oxygen exchange was also significantly
greater.
[0145] The blood viscosity resembling solution provided a reference
for estimating impeller prototype efficiency enhancement in blood
by incorporating relevant viscous effects into the system. The
expected increase in efficiency as a result of blood CO.sub.2
capacity was similar to the magnitude of efficiency reduction
caused by viscosity. Therefore, it was hypothesized that CO.sub.2
removal efficiencies in vivo would achieve rates similar to those
observed in standard DI water.
[0146] A common empirical mass transfer correlation for analyzing
oxygenator gas exchange was determined from an analogous heat
transfer correlation for perpendicular tube bank flow as follows:
Sh=aRe.sup.bSc.sup.1/3, wherein Sh is the Sherwood number, Re is
the Reynolds number, and Sc is the Schmidt number. The coefficients
a, b are dependant on fiber bundle geometry and are determined
experimentally, however, coefficient b can be loosely approximated
as b=0.5. If the velocity term in the Reynolds number is assumed
proportional to rotation rate, VCO.sub.2 can be compared to
RPM.sup.1/2 to assess mass transfer. FIGS. 9F(1) through 9F(9) show
the correlation for the nine impeller geometries tested on the
bench. The linear relationship between VCO.sub.2 and RPM.sup.1/2
indicated that the CO.sub.2 driving gradient for gas exchange at
higher RPM was not reduced, and therefore was not likely an
explanation for the observed leveling trend of gas exchange. The
trend was not investigated further. However, without limitation to
any mechanism, a second hypothesis to explain the plateau was that
specific fiber layers or fiber regions within the bundle were not
fully utilized in exchanging gas as result of an inefficient fluid
mixing mechanism.
[0147] Experiments were also performed to test the effectiveness of
the heparinized saline purge by first testing devices of the
present invention without saline purge. The tests were conducted in
a bench blood circuit to identify whether blood infiltration and
subsequent accumulation was responsible for the torsional failures
witnessed in the initial in vivo studies described above. Mock
devices including only the rotational components were first tested,
and results warranted further testing with full prototypes. Two
devices were tested in the final fatigue study. The first was a
device absent saline purge that served as the test control; while
the second was a device with saline purge system.
[0148] Both devices were placed in parallel 7/8'' test sections
within the blood circuit. Blood was re-circulated through each test
section at a flowrate of 1.5 LPM and maintained at 37.degree. C.
Two liters of blood (Heparinized bovine blood, Hemostat
Laboratories) were used in the circuit. The blood was characterized
at the beginning for ACT and fibrinogen content. A schematic of the
test system is shown in FIG. 9B.
[0149] Saline was pumped into each of medial manifold 400 and
distal manifold 500 at a flowrate of 12 ml/hr using a peristaltic
pump. The drive consoles were programmed to record elapsed time and
11/4 amp fuses were installed into each. The fuses automated the
system to shutdown upon failure and permitted the fatigue testing
to be conducted overnight under no supervision. The prototypes were
operated at a constant speed of 10,000 RPM for the test duration.
Motor torque requirements and elapsed time were recorded. All
equipment used for fatigue testing is described in Table 4.
TABLE-US-00003 TABLE 4 Component Model Manufacturer Blood 2 L
heparinized bovine Hemostat Laboratories blood Ultrasonic flow T110
Transonic Systems Inc. probe Reservoir bag AFFINITY Reservoir
Medtronic Inc. Bag #321 Blood pump BIO-MEDICUS .RTM. Medtronic
Bio-Medicus Inc. biopump BPX-80 Pump console 540 BIO-CONSOLE .RTM.
Medtronic Bio-Medicus Inc. Pediatric heat D1078 Medtronic
Biotronics exchanger Thermometer DIGI-SENSE .RTM. type T
Cole-Parmer Instrument Co. thermocouple Temperature pump T/Pump
TP-406 Gaymar Industries Peristaltic pump MASTERFLEX .RTM. Barnant
Company C/L 77120-62
[0150] Results of preliminary component and full prototype blood
fatigue tests showed the saline purge to be an effective means for
preventing torsional impeller tubing failures. In the full
prototype experiment, device 10 with saline purge was operated for
24 hours and did not show indication of potential failure. The test
was discontinued because the operating time far exceeded that
necessary to complete an acute bovine study. The device without
saline purge, however, failed two hours after the test start. A
plot of motor current requirement versus elapsed time is shown in
FIG. 10. It is observed that the device with saline purge operated
at a constant torque until test completion (motor torque is
proportional to motor current). An increase in torque to failure
was observed for the device without saline purge as a result of
build-up of blood constituents within the annular space between
materials/sleeves 408 and 508 and connector sections 122 and 124,
respectively.
[0151] Blood accumulation was confirmed in the control device
(without saline purge) by dissecting the prototypes and examining
rotational components. The observed build-up of blood constituents
replicated observations of failed devices earlier studied in vivo.
In contrast, observation of the device with saline purge revealed
seals and material/sleeves 408 and 508 free from blood contact. The
saline purge was implemented for subsequent in vivo
evaluations.
[0152] Testing was transitioned from the bench to an in vivo
environment to evaluate device performance in a physiological
setting. All in vivo studies performed were acute studies with the
objective of obtaining CO.sub.2 exchange measurements in a bovine
model. The three impellers determined from bench gas exchange
characterizations to provide the best CO.sub.2 efficiency were the
4-flat, RAX, and sawtooth impeller unit designs. Those three
impeller unit designs were used in all further in vivo testing. Gas
exchange was measured in two different vena cava positions. The
first location was a position spanning the right atrium where
longitudinal blood flow in the IVC and a cross-flow into the right
atrium were experienced. The second position evaluated the catheter
solely in the IVC where only longitudinal flow past the device
occurred.
[0153] All experiments were performed at a University of Pittsburgh
animal facility possessing the proper staff and surgical equipment
to provide full animal care. The surgical procedure and testing
protocol were authorized by the University of Pittsburgh IACUC.
Calves weighing approximately 100-125 kg were used in the
experiments. Each animal was fitted with a Swan-Ganz catheter and
femoral pressure line following anaesthetization. Heparin infusion
was administered into the femoral vein to maintain activated
clotting times (ACTs) above 500. Guidewires for device insertion
were laid through the jugular vein and femoral vein using a
2-cupped magnet system. The jugular vein was exposed and an
incision was made. Upon insertion the device was extended through
the jugular vein into the vena cava to situate in either of the two
locations described above.
[0154] All equipment used in the gas circuit for bench exchange
characterization was used. CO.sub.2 removal efficiency was
calculated in accordance to equations (2) and (3). The purge
flowrate was 12 ml/hr of 1:50 heparin to saline solution. The
femoral blood pCO.sub.2 was maintained in the range of 45-55 mmHg
by adjusting ventilator settings.
[0155] Testing protocol began when calf femoral pCO.sub.2
stabilized within the requested range. The prototype was tested
over the full RPM range of 200-20,000 RPM in a randomized fashion.
CO.sub.2 exhaust concentration, sweep gas flowrate, sweep gas
pressure drop, and torque were acquired at rates of 200, 5,000,
10,000, 15,000 and 20,000 RPM. Each data point was obtained at
least twice to ensure repeatability. The protocol was initiated in
both vena cava locations before a different impeller prototype was
inserted. Calve arterial and femoral pressures were monitored and
plasma free hemoglobin samples were obtained through the course of
testing. Following the experiment, a necropsy was performed to view
the vasculature. Prototypes were later dissected on the bench to
observe any thrombus formation, failure sites, and seal
wear/appearance. A concise outline of the protocol is set forth in
Appendix C.
[0156] Impeller catheter gas exchange data was obtained in four
different calves. The results of impeller geometry influence
matched the trend identified in bench-testing. FIG. 11 shows the
maximum average CO.sub.2 removal efficiencies for 4-flat, RAX, and
sawtooth impellers. The 4-flat geometry produced the most favorable
efficiency level and was found to be significantly higher than the
sawtooth geometry. The corresponding in vivo removal rates were: 36
ml CO.sub.2/min for 4-flat, 34 ml CO.sub.2/min for RAX, and 33 ml
CO.sub.2/min for sawtooth.
[0157] A typical plot of efficiency versus RPM is displayed in FIG.
12. Gas exchange increased as a function of rotational rate. The
slope of the curve, however, began to trend towards zero near
20,000 RPM indicating a maximum efficiency plateau. An inspection
of all data did not show device location spanning the right atrium
to be more beneficial in comparison to the IVC. Any conclusion from
the above observations may be limited by the necessity to configure
the prototype in vena cava positions avoiding severe tortuousness
which may have exposed the catheter to the right atrium mixing in
both intended locations.
[0158] Maximum removal efficiencies in blood were very similar to
bench gas exchange results obtained using normal DI water. Without
limitation to any mechanism, it was hypothesized that an increase
in CO.sub.2 removal over DI water was not observed in blood because
the ability of blood to carry a higher CO.sub.2 content was offset
by the increased bundle drag forces and consequent reduction in
relative fluid velocity resulting from a higher fluid viscosity.
Bench-testing of devices 10 in standard DI water should thus
provide an accurate prediction of maximum CO.sub.2 exchange. Oxygen
exchange in blood are observed to be 2-3 time greater than that
observed in deionized water.
[0159] Once again, plasma free hemoglobin (PFHB) samples were
measured during testing. At least three devices 10 were inserted
per calf. Data from two studies in which a specific impeller was
evaluated over three hours is shown in FIG. 13. FIG. 17 illustrates
a plot of PFHB versus elapsed time for all experiments. The
prototypes appeared to generate consistent levels between 20-40 mg
PFHB/dL that trend towards a constant or decreasing rate. Although
devices 10 and impeller units 100 were not optimized, undesirable
levels of hemolysis were not observed during testing.
[0160] The non-optimized impeller catheters of the present
invention enhanced in vivo CO.sub.2 removal efficiency by 70% over,
for example, the HATTLER Catheter and confirmed the effectiveness
of rotation. The device was capable of operating in tortuous
vasculature while protecting the vena cava wall. The tradeoff in
HFM surface area to reduce size however, reduced overall CO.sub.2
removal rates. Removal rate was ultimately higher in the HATTLER
Catheter because of its 21/2 fold greater surface area. Table 4
sets forth an overview of catheter performance for the HATTLER
catheter and impeller device 10 of the present invention.
TABLE-US-00004 TABLE 4 Surface In vitro In vivo CO.sub.2 Removal
Area Efficiency Efficiency Rate Catheter (m.sup.2) (ml
CO.sub.2/min/m.sup.2) (ml CO.sub.2/min/m.sup.2) (ml CO.sub.2/min)
HATTLER 0.17 230 300 51 Catheter Impeller 0.07 529 513 36
[0161] In the devices, systems and methods of the present
invention, CO.sub.2 removal increases in proportion to CO.sub.2
tension in blood. Respiratory catheter data from acute animal
implants (calves and sheep) in which ventilatory challenges were
done to vary the venous PCO.sub.2 were regressed. The resulting
correlation (r.sup.2=0.993) from 25 mmHg to 70 mmHg: {dot over
(V)}.sub.CO2{dot over
(V)}.sub.CO2.sup.50mmHg=1.32(10.sup.-4)P.sub.CO2.sup.2+8.05(10.sup.-3)P.s-
ub.CO2+0.26 predicts an increase in the CO.sub.2 removal rate of
50% from normocapnia (45-50 mmHg) to permissive hypercapnia
(.about.70 mmHg). Thus, removal rate of CO.sub.2 can be increased
approximately 50% by operation under conditions of permissive
hypercapnia.
[0162] Immobilizing carbonic anhydrase enzyme onto the hollow fiber
membranes of the devices of the present invention as described in
U.S. patent application Ser. No. 11/811,265 and PCT International
Patent Application No. PCT/US2007/013563 (PCT Publication No. WO
2007/146162), the disclosure of which are incorporated herein by
reference, further increases CO.sub.2 efficiency. Preliminary data
using current (baseline) immobilization of CA on hollow fibers
(activity .about.0.002 U/cm.sup.2) indicated a 25% increase in the
CO.sub.2 removal rate in blood. Improved immobilization methods
have increased CA activity by 250%.
[0163] Agitation mechanisms such as rotating impellers increase gas
exchange by reducing the size of diffusional boundary layers that
dictate gas exchange Immobilized CA increases CO.sub.2 exchange by
creating facilitated diffusion of CO.sub.2 as bicarbonate ion,
resulting in a larger effective diffusion coefficient for CO.sub.2.
As such, the effects would be additive. However, such effects may
be more than additive because as the boundary layer is reduced, so
is the amount of native CA (i.e. in red cells) within the boundary
layer, with which immobilized CA competes. Preliminary studies
indicate that without native CA (i.e. using bicarbonate buffer as
the test fluid), the same immobilized CA fibers increased the
CO.sub.2 removal rate by 75%.
[0164] As set forth above, carbonic anhydrase catalyzes blood
bicarbonate into carbon dioxide and increases the fundamental
CO.sub.2 exchange driving gradient. In that regard, carbon dioxide
is present in blood in three primary forms: CO.sub.2 (dissolved),
bicarbonate (HCO.sub.3.sup.-), or carbamate. As known in the
chemical arts, CO.sub.2 is interconvertible among these forms and
the various forms can be in equilibrium with each other as
described by a CO.sub.2 dissociation curve. Most of the CO.sub.2 in
blood, however, exists in the form of HCO.sub.3.sup.- in plasma and
in red blood cells. Colton C K. 1976. Fundamentals of gas transport
in blood. In: Zapol W M and Qvist J, editor. Artificial lungs for
acute respiratory failure. Washington D.C.: Hemisphere Publishing
Corporation. p 3-43. In that regard, approximately 94% of plasma
CO.sub.2 and 82% of red blood cell CO.sub.2 is in the form of
HCO.sub.3.sup.-. The two species are interconvertible via the
reaction:
CO.sub.2+H.sub.2O.revreaction.H.sup.++HCO.sub.3.sup.-
[0165] The CO.sub.2 generates via metabolic pathways in tissue and
diffuses into red blood cells (RBCs), where it is hydrated into
HCO.sub.3.sup.- and hydrogen ions (H.sup.+) by intracellular
carbonic anhydrase (CA). The hydrogen ions formed are bound to
hemoglobin while HCO.sub.3.sup.- is diffused into plasma. However,
very little CO.sub.2 is hydrated in plasma because of a lack of CA
in plasma. In lungs, the reaction is reversed. HCO.sub.3.sup.- is
converted into CO.sub.2 via CA in red blood cells, and then
exhaled. Some CA exists in lung tissue.
[0166] With reference to FIG. 14A, the immobilized enzyme
facilitates diffusion of the specific CO.sub.2 through the membrane
by generating diffusion of HCO.sub.3 to the surface of the
membrane. At the membrane surface the enzyme converts HCO.sub.3 to
CO.sub.2.
[0167] CA maintains substantial activity upon immobilization (for
example, via covalent bonding to a polymeric material). As set
forth in FIG. 14B, radio frequency glow discharge (RFGD) and
cyanogen bromide (CNBr) activation chemistry can, for example, be
used to immobilize CA on (or in the vicinity of)at least a first or
outer surface of the fibers of the HFM via covalent bonding as
described in U.S. patent application Ser. No. 11/811,265 and PCT
International Patent Application No. PCT/US2007/013563 (PCT
Publication No. WO 2007/146162).
[0168] The activity of immobilized CA on fiber surfaces can, for
example, be increased by incorporating a base layer of
poly(L-lysine) (PLL) or dendritic polyamidoamine (PAMAM) on hollow
fibers prior to immobilizing CA. Successive "layers" of CA can be
added onto the first immobilized layer using glutaraldehyde-linked
"layering".
EXPERIMENTAL SECTION
[0169] A. Tortuous Loop Component Tests.
[0170] The following provides a brief discussion of tests and
results of component function in a tortuous environment. The tests
were also beneficial for determining a length of impeller units
suitable to retain flexibility. Three impeller shaft tubings were
tested in the tortuous loop setup to evaluate torque requirements,
fatigue, and impeller length. The sample group is set forth in
Table A-1 below.
[0171] The testing device was placed in a loop and water was
circulated to provide a fluid environment. The test device was a
"skeleton" impeller prototype in which tubing samples could be
easily substituted. Testing was performed without a HFM bundle
except in one instance to evaluate the effect of the bundle
decreasing bending radii. Torque was calculated by measuring the
current into the drive system and converting by means of the motor
torque constant.
TABLE-US-00005 TABLE 0-1 Selected tubings for tortuous testing.
Sealant Manufacturer Part # Material Mesh Topcoat (A) New NEC00029
Clear PEBAX 40 AWG NA England Wire 6333 304 stainless Technologies
steel (B) New N98-TOL- Blue 44 AWG NA England Wire 664 PEBAX 304
stainless Technologies braid (C) S.S. White 067N133C Stainless NA
In-house Bi- directional Steel polyurethane hollow shaft coil
[0172] The final results of all testing indicated that candidate B
of Table A-1 was a preferred candidate. In that regard, the tubing
could withstand rotation at 20,000 RPM in the tortuous loop and
also provided the largest inner diameter to reduce resistance
against sweep gas flow.
[0173] A.1. Effect of Fluid Viscosity on Torque.
[0174] Testing was performed in a straight test section to evaluate
whether fluid viscosity would have a noticeable effect on torque
transmission. This was done to ensure that testing in water was
suitable and that an increase in viscosity to that of blood would
not add excessive torque to the motor or tubing that should be
considered. The test was performed using tubing sample A in both
water and air. FIG. 15A shows that fluid viscosity had a negligible
effect in our applicable ranges.
[0175] A.2. Impeller Unit Length.
[0176] All impeller units were fabricated from rigid materials and
thus constrained tubing flexibility. Fixing the units to the
impeller tubing required both a unit length and spacing to be
determined. The spacing between units was arbitrarily set at 0.5''
(inches). Sample C was tested in the loop with two impeller
configurations consisting of either 1'' impeller units or 0.5''
impeller units. The test was conducted at 20,000 RPM and 1'' units
were concluded to provide inadequate flexibility. The longer units
constrained the tubing to greater bending stresses in between
units. A portion of the testing configuration is represented in
FIG. 15B.
[0177] A.3. Fatigue and Torque Requirement.
[0178] Test devices were placed in the tortuous configuration
represented in FIG. 15B and operated at 20,000 RPM for fatigue
life. Elapsed time and motor torque were monitored. A HFM bundle
was also added to the N98-TOL-664 tubing test device to observe the
effects on torque as bending radii decreased. Torque versus RPM
results for tubing samples A and B are shown in FIG. 15C. Both
samples ran for over 6 hours without failure and were deemed
useable in the prototype. The samples performed similarly in
regards to torque requirement in the loop and the HFM bundle was
observed to increase motor torque by an insignificant 0.5-1 mN-m.
The recommended continuous torque application for the brushless
servomotor was 13.5 mN-m and thus torque required in the
"worst-case" scenario tortuous loop was still 3 times lower
indicating motor-prototype compatibility.
[0179] B. Impeller Catheter Design Calculations.
[0180] The blood shear stress calculations indicated a plausible
cause for build-up observed in the external VESPEL sleeves during
initial in vivo testing. A flushing fluid (saline) purge
calculation was used to meet a design requirement of 200 mmHg gauge
pressure at the seal interface.
[0181] B.1. Blood Shear Stress Calculation.
[0182] Shear stress calculation for fully-developed, steady flow
between concentric cylinders with outer cylinder fixed
(.omega..sub.2=0) (as graphically represented in FIG. 16A):
Laminar Flow (Re<1700) between Rotating Cylinders when
.omega..sub.2=0:
Re = .omega. 1 R 1 .delta. v where .delta. = R 2 - R 1
##EQU00004##
Shear stress .tau..sub.1, on the inner cylinder is:
.tau. r .theta. = - [ .mu. r r ( V .theta. r ) ] r = R 1 where
##EQU00005## V .theta. = R 1 2 .omega. 1 R 2 2 - R 1 2 ( R 2 2 r -
r ) .thrfore. .tau. r .theta. = 2 .mu. R 2 2 .omega. 1 R 2 2 - R 1
1 ##EQU00005.2##
Calculation for Impeller Rotation at 20,000 RPM:
[0183] .omega. 1 = 20 , 000 2 .pi. rad 60 s = 2094 rad s
##EQU00006## .mu. blood = 0.033 P ##EQU00006.2## v blood 0.031 cm 2
s ##EQU00006.3## R 1 = 0.142 cm ##EQU00006.4## R 2 = 0.144 cm
##EQU00006.5##
At 20,000 RPM,
[0184] Re=19.2<1700.thrfore.flow is laminar
.tau. r .theta. = 5010 dyne cm 2 ##EQU00007##
[0185] B.2. Saline Purge Flow Rate Calculation.
[0186] Flow through an annulus (see FIG. 16B) is calculated
according to:
Q = .pi. 8 .mu. ( - P x ) [ a 4 - b 4 - ( a 2 - b 2 ) 2 ln ( a / b
) ] ##EQU00008##
Assume linear pressure gradient:
- P x = P 2 - P 1 L ##EQU00009##
P.sub.1=26,664 Pa
P.sub.2=1,333 Pa
.mu.=0.001 Pa-s
L=0.0127 m
[0187] a=0.001435 m b=0.001422 m
Q = 3.28 .times. 10 - 9 m 3 s = 12 ml hr ##EQU00010##
[0188] C. Impeller Prototype In Vivo Protocol.
[0189] C1. Impeller Catheter Acute In vivo Test Protocol.
[0190] In several studies the impeller catheter of the present
invention was evaluated in a calf model for gas exchange and
hemodynamics at various rotation rates. The results were compared
with gas exchange results to in vitro water data.
[0191] Devices: [0192] 1. Impeller: Vadnais coil tubing with PEBAX
heat-shrink/4-flat impeller units/saline purge [0193] 2. Impeller:
Vadnais coil tubing with PEBAX heat-shrink/RAX impeller
units/saline purge [0194] 3. Impeller: Vadnais coil tubing with
PEBAX heat-shrink/sawtooth impeller [0195] units/saline purge
Protocol:
Pre-Test Preparation (Previous Day):
[0196] 1) Calibrate all equipment (Gas Flowmeter, Pressure
Transducer) 2) Setup equipment and make albumin soaking solution
and heparinized saline (0.02 v/v)
Day of Test:
[0197] 1) Calibrate CO.sub.2 analyzer. [0198] 2) Soak device in
albumin/heparin solution. (50 cc bovine albumin, 10 cc heparin, and
saline) [0199] 3) Fill moisture trap container with ice. [0200] 4)
Take baseline blood samples. [0201] 5) Insert all necessary
pressure lines and take baseline measurements (CO, pressures):
[0202] Left femoral vein [0203] Carotid artery [0204]
Swan-Ganz--CVP and CO [0205] 6) Measure diameter of right external
jugular vein once exposed. [0206] 7) Measure initial hematocrit and
CO, then begin volume loading with lactated ringers at 15 ml/kg.
After volume loading, measure hematocrit and CO again. [0207] 8)
The magnetic guide system will be inserted via the right femoral
vein and right external jugular vein, using Fluoroscopy. [0208] 9)
Pull wires back and attach distal tip of catheter to femoral guide
wire. [0209] 10) Allow 2 minutes for saline flush of device to
initiate. [0210] 11) Administer a heparin bolus through the left
femoral line. [0211] 12) Insert the catheter into the right jugular
with animal positioned on its side. [0212] 13) Pull catheter
through the jugular vein and position spanning the RA. Verify with
Fluoroscopy. [0213] 14) Set the gas flow to maximum allowed and
begin rotating at 5000 RPM. A post-insertion CO and several gas
exchange values should be acquired during the equilibrium phase.
[0214] 15) When exchange stabilizes, rotation rate protocol will
begin. [0215] 16) Test the following rotation rates in a random
order: 5000, 10000, 15000, 20000 RPM, repeating each point twice.
Record % CO.sub.2, rotation rate, current, gas flow rate, and gas
pressure drop. [0216] 17) Reduce the rotation rate and record 200
RPM exchange rate. Obtain a CO at 200 RPM. [0217] 18) When finished
with all rotation rates, move the device into the IVC. [0218] 19)
Allow .about.15 minute stabilization period with rotation at 5000
RPM. Take another CO. [0219] 20) Repeat steps 18-19 in this new
location. [0220] 21) Remove device and place additional impeller
devices. Repeat steps 13-21. [0221] 22) Keep the device in the
second location for necropsy. [0222] 23) Pictures and anatomic
measurements will be taken at necropsy. [0223] 24) Perform gross
examination of organs (lung especially) and send out for histology
samples if needed.
[0224] C.2. In Vivo Hemolysis Studies.
[0225] FIG. 17 illustrates measured PFHB during all in vivo testing
of the impeller prototype. Each line represents a separate calve
experiment. PFHB ranged between 20-40 mg/dL for each animal during
prototype testing.
[0226] The foregoing description and accompanying drawings set
forth the preferred embodiments of the invention at the present
time. Various modifications, additions and alternative designs
will, of course, become apparent to those skilled in the art in
light of the foregoing teachings without departing from the scope
of the invention. The scope of the invention is indicated by the
following claims rather than by the foregoing description. All
changes and variations that fall within the meaning and range of
equivalency of the claims are to be embraced within their
scope.
* * * * *