Hemant S. Thatte, PhD, Kunda S. Biswas, MD, Samer F. Najjar, MD,
Vladimir Birjiniuk, MD, Michael D. Crittenden, MD, Thomas Michel, MD, PhD, and
Shukri F. Khuri, MD
Department of Surgery and Cardiology Division, VA Boston Healthcare System, Brigham and Women’s Hospital and Harvard
Medical School, Boston, Massachusetts
Background. Injury to endothelium can compromise
the patency of bypass grafts harvested during coronary
artery bypass graft (CABG) surgery. Maintaining structural and functional viability of endothelium in grafts
may lead to improved long-term patency. The information gained from the application of multi-photon microscopy in transmission and epifluorescence mode was used
to assess the structural and functional integrity of human
saphenous vein segments stored in multiple preservation
solutions, and to design a superior storage solution.
Methods. Multi-photon microscopy was used to image
deep within saphenous vein tissue harvested from patients undergoing CABG for analysis of endothelial
structure and function. Endothelial cell structural viability, calcium mobilization, and nitric oxide generation
were determined using specific fluorescence markers.
Results. Within 60 minutes of harvest and storage in
standard preservation solutions, calcium mobilization
and nitric oxide generation were markedly diminished
with more than 90% of endothelial cells no longer viable
in the vein. In contrast, veins could be stored for 24 hours
without substantial loss in cell viability in a newly
formulated heparinized physiologic buffered salt solution containing glutathione, ascorbic acid, and L-arginine
(GALA).
Conclusions. Standard solutions in clinical use today
led to a profound decline in saphenous vein endothelial
cell viability, whereas the newly designed physiologic
salt solution (GALA) maintained endothelial function
and structural viability for up to 24 hours. The improvements seen from using GALA as a vessel storage medium
may lead to greater long-term vein graft patency following CABG surgery.
T
Until recently the intracellular events transpiring
within endothelial cells in living vessels could not be
directly evaluated because of the inability of conventional fluorescence microscopy to image deep into the
tissues. With the recent advent of fluorescent multiphoton microscopy [9], it is now possible to overcome
these limitations. Excitation of a fluorescent molecule by
the simultaneous absorption of two (near) infrared photons of longer wavelength facilitates deeper penetration
into tissues [10, 11]. Optical sectioning and construction
of three-dimensional images has become possible without dissecting the tissue. Multi-photon excitation localized to a limited volume at the focal plane minimizes
photo-damage and out-of-plane fluorescence and results
in high-contrast, well-resolved images from deep within
tissues [10, 11].
In this study, we utilized multi-photon imaging techniques in real time to characterize the endothelial cell
viability in saphenous veins harvested from patients
undergoing CABG surgery. We evaluated the comparative efficacy of a variety of storage solutions in preserving
the structure and function of endothelium, including a
newly formulated solution (GALA), which we hypothe-
he saphenous vein represents the most commonly
used venous conduit for coronary artery bypass
graft (CABG) surgery and other peripheral artery bypass
procedures. However, the long-term patency of the saphenous vein graft is limited, with an occlusion rate of
15% to 26% in the first year [1]. Pathologic changes in the
saphenous vein related to graft occlusion are well documented [2, 3]. Endothelial cell damage following vein
harvest results in platelet activation and may be an
important element in early graft occlusion. Therefore, the
integrity of the endothelium is essential for successful
short and long-term patency [2, 3]. Short-term storage of
free vascular grafts is routine in CABG operations, where
one to two hours may elapse between the vein harvest
and reperfusion [2– 4]. This interval may affect both the
structure and function of the graft, depending upon the
composition and temperature of the storage solution and
the duration of tissue ischemia before vein reimplantation [2– 8].
Accepted for publication Oct 24, 2002.
Address reprint requests to Dr Khuri, Department of Surgery (112),
VA Boston Healthcare System, 1400 V. F. W. Parkway, West Roxbury,
MA 02132; e-mail: [email protected].
© 2003 by The Society of Thoracic Surgeons
Published by Elsevier Science Inc
(Ann Thorac Surg 2003;75:1145–52)
© 2003 by The Society of Thoracic Surgeons
0003-4975/03/$30.00
PII S0003-4975(02)04705-7
CARDIOVASCULAR
Multi-Photon Microscopic Evaluation of Saphenous
Vein Endothelium and Its Preservation With a New
Solution, GALA
1146
THATTE ET AL
MICROSCOPIC EVALUATION OF SAPHENOUS VEIN ENDOTHELIUM
CARDIOVASCULAR
sized would provide optimum endothelial cell protection
before CABG surgery. This newly designed solution was
found to maintain the vascular structure and key endothelial cell regulatory pathways of calcium mobilization
and nitric oxide generation for extended periods of time.
Material and Methods
Saphenous Veins
Saphenous vein segments (inner diameter 0.05 to 0.2 mm,
outer diameter 0.2 to 1.0 mm) were obtained before
distension or any other manipulation from male patients
of age 67.13 ⫾ 9.78 (mean ⫾ SD) years undergoing CABG
surgery at the VA Boston Healthcare System, according
to an approved Human Studies Subcommittee protocol.
The vein segments were placed in the preservation
solutions described below at 21°C and processed within
30 minutes. Excess adipose and adventitia were gently
excised. Five different preservation solutions were used
for the ex vivo preservation of explanted saphenous
veins. The segments from the saphenous veins were
stored from 1 to 24 hours at 21°C in one of the following:
HLS (heparin, 40 U/ml; 0.25% lidocaine; 0.9% sodium
chloride), a commonly used storage solution; AHB, autologous heparinized (40 U/ml) blood, also routinely used
in the operating rooms; TCM, tissue culture medium (1:1
MEM [minimum essential medium], RPMI 1640 medium), a solution that is a standard for in vitro tissue
culture and used in the laboratories to provide optimum
conditions for cell growth in tissue propagation; HBSS,
Hank’s balanced salt solution, a representative physiologic salt solution (HBSS consists of, all in gm/L, 0.14
calcium chloride, 0.4 potassium chloride, 0.06 potassium
phosphate [monobasic], 0.1 magnesium chloride [hexahydrate], 0.1 magnesium sulfate [heptahydrate], 0.8 sodium chloride, 0.35 sodium bicarbonate, 0.05 sodium
phosphate [dibasic; hepatahydrate] and 1 D-Glucose);
GALA, an easily prepared, newly formulated storage
solution, which we hypothesized will preserve the endothelial structural and functional viability of the surgical
conduits during storage for an extended period of time.
Hank’s balanced salt solution was modified by adding
0.09 gm/L of reduced glutathione (1000 ␮mol/L final
concentration) and 0.31 gm/L of L-ascorbic acid (500
␮mol/L final concentration) as antioxidants and reducing
agents, with 0.15 gm/L of L-arginine (500 ␮mol/L final
concentration) as a substrate for endothelial nitric oxide
synthase (eNOS). Heparin (50 U/ml) was added as an
anticoagulant and the pH was adjusted to 7.4 using 8.4%
sodium bicarbonate solution. This new solution is referred
to as the GALA solution (glutathione, ascorbic acid, Larginine). The saphenous vein graft segments were collected from 9 different patients. Each segment was divided
into five sections and each section was separately stored in
one of the five preservative solutions described above.
Cell Viability Assay
Structural and functional viability of the endothelial cells
was assessed with a fluorescence-based Live-Dead assay
Ann Thorac Surg
2003;75:1145–52
(calcein AM/ethidium homodimer; Molecular Probes
[Eugene, OR]) [15, 16]. Vein segments were stored in one
of the storage solutions as indicated and then incubated
with calcein AM and ethidium homodimer dyes (10
␮mol/L, final concentration) in 1.5-ml HBSS, pH 7.4, for
30 minutes at 21°C. After incubation, segments were
washed three times with the preservation solution,
mounted on the multi-photon microscope stage in an
imaging chamber, and imaged as described below. A
viability score was assigned to each experiment as follows: a score of 4⫹ ⫽ 76% to 100% viability, 3⫹ ⫽ 51% to
75% viability, 2⫹ ⫽ 26% to 50% viability, and 1⫹ ⫽ ⱕ 25%
viability based on green and red fluorescence observed.
Determination of Intracellular Calcium Mobilization
and Nitric Oxide Generation
Calcium mobilization in the endothelial cells of the
saphenous veins was measured using calcium-sensitive
calcium orange dye (Molecular Probes, Inc., Eugene, OR).
Nitric oxide production in the segments was determined
using the nitric oxide specific indicator dye DAF-2DA (5,
12, 13 [Calbiochem, La Jolla, CA]). Vein segments were
incubated for 60 minutes at 37°C with calcium orange
and DAF in HBSS (10 and 15 ␮mol/L final concentration,
respectively). Under these incubation conditions the fluorescence dyes label endothelial cells and not the smooth
muscle cells. After washing with HBSS to remove excess
dye, the segments were imaged as described below.
Calcium mobilization and eNOS activity were stimulated
by adding bradykinin in HBSS (final concentration of 10
␮mol/L), to the imaging chamber. Changes in calcium
orange and DAF fluorescence were recorded in real time
for 10 minutes at 21°C after bradykinin stimulation. To
measure the specificity of eNOS activity, vein segments
were preincubated for 30 minutes at 37°C with 100
␮mol/L of eNOS inhibitor Nw-nitro-L-arginine (L-NNA)
in 1.5 ml of HBSS, before incubation with DAF [5, 12, 13].
Resting calcium levels and basal activity of eNOS were
measured in the absence of bradykinin stimulation.
Multi-Photon Imaging
Imaging and semiquantitative fluorescence measurements were done with a BioRad MRC 1024ES multiphoton imaging system (BioRad, Hercules, CA) coupled with
a mode-locked titanium:sapphire laser (Spectra-Physics,
Mountain View, CA) operating at 82 MHz repetition
frequency, 80-fs pulse duration with a wavelength tuned
to 820 nm, in transmission and epifluorescence mode. A
Zeiss Axiovert S100 inverted microscope (Carl Zeiss, Inc.,
Thornwood, NY) equipped with a high quality water
immersion 40x/1.2 NA, C-apochroma objective was used
to image the segments and quantitate fluorescence. The
512 ⫻ 512 pixel images were collected in direct detection
configuration at a pixel resolution of 0.484 ␮m with a
Kalman 5 collection filter. The lumen and endothelial cell
layer were identified by XYZ scanning and imaged,
generally at depths of 50 to 200 ␮m in longitudinal vein
segments depending on the size of the vein [5, 14], and at
a depth of 100 ␮m from the site of excision in 10-mm
cross sections of the veins.
THATTE ET AL
MICROSCOPIC EVALUATION OF SAPHENOUS VEIN ENDOTHELIUM
1147
Fig 1. This figure illustrates changes in
structural integrity of vein segments stored
in different preservation solutions. Vein
segments were stored in heparinized lidocaine saline (HLS), autologous heparinized
blood (AHB), tissue culture medium
(TCM) for 60 minutes; and in Hank’s balanced salt solution (HBSS) and the newly
designed glutathione, ascorbic acid, and
L-arginine (GALA) preservation solution
for a total of 24 hours. Green fluorescence indicates cell viability; red fluorescence indicates compromised cells. Extensive cell membrane damage and
compromised endothelial cell integrity in
the vessels was observed after short-term
storage in HLS, AHB, and TCM. Cell viability was well preserved in veins during
short-term storage in HBSS but resulted in
cell death upon extended storage. In contrast, endothelial cells remained viable in
vessels preserved in GALA solution
throughout the extended time of storage.
Representative image, at ⫻400 magnification, n ⫽ 9. (EC ⫽ endothelial cells.)
Quantitative Analysis of Calcium and Nitric Oxide
[5, 12, 14 –16]. The data represents an average of blinded
experiments performed on different days and expressed
as mean ⫾ standard error of the mean (SEM; n ⫽ 9).
Calcium mobilization and nitric oxide generation [5, 12,
13] were measured by recording changes in calcium
orange and DAF fluorescence before and after bradykinin treatment in real time. Typically three to six specific
regions were drawn along the endothelial region of the
lumen identified by XYZ scanning in a field of view at
400⫻ magnification using image processing software
(MetaMorph Imaging Series [Universal Imaging Corp.,
West Chester, PA]). The change in fluorescence intensity
integrated over all pixels within each boundary was
monitored over time and quantitated separately using
MetaMorph in red (calcium) and green (nitric oxide)
fluorescence channels, respectively. Because of the variable size and shape of the boundaries and vein sizes, and
in order to eliminate effects due to variation in fluorescence dye loading, the fluorescence intensity from each
image was normalized by values determined from a
reference image recorded before bradykinin treatment
Results
Effects of Preservation Solutions on Endothelial Cell
Viability in Explanted Saphenous Vein
Vein segments stored in heparinized lidocaine saline
(HLS), autologous heparinized blood (AHB) or tissue
culture medium (TCM) exhibited a red fluorescence
pattern in the lumenal region within 1 hour of storage
indicative of extensive cell membrane damage and compromised viability of endothelial cells (Fig 1, Table 1). In
contrast, the endothelial cells were structurally intact
(green fluorescence) after 1 hour of storage in Hank’s
balanced salt solution (HBSS; Fig 1), but lost cellular
integrity during extended storage (Table 1). Similarly,
smooth muscle cell architecture of the vessels also de-
Table 1. Viability Scores of Endothelial Cells of Veins Stored in Preservative Solutions
Storage Time (minutes)
Solution
60
90
120
150
180
240
300
1440
HLS
AHB
TCM
HBSS
GALA
⫹
⫹
⫹
‡
§
⫹
⫹
⫹
§
§
nt
nt
nt
‡
§
nt
nt
nt
‡
§
nt
nt
nt
‡
§
nt
nt
nt
⫹
§
nt
nt
nt
⫹
§
nt
nt
nt
nt
§
Scoring system: ⫹ ⫽ 0%–25% viable, ‡ ⫽ 51%–75% viable, § ⫽ 76%–100% viable. Each viability score is mean of 9 different patients. nt ⫽ not tested
secondary to lack of predicted viability.
Human saphenous vein segments stored in various preservation solutions: AHB ⫽ blood with added heparin (40 Units/mL);
GALA ⫽ physiological
salt solution with glutathione, ascorbic acid, L-arginine, and heparin;
HBSS ⫽ Hanks’ balanced salt solution;
HLS ⫽ heparin⫹lidocaine⫹saline;
TCM ⫽ RPMI/M199culture media in 1:1 ratio.
CARDIOVASCULAR
Ann Thorac Surg
2003;75:1145–52
1148
THATTE ET AL
MICROSCOPIC EVALUATION OF SAPHENOUS VEIN ENDOTHELIUM
Ann Thorac Surg
2003;75:1145–52
CARDIOVASCULAR
Fig 2. Transverse sections of the saphenous veins imaged using multiphoton microscopy in transmission mode. The segments were imaged using XYZ scanning ⬃100-␮m deep from the site of excision. The vessel lumen, endothelial cell layer, and smooth muscle layer are clearly visible. Endothelial and smooth muscle cell damage was observed in segments stored for 60 minutes in heparinized lidocaine saline (HLS) and
autologous heparinized blood (AHB). Structural integrity of the vessel was well preserved in vessels stored in glutathione, ascorbic acid, and
L-arginine (GALA). Representative image at ⫻400 magnification, n ⫽ 9.
generated completely during prolonged storage in HBSS
(Fig 1) and other storage solutions (not shown).
The protective effect of GALA solution on endothelial
structural integrity was compared to other storage solutions. Vein segments were stored in GALA for 5 hours at
21°C, then for 19 hours at 4°C for a total storage time of 24
hours. In contrast to other solutions, segments stored in
GALA solution exhibited a robust green fluorescence
indicative of structural viability of the endothelium (Fig
1). Dead cells (red fluorescence) were rarely observed
during 24 hours of preservation. Similarly, smooth muscle cell architecture and viability of the vessels were well
maintained during extended storage in GALA (Fig 1,
Table 1). Transmitted light images of cross sections of the
saphenous vein stored for 60 minutes in HLS, AHB and
GALA are illustrated in Figure 2. The sections were
imaged at 100 ␮m from the site of excision. The endothelial monolayer in juxtaposition to the lumen showed a
substantial amount of disruption in vein segments stored
in HLS and AHB. Similarly, thinning and disruption of
smooth muscle layers was also observed in these vein
segments. In contrast, well-preserved endothelium as
well as smooth muscle layers remained morphologically
intact in vein segments stored in GALA, Figure 2.
Effect of Preservation Solutions on Endothelial
Calcium Mobilization and Nitric Oxide Generation in
Stored Saphenous Veins
The functionality of the vein segment endothelium was
determined by the increase in internal calcium concentration and subsequent production of nitric oxide, a key
endothelium-derived mediator, as a marker for endothelial cell function [17–19]. A time-dependent two- to threefold increase in calcium and nitric oxide fluorescence was
observed upon bradykinin stimulation (Fig 3). As ex-
pected, the calcium signal (orange/red fluorescence) and
the nitric oxide signal (green fluorescence) were found to
colocalize (orange-yellow fluorescence) in the endothelium. The effect of the various preservation solutions on
saphenous vein endothelial function is illustrated in
Fig 3. Bradykinin mediated calcium mobilization and nitric oxide
generation in saphenous veins stored in glutathione, ascorbic acid,
and L-arginine. Segments (similar to those illustrated in Fig 2) were
labeled with calcium orange and diaminofluorescein, stimulated with
bradykinin and imaged using multiphoton microscopy. A robust two- to
threefold increase in calcium (red) and nitric oxide (green) fluorescence
signal in the endothelial region was observed 10 minutes after bradykinin stimulation, that found to colocalize (orange-yellow fluorescence). Representative image of nine independent experiments.
THATTE ET AL
MICROSCOPIC EVALUATION OF SAPHENOUS VEIN ENDOTHELIUM
Table 2. Bradykinin Mediated Calcium Mobilization in
Endothelial Cells of Veins Stored in Preservative Solutions
Storage Time (minute)
Solution
HLS
AHB
TCM
HBSS
GALA
60
180
300
1.31 ⫾ 0.11
1.30 ⫾ 0.15
1.35 ⫾ 0.04
2.30 ⫾ 0.03
2.75 ⫾ 0.25
nt
nt
nt
1.25 ⫾ 0.25
2.86 ⫾ 0.30
nt
nt
nt
1.01 ⫾ 0.05
2.90 ⫾ 0.45
Values represent mean ⫾ SEM of normalized calcium fluorescence counts
in arbitrary units: n ⫽ 9; nt ⫽ not tested secondary to lack of predicted
viability.
Human saphenous vein segments stored in various preservation solutions: AHB ⫽ blood with added heparin (40 Units/mL);
HBSS ⫽
Hanks’ balanced salt solution;
GALA ⫽ physiological salt solution
with glutathione, ascorbic acid, L-arginine, and heparin;
TCM ⫽
RPMI/M199culture media in 1:1 ratio.
Table 2 and Figure 4. The bradykinin-dependent calcium
mobilization was impaired in vein segments stored for 60
minutes in HLS, AHB and TCM (Table 2). Correspondingly, activation of eNOS was also severely attenuated
during 60 minutes of storage in HLS, AHB and TCM (Fig
4). Initially, a robust bradykinin-mediated calcium mobilization (Table 2), activation of eNOS’ and generation of
nitric oxide (Fig 4) was observed in segments stored in
HBSS solution. As expected, production of nitric oxide
was completely inhibited by treating the segments with
the eNOS inhibitor, L-NNA (1.0 ⫾ 0.012, mean ⫾ SEM of
normalized nitric oxide fluorescence, arbitrary units, n ⫽
1149
6) indicating specificity. However, a time-dependent decay in calcium mobilization (Table 2) and nitric oxide
generation was observed in vessels stored in HBSS as
shown in Figure 4, and as reported previously [20]. In
contrast to all the storage solutions tested, endothelial
cells remained robustly functional in veins stored in the
GALA preservation solution for the total duration of the
study (Table 2, Fig 4). The bradykinin-mediated mobilization of calcium (Table 2) and activation of eNOS lead to
sustained synthesis and release of nitric oxide by the
endothelial cells of the vessels during prolonged storage
in GALA (Fig 4). These results clearly demonstrate that
the GALA solution can preserve both the structure and
function of endothelium in ex vivo stored veins. To
demonstrate that the structural integrity of the vein was
maintained during 24 hours of storage in GALA, the vein
was labeled with live/dead assay reagents and imaged
using multiphoton microscopy. The vessel was imaged at
increasing depths along the Z axis throughout the intact
longitudinal vein segment. As illustrated in Figure 5,
living cells (green fluorescence) were clearly observed
throughout all the regions of the vein, including media
and intima, demonstrating structural viability of the
GALA preserved vessel.
Comment
The preservation of saphenous vein endothelial cells
during the course of CABG surgery is essential for the
long-term patency of the grafts. Endothelial cells are
important mediators in regulating platelet function and
in determining the coagulant and fibrinolytic pathways
Fig 4. Vein segments were stored in preservative solutions for various time points,
then incubated with diaminofluorescein
(DAF) to measure nitric oxide generation. The
integrated DAF fluorescence intensity in the
endothelial region of each vein was quantitated using multi-photon microscopy after a
10-minute treatment with bradykinin (10
␮mol/L) and was then normalized to the fluorescence intensity measured before the drug
treatment (dashed line ⫽ 1). Each bar represents the mean ⫾ SEM of independent experiments (n ⫽ 9). Nitric oxide generation
was severely impaired in veins stored in heparinized lidocaine saline (shaded bars), autologous heparinized blood (black-spotted
bars), and tissue culture medium (whitespotted bars). Nitric oxide production was
maintained in vessels during short-term storage in Hank’s balanced salt solution (black
bars); however, a temporal decrease in endothelial nitric oxide synthase (eNOS) activity
was observed. In contrast, eNOS activation
and nitric oxide generation was well preserved in vessels stored in glutathione, ascorbic acid, and L-arginine (GALA) solution
(white bars). A sustained increase in nitric
oxide production was observed during extended storage of the vessels in GALA.
CARDIOVASCULAR
Ann Thorac Surg
2003;75:1145–52
1150
THATTE ET AL
MICROSCOPIC EVALUATION OF SAPHENOUS VEIN ENDOTHELIUM
Ann Thorac Surg
2003;75:1145–52
CARDIOVASCULAR
Fig 5. In-depth imaging of a saphenous
vein segment stored in GALA solution using multi-photon microscopy. The vein was
imaged at 4-␮m steps along the Z axis, and
the figure is a representative montage of
two-dimensional images taken at depths
mentioned in the figure. Green fluorescence indicates living cells; red fluorescence indicates dead cells. The smooth
muscle cell layer (SMC; adventitia/media)
is identifiable starting at ⬃4 ␮m; vessel
lumen and endothelial cell layer becomes
clearly visible at ⬃120 ␮m. This figure illustrates that the cell viability was well
maintained throughout the vessel stored in
GALA for 1440 minutes. ⫻400 magnification.
in the vessel wall. Endothelial injury can lead to acute
alterations in these essential pathways, leading to thrombosis and stenosis [1–5]. Histochemical analyses have
suggested that structural derangements in the harvested
veins can be detected within two hours of storage [2–7].
In this study, we used multi-photon microscopy to demonstrate that endothelial cell viability is compromised
within minutes of storage in standard preservation solutions. Current storage solutions, which vary from physiologic salt solutions to heparinized blood, may not represent a medium sufficient for endothelial or smooth
muscle cell support [5– 8]. For example, standard saline
solutions lack an energy source, such as glucose, and
have a low pH that may be injurious to the fragile
endothelial cells [5, 8]. Moreover, the currently available
solutions are deficient in free radical scavengers, antioxidants, and nitric oxide synthase substrate that might help
sustain endothelial cell function during the storage period [5].
Endothelial injury may occur at the time of surgery as
a result of graft preservation, graft insertion and turbulent blood flow, or with local stasis [5]. Defining the time
and nature of this injury is of prime importance. Our
ability to evaluate the structural viability of an intact vein
segment and simultaneously measure calcium mobilization and nitric oxide production in real time in the same
segment using multiphoton microscopy offers a new and
powerful approach into the examination of endothelial
cell structure and function of stored grafts [5, 14]. We
were able to image to a depth of 200 ␮m into preserved
intact saphenous vein segments because of the robust
fluorescence signals of the calcein-ethidium homodimer
dyes. Imaging was limited to a depth of 50 to 70 ␮m using
nitric oxide dye due to degradation of the weak fluorescence signal by the thick tissue [11, 14]. Thus, signal
decay was a limiting factor in using the intact vessel
segments for functional studies. To compensate for this
limitation, and to consistently use the same vessel for
both structural and functional studies, calcium and nitric
oxide assays were performed on cross sections of the
intact vein segments that were used in structural
evaluation.
We have demonstrated that the structural viability, the
bradykinin-mediated increase in internal calcium concentration, and the activation of eNOS leading to nitric
oxide generation in the endothelium, are lost in saphenous veins within minutes of storage in HLS, AHB, and
TCM (Figs 1– 4, Tables 1 and 2). Similarly, HBSS provides
only short-term protection to the vein endothelium
against structural and functional decay and does not
protect against disruption-of-flow-induced apoptosis [21]
in explanted vessels (manuscript in preparation). These
results are in agreement with other published studies
that have demonstrated that storage of vessels in HLS
and AHB and other storage solutions leads to damaged
endothelium and depressed vasomotor function, perhaps
due to free radical damage, cessation of flow induced
apoptosis, low pH, storage media composition, and a
hostile environment [2– 8, 17, 21, 22].
The importance of endothelial calcium and nitric oxide
in regulating vasomotor function has been documented
[2– 8, 17–19, 22]. Nitric oxide is known to inhibit smooth
muscle cell proliferation, as well as platelet adhesion and
activation [18, 19]. Therefore, a loss of endotheliumderived nitric oxide may contribute both to acute thrombosis and intimal proliferation in the graft [2– 6]. Endothelial damage also reduces the production of
prostacyclin, a powerful inhibitor of platelet aggregation,
the loss of which may further promote platelet activation.
Platelet activation may precipitate early thrombosis and
lead to neointimal proliferation during the first few
months following bypass surgery [2– 6]. The loss in
eNOS activity and in endothelial cells’ ability to generate
nitric oxide which we observed in veins stored in standard solutions may adversely affect the vaso-reactivity
and long-term patency of the grafts.
THATTE ET AL
MICROSCOPIC EVALUATION OF SAPHENOUS VEIN ENDOTHELIUM
We hypothesized that the rapid loss of endothelial cell
structural and functional integrity in saphenous vein
stored in standard storage solutions can be avoided by
developing a physiologic salt solution, GALA, that combines free radical scavengers and antioxidants (glutathione, ascorbic acid,) and nitric oxide synthase substrate
(L-arginine). These components provide a favorable environment and cellular support during ex vivo storage.
The protection of structure and function of the endothelium, which we observed in vessels stored over a prolonged period in GALA, confirms this hypothesis (Figs
1–5, Tables 1 and 2).
The three key ingredients added to the GALA were
chosen because of their putative effect on endothelial cell
function. The role of glutathione as a cellular reducing
agent has been extensively investigated. Glutathione has
been found to increase L-arginine transport in endothelial cells and may lead to the formation of biologically
active S-nitrosoglutathione and to the stimulation of
eNOS activity, nitric oxide generation, and coronary
vasodilatation [20, 23]. Ascorbic acid is an antioxidant
known to scavenge reactive oxygen species and thus
demonstrate a sparing action on cellular glutathione and
alpha tocopherol in the plasma membranes [20, 23].
Ascorbic acid also increases eNOS activity by preserving
endothelium-derived nitric oxide bioactivity by possibly
scavenging superoxide anions and preventing oxidative
destruction of tetrahydrobiopterin, an eNOS cofactor [20,
23]. The presence of ascorbic acid in GALA may prevent
the oxidation of this cofactor during vessel storage and
help maintain eNOS function and nitric oxide generation
in vascular endothelium [23]. Nitric oxide also reacts with
thiols and can be trapped inside cells in the form of
S-nitrosoglutathione and other nitrosothiols. These nitrosylated compounds can facilitate long-term availability of nitric oxide in the cell [20, 23, 24]. Therefore,
ascorbic acid, by its reducing property, may assist sustained long-term release of nitric oxide from these compounds in vessels preserved in GALA, and thus help
maintain the patency and tone of the vessels during
storage. Additionally, ascorbic acid mediated reversal of
endothelial dysfunction, reduced platelet activation and
leukocyte adhesion, inhibition of smooth muscle cell
proliferation and lipid peroxidation, and increased prostacyclin production have been demonstrated in numerous cardiovascular pathologies [2, 25]. L-arginine is a
known substrate of nitric oxide synthase [19, 23] and has
been shown to decrease neutrophil-endothelial cell interactions in inflamed vessels [23]. Therefore, the use of
glutathione, ascorbic acid, and L-arginine may act synergistically to enhance the cell preservation properties of
the GALA solution.
Preservation of endothelial viability in GALA-stored
grafts may help counterbalance detrimental effects of
vein graft arterialization [2, 5], the ill effects of which may
be exacerbated when grafts with already damaged endothelium are used for implantation. We have completed
preliminary studies in which GALA preserved grafts
were exposed to arterial pressures in vitro, and the
endothelial structure and function remained intact for
1151
extended periods of time. GALA also prevented disruption-of-flow induced apoptosis in these vessels (manuscript in preparation). It is well established that the
improved patency of internal mammary artery grafts
may be due to the fact that they are not placed in ex vivo
storage and as such maintain their endothelial and
smooth muscle cell functions [2– 6]. The findings from
this study prompt us to speculate that storage in GALA
during conduit harvesting may render a long-term protective effect on the saphenous vein and may improve its
long-term patency to a degree approaching to that of the
internal mammary artery graft.
We gratefully acknowledge the expert technical assistance of
Jin-Hwa Rhee, Thomas McGarry and Sofija Zagarins. We would
like to thank Nancy Healey for editorial assistance and Aditi
Thatte for her encouragement. This work was supported by
National Institutes of Health grants (TM), a Veteran’s Affairs
Merit Review Grant (SFK), and the Richard Warren Surgical
Research and Educational Fund (HST, SFK).
References
1. Sharma GVRK, Deupree RH, Khuri SF, et al. Coronary
bypass graft surgery improves survival in high risk unstable
angina: results of a VA co-operative study with a eight year
followup. Circulation 1991;84:260 –7.
2. Mills NL, Everson CT. Vein graft failure. Curr Opin Cardiol
1995;10:562–8.
3. Davies MG, Hagen PO. Pathophysiology of vein graft failure:
a review. Eur J Vasc Endovasc Surg 1995;9:7–18.
4. Edmunds LH. Techniques of myocardial revascularization.
In: Edmunds LH, ed. Cardiac surgery in the adult. New
York: McGraw-Hill, 1997:481–534.
5. Thatte HS, Khuri SK. The coronary artery bypass conduit: I.
Intraoperative endothelial injury and its implication on graft
patency. Ann Thorac Surg 2001;72:S2245–52.
6. Sellke FW, Boyle EM Jr, Verrier ED. Endothelial cell injury in
cardiovascular surgery: the pathophysiology of vasomotor
dysfunction. Ann Thorac Surg 1996;62:1222–8.
7. Anastasiou N, Allen S, Paniagua R, et al. Altered endothelial
and smooth muscle cell reactivity caused by University of
Wisconsin preservation solution in human saphenous vein.
J Vasc Surg 1997;25:713–21.
8. Wagner R. Intimal protection of bypass-veins during intraoperative storage in blood or Euro-Collins solution: the role
of medium, temperature, and time. J Thorac Cardiovasc
Surg 1990;38:151–6.
9. Denk W, Strickler JH, Webb WW. Two-photon laser scanning fluorescence microscopy. Science 1990;248:73–6.
10. Masters BR, So PT, Gratton E. Multi-photon excitation
fluorescence microscopy and spectroscopy of in vivo human
skin. Biophys J 1997;72:2405–12.
11. Centonze VE, White JG. Multi-photon excitation provides
optical sections from deeper within scattering specimens
than confocal imaging. Biophys J 1998;75:2015–24.
12. Goetz RM, Thatte HS, Prabhakar P, et al. Estradiol induces
the calcium-dependent translocation of endothelial nitric
oxide synthase. Proc Natl Acad Sci 1999;96:2788 –93.
13. Nakatsubo N, Kojima H, Kikuchi K, et al. Direct evidence of
nitric oxide production from bovine aortic endothelial cells
using new fluorescence indicators: diaminofluoresceins.
FEBS Lett 1998;427:263–6.
14. Biswas KS, Thatte HS, Najjar SF, et al. Multi-photon microscopy in the evaluation of human saphenous vein. J Surg Res
2001;95:37–43.
CARDIOVASCULAR
Ann Thorac Surg
2003;75:1145–52
1152
THATTE ET AL
MICROSCOPIC EVALUATION OF SAPHENOUS VEIN ENDOTHELIUM
CARDIOVASCULAR
15. Monette R, Small DL, Mealing G, et al. A fluorescence
confocal assay to assess neuronal viability in brain slices.
Brain Res Protoc 1998;2:99 –108.
16. Hahnel C, Somodi S, Weiss DG, et al. The keratocyte
network of human cornea: a three-dimensional study using
confocal laser scanning fluorescence microscopy. Cornea
2000;19:185–93.
17. Angelini GD, Christie MI, Bryan AJ, Lewis MJ. Surgical
preparation impairs release of endothelium derived relaxing
factor from human saphenous veins. Ann Thorac Surg
1989;48:417–21.
18. Loscalzo J, Welch G. Nitric oxide and its role in the cardiovascular system. Prog Cardiovasc Dis 1995;38:87–104.
19. Michel T, Feron OJ. Nitric oxide synthase: which, where,
how and why? J Clin Invest 1997;100:2146 –52.
20. Siow RC, Sato H, Leake DS. Vitamin C protects human
arterial smooth muscle cells against atherogenic lipopro-
21.
22.
23.
24.
25.
Ann Thorac Surg
2003;75:1145–52
teins: effects of antioxidant vitamin C and E on oxidized LDL
induced adaptive increases in cystine transport and glutathione. Arterioscler Thromb Vasc Biol 1998;18:1662–70.
Dimmeler S, Zeiher AM. Endothelial cell apoptosis in angiogenesis and vessel regression. Cir Res 2000;87:434 –9.
Cavallari N, Abebe W, Mingoli A, et al. Functional and
morphological evaluation of canine veins following
preservation in different storage media. J Surg Res 1997;68:
106 –15.
Tomasian D, Keaney JF, Vita JA. Antioxidants and bioactivity
of endothelial derived nitric oxide. Cardiovasc Res 2000;47:
426 –35.
Hofman H, Schmidt HH. Thiol dependence of nitric oxide
synthase. Biochemistry 1995;34:13443–52.
Heller R, Munscher-Paulig F, Grabner R. L-Ascorbic acid
potentiates nitric oxide synthesis in endothelial cells. J Biol
Chem 1999;274:8254 –60.
INVITED COMMENTARY
Meticulous harvesting, pharmacologic dilatation, and
avoidance of overdistention of a coronary artery bypass
graft have been recognized as important steps during
procurement, which may preserve graft function. Nonetheless, some vein grafts develop early complications
such as thrombosis and neointimal hyperplasia which
lead to atherosclerosis. Preservation during storage of a
coronary artery graft between the time of harvest and
implantation has received little attention, possibly because conduits are thought by many surgeons to be inert
or robust and not particularly sensitive to hypoxic damage. Protection of the endothelium is of major importance in maintaining smooth muscle function and integrity. Vasodilatation is dependent on cytosolic Ca⫹⫹
modulated by endothelial and nonendothelial dependent
pathways such as nitric oxide cyclic guanosine monophosphate (NO-cGMP), prostacyclin and endothelium
derived hyperpolarization. This paper demonstrates the
importance of preserving vein grafts during storage to
maintain structure and function, which the authors suggest may lead to improved patency.
Thatte and colleagues have explored an ingenious
approach using near-infrared multiphoton microscopy in
transmission and epifluorescent modes to assess the
integrity of the human saphenous vein segments stored
in different preservation solutions with the aim of designing more effective storage medium. Specific fluorescene
stains were used to assess calcium and nitric oxide
metabolism. Surprisingly, they demonstrated that using
standard preservation solutions, calcium mobilization,
and nitric oxide function diminished rapidly, and nearly
© 2003 by The Society of Thoracic Surgeons
Published by Elsevier Science Inc
all endothelial cells were nonviable after the first hour. In
contrast, they discovered that veins can be stored up to 24
hours without substantial loss of viability or function
using a newly formulated heparinized physiologic buffer
solution (GALA) containing the antioxidant and reducing
agents glutathione and ascorbic acid and L-arginine as a
substrate for nitric oxide synthase.
Arterial grafts, which contain a high proportion of
smooth muscle compared with saphenous veins may be
even more prone to vasospasm from intimal and smooth
muscle damage. In patients who had reoperation with a
prolonged interval between the time of harvest and graft
reimplantation, we have observed arterial grafts, which
were dilated at the time of leaving the operating room,
develop delayed vasoconstriction despite the use of systemic vasodilators. Storage in GALA solution following
harvesting and vasodilatation is a logical development as
demonstrated by this elegant research. GALA solution
may protect the endothelium and the underlying smooth
muscle in both arterial and venous grafts from ischemic
damage and thus deserves clinical evaluation.
Brian Buxton, MD
Director of Cardiac Surgery
Austin and Repatriation Medical Centre
Level 5
Studley Road
Heidelberg, Victoria 3084
Australia
e-mail: [email protected]
0003-4975/03/$30.00
PII S0003-4975(02)05010-5