Full text of DOI:10.1002/clc.4960250606

Clin. Cardiol. 25, 271–278 (2002)
The Effect of Intracoronary Fibroblast Growth Factor-2 on Restenosis after
Primary Angioplasty or Stent Placement in a Pig Model of Atherosclerosis
MARK J. POST, M.D., PH.D., ROGER J. LAHAM, M.D.,* RICHARD E. KUNTZ, M.D., FACC,† DEBORAH NOVICKI, PH.D.,‡
MICHAEL SIMONS, M.D., FACC
Dartmouth Medical School, Hanover, New Hampshire; *Beth Israel Deaconess Medical Center, and †Clinical Trial Center, Brigham
and Women’s Hospital, Boston, Massachusetts; ‡Chiron Corporation, Emeryville, California, USA
Summary
Background: Therapeutic angiogenesis, if combined with
primary percutaneous transluminal coronary angioplasty or
stent placement, could improve the outcome of patients suffer-
ing from multifocal coronary disease.
angiographic late lumen loss, a trend toward increased intimal
hyperplasia and decreased remodeling.
Conclusion: We conclude that rFGF-2 does not aggravate
restenosis after balloon dilation or stenting in this pig model of
coronary atherosclerosis. Future combinations of angioplasty
and therapeutic angiogenesis in a single session should be pur-
sued as a feasible and safe strategy.
Hypothesis: Because of the concern that angiogenic growth
factors might promote restenosis, we studied the effect of a sin-
gle intracoronary administration of recombinant fibroblast
growth factor (rFGF)-2 on restenosis after balloon angioplasty
and stent placement in a pig model of coronary atherosclerosis.
Methods: In 24 Yucatan minipigs, coronary lesions were in-
duced by arterial injury and 3 months of atherogenic diet.
After 3 months, repeat catheterization was performed with
balloon dilation or stent placement at the injured sites, with a
follow-up of 6 weeks. Results were monitored using quantita-
tive angiography, intravascular ultrasound (IVUS), and histo-
morphometry.
Results: Intracoronary rFGF-2 2 µg/kg did not affect neoin-
tima formation or remodeling in this model. A small but sig-
nificant aggravation of late lumen loss was observed in the ref-
erence segments of the rFGF-2-treated group. Angiographic
and echographic late lumen loss, intimal hyperplasia, and arte-
rial remodeling, as well as histologic neointima were all simi-
lar in the rFGF-2- and the vehicle-treated group. Confirming
earlier studies from our group and those of others, stented ar-
teries compared with balloon-dilated arteries had increased
Key words: angiogenesis, pig, balloon angioplasty, stent,
restenosis, growth factors, coronary disease
Introduction
Therapeutic angiogenesis has gained considerable attention
as an alternative strategy for revascularizing ischemic my-
ocardium that is not amenable to catheter-based revasculariza-
tion or coronary artery bypass grafting (CABG).1–4 Recently,
therapeutic angiogenesis with vascular endothelial growth
factor (VEGF) or basic fibroblast growth factor (FGF-2) has
entered clinical trials as an adjunctive therapy to CABG.5–7
Likewise, patients who are treated with stent placement or per-
cutaneous transluminal coronary angioplasty (PTCA), might
benefit from adjunct angiogenic growth factor treatment to
target ischemic areas not covered by the intervention. How-
ever, since FGF-2 is a nonselective mitogen, stimulates prolif-
eration of smooth muscle cells,8 and was associated with inti-
mal hyperplasia,9 FGF-2 might enhance restenosis. Studies in
nonatherosclerotic rat, rabbit, or canine animal models on the
effects of recombinant fibroblast growth factor (rFGF-2) on
intimal hyperplasia after balloon angioplasty are contradicto-
ry10–12 and might have limited relevance to the response of a
coronary artery with an advanced atherosclerotic lesion to
stents. Therefore, we decided to study the healing response of
coronary arteries after PTCA and stent placement with ad-
junctive treatment of rFGF-2 in an established pig model of
atherosclerosis.13–15
This study was supported in part by grants from the NIH grants
HL53793, HL56993 to MS, and a grant from Chiron Corporation.
Address for reprints:
Mark J. Post, M.D., Ph.D.
Associate Professor of Medicine and Physiology
Dartmouth Medical School
Director of Preclinical Research
Angiogenesis Research Center
Methods
Dartmouth Hitchcock Medical Center, HB 7700
1 Medical Center Drive, Lebanon, NH 03756, USA
e-mail: Mark.J.Post@Hitchcock.org
Yucatan Model of Atherosclerosis
All animals underwent three procedures: initial injury, an
intervention and treatment session, and the final study.
Received: March 28, 2001
Accepted with revision: October 5, 2001

272
Clin. Cardiol. Vol. 25, June 2002
Start atherogenic diet
2 weeks
intervention.” Following intervention, 12 animals were in-
fused (10 min per artery, #2.5 Fr coronary infusion catheter,
ACS, Temecula, Calif., USA) with bovine recombinant fi-
broblast growth factor (rFGF-2, Chiron Corporation, Emery-
ville, Calif., USA) at a total dose of 2 µg/kg, equally distribut-
ed over the two treated arteries. Twelve animals received
vehicle. The rFGF-2 vehicle consisted of 10 mM sodium cit-
rate, 10 mM thioglycerol, 135 mM sodium chloride, and 100
mM EDTA, pH 5.0.
Fogarty injury
13 weeks
Vehicle
FGF-2
24 arteries (n): 24 arteries (n):
Control (6)
PTCA (10)
Stent (8)
Control (5)
PTCA (10)
Stent (9)
Balloon Angioplasty and Stent Placement
4 weeks
Sacrifice
A #8 Fr guiding catheter (Cordis) was placed in the left or
right coronary ostium. For PTCA, a standard angioplasty bal-
loon, 3–4.5 mm wide, 20 mm long, (Guidant) was introduced
over a 0.014" guidewire to the selected site of angioplasty.
Three 1 min inflations were performed at a nominal pressure
to a dilation ratio of 1.2 according to the manufacturers table
and artery diameter on IVUS.
FIG. 1 Layout of the study. PTCA = percutaneous transluminal
coronary angioplasty, FGF-2 = fibroblast growth factor-2.
The animals were preanesthetized with ketamine 10 mg/kg,
intubated, and ventilated with room air and isoflurane 2–3%.
Peri- and postoperative treatment consisted of bretylium tosy-
late 4 mg/kg intravenously (IV), IV heparin to maintain an ac-
tivated clotting time > 300 s, a single intramuscular dose of
0.02 mg/kg buprenorphine, and cefazolin 0.5 gm IV. Anti-
platelet therapy started 1 day before intervention with acetyl-
salicylic acid (325 mg, 2 days) and ticlopidine hydrochloride
(250 mg, b.i.d., 14 days). A #8 Fr sheath (Cordis, Miami, Fla.,
USA) was advanced through a femoral cutdown, which was
repaired for second use during the intervention. All animals re-
ceived humane care in compliance with the Institutional Ani-
mal Care and Use Committee at Beth Israel Deaconess Med-
ical Center and the Guide for the Care and Use of Laboratory
Animals published by the U.S. National Institutes of Health
(NIH Publication No. 85-23, revised 1996).
Male Yucatan minipigs (10–15 kg) received an atherogenic
diet (Purina minipig starter diet with 1.5% cholesterol and 6%
peanut oil) that was continued until final study. Two weeks lat-
er, coronary angiograms were made and the left coronary cir-
cumflex (LCx) and the right coronary artery (RCA) were in-
jured with a standard angioplasty balloon (Guidant, Santa
Clara, Calif., USA), at 8 atm with a dilation ratio of 1.2.
Thirteen weeks later, animals received intervention and
treatment randomized for artery and animal, respectively, pri-
or to baseline studies. Baseline selective quantitative coronary
angiography (QCA) and intravascular ultrasound (IVUS) of
the LCx and RCA were made. Stent placement or PTCA was
performed at the site of maximum angiographic stenosis or
maximal plaque load on IVUS. After intervention, QCA and
IVUS were repeated.
Palmaz-Schatz biliary stents (5–9 mm wide, 15 mm,
Johnson and Johnson, Warren, N.J., USA) were crimped on a
angioplasty balloon. After placement over a 0.014" guidewire,
the stent was deployed by a 10 s inflation at 4 atm. If necessary
for a dilation ratio of 1.2 or optimal apposition of the stent,
postdilatation was performed.
Coronary Angiography
Coronary angiograms were made at initial injury, before
and after intervention, and at follow-up (FU). Before angiog-
raphy, 2 ml of a 125 µg/ml nitroglycerin and 2.5 mg/ml pa-
paverine cocktail (NP cocktail) was injected intracoronary.
Side branches were used to localize the site of injury and in-
tervention during follow-up angiography and intervention.
Cine QCA (Model LU, GE Medical, Milwaukee, Wis., USA)
was performed in standard right and left anterior oblique pro-
jections (contrast: Renografin, Squibb, Princeton, N.J., USA)
using a #7-FR JR4 diagnostic angiography catheter (Cordis).
The cine films were analyzed off-line, using 4ꢀ magnified
end-diastolic frames and digital calipers. The tip of the guid-
ing catheter was used for calibration.
The site with minimum lumen diameter (MLD) prior to the
intervention (PRE) was used for calculation of acute gain, late
lumen loss, angiographic stenosis, and dilation ratio. The ref-
erence lumen diameter (RLD) was calculated as the average of
proximal and distal reference. Angiographic acute gain was
defined as post minus preintervention MLD (MLDPOSTꢁ
MLDPRE), and late lumen loss was defined as MLDPOSTꢁ
MLDFU or MLDPREꢁMLDFU for “no intervention” arteries.
Angiographic stenosis was calculated as (1-MLD/RLD) ꢀ
100. Dilation ratio = balloon diameter on fluoroscopy/RLD.
At the final study 6 weeks after intervention, QCA and
IVUS of the LCx and RCA were repeated, the animals eutha-
nized, and the hearts excised.
Coronary Intravascular Ultrasound
Study Design and Drug Administration (Fig. 1)
For IVUS, we used a #2.9 F Microrail 30 MHz coronary
imaging catheter (CVIS Inc., Sunnyvale, Calif., USA). The
NP cocktail was injected into each artery. The catheter was ad-
The RCA and LCx were randomly assigned to PTCA or
stenting, and in each group four arteries were assigned to “no

M. J. Post et al.: Intracoronary FGF-2 and restenosis
273
vanced over a 0.014" guidewire, with the tip distal from the le-
sion/intervention site and manually withdrawn while imaging.
At MLD, two peripheral sites of the lesion, and at the distal
and proximal reference, the catheter was held for still images.
Off-line video-tape analysis and quantification of a proximal
and a distal reference and three evenly distributed cross sec-
tions at the level of intervention were performed using soft-
ware available in the CVIS console. Lumen area (LA), media-
bounded area (MBA), and stent area (SA) were manually
traced. Precise sites of measurements were localized by fluo-
roscopy of the IVUS catheter tip with reference to side branch-
es. Prior to stent placement or PTCA, LA and MBA were used
to calculate area stenosis (100 ꢀ LAMLD/MBAMLD), LA
stenosis (100 ꢀ {1-LAMLD/LAREF}), and plaque area (PA =
MBA ꢁ LA). The immediate gain of IVUS was defined as
LAPOST – LAPRE and late loss as LAPOST ꢁ LAFU (mm2) or
LAPRE ꢁLAFU for “no intervention” arteries. Intima gain was
calculated from MBA or SA and LA at follow-up and at in-
tervention, assuming constant plaque load and within-stent
growth of neointima: for PTCA, intima gain = (MBA ꢁ
LA)FU ꢁ PA and for stent, intima area gain = (SA ꢁ LA).
Optimas 6.0, Edwards, Wash., USA) were performed by man-
ually tracing perimeters of lumen, intima, preexistent plaque
(in stents), media, and adventitia.
Data Analysis and Statistical Analysis
All measurements and data analyses were performed by ob-
servers blinded to treatment. Data in text, tables, and charts are
means(standard error of the mean. “No intervention” arteries
were too few to incorporate into the statistical analysis. Anal-
ysis of variance (ANOVA) was used to study the influence of
treatment and intervention. Bonferroni correction on post hoc
tests was used to evaluate differences between groups. Since
late loss (angiographic or echographic) usually correlates with
immediate gain,16 we tested regression models with immedi-
ate gain, intervention, and treatment as independent and late
loss as dependent variables.
Results
Hemodynamics
Intracoronary rFGF-2 infusion at 2 µg/kg did not affect
heart rate or blood pressure. Mean arterial blood pressure
(MAP) was 81 ± 5 mmHg before and 82 ± 5 mmHg 20 min af-
ter start of FGF-2 infusion. In the vehicle group, MAP was 73
± 4 mmHg before and during the infusion. Heart rate was also
constant during infusion.
Tissue Analysis
After excision, hearts were pressure perfused with parafor-
maldehyde 4% for 2 h. Arterial segments were dissected and
stored overnight in 20% sucrose at 4ºC. Five mm tissue blocks
were embedded in OCT, frozen, and stored at ꢁ80ºC. Serial
sections (5 µm) were made with hematoxylin and eosin (H&E)
and elastin van Gieson stains. Stent struts were microscopical-
ly removed. Photomicrographs were digitally acquired (Spot
camera, Diagnostic Instruments, Sterling Heights, Mich.,
USA) and stored. Morphometric measurements (Bioscan
Angiography
In five animals (vehicle/stent: two, vehicle/control: two,
FGF-2/control: one), the angiographic data were not available.
The size of the arteries (RLD before intervention, Table I) was
TABLE I Angiographic parameters
Intervention
Treatment
No intervention
PTCA
Stent
Vehicle
rFGF-2
Vehicle
9
rFGF-2
Vehicle
6
rFGF-2
N
4
4
2.91 ± 0.08
2.78 ± 0.17
4.6 ± 5.7
NA
10
9
Pre RLD (mm)
2.80 ± 0.15
2.40 ± 0.17
9.6 ± 3.6
NA
2.75 ± 0.11
2.57 ± 0.11
5.7 ± 1.9
1.16 ± 0.06
2.54 ± 0.13
2.81 ± 0.06
ꢁ10.1 ± 0.7
2.77 ± 0.07
2.57 ± 0.12
8.2 ± 2.9
2.62 ± 0.14
2.49 ± 0.13
4.2 ± 3.1
1.22 ± 0.13
2.65 ± 0.12
2.59 ± 0.11
2.3 ± 0.5
2.39 ± 0.10
2.26 ± 0.13
5.7 ± 3.7
0.26 ± 0.08 a
0.08 ± 0.11
0.33 ± 0. 10
2.67 ± 0.07
2.58 ± 0.10
3.1 ± 1.4
1.23 ± 0.01
2.28 ± 0.24
2.78 ± 0.11
ꢁ4.1 ± 0.6
2.25 ± 0.21
1.76 ± 0.20
20.2 ± 7.4
2.44 ± 0.15
2.25 ± 0.09
7.6 ± 3.5
Pre-MLD (mm)
Pre-diam stenosis (%)
Dilation ratio
1.19 ± 0.06
2.48 ± 0.27
2.52 ± 0.20
ꢁ2.1 ± 1.1
2.27 ± 0.25
1.54 ± 0.24
33.0 ± 15.2
0.21 ± 0.11 a
0.26 ± 0.09
0.98 ± 0.11 b
Post-RLD (mm)
Post-MLD (mm)
Post-diam stenosis (%)
FU-RLD (mm)
FU-MLD (mm)
FU-diam stenosis (%)
Ref late loss (mm)
Immediate gain (mm)
Late loss (mm)
NA
NA
NA
NA
NA
NA
2.79 ± 0.39
2.61 ± 0.36
9.4 ± 4.7
NA
2.99 ± 0.23
2.68 ± 0.23
10.4 ± 6.0
NA
ꢁ0.22 ± 0.15
0.25 ± 0.08
0.25 ± 0.14
0.03 ± 0.2
NA
NA
0.19 ± 0.05
1.02 ± 0.11 b
0.09 ± 0.09
ꢁ0.04 ± 0.04
a p = 0.033 for treatment effect (ANOVA).
b p<0.001 (ANOVA).
For all unmarked data, p > 0.2.
Abbreviations: PTCA = percutaneous transluminal coronary angioplasty, NA = not applicable, RLD = reference lumen diameter, MLD = mini-
mum lumen diameter, FU = follow up, rFGF-2 = recombinant fibroblast growth factor-2, N = number of arteries.

274
Clin. Cardiol. Vol. 25, June 2002
not different between the treatment groups (ANOVA, p =
0.364). Stenoses were also comparable (p = 0.672). A small
angiographic late loss in the reference segments of treated ar-
teries due to transient vasospasm was maximally 0.03 mm in
the vehicle group and 0.21 to 0.26 mm in the FGF-2 group
(Table I, ANOVA, p = 0.033 for treatment effect, not signifi-
cant for intervention effect).
Dilation ratio and immediate gain determine the extent of
arterial wall injury and thus the resultant late loss. Dilation ra-
tios were very similar in the treatment and intervention groups
(Table I, p = 0.847 and p = 0.789, respectively). Immediate
gain was also not different for either intervention (p = 0.593) or
treatment (p = 0.521). Angiographic late loss was higher in the
stent than in the PTCA groups (Table I, Fig. 2, p < 0.001) but
similar in the vehicle or FGF-2 groups (p = 0.869). In a regres-
sion model with immediate gain, intervention and treatment
effects on late loss, immediate gain, and intervention were sig-
nificant determinants of late loss (beta coefficients 0.55, p =
0.05 and 0.67, p< 0.001, respectively) and the treatment effect
was not (beta: 0.07, p = 0.63). Power calculation revealed that
the study had a 80% power to detect a difference in late loss of
0.4 mm between treatment groups.
*
1.2
1.0
0.8
0.6
0.4
0.2
0.0
Acute
gain
Late
loss
Acute
gain
Late
loss
PTCA
Stent
FIG. 2 Angiographic results of PTCA and stented arteries from the
vehicle group (black bars) and the rFGF-2 group (hatched bars).
* p <0.001. Abbreviation as in Fig. 1.
vention and p = 0.956 for treatment effect). No treatment effect
was seen on late loss (p = 0.585), intima gain (p = 0.717), or
MBA loss (p = 0.765). Regression of echographic immediate
gain and treatment on echographic late loss again showed a
significant relation with immediate gain (beta: 0.87, p<0.001),
but not with treatment (beta: 0.34, p = 0.579). Stent placement
resulted in significantly less MBA loss (Fig. 4, p = 0.05) than
PTCA, but showed a trend toward increased intima gain (p =
0.108). Consequently, late loss was similar in the stented and
PTCA-treated arteries (p = 0.744). Late loss in the reference
segments was similar in the vehicle and FGF-treated animals
and was attributed for 60–75% and 100% to neointima forma-
tion in PTCA and stent-treated arteries, respectively.
Intravascular Ultrasound
Forty-three arterial segments had complete serial scans
(Fig. 3). Arteries in the vehicle/PTCA group were slightly big-
ger than in other groups (Table II, ANOVA: p = 0.046), but
area stenosis and lumen stenosis were similar.
Immediate gain was slightly but not significantly higher af-
ter stenting than after PTCA, and similar for the vehicle and
FGF-2 groups (Table II, Fig. 4, ANOVA, p = 0.108 for inter-
FIG. 3 Intravascular ultrasound (IVUS) cross sections from coronary arteries treated by PTCA (A, B) or stent placement (C, D), and with vehi-
cle- (A, C) or rFGF-2- (B, D) treated coronary artery immediately after (left panels, post) intervention and at follow-up (right panels, follow-up
[FU]) 6 weeks later. Lumen and media-bounded areas were manually traced and their measures displayed (A1 and A2, respectively) in the in-
sets. Abbreviation as in Fig. 1.

M. J. Post et al.: Intracoronary FGF-2 and restenosis
275
TABLE II Intravascular ultrasound parameters
Intervention
Treatment
No intervention
PTCA
Stent
Vehicle
rFGF-2
Vehicle
9
rFGF-2
Vehicle
rFGF-2
9
N
4
4
10
7
Pre ref area (mm2)
Pre MLA (mm2)
Pre LA-stenosis (%)
Pre area stenosis (%)
Post ref area (mm2)
Post MLA (mm2)
Post LA stenosis (%)
FU ref area (mm2)
FU MLA (mm2)
FU LA stenosis (%)
Ref late loss (mm2)
Ref MBA loss (mm2)
Immediate gain (mm2)
Late loss (mm2)
Intima area gain (mm2)
MBA loss (mm2)
5.93 ± 1.20
5.43 ± 1.19
8.6 ± 6.5
14.2 ± 5.9
NA
NA
NA
6.88 ± 1.24
6.47 ± 1.04
3.9 ± 4.4
6.45 ± 0.70
6.39 ± 0.86
0.4 ± 7.0
15.7 ± 6.3
NA
NA
NA
4.89 ± 0.27
5.30 ± 0.40
ꢁ8.2 ± 5.4
8.83 ± 0.59 a
8.06 ± 0.53
8.0 ± 3.7
6.98 ± 0.60
6.92 ± 0.74
1.4 ± 4.9
6.94 ± 0.23
5.97 ± 0.46
13.9 ± 6.0
22.8 ± 7.5
7.45 ± 1.09
7.18 ± 0.51
1.5 ± 10.8
6.26 ± 0.77
4.75 ± 0.79
20.0 ± 24.0
1.19 ± 0.83
1.03 ± 0.28
1.22 ± 0.46
2.43 ± 0.76
2.19 ± 0.77
6.70 ± 0.58
5.87 ± 0.43
11.2 ± 4.5
15.7 ± 4.5
7.94 ± 0.90
7.64 ± 0.88
1.6 ± 8.2
6.84 ± 0.50
4.07 ± 0.73
42.3 ± 7.3
1.10 ± 0.71
1.28 ± 0.28
1.77 ± 0.76
3.57 ± 0.70
3.19 ± 0.73
0.38 ± 0.21 b
13.2 ± 4.3
9.3 ± 4.1
8.42 ± 0.93
8.88 ± 0.78
ꢁ11.4 ± 10.1
7.47 ± 0.51
6.01 ± 0.53
19.0 ± 4.7
0.93 ± 0.52
0.70 ± 0.35
0.82 ± 0.44
2.87 ± 0.85
1.47 ± 0.55
1.41 ± 0.38
7.09 ± 0.58
7.25 ± 0.77
ꢁ3.5 ± 7.1
5.40 ± 0.56
4.64 ± 0.37
7.0 ± 9.2
1.69 ± 0.55
1.04 ± 0.45
0.34 ± 0.38
2.61 ± 0.74
0.98 ± 0.42
1.63 ± 0.89
NA
NA
NA
NA
NA
NA
NA
NA
0.24 ± 0.25 b
a p = 0.046 vehicle versus rFGF-2.
b p = 0.02 stent versus PTCA.
For all unmarked data, p > 0.2.
Abbreviations: Ref = reference, MLA = minimum lumen area, LA = lumen area, FU = follow up, MBA = media bounded area, N = number of ar-
teries, NA = not applicable.
Histology
Histomorphometric analysis of harvested arteries con-
(Pearson, r = 0.636, p < 0.001) and intima area (r = 0.580,
p < 0.001), correlated fairly well.
firmed IVUS findings (Table III, Fig. 5). Treatment or inter-
vention had no effect on morphometric parameters; however,
in the rFGF-2-treated group, media area showed a trend to-
ward reduction (p = 0.06, ANOVA). Ultrasound and histo-
morphometric parameters at follow-up, such as lumen area
Discussion
Clinical Background
In numerous preclinical experiments with either
VEGF16511, 17, 18 or with FGF-1,19, 20 FGF-2,21, 22 or FGF-5,23
blood flow to ischemic myocardium could be enhanced by
angiogenesis, which resulted in improved myocardial func-
tion in some studies.17, 20, 23 Surprisingly, this functional an-
giogenesis was triggered and sustained by a single adminis-
Late loss
Intima gain
MBA loss
4.5
4
3.5
3
*
2.5
2
TABLE III Morphometric results of recombinant fibroblast growth
factor (rFGF-2) treatment and intervention
1.5
1
Vehicle
PTCA
rFGF-2
PTCA
10
0.5
0
Treatment
N
Stent
6
Stent
9
9
PTCA Stent
PTCA Stent PTCA Stent
Adventitia (mm2) 3.17 ± 0.39 1.81 ± 0.23 2.64 ± 0.49 2.09 ± 0.21
Media (mm2)
Intima (mm2)
1.61 ± 0.24 1.31 ± 0.22 1.09 ± 0.15 1.01 ± 0.15
1.75 ± 0.65 1.96 ± 0.51 2.01 ± 0.47 3.19 ± 0.59
Lumen (mm2) 4.60 ± 0.36 2.72 ± 0.31 3.16 ± 0.37 3.50 ± 0.35
FIG. 4 Intravascular ultrasound results of immediate gain and late
loss in the PTCA and stented arteries from the vehicle group (black
bars) and the rFGF-2 group (hatched bars). The data correspond
with those in Table V. * p = 0.05. MBA = media-bounded area; other
abbreviations as in Fig. 3.
Abbreviations as in Table I.

276
Clin. Cardiol. Vol. 25, June 2002
(A)
(B)
FIG. 5 Photomicrographs of stented arteries from a vehicle- (A) and rFGF-2- (B) treated animal stained with elastin van Gieson. The arteries
are the same as used for Figure 2. A = adventitia, M = media, I = intima. The arrowheads indicate voids of stent struts. Bar = 750 µm.
tration of FGF-2.24 During the last 2 years, the clinical safety
of growth factor administration was demonstrated for FGF-1,
FGF-2, and VEFG165 either as adjunct to CABG with local
delivery or as stand-alone systemic therapy,5, 7, 25 allowing the
implementation of larger scale phase II/phase III trials. In a
phase I trial with single intracoronary FGF-2 administration,
1.5 µg/kg/artery (to a total dose of 3 µg/kg) was the lowest that
improved perfusion on single-photon emission computed to-
mography (SPECT).26
fusion protein,32 or antisense FGF-210 resulted in reduction of
neointima formation, also suggesting that the mitogenic ef-
fect of FGF-2 on smooth muscle cells might be dominant.
However, these studies have been performed primarily in the
rat carotid artery in the absence of preexistent intima and with
sustained delivery of FGF-1, FGF-2, or FGF-2 inhibitors, and
the present study suggests that this effect may be of little
practical relevance.
In this study, we further support the feasibility and safety of
such a strategy by showing that restenosis is not enhanced by a
single intracoronary administration of 2 µg/kg FGF-2. This
was true both for PTCA and for in-stent restenosis in an
atherosclerotic environment. The higher neointima in the stent-
ed arteries treated with FGF-2 was attributed to a difference in
acute gain, since the treatment effect was not significant in a re-
gression model that corrected for the influence of acute gain.
Potential Side Effects of Fibroblast Growth Factor-2
A second potential side effect of FGF-2 in atherosclerotic
plaque is the acceleration of plaque angiogenesis with the po-
tential of creating a vulnerable plaque that may rupture and
hemorrhage, causing an acute coronary syndrome. Indeed,
hemorrhages in the vicinity of newly formed vessels of athero-
sclerotic plaques have frequently been identified,33 and this
may be exaggerated by FGF-2-stimulated interstitial collage-
nase activity.34 We observed no plaque ruptures or hemor-
rhages in this model of mild disease.
Animal Model of Atherosclerosis
The study was performed in an established model of coro-
nary atherosclerosis,13, 14 which closely mimics human pathol-
ogy and the response to arterial injury13, 15, 27, 28 in spite of mild
disease. A dilation ratio of 1.2 was used to avoid overdisten-
sion, which resulted in a healing response comparable to that
after PTCA or stenting in human atherosclerotic arteries.28
Fibroblast Growth Factor-2 and Geometric Remodeling
In this study, neointima formation accounted for 51 and
38% of lumen renarrowing in the vehicle and FGF-2 PTCA
group, respectively, and for virtually all of the restenosis in the
stent groups. This finding is consistent with previous studies
using the iliac and femoral arteries in the same model.13, 35 The
complementing portion of lumen renarrowing is attributed to
remodeling, which was not influenced by FGF-2.
Fibroblast Growth Factor-2 and Arterial Healing
In stented arteries, neointima formation is the primary cause
of restenosis. Since FGF-2 is a mitogen for endothelial cells,
smooth muscle cells, and other mesenchymal cells, its effect
on neointima formation is not obvious. FGF-1 and FGF-2
may reduce neointima formation by accelerating endothelial
regeneration.11, 29, 30 FGF-1 reduced intimal hyperplasia in a
rat carotid artery model,29 but no inhibitory effect of FGF-2
on intimal hyperplasia has been shown yet. Instead, in a bal-
loon injury model in the rabbit, a single injection of FGF-2
has lead to enhancement of intimal hyperplasia.12 Several
strategies aiming at inhibition of the FGF-2/FGF-R1 pathway
either through antibodies against FGF-2,31 an FGF-2-toxin
Limitations
This study was performed in a small number of animals.
Higher numbers of observations were obtained by using both
branches of the left coronary artery. Although statistical anal-
ysis accounted for more observations in one pig, the overall
power of analysis was borderline (0.8 or higher). The FGF-2
was given as a single intracoronary infusion and recent clin-
ical data (FGF Initiating Revascularization Trial [FIRST])
indicate that this mode of delivery has limited efficacy.36 Al-

M. J. Post et al.: Intracoronary FGF-2 and restenosis
277
8. Schwartz SM, deBlois D, O’Brien ER: The intima. Soil for athero-
sclerosis and restenosis. Circ Res 1995;77:445–465
though double-blind randomized phase II trials of intracor-
onary FGF-2 in patients with CAD (FIRST) and intra-arterial
FGF-2 in patients with peripheral vascular disease (Thera-
peutic Angiogenesis with FGF-2 for Intermittent Claudica-
tion [TRAFFIC]) demonstrated significant symptomatic
relief and functional improvement in subgroups of FGF-2-
treated patients, this effect was temporary. The final proof of
efficacy of this approach, however, must await pivotal phase
III trials. It remains to be shown whether longer exposures to
these growth factors are more efficacious and equally safe
with respect to restenosis and atherogenesis.
9. Lindner V, Reidy MA: Expression of basic fibroblast growth factor
and its receptor by smooth muscle cells and endothelium in injured
rat arteries: An en face study. Circ Res 1993;73:589–595
10. Hanna AK, Fox JC, Neschis DG, Safford SD, Swain JL, Golden
MA: Antisense basic fibroblast growth factor gene transfer reduces
neointimal thickening after arterial injury. J Vasc Surg 1997;25:
320–325
11. Lazarous DF, Shou M, Scheinowitz M, Hodge E, Thirumurti V,
Kitsiou AN, Stiber JA, Lobo AD, Hunsberger S, Guetta E, Epstein
SE, Unger EF: Comparative effects of basic fibroblast growth fac-
tor and vascular endothelial growth factor on coronary collateral
development and the arterial response to injury. Circulation 1996;
94:1074–1082
12. Parish MA, Grossi EA, Baumann FG, Asai T, Rifkin DB, Colvin
SB, Galloway AC: Effects of a single administration of fibroblast
growth factor on vascular wall reaction to injury. Ann Thorac Surg
1995;59:948–954
13. de Smet BJ, van der Zande J, van der Helm YJ, Kuntz RE, Borst C,
Post MJ: The atherosclerotic Yucatan animal model to study the ar-
terial response after balloon angioplasty: The natural history of re-
modeling. Cardiovasc Res 1998;39:224–232
14. Gal D, Rongione AJ, Slovenkai GA, DeJesus ST, Lucas A, Fields
CD, Isner JM: Atherosclerotic Yucatan microswine: An animal mod-
el with high-grade, fibrocalcific, nonfatty lesions suitable for testing
catheter-based interventions. Am Heart J 1990;119:291–300
15. Reitman JS, Mahley RW, Fry DL: Yucatan miniature swine as a
model for diet-induced atherosclerosis. Atherosclerosis 1982;43:
119–132
16. Kuntz RE, Safian RD, Carrozza JP, Fishman RF, Mansour M, Baim
DS: The importance of acute luminal diameter in determining rest-
enosis after coronary atherectomy or stenting. Circulation 1992;
86:1827–1835
Clinical Implications
These results imply that intracoronary FGF-2 may be used
as an adjunct to PTCA or stent placement with the intention to
improve perfusion and function of ischemic areas that are not
directly targeted by the intervention. In the absence of adverse
effects on restenosis, a larger group of patients will benefit
from such a strategy.
Conclusions
We conclude that a single intracoronary administration of
rFGF-2 does not aggravate restenosis after balloon angioplas-
ty or stent placement in a relevant model of coronary athero-
sclerosis. The rFGF-2 affects neither neointima formation nor
remodeling to a clinically significant extent. This opens possi-
bilities for the implementation of trials in which angioplasty is
combined with rFGF-2 angiogenic therapy.
17. Harada K, Friedman M, Lopez JJ, Wang SY, Li J, Prasad PV,
Pearlman JD, Edelman ER, Sellke FW, Simons M: Vascular en-
dothelial growth factor administration in chronic myocardial is-
chemia. Am J Physiol 1996;270:H1791–1802
18. Hariawala MD, Horowitz JR, Esakof D, Sheriff DD, Walter DH,
Keyt B, Isner JM, Symes JF: VEGF improves myocardial blood
flow but produces EDRF-mediated hypotension in porcine hearts.
J Surg Res 1996;63:77–82
References
19. Banai S, Jaklitsch MT, Casscells W, Shou M, Shrivastav S, Correa
R, Epstein SE, Unger EF: Effects of acidic fibroblast growth factor
on normal and ischemic myocardium. Circ Res 1991;69:76–85
20. Lopez JJ, Edelman ER, Stamler A, Hibberd MG, Prasad P, Thomas
KA, DiSalvo J, Caputo RP, Carrozza JP, Douglas PS, Sellke FW,
Simons M: Angiogenic potential of perivascularly delivered aFGF
in a porcine model of chronic myocardial ischemia. Am J Physiol
1998;274:H930–936
21. Harada K, Grossman W, Friedman M, Edelman ER, Prasad PV,
Keighley CS, Manning WJ, Sellke FW, Simons M: Basic fibroblast
growth factor improves myocardial function in chronically is-
chemic porcine hearts. J Clin Invest 1994;94:623–630
22. Lopez JJ, Edelman ER, Stamler A, Hibberd MG, Prasad P, Caputo
RP, Carrozza JP, Douglas PS, Sellke FW, Simons M: Basic fi-
broblast growth factor in a porcine model of chronic myocardial
ischemia: A comparison of angiographic, echocardiographic and
coronary flow parameters. J Pharmacol Exp Ther 1997;282:
385–390
23. Giordano FJ, Ping P, McKirnan MD, Nozaki S, DeMaria AN,
Dillmann WH, Mathieu-Costello O, Hammond HK: Intracoronary
gene transfer of fibroblast growth factor-5 increases blood flow and
contractile function in an ischemic region of the heart (see com-
ments). Nat Med 1996;2:534–539
1. Isner JM, Asahara T: Angiogenesis and vasculogenesis as thera-
peutic strategies for postnatal neovascularization. J Clin Invest
1999;103:1231–1236
2. Laham R, Leimbach M, Chronos N, Vansant J, Pearlman J, Petti-
grew R, Guler H, Whitehouse M, Hung D, Baim D, King S, Simons
M: Intracoronary administration of recombinant fibroblast growth
factor-2 (rFGF-2) in patients with severe coronary artery disease:
Results of Phase I (abstr). J Am Coll Cardiol 1999;33:383A
3. Ware JA, Simons M: Angiogenesis and Cardiovascular Disease, p.
360. New York: Oxford University Press, 1999
4. Ware JA, Simons M: Angiogenesis in ischemic heart disease. Nat
Med 1997;3:158–164
5. Henry T, Annex B, Azrin M, McKendall G, Willerson J, Hendel R,
Giordano F, Klein R, Gibson M, Berman D, Luce C, McCluskey E:
Final results of the VIVA trial of rhVEGF Human Therapeutic
Angiogenesis. Circulation 1999;100:I-476
6. Laham RJ, Chronos NA, Pike M, Leimbach ME, Udelson JE,
Pearlman JD, Pettigrew RI, Whitehouse MJ, Yoshizawa C, Simons
M: Intracoronary basic fibroblast growth factor (FGF-2) in patients
with severe ischemic heart disease: Results of a phase I open-label
dose escalation study (in process citation). J Am Coll Cardiol
2000;36:2132–2139
7. Schumacher B, Pecher P, von Specht BU, Stegmann T: Induction of
neoangiogenesis in ischemic myocardium by human growth fac-
tors: First clinical results of a new treatment of coronary heart dis-
ease. Circulation 1998;97:645–650
24. Devlin GP, Fort S, Yu E, Cusimano RJ, Wei K, Rakowski H,
Butany J, Richards JA, Sole MJ, Schwartz L: Effect of a single bo-
lus of intracoronary basic fibroblast growth factor on perfusion in
an ischemic porcine model. Can J Cardiol 1999;15:676–682

278
Clin. Cardiol. Vol. 25, June 2002
25. Laham RJ, Sellke FW, Edelman ER, Pearlman JD, Ware JA, Brown
DL, Gold JP, Simons M: Local perivascular delivery of basic fi-
broblast growth factor in patients undergoing coronary bypass
surgery: Results of a phase I randomized, double-blind, placebo-
controlled trial. Circulation 1999;100:1865–1871
31. Rutherford C, Martin W, Salame M, Carrier M, Anggard E, Ferns
G: Substantial inhibition of neo-intimal response to balloon injury
in the rat carotid artery using a combination of antibodies to plate-
let-derived growth factor-BB and basic fibroblast growth factor.
Atherosclerosis 1997;130:45–51
26. Udelson JE, Dilsizian V, Laham RJ, Chronos N, Vansant J, Blais M,
Galt JR, Pike M, Yoshizawa C, Simons M: Therapeutic angiogene-
sis with recombinant fibroblast growth factor-2 improves stress and
rest myocardial perfusion abnormalities in patients with severe
symptomatic chronic coronary artery disease (in process citation).
Circulation 2000;102:1605–1610
27. de Smet BJ, Pasterkamp G, van der Helm YJ, Borst C, Post MJ:
The relation between de novo atherosclerosis remodeling and an-
gioplasty-induced remodeling in an atherosclerotic Yucatan mi-
cropig model. Arterioscler Thromb Vasc Biol 1998;18:702–707
28. Post MJ, de Smet BJ, van der Helm Y, Borst C, Kuntz RE: Arterial
remodeling after balloon angioplasty or stenting in an atheroscle-
rotic experimental model. Circulation 1997;96:996–1003
32. Chen C, Mattar SG, Hughes JD, Pierce GF, Cook JE, Ku DN,
Hanson SR, Lumsden AB: Recombinant mitotoxin basic fibroblast
growth factor-saporin reduces venous anastomotic intimal hyper-
plasia in the arteriovenous graft. Circulation 1996;94:1989–1995
33. Kumamoto M, Nakashima Y, Sueishi K: Intimal neovasculariza-
tion in human coronary atherosclerosis: Its origin and pathophysio-
logical significance. Hum Pathol 1995;26:450–456
34. Kennedy SH, Rouda S, Qin H, Aho S, Selber J, Tan EM: Basic
FGF regulates interstitial collagenase gene expression in human
smooth muscle cells. J Cell Biochem 1997;65:32–41
35. Post MJ, Borst C, Kuntz RE: The relative importance of arterial re-
modeling compared with intimal hyperplasia in lumen renarrowing
after balloon angioplasty. A study in the normal rabbit and the hy-
percholesterolemic Yucatan micropig (see comments). Circulation
1994;89:2816–2821
29. Bjornsson TD, Dryjski M, Tluczek J, Mennie R, Ronan J, Mellin
TN, Thomas KA: Acidic fibroblast growth factor promotes vascu-
lar repair. Proc Natl Acad Sci U S A 1991;88:8651–8655
36. Simons M, Post MJ: Coronary artery disease: Vascular endothelial
growth factor and fibroblast growth factor. Curr Interv Cardiol Rep
2001;3:185–191
30. Lindner V, Majack RA, Reidy MA: Basic fibroblast growth factor
stimulates endothelial regrowth and proliferation in denuded arter-
ies. J Clin Invest 1990;85:2004–2008