Preventive Medicine 49 (2009) 101–107

Contents lists available at ScienceDirect

Preventive Medicine
j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y p m e d

Aged garlic extract supplemented with B vitamins, folic acid and L-arginine retards
the progression of subclinical atherosclerosis: A randomized clinical trial
Matthew J. Budoff a,⁎, Naser Ahmadi a, Khawar M. Gul a, Sandy T. Liu a, Ferdinand R. Flores a, Jima Tiano a,
Junichiro Takasu a, Elizabeth Miller b, Sotirios Tsimikas b
a
Division of Cardiology, Los Angeles Biomedical Research Institute at Harbor-UCLA, Torrance, CA, USA
b
Division of Cardiology, Vascular Medicine Program, University of California San Diego, CA, USA

a r t i c l e i n f o a b s t r a c t

Available online 30 June 2009 Objectives. Previous studies demonstrated that aged garlic extract reduces multiple cardiovascular risk
factors. This study was designed to assess whether aged garlic extract therapy with supplements (AGE + S)
Keywords: favorably affects inflammatory and oxidation biomarkers, vascular function and progression of athero-
Progression
sclerosis as compared to placebo.
Atherosclerosis
Methods. In this placebo-controlled, double-blind, randomized trial (conducted 2005–2007), 65
Cardiac CT
Coronary artery calcium
intermediate risk patients (age 60 ± 9 years, 79% male) were treated with a placebo capsule or a capsule
Complementary medicine containing aged garlic extract (250 mg) plus Vitamin B12 (100 μg), folic acid (300 μg), Vitamin B6 (12.5 mg)
Randomized clinical trial and L-arginine (100 mg) given daily for a 1 year. All patients underwent coronary artery calcium scanning
(CAC), temperature rebound (TR) as an index of vascular reactivity using Digital Thermal Monitoring (DTM),
and measurement of lipid profile, autoantibodies to malondialdehyde (MDA)-LDL, apoB-immune complexes,
oxidized phospholipids (OxPL) on apolipoprotein B-100 (OxPL/apoB), lipoprotein (a) [Lp (a)], C-reactive
protein (CRP), homocysteine were measured at baseline and 12 months. CAC progression was defined as an
increase in CAC N 15% per year and an increase in TR above baseline was considered a favorable response.
Results. At 1 year, CAC progression was significantly lower and TR significantly higher in the AGE + S
compared to the placebo group after adjustment of cardiovascular risk factors (p b 0.05). Total cholesterol,
LDL-C, homocysteine, IgG and IgM autoantibodies to MDA-LDL and apoB-immune complexes were
decreased, whereas HDL, OxPL/apoB, and Lp (a) were significantly increased in AGE + S to placebo.
Conclusion. AGE + S is associated with a favorable improvement in oxidative biomarkers, vascular
function, and reduced progression of atherosclerosis.
© 2009 Elsevier Inc. All rights reserved.

Background garlic extract has been demonstrated to improve vascular function

(Okuhara, 1994), inhibit endothelial cell damage and transform smooth
Several clinical reports, including meta-analyses, have revealed muscle cells (Ho et al., 2001), suggesting that AGE may have an effect of
cholesterol-lowering effects of garlic supplementation in humans (Neil controlling arterial function through inhibiting the damage of nitric
et al., 1996; Warshafsky et al., 1993). Such reports have additionally oxide synthesis (Morihara et al., 2002).
impacted public awareness on the potential cardiovascular benefits of A variety of oxidative biomarkers have been developed and valida-
garlic. Garlic and garlic extracts have been postulated to impart ted to reflect the presence of oxidation-specific epitopes on lipopro-
cardiovascular benefits through multiple mechanisms. For example, teins (Tsimikas et al., 2006). These biomarkers can be direct,
recent studies of aged garlic extract (AGE) have shown it as a modulator measuring specific epitopes such as oxidized phospholipids (OxPL)
of multiple cardiovascular risk factors, in addition to LDL lowering or malondialdehyde (MDA) epitopes on apolipoprotein B-100 or lipo-
(Dillon et al., 2003), including lowering blood pressure, reducing protein (a), on indirect, measuring autoantibodies and immune
platelet aggregation and adhesion, preventing LDL oxidation, smoking- complexes to oxidized LDL (Choi et al., 2008; Holvoet et al., 2001;
caused oxidative damage, and also directly suppressing atherosclerosis Itabe and Ueda, 2007). These measures have been shown to reflect
(Steiner et al.,1996; Campbell et al., 2001; Okuhira et al., 2000; Rahman various manifestations of anatomical cardiovascular disease and to
and Billington, 2000; Gonen et al., 2005; Lau, 2001). In addition, aged predict new cardiovascular events (Fraley and Tsimikas, 2006). More
recently, active investigation is being pursued in assessing the role of
⁎ Corresponding author. Division of Cardiology, Harbor-UCLA Medical Center, 1124
such biomarkers following therapeutic interventions (Tsimikas,
West Carson Street, Torrance, CA, USA. Fax: +1 310 787 0448. 2007a). In particular, the OxPL/apoB measure, which detects OxPL
E-mail address: [email protected] (M.J. Budoff). on individual apoB particles, but not total OxPL in plasma, has been

0091-7435/$ – see front matter © 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.ypmed.2009.06.018
102 M.J. Budoff et al. / Preventive Medicine 49 (2009) 101–107

shown to increase in patients following therapeutic interventions Study population

including statin therapy (Choi et al., 2008; Ky et al., 2008; Rodenburg
et al., 2006; Tsimikas et al., 2004a,b), as well as in patients (Choi et al., We enrolled 65 asymptomatic patients (age 60± 9 years, 79% male) with
CAC N 30, intermediate to high risk for coronary artery disease, with a 10-year
2008; Rodenburg et al., 2006; Silaste et al., 2004) and animals
Framingham risk of developing coronary artery disease (Wilson et al., 1998) of
following low fat diets (Tsimikas et al., 2007a,b). Thus, it may serve as a
10–20%. Subjects with established cardiovascular disease, stroke, diabetic
new biomarker of early benefit of therapeutic interventions. retinopathy, peripheral vascular disease, underlying infection, cancer, systemic
The effect of garlic on coronary atherosclerosis progression in inflammation status, immunosuppression, end-stage renal or liver disease,
humans has not been widely studied (Budoff et al., 2006). In addition, creatinine N 1.4 mg/dl, triglycerides N 400 mg/dl, known hypersensitivity to
the effect of garlic on oxidative biomarkers and their relationship to garlic therapy; weight N 300 lb; drug or alcohol abuse, partial ileal bypass or
coronary calcium and vascular function is not known. This randomized, known gastrointestinal disease limiting drug absorption; and patients currently
placebo-controlled study was designed to assess the effect of aged garlic using garlic supplements were excluded.
extract plus homocysteine-lowering vitamins and L-arginine on cor- All the patients signed informed written consent after careful explanation
onary calcium, vascular function and lipid and oxidative biomarkers. and review of the protocol. Patients underwent CAC, Digital Thermal
Monitoring (DTM) of vascular function assessment and oxidative biomarker
measurement at baseline and after 12 months (04/01/2006 to 04/02/2007).
Methods They were randomized to garlic plus supplements or placebo in a double-
blinded manner, using numbered containers, assigned to a computer
The study is a placebo-controlled, double-blind, randomized trial (con- generated randomization chart by the nurse coordinator. All participants,
ducted 2005–2007) to determine whether commercially available aged garlic personnel administering interventions and physicians involved in the study
extract plus B vitamins, L-arginine and folate (AGE+ S) (Kyolic 108, Wakunaga were blinded to group assignment. A capsule of placebo or commercially
Nutritional Supplement, CA, USA) can influence the rate of atherosclerotic available AGE (250 mg) plus vitamin B12 (100 μg), folic acid (300 μg), vitamin
plaque burden measured by coronary artery calcium, improve vascular B6 (12.5 mg) and L-arginine (100 mg) was given daily for the duration of
function and favorably change biomarkers of oxidative stress. 1 year. All participants were educated on a low-cholesterol diet and instructed
The CONSORT (Consolidated Standards of Reporting Trials) statement is to avoid any direct form of garlic and antioxidant supplementation.
outlined in Fig. 1 (Moher et al., 2001). Aged Garlic Extract (AGE, Kyolic®), Body mass index, hip circumference, blood pressure, fasting blood
provided by Wakunaga of America (Mission Viejo, CA), was formulated by glucose, high sensitivity C-reactive protein (CRP), homocysteine, and lipid
soaking sliced raw garlic in aqueous ethanol for up to 20 months at room profile were obtained quarterly by standard techniques. All subjects were on
temperature. The extract was then filtered and concentrated at low temperature. background statin therapy as prescribed by their primary physician.
The AGE used in this trial contained 305 g/l of extracted solids. AGE contains After randomization, participants returned at 3, 6, 9 and 12 months to
primarily water-soluble organosulfur compounds such as S-allylcysteine and assess their compliance with medication. 58 patients completed the study
S-allylmercaptocysteine and small amounts of oil-soluble organosulfur com- protocol. Drops outs were due to 3 patients relocating, and 4 patients with-
pounds (Budoff et al., 2006). The finished product (Kyolic) used in this clinical drawing consent during the treatment period.
study was commercially available and provided by Wakunaga of America Co., Ltd
(Mission Viejo, CA). The Investigational Review Board of Los Angeles Biomedical Demographics and laboratory measures
Research Institute at Harbor-UCLA approved this research project. The trial was
initiated prior to requirements for registering trials at clinicaltrials.gov. Information about age, gender, ethnicity, and medical history were
obtained by questionnaires. Current smoking was defined as having smoked
a cigarette in the last 30 days and quantitated as pack years. Alcohol use was
defined as never, former, or current. Diabetes was defined as a fasting glucose
N 126 mg/dl or use of hypoglycemic medications. Use of antihypertensive and
other medications were based on clinical staff entry of prescribed medications.
Resting blood pressure was measured two times in sitting position after a
5 min rest period using a Dinamap model Pro 100 automated oscillometric
sphygmomanometer (Critikon, Tampa, Florida) and the average of the
readings was recorded. Hypertension was defined as a systolic blood pressure
≥ 140 mm Hg, diastolic blood pressure ≥ 90 mm Hg, or use of medication
prescribed for hypertension. Body mass index was calculated from the
equation weight (kg)/height2 (m2).
Total and HDL cholesterol were measured from blood samples obtained
after a 12-hour fast using the cholesterol oxidase cholesterol method and
triglycerides measured from spectrophotometric reading after endogenous
glycerol has reacted and before lipase is added to release the glycerol from the
triglyceride. LDL cholesterol is calculated in plasma specimens having a
triglyceride value b 400 mg/dl using the formula of Friedewald et al. (1972).
CRP was measured using the BNII nephelometer (N High Sensitivity CRP;
Dade Behring Inc., Deerfield, IL). Analytical intra-assay CVs ranged from 2.3–4.4%
and inter-assay CVs ranged from 2.1–5.7%. S-allylcysteine (SAC) was measured
in serum as previously described (Rosen et al., 2001) to determine compliance
with garlic therapy.

Measures of coronary artery calcium

The electron beam tomography (EBT) studies were performed with an

Imatron C-150XL Ultrafast computed tomographic scanner (Imatron, South San
Francisco, California) in high-resolution volume mode using a 100-ms exposure
time as previously described (28). Electrocardiographic triggering was used so
that each image was obtained at the same point in diastole, corresponding to
40% of the RR interval. Proximal coronary artery was obtained with at least 30
consecutive non-enhanced images at 3-mm intervals. Total radiation exposure
using this technique was b1 Rad per patient. Patients underwent similar repeat
examination at the end of the study protocol. In the acquired images from non-
Fig. 1. Flow chart for patient enrollment and follow up. enhancement, calcification was defined as a plaque of at least 3 contiguous
 M.J. Budoff et al. / Preventive Medicine 49 (2009) 101–107 103

pixels (area 1.02 mm2) with a density of N 130 Hounsfield units. The amount of the groups with group sizes of at least 20 participants per study. To
calcium was quantified using 2 different methods. First, the calcium score as accommodate for dropout, we enrolled 65 patients, with final retention of
described by Agatston et al. (1990) was calculated by multiplying the lesion 58 participants. All continuous data are presented as a mean value ± SD, and
area by a coefficient based on the peak density within that plaque. The total all categorical data are reported as a percentage or absolute number. Student's
Agatston score was determined by summing the scores obtained for each lesion. t tests and Chi-square tests were used to assess differences between groups.
The voxel size on EBT scans obtained with a 3-mm section thickness and a Comparisons of all parameters between the active therapy and placebo were
typical field of view of 30 cm2 (pixel size= 0.586 mm) corresponded to made with the Student's t test. Trends of all parameters during follow-up
0.586 × 0.586 × 3 = 1.03 mm3. The original voxel was divided into the smaller periods were examined using matched pair t tests. The association between
voxels whose size is 0.201 mm3 with the technique of isotropic interpolation for the change of CAC, lipids, and endothelial function were made with
measuring a more precise volume. All voxels with a value greater than 130 correlation coefficients. The associations between changes in the two
Hounsfield units were defined as the calcified lesion. A total score in coronary treatment groups over 1 year between groups (active therapy and placebo)
artery by the Agatston and volumetric methods was determined by summing for risk factors, including lipid profile, oxidation biomarkers, TR, and CAC were
individual lesion scores from each of 4 anatomic sites (left main, left anterior analyzed by logistic regression analyses. These analyses were adjusted for
descending, circumflex, and right coronary arteries). demographics, age, gender, and traditional cardiac risk factors. The results are
A single experienced investigator (JK), blinded to clinical status of the reported as odds ratios (OR) for the logistic regression. Odds ratios were
participant and temporal relation of the scans, interpreted all studies on a calculated for the highest vs. 2 lower tertiles of changes in temperature
commercially available software package (Neo Imagery Technologies, City of rebound (TR), LDL-C, HDL-C, Triglyceride, total cholesterol, homocysteine,
Industry, California). Patients were classified as CAC progressors if the CAC CRP, Lipoprotein (a), OxPL/apoB, IgG and IgM autoantibodies to MDA-LDL and
increased N15% over the one year study period, and as CAC non-progressors, if IC/apoB.
progression was b 15%, as previously described (Raggi et al., 2004).
Results
Digital Thermal Monitoring (DTM) of vascular function

After an overnight fast and abstinence from tobacco, alcohol, caffeine, and Patient characteristics
vasoactive medications, the left arm blood pressure was recorded in a sitting
position 15 min before the DTM test (Omron HEM 705 CP semi-automated The baseline clinical characteristics, lipid variables and oxidation
sphygmomanometer, Bannockburn, IL, USA). After resting in a supine position in biomarkers including OxPL/apoB, Lp (a), autoantibodies to MDA-LDL
a room with temperature 23° to 25 °C for 30 min, DTM of both hands was and immune complexes, as well as TR of vascular function and CAC
obtained during 3 min stabilization; 5 min cuff inflation to 50 mm Hg greater were similar between AGE + S and placebo groups (Tables 1 and 2).
than systolic blood pressure, and 5 min deflation using an automated, operator-
independent protocol (VENDYS-5000, Endothelix Inc., Houston, TX). DTM The effect of aged garlic extract and supplement therapy in CAC
thermal probes, designed to minimize the area of skin probe contact and
progression
fingertip pressure, were attached to the pulp of the index finger, and thermal
changes were traced continuously and digitalized automatically using VENDYS
software (a computer based thermometry system with 0.006 °C thermal After one year follow up, the volumetric CAC score increased from
resolution and an automated compressor for measurement of blood pressure 347 ± 67 to 439 ± 77 in the placebo group compared to 291 ± 50 to
and controlled occlusion hyperemia). 311 ± 48 in the AGE + S group, p = 0.03 for comparison of final scores
Using the cuff occlusive reactive hyperemia procedure, temperature between groups. After one year follow up, the AGE + S group had
rebound (TR), calculated as [post cuff-deflation maximum temperature significantly less CAC progression compared to the placebo group
− baseline temperature] was measured an index of vascular reactivity. The (6.8% vs. 26.5%, p = 0.005, Table 2). In multivariable analysis adjusting
coefficient of variation (CV) after a 1 day interval measured by Bland Altman for age, gender, hypertension, hypercholesterolemia, statin therapy,
model for the TR was 2.4% (Ahmadi et al., 2009).
diabetes, family history of premature CHD and smoking status, CAC
Determination of oxidative biomarkers: OxPL/apoB, IgG and IgM
progression was significantly lower in the aged garlic extract group
IC/apoB, and IgG and IgM MDA-LDL autoantibody titers compared to the placebo group (OR: 0.35, 95% CI 0.1–0.85, p = 0.03).

Oxidized phospholipids on apolipoprotein B-100 particles (OxPL/apoB) The effect of aged garlic extract and supplement therapy in vascular
were measured as described in detail previously (Tsimikas et al., 2005, function
Tsimikas et al., 2004a,b) by chemiluminescent enzyme-linked immunosor-
bent assay using the murine monoclonal antibody E06 which binds to the The mean temperature rebound (TR) index at baseline was
phosphorylcholine head group of oxidized but not native phospholipids. In 0.57 and 0.56 on active therapy and placebo, respectively (p = 0.9)
brief, a 1:50 dilution of plasma was added to microtiter wells coated with the
(Table 2). TR significantly increased in the AGE + S group compared
murine monoclonal antibody MB47 (5 μg/ml), which specifically binds apoB.
with placebo (138% vs. 26.7%, p = 0.001).
Under these conditions, a saturating amount of apoB is added to each wall,
and consequently, equal numbers of apoB particles were captured in each
well for all samples. The content of OxPL/ apoB is then determined with
biotinylated E06. The data are presented as relative light units per 100 ms; the
OxPL/apoB measurement is independent of apoB (and LDL cholesterol) levels Table 1
(Tsimikas et al., 2004a,b, 2006) and reflects OxPL on individual apoB particles, Baseline demographic and clinical characteristics of study subjects.
but not total OxPL in plasma.
Variable Garlic + S Placebo p value
Chemiluminescence enzyme-linked immunosorbent assays were also
(N = 33) (N = 32)
used to measure IgG and IgM autoantibodies to MDA-LDL and apoB-IC as
Age in years — mean (S.D.) 60 (8) 61 (10) 0.40
described previously (Tsimikas et al., 2007a,b). Plasma lipoprotein (a) [Lp (a)]
Gender (% Male) 26 (79%) 25 (78%) 0.90
levels were measured by a chemiluminescent enzyme-linked immunosorbent
Hypertension (%) 18 (55%) 12 (38%) 0.20
assay with monoclonal antibody LPA4, as described previously (Tsimikas et al., Antihypertensive medication (%) 31 (94%) 32 (100) 0.85
2004a,b). Hypercholesterolemia (%) 24 (73%) 26 (81%) 0.20
Statin therapy (%) 33 (100%) 32 (100%) 1.00
Statistical analysis Diabetes mellitus (%) 1 (3%) 2 (6%) 0.70
Current smoker (%) 11 (33%) 12 (38%) 0.70
All statistical analyses were performed using SPSS V 15.0 (SPSS Institute, Family history of CVD (%) 21 (64%) 23 (72%) 0.50
Chicago, IL). Initial power analysis was based upon the results of a our Framingham risk score 10 year risk (S.D.) 13 (5) 15 (7) 0.70
previous placebo double-blind aged garlic extract feasibility study (Budoff Sample sizes in parenthesis, S.D. = standard deviation.
et al., 2004). This study had over an 80% power to detect differences between CVD = Cardiovascular disease; S = supplements.
104 M.J. Budoff et al. / Preventive Medicine 49 (2009) 101–107

Table 2
Comparison between cardiovascular risk factors among aged garlic extract (+ supplements) to placebo groups at baseline and 1 year.

Variable Baseline Follow up Mean annual changes (%)

Garlic + S Placebo p Garlic + S Placebo p Garlic + S Placebo p
N = 33 N = 32 N = 30 N = 28
Coronary artery calcification 291 (50) 347 (67) 0.3 311 (48) 439 (77) 0.03 6.8 26.5 0.005
Total cholesterol, mg/dl 180.9 (41) 180.6 (31) 0.9 165 (29) 182 (30) 0.03 − 8.8 0.02 0.008
HDL-C, mg/dl 46.9 (9) 46.7 (12) 0.9 54.3 (7.8) 49.6 (8.9) 0.03 15.4 6.4 0.01
LDL-C, mg/dl 104 (34) 112 (28) 0.3 90.1 (29) 113 (27) 0.05 − 13.6 0.01 0.009
Triglyceride, mg/dl 109.3 (18) 105 (19) 0.8 89.9 (32) 94.2 (36) 0.36 − 17.8 − 10 0.08
Temperature rebound 0.57 (0.2) 0.56 (0.15) 0.9 1.36 (0.2) 0.71 (0.2) 0.04 138 26.7 0.001
Homocysteine, mg/dl 10.1 (2.4) 10.7 (2.2) 0.3 8.3 (2.7) 10.6 (2.6) 0.02 − 17.8 0.01 0.004
CRP, mg/L 0.31 (40) 0.22 (0.33) 0.4 0.21(0.42) 0.13 (0.37) 0.9 − 41 − 40.9 0.9
Lp (a), mg/dl 13.2 (10.7) 20.2 (19) 0.1 23.6 (9.8) 25.6 (14) 0.1 79 28 0.001
OxPL/apoB, RLU 4139 (913) 3231 (719) 0.4 6397 (522) 4419 (315) 0.008 54 36 0.001
IgG MDA-LDL, RLU 4054 (353) 3745 (271) 0.8 2544 (635) 3650 (410) 0.009 − 37 − 2.5 0.001
IgM MDA-LDL, RLU 2004 (206) 1800 (234) 0.3 1116 (2845) 2264 (117) 0.001 − 44 25 0.008
IgG IC/apoB, RLU 4840 (530) 4668 (392) 0.5 2179 (726) 4551 (663) 0.001 − 55 − 2.5 0.001
IgM IC/apoB, RLU 2553 (306) 2101 (258) 0.5 1845 (358) 1856 (389) 0.3 − 28 − 11.7 0.001

HDL = high density lipoprotein; LDL = low density lipoprotein; Lp (a) = lipoprotein (a); RLU = relative light units; S = supplements.
Mean (standard deviation).

The effect of aged garlic extract and garlic supplement therapy on Compliance with garlic therapy
lipid parameters, oxidative biomarkers, Lp (a), homocysteine
and hsCRP SAC serum measurement is currently the only reliable human
compliance marker used for studies involving garlic consumption
After one year follow up, total cholesterol (180.9 to 165 mg/dl, since it is detectable and quantitatively increases in the blood after
−8.8%, p = 0.008) and LDL-C (104 to 90 mg/dl, − 13.6%, p = 0.009) oral intake of garlic capsules (Rosen et al., 2001; Steiner and Lin, 1998).
decreased significantly, while HDL-C increased significantly (46.9 to The average serum SAC level was significantly higher in patients
54.3 mg/dl, 15.4%, p = 0.01) in the AGE + S group. Lipid values did not randomized to aged garlic therapy as compared to those randomized
significantly change in the placebo group over 1 year (Table 2). to placebo, (36 parts per billion (ppb) vs. 15 ppb, p b 0.05).
After adjustment for cardiovascular risk factors, IgG and IgM
autoantibodies to MDA-LDL and IC/apoB levels decreased significantly Discussion
in the AGE + S group compared to placebo (Fig. 2 and Table 3). In
contrast, OxPL/apoB and Lp (a) increased significantly more in the The current study demonstrates that aged garlic extract supple-
aged garlic extract and supplement group compared to the placebo mented with vitamins B6 and B12, folic acid and L-arginine was
group (Fig. 3 and Table 3). There was no association between aged associated with not only decreased CAC progression but also with
garlic therapy use and CRP (Table 3). concomitant improvement in vascular function following 1 year of
treatment of asymptomatic intermediate risk patients. In addition,
evidence of favorable effects on oxidative biomarkers were noted with
increased levels of OxPL/apoB and Lp (a), which have been previously
noted to increase in response to statins, low fat diets and athero-

Table 3
Odds ratios of progression of coronary artery calcification, vascular function and changes in
lipid and oxidative biomarkers in AGE+ S vs. placebo after 1 year of treatment.

Outcome Aged garlic extract p value

OR (95% CI)
CAC progression 0.35 (0.1–0.85) 0.03
TR increase 2.21(1.12–4.9) 0.04
LDL-C 0.87 (0.71–0.90) 0.04
HDL-C 1.09 (1.03–1.13) 0.04
Triglyceride 0.97 (0.33–3.8) 0.60
Total cholesterol 0.92(0.81–0.98) 0.04
Homocysteine 0.87 (0.46–0.96) 0.03
C-reactive protein 0.92(0.35–3.5) 0.40
Lipoprotein (a) 1.78 (1.1–2.2) 0.02
OxPL/apoB 1.30 (1.1–1.8) 0.01
IgG MDA-LDL 0.74 (0.16–0.87) 0.02
IgM MDA-LDL 0.80 (0.1–0.9) 0.02
IgG IC/apoB 0.82 (0.05–0.9) 0.03
IgM IC/apoB 0.78 (0.1–0.9) 0.02

Logistic regression model adjusted for age, gender, DM, Hypertension, Hypercholes-
terolemia, Family History of CHD, Smoking status, Statin therapy.
Abbreviations: OR = Odds ratio, CI = Confidence intervals, LDL = low density
lipoprotein, HDL = high density lipoprotein, CAC = coronary artery calcium.
Odds ratio of CAC progression (increase in CAC ≥ 15%/year) vs. CAC non-progression.
Odds ratio of highest vs. 2 lower tertiles of temperature rebound (TR), LDL-C, HDL-C,
Fig. 2. Percent changes in CAC, biomarkers of oxidative stress and homocysteine (mean±SD) Triglyceride, Total Cholesterol, Homocysteine, C-reactive protein, Lipoprotein (a), OxPL/
from baseline to one year in the aged garlic extract and placebo groups. apoB, IgG and IgM autoantibodies to MDA-LDL and IC/apoB.
 M.J. Budoff et al. / Preventive Medicine 49 (2009) 101–107 105

We have also demonstrated that increased homocysteine levels are

associated with increased plaque progression (Rasouli et al., 2005).
Our previous study of aged garlic extract alone demonstrated signi-
ficant reduction in homocysteine (Budoff et al., 2004). The current
study reconfirms our previous work and demonstrated more reduc-
tion in homocysteine with the combination of aged garlic extract and
the supplements. Ide et al. (2006) studied the effects of homocysteine
on CD36 expression and foam cell formation in human monocytes/
macrophages (THP-1) using flow cytometry and also evaluated the
impact of aged garlic therapy on this process. They demonstrated that
CD36 expression increased with incubation of THP-1 cells with ho-
mocysteine, also AGE therapy inhibits CD36 expression and OxLDL
uptake in macrophages and suggested that AGE could modulate the
formation of early atherosclerosis.
In addition to garlic and homocysteine-lowering vitamins, active
therapy included arginine supplementation. Arginine has been shown to
increase NO bioavailability, which has been shown to reduce vascular
dysfunction which underlies vascular stiffness, independent of other
age-related vascular pathologies such as atherosclerosis. The activation/
upregulation of arginase appears to be an important contributor to age-
related vascular dysfunction by a mechanism that involves substrate
(L-arginine) limitation for NOS3 and therefore NO synthesis (NOS)
(Briley-Saebo et al., 2008). The activation/upregulation of arginase
Fig. 3. Percent changes in vascular function measured by temperature rebound, OxPL/ impairs NO production, also enhances production of reactive oxygen
apoB and Lp (a) (mean ± SD) from baseline to one year in the aged garlic extract and species by NOS. While arginase abundance is increased in vascular aging
placebo groups.
models, it appears that post-translational modification by S-nitrosylation
of the enzyme enhances its activity as well (Tsimikas, 2008). Further-
sclerotic plaque regression. Paralleling this were decreased levels of more, arginase activation contributes to aging-related vascular changes
indirect measures of oxidative stress as measured by autoantibodies to by mechanisms that are not directly related to changes in NO signaling,
MDA-LDL and apoB-immune complexes. including polyamine-dependent vascular smooth muscle proliferation
We previously studied the impact of aged garlic extract alone and collagen synthesis.
without the supplements evaluated in this study on the atherosclerotic The potent effect of aged garlic extract, B vitamins and L-arginine on
plaque burden measured by changes in CAC (Budoff et al., 2004). After a biomarkers of oxidative stress is a novel observation. In this study a
one year follow up, changes in CAC were less in aged garlic extract alone significant reduction in autoantibodies to MDA-LDL and apoB-immune
as compared to placebo group (7.5 ± 9.4% vs. 22.2± 18.5% and 45.2 ± complexes were noted in the AGE + S group compared to placebo.
57.2 vs. 129.0 ± 102 volumetric score, respectively (p b 0.05)). Our Previous studies have shown that OxLDL autoantibodies correlate with
current study reconfirms our previous findings and provides evidence both the progression and regression of atherosclerosis in LDLR−/− mice
of a larger decrease in the progression of CAC with the combination of in response to low fat diets and antioxidant vitamins (Tsimikas et al.,
aged garlic therapy plus vitamins and L-arginine. 2006). In addition, a recent study in apoE−/− mice with magnetic
CAC is an excellent marker of overall coronary atherosclerotic burden resonance using micelles loaded with oxidation-specific antibodies
(Budoff et al., 2006; Simons et al., 1992) correlating with both the total demonstrated OxLDL in the vessel wall (Briley-Saebo et al., 2008).
plaque burden and the presence of obstructive disease. In addition, there Furthermore, a recent study in humans has shown that autoantibodies
are numerous studies confirming the strong predictive power of CAC for to MDA-LDL and apoB-immune complexes are correlated with the
clinical events in asymptomatic patients (Budoff, 2006; Detrano et al., extent of angiographically determined coronary artery disease (Choi
2008; Budoff et al., 2007). Mechanisms which may explain a significant et al., 2008). Reduction in OxLDL biomarkers with concomitant reduced
decrease in the progression of CAC in AGE plus supplement group progression of CAC and improved vascular function in response to aged
include a variety of potential anti-atherosclerotic effects, as shown garlic extract plus B vitamins and L-arginine, is consistent with the
recently by studies which have shown aged garlic therapy as a mo- previous studies and suggests that part of the benefit may be mediated
dulator of cardiovascular risk factors, (Steiner and Lin, 1998) including through reduction of oxidative stress and oxidative pathways.
blood pressure (Dhawan and Jain, 2004), platelet aggregation and Consistent with prior data on a variety of statins and low fat diets,
adhesion, total cholesterol, low density lipoprotein (LDL), high density aged garlic extract plus supplement also induced an increase in OxPL/
lipoprotein (HDL), LDL oxidation, smoking-caused oxidative damages, apoB and Lp (a) levels (18). These data suggest that OxPL and Lp (a) are
and microcirculation. physically associated and that Lp (a) has a strong affinity for OxPL
A study investigated molecular mechanism of aged garlic extract (Bergmark et al., 2008). Previous studies postulated that Lp (a) may
reported that it directly suppressed atherosclerosis (Steiner and Lin, function at a primordial level to bind and perhaps mediate the removal
1998). In addition, inhibiting damage of the endothelial cells and of oxidized fatty acids from OxPL/apoB, particularly as it is enriched in
transforming smooth muscle cells by aged garlic extract suggest that this lipoprotein-associated phospholipase A2, an enzyme with the ability to
compound may have an effect of controlling vascular function through hydrolyze oxidized fatty acids from OxPL/apoB (Choi et al., 2008).
inhibiting the damage of nitric oxide (NO) synthesis (Morihara et al., Furthermore, previous studies have shown a 50–100% increase of plas-
2002). Our study demonstrated improvement in vascular function under ma OxPL/apoB levels in rabbits and monkeys, along with concomitant
the influence of garlic extract along with B vitamins, folic acid and L- evidence of loss of OxPL from the vessel wall, suggesting that as OxPL/
arginine, with an improvement in fingertip temperature rebound. apoB rises, OxPL epitopes disappear in the vessel wall in response to
Similarly, Williams et al. (2005), in a randomized, placebo-controlled therapeutic interventions (Miyazaki et al., 2005; Nakaishi et al., 2000).
trial on 15 men with angiographically proven CAD, flow mediated Finally, previous studies have shown that vascular dysfunction
vasodilation (FMD) through reactive hyperemia test increased signifi- measured by DTM correlates well with the severity and extent of
cantly in aged garlic extract (44%) as compared to the placebo (p = 0.04). coronary artery disease measured by CAC and CT angiography,
106 M.J. Budoff et al. / Preventive Medicine 49 (2009) 101–107

Framingham risk score, insulin resistance, metabolic syndrome and Ahmadi, N., Shim, J., Vasinrapee, P., et al., 2008b. Vascular dysfunction measured by digital
thermal monitoring is associated with myocardial perfusion defect. Circulation 117,
diabetes (Ahmadi et al., 2007, 2008a,b). Similar to statin therapies e409–e468.
(Choi et al., 2008; Davignon and Ganz, 2004), current study provides Ahmadi, N., Usman, N., Shim, J., et al., 2009. Vascular dysfunction measured by fingertip
evidence that aged garlic therapy plus supplement results in thermal monitoring is associated with the extent of myocardial perfusion defect.
J. Nucl. Cardiol.
significant increases in OxPL/apoB and lipoprotein (a) levels, Bergmark, C., Dewan, A., Orsoni, A., et al., 2008. A novel function of lipoprotein [a] as a
improves vascular function and decreases in apoB-immune complexes preferential carrier of oxidized phospholipids in human plasma. J. Lipid Res. 49,
and OxLDL autoantibodies. In addition suggests that favorable effect of 2230–2239.
Briley-Saebo, K.C., Shaw, P.X., Mulder, W.J., et al., 2008. Targeted molecular probes for
aged garlic therapy plus supplement with increases in OxPL/apoB, Lp imaging atherosclerotic lesions with magnetic resonance using antibodies that
(a), and vascular function may be associated with the reduction in the recognize oxidation-specific epitopes. Circulation 117, 3206–3215.
rate of CAC progression. Budoff, M., 2006. Aged garlic extract retards progression of coronary artery calcification.
J. Nutr. 136, 741S–744S.
Budoff, M.J., Takasu, J., Flores, F.R., et al., 2004. Inhibiting Progression of coronary
Limitations calcification using aged garlic extract in patients receiving statin therapy: a preli-
minary study. Prev. Med. 39, 985–991.
The present study has several limitations. In our study CAC Budoff, M.J., Achenbach, S., Blumenthal, R.S., et al., 2006. Assessment of coronary artery
disease by cardiac computed tomography, a scientific statement from the American
progression was defined as absolute changes in CAC as reported by Heart Association Committee on Cardiovascular Imaging and Intervention, Council
Raggi et al. (2004) as a minimum annualized percent change of CAC on Cardiovascular Radiology and Intervention, and Committee on Cardiac Imaging,
score ≥ 15%. Analyses of the progression of CAC are statistically Council on Clinical Cardiology. Circulation 114 (16), 1761–1791.
Budoff, M.J., Shaw, L.J., Liu, S.T., et al., 2007. Long-term prognosis associated with
challenging because of a high degree of dependence on the baseline coronary calcification: observations from a registry of 25,253 patients. J. Am. Coll.
CAC score as well as inter-scan variability. However, Raggi et al. Cardiol. 49, 1860–1870.
demonstrated a 17-fold increase in CVD events among progressors. Campbell, J.H., Efendy, J.L., Smith, N.J., Campbell, G.R., 2001. Molecular basis by which
garlic suppresses atherosclerosis. J. Nutr. 131, 1006S–1009S.
Furthermore, this study was not designed to assess which Choi, S.H., Chae, A., Miller, E., et al., 2008. Relationship between biomarkers of oxidized
components of the AGE + S capsule were most responsible for the low-density lipoprotein, statin therapy, quantitative coronary angiography, and
observed benefits. A much larger study randomizing patients to atheroma: volume observations from the REVERSAL (Reversal of Atherosclerosis
with Aggressive Lipid Lowering) study. J. Am. Coll. Cardiol. 52, 24–32.
individual components of the therapy would be needed. Both the Davignon, J., Ganz, P., 2004. Role of endothelial dysfunction in atherosclerosis. Circulation
placebo and active therapy groups were advised to adhere to a 109 (III), 27–32.
healthy lifestyle behaviors, but this was not formally assessed as to Detrano, R., Guerci, A.D., Carr, J.J., et al., 2008. Coronary calcium as a predictor of
coronary events in four racial or ethnic groups. N. Engl. J. Med. 358, 1336–1345.
adherence or independent effects on the endpoints. A randomized
Dhawan, V., Jain, S., 2004. Effect of garlic supplementation on oxidized low density
trial of diet and lifestyle would be needed to better assess these lipoproteins and lipid peroxidation in patients of essential hypertension. Mol. Cell.
potential interventions. Further studies addressing these relation- Biochem. 266, 109–115.
ships to clinical outcomes await results of large therapeutic studies Dillon, S.A., Burmi, R.S., Lowe, G.M., Billington, D., Rahman, K., 2003. Antioxidant
properties of aged garlic extract: an in vitro study incorporating human low density
in progress. lipoprotein. Life Sci. 72 (14), 1583–1594.
Fraley, A.E., Tsimikas, S., 2006. Clinical applications of circulating oxidized low-density
Conclusion lipoprotein biomarkers in cardiovascular disease. Curr. Opin. Lipidol. 17, 502–509.
Friedewald, W.T., Levy, R.I., Fredrickson, D.S., 1972. Estimation of the concentration of
low-density lipoprotein cholesterol in plasma, without use of the preparative
This study demonstrates that commercially available aged garlic ultracentrifuge. Clin. Chem. 18, 499–502.
extract with B vitamins, folate and L-arginine: 1) retards progression Gonen, A., Harats, D., Rabinkov, A., et al., 2005. The antiatherogenic effect of allicin:
possible mode of action. Pathobiology 72, 325–334.
of CAC; 2) improves vascular function measured by DTM; 3) favorably Holvoet, P., Mertens, A., Verhamme, P., et al., 2001. Circulating oxidized LDL is a useful
influences markers of oxidative stress. Further longitudinal studies are marker for identifying patients with coronary artery disease. Arterioscler. Thromb.
warranted to evaluate whether these dietary supplements can Vasc. Biol. 21, 844–848.
Ho, S.E., Ide, N., Lau, B.H., 2001. S-allyl cysteine reduces oxidant load in cells involved in
decrease long term adverse cardiovascular events, particularly in the atherogenic process. Phytomedicine 8, 39–46.
high-risk patients. Ide, N., Keller, C., Weiss, N., 2006. Aged garlic extract inhibits homocysteine-induced
CD36 expression and foam cell formation in human macrophages. J. Nutr. 136,
755S–758S.
Conflict of interest statement
Itabe, H., Ueda, M., 2007. Measurement of plasma oxidized low-density lipoprotein and
Dr Budoff has received a grant to support this research from Wakunaga of America, the
its clinical implications. J. Atheroscler. Thromb. 14, 1–11.
manufacturer of the garlic formulation used in this study. None of the other authors has
Ky, B., Burke, A., Tsimikas, S., et al., 2008. The influence of pravastatin and atorvastatin
relationships to disclose.
on markers of oxidative stress in hypercholesterolemic humans. J. Am. Coll. Cardiol.
51, 1653–1662.
Acknowledgments Lau, B.H., 2001. Suppression of LDL oxidation by garlic. J. Nutr. 131, 985S–988S.
Miyazaki, T., Shimada, K., Sato, O., et al., 2005. Daida H. Circulating malondialdehyde-modified
This study was supported by a Grant from Wakanuga Inc. of LDL and atherogenic lipoprotein profiles measured by nuclear magnetic resonance
spectroscopy in patients with coronary artery disease. Atherosclerosis 179, 139–145.
America and the Foundation Leducq. The only conflict of interest is Dr Moher, D., Schulz, K.F., Altman, D.G., 2001. The CONSORT statement: revised
Budoff has received grant support from Wakanuga. All other authors recommendations for improving the quality of reports of parallel-group rando-
declare that there are no conflicts of interest. mised trials. Lancet 357 (9263), 1191–1194.
Morihara, N., Sumioka, I., Moriguchi, T., Uda, N., Kyo, E., 2002. Aged garlic extract
enhances production of nitric oxide. Life Sci. 71, 509–517.
Appendix A. Supplementary data Nakaishi, T., Tamura, A., Watanabe, T., Mikuriya, Y., Nasu, M., 2000. Relationship
between plasma oxidized low-density lipoprotein and the coronary vasomotor
response to acetylcholine in patients with coronary artery disease. Jpn. Circ. J. 64,
Supplementary data associated with this article can be found, in
856–860.
the online version, at doi:10.1016/j.ypmed.2009.06.018. Neil, A., Silagy, C., Lancaster, T., et al., 1996. Garlic powder in the treatment of moderate
hyperlipidaemia: a controlled trial and meta-analysis. J. R. Coll. Physicians Lond. 30,
References 329–334.
Okuhara, T., 1994. A clinical study of garlic extract on peripheral circulation. Jpn.
Agatston, A.S., Janowitz, W.R., Hildner, F.J., Zusmer, N.R., Viamonte Jr., M., Detrano, R., 1990. Pharmacol. Ther. 22, 3695–3701.
Quantification of coronary artery calcium using ultrafast computed tomography. J. Am. Okuhira, M., Holbert, H., Sandhu, R., Steiner, M., 2000. Modification of Cardiovascular
Coll. Cardiol. 15, 827–832. Risk Factors by Aged Garlic Extract (AGE). Phytomedicine 7, 49–50.
Ahmadi, N., Gul, K., Hajsadeghi, F., et al., 2007. Fingertip thermal reactivity correlated Raggi, P., Callister, T.Q., Shaw, L.J., 2004. Progression of coronary artery calcium and risk
with Framingham risk score. J. Cardiovasc. Comput. Tomogr. 1 (1), S13. of first myocardial infarction in patients receiving cholesterol-lowering therapy.
Ahmadi, N., Hajsadeghi, F., Gul, K., et al., 2008a. Digital thermal monitoring of vascular Arterioscler. Thromb. Vasc. Biol. 24, 1272–1277.
function is associated with coronary artery calcification in asymptomatic adults. Rahman, K., Billington, D., 2000. Dietary supplementation with aged garlic extract
Circulation 117, e409–e468. inhibits ADP-induced platelet aggregation in humans. J. Nutr. 130, 2662–2665.
 M.J. Budoff et al. / Preventive Medicine 49 (2009) 101–107 107

Rasouli, L., Budoff, M.J., Park, R., Aziz, D.C., 2005. Plasma homocysteine predicts pro- Tsimikas, S., Witztum, J.L., Miller, E.R., et al., 2004b. High-dose atorvastatin reduces total
gression of atherosclerosis. Atherosclerosis 181, 159–165. plasma levels of oxidized phospholipids and immune complexes present on
Rodenburg, J., Vissers, M.N., Wiegman, A., et al., 2006. Oxidized low-density lipoprotein apolipoprotein B-100 in patients with acute coronary syndromes in the MIRACL
in children with familial hypercholesterolemia and unaffected siblings: effect of trial. Circulation 110, 1406–1412.
pravastatin. J. Am. Coll. Cardiol. 47, 1803–1810. Tsimikas, S., Brilakis, E.S., Miller, E.R., et al., 2005. Oxidized phospholipids, Lp(a)
Rosen, R.T., Hiserodt, R.D., Fukuda, E.K., et al., 2001. Determination of allicin, S-allylcysteine lipoprotein, and coronary artery disease. N. Engl. J. Med. 353, 46–57.
and volatile metabolites of garlic in breath, plasma or simulated gastric fluids. J. Nutr. Tsimikas, S., Kiechl, S., Willeit, J., et al., 2006. Oxidized phospholipids predict the
131, 968S–971S. presence and progression of carotid and femoral atherosclerosis and symptomatic
Silaste, M.L., Rantala, M., Alfthan, G., et al., 2004. Changes in dietary fat intake alter cardiovascular disease: five-year prospective results from the Bruneck study. J. Am.
plasma levels of oxidized low-density lipoprotein and lipoprotein(a). Arterioscler. Coll. Cardiol. 47, 2219–2228.
Thromb. Vasc. Biol. 24, 498–503. Tsimikas, S., Aikawa, M., Miller Jr., F.J., et al., 2007a. Increased plasma oxidized phospholipid:
Simons, D.B., Schwartz, R.S., Edwards, W.D., Sheedy, P.F., Breen, J.F., Rumberger, J.A., apolipoprotein B-100 ratio with concomitant depletion of oxidized phospholipids
1992. Noninvasive definition of anatomic coronary artery disease by ultrafast from atherosclerotic lesions after dietary lipid-lowering: a potential biomarker of early
computed tomographic scanning: a quantitative pathologic comparison study. atherosclerosis regression. Arterioscler. Thromb. Vasc. Biol. 27, 175–181.
J. Am. Coll. Cardiol. 20, 1118–1126. Tsimikas, S., Brilakis, E.S., Lennon, R.J., et al., 2007b. Relationship of IgG and IgM
Steiner, M., Lin, R.I., 1998. Changes in platelet function and susceptibility of lipoproteins autoantibodies to oxidized low density lipoprotein with coronary artery disease
to oxidation associated with administration of aged garlic extract. J. Cardiovasc. and cardiovascular events. J. Lipid Res. 48, 425–433.
Pharmacol. 31, 904–908. Warshafsky, S., Kamer, R.S., Sivak, S.L., 1993. Effect of garlic on total Cholesterol, a meta-
Steiner, M., Kham, A.H., Holbert, D., Lin, R.I.S., 1996. A double-blind crossover study in analysis. Ann. Int. Med. 119, 599–605.
moderately hypercholesteremic men that compared the effect of aged garlic extract Williams, M.J., Sutherland, W.H., McCormick, M.P., Yeoman, D.J., de Jong, S.A., 2005.
and placebo administration on blood lipids. Am. J. Clin. Nutr. 64, 866–870. Aged garlic extract improves vascular function in men with coronary artery disease.
Tsimikas, S., Lau, H.K., Han, K.R., et al., 2004a. Percutaneous coronary intervention Phytother. Res. 19, 314–319.
results in acute increases in oxidized phospholipids and lipoprotein(a): short-term Wilson, P.W., D'Agostino, R.B., Levy, D., Belanger, A.M., Silbershatz, H., Kannel, W.B.,
and long-term immunologic responses to oxidized low-density lipoprotein. 1998. Prediction of coronary heart disease using risk factor categories. Circulation
Circulation 109, 3164–3170. 97, 1837–1847.