Content uploaded by Ethan Basch
Author content
All content in this area was uploaded by Ethan Basch on May 23, 2014
Content may be subject to copyright.
NATURAL STANDARD REVIEW
Catherine Ulbricht, PharmD, MBA(C), Column Editor
Boswellia:
An Evidence-Based Systematic Review
by the Natural Standard Research
Collaboration
Ethan Basch, MD
Heather Boon, BScPhm, PhD
Theresa Davies-Heerema, PhD
Ivo Foppo, MD, ScD
Sadaf Hashmi, MD, MPH
Jens Hasskarl, MD
David Sollars, MAc, HMC
Catherine Ulbricht, PharmD, MBA(C)
Ethan Basch is affiliated with the Natural Standard Research Collaboration. Heather
Boon is affiliated with the University of Toronto, Theresa Davies-Heerema is affiliated
with Boston University, Ivo Foppa is affiliated with Harvard University, Sadaf Hashmi is
affiliated with the Natural Standard Research Collaboration, Jens Hasskarl is affiliated with
Harvard Medical School, David Sollars is affiliated with New England School of Acu
-
puncture, Catherine Ulbricht is affiliated with Massachusetts General Hospital. All are
members of Natural Standard Research Collaboration (www.naturalstandard.com).
The information in this monograph is intended for informational purposes only, and
is meant to help users better understand health concerns. Information is based on re
-
view of scientific research data, historical practice patterns, and clinical experience.
This information should not be interpreted as specific medical advice. Users should
consult with a qualified healthcare provider for specific questions regarding therapies,
diagnosis and/or health conditions, prior to making therapeutic decisions.
Copyright 2004 Natural Standard Inc. Reprinted with permission.
Journal of Herbal Pharmacotherapy, Vol. 4(3) 2004
http://www.haworthpress.com/web/JHP
Digital Object Identifier: 10.1300/J157v04n03_06 63
ABSTRACT. An evidence-based systematic review including written
and statistical analysis of scientific literature, expert opinion, folkloric
precedent, history, pharmacology, kinetics/dynamics, interactions, ad
-
verse effects, toxicology, and dosing.
[Article copies available for a fee from
The Haworth Document Delivery Service: 1-800-HAWORTH. E-mail address:
<docdelivery@haworthpress.com> Website: <http://www.HaworthPress. com>]
KEYWORDS. Boswellia, Boswellia serrata, Articulin-F, RA-1
SYNONYMS/COMMON NAMES/RELATED SUBSTANCES
African elemi (Boswellia frereana), Bibal incense (Boswellia carterii),
Burseraveae, carterii, Dhup, Frankincense, guggals, H15, indish incence,
olibanum, S-compound
®
, sacra, salai guggal, sallai guggul, Sallaki
®
.
Selected combination products which include boswellia: Articulin-F
(Withania somnifera [ashwagandha], Boswellia serrata, Curcuma longa [tur-
meric], zinc complex); RA-1 (Withania somnifera [ashwagandha], Boswellia
serrata, Zingiber officinale [ginger], Curcuma longa [turmeric]).
CLINICAL BOTTOM LINE/EFFECTIVENESS
Brief Background
Resin extracts from the Boswellia serrata tree have been found to in-
hibit the synthesis of pro-inflammatory mediators, including leukotrienes.
Animal and in vitro studies suggest possible efficacy for inflammatory
conditions such as inflammatory bowel disease, rheumatoid arthritis,
and osteoarthritis, although high-quality human data are lacking. Initial
human evidence from one well-designed trial suggests the efficacy of
boswellia as a chronic therapy for asthma (but not for the relief of acute
asthma exacerbations). Further studies are warranted in this area.
As opposed to non-steroidal anti-inflammatory drugs (NSAIDs),
long-term use of boswellia has not been shown to cause gastrointestinal
irritation or ulceration, although adverse effects have not been system
-
atically studied in humans.
64 JOURNAL OF HERBAL PHARMACOTHERAPY
Scientific Evidence for Common/Studied Uses
Historical or Theoretical Uses Which Lack Sufficient Evidence
Acne, amenorrhea, analgesic,
1
antifungal,
2
anti-inflammatory,
3,4
an-
tiseptic, astringent, belching, blood “purification,” breast cysts, bursitis,
cancer,
5
carminative, cervical spondylosis,
6
chronic obstructive pulmo-
nary disease (COPD), cicatrizant (scar formation), cystitis, digestive,
diuretic, dyspepsia, emmenagogue (induces menstruation), expecto-
rant, genital area infections, hyperlipidemia,
7
multiple sclerosis, nephri-
tis, peptic ulcer disease, pimples, sedative,
1
sexually transmitted diseases
(STDs), skin ulcers/sores, syphilis, tendonitis, toxin-induced liver dam-
age,
4
uterine infections, wound healing.
Expert Opinion and Historic Precedent
Boswellia has traditionally been used for a number of topical applica-
tions, including treatment of acne, bacterial and fungal infections, boils,
wound healing, scars, and varicose veins. It is used cosmetically as a fa-
cial toner and to smooth wrinkles.
Chinese herbalists use boswellia in powder form and in teas for rheu-
matism, menstrual pain, and as an external wash for sores and bruises.
Historically, boswellia has been utilized to improve emotional well-
being and as part of religious rituals. It has been reported that it has the
ability to enhance spirituality, mental perception, meditation, prayer,
and consciousness when burned (burning is said to produce a psychoac
-
tive substance, trans-hydrocannabinole).
6
“Olibanum” oil from boswellia is used in food products, alcoholic
and non-alcoholic beverages, frozen dairy desserts, baked goods, pud
-
ding, and gelatins. The level often found in these meat products is
0.001%. Olibanum oil and extracts are used as fixative and/or fragrance
components in soaps, detergents, creams, lotions and perfumes (com
-
monly 0.8%).
Natural Standard Review 65
Indication Evidence Grade
Asthma (chronic therapy) B
Crohn’s disease C
Osteoarthritis C
Rheumatoid arthritis C
Ulcerative colitis C
A
B
C
D
F
GRADING
SYSTEM
LINK
Brief Safety Summary
Likely Safe: When consumed in amounts found in foods (maximum
levels 0.001% in meat products).
Possibly Safe: When used in recommended doses as an oral agent to
treat arthritis, inflammatory bowel disease, and asthma.
8-13
Likely Unsafe: When used in pregnant women, based on reports in
the Indian literature that resin from boswellia may be an emmenagogue
and induce abortion.
14
DOSING/TOXICOLOGY
General
Recommended doses are based on those most commonly used in
available trials, or on historical practice. However, with natural prod-
ucts it is often not clear what the optimal doses are to balance efficacy
and safety. Preparation of products may vary from manufacturer to
manufacturer, and from batch to batch within one manufacturer. Be-
cause it is often not clear what are the active components of a product,
standardization may not be possible, and the clinical effects of different
brands may not be comparable.
Standardization
The gum resin typically contains 30% boswellic acids, while ethanol
extracts contain 43% boswellic acids.
15
Some commercial sources con-
tain up to 65% boswellic acids. Oral doses of 200-400 mg are often stan
-
dardized to contain 37.5% boswellic acids per dose.
8
The standardized boswellia products Sallaki
®
(India) and H15
®
(Swit-
zerland) contain 11-keto-β-boswellic acid (1.8%), acetyl-11-keto-β-bos
-
wellic acid (1.4%), and acetyl-β-boswellic acid/β-boswellic acid (2%).
12
The standardized boswellia product S-compound
®
contains 11-keto-
β-boswellic acid (0.63%), acetyl-11-keto-β-boswellic acid (0.7%), acetyl-
β-boswellic acid/β-boswellic acid (1.5%).
13
Adult Dosing (18 Years and Older)
Oral Tablets/Capsules
Asthma: 300 mg three times a day of boswellia powdered gum resin
capsules (S-compound
®
was used in one trial),
13
or 400 mg three times
66 JOURNAL OF HERBAL PHARMACOTHERAPY
daily (extract standardized to 37.5% boswelilic acids per dose) (anec
-
dotal).
Crohn’s Disease: 1200 mg three times daily of standardized Bos
-
wellia serrata gum resin H15
®
, for up to eight weeks.
16
Osteoarthritis: Two capsules, 3 times daily of Articulin-F
®
(combi
-
nation formula containing 100 mg Boswellia serrata, 450 mg Withania
somnifera [ashwagandha], 50 mg Curcuma longa [turmeric], and 50 mg
zinc complex).
10
Rheumatoid Arthritis: 400 mg three times daily of standardized Bos
-
wellia serrata gum resin H15
®
,
8
or two capsules, three times daily of
Aticulin-F (combination formula containing 100 mg Boswellia serrata,
450 mg Withania somnifera [ashwagandha], 50 mg Curcuma longa
[turmeric], and 50 mg zinc complex).
11
Ulcerative Colitis: 350-400 mg three times daily (extract standard
-
ized to 37.5% boswelilic acids per dose).
12
Pediatric Dosing (Younger Than 18 Years)
Review of the literature reveals no adverse events specifically related
to the use of boswellia in children. However, safety, efficacy, and dos-
ing have not been systematically studied. Some experts believe that reg-
ular use of boswellia may mask the symptoms of asthma in children,
and may delay diagnosis. Use in children should be supervised by an ap-
propriately licensed healthcare professional.
Toxicology
The LD
50
of boswellic acids is > 2 g/kg in rats and mice; doses of 2 g/
kg in mice, rats, and monkeys have not caused death; high doses have
reportedly not yielded significant effects on behavior or clinical, hema-
tological, biochemical and pathological data.
17,15
PRECAUTIONS/CONTRADICTIONS
Allergy
No allergic or hypersensitivity reactions to boswellia have been reported
in the available literature, although this area lacks systematic study.
Adverse Effects
General: Boswellia is generally believed to be safe when used as di
-
rected, although safety and toxicity have not been systematically stud
-
Natural Standard Review 67
ied in humans. The most common complaints in trials have been nausea
and acid reflux. A licensed healthcare provider should be consulted
prior to use.
Dermatologic: Dermatitis was reported in 4 of 62 patients (6%) in
two clinical trials using Articulin-F
®
, a combination product containing
gum resin from Boswellia serrata as well as Withania somnifera (ash
-
wagandha), Curcuma longa (turmeric), and zinc complex.
10,11
The in
-
dependent effects of boswellia are not clear.
Gastrointestinal: Boswellia extract has been associated with mild
gastrointestinal upset in a randomized controlled trial of patients with
osteoarthritis (18). In a study of patients with ulcerative colitis, abdomi
-
nal fullness, epigastric pain, gastroesophageal reflux symptoms, diar-
rhea, and nausea were reported by 6 of 34 patients (18%) receiving 350
mg three times daily of boswellia for 6 weeks.
12
It is not clear to what
extent these symptoms were related to the patients’ underlying colitis.
Two of 80 patients (3%) receiving the boswellia powdered gum resin
formulation S-compound
®
complained of epigastric pain, hyperacidity,
and nausea.
13
Nausea was reported in 3 of 62 patients (5%), and abdom-
inal fullness in 7 of 62 patients (11%) in two clinical trials using
Articulin-F
®
(combination product containing Boswellia serrata, Withania
somnifera [ashwagandha], Curcuma longa [turmeric], and zinc com-
plex).
10,11
However, the independent effects of boswellia could not be
determined.
Precautions/Warnings/Contraindications
Use cautiously in patients with pre-existing gastritis or gastroesoph-
ageal reflux disease (GERD), since reflux and epigastric pain have been
associated with the use of boswellia.
Use cautiously in patients taking lipid-soluble medications, since the
gum resin of boswellia has been reported to lower cholesterol and
triglyceride levels,
7
and may bind to/impair absorption of these medica
-
tions.
Pregnancy and Lactation
Reports in the Indian literature suggest that resin from boswellia is an
emmenagogue and may induce abortion.
14
Safety of boswellia during
pregnancy has not been systematically studied, and therefore cannot be
recommended.
68 JOURNAL OF HERBAL PHARMACOTHERAPY
INTERACTIONS
Boswellia/Drug Interactions
Leukotriene Inhibitors: Boswellia has been found in animal and in vi
-
tro studies to inhibit 5-lipoxygenase, thereby reducing the production of
leukotrienes.
3,6,4,19
Boswellia therefore may potentiate the actions of
pharmaceutical leukotriene inhibitors such as zafrilukast (Accolate
®
)
and montelukast (Singulair
®
), which are used in the treatment of asthma.
Anti-Neoplastic Agents: Boswellic acids have been found in vitro to
inhibit protein synthesis via effects on nucleic acids, and to inhibit pro
-
liferation of human leukemic HL-60 cells.
20,5
Theoretically, concomi-
tant use with other anti-proliferative agents may potentiate effects or
toxicity.
Lipid-Lowering Agents: The gum of boswellia has been reported to
lower cholesterol and triglyceride levels in rats,
7
and may potentiate the
effects of lipid lowering agents.
Fat-Soluble Medications: The gum resin of boswellia has been re-
ported to lower cholesterol and triglyceride levels,
7
and may bind to/im-
pair absorption of lipid-soluble agents.
Non-Steroidal Anti-Inflammatory Agents (NSAIDs), COX-2 Inhibi-
tors: In the treatment of arthritic conditions, the purported mechanism
of boswellia’s activity is reduction of glycosaminoglycan (GAG) deg-
radation, based on rat studies.
21
This potentially beneficial mechanism
may theoretically be disrupted by concomitant use of NSAIDs.
Anti-Fungal Agents: Potential Positive Interaction: The essential oil
from Boswellia serrata has been reported to possess anti-fungal activ-
ity, with weak activity against human fungal pathogens in vitro (but
greater effect against plant fungal pathogens).
2
Boswellia/Herb/Supplement Interactions
Glycosaminoglycans (GAGs), Chondroitin Sulfate, Glucosamine:
Boswellia has been reported to reduce the degradation of glycosamino
-
glycans in rats,
21
and may act addictively or synergistically with agents
shown to be efficacious in the treatment of osteoarthritis, such as
glucosamine and chondroitin.
Anti-Proliferative Agents: Boswellic acids have been found in vitro
to inhibit protein synthesis via effects on nucleic acids, and to inhibit
proliferation of human leukemic HL-60 cells.
20,5
Theoretically, con
-
Natural Standard Review 69
comitant use with other anti-proliferative agents may potentiate effects
or toxicity.
Anti-Fungal Agents: Potential Positive Interaction: The essential oil
from Boswellia serrata has been reported to possess anti-fungal activ
-
ity, with weak activity against human fungal pathogens in vitro (but
greater effect against plant fungal pathogens).
2
Lipid-Lowering Agents: The gum of boswellia has been reported to
lower cholesterol and triglyceride levels in rats,
7
and may potentiate the
effects of lipid lowering agents such as garlic.
Boswellia/Food Interactions
Literature review reveals no reported interactions.
Boswellia/Lab Interactions
Serum Lipids: The gum of boswellia has been reported to lower cho-
lesterol and triglyceride levels in rats,
7
although human studies are lack-
ing.
Liver Function Tests (Transaminases): Toxin-induced transaminitis
in mice was reduced by administration of boswellia, although effects on
normal liver or on humans are not clear.
4
MECHANISM OF ACTION
Pharmacology
Boswellia serrata is a branching tree found in India, North Africa,
and the Middle East. A gummy oleo-resin is found under the bark, which
contains oil, resins, and gum. Extracts of this gummy exudate have been
used medicinally and scientifically evaluated.
In vitro and rat studies have reported that acetyl-11-keto-β-boswellic
acid from boswellia inhibits the enzyme 5-lipoxygenase, which produces
5-hydroxyeicosatetraenoic (5-HETE) and leukotriene B4 (LTB4).
3,6,22,23,
These products are involved with the induction of bronchoconstriction,
chemotaxis, and vascular permeability.
3,4,6,19
Additional studies have
found that boswellia inhibits human leukocyte elastase (HLE), which is
involved in the pathogenesis of emphysema, cystic fibrosis, chronic
bronchitis, and acute respiratory distress syndrome.
3,6
Multiple penta
-
cyclic triterpenic acids have been isolated from boswellia.
24-26
70 JOURNAL OF HERBAL PHARMACOTHERAPY
Anti-inflammatory effects of boswellic acids have been reported in
animal studies.
22, 27
Doses of 50-200 mg/kg given orally to mice, after
intra-pleural injection of carrageenan, inhibited polymorphonuclear
leukocyte (PMN) infiltration into the pleural cavity. This response was
similar to indomethacin (1.25 to 5 mg/kg). Alcoholic extracts of boswellia
in doses of 50-200 mg/kg orally inhibited carrageenan-induced paw
edema in rats similar to phenylbutazone (50-100 mg/kg), and improved
blood supply to joint tissues.
17
Mixed acetylboswellic acids extracted
from the gum resin of Boswellia serrata significantly inhibited iono
-
phone-stimulated release of leukotrienes B4 and C4 from intact human
PMNs.
28
Boswellic acids have demonstrated anti-inflammatory and
anti-arthritic activity in chronic models of adjuvant-induced polyar-
thritis and formaldehyde arthritis in rats,
17
and in BSA-induced arthritis
in rabbits.
29
Boswellic acids produced a protective effect in sodium
urate gouty arthritis in dogs, reduced exudate volume and inhibited leu-
kocyte migration in carrageenan-induced pleurisy in rats, and was
antipyretic in rats and rabbits.
27
In animals, ingestion of defatted alcoholic extracts of boswellia de-
creases PMN infiltration and migration, decreases primary antibody
synthesis, and inhibits the classical complement pathway.
26,29,30
Humoral
responses are also inhibited by oral boswellia extract 25-200 mg/kg in
mice (similar to the effect of azathioprine 100 mg/kg orally).
17
Pro-
longed administration of boswellic acids (25-100 mg/kg for 21 days) in-
creases body weight and total leukocyte counts in rats.
30,31
The non-phenolic ration of Boswellia serrata gum resin (20-300 mg/kg)
exhibits an analgesic effect in rats similar to morphine (4.5 mg/kg), and a
sedative effect (55-300 mg/kg) comparable to chlorpromazine (7.5 mg/kg).
1
In biochemical studies, boswellic acids have acted to reduce arthri-
tis-associated elevated enzymes such as glutamic pyruvic transaminase,
glycohydrolase, and β-glucuronidase.
32,33
Inhibition of glycosamino
-
glycan (GAG) synthesis and urinary excretion of connective tissue me
-
tabolites by boswellic acids have been proposed as support for the
purported beneficial effects of boswellia in preventing the degradation
of connective tissue in inflammatory arthritic conditions.
21
In anti-hyperlipidemic studies performed in rats, boswellic acids
have been found to reduce serum cholesterol and triglycerides.
7
Boswellic acids have not been found to act as antioxidants.
4
Pharmacodynamics/Kinetics
The LD
50
of boswellic acid is > 2 g/kg in rats and mice when adminis
-
tered orally or intraperitoneally. Subacute toxicity studies in rabbits
Natural Standard Review 71
over three months, and chronic toxicity studies in rats and monkeys
over six months, have found no toxic effects of boswellic acids at high
doses.
27,34
Pentacylic triterpene boswellic acids from Boswellia serrata
Roxb. inhibited leukotriene B4 and C4 biosynthesis in intact PMNs.
28
Acetyl-11-keto-β-boswellic acid induced inhibition of 5-lipoxygenase
product formation non-competitively and reversibly.
Twelve hundred milligrams of boswellia resulted in plasma concen
-
trations of 10-32 µl M of 11-keto-β-boswellia acid and 18-20 µl M
acetyl-11-keto-β-boswellic acid, measured 2-3 hours following admin
-
istration.
12
HISTORY
Boswellia has traditionally been used for numerous medicinal pur
-
poses, including skin disorders, infections, wound healing, and varicos-
ities. In Traditional Chinese Medicine, powders and teas from boswellia
were used to treat rheumatic diseases, menstrual disorders, and bruises.
Cosmetically, it has been recommended as a facial toner and to smooth
wrinkles.
The use of boswellia dates to ancient Egypt, where “Olibanum” was
used an ingredient in embalming liquids for mummification. Boswellia
has historically been utilized to improve emotional well-being and as a
part of religious rituals. It has been reported to enhance spirituality, mental
perception, meditation, prayer, and consciousness when burned (burning
is said to produce a psychoactive substance, trans-hydrocannabinole).
6
Recent interest in the use of boswellia for inflammatory diseases
such as arthritis, inflammatory bowel disease, and asthma stems from
scientific reports that boswellia inhibits leukotriene synthesis.
EVIDENCE TABLE
Condition Study
Design
Author,
Year
N Statistically
Significant?
Quality of
Study
0-2 = poor
3-4 = good
5 = excellent
Magnitude
of Benefit
ARR NNT Comments
Asthma Randomized
Controlled
Gupta,
1998
80 Yes 4 Medium NA NA 70% showed
improvement.
Crohn’s
disease
Randomized
equivalence
study,
double-blind
Gerhardt,
2001
102 No 3 None NA NA 19% dropout. No
difference between
boswellia (H15)
and mesalazine.
Unlcear if
equivalent, or
inadequately
powered to detect
differences.
72 JOURNAL OF HERBAL PHARMACOTHERAPY
Condition Study
Design
Author,
Year
N Statistically
Significant?
Quality of
Study
0-2 = poor
3-4 = good
5 = excellent
Magnitude
of Benefit
ARR NNT Comments
Osteoarthritis Randomized
controlled,
crossover
Kulkarni,
1991
42 Yes 4 Large NA NA Combination
product used
(Articulin-F);
effects of
boswellia alone
not clear.
Osteoarthritis
(knee)
Randomized
placebo
controlled,
crossover
Kimmatkar,
2003
30 Yes 2 Medium NA NA
Boswellia serrata
extract (BSE)
associated with
improved pain,
flexion, walking
distance after 8
weeks.
Rheumatoid
arthritis
Randomized
controlled
Chopra,
2000
182 Yes 5 Small NA NA Combination
product used
(“RA-1”). Strong
effect seen in
placebo group.
Rheumatoid
arthritis
Before and
after
comparison
Sander,
1998
37 No 3 NA NA NA No improvement
in pain with
boswellia product
“H15,” but no
power calculation
done.
Rheumatoid
arthritis
Review of 11
unpublished
studies
Etzel, 1996 > 260 NA NA NA NA NA Poorly described
review with no
numeric or
statistical data;
inclusion and
exclusion criteria
not noted.
Rheumatoid
arthritis
Randomized
controlled,
crossover
Kulkarni,
1992
20 Yes 3 Large NA NA Combination
product used
(Articulin F);
effects of
boswellia alone
not clear.
Ulcerative
colitis
Non-
randomized,
open,
equivalence
study
Gupta,
2001
30 No 0 NA NA NA Improved signs &
symptoms with
boswellia vs.
baseline, but no
difference from
control
(sulfasalazine).
No placebo.
Methodologically
weak.
Ulcerative
colitis
Non-
randomized,
open,
equivalence
study
Gupta,
1997
42 No 0 NA NA NA Improved signs &
symptoms with
boswellia vs.
baseline, but no
difference from
control
(sulfasalazine).
No placebo.
Methodologically
weak.
Natural Standard Review 73
EVIDENCE DISCUSSION
Asthma (Chronic Therapy)
Summary: Boswellia has been proposed as a potential asthma chronic
therapy, based on its known properties as an inhibitor of leukotriene
biosynthesis (a mechanism known to be involved with the progression
of asthmatic bronchoconstriction). One randomized, controlled trial of
good quality in 80 subjects has demonstrated improvements in FEV1
(forced expiratory volume in 1 second), FVC (forced vital capacity),
number of asthma exacerbations, and wheezing following 41 days of
boswellia therapy. However, baseline characteristics between patients
in this study may not have been comparable. Nonetheless, the existing
data provide good initial evidence in favor of this use of boswellia. Fu-
ture studies are warranted to assess the long-term efficacy and safety of
boswellia, the temporality of effects, and the efficacy of boswellia vs.
standard therapies. Boswellia should not be used for the relief of acute
asthma exacerbations.
Evidence: A 6-week double-blind, placebo controlled study, Gupta
et al. examined the use of boswellia gum resin in 80 patients with bron-
chial asthma.
13
Subjects were randomized to receive either 300 mg
boswellia powdered gum resin capsules (S-compound
®
) or 300 mg lac-
tose (as placebo), orally three times daily. Baseline patient characteris-
tics were compared between groups in a Table 1. Overall, the boswellia
group was older than the placebo group (mean age 37.7 vs. 33.0 respec-
tively), although this difference was not statistically significant. However,
there were significant differences in quantitative baseline characteris-
tics in the boswellia subjects vs. placebo, including a lower mean FEV1
(1.6 L vs. 2.0 L), a lower FVC (1.9 L vs. 2.3 L), and a lower PEFR (peak
expiratory flow rate) (244 L/min vs. 306 L/min). Clinical and laboratory
assessments were performed on days 1 and 42, and a log of asthma ex
-
acerbations was documented. Mean improvements were seen in both
groups for multiple parameters, and improvements in the boswellia
group were greater. The median improvement in FEV1 was 25% in the
boswellia group vs. 5% in placebo, the difference being statistically sig
-
nificant. However, it should be noted that the mean FEV1 in the
boswellia group was 20% lower than in the control group at baseline,
and therefore at the study’s end, the values for the two groups were near
equivalent. The mean FVC improved by 21% in the boswellia group vs.
9% in the control group, with statistical significance. Again, however,
there were significant differences in baseline values. Secondary effects
74 JOURNAL OF HERBAL PHARMACOTHERAPY
such as reduced dyspnea and eosinophilia, and absence of rhonchi after
treatment, were also apparent in the boswellia group vs. placebo. In ad
-
dition, the number of asthma exacerbations was significantly reduced in
the boswellia group (p < 0.0001). Overall, these results are promising,
and have a basis in basic science studies which demonstrate boswellia to
inhibit leukotriene biosynthesis (a mechanism known to be involved in
the biochemical progression of asthmatic bronchoconstriction). The
principal weakness of this study is the significant difference in baseline
characteristics between groups, raising questions about appropriate ran
-
domization, and about whether the groups were comparable. Future
studies should assess the long-term efficacy and safety of boswellia, the
temporality of boswellia’s effects, and compare boswellia with standard
therapies.
Crohn’s Disease
Summary: Boswellia has been noted in animal and in vitro studies to
possess anti-inflammatory properties. Based on these observations,
boswellia has been suggested as a potential treatment for Crohn’s dis-
ease. However, limited human data exist, and there is inadequate evi-
dence in favor of or against this use of boswellia.
Evidence: Gerhardt et al. conducted an eight-week randomized, dou-
ble-blind equivalence study using the Boswellia serrata standardized
extract H15
®
or mesalazine in 102 patients with active Crohn’s dis-
ease.
16
Three times daily, subjects were administered either 1200 mg of
H15
®
or 1.5 g of mesalazine. The primary outcome measure used was
the validated Crohn’s Disease Activity Index (CDAI) which takes into
account multiple signs and symptoms. At the end of eight weeks, mod-
erate improvements in CDAI scores were seen in both groups, with
greater improvements in the H15
®
group. However, the difference in
mean scores between groups was not statistically significant. Since no
power calculation was conducted, it remains unclear if the lack of sig
-
nificance of results between groups is due to actual equivalence be
-
tween therapies, or to an inadequate sample size (although a sample size
of 102 may have been adequate). Because there was no placebo group,
the improvements in the H15
®
group vs. baseline cannot be discerned
from the natural history of the disease. The procedures for blinding and
randomization were not described. Nineteen subjects (19%) dropped
out of the study (6 in the boswellia group and 13 in the mesalazine
group), and it is not clear that there was an intent-to-treat analysis. Al
-
though the improvements observed in the H15
®
group appear promis
-
Natural Standard Review 75
ing, due to the methodological weaknesses of this study, the results
cannot be considered clinically relevant. Further investigation is war
-
ranted in this area.
Osteoarthritis
Summary: Boswellia has been noted in animal and in vitro studies to
possess anti-inflammatory properties. Based on these observations,
boswellia has been suggested as a potential treatment for osteoarthritis.
To date, there is limited clinical evaluation of boswellia for this indica
-
tion. Although promising results are reported, due to methodological
problems with available data, no clear conclusion can be drawn.
Evidence: Kulkarni et al. conducted a randomized, double-blind, pla-
cebo controlled, crossover study in 42 patients with osteoarthritis.
10
Patients were randomized to receive either the combination product
Articulin-F
®
(two capsules, 3 times daily) or placebo for 3 months. Each
tablet of Articulin-F
®
contains extracts of 100 mg Boswellia serrata,
450 mg Withania somnifera (ashwagandha), 50 mg Curcuma longa
(turmeric), and 50 mg zinc complex. Pain severity and disability scores
were tabulated using validated instruments (Ritchie articular index,
American Rheumatism Association joint score). At the study’s end,
treatment with Articulin-F
®
was found to have significantly improved
the mean pain severity score (p < 0.001) and mean disability score (p <
0.05). Other parameters such as morning stiffness, grip strength and
joint score also showed improvement, but without statistical signifi-
cance. The lack of significance of these results may reflect an absence
of true benefit, or inadequate sample size to detect true benefits. Weak-
nesses of this study also include poor description of randomization and
blinding methods, and unclear diagnostic criteria by which patients
were judged to have osteoarthritis (unclear inclusion criteria). As a re
-
sult, the patient population may not have been uniform. The mixed sta
-
tistical significance of results and the isolated nature of this study leave
open the question of the efficacy of Articulin-F
®
for osteoarthritis. Be
-
cause Articulin-F
®
is a combination product, no firm conclusions can be
drawn regarding boswellia specifically.
Kimmatkar et al. conducted a randomized, double-blind, placebo con
-
trolled, crossover trial in 30 patients with osteoarthritis of the knee.
18
Subjects were administered either placebo or a formulation of
Boswellia serrata extract (BSE) for eight weeks, followed by a washout
period and crossover for an additional eight weeks. The authors re
-
ported statistically significant mean improvements in the BSE group
76 JOURNAL OF HERBAL PHARMACOTHERAPY
compared to placebo in terms of pain, flexion, and walking distance. Al
-
though these results are promising, the descriptions of blinding, random
-
ization, and statistical analysis were not well delineated, diminishing the
quality of this publication overall. Better-quality study is necessary to
confirm these findings.
Rheumatoid Arthritis
Summary: Boswellia has been noted in animal and in vitro studies to
possess anti-inflammatory properties. Based on these observations,
boswellia has been suggested as a potential treatment for rheumatoid ar
-
thritis (RA). Two methodologically weak publications have reported
conflicting results for boswellia monotherapy. Two positive studies of
combination products containing boswellia (RA-1
®
and Articulin-F
®
)
have not provided adequate data regarding the effects of boswellia
alone. Therefore, there is currently insufficient evidence to recommend
for or against the use of boswellia for rheumatoid arthritis.
Evidence (boswellia monotherapy): In a German double-blind, pla-
cebo controlled, pilot study, a 37 patient single-center subset of 78 rheu-
matoid arthritis patients recruited for a multi-center trial were treated
with the standardized boswellia extract H15
®
.
9
For 12 weeks, a 3600
mg daily dose of H15
®
or placebo was given to subjects, in addition to
their baseline medications (NSAIDs and/or steroids). At the study’s
conclusion, a small, non-statistically significant reduction in the usage
of NSAIDs was observed in the boswellia group vs. placebo (5.8% vs.
3.1%, respectively). H15
®
was not found to be effective in reducing
pain or improving function as measured by validated scales, and no dif-
ferences in C-reactive protein were observed between groups. Because
no power calculation was performed, it is not clear that the sample size
of this study was adequate to discern true benefits. In addition, methods
of randomization and blinding were not well described. Therefore, the
results cannot be considered definitive.
Etzel et al. reported the cumulative findings of 11 unpublished stud
-
ies conducted between 1985-1990, ranging from 1-6 months in dura
-
tion, in which > 260 patients with rheumatoid arthritis were treated with
the standardized boswellia extract H15
®
.
8
Results were tabulated for a
total of 375 subjects, but apparently some of the described patients were
the same individuals, assessed in different investigations. Study designs
varied from direct observation (most common) to placebo controlled
and double-blind (although methodological details of each study were
not provided in detail). The authors concluded that across investiga
-
Natural Standard Review 77
tions, H15
®
was not efficacious for the relief of acute pain, but may im
-
prove chronic symptoms such as joint swelling and stiffness, and may
reduce NSAID intake. However, no numeric or statistical data were
provided to support these assertions. Benefits were reported to be addi
-
tive to other therapies, such as NSAIDs. Adverse effects were reported
as being minimal. No statistical analysis was reported in this review.
The primary data are not available for analysis, and the majority of evi
-
dence was derived from non-controlled studies. Therefore, no evi
-
dence-based conclusions can be drawn from this report.
Evidence (combination products that include boswellia): Equivocal
results from a well-designed study were reported by Chopra et al.
35
The
authors conducted a 16-week randomized, double-blind, placebo con-
trolled trial to assess the efficacy of the standardized combination prod-
uct RA-1
®
in 182 patients with rheumatoid arthritis (RA). RA-1
®
is an
herbal mixture of Boswellia serrata, Withania somnifera (ashwagandha),
Zingiber officinale (ginger), and Curcuma longa (turmeric). Subjects
received two tablets three times daily (444 mg/day), and discontinued
other RA therapies. Signs and symptoms associated with RA were as-
sessed by validated instruments, taking into account joint tenderness,
pain, swelling, stiffness, rheumatoid factor, C-reactive protein, and
interleukin 6 (IL-6). Baseline patient characteristics in the two groups
were comparable. An initial calculation was performed to assure at least
80% power (required total n = 130), and result tabulation was designed
as an intent-to-treat analysis. Seventeen patients withdrew, but none
due to drug toxicity. Significant improvements over baseline were ob-
served in both the RA-1 and placebo groups (p < 0.001). Although re-
sults for the RA-1 group were numerically superior to placebo across
outcomes, these differences were not statistically significant except for
three measures: (1) a slightly increased proportion of subjects with a
50% reduction in the number of swollen joints/swollen joint score;
(2) reduced rheumatoid factor (30% reduction vs. 0%); and (3) a rela
-
tive improvement in the “ACR 20” (American College of Rheumatism)
assessment score (39% vs. 30%). Although suggestive, these results
cannot be applied to clinical practice due to their mixed statistical sig
-
nificance. It is not clear if the lack of significance for most results re
-
flects a lack of benefit of RA-1 vs. placebo, or if the study was not
adequately powered to detect differences. Although a power calculation
was performed, due to large improvements in multiple parameters in the
placebo group, true differences across categories may not have been de
-
tected with statistical significance. Notably, rheumatoid factor, a sero
-
logic marker which is less subject to a “placebo effect” did improve in
78 JOURNAL OF HERBAL PHARMACOTHERAPY
the RA-1 group over placebo, although C-reactive protein did not. In
addition, because this study examined a combination product, the rela
-
tive contributions of each constituent are not clear, and the amounts of
each may have been lower than in common monotherapy doses. The re
-
sults remain equivocal, and may merit confirmation by additional stud
-
ies.
In a randomized, double-blind, placebo controlled, crossover study,
Kulkarni et al. studied the combination product Articulin-F
®
in 20 pa
-
tients with RA (11). Each tablet of Articulin-F
®
contains extracts of 100
mg Boswellia serrata, 450 mg Withania somnifera (ashwagandha), 50
mg Curcuma longa (turmeric), and 50 mg zinc complex. Treatment in
-
cluded Articulin-F
®
(two capsules, 3 times a day) or placebo for 3
months. All previous drugs were withdrawn 1 month prior to the study.
Inclusion criteria for subjects included morning stiffness, joint swell-
ing, pain severity, disability/loss of function, spells of remission, and
serologically positive rheumatoid factor. Assessment was based on
measures of stiffness and pain (by both validated [Ritchie articular in-
dex, American Rheumatism Association joint score] and non-validated
scales), erythrocyte sedimentation rate (ESR), and rheumatoid factor.
Treatment with Articulin-F
®
was reported to moderately improve all
measurements of pain, morning stiffness duration, grip strength, and
disability score vs. placebo (p < 0.001). Benefits were noted after two
weeks of therapy, and persisted throughout the study. After three
months, rheumatoid factor seroconversion (to seronegativity) occurred
in 9 patients taking Articulin-F
®
and in none taking placebo. Non-
steroidal anti-inflammatory drugs were required to control symptoms in
3 patients taking Articulin-F
®
vs. 18 taking placebo (p < 0.05). Overall,
these results are promising, although procedures for randomization and
blinding were not reported. As a result, bias and confounders may have
been introduced that affected outcomes. To some extent, the crossover
design would reduce the potential effects of confounders. Since a com
-
bination product was used, the isolated effects of boswellia cannot be
assessed.
Ulcerative Colitis
Summary: Boswellia has been noted in animal and in vitro studies to
possess anti-inflammatory properties. Based on these observations,
boswellia has been suggested as a potential treatment for ulcerative co
-
litis. At this time, however, only a limited number of poor-quality hu
-
man trials have evaluated this use of boswellia, with inconclusive
Natural Standard Review 79
results. Therefore, there is inadequate evidence for or against this use of
boswellia.
Gupta et al. conducted a poor-quality, open, non-randomized equiva
-
lence study in 30 patients with chronic colitis, in which subjects were
administered either Boswellia serrata gum resin (S compound
®
manu
-
factured in India) or sulfasalazine.
36
Subjects between the ages of 18 to
48 years were treated for six weeks with either boswellia 900 mg daily
in three divided doses (n = 20) or sulfasalazine 3 g daily in three divided
doses (n = 10). Measured outcomes included sigmoidoscopic examina
-
tion, rectal biopsy histopathology, stool characteristics, and serum val
-
ues (hemoglobin, iron, calcium, phosphorus, proteins, total leukocytes
and eosinophils). These values were entered into a formula to determine
a “remission rate,” although the specifics of the formula and calculation
were not adequately described. The authors noted that 18 of 20 boswellia
patients entered “remission,” vs. 6 of 10 in the sulfasalazine group, but
the difference between groups was not statistically significant. Histo-
logical improvement of biopsies was noted in 75% of boswellia subjects
vs. 40% of sulfasalazine subjects, but again, no statistically significant
differences between groups was found. These results cannot be clini-
cally interpreted. Although the authors report statistically significant
improvements in the “remission rate” and biopsy histolopathology of
boswellia subjects vs. baseline, comparisons with controls were not sta-
tistically significant. Since no power calculation was conducted, it re-
mains unclear if the lack of significance is due to actual equivalence
between therapies, or to an inadequate sample size. Because there was
no placebo group, the improvements in the boswellia group vs. baseline
cannot be discerned from the natural history of the disease. The lack of
blinding introduces the possibility of bias, and lack of randomization af-
fords an opportunity for confounding.
In an earlier open, non-randomized trial, the same group of authors
administered encapsulated powdered Boswellia serrata gum resin
(350 mg three times daily) or sulfasalzine (1 g three times daily) for six
weeks to 42 patients with ulcerative colitis.
12
Patients were allowed to
select their choice of therapy: 34 chose boswellia and 8 chose sulfa
-
salazine. Measured outcomes included symptoms improvement (ab
-
dominal pain, diarrhea), sigmoidoscopic examination, rectal biopsy
histopathology, stool characteristics, and serum values (hemoglobin,
iron, calcium, phosphorus, proteins, total leukocytes and eosinophils).
These values were entered into a formula together to determine a “re
-
mission rate,” although the specifics of the formula and calculation
were not adequately described. The authors reported improved abdomi
-
nal pain and diarrhea in all sulfasalazine patients, and in approximately
80 JOURNAL OF HERBAL PHARMACOTHERAPY
90% of boswellia patients, without statistical significance between
groups. Histopathological improvements were observed in approxi
-
mately 75% of both boswellia and sulfasalazine patients, again without
significant differences between groups. “Remission” occurred in 82.4%
of boswellia patients and 75% of sulfasalazine patients with no signifi
-
cant differences between groups. Since no power calculation was con
-
ducted, it remains unclear if the lack of significance of results between
groups is due to actual equivalence between therapies, or to an inade
-
quate sample size. Because there was no placebo group, the improve
-
ments in the boswellia group vs. baseline cannot be discerned from the
natural history of the disease. The lack of blinding or randomization
with the allowance for patients to choose their own therapies increases
the risk of bias or confounding. No firm conclusions can be drawn due
to the methodological weaknesses of this study.
FORMULARY: BRANDS USED IN CLINICAL TRIALS/
THIRD-PARTY TESTING
Brands Used in Clinical Trials
Sallaki
®
(India) and H15
®
(Switzerland) are standardized extracts of
Boswellia serrata, marketed by M/S Gufic Ltd., India. This compound
contains 11-keto-β-boswellic acid (1.8%), acetyl-11-keto-β-boswellic
acid (1.4%), and acetyl-β-boswellic acid/β-boswellic acid (2%).
12
S-compound
®
is manufactured by Rahul Pharma, Jammu Tawi, India.
This compound contains 11-keto-β-boswellic acid (0.63%), acetyl-11-
keto-β-boswellic acid (0.7%), acetyl-β-boswellic acid/β-boswellic acid
(1.5%).
13
RA-1
®
is a herbal mixture of Boswellia serrata, Withania somnifera
(ashwagandha), Zingiber officinale (ginger), and Curcuma longa (tur
-
meric).
35
Articulin-F
®
is a herbomineral combination containing 100 mg Bos
-
wellia serrata, 450 mg Withania somnifera (ashwagandha), 50 mg Curcuma
longa (turmeric), and 50 mg zinc complex.
10
REFERENCES
1. Menon MK, Kar A. Analgesic and psychopharmacological effects of the gum
resin of Boswellia serrata. Planta Med 1971;19(4):333-341.
Natural Standard Review 81
2. Gangwal ML, Vardhan DK. Antifungal studies of volatile constituents of
Boswellia serrata. Asian J Chem 1995;7:675-676.
3. Ammon HP, Mack T, Singh GB, et al. Inhibition of leukotriene B4 formation in
rat peritoneal neutrophils by an ethanolic extract of the gum resin exudate of Boswellia
serrata. Planta Med 1991;57(3):203-207.
4. Safayhi H, Mack T, Sabieraj J, et al. Boswellic acids: Novel, specific, nonredox
inhibitors of 5-lipoxygenase. J Pharm Exper Ther 1992;261(3):1143-1146.
5. Shao Y, Ho CT, Chin CK, et al. Inhibitory activity of boswellic acids from
Boswellia serrata against human leukemia HL-60 cells in culture. Planta Med 1998;
64(4):328-331.
6. Ammon HP. Salai Guggal–Boswellia serrata: From a herbal medicine to a
non-redox inhibitor of leukotriene biosynthesis. Eur J Med Res 1996;1(8):369-370.
7. Atal CK, Gupta OP, Singh GB. Salai guggal: A promising anti-arthritic and
anti-hyperlipidemic agent. Proc BPS 1981;203P-204P.
8. Etzel R. Special extract of Boswellia serrata (H15) in the treatment of rheuma
-
toid arthritis. Phytomed 1996;3(1):91-94.
9. Sander O, Herborn G, Rau R. [Is H15 (resin extract of Boswellia serrata, “in
-
cense”) a useful supplement to established drug therapy of chronic polyarthritis? Re-
sults of a double-blind pilot study]. Z Rheumatol 1998;57(1):11-16.
10. Kulkarni RR, Patki PS, Jog VP, et al. Treatment of osteoarthritis with a
herbomineral formulation: A double-blind, placebo-controlled, cross-over study. J
Ethnopharm 1991;33(1-2):91-95.
11. Kulkarni RR, Patki PS, Jog VP, et al. Efficacy of an Ayurvedic formulation in
rheumatoid arthritis: A double-blind, placebo-controlled, cross-over study. Indian J
Pharm 1992;24:98-101.
12. Gupta I, Parihar A, Malhotra P, et al. Effects of Boswellia serrata gum resin in
patients with ulcerative colitis. Eur J Med Res 1997;2(1):37-43.
13. Gupta I, Gupta V, Parihar A, et al. Effects of Boswellia serrata gum resin in pa-
tients with bronchial asthma: Results of a double-blind, placebo-controlled, 6-week
clinical study. Eur J Med Res 1998;3(11):511-514.
14. Kamboj VP. A review of Indian medicinal plants with interceptive activity. In
-
dian J Med Res 1988;87:336-355.
15. Singh GB, Bani S, Singh S. Toxicity and safety evaluation of boswellic acids.
Phytomed 1996;3(1):87-90.
16. Gerhardt H, Seifert F, Buvari P, et al. [Therapy of active Crohn disease with
Boswellia serrata extract H 15]. Z Gastroenterol 2001;39(1):11-17.
17. Singh GB, Atal CK. Pharmacology of an extract of salai guggal ex-Boswellia
serrata, a new non-steroidal anti-inflammatory agent. Agents Actions 1986;18(3-4):
407-412.
18. Kimmatkar N, Thawani V, Hingorani L, et al. Efficacy and tolerability of
Boswellia serrata extract in treatment of osteoarthritis of knee–A randomized double
blind placebo controlled trial. Phytomedicine 2003;10(1):3-7.
19. Safayhi H, Sailer ER, Amnon HP. 5-lipoxygenase inhibition by acetyl-11-
keto-beta-boswellic acid (AKBA) by a novel mechanism. Phytomed 1996;3(1):71-72.
20. Jing Y, Nakajo S, Xia L, et al. Boswellic acid acetate induces differentiation and
apoptosis in leukemia cell lines. Leuk. Res 1999;23(1):43-50.
82 JOURNAL OF HERBAL PHARMACOTHERAPY
21. Reddy GK, Chandrakasan G, Dhar SC. Studies on the metabolism of glyco
-
saminoglycans under the influence of new herbal anti-inflammatory agents. Biochem
Pharmacol 1989;38(20):3527-3534.
22. Ammon HP. [Boswellic acids (components of frankincense) as the active prin
-
ciple in treatment of chronic inflammatory diseases]. Wien Med Wochenschr 2002;
152(15-16):373-378.
23. Ammon HP, Safayhi H, Mack T, et al. Mechanism of antiinflammatory actions
of curcumine and boswellic acids. J Ethnopharmacol. 1993;38(2-3):113-119.
24. Buchele B, Simmet T. Analysis of 12 different pentacyclic triterpenic acids
from frankincense in human plasma by high-performance liquid chromatography and
photodiode array detection. J Chromatogr B Analyt Technol Biomed Life Sci 2003;
795(2):355-362.
25. Buchele B, Zugmaier W, Simmet T. Analysis of pentacyclic triterpenic acids
from frankincense gum resins and related phytopharmaceuticals by high-performance
liquid chromatography. Identification of lupeolic acid, a novel pentacyclic triterpene. J
Chromatogr B Analyt Technol Biomed Life Sci 2003;791(1-2):21-30.
26. Knaus U, Wagner H. Effects of boswellic acid of Boswellia serrata and other
triterpenic acids on the complement system. Phytomedicine 1996;3(1):77-80.
27. Singh GB, Singh S, Bani S. Anti-inflammatory actions of boswellic acids.
Phytomed 1996;3(1):81-85.
28. Wildfeuer A, Neu IS, Safayhi H, et al. Effects of boswellic acids extracted from
a herbal medicine on the biosynthesis of leukotrienes and the course of experimental
autoimmune encephalomyelitis. Arzneimittelforschung 1998;48(6):668-674.
29. Sharma ML, Bani S, Singh GB. Anti-arthritic activity of boswellic acids in bo-
vine serum albumin (BSA)-induced arthritis. Int J Immunopharmacol. 1989;11(6):
647-652.
30. Sharma ML, Khajuria A, Kaul A, et al. Effect of salai guggal ex-Boswellia
serrata on cellular and humoral immune responses and leucocyte migration. Agents
Actions 1988;24(1-2):161-164.
31. Sharma ML, Kaul A, Khajuria A, et al. Immunomodulatory activity of boswellic
acids (pentacyclic triterpene acids) from Boswellia serrata. Phytother Res 1996;
10:107-112.
32. Kesava RG, Dhar SC. Effect of a new non-steroidal anti-inflammatory agent on
lysosomal stability in adjuvant induced arthritis. Ital J Biochem 1987;36(4):205-217.
33. Kesava RG, Dhar SC, Singh GB. Urinary excretion of connective tissue metab
-
olites under the influence of a new non-steroidal anti-inflammatory agent in adjuvant
induced arthritis. Agents Actions 1987;22(1-2):99-105.
34. Singh GB, Singh S, Bani S. Alcoholic extract of salai-guggal ex-Boswellia
serrata, a new natural source NSAID. Drugs Today 1996;32(2):109-112.
35. Chopra A, Lavin P, Patwardhan B, et al. Randomized double blind trial of an
ayurvedic plant derived formulation for treatment of rheumatoid arthritis. J Rheumatol
2000;27(6):1365-1372.
36. Gupta I, Parihar A, Malhotra P, et al. Effects of gum resin of Boswellia serrata
in patients with chronic colitis. Planta Med 2001;67(5):391-395.
Natural Standard Review 83
