Risk assessment for glucosamine and chondroitin sulfate.

Abstract

Glucosamine and chondroitin sulfate are two popular dietary ingredients present in dietary supplements intended to support joint health. A large body of human and animal research suggests that oral intakes of these ingredients, either alone or in combination, reduces joint pain and improves mobility in persons with osteoarthritis. The increased awareness and use of these ingredients in dietary supplements warrant a comprehensive review of their safety. Systematic evaluation of the research designs and data do not provide a basis for risk assessment and the usual safe upper level of intake (UL) derived from it unless the newer methods described as the observed safe level (OSL) or highest observed intake (HOI) are utilized. The OSL risk assessment method indicates that the evidence strongly supports safety at intakes up to 2000 mg/d for glucosamine, and 1200 mg/d for chondroitin sulfate, and these levels are identified as the respective OSL. These values represent the highest levels tested in human clinical trials. The complete absence of adverse effects at these levels supports a confident conclusion of their long-term safety.

Full text

Page 1

Regulatory Toxicology and Pharmacology 47 (2007) 78–83
www.elsevier.com/locate/yrtph
0273-2300/$ - see front matter © 2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.yrtph.2006.07.004
Risk assessment for glucosamine and chondroitin sulfate?
John N. Hathcock¤, Andrew Shao
Council for Responsible Nutrition, 1828 L Street, NW, Suite 900, Washington, DC 20036-5114, USA
Received 19 May 2006
Available online 30 August 2006
Abstract
Glucosamine and chondroitin sulfate are two popular dietary ingredients present in dietary supplements intended to support joint
health. A large body of human and animal research suggests that oral intakes of these ingredients, either alone or in combination, reduces
joint pain and improves mobility in persons with osteoarthritis. The increased awareness and use of these ingredients in dietary supple-
ments warrant a comprehensive review of their safety. Systematic evaluation of the research designs and data do not provide a basis for
risk assessment and the usual safe upper level of intake (UL) derived from it unless the newer methods described as the observed safe level
(OSL) or highest observed intake (HOI) are utilized. The OSL risk assessment method indicates that the evidence strongly supports safety
at intakes up to 2000mg/d for glucosamine, and 1200mg/d for chondroitin sulfate, and these levels are identiWed as the respective OSL.
These values represent the highest levels tested in human clinical trials. The complete absence of adverse eVects at these levels supports a
conWdent conclusion of their long-term safety.
© 2006 Elsevier Inc. All rights reserved.
Keywords: Glucosamine; Chondroitin sulfate; Upper level of intake (UL); Observed safe level (OSL)
1. Introduction
Glucosamine is an aminomonosaccharide and it is the
principal component of O-linked and N-linked glycosami-
noglycans, which form the matrix of all connective tissues,
including cartilage (Harris et al., 2005). The evidence that
orally administered glucosamine compounds may be
eVective in ameliorating pain due to osteoarthritis has
been accumulating for 30–40 years (Deal and Moskowitz,
1999). Dietary supplements in the United States may con-
tain glucosamine hydrochloride, glucosamine sulfate, or
N-acetyl-glucosamine. The raw material for glucosamine
supplements has historically been derived from extraction
of chitin, a component of shellWsh (shrimp, crab, and lob-
ster). Recent technological advances have led to a more
eYcient means of production of a vegetarian source by
fermentation (Almada, 2003). Numerous clinical trials
have investigated the eYcacy of oral glucosamine com-
pounds, often in combination with chondroitin sulfate, in
individuals with osteoarthritis. Long-term (3 year) clinical
trials (Reginster et al., 2001; Pavelka et al., 2002), clinical
trials of various durations (Drovanti et al., 1980; Pujalte
et al., 1980; Muller-Fassbender et al., 1994; Noack et al.,
1994; McAlindon et al., 2004; Clegg et al., 2006) and meta-
analyses (McAlindon et al., 2000; Richy et al., 2003; Zer-
kak and Dougados, 2004); (Poolsup et al., 2005; Towheed
et al., 2005) support the eYcacy and safety of oral gluco-
samine in osteoarthritis. In general, the evidence suggests
that glucosamine, at the range of dosages commonly con-
sumed, is not toxic and produces no recognizable pattern
of adverse eVects. Most of the data relate to a single
intake level, namely 1500mg/d, although this is sometimes
divided into three or more individual doses. While most
published studies have been on the sulfate form, a few
have used the hydrochloride form. One 12-week clinical
trial involved a daily dose of 2000mg of glucosamine
hydrochloride (Braham et al., 2003).
Chondroitin sulfate is a glycosaminoglycan with a poly-
merized disaccharide base linked to a sulfate group, and is
?No funding was speciWc to the production of this manuscript. The sala-
ries for authors were provided by the aYliated organization.
*Corresponding author. Fax: +1 202 204 7980.
E-mail address: jhathcock@crnusa.org (J.N. Hathcock).

Page 2

J.N. Hathcock, A. Shao / Regulatory Toxicology and Pharmacology 47 (2007) 78–83
79
found in the proteoglycans of articular cartilage (Harris
et al., 2005). As a dietary supplement, chondroitin sulfate is
usually derived from bovine trachea, although other
sources such as ovine or porcine trachea and shark skele-
tons (shark cartilage) are also used in some dietary supple-
ments (Nagib, 2003). The concept that orally administered
chondroitin sulfate, along with glucosamine, might slow the
process of osteoarthritis has also been recognized for
decades. Numerous clinical trials have investigated the
eYcacy of oral chondroitin sulfate and/or glucosamine in
individuals with osteoarthritis. Meta-analysis tends to sup-
port the assertion of eYcacy (Rovetta and Monteforte,
1996; Bourgeois et al., 1998; McAlindon et al., 2000; Richy
et al., 2003). In general, the evidence suggests that chondroi-
tin sulfate, at the range of dosages commonly consumed
(usually 1200mg/d), is not toxic and produces no recogniz-
able pattern of adverse eVects. Most of the data relate to
single intake levels, and no systematic study of the dose–
response relationship has been conducted.
The increase in both public awareness and usage of these
ingredients in dietary supplements warrants a comprehen-
sive review of their safety. Most upper safe levels of nutri-
ents and related substances are based on widely applicable
risk assessment models used by the US Food and Nutrition
Board (FNB) in its Dietary Reference Intakes documents
in 1997 and after (Food and Nutrition Board, Institute of
Medicine, 1997, 1998a,b, 2000, 2001). The FNB method and
reviews are a formalization and extension of the quantita-
tive methods widely used earlier in risk assessment of other
substances, and by the food and dietary supplement indus-
tries. Because of the systematic, comprehensive and author-
itative character of the FNB risk assessment method for
nutrients, this approach has gathered widespread support
and adoption by others such as the European Commission
ScientiWc Committee on Food (SCF) (European Commis-
sion, 2001), the United Kingdom Expert Group on Vita-
mins and Minerals (EVM) (Food Standards Agency, 2003)
and more recently by the Food and Agriculture Organiza-
tion/World Health Organization project report A Model for
Establishing Upper Levels of Intake for Nutrients and
Related Substances (FAO/WHO, 2006) with some slight
modiWcations. All these reports reXect the concepts and
procedures established much earlier for the risk assessment
of non-carcinogenic chemicals (National Research Council,
1983).
2. Methods
The safety evaluation method applied to orally administered glucosa-
mine or chondroitin sulfate is that of the Council for Responsible Nutri-
tion (CRN) Vitamin and Mineral Safety, 2nd ed. (Hathcock, 2004), which
contains the basic features of the FNB method and also the observed safe
level (OSL) modiWcation recently adopted as a highest observed intake
(HOI) in the FAO/WHO report.
Overall, this risk analysis was derived from the human clinical trial
database through the following major steps:
1. Derive a safe upper level of intake (UL) if the data are appropriate:
(a) Search for data that identify a hazard related to excessive intake.
(b) Assess the dose–response relationship for the identiWed hazard.
(c) Consider uncertainty and assign an uncertainty factor (UF).
(d)Derive a UL from the no observed adverse eVect level (NOAEL) or
lowest observed adverse eVect level (LOAEL), and the UL D
NOAEL/UF.
2. If no data establish adverse eVects in humans, the above procedure
cannot be used. In these circumstances, the highest intake level with
suYcient evidence of safety is identiWed as a value named the OSL by
CRN and the HOI by FAO/WHO. Uncertainty is considered in selec-
tion of the OSL value and the selection is made with suYcient conser-
vatism to justify assignment of UFD1.0.
We applied the Wrst procedure to the glucosamine and chondroitin sul-
fate human trial data and found no basis for a NOAEL or LOAEL, and
thus could not derive a classical UL. Consequently, we applied the OSL
procedure to the clinical trial data, with the results described in the sec-
tions below.
Due to the nature of the raw material sources for these two ingredi-
ents, there is little, if any dietary contribution, and therefore the OSL
value identiWed from the trials does not require correction for dietary
intakes, and the OSL can be identiWed as a safe upper level for supple-
ments (ULS).
3. ScientiWc evidence related to safety—glucosamine
3.1. Human studies
Publications on the clinical trials of glucosamine for eVec-
tiveness in osteoarthritis also contained much useful infor-
mation relating to safety. None of the clinical trials have
found signiWcant patterns of adverse eVects related to gluco-
samine consumption (Deal and Moskowitz, 1999; Reginster
et al., 2001; Pavelka et al., 2002; Richy et al., 2003; Zerkak
and Dougados, 2004; Clegg et al., 2006). In the clinical trials
of three years duration, substantial numbers of several
diVerent adverse health events occurred in both the placebo
and the treatment groups, but none of the small diVerences
in adverse event frequency approached statistical signiW-
cance (Reginster et al., 2001; Pavelka et al., 2002). The con-
clusions from these studies are further supported by the
absence of signiWcant adverse eVects in other clinical trials
(Drovanti et al., 1980; Pujalte et al., 1980; Muller-Fassb-
ender et al., 1994; Noack et al., 1994; Braham et al., 2003;
McAlindon et al., 2004; Clegg et al., 2006). Human clinical
trial data have shown no cause for concern about the safety
of oral glucosamine at current and plausible intakes (Mathe-
son and Perry, 2003; Bruyere et al., 2004).
Speculation over a causal relationship between gluco-
samine intake and diabetes has led to the investigation of
the possible eVects on insulin function and glucose metab-
olism, but not always with an appropriate experimental
protocol. Infused glucosamine can increase the hexosa-
mine pathway Xux, suggesting a potential adverse eVect of
this supplement on glucose homeostasis (Monauni et al.,
2000). The hexosamine pathway activation leads to deteri-
oration of pancreatic ?-cell function, thereby posing the
possibility that glucosamine could enhance the risk of dia-
betes (Kaneto et al., 2001; Yoshikawa et al., 2002).
Concerns about a possible adverse eVect of glucosamine
on glucose homeostasis or diabetes have prompted direct

Page 3

80
J.N. Hathcock, A. Shao / Regulatory Toxicology and Pharmacology 47 (2007) 78–83
evaluation of these endpoints in clinical trials. One clinical
trial administered a daily dose of 1500mg glucosamine
hydrochloride for 90 days and found no eVects on hemo-
globin A1c concentrations in diabetic subjects (Scroggie
et al., 2003), and another found no eVects of 1500mg/d of
glucosamine sulfate on blood glucose or serum insulin in
normal volunteers after 12 weeks (Tannis et al., 2004).
Thorough review of the evidence on this relation of gluco-
samine to glucose metabolism and function reveals no
adverse eVects (Anderson et al., 2005). Thus, concerns
about a possible diabetogenic eVect of glucosamine that
arose from biochemical studies have been investigated in
clinical trials, and the human data directly demonstrate
that this eVect does not occur in normal or diabetic subjects
who consume 1500mg/d of glucosamine for up to 12 weeks.
Because of the small size of the clinical trials involved, the
possibility of an eVect in sensitive individuals cannot be
entirely excluded.
The highest glucosamine dosage utilized in a double-
blind, placebo-controlled, randomized clinical trial was
2000mg of glucosamine hydrochloride/d for 12 weeks in
subjects with osteoarthritis of the knee (Braham et al.,
2003). Subjects (24 assigned to glucosamine and 22 to pla-
cebo) were monitored for the side eVects of nausea/vomit-
ing, gastrointestinal upset/cramps, headache, bloating, dry
mouth, and tenderness in the knee. The total side eVects
reported were similar, with 11 among the 24 subjects in the
glucosamine group and 10 among the 22 placebo controls,
with no signiWcant diVerences in any category.
The NIH-sponsored glucosamine/chondroitin arthritis
intervention trial (GAIT) involved more than 1500 osteoar-
thritis patients who ingested 1500mg/d glucosamine hydro-
chloride, 1200mg/d chondroitin sulfate, the combination of
the two, 200mg/d of the prescription pain medication
Celebrex™, or placebo for 24 weeks (Clegg et al., 2006).
Adverse eVects were closely monitored throughout the
study period. A total of 634 patients were exposed to gluco-
samine hydrochloride. Results showed no signiWcant diVer-
ence in the incidence of adverse eVects between any of the
treatment arms.
The glucosamine fraction of total weight is higher with
the hydrochloride than with the sulfate; the bioavailability
of both forms exceeds 90%, with glucosamine hydrochlo-
ride approaching 100% (Deal and Moskowitz, 1999; Math-
eson and Perry, 2003; Institute of Medicine, 2005;
Anderson et al., 2005). With these chemical diVerences and
bioavailability similarities, the safety conclusions reached
for hydrochloride can be appropriately and conWdently
extrapolated to the sulfate.
3.2. Animal and in vitro studies
The large number of animal and in vitro studies
addressing the safety as well as the metabolism and meta-
bolic eVects of glucosamine have been reviewed in detail
(Anderson et al., 2005). The LD50 of glucosamine hydro-
chloride is greater than 5000mg/kg, and the NOAEL is
2700mg/kg in rats and 2149mg/kg in dogs (Anderson
et al., 2005). Assuming a 60kg adult body weight, the
1500mg daily dose in humans amounts to 25mg/kg, and
the 2000mg dose equals 33mg/kg. Thus, extrapolation of
the extensive data obtained from the animal and in
in vitro toxicology studies suggests that adverse eVects are
unlikely in humans.
4. Human NOAEL or OSL (HOI)—glucosamine
4.1. NOAEL or LOAEL
None of the clinical trials found adverse eVect related to
glucosamine administration in any form, and therefore
there is, by deWnition, no basis for identifying a LOAEL. In
the absence of a LOAEL, a NOAEL is not usually set.
Without either of these two values the establishment of a
UL is not appropriate (Food and Nutrition Board, Insti-
tute of Medicine, 1998b).
4.2. OSL
The glucosamine dosage that was utilized in most clini-
cal trials seems to be 1500mg/d. The one clinical trial that
used 2000mg of glucosamine hydrochloride found no
adverse eVects. There are ample data to identify 1500mg
of glucosamine sulfate as the OSL. The absence of adverse
eVects in clinical trial at 2000mg of glucosamine hydro-
chloride, together with the huge margins of safety indi-
cated by animal studies and the direct evidence against a
diabetogenic eVect in humans is suYcient grounds for set-
ting the OSL at 2000mg of glucosamine hydrochloride.
Further, the diVerences in glucosamine content and bio-
availability allow this OSL to be applied to glucosamine
sulfate as well.
In one placebo-controlled, double-blind randomized
clinical trial of glucosamine hydrochloride (1500mg) in
combination with chondroitin sulfate (1200mg) two sub-
jects in the active group experienced allergic responses,
compared with none in the placebo group (Nguyen et al.,
2001). The combined treatment prevents attribution of the
eVect to a speciWc ingredient, but allergic responses related
to glucosamine of shell Wsh origin have been reported previ-
ously (Anderson et al., 2005). In the US, supplement prod-
ucts containing glucosamine from this source are required
carry an allergy warning statement (US Food and Drug
Administration, 2004). Glucosamine derived from plant
sources would not need such warnings.
IdentiWcation of 2000mg/d as the OSL for oral con-
sumption of glucosamine (either the hydrochloride or the
sulfate) carries little uncertainty, due to the conWdence
gained from substantial safety in animal and in vitro tests.
The subjects in the clinical trials would have been con-
suming little-to-no glucosamine in their diet, and there-
fore the quantities of glucosamine added in clinical trials
discussed were supplemental amounts well above the very
small quantities potentially consumed in foods. Therefore,

Page 4

J.N. Hathcock, A. Shao / Regulatory Toxicology and Pharmacology 47 (2007) 78–83
81
no correction is required for glucosamine in the food sup-
ply and this risk assessment represents a direct approach
to the safe upper level for supplements (ULS). No correc-
tion is needed for the glucosamine in the food supply. The
OSL for glucosamine set at 2000mg is also identiWed as
the ULS. Allergic warnings are appropriate and required
only for products including glucosamine of shell Wsh
origin.
NOAEL and LOAEL: No toxicological basis.
OSL: 2000mg glucosamine compound (hydrochloride
or sulfate).
ULS: 2000mg/d.
5. ScientiWc evidence related to safety—chondroitin sulfate
5.1. Human studies
Several clinical trials have involved the oral administra-
tion of chondroitin sulfate (Rovetta and Monteforte, 1996;
Bourgeois et al., 1998; Mazieres et al., 2001; Verbruggen
et al., 2002; Rovetta et al., 2004; Uebelhart et al., 2004;
Clegg et al., 2006). The age, health conditions, dosage, dura-
tion, and monitoring and evaluation methods have diVered
greatly. For conWdence in the results, those studies with
stronger designs carry more weight regarding a conclusion
of safety at that dosage. In a risk assessment, the studies
with strong designs and involving higher dosages deserve
greater weight in identifying the highest dosage that can be
conWdently concluded to carry no identiWable risk of
adverse eVects.
The highest oral chondroitin sulfate dosage adminis-
tered in published clinical trials is 1200mg/d (Bourgeois
et al., 1998; Verbruggen et al., 2002; Clegg et al., 2006).
Other trials have utilized dosages of 1000mg (Mazieres
et al., 2001) and 800mg (Rovetta and Monteforte, 1996;
Rovetta et al., 2004; Uebelhart et al., 2004). The number
of subjects in the trials varied from 12 to 635, and the clin-
ical monitoring capable of detecting adverse eVects
ranged from sparse to extensive (e.g., self-reports of possi-
ble adverse eVect to clinical evaluation combined with
extensive hematological and clinical chemistry indices).
None of these clinical trials found any signiWcant adverse
eVects.
The clinical trial by Verbruggen and coworkers (Verb-
ruggen et al., 2002) is especially convincing regarding the
safety of oral chondroitin sulfate. It involved 165 subjects
treated for 3 years with an oral dose of 1200mg/d. The
monitoring included examination by three physicians. The
only adverse eVect reported was a single case of gastritis in
one chondroitin sulfate-treated subject. The subject with-
drawals were fewer among those treated with chondroitin
sulfate compared with placebo-treated controls. These
results are consistent with the most recent trial conducted
on chondroitin sulfate. The NIH-sponsored glucosamine/
chondroitin arthritis intervention trial (GAIT) involved
more than 1500 osteoarthritis patients who ingested
1500mg/d glucosamine hydrochloride, 1200mg/d chon-
droitin sulfate, the combination of the two, 200mg/d of the
prescription pain medication Celebrex™, or placebo for 24
weeks (Clegg et al., 2006). Adverse eVects were closely mon-
itored throughout the study period. A total of 635 patients
were exposed to chondroitin sulfate. Results showed no sig-
niWcant diVerence in the incidence of adverse eVects
between any of the treatment arms. None of the clinical tri-
als found any adverse eVects on clinical chemistry (blood
and urine) or hematological measurements resulting from
oral chondroitin sulfate.
The clinical trial evidence has been the subject of four
published meta-analyses (Leeb et al., 2000; McAlindon
et al., 2000; Edelist and Evans, 2001; Richy et al., 2003)
and one review/commentary (McAlindon, 2001). These
publications focused primarily on the beneWts of oral
chondroitin sulfate in limiting the progression of osteoar-
thritis, but they also have relevance to the safety of this
ingredient. The meta-analyses support the safety of oral
chondroitin sulfate at 1200mg/d, the highest intake sys-
tematically studied.
5.2. Human NOAEL
The absence of adverse eVects at any of the dosages used
in the clinical trials does not support identiWcation of a
LOAEL or NOAEL. The evidence indicates no adverse
eVect of 1200mg/d of oral chondroitin sulfate, but does not
suggest at what dosage adverse eVects might occur. There-
fore there is, by deWnition, no basis for identifying a
LOAEL. In the absence of a LOAEL, a NOAEL is not usu-
ally set. Without either of these two values the establish-
ment of a UL is not appropriate.
6. Human NOAEL or OSL (HOI)—Chondroitin sulfate
6.1. OSL
The highest chondroitin sulfate dosage that has been
utilized in clinical trials is 1200mg/d. There are suYcient
data at this level to identify it as the OSL. The nearly com-
plete absence of any adverse eVects of chondroitin sulfate
within the range of the clinical trials reviewed (800–
1200mg/d) suggest that the highest level, 1200mg/d, is not
a true NOAEL and that any LOAEL is likely to be much
higher. The single case of gastritis among hundreds of
subjects treated suggests that this one case is not causally
related to chondroitin sulfate, or that the individual had a
very unusual sensitivity, and should not inXuence the out-
come of the risk assessment. IdentiWcation of 1200mg/d as
the OSL for oral consumption of chondroitin sulfate up
to three years by adults carries little uncertainty—there
are no known adverse eVects to be avoided. Due to the rel-
atively low consumption of chondroitin sulfate in the diet,
this risk assessment represents a direct approach to the
ULS. Therefore, the OSL of 1200mg is identiWed as the
ULS.

Page 5

82
J.N. Hathcock, A. Shao / Regulatory Toxicology and Pharmacology 47 (2007) 78–83
6.2. Animal studies
Because the human clinical trial data are judged suY-
cient to support identiWcation of an OSL value, the data
from animal experiments were not reviewed.
NOAEL and LOAEL: no toxicological basis
OSL: 1200mg chondroitin sulfate
ULS: 1200mg/d
7. Conclusions
Glucosamine and chondroitin sulfate are two dietary
ingredients whose use continues to increase among Ameri-
cans in the form of dietary supplements. While a substantial
body of evidence exists supporting the beneWcial eVects of
these on joint health, oYcial documentation of their safety
using risk assessment has not, to our knowledge, yet been
published. The present review of the available published
human clinical trial data involving glucosamine and chon-
droitin sulfate, along with several published meta-analyses,
provide for a high level of conWdence in the safety of these
ingredients. The absence of a well-deWned critical eVect for
either ingredient precludes the selection of a NOAEL or
LOAEL, and therefore required use of the observed safe
level (OSL) or highest observed intake (HOI) approach
established by FAO/WHO to conduct this risk assessment.
Evidence from well-designed randomized, controlled
human clinical trials indicates that the ULS for is 2000mg/
d for glucosamine and 1200mg/d for chondroitin sulfate.
References
Almada, A.L., 2003. Glucosamine Shell Game Revisited. In: Functional
Foods and Nutraceuticals.
Anderson, J.W., Nicolosi, R.J., Borzelleca, J.F., 2005. Glucosamine eVects
in humans: a review of eVects on glucose metabolism, side eVects,
safety considerations and eYcacy. Food Chem. Toxicol. 43, 187–201.
Bourgeois, P., Chales, G., Dehais, J., Delcambre, B., Kuntz, J.L., Rozen-
berg, S., 1998. EYcacy and tolerability of chondroitin sulfate 1200 mg/
day vs. chondroitin sulfate 3£ 400mg/day vs. placebo. Osteoarthritis
Cartilage 6 (Suppl. A), 25–30.
Braham, R., Dawson, B., Goodman, C., 2003. The eVect of glucosamine
supplementation on people experiencing regular knee pain. Br. J.
Sports Med. 37, 45–49. discussion 49.
Bruyere, O., Pavelka, K., Rovati, L.C., Deroisy, R., Olejarova, M., Gatter-
ova, J., Giacovelli, G., Reginster, J.Y., 2004. Glucosamine sulfate
reduces osteoarthritis progression in postmenopausal women with
knee osteoarthritis: evidence from two 3-year studies. Menopause 11,
138–143.
Clegg, D.O., Reda, D.J., Harris, C.L., Klein, M.A., O’Dell, J.R., Hooper,
M.M., Bradley, J.D., Bingham 3rd, C.O., Weisman, M.H., Jackson,
C.G., Lane, N.E., Cush, J.J., Moreland, L.W., Schumacher Jr., H.R.,
Oddis, C.V., Wolfe, F., Molitor, J.A., Yocum, D.E., Schnitzer, T.J.,
Furst, D.E., Sawitzke, A.D., Shi, H., Brandt, K.D., Moskowitz, R.W.,
Williams, H.J., 2006. Glucosamine, chondroitin sulfate, and the two in
combination for painful knee osteoarthritis. N Engl. J. Med. 354, 795–
808.
Deal, C.L., Moskowitz, R.W., 1999. Nutraceuticals as therapeutic agents in
osteoarthritis. The role of glucosamine, chondroitin sulfate, and colla-
gen hydrolysate. Rheum. Dis. Clin. North Am. 25, 379–395.
Drovanti, A., Bignamini, A.A., Rovati, A.L., 1980. Therapeutic activity of
oral glucosamine sulfate in osteoarthrosis: a placebo-controlled dou-
ble-blind investigation. Clin. Ther. 3, 260–272.
Edelist, D.D., Evans, M.F., 2001. Do glucosamine and chondroitin treat
the symptoms of osteoarthritis? Can. Fam. Physician 47, 275–277.
ScientiWc Committee on Food, 2001. Guidelines of the ScientiWc Commit-
tee on Food for the Development of Tolerable Upper Intake Levels for
Vitamins and Minerals. European Commission, Brussels, Belgium.
FAO/WHO, 2006. A Model for Establishing Upper Levels of Intake for
Nutrients and Related Substances. In: FAO/WHO Technical Work-
shop on Nutrient Risk Assessment, Geneva, Switzerland.
Food Standards Agency, 2003. Expert Group on Vitamins and Minerals,
2003. Safe Upper Levels for Vitamins and Minerals. Food Standards
Agency, United Kingdom.
Food and Nutrition Board, Institute of Medicine, 1997. Dietary Reference
Intakes for Calcium, Phosphorus, Magnesium, Vitamin D, and Fluo-
ride. National Academy Press, Washington, DC.
Food and Nutrition Board, Institute of Medicine, 1998a. Dietary Refer-
ence Intakes for Thiamin, RiboXavin, Niacin, Vitamin B6, Folate, Vita-
min B12, Pantothenic Acid, Biotin and Choline. National Academy
Press, Washington, DC.
Food and Nutrition Board, Institute of Medicine, 1998b. Dietary Refer-
ence Intakes: A Risk Assessment Model for Establishing Upper Intake
Levels for Nutrients. National Academy Press, Washington, DC.
Food and Nutrition Board, Institute of Medicine, 2000. Dietary Reference
Intakes for Vitamin C, Vitamin E, Selenium and Carotenoids. National
Academy Press, Washington, DC.
Food and Nutrition Board, Institute of Medicine, 2001. Dietary Reference
Intakes for Vitamin A, Vitamin K, Arsenic, Boron, Chromium, Cop-
per, Iodine, Iron, Manganese, Molybdenum, Nickel, Silicon, Vanadium
and Zinc. National Academy Press, Washington, DC.
Harris, E.D., Budd, R.C., Firestein, G.S., Genovese, M.C., Sergent, J.S.,
Ruddy, S., Sledge, C.B. 2005. Structure and Function of Bone, Joints,
and Connective Tissue, Kelley’s Textbook of Rheumatology.
Hathcock, J., 2004. Vitamin and Mineral Safety. Council for Responsible
Nutrition, Washington, DC.
Institute of Medicine, 2005. Glucosamine: Prototype Monograph Sum-
mary. Dietary Supplements: A Framework for Evaluating Safety.
National Academy Press, Washington, DC, pp. 363–366, Appendix E.
Kaneto, H., Xu, G., Song, K.H., Suzuma, K., Bonner-Weir, S., Sharma, A.,
Weir, G.C., 2001. Activation of the hexosamine pathway leads to dete-
rioration of pancreatic beta-cell function through the induction of oxi-
dative stress. J. Biol. Chem. 276, 31099–31104.
Leeb, B.F., Schweitzer, H., Montag, K., Smolen, J.S., 2000. A metaanalysis
of chondroitin sulfate in the treatment of osteoarthritis. J. Rheumatol.
27, 205–211.
Matheson, A.J., Perry, C.M., 2003. Glucosamine: a review of its use in the
management of osteoarthritis. Drugs Aging. 20, 1041–1060.
Mazieres, B., Combe, B., Phan Van, A., Tondut, J., Grynfeltt, M., 2001.
Chondroitin sulfate in osteoarthritis of the knee: a prospective, double
blind, placebo controlled multicenter clinical study. J. Rheumatol. 28,
173–181.
McAlindon, T.E., LaValley, M.P., Gulin, J.P., Felson, D.T., 2000. Glucosa-
mine and chondroitin for treatment of osteoarthritis: a systematic
quality assessment and meta-analysis. JAMA 283, 1469–1475.
McAlindon, T., 2001. Glucosamine and chondroitin for osteoarthritis?
Bull. Rheum. Dis. 50, 1–4.
McAlindon, T., Formica, M., LaValley, M., Lehmer, M., Kabbara, K.,
2004. EVectiveness of glucosamine for symptoms of knee osteoarthri-
tis: results from an internet-based randomized double-blind controlled
trial. Am. J. Med. 117, 643–649.
Monauni, T., Zenti, M.G., Cretti, A., Daniels, M.C., Targher, G., Caruso,
B., Caputo, M., McClain, D., Del Prato, S., Giaccari, A., Muggeo, M.,
Bonora, E., Bonadonna, R.C., 2000. EVects of glucosamine infusion on
insulin secretion and insulin action in humans. Diabetes 49, 926–935.
Muller-Fassbender, H., Bach, G.L., Haase, W., Rovati, L.C., Setnikar, I.,
1994. Glucosamine sulfate compared to ibuprofen in osteoarthritis of
the knee. Osteoarthritis Cartilage 2, 61–69.