Copyright 2005 by Humana Press Inc.

All rights of any nature, whatsoever, reserved.

0163-4984/05/106020123 $30.00

Synovial Fluid and Plasma Selenium,

Copper, Zinc, and Iron
Concentrations in Patients
with Rheumatoid Arthritis
and Osteoarthritis
M. YAZAR,1 S. SARBAN,*,1 A. KOCYIGIT,2 AND U. E. ISIKAN1
Departments of 1Orthopaedic Surgery and 2Biochemistry,
Harran University, Medical Faculty, 63200 Sanliurfa-Turkey
Received October 21, 2004; Accepted October 29, 2004

ABSTRACT
In recent years, a great number of studies have investigated the possible
role of trace elements in the etiology and pathogenesis of rheumatoid arthri-
tis (RA) and osteoartritis (OA). We studied synovial fluid and plasma con-
centrations of selenium (Se), zinc (Zn), copper (Cu), and iron (Fe) in patients
with RA and OA and compared them with sex- and age-matched healthy
subjects. Plasma albumin levels were measured as an index of nutritional sta-
tus. Plasma Se, Cu, and Zn concentrations were determined by atomic
absorption spectrophotometry and Fe concentrations were determined by the
colorimetric method. Although plasma and synovial fluid Se concentration
were found to be significantly lower (p < 0.05, and p < 0.05, respectively), Cu
concentrations were significantly higher in patients with RA than those of
healthy subjects and OA (p < 0.05 and p < 0.05, respectively). There were no
significant differences in plasma and synovial fluid Zn concentrations and
albumin levels among three groups (p > 0.05). On the other hand, synovial
fluid Cu and Fe concentrations were significantly higher in patients with OA
than those of healthy subjects (p < 0.05). There was a significantly positive cor-
relation between synovial fluid SeCu values and ZnFe values in patients
with RA. Our results showed that synovial fluid and plasma trace element
concentrations, excluding Zn, change in inflammatory RA, but not in OA.
These alterations in trace element concentrations in inflammatory RA might
be a result of the changes of the immunoregulatory cytokines.
Index Entries: Rheumatoid arthritis; osteoarthritis; selenium; zinc;
copper; iron.

*Author to whom all correspondence and reprint requests should be addressed.

Biological Trace Element Research 123 Vol. 106, 2005

124 Yazar et al.

INTRODUCTION
Rheumatoid arthritis (RA) is a chronic inflammatory arthropathy
that can affect most synovial joints and has manifestations in many
organs other than the locomotor system. Chronic inflammation is the
defining feature of RA. There is a constant traffic of cells into the
inflammed synovium and the joint space. Large numbers of neutrophils
from the circulatory system move through the synovial tissue into the
synovial fluid during active phase of the disease. Other immune system
cells invade the synovium; cellular invasion is also evident at the pan-
nuscartilage junction. Although the etiology and pathogenesis of RA is
still unknown despite progress in medicine, especially in immunogenet-
ics, the pro-inflammatory (tumor necrosis factor [TNF]-, interleukin
[IL]-1, IL-6, TNF-) and immunoregulatory cytokines (IL-2, interferon
[IFN]-, IL-4, IL-5, IL-12, IL-10, IL-15, IL-17) could play a major role.
Thus, the most widely accepted model to explain the pathogenesis of RA
invokes antigen-specific T-cells that orchestrate chronic synovial inflam-
mation. The final outcome is usually total destruction of the articular car-
tilage and loss of joint function (1,2). Osteoarthritis (OA), also referred to
as degenerative joint disease, consists of generally progressive loss of
articular cartilage accompanied by attempted repair of articular carti-
lage, remodeling, and sclerosis of subchondral bone, and, in many
instances, the formation of subchondral bone cysts and marginal osteo-
phytes. Pathologically, the disease is characterized by fissuring and focal
erosive cartilage lesions, cartilage loss, and destruction. Although OA
accepted as a noninflammatory arthritis, the mechanisms responsible for
it remain poorly understood (3).
Selenium (Se), zinc (Zn), copper (Cu), and iron (Fe) are essential trace
elements and the plasma contents of these nutrients change during the
course of most infection and inflammation (4). However, it is not exactly
known yet whether the causes of these changes are as a result of specific
deficiency from dietary inadequacies and imbalances or a part of inflam-
matory response of an organism that is regulated by some cytokines.
Despite progress in the research, there is also relatively little known
regarding the pathogenesis of OA and RA diseases. In generally, OA is
accepted as a noninflammatory degenerative joint disease, whereas RA is
accepted as an inflammatory one. Articular cartilage damage is an impor-
tant pathologic feature of OA; on the other hand, synovial proliferation,
inflammation, and pro-inflammatoryimmunoregulatory cytokines play a
central part in RA expression (13). Because of the metabolically active
nature of RA, trace element alterations in plasma and synovial fluid can be
expected.
Several investigators have found depressed plasma or serum Se val-
ues in RA (59), whereas Peretz et al. (10) and Gambhir and Lali (11) have
reported normal plasma Se concentrations. On the other hand, normal (12)
and decreased (1317) Zn concentrations of plasma and red cells in RA

Biological Trace Element Research Vol. 106, 2005

Essential Elements in Rheumatoid Arthritis 125

patients have been reported by several investigators. The plasma Cu level

increases in all inflammatory processes as well as in RA. Scudder et al. (18)
and Tuncer et al. (19) reported high concentrations of plasma Cu in RA
patients. As seen in the literature, the alterations in trace element concen-
trations in the plasma and synovial fluid of patients with RA are inconsis-
tent and, to our knowledge, there are no available reports of the synovial
fluid selenium level in these patients.
The purpose of this study is to evaluate the status of plasma and syn-
ovial fluid essential trace elements Se, Zn, Cu, and Fe concentrations and
to assess any association and correlation between synovial fluid and
plasma concentrations of these elements in patients with RA and OA and
compare them age-matched healthy subjects.

MATERIALS AND METHODS

Subjects
The study plan was approved by the Ethics Committee of the medical
faculty and all subjects volunteered for the trial. The control group of
healthy subjects consisted of 25 individuals (13 females, 12 males), aged
3957 yr (mean age: 48.2 6.2). The RA subjects were 25 individuals (14
females, 11 males), aged 4259 yr (mean age: 47.4 7.3). The OA patients
consisted of 25 individuals (13 females, 12 males), aged 4262 yr (mean
age: 49.3 8.2). The patients with RA were diagnosed by the diagnostic cri-
teria of the American Rheumatism Association, and those who had been
receiving corticosteroid agents, gold preparations, and/or D-penicillamine
for at least 3 mo before the consultation were excluded from this study.
However, patients who had been receiving ordinary dosages of non-
steroidal anti-inflammatory drugs (NSAIDs) were not excluded, as almost
all the patients with OA and the vast majority of the patients with RA had
been undergoing NSAID treatment.
All of the materials (glass and plastic) employed were thoroughly
cleaned with a hot solution of nitric acid (20% v/v) for 48 h and rinsed
three times with ultradeionized water. Disinfectant solution was used for
cleaning all containers and glassware in which the plasma and synovial
samples were placed.

Blood Samples
About 10 mL of whole blood was collected from the cubital vein of
each patient by a heparinized vacuum syringe in the morning before
breakfast. The samples were centrifuged at 3000g for 10 min at 4C, and the
plasma and buffy coat were carefully separated. Plasma were separated to
determine Se, Cu, Zn, and Fe concentrations and albumin level and kept
at 80C until they were assayed.

Biological Trace Element Research Vol. 106, 2005

126 Yazar et al.

Table 1
Furnace Temperature Program for Se

Synovial Fluid Samples

All of the synovial fluids used in this investigation had been aspi-
rated aseptically from the joint of patients attending the orthopedic and
physical medicine clinics of the hospital. The particulate material and cel-
lular content of all fluids were removed by centrifugation at 3000g for 15
min. The supernatants were collected and maintained at 80C until used.
Plasma and synovial fluid Se determination was performed by an
atomic absorption spectrophotometer (AAS) with a graphite furnace (Var-
ian GTA-97, Australia), with a deuterium background correction. A Varian
hollow-cathode lamp was employed at the 196-nm wavelength and 1.0-
nm bandpass. Pyrolytically coated graphite tubes with pyrolytic graphite
platforms (Varian, Australia) were used as the furnace. Se concentrations
were determined by an internal standard addition method, as previously
described (20). Plasma was diluted (1 : 4) with 0.05% Triton X-100 in
0.125% (v/v) nitric acid. All determinations were run in duplicate, and
individual values were averaged. By means of an autosampler, 10 L of the
solution was dispensed on the atomizer platform together with 10 L of 1
g/mL nickel nitrate and 2 L 2% (w/v) of ascorbic acid solution. The
temperature program of the furnace is shown Table 1. Absorption readings
were measured for the peak height. The variation coefficient for replicate
measurement was <3%. The lowest threshold Se detection of the instru-
ment was 10 L. The accuracy and precision of the procedure were regu-
larly checked with commercial samples with recommended Se contents
(Seronorm serum; Nycomed AS, Oslo, Norway). Plasma and synovial Se
values are expressed as microgram of Se per liter (g/L).
Plasma samples were diluted with deionized water for Cu and Zn
measurements. Cu and Zn concentrations were determined by a Zeeman
AAS (Varian Spectr AA 250 Plus, Australia) with a deuterium background
correlation. Plasma and synovial fluid Zn and Cu values are expressed as
micrograms per deciliter.
Plasma Fe concentration and albumin levels were determined by a
colorimetric method with commercial kits (Abbott USA) using an auto-
matic analyzer (Aeroset, USA).

Biological Trace Element Research Vol. 106, 2005

Essential Elements in Rheumatoid Arthritis 127

Table 2
Physical Characteristics of RA and OA Patients and Healthy Subjects

Note: The values represent the mean SD.

Statistical Analysis
Results were expressed as mean and standard deviation (SD). Statis-
tical analysis were carried out using the SPSS program (version 10.0 soft-
ware; SPSS Inc. Chicago, IL, USA). For comparison of groups, variance
analysis (one-way ANOVA), post hoc LSD test, and Pearson correlation
test were used. p-Values of less than 0.05 were regarded as significant.

RESULTS
Rheumatoid arthritis and OA patients and healthy subjects were similar
in age, height, body weight, and body mass index (BMI), as seen in Table 2.
Plasma Se concentrations were significantly lower (p < 0.05) in
patients with RA than in healthy subjects and OA patients. Also, plasma
Cu concentrations were significantly higher (p < 0.05) in RA patients than
in healthy subjects and OA patients. The plasma Zn concentrations were
not statistically different (p > 0.05). Plasma Fe concentrations were lower in
RA patients than in healthy subjects and OA patients, but the difference
was not significant (p > 0.05) (see Table 3).
As shown in Table 4, synovial fluid Se concentrations were found to
be significantly lower in RA patients than those in the healthy and OA
patients (p < 0.05). The synovial fluid Zn concentrations were not statisti-
cally different (p > 0.05). Synovial fluid Cu concentrations of both RA and
OA patients were significantly higher than those in the healthy group (p
< 0.05). Synovial fluid Fe concentrations were increased significantly in
OA patients than those in RA patients and healthy subjects (p < 0.05).
There were positive correlations between synovial fluid Fe and Zn concen-
trations (r = 0.730, p = 0.026) and Se and Cu concentrations (r = 0.418, p = 0.016)
in RA patients. Also, there was a positive correlation between synovial fluid Cu
and Zn concentrations (r = 0.463, p = 0.023) in OA patients (see Table 5).

Biological Trace Element Research Vol. 106, 2005

128 Yazar et al.

Table 3
Plasma Se, Zn, Cu, and Fe Concentrations and Albumin Levels in Patients
with RA and OA and Healthy Subjects

Note: The values represent the mean SD.

* Significance was defined as p < 0.05.
a Statisticaly significant, compares RA patients (p < 0.05).

Table 4
Synovial Fluid Se, Zn, Cu, and Fe Concentrations in Patients
with RA and OA and Healthy Subjects

Note: The values represent the mean SD.

* Significance was defined as p < 0.05.
a Statisticaly significant, compares healthy subjects and RA patients (p < 0.05).
b Statisticaly significant, compares healthy subjects and OA patients (p < 0.05).
c Statisticaly significant, compares RA and OA patients (p < 0.05).

DISCUSSION
Trace elements are the cofactors of the most enzymes in the organism.
They are found very less in the body, but deficiencies of them could cause
serious problems. Two general classes of abnormality associated with trace
elements are encountered: one is a result of a specific deficiency from

Biological Trace Element Research Vol. 106, 2005

Essential Elements in Rheumatoid Arthritis 129

Table 5
Correlations Among Synovial Fluid Se, Zn, Cu, and Fe Concentrations
in Patients with RA and OA and Healthy Subjects

* Significance was defined as p < 0.05.

dietary inadequancies and imbalances and the second is abnormality

related to other diseases. Both kinds of abnormality can be diagnosed by
analyses of trace elements in plasma or other tissues. Secondary changes
of trace elements that occur in inflammatory and noninflammatory arthri-
tis were studied since the 1970s, but the changes are not exactly under-
stood (519). Therefore, we have studied the status of plasma and synovial
fluid concentrations of Se, Zn, Cu, and Fe in patients with RA and OA.
In our study, synovial fluid and plasma Se concentrations in the RA
group were significantly lower than those in the healthy and OA groups of
patients (p < 0.05), and malnutrition was not a problem in the RA and OA
group because plasma albumin levels, BMI, weight, and height were not
statistically different between the patients and healthy subjects. This can
be a sign of depletion or redistribution of Se from the plasma pool into the
other tissues as a defense mechanism that it might be modulated by pro-
inflammatory and immunoregulatory cytokines. Our result that lower
plasma Se level is in agreement with most other studies previously pub-
lished (513), except that by Peretz et al. (10) and Gambhir and Lali (11).
They reported normal plasma Se concentrations in RA.
Selenium concentrations were rarely were measured in other body
compartments, although such studies would give additional information
on the status and distribution of this element. Synovial fluid Se concentra-
tions in the RA group were significantly lower than those in the healthy
subjects and OA groups of patients (p < 0.05). To the best of our knowl-
edge, there is no available document on synovial fluid Se concentration in
English literature. Erythrocyte Se was assessed in only a few studies; low
or normal values were reported. McConnell et al. (21) reported that syn-
ovial tissue Se concentrations in the RA group were increased significantly.
Plasma Cu level increases in all inflammatory processes, as well as in
RA. Acute-phase reactants, one of them ceruloplasmin, are released during
inflammation and the ceruloplasmin release is associated with the
increased plasma Cu level (22,23). We found a significant increase in the
plasma Cu concentration in patients with RA compared to healthy subjects
(p < 0.05). Our result related to plasma Cu level showed correspondence
with the results of Scudder et al. (18) and Tuncer et al. (19). We also found

Biological Trace Element Research Vol. 106, 2005

130 Yazar et al.

that synovial fluid Cu concentrations were significantly higher in RA

patients when compared to healthy subjects and OA (p < 0.05 and p < 0.05
respectively). The increased plasma/synovial Cu could be attributable to
inflammation associated with the RA disaese.
The alterations in the plasma and the erythrocyte Zn concentrations in
RA patients were reported by several authors (12,24). In general, during
the acute-phase response of inflammation, the serum Zn level decreases. It
is believed that this occurs secondary to the sequestration by the intracel-
lular metal-binding protein in the liver and probably in other tissues.
Balogh et al. (25) have reported that the treatment with NSAIDs decreases
the plasma Zn concentrations; similar changes are seen with levamisole
and penicillamine. The authors reported that these alterations were related
to the different disease activity and pharmacological treatment. In our
study, the plasma Zn concentrations of the patients were similar in all
three groups. Our results related to plasma Zn concentrations showed cor-
respondence with the results of Alegre et al. (12). We thought that this was
correlated with the chronicity of the RA patients and the cessation of the
drugs 3 mo before taking the blood samples.
In our study, plasma and synovial fluid Fe concentrations of RA
patients showed slight and statistically insignificant decreases when com-
pared to healthy subjects (p > 0.05). This might explain why none of our
patients had deep anemia. We have found that synovial fluid Fe is
increased significantly in OA patients compared with levels in RA patients
and healthy subjects. We conclude that local factors related to articular car-
tilage degradation are of major importance in the deposition of Fe. Ogilvie-
Harris and Fornaiser (26) had examined the Fe deposition in the synovia
of patients with RA and OA. They reported synovial tissue Fe concentra-
tions in both RA and OA groups were increased significantly.
Although there are important alterations in plasma and synovial con-
centrations of Se and Cu in RA, these alterations are not seen in OA. The
steady-state status of plasma essential trace element concentrations in
patients with OA can be explained by the noninflammatory nature of the
disease. The causes of these changes in RA might not be a result of a spe-
cific deficiency/excess from dietary inadequacies or imbalances, but could
be a result of the inflammatory status, which is regulated by pro-inflam-
matory and immunoregulatory cytokines. The increase in levels of these
cytokines, especially IL-1, TNF-, and IL-6, in the setting of active RA might
alter the plasma and synovial fluid concentrations of these elements by
inducing the production of metal-binding proteins and metallothioneins.

REFERENCES
1. A. D. Recklies, A. R. Poole, S. Banerjee, et al., Pathophysiologic aspects of inflammation
in diarthroidal joints, in Orthopaedic Basic Science, 2nd ed., J. A. Buckwalter, T. A. Ein-
horn, and S. R. Simon, eds., AAOS, Rosemont, IL, pp. 489530 (2000).

Biological Trace Element Research Vol. 106, 2005

Essential Elements in Rheumatoid Arthritis 131

2. E. D. Harris, Pathogenesis of rheumatoid arthritis, in Textbook of Rheumatology, W. N.

Kelley, ed., W. B. Saunders, Philadelphia, pp. 905942 (1989).
3. H. J. Mankin, V. C. Mow, and J. A. Buckwalter, Articular cartilage repair and osteoarthri-
tis, in Orthopaedic Basic Science, 2nd ed., J. A. Buckwalter, T. A. Einhorn, and S. R. Simon,
eds., AAOS, Rosemont, IL, pp. 471488 (2000).
4. A. J. Mastousek, J. L. Burguera, M. Burguera, and N. Anez, Change in total content of
iron, copper and zinc in serum, heart, liver, spleen and skeletal muscle tissues of rats
infected with Trypanosoma Cruzi, Biol. Trace Element Res. 37, 5169 (1993).
5. J. R. ODell, S. Lemley-Gillespie, W. R. Palmer, A. L. Weaver, G. F. Moore, and L. W.
Klassen, Serum selenium concentrations in rheumatoid arthritis, Ann. Rheum. Dis. 50,
376378 (1991).
6. J. Aaseth, E. Munthe, O. Forre, and E. Steinnes, Trace elements in serum and urine of
patients with rheumatoid arthritis, Scand. J. Rheumatol. 7, 237240 (1978).
7. U. Tarp, Selenium glutathione peroxidase in rheumatoid arthritis, Br. J. Rheumatol. 29,
158 (1990).
8. U. Tarp, Selenium in rheumatoid arthritis, Analyst 120, 877881 (1995).
9. K. Kse, P. Dogan, Y. Kardas, and R. Saraymen, Plasma selenium levels in rheumatoid
arthritis, Biol. Trace Element Res. 53, 5156 (1996).
10. A. Peretz, J. Neve, F. Vertongen, J. P. Famaey, and L. Molle, Selenium status in relation
to clinical variables and corticosteroid treatment in rheumatoid arthritis, J. Rheumatol.
14, 11041147 (1987).
11. J. K. Gambhir and P. Lali, Blood selenium levels in healthy Indian subjects and patients
with rheumatoid arthritis, Clin. Biochem. 32, 665666 (1999).
12. C. Alegre, J. Baro, and J. Obach, Zinc and rheumatic disease, Arthritis Rheum. 27, 10731074
(1984).
13. H. Morgenstern and I. Machtey, Serum zinc and copper levels in rheumatoid arthritis,
Arthritis Rheum. 26, 933934 (1983).
14. K. L. Svenson, R. Hallgren, E. Johansson, and U. Lindh, Reduced zinc in peripheral
blood cells from patients with inflammatory connective tissue diseases, Inflammation
9(2), 189199 (1985).
15. Y. Naveh, D. Schapira, Y. Ravel, E. Geller, and Y. Scharf, Zinc metabolism in rheuma-
toid arthritis: plasma and urinary zinc and relationship to disease activity, J. Rheumatol.
24, 643646 (1997).
16. H. Mussalo-Rauhamaa, Y. T. Konttinen, J. Lehto, and V. Honkanen, Predictive clinical
and laboratory parameters for serum zinc and copper in.rheumatoid arthritis, Ann.
Rheum. Dis. 47, 816819 (1988).
17. P. Dore-Duffy, M. Peterson, F. Catalanotto, et al., Zinc profiles in rheumatoid arthritis,
Clin. Exp. Rheumatol. 8, 541546 (1990).
18. P. R. Scudder, D. Al-Timimi, W. McMurray, A. G. White, B. C. Zoob, and T. L. Dor-
mandy, Serum copper and related variables in rheumatoid arthritis, Ann. Rheum. Dis.
37, 6770 (1978).
19. S. Tuncer, A. Kamanli, E. Akcil, G. O. Kavas, B. Seckin, and M. B. Atay, Trace element
and magnesium levels and superoxide dismutase activity in rheumatoid arthritis, Biol.
Trace Element Res. 68, 137142 (1999).
20. V. Korunova, Z. Skodova, J. Dedina, et al., Serum selenium in adult Czechoslovak (cen-
tral Bohemia) population, Biol. Trace Element Res. 37, 9199 (1993).
21. K. P. McConnell, W. L. Broghamer, Jr., A. J. Blotcky, and O. J. Hurt, Selenium levels in
human blood and tissues in health and in disease, J. Nutr. 105, 10261031 (1975).
22. M. Marrella, U. Moretti, M. Pasqualicchio, et al., Plasma and total blood cell copper in
rheumatoid arthritis, Agents Actions 29, 120121 (1990).
23. H. Ozgunes, H. Gurer, and S. Tuncer, Correlation between plasma malondialdehyde and
ceruloplasmin activity values in rheumatoid arthritis, Clin. Biochem. 28, 193194 (1995).

Biological Trace Element Research Vol. 106, 2005

132 Yazar et al.

24. H. Morgenstern and I. Machtey, Serum zinc and copper levels in rheumatoid arthritis,
Arthritis Rheum. 26, 933934 (1983).
25. Z. Balogh, A. F. El-Ghobarey, G. S. Fell, D. H. Brown, J. Dunlop, and W. C. Dick, Plasma
zinc and its relationship to clinical symptoms and drug treatment in rheumatoid arthri-
tis, Ann. Rheum. Dis. 39, 329332 (1980).
26. D. J. Ogilvie-Harris and V. L. Fornaiser, Synovial iron deposition in osteoarthritis and
rheumatoid arthritis, J. Rheumatol. 7, 3036 (1980).

Biological Trace Element Research Vol. 106, 2005

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.