Clin Rheumatol
DOI 10.1007/s10067-016-3363-1
ORIGINAL ARTICLE
Improved WOMAC score following 16-week treatment
with bromelain for knee osteoarthritis
Thitima Kasemsuk 1 & Nadhaporn Saengpetch 2 & Nathawut Sibmooh 1 &
Supeenun Unchern 1
Received: 28 January 2016 / Revised: 12 July 2016 / Accepted: 17 July 2016
# International League of Associations for Rheumatology (ILAR) 2016
Abstract Treatment with bromelain-containing enzyme
preparation for 34 weeks is effective for treatment of knee
osteoarthritis (OA). Here, we aimed to assess 16-week treat-
ment with bromelain in mild-to-moderate knee OA patients.
We performed a randomized, single-blind, active-controlled
pilot study. Forty knee OA patients were randomized to re-
ceive oral bromelain (500 mg/day) or diclofenac (100 mg/
day). Primary outcome was the Western Ontario and
McMaster Universities Osteoarthritis Index (WOMAC) ana-
lyzed by Wilcoxon signed rank test. Secondary outcome was
the short-form 36 (SF-36). Plasma malondialdehyde (MDA)
and nitrite were measured as oxidative stress markers. There
was no difference in WOMAC and SF-36 scores compared
between bromelain and diclofenac groups after 4 weeks. At
week 4, the improvement of total WOMAC and pain sub-
scales from baseline was observed in both groups; however,
two patients given diclofenac had adverse effects leading to
discontinuation of diclofenac. However, observed treatment
difference was inconclusive. At week 16 of bromelain treat-
ment, the patients had improved total WOMAC scores (12.2
versus 25.5), pain subscales (2.4 versus 5.6), stiffness sub-
scales (0.8 versus 2.0), and function subscales (9.1 versus
17.9), and physical component of SF-36 (73.3 versus 65.4)
as compared with baseline values. OA patients had higher
plasma MDA, nitrite, and prostaglandin E2 (PGE2) in lipo-
polysaccharide (LPS)-stimulated whole blood but lower
* Supeenun Unchern
supeenun.unc@mahidol.ac.th
1
Department of Pharmacology, Faculty of Science, Mahidol
University, 272 Rama 6 Road, Bangkok 10400, Thailand
2
Department of Orthopedics, Faculty of Medicine, Ramathibodi
Hospital, Mahidol University, Rama 6 Road, Bangkok 10400,
Thailand
plasma -tocopherol than control subjects. Plasma MDA
and LPS-stimulated PGE2 production were decreased at week
16 of bromelain treatment. Bromelain has no difference in
reducing symptoms of mild-to-moderate knee OA after
4 weeks when compared with diclofenac.
Keywords Bromelain . Diclofenac . Knee osteoarthritis .
Malondialdehyde . SF-36 . WOMAC
Introduction
Osteoarthritis (OA) commonly affects the knee joints causing
joint pain in older adults. It is a degenerative disease of the
cartilage but also involves synovial inflammation. Increased
mononuclear infiltration and expression of inflammatory me-
diators in synovial tissue are reported in OA [1].
Non-steroidal anti-inflammatory drugs (NSAIDs) can re-
duce joint swelling and pain effectively in OA patients. The
incidence of adverse events especially gastrointestinal (GI)
disturbances associated with prolonged NSAID use is high
[2]. GI symptoms such as dyspepsia, nausea, vomiting, ab-
dominal pain, and heartburn are common adverse GI effects,
which are reported in up to 40 % of NSIAD users [3].
Advanced age (65 years), previous GI symptoms, and con-
current aspirin use increase the risk of upper GI complications
by three- to fourfold in OA patients receiving NSAIDs [4].
Thereby, an effective and safer alternative is necessary for
long-term treatment in OA patients.
Bromelain is derived from the stem and fruit of pineapple
plant. It contains a mixture of proteolytic enzymes with anti-
inflammatory and analgesic properties [5, 6]. Bromelain was
first reported to be useful for treatment of rheumatoid arthritis
and OA in 1964 [7]. Currently, bromelain is a food supple-
ment used for treatment of sport injuries and acute

inflammation [8]. Treatment of knee OA by oral enzyme mix-
ture containing bromelain for 34 weeks is as effective as
diclofenac [912]. Bromelain treatment for 4 weeks is also
effective for treatment of mild knee pain in healthy adults
[13]. However, treatment with bromelain (800 mg/day) for
12 weeks is not effective for moderate-to-severe knee OA
[14]. Here, the clinical efficacy and safety of 16-week treat-
ment with 500-mg/day bromelain were investigated in mild-
to-moderate knee OA. The effects of bromelain on oxidative
stress and LPS-induced PGE2 production were also tested.
Methods
Patients
Forty-three patients who were diagnosed with primary knee
OA according to the American College of Rheumatology
(ACR) Classification Criteria were recruited [15] from ortho-
pedic OPD, Ramathibodi Hospital between September 2010
and December 2011. The inclusion criteria were (1) the sub-
jects had knee pain and (2) radiographic osteophytes appeared
at the margins of the joint with at least one of the following
features: (i) age 50 years, (ii) morning stiffness <30 min, and
(iii) crepitus on motion. The exclusion criteria were history of
peptic ulcer, intra-articular injection of steroid/hyaluronic acid
within 3 months, inability to walk, hypersensitivity to
NSAIDs, or concomitant use of anti-inflammatory or analge-
sic drugs. According to the Kellgren-Lawrence Grading
System [16], 57.5 and 42.5 % of our patients were classified
as score 1 and 2, respectively. Thereby, these patients were
classified as having mild-to-moderate knee OA. In addition,
20 age and sex matched healthy volunteers were enrolled.
Sample size calculation was based on mean and standard de-
viation of serum nitrite concentration in knee OA patients
(0.213 0.21 mol/l) and controls (0.142 0.06 mol/l)
[17]. A sample size of n = 13 was calculated with levels
of 0.05 and a statistical power of 0.80.
This study was approved by the Ramathibodi Hospital
Ethics Committee, Faculty of Medicine Ramathibodi
Hospital, Mahidol University, in accordance with
Declaration of Helsinki. All patients provided the written in-
formed consent before randomization.
Treatment
Forty-three patients were recruited, and three patients
discontinued the study (Fig. 1). Forty patients were blindly
randomized to receive bromelain (General Nutrition
Corporation, Pittsburgh, PA, USA) at dose 500 mg (2000
gelatin digestion units/g) per day, once daily for 16 weeks
(n = 20), or diclofenac sodium (Remedica, Bangkok,
Thailand) at dose 100 mg per day, once daily for 4 weeks
Clin Rheumatol
(n = 20). Participants, examining physician, and staff nurse
were blinded to group assignment. The researcher was un-
blinded. The random allocation sequence was generated using
a computer program by researcher. Patients were randomly
assigned to either diclofenac or bromelain group in a 1:1 ratio.
Doctor enrolled participants and then researcher assigned par-
ticipants to intervention. All subjects were asked to stop
NSAIDs at least 1 week prior to participating in the trial.
Before the end of week 4, two patients given diclofenac
dropped out of the trial because of severe dyspnea and heart-
burn. Therefore, the diclofenac arm was ethically discontinued
at week 4.
Physical examination and measurement of joint space
width were conducted in orthopedic OPD at Ramathibodi
Hospital. WOMAC index and SF-36 questionnaire were re-
corded by a staff nurse.
Outcomes
Primary outcome
The clinical efficacy outcomes were monitored at baseline,
week 4 and week 16 of treatment. Treatment compliance
was checked at each visit by pill counting. The primary out-
come was change in WOMAC scores from baseline.
WOMAC includes 24 items of pain (020), stiffness (08),
and physical function (068) with the total scores of 96 [18].
The lower WOMAC score represents the better clinical
outcome.
For non-inferiority testing, non-inferiority margin of 12 %
was from a difference of percent change from baseline at week
12 in WOMAC function score between bromelain (25 %)
and placebo (13 %) groups of knee OA [14]. Mean differ-
ence at week 4 was calculated as %change from baseline
values in bromelain group%change from baseline values
in diclofenac group. Bromelain would be judged to non-
inferior to diclofenac if the 95 % confidence interval of the
different in percent change from baseline at week 4 in
WOMAC score (bromelain-diclofenac) lied to the left of
non-inferiority margin (12 %) but was not greater than
+12 % [19].
Secondary outcome and others
Secondary outcome The secondary clinical outcome was the
quality of life assessment (SF-36) [20]. SF-36 was assessed
using two summary scores (physical and mental component
summary). SF-36 was scored from 0 to 100; 0 score indicated
extreme problems and 100 score indicated no problems.
Safety assessment Adverse effects reported by patients were
recorded by the staff nurse at week 4 and 16 of trial.
Clin Rheumatol
Fig. 1 Flow of the study
Oxidative stress Plasma MDA (a lipid peroxidation marker)
and nitrite were measured as oxidative stress markers. Total glu-
tathione in erythrocytes and plasma -tocopherol were also de-
termined. Plasma MDA was evaluated by fluorometric method
based on the reaction between MDA and thiobarbituric acid [21].
Blood (10 ml) was drawn from overnight fast patients and
healthy subjects and collected into lithium heparin tubes.
Blood (8 ml) was centrifuged at 2000g for 10 min at 4 C.
Packed red blood cells were used freshly for total glutathione
measurement. Plasma was kept at 80 C until measurement
of MDA and -tocopherol. For nitrite measurement, plasma
was immediately separated from 2 ml of whole blood by cen-
trifugation at 5000 g for 1 min at 4 C.
Plasma nitrite was measured by the triiodide-based chemi-
luminescence method [22] using the nitric oxide (NO) analyz-
er (Eco Medics Analyzer CLD88, Switzerland).
For determination of total glutathione, erythrocytes were
washed three times with threefold volume of 0.9 % NaCl
solution. One hundred microliters of packed erythrocytes were
used [23].
Plasma -tocopherol was measured by reverse phase high-
performance liquid chromatography [24]. The chromato-
graphic analyses were performed with a chromatographic sys-
tem (Alliance 2695, Waters, Milford, MA, USA) equipped
with a fluorescence detector (ex = 295, and em = 370 nm)
(Model 2475, Waters, Milford, MA, USA). The column was a
Novapak-C18, 3.9 150 mm. The mobile phase was 100 %
methanol at a flow rate of 1 ml/min. Five hundred microliters
of plasma was extracted with 1 ml of sodium dodecyl sulfate
(0.1 M), 2 ml of ethanol, and 4 ml of hexane. The samples
were evaporated under nitrogen gas and re-dissolved in 1 ml
of methanol. A 10 or 20 l of samples or standards were
injected into the column.
PGE2 production in LPS-stimulated whole blood LPS
(10 g/ml) was added to tubes containing 800 l of RPMI
1640 medium (supplemented with 100 IU/ml penicillin,
100 g/ml streptomycin, and 2.5 IU/ml heparin) and 200 l
of fresh heparinized whole blood. After incubation at 37 C in
a humidified atmosphere of 95 % air and 5 % CO2 for 24 h, the
 Clin Rheumatol

Table 1 Baseline demographics and clinical characteristics of patients in diclofenac and bromelain groups

Characteristics Total (n = 40) Diclofenac (n = 20) Bromelain (n = 20)

Age (years) 61.7 1.1 61.3 1.6 62.1 1.6

Male/female 9/31 2/18 7/13
BMI (kg/m2) 25.3 0.6 26.2 0.8 24.5 0.8
Duration of knee pain (months) 18.7 3.4 22.1 5.9 15.3 3.3
Bilateral knee OA (number) 14 8 6
Joint space width (mm) 5.0 0.2 4.7 0.2 5.3 0.2
X-ray assessment of OA severity (number)
Kellgren and Lawrence score 1 23 9 14
Kellgren and Lawrence score 2 17 11 6
WOMAC index score (points)
Total 25.7 2.5 25.9 3.3 25.5 3.8
Pain subscale 5.8 0.6 5.9 0.9 5.6 0.8
Stiffness subscale 1.5 0.3 1.1 0.3 2.0 0.4
Function subscale 18.4 1.8 18.9 2.4 17.9 2.7
SF-36 score (points)
PCS 65.7 1.9 66.0 2.5 65.4 2.9
MCS 84.4 2.0 82.6 2.6 86.2 3.0

Data are presented as means SEM or number. Comparison between diclofenac and bromelain groups was done by Mann-Whitney test and 2 test
SF-36 short-form 36, WOMAC Western Ontario and McMaster Universities Osteoarthritis Index, BMI body mass index, MCS mental component
summary, PCS physical component summary

samples were centrifuged, and the supernatant was frozen at enzyme-linked immunosorbent assay kit (Assay Designs,
70 C until measurement [25]. PGE2 was measured by Ann Arbor, MI, USA).

Table 2 Comparison of mean a

change from baseline in Parameters Mean change (95 % CI) P-value Effect size (95 % CI)
WOMAC and SF-36 scores
between diclofenac (n =18) and Total WOMAC
bromelain (n = 20) treatment at Diclofenac 15.06 (22.13, 7.98) 0.299 0.29 (0.36, 0.92)
week 4 in knee osteoarthritis Bromelain 10.45 (18.51, 2.39)
patients
Pain subscale
Diclofenac 3.78 (5.74, 1.81) 0.370 0.31 (0.34, 0.95)
Bromelain 2.55 (4.40, 0.70)
Stiffness subscale
Diclofenac 0.50 (1.27, 0.27) 0.459 0.23 (0.86, 0.42)
Bromelain 0.85 (1.57, 0.13)
Function subscale
Diclofenac 10.78 (15.80, 5.75) 0.156 0.32 (0.33, 0.95)
Bromelain 7.05 (13.05, 1.05)
SF-36 PCS
Diclofenac 11.03 (4.98, 17.08) 0.193 0.36 (0.99, 0.29)
Bromelain 5.88 (1.74, 13.50)
SF-36 MCS
Diclofenac 10.16 (4.61, 15.70) 0.052 0.56 (1.20, 0.10)
Bromelain 1.08 (8.09, 10.24)

Values are presented as means (95 % CI)

CI confidence interval, MCS mental component summary, PCS physical component summary, SF-36 short-form
36, WOMAC Western Ontario and McMaster Universities Osteoarthritis Index
a
Compared between diclofenac and bromelain groups (Mann-Whitney test)
Clin Rheumatol

high-dose and low-dose bromelain in mild acute knee pain

Fig. 2 Mean difference and 95 % confidence interval in percent change
from baseline in WOMAC score after 4-week treatment with diclofenac
(n = 18) and bromelain (n = 20)
was used to calculate [13].
Data are presented as number of patients, means standard
error of the mean (SEM) or means (95 % confidence interval).
Analysis was performed by GraphPad Prism version 5
(GraphPad software Inc., San Diego, CA, USA). Comparison
of clinical outcomes and biochemical markers at baseline, week
4 and week 16 was performed by Wilcoxon signed rank test
with Bonferroni correction in which P < 0.025 was considered
as significant difference. Comparison of mean change from
baseline between diclofenac and bromelain groups and param-
eters between healthy subjects and OA patients were performed
by Mann-Whitney test in which P < 0.05 was considered as
significant difference. The chi-square test was used for testing
categorical data between diclofenac and bromelain groups.

Statistical analysis
The sample size calculation was based on a non-inferiority
margin of 12 % which was a difference of percent change
from baseline in WOMAC function score between bromelain
and placebo groups in knee OA [14]. The 33 patients per
group required for 80 % power to show the two-sided alpha
of 0.05 of the treatment difference. In addition, the sample size
was 17 patients per group with a two-sided alpha of 0.05 and
80 % power. A difference of WOMAC stiffness score between
Results
Demographics and baseline clinical characteristics
The mean age of 40 patients with knee OAwas 61.7 1.1 years,
and the mean duration of OA pain was 18.7 3.4 months
(Table 1). Fourteen patients had bilateral knee OA at enroll-
ment. The mean joint space width at enrollment was
5.0 0.2 mm. There was no significant difference between

Table 3 WOMAC and SF-36 a a

scores before and after treatment Parameters Baseline 4 weeks P value 16 weeks P value
with diclofenac (n = 18) and
bromelain (n = 20) in knee Total WOMAC
osteoarthritis patients Diclofenac 27.3 3.5 12.3 3.0 0.002
Bromelain 25.5 3.8 15.0 2.1 0.022 12.2 2.2 0.002
Pain subscale
Diclofenac 6.3 0.9 2.5 0.6 0.001
Bromelain 5.6 0.8 3.1 0.5 0.012 2.4 0.4 0.001
Stiffness subscale
Diclofenac 1.2 0.3 0.7 0.3 0.170
Bromelain 2.0 0.4 1.1 0.3 0.028 0.8 0.3 0.005
Function subscale
Diclofenac 19.8 2.5 9.1 2.3 0.002
Bromelain 17.9 2.7 10.9 1.5 0.028 9.1 1.7 0.005
SF-36 PCS
Diclofenac 65.0 2.7 76.0 2.6 0.002
Bromelain 65.4 2.9 71.3 2.6 0.078 73.3 2.3 0.005
SF-36 MCS
Diclofenac 82.1 2.8 92.2 1.6 0.0004
Bromelain 86.2 3.0 87.3 2.9 0.640 88.7 2.8 0.180

Values are presented as means SEM

MCS mental component summary, PCS physical component summary, SF-36 short-form 36, WOMAC Western
Ontario and McMaster Universities Osteoarthritis Index
a
Compared with baseline (Wilcoxon signed rank test with Bonferroni correction)
 Clin Rheumatol

Table 4 Number of adverse events in patients given diclofenac (n = 20) Mean difference (95 % confidence interval) of percent change
and bromelain (n = 20)
from baseline between treatment groups was shown in Fig. 2
Adverse events Diclofenac Bromelain Bromelain for total WOMAC score (37.00 (26.33, 100.34)), pain (25.21
(4 weeks) (4 weeks) (16 weeks) (22.76, 73.18)), stiffness (0.67 (45.11, 43.77)), and func-
tion subscale (54.77 (28.39, 137.93)). With the non-
Total number of 7 5 3
patients
inferiority margin set at 12 %, the trial was inconclusive.
with adverse events Confidence interval was mostly at the right and partly at the
GI symptoms left of non-inferiority margin (wide confidence interval).
Nausea 2 2 We compared the values at baseline with those at the end of
Abdominal pain 2 study. At week 4 of treatment, total WOMAC scores were
Constipation 2 decreased in both diclofenac and bromelain groups (Table 3).
Flatulence 1 1 At week 4, the improved pain and functional subscales were
Diarrhea 1 1 observed in diclofenac group, while only the pain subscale was
Heartburn 1 improved in bromelain group. Stiffness subscales were un-
Others changed in both groups. At week 16 of bromelain treatment,
Dyspnea 2 total WOMAC scores and subscales (pain, stiffness and func-
Dry mouth 1 1 tion) were improved.
Headache 1 1
Tiredness 1 Secondary outcome and others
Swollen eyelids 1
Itchy eyes 1 Secondary outcome There was no difference in mean change
Urticaria 1 from baseline of SF-36 score between diclofenac and bromelain
groups at week 4 (Table 2). Small effect size was found in SF-36
GI gastrointestinal
physical health between diclofenac and bromelain groups.
Moderate effect size was found in SF-36 mental health (Table 2).
At week 4 of treatment, significant improvement was ob-
diclofenac and bromelain groups in terms of demographics or
served in both SF-36 physical health and mental health sum-
baseline characteristics.
mary scores in diclofenac group, but not in bromelain group
(Table 3). At week 16, significant improvement was noted in
Outcomes SF-36 physical health in bromelain group.

Primary outcome
There was no difference in mean change from baseline of
WOMAC score between diclofenac and bromelain groups at
week 4 (Table 2). Effect size was calculated from a difference
in mean change score from baseline between bromelain and
diclofenac groups divided by the pooled standard deviation.
Effect size <0.20.5 is considered as small, and effect size
>0.50.8 is moderate [26]. Small effect size was found in
WOMAC score between diclofenac and bromelain groups.
Safety assessment Adverse effects reported by patients are
presented in Table 4. Before the end of week 4, two patients
in diclofenac group dropped out of trial because of intolerable
dyspnea and heartburn. Bromelain was well tolerated. Its ad-
verse effects were mild nausea, constipation, flatulence, diar-
rhea, dry mouth, headache, and tiredness.
Oxidative stress OA patients had higher plasma nitrite and
MDA concentrations than healthy subjects (Table 5). Plasma

Table 5 Baseline biochemical a

markers of the healthy subjects Biochemical markers Healthy subjects Patients P value
(n = 20) and osteoarthritis patients
(n = 38) Plasma MDA (M) 0.28 0.01 0.31 0.01 0.018
Plasma nitrite (nM) 64.92 6.22 96.52 7.52 0.015
Plasma -tocopherol (g/ml) 16.95 0.89 14.32 0.47 0.009
Erythrocytic glutathione (mM) 10.55 1.07 7.99 0.81 0.051
LPS-induced PGE2 production in whole blood (ng/ml) 14.68 2.12 25.58 3.61 0.049

Values are presented as means SEM

LPS lipopolysaccharide, MDA malondialdehyde, PGE2 prostaglandin E2
a
Compared with healthy subjects (Mann Whitney test)
Clin Rheumatol

Fig. 3 Oxidative stress markers

in knee osteoarthritis patients. a
and b Plasma malondialdehyde
(MDA) levels before and after
treatment with diclofenac or
bromelain. c and d Plasma nitrite
levels before and after treatment
with diclofenac or bromelain. e
and f Erythrocytic total
glutathione levels before and after
treatment with diclofenac or
bromelain. g and h Plasma -
tocopherol levels before and after
treatment with diclofenac or
bromelain. *P < 0.025 compared
with baseline values (Wilcoxon
signed rank test with Bonferroni
correction)

-tocopherol concentrations were lower in OA patients.
Erythrocytic total glutathione was not different between two
groups.
At week 4, there was no significant alteration in plasma
MDA in both bromelain and diclofenac groups (Fig. 3ab).
However, at week 16, a reduction of plasma MDA was ob-
served in bromelain group (P < 0.025, Fig. 3b). The plasma
nitrite, erythrocytic total glutathione, and plasma -tocopherol
levels were unchanged in both diclofenac and bromelain
groups at any time of treatment (Fig. 3ch).
PGE2 production by LPS-stimulated whole blood LPS-in-
duced PGE2 production in whole blood of OA patients was
higher than that of healthy controls (Table 5). At week 4, a
reduction of PGE2 production was found in diclofenac group

Fig. 4 Lipopolysaccharide-
induced prostaglandin (PG) E2
production in whole blood of knee
osteoarthritis patients before and
after treatment with diclofenac (a)
or bromelain (b) *P < 0.025
compared with baseline values
(Wilcoxon signed rank test with
Bonferroni correction)
(Fig. 4a). At week 4 and 16, a decrease in LPS-induced PGE2
production was observed in bromelain group (Fig. 4b).
Discussion
There was no difference in symptomatic relief (WOMAC
scores) and quality of life (SF-36 scores) of knee OA com-
pared between bromelain and diclofenac treatments at week 4.
Both treatments demonstrated the improvement of total
WOMAC scores and WOMAC pain subscales in knee OA
at week 4. Functional subscales and quality of life (SF-36)
were improved only in diclofenac group. At week 16, patients
given bromelain showed additional improvement in physical
function and stiffness, and SF-36 physical health. Sixteen-
week treatment with 500-mg/day of bromelain was safe.
Oxidative stress and PGE2 production were decreased at 16-
week bromelain treatment.
Treatment with enzyme preparation containing bromelain
(90 mg three times a day), trypsin, and rutin for 3 weeks
showed 77.1 % decrease in symptom of knee pain in severe
OA [10]. Four-week treatment with bromelain reduced symp-
toms of mild knee pain and improved well-being in healthy
adults [13]. Improvements were higher in the subjects receiving
400-mg/day bromelain compared with 200-mg/day bromelain.
In our study, higher dose of bromelain (500-mg/day) was ad-
ministered to mild-to-moderate knee OA patients for 16 weeks.
However, bromelain at dose 800-mg/day for 12 weeks was
ineffective for moderate-to-severe knee OA [14]. Thus, brome-
lain can ameliorate symptoms in patients with mild-to-
moderate OA. Bromelain was observed to be well tolerated,
while two patients in the diclofenac group dropped out from the
study due to dyspnea and heartburn. When compared between
two groups, there was no difference in symptom improvement
at week 4. The limitation of our study is the small sample size
which may give rise to overestimation of treatment effect [27].
In addition, self-reported outcome and continuous endpoints in
our study would lead to potential bias [28, 29].
The potential mediators responsible for cartilage destruc-
tion in OA are reactive oxygen species (ROS), NO, and in-
flammatory mediators such as PGE2 [30, 31]. ROS contribute
to cartilage matrix degradation in OA through lipid
Clin Rheumatol
peroxidation [32]. Arachidonic acid is oxidized by ROS into
three primary products: F2-isoprostanes, 4-hydroxynonenal,
and MDA. These molecules are known to be toxic to hyaline
cartilage [33]. We have noted that plasma nitrite and MDA
levels were higher, and -tocopherol level was lower in pa-
tients with knee OA compared to controls. These findings
suggest that knee OA patients have higher oxidative stress.
Plasma MDA concentration was decreased after 16-week
treatment with bromelain. Reduction of plasma MDA level
by anti-inflammatory drugs such as tiaprofenic acid was
shown in knee OA and effectively reduced OA symptoms
[34]. Thus, the effect of bromelain on OA symptoms may be
due to its role in reducing MDA level. Bromelain-treated mice
had increased serum antioxidant enzyme activities (superox-
ide dismutase and glutathione peroxidase) and reduced gluta-
thione compared to control mice [35]. Increased antioxidant
activities by bromelain may be the cause of MDA reduction.
PGE2 has catabolic effect on proteoglycan turnover in carti-
lage [31]. Levels of PGE2 production after LPS stimulation in
whole blood samples of knee OA patients were higher com-
pared to controls, suggesting an increased cyclooxygenase-2
expression in OA patients [36]. At week 4 and 16, a decrease
in LPS-induced PGE2 production was observed in bromelain
group. This is consistent with the report that bromelain in-
hibits cyclooxygenase-2 mRNA expression induced by LPS
in human macrophage cell line [37]. Reduction in WOMAC
pain and total WOMAC scores by bromelain may be associ-
ated with reduction of PGE2 level at week 4. In addition, later
reduction in pain and stiffness and improved physical function
and SF-36 physical health after 16-week bromelain treatment
may be associated with reduction of lipid peroxidation and
PGE2 levels.
Conclusion
Treatments with bromelain and diclofenac for 4 weeks have
similar effects in reducing knee OA symptoms. Further treat-
ment with bromelain for 16 weeks improves pain, stiffness,
and physical function compared with baseline. Reduction in
lipid peroxidation and LPS-induced PGE2 production was ob-
served at week 16 in bromelain group.
Clin Rheumatol
Acknowledgments This work was supported by the Program Strategic
Scholarships for Frontier Research Network for the Joint Ph.D. Program,
the Office of the Higher Education Commission, Thailand, and
Department of Pharmacology, Faculty of Science, Mahidol University.
We thank all patients and healthy subjects who participated in this study.
Compliance with ethical standards This study was approved by the
Ramathibodi Hospital Ethics Committee, Faculty of Medicine
Ramathibodi Hospital, Mahidol University, in accordance with
Declaration of Helsinki. All patients provided the written informed con-
sent before randomization.
Disclosures None.
References
1. Benito MJ, Veale DJ, FitzGerald O, van den Berg WB, Bresnihan B
(2005) Synovial tissue inflammation in early and late osteoarthritis.
Ann Rheum Dis 64(9):12631267
2. Silverstein FE, Faich G, Goldstein JL, Simon LS, Pincus T,
Whelton A, Makuch R, Eisen G, Agrawal NM, Stenson WF
(2000) Gastrointestinal toxicity with celecoxib vs nonsteroidal
anti-inflammatory drugs for osteoarthritis and rheumatoid arthritis.
JAMA 284(10):12471255
3. Brun J, Jones R (2001) Nonsteroidal anti-inflammatory drugasso-
ciated dyspepsia: the scale of the problem. Am J Med 110(1):S12
S13
4. Laine L, Curtis S, Cryer B, Kaur A, Cannon C (2010) Risk factors
for NSAID-associated upper GI clinical events in a long-term pro-
spective study of 34 701 arthritis patients. Aliment Pharmacol Ther
32(10):12401248
5. Mller S, Mrz R, Schmolz M, Drewelow B, Eschmann K, Meiser
P (2013) Placebo-controlled randomized clinical trial on the
immunomodulating activities of low- and high-dose bromelain after
oral administrationnew evidence on the antiinflammatory mode
of action of bromelain. Phytother Res 27(2):199204
6. Majid OW, Al-Mashhadani BA (2014) Perioperative bromelain re-
duces pain and swelling and improves quality of life measures after
mandibular third molar surgery: a randomized, double-blind,
placebo-controlled clinical trial. J Oral Maxillofac Surg 72(6):
10431048
7. Cohen A, Goldman J (1964) Bromelain therapy in rheumatoid ar-
thritis. Penn Med J 67:2730
8. Brien S, Lewith G, Walker A, Hicks SM, Middleton D (2004)
Bromelain as a treatment for osteoarthritis: a review of clinical
studies. Evid Based Complement Alternat Med 1(3):251257
9. Singer F, Oberleitner H (1995) Drug therapy of activated arthrosis.
On the effectiveness of an enzyme mixture versus diclofenac. Wien
Med Wochenschr (1946) 146(3):5558
10. Klein G, Kullich W (2000) Short-term treatment of painful osteo-
arthritis of the knee with oral enzymes: a randomised, double-blind
study versus diclofenac. Clin Drug Investig 19(1):1523
11. Singer F, Singer C, Oberleitner H (2001) Phlogenzym versus
diclofenac in the treatment of activated osteoarthritis of the knee.
A double-blind prospective randomized study. Int J Immunother
17(2/4):135142
12. Tilwe G, Beria S, Turakhia N, Daftary G, Schiess W (2001)
Efficacy and tolerability of oral enzyme therapy as compared to
diclofenac in active osteoarthrosis of knee joint: an open random-
ized controlled clinical trial. JAPI 49:617621
13. Walker A, Bundy R, Hicks S, Middleton R (2002) Bromelain re-
duces mild acute knee pain and improves well-being in a dose-
dependent fashion in an open study of otherwise healthy adults.
Phytomedicine 9(8):681686
14. Brien S, Lewith G, Walker A, Middleton R, Prescott P, Bundy R
(2006) Bromelain as an adjunctive treatment for moderate-to-severe
osteoarthritis of the knee: a randomized placebo-controlled pilot
study. QJM 99(12):841850
15. Altman R, Asch E, Bloch D, Bole G, Borenstein D, Brandt K,
Christy W, Cooke T, Greenwald R, Hochberg M (1986)
Development of criteria for the classification and reporting of oste-
oarthritis: classification of osteoarthritis of the knee. Arthritis
Rheum 29(8):10391049
16. Kellgren J, Lawrence J (1963) Atlas of standard radiographs
Oxford. Blackwell, London
17. Farrell AJ, Blake DR, Palmer R, Moncada S (1992) Increased con-
centrations of nitrite in synovial fluid and serum samples suggest
increased nitric oxide synthesis in rheumatic diseases. Ann Rheum
Dis 51(11):12191222
18. Bellamy N, Buchanan WW, Goldsmith CH, Campbell J, Stitt LW
(1988) Validation study of WOMAC: a health status instrument for
measuring clinically important patient relevant outcomes to anti-
rheumatic drug therapy in patients with osteoarthritis of the hip or
knee. J Rheumatol 15(12):18331840
19. Piaggio G, Elbourne DR, Pocock SJ, Evans SJ, Altman DG, Group
C (2012) Reporting of noninferiority and equivalence randomized
trials: extension of the CONSORT 2010 statement. JAMA 308(24):
25942604
20. Ware J, Snow K, Kosinski M, Gandek B (1993) SF-36 health sur-
vey: manual and interpretive guide. The health institute. New
England Medical Center, Boston
21. Ohkawa H, Ohishi N, Yagi K (1978) Reaction of linoleic acid
hydroperoxide with thiobarbituric acid. J Lipid Res 19(8):1053
1057
22. Pelletier MM, Kleinbongard P, Ringwood L, Hito R, Hunter CJ,
Schechter AN, Gladwin MT, Dejam A (2006) The measurement of
blood and plasma nitrite by chemiluminescence: pitfalls and solu-
tions. Free Radic Biol Med 41(4):541548
23. Beutler E, Duron O, Kelly BM (1963) Improved method for the
determination of blood glutathione. J Lab Clin Med 61:882888
24. Zaspel BJ, Csallany AS (1983) Determination of alpha-tocopherol
in tissues and plasma by high-performance liquid chromatography.
Anal Biochem 130(1):146150
25. von Aulock S, Hermann C, Hartung T (2003) Determination of the
eicosanoid response to inflammatory stimuli in whole blood and its
pharmacological modulation ex vivo. J Immunol Methods 277(1):
5363
26. Sullivan GM, Feinn R (2012) Using effect size-or why the P value
is not enough. J Grad Med Educ 4(3):279282
27. Nesch E, Trelle S, Reichenbach S, Rutjes AW, Tschannen B,
Altman DG, Egger M, Jni P (2010) Small study effects in meta-
analyses of osteoarthritis trials: meta-epidemiological study. BMJ
341:c3515
28. Hrbjartsson A, Gtzsche PC (2001) Is the placebo powerless? An
analysis of clinical trials comparing placebo with no treatment. N
Engl J Med 344(21):15941602
29. Hrbjartsson A, Thomsen ASS, Emanuelsson F, Tendal B, Hilden
J, Boutron I, Ravaud P, Brorson S (2012) Observer bias in
randomised clinical trials with binary outcomes: systematic review
of trials with both blinded and non-blinded outcome assessors. BMJ
344:e1119
30. Del Carlo M, Schwartz D, Erickson EA, Loeser RF (2007)
Endogenous production of reactive oxygen species is required for
stimulation of human articular chondrocyte matrix metalloprotein-
ase production by fibronectin fragments. Free Radic Biol Med
42(9):13501358
31. Mastbergen SC, Bijlsma JW, Lafeber FP (2008) Synthesis and re-
lease of human cartilage matrix proteoglycans are differently

regulated by nitric oxide and prostaglandin-E2. Ann Rheum Dis
67(1):5258
32. Shah R, Raska K, Tiku ML (2005) The presence of molecular
markers of in vivo lipid peroxidation in osteoarthritic cartilage: a
pathogenic role in osteoarthritis. Arthritis Rheum 52(9):27992807
33. Morquette B, Shi Q, Lavigne P, Ranger P, Fernandes JC,
Benderdour M (2006) Production of lipid peroxidation products
in osteoarthritic tissues: new evidence linking 4-hydroxynonenal
to cartilage degradation. Arthritis Rheum 54(1):271281
34. Tuzun S, Uzun H, Aydin S, Dinc A, Sipahi S, Topcuoglu MA,
Yucel R, Belce A (2005) Effects of flurbiprofen and tiaprofenic acid
on oxidative stress markers in osteoarthritis: a prospective,
Clin Rheumatol
randomized, open-label, active-and placebo-controlled trial. Curr
Therap Res 66(4):335344
35. Agarwal S, Chaudhary B, Bist R (2016) Bacoside a and bromelain
relieve dichlorvos induced changes in oxidative responses in mice
serum. Chem Biol Interact 254:173178
36. Hinz B, Brune K, Pahl A (2000) Cyclooxygenase-2 expression in
lipopolysaccharide-stimulated human monocytes is modulated by
cyclic AMP, prostaglandin E2, and nonsteroidal anti-inflammatory
drugs. Biochem Biophys Res Commun 278(3):790796
37. Huang J-R, Wu C-C, Hou RC-W , Jeng K-C (2008) Bromelain
inhibits lipopolysaccharide-induced cytokine production in human
THP-1 monocytes via the removal of CD14. Immunol Investig
37(4):263277