Journal of the Formosan Medical Association (2012) 111, 542e549

Available online at www.sciencedirect.com

journal homepage: www.jfma-online.com

ORIGINAL ARTICLE

Initial manifestations and clinical course of systemic

onset juvenile idiopathic arthritis: A ten-year
retrospective study
Hu-Yuan Tsai, Jyh-Hong Lee, Hsin-Hui Yu, Li-Chieh Wang, Yao-Hsu Yang,
Bor-Luen Chiang*

Department of Pediatrics, National Taiwan University Hospital, Taipei, Taiwan, Republic of China

Received 15 January 2011; received in revised form 22 June 2011; accepted 25 June 2011

KEYWORDS Background/Purpose: The diagnosis of systemic onset juvenile idiopathic arthritis (SoJIA) on
systemic juvenile disease onset is challenging and made mainly by exclusion. This study aimed to investigate
idiopathic arthritis; the initial clinical and laboratory features of children with SoJIA in Taiwan.
systemic onset Methods: Patients diagnosed with SoJIA at the National Taiwan University Hospital between
juvenile 1997 and 2007 were evaluated and data were collected by retrospective chart review. Inferen-
rheumatoid tial statistics were used to compare features of patients with steroid use for <6 months or >6
arthritis months.
Results: Twenty-eight (28) patients (13 boys and 15 girls) were included in this study. The
mean age of onset was 8.7 years old. The most common presentations were fever (100%),
arthritis (89.3%), and skin rash (67.9%). The patterns of arthritis in affected patients were
50% oligoarticular type and 39% polyarticular type. The most common joints involved were
the knee (76% of patients with arthritis), ankle (56%), and elbow and proximal interphalangeal
joints (28%). The most common pattern of fever during first week was intermittent (53%). Pro-
longed use of steroid was associated with leukocytosis (17.63
7.71 vs. 11.93
4.43  109
leukocytes/L, p < 0.05) and higher aspartate aminotransferase (89.4 vs. 31.2 U/L, p < 0.05)
on initial presentation.
Conclusion: In SoJIA, extra-articular features such as fever, rash, and lymphadenopathy are
most prominent. Leukocytosis and polyarticular pattern on presentation may indicate a refrac-
tory clinical course.
Copyright ª 2012, Elsevier Taiwan LLC & Formosan Medical Association. All rights reserved.

* Corresponding author. Department of Pediatrics, National Taiwan University Hospital, 7 Chung-Shan South Road, Taipei 10002, Taiwan,
Republic of China.
E-mail address: gicmbor@ntu.edu.tw (B.-L. Chiang).

0929-6646/$ - see front matter Copyright ª 2012, Elsevier Taiwan LLC & Formosan Medical Association. All rights reserved.
doi:10.1016/j.jfma.2011.06.013
Systemic onset juvenile idiopathic arthritis
Introduction
Juvenile idiopathic arthritis (JIA), also known as juvenile
rheumatoid arthritis or juvenile chronic arthritis, is the
most common rheumatic disease in children.1 JIA is defined
as persistent arthritis for 6 weeks if certain exclusionary
conditions have been eliminated.2 The disease onset
subtype is defined by clinical symptoms in the first 6 months
of the disease.
In 1897, Still outlined his observations on 12 children
with systemic juvenile idiopathic arthritis.3 Since then,
systemic-onset juvenile idiopathic arthritis (SoJIA) has been
defined as a subtype of JIA and characterized by 6 weeks of
arthritis accompanied by intermittent fever daily or twice
daily for at least 2 weeks in children aged <16 years in the
absence of an identifiable underlying etiology.4 By the
newer International League of Associations for Rheuma-
tology (ILAR) criteria, SoJIA is now more strictly defined as
daily quotidian fever of 39 C and requires the presence of
at least one of following clinical features of systemic
inflammation: characteristic evanescent erythematous skin
rash; generalized lymphadenopathy; hepato-splenomegaly;
and serositis.5
The clinical manifestations of SoJIA have little similarity
to those of the other JIA subtypes. Unlike other JIA patients
whose joint manifestations are more prominent than their
general symptoms, the prominent clinical features that
distinguish SoJIA are extra-articular manifestations such as
fever, rash, and lymphadenopathy. It is usually character-
ized by laboratory evidence of inflammation, including
granulocytosis, thrombocytosis, and the up-regulation of
acute phase reactants, including erythrocyte sedimenta-
tion rate (ESR) and C-reactive protein (CRP). Secondary
macrophage activation syndrome (MAS) is a well-known
complication most often seen in children with SoJIA.
SoJIA accounts for 10% to 30% of children with JIA and is
the most common JIA subtype in hospitalized children.6 The
wide variation of signs and symptoms make diagnosis diffi-
cult. Thus, the diagnosis of SoJIA on disease onset is chal-
lenging and made mainly by exclusion. This retrospective
study aimed to summarize the initial clinical and laboratory
features, and clinical course of SoJIA children in a single
tertiary care referral centre in Taiwan.
Patients and methods
Patients newly diagnosed with SoJIA at the National Taiwan
University Hospital between 1997 and 2007 were identified.
The diagnosis of SoJIA was based on the International
League of Associations for Rheumatology (ILAR) criteria.5
All these patients were admitted for evaluation and
management, and did not receive any specific treatment
for SoJIA before admission. Medical records were reviewed
retrospectively and data from the first consult with rheu-
matologist were collected. The initial symptoms on
presentation during the first 2 weeks with regard to fever
were recorded and analyzed. Medical records of the follow-
up period up to the end of 2008 were retrospectively
examined. Informed consents were obtained before labo-
ratory examination of each patient. The study was
approved by the ethics review boards (No. 200905019R).
543
Since all patients essentially developed daily quotidian
fever for at least 2 weeks, the fever pattern during the first
week was analyzed and intermittent fever was defined as
regular periodic fever lasting >1 day, while continuous
fever was defined as an unremitting fever on presentation.
The location and number of involved joints at the time of
diagnosis were recorded. The oligoarticular type was
defined as fewer than four joints in the first 6 months of the
disease, while involvement of more than five joints was
defined as the polyarticular type.
In order to summarize the therapeutic response, these
patients were classified into three groups by the clinical
course. Clinical remission course was defined as the
absence of active arthritis and systemic features for at
least 2 consecutive months and complete discontinuation of
medications for >2 months, and without relapse of symp-
toms and restart of medications. Drug-dependent course
was defined as without clinical remission after the onset, or
one clinical remission but without complete cessation of
medications for >2 months. Relapse course was defined as
clinical remission and complete discontinuation of medi-
cations for >2 months, but relapse of symptoms and restart
of medications.
Laboratory records
Laboratory data, including: complete blood cell count;
complements C3 and C4; erythrocyte sedimentation rate
(ESR); C-reactive protein (CRP); serum concentrations of
immunoglobulins IgG, IgA and IgM; liver enzymes aspartate
aminotransferase (AST) and alanine aminotransferase
(ALT); iron profile (iron, total iron binding capacity, and
ferritin); and lactic dehydrogenase (LDH) at the time of
diagnosis were gathered, mainly within the first 2 weeks of
symptom onset. The status of anti-nuclear antibody (ANA),
rheumatoid factor (RF), and human leukocyte antigen B27
(HLA-B27) were also collected.
Statistical analysis
Comparisons of various laboratory values were done using
the Mann-Whitney test, Student’s t test, and c2 test
accordingly. Inferential statistics with binary logistic
regression were used to compare features of patients with
and without steroid use for >6 months. A p value <0.05 was
considered statistically significant.
Results
Basic profile and clinical manifestations on
presentation
Twenty-eight patients, including 13 boys and 15 girls, were
enrolled. The mean follow-up period was 69 months. The
mean hospitalized stay was 15.8 days. The mean age at
onset was 8.7 years (range 1.3e15 years; median 7 years;
Fig. 1). Among the symptoms on initial presentation, the
most common were fever (100%), arthritis (89.3%), and skin
rash (67.9%; Fig. 2). Lymphadenopathy (46.4%), as well as
hepato-splenomegaly and headache (21.4%) were also
544 H.-Y. Tsai et al.

Age at presentation involved were the knee (76%), ankle (56%), and elbow and
proximal interphalangeal (PIP) joints (28%), followed by the
5
wrist (24%). Arthritis of small joints of the foot was found in
only two patients (8%; Fig. 3B).
4 All of the patients had fever on presentation. The most
Number of patients

common fever pattern during the first week was intermit-

3 tent (53%), while 18% had daily evening fever. The other
fever patterns were split equally among daily morning
fever, twice-daily fever, and continuous unremitting fever
2
(Fig. 3C).

1 Laboratory characteristics on presentation

0 Laboratory data on presentation are summarized in Table 1.

1 2
Figure 1 The patients’ age on disease onset. The onset age
of the children in this study is evenly distributed.
prominent. Few patients presented with serositis (2/28,
7.1%), seizure (2/28, 7.1%), uveitis (1/28, 3.6%) with
manifestation of red eye, and central retinal vein occlusion
(1/28, 3.6%). One patient had left centroparietal focal
seizure recorded by electroencephalography and central
retinal vein occlusion simultaneously, with manifestation of
blurred vision at 1 month after disease onset and the eye-
ground revealed bilateral flame-shaped hemorrhage.
The patterns of arthritis (Fig. 3A) were oligoarticular
(50%) and polyarticular type (39%). Interestingly, 11% of
SoJIA patients did not present with arthritis on disease
onset. Of these 3 patients, the mean period from the first
day of clinical symptoms to that of developing arthritis was
75 days (62 days, 80 days, and 83 days respectively), and
the patterns of arthritis were all oligoarticular type. The
mean number of joints involved was 7 (range 0e37).
Regarding location of the arthritis, the most common joints
3 4 5 6 7 8 9 10 11 12 13 14 15 Forty percent of patients had leukocytosis and 64.3% had
Age (years) thrombocytosis. White blood and platelet counts were
elevated, with a mean of 14.78  109 cells/L and
458.86  109 cells/L, respectively, while 35.7% had anemia
with mean hemoglobin of 10.12 g/dL. The majority of
inflammatory parameters, including ESR, CRP, and C3/C4,
were elevated. All patients had elevated CRP level on
presentation.
There were elevated liver enzymes: 40.7% had elevated
AST, 33.3% had elevated ALT, and 90.9% had elevated LDH.
Elevated serum ferritin levels were present in 71.4% of
patients, with a mean ferritin level of 3075.5 ng/ml. Only
two of 28 (7.14%) patients had positive ANA, two of 25 (8%)
patients had positive rheumatoid factor, and all patients
had negative HLA-B27.
Hemophagocytosis on presentation
In this study, bone marrow aspirates and biopsies were
performed in 26 (93%) patients. Positive hemophagocytosis
were present in only three (11.5%) patients. The other bone
marrow biopsies revealed reactive bone marrow, presented
with erythroid and granulocytic hyperplasia without
evidence of hymophagocytosis.

100
Therapeutic strategies and response to treatment
% of patients presenting with symptoms

90
The combination of nonsteroidal anti-inflammatory drugs
80
(NSAIDs) and corticosteroids is often needed to control the
70
extra-articular features. All of the 28 patients had NSAIDs
60 as their primary anti-inflammatory therapy. However, 25
50 (89.3%) initially required additional corticosteroids, while
40 27 patients (96.4%) took other disease-modifying antirheu-
30 matic drugs (DMARDs). Eleven patients (39.3%) further
received etanercept, a biological agent of TNF-a antago-
20
nist. Among patients taking etanercept, six presented with
10
polyarticular pattern, four with oligoarticular pattern, and
0
one patient was initially without articular symptoms.
The corticosteroid treatment was mainly used for the
ti s
is

sh
r

tis

VD
re
he
ve

y
hy
rit

al

ei
izu
ra

si
ac
fe

CR
t

eg
th

pa

uv
ro

acute phases of the systemic disease or during acute flares.

Se
ad
ar

se
no

He
no
e
ad

Once the symptoms were controlled, the dose was gradu-

ga
ph

or
m

ally tapered as much as possible to minimize the side

ly

*CRVD: central retinal vein occlusion

effects. Thus, in order to clarify the predictive factors for
a refractory disease course, the patients were classified
Figure 2 Initial manifestations of SoJIA. The most common into two groups according to the period of steroid use (>6
presentations were fever (100%), arthritis (89.3%), and skin months and <6 months) and were compared by basic profile
rash (67.9%). and laboratory parameters on diagnosis. Only two items,
Systemic onset juvenile idiopathic arthritis 545

Figure 3 (A) Pattern of arthritis. Oligoarticular pattern was the most common and 11% of SoJIA patients did not present with
arthritis on disease onset. (B) Location of involved joints. The most common joints involved were the knee (76%), ankle (56%), and
elbow and PIP joints (28%). (C) Pattern of fever. The most common pattern was intermittent (53%).

Table 1 Basic laboratory data at initial presentation; and patient numbers and percentage of abnormal laboratory data.
Laboratory values Mean (range) Laboratory data Patient numbers (%)
Hb (g/dl) 10.12 (5.2e13.3) Anemia (Hb<10 g/dL) 10/28 (35.7)
Platelet ( 109/L) 458.86 (114e874) Leukocytosis (WBC>15  109/L) 11/28 (39.3)
WBC ( 109/L) 14.781 (5.0e28.2) Thrombocytosis (platelets >400  109/L) 18/28 (64.3)
AST (U/L; n Z 27) 59.2 (12e331) Elevated AST 11/27(40.7)
ALT (U/L; n Z 27) 50.6 (6e249) Elevated ALT 9/27 (33.3)
C3 (mg/dL; n Z 27) 167.1 (91e315) Elevated CRP (>0.8 mg/dL) 28/28 (100)
C4 (mg/dL; n Z 27) 33.4 (14e65.6) Elevated LDH (>460 U/L) 20/22 (90.9)
ESR 1hr (mm; n Z 20) 86.45 (31e140) Elevated ferritin 10/14 (71.4)
ESR 2hr (mm; n Z 20) 110.95 (55e150) Positive ANA 2/28 (7.14)
CRP (mg/dL) 10.7 (2.71e33.3) Positive RF 2/25 (8)
IgG (mg/dL; n Z 14) 1774.6 (671e3624) Positive HLA B-27 0/11 (0)
IgA (mg/dL; n Z 13) 257.2 (25e501)
IgM (mg/dL; n Z 13) 154.8 (66e325)
Iron (mg/dL; n Z 12) 27.4 (9e77)
TIBC (mg/dL; n Z 12) 252.5 (198e334)
Ferritin (ng/mL; n Z 14) 3075.5 (73e26900)
LDH (U/L; n Z 22) 938.9 (306e3707)
TNFa (n Z 2) 24.2 (12.7e35.7)
ALT Z alanine aminotransferase; AST Z aspartate aminotransferase; ESR Z erythrocyte sedimentation rate; LDH Z Lactate dehy-
drogenase; TIBC Z Total iron-binding capacity.
546 H.-Y. Tsai et al.

white blood cells (WBC) count (17.63
7.71 vs.
11.93
4.43  109/L, p < 0.05) and AST level (89.4 vs.
31.2 U/L, p < 0.05) on initial presentation, showed signifi-
cant differences between the two groups (Table 2). After
comparing features with inferential statistics using binary
logistic regression, corrected by age and sex, WBC count
was the only predictive factor for prolonged steroid use >6
months (p Z 0.039, odds ratio Z 1.00013, 95% confidence
interval 1.000011e1.0000256).
On the relationship between initial articular pattern and
duration of steroid use, prolonged steroid use >6 months
was more common in children with polyarticular onset
(63.6%) than with oligoarticular onset (42.9%) and without
articular involvement on onset (33.3%), although this was
not statistically significant (Table 3).
Clinical course
Eight patients (28.6%) achieved complete remission without
needing medication for >2 months after a mean of 3.1
months of medication (range 2e7 months). Seven patients
(25%) had drug-dependent courses with chronic persistent
arthritis, who did not achieve clinical remission after the
onset, or gained clinical remission but could not dis-
continue medications for >2 months. Thirteen (46.4%) had
relapsing courses, characterized by flares of articular or/
and extra-articular symptoms, mainly fever. Among those
presenting with unusual symptoms, one patient had
a relapse with CNS vasculitis and acute respiratory distress
syndrome, another relapsed with serositis and had mani-
festations of pleural effusion. Three patients underwent
bilateral total hip replacement and one additional total
knee replacement. Of the three MAS patients, two had
relapsing courses and the other had drug-dependent cour-
ses, indicating a refractory disease outcome. The clinical
course and pattern of articular involvement are summa-
rized in Table 3. Children with polyarticular onset (63.6%)
were more prone to develop frequent systemic flares than
those with oligoarticular onset (28.6%), although this was
not statistically significant.
Discussion
Systemic onset juvenile idiopathic arthritis is heteroge-
neous in severity and disease course. In contrast to other
JIA subtypes, where joint manifestations are more prom-
inent than general symptoms, SoJIA more often presents
with extra-articular features such as fever, rashes, and
lymphadenopathy. Furthermore, because the presentation
of SoJIA is highly variable, this study intended to charac-
terize the features of initial presentation of SoJIA. Leuko-
cytosis and polyarticular pattern on presentation may
indicate a refractory clinical course.
Consistent with the previous studies, SoJIA occurs in
children with a peak age of onset between 18 months and 2
years in two UK series, but can occur in children of any
age.7e9 The onset age of the children in this study is evenly
distributed. However, there should be more caution in
diagnosing SoJIA in children aged <1 year because the
incidence is relatively low. In contrast to the other JIA
subtypes that are female predominant, the incidence of
SoJIA is equal in both sexes among Caucasian populations of
Europe and North America,10,11 which is the same as the
results here.

Table 2 Comparison of basic profiles and laboratory data at diagnosis between the two patient groups with and without
prolonged steroid use.
Initial presentation Mean values
S.D. p value
Steroid use > 6 mo (n Z 14) Steroid use < 6 mo (n Z 14)
Age 8.33
4.15 8.97
3.82 NS
Admission days 19.64
13.02 11.86
6.69 NS
Total joints 9.29
10.48 6.21
9.45 NS
Hb (g/dL) 9.8
1.96 10.44
1.5 NS
Platelet (109/L) 445
211.53 472.71
170.82 NS
WBC (109/L) 17630
7710.27 11932.86
4429.16 0.024
AST (U/L) 89.38
107.62 (n Z 13) 31.21
16.58 0.05
ALT (U/L) 59.08
67.83 (n Z 13) 42.79
63.13 NS
C3 (mg/dL) 162.32
59.85 (n Z 13) 171.57
33.71 NS
C4 (mg/dL) 29.78
8.2 (n Z 13) 36.76
16.3 NS
ESR 1hr (mm) 87.63
39.14 (n Z 8) 85.67
26.75 (n Z 12) NS
ESR 2hr (mm) 106
32.83 (n Z 8) 114.25
21.6 (n Z 12) NS
CRP (mg/dL) 12.1
8.06 9.37
5.39 NS
IgG (mg/dL) 1736.13
886.85 (n Z 8) 1826
508 (n Z 6) NS
IgA (mg/dL) 280.66
165.71 (n Z 7) 229.83
144.97 (n Z 6) NS
IgM (mg/dL) 170.57
101.83 (n Z 7) 136.5
65.8 (n Z 6) NS
Iron (mg/dL) 23.80
16.54 (n Z 5) 30.00
23.25 (n Z 7) NS
TIBC (mg/dL) 259.40
50.06 (n Z 5) 247.57
23.24 (n Z 7) NS
Ferritin (ng/mL) 6094.02
10312.35 (n Z 6) 811.59
1039.15 (n Z 8) NS
LDH (U/L) 1115.80
1020.64 (n Z 10) 791.50
400.07 (n Z 12) NS
ALT Z alanine aminotransferase; AST Z aspartate aminotransferase; ESR Z erythrocyte sedimentation rate; LDH Z Lactate dehy-
drogenase; TIBC Z Total iron-binding capacity.
Systemic onset juvenile idiopathic arthritis
Table 3 Association between pattern of arthritis and clinical course.
Non-articular (n Z 3)
y
Group I (n Z 8) 0 5
Group IIy (n Z 7) 1 (33.3%)
Group IIIy (n Z 13) 2 (66.7%)
Steroid use > 6 mo 1 (33.3%)
y
Group I: remission course; Group II: drug-dependent course; Group III: relapse course.
a
Expressed as the number of patients (percentage).
Although the classic pattern of quotidian fever is high
spiking fever in the evenings,12 it is only the second most
frequent fever pattern in this study. Instead, intermittent
fever is the most common on presentation. Therefore,
fever pattern is highly variable and does not necessarily
fulfill the criteria of ILAR even though it is universally
present at disease onset (Fig. 2). Similarly, arthritis is
common on initial presentation (89.3%), with knee, ankle,
elbow, and PIP joints being the most commonly affected. In
contrast to previous report where the wrist is the most
common joint affected in SoJIA,9 the wrist is the fourth
most common affected joint in this study (24%). However,
the incidence is still much higher than pauciarticular JIA,
where the wrist is only involved in 4% of patients.13 The
sacroiliac joint is the least joint affected, which is the same
result as seen in a previous study.9 Interestingly, 11% of
patients did not have arthritis on initial presentation,
indicating that the lack of arthritis during disease onset
does not exclude the diagnosis of SoJIA. Uveitis complicates
several forms of juvenile idiopathic arthritis (oligoarthritis,
rheumatoid factor-negative polyarthritis, psoriatic
arthritis, and enthesitis-related arthritis), but is *rare in
SoJIA.1,14 Central retinal vein occlusion was an extremely
rare complication that might be related to thrombosis
associated with antiphospholipid antibody,15 even though
there was no similar finding in our patient.
Instead of specific laboratory tests for SoJIA, the char-
acteristic patterns of laboratory abnormalities indicate
systemic inflammation, including the up-regulation of ESRs,
CRP levels,16 granulocytosis, and thrombocytosis. In this
study, serum ferritin and CRP levels were elevated in most
patients, which is compatible with previous reports that
described those parameters to be correlated with systemic
activity.17,18 Elevated ferritin levels may be due to
augmented interleukin 1b signaling9 since IL-1b is increas-
ingly recognized in SoJIA patients and induces ferritin
mRNA translation.19,20 LDH is not a valuable parameter for
differentiating malignancy from systemic inflammation
because it is increased in most SoJIA patients. Liver
enzymes may be abnormal in more severe cases, and
together with leukocytosis, can explain the need for
corticosteroids for a longer period. While reviewing charts
retrospectively, patients who were HLA-B27 seropositive,
which is regarded as an exclusion criterion of SoJIA, were
excluded from this study. However, HLA typing may be
valuable for predicting joint destruction and severity of
systemic manifestation of SoJIA. The statistically significant
increase in the incidence of HLA-DR4 and HLA-DR3 was
observed in the subgroup of patients with radiological
547
Onset typea
Oligoarticular (n Z 14) Polyarticular (n Z 11)
(35.7%) 3 (27.3%)
5 (35.7%) 1 (9.10%)
4 (28.6%) 7 (63.6%)
6 (42.9%) 7 (63.6%)
evidence of joint destruction and with severe systemic
symptoms persisting or recurring for >2 years in comparison
with controls respectively.21 HLA-DQA1*05 and DRB1*11
were associated with systemic arthritis. The DRB1*11-
DQA1*05-DQB1*03 haplotype was also increased in
systemic arthritis.22
Macrophage activation syndrome is a serious complica-
tion of SoJIA. Its strongest laboratory discriminators are
decreased platelet count, increased AST, leukopenia, and
hypofibrinogenemia.23 Behren et al9 reported that patients
with MAS have lower WBC and ESR than those with a normal
bone marrow. In the current study, the number of MAS
patients (n Z 3) is too small to allow for statistical
comparisons. All three MAS patients needed steroid treat-
ment for >6 months. They presented leukocytosis at
disease onset and evolved to cytopenia while MAS devel-
oped. Therefore, the effect of cytopenia on the prediction
for poor outcome was considered to be independent of our
statistical comparison.
The aim of therapy in SoJIA is the suppression of unre-
strained systemic and local inflammation in order to ach-
ieve disease remission and avoid disease relapse and
chronic morbidity. NSAIDs, with corticosteroids, are still the
appropriate initial therapy.24 NSAIDs also remain the major
treatment for SoJIA in this study. Nevertheless, in the last
decade, there have been many therapeutic advances in
terms of treatment. Aside from extended use of cortico-
steroids and methotrexate, there is a trend for the early
introduction of other DMARDs.25e27 Recently, the avail-
ability of newer biologic agents, such as anti-TNF-alpha
agents,28,29 anakinra (an interleukin-1 receptor antago-
nist),30,31 and tocilizumab (an anti-interleukin-6-receptor
monoclonal antibody),32 has provided additional options
for refractory cases, although these have not yet been
proven consistently effective. In this study, 89.3% of the
patients received corticosteroids for controlling acute
symptoms. DMARDs are introduced early in the disease
course in most of these patients, while etanercept is
applied for some patients, mostly those with a polyarticular
pattern at onset. Proper and early application of these
agents may prevent morbidity related to long-term corti-
costeroids use. Considerable improvement in the prognosis
of SoJIA is attributed to this progress in therapy.
The clinical course and outcomes of SoJIA are highly
variable. The previous study reports that about 20% to 30%
of children with SoJIA develop destructive polyarthritis
with long-term morbidity and disability.33,34 However,
prognosis is highly variable among previous reports in
Taiwan. Lin et al17 reported that 43% (9/21) of children with
548
SoJIA had chronic polyarthritis with duration >6 months,
five of whom had progressive articular damage leading to
disability. In contrast, Wang et al35 showed that none of
12 children with Still’s disease developed destructive
arthritis. It was consistent with the current study that only
three patients developed chronic disability and needed
total joint replacement.
About 50% of SoJIA patients have persistent disease
activity or remitting flares, or depend on high-dose
steroids.29,36 Some studies show that steroid dependence
and the associated morbidity occur in nearly one-third of
patients.12,34,37 In this study, 50% of patients were depen-
dent on steroids for >6 months. Eventually, 28.6% of
patients developed a monocyclic pattern with complete
remission, 46.4% children had acute flares, and 25% devel-
oped drug-dependent persistent arthritis. Lomater et al34
reported that among patients with SoJIA, the frequency
of remission was 100% in those with a monocyclic course,
37% in those with intermittent course, and 23% in those with
persistently active course. In Taiwan, Lin et al17 reported
that 71% of patients had monocyclic pattern, while Hsu
et al38 showed that 44% of 16 SoJIA children had clinical
remission without relapse, 19% had relapse, and 37%
became drug-dependent.
Some studies have clarified the predictive factors for the
prognosis of SoJIA. Polyarticular pattern with hip involve-
ment at 6 months predicts a high articular index at 3 years
after onset.16 The presence of active systemic disease at 6
months, as characterized by fever or the need for cortico-
steroids, and thrombocytosis are strongly correlated with
a high CHAQ score at a minimum of 2 years after onset, and
predicts poor functional outcome.37 Schneider et al33
described that the development of destructive changes
within 2 years after disease onset is associated with
persistent systemic features and thrombocytosis at 6
months. The current study reports that leukocytosis and
polyarticular pattern at onset may indicate a refractory
clinical course, may require corticosteroids for more than 6
months, and have higher relapse rate. Leukocytosis on
diagnosis suggests a more severe inflammation and poly-
articular involvement implies broad extension of disease
severity, which may explain the longer duration of treat-
ment and higher rate of flares. Thus, common predictors of
poor outcome in SoJIA are associated with greater severity
and extension of arthritis at onset, together with prolonged
active disease.
Adult-onset Still’s disease (AOSD) is an inflammatory
multisystemic disorder characterized by high spiking fever,
evanescent rash, arthralgia or arthritis, sore throat,
lymphadenopathy, hepatosplenomegaly, and laboratory
abnormalities including neutrophilic leukocytosis,
abnormal liver function tests, and increase in acute phase
reactants and serum ferritin.39 There are some possible
differences in the clinical characteristics between SoJIA
and AOSD. In Taiwan, Chen et al40 reported that the most
common clinical manifestations of AOSD were fever (100%),
articular symptoms (100%), evanescent rash (87%), and sore
throat (84%). However, among articular symptoms of AOSD,
the incidence of arthritis was only 76%, which was lower
than that of the children population. Otherwise, the inci-
dence of thrombocytosis was higher and the incidences of
sore throat and myalgia were lower in children from this
H.-Y. Tsai et al.
study and Lin et al’s study,17 compared to Chen et al’s
study.40 The disease course and outcome were similar
between SoJIA and AOSD, where the polycyclic systemic
course was the most common (45%), followed by mono-
cyclic systemic course (34%), and only 20% of patients
progressed to chronic arthropathy.
This study has some limitations, including its retro-
spective nature and limited patient numbers because of the
hospital-based population. The positive rheumatoid factors
have not been confirmed at least 3 months after the first
test. The family histories were not all specifically excluded
due to insufficient data. The limited numbers also
decreased the strength of analysis. However, the study has
been able to demonstrate the gross manifestations and
therapy of SoJIA, as well as the risk factors associated with
therapeutic response and flares. There has been only one
recent report17 about SoJIA among children in Taiwan, and
no report regarding the predictors for clinical course in
a Taiwanese population.
Conclusion
The clinical manifestations of SoJIA are different from
those of the other JIA subtypes. Its major initial features
are fever, arthritis, and skin rash, and a significant portion
(11%) of patients do not present with arthritis initially.
Leukocytosis and polyarticular pattern on presentation
indicate a refractory clinical course with longer steroid use
and higher rate of flares. The findings here may provide
valuable information for clinicians in decision-making for
the diagnosis and management of SoJIA.
Acknowledgments
This study was supported by the grants from National
Taiwan University Hospital.
References
1. Cassidy JTPR. Juvenile rheumatoid arthritis. Textbook of
pediatric rheumatology. 4th ed. Philadelphia: WB Saunders;
2001. p.219e220.
2. Brewer Jr EJ, Baum J, Baum J, Cassidy JT, Fink C, Jacobs J,
et al. Current proposed revision of JRA criteria. Arthritis
Rheum 1977;20(Suppl. 2):195e9.
3. Still GF. On a form of chronic joint disease in children. Am J Dis
Child 1978;132:195e200.
4. Cassidy JT, Levinson JE, Bass JC, Baum J, Brewer Jr EJ, Fink CW,
et al. A study of classification criteria for a diagnosis of juvenile
rheumatoid arthritis. Arthritis Rheum 1986;29:274e81.
5. Petty RE, Southwood TR, Manners P, Baum J, Glass DN,
Goldenberg J, et al. International League of Associations for
Rheumatology classification of juvenile idiopathic arthritis:
second revision, Edmonton, 2001. J Rheumatol 2004;31:390e2.
6. Lee FH, Hseih HK. Clinical observation of juvenile rheumatoid
arthritis. Acta Paediatr Sin 1984;25:135e43.
7. Symmons DP, Jones M, Osborne J, Sills J, Southwood TR, Woo P.
Pediatric rheumatology in the United Kingdom: data from the
British Pediatric Rheumatology Group National Diagnostic
Register. J Rheumatol 1996;23:1975e80.
8. Fishman D, Faulds G, Jeffery R, Mohamed-Ali V, Yudkin JS,
Humphries S, et al. The effect of novel polymorphisms in the
Systemic onset juvenile idiopathic arthritis
interleukin-6 (IL-6) gene on IL-6 transcription and plasma IL-6
levels, and an association with systemic-onset juvenile
chronic arthritis. J Clin Invest 1998;102:1369e76.
9. Behrens EM, Beukelman T, Gallo L, Spangler J, Rosenkranz M,
Arkachaisri T, et al. Evaluation of the presentation of systemic
onset juvenile rheumatoid arthritis: data from the Pennsylva-
nia Systemic Onset Juvenile Arthritis Registry (PASOJAR). J
Rheumatol 2008;35:343e8.
10. Woo P. Systemic juvenile idiopathic arthritis: diagnosis,
management, and outcome. Nat Clin Pract Rheumatol 2006;2:
28e34.
11. Feldman BM, Birdi N, Boone JE, Dent PB, Duffy CM,
Ellsworth JE, et al. Seasonal onset of systemic-onset juvenile
rheumatoid arthritis. J Pediatr 1996;129:513e8.
12. Schneider R, Laxer RM. Systemic onset juvenile rheumatoid
arthritis. Baillieres Clin Rheumatol 1998;12:245e71.
13. Sharma S, Sherry DD. Joint distribution at presentation in
children with pauciarticular arthritis. J Pediatr 1999;134:
642e3.
14. Petty RE, Smith JR, Rosenbaum JT. Arthritis and uveitis in
children. A pediatric rheumatology perspective. Am J Oph-
thalmol 2003;135:879e84.
15. Andrews A, Hickling P. Thrombosis associated with anti-
phospholipid antibody in juvenile chronic arthritis. Lupus 1997;
6:556e7.
16. Modesto C, Woo P, Garcia-Consuegra J, Montecucco C,
Martini A. Systemic onset juvenile chronic arthritis, poly-
articular pattern and hip involvement as markers for a bad
prognosis. Clin Exp Rheumatol 2001;19:211e7.
17. Lin SJ, Huang JL, Chao HC, Lee WY, Yang MH. A follow-up study
of systemic-onset juvenile rheumatoid arthritis in children.
Acta Paediatr Taiwan 1999;40:176e81.
18. Pelkonen P, Swanljung K, Siimes MA. Ferritinemia as an indi-
cator of systemic disease activity in children with systemic
juvenile rheumatoid arthritis. Acta Paediatr Scand 1986;75:
64e8.
19. Rogers JT, Bridges KR, Durmowicz GP, Glass J, Auron PE,
Munro HN. Translational control during the acute phase
response. Ferritin synthesis in response to interleukin-1. J Biol
Chem 1990;265:14572e8.
20. Pascual V, Allantaz F, Arce E, Punaro M, Banchereau J. Role of
interleukin-1 (IL-1) in the pathogenesis of systemic onset
juvenile idiopathic arthritis and clinical response to IL-1
blockade. J Exp Med 2005;201:1479e86.
21. Musiej-Nowakowska E, Ma ska-Rusiniak B, Ko1odziejczyk B,
˛czyn
Gazda A. HLA-B27 and HLA-DR:prognostic factors in systemic
onset juvenile idiopathic arthritis. Reumatologia 2011;49:
23e31.
22. Thomson W, Barrett JH, Donn R, Pepper L, Kennedy LJ,
Ollier WE, et al. Juvenile idiopathic arthritis classified by the
ILAR criteria: HLA associations in UK patients. Rheumatology
2002;41:1183e9.
23. Ravelli A, Magni-Manzoni S, Pistorio A, Besana C, Foti T,
Ruperto N, et al. Preliminary diagnostic guidelines for macro-
phage activation syndrome complicating systemic juvenile
idiopathic arthritis. J Pediatr 2005;146:598e604.
24. Kimura Y, Fieldston E, Devries-Vandervlugt B, Li S,
Imundo L. High dose, alternate day corticosteroids for
549
systemic onset juvenile rheumatoid arthritis. J Rheumatol
2000;27:2018e24.
25. Levinson JE, Wallace CA. Dismantling the pyramid. J Rheu-
matol 1992;Suppl 33:6e10.
26. Giannini EH, Cawkwell GD. Drug treatment in children with
juvenile rheumatoid arthritis. Past, present, and future.
Pediatr Clin North Am 1995;42:1099e125.
27. Ilowite NT. Current treatment of juvenile rheumatoid arthritis.
Pediatrics 2002;109:109e15.
28. Lovell DJ, Giannini EH, Reiff A, Cawkwell GD, Silverman ED,
Nocton JJ, et al. Etanercept in children with polyarticular
juvenile rheumatoid arthritis. Pediatric Rheumatology Collab-
orative Study Group. N Engl J Med 2000;342:763e9.
29. Kimura Y, Pinho P, Walco G, Higgins G, Hummell D, Szer I, et al.
Etanercept treatment in patients with refractory systemic onset
juvenile rheumatoid arthritis. J Rheumatol 2005;32:935e42.
30. Swart JF, Barug D, Möhlmann M, Wulffraat NM. The efficacy
and safety of interleukin-1-receptor antagonist anakinra in the
treatment of systemic juvenile idiopathic arthritis. Expert
Opin Biol Ther 2010;10:1743e52.
31. Quartier P, Allantaz F, Cimaz R, Pillet P, Messiaen C, Bardin C,
et al. A multicentre, randomised, double-blind, placebo-
controlled trial with the interleukin-1 receptor antagonist
anakinra in patients with systemic-onset juvenile idiopathic
arthritis (ANAJIS trial). Ann Rheum Dis 2011;70:747e54.
32. Yokota S, Imagawa T, Mori M, Miyamae T, Aihara Y, Takei S,
et al. Efficacy and safety of tocilizumab in patients with
systemic-onset juvenile idiopathic arthritis:a randomised,
double-blind, placebo-controlled, withdrawal phase III trial.
Lancet 2008;371:998e1006.
33. Schneider R, Lang BA, Reilly BJ, Laxer RM, Silverman ED,
Ibanez D, et al. Prognostic indicators of joint destruction in
systemic-onset juvenile rheumatoid arthritis. J Pediatr 1992;
120:200e5.
34. Lomater C, Gerloni V, Gattinara M, Mazzotti J, Cimaz R,
Fantini F. Systemic onset juvenile idiopathic arthritis: a retro-
spective study of 80 consecutive patients followed for 10
years. J Rheumatol 2000;27:491e6.
35. Wang YJ, Lee YP, Chi CS. Still’s disease: experience in 12
children. Zhonghua Min Guo Xiao Er Ke Yi Xue Hui Za Zhi 1993;
34:390e7.
36. Bowyer SL, Roettcher PA, Higgins GC, Adams B, Myers LK,
Wallace C, et al. Health status of patients with juvenile
rheumatoid arthritis at 1 and 5 years after diagnosis. J Rheu-
matol 2003;30:394e400.
37. Spiegel LR, Schneider R, Lang BA, Birdi N, Silverman ED,
Laxer RM, et al. Early predictors of poor functional outcome in
systemic-onset juvenile rheumatoid arthritis: a multicenter
cohort study. Arthritis Rheum 2000;43:2402e9.
38. Hsu CT, Lin YT, Yang YH, Chiang BL. Factors affecting clinical
and therapeutic outcomes of patients with juvenile rheuma-
toid arthritis. Scand J Rheumatol 2004;33:312e7.
39. Ohta A, Yamaguchi M, Kaneoka H, Nagayoshi T, Hiida M. Adult
Still’s disease: Review of 228 cases from the literature. J
Rheumatol 1987;14:1139e46.
40. Chen DY, Lan JL, Hsieh TY, Chin YH. Clinical manifestations,
disease course, and complications of adult-onset Still’s disease
in Taiwan. J Formos Med Assoc 2004;103:844e52.