Juvenile arthritis and the short-term risk for cardiometabolic outome

EXPERIMENTAL & CLINICAL CARDIOLOGY

Volume 20, Issue 8, 2014

Title: "Juvenile Arthritis and the Short-term Risk for Cardiometabolic Outcomes "

Authors: Victor C. Kok, Jorng-Tzong Horng and Jing-Long Huang

How to reference: Juvenile Arthritis and the Short-term Risk for Cardiometabolic Outcomes /VictorC. Kok, Jorng-Tzong Horng and Jing-Long Huang/Exp Clin Cardiol Vol 20 Issue8 pages 2832-2840/ 2014

1

)

EXPERIMENTAL & CLINICAL CARDIOLOGY

Juvenile Arthritis and the Short-term Risk for Cardiometabolic Outcomes

)

Original Article

)

18pt) Victor C. Kok1, 2, Jorng-Tzong Horng3, 2, Jing-Long Huang4, 5

1 Department of Internal Medicine, Kuang Tien General Hospital, Taichung, Taiwan; 2 Department of Biomedical Informatics, Asia University, Taichung, Taiwan; 3 Department of Computer Science and Information Engineering, National Central University,

Jhongli, Taiwan; 4 Division of Pediatric Allergy Asthma and Rheumatology, Department of Pediatrics, Chang

Gung Memorial Hospital at Linkou, Taiwan;

5 Chang Gung University College of Medicine, Taoyuan, Taiwan

Corresponding authors ([email protected]; [email protected])

© 2014 Kok VC et al.; licensee Cardiology Academic Press. This is an open access article distributed under the terms of the

Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution,

and reproduction in any medium, provided the original work is properly cited.

15pt)

)

Abstract Chronic inflammatory state is associated with

accelerated atherosclerosis and resultant cardiovascular

diseases (CVD) in the long run. The magnitude of the risk

for CVD in patients with juvenile idiopathic arthritis (JIA)

needs more research. We conducted a population-based

retrospective cohort study to investigate the association of

JIA and its anti-rheumatic therapy and subsequent

cardiometabolic outcomes. Cox model was adopted to

derive adjusted hazard ratios (aHR). We found that

children with JIA (n = 2,399) harbors an increased risk for

extracranial non-coronary arterial disorders with an aHR at

2.77 (95% confidence interval, 1.38-5.54, p < 0.004) than

subjects without JIA (n = 11,456) after a mean follow-up

duration of 6.53 (±1.22) years. This study also found

elevated risks for dyslipidemia (aHR 1.69, 1.13 – 2.54),

adolescent diabetes (2.06, 1.17 – 3.62) and gout (2.94, 2.13 –

4.04) in children with JIA compared to non-JIA cohort. The

risk for gout was elevated in every group of the JIA cohort

with the greatest risk at 6.42 (2.61 – 15.75) in children who

received anti-TNF. With this short follow-up period, there

were no increased risks for clinically evident myocarditis,

hypertension, ischemic heart disease, heart failure, or

cerebrovascular disease.

)

Keywords cardiometabolic outcomes; juvenile idiopathic

arthritis; retrospective cohort study; population-based

study; NHIRD; follow-up study. (

Introduction

In adults with rheumatoid arthritis, a chronic inflammatory

disease, premature atherosclerosis occurs early in the

course of illness contributing to an increased risk for

cardiovascular disease (CVD).[1-3] The magnitude of the

risk for CVD has been shown to be comparable to that of

type 2 diabetes, which is approximately 2 fold higher than

the general population in terms of the adjusted hazard ratio

(aHR).[4, 5] The explanatory mechanisms are thought to be

from the classical risk factors, such as hypertension,


Exp Clin Cardiol, Volume 20, Issue 8, 2014 - Page 2832

2

diabetes and dyslipidemia; the underlying systemic

autoimmune inflammatory state; as well as the use of non-

biologic (methotrexate, MTX), biologic (tumor necrosis

factor inhibitor, anti-TNF) disease-modifying anti-

rheumatic drugs (DMARDs) and corticosteroid for arthritis

control.[6, 7] Not only will the risk for coronary heart

disease and cerebrovascular event increase, but also the

risk of extracranial non-coronary arterial disease will be

elevated in patients with rheumatoid arthritis.[3, 8-11]

Juvenile idiopathic arthritis (JIA) defined as a spectrum of

several heterogeneous chronic inflammatory arthritis with

no apparent etiology with an onset before the age of 16, is

previously known as juvenile rheumatoid arthritis and

juvenile chronic arthritis. Since the disease occurs decades

earlier than adult rheumatoid arthritis, it is important to

pay attention to its association with short-term and long-

term cardiovascular outcomes. Recent research has

provided insights into this aspect.[12-14] These

investigations suggest that endothelial function measured

commonly as brachial artery flow-mediated dilation (FMD)

is impaired in children with JIA at a very young age. The

endothelial dysfunction coupled with hypercoagulable

state and resultant platelet reactivity might induce a state

of proatherothrombotic susceptibility contributing to

certain subsequent cardiometabolic outcomes.[15]

Nevertheless, there is still lacking of a population-based

cohort study examining the precise magnitude of the risk

for cardiometabolic outcomes in patients with JIA and

whether different forms of anti-rheumatic treatment will

independently affect the risk. We thus want to investigate

the association between JIA and clinically evident

cardiometabolic outcomes by designing a nationwide

population-based retrospective cohort study.

Patients and Method

¶

1. Data Source

This study used an administrative claim database, NHIRD,

derived from the Taiwan National Health Insurance

(TNHI). The TNHI program started on March 1, 1995 and

as of 2007, the coverage has reached 98.4% of Taiwan

population of 23 million people. The NHIRD consists of

four main database files: ambulatory expenditures by visit,

inpatient expenditures by admission, detailed files of

ambulatory care orders, and detailed files of inpatient

orders. These data files were de-identified by scrambling

the identification codes of both enrollees and medical

facilities. Detailed descriptions of the content of the NHIRD

data files, validity, and research use are well documented

in the literature. The authors of this study have used this

database for several population-based research.[16-21]

2. Study Design

The study design to examine the association of JIA and

selected cardiometabolic outcomes was retrospective

cohort study. We assembled the comparator non-JIA cohort

by 1 to 4 matching by age and sex. The study was approved

of exemption from review by the Chang Gung Memorial

Hospital Ethics Committee.

3. Case definition and definition of outcomes

Pediatrician in Taiwan commonly diagnose children under

the age of 16 as JIA using the International League of

Associations for Rheumatology (ILAR) classification

criteria, in which, the corresponding groupings of the

International Classification of Diseases-ninth edition-

Clinical Modification (ICD-9-CM) codes are as follows:

714.2 systemic arthritis; 714.30 + 714.31 polyarthritis; 714.32

+ 714.33 (pauciarthritis or oligoarthritis, persistent and

extended; 720.0 alone or (720.0 + 714.3) enthesitis-related

arthritis; and 696.0 (psoriatic arthritis).[22] A diagnosis of

JIA was qualified in our study when three requirements can

be fulfilled: age under 16, at least three times claims filed by

a pediatrician(s) or a pediatric rheumatologist(s) using any

of the above ICD-9-CM codes with an interval of at least 6

months between the first and the third claims. The first

claim of an eligible subject became the index date.

The CVD outcomes were ischemic heart disease, angina,

heart failure, myocarditis, hypertension, cerebral stroke,

and peripheral arterial disease. The selected metabolic

disease outcomes were dyslipidemia, type 2 diabetes, and

gout. The corresponding ICD-9-CM codes are well known

to the medical community and distinctive enough to

recognize and three times appearance should be present if

given in the outpatient setting and only one time can

qualify if in the discharge diagnosis.

4. Study Cohorts

4.1 The formation of JIA cohort according to outcomes

From the NHIRD, we retrieved children who were aged 16

or under in 2003 to 2005. Then subjects with unknown age

or gender, type 1 diabetes, chronic kidney failure, organ

transplantation history, and other autoimmune disease

were excluded. Separate cohorts of JIA for CVD outcomes

and metabolic disease outcomes were then formed after

further excluding of subjects with pre-existing CVD or

metabolic disease respectively before the index date.

4.2 The formation of the comparison non-JIA cohort



3

The non-JIA cohort was formed after matching 1:4 by age

and gender of the index cases.

5. Longitudinal Follow-up of the Cohorts for Endpoints

All subjects from both JIA and non-JIA cohorts were

followed up until the occurrence of the cardiometabolic

outcomes, drop-out from the NHI, or December 31, 2010

(Figure 1).

Figure 1. Study flowchart showing the assembly of the JIA and non-JIA cohorts. JIA: juvenile

idiopathic arthritis; Met-D: metabolic disease.



4

6. Statistical Analysis

For estimation of the risk for subsequent cardiometabolic

outcomes adjusted by gender, age and follow-up duration,

a Cox proportional hazard model was chosen. Adjusted

HRs with 95% confidence intervals (CI) were presented.

Stratification of the JIA cohort was performed by different

treatment received into three groups: methotrexate (MTX)

use but no anti-TNF therapy group, MTX followed by anti-

TNF group and neither MTX nor anti-TNF group.

Results

In 2003 to 2005, there were 2,912 eligible children aged 16

or under for the accrual for further study. After excluding

subjects with pre-existing cardiometabolic conditions

before the index date, we sorted out two separate JIA

cohorts for CVD and metabolic disease outcomes. We then

assembled two corresponding comparator non-JIA cohorts

with 1 to 4 matching by age and gender of the children

in the JIA cohort. For the analysis of the CVD outcomes,

the JIA cohort had 2,864 children, and the non-JIA cohort

11,456 children. For the metabolic outcomes, the JIA cohort

had 2,787 children and the age- and gender-matched non-

JIA cohort 11,148 children.

1. Demographic data of the JIA and non-JIA cohorts

Table 1 demonstrated the demographic details of the JIA

cohort grouped by treatment and non-JIA cohort for the

cardiovascular disease outcome analysis. In our series,

there were more boys than girls in every group, 59.3% male

in the “MTX without anti-TNF group”, 62.2% in the “MTX

+ anti-TNF group” and 55.4% in the “neither MTX nor anti-

TNF group”. The mean age of the studied children was

approximately 11 years old. The mean duration of exposure

to MTX only was 2.9 years, whereas, it was 3.1 years for

those children exposed to MTX followed by anti-TNF

inhibitor(s).

JIA cohort Non-JIA cohort

MTX without anti-

TNF

MTX + anti-TNF Neither MTX nor

anti-TNF

Total patients 354 111 2,399 11,456

Sex, n (%)

Girl 144 (40.7) 42 (37.8) 1,071 (44.6) 5,028 (43.9)

Boy 210 (59.3) 69 (62.2) 1,328 (55.4) 6,428 (56.1)

Age, mean (std) 11.2 (3.2) 10.5 (3.7) 11.5 (3.5) 10.9 (3.7)

Age group, n (%)

1 – 5 20 (5.6) 12 (10.8) 200 (8.3) 928 (8.1)

6 – 10 111 (31.4) 38 (34.2) 552 (23.0) 2,804 (24.5)

11 – 15 223 (63.0) 61 (55.0) 1647 (68.7) 7,724 (67.4)

Duration of drug

exposure in year,

mean (std)

2.9 (2.8) 3.1 (1.8) NA NA

Follow-up in year,

mean (std)

6.7 (1.7) 6.9 (1.2) 6.5 (1.2) 7.7 (0.8)

Table 1. Demographic details of the JIA grouped by treatment and non-JIA cohorts for the cardiovascular disease

outcome analysis

anti-TNF: anti-tumor necrosis factor inhibitor; JIA: juvenile idiopathic arthritis; MTX: methotrexate; NA: not

applicable; std: standard deviation



5

Likewise, in the Table 2, the specific demographic details

can be found for the two cohorts followed up for the

metabolic outcome analysis.

2. Risk for the cardiometabolic outcomes

2.1 Relative risk for cardiometabolic outcomes

The relative risk (RR) for cardiovascular outcomes as a

whole was 2.13 (95% CI, 1.66 – 2.71) in the “Neither MTX

nor anti-TNF” group as compared with the non-JIA

group. It was 2.10 (1.22 – 3.60) in the “MTX without anti-

TNF” group (Table 3).

The population attributable risk (PAR) was estimated at

16.3% (95% CI, 9.95 – 22.66) in the “Neither MTX nor anti-

TNF” group versus the non-JIA cohort. PAR estimates the

proportion of CVD outcomes in the study population that

is attributable to the exposure (JIA). That means

somewhere between 10% and 23% of the cases with

cardiovascular outcomes in the study cohort are

associated with JIA (Table 3).

The RR for metabolic outcomes as a whole was 2.06 (1.64 –

2.59) and the PAR(%) was 15.6% (9.8 – 21.4) (Table 3).

2.2 CVD outcomes

With a mean follow-up period of approximately 7 years,

when compared to non-JIA children, there were no

increased risks for most types of CVD including ischemic

heart disease (aHR 0.83, 95% CI, 0.31 – 2.24), angina (1.65,

0.74 – 3.71), heart failure (0.75, 0.21 – 2.71), myocarditis

(1.58, 0.89 – 2.81), hypertension (1.40, 0.83 – 2.36), and

cerebral stroke (1.81, 0.95 – 3.43) in children with JIA with

neither MTX nor anti-TNF exposure which was the largest

group of the JIA cohort. But notably, there was statistically

significant increase of the risk for extracranial non-

coronary arterial disease with an aHR at 2.77 (1.38 – 5.54),

p = 0.004. (Table 4)

2.3 Metabolic outcomes

In contrast to the CVD outcomes analysis where except for

peripheral arterial disease, there was no increase of the risk

for most types of CVD, the Cox analysis of the metabolic

outcomes as a whole revealed elevated risk across the

board. There was seemingly escalation of the magnitude of

risk from 2.36 (1.87 – 2.97) in the neither MTX nor anti-TNF

group to 2.73 (1.67 – 4.47) in the MTX without anti-TNF

group and further to 3.99 (1.88 – 8.47) in the MTX + anti-

TNF group. This study also found elevated risks for

dyslipidemia (aHR 1.69, 1.13 – 2.54), adolescent diabetes

(aHR 2.06, 1.17 – 3.62) and gout (2.94, 2.13 – 4.04) in children

with JIA under the neither MTX nor anti-TNF group when

compared to non-JIA cohort. The risk for gout was elevated

in every group of the JIA cohort with the greatest risk at

6.42 (2.61 – 15.75) in children received anti-TNF. (Table 4)


MTX without anti-

TNF


anti-TNF

Total patients 341 107 2,339 11,148

Sex, n (%)

Girl 143 (41.9) 42 (39.3) 1,048 (44.8) 4,932 (44.2)

Boy 198 (58.1) 65 (60.7) 1,291 (55.2) 6,216 (55.8)

Age, mean (std) 11.1 (3.2) 10.4 (3.7) 11.5 (3.5) 10.9 (3.7)

Age group, n (%)

1 – 5 20 (5.9) 12 (11.2) 203 (8.7) 940 (8.4)

6 – 10 107 (31.4) 38 (35.5) 543 (23.2) 2,752 (24.7)

11 – 15 214 (62.8) 57 (53.3) 1,593 (68.1) 7,456 (66.9)

Duration of drug

exposure in year,

mean (std)

2.9 (2.8) 3.1 (1.9) NA NA

Follow-up in year,

mean (std)

6.6 (1.9) 6.7 (1.7) 6.5 (1.3) 7.7 (0.8)

Table 2. Demographic details of the JIA grouped by treatment and non-JIA cohorts for the metabolic disease outcome

analysis

anti-TNF: anti-tumor necrosis factor inhibitor; JIA: juvenile idiopathic arthritis; MTX: methotrexate; NA: not applicable;

std: standard deviation



6

Table 4. The risk for cardiometabolic outcomes represented as adjusted hazard ratio (aHR) derived from the Cox model comparing

subjects with juvenile idiopathic arthritis (JIA) grouped by treatment and non-JIA subjects (N = 11,456 for CVD; 11,148 for MetD).

The treatment for the management of JIA was categorized as whether methotrexate (MTX) or a tumor necrosis factor-alpha (TNF)

inhibitor was used. The reference group is children without JIA.

JIA cohort for CVD outcomes

MTX without anti-TNF (n = 354) MTX + anti-TNF (n = 111) Neither MTX nor anti-TNF (n =

2,399)

aHR (95% CI) P-value aHR (95% CI) P-value aHR (95% CI) P-value

Overall 2.40 (1.37-4.20) 0.002 1.82 (0.58-5.69) 0.30 2.38 (1.85-3.06) <0.0001

Ischemic

Heart Disease

2.30 (0.54-9.92) 0.263 NA NA 0.83 (0.31-2.24) 0.715

Angina 1.21 (0.16-9.06) 0.853 NA NA 1.65 (0.74-3.71) 0.222

Heart failure 1.56 (0.20-12.19) 0.669 NA NA 0.75 (0.21-2.71) 0.658

Myocarditis 1.06 (0.25-4.42) 0.936 NA NA 1.58 (0.89-2.81) 0.120

Hypertension 1.49 (0.54-4.12) 0.446 2.92 (0.39-21.93) 0.296 1.40 (0.83-2.36) 0.205

Cerebral

Stroke

NA NA 3.42 (0.45-25.88) 0.233 1.81 (0.95-3.43) 0.070

Peripheral

arterial

2.32 (0.54-10.01) 0.260 NA NA 2.77 (1.38-5.54) 0.004

JIA cohort for metabolic disease (MetD) outcomes

MTX without anti-TNF (n = 341) MTX + anti-TNF (n = 107) Neither MTX nor anti-TNF (n =

2,339)

aHR (95% CI) P-value aHR (95% CI) P-value aHR (95% CI) P-value

Overall 2.73 (1.67-4.47) <0.0001 3.99 (1.88-8.47) 0.0003 2.36 (1.87-2.97) <0.0001

Dyslipidemia 1.50 (0.55-4.08) 0.425 1.32 (0.18-9.46) 0.783 1.69 (1.13-2.54) 0.011

Type 2

diabetes

2.56 (0.80-8.27) 0.115 2.90 (0.40-21.09) 0.293 2.06 (1.17-3.62) 0.012

Gout 3.21 (1.62-6.34) 0.001 6.42 (2.61-15.75) <0.0001 2.94 (2.13-4.04) <0.0001

Table 3. The relative risks and population attributable risks of both cardiovascular and metabolic outcomes comparing

different groups of JIA cohort and non-JIA cohort


MTX without

anti-TNF


anti-TNF

No. of patients 354 111 2,399 11,456

Cardiovascular

Outcome, n (%)

Yes 13 (3.7) 3 (2.7) 89 (3.7) 200 (1.7)

No 341 (96.3) 108 (97.3) 2,310 (96.3) 11,256 (98.3)

RR (95%

CI)

2.10 (1.22–3.60) 1.55 (0.53–4.42) 2.13 (1.66–2.71) reference

PAR* (%) 3.2% (-0.07–6.47) 0.52% (-1.13–2.18) 16.3% (9.95–22.66) reference

Metabolic

Outcome, n (%)

Yes 17 (5) 7 (6.5) 103 (4.4) 238 (2.1)

No 324 (95) 100 (93.5) 2,236 (95.6) 10,910 (97.9)

RR (95%

CI)

2.34 (1.45–3.73) 3.06 (1.49–6.11) 2.06 (1.64 –2.59) reference

PAR (%) 3.8% (0.71–6.91) 1.9% (-0.14 –3.99) 15.6% (9.8–21.4) reference

anti-TNF: anti-tumor necrosis factor inhibitor; CI: confidence intervals; JIA: juvenile idiopathic arthritis; MTX:

methotrexate; NA: not applicable; No.: number; PAR: population attributable risk; RR: relative risk; std: standard

deviation



7

Discussion

With this population-based cohort study on Taiwanese JIA

subjects with a mean follow-up duration of seven years,

our study reveals an increased risk of clinically evident

cardiometabolic disorder in the aspect of peripheral arterial

disease, (secondary) dyslipidemia, type 2 diabetes and

gout as compared to non-JIA subjects. Previously there is

no such a study on the estimation of the magnitude of the

cardiometabolic risks in this population. The study design

is notable in that we deliberately exclude subjects with type

1 diabetes, chronic kidney failure, organ transplantation

and other type of systemic autoimmune disease at the

outset in both cohorts. Most importantly, subjects with pre-

existing clinically evident cardiometabolic disorders are

excluded before starting the follow-up. This pre-hoc

consideration in the study design makes further statistical

inferential analysis rather robust.

This study is also able, for the first time, to reveal that

physician-diagnosed gout has approximately three to six-

fold increase in patients with JIA with escalation of the

magnitude of risk by treatment incorporation of more

DMARDs for disease control. The group of JIA subjects

under the MTX + anti-TNF group theoretically is a group

of patients who require more intensive treatment for

disease control indicating more health problems. Second

diagnosis of gout in patients with JIA may be overlooked

since the joint pain could be attributed to the prior

idiopathic arthritis rather than an coexistence of metabolic

gout.[23] This study shows that the coexistence of gout can

be up to 4-fold increase in patients receiving more

subsequent lines of treatment when compared to non-JIA

patients. Therefore, clinicians should maintain a high index

of suspicion and resort to timely investigation for making

correct differential diagnosis.[23, 24]

The use of DMARDs for disease control and its association

with subsequent CVD or cardiometabolic risk is largely

unknown. A recent longitudinal study by Shen C-C et al

aiming to investigate whether serum lipid levels and

atherogenic indices including the ratios of total cholesterol

(TC) to high density lipoprotein-cholesterol (HDL-C) and

low density lipoprotein-cholesterol (LDL-C) to HDL-C

were associated with disease activity and the impact of

anti-rheumatic treatment.[25] The study nicely shows that

children with newly-diagnosed JIA who are corticosteroids

or DMARDs naïve have dyslipidemia and aberrant

atherogenic indices, which will improve after effective anti-

rheumatic treatment. Our study additionally provides an

estimation of the magnitude of the risk for dyslipidemia

being at 1.69 (1.13-2.54) during a follow-up duration of

seven years when compared to non-JIA subjects.

A few of the studies have demonstrated that subclinical

atherosclerosis, evidenced by abnormal carotid intima-

media wall combined thickness (IMT) index values or

abnormal aortic distensibility and pulse wave velocity

(PWV) of thoracic aorta on the phase contrast magnetic

resonance imaging, is present in prepubertal children with

JIA.[12, 13, 25, 26] One research has shown that the

systemic type JIA was associated with greater carotid IMT

compared to children with oligoarticular disease,

polyarticular disease, or controls.[13] This study for the

first time suggests that the risk for extracranial non-

coronary peripheral arterial disease in children with JIA is

approximately three fold higher than the general

population (Table 3). The result of our study strongly

echoes the assertion that the cardiometabolic risks have

become a major health concerns for subjects with

inflammatory rheumatic disease.

The strength of this study lies on the success to avoid

allocation bias when the non-JIA cohort is completely

checked for being free of JIA after the baseline up to the

longest follow-up as possible. Reverse causation

(protopathic) bias is also not the concern inasmuch as we

have excluded pre-existing cardiometabolic disorders

present before the index date. There is no cross-over of

subjects between the two cohorts and tracking of the

enrollees has been perfect in both cohorts. However, there

are some limitations to be noted. Owing to the personal

data de-identified nature of the database, we are not able

to perform medical record retrieval, interviews, or

questionnaire research. Similarly, the biochemistry data,

body mass index, and smoking history are not available for

review.

We have confirmed the presence of statistically significant

increase of the risk of clinically evident cardiometabolic

disorders in terms of peripheral arterial disease,

dyslipidemia, type 2 diabetes and gout during a follow-up

duration of seven years in children with juvenile idiopathic

arthritis compared to non-JIA individuals. This should

prompt for further research into this arena since the

effectiveness of anti-rheumatic therapy will render more

children into long-term remission of the idiopathic arthritis

and thus survive into adulthood.

Acknowledgments

The authors thank the National Health Research Institute

for her kind provision of access to the NHIRD. We also

thank Mr. Neil Chuang, MSc, for his technical support as a

research assistant at the National Central University. The



8

interpretation and conclusions contained herein do not

represent those of the aforementioned institutions.

References

t)

[1] Maradit-Kremers H, Nicola PJ, Crowson CS, Ballman

KV, Gabriel SE. Cardiovascular death in rheumatoid

arthritis: a population-based study. Arthritis and

rheumatism. 2005;52:722-32.

[2] Gkaliagkousi E, Gavriilaki E, Doumas M, Petidis K,

Aslanidis S, Stella D. Cardiovascular risk in rheumatoid

arthritis: pathogenesis, diagnosis, and management.

Journal of clinical rheumatology : practical reports on

rheumatic & musculoskeletal diseases. 2012;18:422-30.

[3] Mohammad A, Lohan D, Bergin D, Mooney S, Newell

J, O'Donnell M, et al. The prevalence of aortic calcification

on vertebral fracture assessment imaging among patients

with rheumatoid arthritis. Journal of clinical

densitometry : the official journal of the International

Society for Clinical Densitometry. 2014;17:72-7.

[4] Peters MJ, van Halm VP, Voskuyl AE, Smulders YM,

Boers M, Lems WF, et al. Does rheumatoid arthritis equal

diabetes mellitus as an independent risk factor for

cardiovascular disease? A prospective study. Arthritis and

rheumatism. 2009;61:1571-9.

[5] van Halm VP, Peters MJ, Voskuyl AE, Boers M, Lems

WF, Visser M, et al. Rheumatoid arthritis versus diabetes

as a risk factor for cardiovascular disease: a cross-sectional

study, the CARRE Investigation. Annals of the rheumatic

diseases. 2009;68:1395-400.

[6] Roubille C, Martel-Pelletier J, Haraoui B, Tardif JC,

Pelletier JP. Biologics and the cardiovascular system: a

double-edged sword. Anti-inflammatory & anti-allergy

agents in medicinal chemistry. 2013;12:68-82.

[7] Solomon DH, Curtis JR, Saag KG, Lii J, Chen L,

Harrold LR, et al. Cardiovascular risk in rheumatoid

arthritis: comparing TNF-alpha blockade with nonbiologic

DMARDs. The American journal of medicine.

2013;126:730.e9-.e17.

[8] Stamatelopoulos KS, Kitas GD, Papamichael CM,

Kyrkou K, Zampeli E, Fragiadaki K, et al. Subclinical

peripheral arterial disease in rheumatoid arthritis.

Atherosclerosis. 2010;212:305-9.

[9] del Rincon I, Haas RW, Pogosian S, Escalante A. Lower

limb arterial incompressibility and obstruction in

rheumatoid arthritis. Annals of the rheumatic diseases.

2005;64:425-32.

[10] Roman MJ, Devereux RB, Schwartz JE, Lockshin MD,

Paget SA, Davis A, et al. Arterial stiffness in chronic

inflammatory diseases. Hypertension. 2005;46:194-9.

[11] Henke PK, Sukheepod P, Proctor MC, Upchurch GR,

Jr., Stanley JC. Clinical relevance of peripheral vascular

occlusive disease in patients with rheumatoid arthritis

and systemic lupus erythematosus. Journal of vascular

surgery. 2003;38:111-5.

[12] Breda L, Di Marzio D, Giannini C, Gaspari S, Nozzi

M, Scarinci A, et al. Relationship between inflammatory

markers, oxidant-antioxidant status and intima-media

thickness in prepubertal children with juvenile idiopathic

arthritis. Clinical research in cardiology : official journal of

the German Cardiac Society. 2013;102:63-71.

[13] Vlahos AP, Theocharis P, Bechlioulis A, Naka KK,

Vakalis K, Papamichael ND, et al. Changes in vascular

function and structure in juvenile idiopathic arthritis.

Arthritis care & research. 2011;63:1736-44.

[14] Jednacz E, Rutkowska-Sak L. Atherosclerosis in

juvenile idiopathic arthritis. Mediators of inflammation.

2012;2012:714732.

[15] Palomo I, Moore-Carrasco R, Alarcon M, Rojas A,

Espana F, Andres V, et al. Pathophysiology of the

proatherothrombotic state in the metabolic syndrome.

Frontiers in bioscience. 2010;2:194-208.

[16] Chang CW, Kok VC, Tseng TC, Horng JT, Liu CE.

Diabetic patients with severe sepsis admitted to intensive

care unit do not fare worse than non-diabetic patients: a

nationwide population-based cohort study. PloS one.

2012;7:e50729.

[17] Chen TA, Kang HY, Chang HC, Lin WC, Chao TM,

Horng JT. Gender differences in colorectal cancer during

the past 20 years in Taiwan. International journal of

colorectal disease. 2012;27:345-53.

[18] Kok VC, Horng JT, Lin HL, Chen YC, Chen YJ, Cheng

KF. Gout and subsequent increased risk of cardiovascular

mortality in non-diabetics aged 50 and above: a

population-based cohort study in Taiwan. BMC

cardiovascular disorders. 2012;12:108.

[19] Chen TA, Horng JT, Lin WC. Metachronous colorectal

cancer in Taiwan: analyzing 20 years of data from Taiwan

Cancer Registry. International journal of clinical oncology.

2013;18:267-72.

[20] Kok VC, Horng JT, Agustriawan D. Statins use

increases the risk of urinary tract cancer: preliminary

results of a nationwide population-based case-control

study. Asia-Pacific journal of clinical oncology. 2013;9:190-

1.

[21] Kok VC, Lee C-K, Horng J-T, Lin C-C, Sung F-C.

Reappraisal of the Etiology of Extracorpuscular Non-

Autoimmune Acquired Hemolytic Anemia in 2657

Hospitalized Patients with Non-Neoplastic Disease.

Clinical Medicine Insights: Pathology. 2014;7:11-4.

[22] Yu HH, Chen PC, Wang LC, Lee JH, Lin YT, Yang

YH, et al. Juvenile idiopathic arthritis-associated uveitis: a

nationwide population-based study in Taiwan. PloS one.

2013;8:e70625.

[23] Morris H, Grant K, Khanna G, White AJ. Gout in a 15-

year-old boy with juvenile idiopathic arthritis: a case

study. Pediatric rheumatology online journal. 2014;12:1.

[24] Chen SY, Shen ML. Juvenile gout in Taiwan

associated with family history and overweight. The

Journal of rheumatology. 2007;34:2308-11.



9

[25] Shen CC, Yao TC, Yeh KW, Huang JL. Association of

disease activity and anti-rheumatic treatment in juvenile

idiopathic arthritis with serum lipid profiles: a

prospective study. Seminars in arthritis and rheumatism.

2013;42:590-6.

[26] Argyropoulou MI, Kiortsis DN, Daskas N, Xydis V,

Mavridis A, Efremidis SC, et al. Distensibility and pulse

wave velocity of the thoracic aorta in patients with

juvenile idiopathic arthritis: an MRI study. Clinical and

experimental rheumatology. 2003;21:794-7.



Juvenile arthritis and the short-term risk for cardiometabolic outome

Documents