Annals of Internal Medicine Review

Meta-analysis: Diagnostic Accuracy of Anti–Cyclic Citrullinated Peptide

Antibody and Rheumatoid Factor for Rheumatoid Arthritis
Kunihiro Nishimura, MD, MPH; MS; Daisuke Sugiyama, MD, MPH; Yoshinori Kogata, MD; Goh Tsuji, MD, PhD; Takashi Nakazawa, MD, PhD;
Seiji Kawano, MD, PhD; Katsuyasu Saigo, MD, PhD; Akio Morinobu, MD, PhD; Masahiro Koshiba, MD, PhD; Karen M. Kuntz, ScD;
Isao Kamae, MD, DrPH; and Shunichi Kumagai, MD, PhD

Background: Rheumatoid factor (RF) and autoantibodies against 50 studies of RF. The pooled sensitivity, specificity, and positive and
cyclic citrullinated peptide (CCP) are markers that might help phy- negative likelihood ratios for anti-CCP antibody were 67% (95%
sicians diagnose rheumatoid arthritis. CI, 62% to 72%), 95% (CI, 94% to 97%), 12.46 (CI, 9.72 to
15.98), and 0.36 (CI, 0.31 to 0.42), respectively. For IgM RF, the
Purpose: To determine whether anti-CCP antibody more accurately values were 69% (CI, 65% to 73%), 85% (CI, 82% to 88%), 4.86
identifies patients with rheumatoid arthritis and better predicts ra- (CI, 3.95 to 5.97), and 0.38 (CI, 0.33 to 0.44). Likelihood ratios
diographic progression than does RF. among IgM RF, IgG RF, and IgA RF seemed to be similar. Results
Data Sources: MEDLINE through September 2006 and reference from studies of patients with early rheumatoid arthritis were similar
lists of retrieved studies and review articles. to those from all studies. Three of 4 studies found that risk for
radiographic progression was greater with anti-CCP antibody pos-
Study Selection: Studies in any language that enrolled at least 10 itivity than with IgM RF positivity.
participants and that examined the role of anti-CCP antibody and
RF in the diagnosis or prognosis of known or suspected rheumatoid Limitations: Many studies had methodological limitations. Studies
arthritis. of RF were heterogeneous and had wide ranges of sensitivity and
specificity.
Data Extraction: Two authors independently evaluated studies for
inclusion, rated methodological quality, and abstracted relevant Conclusions: Anti-CCP antibodies are more specific than RF for
data. diagnosing rheumatoid arthritis and may better predict erosive
disease.
Data Synthesis: The DerSimonian–Laird random-effects method
was used to summarize sensitivities, specificities, and positive and Ann Intern Med. 2007;146:797-808. [Link]
negative likelihood ratios from 37 studies of anti-CCP antibody and For author affiliations, see end of text.

R heumatoid arthritis is the most common autoimmune

disease, affecting approximately 1% of the world’s
population (1). It causes persistent synovitis, pain, joint
data on the sensitivity, specificity, and likelihood ratios of
RF and anti-CCP antibodies for diagnosing rheumatoid
arthritis. We also summarize results of studies that assessed
destruction, and functional disability. Because irreversible the associations of these markers with development and
joint destruction can be prevented by intervention during radiographic progression of rheumatoid arthritis.
the first months of disease, early diagnosis of rheumatoid
arthritis is important (2– 4). METHODS
Rheumatoid factor (RF) is an antibody directed Data Sources and Searches
against the Fc region of IgG that has been used as a diag-
We developed a protocol for the review and followed
nostic marker for rheumatoid arthritis. However, it is non-
standard reporting guidelines (13, 14). We searched MED-
specific and may be present in healthy elderly persons or in
LINE for studies published in any language through Sep-
patients with other autoimmune and infectious diseases
tember 2006 that examined autoantibodies against citrul-
(5). Other rheumatoid arthritis–associated autoantibodies
linated proteins, rheumatoid factor, or both for the
known to be specific for rheumatoid arthritis include anti-
diagnosis of rheumatoid arthritis. Our searches (available
perinuclear factor and antikeratin antibodies (6, 7). Be-
cause of rigorous technical requirements for their detec-
tion, antiperinuclear factor and antikeratin antibodies have
never been widely used as markers for rheumatoid arthritis, See also:
despite their high specificity. The epitopes of their antigens
Print
are arginyl residues citrullinated by peptidyl arginine de-
Editors’ Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 798
iminase (8 –10). Some enzyme-linked immunosorbent as-
Editorial comment. . . . . . . . . . . . . . . . . . . . . . . . . . 816
says (ELISAs) use linear citrulline-containing peptides that
have similar sensitivity to and higher specificity than RF Web-Only
for diagnosing rheumatoid arthritis (11). To improve sen- Appendix Tables
sitivity, assays that use cyclic citrullinated peptide (CCP) Appendix Figure
were developed to detect anti-CCP antibody (12). Conversion of figures and table into slides
In this systematic review, we summarize published
© 2007 American College of Physicians 797

Downloaded From: [Link] by a Tulane University User on 09/03/2013

 Review Anti-CCP Antibody and Rheumatoid Factor for Diagnosis of Rheumatoid Arthritis

anti-CCP antibody test, technical quality of the RF test,

Context application of the reference or index test, blinding of ob-
Are autoantibodies against cyclic citrullinated peptide servers, description of the study sample, and cohort assem-
(CCP) better serum markers for rheumatoid arthritis than bly. We used ␬ coefficients to examine interrater agree-
rheumatoid factor (RF)? ment for our initial overall quality score (19) and resolved
Contribution
any item discrepancies through discussion.
This meta-analysis of 86 studies found that the positive Data Analysis
likelihood ratio for anti-CCP antibody was greater than We used a random-effects model to combine estimates
that for IgM RF for identifying patients with rheumatoid of sensitivity, specificity, and positive and negative likeli-
arthritis (12.5 vs. 4.9). Sensitivity was similar for the 2 hood ratios (19 –21). We planned analyses that were strat-
tests, although specificity of anti-CCP antibody (95%) was ified by generation of anti-CCP antibody assay (first [anti-
higher than specificity of IgM RF (85%). CCP1] second [anti-CCP2]) and by RF subtype (IgA, IgG,
Cautions and IgM). We analyzed subgroups of relevant studies that
included patients with early rheumatoid arthritis and that
Fewer studies evaluated anti-CCP antibody than RF. There
evaluated combination testing for anti-CCP antibody and
was possible publication bias for reporting positive findings
RF. We conducted a stratified analysis for different thresh-
regarding anti-CCP antibody.
old and measurement methods when we suspected hetero-
Implication geneity among studies. We also conducted threshold anal-
Anti-CCP antibody positivity seems to be more specific yses and metaregression to assess whether the threshold
than IgM RF positivity for identifying patients with rheu- effect and heterogeneity among studies existed (22).
matoid arthritis. We examined funnel plots for diagnostic odds ratios to
explore the possibility of publication bias (23). For analy-
—The Editors ses, we used MetaDiSc, version 1.1.4 (Hospital Universi-
tario Ramón y Cajal, Madrid, Spain); Stata, version 8.2
(Stata Corp., College Station, Texas); and R, version 2.21
on request) were based on combinations of the following (R Foundation for Statistical Computing, Vienna, Austria).
index terms: rheumatoid arthritis, antiperinuclear factor,
antikeratin antibody, citrullinated protein, anti– cyclic Role of the Funding Sources
citrullinated peptide, rheumatoid factor, sensitivity, specificity, This study was supported in part by a Grant-in-Aid
mass screening, predictive value of tests, receiver-operating for Young Scientists from the Ministry of Education, Cul-
characteristic curve, and accuracy. We also reviewed refer- ture, Sports, Science and Technology, Japan, and a grant
ence lists of retrieved studies and review articles. for Research on Allergic Disease and Immunology from
the Ministry of Health, Labor and Welfare, Japan. The
Study Selection
funding sources had no role in the collection, analysis, or
Two reviewers independently scanned abstracts that interpretation of the data or in the decision to submit the
met the inclusion criteria. We included studies that evalu- manuscript for publication.
ated the utility of assaying anti-CCP antibody or RF for
diagnosis of known or suspected rheumatoid arthritis, en-
rolled at least 10 participants, were published after 1987, RESULTS
and provided enough data to allow calculation of sensitiv- Search Results and Characteristics of Studies
ity and specificity for diagnosis of rheumatoid arthritis. We We identified 302 reports, of which 86 met the inclu-
used the 1987 revised American College of Rheumatology sion criteria (11, 12, 24 –106) (Appendix Figure, available
(ACR) criteria as the reference standard of rheumatoid ar- at [Link]). Thirty-seven studies in 14 949 patients
thritis (15). In general, we regarded reports of patients with (11, 12, 24, 26, 29 –38, 40 – 42, 44, 45, 47, 48, 50, 52, 54,
symptom duration of less than 1 year as studies of early 56, 58, 60 – 62, 64, 66, 67, 70, 74, 76, 97, 99, 100) re-
rheumatoid arthritis, although we also used the researchers’ ported on the diagnostic accuracy of anti-CCP antibody,
definitions of early rheumatoid arthritis. whereas 50 studies in 15 286 patients (12, 24, 27, 29, 30,
Data Extraction and Study Quality Assessment 32–37, 39, 40, 42– 44, 47, 48, 50, 52, 54, 55, 60 – 62, 64,
We extracted data by using a standard form that in- 66, 70, 72–74, 76, 80 – 85, 88 –98, 100) reported on the
cluded the demographic characteristics of the participants, diagnostic accuracy of RF.
inclusion and exclusion criteria, number of participants Appendix Table 1 (available at [Link]) (11,
who were evaluated with the index test, and methods of 24, 26, 29 –38, 40 – 42, 44, 45, 47, 48, 50, 52, 54, 56, 58,
antibody testing. Two investigators independently assessed 60 – 62, 64 – 67, 74, 76, 97, 99, 100) and Appendix Table
the design of the studies by using previously developed 2 (available at [Link]) (12, 24, 27, 29, 33–35, 37,
quality criteria for studies of diagnostic tests (16 –18). 39 – 43, 45, 47, 48, 52, 65, 66, 72–74, 76, 80, 81, 88,
These assessments addressed the technical quality of the 90 –92, 94 –98, 100) summarize the characteristics of the
798 5 June 2007 Annals of Internal Medicine Volume 146 • Number 11 [Link]

Downloaded From: [Link] by a Tulane University User on 09/03/2013

 Anti-CCP Antibody and Rheumatoid Factor for Diagnosis of Rheumatoid Arthritis Review

included studies. In anti-CCP antibody and IgM RF stud- IgM RF (Figure 1 and Figure 2). The pooled sensitivity
ies, respectively, the median numbers of participants were and specificity were 67% (CI, 65% to 68%) and 95% (CI,
404 and 226, their median ages were 57 years and 53 years, 95% to 96%), respectively, for anti-CCP antibody and
and the median proportions of women were 59% and 69% (CI, 68% to 70%) and 85% (CI, 84% to 86%) for
68%. Studies of anti-CCP antibody that were published IgM RF. Data that were limited to studies of patients with
after 2000 usually addressed anti-CCP2 assays. early rheumatoid arthritis were similar to those from all
Characteristics of control groups varied. Among the studies (data available from the authors on request).
anti-CCP antibody studies, 5 used patients with undiffer- Studies published before 2000 tended to report high
entiated arthritis, 13 used patients with other rheumatic sensitivity and specificity for RF compared with studies
diseases, 1 used healthy persons, 1 used hepatitis C carriers, published from 2000 onward. More recent studies re-
and 17 used a mix of healthy persons and patients with ported favorable specificities for anti-CCP antibody. Sum-
other diseases. Among the IgM RF studies, 5 used patients mary likelihood ratios for studies that directly compared
with undifferentiated arthritis, 16 used patients with other anti-CCP antibody and IgM RF (11, 12, 24, 26, 29 –38,
rheumatic diseases, 2 used healthy persons, 1 used hepatitis 40 – 42, 44, 45, 47, 48, 50, 52, 54, 56, 58, 60 – 62, 64, 66,
C carriers, 1 used patients with polymyalgia rheumatica, 67, 70, 74, 76, 97, 99, 100) were similar to summary data
and 22 used a mix of healthy persons and patients with from all studies. Positive likelihood ratios for anti-CCP
other diseases. Three studies did not report details on the antibody and IgM RF were 12.32 and 3.86, respectively.
control group. Negative likelihood ratios for anti-CCP antibody and IgM
RF were 0.40 and 0.41, respectively. Positive and negative
Study Quality likelihood ratios for IgA RF and IgG RF seemed to be
Only 1 study satisfied all criteria on our quality check- qualitatively similar to those for IgM RF (Figure 3). Strat-
list. Twenty-two studies (30%) met at least 70% of the ified analyses for IgM RF showed no major differences for
criteria, and 9 studies (10%) met fewer than 50% of the positive summary likelihood ratios or negative likelihood
criteria. The ␬ coefficient for interrater agreement was 0.92 ratios across the strata of cutoff values and measurement
on the quality score. methods (Table). The threshold effect for IgM RF is not
Most studies adequately described the technical as- statistically significant, and no covariate was statistically
pects of assaying anti-CCP antibody and RF. In 86% (32 significant in the metaregression model.
of 37) of anti-CCP antibody studies and 82% (41 of 50) of
RF studies, the 1987 revised ACR criteria were used as the Diagnostic Accuracy of Anti-CCP1, Anti-CCP2, and Both
reference standard for rheumatoid arthritis. Most studies Anti-CCP Antibody and IgM RF
did not explicitly mention blinding of investigators to the Twenty-nine studies in 11 821 patients (24, 26, 29 –
clinical assessment or to the reference standard. Most stud- 38, 40 – 42, 44, 45, 47, 48, 50, 52, 54, 56, 60, 62, 64, 97,
ies (90%) enrolled patients with known or suspected rheu- 99, 100) assessed anti-CCP2, whereas 5 studies in 2098
matoid arthritis. Characteristics of these patients were fully patients (61, 66, 67, 70, 74) assessed anti-CCP1.
described in just over half of the studies. Enrollment was Although the sensitivities and specificities were similar
prospective in 18 of 37 anti-CCP antibody studies and 25 to those in the anti-CCP1 studies, 3 studies (12, 58, 76)
of 50 RF studies. that used an in-house ELISA were excluded because incor-
Studies of RF showed a wide range of sensitivities and porating them introduced a positive threshold effect and
specificities (Appendix Table 1, available at [Link] caused heterogeneity among the studies. The summary
.org). One study (35) reported very low sensitivity and positive and negative likelihood ratios were 12.77 (CI,
specificity. In this study, 57% of control patients had con- 9.62 to 16.94) and 0.32 (CI, 0.27 to 0.38), respectively,
ditions that can present with RF-positive arthritis (primar- for anti-CCP2 and 13.03 (CI, 5.74 to 29.04) and 0.53
ily the Sjögren syndrome or Wegener granulomatosis). (CI, 0.46 to 0.61) for anti-CCP1 (Figure 4).
Laboratory techniques for measuring RF varied across Six studies in 1753 patients (12, 30, 37, 50, 64, 74)
studies. Fifteen studies used nephelometry, 16 used latex simultaneously measured anti-CCP antibody and RF,
agglutination, and 16 used ELISA. Twenty-two studies whereas 8 studies in 2837 patients (12, 30, 37, 42, 50, 64,
used less than 20 U/mL as the cutoff value for negative test 70, 74) performed 1 of the tests only when the results on
results, 11 used less than 40 U/mL as the cutoff value, and the other test were positive. For studies that required the
17 did not report cutoff values. presence of both anti-CCP antibody and IgM RF for a
positive result, the summary positive and negative likeli-
Diagnostic Accuracy of Anti-CCP Antibody and IgM RF, hood ratios were 15.72 (CI, 8.30 to 29.75) and 0.46 (CI,
IgA RF, and IgG RF 0.35 to 0.61), respectively. For studies that considered a
The summary positive and summary negative likeli- result positive if either anti-CCP antibody or IgM RF was
hood ratios, respectively, were 12.46 (95% CI, 9.72 to detected, the positive and negative summary likelihood ra-
15.98) and 0.36 (0.31 to 0.42) for anti-CCP antibody and tios were 4.32 (CI, 2.71 to 6.90) and 0.32 (CI, 0.25 to
4.86 (CI, 3.95 to 5.97) and 0.38 (CI, 0.33 to 0.44) for 0.42), respectively.
[Link] 5 June 2007 Annals of Internal Medicine Volume 146 • Number 11 799

Downloaded From: [Link] by a Tulane University User on 09/03/2013

 Review Anti-CCP Antibody and Rheumatoid Factor for Diagnosis of Rheumatoid Arthritis

Figure 1. Likelihood ratio (LR) for autoantibodies against cyclic citrullinated peptides (anti-CCP).

Study, Year (Reference) Positive LR (95% CI) Negative LR (95% CI)

Anti-CCP
Quinn et al., 2006 (24) 9.37 (5.16–17.02) 0.21 (0.16–0.28)
Fernández-Suárez et al., 2005 (36) 88.67 (5.55–1417.64) 0.42 (0.31–0.58)
Kwok et al., 2005 (33) 18.71 (4.73–73.96) 0.46 (0.38–0.56)
Greiner et al., 2005 (35) 37.49 (15.66–89.79) 0.20 (0.13–0.31)
Sauerland et al., 2005 (29) 13.35 (9.12–19.55) 0.27 (0.22–0.34)
Kamali et al., 2005 (34) 32.22 (4.54–228.53) 0.44 (0.32–0.62)
Aotsuka et al., 2005 (38) 4.65 (3.01 –7.16) 0.15 (0.09–0.24)
Choi et al., 2005 (37) 9.14 (5.97–13.99) 0.30 (0.25–0.35)
García-Berrocal et al., 2005 (99) 4.56 (2.41–8.64) 0.25 (0.16–0.39)
Nell et al., 2005 (32) 20.18 (5.02–81.10) 0.60 (0.51–0.71)
Raza et al., 2005 (30) 15.62 (4.99–48.89) 0.44 (0.31–0.63)
van Gaalen et al., 2005 (26) 12.95 (7.45–22.49) 0.48 (0.41–0.57)
Correa et al., 2004 (56) 11.57 (6.53–20.49) 0.11 (0.05–0.20)
De Rycke et al., 2004 (54) 27.53 (10.42–72.76) 0.25 (0.18–0.35)
Girelli et al., 2004 (50) 15.00 (3.82–58.95) 0.30 (0.18–0.51)
Grootenboer-Mignot et al., 2004 (48) 7.56 (3.87–14.78) 0.40 (0.34–0.48)
Hitchon et al., 2004 (47) 1.82 (0.99–3.34) 0.56 (0.34–0.93)
Kumagai et al., 2004 (45) 17.76 (10.53–29.96) 0.20 (0.13–0.31)
Lopez-Hoyos et al., 2004 (97) 21.72 (7.80–60.48) 0.01 (0.00–0.21)
Bombardieri et al., 2004 (100) 60.65 (3.83–959.73) 0.24 (0.13–0.46)
Nielen et al., 2005 (31) 9.98 (4.83–20.64) 0.45 (0.39–0.52)
Dubucquoi et al., 2004 (52) 42.11 (10.58–167.51) 0.36 (0.29–0.45)
Söderlin et al., 2004 (44) 11.59 (2.67–50.36) 0.58 (0.38–0.90)
Vallbracht et al., 2004 (42) 22.54 (12.82–39.63) 0.37 (0.31–0.43)
van Venrooij et al., 2004 (41) 22.52 (18.09–28.03) 0.23 (0.21–0.26)
Vittecoq et al., 2004 (40) 10.82 (4.49–26.09) 0.63 (0.56–0.71)
Bas et al., 2003 (66) 5.59 (3.75–8.33) 0.49 (0.41–0.57)
Lee and Schur, 2003 (64) 6.88 (4.11–11.55) 0.38 (0.29–0.49)
Rantapää-Dahlqvist et al., 2003 (62) 38.28 (18.08–81.08) 0.30 (0.21–0.44)
Saraux et al., 2003 (61) 6.64 (3.60–12.26) 0.58 (0.47–0.70)
Suzuki et al., 2003 (60) 7.92 (5.38–11.66) 0.14 (0.11–0.17)
Zeng et al., 2003 (58) 21.54 (10.20–45.51) 0.54 (0.47–0.62)
Jansen et al., 2003 (65) 17.20 (5.58–53.04) 0.59 (0.53–0.66)
Vincent et al., 2002 (67) 38.97 (18.53–81.93) 0.43 (0.37–0.50)
Bizzaro et al., 2001 (74) 18.94 (7.71–46.55) 0.60 (0.51–0.71)
Goldbach-Mansky et al., 2000 (76) 4.46 (2.48–8.02) 0.65 (0.55–0.77)
Schellekens et al., 1998 (11) 10.77 (6.29–18.45) 0.54 (0.46–0.63)
Total 12.46 (9.72–15.98) 0.36 (0.31–0.42)

0.01 1.00 100.00 0.01 1.00 100.00

Positive LR Negative LR

Prognostic Value of Anti-CCP Antibody and IgM RF tio for rheumatoid arthritis was 16.1 to 38.99 for anti-
Appendix Table 3 (available at [Link]) sum- CCP antibody positivity and 1.2 to 8.7 for RF positivity.
marizes the results of 5 studies of the association between Fifteen studies examined associations between marker
rheumatoid arthritis and anti-CCP antibody. The odds ra- positivity and radiographic progression (Appendix Table 4,
800 5 June 2007 Annals of Internal Medicine Volume 146 • Number 11 [Link]

Downloaded From: [Link] by a Tulane University User on 09/03/2013

 Anti-CCP Antibody and Rheumatoid Factor for Diagnosis of Rheumatoid Arthritis Review

Figure 2. Likelihood ratio (LR) for autoantibodies against IgM rheumatoid factor (RF).

Study, Year (Reference) Positive LR (95% CI) Negative LR (95% CI)

IgM RF
Quinn et al., 2006 (24) 1.38 (1.10–1.74) 0.68 (0.53–0.87)
Fernández-Suárez et al., 2005 (36) 21.23 (5.30–85.01) 0.45 (0.33–0.61)
Kwok et al., 2005 (33) 2.54 (1.62–3.98) 0.53 (0.41–0.68)
Greiner et al., 2005 (35) 4.78 (3.59–6.37) 0.17 (0.10–0.29)
Sauerland et al., 2005 (29) 4.83 (4.00–5.84) 0.05 (0.03–0.11)
Kamali et al., 2005 (34) 0.77 (0.52–1.16) 1.29 (0.87–1.90)
Anuradha and Chopra, 2005 (39) 66.23 (16.71–262.57) 0.15 (0.12–0.18)
Thammanichanond et al., 2005 (27) 4.92 (3.42–7.08) 0.12 (0.05–0.25)
Choi et al., 2005 (37) 3.74 (2.94–4.77) 0.25 (0.20–0.31)
Nell et al., 2005 (32) 4.89 (2.73–8.77) 0.51 (0.41–0.64)
Raza et al., 2005 (30) 21.48 (5.30–87.00) 0.49 (0.35–0.67)
Das et al., 2004 (55) 2.82 (2.11–3.77) 0.34 (0.21–0.54)
De Rycke et al., 2004 (54) 4.11 (2.91–5.81) 0.26 (0.18–0.37)
Girelli et al., 2004 (50) 1.32 (1.06–1.66) 0.28 (0.09–0.89)
Grootenboer-Mignot et al., 2004 (48) 3.48 (2.25–5.38) 0.39 (0.28–0.53)
Hitchon et al., 2004 (47) 1.80 (1.10–2.94) 0.39 (0.20–0.77)
Lopez-Hoyos et al., 2004 (97) 21.37 (7.01–65.09) 0.13 (0.06–0.29)
Bombardieri et al., 2004 (100) 5.85 (2.78–12.33) 0.12 (0.04–0.35)
Dubucquoi et al., 2004 (52) 1.92 (1.44–2.56) 0.58 (0.46–0.74)
Söderlin et al., 2004 (44) 4.14 (1.26–13.61) 0.74 (0.53–1.04)
Spiritus et al., 2004 (43) 7.18 (3.77–13.66) 0.40 (0.30–0.53)
Vallbracht et al., 2004 (42) 3.72 (2.98–4.64) 0.41 (0.35–0.48)
Vittecoq et al., 2004 (40) 4.42 (2.42–8.06) 0.70 (0.62–0.79)
Bas et al., 2003 (66) 4.06 (3.05–5.39) 0.33 (0.26–0.42)
Lee and Schur, 2003 (64) 3.64 (2.46–5.40) 0.35 (0.26–0.49)
Rantapää-Dahlqvist et al., 2003 (62) 10.57 (6.87–16.26) 0.39 (0.29–0.52)
Saraux et al., 2003 (61) 7.99 (3.88–16.44) 0.62 (0.52–0.75)
Suzuki et al., 2003 (60) 3.82 (2.85–5.12) 0.37 (0.32–0.49)
Jansen et al., 2003 (65) 7.62 (3.86–15.05) 0.53 (0.47–0.61)
Bizzaro et al., 2001 (74) 4.01 (2.86–5.62) 0.45 (0.34–0.58)
Vasiliauskiene et al., 2001 (73) 4.72 (3.15–7.08) 0.26 (0.18–0.39)
Vittecoq et al., 2001 (72) 13.45 (1.92–94.34) 0.57 (0.45–0.73)
Goldbach-Mansky et al., 2000 (76) 2.24 (1.06–3.01) 0.48 (0.36–0.64)
Schellekens et al., 2000 (12) 5.98 (4.08–8.78) 0.51 (0.43–0.61)
Aho et al., 1999 (80) 7.43 (4.57–12.06) 0.33 (0.24–0.45)
Jónsson et al., 1998 (81) 10.46 (6.18–17.71) 0.31 (0.21-0.44)
Swedler et al., 1997 (82) 12.71 (4.27–37.90) 0.11 (0.05–0.19)
Young et al., 1991 (93) 13.46 (1.96–92.47) 0.38 (0.25–0.58)
de Bois et al., 1996 (84) 4.44 (2.40–8.24) 0.07 (0.00–1.05)
Cordonnier et al., 1996 (85) 4.08 (1.05–15.86) 0.66 (0.50–0.87)
Visser et al., 1996 (83) 4.08 (3.55–4.69) 0.40 (0.33–0.48)
Berthelot et al., 1995 (89) 1.28 (1.02–1.60) 0.69 (0.50–0.97)
Saraux et al., 1995 (88) 2.54 (1.02–6.29) 0.86 (0.73–1.02)
Després et al., 1994 (91) 3.01 (2.25–4.02) 0.40 (0.32–0.50)
Gomès-Daudrix et al., 1994 (90) 55.09 (7.76–390.93) 0.46 (0.36–0.58)
Banchuin et al., 1992 (92) 24.86 (10.86–56.87) 0.54 (0.44–0.67)
Carpenter and Bartkowiak, 1989 (98) 11.91 (6.02–23.56) 0.27 (0.18–0.39)
Davis and Stein, 1989 (95) 3.43 (1.11–10.62) 0.71 (0.55–0.91)
Winkles et al., 1989 (94) 20.94 (13.37–32.80) 0.21 (0.15–0.29)
van Leeuwen et al., 1988 (96) 12.80 (7.01–23.36) 0.11 (0.11–0.32)
Total 4.86 (3.95–5.97) 0.38 (0.33–0.44)

0.01 1.00 100.00 0.01 1.00 100.00

Positive LR Negative LR

[Link] 5 June 2007 Annals of Internal Medicine Volume 146 • Number 11 801

Downloaded From: [Link] by a Tulane University User on 09/03/2013

 Review Anti-CCP Antibody and Rheumatoid Factor for Diagnosis of Rheumatoid Arthritis

Figure 3. Likelihood ratio (LR) for IgA rheumatoid factor (RF) and IgG RF.

Study, Year (Reference) Positive LR (95% CI) Negative LR (95% CI)

IgA RF
Greiner et al., 2005 (35) 10.52 (6.18–17.92) 0.39 (0.30–0.52)
García-Berrocal et al., 2005 (99) 2.31 (1.17–4.57) 0.72 (0.58–0.90)
Berglin et al., 2004 (105) 2.71 (1.09–6.76) 0.80 (0.78–1.00)
Vallbracht et al., 2004 (42) 4.36 (3.27–5.80) 0.56 (0.49–0.63)
Rantapää-Dahlqvist et al., 2003 (62) 12.76 (8.18–19.90) 0.32 (0.22–0.46)
Saraux et al., 2003 (61) 2.93 (1.90–4.54) 0.66 (0.55–0.80)
Jansen et al., 2003 (65) 9.11 (4.40–18.88) 0.50 (0.44–0.58)
Bas et al., 2003 (66) 5.42 (2.90–10.12) 0.40 (0.31–0.50)
Vasiliauskiene et al., 2001 (73) 4.89 (3.23–7.43) 0.27 (0.19–0.40)
Jónsson et al., 1998 (81) 8.79 (5.12–15.09) 0.43 (0.32–0.57)
Swedler et al., 1997 (82) 4.03 (2.91–5.58) 0.25 (0.17–0.38)
Visser et al., 1996 (83) 2.62 (2.20–3.12) 0.66 (0.59–0.75)
van Leeuwen et al., 1988 (96) 5.31 (3.70–7.62) 0.14 (0.09–0.21)
Total 5.01 (3.68–6.83) 0.44 (0.35–0.55)
IgG RF
Berglin et al., 2004 (105) 2.81 (1.13–6.99) 0.88 (0.77–1.00)
Das et al., 2004 (55) 2.67 (1.04–6.87) 0.92 (0.84–1.02)
Vallbracht et al., 2004 (42) 4.83 (3.48–6.72) 0.62 (0.56–0.69)
De Rycke et al., 2004 (54) 4.11 (2.91–5.81) 0.26 (0.18–0.37)
Rantapää-Dahlqvist et al., 2003 (62) 8.42 (5.16–13.73) 0.57 (0.45–0.71)
Jansen et al., 2003 (65) 13.44 (4.34–41.65) 0.68 (0.62–0.75)
Saraux et al., 1995 (88) 1.36 (1.04–1.78) 0.77 (0.61–0.98)
Jónsson et al., 1998 (81) 3.78 (2.39–5.98) 0.63 (0.51–0.78)
Swedler et al., 1997 (82) 11.16 (5.55–22.44) 0.47 (0.38–0.59)
Visser et al., 1996 (83) 1.48 (1.35–1.62) 0.54 (0.43–0.67)
Aupperle et al., 1996 (106) 4.80 (3.40–6.78) 0.39 (0.32–0.47)
Banchuin et al., 1992 (92) 7.62 (5.02–11.59) 0.44 (0.33–0.57)
van Leeuwen et al., 1988 (96) 8.15 (5.11–12.99) 0.15 (0.10–0.21)
Total 4.52 (2.64–7.73) 0.52 (0.42–0.66)

0.01 1.00 100.00 0.01 1.00 100.00

Positive LR Negative LR

available at [Link]). Six studies assessed associations specific than IgM RF, IgG RF, or IgA RF positivity for
with anti-CCP antibody positivity; 3 of these studies used an rheumatoid arthritis and was more specific than IgM RF
anti-CCP1 assay. All 6 studies reported that anti-CCP anti- for early rheumatoid arthritis. Because pooled sensitivities
body positivity was a statistically significant risk factor for were similar for anti-CCP antibody and RF, the better
radiographic progression. Of the 4 studies that examined anti- diagnostic accuracy of anti-CCP antibody was mainly due
CCP antibody and RF, 3 reported that the risk for radio- to its higher specificity. Anti-CCP2 was a more sensitive
graphic progression was greater for patients with anti-CCP marker than anti-CCP1. Assaying anti-CCP antibody alone
antibody positivity than for those with IgM RF positivity. and assaying combinations of anti-CCP antibody and IgM
RF provided similar results. Anti-CCP antibody positivity,
DISCUSSION especially anti-CCP2, was superior to IgM RF positivity
We identified several issues that had not been ad- for predicting development of rheumatoid arthritis and ra-
dressed systematically or quantitatively in past narrative re- diographic progression.
views (107, 108). Anti-CCP antibody positivity was more Some experts believe that immunity against citrulline
802 5 June 2007 Annals of Internal Medicine Volume 146 • Number 11 [Link]

Downloaded From: [Link] by a Tulane University User on 09/03/2013

 Anti-CCP Antibody and Rheumatoid Factor for Diagnosis of Rheumatoid Arthritis Review

plays a crucial role in the pathogenesis of rheumatoid ar- body and RF positivity combined and is a less accurate
thritis (109). Anti-CCP antibodies and anticitrullinated indicator than positivity for either antibody alone.
filaggrin antibodies are locally produced in inflamed joints, In clinical practice, most rheumatologists recommend
and citrullinated fibrin is found in the synovia of patients measuring anti-CCP antibody and RF because anti-CCP
with rheumatoid arthritis (110). antibody has moderate sensitivity, and clinicians try to
Anti-CCP antibody is present before symptoms de- maximize sensitivities by combining the 2 markers, espe-
velop, which suggests that citrullination and production of cially for early rheumatoid arthritis (32, 47, 48, 52, 59, 61,
anti-CCP antibody are early processes in rheumatoid ar- 63, 64, 66). Also, rheumatologists measure RF because it is
thritis (62). As we show, anti-CCP antibody is highly spe- included in the 1987 ACR criteria, and both anti-CCP
cific for rheumatoid arthritis. However, the biological antibody and RF are recommended screening tests for
function of RF is unclear: It is found in some apparently rheumatoid arthritis (114). In any case, comparison of
healthy elderly persons and in persons who have conditions anti-CCP antibody only with testing for anti-CCP anti-
other than rheumatoid arthritis (111). Substantial differ- body and RF involves a tradeoff between overall sensitivity
ences exist among RF test kits, and the reliability of some and specificity. If we want to maximize sensitivity, then
RF assays is questionable (112). The varying techniques for both tests are better, although this may prompt us to treat
measuring RF might partly explain the heterogeneous patients who are anti-CCP antibody negative but RF pos-
study results for RF. itive. Because it is harmful and costly to treat persons with
Some studies have reported conflicting results on false-positive results who do not have rheumatoid arthritis,
whether IgG RF and IgA RF are better diagnostic markers we need to consider the risks and the benefits of such an
than IgM RF (82, 87, 92). We found no major diagnostic approach. Clinical trials and cost-effectiveness studies of
differences among IgG RF, IgA RF, and IgM RF, whereas these tradeoffs are needed.
anti-CCP antibody was a better diagnostic marker than all When should we measure both anti-CCP antibody
3 RF subclasses. Our findings are compatible with those of and RF? If the prior probability of rheumatoid arthritis is
earlier studies of the sensitivity of different generations of relatively low, such as in patients who have knee pain only
in primary care or those who meet no other ACR criteria,
anti-CCP antibody assays. Filaggrin-derived cyclic peptide
measuring anti-CCP antibody alone seems to be a reason-
anti-CCP1 assays had very high specificity (98%) and
able strategy that avoids too many false-positive results. If,
moderate sensitivity that was lower than that of RF (12,
however, the prior probability of rheumatoid arthritis is
113). To overcome this problem, various cyclic epitopes
relatively high, such as in patients seen in rheumatology
that mimic true conformational epitopes were selected
clinics or those who meet other ACR criteria, measuring
from libraries of citrullinated peptides to develop more
anti-CCP antibody or IgM RF seems to be a reasonable
sensitive anti-CCP2 assays (41, 62). strategy that avoids missing potentially treatable patients.
Some studies suggested that the diagnostic accuracy of We found that the presence of anti-CCP antibody is
both anti-CCP antibody and IgM–RF positivity was not associated with development of rheumatoid arthritis and
markedly better than that of anti-CCP antibody positivity greater radiographic progression, and we confirmed that
alone. The combination of anti-CCP antibody and IgM RF is a major predictor of bone damage (58, 88).
RF positivity improved specificity over RF positivity alone. Our review has several limitations. We may have
Persons without rheumatoid arthritis who had false-posi- missed some pertinent studies, because we included only
tive results for RF did not have positive results for anti- diagnostic studies that provided information on sensitivity
CCP antibody and were regarded as healthy. The sensitiv-
ity, however, was reduced because positivity for anti-CCP
antibody and RF is a more stringent criterion than is pos- Table. Summary Likelihood Ratios of IgM Rheumatoid
itivity for anti-CCP antibody or IgM RF alone. As a result, Factor*
combining anti-CCP antibody and RF testing offered little
improvement. Variable Positive LR Negative LR
However, anti-CCP antibody positivity or IgM RF (95% CI) (95% CI)
positivity is more permissive in terms of sensitivity because All studies 4.86 (3.96–5.97) 0.38 (0.33–0.44)
the antibodies complement each other in patients with
Cutoff value
false-negative results. In this case, specificity is reduced ⱖ20 U/mL 4.42 (3.02–6.47) 0.39 (0.31–0.50)
substantially because all persons with false-positive results ⱖ40 U/mL 5.49 (2.25–13.38) 0.50 (0.37–0.69)
for RF are counted as having positive results for rheuma- ⱖ80 U/mL 4.57 (1.36–15.09) 0.44 (0.29–0.68)

toid arthritis. Because the improvement and deterioration Measurement method

of sensitivity were balanced, the overall diagnostic accuracy Nephelometry 4.15 (2.95–5.84) 0.32 (0.25–0.41)
Latex agglutination 5.05 (3.01–8.50) 0.39 (0.27–0.56)
of RF is less than that of anti-CCP antibody alone. To- ELISA 6.13 (4.60–8.17) 0.42 (0.34–0.51)
gether, these results show that anti-CCP antibody positiv-
ity is as effective a diagnostic indicator as anti-CCP anti- * ELISA ⫽ enzyme-linked immunosorbent assay; LR ⫽ likelihood ratio.

[Link] 5 June 2007 Annals of Internal Medicine Volume 146 • Number 11 803

Downloaded From: [Link] by a Tulane University User on 09/03/2013

 Review Anti-CCP Antibody and Rheumatoid Factor for Diagnosis of Rheumatoid Arthritis

Figure 4. Pooled likelihood ratio (LR) for first-generation assays for autoantibodies against cyclic citrullinated peptide (CCP1),
second-generation assays (CCP2), anti-CCP antibody and rheumatoid factor (RF), and anti-CCP antibody or RF.

804 5 June 2007 Annals of Internal Medicine Volume 146 • Number 11 [Link]

Downloaded From: [Link] by a Tulane University User on 09/03/2013

 Anti-CCP Antibody and Rheumatoid Factor for Diagnosis of Rheumatoid Arthritis Review

and specificity. Our funnel plots suggested some publica- T, et al. Rheumatoid arthritis specific anti-Sa antibodies target citrullinated vi-
mentin. Arthritis Res Ther. 2004;6:R142-50. [PMID: 15059278]
tion bias for favorable anti-CCP antibody studies (data not 9. Vossenaar ER, Radstake TR, van der Heijden A, van Mansum MA, Dieteren
shown). Because RF is incorporated into the current diag- C, de Rooij DJ, et al. Expression and activity of citrullinating peptidylarginine
nostic criteria of rheumatoid arthritis, diagnostic studies of deiminase enzymes in monocytes and macrophages. Ann Rheum Dis. 2004;63:
IgM RF might have some incorporation bias that could 373-81. [PMID: 15020330]
10. Vossenaar ER, Zendman AJ, van Venrooij WJ, Pruijn GJ. PAD, a growing
have increased the apparent sensitivity of this marker family of citrullinating enzymes: genes, features and involvement in disease.
(115). Bioessays. 2003;25:1106-18. [PMID: 14579251]
In conclusion, anti-CCP antibody positivity is more 11. Schellekens GA, de Jong BA, van den Hoogen FH, van de Putte LB, van
specific than IgM RF positivity for diagnosing rheumatoid Venrooij WJ. Citrulline is an essential constituent of antigenic determinants
recognized by rheumatoid arthritis-specific autoantibodies. J Clin Invest. 1998;
arthritis and early rheumatoid arthritis. Anti-CCP anti- 101:273-81. [PMID: 9421490]
body positivity should be included among the diagnostic 12. Schellekens GA, Visser H, de Jong BA, van den Hoogen FH, Hazes JM,
criteria of these 2 conditions. Breedveld FC, et al. The diagnostic properties of rheumatoid arthritis antibodies
recognizing a cyclic citrullinated peptide. Arthritis Rheum. 2000;43:155-63.
From Harvard School of Public Health, Boston, Massachusetts, and [PMID: 10643712]
Hyogo College of Medicine and Kobe University Graduate School of 13. Irwig L, Tosteson AN, Gatsonis C, Lau J, Colditz G, Chalmers TC, et al.
Medicine, Hyogo, Japan. Guidelines for meta-analyses evaluating diagnostic tests. Ann Intern Med. 1994;
120:667-76. [PMID: 8135452]
14. Standards for Reporting of Diagnostic Accuracy. Towards complete and ac-
Acknowledgments: The authors thank their colleagues from Kobe Uni- curate reporting of studies of diagnostic accuracy: The STARD Initiative. Ann
versity Graduate School of Medicine: Yumiko Nobuhara, MD, for lan- Intern Med. 2003;138:40-4. [PMID: 12513043]
guage assistance and Yasuko Nakayama for logistic assistance. They also 15. Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS,
thank Yasushi Kaji, Dokkyo Medical University, Department of Radiol- et al. The American Rheumatism Association 1987 revised criteria for the classi-
ogy, for methods assistance and Dr. William F. Goldman (MST Editing fication of rheumatoid arthritis. Arthritis Rheum. 1988;31:315-24. [PMID:
Company, Baltimore, Maryland) for editorial assistance and preparation 3358796]
of the manuscript. 16. Owens DK, Holodniy M, Garber AM, Scott J, Sonnad S, Moses L, et al.
Polymerase chain reaction for the diagnosis of HIV infection in adults. A meta-
analysis with recommendations for clinical practice and study design. Ann Intern
Grant Support: In part by a Grant-in-Aid for Young Scientists from the Med. 1996;124:803-15. [PMID: 8610949]
Ministry of Education, Culture, Sports, Science and Technology, Japan, 17. Hoffman RM, Kent DL, Deyo RA. Diagnostic accuracy and clinical utility
and a grant for Research on Allergic Disease and Immunology from the of thermography for lumbar radiculopathy. A meta-analysis. Spine.
Ministry of Health, Labor and Welfare, Japan. 1991;16:623-8. [PMID: 1830690]
18. Kent DL, Larson EB. Disease, level of impact, and quality of research meth-
Potential Financial Conflicts of Interest: Grants received: S. Kumagai ods. Three dimensions of clinical efficacy assessment applied to magnetic reso-
(Japanese Ministry of Health). nance imaging. Invest Radiol. 1992;27:245-54. [PMID: 1551777]
19. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials.
1986;7:177-88. [PMID: 3802833]
Requests for Single Reprints: Kunihiro Nishimura, MD, MPH, MS, 20. Midgette AS, Stukel TA, Littenberg B. A meta-analytic method for summa-
Department of Health Policy Management, Harvard School of Public rizing diagnostic test performances: receiver-operating-characteristic-summary
Health, 718 Huntington Avenue, Boston, MA 02215; e-mail, point estimates. Med Decis Making. 1993;13:253-7. [PMID: 8412556]
knishimu@[Link]. 21. Deeks J. Systematic reviews of evaluations of diagnostic and screening tests.
In: Egger M, Davey Smith G, Altman DG, eds. Systematic Reviews in Health
Current author addresses are available at [Link]. Care: Meta-Analysis in Context. 2d ed. London: BMJ; 2001.
22. Moses LE, Shapiro D, Littenberg B. Combining independent studies of a
diagnostic test into a summary ROC curve: data-analytic approaches and some
additional considerations. Stat Med. 1993;12:1293-316. [PMID: 8210827]
References 23. Sutton AJ, Jones DR, Sheldon TA, Song F. Methods for Meta-Analysis in
1. Lee DM, Weinblatt ME. Rheumatoid arthritis. Lancet. 2001;358:903-11. Medical Research. Chichester, United Kingdom: Wiley; 2000.
[PMID: 11567728] 24. Quinn MA, Gough AK, Green MJ, Devlin J, Hensor EM, Greenstein A,
2. Landewé RB. The benefits of early treatment in rheumatoid arthritis: con- et al. Anti-CCP antibodies measured at disease onset help identify seronegative
founding by indication, and the issue of timing [Editorial]. Arthritis Rheum. rheumatoid arthritis and predict radiological and functional outcome. Rheuma-
2003;48:1-5. [PMID: 12528097] tology (Oxford). 2006;45:478-80. [PMID: 16287917]
3. Lard LR, Visser H, Speyer I, vander Horst-Bruinsma IE, Zwinderman AH, 25. Meyer O, Nicaise-Roland P, Santos MD, Labarre C, Dougados M,
Breedveld FC, et al. Early versus delayed treatment in patients with recent-onset Goupille P, et al. Serial determination of cyclic citrullinated peptide autoantibod-
rheumatoid arthritis: comparison of two cohorts who received different treatment ies predicted five-year radiological outcomes in a prospective cohort of patients
strategies. Am J Med. 2001;111:446-51. [PMID: 11690569] with early rheumatoid arthritis. Arthritis Res Ther. 2006;8:R40. [PMID:
4. Bukhari MA, Wiles NJ, Lunt M, Harrison BJ, Scott DG, Symmons DP, 16469118]
et al. Influence of disease-modifying therapy on radiographic outcome in inflam- 26. van Gaalen FA, Visser H, Huizinga TW. A comparison of the diagnostic
matory polyarthritis at five years: results from a large observational inception accuracy and prognostic value of the first and second anti-cyclic citrullinated
study. Arthritis Rheum. 2003;48:46-53. [PMID: 12528102] peptides (CCP1 and CCP2) autoantibody tests for rheumatoid arthritis. Ann
5. Dörner T, Egerer K, Feist E, Burmester GR. Rheumatoid factor revisited. Rheum Dis. 2005;64:1510-2. [PMID: 15800005]
Curr Opin Rheumatol. 2004;16:246-53. [PMID: 15103252] 27. Thammanichanond D, Kunakorn M, Kitiwanwanich S, Attamasirikul K,
6. Nienhuis RL, Mandema E. A new serum factor in patients with rheumatoid Nantiruj K. Raising rheumatoid factor cutoff helps distinguish rheumatoid ar-
arthritis: the antiperinuclear factor. Ann Rheum Dis. 1964;23:302-5. [PMID: thritis. Asian Pac J Allergy Immunol. 2005;23:165-8. [PMID: 16252848]
14178016] 28. Tanaka N, Sakahashi H, Ishii S, Sato E, Hirose K, Ishima T. Synovial
7. Young BJ, Mallya RK, Leslie RD, Clark CJ, Hamblin TJ. Anti-keratin membrane enhancement and bone erosion by magnetic resonance imaging for
antibodies in rheumatoid arthritis. Br Med J. 1979;2:97-9. [PMID: 111762] prediction of radiologic progression in patients with early rheumatoid arthritis.
8. Vossenaar ER, Després N, Lapointe E, van der Heijden A, Lora M, Senshu Rheumatol Int. 2005;25:103-7. [PMID: 14648111]

[Link] 5 June 2007 Annals of Internal Medicine Volume 146 • Number 11 805

Downloaded From: [Link] by a Tulane University User on 09/03/2013

 Review Anti-CCP Antibody and Rheumatoid Factor for Diagnosis of Rheumatoid Arthritis

29. Sauerland U, Becker H, Seidel M, Schotte H, Willeke P, Schorat A, et al. 48. Grootenboer-Mignot S, Nicaise-Roland P, Delaunay C, Meyer O, Chollet-
Clinical utility of the anti-CCP assay: experiences with 700 patients. Ann N Y Martin S, Labarre C. Second generation anti-cyclic citrullinated peptide (anti-
Acad Sci. 2005;1050:314-8. [PMID: 16014547] CCP2) antibodies can replace other anti-filaggrin antibodies and improve rheu-
30. Raza K, Breese M, Nightingale P, Kumar K, Potter T, Carruthers DM, matoid arthritis diagnosis. Scand J Rheumatol. 2004;33:218-20. [PMID:
et al. Predictive value of antibodies to cyclic citrullinated peptide in patients with 15370715]
very early inflammatory arthritis. J Rheumatol. 2005;32:231-8. [PMID: 49. Goronzy JJ, Matteson EL, Fulbright JW, Warrington KJ, Chang-Miller A,
15693082] Hunder GG, et al. Prognostic markers of radiographic progression in early rheu-
31. Nielen MM, van der Horst AR, van Schaardenburg D, van der Horst- matoid arthritis. Arthritis Rheum. 2004;50:43-54. [PMID: 14730598]
Bruinsma IE, van de Stadt RJ, Aarden L, et al. Antibodies to citrullinated 50. Girelli F, Foschi FG, Bedeschi E, Calderoni V, Stefanini GF, Martinelli
human fibrinogen (ACF) have diagnostic and prognostic value in early arthritis. MG. Is anti cyclic citrullinated peptide a useful laboratory test for the diagnosis of
Ann Rheum Dis. 2005;64:1199-204. [PMID: 15640269] rheumatoid arthritis? Allerg Immunol (Paris). 2004;36:127-30. [PMID:
32. Nell VP, Machold KP, Stamm TA, Eberl G, Heinzl H, Uffmann M, et al. 15180353]
Autoantibody profiling as early diagnostic and prognostic tool for rheumatoid 51. Forslind K, Larsson EM, Eberhardt K, Johansson A, Svensson B. Magnetic
arthritis. Ann Rheum Dis. 2005;64:1731-6. [PMID: 15878904] resonance imaging of the knee: a tool for prediction of joint damage in early
33. Kwok JS, Hui KH, Lee TL, Wong W, Lau YL, Wong RW, et al. Anti- rheumatoid arthritis? Scand J Rheumatol. 2004;33:154-61. [PMID: 15228185]
cyclic citrullinated peptide: diagnostic and prognostic values in juvenile idiopathic 52. Dubucquoi S, Solau-Gervais E, Lefranc D, Marguerie L, Sibilia J, Goetz J,
arthritis and rheumatoid arthritis in a Chinese population. Scand J Rheumatol. et al. Evaluation of anti-citrullinated filaggrin antibodies as hallmarks for the
2005;34:359-66. [PMID: 16234183] diagnosis of rheumatic diseases. Ann Rheum Dis. 2004;63:415-9. [PMID:
34. Kamali S, Polat NG, Kasapoglu E, Gul A, Ocal L, Aral O, et al. Anti-CCP 15020336]
and antikeratin antibodies in rheumatoid arthritis, primary Sjögren’s syndrome, 53. Dixey J, Solymossy C, Young A; Early RA Study. Is it possible to predict
and Wegener’s granulomatosis [Letter]. Clin Rheumatol. 2005;24:673-6. radiological damage in early rheumatoid arthritis (RA)? A report on the occur-
[PMID: 15926038] rence, progression, and prognostic factors of radiological erosions over the first 3
35. Greiner A, Plischke H, Kellner H, Gruber R. Association of anti-cyclic years in 866 patients from the Early RA Study (ERAS). J Rheumatol Suppl.
citrullinated peptide antibodies, anti-citrullin antibodies, and IgM and IgA rheu- 2004;69:48-54. [PMID: 15053454]
matoid factors with serological parameters of disease activity in rheumatoid ar- 54. De Rycke L, Peene I, Hoffman IE, Kruithof E, Union A, Meheus L, et al.
thritis. Ann N Y Acad Sci. 2005;1050:295-303. [PMID: 16014545] Rheumatoid factor and anticitrullinated protein antibodies in rheumatoid arthri-
36. Fernández-Suárez A, Reneses S, Wichmann I, Criado R, Núñez A. Efficacy tis: diagnostic value, associations with radiological progression rate, and extra-
of three ELISA measurements of anti-cyclic citrullinated peptide antibodies in the articular manifestations. Ann Rheum Dis. 2004;63:1587-93. [PMID: 15547083]
early diagnosis of rheumatoid arthritis. Clin Chem Lab Med. 2005;43:1234-9. 55. Das H, Atsumi T, Fukushima Y, Shibuya H, Ito K, Yamada Y, et al.
[PMID: 16232091] Diagnostic value of antiagalactosyl IgG antibodies in rheumatoid arthritis. Clin
37. Choi SW, Lim MK, Shin DH, Park JJ, Shim SC. Diagnostic performances Rheumatol. 2004;23:218-22. [PMID: 15168148]
of anti-cyclic citrullinated peptides antibody and antifilaggrin antibody in Korean 56. Correa PA, Tobón GJ, Citera G, Cadena J, Schneeberger E, Camargo JF,
patients with rheumatoid arthritis. J Korean Med Sci. 2005;20:473-8. [PMID: et al. [Anti-cyclic citrullinated peptide antibodies in rheumatoid arthritis: relation
15953872] with clinical features, cytokines and HLA-DRB1]. Biomedica. 2004;24:140-52.
38. Aotsuka S, Okawa-Takatsuji M, Nagatani K, Nagashio C, Kano T, Naka- [PMID: 15495594]
jima K, et al. A retrospective study of the fluctuation in serum levels of anti-cyclic 57. Cook AD, Gray R, Ramshaw J, Mackay IR, Rowley MJ. Antibodies against
citrullinated peptide antibody in patients with rheumatoid arthritis. Clin Exp the CB10 fragment of type II collagen in rheumatoid arthritis. Arthritis Res Ther.
Rheumatol. 2005;23:475-81. [PMID: 16095115] 2004;6:R477-83. [PMID: 15380047]
39. Anuradha V, Chopra A. In the era of nephelometry, latex agglutination is still 58. Zeng X, Ai M, Tian X, Gan X, Shi Y, Song Q, et al. Diagnostic value of
good enough to detect rheumatoid factor. J Rheumatol. 2005;32:2343-4. anti-cyclic citrullinated peptide antibody in patients with rheumatoid arthritis. J
[PMID: 16331760] Rheumatol. 2003;30:1451-5. [PMID: 12858440]
40. Vittecoq O, Incaurgarat B, Jouen-Beades F, Legoedec J, Letourneur O, 59. Vencovský J, Machácek S, Sedová L, Kafková J, Gatterová J, Pesáková V,
Rolland D, et al. Autoantibodies recognizing citrullinated rat filaggrin in an et al. Autoantibodies can be prognostic markers of an erosive disease in early
ELISA using citrullinated and non-citrullinated recombinant proteins as antigens rheumatoid arthritis. Ann Rheum Dis. 2003;62:427-30. [PMID: 12695154]
are highly diagnostic for rheumatoid arthritis. Clin Exp Immunol. 2004;135:173- 60. Suzuki K, Sawada T, Murakami A, Matsui T, Tohma S, Nakazono K, et al.
80. [PMID: 14678280] High diagnostic performance of ELISA detection of antibodies to citrullinated
41. van Venrooij WJ, Vossenaar ER, Zendman AJ. Anti-CCP antibodies: the antigens in rheumatoid arthritis. Scand J Rheumatol. 2003;32:197-204. [PMID:
new rheumatoid factor in the serology of rheumatoid arthritis. Autoimmun Rev. 14626625]
2004;3 Suppl 1:S17-9. [PMID: 15309772] 61. Saraux A, Berthelot JM, Devauchelle V, Bendaoud B, Chalès G, Le Henaff
42. Vallbracht I, Rieber J, Oppermann M, Förger F, Siebert U, Helmke K. C, et al. Value of antibodies to citrulline-containing peptides for diagnosing early
Diagnostic and clinical value of anti-cyclic citrullinated peptide antibodies com- rheumatoid arthritis. J Rheumatol. 2003;30:2535-9. [PMID: 14719190]
pared with rheumatoid factor isotypes in rheumatoid arthritis. Ann Rheum Dis. 62. Rantapää-Dahlqvist S, de Jong BA, Berglin E, Hallmans G, Wadell G,
2004;63:1079-84. [PMID: 15308516] Stenlund H, et al. Antibodies against cyclic citrullinated peptide and IgA rheu-
43. Spiritus T, Verschueren P, Westhovens R, Bossuyt X. Diagnostic charac- matoid factor predict the development of rheumatoid arthritis. Arthritis Rheum.
teristics of a gelatin based Waaler-Rose assay (Serodia-RA) for the detection of 2003;48:2741-9. [PMID: 14558078]
rheumatoid factor. Ann Rheum Dis. 2004;63:1169-71. [PMID: 15308531] 63. Meyer O, Labarre C, Dougados M, Goupille P, Cantagrel A, Dubois A, et
44. Söderlin MK, Kastbom A, Kautiainen H, Leirisalo-Repo M, Strandberg G, al. Anticitrullinated protein/peptide antibody assays in early rheumatoid arthritis
Skogh T. Antibodies against cyclic citrullinated peptide (CCP) and levels of for predicting five year radiographic damage. Ann Rheum Dis. 2003;62:120-6.
cartilage oligomeric matrix protein (COMP) in very early arthritis: relation to [PMID: 12525380]
diagnosis and disease activity. Scand J Rheumatol. 2004;33:185-8. [PMID: 64. Lee DM, Schur PH. Clinical utility of the anti-CCP assay in patients with
15228190] rheumatic diseases. Ann Rheum Dis. 2003;62:870-4. [PMID: 12922961]
45. Kumagai S, Nishimura K, Hayashi N. [Topics on immunological tests for 65. Jansen LM, van Schaardenburg D, van der Horst-Bruinsma I, van der Stadt
rheumatoid arthritis]. Rinsho Byori. 2004;52:836-43. [PMID: 15624500] RJ, de Koning MH, Dijkmans BA. The predictive value of anti-cyclic citrulli-
46. Jansen TL, Bruyn GA. Diagnostic value of anticyclic citrullinated peptide nated peptide antibodies in early arthritis. J Rheumatol. 2003;30:1691-5.
antibody in rheumatoid arthritis [Letter]. J Rheumatol. 2004;31:1012-3. [PMID: [PMID: 12913923]
15124273] 66. Bas S, Genevay S, Meyer O, Gabay C. Anti-cyclic citrullinated peptide
47. Hitchon CA, Alex P, Erdile LB, Frank MB, Dozmorov I, Tang Y, et al. antibodies, IgM and IgA rheumatoid factors in the diagnosis and prognosis of
A distinct multicytokine profile is associated with anti-cyclical citrullinated pep- rheumatoid arthritis. Rheumatology (Oxford). 2003;42:677-80. [PMID:
tide antibodies in patients with early untreated inflammatory arthritis. J Rheuma- 12709545]
tol. 2004;31:2336-46. [PMID: 15570632] 67. Vincent C, Nogueira L, Sebbag M, Chapuy-Regaud S, Arnaud M, Letour-

806 5 June 2007 Annals of Internal Medicine Volume 146 • Number 11 [Link]

Downloaded From: [Link] by a Tulane University User on 09/03/2013

 Anti-CCP Antibody and Rheumatoid Factor for Diagnosis of Rheumatoid Arthritis Review
neur O, et al. Detection of antibodies to deiminated recombinant rat filaggrin by Diagnostic value of anti-RA33 antibody, antikeratin antibody, antiperinuclear
enzyme-linked immunosorbent assay: a highly effective test for the diagnosis of factor and antinuclear antibody in early rheumatoid arthritis: comparison with
rheumatoid arthritis. Arthritis Rheum. 2002;46:2051-8. [PMID: 12209508] rheumatoid factor. Br J Rheumatol. 1996;35:620-4. [PMID: 8670593]
68. Union A, Meheus L, Humbel RL, Conrad K, Steiner G, Moereels H, et al. 86. Brennan P, Harrison B, Barrett E, Chakravarty K, Scott D, Silman A, et al.
Identification of citrullinated rheumatoid arthritis-specific epitopes in natural A simple algorithm to predict the development of radiological erosions in patients
filaggrin relevant for antifilaggrin autoantibody detection by line immunoassay. with early rheumatoid arthritis: prospective cohort study. BMJ. 1996;313:471-6.
Arthritis Rheum. 2002;46:1185-95. [PMID: 12115222] [PMID: 8776318]
69. Saraux A, Berthelot JM, Chalès G, Le Henaff C, Mary JY, Thorel JB, et al. 87. van Leeuwen MA, Westra J, van Riel PL, Limburg PC, van Rijswijk MH.
Value of laboratory tests in early prediction of rheumatoid arthritis. Arthritis IgM, IgA, and IgG rheumatoid factors in early rheumatoid arthritis predict-
Rheum. 2002;47:155-65. [PMID: 11954009] ive of radiological progression? Scand J Rheumatol. 1995;24:146-53. [PMID:
70. Jansen AL, van der Horst-Bruinsma I, van Schaardenburg D, van de Stadt 7777825]
RJ, de Koning MH, Dijkmans BA. Rheumatoid factor and antibodies to cyclic 88. Saraux A, Valls I, Voisin V, Koreichi A, Baron D, Youinou P, et al. How
citrullinated peptide differentiate rheumatoid arthritis from undifferentiated poly- useful are tests for rheumatoid factors, antiperinuclear factors, antikeratin anti-
arthritis in patients with early arthritis. J Rheumatol. 2002;29:2074-6. [PMID: body, and the HLA DR4 antigen for the diagnosis of rheumatoid arthritis? Rev
12375314] Rhum Engl Ed. 1995;62:16-20. [PMID: 7540486]
71. Bas S, Perneger TV, Seitz M, Tiercy JM, Roux-Lombard P, Guerne PA. 89. Berthelot JM, Maugars Y, Audrain M, Youinou P, Prost A. Specificity of
Diagnostic tests for rheumatoid arthritis: comparison of anti-cyclic citrullinated antiperinuclear factor for rheumatoid arthritis in rheumatoid factor-positive sera.
peptide antibodies, anti-keratin antibodies and IgM rheumatoid factors. Rheu- Br J Rheumatol. 1995;34:716-20. [PMID: 7551653]
matology (Oxford). 2002;41:809-14. [PMID: 12096232] 90. Gomès-Daudrix V, Sebbag M, Girbal E, Vincent C, Simon M, Rakotoar-
72. Vittecoq O, Salle V, Jouen-Beades F, Krzanowska K, Ménard JF, Gayet A, ivony J, et al. Immunoblotting detection of so-called ‘antikeratin antibodies’:
et al. Autoantibodies to the 27 C-terminal amino acids of calpastatin are detected a new assay for the diagnosis of rheumatoid arthritis. Ann Rheum Dis. 1994;53:
in a restricted set of connective tissue diseases and may be useful for diagnosis of 735-42. [PMID: 7529986]
rheumatoid arthritis in community cases of very early arthritis. Rheumatology 91. Després N, Boire G, Lopez-Longo FJ, Ménard HA. The Sa system: a novel
(Oxford). 2001;40:1126-34. [PMID: 11600742] antigen-antibody system specific for rheumatoid arthritis. J Rheumatol. 1994;21:
73. Vasiliauskiene L, Wiik A, Høier-Madsen M. Prevalence and clinical signif- 1027-33. [PMID: 7932409]
icance of antikeratin antibodies and other serological markers in Lithuanian pa- 92. Banchuin N, Janyapoon K, Sarntivijai S, Parivisutt L. Re-evaluation of
tients with rheumatoid arthritis. Ann Rheum Dis. 2001;60:459-66. [PMID: ELISA and latex agglutination test for rheumatoid factor detection in the diag-
11302867] nosis of rheumatoid arthritis. Asian Pac J Allergy Immunol. 1992;10:47-54.
74. Bizzaro N, Mazzanti G, Tonutti E, Villalta D, Tozzoli R. Diagnostic accu- [PMID: 1418183]
racy of the anti-citrulline antibody assay for rheumatoid arthritis. Clin Chem. 93. Young A, Sumar N, Bodman K, Goyal S, Sinclair H, Roitt I, et al. Agalac-
2001;47:1089-93. [PMID: 11375296] tosyl IgG: an aid to differential diagnosis in early synovitis. Arthritis Rheum.
75. Kroot EJ, de Jong BA, van Leeuwen MA, Swinkels H, van den Hoogen 1991;34:1425-9. [PMID: 1953820]
FH, van’t Hof M, et al. The prognostic value of anti-cyclic citrullinated peptide 94. Winkles JW, Lunec J, Gray L. Automated enhanced latex agglutination assay
antibody in patients with recent-onset rheumatoid arthritis. Arthritis Rheum. for rheumatoid factors in serum. Clin Chem. 1989;35:303-7. [PMID: 2914380]
2000;43:1831-5. [PMID: 10943873] 95. Davis P, Stein M. Evaluation of criteria for the classification of SLE in
76. Goldbach-Mansky R, Lee J, McCoy A, Hoxworth J, Yarboro C, Smolen JS, Zimbabwean patients [Letter]. Br J Rheumatol. 1989;28:546-7. [PMID:
et al. Rheumatoid arthritis associated autoantibodies in patients with synovitis of 2686803]
recent onset. Arthritis Res. 2000;2:236-43. [PMID: 11056669] 96. van Leeuwen MA, Westra J, Limburg PC, de Jong HJ, Marrink J, van
77. Bas S, Perneger TV, Mikhnevitch E, Seitz M, Tiercy JM, Roux-Lombard Rijswijk MH. Quantitation of IgM, IgA and IgG rheumatoid factors by ELISA
P, et al. Association of rheumatoid factors and anti-filaggrin antibodies with in rheumatoid arthritis and other rheumatic disorders. Scand J Rheumatol Suppl.
severity of erosions in rheumatoid arthritis. Rheumatology (Oxford). 2000;39: 1988;75:25-31. [PMID: 3070726]
1082-8. [PMID: 11035127] 97. Lopez-Hoyos M, Ruiz de Alegria C, Blanco R, Crespo J, Peña M, Rodri-
78. Aman S, Paimela L, Leirisalo-Repo M, Risteli J, Kautiainen H, Helve T, guez-Valverde V, et al. Clinical utility of anti-CCP antibodies in the differential
et al. Prediction of disease progression in early rheumatoid arthritis by ICTP, RF diagnosis of elderly-onset rheumatoid arthritis and polymyalgia rheumatica.
and CRP. A comparative 3-year follow-up study. Rheumatology (Oxford). 2000; Rheumatology (Oxford). 2004;43:655-7. [PMID: 14970400]
39:1009-13. [PMID: 10986307] 98. Carpenter AB, Bartkowiak CD. Rheumatoid factors determined by fluores-
79. Vincent C, de Keyser F, Masson-Bessière C, Sebbag M, Veys EM, Serre G. cence immunoassay: comparison with qualitative and quantitative methods. Clin
Anti-perinuclear factor compared with the so called “antikeratin” antibodies and Chem. 1989;35:464-6. [PMID: 2646034]
antibodies to human epidermis filaggrin, in the diagnosis of arthritides. Ann 99. Garcı́a-Berrocal B, González C, Pérez M, Navajo JA, Moreta I, Dávila C,
Rheum Dis. 1999;58:42-8. [PMID: 10343539] et al. Anti-cyclic citrullinated peptide autoantibodies in IgM rheumatoid factor-
80. Aho K, Palosuo T, Lukka M, Kurki P, Isomäki H, Kautiainen H, et al. positive patients. Clin Chim Acta. 2005;354:123-30. [PMID: 15748608]
Antifilaggrin antibodies in recent-onset arthritis. Scand J Rheumatol. 1999;28: 100. Bombardieri M, Alessandri C, Labbadia G, Iannuccelli C, Carlucci F,
113-6. [PMID: 10229141] Riccieri V, et al. Role of anti-cyclic citrullinated peptide antibodies in discrimi-
81. Jónsson T, Steinsson K, Jónsson H, Geirsson AJ, Thorsteinsson J, Valdi- nating patients with rheumatoid arthritis from patients with chronic hepatitis C
marsson H. Combined elevation of IgM and IgA rheumatoid factor has high infection-associated polyarticular involvement. Arthritis Res Ther. 2004;6:R137-
diagnostic specificity for rheumatoid arthritis. Rheumatol Int. 1998;18:119-22. 41. [PMID: 15059277]
[PMID: 9833253] 101. Nielen MM, van Schaardenburg D, Reesink HW, van de Stadt RJ, van
82. Swedler W, Wallman J, Froelich CJ, Teodorescu M. Routine measurement der Horst-Bruinsma IE, de Koning MH, et al. Specific autoantibodies precede
of IgM, IgG, and IgA rheumatoid factors: high sensitivity, specificity, and pre- the symptoms of rheumatoid arthritis: a study of serial measurements in blood
dictive value for rheumatoid arthritis. J Rheumatol. 1997;24:1037-44. [PMID: donors. Arthritis Rheum. 2004;50:380-6. [PMID: 14872479]
9195506] 102. Jansen LM, van der Horst-Bruinsma IE, van Schaardenburg D, Bezemer
83. Visser H, Gelinck LB, Kampfraath AH, Breedveld FC, Hazes JM. Diag- PD, Dijkmans BA. Predictors of radiographic joint damage in patients with early
nostic and prognostic characteristics of the enzyme linked immunosorbent rheu- rheumatoid arthritis. Ann Rheum Dis. 2001;60:924-7. [PMID: 11557647]
matoid factor assays in rheumatoid arthritis. Ann Rheum Dis. 1996;55:157-61. 103. van Jaarsveld CH, ter Borg EJ, Jacobs JW, Schellekens GA, Gmelig-
[PMID: 8712877] Meyling FH, van Booma-Frankfort C, et al. The prognostic value of the anti-
84. de Bois MH, Arndt JW, Speyer I, Pauwels EK, Breedveld FC. Technetium- perinuclear factor, anti-citrullinated peptide antibodies and rheumatoid factor in
99m labelled human immunoglobulin scintigraphy predicts rheumatoid arthritis early rheumatoid arthritis. Clin Exp Rheumatol. 1999;17:689-97. [PMID:
in patients with arthralgia. Scand J Rheumatol. 1996;25:155-8. [PMID: 10609067]
8668958] 104. van Zeben D, Hazes JM, Zwinderman AH, Vandenbroucke JP, Breedveld
85. Cordonnier C, Meyer O, Palazzo E, de Bandt M, Elias A, Nicaise P, et al. FC. Factors predicting outcome of rheumatoid arthritis: results of a followup

[Link] 5 June 2007 Annals of Internal Medicine Volume 146 • Number 11 807

Downloaded From: [Link] by a Tulane University User on 09/03/2013

 Review Anti-CCP Antibody and Rheumatoid Factor for Diagnosis of Rheumatoid Arthritis

study. J Rheumatol. 1993;20:1288-96. [PMID: 8230007] antifilaggrin autoantibodies are deiminated forms of the alpha- and beta-chains of
105. Berglin E, Padyukov L, Sundin U, Hallmans G, Stenlund H, Van Ven- fibrin. J Immunol. 2001;166:4177-84. [PMID: 11238669]
rooij WJ, et al. A combination of autoantibodies to cyclic citrullinated peptide 111. Shmerling RH, Delbanco TL. How useful is the rheumatoid factor? An
(CCP) and HLA-DRB1 locus antigens is strongly associated with future onset of analysis of sensitivity, specificity, and predictive value. Arch Intern Med. 1992;
rheumatoid arthritis. Arthritis Res Ther. 2004;6:R303-8. [PMID: 15225365] 152:2417-20. [PMID: 1456851]
106. Aupperle KR, Alsalameh S, Stock KP, Burmester GR, Kalden JR. 112. Bas S, Perneger TV, Kunzle E, Vischer TL. Comparative study of different
[Comparison of rheumatoid test procedures—value and critical interpretation of enzyme immunoassays for measurement of IgM and IgA rheumatoid factors.
sensitivity and specificity and their effect on pre-test and post-test probability]. Ann Rheum Dis. 2002;61:505-10. [PMID: 12006322]
Z Rheumatol. 1996;55:158-67. [PMID: 8768145] 113. van Boekel MA, Vossenaar ER, van den Hoogen FH, van Venrooij WJ.
107. Riedemann JP, Muñoz S, Kavanaugh A. The use of second generation
Autoantibody systems in rheumatoid arthritis: specificity, sensitivity and diagnos-
anti-CCP antibody (anti-CCP2) testing in rheumatoid arthritis—a systematic
tic value. Arthritis Res. 2002;4:87-93. [PMID: 11879544]
review. Clin Exp Rheumatol. 2005;23:S69-76. [PMID: 16273788]
114. IUIS/WHO/AF/CDC Committee for the Standardization of Autoanti-
108. van Venrooij WJ, van de Putte LB. [Early diagnosis of rheumatoid arthritis
with a test based upon a specific antigen: cyclic citrullinated peptide]. Ned Tijd- bodies in Rheumatic and Related Diseases. Cutting edge diagnostics in rheuma-
schr Geneeskd. 2003;147:191-4. [PMID: 12645351] tology: the role of patients, clinicians, and laboratory scientists in optimizing the
109. Cantaert T, De Rycke L, Bongartz T, Matteson EL, Tak PP, Nicholas use of autoimmune serology. Arthritis Rheum. 2004;51:291-8. [PMID:
AP, et al. Citrullinated proteins in rheumatoid arthritis: crucial . . . but not suf- 15077275]
ficient! Arthritis Rheum. 2006;54:3381-9. [PMID: 17075816] 115. Ransohoff DF, Feinstein AR. Problems of spectrum and bias in evaluating
110. Masson-Bessière C, Sebbag M, Girbal-Neuhauser E, Nogueira L, Vincent the efficacy of diagnostic tests. N Engl J Med. 1978;299:926-30. [PMID:
C, Senshu T, et al. The major synovial targets of the rheumatoid arthritis-specific 692598]

EXPEDITED REVIEW

Annals invites authors of clinically important randomized, controlled trials

to request expedited review and publication of their manuscripts. Send
requests to Harold Sox (hsox@[Link]), Christine Laine (chrisl
@[Link]), Michael Berkwits (mberkwits@[Link]), or
Cynthia Mulrow (cmulrow@[Link]). We take extra efforts to pro-
vide thorough, high-quality, and timely critiques of trials that we expe-
dite. We publish expedited trials that are accepted early online. We also
provide readers ancillary material about selected trials, including regis-
tered protocols, lists of other ongoing and published relevant trials, lists
of relevant published systematic reviews, and links to clinical sources that
provide physicians and patients information about the topic of the trial.

808 5 June 2007 Annals of Internal Medicine Volume 146 • Number 11 [Link]

Downloaded From: [Link] by a Tulane University User on 09/03/2013

 ONLINE
Annals of Internal Medicine ˜ONLY
Current Author Addresses: Drs. Nishimura and Kuntz: Department of
Health Policy Management, Harvard School of Public Health, 718 Hun- Appendix Figure. Study flow diagram. WEB-ONLY
tington Avenue, Boston, MA 02215.
Drs. Sugiyama, Kogata, Tsuji, Nakazawa, Kawano, Saigo, Morinobu,
and Kumagai: Department of Clinical Pathology and Immunology, Potentially eligible reports (n = 302)
Kobe University Graduate School of Medicine, 7-5-2 Kusunokicho,
Kobe, Hyogo, Japan 650-0017.
Reports excluded on basis of title or abstract (n = 126)
Dr. Koshiba: Hyogo College of Medicine, 1-1 Mukogawacho, Nishi- Published unknown: 40
nomiya, Hyogo, Japan 663-8501. Review or editorial articles: 41
Dr. Kamae: Department of Applied Biostatistics, Kobe University Grad- Not about anti-CCP or RF: 45
uate School of Medicine, 7-5-2 Kusunokicho, Kobe, Hyogo, Japan 650-
Ta1
Ta2 001.
Full-text reports retrieved for detailed evaluation (n = 176)

Reports excluded on basis of detailed evaluation (n = 92)

Articles about other conditions or other techniques: 19
Insufficient data to calculate sensitivity and
specificity: 67
Overlap of the cohort: 6

Reports included in the review (n = 86)

CCP ⫽ cyclic citrullinated peptide; RF ⫽ rheumatoid factor.

W-182 5 June 2007 Annals of Internal Medicine Volume 146 • Number 11 [Link]

Downloaded From: [Link] by a Tulane University User on 09/03/2013

 Appendix Table 1. Characteristics of Studies of Autoantibodies against Cyclic Citrullinated Peptide*

Study, Year (Reference) Location Language Setting Generation Assay Reference Design‡ Blind Interpretation Interval between Test Technical Quality Clinical Mean or Women, Mean Duration Control Participants Result Sensitivity, Specificity, Positive Negative
of CCP Manu- Standard‡ of Test Result‡ and Reference of Anti-CCP Description of Median % of Illness, y % % LR LR
facturer† Standard Antibody Sample Age, y
TP FP FN TN
Reported‡ Reported‡
Quinn et al., 2006 (24) England English Rheumatology CCP2 Axis-Shield ACR Prospective Not reported Not reported Yes Yes 58 64.2 0.58 Other rheumatic diseases 147 10 35 106 80.8 91.4 9.37 0.21
clinic (n ⫽ 116)
Fernández-Suárez Spain English Primary care CCP2 Inova ACR Prospective Not reported Not reported Yes Yes 52 45.5 NA Other rheumatic diseases 31 0 22 75 58.5 100 88.67 0.42
et al., 2005 (36) (n ⫽ 25), healthy persons
(n ⫽ 50)
Kwok et al., 2005 (33) Korea English Rheumatology CCP2 Inova ACR Retrospective Not reported NA Yes Yes 56 86.8 13.2 Other rheumatic diseases 71 2 58 66 55.0 97.1 18.71 0.46
clinic (n ⫽ 68), healthy persons
(n ⫽ 60)
Greiner et al., Germany English Teaching CCP2 Euro-Diag- ACR Not reported Not reported NA Yes Yes 54.8 NA NA Other rheumatic diseases 70 5 17 228 80.5 97.9 37.49 0.20
2005 (35) hospital nostica (n ⫽ 233)
Sauerland et al., Germany English Teaching CCP2 Euroimmun ACR Prospective Not reported NA Yes Yes NA NA NA Other rheumatic diseases 171 26 60 443 74.0 94.5 13.35 0.28
2005 (29) hospital (n ⫽ 469)
Kamali et al., 2005 (34) Turkey English Teaching CCP2 Euroimmun ACR Not reported Not reported Not reported Yes No NA NA NA Progressive systemic sclerosis 26 1 20 56 56.5 98.2 32.20 0.44
hospital (n ⫽ 32), Wegener granulo-
matosis (n ⫽ 22)
Aotsuka et al., Japan English Teaching CCP2 Axis-Shield ACR Retrospective Not reported 0–24 y Yes No NA NA NA Other rheumatic diseases 115 17 16 73 87.8 81.1 4.65 0.15
2005 (38) hospital (n ⫽ 90), healthy persons
(n ⫽ 200)
Choi et al., 2005 (37) Korea English Primary care CCP2 Tosho ACR Not reported Not reported Not reported Yes Yes NA NA NA Other rheumatic diseases 236 20 88 231 72.8 92.0 9.14 0.25
(n ⫽ 251)
García-Berrocal et al., Spain English Teaching CCP2 Euro-Diag- ACR Retrospective Not reported Not reported Yes Yes NA NA NA Other diseases (n ⫽ 49) 69 8 18 38 79.3 82.6 4.56 0.25
2005 (99) hospital nostica
Nell et al., 2005 (32) Austria English Cohort study CCP2 Axis-Shield ACR Prospective Not reported ⬍12 mo Yes No NA NA 0.125 UA (n ⫽ 98) 42 2 60 96 41.2 98.0 20.18 0.60
Raza et al., 2005 (30) England English Rheumatology CCP2 Axis-Shield ACR Prospective Not reported ⬍18 mo Yes Yes 59.5 53.7 0.1 Osteoarthritis (n ⫽ 10), 24 3 18 79 57.1 96.3 15.62 0.45
clinic hyperlipidemia (n ⫽ 20),
other rheumatic diseases
(n ⫽ 52)
van Gaalen et al., Netherlands English Cohort study CCP2 Euro-Diag- ACR Prospective Not reported ⬍12 mo Yes Yes 49 0.55 3 UA (n ⫽ 107), other rheumatic 82 13 71 301 53.6 95.9 12.95 0.48
2005 (26) nostica diseases (n ⫽ 207)
Correa et al., 2004 (56) Colombia Spanish Teaching CCP2 Inova Di- ACR Retrospective Not reported Not reported Yes Yes NA NA NA Other rheumatic diseases 74 11 8 130 90.2 92.2 11.57 0.11
hospital agnostics (n ⫽ 131), healthy persons
(n ⫽ 10)
De Rycke et al., Belgium English Rheumatology CCP2 Euro-Diag- ACR Prospective Not reported Same period Yes Yes 63.5 34.7 5 Other rheumatic diseases 89 4 29 142 75.4 97.3 27.53 0.25
2004 (54) clinic nostica (n ⫽ 146)
Girelli et al., 2004 (50) Italy English Rheumatology CCP2 Axis-Shield ACR Prospective Not reported Same period Yes Yes 62.9 0.779 NA HCV infection (n ⫽ 14), other 25 2 10 40 71.4 95.2 15.00 0.30
clinic rheumatic diseases (n ⫽ 28)
Grootenboer-Mignot, France English Teaching CCP2 Euro-Diag- Not Not reported Not reported Not reported Yes No NA NA NA Other rheumatic diseases 167 8 98 88 63.0 91.7 7.56 0.40
et al., 2004 (48) hospital nostica reported (n ⫽ 91)
Hitchon et al., Canada English Teaching CCP2 Inter- ACR Prospective Not reported Not reported Yes Yes NA NA NA UA (n ⫽ 23) 26 8 15 15 63.4 65.2 1.82 0.56
2004 (47) hospital medico
Kumagai et al., Japan Japanese Teaching CCP2 Axis-Shield ACR Retrospective Not reported Not reported Yes No NA NA NA Other rheumatic diseases 64 14 15 293 81.0 95.4 17.77 0.20
2004 (45) hospital (n ⫽ 307)
Lopez-Hoyos et al., Spain English Teaching CCP2 Euro-Diag- ACR Prospective Not reported Not reported Yes Yes 62.5 64.8 NA Polymyalgia rheumatica 38 3 0 73 100 96.1 21.72 0.01
2004 (97) hospital nostica (n ⫽ 48)
Bombardieri et al., Italy English Teaching CCP2 Axis-Shield ACR Prospective Not reported Not reported Yes Yes 58.8 NA 10 HCV infection (n ⫽ 10) 23 0 7 39 76.7 100 60.65 0.25
2004 (100) hospital
Nielen et al., 2005 (31) Netherlands English Rheumatology CCP2 Euro-Diag- ACR Prospective Yes 1y Yes Yes 56.1 0.686 0.4 UA (n ⫽ 121) 149 7 109 114 57.8 94.2 9.98 0.45
clinic nostica
Dubucquoi et al., France English Teaching CCP2 Axis-Shield ACR Retrospective Not reported 6–18 mo Yes No NA NA NA Other rheumatic diseases 90 2 50 129 64.3 98.5 42.10 0.36
2004 (52) hospital (n ⫽ 98), healthy persons
(n ⫽ 33)
Söderlin et al., Sweden English Health care CCP2 Euro-Diag- Clinical Prospective Yes 2y Yes Yes 49.6 63.7 0.3 Reactive arthritis (n ⫽ 28), UA 7 2 9 51 43.8 96.2 11.59 0.59
2004 (44) centers nostica diagnosis (n ⫽ 10), other arthritis
(n ⫽ 15)
Vallbracht et al., Germany English Teaching CCP2 Euro-Diag- ACR Not reported Not reported Not reported Yes Yes 56.8 0.712 8.3 Degenerative joint disease 190 12 105 408 64.4 97.1 22.54 0.37
2004 (42) hospital nostica (n ⫽ 163), other rheumatic
diseases (n ⫽ 103), healthy
persons (n ⫽ 154)
van Venrooij et al., Netherlands English Teaching CCP2 Not Not Not reported Not reported Not reported No Yes NA NA NA Other rheumatic diseases 865 79 252 2218 77.4 96.6 22.51 0.23
2004 (41) hospital reported reported (n ⫽ 2297)
Vittecoq et al., France English Cohort study CCP2 Euroimmun ACR Prospective Not reported Not reported Yes Yes 51.7 10.5 0.33 Other rheumatic diseases 69 5 107 133 39.2 96.4 10.80 0.63
2004 (40) (n ⫽ 225)
Bas et al., 2003 (66) Switzerland English Teaching CCP1 Euro-Diag- ACR Cross- Not reported Not reported Yes Yes 62 0.71 NA Other rheumatic diseases 110 24 86 215 56.1 90.0
hospital nostica sectional (n ⫽ 160), spondylo-
arthropathies (n ⫽ 79)
Lee and Schur, United States English Teaching CCP2 Axis-Shield ACR Retrospective Not reported Not reported Yes Yes 47.5 79.1 NA Other rheumatic diseases 68 14 35 132 66.0 90.4 6.89 0.38
2003 (64) hospital (n ⫽ 113), noninflammatory
arthritis (n ⫽ 23)
Rantapää-Dahlqvist, Netherlands English Cohort study CCP2 Euro-Diag- ACR Nested Not reported 6.1 y Yes Yes NA NA 3 Healthy age- and sex-matched 47 7 20 375 70.1 98.2 38.28 0.30
et al., 2003 (62) nostica case– persons (n ⫽ 382)
control
Saraux et al., 2003 (61) France English Cohort study CCP1 Euro-Diag- Clinical Prospective Not reported Not reported Yes Yes 49.4 66.6 NA UA (n ⫽ 157) 40 11 46 146 46.5 93.0 6.64 0.58
nostica diagnosis

Continued on following page

[Link] 5 June 2007 Annals of Internal Medicine Volume 146 • Number 11 W-183

Downloaded From: [Link] by a Tulane University User on 09/03/2013

 Appendix Table 1—Continued

Study, Year (Reference) Location Language Setting Generation Assay Reference Design‡ Blind Interpretation Interval between Test Technical Quality Clinical Mean or Women, Mean Duration Control Participants Result Sensitivity, Specificity, Positive Negative
of CCP Manu- Standard‡ of Test Result‡ and Reference of Anti-CCP Description of Median % of Illness, y % % LR LR
facturer† Standard Antibody Sample Age, y
TP FP FN TN
Reported‡ Reported‡
Suzuki et al., 2003 (60) Japan English Teaching CCP2 Axis-Shield ACR Retrospective Not reported Not reported Yes Yes 57.18 85.2 9.4 Other rheumatic diseases 481 23 68 185 87.6 88.9 7.92 0.14
hospital (n ⫽ 208)
Zeng et al., 2003 (58) China English Teaching CCP1 In-house ACR Not reported Not reported Not reported Yes Yes 46.14 71.7 NA Other rheumatic diseases 90 7 101 313 47.1 97.8 21.50 0.54
hospital ELISA (n ⫽ 132), nonrheumatic
diseases (n ⫽ 98), healthy
persons (n ⫽ 90)
Jansen et al., 2003 (65) Netherlands English Rheumatology CCP1 Euro-Diag- Clinical Prospective Not reported Not reported Yes Yes 57 69 NA UA (n ⫽ 121) 110 3 148 118 42.6 97.5 17.20 0.59
clinic nostica diagnosis
Vincent et al., France English Rheumatology CCP1 Euro-Diag- ACR Not reported Not reported Not reported Yes Yes 58.06 79.6 NA Other rheumatic diseases 139 7 101 464 57.9 98.5 38.9 0.43
2002 (67) clinic nostica (n ⫽ 157), nonrheumatic
arthritis (n ⫽ 314)
Bizzaro et al., Italy English Rheumatology CCP1 Euro-Diag- ACR Prospective Yes Not reported No Yes 65 89.7 NA Other rheumatic diseases 40 5 58 227 40.8 97.8 18.94 0.61
2001 (74) clinic nostica (n ⫽ 174), healthy persons
(n ⫽ 58)
Goldbach-Mansky United States English Cohort study CCP1 In-house ACR Prospective Not reported 12 mo Yes Yes 42.1 0.66 NA UA (n ⫽ 85), other rheumatic 43 12 63 120 40.6 90.9 4.46 0.65
et al., 2000 (76) ELISA diseases (n ⫽ 57)
Schellekens et al., Netherlands English Teaching CCP1 In-house ACR Retrospective Yes Not reported Yes Yes NA NA NA Other rheumatic diseases 72 14 77 298 48.3 95.5 10.77 0.54
1998 (11) hospital ELISA (n ⫽ 329), infectious dis-
eases (n ⫽ 366), healthy
persons (n ⫽ 120)

* ACR ⫽ American College of Rheumatology criteria; CCP ⫽ cyclic citrullinated peptide; ELISA ⫽ enzyme-linked immunosorbent assay; FN ⫽ false negative; FP ⫽ false positive; HCV ⫽ hepatitis C virus; LR ⫽ likelihood ratio; NA ⫽ not available; TN ⫽ true negative; TP ⫽ true positive; UA ⫽ undifferentiated arthritis.
† The locations of the assay manufacturers are as follows: Axis-Shield (Dundee, United Kingdom), Euro-Diagnostica (Arnhem, the Netherlands), Euroimmun (Luebeck, Germany), Inova Diagnostics (San Diego, California), Intermedico (Markham, Ontario, Canada), and Tosho (Tokyo, Japan).
‡ Included for evaluation of study quality.

W-184 5 June 2007 Annals of Internal Medicine Volume 146 • Number 11 [Link]

Downloaded From: [Link] by a Tulane University User on 09/03/2013

 Appendix Table 2. Characteristics of Studies of Rheumatoid Factor*

Study, Year (Reference) Location Language Setting Method of Assay Cutoff Value, Reference Design‡ Blind Interpretation Interval between Test Technical Clinical Mean or Women, Mean Control Participants Result Sensitivity, Specificity, Positive Negative
Measurement Manufacturer† U/mL Standard‡ of Test Result‡ and Reference Quality Description of Median % Duration of % % LR LR
Standard of RF Sample Age, y Illness, y
TP FP FN TN
Reported‡ Reported‡
Quinn et al., 2006 (24) England English Rheumatology Not reported Not reported Not reported ACR Prospective Not reported Not reported No Yes 58 64.2 7 mo Other rheumatic diseases 115 53 67 63 63.0 54.00 1.38 0.68
clinic (n ⫽ 116)
Fernández-Suárez Spain English Primary care Nephelometry Dade Behring 50 ACR Prospective Not reported Not reported Yes Yes 52 45.5 NA Other rheumatic diseases (n ⫽ 25), 30 2 23 73 56.6 97.1 21.23 0.45
et al., 2005 (36) healthy persons (n ⫽ 50)
Kwok et al, 2005 (33) Korea English Rheumatology Nephelometry Dade Behring 15 ACR Retrospective Not reported NA Yes Yes 56 86.8 13.2 Other rheumatic diseases (n ⫽ 68), 77 16 52 52 – – 2.54 0.53
clinic healthy persons (n ⫽ 60)
Greiner et al., Germany English Teaching Nephelometry Dade Behring Not reported ACR Not reported Not reported NA Yes Yes 54.8 NA NA Other rheumatic diseases 75 42 12 191 – – 4.78 0.17
2005 (35) hospital (n ⫽ 233)
Sauerland et al., Germany English Teaching Nephelometry Dade Behring 20 ACR Prospective Not reported NA Yes Yes NA NA NA Other rheumatic diseases 161 89 7 360 69.7 81 4.84 0.05
2005 (29) hospital (n ⫽ 469)
Kamali et al., 2005 (34) Turkey English Teaching LA Not reported 20 ACR Not reported Not reported Not reported Yes No NA NA NA Progressive systemic sclerosis 20 32 26 25 43.5 43.9 1.29 0.78
hospital (n ⫽ 32), Wegener
granulomatosis (n ⫽ 22)
Anuradha and Chopra, India English Rheumatology LA Tulip Diagnostics 8 ACR Not reported Not reported Not reported Yes No NA NA NA Healthy persons (n ⫽ 155) 482 2 82 153 85.5 98.7 66.20 0.15
2005 (39) clinic
Thammanichanond Thailand English Teaching LA Dade Behring 20 ACR Retrospective Not reported Not reported Yes Yes NA OA (n ⫽ 15), other rheumatic 57 25 6 111 90.5 97.3 4.92 0.12
et al., 2005 (27) hospital diseases (n ⫽ 10), healthy
persons (n ⫽ 110)
Choi et al., 2005 (37) Korea English Primary care LA Hitachi 9 ACR Not reported Not reported Not reported Yes Yes NA NA 14.6 Other rheumatic diseases 261 54 63 197 80.6 78.5 3.74 0.25
(n ⫽ 251)
Nell et al., 2005 (32) Austria English Cohort study Not reported Not reported Not reported ACR Prospective Not reported ⬍12 mo Yes No NA NA 0.125 UA (n ⫽ 98) 56 11 46 87 54.9 88.8 4.89 0.51
Raza et al., 2005 (30) England English Rheumatology LA Mast Diagnostics 30 ACR Prospective Not reported ⬍18 mo Yes Yes 59.5 53.7 0.1 OA (n ⫽ 10), hyperlipidemia 22 2 20 80 52.4 97.6 21.48 0.49
clinic (n ⫽ 20), other rheumatic
diseases (n ⫽ 52)
Das et al., 2004 (55) Japan English Teaching Nephelometry Dade Behring 16.3 Prospective Not reported Same period Yes Yes 47.24 93 NA Other rheumatic diseases 42 46 14 127 75.0 73.4 2.82 0.34
hospital (n ⫽ 206)
De Rycke et al., Belgium English Rheumatology LA Difco Laboratories 3.125 ACR Prospective Not reported Same period Yes Yes 63.5 34.7 5 Other rheumatic diseases 93 28 25 118 78.8 80.8 4.11 0.26
2004 (54) clinic (n ⫽ 146)
Girelli et al., 2004 (50) Italy English Rheumatology Nephelometry Dade Behring 20 ACR Prospective Not reported Same period Yes Yes 62.9 77.9 NA HCV infection (n ⫽ 14), other 32 29 3 13 91.4 31.0 1.32 0.28
clinic rheumatic diseases (n ⫽ 28)
Grootenboer-Mignot France English Teaching Nephelometry Dade Behring 20 Not reported Not reported Not reported Not reported Yes No NA NA NA Other rheumatic diseases (n ⫽ 91) 64 18 29 73 68.8 80.2 3.48 0.39
et al., 2004 (48) hospital
Hitchon et al., Canada English Teaching Nephelometry Intermedico 20 ACR Prospective Not reported Not reported Yes Yes NA NA NA UA (n ⫽ 23) 32 10 9 13 78.0 43.0 1.80 0.39
2004 (47) hospital
Lopez-Hoyos et al., Spain English Teaching Nephelometry Dade Behring 22 ACR Prospective Not reported Not reported Yes Yes 62.5 64.8 NA Polymyalgia rheumatica (n ⫽ 48) 36 3 5 70 88.0 96.0 21.37 0.13
2004 (97) hospital
Bombardieri et al., Italy English Teaching Nephelometry Dade Behring 15 ACR Prospective Not reported Not reported Yes Yes 58.8 NA 10 HCV infection (n ⫽ 10) 27 6 3 33 90.0 85.0 5.85 0.12
2004 (100) hospital
Dubucquoi et al., France English Teaching ELISA Biomedical 20 ACR Retrospective Not reported 6–18 mo Yes No NA NA NA Other rheumatic diseases (n ⫽ 98), 84 41 56 90 60.0 69.0 1.92 0.58
2004 (52) hospital Diagnostics healthy persons (n ⫽ 33)
Söderlin et al., Sweden English Health care LA Not reported Not reported Clinical Prospective Yes 2y No Yes 49.6 63.7 0.3 Reactive arthritis (n ⫽ 28), UA 5 4 11 49 31.0 93.0 4.14 0.74
2004 (44) centers diagnosis (n ⫽ 10), other arthritis (n ⫽ 15)
Spiritus et al., Belgium English Teaching Nephelometry Beckman 20 ACR Prospective Yes Not reported Yes Yes 50.75 62 NA Other rheumatic diseases 57 9 33 93 63.0 91.0 7.18 0.40
2004 (43) hospital Instruments (n ⫽ 102)
Vallbracht et al., Germany English Teaching ELISA [Link] 15 ACR Not reported Not reported Not reported Yes Yes 56.8 71.2 8.3 Degenerative joint disease 196 75 99 345 66.0 82.0 3.72 0.41
2004 (42) hospital Diagnostika (n ⫽ 163), other rheumatic
diseases (n ⫽ 103), healthy
persons (n ⫽ 154)
Vittecoq et al., France English Cohort study ELISA In-house 16 ACR Prospective Not reported Not reported Yes Yes 51.7 10.5 0.33 Other rheumatic diseases 62 11 114 127 35.2 92.0 4.42 0.70
2004 (40) (n ⫽ 225)
Bas et al., 2003 (66) Switzerland English Teaching ELISA In-house Not reported ACR Cross- Not reported Not reported Yes Yes 62 71 NA Other rheumatic diseases 143 43 53 196 73.0 82.0 4.06 0.33
hospital sectional (n ⫽ 160), spondylo-
arthropathies (n ⫽ 79)
Lee and Schur, United States English Teaching LA Difco Laboratories 80 ACR Retrospective Not reported Not reported Yes Yes 47.5 79.1 NA Other rheumatic diseases 73 22 29 90 71.6 80.4 3.64 0.35
2003 (64) hospital (n ⫽ 113), noninflammatory
arthritis (n ⫽ 23)
Rantapää-Dahlqvist Netherlands English Cohort study ELISA In-house 20 ACR Nested Not reported 6.1 y Yes Yes NA NA 3 Healthy persons (n ⫽ 382) 49 23 28 359 63.6 94.0 10.57 0.39
et al, 2003 (62) case–
control
studies
Saraux et al., 2003 (61) France English Cohort study ELISA Not reported Not reported Clinical Prospective Not reported Not reported Yes Yes 49.4 66.6 NA UA (n ⫽ 157) 35 8 51 149 41.0 95.0 2.54 0.63
diagnosis
Suzuki et al., 2003 (60) Japan English Teaching Nephelometry Dade Behring 15 ACR Retrospective Not reported Not reported Yes Yes 57.18 85.2 9.4 Other rheumatic diseases 383 38 166 170 69.8 81.7 3.82 0.37
hospital (n ⫽ 208)
Jansen et al., 2003 (65) Netherlands English Rheumatology Nephelometry Dako Diagnostics 30 Clinical Prospective Not reported Not reported Yes yes 57 69 NA UA (n ⫽ 121) 130 8 128 113 50.4 93.4 7.62 0.53
clinic diagnosis
Bizzaro et al., Italy English Rheumatology Nephelometry Not reported Not reported ACR Prospective Yes Not reported No yes 65 89.7 NA Other rheumatic diseases 61 36 37 196 62.2 84.5 4.01 0.45
2001 (74) clinic (n ⫽ 124), healthy persons
(n ⫽ 58)
Vasiliauskiene et al., Lithuania English Teaching ELISA In-house 17 ACR Prospective Not reported Not reported Yes Yes 53.54 75.8 7.8 Other rheumatic diseases (n ⫽ 90), 75 21 21 106 78.1 83.5 4.73 0.26
2001 (73) hospital healthy persons (n ⫽ 37)
Vittecoq et al., France English Rheumatology LA Not reported 80 ACR Prospective Not reported Not reported Yes Yes 51.4 68 0.33 Other rheumatic diseases (n ⫽ 30) 26 1 32 29 44.8 96.7 0.57 13.45
2001 (72) clinic

Continued on following page

[Link] 5 June 2007 Annals of Internal Medicine Volume 146 • Number 11 W-185

Downloaded From: [Link] by a Tulane University User on 09/03/2013

 Appendix Table 2—Continued

Study, Year (Reference) Location Language Setting Method of Assay Cutoff Value, Reference Design‡ Blind Interpretation Interval between Test Technical Clinical Mean or Women, Mean Control Participants Result Sensitivity, Specificity, Positive Negative
Measurement Manufacturer† U/mL Standard‡ of Test Result‡ and Reference Quality Description of Median % Duration of % % LR LR
Standard of RF Sample Age, y Illness, y
TP FP FN TN
Reported‡ Reported‡
Goldbach-Mansky United States English Cohort study Nephelometry Not reported 20 ACR Prospective Not reported 12 mo Yes Yes 42.1 66 NA UA (n ⫽ 85), other rheumatic 70 39 36 93 66.0 70.5 2.24 0.48
et al., 2000 (76) diseases (n ⫽ 57)
Schellekens et al., Netherlands English Teaching ELISA Not reported Not reported ACR Retrospective Yes Same period Yes Yes NA NA NA Other rheumatic diseases 80 28 69 284 53.7 91.0 5.98 0.51
2000 (12) hospital (n ⫽ 329), infectious diseases
(n ⫽ 366), healthy persons
(n ⫽ 120)
Aho et al., 1999 (80) Finland English Rheumatology LA Not reported Not reported ACR Retrospective Not reported Not reported Yes NA NA NA Other rheumatic diseases 64 16 27 153 70.3 90.5 7.43 0.33
clinic (n ⫽ 108), miscellaneous
disorders (n ⫽ 56)
Jónsson et al., Iceland English Teaching ELISA In-house Not reported ACR Not reported Not reported Not reported Yes Yes NA NA NA OA (n ⫽ 50), UA (n ⫽ 74), other 50 14 20 191 71.4 93.2 10.46 0.31
1998 (81) hospital rheumatic diseases (n ⫽ 81)
Swedler et al., United States English Rheumatology Nephelometry Dade Behring 20 ACR Retrospective Not reported Not reported Yes NA NA NA Mixed 89 3 9 39 90.8 92.9 12.71 0.10
1997 (82) clinic
Young et al., 1991 (93) England English Rheumatology Rheumatoid Not reported 40 ACR Prospective Not reported Not reported Yes Yes 51.1 66.6 0.66 Other arthritis (n ⫽ 21) 25 1 14 20 64.1 95.2 13.50 0.38
clinic arthritis
hemaggluti-
nation
de Bois et al., Netherlands English Teaching ELISA Not reported 3 ACR Not reported Not reported Not reported Yes Yes 42 84.5 NA UA (n ⫽ 39) 8 8 0 31 100 79.5 4.44 0.07
1996 (84) hospital
Cordonnier et al., France English Teaching LA Pasteur 40 ACR Prospective Not reported 12–24 mo Yes Yes 50 75.4 0.5 UA (n ⫽ 15), other arthritis (n ⫽ 5) 20 2 29 18 40.8 90.0 4.08 0.66
1996 (85) hospital Production
Visser et al., 1996 (83) Netherlands English Teaching ELISA In-house Not reported Clinical Retrospective Not reported ⬍2 mo Yes Yes 48 67.3 12 Mixed 157 287 78 1466 66.8 83.6 4.08 0.40
hospital diagnosis
Berthelot et al., France English Teaching LA Fumouze 100 ACR Prospective Yes Not reported Yes NA NA NA Not reported 80 50 39 45 67.2 47.4 1.23 0.69
1995 (89) hospital Diagnostics
Saraux et al., 1995 (88) France English Teaching LA Biolyon 40 ACR Retrospective Not reported Not reported Yes Yes 51.98 59 NA Other rheumatic diseases (n ⫽ 99) 8 8 31 91 20.5 91.9 2.54 0.87
hospital
Després et al., Canada English Teaching LA Not reported Not reported ACR Prospective Not reported Not reported Yes NA NA NA Other rheumatic diseases 143 39 63 130 69.4 76.9 3.01 0.40
1994 (91) hospital (n ⫽ 165), other arthritis
(n ⫽ 65), healthy persons
(n ⫽ 36), infectious
mononucleosis (n ⫽ 10)
Gomès-Daudrix et al., France English Teaching ELISA Cogent Not reported ACR Not reported Not reported Not reported Yes NA NA NA Other rheumatic diseases 48 1 40 99 54.5 99.0 55.1 0.46
1994 (90) hospital Diagnostics (n ⫽ 100)
Banchuin et al., Thailand English Teaching ELISA In-house Not reported ACR Not reported Not reported Not reported Yes Yes NA NA NA Healthy persons (n ⫽ 200), cancer 36 6 41 313 46.8 98.1 24.9 0.54
1992 (92) hospital (n ⫽ 30), infectious diseases
(n ⫽ 56), other rheumatic
diseases (n ⫽ 29)
Carpenter and United States English Teaching ELISA In-house 87 Not reported Not reported Not reported Not reported Yes NA NA NA OA (n ⫽ 56), healthy persons 60 8 20 119 75.0 93.7 11.91 0.27
Bartkowiak, hospital (n ⫽ 76)
1989 (98)
Davis and Stein, Zimbabwe English Teaching ELISA Dade MicroScan Not reported ACR Prospective Not reported Same period Yes NA 70 NA Other rheumatic diseases (n ⫽ 55) 18 3 31 25 36.7 89.3 3.43 0.71
1989 (95) hospital
Winkles et al., England English Rheumatology LA Polysciences Not reported Not reported Not reported Not reported Not reported Yes NA NA NA Not reported 113 19 29 481 79.6 96.2 20.94 0.21
1989 (94) clinic
van Leeuwen et al., Netherlands English Teaching ELISA In-house Not reported Not reported Not reported Not reported Not reported Yes NA NA NA Not reported 163 10 28 140 85.3 93.3 12.80 2.07
1988 (96) hospital

* ACR ⫽ American College of Rheumatology criteria; ELISA ⫽ enzyme-linked immunosorbent assay; FN ⫽ false negative; FP ⫽ false positive; HCV ⫽ hepatitis C virus; LA ⫽ latex agglutination; LR ⫽ likelihood ratio; NA ⫽ not available; OA ⫽ osteoarthritis; RF ⫽ rheumatoid factor; TN ⫽ true negative; TP ⫽ true positive; UA ⫽ undifferentiated arthritis.
† The manufacturers of the assays are as follows: [Link] Diagnostika (Wendelsheim, Germany), Beckman Instruments (Fullerton, California), Biolyon (Lyon, France), Biomedical Diagnostics (Marne-la-Vallée, France), Cogent Diagnostics (Penicuik, United Kingdom), Dade Behring (Marburg, Germany), Dade MicroScan (West Sacramento, California), Dako Diagnostics (Glostrup, Denmark),
Difco Laboratories (Detroit, Michigan), Fumouze Diagnostics (Asnières, France), Hitachi (Tokyo, Japan), Intermedico (Markham, Ontario, Canada), Mast Diagnostics (Bootle, United Kingdom), Pasteur Production (Paris, France), Polysciences (Northampton, United Kingdom), Tulip Diagnostics (Goa, India).
‡ Included for evaluation of study quality.

W-186 5 June 2007 Annals of Internal Medicine Volume 146 • Number 11 [Link]

Downloaded From: [Link] by a Tulane University User on 09/03/2013

 Appendix Table 3. Value of Autoantibodies against Cyclic Citrullinated Peptide and Rheumatoid Factor in Predicting Development of Rheumatoid Arthritis*

Study, Year (Reference) Location Language Setting Design Cohort Outcome Measure Mean or Median Follow-up, Patients Treatments Mean or Women, % Reference Diagnostic Test Effect Size
Duration of y Who Received Median Age, Standard
Illness before Completed y
Study Follow-up,
n/n
van Gaalen et al., Netherlands English Population-based Prospective cohort 936 patients with UA OR for RA, adjusted for 3 mo 3 127/318 Not reported 49 55 ACR Anti-CCP OR, 38.9 (95% CI, 9.9–151.0)
2005 (26) cohort study other ACR criteria IgM RF OR, 8.7 (CI, 2.4–31.2)
Arthritis in ⬎3 OR, 5.0 (CI, 1.8–13.2)
joints
Erosion on OR, 8.7 (CI, 2.4–31.2)
radiographs
Nielen et al., Netherlands English Population-based Retrospective cohort 79 blood donors with RA Cumulative percentage of 7.5 y 15 (maximum)
79/79 Not reported 51.4 62 ACR Anti-CCP 40.5% (32/79)
2004 (101) cohort study positive test results IgM RF 27.8% (22/79)
before onset of
symptoms IgM RF or anti-CCP 49.4% (39/79)
Söderlin et al., Sweden English Population-based Prospective cohort 69 patients with RA PPV for rheumatoid 3 mo 2 69/69 Not reported 49.6 63.7 Clinical Anti-CCP PPV, 78
2004 (44) cohort study factor judgment
Rantapää-Dahlqvist, Netherlands English Population-based Nested case–control 83 blood donors with RA OR for RA 3 10.9 Not reported Not reported Not available Not available ACR Anti-CCP (ⱕ1.5 y OR, 28.9 (CI, 4.3–192.6)
2003 (62) cohort studies adjusted in multivariate before symptom
logistic regression onset)
IgM RF (ⱕ1.5 y OR, 1.2 (CI, 0.1–22.1)
before symptom
onset)
IgG RF (ⱕ1.5 y OR, 6.2 (CI, 0.6–61.0)
before symptom
onset)
IgA RF (ⱕ1.5 y OR, 11.4 (CI, 1.3–98.0)
before symptom
onset)
Anti-CCP (⬎1.5 y OR, 16.1 (CI, 3.3–76.7)
before symptom
onset)
IgM RF (⬎1.5 y OR, 1.3 (CI, 0.6–2.4)
before symptom
onset)
IgG RF (⬎1.5 y OR, 0.5 (CI, 0.1–2.4)
before symptom
onset)
IgA RF (⬎1.5 y OR, 5.1 (CI, 1.6–16.0)
before symptom
onset)
Saraux et al., 2002 (69) France English Primary care 184 patients with Prediction of RA ⬍1 2.5 (median) Not Not reported 49.5 68.9 ACR Prospective IgM RF (latex OR, 3.6 (CI, 1.2–10.4)
UA reported cohort agglutination)
study
IgM RF (ELISA) OR, 4.0 (CI, 1.6–10.1)
IgG AKA OR, 6.6 (CI, 2.7–16.0)

* ACR ⫽ American College of Rheumatology criteria; AKA ⫽ antikeratin antibody; CCP ⫽ cyclic citrullinated peptide; ELISA ⫽ enzyme-linked immunosorbent assay; OR ⫽ odds ratio; PPV ⫽ positive predictive value; RA ⫽ rheumatoid arthritis; RF ⫽ rheumatoid factor; UA ⫽ undifferentiated arthritis.

[Link] 5 June 2007 Annals of Internal Medicine Volume 146 • Number 11 W-187

Downloaded From: [Link] by a Tulane University User on 09/03/2013

 Appendix Table 4. Value of Autoantibodies against Cyclic Citrullinated Peptide and Rheumatoid Factor in Predicting Radiographic Progression of Rheumatoid Arthritis*

Study, Year (Reference) Location Language Setting Design Outcome Measure Mean Duration Follow-up, Patients Treatment Received Mean or Women, % Reference Diagnostic Test Effect Size
of Illness y Who Median Standard
before Study, Completed Age, y
mo Follow-up,
n/n
Nell et al., 2005 (32) Austria English Population-based Prospective cohort Larsen score ⬍0.25 3 66/102 Not reported 50 Not reported ACR CCP2 PPV, 88%
cohort study study RF PPV, 78%
Meyer et al., 2006 (25) France English Population-based Prospective cohort Sharp–van der Heijde score 4.3 5 99/99 MTX (n ⫽ 38), 50 73 ACR CCP2 (first 3 y) OR, 3.17 (95% CI, 1.3–7.7)
cohort study study sulfasalazine IgM RF (first 3 y) OR, 0.88 (CI, 0.30–2.58)
(n ⫽ 31), both
MTX and
sulfasalazine
(n ⫽ 27),
corticosteroids
(n ⫽ 33)
Tanaka et al., Japan English Teaching hospital Prospective cohort Sharp–van der Heijde score ⬍1 10 114/130 DMARDs (95%), 54 69 ACR RF OR, 2.07 (CI, 1.01–3.11)
2005 (28) study MTX (35%), CRP OR, 2.86 (CI, 1.01–5.88)
sulfasalazine Synovial membrane OR, 3.59 (CI, 1.53–8.39)
(47%), bucil- enhancement on
lamine (13%), MRI
gold (7%),
auranofin (2%)
Dixey et al., 2004 (53) United English Rheumatology Larsen erosive scores at 3 y ⬍2 3 866/866 Followed UK NA 64 ACR RF OR, 2.43 (CI, 1.72–3.44)
Kingdom clinic guidelines for RA Nodules OR, 2.09 (CI, 1.08–4.05)
HLA-DR shared OR, 2.57 (CI, 1.72–3.85)
epitopes
Goronzy et al., United English Teaching hospital Prospective cohort Sharp–van der Heijde score ⬍1 2 94/111 Followed the 51.5 70.3 ACR RF OR, 3.14 (CI, 1.41–7.00)
2004 (49) States study algorithm created Age OR, 1.45 (CI, 1.08–1.94)
by the authors Number of erosions OR, 4.31 (CI, 1.78–10.42)
Forslind et al., Sweden English Population-based Prospective cohort Larsen score ⬍1 2 333/379 DMARDs (66%), 55 65 ACR Larsen score ⬎3 OR, 9.3 (CI, 5.3–16.1)
2004 (51) cohort study study MTX (36%), sul- Anti-CCP2 OR, 3.0 (CI, 1.7–5.2)
fasalazine (51%) ESR ⬎28 mm/h OR, 1.8 (CI, 1.0–3.1)
Meyer et al., 2003 (63) France English Teaching hospital Prospective cohort Progression of total Sharp score ⬍1 5 156/191 DMARDs or 50.5 73 ACR CCP1 OR, 2.5 (CI, 1.2–5.0)
study Progression of erosion Sharp score NSAIDs (100%) CCP1 OR, 3.4 (CI, 1.6–7.2)
Progression of total Sharp score RF OR, 0.7 (CI, 0.3–1.5)
Progression of erosion Sharp score RF OR, 1.2 (CI, 0.5–2.8)
Vencovský et al., Czech English Prospective cohort Larsen score progression ⬎10 vs. ⬍10 ⬍2 2 104/104 Not reported NA NA ACR CCP1 OR, 4.8 (CI, 2.0–11.4)
2003 (59) Republic study Larsen score progression ⬎10 vs. ⬍10 IgM RF OR, 2.7 (CI, 1.2–6.2)
Larsen score progression ⬎10 vs. ⬍10 IgG RF OR, 2.7 (CI, 1.2–6.2)
Larsen score progression ⬎10 vs. ⬍10 IgA RF OR, 2.9 (CI, 1.2–6.7)
Jansen et al., Netherlands English Population-based Prospective cohort Difference of Sharp–van der Heijde 0.25 1 114/130 Not reported 64 68 ACR IgM RF OR, 2.58 (CI, 1.11–5.97)
2001 (102) cohort study study score from multiple logistic CRP OR, 3.59 (CI, 1.53–8.39)
regression Joint damage at OR, 1.07 (CI, 1.02–1.12)
study entry
Aman et al., 2000 (78) Finland English Rheumatology Prospective cohort Larsen score progression ⬎20 ⬍1 3 63/63 Gold (83%), 43.5 0.83 Clinical RF OR, 3.9 (CI, 1.0–15.5)
clinic study sulfasalazine judgment Cross-linked carboxyl OR, 3.9 (CI, 1.3–11.9)
(12%), hydroxy- telopeptide of type
chloroquine (5%) 1 collagen
Bas et al., 2000 (77) Switzerland English Rheumatology Prospective cohort Larsen score progression/y Not reported 12 (by Not reported Not reported 59 71.6 ACR RF 2.0 points/y (CI, 1.3–2.6)
clinic study exploration Antifilaggrin antibody 1.6 points/y (CI, 1.1–2.2)
of linear
regression
model)
Kroot et al., 2000 (75) Netherlands English Teaching hospital Prospective cohort Sharp–van der Heijde score ⬍1 6 Not reported Not reported 51.5 65.9 ACR Anti-CCP1 (damage Regression coefficient,
study score at 6 y) 0.918 (P ⬍ 0.05)
Anti-CCP1 (damage Regression coefficient,
score at 3 y) 0.209 (NS)
IgM RF (damage Regression coefficient,
score at 6 y) 2.477 (P ⬍ 0.001)
IgM RF (damage Regression coefficient,
score at 3 y) 1.964 ( P ⬍ 0.001)
van Jaarsveld et al., Netherlands English Teaching hospital Prospective cohort Modified Sharp score ⬍1 3 Not reported Not reported NA NA ACR Anti-CCP vs. IgM RF NS
1999 (103) study
Brennan et al., United English Teaching hospital Retrospective Probability of predicting erosion 3 1 175/175 Not reported 59 71 ACR RF Probability of developing
1996 (86) Kingdom cohort study (Larsen score ⱖ grade 2) erosions, 0.46
Disease duration ⱖ3 Probability of developing
mo erosions, 0.26
Involvement of ⱖ2 Probability of developing
joints erosions, 0.37
van Zeben et al., Netherlands English Teaching hospital Prospective cohort Physician opinion 1.6 6 (median) Not reported Not reported NA NA ACR RF OR, 8.26 (CI, 2.8–24.3)
1993 (104) study Agalactosyl IgG OR, 3.34 (CI, 1.03–10.9)
Ritchie score OR, 1.09 (CI, 1.01–1.19)

* ACR ⫽ American College of Rheumatology criteria; CCP ⫽ cyclic citrullinated peptide; CRP ⫽ C-reactive protein; DMARDs ⫽ disease-modifying antirheumatic drugs; ESR ⫽ erythrocyte sedimentation rate; MRI ⫽ magnetic resonance imaging; MTX ⫽ methotrexate; NA ⫽ not available; NS ⫽ not significant; NSAIDs ⫽ nonsteroidal anti-inflammatory drugs; OR ⫽ odds ratio; PPV ⫽
positive predictive value; RA ⫽ rheumatoid arthritis; RF ⫽ rheumatoid factor.

W-188 5 June 2007 Annals of Internal Medicine Volume 146 • Number 11 [Link]

Downloaded From: [Link] by a Tulane University User on 09/03/2013