Meta-analysis: Diagnostic Accuracy of Anti–Cyclic Citrullinated Peptide Antibody and Rheumatoid Factor for Rheumatoid Arthritis

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

From Harvard School of Public Health, Boston, Massachusetts, and Hyogo College of Medicine and Kobe University Graduate School of Medicine, Hyogo, Japan.

For author affiliations, see end of text.

Acknowledgments: The authors thank their colleagues from Kobe University Graduate School of Medicine: Yumiko Nobuhara, MD, for language assistance and Yasuko Nakayama for logistic assistance. They also thank Yasushi Kaji, Dokkyo Medical University, Department of Radiology, for methods assistance and Dr. William F. Goldman (MST Editing Company, Baltimore, Maryland) for editorial assistance and preparation of the manuscript.

Grant Support: In part by a Grant-in-Aid for Young Scientists from the Ministry of Education, Culture, Sports, Science and Technology, Japan, and a grant for Research on Allergic Disease and Immunology from the Ministry of Health, Labor and Welfare, Japan.

Potential Financial Conflicts of Interest:Grants received: S. Kumagai (Japanese Ministry of Health).

Requests for Single Reprints: Kunihiro Nishimura, MD, MPH, MS, Department of Health Policy Management, Harvard School of Public Health, 718 Huntington Avenue, Boston, MA 02215; e-mail, [email protected].

Current author addresses are available at www.annals.org.

Current Author Addresses: Drs. Nishimura and Kuntz: Department of Health Policy Management, Harvard School of Public Health, 718 Huntington Avenue, Boston, MA 02215.

Drs. Sugiyama, Kogata, Tsuji, Nakazawa, Kawano, Saigo, Morinobu, and Kumagai: Department of Clinical Pathology and Immunology, Kobe University Graduate School of Medicine, 7-5-2 Kusunokicho, Kobe, Hyogo, Japan 650-0017.

Dr. Koshiba: Hyogo College of Medicine, 1-1 Mukogawacho, Nishinomiya, Hyogo, Japan 663-8501.

Dr. Kamae: Department of Applied Biostatistics, Kobe University Graduate School of Medicine, 7-5-2 Kusunokicho, Kobe, Hyogo, Japan 650-001.

Annals of Internal Medicine 146(11):p 797-808, June 5, 2007.

Background:

Rheumatoid factor (RF) and autoantibodies against cyclic citrullinated peptide (CCP) are markers that might help physicians diagnose rheumatoid arthritis.

Purpose:

To determine whether anti-CCP antibody more accurately identifies patients with rheumatoid arthritis and better predicts radiographic progression than does RF.

Data Sources:

MEDLINE through September 2006 and reference lists of retrieved studies and review articles.

Study Selection:

Studies in any language that enrolled at least 10 participants and that examined the role of anti-CCP antibody and RF in the diagnosis or prognosis of known or suspected rheumatoid arthritis.

Data Extraction:

Two authors independently evaluated studies for inclusion, rated methodological quality, and abstracted relevant data.

Data Synthesis:

The DerSimonian–Laird random-effects method was used to summarize sensitivities, specificities, and positive and negative likelihood ratios from 37 studies of anti-CCP antibody and 50 studies of RF. The pooled sensitivity, specificity, and positive and negative likelihood ratios for anti-CCP antibody were 67% (95% 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 values were 69% (CI, 65% to 73%), 85% (CI, 82% to 88%), 4.86 (CI, 3.95 to 5.97), and 0.38 (CI, 0.33 to 0.44). Likelihood ratios among IgM RF, IgG RF, and IgA RF seemed to be similar. Results from studies of patients with early rheumatoid arthritis were similar to those from all studies. Three of 4 studies found that risk for radiographic progression was greater with anti-CCP antibody positivity than with IgM RF positivity.

Limitations:

Many studies had methodological limitations. Studies of RF were heterogeneous and had wide ranges of sensitivity and specificity.

Conclusions:

Anti-CCP antibodies are more specific than RF for diagnosing rheumatoid arthritis and may better predict erosivedisease.