Seronegative Rheumatoid Arthritis: A Case Control Study

Abstract

Background: Seronegative Rheumatoid Arthritis (RA) is a disorder associated with considerable diagnostic, prognostic and therapeutic uncertainty for many clinicians.

Objectives: The aim of this study is to elucidate clinical features at diagnosis, manifestations and treatment of patients with RA with negative serologies, as compared to a control group of patients presenting with similar polyarthralgias but diagnosed with alternative (non-RA) musculoskeletal disorders.

Methods: The study was a retrospective chart review of electronic medical records from January 2003 to December 2012. Patients were identified using ICD-9 code Rheumatoid Arthritis 714.0 and at least two rheumatology clinic visits during the specified time. Charts were reviewed individually by two investigators. The inclusion criteria were a diagnosis of RA confirmed by a rheumatologist and normal values for both rheumatoid factor (RF) and anti-cyclic citrullinated protein antibodies (ACPA, third generation assay). Charts were also reviewed for eventual final diagnosis, either seronegative RA or alternate diagnosis (control group). Data were collected on demographics (sex, race, smoking status), family history of RA, and laboratory values (presence of anemia, inflammatory markers) at the time of diagnosis. The presence of erosions and synovitis identified by imaging studies was assessed. In addition, the presence of extra-articular manifestations of RA including nodules, pleural or parenchymal lung disease, eye involvement and osteoporosis was recorded. The therapies (disease modifying anti-rheumatic drug (DMARD), biologic) used to treat the seronegative RA were also reviewed. The family & smoking history and laboratory values of the seronegative RA patients were compared to the control group and analysis was done using Fisher’s exact test.

Results: Charts from 107 patients were reviewed. Forty-four patients were eventually classified as having an alternate diagnosis and were considered the control group. Sixty-three patients were considered to have an established clinical diagnosis of seronegative RA. Among all patients at the time of diagnosis, 25% were smokers, 13% had a family history of RA, 54% were anemic, and 76% had abnormal ESR or CRP. The RA patients had statistically higher proportion with anemia compared to controls at presentation, and statistically lower proportion with ESR elevation compared to controls (p=0.033 and p=0.013, respectively). Seven of the 59 (11%) patients who had hand/wrist films during their care had erosions on radiography, and 6 of 13 (46%) patients who had an MRI of an extremity had findings of synovitis. Extra-articular manifestations were infrequent in this group. Forty-eight of the 62 were initiated on a DMARD, most commonly hydroxychloroquine (16% patients) or methotrexate (29% patients) or a combination of methotrexate and hydroxychloroquine therapy (35%). Of the 63 patients, 17 (27%) patients required a biologic therapy during treatment course.

Conclusions: This study supports the hypothesis that clinical history and physical examination can be important determinants in helping to diagnose seronegative RA and distinguish it from other polyarthopathies. In addition to characteristic symptoms, factors which might contribute to diagnosis of RA in a patient without seropositivity include presence of anemia, and results of imaging studies.