TUESDAY, Jan. 30, 2007 (HealthDay News) -- Dutch researchers may have a new method of predicting whether patients with arthritic symptoms will progress to the autoimmune form of the disease, rheumatoid arthritis.
By differentiating those patients who will develop full-blown rheumatoid arthritis from those who will not, the new formula could speed earlier treatment of rheumatoid arthritis patients, reducing damage to their joints while sparing those who will not develop the disease the side effects sometimes associated with rheumatoid arthritis drugs.
"You don't want to give treatment to patients who will spontaneously remit, because they will not get the benefit," explained lead researcher Dr. Annette van der Helm-van Mil, a rheumatologist at Leiden University Medical Center in the Netherlands. "You want to give it only to the patients who have a high chance of progressing to rheumatoid arthritis."
The findings are published in the February issue of Arthritis & Rheumatism.
According to the Arthritis Foundation, rheumatoid arthritis is an autoimmune disease that affects some 2.1 million Americans, most of them women. The disease often presents first as "undifferentiated arthritis," a condition that lacks the criteria for a more definitive diagnosis. Up to 50 percent of patients with undifferentiated arthritis will spontaneously go into remission, while another third will progress to rheumatoid arthritis.
The problem, said van der Helm-van Mil, is that treatment of rheumatoid arthritis with the drug methotrexate at this point can reduce future joint damage but is also potentially toxic. That's why spotting patients with true rheumatoid arthritis early is so important.
In their study, the Dutch group studied a cohort of 570 patients who presented to the Leiden Early Arthritis Clinic with undifferentiated arthritis, 177 of whom progressed to rheumatoid arthritis within one year.
They identified nine variables, including gender, age, the number and distribution of stiff and swollen joints, and three laboratory tests. When factored into an algorithm, these factors could predict the likelihood of developing rheumatoid arthritis with nearly 90 percent accuracy.
Scores from this "prediction rule" ranged from zero to 14. Patients who score six or below have a 91 percent chance of not developing rheumatoid arthritis, the researchers said, while those who score above 8 have an 84 percent chance of progressing to the autoimmune disease. Those who score seven (about 25 percent of patients) have a 50/50 chance of developing rheumatoid arthritis, while those who score above 10 have a 100 percent chance of developing the disease.
"Using information like this can be extremely helpful in managing patients," said Dr. Clifton Bingham III, assistant professor of medicine in the divisions of rheumatology and allergy and clinical immunology at the Johns Hopkins Arthritis Center in Baltimore, Md. "One of the large questions we face in patients who present with undifferentiated arthritis is knowing which of those patients should receive more aggressive therapy to minimize the long-term consequences of the disease or to decrease the likelihood of going on to develop rheumatoid arthritis," he explained.
Bingham noted, however, that this information may be more useful in the United States for primary care physicians than for rheumatologists. The formula already reflects common practice among rheumatologists, he said. Plus, health care differences between the Netherlands and the United States mean that rheumatologists in the U.S. may be less likely to see patients with undifferentiated arthritis than their counterparts in Leiden, because in the U.S., these patients are more likely to present to primary care doctors first. By the time the patient gets to a rheumatologist, he or she has often already developed more-definite rheumatoid arthritis, Bingham said.
"So, it provides a decision tool for primary care doctors to use in determining which patients are most appropriate for early referral to a rheumatologist," he said.
Bingham cautioned that several caveats must be considered before implementing this prediction score in the United States. First, it needs to be validated in other locales and with other patient populations. Second, he cautioned against using this test to produce strict cutoff values for treatment, since what might be true in a population isn't always true for an individual patient. Finally, he noted that the study doesn't address which treatment regimen is most effective once a patient actually develops rheumatoid arthritis.
"We face that question [of treatment] perhaps more often than the question being raised in this study," Bingham said. "We don't yet know how to answer that question."
The methods used in this study could possibly be used to help solve that puzzle, he added.
More information
For more information on rheumatoid arthritis, visit the Arthritis Foundation.
