Postmenopausal women with rheumatoid arthritis were almost three times more likely to die over a 10-year period than were those who did not have the disease in a subanalysis of participants in the Women’s Health Initiative study.
Using stored serum from Women’s Health Initiative (WHI) subjects, investigators found that the increased risk was mostly attributable to women with anti-cyclic citrullinated peptide (anti-CCP) antibodies – a marker used to identify those in the sample who probably had true rheumatoid arthritis (RA). This finding stayed significant even when the analysis controlled for the use of disease-modifying antirheumatic drugs (DMARDs), and also in several models that controlled for health and lifestyle risk factors, Dr. Lewis H. Kuller and his colleagues reported in Arthritis & Rheumatism.
"The increased risk of death for anti-CCP positive women was not explained by age, rheumatoid factor positivity, ANA [antinuclear antibody] positivity, or DMARD use," wrote Dr. Kuller of the University of Pittsburgh and his coauthors (Arthritis Rheum. 2014 Dec. 23 [doi:10.1002/art.38268]).
Dr. Lewis H. Kuller
"Traditional risk factors (e.g., smoking, diabetes), kidney disease, joint pain, health status or [white blood cell] count also did not statistically explain the increased risk for anti-CCP-positivity, despite their independent associations with higher mortality," they said.
The study focused on 9,988 women from the WHI, all of whom self-reported having RA at baseline or at follow-up visits. To further refine the sample, the investigators defined probable RA as the presence of anti-CCP antibodies, with or without the use of DMARDs (8% of the sample), or the absence of anti-CCP antibodies reported with DMARD use (7%). Thus, a total of 1,029 women were considered to have probable RA. The remainder of the sample probably had osteoarthritis, the authors suggested.
The mortality analysis further considered the timing of RA onset by whether it was reported at baseline or during follow-up.
During the 10-year study period, 13% of the group died, including 14% of those with RA at baseline and follow-up, 16% of those with RA only at baseline, and 10% of those who reported it only at follow-up. The median times to death were 8 years for those with baseline RA, 7 years for those with baseline and follow-up RA, and 6 years for those with RA at follow-up.
Women with anti-CCP antibodies had the highest age-adjusted death rate (19 per 1,000 person-years), followed by those who did not have anti-CCP antibodies but were taking DMARDs (18 per 1,000 person-years) and those who probably did not have RA (no antibodies and no DMARDS; 10 per 1,000 person-years).
ANA positivity occurred in 16% of the women, but two-thirds of them were negative for anti-CCP and rheumatoid factor and "therefore probably did not have clinical RA." Age-adjusted death rates were lower for ANA-positive women than for anti-CCP–positive and rheumatoid factor–positive women and were not higher for anti-CPP–positive women with or without rheumatoid factor, or for anti-CPP–negative women with rheumatoid factor. This suggested to the investigators that "anti-CCP positivity rather than rheumatoid factor positivity was the correlate of increased death rates."
Mortality was similar for those using DMARDs and those not using them, and for the use of methotrexate, compared with other DMARDs.
There were no between-group differences in causes of death, which were most frequently cardiovascular disease and cancers. Smoking, diabetes, less physical activity, poor overall health, and a history of coronary heart disease all increased the death rate among those with anti-CCP antibodies. About a quarter of those with the antibodies reported poor health status, with a death rate four times higher than those who probably didn’t have RA and who also said they had excellent health.
Joint pain in the 4 weeks before the baseline interview was another indicator of mortality. Among all antibody-positive women, those who reported severe pain were twice as likely to die during the follow-up as were those who reported no pain or mild pain. Additionally, among all women with severe joint pain, the death rates were significantly higher for those with the antibodies (35.5 per 1,000 person-years) than for those who reported RA but were antibody negative (17.2 per 1,000 person-years), those who reported arthritis but not RA (12.1 per 1,000 person-years), and those who reported no arthritis (11.4 per 1,000 person-years).
In a multivariate, age-adjusted model, anti-CCP–positive women using DMARDs were almost three times more likely to die (hazard ratio [HR], 2.8) than were women without RA. The risk was also significantly elevated in anti-CCP–positive women who weren’t taking DMARDs (HR, 2.2) and for those who were anti-CCP–negative but who used DMARDs (HR, 1.8).