Background/Purpose

Low body mass index (BMI) has been linked to greater mortality among patients with Rheumatoid Arthritis (RA). Weight loss has also been associated with a greater risk of death among the elderly. The purpose of this study was to determine if weight loss is a predictor of death in RA.

Methods

Our sample consists of 1634 subjects from the Veterans Affairs Rheumatoid Arthritis (VARA) Registry. Dates of death were identified through review of the VA Computerized Patient Record System. BMI was extracted within 14 days of each visit and the change in BMI from the previous visit was determined. BMI category and weight change were considered time-varying. Weight loss at each visit was defined as a decrease in BMI of 1 kg/m² from the preceding visit. Rate of loss (per 1 year) was defined as the change BMI from the preceding visit divided by the preceding interval length. Potential confounding co-variables associated with mortality in this cohort were identified. Cox proportional hazard models were used to assess associations between time-variant and time-invariant predictors of survival including the change in BMI and the rate of decline in BMI from the preceding visit adjusting for age, gender, race, BMI, diabetes, cardiovascular disease, chronic obstructive pulmonary disease, chronic kidney disease, smoking, and current use of methotrexate, prednisone, and anti-Tumor Necrosis Factor (TNF-α) medications. 

Results

Among 1634 subjects (280 deaths, 8102 patient-years, 17,057 unique observations), weight loss of 1 kg/m² of BMI occurred in 2,308 observation periods (13.5%). Weight loss over the preceding interval was associated with an increased risk of subsequent death [HR: 2.00 (1.54, 2.60) p<0.001] (Table 1). In a subset of 1520 subjects with available data (223 deaths, 6650 patient-years), weight loss remained associated with an increased risk of death after further adjusting for CRP [HR 1.78 (1.33, 2.38) p<0.001] (full model not shown). In similar models, a rate of weight loss of >1 kg/m² of BMI over a 6-month period was associated with a greater risk of death [HR: 1.74 (1.31, 2.32) p<0.001] while a slower rate of weight loss was not associated with an increased risk compared to those who did not lose weight [HR: 0.94 (0.70, 1.27) p=0.7].

Conclusion

Recent weight loss, particularly a loss of more than 1 kg/m² of BMI (approximately 3.1 kg on average) over a 6-month period, is an independent predictor of death in RA. Changing weight may be a marker of poor functional status, poor nutrition, ongoing inflammation, and/or underlying malignancy and may help risk-stratify patients for more aggressive interventions.

Table 1:  Multivariable-adjusted risk of death among subjects with rheumatoid arthritis. (N=1634, Deaths=280, Person-Years=8,102)

*p<0.05; †p<0.01; cm<0.001

Abbreviations: BMI= Body Mass Index; TNF= anti-Tumor Necrosis Factor Blocker Therapy; COPD= Chronic Obstructive Pulmonary Disease