Background/Purpose: The role of body mass index (BMI) in rheumatoid arthritis (RA) disease activity and response to treatment has been difficult to determine.¹ As a measure of physiologic function, basal metabolic rate (BMR) may be a better indicator than BMI for clinical assessment and treatment response in RA patients. In this posthoc analysis, the relationship between BMR, BMI, disease activity and responses to etanercept (ETN)-methotrexate (MTX) treatment were assessed in subjects with moderately active RA in the PRESERVE trial.²

Methods: Subjects with DAS28 >3.2 and ≤5.1 despite stable doses of oral MTX received open-label ETN 50 mg once weekly (QW) plus MTX (titration to ≤25 mg/week through week 28) for 36 weeks (Period 1). Posthoc analyses of disease activity and treatment response by BMR and BMI categories at baseline and Week 36 of treatment were conducted in subjects who received ≥1 treatment dose and had Week-36 assessments. The Mifflin-St Jeor³ and revised Harris-Benedict⁴ formulas were used to calculate BMR. Adjusted mean DAS28 scores were calculated using ANCOVA with BMR categories and weight as predictors at baseline and using weight and baseline scores as predictors at Week 36. Similar analysis performed on BMI categories but weight was excluded because it was not significant. Pearson correlations were used to determine overall associations between BMR/BMI and DAS28.

Results: Of 829 subjects analyzed, the proportions of patients with each category of BMR (Mifflin-St Jeor) were evenly distributed and half had a BMI in the low to normal range (table). At baseline, higher BMR was significantly associated with lower DAS28 scores after adjusting for weight (r=-0.127, p=0.008), and lower BMI was associated with lower DAS28 scores (r=0.124, p<0.001). At week 36, higher BMR was significantly associated with lower DAS28 scores (r=-0.233, p<0.001) and were also associated with a greater likelihood of remission (DAS28<2.6, p<0.001). Similar results were seen using the revised Harris-Benedict⁶ formula. Higher BMI was significantly associated with higher week 36 DAS28 scores (p=0.002) and with decreased likelihood of remission (p=0.047).

Conclusion: Subjects with moderate RA and higher BMRs demonstrated a better clinical response after 36 weeks of etanercept-MTX treatment compared to their counterparts with lower BMRs after adjusting for weight. Our data confirm previous research that BMI is suggestive of treatment outcomes in RA, but may not be clinically meaningful. In addition, the data suggest that BMR may be a more sensitive predictor than BMI. Further research is needed to evaluate the relationships between BMR, BMI, disease activity and treatment outcomes in RA.

References: 1. Westhoff G, et al. Arthritis Rheum. 2007;56:3575–82. 2. Smolen J, et al. Lancet. 2012; Accepted for publication. 3. Mifflin MD, et al. J Am Diet Assoc. 2005;51:241-7. 4. Roza AM, et al. Am J Clin Nutr. 1984; 40:168-82.