Statins and Physical Function Among 6,265 Patients

1
Statins and Physical Function Among 6,265
Patients with Rheumatoid ArthritisEric J.
Hochman MD1, Frederick Wolfe MD2, and Hyon K Choi
MD DrPH3 Division of Rheumatology, Washington
University School of Medicine, St. Louis, Mo1,
National Data Bank for Rheumatic Diseases,
Wichita, KS2 , Massachusetts General Hospital,
Harvard Medical School, Boston, MA3

• Results
• The mean age of 6,265 RA patients was 61 years,
  female proportion 22, RA duration 16 years,
  total income 46,600, education level 14 years,
  total comorbidity score 4 (0-22 scale),
  prednisone use 35, biologic use 47, and other
  DMARD use 78.
• Statin use was reported in 968 patients (15) and
  non-statin lipid agent use was reported in 126
  patients (2).
• Demographic and clinical characteristics of
  6,265 RA patients according to statin use are
  summarized in the table below.

• Background
• Recently the TARA (Trial of Atorvastatin in
  Rheumatoid Arthritis) study demonstrated an
  anti-inflammatory effect of atorvastatin based on
  116 patients with Rheumatoid Arthritis (RA) from
  a single center.
• The trial showed a marked suppression with
  atorvastatin of acute-phase variables and a
  significant reduction in swollen joint count but
  no effect was seen on other clinical measures of
  disease activity, particularly health assessment
  questionairre (HAQ). The authors acknowledged
  that large high-powered studies would be
  important to confirm the relevance of these
  results to the wider RA population.
• We performed a cross-sectional analysis (n
  6,265) to examine the relation between statin use
  and HAQ disability score based on a large cohort
  of RA patients from multiple regions in the
  United States (the National Data Bank for
  Rheumatic Diseases NDB).

• After adjusting for age, statin use was
  associated with a significantly lower HAQ score
  (difference -0.10 95 CI, -0.15, -.05). After
  further adjustment for other covariates, the
  magnitude of difference slightly increased
  (difference -0.12 95 CI, -0.17, -.08).
• In contrast, non-statin lipid agent use was not
  associated with HAQ score. Other covariates
  showed expected associations with HAQ scores.
• The table below summarizes the difference in HAQ
  score according to statin use and covariates
  adjusted in the linear regression models.

• Methods
• We studied the NDB data collected over a 6-month
  period (7-12/2002) during which the first
  specific data on statin use (simvastatin,
  lovastatin, pravastatin, atorvastatin,
  fluvastatin, and cerivastatin) were available.
• NDB questionnaires inquired about HAQ scores and
  relevant covariates including age, sex, education
  level, total income, RA duration, total
  comorbidity, and use of prednisone, biologics
  (etanercept, infliximab, and anakinra), other
  DMARDs, and non-statin lipid agents.
• Linear regression was performed to examine the
  relation between statin use and HAQ score among
  those with complete information available for all
  variables (n 6265).

• Conclusions
• These large data indicate that statin use may be
  associated with modestly improved physical
  function assessed by HAQ in RA patients.
• If confirmed by longitudinal studies, it may
  expand the benefits of statins in RA patients
  documented in the TARA trial.