Arthritis and back pain impact respiratory-specific quality of life measures in smokers with and without COPD

Abstract

Purpose: COPD affects more than 10 million people in the US and approximately 49% of the US population over age 45 is a current or exsmoker. Comorbid arthritis and degenerative spine disease occur in aging smokers but the prevalence of these conditions in smokers and their impact on respiratory-specific quality of life measures has not been studied. We hypothesized that either condition might impact the St George Respiratory Questionnaire (SGRQ) - which is the most commonly used instrument in both COPD clinical trials and the assessment of COPD progression. We have previously found strong associations of osteoporosis to smoking and COPD severity. Methods: COPDGene is a longitudinal cohort of 10,192 current and exsmokers, age 45 -80, designed to assess subtypes of COPD and their genetic associations. Baseline cross-sectional data from the smoker controls and from COPD subjects was used. Study data included selfreports of musculoskeletal diseases and physical complaints. We identified those with osteoarthritis (OA) based on self-report of physician-diagnosis. Chronic back pain was identified by a positive response to "Do you have lower back, buttock, or radiating leg pain most days of the month?" We determined the prevalence of OA and Chronic Back Pain (CBP) by age groups and by severity of lung disease. Using multivariable regression adjusting for age, gender, obesity and race, we tested the association of current smoking and pack years to OA and chronic back pain. We compared these results to similar analyses of self-reported osteoporosis, osteoporosis based on volumetric QCT measures, and rheumatoid arthritis based on self-report and disease-modifying anti-rheumatic drugs (DMARD) use. Multivariable regression (adjusting for age, gender, race, smoking status and pack years) was used to assess the impact of each of these conditions on SGRQ scores and on the SF-36. Results: Self-reported OA is slightly more prevalent in COPD cases (21%) than controls (17%) in univariate analysis but the association is not present after adjustment for age, gender, and race (OR 0.9, 95% CI 0.8- 1.0, p = 0.07). There are expected associations to female gender, obesity and increased age in univariate and multivariable analyses. In this smoker cohort, the overall prevalence of OA is 19% but increases by age group up to 40% in the 60-70 year old age group. Neither current smoking nor pack years were associated with OA in multivariable models. Chronic back pain is also not associated with COPD, but is strongly associated with African American race, female gender, obesity, current smoking and pack years in multivariable models. Both OA and back pain have significant negative effects on respiratory health status exceeding the minimal clinically significant difference of 4 units in COPD patients), but the magnitude of the effect from chronic back pain is three-fold greater than OA (range 11 -20 units). The generic quality of life instrument, SF-36 showed similar results with both conditions demonstrating significant impairment, but chronic back pain had a 3 to 5-fold greater effects on physical function, bodily pain and physical component scores. Conclusions: Osteoarthritis and chronic back pain are common in a smoking cohort and in COPD patients, but show divergent associations with smoking. Chronic back pain has much greater impact on quality of life than osteoarthritis of the lower extremities. Both conditions affect quality of life measurements using a respiratory-specific instrument. Clinical trials that use the SGRQ to measure outcomes of treatment in COPD should assess subjects for OA and chronic spine disease to avoid confounding. Drug treatments for COPD that are directed toward controlling inflammation may improve respiratory-specific quality of life scores as a result of improved musculoskeletal function. (Figure Presented).