THE RELATIONSHIP BETWEEN PHYSICAL ACTIVITY AND SUBSEQUENT COPD EXACERBATIONS: A PROSPECTIVE STUDY

Abstract

PURPOSE: COPD patients are often physically inactive, and this inactivity is correlated with increases in mortality and subsequent health resource utilization. We explored this relationship further by assessing physical activity at baseline and then determining whether this predicts COPD exacerbations. METHOD(S): Fifty-eight clinically-stable COPD patients were studied. All had the following initial evaluations: 1) Clinical/demographic assessment; 2) FEV1 percent-predicted; 3) Six minute walk distance (6MWD); 4) The Pulmonary Functional Status and Dyspnea Questionnaire (PFSDQ-M) activity score; and 5) Direct activity monitoring over 7 consecutive days using a tri-axial accelerometer worn at the waist. Accelerometer data were stored as vector magnitude units (VMU) over one-minute epochs; non-use data were removed using a computer algorithm designed for this purpose. Mean VMU (counts/minute) and the total number of minutes when VMU were > 250 (VM250) were analyzed. COPD exacerbations were defined as increases in respiratory symptoms that resulted in new therapy or changes in therapy. Exacerbations were determined from patient contacts, record review, and pharmacy contacts. A Cox proportional hazards analyses model was used to relate activity and other variables to time to first COPD exacerbation. RESULT(S): Fifty-eight patients (50% male) were studied; the FEV1 was 54 +/-18% of predicted and the six minute walk distance 320 +/-103 m. The mean VM250 was 1146 +/-544 minutes/week. Higher levels of physical activity by the VM250 and PFSDQ-M correlated with the 6MWD (r = 0.59 and-0.45, resp., both p < 0.001). Thirty-four percent had one or more exacerbations over 348 +/-67 days. In multivariate analysis, the following three variables predicted exacerbations: lower FEV1 (p = 0.005), a history of exacerbations in the preceding year (p = 0.01), and long-acting anticholinergic bronchodilator use (p = 0.001). Questionnaire-rated or directly measured physical activity did not predict exacerbations in univariate or multivariate models. CONCLUSION(S): Physical activity, measured directly or by questionnaire, did not predict exacerbations in our COPD patients. CLINICAL IMPLICATIONS: The nature of the relationship between physical activity and COPD exacerbations remains unclear.