Physical activity is the strongest predictor of all-cause mortality in patients with COPD: a prospective cohort study

Systemic effects of COPD are incompletely reflected by established prognostic assessments. We determined the prognostic value of objectively measured physical activity in comparison with established predictors of mortality and evaluated the prognostic value of noninvasive assessments of cardiovascular status, biomarkers of systemic inflammation, and adipokines.

In a prospective cohort study of 170 outpatients with stable COPD (mean FEV(1), 56% predicted), we assessed lung function by spirometry and body plethysmography; physical activity level (PAL) by a multisensory armband; exercise capacity by 6-min walk distance test; cardiovascular status by echocardiography, vascular Doppler sonography (ankle-brachial index [ABI]), and N-terminal pro-B-type natriuretic peptide level; nutritional and muscular status by BMI and fat-free mass index; biomarkers by levels of high-sensitivity C-reactive protein, IL-6, fibrinogen, adiponectin, and leptin; and health status, dyspnea, and depressive symptoms by questionnaire. Established prognostic indices were calculated. The median follow-up was 48 months (range, 10-53 months).

All-cause mortality was 15.4%. After adjustments, each 0.14 increase in PAL was associated with a lower risk of death (hazard ratio [HR], 0.46; 95% CI, 0.33-0.64; P < .001). Compared with established predictors, PAL showed the best discriminative properties for 4-year survival (C statistic, 0.81) and was associated with the highest relative risk of death per standardized decrease. Novel predictors of mortality were adiponectin level (HR, 1.34; 95% CI, 1.06-1.71; P = .017), leptin level (HR, 0.81; 95% CI, 0.65-0.99; P = .042), right ventricular function (Tei-index) (HR, 1.26; 95% CI, 1.04-1.54; P = .020), and ABI < 1.00 (HR, 3.87; 95% CI, 1.44-10.40; P = .007). A stepwise Cox regression revealed that the best model of independent predictors was PAL, adiponectin level, and ABI. The composite of these factors further improved the discriminative properties (C statistic, 0.85).

We found that objectively measured physical activity is the strongest predictor of all-cause mortality in patients with COPD. In addition, adiponectin level and vascular status provide independent prognostic information in our cohort.