ST GEORGE'S RESPIRATORY QUESTIONNAIRE (SGRQ) SCORES MAY HELP IDENTIFY COPD PATIENTS AT INCREASED RISK OF DEATH OVER 1 YEAR

Paul Jones1, Peter Calverley2, Thomas Larsson3 and Stefan Peterson3 1St George's Hospital Medical School, London, UK; 2 Aintree Chest Centre, University of Liverpool, Liverpool, UK; 3AstraZeneca, Lund, Sweden.

Background:
Patients with advanced COPD have a poor prognosis, and several mortality risk factors have been identified. Results from two large well-controlled studies of budesonide/formoterol (Symbicort®) vs budesonide, formoterol or placebo in severe COPD (Szafranski, et al. ERJ 2003;21:74–81; Calverley, et al. ERJ 2003;22:912–919) were used to evaluate baseline data as predictors of mortality.

Methods:
Data were obtained from patients treated with budesonide/formoterol or budesonide (n=917) and formoterol or placebo (n=917). All were given terbutaline prn; 77% were male, mean age 64 years, mean pack-years 41, mean FEV1 36% predicted normal and BMI 24.6 kg/m2. Evaluation of baseline covariates was obtained by Cox regression.

Results:
Baseline information was available for 56 deaths (22, budesonide/formoterol or budesonide group; 34, formoterol or placebo group; p=0.036). Baseline SGRQ total score and SGRQ impact domains were the strongest predictive covariates for mortality (p<0.0001), both demonstrating linear associations. Activity (p=0.0002) and SF-36MCS, SF-36PCS and SGRQ symptom domains (all p<0.05) were also predictors. Age, FEV1 and BMI were not predictive of mortality. Adjusting for SGRQ total score or impact domain removed significance for all other predictors.

Conclusions: SGRQ scores may have a role in identifying patients at increased risk of mortality over 1 year.

Back to Abstracts Menu