PT - JOURNAL ARTICLE AU - Hana Mullerova AU - Nicholas Locantore AU - Paul Jones TI - GOLD assessment of COPD patients: Impact of symptoms assessment choice DP - 2012 Sep 01 TA - European Respiratory Journal PG - 1650 VI - 40 IP - Suppl 56 4099 - http://erj.ersjournals.com/content/40/Suppl_56/1650.short 4100 - http://erj.ersjournals.com/content/40/Suppl_56/1650.full SO - Eur Respir J2012 Sep 01; 40 AB - Background: The 2011 GOLD guidelines recommend combined COPD assessment using symptoms (modified Medical Research Council Dyspnoea [mMRC] ≥2 or COPD Assessment Test [CAT] ≥10) combined with a history of exacerbations in the past 12mo {0,1} vs 2+ and spirometric classification GOLD I/II vs III/IV. Four groups are identified, A: low symptoms+low risk; B: high symptoms+low risk; C: low symptoms+high risk; D: high symptoms+high risk.Objectives: Characterize the 4 groups using the ECLIPSE (Evaluation of COPD Longitudinally to Identify Predictive Surrogate Endpoints) cohort.Methods: 2028 COPD patients, FEV1<80%pred, provided baseline mMRC, SGRQ and previous 12mo history of exacerbations. SGRQ score ≥25 was used to replace the CAT ≥10, using a validated conversion (Jones: BMCPulmMed 2011).Results: The 4 groups were comparable on age and gender, but had different characteristics. Size of patient groups classified by mMRC were A: 23%, B: 14%, C: 23%, D: 40%; by SGRQ, A: 9%, B: 28%, C: 3%, D: 60%. Compared to the SGRQ, patients classified as 'low symptoms' (GOLD A & C) using mMRC had worse health status, more fatigue and lower exercise capacity (6MWD). Categorising mMRC as 0 vs. ≥1 produced groups of similar size to those classified by SGRQ: A: 9%, B: 29%, C: 4%, and D: 59%. The kappa of agreement for group membership defined by SGRQ and mMRC increased from 0.2 (mMRC ≤1 vs ≥2) to 0.5 (mMRC 0 vs ≥1).Conclusions: The new assessment permits classification of COPD patients beyond airflow obstruction. GOLD recommends either CAT ≥10 or mMRC ≥2 as the symptomatic cut-point, but this analysis suggests that mMRC ≥1 will classify patients more closely to using the CAT.Clinicaltrials.gov NCT00292552; GSK study SCO104960.