de-Torres, J. P. (Autor de correspondencia); O'Donnell, D. E.; Marín, J. M.; Cabrera, C.; Casanova, C.; Marin Oto, Marta
; Ezponda Casajús, Ana
; Cosío, B. G.; Martínez, C.; Solanes, I.; Fuster, A.; Neder, J. A.; González-Gutiérrez, J.; Celli, B. R.
BACKGROUND: The Global Initiative for Obstructive Lung Disease (GOLD) does not promote diffusing capacity for carbon monoxide (DLCO) values in the evaluation of COPD. In GOLD spirometric stage I COPD patients, the clinical and prognostic impact of a low DLCO has not been explored. RESEARCH QUESTION: Could a DLCO threshold help define an increased risk of death and a different clinical presentation in these patients? STUDY DESIGN AND METHODS: GOLD stage I COPD patients (n = 360) were enrolled and followed over 109 +/- 50 months. Age, sex, pack-years' history, BMI, dyspnea, lung function measurements, exercise capacity, BODE index, and history of exacerbations were recorded. A cutoff value for DLCO was identified for all-cause mortality and the clinical and physiological characteristics of patients above and below the threshold compared. Cox regression analysis explored the predictive power of that cutoff value for all-cause mortality. RESULTS: A DLCO cutoff value of <60% predicted was associated with all-cause mortality (DLCO >= 60%: 9% vs DLCO < 60%: 23%, P = .01). At a same FEV1% predicted and Charlson score, patients with DLCO < 60% had lower BMI, more dyspnea, lower inspiratory capacity (IC)/total lung capacity (TLC) ratio, lower 6-min walk distance (6MWD), and higher BODE. Cox multiple regression analysis confirmed that after adjusting for age, sex, pack-years history, smoking status, and BMI, a DLCO < 60% is associated with all-cause mortality (hazard ratio [HR], 95% CI = 3.37, 1.35-8.39; P = .009) INTERPRETATION: In GOLD I COPD patients, a DLCO < 60% predicted is associated with increased risk of death and worse clinical presentation. What the cause(s) of this association are and whether they can be treated need to be determined.