Dyspnea is a Better Predictor of 5-year Survival than Airway Obstruction in COPD

Abstract & Commentary

Synopsis: In this prospective study of 183 patients, this study found that categorizing COPD patients on the basis of level of dyspnea correlated better with 5-year survival than categorizing on the basis of percentage of predicted FEV1. The study also found no statistically significant difference in dyspnea grade among patients graded by FEV1 criteria.

Source: Nishimura K, et al. Chest. 2002;121:1434-1440.

Chronic obstructive pulmonary disease (COPD) is characterized by slowly progressive airway obstruction and is a major cause of mortality in developed countries. FEV1 is the measure used to assess disease severity for staging and is reported to be the best single correlate of mortality. However, there is evidence that FEV1 may not be the best single parameter and that other measures may be more helpful. Dyspnea is believed by some to be more discriminating with respect to health-related quality of life. The purpose of this study was to compare level of dyspnea vs. disease severity, as defined by FEV1, as predictors of 5-year mortality in patients with COPD.

This was a prospective, observational, multicenter study in Japan following 183 COPD patients (90% male, 10% female) for 5 years. To be eligible, the patients had to have a clinical diagnosis of COPD or the presence of emphysema. Patients with respiratory diseases other than COPD or any severe systemic diseases that could affect prognosis were excluded. On each patient, initial clinical, physiologic, and radiographic features were examined in the stable state. The severity of obstruction was assessed based on American Thoracic Society guidelines. The level of dyspnea was recorded on a scale of I (breathlessness only on strenuous exertion) to V (too breathless to leave the house or breathlessness after undressing). Five years after entry, the patients were followed up with regard to their clinical course including any objective data and cause of death.

At the end of the 5-year period, 134 of 183 patients (73%) were alive. Of the 49 nonsurvivors 22 (45%) died of COPD and 10 (20%) died of malignancy. Between the group of survivors and nonsurvivors, there were significant differences in age (67 yr vs 71 yr, respectively; P < 0.001); dyspnea (grade 2.6 vs 3.3; P < 0.001); vital capacity (2.81 L vs 2.34 L; P < 0.001); FEV1 (1.09 L vs 0.83 L; P < 0.001); DLCO/VA (2.66 mL/min/L/mm Hg vs 1.94 mL/min/ L/mm Hg; P = 0.004); RV/TLC (51.6% vs 60.5%; P < 0.001); PaO2 (75.2 mm Hg vs 67.6 mm Hg; P < 0.001); and PaCO2 (41.2 mm Hg vs 43.9 mm Hg; P = 0.017). There were no significant differences in smoking history, presence of symptoms of chronic bronchitis and FEV1/FVC%.

The staging of disease severity based on ATS guidelines did not significantly correlate with 5-year survival (stage I 5-year survival = 86%, stage II = 75%, stage III = 66%; P = 0.08). When the researchers applied British Thoracic Society and European Respiratory Society guidelines, no significant correlation was found (data not shown). When looking at patients classified according to dyspnea grade, there was a significant correlation with 5-year survival (grade II 5-year survival = 90%, grade III = 76%, grade IV = 35%, grade V = 0%; P < 0.001). A Cox proportional hazards model analysis revealed categorization by the level of dyspnea (with grade II as reference: grade III relative risk = 2.21, grade IV = 8.31, grade V = 61.3 [P < 0.001]) was more significantly related to survival than classification based on FEV1 (with stage I as reference: stage II relative risk = 2.09, stage III = 2.51; P = 0.20). Comparing the backgrounds of groups delineated on the basis of ATS staging and dyspnea score, there was no significant difference in dyspnea among patients graded by ATS stage (dyspnea score stage I = 2.6, stage II = 2.7, stage III = 3.0; P = 0.051).

Comment by David Ost, MD, & Charles Scott Hall, MD

Previous researchers have found FEV1 to be the most accurate predictor of mortality in COPD.1 However, some researchers feel there needs to be a more comprehensive staging system that would allow for better categorization of patients with COPD.2 Hajiro et al found the level of dyspnea correlated with health-related quality of life better than staging according to ATS guidelines.3 Nishimura and colleagues in this study focused on mortality and found the dyspnea score correlated better than the ATS staging. Despite the lack of standardization in the PFT readings among the multiple hospitals involved and the questionably high proportion of men vs. women, these results confirm that FEV1 cannot be used as the sole prognostic marker and that grade of dyspnea should be included in evaluating patients with COPD.

Dr. Ost, Assistant Professor of Medicine, NYU School of Medicine, Director of Interventional Pulmonology, Division of Pulmonary and Critical Care Medicine, Northshore University Hospital, Manhasset, NY, is Associate Editor of Internal Medicine Alert. Dr. Hall is a Fellow in Pulmonary and Critical Care Medicine, North Shore University Hospital-NYU School of Medicine, Manhasset, NY.


1. Anthonisen NR, et al. Am Rev Respir Dis. 1986;133: 14-20.

2. Celli BR. Chest. 2000;117(2 Suppl):S15-S19.

3. Hajiro T, et al. Chest. 1999;116:1632-1637.