The purpose of this study was to reveal predictors for the long-term effects of pulmonary rehabilitation for chronic obstructive pulmonary disease (COPD) patients, in terms of health-related quality of life (HRQoL). We investigated the long-term effects of pulmonary rehabilitation in 53 COPD patients who had completed the outpatient program and could be evaluated continuously for 1 year. We also investigated factors related to long-term maintenance of HRQoL assessed by the St George's Respiratory Questionnaire (SGRQ). In the year following the program, the only items that retained a significant improvement compared with the pre-program levels were respiratory muscle strength and 6-min walking distance. Patients whose total SGRQ score showed improvements that were maintained above the minimal clinically important difference were placed in a maintained-improvements group (n = 18, 34.0%), and the others in a non-maintained group (n = 35, 66.0%). A comparison of the groups revealed that the maintained-improvements group had significantly lower forced vital capacity (FVC), inspiratory capacity (IC), and tidal volume (TV) at rest; higher PaCO(2); greater initial impairments in HRQoL; and more frequent attendance in a maintenance program. In a multiple logistic regression model, only PaCO(2) was identified as predictor for the maintenance of improvement in HRQoL over a long term. In conclusion, higher baseline PaCO(2) is predictor of maintained, long-term improvement in HRQoL after pulmonary rehabilitation. Frequent attendance in a maintenance program is another predictor.