Figure 4.

Flow diagram summarizing the main results of our analysis. Risk ratios (RRs) for mortality were estimated on the basis of the overall mortality in each subgroup after a median follow-up of 6.0 years. The RR for improvement in exercise capacity was assessed by maximal workload achieved on a cycle ergometer 2 years after randomization (less than 10 W vs. at least 10 W).*High risk were patients with FEV₁ not exceeding 20% predicted and either diffusing capacity of carbon monoxide not exceeding 20% predicted or nonheterogeneous distribution of emphysema on computed tomography. ^†Patients with non–upper lobe–predominant emphysema and low exercise capacity can experience more frequent improvement in functional outcomes with lung volume reduction surgery (LVRS) rather than optimal medical management though these are not durable beyond the first 2 years after LVRS, unlike in patients with upper lobe–predominant emphysema. There is no improvement in survival with surgery. In the current analysis upper zone perfusion did not help further define prognosis in this group of patients.