Pulmonary Rehabilitation in Interstitial Lung Disease: Benefits and Predictors of Response

Alicia Ferreira, MD; Chris Garvey, FNP; Gerilynn L. Connors, RRT; Lana Hilling, RCP; Julia Rigler, RRT; Susan Farrell, RRT; Cindy Cayou, RCP; Cyrus Shariat, MD; Harold R. Collard, MD, FCCP


CHEST. 2009;135(2):442-447. 

In This Article


Study Population

One hundred thirteen patients were identified. Nine patients had incomplete data due to not finishing the PR program, and five patients had missing or inconsistent data points. The most common reason for not completing the PR program was undergoing lung transplantation (44%). Other reasons included cardiac complications (11%), back pain (11%), patient relocation (11%), noncompliance (11%), and unspecified health issues (11%). Complete data were available for 99 patients. These patients comprised the main study set.

All patients had ILD diagnosed, which was recorded as idiopathic pulmonary fibrosis (n = 50), unspecified ILD (n = 42), scleroderma (n = 3), nonspecific interstitial pneumonia (n = 2), sarcoidosis (n = 1), and lymphangioleiomyomatosis (n = 1). More detailed diagnostic information was not available. The mean patient age was 66 years, and 55% were men. A majority reported previous smoking, but there were no active smokers. There were substantial gas exchange abnormalities present with a mean DLCO of 40% of the predicted value, and 66% of patients required oxygen therapy. Additional baseline features are described in Table 1 .

Change in Borg Score and 6MWT Distance

A statistically significant difference was seen in both the change in Borg score and 6MWT distance after PR ( Table 2 ). The Borg score decreased by an average value of 1 unit (p < 0.0001), and 50% of patients decreased their Borg score by the established minimal clinically significant difference of 1 unit (p < 0.0001). The 6MWT distance increased by an average of 56 m (p < 0.0001), and 49% of patients increased their 6MWT distance by the established minimal clinically significant difference of 54 m (p < 0.0001). The median percentage change in the 6MWT distance was 14% (p < 0.0001).

Change in UCSD and CES-D Scores

UCSD shortness of breath scores and CES-D scores were available for 29 and 27 patients, respectively. A statistically significant difference was observed in both outcomes after PR ( Table 2 ). The UCSD shortness of breath score decreased by an average of 8.3 units (p = 0.005), and 59% of patients improved their UCSD shortness of breath score by the established minimal clinically significant difference of 5 units. For the CES-D, there was an average reduction of 2.2 units (p = 0.046); there is no clearly established minimal clinically significant difference.

Predictors of Change

Of the variables tested (age, gender, baseline FVC, baseline DLCO, smoking history, use of LTOT, baseline Borg score, baseline 6MWT distance, and PR center), only baseline 6MWT distance was a significant predictor of change in 6MWT distance (p < 0.0001), with increasing baseline 6MWT distance predicting a smaller improvement after PR (Figure 1). However, post hoc analysis was unable to identify a baseline 6MWT distance value above which PR seemed ineffective. Lower baseline 6MWT distance was associated with lower baseline FVC (p < 0.0001), lower baseline DLCO (p < 0.0001), and the use of LTOT (p = 0.002).

Relationship of baseline 6-min walk distance to change in 6-min walk distance after PR.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.