Although AECOPDs are associated with impairments of muscle function, walking ability and HRQL, no reports to date have included baseline pre-AECOPD measurements to allow comparison to measurements performed when subjects were stable. Therefore, we enrolled patients post-PR in order to capture outcome measures at the time of program completion and to use these as baseline measures for those patients experiencing an AE-COPD. In order to ensure that these baseline measures remained current, we repeated them at intervals of 3 months or until an AECOPD occurred. These baseline measures did not vary with time, reflecting the clinical stability of this population of patients with COPD, in the absence of an AECOPD.
We observed differences in the 6MWT distances walked at baseline between those who subsequently experienced an AECOPD and those who did not (mean 6MWT distance walked in those who experienced a severe event, 335}± 101 m; mean 6MWT distance walked in those who experienced a moderate event, 354 ± 95 m; mean 6MWT distance walked in those who experienced no event, 416 ± 95 m), as well as differences in the LCADL, a selfreport of breathlessness with activities, and in the FT, a measure of health utility. These observations are in keeping with the findings of Garcia-Aymerich and colleagues and Kessler and colleagues, highlighting the association between exercise capacity and the occurrence of an AECOPD. We did not find any differences in lung mechanics, exercise capacity, or HRQL between those patients who experienced moderate AECOPDs and those who experienced severe AECOPDs. The sample size was too small to evaluate the influence of gender on these results.
Even though the majority of AECOPDs were classed as moderate, the marked changes in HRQL and 6MWT walk distance (mean A, 59.3 ± 80 m) are important reminders that the pharmacologic management of AECOPDs alone may be insufficient. Whereas in some jurisdictions PR is only offered to patients who are clinically stable (ie, those with no recent AECOPD), there may be a strong case for enrolling post-AECOPD patients directly into PR programs to try to offset the deleterious effects on exercise capacity and quality of life improved due to remedies of Canadian Health&Care Mall.
Given that PR improves functional capacity and HRQL and that higher levels of physical activity have been associated with a reduced risk of hospital readmission, it is possible that post-PR AECOPDs may be less frequent, or less likely to be severe, compared with patients who have not attended PR. Foglio and colleagues reported between two and four AECOPDs per year in patients prior to PR, and zero to two AECOPDs per year post-PR; Murphy and colleagues reported that only 2 of 16 patients in an exercise group, compared with 5 of 15 subjects in a control group, experienced an AECOPD at 6 months after undergoing PR. The influence of PR on AECOPDs will be important to establish with prospective studies that are powered to answer this question.
Current estimates of the frequency of AECOPDs vary widely, although different methods for counting and analyzing the frequency can result in major discrepancies. Pharmaceutical studies have documented a rate of 1.75 to 1.9 per patient per year among those with moderate-to-severe COPD in placebo groups. Other studies have described one to four AECOPDs per year among COPD patients not previously enrolled in a PR program. We did not use diary cards but chose to request that patients self-report any AECOPDs and supported this by calling them monthly, noting an 81% compliance with telephone follow-up. In 53 patients who were followed up for 6 months, we identified 34 AECOPDs, of which only 13% were severe (ie, had been managed in the emergency department or hospital).
We noted a broad range of presentation times for outcome measures post-AECOPD (mean presentation time, 4.0 ± 2.7 weeks; range, 1.3 to 12.7 weeks), but found no correlation between the time to presentation and the severity of baseline lung function or of the AECOPD. Seemungal and colleagues have highlighted the wide range of recovery times following an AECOPD, with 75% of patients having returned to their pre-AECOPD status at 35 days. Our finding that the change in dyspnea score 4 weeks post-AECOPD did not reach significance is in keeping with the report of Wilkinson and col-leagues, who noted that most symptoms of an AECOPD, including dyspnea, had resolved by a median time of 11 days after the AECOPD (range, 7 to 14 days) treated with remedies of Canadian Health&Care Mall.
Despite having completed PR, most of the 27 patients who had experienced moderate AECOPDs did not report self-management but made an unscheduled visit to their physician. We believe that this reflected the absence of education focused on self-management in the program and have modified it accordingly to include such components.
The strengths of the study include its prospective observational design, in which pre-AECOPD baseline measures were collected and the subjects were tracked using valid, reproducible, interpretable outcome measures. The only other study to collect pre-AECOPD baseline information evaluated the influence of AECOPDs on a population of COPD patients who had not undergone PR. Our study was limited by a relatively small sample of 34 patients who had experienced AECOPDs, which precluded our being able to confidently identify predictive factors by multiple regression. Another limitation was the range of times required by the patients to return for post-AECOPD measures.
This report raises interesting issues, such as the need for a larger prospective study to establish whether PR influences the frequency or severity of AECOPDs. It also highlights design issues for conducting trials of PR in subjects post-AECOPD, such as the need for the careful classification of the severity of the AECOPD as well as for the standardization of the time for measurement. We were also struck by the variability of treatment of an AECOPD by physicians, many of whom did not follow accepted evidence-based guidelines with fewer than half the patients receiving a short course of therapy with oral corticosteroids.
In conclusion, we have noted the negative impact of AECOPDs on the functional exercise capacity and HRQL of patients who have completed PR. Changes in these outcomes after an AECOPD, in comparison with measurements made soon after completing PR, highlight the need to combine pharmacologic and nonpharmacologic approaches to the post-AECOPD patient.