Quality of the assessment of primary and secondary endpoints in claudication and critical leg ischemia trials.

摘要

Clinical trials in peripheral arterial disease (PAD) require an accurate definition of the disease for inclusion; they typically use treadmill testing, questionnaires and hemodynamic measures as primary and secondary endpoints. Trials of new pharmacologic therapies for PAD often employ multiple clinical sites with presumed expertise in the diagnosis and management of PAD as well as in clinical trials. However, considerable variability has been observed in the assessment of endpoints used in PAD trials, as well as a marked placebo response with treadmill testing. This variability and placebo response impact adversely on overall trial integrity, necessitate an inflated sample size, and may contribute to the large number of recently negative claudication trials. We hypothesized that site monitoring visits for evaluating testing methods would identify and characterize several critical issues that would contribute to poor testing quality. One hundred sites participating in three claudication trials for which peak walking time on the treadmill was the primary endpoint, and 16 sites participating in a critical leg ischemia study for which transcutaneous oxygen tension (TcPO2) was the primary endpoint were evaluated. Each site was visited one or more times by a clinical monitor trained in conducting a 'site endpoint evaluation visit' focusing on equipment, physical set-up of the room in which testing was to be conducted, and the site staff's ability to conduct each of the specific measurements. Full reports were generated that covered a number of technical issues for each measurement and data were extracted from these reports to summarize the testing problems encountered at each site. Problems with treadmill testing were common. For example, 92% of sites had problems with their treadmill equipment, 58% did not perform proper treadmill familiarization, 24% did not start the treadmill test appropriately, 24% did not conduct the test properly, and 15% did not properly conclude the test to determine the peak walking time of the participant. Similar problems were encountered with the ankle-brachial index test, the administration of questionnaires and measurement of the TcPO2. Major deficiencies were identified at the majority of sites in the assessment of primary and secondary endpoints in PAD trials. These errors and improper testing provide a potential explanation for the wide variability and placebo responses observed in claudication and critical leg ischemia trials. Site interventions need to address these deficiencies in measurement to improve the quality of PAD trials.