Background: Peak oxygen uptake (VO2) measured during cardiopulmonary exercise testing (CPX) is often used as an outcome measure in clinical trials. The purposes of this study are (a) to report the outcomes of a quality assurance (QA) procedure instituted in multisite clinical trials by a CPX data core laboratory and (b) to report a normative VO2 reference dataset for future use.
Methods: The CPX laboratory at each site participating in a multisite clinical trial in which Henry Ford Hospital served as the CPX data core laboratory was required to pass a standardized QA procedure before site activation and regularly thereafter. Data were compared with a VO2 reference dataset (pilot data) and assessed for test-retest reproducibility. VO2 data that represented a normal physiologic response were used to develop a final normative VO2 reference dataset.
Results: Between 2003 and 2006, 81 laboratories submitted 144 baseline QA tests. Of these, 34% did not initially meet the passing criteria, largely because of poor test-retest reproducibility. Among all QA tests submitted to the core laboratory, 159 unique volunteers had exercise data that met the criteria to be entered into the final normative VO2 reference dataset. Within this dataset, the mean coefficient of variation for VO2 between the test and retest was 5.1%.
Conclusion: A standardized QA procedure can be used to identify aberrant data and minimize the variability of VO2 measured in a clinical trial or the routine evaluation of patients.