Exercise testing in clinical context: Reference ranges for interpreting anaerobic threshold as an outcome for congenital heart disease patients

Change in the oxygen consumption (VO 2 ) at the ventilatory anaerobic threshold (VAT) is an important outcome in research studies of children with congenital heart disease (CHD). The range of values reported by different raters for any given VAT is needed to contextualize a change in VAT in intervention studies. Sixty maximal cardiopulmonary exercise tests (CPET) for CHD patients 8–21 years old were independently reviewed by six exercise physiologists and four pediatric cardiologists. For each of the unique rater pairs for the 60 CPETs, the absolute difference in VAT was calculated and displayed on a histogram to demonstrate the distribution of inter-rater variability. This method was repeated for subgroups of test modality (cycle/treadmill), patient factors (diagnoses, exercise capacity), and rater factors (cardiologist/physiologist, years of experience). Rater agreement was good with an intraclass correlation coefficient of 0.79–0.91 but the distribution of differences was broad. The median difference was 2.7 % predicted peak VO 2 (60 mL/min, 1.0 mL/kg/min), the 75th percentile was 6.4 % (140 mL/min, 2.5 mL/kg/min), and the 95th percentile was 16.3 % (421 mL/min, 6.5 mL/kg/min). Distributions were similar for CPET modality and years of rater experience, but differed for other factors. The baseline distribution of reported VAT is relatively broad, varied by units, and was not explained by differences in rater experience or test modality, but varies by patient factors. When evaluating clinical relevance, a change in the VO 2 at VAT in response to an intervention of <6.5 % predicted falls within the majority (75th percentile) of expected variability and should be interpreted with caution.