Study design: Cross-sectional study. Objectives: To measure the reproducibility and responsiveness of heart rate indexes in measuring energy expenditure during wheelchair locomotion of adults with spinal cord injury (SCI). Setting: SARAH Rehabilitation Hospital Network-Brasilia, Brazil. Methods: The study consecutively enrolled 35 individuals with SCI. Volunteers propelled their wheelchairs at a self-selected and comfortable velocity around a circular track. The first assessment day was a responsiveness study, in which three tests were performed, with the wheelchair weight increased randomly during the last two tests. On the second assessment day, the test was conducted without extra weight (that is, baseline conditions) for reproducibility analyses. Results: All indexes demonstrated reproducibility on all statistical analyses. The Propulsion Cardiac Cost Index (PCCI) had a lower variation coefficient than the Physiological Cost Index (PCI; 8.5%, 95% confidence interval (CI) 5.8-11.2 vs 24.2%, 95% CI 16.0-32.4) and lower effect size than the Total Heart Beat Index (THBI; 0.16 vs 0.26). The PCCI also had the best responsiveness results on all statistical tests. The PCCI correlation was the largest (r ¼ 0.63, Po0.01, 95% CI 0.43-0.83, followed by PCI (r ¼ 0.53, Po0.01, 95% CI 0.34-0.72) and then the THBI (r ¼ 0.50, Po0.01, 95% CI 0.29-0.71). Conclusions: All three indexes were reproducible and responsive. However, the PCCI gave better statistical results. This index may be useful in functional diagnosis and follow-up of rehabilitation.