Intra-test and test-retest reliability of exercise oximetry in arterial claudication

Transcutaneous oxygen pressure (tcpO2) reliability is blunted by an unpredictable transcutaneous gradient through the skin. We hypothesized that the "Decrease from Rest of Oxygen pressure (DROP: subtraction of limb-changes from chest-changes from the respective starting values) would show a good to excellent reliability during Exercise -tcpO2 investigations.

In three different experiments we tested: The intra-test variability at the peripheral level (Experiment A: 32 patients, 16 at each location), at the chest level (Experiment B: 45 patients) and the test-retest reproducibility within 3 months (Experiment C: 67 patients). We calculated the intra-class coefficient of correlation (ICC) with 95% confidence interval [Lower/upper limit]. ICC between 0.60 and 0.749 indicate a good agreement. ICC above 0.750 indicates an excellent agreement.

ICC values for DROP-min were 0.848 [0.723/0.935] at the buttock and 0.920 [0.846/0.967] at the calf levels, in experiment A; ICC were 0.873 [0.799/0.923] at the buttock and 0.898 [0.790/0.953] at the calf levels, in experiment B; 0.806 [0.716/0.871] at then buttock level (n = 67) and 0.807 [0.722/0.868] at the calf in experiment C.

Intra-test and test-retest reliability is excellent using the DROP calculation for exercise-tcpO2 investigations.