ABSTRACT
UNLABELLED: Poor correlation among the various methods for measuring anaerobic threshold (AT) has been reported, and some authors have even reported interobserver variability within a single method. It is unknown whether such variability exists in patients with metabolic myopathies (MM). OBJECTIVE: To determine intra- and interobserver variability in the measurement of AT in patients with MM and analyze the differences for each method used. We enrolled 16 patients with differing forms of MM for exercise testing. AT was determined by 4 methods: lactic threshold (LT), semi-log LT, respiratory equivalent and V-slope. AT was assessed by 2 observers, who analyzed plots in random order on 2 different occasions, locating the AT in each plot. The respiratory equivalent method gave the best intraobserver correlation (r' = 0.95; p < 0.05) and was superior to the LT method (r' = 0.68; p < 0.05). Interobserver correlation was equally acceptable for all except the V-slope method (r' = 0.36; p > 0.05). The most important finding after comparison of the methods was the significant difference in VO2 (ml/min) in AT between the LT method and the non invasive methods (LT = 1,006; respiratory equivalent 1,312; p < 0.05; LT = 1,095; V-slope = 1,251; p < 0.05). CONCLUSION: Intra- and interobserver variability is slight in patients with MM; the best method in this respect is that of the V-slope. AT appears later with non invasive methods and for that reason, invasive measurement (LT) is preferable.
