Statistical analysis of fiber area in human skeletal muscle.

Previous research has indicated that 50 fiber measurements per individual for type I and II fibers would be sufficient to characterize the fiber areas. This study replicated the work of McCall et al. (1998) using the three major fiber types (I, IIA, and IIB) and sampling larger populations of fibers. Random blocks of fibers were also examined to investigate how well they correlated with the overall mean average fiber area. Using random blocks of 50 fibers provided an accurate reflection of the type IIB fibers (r = 0.96-0.98) but not for the type I (r = 0.85-0.94) or IIA fibers (r = 0.80-0.91). Type I fibers were consistently reflected by a random block of 150 fibers (r = 0.95-0.98) while type IIA fibers required random blocks of 200 fibers (r = 0.94-0.98), which appeared to provide an accurate reflection of the cross-sectional area. These results indicate that for a needle biopsy different numbers of fibers are needed depending on the fiber type to accurately characterize the mean fiber population.