DeFreitas, J. M., Beck, T. W., Stock, M. S., Dillon, M. A., Sherk, V. D., Stout, J. R., & Cramer, J. T. (2010). A comparison of techniques for estimating training-induced changes in muscle cross-sectional area. The Journal of Strength & Conditioning Research, 24(9), 2383-2389.
Premise: Increases in muscular force production (i.e., strength) are attributable to enhancements in neural efficiency of skeletal muscle activation and increases in muscular cross-sectional area (CSA)1,2,3. Determining whether these changes are attributable to ‘neural factors’ or muscular hypertrophy is not practical without the use of imaging technology for estimation of muscular cross-sectional area, or electromyography for estimation of motor unit activity. In addition, where limb circumference measurements are used to measure the effectiveness of a training program for muscular hypertrophy or fat loss, it is not readily apparent whether changes in limb circumference are attributable to one or the other without more technically-intensive laboratory techniques. Consequently, field-based methods of estimating skeletal muscle CSA involving the combination of circumference and skin-fold measurements have been developed. Early work by Housh et al. showed that estimation of total thigh muscle cross-sectional area using these methods produce prediction errors in cross-validation groups of 16.8% (24.8 cm2) and 17.6% (25.4 cm2) in the dominant and non-dominant thighs respectively of forty-three untrained adult males (age: 25 ± 5 years) when compared to criterion values obtained via magnetic resonance imaging2. In both limbs, the prediction equation systematically underestimated the criterion measurement, but shared correlations with the criterion measurement of r = 0.85 (dominant) and 0.88 (non-dominant). These authors suggested that the magnitude of prediction errors may preclude use of this method in detecting training-related changes in muscular CSA since training-induced changes in thigh muscle CSA may be smaller than the error associated with prediction. The original research reviewed here by DeFreitas et al. aimed to establish the reliability of this method, and to evaluate whether it was sensitive enough to detect changes in thigh muscle CSA during eight weeks of resistance training.
Participants & Procedures: Twenty-five healthy men (age: 21.5 ± 3.6 years) with no recent (within 6 months) exposure to an organized weight training program were recruited to participate in two pre-testing sessions, and eight weeks of strength training. In two pre-training sessions separated by 48 hours, total thigh muscular CSA was estimated using anthropometric methods and compared with criterion muscle CSA obtained via peripheral quantitative computed tomography scans (pQCT). The equation for anthropometric estimation of CSA from the original research is as follows2:
- Thigh CSA = (4.68 x Thigh Circumference) – (2.09 x Anterior Skinfold) – 80.99
- Measurements are midway between inguinal crease and proximal border of the patella.
- Thigh circumference is measured to the nearest 0.1cm.
- Anterior skinfold is measured to the nearest 0.5mm.
The training sessions took place three days per week with sessions separated by 48 hours. Loads were adjusted so that subjects achieved failure between eight and twelve repetitions for three sets on each of: bilateral leg press, leg extensions, and bench press exercises. Sets were separated by two minutes. Thigh muscle CSA was evaluated during the initial pre-testing period and then prior to training on the third day of weeks two, four, and six, with post-testing taking place on the seventh day of week 8.
Results: In the two pre-training test sessions, an intraclass correlation coefficient of 0.96 was found for the anthropometric estimation of thigh muscle CSA (compared to 0.995 for pQCT) and a significant correlation of r = 0.95 was reported between these two methods. Over eight weeks of training, the anthropometric method significantly underestimated the measurements obtained by pQCT, but both methods detected significant pre- to post-test changes in thigh muscle CSA. Concerning the pattern of changes detected by each method:
- pQCT detected increases in CSA at week 2, 4, and 6. No change between weeks 6 and 8.
- Anthropometric prediction detected increases in CSA at weeks 2 and 8.
- Total change (pre-post) in CSA was 9.1% (pQCT) , and 9.9% (anthropometric prediction).
Practical Significance (+): Compared to time-, technical-, and cost-prohibitive research-grade estimation of thigh muscle CSA, anthropometric estimates may be sensitive enough to detect changes in muscle CSA over eight weeks of training in relatively untrained, healthy young males.
Practical Significance (-): While the anthropometric method of CSA estimation was significantly correlated to the pQCT method (r = 0.95), it consistently underestimates the likely true values of muscle CSA. Despite this, the pattern of increase detected in muscle CSA were similar between the two methods over eight weeks, differing by only 0.8%. While the strength training range of 8-12 repetitions agrees well with recommendations for muscular hypertrophy, it may be that the rest intervals of two minutes between sets are not optimal (i.e., are too long) for inducing skeletal muscle hypertrophy1. The scope of this research was not necessarily to ‘maximize’ muscular hypertrophy, but it would be useful to know if the sensitivity of the anthropometric estimates might be greater if the rest period was reduced, and a dietary intervention aimed at maximizing hypertrophy were followed.
1Haff, G. G., & Triplett, N. T. (Eds.). (2015). Essentials of strength training and conditioning 4th edition. Human kinetics.
2Housh, D. J., Housh, T. J., Weir, J. P., Weir, L. L., Johnson, G. O., & Stout, J. R. (1995). Anthropometric estimation of thigh muscle cross-sectional area. Medicine & Science in Sports & Exercise, 27(5), 784-791.
3Moritani, T. (1979). Neural factors versus hypertrophy in the time course of muscle strength gain. American journal of physical medicine, 58(3), 115-130.