Reliability of BMode Ultrasound for Measurement of Transversus Abdominis Thickness: 1497: Board #260 May 30 11:00 AM 12:30 PM

Moran, Robert W.; Stewart, Andrew M.; Spurdle, Anastasia

Medicine & Science in Sports & Exercise: May 2007 - Volume 39 - Issue 5 - p S226
doi: 10.1249/
A-36 Free Communication/Poster - Measurement of Muscle Mass: MAY 30, 2007 7:30 AM - 12:30 PM ROOM: Hall E

Unitec, Auckland, New Zealand.


Ultrasound imaging is a non-invasive tool that can be used to generate measurements of muscle size, therefore providing a method of direct assessment of muscle atrophy and hypertrophy in response to various pain states or rehabilitative stimuli. Despite considerable clinical research interest in the anterior abdominal wall, especially transversus abdominis (TA), there has been little published research investigating the reliability of measurement of this muscle.

PURPOSE: To determine reliability of B-mode ultrasound for the measurement of TA muscle thickness in volunteers with a history of low back pain.

METHODS: Eight subjects (four male and four female) ranging in age from 21 to 45 yr (mean ± SD = 29.9 ± 7.6 yr) were recruited. An experienced registered sonographer performed five serial measurements of TA thickness for each subject in the supine position during two sessions at an interval of seven days.

RESULTS: The mean coefficients of variation (%) were 5.7 (90% CI=4.7 to 7.4) and 5.5 (4.2 to 6.6) for session one and two respectively. The mean intraclass correlation coefficients (ICC) across the five trials for all subjects were 0.99 (0.97 to 1.0) and 0.99 (0.97 to 1.0) for session one and session two respectively. The range of ICCs (and confidence intervals) across the five trials for all subjects was 0.94 to 0.98 (CI range 0.75 to 0.99) for session one and 0.82 to 0.98 (CI range 0.4 to 1) for session two. The mean (SD) thickness of TA taken at the first and second measurement sessions was 0.37 cm (0.11) and 0.40 cm (0.10) respectively. The mean typical error between the two measurement sessions was 8.1% (5.7 to 15.1). The mean intraclass correlation coefficient for measurement of TA thickness between sessions was 0.93 (0.74 to 0.99).

CONCLUSION: The use of B-mode ultrasound to determine TA thickness as an outcome measure appears to be a highly reliable measurement technique. Both intra and inter-session reliability is excellent, however, the wider confidence interval associated with inter-session reliability indicates more caution should be applied in interpreting the results of multiple session B-mode ultrasound in intervention studies.

© 2007 American College of Sports Medicine