ABSTRACT
STUDY DESIGN: Prospective test-retest, intrarater reliability study. OBJECTIVES: To estimate the intrarater reliability, asymmetry, and associated error with measurement of the cross-sectional area (CSA) of the bilateral S1 multifidi when measured by a physical therapist following a short course of self-directed training in ultrasound imaging. BACKGROUND: There is increasing interest in the assessment of the lumbar multifidus during the recovery from low back injury. It is important to know the error associated with the CSA measurements obtained by a physical therapist with limited experience in ultrasound imaging when using a portable unit. METHODS AND MEASURES: Thirty healthy females (mean +/- SD age, 23 +/- 2 years; mean +/- SD mass, 63.1 +/- 9.2 kg; mean +/- SD height, 1.63 +/- 0.06 m) participated. Fourteen subjects returned within 1 to 4 days for repeated measurements. RESULTS: For all 30 subjects, the average (+/- SD) CSA of the left S1 multifidus (4.18 +/- 0.55 cm2) was larger (P<.05) than the right (4.11 +/- 0.57 cm2), with a standard error of the measurement (SEM) of 0.13 cm2 and average +/- SD asymmetry of 3.5% +/- 3.4%. For a subset of 14 subjects, the between-day intrarater reliability for the right S1 multifidus muscle was ICC3,1 = 0.80 (95% CI, 0.49-0.93), while the ICC for the left side was 0.72 (95% CI, 0.34-0.90). The day-to-day average differences for the left and right side were 0.02 cm2 and 0.04 cm2, respectively. For the most conservative estimate, the between-day SEM was 0.37cm2. CONCLUSIONS: A physical therapist, newly trained in ultrasound imaging, obtained reasonable between-day intrarater reliability when imaging the S1 multifidus. A high degree of symmetry was found between the bilateral S1 multifidi in a sample of healthy subjects, which is consistent with previous reports from measurements by skilled ultrasonographers.
