Reliability of ultrasound imaging of erector spinae and gluteus medius muscle function using linear regression estimation of contraction

BackgroundDespite the relation of the erector spinae and gluteus medius muscles to low back pain, only a few studies to date have attempted to quantify their dynamic function using ultrasound imaging.ObjectiveTo assess intra-rater reliability of ultrasound imaging measurement of erector spinae and gluteus medius muscle thickness, at rest and during submaximal contraction, and to compare the reliability of percent thickness change calculations with a novel method using linear regression analysis to estimate residualized thickness change.MethodsErector spinae and gluteus medius muscles of 30 participants were imaged at rest and during submaximal contraction. Intra-rater reliability of thickness measurements was estimated both based on a single measurement and on an average of three measurements, with intraclass correlation coefficients (ICCs), standard error of measurement (SEM) and minimal detectable change (MDC).ResultsReliability was high for all rested and contracted thickness measurements with ICCs ranging between 0.85 to 0.98, SEM 0.08 to 0.16 cm, and MDC 0.22 to 0.47 cm. Although less reliable than simple thickness measurements, reliability of residualized change measures was consistently higher than percent thickness change calculations.ConclusionUltrasound imaging of the erector spinae and gluteus medius muscles is reliable. Measurement of simple resting and contracted thickness is highly reliable and can be based on a single measurement. Dynamic measures are less reliable and should be based on an average of three measurements. Measurements of residualized thickness change are consistently more reliable than percent thickness change calculations and its use should be explored in future dynamic ultrasound imaging research.