Finite Sample Rates for Logistic Regression with Small Noise or Few Samples.

BACKGROUND

Alteration in the strain properties of the Achilles tendon may lead to adaptations such as pathological stiffening. Stiff tendons have reduced adaptive ability, which may increase the risk for developing tendinopathy. Strain can be measured using musculoskeletal ultrasound imaging. A two-probe ultrasound procedure may reduce the measurement error associated with a one-probe procedure. However, the reliability of the two-probe procedure has not been established. This study aimed to determine the within-session intra- and inter-rater reliability and between-session reliability of a two-probe ultrasound procedure to measure Achilles tendon strain.

METHODS

Participants were 29 healthy individuals (19 females, 10 males; mean age 33.6 years). Achilles tendon images were acquired with a two-probe ultrasound procedure as the ankle moved through a standardised range of motion (20° plantarflexion to 10° dorsiflexion). Both probes were positioned longitudinally, one over the musculotendinous junction and the second over the calcaneal insertion of the Achilles tendon. Repeat measurements were taken for all participants at the initial study visit, and for 10 participants in a second measurement session 4 weeks later. Strain measures were calculated from pre-captured images using Motion Analysis 2014v1 software by two independent raters. Within-session intra- and inter-rater reliability and between-session intra-rater reliability were calculated using intraclass correlation coefficients (ICC) with 95% confidence intervals. The standard error of measurement was also calculated.

RESULTS

The two-probe procedure to measure Achilles tendon strain showed excellent within-session intra-rater (ICC = 0.84, p < 0.001) and inter-rater reliability (ICC = 0.88, p = 0.003), but poor between-session intra-rater reliability (ICC = 0.18, p = 0.397).

CONCLUSION

The two-probe procedure to measure Achilles tendon strain is reliable for repeated measurements on the same day. However, measurement error increased when strain was measured on different days, which may be attributable to a combination of examiner error and participant factors. Measurement of Achilles tendon strain offers an additional tool for evaluating the tendon's mechanical characteristics. The ability to reliably quantify strain may allow clinicians to identify those at risk for Achilles tendinopathy and formulate more effective management plans.