Measurement properties of isokinetic dynamometry for assessment of shoulder muscle strength: a systematic review.

Abstract Objective To investigate the evidence of measurement properties of isokinetic dynamometry (ID) for assessment of shoulder muscle strength in healthy individuals and patients with non-neurological shoulder pathology. Data Sources Cochrane Central Register of Controlled Trials (CENTRAL), Pubmed, EMBASE, and PEDro were searched up to February 2020 without restrictions. Reference lists and citations were hand-searched. Study selection Two review authors independently included studies that met the following criteria: 1) evaluating measurement properties of ID when used on the glenohumeral joint and 2) included individuals’ ≥ 18 years. Studies including patients with neurologic, neuromuscular, or systemic diseases or critical illness were excluded. Data Extraction The quality assessment and data synthesis were performed according to the COnsensus-based Standards for the selection of health Measurement INstruments (COSMIN) methodology. Data Synthesis Twenty-one studies with a total of 597 participants were included. The results were combined separately for isometric, concentric, and eccentric test mode, for the velocities 30-60°/s, 90°/s, 120°/s and 240°/s, for the seated, supine and standing position, and for internal rotation (IR), external rotation (ER), and the ER/IR ratio. The reliability of ID was overall sufficient with the majority of intraclass correlation coefficients (ICC) ≥0.70. The quality of evidence was moderate or low for 20 out of 30 strata examined. The measurement error results were rated as insufficient for all strata. The standard error of measurement in percent (%SEM) ranged from 4% to 28%. The quality of evidence varied depending of strata examined. Conclusion The reliability of ID for measurement of shoulder strength was overall sufficient for all positions, velocities and mode of strength. The measurement error was not sufficient. As most studies used the seated position, the velocities 30-60°/s or 120°/s, and the concentric test mode, the quality of evidence was highest for these conditions.