The reliability of rehabilitative ultrasound to measure lateral abdominal muscle thickness: A systematic review and meta-analysis.

Abstract

BACKGROUND\nVariations in rehabilitative ultrasound imaging (RUSI) protocols may alter lateral abdominal muscle (LAM) thickness measurements. A standardised protocol is required for clinicians to accurately compare LAM thickness changes.\n\n\nOBJECTIVE\nIn healthy and lower back pain (LBP) populations, to assess the 1) overall reliability of RUSI to diagnose LAM thickness via meta-analysis, 2) reliability of ultrasound variables to diagnose LAM thickness via systematic review, and 3) propose a RUSI protocol for the LAM using variables associated with excellent reliability (intraclass correlation coefficient [ICC] >0.9).\n\n\nDESIGN\nSystematic review and meta-analysis.\n\n\nMETHOD\nDatabases were searched from January 2000 for studies reporting the reliability of RUSI on the LAM at rest. Title, abstract and full-text screening were performed. Reference lists of reviews and included full-text articles were scanned for further articles. Study characteristic, ultrasound procedure and reliability data were extracted, and article quality assessed. Data was synthesised using meta-analysis to determine the overall reliability for RUSI in different subgroups; calculation of the mean ICCs and standard error of measurements of protocol variables; and narrative synthesis of protocols to contrast those of differing reliability.\n\n\nRESULTS\nTwenty-seven articles, involving 884 participants were included. Reliability ranged from good-to-excellent (ICC 0.859-0.958) in all subgroups. Protocols ranged in subject selection and position, examiner experience, transducer position with the comprehensiveness of protocol description the main limitation of the reviewed literature. Based on the findings an ultrasound protocol was proposed.\n\n\nCONCLUSIONS\nRUSI variables for the LAM at rest show moderate-to-excellent reliability; future research should explore reliability following the proposed protocol.