The accuracy of screening instruments for sarcopenia: a diagnostic systematic review and meta-analysis

Abstract

Abstract Objective This review aimed to summarise the diagnostic accuracy of screening tools for sarcopenia. Methods We conducted a systematic review along with a critical appraisal of published studies on screening tools for sarcopenia. We assessed the measurement properties of screening instruments using the consensus-based standards for selecting health measurement instruments (COSMIN) checklist. We evaluated the risk bias of the included studies using the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool. The diagnostic test accuracy of instruments for sarcopenia was reported using sensitivity, specificity, positive likelihood ratio (PLR) and negative likelihood ratio (NLR). Results We screened 7,120 titles and abstracts; 42 studies including five screening tools for sarcopenia were included. The overall study quality assessed by the QUADAS-2 tool was moderate to good. Of the five screening tools, three instruments had specificities ≥85%: 92% [95% confidence interval (CI): 63–99%] for the SARC-F modified version, 87% (95% CI: 82–90%) for the SARC-F and 85% (95% CI: 77–90%) for the Ishii score. Three tools had sensitivity ≥75%, namely, MSRA 82% (95% CI: 69–90%), Ishii score 79% (95% CI: 62–89%) and U-TEST 76%. PLR higher than 5.0 were present for the Ishii score and SARC-F modified versions; the Ishii score also had the best NLR of 0.25 of all scales. Conclusion The MSRA and Ishii score had excellent sensitivity for sarcopenia screening at an early stage; SARC-F modified versions and Ishii score had superior specificity for sarcopenia diagnosis.