Clinical predictive value of manual muscle strength testing during critical illness: an observational cohort study

INTRODUCTION: Impaired skeletal muscle function has important clinical outcome implications for survivors of critical illness. Previous studies employing volitional manual muscle testing for diagnosing intensive care unit-acquired weakness (ICU-AW) during the early stages of critical illness have only provided limited data on outcome. This study aimed to determine inter-observer agreement and clinical predictive value of the Medical Research Council sum score (MRC-SS) test in critically ill patients. METHODS: Study 1: Inter-observer agreement for ICU-AW between two clinicians in critically ill patients within ICU (n = 20) was compared with simulated presentations (n = 20). Study 2: MRC-SS at awakening in an unselected sequential ICU cohort was used to determine the clinical predictive value (n = 94) for outcomes of ICU and hospital mortality and length of stay. RESULTS: Although the intra-class correlation coefficient (ICC) for MRC-SS in the ICU was 0.94 (95% CI 0.85-0.98), κ statistic for diagnosis of ICU-AW (MRC-SS <48/60) was only 0.60 (95% CI 0.25-0.95). Agreement for simulated weakness presentations was almost complete (ICC 1.0 (95% CI 0.99-1.0), with a κ statistic of 1.0 (95% CI 1.0-1.0)). There was no association observed between ability to perform the MRC-SS and clinical outcome and no association between ICU-AW and mortality. Although ICU-AW demonstrated limited positive predictive value for ICU (54.2%; 95% CI 39.2-68.6) and hospital (66.7%; 95% CI 51.6-79.6) length of stay, the negative predictive value for ICU length of stay was clinically acceptable (88.2%; 95% CI 63.6-98.5). CONCLUSIONS: These data highlight the limited clinical applicability of volitional muscle strength testing in critically ill patients. Alternative non-volitional strategies are required for assessment and monitoring of muscle function in the early stages of critical illness.