Myths and methodologies: Assessment of dynamic cerebral autoregulation by the mean flow index

Abstract

AbstractThe mean flow index—usually referred to as Mx—has been used for assessing dynamic cerebral autoregulation (dCA) for almost 30 years. However, concerns have arisen regarding methodological consistency, construct and criterion validity, and test–retest reliability. Methodological nuances, such as choice of input (cerebral perfusion pressure, invasive or non‐invasive arterial pressure), pre‐processing approach and artefact handling, significantly influence mean flow index values, and previous studies correlating mean flow index with other established dCA metrics are confounded by inherent methodological flaws like heteroscedasticity, while the mean flow index also fails to discriminate individuals with presumed intact versus impaired dCA (discriminatory validity), and its prognostic performance (predictive validity) across various conditions remains inconsistent. The test–retest reliability, both within and between days, is generally poor. At present, no single approach for data collection or pre‐processing has proven superior for obtaining the mean flow index, and caution is advised in the further use of mean flow index‐based measures for assessing dCA, as current evidence does not support their clinical application.