Cutoffs, sensitivity and specificity of the Ewing battery in evaluating autonomic nervous system disorders: a systematic review.

Abstract

Purpose: Many methods have been developed for the assessment of dysautonomia, but they are often unreliable and not readily applicable to daily practice. Among the available investigation methods, the Ewing battery, which includes five non-invasive cardiovascular reflex tests, have become the reference standard in assessing dysautonomia and have been utilized for more than four decades. This systematic review evaluates the diagnostic thresholds and diagnostic performance of Ewing tests across studies in identifying autonomic nervous system (ANS) disorders.

Methods: We conducted a comprehensive multi-database literature search, including PubMed, EMBASE, Web of Science, Scopus, and the Cochrane Library, using a pre-defined search strategy, for studies that applied the Ewing tests to assess autonomic dysfunction. We focused on studies that used continuous recordings and those reporting extractable numerical data, including either normative values or diagnostic cutoff thresholds for the Ewing tests. The Ewing parameters that were tested were Valsalva manoeuvre, heart rate variability during deep breathing (E/I ratio), the isometric handgrip, heart rate response to standing (30:15 ratio) and blood pressure response to standing.

Results: Our analysis demonstrates good diagnostic performance of the Ewing tests, with sensitivity for individual components reaching up to 91% for Valsalva ratio (VR) in some cohorts. The normative data and respective cutoff points are influenced by age and sex. Optimal diagnostic performance was achieved when the Ewing battery was interpreted using the conventional criterion of ≥ 2/5 abnormal tests particularly when age-adjusted cutoffs were applied.

Conclusion: The Ewing battery remains a highly effective tool for diagnosing ANS disorders, especially when age-adjusted normative thresholds are used. To further enhance its diagnostic performance, each laboratory should establish its own normative data reflective of the specific population it serves.