Recurrence Quantification Analysis of F-Waves and the Evaluation of Neuropathies



1. Introduction

The clinical classification of neuropathies has depended on electrodiagnostic (EDX) studies based on distinctions between axonal and demyelinating processes. Such an approach has limitations. Axonal and demyelinating injury is not dichotomous since axons andmyelin are in fact intimately connected functionally. In addition, structural injury to nerves in a pathological sense is not the only basis for altered nerve conduction. Functional changes in ion channels can, for example, produce similar effects without disruption of the structural integrity of nerves.

As amethodology, axonal/demyelinating paradigms have been variable and have involved consensus criteria. Based on sensitivity and specificity, criteria sets using such paradigms

have not produced a satisfactory EDX separation of acute (AIDP) and chronic (CIDP) inflammatory demyelinating polyneuropathies from other common neuropathies [1–3].

This is true despite almost 30 years of effort and at least 16 proposed criteria sets. This large number of criteria sets has been used to argue for the limited utility of the method in general [2]. A recent article reports that two of the proposed criteria sets are associated with a clinical diagnosis of CIDP with a reasonable degree of sensitivity and specificity [4]. This report, however, depends on prior clinical analysis and therefore retains the fundamental problem of all such studies. As has been argued elsewhere [3], it would be preferable to calculate the likelihood that a neuropathy has specific features based on the EDX data itself without dependence on a clinical diagnosis as the primary standard.These concerns about current approaches to EDX evaluation of neuropathies are compounded by the fact that adequate reproducibility of nerve conduction studies is present only if such studies are

performed by the same electromyographer [5].