The evaluation of polyneuropathies.

P olyneuropathy has an estimated prevalence of 2%–3% in the general population and a prevalence as high as 8% in people over the age of 55 years.1 Roughly one-third of polyneuropathies will have a genetic cause, one-third an acquired etiology, and onethird will be idiopathic, despite appropriate diagnostic evaluation.2 There are over 100 known acquired and inherited disorders that may cause polyneuropathy, a fact that presents challenges and can contribute to uncertainty about the scope, direction, and level of aggressiveness of any evaluation.3 This sometimes leads to a one-size-fits-all diagnostic strategy, a strategy that is unfocused, inefficient, and costly, and sometimes places the patient at unnecessary risk of a procedure-related complication (e.g., nerve biopsy). In this article, we present a simple and easy-toremember algorithm for diagnosing polyneuropathy, based on first answering 4 clinical questions: what, where, when, and what setting (figure 1).3 The 4-step clinical characterization should almost always be followed by electrodiagnostic (EDX) characterization with appropriate nerve conduction and needle EMG. The clinical and EDX characterization can then be combined, as necessary, with a consultation of appropriate tables and lists of differentials or the figure we provide in this article, allowing for the generation of a focused differential diagnosis and appropriate and efficient evaluation.