Current electrophysiologic criteria for chronic inflammatory demyelinating polyneuropathy (CIDP) are highly specific but poorly sensitive. The required extensiveness and best practical way of performing nerve conduction studies to achieve optimal sensitivity remain unknown. We here initially retrospectively analyzed the motor nerve conduction study results of 20 consecutive patients with a clinical diagnosis of CIDP (four performed prior to, and 16 after, treatment initiation) to assess the sensitivity of six published sets of criteria (Nicolas et al., 2002; Thaisetthawatkul et al., 2002; Ad Hoc Subcommittee of the American Academy of Neurology AIDS Taskforce, 1991; Magda et al., 2003; Hughes et al., 2001; Saperstein et al., 2001), as well as four combinations (Nicolas et al., 2002; Ad Hoc Subcommittee of the American Academy of Neurology AIDS Taskforce, 1991; Hughes et al., 2001; Saperstein et al., 2001, each individually combined with Thaisetthawatkul et al., 2002). Sensitivity was highest for the combination of Nicolas et al. (2002) and Thaisetthawatkul et al. (2002) (100%). We then determined the sensitivity of this combined criteria, using five different, hypothetical, nerve conduction study protocols, applied retrospectively to the neurophysiologic data of our 20 patients (exclusive upper limb studies with proximal stimulations; exclusive lower limb studies; full forearm and foreleg studies without proximal stimulations; right‐sided studies with proximal stimulations; and left‐sided studies with proximal stimulations). The findings showed that exhaustive upper limb or, alternatively, four‐limb forearm and foreleg testing would have proved considerably more sensitive than unilateral or lower limb studies to achieve an electrophysiologic diagnosis of CIDP.