Re-evaluating the accuracy of optimized electrodiagnostic criteria in very early Guillain-Barré syndrome: a sequential study.

Using recent optimized electrodiagnostic criteria sets, we aimed at verifying the accuracy of initial nerve conduction studies (NCS) in classic very early Guillain-Barré syndrome (VEGBS), ≤ 4 days after onset, compared with the results of serial NCS. This is a retrospective study based on unreported and consecutive VEGBS patients admitted to two university hospitals between 2015 and 2019. Each patient had serial NCS in at least four nerves. Initial NCS studies were done within 4 days after onset, and serial ones from days 20 to 94. Electrophysiological recordings were blinded evaluated by four of the authors, GBS subtype being established accordingly. Seven adult classic VEGBS patients were identified with a median age of 58 years. At first NCS, GBS subtyping was only possible in 1 case that exhibited an axonal pattern, the remaining patterns being equivocal in 3, and mixed (combining axonal and demyelinating criteria) in the remaining 3. Upon serial NSC there was a rather intricate evolution of electrophysiological GBS patterns, 3 of them being classified as axonal or demyelinating, and the remaining 4 as equivocal or mixed. NCS in VEGBS systematically allows detection of changes suggestive of peripheral neuropathy, though even after serial studies accurate GBS subtyping was only possible in 43% of cases. We provide new pathophysiological insights for better understanding of the observed electrophysiological changes.