LAUSR

Introduction Nerve conduction study (NCS) is the important diagnostic tool in Guillain Barre syndrome (GBS). Many electrodiagnostic criteria sets have been proposed to distinguish demyelinating from axonal pathology. Misclassification often occurs in early course of the disease. Moreover, timing and selection of the nerves are crucial. Repeated NCS may provide more definite diagnosis. Therefore, the aim of this study is to describe NCS results and their serial changes in specific nerves and related parameters. Methods Retrospective and prospective review of medical records of GBS patients, who were 15 years of age or older, visited to Thammasat University Hospital and Bangkok Hospital Medical Center in Thailand between 1st January 2009 and 31st October 2017. Every patient had at least one nerve conduction study at the time of diagnosis and confirmed by the neuromuscular specialist or treating neurologist. Results Forty-four patients were recruited, 23 (52.3%) of these were female. The mean age at presentation was 49.2 years old. Most patients are Asian (59.1%) or European (15.9%). Antecedent infection mainly were upper respiratory tract and GI infection. The most common clinical diagnostic subtype was AIDP (63%). Other subtypes were AMAN (15.9%), MFS (11%), pharyngeal-cervical-brachial weakness (9%) and AMSAN (2%). Comparing the difference of demyelinating criteria sets revealed that Albers criteria showed the highest sensitivity (98%) to defined demyelinating pattern, whereas Cornblath criteria was the lowest (53%). On the other hand, Ho (94%) and Hadden (91%) criteria sets demonstrated indifferent diagnostic results. Initial NCS was abnormal in 34 of 44 patients (78%). Demyelinating pattern was found in 28 patients (64%), mixed axonal and demyelinating pattern in 6 patients (14%) and the rest was equivocal. Second NCS was performed in 20 patients and the classification was changed in 2 of 20 patients (10%), shifting from demyelinating to axonal subtypes in final NCS. This might reflect reversible conduction block in early axonal GBS. Demyelinating pattern was more prevalent in lower limbs (tibial and peroneal nerves) which were most defined by slow conduction velocity and conduction block. While in upper limb, median nerve was the most commonly affected, based on the conduction velocity, distal latency and temporal dispersion. Sural sparing pattern was common and it was present in 44 studies (80%) in AIDP according to Albers criteria. Conclusion AIDP is the most common subtype of GBS in Thai tertiary care centers. Albers criteria is the most sensitive criteria, whereas Cornblath criteria is the lowest. Initial NCS reveals abnormality in 78% of patients while serial NCS demonstrates subtype change from demyelinating to axonal pattern in 10% of patients. Tibial and peroneal nerves are most frequently affected. Appropriate selection of the studied nerves and classification criteria, combined with serial NCS are important to identify accurate GBS subtypes.