Submitted by Elliot Bodofsky, MD
Edited by Hristelina Ilieva, MD, PhD
A Model to Predict the Probability of Acute Inflammatory Demyelinating Polyneuropathy.
Tan C, Yukari S, Khena-Jin G, et al, Clinical Neurophysiology, 131 (2020): 63-69
The goal of this research was developing a predictive model to differentiate acute inflammatory demyelinating polyneuropathy (AIDP) from non-demyelinating Guillain-Barré syndrome (GBS), based on NCSs performed at 1-20 days after onset and at 3-8 weeks. There were 90 patients, 40 with AIDP and 50 with non-AIDP GBS. Testing included median and ulnar motor and sensory, tibial motor, and sural sensory. There were significant differences between the two groups on almost all parameters, except for the sural nerve. A predictive model was constructed using just 3 parameters; median motor nerve conduction velocity, ulnar distal motor latency, and sural sparing (abnormal ulnar/normal sural). A 0-6 total point scoring system for these criteria yielded an AUC of 0.862 and 0.885 for the early and late NCS. The later NCS test had a positive predictor value for AIDP of 93% for a score of 2 or greater and 98% for 3 or greater.
This article outlines a simple and relatively straightforward way of determining long term prognosis in GBS, a common disorder. The amount of needed testing is rather modest. Accuracy is high. This appears to be a rather efficient way of evaluating these cases.