Optimizing the electrodiagnostic accuracy in Guillain-Barré syndrome subtypes: Criteria sets and sparse linear discriminant analysis.

OBJECTIVE

To optimize the electrodiagnosis of Guillain-Barré syndrome (GBS) subtypes at first study.

METHODS

The reference electrodiagnosis was obtained in 53 demyelinating and 45 axonal GBS patients on the basis of two serial studies and results of anti-ganglioside antibodies assay. We retrospectively employed sparse linear discriminant analysis (LDA), two existing electrodiagnostic criteria sets (Hadden et al., 1998; Rajabally et al., 2015) and one we propose that additionally evaluates duration of motor responses, sural sparing pattern and defines reversible conduction failure (RCF) in motor and sensory nerves at second study.

RESULTS

At first study the misclassification error rates, compared to reference diagnoses, were: 15.3% for sparse LDA, 30% for our criteria, 45% for Rajabally's and 48% for Hadden's. Sparse LDA identified seven most powerful electrophysiological variables differentiating demyelinating and axonal subtypes and assigned to each patient the diagnostic probability of belonging to either subtype. At second study 46.6% of axonal GBS patients showed RCF in two motor and 8.8% in two sensory nerves.

CONCLUSIONS

Based on a single study, sparse LDA showed the highest diagnostic accuracy. RCF is present in a considerable percentage of axonal patients.

SIGNIFICANCE: Sparse LDA, a supervised statistical method of classification, should be introduced in the electrodiagnostic practice.