Modified Erasmus Guillain-Barré Syndrome Outcome Score May Predict Inability to Walk

Researchers sought to validate the modified Erasmus GBS Outcome Score, a clinical model that predicts the inability to walk in patients with GBS, in the International GBS Outcome Study cohort, observe any regional variations in performance, and improve predictions by adjusting accordingly.

The modified Erasmus Guillain-Barré syndrome (GBS) Outcome Score (mEGOS) can be used in its original form in patients with GBS to predict the risk of poor outcomes, according to new research from Neurology. The researchers created a variation for predicting these outcomes in patients from European countries and North America.

GBS management is challenged by regional variations in clinical course and outcome, as the researchers of the current study previously found when they analyzed 1000 patients in the ongoing prospective International GBS Outcome Study (IGOS). The mEGOS forecasts based on age, muscle strength, and preceding diarrhea which patients may lose the ability to walk independently within six months, but it has only been validated in a primarily Dutch cohort and two Asian cohorts.

The objective of the current study was to validate the mEGOS in the global cohort, observe any regional variations in performance, and improve predictions by adjusting accordingly.

The researchers analyzed data of the first 1500 patients in the IGOS who enrolled from 2012 to 2017 in 19 countries, including the United States and Canada. They included patients with GBS who lost the ability to walk (GBS disability score greater than 2) at entry and at day 7. Exclusion factors included alternative diagnosis, protocol violation, insufficient data, missing age, age under 6 years, and ability to walk independently. Patients with Miller Fisher syndrome (MFS) were included in the IGOS validation cohort

The validation cohort at entry (n=809) included 677 patients from European or North American countries. The cohort at week 1 was 671 patients, including 563 individuals from European or North American countries.

Patients in the validation cohorts tended to be older and have mild muscle weakness (MRC sum score 51-60) compared with patients in the Danish-centric development cohort, which did not include individuals with MFS.

The area under the ROC curve (AUC)-values for the full cohort and European/North American (Eu/NA) subgroup were lower compared with AUC-values in the development cohort. Researchers found that the mEGOS model and calibration plots tended to be more conservative in risk prediction compared with observed frequencies in the full cohort and the Eu/NA subgroup but less conservative compared with observed frequencies in Asia. The differences were greater between observations and predictions at 4 weeks compared with at 6 months. After the researchers recalibrated the original model to address the differences, they found predictions improved.

The researchers said differences between observed and predicted risks and performance of mEGOS between continents may be due to regional variations or differences in treatment and health care facilities.

They recommended applying the mEGOS-Eu/NA model at hospital admission or day 7 of admission to patients from Europe or North America diagnosed with GBS or its variants who cannot walk independently.

Study limitations include missing GBS disability scores, applicability only to patients who can no longer walk, and other generalization challenges.

“The mEGOS is a validated tool to predict the inability to walk unaided at 4 and 26 weeks in GBS patients, also in countries outside The Netherlands,” the researchers concluded.

Disclosure: Some study authors declared affiliations with biotech, pharmaceutical, and/or device companies. This research was supported by CSL Behring, Grifols, Annexon and Hansa Biopharma. Please see the original reference for a full list of disclosures.


Doets AY, Lingsma HF, Walgaard C, et al. Predicting outcome in Guillain-Barré Syndrome: international validation of the modified Erasmus GBS outcome score. Neurol. Published online December 22, 2021. doi:10.1212/WNL.0000000000013139