Simple bedside predictors of mechanical ventilation in patients with Guillain-Barre syndrome

Objective: The objective of the study is to develop and validate a predictor score for assessing the requirement of mechanical ventilation (MV) in patients with Guillain-Barre syndrome (GBS).

Study design: The study was conducted in patients admitted with GBS in neurointensive care unit in a tertiary care hospital. The demographic, clinical factors, electrophysiological, and spirometric data of all consecutive patients were prospectively collected. The study was undertaken in 2 stages. In the first stage, data were collected for development of a predictor score. In the second stage, the score developed was validated on a separate set of patient data.

Results: The data collected were compared between the 2 groups (ventilated vs nonventilated). On univariate analysis, time taken to reach maximum deficit, neck weakness, bulbar weakness, facial weakness, single breath count (SBC), forced vital capacity, and phrenic nerve latency predicted the need for MV. On multivariate analysis, only neck weakness, bulbar weakness, SBC, and forced vital capacity were independent predictors of MV. There was a good correlation between SBC and the spirometric tests and phrenic nerve distal motor latency, as reflected in receiver operating characteristics curve. The predictor score developed using the regression coefficient of independent predictors showed that the best cutoff score for prediction of ventilation was 60 (sensitivity, 0.95; 1--specificity, 0.065). Internal cross validation of the neck weakness, SBC, and bulbar palsy (NSB) score showed good correlation (Pearson R = 0.76; P = .00). There was no statistically significant difference between predicted and observed outcomes (sensitivity, 95%; specificity, 93%).

Conclusion: Several independent risk factors were found to predict the requirement for MV in patients with GBS at admission. However, after scoring and analyzing them, it was found that combining a few of them was more useful to predict the need for MV. A model using NSB score, developed using clinical variables, accurately predicted the requirement of MV. In addition, among the NSB score parameters, simple bedside SBC could adequately assess the adequacy of vital capacity.