Sensitivity and specificity of phrenic nerve electrophysiology to predict mechanical ventilation in the <scp>Guillain‐Barré</scp> syndrome

Abstract

INTRODUCTION/AIMS: In Guillain-Barré syndrome (GBS), the sensitivity and specificity of phrenic compound muscle action potential (CMAP) measurements to predict endotracheal mechanical ventilation are unknown. Hence, we sought to estimate sensitivity and specificity. METHODS: We performed a 10-year retrospective analysis of adult GBS patients from our single-center laboratory database (2009 to 2019). The phrenic nerve amplitudes and latencies before ventilation were recorded along with other clinical and demographic features. Receiver operating curve (ROC) analysis with area under the curve (AUC) was used to determine the sensitivity and specificity with 95% confidence interval (CI) for phrenic amplitudes and latencies in predicting the need for mechanical ventilation. RESULTS: Two hundred five phrenic nerves were analyzed in 105 patients. The mean age was 46.1 ± 16.2 years, with 60% of them being male. Fourteen patients (13.3%) required mechanical ventilation. The average phrenic amplitudes were lower in the ventilated group (P = .003), but average latencies did not differ (P = .133). ROC analysis confirmed that phrenic amplitudes could predict respiratory failure (AUC = 0.76; 95% CI, 0.61 to 0.91; P < .002), but phrenic latencies could not (AUC = 0.60; 95% CI, 0.46 to 0.73; P = .256). The best threshold for amplitude was ≥0.6 mV, with sensitivity, specificity, and positive and negative predictive values of 85.7%, 58.2%, 24.0%, and 96.4%, respectively. DISCUSSION: Our study suggests that phrenic CMAP amplitudes can predict the need for mechanical ventilation in GBS. In contrast, phrenic CMAP latencies are not reliable. The high negative predictive value of phrenic CMAP amplitudes ≥0.6 mV can preclude mechanical ventilation, making these a useful adjunct to clinical decision-making.