Clinical characteristics of Guillain-Barré syndrome combined with hyponatremia in Southern China and its risk factors for prognosis: a multi-center retrospective analysis / 中华神经医学杂志

ABSTRACT

Objective:To investigate the clinical characteristics of Guillain-Barré syndrome (GBS) combined with hyponatremia in Southern China and its risk factors for prognosis.Methods:The retrospective cohort study involved patients who met the diagnostic criteria of GBS from 18 upper first-class hospitals of 6 provinces/cities in southern China (south of Huaihe River) from January 1, 2013 to September 30, 2016. The clinical data of these patients were collected. According to serum sodium levels, they were divided into hyponatremia group (serum sodium concentration<135 mmol/L) and normal serum sodium group (serum sodium concentrations≥135 mmol/L). Based on Medical Research Coucil sum scores at nadir, these patients were divided into mild GBS group (>40), moderate GBS group (30-40), and severe GBS group (<30). Furthermore, according to the Hughes GBS disability scale (H-GBS-DS) scores at discharge, these GBS patients with hyponatremia were divided into favorable prognosis group (H-GBS-DS<3) and poor prognosis group (H-GBS-DS≥3). The incidence of hyponatremia in patients from the mild GBS group, moderate GBS group, and severe GBS group were compared. Multivariate Logistic regression analysis was performed to determine the clinical risk factors for hyponatremia in GBS patients. The clinical data of hyponatremia patients from favorable prognosis group and poor prognosis group were compared; multivariate Logistic regression analysis was used to determine the risk factors for poor prognosis in GBS patients with hyponatremia.Results:(1) Among the 570 patients, 354 had mild GBS, 94 had moderate GBS, and 122 had severe GBS; 134 GBS patients were combined with hyponatremia, 436 GBS patients had normal serum sodium. The hyponatremia incidence in mild, moderate and severe GBS groups increased successively, ( P<0.05). Multivariate Logistic regression analysis showed that facial paralysis ( OR=1.979, 95%CI: 1.172-3.342, P=0.011), respiratory muscle paralysis ( OR=3.218, 95%CI: 1.611-6.428, P=0.001), secondary pulmonary infection ( OR=4.822, 95%CI: 2.835-8.201, P=0.000), severe GBS ( OR=2.611, 95%CI: 1.444-4.721, P=0.001) and length of hospital stay ( OR=1.029, 95%CI: 1.009-1.050, P=0.004) were risk factors for hyponatremia in GBS patients. (2) Among 134 GBS patients with hyponatremia, 80 had poor prognosis and 54 had favorable prognosis. As compared with the favorable group, the poor prognosis group had significantly lower proportion of patients with extraocular muscle paralysis, statistically higher proportions of patients with respiratory muscle paralysis and secondary pulmonary infection, significantly different severities of GBS, signficantly higher proportion of patients accepted intravenous immunoglobulin (IVIG) and hormone treatments, statistically longer length of hospital stay ( P<0.05). Respiratory muscle paralysis ( OR=25.590, 95%CI: 9.433-69.423, P=0.000), moderate GBS ( OR=17.030, 95%CI: 8.441-34.361, P=0.000), and severe GBS ( OR=51.042, 95%CI: 24.596-105.926, P=0.000) were independent risk factors for poor short-term prognosis of GBS patients with hyponatremia. Conclusions:Severe GBS patients with facial paralysis, respiratory muscle palsy, secondary pulmonary infection, and long hospital stay trend to have hyponatremia. Hyponatremia patients with respiratory muscle paralysis and moderate/severe GBS have poor short-term prognosis.