Epidemiological and public health studies on Guillain-Barré syndrome in Sweden

Abstract

The Guillain-Barré syndrome (GBS) is an acute or subacute peripheral neuropathy affecting motor, sensory and autonomic nerves as well as spinal roots. Although the pathogenesis of GBS remains incompletely known, it is widely accepted that GBS is an autoimmune disease triggered by a preceding bacterial or viral infection. The targets of the humoral and cellular immune reactions can be epitopes in myelin or the axonal membrane. GBS as a clinical entity is characterised by acute progressive motor weakness and tendon areflexia, with or without sensory signs and autonomic dysfunction. During recent decades, epidemiological studies from different populations have reported similar GBS incidence rates of around I to 2 per 100,000 person-years. Outbreaks of GBS have been related to vaccinations or infections.

In order to investigate the need for public health surveillance (PHS) of GBS in Sweden and to establish the corresponding basis, historical GBS data from the national hospital inpatient registry were analysed by performing time-series analyses with autoregressive integrated moving average (ARIMA) models. When the forecasts of GBS incidence for 1993 were used for examination of the retrospective data, significant variations of GBS incidences were found for certain time periods. The results suggested that monthly or bimonthly forecasts of GBS incidences obtained from the models could have detected the increases of GBS incidence of unknown cause and the zimeldine-induced GBS epidemic in 1978 and 1983, respectively.

We set up a multi-centre network of 18 neurologists in Sweden, and carried out PHS of GBS across the country during the period 1996-1997, in a general population with around 4.5 million inhabitants (51 % of the whole nation). While 117 incident GBS patients were identified, no alarm signals were found during the period. Threshold values for monthly GBS incidences in the general and selected populations were obtained.

The incidence of GBS in 1996 in the surveillance population was described by using information from patients prospectively observed and data from the national hospital inpatient registry. GBS diagnoses for all patients were validated by neurologists. The GBS incidence, age-adjusted to the European standard population, was 1.51 (95% CI 1.18-1.90) per 100,000 person-years in 1996, was higher in males and increased with age.

Clinical manifestations, results of laboratory analysis, outcome, and prognostic indicators were analysed based on the information from 53 GBS patients, who were prospectively followed up. At one year after onset, 46% of the patients were fully recovered, 42% had only minor residual signs or symptoms, 4% had moderate and 6% severe disabilities, and one patient (2%) had died. Sum Medical Research Council (MRC) score at nadir was significantly associated with residual signs at one year. Three clinico-epidemiological subgroups of GBS patients were identified by using cluster analysis. In two of the subgroups, GBS onset was preceded mainly by respiratory infections; encompassed 18% and 47% of cases and differed with regard to age, severity and outcome. A third subgroup was comprised mostly of elderly patients with disease of moderate severity and frequent residual signs.

The relationship between GBS and pregnancy or delivery was investigated by linking registered data for Swedish females aged 15-49 years during the period 1978-1993. Poisson regression analysis yielded age-adjusted rate ratios of 2.93 (95% CI 1.20-7.11) for GBS incidence during the first 30 days after delivery and 0.89 (95% CI 0.52-1.53) for pregnant women, when compared with that in females neither pregnant nor in the 90-day postpartum period.

In conclusion: 1) PHS of GBS in Sweden is practical and feasible with a network of neurologists monitoring incident GBS patients. 2) The incidence of GBS in Sweden is in line with those reported from other populations. 3) In general, GBS in Sweden is frequently preceded by a respiratory infection, is often treated with immunomodulatory therapies and exhibits a high recovery rate and a low fatality rate. 4) Severe motor disability at nadir is associated with residual signs at one year after onset. 5) Three subgroups of GBS patients with different clinico-epidemiological characteristics can be identified. 6) There is an increased risk for GBS onset in the immediate postpartum period, but not in successive months or earlier during gestation.