Temporal variation in Guillain-Barré syndrome (GBS) warrants monitoring in certain situations. This study sought to describe a public-health-based GBS surveillance service in Spain and conduct pilot surveillance in the period 1998-1999. Neurologists from 11 hospitals countrywide, serving a population of 3.9 million, reported all patients, ages 20 years or over, admitted to hospital with suspected GBS. Cases that did not belong to the designated hospital catchment area or failed to fulfill diagnostic criteria after follow- up were excluded. Reported monthly incidence was compared against predicted incidence obtained from retrospective data (1985-1997) using a reported method based on 97.5% percentile values. Alarm thresholds for 2000 onwards were obtained by applying the same method to the updated 1985-1999 series. During the 2-year period, 98 GBS cases were reported, yielding an overall age-adjusted incidence of 1.26 per 100 000 population, with a breakdown by sex of 1.83 for males and 0.76 for females. Monthly incidence remained below or was similar to the corresponding threshold limit value. Seasonality with highest incidence in winter was more pronounced in the elderly. Preceding events, mainly respiratory infections, were identified in 71% of patients. Pilot two-year GBS surveillance in Spain resulted neither in alarm nor in preventive measures. Adult GBS incidence in Spain might be monitored by a surveillance system set up at short notice when a possible threat is perceived. A monthly incidence of over 3 per 100 000 person-years in the population aged 20 years or older would exceed threshold values.