Een legionella -epidemie onder bezoekers van een beurs in Bovenkarspel 1. Beschrijving van de epidemie 2. Patient-controle onderzoek naar de bron

In March 1999 an unexpected high number of patients with atypical pneumonia was hospitalised in Hoorn. The diagnosis legionellosis was confirmed by the urine antigen test. An exploratory case control study suggested that the source should be found at the Westfriesian Flora (WF, a yearly exhibition organised from February 19 to 28 at Bovenkarspel with about 80,000 visitors). Subsequently, a national alert was sent out through the media, a case register was established and the outbreak investigation was started. This report describes the outbreak using the data in the case-register, as well as the results of the case-control study. Separate reports are published on the environmental investigation and the cohort study.The case-register includes 318 persons among whom 133 confirmed cases and 55 probable cases. The median age was 66 years (range 20-91 years), the male:female ratio 1.4. The day of onset was between February 25 and March16; the longest incubation period was 19 days, considerably longer than reported in literature. Except for one patient with COPD and recurrent pneumonia, all cases -regardless of earlier visits- visited the Flora on February 23 or later. The attack-rate per 100,000 visitors according to day of visit increased from 1.1 (February 21) to 53.4 (February 27). The highest number of hospital admissions was on March 12, the day of the national alert. The hospitalisation frequency was 86.7%; 19.1% of the patients required artificial respiration. Twenty-nine persons who were included in the case-register and had visited the Flora died, among whom 17 confirmed and four probable cases. The case-fatality rate for confirmed and probable cases amounted to 0,11.The findings strongly indicate that one or more sources of infection were contaminated with Legionella spp on or before February 21; subsequently multiplication of the agent in these sources led to increasing concentrations of Legionella spp in the exhibition halls. Three variables were included in the final models of all three measurements (questionnaire, sketches and floorplan) in the case-control study: the number of hours visiting the WF (OR 1.6, 2.2 and 1.8 respectively), smoking (OR 3.8, 6.0 and 10.9 respectively) and the number of hours visiting the consumer products fair with equal duration of the total visit (OR 0.7, 0.5 and 0.5 respectively). Age was included in the final model of questionnaire data (OR 1.1) and of floorplan data (OR 1.2). Halting at the whirlpool in hall 3 was an explanatory variable in the model of the sketch data (OR 2.6) as well as the floorplan data (OR 5.8). The variable 'whirlpool' was also associated with legionellosis in the analysis of questionnaire data. However, no distinction was made between the two whirlpools and the two bubblemats.Both from these results and from the results of the environmental investigation and the cohort study, it was concluded with reasonable certainty that the whirlpool in hall 3 was the source of the outbreak.