[Prevention of Legionella pneumonia in the Netherlands: results from the Legionella Source Identification Unit, 2002-2012].

Objective: To study the effectiveness of a Legionella pneumonia (LP) prevention programme.

Design: Observational study.

Method: We evaluated the effectiveness of the current LP prevention programme using two outcome measures, genotype match and cluster, for the period 2002-2012. If patients were associated with a source of infection via a matching or as part of a cluster it could be assumed that prevention of LP was achieved by implementing control measures for this source. By comparing genotypes we were given an indirect impression of the validity of the sampling process.

Results: Legionella pneumophila serogroup 1 was detected in 97 (7%) of the 1484 sampled sources. A likely source of infection was identified for 41 (2%) of the 1991 LP patients, and confirmed by matching. In more than half of these patients, the source was either a residential house or a hospital. Of the 1991 LP patients, 266 (13%) were part of a cluster. Two L. pneumophila serogroup 1 genotypes, ST47 and ST62, were present in 48% of the LP patients, but these genotypes were seldom detected in source sampling (0.9%).

Conclusion: The current method of source detection does not adequately contribute to the prevention of LP, because the presence of L. pneumophila serogroup 1 is not often detected in the source. Other sources than those currently known are probably involved in the transmission of these bacteria. Serial infection via a common source is a substantial cause of LP, which emphasises the importance of cluster registration. It is important to identify as yet unknown alternative infection sources.