Intended for healthcare professionals


New outbreak of legionnaires' disease in the United Kingdom

BMJ 2002; 325 doi: (Published 17 August 2002) Cite this as: BMJ 2002;325:347

A wake up call for continued vigilance

  1. Carol Joseph, consultant public health scientist (cjoseph{at}
  1. Public Health Laboratory Service Communicable Disease Surveillance Centre, London NW9 5EQ

    In the past 25 years an extensive knowledge base has been established about legionnaires' disease, and the measures needed for control and prevention from environmental sources of infection arenow legally enforced in most industrialised countries.1 Yet in spite of these advances in public health knowledge and practice, four major community outbreaks around the world have occurred in the past three years, the most recent in England.2 This outbreak and those in the Netherlands 3 and Australia4 are the largest ever in their countries, and the one in Spain last year ranks as the world's largest.5 These four outbreaks alone account for almost 1000 cases of pneumonia and around 40 deaths. They are an uncomfortable but timely reminder of how breakdowns in control or maintenance procedures can have catastrophic consequences for many people. They also raise interesting questions—namely, why were they so large in comparison with other outbreaks, why did fewer deaths occur, and what do the results of their investigation contribute to our knowledge of legionella infections?

    Legionnaires' disease is usually seen as a serious form of pneumonia that carries with it a fatality rate in the order of 10-15% in otherwise healthy individuals. It is a rare disease and estimated to cause 2-16% of community acquired pneumonia in industrialised countries.6 It stems from environmental sources linked to industrial, commercial, hospital, or domestic settings, and infection occurs when people breathe in air that contains the bacteria in an aerosol. Illness normally begins 2-10 days after inhalation of the bacteria and includes flu-like symptoms, fever, headache, dry cough, and often progresses to pneumonia. Risk factors forthe disease include being of an older age group, male, and a heavy smoker; and having an underlying disease associated with immunodeficiency. The disease is not transmitted from person to person, and microbiological evidence of infection is necessary to establish the diagnosis. Outcome of illness will depend on early treatment with appropriate antibiotics as well as a patient's individual susceptibility.

    Cooling towers can give rise to very large outbreaks because of their ability to disperse contaminated aerosols over large distances (up to 500 metres).7 Environmental and biological factors that affect the size and magnitude of an outbreak are climate(factors such as ambient temperature, direction of wind, or humidity), location of cooling towers in relation to proximity of the population and individual susceptibility of the persons exposed tothe bacteria. Small community outbreaks occur on a regular basis. In Europe, data from the European Working Group for Legionella Infections (EWGLI) shows that 51 such outbreaks were reported between 1997 and 2001, accounting for 20-525 outbreak related cases per year.8

    What is of interest is why the recent outbreaks have involved so many cases but fewer deaths compared with outbreaks 10-20 years ago, given that environmental causes of outbreaks remain unchanged. Four reasons come to mind. Firstly, as large community outbreaks become rarer, publicity aboutthe disease declines and managers of cooling towers or other systems may not fully appreciate the implications of insufficient or poor maintenance procedures. Secondly, clinicians have a greater awareness of legionella infection during an outbreak. Thirdly, enhanced surveillance has been introduced in many countries and led to better reporting of cases. Fourthly, the use of the urinary antigen detection test to confirm the diagnosis is now almost universally applied. This diagnostic method, which was not available when the big outbreaks of the 1980s occurred, has enabled rapid confirmation of legionella infection soon after onset of illness and through appropriate clinical management of these patients reduced the number of associated deaths. The test is easy to administer, so a larger pool of potential cases may be tested. In turn, as more people with symptoms are tested, a wider spectrum of infection has been shown than was previously recognised. A significant proportion of non-pneumonic legionella infections has emerged in some of these outbreaks.

    The low mortality in large outbreaks is not mirrored in sporadic cases. Mortality associated with these continues to remain at 10-15% overall, perhaps because of underdiagnosis and under-reporting of sporadic cases, with only the very severely ill patients being detected and reported. By far the largest numbers of cases in many national datasets are reported as sporadic cases.

    The recent major outbreaks show the extent to which many cases associated with other exposures are missed, and how routine surveillance is affected by the “tip of the iceberg” syndrome. Enhanced detection and improved surveillance is called for to reconcile information from sporadic and outbreak cases, along with greater vigilance of control and prevention measures to minimise the risk of infection in people. Together, these should help to put the results from recent major outbreaks into perspective. The wake up call must provide this impetus.


    View Abstract