In June and July 2008, two office workers were admitted to a Dublin hospital with Legionnaires' disease. Investigations showed that cooling towers in the basement car park were the most likely source of infection. However, positive results from cooling tower samples by polymerase chain reaction (PCR) did not correlate with subsequent culture results. Also, many employees reported Pontiac fever-like morbidity following notification of the second case of Legionnaires' disease. In total, 54 employees attended their general practitioner or emergency department with symptoms of Legionnaires' disease or Pontiac fever. However, all laboratory tests for Legionnaires' disease or Pontiac fever were negative. In this investigation, email was used extensively for active case finding and provision of time information to employees and medical colleagues. We recommend clarification of the role of PCR in the diagnosis of legionellosis and also advocate for a specific laboratory test for the diagnosis of the milder form of legionellosis as in Pontiac fever.
Legionellosis presents as two distinct clinical entities: Legionnaires’ disease (LD) comprises pneumonia with severe multisystem disease and Pontiac fever (PF) which is a self-limiting flu-like illness [
Annually, on average, 6 cases of LD were reported to the Health Protection Surveillance Centre by the Departments of Public Health in Ireland between 1994 and 2007. This reported rate of two per million is low compared with other European countries, such as Spain and France where rates of 24.8 per million and 22.8 per million were reported, respectively, in 2007. In 2007, the overall European rate of reported infection was 11.4 cases per million [
LD is transmitted from the environment by inhalation of an infectious aerosol [
Over a ten-day period in June-July 2008, two cases of LD were reported to the Department of Public Health, Health Services Executive East (HSE-E) in Dublin. Both cases worked in a large Dublin-based insurance firm and were based in a newly built office block in the city. The office building consisted of seven floors over a basement car park. Two cooling towers were located in the basement. The staff smoking area was located immediately adjacent to the cooling towers. Air conditioning within the building was confined to the printing/postal room (3rd floor) and the conference/meeting rooms (6th and 7th floor). This paper describes the cases with LD, the outbreak investigation, and subsequent management. A retrospective cohort study of cases with PF will be reported separately [
The first case of LD was a 58-year-old Southern-European male smoker who worked in the postal department on the 3rd floor of the company office, and became unwell in early June. He was admitted on June 25th to a Dublin hospital with a 3-week history of fever, night sweats, chills, diarrhoea, and confusion. His chest X-ray showed right upper and lower lobe consolidation. Liver function tests and renal indices were elevated. Diagnosis of LD was confirmed by
On July 7th, a second case of LD (case
On July 8th, it was reported to the Department of Public Health HSE-E by company management that many employees were absent from work and had reported severe flu-like illness. Consequently, active case finding was carried out on the morning of July 8th in the company. A standardised questionnaire was emailed to all company and contract employees who had worked in the building at any time during the previous 14 days. Information was requested on symptoms (cough, headache, fever, muscle pains, diarrhoea, and flu-like symptoms), potential exposure (office floor, air-conditioning, and shower use), smoking, travel, and demographics (Figure
Questionnaire for office workers distributed by email on July 8th 2008.
We defined a case of LD as an employee of, or visitor to, the company office who had pneumonia with onset since June 1st, and
We defined a case of PF in the initial investigation as an employee of, or visitor to, the company office with symptoms of fever, headache, myalgia, and nonproductive cough with onset since June 1st.
Sickness records were reviewed by Human Resources staff and absent workers were contacted to inform them of the outbreak and to enquire about their health and whether they had sought medical attention. Respondents who met our case definition for LD or PF were promptly telephoned by Department of Public Health medical or nursing staff and advised on appropriate medical follow-up.
Employees who attended the emergency department (ED) or their general practitioner (GP) were followed up to determine their symptomatology, investigation, results, and treatment. Subsequently, a designated clinic was arranged for all employees and contract workers who (a) met our case definition for PF or (b) had attended the ED or their GP with concerns regarding LD. All were invited for
In total, 54 employees attended EDs in the Republic of Ireland and Northern Ireland or their general practitioner with symptoms and/or concerns about LD or PF; each had a UAT, all of which were reported negative. Of these at least 38 were reported as having been treated with an antibiotic. Subsequently, 37 of these employees with possible symptoms of PF attended a special clinic for
On July 8th, Public Health doctors met on site with the company’s Health and Safety manager, Risk Assessment manager, and Human Resources manager. Public Health doctors, Environmental Health officers of the Health Services Executive, and Environmental Health officers of Dublin City Council conducted an inspection and environmental risk assessment of the building. On July 9th, the Health, and Safety Authority, the organisation responsible for ensuring safety, health and welfare at work, inspected the building. There were two separate water-containing systems supplying the office. The primary system consisted of two cooling towers located in the large basement carpark of the building. Both cooling towers were immediately adjacent to the staff smoking area. This system was used to chill incoming water. The secondary system was a hot and cold water system used to service sinks and showers. There was no physical connection between the primary and secondary systems. Both systems were supplied with mains water. Managers reported the use of an alarmed temperature control system for the water supply to the building. There was no fountain in the building.
As a precautionary measure, after case
Microbiological results from cooling towers 1 and 2 by PCR and culture, Dublin, July 2008.
Cooling Tower | PCR (GU/L) | Culture (CFU/L) | |||
1 | 61,800,000 | 73,200,000 | < Detection level | < Detection level | < Detection level |
2 | 174,000 | <3,300 | 100 | 100 | < Detection level |
Both cooling towers were immediately deactivated, cleaned, and disinfected as recommended [
Both patients admitted to hospital were confirmed cases of LD by
After the first case of LD was notified, an email alert was sent to 600 company employees informing them of the symptoms of LD and advising them to seek medical attention from their general practitioner if they had any concerns regarding illness. This alert was repeated to all employees three days later. In addition, every line manager endeavoured to contact their employees to confirm they had received relevant public health information of LD and were not symptomatic. Similar efforts were also made to check on contract employees and visitors to the company since June 1st. Following reporting of the second case of LD, an alert was sent to all general practitioners and ED consultants in the region by fax and email. In addition, due to the high level of concern among the employees in the company, two educational/information sessions on LD were held in the office by medical doctors.
This paper describes a cluster of LD among office workers in Dublin. The only common link among the two cases was their place of employment. Consequently, the most likely source of this cluster was the cooling towers in the basement car park. Definitive microbiological confirmation of the cooling towers as the source of infection required an isolate from one of the patients to match the environmental isolate. As
Early identification and treatment of patients with LD have important implications for clinical management of patients and outcome [
Few studies have addressed the practical value of PCR in routine clinical and environmental microbiology laboratories. PCR offers theoretical advantages, such as a high sensitivity, rapid availability of results, and the potential to detect infections caused by various serogroups of
The affected premises in this cluster ware a new purpose built “green building” which was occupied for three months prior to the reported onset of illness. A green building is defined as “the practice of increasing the efficiency with which buildings use resources—energy, water, and material—while reducing the building’s impact on human health and the environment during the building’s lifecycle, through better siting, design construction, operation, maintenance, and removal” [
The nonspecific presentation of LD makes clinical diagnosis very difficult in most patients with community-acquired pneumonia of uncertain aetiology. Thus, the key to diagnosis is appropriate microbiological testing [
This investigation demonstrated that timely intervention is vital in control of a cluster of legionellosis. In order to control, and ideally prevent, future clusters of legionellosis we recommend the statutory notification of high risk sites and also improved diagnostic tests.
The authors gratefully acknowledge the advice and assistance received from Dr. Tim Harrison Respiratory & Systemic Infection Laboratory HPA Centre for Infections, Colindale, UK.