THE EARLIEST recognized hospital-acquired outbreak of Legionella bacteria was diagnosed in 1965. (1) The role of Legionella-contaminated potable water distribution systems as a source of nosocomial and community-acquired Legionnaires' disease has since been well established. (2) In an outbreak of legionellosis in the Jesenice General Hospital, Jesenice, Slovenia, in 1991, more healthy personnel acquired the disease than did bed-ridden patients. This discovery, which was in contrast to previous findings, precipitated a study on the extent and the mode of transmission in the outbreak.
Materials and Method
Epidemiological design. This study was conducted as a cross-sectional survey design. The epidemiological data comprised information on the technical characteristics of the water supply, including boiler temperature. We obtained information from patients regarding whether they were able to move from their beds, and from staff on their daily tasks and the location of their duty. We diagnosed Legionella infection using standard protocols. We also investigated the distance from, and the duration of, daily exposure to water from the suspected water source of Legionella.
Technical characteristics. A schematic plan of the hospital wards, the location of Legionella cases, and the water supply system of the Jesenice General Hospital are shown in Figure 1.
[FIGURE 1 OMITTED]
There were no air conditioners in the wards. The ventilation system had exhaust fans, and the windows were closed most of the time. The water supply consisted of two systems. An old system with little water circulation and a dead-end pipe supplied the intensive care and coronary units. The other wards (medical and surgical) were supplied by the new water system with circulating water. The water systems had a common heating source.
In the old system, two water faucets each were located in the intensive care and the coronary units (60- and 40-cm-high jets, respectively). Two additional faucets were located in the adjacent sanitary units. No one drank this water; it was used only for washing and cleaning by nurses and by some ambulatory patients for their personal hygiene. It took about 2 min before water from this system became warm. Early in the morning, even longer waits were necessary. During this time, the tap would be open to full capacity and droplets scattered widely.
To estimate the range of spray around the washstands, we measured relative humidity at different distances using a hygrometer (Fisher art. No. 115, Fisher Scientific GmgH, Schwerte, Germany). We measured the humidity and temperature around all four water faucets in early morning and later in the day. For comparison, we performed similar measurements in different parts of the hospital with the new water system. We took measurements close to the water jet at 1, 2, 3, 4, and, in certain rooms, up to 5 m from the water faucets. We took the measurements before the tap was turned on, for 2-min while the water was running, and just before turning off the tap. We repeated this process with measurements taken at 2-min intervals for 10 min.
Study population. We studied hospital personnel and patients hospitalized during the epidemic (from May to July 1991). Subjects were divided into three groups, according to hospital ward location (see Fig. 1).
1. Medical ward staff who shared tasks in all medical wards (60 women and 11 men, mean age = 33.9 years).
2. Medical ward patients (49 women and 62. men, mean age = 62.1 years), 37 of whom were located in the intensive care or the coronary units.
3. Surgical ward staff (17 women, 4 men, mean age = 37.8 years) who worked in surgical but not medical wards.
Criteria for diagnosis of legionellosis. We defined legionellosis according to the criteria proposed by the U.S. Centers for Disease Control and Prevention. (3,4) We considered a person to have legionellosis in case of febrile illness and demonstration of Legionella infection by at least one of the following tests: (a) isolation of Legionella from a clinical specimen (sputum, bronchoalveolar aspirate, transbronchial biopsy, pleural fluid); (b) demonstration of at least 4-fold elevation of Legionella serum antibody titers to 1:64, or a titer = 1:128 (without 4-fold change), established by indirect immunofluorescence assay; (c) a positive direct fluorescence assay (DFA) result in a clinical specimen; or (d) demonstration of the presence of Legionella in a clinical specimen by radiolabelled DNA probe (GEN-PROBE, Gene Probe Inc., San Diego, CA).
We performed serologic testing at the beginning of the study, 2-6 wk later, and in individual cases after 10-15 wk, in all medical personnel and inpatients in the medical wards. We performed other tests on persons with clinical features indicating a possibility of legionellosis.
Results
The temperature in the central boiler was usually 60-70[degrees]C, except during the 2 wk prior to the outbreak, when it was 50-55[degrees]C. At that time, the water was also discolored. Water supplied by the new water system was 52 and 64[degrees]C when the faucet was turned on and after it ran for 2 min, respectively. Water from the old system was 21-26[degrees]C when turned on and 42-51[degrees]C after running 2 min. Temperature in the old water supply never exceeded 51 [degrees]C. We isolated Legionella pneumophila serogroup 1 from specimens of faucet water from the coronary unit obtained early in the epidemic, twice in culture and once by DFA, indicating that the old water system was the source of infection. With the exception of night hours, this water was used frequently.
We detected legionellosis cases among medical, but not among surgical, hospital staff, whereas among the patients, only those from the coronary and intensive care units developed the disease. Seven of 111 (6.3%) patients hospitalized in medical wards and 31 of 71 (43.7%) medical ward personnel acquired legionellosis during the outbreak. All 7 patients with legionellosis were from the intensive or coronary care units. These were the only patients from these units who were mobile and were able to use water from the contaminated faucets for their personal hygiene.
Table 1 shows the minimum distance from the water faucet of patients and personnel who spent at least 30 min per day near the faucets. All cases appeared among patients and staff who spent at least 30 min a day at 1-2 m from the faucet.
Table 2 shows the results from measurements of air humidity around the faucets in the intensive care unit. We found similar results in the coronary unit.
The humidity and temperature field around the faucets in the washstand area reached a quasi-steady state with strong gradients of both measured variables within 2 min, decreasing with the distance from the faucet. At a distance of 4 m or more, there was no increase in humidity. After the faucet was turned off, the humidity and temperature slowly returned to background values.
Discussion
This outbreak of legionellosis was classic because it originated in a contaminated hot water source that was not sufficiently heated. This environment, along with stagnant water in the old supply system, was favorable to Legionella growth. The epidemiological data showed that ingestion, (4) aspiration, (5,6) or human-to-human transmission could be ruled out. Patients with nasogastric tubes did not acquire Legionella infection.
Compared with previous reports, the pattern of morbidity was unusual. (2,7-10) In this outbreak, the majority of cases were healthy personnel, whereas previous observations found that healthy personnel acquired legionellosis less frequently than did bed-ridden patients. (11,12) The pattern of morbidity together with the risk in relation to distance to the source (Table 2) suggest that the exposure dose, rather than predisposing factors, was the critical risk factor.
The exposure pattern was a spread from the water faucets, but there was no increase in humidity at 4 m or farther from the faucets. This corresponds to the distance at which we found no increased risk for staff and patients. Airborne Legionella remains viable for 2 hr at 80% relative humidity, but for less than 3 min at 30% relative humidity; only 0.1% of organisms survive 5 min. (13,14) The results from this study suggest that conditions for survival of Legionella in the air were not present at more than 4 m from the faucets, and the aerosol arriving at the beds of the patients in the wards would be considerably less infectious than closer to the source. This is in contrast to a recent report on a severe acute respiratory syndrome (SARS) virus outbreak in Hong Kong, for which indications of an infectious airborne transmission were found up to 60 m away. (15) The possible source of the SARS virus in that epidemic was agitated sewage water. This suggests that the SARS virus is more resistant to desiccation than Legionella, or that the humidity in the Hong Kong air was above the level critical for viability of the SARS virus.
Conclusion
The humidity level relative to distance from the contaminated water source proved to be the most important risk factor for acquiring legionellosis in this outbreak. High relative humidity was present close to the faucets with contaminated water, causing a high morbidity among hospital staff as well as patients.
References
(1.) Thacker SB, Bennet JV, Tsai TF, et al. An outbreak in 1965 of severe respiratory illness caused by the Legionnaires' disease bacterium. J Infect Dis 1978; 138:512-9.
(2.) Joseph CA, Watson JM, Harrison TG, et al. Nosocomial Legionnaires' disease in England and Wales, 1980-92. Epidemiol Infect 1994; 112:329-45.
(3.) Anon. Epidemiology, prevention and control of legionellosis: memorandum from a WHO meeting. Bull WHO 1990; 68:155-64.
(4.) U.S. Centers for Disease Control and Prevention. Guidelines for prevention of nosocomial pneumonia. Morb Mort Wkly Rep 1997; 46:28-79.
(5.) Muder RR, Yu VL, Fang GD. Community-acquired Legionnaires' disease. Semin Respir Infect 1989; 4:32-9.
(6.) Blatt SP, Parkinson MD, Pace E, et al. Nosocomial Legionnaires' disease: aspiration as a primary mode of disease acquisition. Am J Med 1993; 95:16-22.
(7.) Stout JE, Yu VL. Legionellosis. N Engl J Med 1997; 337:682-7.
(8.) Hutchinson DN. Nosocomial legionellosis. Rev Med Microbiol 1990; 1:108-15.
(9.) Edelstein PH. Nosocomial Legionnaires' disease: a global perspective. J Hosp Infect 1988; 11 (suppl A):182-8.
(10.) Lowry PW, Tompkins LS. Nosocomial legionellosis: a review of pulmonary and extrapulmonary syndromes. Am J Infect Control 1993; 21:21-7.
(11.) Strausbaugh JL. Legionnaires' disease. Int J Dermatol 1983; 22:239-44.
(12.) Wenzel RP. Hospital-acquired pneumonia: overview of the current state of the art for prevention and control. Eur J Clin Microbiol Infect Dis 1989; 8:56-60.
(13.) Berendet RF. Survival of Legionella pneumophila in aerosols: effect of relative humidity. J Infect Dis 1980; 141:689.
(14.) Dennis PJ, Lee JV. Differences in aerosols survival between pathogenic and non-pathogenic strains of Legionella pneumophila serogroup 1. Appl Bacteriol 1988; 65:135-41.
(15.) Yu ITS, Li Y, Wong TW, et al. Evidence of airborne transmission of the severe acute respiratory syndrome virus. New Engl J Med 2004; 350:1731-9.
MARJETA TERCELJ-ZORMAN
Unit of Pulmonary Diseases and Allergy
Division of Internal Medicine
University Medical Center
Ljubljana, Slovenia
MARIJA SELJAK
Institute of Public Health
Kranj, Slovenia
JANEZ STARE
Institute for Biomedical Informatics
Medical Faculty
University of Ljubljana
Ljubljana, Slovenia
JOZE MENCINGER
Jesenice General Hospital
Jesenice, Slovenia
JOZE RAKOVEC
University of Ljubljana
Ljubljana, Slovenia
RAGNAR RYLANDER
Department of Environmental Medicine
University of Gothenburg
Gothenburg, Sweden
FRANC STRLE
Department of Infectious Diseases
University Medical Center
Ljubljana, Slovenia
Submitted for publication August 11, 2004; revised; accepted for publication January 12, 2005.
Reprint requests should be sent to Marjeta Tercelj-Zorman; Unit of Pulmonary Diseases and Allergy, Division of Internal Medicine, University Medical Center, Zaloska 7, 1525 Ljubljana, Slovenia.
E-mail: marjeta.tercelj@kclj.si
COPYRIGHT 2005 Heldref Publications
COPYRIGHT 2005 Gale Group