By Arjun Srinivasan, MD

• State of Maryland study on Legionella in Water Systems of Healthcare Institutions
• Comparison Chart of Water Disinfection Methods

The Organism

• More than 30 species of Legionella have been identified but 90% human disease is caused by Legionella pneumophilia. There are 14 serogroups of Legionella pneumophilia though more than 80% of reported cases of Legionellosis are caused by serogroup 1 (1,2).
• Legionella micdadei (the Pittsburgh Pneumonia Agent) is the second most common species isolated in humans and is found primarily in organ transplant patients (3).
• Legionellacea are gram-negative aerobic bacilli. They are difficult to visualize on Gram's stain. Gimenez and silver stains allow better direct visualization although the latter can only be used on tissue samples (4).
• Water is the primary reservoir of the organism. Warm temperatures (45-50 oC), nutrient biofilms and commensal bacteria, all of which are found in large water distribution systems, promote growth of Legionella (9).
• Isolation can be difficult as Legionella species do not grow on standard bacteriologic media. However, the organism can be grown on buffered charcoal yeast extract supplemented with alpha-ketoglutarate and antimicrobial agents (5).

Epidemiology

Clinical:

• Legionella is one of the top 3 or 4 causes of community acquired pneumonia accounting for 2-15% of pneumonias in hospitalized patients (6). Cases are likely underreported owing to the difficulties with recovering the organism (7).
• The incidence of nosocomial disease likewise varies depending on the techniques used to make the diagnosis and on the patient population studied.
• Several risk factors for acquisition of nosocomial Legionella have been identified (1,3) and include:

• Cigarette smoking
• Chronic lung or heart disease
• Immunosuppression (especially with corticosteroids)
• Surgery, especially transplant and head and neck surgery
• Bone marrow transplant
• Hairy Cell Leukemia
• Intubation and mechanical ventilation
• Nasogastric tubes

• Interestingly, patients with neutropenia, leukemia (other than Hairy Cell) and HIV do not seem to be at increased risk (8).

Top of page

Environmental

• Though the first Legionella outbreak was ascribed to water from an air-conditioning cooling tower it is now felt that most cases are acquired from contamination of potable water supplies (9).
• Environmental contamination appears to be essential for nosocomial transmission as several studies have shown that nosocomial cases do not occur in hospitals where Legionella is not in the water supply (10).

Diagnosis

Clinical:

• Classically, cases of Legionnaires' disease were characterized by findings including: non-productive cough, pulse-temperature dissociation, diarrhea, myalgia and confusion and lab abnormalities including elevated liver enzymes, hyponatremia and hypophosphotemia (11).
• While these finding are seen in some cases, it is now widely recognized that Legionnaires' disease may present in a variety of ways. Several studies have shown that it can mimic pneumococcal pneumonia, especially on initial presentation (11).
• Radiographs are likewise non-specific although the majority of infiltrates are alveolar and not interstitial. Cavitation can occur but is seen almost exclusively in immunocompromised hosts (11).

Laboratory

• There are several methods for diagnosing Legionella in addition to culture and these are summarized in the table below. It should be noted, however, that culture remains the gold standard and is important in that it allows speciation and typing of the isolate.
• Urinary antigen testing is becoming increasingly popular but can ONLY detect infections due to serogroup 1.
• PCR detection shows promise but is not yet available clinically.

Test	Sensitivity	Specificity	Comments
Antibody - Paired Samples (acute and convalescent)	75%	95-99%	Only well validated for serogroup 1. Both IgG and IgM should be sent.
Antibody - Single Sample	No Data	50-70%	Single titer of 1:256 or more is suggestive of infection. Only well validated for serogroup 1.
clavage (BAL) specimen	25-75%	95-99%	Monoclonal anitbody yields highest specificity. Remain positive for several days after treatment is started.
DFA- Lung biopsy specimen	80-90%	99%	As above
Culture- Sputum or BAL	80-90%	100%	Often remains positive soon after antibiotics are started though sensitivity is diminished.
Culture- Lung biopsy	10-30%	100%	As above
Culture- Blood	10-30%l	100%
DNA Probe	50-70%	95-99%	May remain positive for several days after antibiotics are started.
Urinary Antigen	80-99%	99%	Only able to detect serogroup 1. Remains positive for weeks to months after infection.

Top of page

Treatment

• Several antibiotics have demonstrated both in vitro and in vivo activity against Legionella. Several treatment options are presented below (dosages given are for patients with normal renal and hepatic function).
• Quinolones and newer macrolides are considered drugs of choice (12).
• Optimal duration of treatment is unknown but most treat for 14 days in normal hosts and 21 days in immunosuppressed patients.

Adapted from: Stout, J.E. and Yu, V.L. Legionellosis. NEJM 1997;337(10)682.

Prevention

• Prevention of nosocomial Legionnaires' disease is important as cases generate enormous publicity in the lay press and are frequently the subject of lawsuits
• Environmental control of Legionella in water supplies is one of the most controversial and hotly debated topics in infection control.
• Much of the debate in recent years has focused on the merits or routine environmental cultures for Legionella.
• The CDC currently does not recommend culturing of the environment until at least 2 nosocomial cases are detected (13). They argue that colonization of water supplies with Legionella is nearly ubiquitous and cite studies that show it can be present in hospital water supplies without causing disease. They also point out that colony counts in surveillance cultures do not correlate with the risk of disease. The CDC recommends that Legionella be considered in all cases of nosocomial pneumonia and cultured when appropriate.
• Opponents of this point to several prevalence studies that show the incidence of Legionella colonization in hospital water supplies varies from 12-70%. Because nosocomial Legionnaires' disease does not appear to occur in hospitals that are not colonized they argue that environmental testing will save money by preventing unnecessary clinical testing in hospitals that are not colonized. They also question the validity of the studies that show nosocomial Legionella does not occur in some hospitals that are colonized, arguing that case finding in most of these studies was not rigorous (10).
• Whether or not to decontaminate water supplies is another area of controversy but experts recommend that hospitals with contamination rates of >30% give serious thought to decontamination (14).
• There are several options available for decontaminating water supplies that contain Legionella. The advantages, disadvantages and costs of the currently available methods are summarized in Appendix II.

The following link will allow you to view the: Report of the Maryland Scientific Working Group to Study Legionella in Water Systems in Healthcare Institutions .

Appendix II

Table on options for water decontamination. (HTML Version). (Excel Version 24kb)

References

1. Stout, J.E. and Yu, V.L. Legionellosis. New Engl Jnl Med 1997;337(10):68
2. Marston, B.J. et.al. Surveillance for Legionnaires' Disease: Risk Factors for Morbidity and Mortality. Arch Intern Med 1994;154:2417.
3. Yu, V.L. and Nguyen, M.L.T. Legionnaires' Disease: New Insights. Contemporary Internal Medicine 1992; June:49.
4. Yu, V.L. Legionella Pneumophilia (Legionnaires' Disease) in Mandell, Douglas and Bennett's Principles and Practice of Infectious Diseases. Philadelphia, Churchill Livingstone, 2000.
5. Procedures for the Recovery of Legionella from the environment. CDC Guidelines, 1994.
6. Muder, R.R. et.al. Community-Acquired Legionnaires' Disease. Semin Respir Infect 1989;4:32.
7. Marston, B.J. et.al. Preliminary Findings of a Community-Based Pneumonia Incidence Study. In: Barbaree, J., Breiman, R.F. and Dufor, A.P. eds. Legionella. Current Status and Emerging Perspective. Washington, D.C.: American Society for Microbiology, 1993:36.
8. Blatt, S.P. et.al. Legionnaires' Disease in Human Immunodeficieny Virus-Infected Patients: Eight Cases and Review. Clin Infect Dis 1994;18:227.
9. Lin, Y.E. et.al. Legionella in Water Distribution Systems. Journal AWWA 1998;90(9):112.
10. Yu, V.L. Resolving the Controversy on Environmental Cultures For Legionella: A Modest Proposal. Infect Ctrl Hosp Epi 1998;19(12):893.
11. Edelstein, Paul. Lgionnaires' Disease. Ciln Infect Dis 1993;16:741.
12. Dedicoat, M. and Venkatesan, P. The Treatment of Legionnaires' Disease. Jnl Antimicro Chem 1999;43:747.
13. Centers for Disease Control and Prevention. Guidelines for the Prevention of Nosocomial Pneumonia. MMWR 1997;46:31.
14. Goetz, A.M. Nosocomial Legionnaires' Disease Discovered in Community Hospitals Following Cultures of the Hot Water System: Seek and Ye Shall Find. Amer Jnl Infect Ctrl 1998 26(1):8.