Secondary Logo

Journal Logo

DIAGNOSIS OF LEGIONELLA INFECTION

Infectious Diseases in Clinical Practice: September-October 2002 - Volume 11 - Issue 7 - p 418-419
SPECIAL ARTICLES: Review of Literature: General Infectious Diseases
Free

This section of IDCP features summaries of publications relevant to the practice of general infectious diseases. In most cases, a comment is provided from the editor concerning interpretation, impact or further relevant information on the topic reviewed. This represents a modification of selected entries in the “What’s News” section of the Johns Hopkins website for ID (reprinted fromwww.hopkins.id.edu, with permission).

DIAGNOSIS OF LEGIONELLA INFECTION

[Murdoch DR. CID 2003;36:64]

The author reviews diagnostic testing for Legionella. Emphasis is placed on the following observations: 1) specialized tests are required; 2) there are substantial differences between performance by research laboratories and service laboratories; 3) there is no “gold standard”; 4) the specific recommendations should be influenced by local Legionella epidemiology which shows that L. pneumophila serogroup one dominates in most areas of the world, but L. longbeachae is the major cause in Australia and New Zealand. Table 1 summarizes the available information about these tests.

Table 1

Table 1

Back to Top | Article Outline

Culture

  • Standard medium is buffered charcoal yeast extract agar supplemented by A-ketoglutarate.
  • A difficulty is that <50% of patients produce sputum.
  • Samples should be processed promptly; cytologic screening should not be done.
  • These bacteria may be grown in blood cultures, but the yield is poor and may not activate the alarm of commercial blood culture machines (JCM 1985;22:422). This means blind subcultures are required.
Back to Top | Article Outline

DFA Staining

  • The advantage is availability of results in 2–4 hours.
  • The technique is demanding and requires expert personnel.
  • The yield varies by species (highest for L. pneumophila) and is higher with BAL (JAMA 1983; 250:1981).
  • Cross-reactions with false-positive results may occur with B. fragilis, Pseudomonas species, Stenotrophomonas and Flavobacterium.
Back to Top | Article Outline

Urinary Antigen

  • The test is easy to perform with results in 15 minutes.
  • Commercial kits use both RIA and EIA methods, which are considered equally good.
  • Sensitivity can be increased by 20% with urine concentration (CID 1999;29:953).
  • The test detects only L. pneumophila serogroup one.
  • The test is positive within one day of onset of symptoms and persists for days to weeks, and in one case persisted over 300 days (JCM 1984;20:605).
  • A false-positive test has been reported with serum sickness (CID 1999;29:953).
Back to Top | Article Outline

Serology

  • The convalescent sample should be at least three weeks after the onset of symptoms and a four-fold increase in titer to >1:128.
  • An advantage is the ability to detect all species and serogroups.
  • IgM is an unreliable marker of acute infection because these antibodies may persist for long periods.
  • Cross-reacting antibodies are occasionally found with pseudomonads, mycobacteria, Bacteroides, and Campylobacter.
Back to Top | Article Outline

PCR

  • This is currently experimental and is not generally available except through research laboratories.
  • Sensitivity is equal or greater than culture (JCM 1992;30:920; JCM 2000;38:1709).
  • PCR may be applied to urine, serum and sputum; the authors speculate that PCR “could be considered the test of choice for patients who produce sputum.”
  • Additional work is required to define standard PCR methods.
Back to Top | Article Outline

Recommendations

  • Risk factors for this infection are: elderly patients, smokers, immunosuppressed, patients with chronic lung disease, patients in hospitals with contaminated water supplies and in persons who are seriously ill with pneumonia.
  • In areas where L. pneumophila serogroup one is predominant or in epidemics involving this organism, the urinary antigen test is preferred.
  • If available, Legionella PCR combined with urinary antigen testing is predicted to be the best test. Culture is important, but has low sensitivity.
Back to Top | Article Outline

Comment:

These recommendations for urinary antigen assay and culture on selective media for detection of Legionella are the recommendation of the IDSA guidelines for pneumonia. The guidelines do not include PCR due to the lack of a commercially available test and variable results with “home brews.” The disappointing facet of Legionella testing is that few hospitals use any test at all.

© 2002 Lippincott Williams & Wilkins, Inc.