In March, two workers at the Ford Motor Plant in Brook Park, Ohio, died of Legionnaire’s disease. Even though the plant was shut down, the source of the legionella bacteria was never identified. Ironically, a paper published in the New England Journal of Medicine in September 1997, reported that “One large-scale study of community acquired pneumonia in Ohio suggested that only 3 percent of sporadic cases of Legionnaires’ disease were correctly diagnosed.” But the under-recognition of legionella bacteria in public facilities and of cases of pneumonia due to such contamination is far from unique to Ohio.

le01.jpg (8835 bytes)(left to right) Direct fluorescent antibody stain of Legionella pneumophila; Gram stain of Legionella; Colonies of L. pneumophila on buffered charcoal yeast extract agar; (bottom) The Binax NOW rapid urinary antigen test

According to the Centers for Disease Control in Atlanta, there are 10,000 to 20,000 cases of legionnaires’ disease a year in the United States, 1,500 to 1,800 of which are reported. These numbers are particularly disturbing when you consider that legionella accounts for 5 to 15 percent of all community-acquired pneumonias (CAPs) and 20 percent of CAP deaths. Of even greater concern, legionella is the culprit in 23 percent of nosocomial pneumonias, the mortality rate for which is 40 percent.

An atypical pathogen
Compared to other pneumonia-causing bacteria, legionella has two important characteristics; it can survive in the environment for long periods; and it is abnormally virulent. Legionella species also get help from other bacteria and protozoa found in most water systems.

Despite its staying power and virulence, legionella could not be the scourge it is without help from humans. For example, the presence of Legionella alone does not mean there will be an outbreak of legionellosis. Not all species are pathogenic, and some are more deadly than others.

So why are “cooling towers,” ice machines and other fabricated water reservoirs so often linked to outbreaks of legionellosis? Researchers say their involvement relates to the fact that water in these devices often stands at temperatures between 77 and 108 ÞF, legionella’s favorite temperature for growth. As one CDC representative put it in an interview with CNN, “Whenever you put together a water system for a building, you’re gonna create a place where Legionella like to set up shop and grow.”

Non-specific presentation delays diagnosis
As with other pneumonias, effective treatment of legionellosis requires rapid response with specific antibiotics. After the infamous 1976 outbreak at the Legionnaire’s convention in Philadelphia that gave the disease its name, clinicians considered the unusual severity seen in those cases a hallmark of legionellosis. Now, however, they realize that the disease profile is much less distinct.

Early in the disease, symptoms are non-specific and very similar to other bacterial pneumonias. An X-ray, for example, cannot distinguish legionnaires’ from other pneumonias. In advanced cases, symptoms can escalate to stupor, respiratory failure and multi-organ failure. Although rare, legionellosis has attacked body systems other than the lungs.

Fortunately, there are several excellent antibiotics, such as erythromycin, that stop legionellosis in its tracks. But these are not necessarily among the frontline, drugs of choice for treating other pneumonias.

“Particularly with hospital-acquired Legionnaires’ disease, the correct treatment is entirely dependent on the physician’s index of suspicion,” said Janet E. Stout, Ph.D., director of the Special Pathogens Laboratory at the VA Medical Center in Pittsburgh. That means a test is not routinely performed unless physicians request it. “Most infectious disease researchers believe that legionella is a very underdiagnosed pneumonia,” Stout said. “And, of course, treatment with the wrong antibiotics means lost time and increased chance of death.”

Once concentrations of the organism are high enough, an outbreak is nearly inevitable. Unfortunately, researchers have not been able to determine the threshold concentration of legionella that triggers an outbreak. Another important aspect of the legionella threat is that it is particularly dangerous for immunosuppressed patients. And where are immunosuppresed patients most likely to be found? In a hospital, of course. As ominous as this may seem, the good news is that both environmental and clinical tests are available for legionella. The problem is that experts disagree on which test to use and when to use it.

Obviously, testing patients for whom there is a high index of suspicion of legionellosis is good medicine. Unfortunately, too many clinicians do not consider legionella soon enough. Some argue that this is a good reason to do environmental testing. Others, including the CDC, argue that the emphasis for testing should be on patients.

Enter the lab
“[Hospitals] need to be testing their patients, that’s essential,” said Dr. Richard Besser of the CDC in an interview with CBS news.

But until recently, testing took time. The gold standard for diagnosis of legionellosis is culture of the organism, but because it does not grow on standard microbiologic media, results can vary. This is true particularly when a laboratory is not accustomed to the procedure. Perhaps more importantly, it takes four to 10 days to get results.

Direct fluorescent-antibody staining tests are available but offer low sensitivity. PCR-based tests for urine, sputum and serum samples offer high specificity, relatively rapid results and the capability to test for species other than L. pneumophila but their sensitivity is relatively low. In addition, PCR-tests are limited by the possible presence of inhibitors in sputum and serum.

One test, the Binax enzyme immunoassay urinary antigen, detects Serogroup 1 of Legionella pneumophila, which accounts for about 90 percent of Legionnaires’ disease cases. The Binax (Portland, Me.) test also is fast (20 minutes) and works on urine samples. So, now that it is possible to quickly test patients, the key is getting physicians to order the test. Laboratorians can help by familiarizing themselves with the test and letting physicians know it is available.

Binax also makes a test that quickly provides results on environmental legionella contamination. Many hospitals routinely use the Binax Equate test as a pre-emptive strike. If legionella is found in the water or cooling system, clinicians are alerted so that they can test pneumonia patients sooner.

Critics of environmental testing argue against it with two important points. First, since no threshold level of contamination for legionella has been established, the mere presence of the bacteria does not mean an outbreak will happen. Second, if a facility’s tests are negative, it could produce a false sense of security. This is part of the reason the CDC has emphasized clinical rather than environmental testing.

Proponents of environmental testing maintain that the risk of nosocomial infection from legionella is reduced when contamination is found and action is taken. This is why JCAHO now requires hospitals to develop a management plan for pathogenic bacteria in water systems and cooling towers. “Essentially, what this JCAHO requirement is referring to is legionella,” Stout said.

“The role that we see for environmental monitoring in hospital situations is that if you do an environmental survey, and you don’t find legionella, then you don’t have to put a lot of money and effort into doing diagnostic tests on all hospital-acquired pneumonias. But if you do have legionella in your water, especially if it is serogroup 1, then we (Allegheny County) recommend you have and use the urinary antigen test in-house to get quick results for clinicians, ” said Stout. To that end, Binax, Stout and her colleagues have developed guidelines and made them available on the Web at http://www.legionella.org. Binax also has developed an algorithm for using its environmental test. “Legionella presents a unique challenge to both physicians and the laboratory,” Stout concluded.

Jonathan Briggs is a freelance writer based in Seneca, S.C.