Hold Your Breath? Legionellosis and Tuberculosis During Travel

Abstracts & Commentary

Synopsis: Two papers published just this year present the findings of investigations into the transmission of respiratory infections: Cruise ship-associated Legionella pneumophila and exposure to tuberculosis while riding a train. Both of these papers were accompanied by interesting editorials that help to put the issue of travel-associated respiratory infections in perspective.

Sources: Pastoris MC, et al. Clin Infect Dis 1999;28:33-38; Moore M, et al. Clin Infect Dis 1999;28:52-56; Edelstein PH, Cetron MS. Clin Infect Dis 1999;28:39-41; Witt MD. Clin Infect Dis 1999;28:57-58.

Pastoris and associates describe a 67-year-old British man who presented with Legionella pneumonia in September 1995, eight days into a Mediterranean cruise on a large Italian ship. He was hospitalized but, despite treatment, died 20 days following the development of symptoms. The patient was a cigarette smoker and had coronary artery disease. The case was reported to the Italian representative for the European Working Group on Legionella Infection (EWGLI) by the British Communicable Disease Surveillance Centre. An investigation was made to detect other cases among crew or passengers and to determine the source of Legionella infection.

None of the 116 crew members reported symptoms, and only three (2.6%) had low-titer serologic evidence of Legionella infection. Among passengers, there were no other cases reported during the cruise with the index case and no cases prior to this cruise from 1986 through 1995. One clinically compatible case was subsequently reported in November 1995 and another documented case in October 1996. Environmental sampling of multiple water sources yielded Legionella pneumophila from the ship’s freshwater system that was identical to the patient’s isolate. In fact, 80% of the fresh water samples yielded L. pneumophila.

The second report was from the United States and the Centers for Disease Control. In January 1996, a 22-year-old male was diagnosed with highly smear-positive, culture-confirmed pulmonary tuberculosis, following hospitalization after requesting medical help while he was traveling on a train. The patient was coughing throughout his trip and frequently experienced hemoptysis. Despite antituberculous therapy in the hospital, he died from a pulmonary hemorrhage two weeks into hospitalization. Because he had used public transportation (train and bus) over a two-day period, an investigation was launched into the possible transmission of tuberculosis to other passengers and crew members. All passengers and crew were contacted within two weeks of the exposure and were asked to undergo a two-step tuberculin skin test (TST) and, if negative, a third TST after three months. Those with positive tests were evaluated for their exposure to the index patient, for symptoms, for other possible exposures to tuberculosis, and whether they had received bacille Calmette-Guérin vaccine. A positive skin test was considered to be 10 mm or more.

The patient had traveled from Chicago to Florida. His first plane trip was 12.3 hours. Because of flooded train tracks, there was an intervening bus trip of 5.5 hours, and the final train trip was 16.8 hours. Of 479 passengers, 368 (77%) were able to be contacted; there were 44 crew members. Final TST was obtained from 228 passengers and 29 of the crew. Seventeen patients were excluded from analysis. For the 240 passengers and crew, there were four TST conversions and 11 positive TSTs. Four of these 15 positive TSTs had exposure to the index patient. All of these exposures occurred in the dining car, three passengers were seated at tables near him, and one had a direct conversation. For two of these four persons, which included the person who had the conversation, there were no other risk factors for a positive TST and, thus, they were presumed to have acquired new infection from the index case. The patient had "briefly" traveled to the dining car but spent most of the time in the passenger cars with his head under a hooded sweatshirt. The train was fitted with high-efficiency particulate air filters, with about 10-15 air exchanges per hour.

Comment by David R. Hill, MD, DTM&H

The reports of legionellosis during a vacation cruise in the Mediterranean and tuberculosis exposure while riding public transport in the United States could be cause for concern and alarm. However, there is a silver lining to both reports. In the first instance, the establishment of EWGLI created the structure to identify the case in England and adequately investigate it in Italy.1 This working group can be credited with raising awareness of the problem of legionellosis during tourist travel throughout many European sites, with developing methods for both detection and surveillance, and in helping to define standards for water purification to help prevent future cases. The Internet web site for information on Legionella outbreaks as well as other infectious disease outbreaks throughout Europe can be accessed at www.outbreak.org.uk/secure/index.html.

In the second case, despite a highly positive case of tuberculosis, transmission was limited (2 TST conversions/240 persons investigated). In the cases that acquired new infection, the risk factor was a brief, proximal contact, either face-to-face, or when seated near the patient, rather than sharing air space over a prolonged period. That the latter was not a risk could be because the patient coughed into his sweatshirt hood or because of the efficiency of air exchanges within the train ventilation system. Nevertheless, transmission did occur and there could have been other, unrecognized cases since only 49% of all passengers and crew were studied. The investigation was hampered by the inability to locate all passengers (77% identified) and the difficulty in having these persons complete a three-stage skin testing procedure over three months.

Travel-associated legionellosis has been well described in both cruise ship passengers and other tourists.2-4 In cruise ships, the source has usually not been determined or has been from the whirlpool spas, as opposed to potable, fresh water.4 The implications of the Legionella case for travel medicine physicians is three-fold. First, they should be aware that elderly or other travelers with health risk factors, such as smoking and chronic illness, may be exposed to Legionella and, thus, at risk for infection. The elderly traveler with chronic illness is just the person who may choose to travel internationally by ship rather than overland. At the International Traveler’s Medical Service at the University of Connecticut, cruise ship travelers had a mean age of 61 years compared to 41 years for all other travelers (P < 0.001), and 55% of them had chronic medical conditions. Second, these travelers should be assessed for their immunization status against other respiratory infections, such as influenza and pneumococcal pneumonia.

Third, the travel medicine provider should be aware that a pulmonary infection in a returned traveler could be Legionella. The travel history should be taken and specialized tests ordered to isolate the organism. If a case of legionellosis is documented, it should be promptly reported to the appropriate health authorities so that other cases may be identified and a potential outbreak explored and contained. It was through the recognition of only a few cases in a community hospital5 that the largest ship-associated outbreak was discovered.4 Prompt recognition and reporting may also be useful in containing other pulmonary infections occurring on ships. The most recent example of this was the outbreak of influenza A among passengers touring Alaska and the Yukon Territory last summer.6

The risk of tuberculosis during public transport has received recent publicity surrounding airline-associated cases. The risk in these situations has been determined to be extremely low. Indeed, out of seven investigations of separate flight exposures to a highly infectious index case, in only two situations was there evidence for transmission to other passengers or crew members,7-9 and, in one of these, the transmission probably occurred with repeated exposures over several flights since the index case was a crew member.7 Active tuberculosis has not occurred as a result of airline transmission in any situation.

The WHO recently convened a consensus committee on airline-associated tuberculosis and, from analysis of these investigations, has recommended that passengers and crew be notified of a potential exposure only if the exposure occurred within three months, if the flight was eight or more hours duration (including ground and waiting time), and if passengers were in close proximity to the index case.10 Because of the low risk of transmission and the difficulty and cost of performing adequate epidemiologic studies, they also state that further epidemiologic studies of airline-associated tuberculosis did not appear warranted.

Exposure to tuberculosis is unlikely during travel, although there could certainly be instances of unknown exposure both during public transport and from face-to-face contact. The data from the train and airline studies indicate that transmission risk is extremely low, and the CDC authors even suggest that expanded contact investigations in situations of a low likelihood of transmission may not be worth the resources required. If a traveler is concerned about potential contact with tuberculosis, then the most prudent action is to check pre- and post-travel TSTs, with the post-travel TST performed approximately two months following exposure.

Thus, it seems that travelers can breath easily; however, they should keep in mind the potential for exposure to unusual pathogens, particularly if they have concomitant health conditions. In addition, the travel health professional should be prepared to recognize and respond to these situations if the traveler returns ill. (Dr. Hill is Associate Professor of Medicine; Director, International Travelers’ Medical Service, University of Connecticut.)


    1. Hutchinson EJ, et al. EWGLI: Eurosurveillance 1996; 1:37-39.

    2. Rosamini F, et al. Am J Epidemiol 1984;119:124-134.

    3. Joseph C, et al. Eur J Epidemiol 1996;12:215-219.

    4. Jernigan DB, et al. Lancet 1996;347:494-499.

    5. Guerrero IC, Filippone C. Infect Control Hosp Epidemiol 1996;17:177-178.

    6. CDC. MMWR Morb Mortal Wkly Rep 1998;47:685-688.

    7. Driver CR, et al. JAMA 1994; 272:1031-1035.

    8. CDC. MMWR Morb Mortal Wkly Rep 1995;44:137-140.

    9. Kenyon TA, et al. N Engl J Med 1996;334:933-938.

    10. Tuberculosis and Air Travel: Guidelines for Prevention and Control. 1998, Geneva: WHO. (report WH)/TB/98.256). http://www.who.int/gtb/publications/aircraft/index.html).

Which of the following statements concerning respiratory infections during travel is correct?

    a. Elderly persons with chronic medical conditions are at increased risk for legionellosis.
    b. Influenza and pneumococcal vaccines may be indicated for certain travelers.
    c. Tuberculosis skin testing may be used to determine exposure to and infection with Mycobacterium tuberculosis.
    d. The risk of infection following train or airline exposure to tuberculosis is extremely low.
    e. All of the above