Filling Gaps on Influenza En Route and the Etiology of Influenza During a Pandemic Year


  • Robert Steffen MD

    Corresponding author
    1. Center for Travel Medicine, Division of Epidemiology and Prevention of Communicable Diseases, Institute for Social and Preventive Medicine, University of Zurich, WHO Collaborating Centre for Travellers' Health, Zurich, Switzerland
    2. Epidemiology and Disease Prevention Division, University of Texas School of Public Health, Houston, TX, USA
    Search for more papers by this author
    • This Editorial refers to the articles by Boggild et al., pp. 4–8; Vilella et al., pp. 9–14; Memish et al., pp. 15–21; and Jauréguiberry et al., pp. 22–27 of this issue.

Prof. Robert Steffen, MD, University of Zurich ISPM/ZRM, Hirschengraben 84/E29, CH-8001 Zurich, Switzerland. E-mail:

This is the first issue of Journal of Travel Medicine with the cross-bar “Influenza” on the cover. In view of the fact that this infection is sometimes labeled the most frequent vaccine-preventable disease in travelers, this is justified. But what missing pieces do the four submitted original articles fill in the epidemiological and etiological puzzle?

Transmission of Influenza During Travel

The contribution by Vilella and colleagues confirms that influenza, particularly pandemic influenza A(H1N1) 2009, is intensely and probably rapidly transmitted among groups with close and prolonged interpersonal contact, such as during a 4-hour bus ride.1 Among the 113 Spanish medical students who traveled for 1 week to the Dominican Republic, 6 (5.3%) developed mild influenza-like illness abroad 1–3 days before return; 62 among 86 (72.1%) who could be interviewed developed illness within 4 days after landing back in Spain. Overall, pandemic influenza A(H1N1) 2009 was confirmed in 39 patients, 2 of them asymptomatic. Also, there is additional evidence that social distancing—far easier back home in the usual environment as compared to during travel—drastically reduces the risk of transmission: among 98 contacts of the 39 students with confirmed infection, only 5 developed probable or confirmed (1 patient) influenza (overall 5.1%). Among possible biases for such a significant difference is that viral shedding may have decreased after the trip, but this is unlikely to have played the decisive role, as viral detection was still demonstrated in a large proportion of students.

Based on anecdotes from families and friends there is common belief that “flu” is frequently transmitted on flights. Vilella and colleagues describe that aboard the flight from Santo Domingo back to Madrid the “students who became ill (upon return) were seated throughout the aircraft with no apparent clustering.”1 Although no information about other passengers could be obtained, that may be additional soft evidence to the observation that the majority of transmissions occurred preflight and that in-flight transmission is rare. Similarly, influenza A(H1N1) 2009 originated from an American spread within a tourist group in China, but only 1 of 87 passengers sharing the same flight outside that group was infected during a 45-minute flight, based on a thorough retrospective cohort investigation by the Chinese authorities.2 That patient was sitting in seat 9A, the index patient nearby in seat 7A. As in the Spanish student group, influenza transmission appears primarily to have occurred any time except during flights.

In the contribution by the GeoSentinel Surveillance Network,3 Boggild and colleagues discuss that “a small but measurable risk of influenza acquisition aboard commercial aircraft has been well documented, with long-haul flights conferring the highest risk of infections.” Pandemic influenza A(H1N1) 2009 was transmitted during a 12-hour 40-minute Los Angeles to Auckland flight from nine laboratory-confirmed members of a school group to 2 of 57 passengers seated within two rows; thus, the risk of infection was estimated to be 3.5% for this particularly exposed population.4 A single additional patient may have been infected during a 13-hour 20-minute Los Angeles to Seoul flight although she was sitting several rows (>5 m) apart from the index patient.5 Surprisingly, there is no documentation of in-flight transmission of seasonal influenza viruses, although the following three reports are often included in reviews6: influenza A/Texas/1/77(H3N2) was transmitted aboard an airliner in Alaska, while the passengers were kept aboard on the ground for 3 hours during repairs on the plane. Transmission was associated with the fact that the ventilation system and thus high-efficiency particulate air (HEPA) recirculation filters were not in use during that period, not with the flight.7,8 An Australian report lacked any laboratory evaluation, thus the title of that article had to be corrected to “Outbreak of influenza-like illness related to air travel” 4 years after publication.9 Lastly, transmission of influenza A/Taiwan/1/86 (H1N1) associated with the transfer of military personnel to Florida may have occurred preflight in the barracks of Puerto Rico.10 In summary, the literature includes four passengers with probable in-flight transmission of the pandemic virus, none with seasonal virus—the risk of transmission is very small; such evidence contradicts common belief. As there is suspected underreporting, additional research is indicated, but by own experience that is difficult as airlines are unlikely to collaborate, except possibly in China. Based on four cases only, there is insufficient evidence to claim that long-haul flights would confer the highest risk of transmission. No study so far has compared transmission on long versus short flights and neither the GeoSentinel report nor the quoted Swedish review11 included additional cases to add evidence. The latter actually seems to have been misquoted as it was referring to the different matter that “Air transportation, and especially long-haul flight, is a key factor for the spread of influenza.”11 Also, a mathematical model trying to calculate within-flight transmission of influenza wrongly used as a basic assumption that the plane in Alaska “managed to infect 72% of passengers during a 3-hour flight on a plane without ventilation,”12 while this aircraft actually was on the ground for that period of time.7 One should, however, not conclude that an aircraft cabin is germ-free; disease transmission of other infectious diseases has been documented.

Acute Respiratory Tract Infections and/or Influenza-like Illness in Travelers During (Pre-) Pandemic Periods

With respect to etiology of acute respiratory infections in a period of pandemic, both the French13 and Saudi14 experiences are instructive. At a Paris hospital returning patients with respiratory tract infections were consecutively investigated for pathogens from April to July 2009; similarly at the King Abdulaziz International Airport in Jeddah random samples of pilgrims were investigated pre- and post-Hajj 2009. The pathogen detection rate was 65.6% among the patients with respiratory disease, while the probably asymptomatic pilgrims had rates of 12.5% on arrival, 14.8% on departure back home, respectively. During the early influenza pandemic phase in Paris, the predominant pathogens to be associated with the respiratory tract infection among the 99 evaluated patients were rhinovirus (20%), influenza A(H1N1) 2009 (18%), and other influenza viruses (14%). Streptococci were cultured from 4.0% of the population; these four patients were among the eight with tonsillitis as a leading symptom. In the pilgrim population a broad variety of viruses was detected, mainly entero- and rhinoviruses, but influenza viruses were a small minority. The lesson learned is that at least during the initial phase of an influenza pandemic other infections may persist even if patients have respiratory tract symptoms. However, the pandemic virus will clearly dominate at its peak, particularly when the case definition is limited to influenza-like illness. Thus, one should not draw the wrong conclusion that immunization against influenza is useless.

The account derived from GeoSentinel,3 in contrast, during a prepandemic period exceeding 10 years, 1997 to 2007, detected only 70 probable or confirmed cases of influenza A and B among the over 37,500 ill-returned travelers. As patients with comparatively trivial illness, such as influenza, may rather consult with their family physician than a GeoSentinel site, this database may be more appropriate to evaluate serious infections, particularly rare ones.

Declaration of Interests

R. S. in the past 4 years has received honoraria from the pharmaceutical industry for lectures on influenza epidemiology, prevention, and therapy. Also, he was paid for participation in influenza vaccine advisory boards and for participation in influenza vaccine trials.


inline image

Wearing respiratory masks is an efficient protection against air transmitted pathogens such as influenza virus. The new pandemic with the virus influenza A (H1N1) 2009 was first detected in Southern California and Mexico during late April 2009 and then extended to the world within a few weeks. In this issue, the reader will find an editorial (pp. 1–3) and 4 articles that refer to influenza: a) carriage of infuenza virus by sick travelers across world hemispheres (pp. 4–8); b) outbreak of influenza A(H1N1) 2009 among medical students visiting the Dominican Republic (pp. 9–14); c) portage of respiratory tract pathogens in pilgrims attending the Hajj, Saudi Arabia (pp. 15–21); and d) etiologies of respiratory tract infections in returning travelers at the onset of the pandemic of influenza A(H1N1) 2009 (pp. 22–27). Setting: Tokyo subway, 2008.

Credit: Eric Caumes