Sequence analysis of viruses from cases on a single flight from the United States to Asia via Europe. Case viruses and 1 virus from a contact of case 3 cluster closely to published sequences from the New York region of the United States from the same period (A/New York/18/2009[H1N1]), and were distinct from other Singapore identified viruses sequenced during the same period (A/SG/08/2009[H1N1] and A/SG/02/2009[H1N1]).
Ooi PL, Lai FYL, Low CL, Lin R, Wong C, Hibberd M, Tambyah PA. Clinical and Molecular Evidence for Transmission of Novel Influenza A(H1N1/2009) on a Commercial Airplane. Arch Intern Med. 2010;170(10):913-915. doi:10.1001/archinternmed.2010.127
Copyright 2010 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2010
Influenza A(H1N1/2009) has spread rapidly throughout the world by international air travel.1 However, in-flight transmission of the virus has not been well documented. We report 6 cases of influenza A(H1N1/2009) associated with a single flight from the United States to Asia via Europe (“Flight A”) linked by molecular epidemiological data.
Five passengers and 1 crew member who had traveled on Flight A presented with acute onset of fever, malaise, cough, sore throat, or rhinorrhea, with the first case presenting symptoms while he was in New York, New York, and the rest within 3 days of the flight's arrival in Singapore. All were discharged well, without sequelae, after hospitalization for isolation. Immediate contact tracing of all passengers was conducted according to the World Health Organisation (WHO) recommendations.2 These define “close contacts” as passengers in the same row or in the 2 rows in front of and behind infected travelers. Twenty-three passengers in Singapore received oseltamivir phosphate (Tamiflu; Roche Laboratories Inc, Nutley, New Jersey) chemoprophylaxis and a strict 7-day quarantine. The WHO and the destination countries of the other passengers were notified. No additional cases of influenza A(H1N1/2009) were identified through active surveillance. We attempted to contact all passengers and crew to determine risk factors for infection in flight using a standardized questionnaire, but only 19 passengers and 7 crew members responded, since many were traveling and not contactable. We also performed molecular typing on the virus from nasopharyngeal specimens of infected patients.
A total of 596 passengers and crew members were on both legs of the flight. The infected crew member and 4 of 5 affected passengers were in the economy class from row 50 onwards, yielding an overall attack rate of 4.7% for that section compared with 0.2% for the rest of the plane (relative risk, 22.9; 95% confidence interval, 2.7-193.6 [P < .001]). Infected passengers and crew were younger than those not infected (mean age, 27.5 years vs 37.1 years; P = .02) (Table). This was consistent with the observation that a higher proportion of infection occurred in younger individuals.3
Viral genome sequencing4 using 1949 bases from segments 4 (HA) and 5 (NP) for 4 of the Flight A cases (A/SG/case 1-4) revealed that all the sequences closely matched sequences from New York. These were distinct from other viruses identified from Singapore and countries outside Flight A's route at that time (as demonstrated by the 98 bootstrap value in the Figure). The phylogenetic tree built from these sequences shows that the crew member was infected by a virus strain virtually identical to other New York strains circulating at the time, while passengers (including the business class passenger [case 2]) were infected by viruses that could be derivatives of this strain. Overall, the molecular and epidemiological data support the evidence of in-flight transmission of influenza A(H1N1/2009), although the precise mode of transmission is difficult to ascertain with certainty.
Modern commercial aircraft with high-efficiency particulate filters and frequent recirculation of cabin air have reduced the risk of transmission of airborne respiratory infections. Spread of respiratory viral infections, however, is thought to be related to infectivity of the source patient(s), proximity, and duration of contacts.
Only 2 of the infected passengers on Flight A would have been detected using the WHO criteria for contact tracing. This was also the case with severe acute respiratory syndrome,5 another emerging viral infection, transmitted predominantly by large particle droplets and direct contact with respiratory secretions or fomites. Perhaps contact tracing all passengers and crew in the same cabin or served by the same crew might be more appropriate in future airline outbreak investigations.
Human activities including onboard interactions may be important in in-flight influenza transmission. Our study showed that the infected passengers slept less on the plane (P = .06; Table). This was also reported in an outbreak of influenza on a delayed Alaskan Airlines flight in 1977.6 Unfortunately, too few of the passengers we studied used hand sanitizers or masks to assess their impact in reducing transmission of respiratory infections in air travel.
The most important limitation of our study is that we were unable to interview the majority of passengers and crew on Flight A or to do airflow, environmental, or seroepidemiological studies. We also depended on reporting from other international agencies to ascertain all infections. We could thus have underestimated the attack rates.
Our clinical, epidemiological, and molecular evidence are, however, highly suggestive that influenza A(H1N1/2009) transmission occurred on board Flight A, possibly through human interaction in a crowded cabin. Efforts to contain future emerging respiratory viral infections spread through international travel will have to include more thorough predeparture screening, perhaps novel decontamination or personal protective technologies, and broader clinical and molecular epidemiologic investigations than currently recommended.
Correspondence: Dr Ooi, Ministry of Health, 16 College Rd, Singapore 169854, Singapore (OOI_Peng_Lim@moh.gov.sg).
Author Contributions:Study concept and design: Ooi, Lai, Hibberd, and Tambyah. Acquisition of data: Ooi, Low, Lin, Wong, and Hibberd. Analysis and interpretation of data: Ooi, Lai, Lin, Wong, Hibberd, and Tambyah. Drafting of the manuscript: Ooi, Lai, Low, Hibberd, and Tambyah. Critical revision of the manuscript for important intellectual content: Ooi, Lin, Wong, Hibberd, and Tambyah. Statistical analysis: Lai and Hibberd. Obtained funding: Wong. Administrative, technical, and material support: Ooi, Low, Lin, Wong, and Hibberd. Study supervision: Ooi, Wong, and Tambyah.
Financial Disclosure: Dr Tambyah received research support from Baxter, Adamas, Merlion, Novartis, Pfizer, Wyeth, Asian Hygiene Council, and Asia-Pacific Advisory Committee on Influenza. None of these commercial entities had any access to the data.
Additional Contributions: The following individuals assisted in the outbreak investigation: Bok Huay Foong, BA, Pei Pei Chan, BSc(Hons), Hwi Kwang, Han, BEOHS, Hai Yin Toh, BEOHS, Suhana Solhan, BSc(Pharm)(Hons), Cheryl Tang, BSc(Hons), Lyn James, MBBS, MMed(PH), FAMS, Jeffery Cutter, MBBS, MMed(PH), MSc, FAMS, and Cui Lin, PhD, from the Ministry of Health, Singapore; Chee Wee Koh, BSc, Wan Yee Leong, BSc, and Charlie Lee, BSc, from the Genome Institute of Singapore; and Sebastian Maurer-Stroh, BSc, PhD, from the Bioinformatics Institute, Singapore.