key: cord-0951522-1j8c98mn
authors: Toyokawa, Takao; Shimada, Tomoe; Hayamizu, Takahiro; Sekizuka, Tsuyoshi; Zukeyama, Yuji; Yasuda, Miyako; Nakamura, Yuko; Okano, Sho; Kudaka, Jun; Kakita, Tetsuya; Kuroda, Makoto; Nakasone, Tadashi
title: Transmission of SARS‐CoV‐2 during a 2‐h domestic flight to Okinawa, Japan, March 2020
date: 2021-10-03
journal: Influenza Other Respir Viruses
DOI: 10.1111/irv.12913
sha: a79db08608b3100766d12154daa990e94e06b6e0
doc_id: 951522
cord_uid: 1j8c98mn

BACKGROUND: Coronavirus disease (COVID‐19), caused by severe acute respiratory syndrome coronavirus (SARS‐CoV‐2), has rapidly spread globally. Potentially infected individuals travel on commercial aircraft. Thus, this study aimed to investigate and test the association between the use of face masks, physical distance, and COVID‐19 among passengers and flight attendants exposed to a COVID‐19 passenger in a domestic flight. METHODS: This observational study investigated passengers and flight attendants exposed to COVID‐19 on March 23, 2020, on board a flight to Naha City, Japan. Secondary attack rates were calculated. Whole‐genome sequencing of SARS‐CoV‐2 was used to identify the infectious linkage between confirmed cases in this clustering. The association between confirmed COVID‐19 and proximity of passengers' seats to the index case and/or the use of face masks was estimated using logistic regression. RESULTS: Fourteen confirmed and six probable cases were identified among passengers and flight attendants. The secondary attack rate was 9.7%. Twelve of 14 SARS‐CoV‐2 genome sequences in confirmed cases were identical to that of the index case or showed only one nucleotide mutation. Risk factors for infection included not using a face mask (adjusted odds ratio [aOR]: 7.29, 95% confidence interval [95% CI]: 1.86‐28.6), partial face mask use (aOR: 3.0, 95% CI: 0.83‐10.8), and being seated within two rows from the index patient (aOR: 7.47, 95% CI: 2.06‐27.2). CONCLUSION: SARS‐CoV‐2 was transmitted on the airplane. Nonuse of face masks was identified as an independent risk factor for contracting COVID‐19 on the airplane.

The risk of in-flight transmission of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection has been reported to be low. 1 Well-documented cases of SARS-CoV-2 transmission in airplanes have mostly occurred on international flights, 2-8 thus suggesting a low risk of infection on domestic flights, which have relatively short flight times. 9 Given that physical distancing is impossible during flight travel, other measures such as the use of face masks play an important role in preventing viral transmission. Recent reports 1, 10 have suggested that face mask use effectively reduces the risk of SARS-CoV-2 infection; however, few reports indicate association between the precautionary measures and infection incidence. We investigated a cluster of coronavirus disease (COVID-19) cases among travelers who were likely exposed to an index patient in an airplane and tested the associated risk of infection according to face mask use and passengers' seats proximity to that of the index patient.

On March 26, 2020, a health center in Okinawa Prefecture was notified by a physician that a man in his 30s, a resident of Naha, Japan, 

This cohort study investigated passengers and flight attendants who boarded the flight at the airport in Prefecture B and traveled to Naha Airport in Okinawa on March 23, 2020.

The airplane was a Boeing 737-800 with 177 economy-class seats. It had an air recirculation system equipped with high-efficiency particulate air (HEPA) filtration. We confirmed that the HEPA filter of this airplane had been changed on February 15, 2020. The airplane was boarded by 148 persons: the index patient, 141 other passengers (seat occupancy: 80.2%), four flight attendants, and two pilots. We verified with the airline office that there was no flight delay and that passengers and crew members boarded and disembarked the plane using the boarding bridge.

F I G U R E 1 Flowchart showing the selection process of the study subjects on board the airplane with the index case of coronavirus disease 2.2 | Investigation of the index case, passengers, and flight attendants With the physician's assistance, we investigated the index patient over the phone. We obtained information, including name, age, sex, seat assignment, and telephone number of the patient's fellow passengers, from the passenger manifest.

At the beginning of our investigation, we traced the passengers who sat within two rows from the index patient. However, because symptomatic passengers with positive PCR test results were reported by the health department of another prefecture, we expanded our contact tracing as shown in Figure 1 : initial investigation and followup. In the initial investigation, we interviewed 82 out of all 141 passengers between March 26 and April 6, 2020. The passengers were notified by telephone about their potential exposure to SARS-CoV-2 infection and were asked to self-quarantine and self-monitor until April 6, which was 14 days after the flight travel. We advised them to consult their local COVID-19 consultation center if they experienced any symptoms. We requested the airline to inform the flight attendants in the same manner.

To investigate and test the association between mask wearing and physical distancing with COVID-19 incidence in passengers and flight attendants exposed to a passenger with COVID-19 on a domestic flight, we carried out the follow-up investigation during April 26-29, 2020. We developed a validated questionnaire and obtained information via a telephone interview. We interviewed 122 out of the 141 passengers, including the 77 who had been interviewed earlier and 45 who had not been contacted in the initial investigation, on their characteristics, seat assignment, face mask use during the flight travel (Always; Most of the time; Never/did not have a face mask; Cannot remember; Others), and symptoms of illness (fever ≥37.5 C, cough, nasal discharge/obstruction, sore throat, shortness of breath/ difficulty breathing, fatigue, nausea/vomiting, headache, myalgia/ arthralgia, hyposmia/hypogeusia) between March 23 and April 6, 2020. We requested the airline to administer the questionnaire to the flight attendants and received the completed forms by email.

Thus, we followed up a total of 126 participants for 34 days after the index case was reported.

Additionally, we requested the local health authorities and public health centers in seven prefectures to share the data of confirmed cases in each epidemiological investigation and send each specimen residue used for diagnosis to the Pathogen Genomics Center, National Institute of Infectious Diseases, Japan.

This study was conducted as an active epidemiological investigation under Article 15 of the Japanese Infectious Disease Law in cooperation with local health authorities, public health centers, and public health institutes. Thus, the study was exempt from the institutional review board review, and informed consent was not required.

We defined a confirmed case as an instance in which a passenger or flight attendant on board had SARS-CoV-2 infection confirmed between March 23 and April 6, 2020, regardless of whether symptoms or signs developed. We defined a probable case as an instance in which a passenger or flight attendant on board this flight had an acute respiratory illness (fever ≥37.5 C and at least one sign/symptom of respiratory disease, e.g., cough, sore throat, nasal discharge/obstruction, and shortness of breath) but did not undergo PCR testing between March 23 and April 6, 2020. The day of symptom onset was defined as the day on which the first symptom occurred in confirmed or probable cases based on epidemiological data. Secondary attack rates were calculated as the number of confirmed cases divided by the number of passengers and flight attendants, excluding the index case.

Nasopharyngeal swab specimens were collected from patients, and real-time reverse-transcription quantitative PCR (RT-qPCR) testing for SARS-CoV-2 11 was performed. An epidemiological investigation was performed using positive RNA samples subjected to whole-genome sequencing according to the PrimalSeq protocol for enriching complementary DNA of the SARS-CoV-2 genome using multiplex RT-PCR, 12 followed by discriminating the linkage between confirmed cases in this clustering based on genome-wide single nucleotide variations.

Confirmed or probable cases were identified as outcome variables.

Univariate and multivariate logistic regression analyses were performed, respectively, to calculate the crude and adjusted relative risk of COVID-19 among passengers seated within two rows from the index case or those wearing masks. We evaluated the proximity of a passenger's seat to the index case according to the World Health Organization guideline for close contact with COVID-19 in travelassociated settings, which is whether a passenger was seated within two rows of the index patient. The analysis included 121 passengers and four flight attendants from whom information could be obtained in the follow-up investigation in addition to the index patient. Regarding face mask use, no respondent selected "Others," and only one passenger selected "Cannot remember." Therefore, we used three categories, namely, "Always," "Most of the time," and "Never/did not have a face mask" as independent variables in the logistic regression analysis. All analyses were performed using SAS, Version 9.4 (SAS Institute Inc., Cary, NC, USA).

The 146 individuals aboard the flight (excluding the two pilots who did not share the cabin space during the flight) had a median age of 26 years (range, 3-85 years), and 74 (50.7%) of them were male.

Among these individuals, 14 confirmed cases and six probable cases were identified (Figures 2 and 3 ; Table 1 ).

Additionally, based on an epidemiological investigation conducted by each municipality, none of the 14 individuals with confirmed COVID-19 were identified as having a history of close contact with a confirmed COVID-19 patient other than the index patient and other confirmed cases on the flight. Furthermore, we were able to determine the genome sequence of the virus in 12 of the 14 confirmed cases that were all either identical to that in the index case-classified as the progeny of the Europe-related Japanese isolates (Pangolin lineage B.1.1) 13 -or differed by only one nucleotide (Table 2 ). In-flight exposure is thus the most likely explanation for this outbreak. As shown in Table 3 , the relative risks and 95% confidence inter- Notes: For each case, the day of symptom onset is shown in red, and the day of testing positive for SARS-CoV-2 using reverse-transcription polymerase chain reaction is shown in yellow. The passengers in seats 25F and 25G lived together (blue box), and the passengers in seats 14B, 14F, 14G, and 14H lived together (green box). Abbreviations: COVID-19, coronavirus disease; NA, not available; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2.

F I G U R E 3 Day of symptom onset for confirmed and probable cases of coronavirus disease (COVID-19) among passengers of Boeing 737-800 from City X, Prefecture B, to Naha, Okinawa, Japan, on March 23, 2020 (N = 21) conditions, including closed spaces, crowded places, and close-contact settings. [28] [29] [30] The cabin environment of a commercial airplane is essentially crowded and enclosed, making it conducive to disease spread through infected aerosols. 31 Our findings suggest that contact tracing for airplane events might be necessary to move beyond the scope of conventional rules and in some cases investigate all passengers and flight attendants. In this example, the clustering occurred on a relatively short-distance domestic flight (2-h travel). Some contributory factors were noted: the date of symptom onset of the index patient, which is thought to be the period of high viral shedding 32,33 ; the high viral load of the index patient's specimen, which was collected on the day after the flight; and the fact that the index patient was not wearing a face mask during the flight, despite his intense cough, which is a common source of aerosols. 31 Additionally, in terms of airflow in the cabin, the aisle seats, such as the index patient's seat, are located upstream of cabin airflow, according to the International Air Transport Association (IATA), which could have led to the diffusion of aerosols into the cabin space. 34 These factors might have contributed to this clustering. Nonetheless, despite these factors, the relatively low secondary attack rate may indicate that the cabin ventilation system worked effectively. Simultaneously, when conducting a risk assessment of an index patient on an airplane, a twotiered approach can be followed: first, the two-row rule as the initial step, and second, if more than one person is positive or suspected of being infected, the investigation should be extended to all passengers. This study has some limitations. First, passengers' information was obtained via telephone interview, and recall bias cannot be excluded. Second, we could not investigate all passengers and evaluate all asymptomatic passengers for SARS-CoV-2 infection using an antibody test; thus, we may have missed some passengers with COVID-19 and underestimated the impact of this event. Third, 19 passengers were lost to follow-up; however, we believe that the loss was randomly distributed among the passengers and was unrelated to their disease status. Therefore, loss to follow-up did not have an epidemiologically important impact on our findings. Fourth, we may have overestimated the attack rate by counting confirmed cases that may have been transmitted within the family at home rather than in the flight and assuming that all passengers classified as probable cases were infected with SARS-CoV-2. As their infection status was not confirmed on RT-PCR testing, we cannot exclude the possibility that their symptoms were due to other conditions or diseases, especially influenza. However, this is unlikely because according to the National Epidemiological Surveillance of Infectious Diseases system, the incidence of influenza was below 10.0 cases/sentinel/week at the timebelow the threshold for an alert. Fifth, those infected with SARS-CoV-2 might not have been infected while they were seated in the cabin of the airplane. Activities during their travel, their close contacts, or contaminated surfaces of seatback or in the aisle or toilets of the airplane, the baggage claim, and departures lobby or lounge in the airport were potential opportunities for exposure to the index patient.

Finally, we did not evaluate how well the ventilation system functioned in the aircraft at the time of this event.

Our study shows that aerosol-based transmission of SARS-CoV-2 may occur during a short domestic flight. Wearing face masks during flight travel was associated with a reduced risk of SARS-CoV-2 infection. Now that the spread of variants of SARS-CoV-2 across regions and countries has become a global threat, awareness on the use of face masks, an effective pre-boarding screening system, and a vigorous investigation system are necessary to minimize the risk of in-flight viral transmission. Additionally, this cluster occurred in March 2020 before the emergence of several variants of SARS-CoV-2 with confirmed increased transmissibility. In the future, it will be necessary to assess the risk of in-flight transmission of these variants.

We thank Dr. Tomoyuki Azuma for his leadership and support as the former director of the Naha City Public Health Center. We are grateful for the cooperation and support of the airline company, medical staff, staff members of related prefectural health authorities and local public health centers, the exposed passengers, and flight attendants in conducting the investigation. We would also like to thank our colleagues for their remarkable contribution and work. 

None declared.

The peer review history for this article is available at https://publons. com/publon/10.1111/irv.12913.

The new sequences have been deposited in the Global Initiative on Sharing All Influenza Data (GISAID) with accession IDs EPI_-ISL_591471 to EPI_ISL_ 591 483 (Table 2) .

https://orcid.org/0000-0003-3872-6977

Tetsuya Kakita https://orcid.org/0000-0002-8991-7397

In-flight transmission of SARS-CoV-2: a review of the attack rates and available data on the efficacy of face masks

Potential transmission of SARS-CoV-2 on a flight from Singapore to Hangzhou, China: an epidemiological investigation

Assessment of SARS-CoV-2 transmission on an international flight and among a tourist group

Transmission of severe acute respiratory syndrome coronavirus 2 during long flight

Flight-associated transmission of severe acute respiratory syndrome coronavirus 2 corroborated by whole-genome sequencing

In-flight transmission of SARS-CoV-2

In-flight transmission cluster of COVID-19: a retrospective case series

Genomic evidence of inflight transmission of SARS-CoV-2 despite predeparture testing

Absence of in-flight transmission of SARS-CoV-2 likely due to use of face masks on board

Development of genetic diagnostic methods for detection for novel coronavirus 2019 (nCoV-2019) in Japan

Disentangling primer interactions improves SARS-CoV-2 genome sequencing by multiplex tiling PCR

A genome epidemiological study of SARS-CoV-2 introduction into

International Air Transport Association. IATA calls for passenger face covering and crew masks

Airports Council International World and International Air Transport Association

Risk of COVID-19 during air travel

Contact tracing in the context of COVID-19: interim guidance

A major outbreak of severe acute respiratory syndrome in Hong Kong

Extensive viable Middle East respiratory syndrome (MERS) coronavirus contamination in air and surrounding environment in MERS isolation wards

Scientific brief: SARS-CoV-2 transmission

High SARS-CoV-2 attack rate following exposure at a choir practice

COVID-19 outbreak associated with air conditioning in restaurant

Community outbreak investigation of SARS-CoV-2 transmission among bus riders in eastern China

Ten scientific reasons in support of airborne transmission of SARS-CoV-2

Early transmission dynamics in Wuhan, China, of novel coronavirus-infected pneumonia

Estimation of the reproductive number of novel coronavirus (COVID-19) and the probable outbreak size on the Diamond Princess cruise ship: a datadriven analysis

Nowcasting and forecasting the potential domestic and international spread of the 2019-nCoV outbreak originating in Wuhan, China: a modelling study

Clustering and superspreading potential of SARS-CoV-2 infections in Hong Kong

Closed environments facilitate secondary transmission of coronavirus disease

What settings have been linked to SARS-CoV-2 transmission clusters?

Aerobiology and its role in the transmission of infectious diseases

Temporal dynamics in viral shedding and transmissibility of COVID-19

Coronavirus-disease 2019 patients in earlier stages exhaled millions of SARS-CoV-2 per hour

International Air Transport Association. Cabin air quality

Casecontrol study of use of personal protective measures and risk for SARS-CoV-2 infection

Respiratory virus shedding in exhaled breath and efficacy of face masks

Effectiveness of face masks in preventing airborne transmission of SARS-CoV-2. mSphere

Transmission of SARS-CoV-2 during a 2-h domestic flight to Okinawa