If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Kyoto MS Center, Kyoto Min-Iren Chuo Hospital, 16-1 Kasuga-Cho Nishinokyo, Nakagyo-Ku, Kyoto 604-8453, JapanDepartment of Neurology, Kaikoukai Josai Hospital, 4-1 Kitahata-Cho, Nakamura-Ku, Nagoya 453-0815, Japan
We investigated impact of BCG vaccination on prevention of local COVID-19 spread.
•
Prefectural data on SARS-CoV-2 prevalence and BCG vaccine coverage were analyzed.
•
The BCG vaccine coverage was higher in uninfected than in prevalent prefectures.
•
Prevalence of the infection significantly correlated to the BCG vaccine coverage.
•
The infant BCG vaccine plays a significant protective role in local COVID-19 spread.
Summary
Objectives
In Japan, the first case of coronavirus disease 2019 (COVID-19) was diagnosed on January 15, 2020 and subsequent infections rapidly increased. The Bacillus Calmette-Guérin (BCG) vaccination program is the principal element of tuberculosis control in Japan. We investigated the impact of routine infant BCG vaccination on prevention of local COVID-19 spread.
Methods
Data on the prevalence of SARS-CoV-2 infection, annual routine infant BCG vaccine coverage (represented by the number of BCG vaccinations per live births), and other candidate factors in each prefecture were obtained from the official notifications database in Japan. We analysed the association of vaccine coverage with the prevalence of SARS-CoV-2 infection.
Results
The BCG vaccine coverage in 1999–2002, 2004, and 2012 in five prefectures with no COVID-19 infections was significantly higher than that in five prefectures with a high prevalence of infections (Mann-Whitney U test, p<0.05). The prevalence of SARS-CoV-2 infection was significantly negatively correlated with BCG vaccine coverage in 2004 and was significantly positively correlated with age groups 20–34 and 40–54 years (Spearman's rank correlation, p<0.01).
Conclusions
Our findings suggest that routine infant BCG vaccination coverage in young generation had a significant impact on prevention of local COVID-19 spread in Japan.
Because of an explosive increase in severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infections since late March, the Japanese Prime Minister declared a state of emergency for seven prefectures on April 7 and for the whole country on April 16. Japan lifted its nationwide state of emergency on May 25, 2020. The cumulative number of COVID-19 cases in Japan had topped 16,623 (13.1 per 100,000 population) on that date; however, this figure was much lower than that in Western countries.
Based on the possible effect of the Bacillus Calmette-Guérin (BCG) vaccination to prevent nonspecific infections other than tuberculosis, several clinical trials have been initiated with the aim to protect health care personnel and older people from COVID-19 infection.
Japanese bacteriologist Kiyoshi Shiga introduced BCG to Japan in 1924. The 172nd passage from the first culture is the origin of the Japanese routine infant BCG vaccine (Tokyo-172).
The BCG vaccination program has been the principal element in tuberculosis control in Japan for more than 70 years.
Here, we investigated the degree of impact of BCG vaccination on local prevalence of patients with COVID-19 infection, SARS-CoV-2 polymerase chain reaction-positive (PCR+) individuals, and deaths owing to COVID-19 in prefectures of Japan. Data on the number of BCG vaccinations, number of live births, and other candidate factors in each prefecture were obtained from the official notifications database of the Japanese government.
Statistics Bureau, Ministry of Internal Affairs and Communications of Japan. e-Stat, Portal Site of Official Statistics of Japan. https://www.e-stat.go.jp. Accessed 24 April2020.
Prevalence of patients with COVID-19, SARS-CoV-2 PCR+ individuals, and deaths in each prefecture
Anonymous data on the number of patients with COVID-19 from January 15 to March 29, 2020, and thereafter the number of SARS-CoV-2 PCR+ individuals and the number of deaths owing to COVID-19 among Japanese residents in each prefecture were obtained from official notification records of the Ministry of Health, Labour and Welfare of Japan.
Data on March 29, April 24, and May 1, 2020 were used. Passengers aboard the Diamond Princess cruise ship anchored in Yokohama were not included. Crude SARS-CoV-2 prevalence in each prefecture was calculated as the number of patients/PCR+ individuals divided by the number of individuals in each prefecture in 2018, estimated using the results of the national census in 2010 and in 2015 taken by the Statistics Bureau, Ministry of Internal Affairs and Communications of Japan.
Statistics Bureau, Ministry of Internal Affairs and Communications of Japan. e-Stat, Portal Site of Official Statistics of Japan. https://www.e-stat.go.jp. Accessed 24 April2020.
Information on the annual number of routine infant BCG vaccinations in each prefecture from 1998 to 2017 was obtained from the Report on Regional Public Health Services and Health Promotion Services, and the annual number of live births in each prefecture in the same year was obtained from the latest version of the Vital Statistics; their ratio was used as a surrogate for BCG vaccine coverage. These surveys were conducted by the Vital, Health and Social Statistics Office, Ministry of Health, Labour and Welfare of Japan.
Statistics Bureau, Ministry of Internal Affairs and Communications of Japan. e-Stat, Portal Site of Official Statistics of Japan. https://www.e-stat.go.jp. Accessed 24 April2020.
We investigated other candidate factors related to the spread of SARS-CoV-2. The number of days from diagnosis of the first patient with COVID-19 infection in each prefecture up to March 29, 2020 was obtained from official notification records.
The elderly population in 2018 and the population by 5-year age groups in 2019 in each prefecture was calculated based on official population estimates.
Statistics Bureau, Ministry of Internal Affairs and Communications of Japan. e-Stat, Portal Site of Official Statistics of Japan. https://www.e-stat.go.jp. Accessed 24 April2020.
The populations that migrated to and from each prefecture in 2018 was obtained from the annual report on internal migration in Japan derived from the basic resident registration; inhabitable area was calculated using data of the Statistics Bureau, Ministry of Internal Affairs and Communications.
Statistics Bureau, Ministry of Internal Affairs and Communications of Japan. e-Stat, Portal Site of Official Statistics of Japan. https://www.e-stat.go.jp. Accessed 24 April2020.
The ratio of day- to night-time population, ratio of households of 5 individuals or more to entire households, and ratio of workers in the primary, secondary and tertiary sectors of industry to entire working population were obtained from results of the national census in 2015.
Statistics Bureau, Ministry of Internal Affairs and Communications of Japan. e-Stat, Portal Site of Official Statistics of Japan. https://www.e-stat.go.jp. Accessed 24 April2020.
The Kolmogorov–Smirnov test revealed that the variables did not have a normal distribution. Therefore we performed non-parametric analyses to compare annual BCG vaccine coverage among prefectures and to examine the correlation of SARS-CoV-2 prevalence with BCG vaccine coverage and other related factors. The significance level was set at p = 0.05 for group comparisons and p = 0.01 for correlation analyses. All statistical analyses were performed using SPSS 16.0 J (IBM Japan, Tokyo, Japan).
Results
Table 1 depicts the prevalence of patients with COVID-19, SARS-CoV-2 PCR+ individuals, and deaths owing to COVID-19 in each prefecture. The prevalence of COVID-19 varied widely among the 47 prefectures. On March 29, no infections were detected in five prefectures (Iwate, Yamagata, Toyama, Tottori, and Shimane), whereas 33.30 individuals/million were diagnosed in Hokkaido. On April 24 and May 1, only Iwate had no infections. PCR+ individuals were the most prevalent in Tokyo on May 1 (298.08 individuals/million). The mortality rate was high in Fukui, Gunma, and Ishikawa prefectures (9.04, 7.68, and 5.25 deaths/million, respectively). The case fatality rate was high in Gunma, Aichi, and Ehime prefectures (10.27, 6.38, and 6.38 deaths/PCR+ individuals (%)).
Table 1Prevalence of patients with COVID-19, SARS-CoV-2 PCR+ individuals, and deaths in each prefecture.
The annual change in routine infant BCG vaccine coverage is shown in Fig. 1 and Table 2. BCG vaccine coverage was relatively low and variable among prefectures between 1999 and 2002. In eight prefectures coverage was less than 50%, and in Tokyo and Kanagawa less than 30% of the live births. In 2004, in accordance with the national policy on improving tuberculosis prevention and change in the target age of vaccination, the nationwide BCG vaccine coverage was 1.179 and that of Iwate was as high as 1.469. In 2006 and thereafter, the vaccine coverage was more than 50%, except for 42% in Miyagi in 2010. The vaccination coverage gradually stabilized, with all prefectures maintaining a greater than 90% coverage rate since 2013.
Fig. 1Annual change in routine infant BCG vaccine coverage. Top: Change in annual number of routine infant BCG vaccines (grey squares) and live births (black diamonds). The number of live births has gradually decreased over the last 20 years. The number of BCG vaccines has fluctuated according to changes in the routine vaccination system. Bottom: Annual change in routine infant BCG vaccine coverage, represented by the number of BCG vaccines per number of live births. *p<0.05, Mann–Whitney U test. aCOVID-19 prevalence on March 29 (patients per million).
Fig. 1 shows a comparison of annual BCG vaccine coverage among the five prefectures with no COVID-19 infections and the top five prefectures with the highest COVID-19 prevalence (Hokkaido, Tokyo, Chiba, Oita, and Hyogo) on March 29, 2020. In 1999, 2000, 2001, 2002, 2004, and 2012, the BCG vaccine coverage of the five prefectures with no infections was significantly higher than that of the prefectures with a high prevalence of COVID-19 infections (p<0.05). Prior to 2005, the target age of vaccination was less than 4 years old; therefore, the vaccine coverage in 1999 was actually relevant to the generation born between 1995 and 1999.
The prevalence of patients with COVID-19, SARS-CoV-2 PCR+ individuals, and deaths owing to COVID-19 showed a significant negative correlation with BCG vaccine coverage in 2004, ratio of the elderly population (age 65 years or more), and the primary sector of industry, and a significant positive correlation to days from diagnosis of the first patient with COVID-19 to March 29 (p<0.01) (Fig. 2, Table 3). Because the elderly population showed an inverse association with prevalence of SARS-CoV-2 infection, additional correlation analyses by age group were performed. A significant positive correlation was shown between the prevalence of SARS-CoV-2 infection and age groups 20–34 and 40–54 years (Fig. 3). As for occupation, workers in the tertiary sector of industry were significantly correlated with SARS-CoV-2 infection in a positive manner (Fig. 3) (Table 3).
Fig. 2Correlation between COVID-19 prevalence and related factors. Scatter plot of COVID-19 prevalence (ordinate) against related factors (abscissa). Prevalence of (a) patients with COVID-19 on March 29, (b) SARS-CoV-2 PCR+ individuals on April 24, and (c) deaths on April 24 showed significant negative correlation with BCG vaccine coverage in 2004, the population age 65 years or more, and workers in the primary sector of industry; and significant positive correlation with days from the first patient diagnosis with COVID-19 up to March 29. Dotted squares indicate significant correlation (Spearman's rank correlation, p<0.01). Regression lines are shown for reference.
Fig. 3Correlation between prevalence of patients with COVID-19 on March 29 and related factors. Scatter plot of the prevalence of patients with COVID-19 (ordinate) against related factors (abscissa). (A) Significant positive correlation of age groups 20–24, 25–29, 30–34, 40–44, 45–49 and 50–54 years with COVID-19 prevalence. (B) Workers in the tertiary sector of industry were significantly correlated with SARS-CoV-2 infection in a positive manner. Dotted squares indicate significant correlation (Spearman's rank correlation, p<0.01). Regression lines are shown for reference.
In this study, we evaluated the relationship between BCG vaccination and SARS-CoV-2 infection in Japanese patients with COVID-19 infection. To the best of our knowledge, this is the first study to demonstrate an impact of routine infant BCG vaccine coverage among younger people, especially those born between 1995 and 2004, on local COVID-19 spread in Japan. The current data are in accordance with the concept of silent spreaders in which asymptomatic or mild cases have a substantial role in the dissemination of SARS-CoV-2.
Moreover, our study findings suggest that infant BCG vaccination has a protective effect against mass infection.
Diagnosis of COVID-19 was made based upon the Infectious Disease Surveillance System in Japan, in accord with the Act on the Prevention of Infectious Diseases and Medical Care for Patients with Infectious Diseases (the Infectious Diseases Control Law). Because in principle, the Act is applicable to individuals with flulike symptoms such as fever, cough, and fatigue, which are suggestive of SARS-CoV-2 infection, all patients officially notified on March 29 were symptomatic. Along with the disease spread, the Act has been applied to asymptomatic ones; the number of official notifications thereafter included PCR+ individuals both with and without clinical symptoms. The number of PCR+ individuals is lower than the actual number because of limited PCR testing in Japan; however, considering the relatively good correlation between the prevalence of PCR+ individuals and the PCR positivity rate, this likely has little involvement in nonparametric analyses (Table 4). Nevertheless, this difference may cause inconsistent results for the case fatality rate among prefectures.
Table 4Prevalence of PCR+ individuals (descending order), number of PCR tests conducted, and positivity rate on April 24 in each prefecture.
(A) Prevalence of PCR+ (per million)
(B) Number of PCR testing (per million)
(C) Positivity ratio (%)
Tokyo
271.09
710.97
38.10
Ishikawa
194.23
1245.84
15.60
Osaka
164.30
782.37
19.20
Toyama
155.24
1795.24
8.60
Fukui
155.04
1696.38
9.10
Chiba
122.78
780.02
14.80
Fukuoka
114.35
1559.23
7.30
Kyoto
110.38
1394.83
7.90
Hyogo
110.32
1178.52
9.40
Hokkaido
106.32
1014.38
10.50
Saitama
104.64
803.41
13.00
Kochi
100.57
1832.86
5.50
Kanagawa
99.38
553.01
16.60
Okinawa
91.85
1184.39
7.80
Gifu
73.11
1137.71
6.40
Gunma
71.72
1246.93
5.80
Shiga
65.86
808.07
8.20
Aichi
62.89
832.16
7.60
Yamanashi
62.42
2352.51
2.70
Yamagata
60.55
1786.24
3.40
Nara
56.76
876.03
6.50
Wakayama
55.61
2701.60
2.10
Ibaraki
54.57
1401.11
3.90
Oita
52.45
2543.71
2.10
Hiroshima
50.41
1504.79
3.30
Saga
39.07
868.13
4.50
Miyagi
36.27
677.03
5.40
Fukushima
35.41
761.80
4.60
Ehime
34.76
741.12
4.70
Nagano
32.48
750.36
4.30
Kagawa
29.11
1483.37
2.00
Niigata
28.94
1109.08
2.60
Tochigi
27.24
859.71
3.20
Mie
25.13
922.95
2.70
Kumamoto
24.47
1492.89
1.60
Shimane
23.53
1102.94
2.10
Yamaguchi
22.63
888.32
2.50
Aomori
17.42
409.34
4.30
Akita
16.31
801.22
2.00
Shizuoka
16.12
646.62
2.50
Miyazaki
15.73
934.32
1.70
Nagasaki
11.93
1288.59
0.90
Okayama
10.54
548.47
1.90
Tokushima
6.79
524.46
1.30
Kagoshima
6.20
687.11
0.90
Tottori
5.36
1658.93
0.30
Iwate
0.00
225.62
0.00
Spearman's rank correlation between (A) and (B): ρ=0.320, p<0.05, between (A) and (C): ρ=0.912, p<0.001.
The significant negative correlation between COVID-19 infection and the population age 65 years and over suggests that in Japan, this population is irrelevant to disease spread. In contrast, the case fatality rate is high among those age 70 years or older (Fig. 4). Acute respiratory distress syndrome subsequent to SARS-CoV-2 respiratory infection is attributable to a high level of inflammatory cytokines, which increase with age, especially in men.
BCG-vaccinated neonates (BCG-Denmark 0.05 mL intradermally within 10 days of birth) showed increased interleukin 6 in an unstimulated state but decreased production of interleukins and chemokines following stimulation with pathogens.
This efficacy possibly contributed to the protection against SARS-CoV-2 spread, infection, and mortality in communities with high BCG vaccine coverage among younger people. In Japan, routine BCG vaccination has been the principal element in tuberculosis control. The incidence of tuberculosis in Japan has drastically decreased from 698.4 in 1951 to 17.7 per 100,000 in 2011.
After the introduction of BCG to the country in 1924, data on tuberculosis prevention and adverse event profiles had been accumulated mainly among young nurses or students.
Group BCG vaccination commenced in 1946 for young people in their teens and 20 s, and in 1949 for all citizens younger than 30 years old; vaccination was repeated each year until positive tuberculin skin test result was obtained.
In 1974, the timing of routine infant vaccination was set at less than 4 years old, followed by revaccination twice during the first year of elementary school and during the first or second year of junior high school after tuberculin skin test screening.
In 1999, a state of emergency concerning tuberculosis was declared because gradual mitigation of tuberculosis prevention measures during previous periods brought about a resurgence of newly infected patients.
The target age of routine infant BCG vaccination was set at younger than 6 months old in 2005 and then at 1 year old in 2013. High COVID-19 mortality among elderly people may be owing to a lack of benefit received from immunization as routine BCG vaccination for infants using qualified freeze-dried BCG commenced in 1949, and BCG vaccination was not administered in those with a positive tuberculin skin test. Current younger generations born in the late 1990s and early 2000s, before establishment of the reinforced tuberculosis control system, have relatively low BCG vaccine coverage. It is possible that their parents, age around 50 years (considering mean age of the mother at the first childbirth was 28 years old in 2000
), represent the significant positive correlation with COVID-19 infection in these age groups.
Fig. 4Number of PCR+ individuals (top), seriously ill patients and deaths owing to COVID-19 (middle), and their proportions (bottom) in each age group on May 1. A drastic increase in case fatality rate was observed with age, especially among individuals age 70 years or older. Shaded squares show age groups born before commencement of routine BCG vaccination for infants using qualified freeze-dried BCG.
Several studies have compared the incidence and severity of COVID-19 among countries with different BCG vaccination policies and found a protective effect of universal BCG vaccination, especially during the past 15 years, and high vaccination coverage in young people under 25 years of age.
; these findings are in accord with those of the present study. Two within-country studies from Israel and Germany failed to demonstrate the relevance of BCG vaccination to COVID-19 prevention; however, those studies do not contradict our study findings considering that universal immunization was discontinued in these countries in 1982 and 1990, respectively, and that the number of asymptomatic silent spreaders could not be evaluated.
In summary, the current study demonstrated that routine infant BCG vaccine coverage in young people showed a protective effect against local COVID-19 spread in Japan. The possible relevance of infant BCG vaccination to high mortality among elderly patients with COVID-19 should be elucidated.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Statistics Bureau, Ministry of Internal Affairs and Communications of Japan. e-Stat, Portal Site of Official Statistics of Japan. https://www.e-stat.go.jp. Accessed 24 April2020.
30547-8&id=gr1.jpg){kind=link}
30547-8&id=gr2.jpg){kind=link}
30547-8&id=gr3.jpg){kind=link}
30547-8&id=gr4.jpg){kind=link}