Clin Exp Vaccine Res. 2020 Jul;9(2):179-182. English.
Published online Jul 31, 2020.
© Korean Vaccine Society.
Brief Communication

Current national policies for infant universal bacille Calmette-Guérin vaccination were associated with lower mortality from coronavirus disease 2019

Risa Ebina-Shibuya,1 Nobuyuki Horita,2 Ho Namkoong,3 and Takeshi Kaneko2
    • 1Laboratory of Molecular Immunology, Immunology Center, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, MD, USA.
    • 2Department of Pulmonology, Yokohama City University School of Medicine, Yokohama, Japan.
    • 3Laboratory of Clinical Immunology and Microbiology, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD, USA.
Received April 19, 2020; Revised April 19, 2020; Accepted July 28, 2020.

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

An exciting debate has emerged whether bacille Calmette-Guérin (BCG) vaccination is effective for the coronavirus disease 2019 (COVID-19) pandemic. Some advocated that BCG-vaccinated people are less suffered from the virus because BCG vaccination is recommendedin COVID-19 high burden countries. However, the others objected because this seemingly attractive relationship is explainable with confounding factors. In a multiple regression with 171 countries adjusting socioeconomical and climatic covariates, countries with current universal pediatric BCG policy were associated with 30-fold (95% confidence interval, 17–52) decrease of COVID-19 mortality per population compared to countries without the policy.

Keywords
Vaccine; Viruses; Pneumonia; Epidemiology

As of April 10th, 2020, 1.7 million population has been diagnosed with coronavirus disease 2019 (COVID-19) caused by a novel coronavirus, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and more than 100 thousand died from it [1]. Both numbers of infected cases and death cases are still increasing exponentially though many cities around the world were locked down to prevent further disaster. An exciting debate has emerged whether bacille Calmette-Guérin (BCG) vaccination is effective for the pandemic. Some advocated that BCG-vaccinated people are less suffered from the virus because COVID-19 high burden countries such as Italy, Spain, USA, and France do not currently recommend universal BCG vaccination [2, 3, 4]. However, the others objected because this seemingly attractive relationship is explainable with confounding factors such as aged society, resource rich medical system to detect COVID-19, lower temperature, and advanced reporting system of COVID-19 high-burden tuberculosis (TB) low-burden countries [5, 6]. We assessed if country-level universal BCG vaccination policy was associated with decreased mortality from COVID-19 after adjusting multiple possible confounding factors.

We selected the COVID-19 mortality per population as a dependent variable because COVID-19 mortality per infected cases and numbers of infected and death cases are easily affected by a number of polymerase chain reaction test conducted and a population size of each country. Gross national income per capita, life expectancy at birth, and infant mortality during the first year after the live birth were selected as independent variables because these parameters reflect how each society is developed, which indirectly explains stature of medical system. Life expectancy at birth was meaningful also because higher mortality from COVID-19 were observed for the elderly. Population density was adopted because this may associate with social distance. Annual average temperature was included as an explanatory variable because confirmed cases and death cases were rare in tropical countries near the equator.

Countries and regions were divided into currently recommended countries (CRC) and currently not recommended countries (CNRC) based on the national policy for the vaccination.

Following data sources were used: (1) number of death cases as of April 10, 2020 from each countries and region: Worldometer “COVID-19 Coronavirus Pandemic” [1]; population size and population density: United Nations “World Population Prospects 2019” [7]; national BCG vaccination policy: “The BCG World Atlas” [6]; gross national income per capita and life expectancy at birth: United Nations Development Programme “Human Development Report 2019” [8]; and infant mortality out of live births and annual average temperature: the World Bank “Data Catalog” [9].

Mann-Whitney test was used to compare continuous variables in two groups. Multiple regression analysis was used to adjust covariables. COVID-19 mortality per population, gross national income per capita, infant mortality, and population density were log10 transformed before the regression analysis because these variables were long tailed. A half of the minimal mortality next to null was substituted instead of null mortality before log10 transformation. BellCurve 2015 (SSRI, Tokyo, Japan) and GraphPad PRISM ver. 7.0 (GraphPad Software, San Diego, CA, USA) were used.

Out of 189 countries and regions presented on Human Development Report 2019, 18 countries were excluded from our analysis due to lack of description of national policy on the BCG World Atlas. One hundred and seventy-one analyzable countries consisted of 146 CRC, most of which recommend the vaccine for infant, and 25 CNRC. Median country level mortalities from COVID-19 per 106 population were 0.42 (interquartile range [IQR], 0–2.02) for CRC and 35.4 (IQR, 11.1–132.0) for CNRC (Fig. 1). This great discrepancy was compatible with previous reports [2, 3, 4]. However, CNRC was also significantly associated with higher gross national income per capita (8 [IQR, 4–16] versus 44 [IQR, 35–49] 1,000 US dollar [USD]; p<0.001), longer life expectancy at birth (72 [IQR, 66–76] versus 82 [81–83] years; p<0.001), lower infant mortality (19.2 [IQR, 8.8–37.4] versus 3.1 [IQR, 2.6–3.7] per 1,000 live births; p<0.001), and lower annual average temperature (24 [IQR, 15–26] versus 9 [IQR, 7–11] ℃; p<0.001) (Fig. 1).

Fig. 1
Univariable comparisons of the mortality from COVID-19 between CRC and CNRC may be confounded by several factors. A box and whiskers indicate interquartile range and 5%–95% range. Median with interquartile range are described over the box and whisker plot. A p-value for Mann-Whitney test is shown. (A) Mortality from COVID-19 per population. (B) Gross national income per capita. (C) Life expectancy at birth. (D) Infant mortality among live birth. (E) Population density. (F) Annual average temperature. COVID-19, coronavirus disease 2019; CRC, currently recommended countries; CNRC, countries not recommended countries; IQR, interquartile range; USD, US dollar.

In a multiple regression analysis using the six parameters as independent variables, CNRC was still related to log10 mortality from COVID-19 (partial regression coefficient, 1.48; 95% confidence interval [CI], 1.24 to 1.72; p<0.001), which is equivalent to 30-fold (95% CI, 17–52) increase of COVID-19 associated mortality per population in CNRC compared to CRC (Table 1). Annual average temperature was another covariable with significant association (p=0.009) (Table 1). Ten degrees Celsius increment of annual average temperature could explain 0.64-fold (95% CI, 0.46 to 0.90) decrease of COVID-19 associated mortality per population (Table 1).

Table 1
Multiple regression analysis to predict the coronavirus disease 2019 mortality in general population

CNRC was further divided into countries that had previously recommended the vaccination and that had never recommended the vaccination [6]. The mortality from COVID-19 did not significantly differ (p=0.400) between countries wherein BCG had previously recommended (N=19; median, 32.7; IQR, 11.1–101.0; per 106 population) and those wherein BCG had never been recommended (N=6; median, 101.6; IQR, 25.5–232.0).

Possible cross-immunization between BCG vaccination and SARS-CoV-2 is a serious concern around the world because it is far from easy to develop a novel vaccine that specifically targets SARS-Cov-2. Biological rationale of this hypothesis is trained immunity by non-specific cross-protection against non-TB pathogens. The large social, economic, and climatic inconsistency between CRC and CNRC made it difficult to derive the conclusion from a cross-sectional study. To date, no peer-reviewed report has presented data that were adjusted for multiple possible confounders. Insufficient adjustment of covariable factors may lead to biased result. Based on our data, infant universal BCG vaccination prevents death from COVID-19.

We need to comment a few limitations. First, our data is preliminary until confirmed by randomized trial because unknown confounders including biological, medical, and social factor may exist. Second, we do not recommend BCG vaccination for adult and children of school age because our data is based on the current policy that recommends the vaccination mainly for infant. Development of novel vaccination for adult is still anticipated. We would like people not to rash for BCG vaccination because this may lead to a supply shortage for infants, increasing their chances of TB meningitis and miliary TB [10].

In conclusion, population based COVID-19 mortality was lower in countries with universal infant BCG vaccination policy. However, a univariable analysis for this topic introduces a strong bias because COVID-19 high-burden countries are TB low-burden developed countries. Based on our analysis of 171 countries and regions, the vaccination policy linked to 30-fold lower mortality after adjusting five socioeconomical covariates.

Notes

No potential conflict of interest relevant to this article was reported.

References

    1. Worldometer. Worldometer COVID-19 coronavirus pandemic [Internet]. [place unknown]: Worldometer; 2020 [cited 2020 Apr 10].
    1. Miller A, Reandelar MJ, Fasciglione K, Roumenova V, Li Y, Otazu GH. Correlation between universal BCG vaccination policy and reduced morbidity and mortality for COVID-19: an epidemiological study [Internet]. Huntington, NY: medRxiv, Cold Spring Harbor Laboratory; 2020 [cited 2020 Apr 11].
    1. Sala G, Miyakawa T. Association of BCG vaccination policy with prevalence and mortality of COVID-19 [Internet]. Huntington, NY: medRxiv, Cold Spring Harbor Laboratory; 2020 [cited 2020 Apr 11].
    1. Berg MK, Yu Q, Salvador CE, Melani I, Kitayama S. Mandated Bacillus Calmette-Guerin (BCG) vaccination predicts flattened curves for the spread of COVID-19 [Internet]. Huntington, NY: medRxiv, Cold Spring Harbor Laboratory; 2020 [cited 2020 Apr 11].
    1. Fukui M, Kawaguchi K, Matsuura H. Does TB vaccination reduce COVID-19 infection?: no evidence from a regression discontinuity analysis [Internet]. Huntington, NY: medRxiv, Cold Spring Harbor Laboratory; 2020 [cited 2020 Apr 11].
    1. The BCG Atlas Team. The BCG World Atlas: a database of global BCG vaccination policies and practices: 2nd edition [Internet]. [place unknown]: The BCG Atlas Team; 2017 [cited 2020 Apr 10].
      Available from: http://www.bcgatlas.org/.
    1. Department of Economic and Social Affairs, Population Dynamics, United Nations. World population prospects 2019 [Internet]. New York, NY: United Nations; 2019 [cited 2020 Apr 10].
    1. United Nations Development Programme. Human Development Report 2019: beyond income, beyond averages, beyond today: inequalities in human development in the 21st century [Internet]. New York, NY: United Nations Development Programme; 2019 [cited 2020 Apr 10].
    1. The World Bank. Data Catalog [Internet]. Washington, DC: The World Bank; 2020 [cited 2020 Apr 10].
    1. Namkoong H, Horita N, Ebina-Shibuya R. Concern over a COVID-19-related BCG shortage. Int J Tuberc Lung Dis 2020;24:642–643.

Metrics
Share
Figures

1 / 1

Tables

1 / 1

ORCID IDs
PERMALINK