Although Bacille Calmette-Guréin (BCG) vaccine is the only currently available vaccine against tuberculosis, polices for BCG vaccination vary between countries. Many countries use the intradermal method of administration for BCG vaccination recommended by World Health Organization (WHO) and United Nations Children's Fund (UNICEF), while some countries such as Japan (and South Africa until 2002, South Korea, Brazil) use the percutaneous method of administration with a multipuncture technique. Brazil was the last country to use oral administration of BCG vaccine until this was discontinued in 1973. Other techniques such as scarification, jet injection and use of bifurcated needles have been tried, but have yielded highly variable and inadequate results mainly because of inconsistent dose delivery, adverse reactions and low tuberculin conversion rate1. In South Korea, the policy is to give BCG vaccination within the first month of birth along with revaccination given to some elementary children aged 6 to 7 years if they have no visible BCG scar without prior tuberculin testing (until the revaccination policy was stopped in 2008). All the BCG vaccine needed in the nation had been produced by the Korean Institute of Tuberculosis (KIT) until the year 2007. However as per Korea's liberalized import action since the 1990s, imported BCG vaccines, such as the BCG Tokyo vaccine given by multipuncture device, have been also introduced to private health care providers. Use of the BCG Tokyo vaccine delivered by multipuncture device has been constantly increasing because it is known to be less scarring and to have lower rates of side effects compared to the BCG Pasteur vaccine given by the intradermal method2,3. Also, the private sector prefers to use the imported vaccines not only due to the higher profits, but also due to the easier means of administration which does not require special training.

Percutaneous administration methods are simpler with a low rate of adverse reactions but are less consistent in terms of delivery dose4. Intradermal injection using a syringe and needle is generally accepted as the most accurate method, because the delivered dose can be precisely measured and controlled, resulting in more consistent mycobacteria-specific immunity5,6. The downside of using the intradermal method is a higher rate of local reactions such as ulcers and lymphadenitis than other methods when other vaccine specific factors are controlled5. The implications of intradermal versus percutaneous administration routes have long been debated but, the relative protective efficacy of the two is not yet known. However, there have been some studies that compared the effect of vaccine route on T cell immunity against tuberculosis. Kemp et al.6 demonstrated that Mycobacterium tuberculosis-specific interferon-γ (IFN-γ) production was increased after the intradermal route of BCG vaccination but not by the percutaneous route in healthy volunteers. Also, Hussey et al.7 in South Africa demonstrated that there was a trend for greater IFN-γ responsiveness following the intradermal route of vaccination compared with percutaneouly administered BCG vaccine. On the other hand, in contrast with those of Kemp et al.6, a recent study in South Africa found that percutaneous vaccination with Japanese BCG induced significantly greater helper T cell (Th)-1 cytokines in whole blood of neonates than did intradermal vaccination with Danish BCG8. These studies with different conclusions suggest that the vaccine route can affect M. tuberculosis-specific T cell immunity in terms of IFN-γ and other T cell-derived cytokines, but it might be important to determine its actual effect on the variable vaccine efficacy and clinical consequences of vaccination in particular populations.

Therefore, it is necessary to investigate the current status of BCG vaccination in young children who have been affected by the diversity of BCG vaccines in use in the public and private sector in South Korea. Using the data from this BCG scar survey, we can evaluate whether proper BCG vaccination delivered by the two different administration methods has been performed in the nation. Findings from this study will provide baseline data to health policy decision makers to assist in evaluating the performance of the existing BCG vaccination program and to provide a better understanding of the BCG vaccines recently used in South Korea.

From this scar survey, 4.4% of responders had no BCG scar. Based on the national tuberculosis survey in 1995, the rate of BCG vaccination under 10 years old was 87.7%9. Comparing these data, it seems the rate of BCG vaccination has improved since 1995. However, the reporting rate of BCG vaccination has been slightly decreasing, taking into consideration the figures of 80.3% in 2002, 77.8% in 2003, and 77.7% in 2004. A lot of individuals on whom data were collected may not have been properly reported to the surveillance program, which has no system of collecting data from all the private clinics. This suggests that the current reporting system should be examined and improved so as to include information on all vaccinated children.

From the data for the age of 0 to 6 years, 64.5% of responders had been vaccinated with BCG Tokyo/multipuncture. It appears that only 11.8% of responders in pediatric clinics and 14.4% of responders in obstetrics/gynecology clinics, had vaccination with BCG Pasteur/intradermal, showing that most of the private clinics prefer to use the BCG Tokyo vaccine. It seems the use of BCG Tokyo by the multipuncture method in private sector has been gradually increasing compared to that of BCG Pasteur given by the intradermal method. In addition to an easier manner of administration, less scarring and higher profits to the providers, the private sector may also favor using the BCG Tokyo strain, considering the fact that BCG Tokyo strain is known to be one of the 'weak' BCG strains with lower frequencies of side effects such as BCG lymphadenitis than 'strong' strains such as BCG Pasteur and BCG Danish10,11.

As the multipuncture method uses a device with 9 needles twice, a maximum of eighteen BCG scars should remain in the skin when properly injected, but previous studies have found that typically only 9 to 11 BCG scars remained a few years after vaccination12,13. In this survey, the data showed that the median of pin number scars was 16 (IQR, 12~18) among 3,684 BCG Tokyo recipients at the age of 0~6 years. In children less than 12 months old, the median of pin number was 18 (IQR, 15~18), while the median of pin number in 6 years old was 16 (IQR, 11~18), showing a decreasing trend of pin scars over time. This may provide further evidence of the disappearance of pin scars over time, although we cannot discount the possibility that as the use of BCG delivered with the multipuncture device increased over time, the injection technique improved, which may have resulted in increased numbers of pin scars in younger children.

The findings in the survey clearly showed a growing preference of parents for the BCG Tokyo vaccine by the multipuncture method in South Korea. This suggests that special training and education to ensure accurate injection in the private sector will have to be planned by the government.