In August 2012, reports emerged that the Japanese government planned to add Haemophilus influenzae type-b (Hib) vaccine, 7-valent pneumococcal conjugate vaccine (PCV7), and human papillomavirus vaccine to the routine schedule in April 2013. Japan moved from its decades-old conservative vaccine policy toward the active vaccine policy. The milestone event was approval of the Hib vaccine in 2007, after nearly 10 years of discussions between the manufacturer and the Japanese government. Not long after the Hib vaccine was approved, PCV7 was approved in 2009, and human papillomavirus vaccines were approved in 2010 and 2011.
Japan is a relatively small, highly populated country with a universal healthcare system that allows all citizens to access any healthcare facility, regardless of place of residence. People often cross prefecture borders to visit and be admitted to hospitals located outside of their place of residence. In addition to these facts, Japan has no centralized disease database; therefore, it is very difficult to obtain the population-based incidence data that serve as the basis for vaccine introduction. Our study area Hokkaido Prefecture is an island, thus one of very few places in Japan where population-based studies are possible.
We conducted a comprehensive prospective study of invasive bacterial diseases among children aged 28 days to 5 years who were admitted to hospitals in Hokkaido Prefecture. All hospitals in the study area with a pediatric admission facility participated in the study. Only children whose blood or cerebrospinal fluid culture was positive and who legally resided in the study area were included in the study. Children clinically diagnosed with meningitis or bacteremia but who had negative culture results were excluded, irrespective of prior antibiotic use. Polymerase chain reaction diagnosis was available for culture-negative cases.
The study of bacterial meningitis throughout Hokkaido Prefecture began on January 1, 2007, after the institutional review boards of all 64 participating institutions had approved the protocol. Doctors were asked to immediately ship all bacteria cultured from cerebrospinal fluid to Kitasato University for serotyping using Transystem transport swabs (Copan Italia S.p.A, Brescia, Italy), which allow for maintenance of bacterial viability at room temperature. Between January 1, 2008, and December 31, 2010, we conducted a study of bacteremia in Eastern Hokkaido. We selected this area because it is geographically isolated from the rest of the prefecture by the Tokachi-Taisetsu Mountains, and therefore, the data remain population-based. Streptococcus pneumoniae specimens were preserved frozen in Microbank containers (Pro-Lab Diagnostics, Richmond Hills, ON, Canada) and shipped periodically to the Nagasaki University Institute of Tropical Medicine for serotyping. The bacteremia study protocol was approved by the Clinical Study Review Board at the Graduate School of Hokkaido University School of Medicine on December 21, 2007. The meningitis/bacteremia target populations were 206,910/35,244 in 2008, 204,247/35,035 in 2009 and 203,366/36,117 in 2010.
There were 87 admissions for bacterial meningitis among children aged 28 days to 5 years between January 1, 2007, and December 31, 2011, across Hokkaido Prefecture. Of these 87 cases, 59 cases were caused by H. influenzae (Hi) and 19 cases by S. pneumoniae. The calculated incidence of bacterial meningitis was 8.48, Hib meningitis was 5.65 and pneumococcal meningitis was 1.85 per 100,000 person-years in Hokkaido Prefecture. Fifteen of the 19 pneumococcal strains were serotyped. Reported serotypes were 6B (5), 23F (3), 19F (2), 14 (2), 6C (2) and 19A (1). The overall serotype coverage of PCV7 was 80% and PCV13 was 87%.
Between 2008 and 2010, 101 admissions for invasive bacterial diseases were reported in Eastern Hokkaido. There were 92 cases of bacteremia and 11 cases of bacterial meningitis; 2 cases of meningitis occurred with bacteremia. S. pneumoniae infection accounted for 70% (64 cases), whereas Hi infection accounted for 22% (21 cases). Before introduction of the Hib vaccine and PCV7, the average annual incidence per 100,000 children aged 28 days to 5 years was 95.87 for invasive bacterial disease, 87.41 for bacteremia and 10.34 for meningitis in Eastern Hokkaido. The average incidence of bacteremia by pathogen was 60.15 (S. pneumoniae) and 18.80 (Hi). We serotyped 45 of 64 S. pneumoniae strains during the study period. Serotypes identified were 23F (12), 6A (10), 6B (8), 14 (5), 9A (3), 19A (2), 19F (1), 22F (1), 23A (1), 28A (1) and 9L (1). The average annual serotype coverage of PCV7 for bacteremia was 58% and PCV13 for bacteremia was 84%. Serotypes included only in PCV13 accounted for 26% (12/45) of all strains serotyped.
We presented the first population-based incidence data of vaccine preventable invasive bacterial diseases in Japan from a study conducted on the island of Hokkaido in this article. Before vaccines were introduced, the incidence of invasive Hi diseases in children <5 years old in Hokkaido Prefecture was 23, comparable with that in Europe (France: 21/100,000; Spain: 12/100,000),1,2 where data primarily reflect hospital admissions. In contrast, the incidence of invasive pneumococcal diseases in Hokkaido was 63, comparable with that in the United States (54.7/100,000).3 It is said that despite a similar socioeconomic status, the incidence is much higher in the United States than in Europe, presumably due to differences in blood culture practice.
To our knowledge, Sakata4,5 conducted the only other population-based studies in Japan, prospectively estimating the incidence of bacterial meningitis among children <5 years old at 6.3/100,000 in 1993 to 2005 and retrospectively estimating the incidence of pneumococcal bacteremia at 30.95 in 1997 to 2004 in Hokkaido. Our estimated incidence in Hokkaido was almost double that of Sakata.
Recently, a population-based study using the same protocol used in our study was conducted in Okinawa Prefecture. Okinawa and Hokkaido are the only island prefectures in Japan. Okinawa Prefecture is subtropical islands situated at the southernmost end of the Japanese archipelago, whereas northernmost Hokkaido is subarctic. The rest of Japan is temperate. Despite some climatic and cultural differences, the socioeconomic status of both prefectures is similar to that of the rest of Japan. Compared with Hokkaido, the incidence of pneumococcal diseases in Okinawa was 2.5–5 times higher for meningitis and 1.5 times higher for bacteremia, although there was no significant difference overweighing the annual fluctuations regarding Hib/Hi diseases (Table 1).6,7
The difference may be due to the climatic difference partly, but probably more to the differences of antibiotics use and blood culture practices, and its consequences. In outpatient clinics in Japan, antibiotics are commonly prescribed without blood cultures, especially in areas with few large medical facilities, such as Eastern Hokkaido. However, a blood culture is generally performed when the patient is admitted for acute febrile syndrome. We assume the frequent use of antibiotics differently affects the epidemiology, antibiotic resistance and possibly the progress of pneumococcal and Hib/Hi diseases.
The same assumption applies to differences in incidence between Japan and other developed countries. Compared with other developed countries, invasive pneumococcal diseases were more detectable than invasive Hib/Hi diseases in Hokkaido. This trend is more prominent when looking only at meningitis. The incidence of pneumococcal meningitis was 0.97–1.97 in Hokkaido, whereas that is 2.1–7.0 in Europe8 and 3.6 in the United States.8 The incidence of Hib meningitis was 6.02 in Hokkaido, whereas that is 12–22 in Europe9 and 16–30 in the Americas.9 Compared with other developed countries, the incidence of Hib meningitis is much lower in Hokkaido, but a substantial burden of this fatal disease was observed. Again, climate/geography may play a role in the relative differences in pneumococcal and Hib incidence, but it may also be due to frequent antibiotic use.
The reported incidence of pneumococcal bacteremia in children <5 years old in prefectures other than Hokkaido and Okinawa in 2008 to 2010 was 6.5–26.16,7 and that of invasive Hib/Hi diseases was 11.2–24.2.6,7 In Hokkaido, the invasive pneumococcal disease incidence was 2- to 5-fold higher and invasive Hib diseases was higher than other prefectures, with the exception of some of the more heavily populated prefectures that might have admitted patients from neighboring prefectures. Differences in the data from the mainland and island prefectures most likely arise from the considerable flow of patients across prefectural borders. Many bacteremia cases also likely go unreported in some areas.
At the end of 2011, the estimated immunization coverage of the Hib vaccine in Hokkaido Prefecture by the manufacturers was 76% and that of PCV7 was 90%, but we saw no significant reduction in incidence of meningitis. However, from January 1 to October 31, 2012, only 1 case of bacterial meningitis (by group B Streptococcus) was reported to us. According to the Hokkaido Prefecture government, utilization of the prefecture’s reimbursement system indicates that as of April 2012, over 95% of children <5 years old had been vaccinated with at least 1 dose of Hib vaccine and/or PCV7 (unpublished data). Although it is too early to state any evidence-based conclusions, we expect that the Hib vaccine and PCV7 will prove highly beneficial in Japan and believe it is important to strengthen surveillance to monitor their impact.
Note: The immunization coverage of PCV7 provided by the manufacturers is calculated as follows:
Vaccination coverage = number of vaccines sold in the area / number of children <5 year old.
1. Georges S, Lepoutre A, Dabernat H, et al. Impact of Haemophilus influenzae type b vaccination on the incidence of invasive Haemophilus influenzae disease in France, 15 years after its introduction. Epidemiol Infect. 2013;21:1–10
2. Watt JP, Levine OS, Santosham M. Global reduction of Hib diseases: what are the next steps? Proceedings of the meeting Scottsdale, Arizona, Sep 22–25, 2002. J Pediatr. 2003;143(suppl 6):S163–S187
3. Black SB, Shinefield HR, Hansen J, et al. Postlicensure evaluation of the effectiveness of seven valent pneumococcal conjugate vaccine
. Pediatr Infect Dis J. 2001;20:1105–1107
4. Sakata H. A study of bacterial meningitis in Hokkaido between 1999 and 2003. Kansenshougaku Zasshi. 2005;79:680–687
5. Sakata H. Clinical study of Streptococcus pneumoniae bacteremia in children. Kansenshougaku Zasshi. 2005;79:1–6
6. Kamiya H. Research Report 2010: Evidence to improve the usefulness of vaccines; Labor and Health Sciences Research Grant. 2010
7. Kamiya H, Ihara T. Research Report 2011: Basic and clinical research of the safety, effectiveness and dosing of newly developed Hib, pneumococcal, Rota virus, and HPV vaccines, Labor and Health Sciences Research Grant. 2011
8. Hausdorff WP, Siber G, Paradiso PR. Geographical differences in invasive pneumococcal disease
rates and serotype frequency in young children. Lancet. 2001;357:950–952
9. Watt JP, Wolfson LJ, O’Brien KL, et al.Hib and Pneumococcal Global Burden of Disease Study Team. Burden of disease caused by Haemophilus influenzae type b in children younger than 5 years: global estimates. Lancet. 2009;374:903–911