国立感染症研究所

Pertussis is caused by Bordetella pertussis.  It is an acute respiratory infection mainly affecting children.  The main symptom is protracted cough.  In Japan, children are recommended to receive four shots (including one supplementary injection) of pertussis-containing vaccine in their infancy.  Instead of “adsorbed diphtheria-tetanus-acellular pertussis (DTaP) combined vaccine”, inactivated poliovirus vaccine (IPV)-containing DTaP, DTaP-IPV, was introduced since November 2012 (see p. 323 of this issue).  It has been found recently that acquired immunity wanes in 4-12 years after vaccination, which permits infection among children and adults who were already vaccinated.  As being often asymptomatic, the infected adolescent and adults transmit the bacteria to unvaccinated infants whose infection tends to become severe.  This trend is a challenge to many developed countries. 

Incidence: Pertussis is a category V infectious disease to be reported by sentinel clinics under the National Epidemiological Surveillance of Infectious Diseases (NESID).  Clinical cases are reported every week from approximately 3,000 pediatric sentinels all over the country (criteria for reporting: http://www.mhlw.go.jp/bunya/kenkou/kekkaku-kansenshou11/01-05-23.html).  While weekly incidence was low during 2001-2007 (<0.040/sentinel), it increased in 2008 with a peak (0.115/sentinel) in week 20/May (Fig. 1).  Pertussis epidemic is known to occur at an interval of about four years (IASR 26: 61-62, 2005).  Year 2008 corresponded the epidemic year as the previous epidemics occurred in 1999-2000 and 2004.  The epidemic started in 2008 continued for three years until 2010.  

The 2008-2010 epidemic was nationwide (Fig. 2).  In 2007 one year before the epidemic, prefectures reporting ≥2.00 cases/sentinel/year were only two, Chiba and Tochigi.  After the onset of the epidemic, prefectures reporting ≥2.00 cases/sentinel/year increased to 16 in 2008, 7 in 2009 and 11 in 2010.  

Immunological status of Japanese population: National Epidemiological Surveillance of Vaccine-Preventable Diseases (NESVPD) conducted sero-prevalence of anti-pertussis antibodies (anti-pertussis toxin and anti-filamentous hemagglutinin antibodies) among people in all age groups in 2003 and 2008.  In 2008, antibody-positive rate was about 80% among children aged 6 months to 2 years, which was higher than in 2003, and early acquisition of immunity among this age group was confirmed (Fig. 5).  Meanwhile, sero-prevalence rates among other age groups remained as low as 50% from 2003 to 2008. 

Outbreaks: Japan experienced large scale pertussis outbreaks in universities and other facilities in 2007, which reconfirmed the easiness of pertussis transmission in enclosed spaces occupied by people for a long time (IASR 29: 65-66, 2008).  Later than 2008, outbreaks continued to occur in nursery schools and junior high schools (IASR 29: 201-202, 2008, and see pp. 325 & 326 of this issue).  In addition, local epidemics involving mainly primary and junior high school children occurred (IASR 32: 340-341, 2011, and see pp. 327 & 329 of this issue). 

Pathogens that cause pertussis-like clinical symptoms:B. pertussis-related microorganisms that cause cough and other pertussis-like symptoms are B. parapertussis and B. holmesii.  B. holmesii is a new species identified by US CDC in 1995.  In Japan, B. holmesii has been isolated since late 2000’s from pericarditis and “pertussis” patients (IASR 33: 15-16, 2012 and see pp. 329 & 332 of this issue).  

Microorganisms other than Bordetella that cause pertussis-like symptoms are Mycoplasma pneumoniae, Chlamydia pneumoniae and human bocavirus.  Rhinovirus primarily responsible of common cold was detected from a suspected pertussis outbreak in a medical college and its affiliated hospital in Japan, which warned necessity of differential diagnosis between B. pertussis and rhinoviruses (IASR 32: 234-236, 2011).  From a pertussis outbreak in a nursery school in 2012, rhinovirus and coxsackievirus A9 were detected together with B. pertussis, which indicates occasional double infection of these pathogens in children (see p. 326 of this issue).  

Laboratory diagnosis of pertussis: For laboratory diagnosis of pertussis, bacterial isolation, serological test, and gene detection are applicable.  In Japan, the bacterial agglutination test is widely used as a simple serological test.  However, it is of low precision, and its application to adult or vaccinated child cases is not always appropriate (IASR 32: 236-237, 2011).  Titration of anti-pertussis toxin IgG antibody can be used but is not useful as a rapid test, because IgG increases one week or later after the onset of cough.  Bacterial isolation is extremely difficult particularly from adolescent or adult patients whose bacterial load is low, on account of limited bacterial growth.  The genetic testing can detect B. pertussis genome with a high sensitivity and widely being used as rapid detection method in the United States and European countries.  In Japan, the loop-mediated isothermal amplification (LAMP) is only used by the prefectural and municipal public health institutes (PHIs) and some medical institutions for research purposes (IASR 33: 104-105, 2012).  

For B. holmesii, both bacterial isolation and genetic testing are feasible (see p. 330 of this issue), but only genome sequencing can give definitive diagnosis of the pathogen.  Therefore, its presence may have been overlooked in clinical settings.  National Institute of Infectious Diseases Japan (NIID) is currently developing a LAMP method specifically detecting B. holmesii as a part of its NIID-PHI joint activities in strengthening national laboratory pathogen surveillance system.

Additional comments: Pertussis cases are increasing among adults in developed countries including Japan.  In Japan, pertussis epidemic still persists among secondary school children and also in communities.  As clinical diagnosis of pertussis particularly among adolescents and adults is difficult, genetic testing instead of bacterial culture should be used more widely as in the United States and European countries.  Introduction of simple genetic testing with high accuracy is needed not only for surveillance but also for evaluating vaccine efficacy in Japan. 

The 2011/12 season's influenza epidemic (from week 36/September of 2011 to week 35/August of 2012) was caused mainly by subtype AH3 and to lesser extent by influenza virus type B.  Influenza A(H1N1)pdm09 that dominated in the 2009/10 epidemic and occupied about a half of the influenza virus isolates in 2010/11 season was rare after April 2011.

Incidence of Influenza: Under the National Epidemiological Surveillance of Infectious Diseases (NESID), 5,000 influenza sentinels (3,000 pediatric and 2,000 internal medicine clinics) report diagnosed influenza cases at weekly basis.  In the 2011/12 season, the epidemic index (number of cases/sentinel) became above 1.0 in week 49 nationwide, and the level was maintained for 22 weeks till week 18 of 2012.  The epidemic attained its peak in week 5 of 2012 with the incidence of 42.7 cases/sentinel (Fig. 1), which was the second highest in the past 10 seasons following the peak observed in 2004/05 season (50.1 cases/sentinel) (http://www.niid.go.jp/niid/en/10/2096-weeklygraph/2572-trend-week-e.html).  The cumulative number of cases per sentinel of this season, 342.5, was also the second highest in the past 10 seasons following 415.4 in 2009/10 season.

At prefecture levels, the epidemic index exceeded 10.0 first in Miyagi and Aichi in week 50 of 2011.  The number of prefectures with the epidemic index exceeding 10.0 increased to 12 in week 2 of 2012 and then to 42 in week 3 resulting in the nationwide influenza epidemic (https://nesid3g.mhlw.go.jp/Hasseidoko/Levelmap/flu/index.html).

The total number of patients who visited medical institutions, which was estimated from sentinel site reports, was about 16,480,000 from week 36 of 2011 to week 18 of 2012 (September 6, 2011-May 6, 2012).  According to the hospitalization surveillance that started in September 2011, total 11,118 patients were hospitalized in the “designated sentinel hospitals” (about 500 hospitals with more than 300 beds), among which 1,487 were clinically severe (http://www.mhlw.go.jp/bunya/kenkou/kekkaku-kansenshou04/pdf/120525-01.pdf).

Isolation/detection of influenza virus: In 2011/12 season, the prefectural and municipal public health institutes isolated total 5,457 influenza viruses (as of October 18, 2012, Table 1).  In addition, there were 1,799 influenza virus detections by PCR alone.  Among the total 7,256 isolated/PCR-detected viruses, 5,755 were derived from influenza sentinels and 1,501 from elsewhere (Table 2).

Influenza viruses of the 2011/12 season consisted mainly of subtype AH3 (71%) and type B (28%).  AH1pdm09 were few (0.2%).  Former seasonal AH1 subtype virus has not been isolated since week 36 of 2009.  Influenza B viruses consisted of Victoria and Yamagata lineages, whose isolation/detection ratio was 2:1.  Viruses isolated/detected from overseas travelers were subtype AH3 (25 cases), type B (5 cases) and AH1pdm09 (2 cases) (Table 2).

Nationwide, subtype AH3 was predominant among the isolates from the beginning till week 9 of 2012, well after the epidemic peak, when type B influenza started to exceed subtype A (Fig. 1　and Fig. 2).  In Okinawa Prefecture, however, the epidemic did not fade; the patients further increased from week 26 of 2012.  The AH3 epidemic in this region lasted from June to September of 2012 (IASR 33: 242, 2012).

As for age distribution of the patients, the peak age was invariably 5-9 years for subtype AH3, B/Victoria and B/Yamagata lineages (Fig. 3 and Fig. 4).

Antigenic characteristics of the 2011/12 season isolates and their drug resistance (see p. 288 of this issue): Of the eight AH1pdm09 isolates tested, six were antigenically similar to A/California/7/2009, the vaccine strain for 2009/10-2012/13 seasons.  Subtype AH3 isolates was antigenically similar to A/Victoria/361/2011 (the 2012/13 season vaccine strain), whose antigenicity was slightly changed from that of A/Victoria/210/2009 (the 2010/11-2011/12 season vaccine strain used in Japan).  Antigenicity of Victoria lineage isolates that occupied 2/3 of all the type B isolates was similar to that of B/Brisbane/60/2008 (the 2009/10-2011/12 season vaccine strain), and that of Yamagata lineage that occupied 1/3 of the type B isolates was similar to that of B/Wisconsin/1/2010 (2012/13 season vaccine strain).

None of the nine AH1pdm09 isolates tested had the H275Y mutation implicated in the oseltamivir resistance, while 2.0% of the tested 2010/11 season isolates had the H275Y mutation.  Among 278 subtype AH3 isolates tested, only one had R292K mutation attributable to oseltamivir/peramivir resistance (http://www.niid.go.jp/niid/en/iasr-inf-e.html#Antiviral).

Immunological status of Japanese population: According to the data of National Epidemiological Surveillance of Vaccine-Preventable Diseases (see p. 294 of this issue) that was obtained with serum samples collected from July to September in 2011, frequency of anti-A/California/7/2009pdm09 HI antibody positives (titer higher than 1:40) was average 49%.  The antibody positive rate was relatively high for 5-24 years of age (64-78%).  For age groups 0-4 years and 50 years or older, positive rates were 24-38%, which were higher than in 2010 (blood samples collected during July-September).  Antibody positive populations for subtype AH3 and B/Victoria lineage were average 50% and 45%, respectively, and were highest among age group of 15-19 years (68% and 57%, respectively).  Antibody positive rate for B/Yamagata lineage was generally low, 18% in average and highest (38%) in age group 15-19 years.

Influenza vaccine: The quantity of trivalent vaccines produced in 2011/12 season was 28,880,000 vials (calculated for 1mL/vial), of which estimated 25,100,000 vials were used for vaccination.  The vaccination coverage of the elderly (older than 65 years) in compliance with the Preventive Vaccination Law was 53.3% (53.1% in 2010/11 season).

The vaccine strain selected for 2012/13 season for AH1 was A/California/7/2009pdm09 which is the same as for 2010/11 and 2011/12 seasons, whereas the vaccine strains for AH3 and type B were changed to A/Victoria/361/2011 and B/Wisconsin/1/2010 (Yamagata lineage), respectively (see p. 297 of this issue).

Avian influenza virus A(H5N1) and swine-origin A(H3N2) variant: From November 2010 to March 2011, highly pathogenic avian influenza (HPAI) virus subtype A(H5N1) spread among wild birds and domestic fowl in various places in Japan.  Epidemic of the virus among birds has been continuously reported from Indonesia, Vietnam and Egypt before and after 2011 and human cases of A(H5N1) infection, too.  More recently in September 2012, HPAI among birds was reported from China and Nepal (http://www.maff.go.jp/j/syouan/douei/tori/index.html).

Since July 2012, United States has reported more than 300 human cases of A(H3N2) variant influenza virus infections that occurred through exposure to pigs.

Act on Special Measures For Pandemic Influenza, etc. Preparedness and Response: In preparation for the case of rapid spread of the novel influenza and other emerging/re-emerging infectious diseases of similar public health concern, and based on the experience of pandemic (H1N1)2009, “Act on Special Measures For Pandemic Influenza etc. Preparedness and Response” was issued on May 11 of 2012 (http://www.cas.go.jp/jp/influenza/120511houritu.html).

Additional comments: Trends of outbreaks should be monitored continuously by sentinel surveillance, school outbreak surveillance, hospitalization surveillance and other possibilities.  The virus isolation should be conducted throughout the year and the antigenic and genetic changes of the epidemic strains should be monitored so as to secure vaccine candidate strains.  Monitoring of the resistance to anti-influenza drugs among isolates should be continued.  These measures are all important for future risk management measures.

Flash reports on the isolation and detection of influenza viruses in 2012/13 season are found in p. 300 of this issue and http://www.niid.go.jp/niid/en/iasr-inf-e.html

Mycoplasmal pneumonia is a category V infectious disease in the National Epidemiological Surveillance of Infectious Diseases (NESID) under the Law Concerning the Prevention of Infectious Diseases and Medical Care for Patients of Infections (the Infectious Diseases Control Law) enforced on April 1, 1999.  Sentinel hospitals* regularly report the weekly number of mycoplasmal pneumonia patients (total of outpatients and inpatients).  In addition to isolation of Mycoplasma pneumoniae or detection of serum antibody to M. pneumoniae, detection of M. pneumoniae genome by PCR or LAMP has been included in the notification criteria (http://www.mhlw.go.jp/bunya/kenkou/kekkaku-kansenshou11/01-05-38.html) since its modification in April 2011.  Recently, the reported number of mycoplasmal pneumonia patients is increasing (Fig. 1).

*There are about 500 sentinel hospitals in Japan, which are selected from those equipped with departments of pediatrics and internal medicine and with more than 300 beds.

Periodicity of mycoplasmal pneumonia epidemics and the epidemic that started in 2011: Periodicity of mycoplasmal pneumonia epidemic at a 3-8 year interval has been observed worldwide.  It is presumably brought about by the herd immunity vs. pathogen interaction but its exact mechanism is unknown.  Seasonally, mycoplasmal pneumonia is prevalent from autumn to winter, and occasionally also in early summer (Fig. 1).

In Japan, the large-scale atypical pneumonia epidemics occurred at four-year intervals from the late 1970s through the 1980s coinciding Olympic years.  Under the former NESID system (July 1981-March 1999), number of “clinically-diagnosed atypical pneumonia” cases peaked in 1984 and 1988 (IASR 28: 31-32, 2007) and it was probably epidemics of mycoplasmal pneumonia, because major cause of atypical pneumonia is M. pneumoniae.  Though almost absent since 1990s (except the epidemic in 2006), mycoplasmal pneumonia cases increased in autumn of 2010 (Fig. 1).  In 2011, number of the reports increased from summer and reached its peak in winter, which was more than twice as high as the past peaks in 2006 and 2010.  The number of the reports per week in 2012 is continuously higher than in 2011.  Since 2010, mycoplasmal pneumonia epidemic is found worldwide, such as, in United Kingdom, France, Northern European countries and Israel.

Age and geographical distribution of mycoplasmal pneumonia patients: Children aged 1-14 years occupied 80% of the patients. Among them, since 2011, age group 10-14 years increased and that below 4 years decreased in proportion (Fig. 2).  As such age group shift has been observed in the past (IASR 28: 31-32, 2007), the present resurgence of mycoplasmal pneumonia may not be related to the age shift.

Regionally, since 2007, Aomori, Miyagi, Fukushima, Gunma, Saitama and Okinawa sentinel points have reported larger number of mycoplasmal pneumonia cases, and since 2010, Iwate, Tochigi, Toyama, Aichi, Gifu, Osaka, Ehime and Saga also do so (Fig. 3).  Since 2011, other prefectures with fewer cases started to report larger number of cases than before.

Mycoplasma pneumoniae: Among species of Mycoplasma of human origin, clear pathogenicity has been found only in Mycoplasma pneumoniae.  M. pneumoniae belongs genus Mycoplasma of class Mollicutes.  Its genome is 800kb and the smallest among organisms that grow in artificial media.  It is entirely devoid of peptide glycan cell wall, and β-lactam antibiotics are ineffective.  It is rod-shaped and 0.3 × 2 µm in size.  It has a cytoplasmic protrusion on one end of the body; it is an organelle used for adherence to the surface of respiratory epithelial cells thus contributing to the bacteria’s pathogenicity (see p. 263 of this issue).  On its surface clustered is a large number of cytadhesin protein P1 (molecular weight 170kDa).  P1 protein is polymorphic, which allows classification of M. pneumoniae into type 1 and 2 and their subtypes by genome sequencing.  So far, type 1, 2, 2a, 2b and 2c have been identified among the Japanese clinical isolates.  Though the type does not affect the pathogenicity, the frequency of the prevalent types is variable from year to year and from region to region.  Recently, multiple-locus variable-number tandem repeat analysis (MLVA) is used for epidemiological investigation in the US and European countries.  So far, more than 30 MLVA types have been reported.

Increase of macrolide resistance: Macrolide antibiotics are used for treatment of mycoplasmal pneumonia.  However, since 2000 when the macrolide-resistant M. pneumoniae was first reported in Japan, macrolide-resistant strains are found continuously increasing in Asia and nearby regions (IASR 32: 337-339, 2011).  Now, more than 50% of clinical isolates in Japan are estimated to be macrolide resistant (see pp. 264, 265 & 267 of this issue).  However, European countries experiencing mycoplasmal pneumonia epidemic similarly as Japan report the macrolide resistance rate below 10%.  Macrolide-resistant strains are more often isolated from pediatric patients rather than adults.  Macrolide resistance itself does not significantly affect the sequel as most cases recover without chemotherapy.  When the patients received chemotherapy, feverish phase may prolong in case of macrolide-resistant M. pneumoniae infection than in case of macrolide-susceptible one (IASR 28: 41-42, 2007 and see p. 266 of this issue).  Macrolide-resistant M. pneumoniae infection can be effectively treated with quinolone and tetracycline antibiotics, and no resistant clinical isolates have been reported in the world including Japan.  However, their use for children should be limited to really serious cases on account of their potential side effects (see p. 268 of this issue).

Laboratory diagnosis of mycoplasmal pneumonia: Culture, serodiagnosis and gene amplification methods are available.  Isolation of M. pneumoniae is the most reliable, but 1-4 weeks are required for obtaining the final data.  PA, EIA and other serodiagnostic kits are preferentially used in clinical settings for the rapid test.  However, it is useful only when the paired serum antibody titers of the patients were obtained.  The current most reliable rapid diagnosis is PCR, LAMP and other gene amplification methods (see p. 268 of this issue).  LAMP method has been covered by the health insurance since October 2011.

Final Comments: The progressing mycoplasmal pneumonia epidemic in Japan (http://www.niid.go.jp/niid/ja/10/2096-weeklygraph/1659-18myco.html) necessitates continued surveillance of patients and continued monitoring of drug-resistance and pathogenicity of the bacterial isolates.