In this article, the status of legionellosis in Japan during 2003-2008 after the previous report (IASR 24: 27-28, 2003) are summarized.
The legionellosis incidence: Legionellosis has been classified as a category IV notifiable infectious disease under the National Epidemiological Surveillance of Infectious Diseases (NESID) based on the Infectious Diseases Control Law and physicians must notify legionellosis cases. From January 2003 to the end of September 2008, 2,460 legionellosis cases were reported (as of October 11, 2008). Since 2005, the reported cases have markedly increased (Table 1). This increase is thought to be influenced by that health insurance has been applicable to antigen-detection in urine by enzyme immunoassay from 2003 and by immunochromatographic assay from 2004 and the Guidelines for the Management of Community-Acquired Pneumonia in Adults (the Japanese Respiratory Society) was issued in October 2005, in which it is described to test moderate/severe/ultra-severe pneumonia cases with a Legionella urinary antigen assay.
The day of the first medical consultation peaked in July every year except 2004 (Fig. 1). In Japan and foreign countries, it is reported that this seasonality relates with humidity (see p. 331 of this issue). The incidence by prefecture (Fig. 2) does not indicate a clustering of cases in particular districts.
The mean age of cases was 65.2 years (60.8 years in the previous report), 64.0 years in males and 71.0 years in females. There were few cases under 20 years (0.6%) and the age of adult cases ranged widely to 102 years with a peak at 55-59 years (Fig. 3). The proportion of males was 83% (the same figure in the previous report) being higher than 61% in the United States during 1990-2005 (Neil K, et al ., CID 47: 591-599, 2008).
Methods of diagnosis: Of 2,458 cases, for which the methods for diagnosis were recorded, 2,315 cases (94%) were diagnosed by antigen detection in urine, followed by culture in 97 cases (3.9%), titration of serum antibody in 67 cases (2.7%), and PCR in 35 cases (1.4%) (including cases diagnosed by multiple methods). Cases diagnosed solely by antigen detection in urine largely increased to 96% in 2008 from 68% in 2003. Accompanied by greater use of the urinary antigen detection test, cases diagnosed on the day of the first medical consultation and those within four days from the day are increased to 46% and 80% in 2008 from 15% and 30% in 2003, respectively (Fig. 4). Early diagnosis is desired because the higher case-fatality rate (at the time of notification) was observed if the more days passed from the first medical consultation to diagnosis; 1.0%, 2.7%, 3.5% and 4.6% among cases diagnosed within 1 day, 2-4 days, 5-7 days and ≥8 days, respectively. The death at the time of notification decreased to 2.6% (65 cases) during 2003-2008 from 7.3% in the previous report, although the influence of amendment on the notification from “within seven days after diagnosis” to “promptly after diagnosis” on November 2003 should be taken into consideration.
Species of detected Legionella : Including the cases in which isolates were provided to the Legionella Reference Center after the notification of legionellosis (see p. 332 of this issue), the etiological agent was recorded in 83 culture-positive cases during 2003-2008; Legionella pneumophila in 73 cases [of which, 50 were serogroup (SG) 1, five SG2, four each SG3 and SG5, three SG6, two SG4 and one each SG8 and SG9; including one case with isolation of SG5 and SG8], L. longbeachae in four (of which one case was reported in IASR 26: 247, 2005), L. micdadei and L. rubrilucens in one each, and unknown Legionella spp. in four cases. Isolation of L. rubrilucens from clinical specimen was first reported (IASR 29: 194-195, 2008). In addition, most outbreaks described below were caused by L. pneumophila SG1. On the other hand, according to a study group of the Ministry of Health, Labour and Welfare headed by Endo T with collaboration of Kura F, 735 strains of Legionella spp. were isolated from environmental sources in 2006 fiscal year; L. pneumophila SG1 was predominant from cooling tower water, followed by L. anisa and L. pneumophila SG7, while from hot spring water and circulating bath water, L. pneumophila SG1 and SG untypable were predominant and SG5 and SG6 were also detected.
Outbreaks: The followings are notable incidents during 2003-2008; three cases caused by cruise-ship spas in January 2003 for the first time in Japan (IASR 25: 40-42, 2004), two cases among workers engaged in maintenance of a cooling system at a waste processing plant in Kyoto City in September 2003, two cases who used a whirlpool spa at a fitness club in Niigata City in December 2006 (IASR 28: 144-145, 2007), two cases at a public bathing facility in Kobe City in January 2008 (see page 329 of this issue) and two cases at a welfare facility for the elderly in Okayama Prefecture in July 2008 (see page 330 of this issue). In addition, five suspected outbreak incidents were recognized during June 2006-July 2008; 2-5 cases occurred after using the same facilities in each incident. There was no large outbreak as ever reported.
Control strategy: Legionella is transmitted to human through inhalation of aerosol mist or dust. A legionellosis case was infected during cleaning of bathtubs for public foot bathing with high-pressure water (IASR 29: 49-50, 2008) and another case was infected during treating homemade leaf mold (IASR 26: 221-222, 2005). During cleaning of the wall of bathtubs on which biofilms attached or treating leaf mold, wearing a mask and careful operations are required. Biofilms growing on natural ores in a bath sometimes become a hotbed of Legionella (IASR 29: 193-194, 2008). It is useful to check removal of biofilms after cleaning by on-site measuring the amount of ATP in swabs of wall surfaces as an indicator of residual biofilms. Obviously from the incident in which a case was infected from an ultrasonic humidifier (IASR 29: 19-20, 2008), it is important to clean the inside of the tank of the humidifier for prevention of legionellosis. Although practically no reports of infection by aerosol derived from cooling towers and biological treatment plants [Eurosurveill. 2008;13(38):pii=18985] in Japan, attention is necessary. At present, the standard Legionella count for hygiene management of bath water is less than 10 cfu per 100 ml (below the detection limit). In the aquatic environments keeping the appropriate temperature, Legionella multiplies rapidly in coexistence of their host ameba. Therefore, adequate drainage, cleaning, and disinfection of artificial water equipment in compliance with the maintenance manual are essential to prevent the legionellosis.
For pulsed-field gel electrophoresis (PFGE), which is essential to identify the source of infection, the time to obtain the results shortened to 2 days from 4 days by improvement of the conventional method (see p. 333 of this issue). It is reported that sequence-based typing (SBT) of L. pneumophila strains can predict the environment harboring the strains to some extent (see p. 332 of this issue). Isolation of Legionella spp. from clinical specimens was failed in many outbreaks. It is important to isolate strains from both clinical and environmental specimens, identify the source of infection using PFGE and SBT, and prevent further infection. For further epidemiological analysis hereafter, additional information will be requested if the bacterium is isolated after the case notification.