HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Miyashita, N. Articles by Oka, M. Search for Related Content PubMed PubMed Citation Articles by Miyashita, N. Articles by Oka, M. Agricola Articles by Miyashita, N. Articles by Oka, M.

Community-acquired pneumonia in Japan: a prospective ambulatory and hospitalized patient study

Division of Respiratory Disease, Department of Internal Medicine, Kawasaki Medical School, 577 Matsushima, Kurashiki City, Okayama 701-0192, Japan

Correspondence Naoyuki Miyashita nao{at}med.kawasaki-m.ac.jp

Received October 9, 2004
Accepted January 10, 2005

In this study the aetiology of community-acquired pneumonia (CAP) in Japan was investigated and the incidence of causative pathogens in ambulatory and hospitalized patients was compared. In addition, the roles of Chlamydophila felis and Chlamydophila pecorum as causes of CAP were investigated. Five hundred and six patients with CAP who visited an outpatient clinic or were admitted to one of three different hospitals were enrolled in this study; 106 of them were outpatients and 400 were hospitalized patients. Among the 506 CAP cases, Mycoplasma pneumoniae was the most common pathogen found in the outpatients and Streptococcus pneumoniae was the most common in the hospitalized patients. No cases of Chlamydophila pecorum pneumonia were observed and only one patient had an antibody titre suggestive of recent infection with the feline strain of Chlamydophila. The incidence of infection with M. pneumoniae and Chlamydophila pneumoniae was higher among the outpatients than among hospitalized patients, whereas the incidence of infection with S. pneumoniae and Haemophilus influenzae was higher among the hospitalized patients. Recognition of these results will allow prompt and appropriate antimicrobial therapy to be provided using Japanese CAP guidelines.

	INTRODUCTION

TOP INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
Community-acquired pneumonia (CAP) continues, even nowadays, to be one of the most common causes of morbidity and hospitalization worldwide, particularly in the elderly. Epidemiologic studies show that pneumonia ranks fourth as the leading cause of death in Japan. Because of this high morbidity and mortality, guidelines for CAP management have been promoted throughout the world during the past decade (British Thoracic Society, 2001; American Thoracic Society, 2001; Mandell et al., 2003). The Japanese Respiratory Society (JRS) has also been developing its guidelines since 1998 and published CAP guidelines in March 2000 (Japanese Respiratory Society, Community-acquired Pneumonia Treatment Guideline Creation Committee, 2000). However, prospective studies on the aetiology of CAP among the Japanese population have been very limited (Matsushima et al., 2002). Further, there have been no studies investigating the aetiology of CAP in an ambulatory setting in Japan. The purposes of this study were to determine the pathogens causing CAP in Japan and to compare the incidence of these pathogens in ambulatory and hospitalized patients. In addition, we also investigated the role of Chlamydophila felis and Chlamydophila pecorum as a cause of CAP.

	METHODS

TOP INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
Study population.

Adult patients with CAP who visited an outpatient clinic at or were admitted to Kawasaki Medical School Hospital, Kawasaki Medical School Kawasaki Hospital or Kurashiki Daiichi Hospital, Okayama, Japan, between April 1998 and December 2003 were enrolled in this study. The diagnosis was based on clinical symptoms and signs (cough, fever, productive sputum, dyspnoea, chest pain or abnormal breathing sounds) and radiographic pulmonary abnormalities that were at least segmental and were not due to pre-existing or other known causes. None of the patients were immunocompromised; that is, there were no patients with HIV infection, patients with neutropenia secondary to chemotherapy, patients on immunosuppressants, patients from nursing homes or patients who had recently (<30 days) been admitted to hospital.

Both outpatients and hospitalized patients with CAP were included. Outpatients were defined as those treated in an ambulatory setting throughout and those initially treated as an outpatient but subsequently admitted to a hospital. Hospitalized patients were defined as those admitted to a hospital at the beginning of the study. Concerning the hospitalized patients, all cases of pneumonia occurring more than 3 days after hospitalization were considered nosocomial and were excluded. Informed consent was obtained from all subjects.

Microbiological laboratory tests.

Blood cultures and nasopharyngeal swab specimens were obtained and if pleural fluid and sputum were available, a Gram stain test and a quantitative culture were obtained. Sputum data were only evaluated when the Gram stain test revealed numerous leukocytes (>25 in a x100 microscopic field) and few squamous epithelial cells (<10 in a x100 microscopic field). Certain invasive methods such as bronchoscopic examination were employed to obtain specimens in some patients after full explanation of the procedures. These specimens were used for culturing of Mycoplasma pneumoniae and Legionella species on pleuropneumonia-like organism agar [70 % Mycoplasma agar base (Becton Dickinson Microbiology Systems), 20 % horse serum, 10 % fresh yeast extract, thallium acetate (final concentration 0.5 mg ml^–1) and sterile penicillin G (final concentration 1000 U ml^–1)] and buffered charcoal-yeast extract alpha agar, respectively.

Nasopharyngeal swab specimens were obtained and used for culturing of Chlamydia and Chlamydophila species on cycloheximide-treated HEp-2 cells and suspended L (SL) cells, as described previously (Miyashita & Matsumoto, 1992; Miyashita et al., 2003). The specimens were placed in a sucrose/phosphate/glutamate (SPG) transport medium. Each specimen in SPG medium was sonicated and briefly centrifuged (900 g for 10 min) and then the supernatant was overlaid on confluent monolayers of HEp-2 cells or SL cells. All specimens were passaged twice. Following incubation three different mAbs were used to stain inclusions: a Chlamydophila- and Chlamydia-specific fluorescein isothiocyanate-conjugated mAb (Chlamydia FA Seiken, Denka Seiken), a Chlamydophila pneumoniae species-specific mAb and a Chlamydophila psittaci species-specific mAb (Miyashita et al., 2003). Inclusion bodies formed in the cells were observed with a Nikon epifluorescence microscope at x200 or x400 magnification. Among all the CAP cases, Chlamydophila pneumoniae was isolated from three patients. However, no patients culture-positive for Chlamydophila psittaci, Chlamydia trachomatis or Chlamydophila pecorum were observed among these cases.

Paired serum samples were collected at intervals of at least 4 weeks after onset. Complement-fixation tests were done in all patients for antibodies to influenza A and B viruses, adenovirus, respiratory syncytial virus, cytomegalovirus, parainfluenza virus types 1, 2 and 3 and M. pneumoniae. Antibodies to Legionella species and Coxiella burnetti were measured by the microagglutination test and the indirect immunofluorescence test, respectively. The microimmunofluorescence test was used for titration of IgG and IgM antibodies against chlamydial species, using formalinized elementary bodies. The Chlamydophila pneumoniae KKpn-15 and TW-183, Chlamydia trachomatis L2/434/Bu, Chlamydophila psittaci Budgerigar-1 (a human psittacosis-related strain) and Fe/Pn-1 (a feline pneumonitis strain, obtained from Hideto Fukushi, Gifu University, Japan) and Chlamydophila pecorum Bo/E58 strains were used as the antigens. Chlamydophila pneumoniae was propagated in HEp-2 cells, Chlamydia trachomatis was propagated in HeLa 229 cells and the other Chlamydophila strains were propagated in SL cells (Miyashita & Matsumoto, 1992; Miyashita et al., 2003). The elementary bodies of all the strains were purified by a method of continuous urografin (Schering AG) gradient centrifugation (40–52 %), as described previously (Miyashita & Matsumoto, 1992). Rheumatoid factors were absorbed with Gullsorb (Gull Laboratories) before IgM titration. In addition to serology and/or culturing, the urinary antigen test (Binax NOW, Binax) was used for detection of Streptococcus pneumoniae and Legionella pneumophila.

Criteria for determination of microbial aetiology.

The microbial aetiology was classified as ‘definitive', ‘presumptive’ or ‘unknown'. Bacteria were considered to be definitive causative agents when isolated from blood or pleural fluid cultures. We considered the results of sputum cultures in combination with Gram stain findings. An organism showing heavy (>10⁷ c.f.u. ml^–1) or moderate (10⁶ c.f.u. ml^–1) growth of a predominant bacterium on a sputum culture was considered to be a presumptive pathogen. Any micro-organism isolated from bronchoalveolar lavage fluid was considered to be a presumptive pathogen when its concentration reached >10⁵ c.f.u. ml^–1 in quantitative culture. If M. pneumoniae or the Legionella species were isolated from a specimen, that specimen was considered to be a definitive pathogen even if the culture showed little growth. L. pneumophila and S. pneumoniae were considered to be a presumptive agent when the urinary antigen test was positive. For serologic tests, a fourfold rise in the antibody titre level between paired sera was considered definitive. Acute Chlamydophila pneumoniae infection was defined as IgM >1 : 32 or a fourfold rise in IgG or IgM titre between acute and convalescent serum samples. Seropositivity to other Chlamydia and Chlamydophila species was defined as a fourfold rise between acute and convalescent serum samples or a stable IgG titre of >1 : 64, as reported by Marrie et al. (2003).

Statistical analysis.

Statistical analysis was done by Fisher's exact test. A mean age comparison was done by Student's t test.

	RESULTS

TOP INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
Patient characteristics

Five hundred and sixty-two CAP patients were enrolled in this study and eventually 506 patients were analysed. One hundred and six of them were outpatients and 400 were in-patients. Patient characteristics are summarized in Table 1. The outpatients were significantly younger than the in-patients (mean age, 46.4 vs. 61.5 years, respectively; P < 0.001) and had a lower incidence of underlying disease (30.2 % vs. 63.7 %, respectively; P < 0.0001).

Microbial aetiology

A microbiological diagnosis was established in 56 outpatients (52.8 %) and 262 in-patients (65.5 %) (Table 2). The most common pathogens were M. pneumoniae (27.4 %) followed by S. pneumoniae (12.3 %) and Chlamydophila pneumoniae (11.3 %) in the ambulatory patients, and S. pneumoniae (26.3 %) followed by Haemophilus influenzae (13.0 %) and M. pneumoniae (9.3 %) in the in-patients (Table 2). The presence of two or more pathogens was detected in eight outpatients (7.5 %) and 56 in-patients (14.0 %).

A comparison of the aetiology of CAP in the ambulatory and hospitalized patients showed the incidence of M. pneumoniae to be significantly higher among the outpatients than among the in-patients (27.4 % vs. 9.3 %; P < 0.0001). The incidence of infection caused by Chlamydophila pneumoniae was also higher among the outpatients than among the hospitalized patients, but not significantly (11.3 % vs. 6.8 %; P = 0.1491). In contrast, the rates of infections caused by S. pneumoniae and H. influenzae were significantly higher in the hospitalized patients.

Chlamydia and Chlamydophila as causes of CAP

Among our 506 CAP cases, Chlamydophila pneumoniae was identified as the aetiologic pathogen in 39 cases. Of the 39 cases of Chlamydophila pneumoniae, 34 cases demonstrated fourfold or greater rises in IgG antibody titres and 13 cases demonstrated positive IgM antibody titres (eight cases met both criteria). In 27 of the 39 cases, Chlamydophila pneumoniae was the only pathogen identified, while one additional agent was found in 10 cases and two additional agents were found in two cases. The additional pathogens were S. pneumoniae in five, M. pneumoniae in four, H. influenzae in two, Moraxella catarrahalis in one, Staphylococcus aureus in one and L. pneumophila in one (Table 3).

Psittacosis was identified in five cases among the in-patients only, and no cases of Chlamydophila pecorum pneumonia were observed in any of the CAP cases. Of the five cases of Chlamydophila psittaci, all demonstrated fourfold or greater rises in IgM and IgG antibody titres and three had a history of bird contact.

Only one patient had an antibody titre suggestive of a recent infection with the feline chlamydial strain. This was the case of a 57-year-old female, whose acute and convalescent serum samples had an IgG titre of 1 : 128 directed against a feline chlamydial strain. She was treated intravenously with sulbactam/ampicillin (6 g b.i.d.) and minocycline (200 mg b.i.d.) for 5 days followed by levofloxacin (400 mg day^–1) orally for 10 days and made an uneventful recovery. She had kept three pet cats in her house for more than 10 years.

	DISCUSSION

TOP INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
In Japan, there have been only three reports on the aetiology of CAP among the Japanese population (Matsushima et al., 2002). In this study, we prospectively investigated the aetiology of CAP in an ambulatory setting for the first time.

M. pneumoniae and Chlamydophila pneumoniae have been reported to be more common in ambulatory patients than in patients requiring hospital admission (Blanquer et al., 1991; Michetti et al., 1995; Wattanathum et al., 2003). Our results support these findings.

The most common aetiologic agent has differed among different outpatient studies. S. pneumoniae has been reported to be the most frequent in some studies (Blanquer et al., 1991; Macfarlane et al., 1993; Bochud et al., 2001), whereas some investigations have found atypical pathogens such as M. pneumoniae (Michetti et al., 1995; Berntsson et al., 1986; Marrie et al., 1996) or Chlamydophila pneumoniae (Wattanathum et al., 2003; Almirall et al., 1993) to be more frequent. These differences may be explained by the differences in diagnostic tests, the lack of consensus on diagnostic criteria, seasonal variation and the epidemic period of atypical pathogens. In our study, M. pneumoniae was the most common pathogen, followed by S. pneumoniae, Chlamydophila pneumoniae and H. influenzae. The lower incidence of S. pneumoniae and H. influenzae may be related to the younger age of the patients, a low incidence of comorbidities and the limitation of the tests performed (sputum specimens were not collected from many of the outpatients) (Michetti et al., 1995; Bochud et al., 2001; Marrie et al., 1996; Almirall et al., 1993).

Some studies demonstrated that Legionella species were also common in an ambulatory setting, accounting for 3–12 % of CAP cases (Blanquer et al., 1991; Michetti et al., 1995; Wattanathum et al., 2003). In our study, however, no Legionella species were detected in an ambulatory setting despite the use of culturing, serology and the urinary antigen test. The incidence of Legionella pneumonia in Japan observed in both the present study and previous Japanese studies is lower than that in Western countries (Matsushima et al., 2002). The reason why Legionella pneumonia remains rare in Japan is uncertain.

In contrast to the ambulatory setting, S. pneumoniae was the most common aetiologic agent causing CAP among the hospitalized patients and the rate of infection caused by atypical pathogens was also high. Our results are quite consistent with those of previous studies reported in Japan and other countries (British Thoracic Society, 2001; American Thoracic Society, 2001; Mandell et al., 2003; Matsushima et al., 2002), in which S. pneumoniae and atypical pathogens were reported to be common organisms. Infections caused by Gram-negative bacteria were more common in the hospitalized patients in our study. This result is also in agreement with those of previous studies (Blanquer et al., 1991; Wattanathum et al., 2003).

The observation that more than one causative pathogen can be identified in a patient with CAP has been demonstrated in several studies, and atypical pathogens, mainly Chlamydophila pneumoniae, seem to be the most common organisms of coinfection (American Thoracic Society, 2001; Mandell et al., 2003). In previous studies, approximately 35 % of the Chlamydophila pneumoniae pneumonia cases had a concomitant infection with other pathogens (Miyashita et al., 2002a, b). In this study, 30 % of the Chlamydophila pneumoniae pneumonia and 16 % of the M. pneumoniae pneumonia cases involved mixed infection with other pathogens (Table 3). However, it has been suggested that Chlamydophila pneumoniae may not be the primary cause of the pneumonia but it might disrupt the normal clearance mechanisms, enabling other pathogens to invade. Chlamydophila pneumoniae has been shown to have a ciliastatic effect on ciliated bronchial epithelial cells in vivo (Shemer-Avni & Liberman, 1995). M. pneumoniae also exerts a toxic effect on the ciliated human epithelium. Further, we observed that there were differences in the clinical presentation of CAP cases with multiple pathogens and cases in which Chlamydophila pneumoniae was the only pathogen identified (Miyashita et al., 2002a, b). Therefore, we also believe that mixed cases of mild or asymptomatic upper respiratory tract infections are probably induced by atypical pathogens and followed by secondary bacterial pneumonia due to another proven aetiology.

Based on the above data, the North American guidelines demonstrated that the treatment of CAP in both outpatients and hospitalized patients must cover atypical pathogens and S. pneumoniae (American Thoracic Society, 2001; Mandell et al., 2003). Therefore, macrolides, doxycycline and fluoroquinolones are recommended for primary empiric monotherapy or combined-therapy with ß-lactams, since each has activity against common bacterial pathogens and atypical pathogens.

The aetiologic agent which most clearly differentiates Japan from Western countries is drug-resistant S. pneumoniae (Felmingham et al., 2002). A recent study found that the frequency of drug-resistant S. pneumoniae has been increasing gradually in Japan. Further, the JRS recommends the restriction of quinolone usage to prevent an increase in quinolone-resistant strains to the high frequency of macrolide- or tetracycline-resistant strains. Based on these facts, the JRS proposed a differential diagnosis between bacterial pneumonia and atypical pneumonia for an appropriate antibiotic to be selected for the management of mild to moderate pneumonia (Japanese Respiratory Society, Community-acquired Pneumonia Treatment Guideline Creation Committee, 2000). The guidelines set up nine headings and criteria for the differential diagnosis, and, in preliminary reports, we have demonstrated the guideline criteria to be a useful tool for distinguishing between mycoplasmal and bacterial pneumonia (Miyashita et al., 2004). Further studies are being carried out all over Japan to determine whether the proposed differential diagnosis can successfully distinguish between atypical pneumonia and bacterial pneumonia.

The Chlamydia and Chlamydophila species are obligate intracellular pathogens, and Chlamydophila pneumoniae, Chlamydophila psittaci and Chlamydia trachomatis are human respiratory pathogens. Lower respiratory infection due to Chlamydia trachomatis has been found mostly in neonates rather than in the adult population. In this study, we confirmed the same results, with no acute Chlamydia trachomatis infection being observed among our adult CAP cases. Chlamydophila psittaci is a pathogen for a wide range of avian and mammal species. Psittacosis is a well-recognized infection in humans, classically transmitted from infected birds (ornithosis). The feline chlamydial strains typed so far have been classified as Chlamydophila psittaci. Feline Chlamydophila strains differ from Chlamydophila psittaci derived from other animal species based on the RFLP of a PCR-amplified DNA, the sequence diversity of the ompA, 16S rRNA and 23S rRNA genes, and the antigenicity of the major outer-membrane protein as well as the antigenic diversity of elementary bodies (Fukushi & Hirai, 1988; Everett et al., 1999).

There have been only a few case reports of feline chlamydiosis in humans: cases of conjunctivitis in cat owners (Johnson, 1992), a case of general malaise, cough and abnormal liver function in an immunosuppressed female (Griffiths et al., 1978) and a case of endocarditis with associated glomerulonephritis in a 40-year-old male (Regan et al., 1979). These cases reported close contact with cats. Furthermore, Cotton & Partridge (1998) reported a pneumonia case in a 48-year-old male with no underlying disease. Recently, Marrie et al. (2003) investigated the role of Chlamydophila species as a cause of CAP in Canada, and found only two cases with possible feline Chlamydophila psittaci pneumonia among 539 episodes. In this study, we found only one case with possible feline Chlamydophila pneumonia. Our results were consistent with Marrie's study, indicating that feline Chlamydophila may cause a few cases of CAP.

However, Fukushi and his colleagues found a high prevalence of feline Chlamydophila infection in cats in Japan, and they also found that the prevalence of anti-feline Chlamydophila antibodies in small-animal clinical veterinarians was significantly higher than in the general population (Yan et al., 2000). From these results, together with the above case reports, they suggested that feline chlamydiosis could be transmitted to people who are in close contact with infected cats. Therefore, further study is needed to determine the role of feline Chlamydophila as a cause of respiratory tract infections, especially in pet owners. Feline Chlamydophila may be a causative agent of upper respiratory tract infection, but is not a significant causative agent of lower respiratory tract infection. Therefore, we are now investigating the role of feline Chlamydophila as a cause of upper respiratory tract infections in two groups of cat owners and non-cat owners.

In conclusion, our study indicates that the S. pneumoniae and the atypical pathogens, Chlamydophila pneumoniae and M. pneumoniae are significant causative micro-organisms in CAP in both outpatients and hospitalized patients in Japan. Further, our study demonstrates that feline Chlamydophila is an uncommon cause of CAP in the general population. Recognition of these results will allow us to promptly treat patients with appropriate antimicrobial therapy, and newly promoted guidelines for the management of CAP in Japan will be available.

	ACKNOWLEDGEMENTS

TOP INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
This work was supported by the 32nd Kanae Foundation for Life & Socio-Medical Science and project research grants from Kawasaki Medical School (13-401, 14-402, 15-405A, 16-405M).

	REFERENCES

TOP INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGEMENTS REFERENCES