|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Division of Pulmonary (Drs. Wattanathum, Chaoprasong, Jatakanon, and Chanthadisai), Department of Medicine, and Department of Microbiology (Ms. Chantaratchada), Phramongkutklao Hospital, Bangkok, Thailand; the Department of Microbiology (Ms. Nunthapisud), Chulalongkorn University, Bankok, Thailand; and the Division of Clinical Pathology (Mr. Limpairojn), Army Institute of Pathology, Bangkok, Thailand.
Correspondence to: Anan Wattanathum, MD, 706-1772 Comox St, Vancouver, BC, V6G 1P8 Canada; e-mail: wattanathum{at}hotmail.com
| Abstract |
|---|
|
|
|---|
Design: A prospective study.
Setting: Three general hospitals in Thailand.
Patients: Two hundred forty-five adult patients fulfilling the clinical criteria of CAP from September 1998 to April 2001.
Interventions: Investigations included sputum Gram stain and culture, blood culture, pleural fluid culture (if presented), urine antigen for Legionella pneumophila and Streptococcus pneumoniae, and serology for Mycoplasma pneumoniae, Chlamydia pneumoniae, and L pneumophila.
Results: There were 98 outpatients and 147 hospitalized patients included in the study, and an organism was identified in 74 of 98 outpatients (75.5%) and 105 of 147 of the hospitalized patients (71.4%). C pneumoniae (36.7%), M pneumoniae (29.6%), and S pneumoniae (13.3%) were the most frequent causative pathogens found in outpatients, while S pneumoniae (22.4%) and C pneumoniae (16.3%) were the most common in hospitalized patients. There was a significantly higher incidence of C pneumoniae (36.7% vs 16.3%, respectively; p < 0.001) and M pneumoniae (29.6% vs 6.8%; p < 0.001, respectively) in the outpatients than in the hospitalized patients. The incidence of S pneumoniae, L pneumophila, and mixed infections was not different between the groups. Mixed infections were presented in 13 of 98 outpatients (13.3%) and 9 of 147 hospitalized patients (6.1%), with C pneumoniae being the most frequent coinfecting pathogen.
Conclusions: The data indicate that the core organisms causing CAP in Southeast Asia are not different from those in the Western countries. The guidelines for the treatment of patients with CAP, therefore, should be the same.
Key Words: community-acquired pneumonia etiology pathogens Southeast Asia Thailand
| Introduction |
|---|
|
|
|---|
| Materials and Methods |
|---|
|
|
|---|
Patients
Adult subjects who were > 15 years of age with a diagnosis of CAP made by a primary physician were recruited within the 24 h of presentation. The diagnosis of CAP required the following to be present2
6
: (1) a new pulmonary infiltrate seen on a chest radiograph that was obtained within 24 h of presentation; and (2) a confirmatory clinical finding of at least one of the major criteria, which included cough, sputum production, or temperature > 37.8°C, or at least two of the minor criteria, which included pleuritic chest pain, dyspnea, altered mental status, pulmonary consolidation by physical examination, and WBC count of > 12,000 cells/µL. Exclusion criteria were as follows: (1) patient transferred from another hospital or hospitalization within the previous 3 weeks; (2) presence of an emerging alternative diagnosis (eg, pulmonary or septic emboli, pulmonary edema, or malignancy) during follow-up; (3) presence of pneumonia caused by tuberculosis or postobstructive pneumonia due to lung cancer; (4) presence of severe immunosuppression including severe neutropenia (ie, < 1.0 x 109 cells/L), HIV infection, and solid-organ or bone marrow transplantation; and (5) receiving corticosteroid treatment with a dosage of > 20 mg prednisolone-equivalent per day for
2 weeks.
Both outpatients and hospitalized patients with CAP were included. Outpatients were defined as those persons treated in an ambulatory setting and those initially treated as an outpatient but subsequently admitted to a hospital. Hospitalized patients were defined as those persons admitted to a hospital at the beginning of the study. The criteria for hospital admission were based on the American Thoracic Society (ATS) guidelines.7 The clinical criteria of severe CAP for ICU admission included the following: (1) shock or sustained hypotension requiring vasopressor administration for > 4 h; (2) multiorgan failure (ie, respiratory, renal, neurologic, cardiac, or hepatic failure); (3) severe respiratory failure (ie, PaO2/fraction of inspired oxygen ratio of < 250) and/or a requirement for mechanical ventilation; or (4) acute renal failure requiring dialysis. The choice of the initial antimicrobial therapy for pneumonia patients was made by the attending physician. At hospital discharge, the patients were referred to the investigators at the pulmonary disease clinic of the hospitals for clinical and radiologic follow-up.
Microbiological Evaluation
The microbiological evaluation included sputum Gram stain, sputum culture, blood culture, pleural fluid culture (if available), urine antigen tests for Legionella pneumophila and Streptococcus pneumoniae, and paired sera (4 to 6-week intervals) for subsequent serology tests. Sputum was cultured immediately on blood agar, chocolate, and MacConkey agar and was incubated at 37°C. However, the results were included for analysis only if the specimen was adequate (ie, if it contained > 25 polymorphonuclear cells and < 10 epithelial cells per low-power field). The identification of microorganisms was accomplished according to standard methods.8
The susceptibility to penicillin of isolated S pneumoniae was subsequently determined with a 1-µg oxacillin disk (BBL Microbiology Systems; Cockeysville, MD) by the disk diffusion method according to the performance standards of the National Committee for Clinical Laboratory Standards (NCCLS). Isolates producing a zone of inhibition of
20 mm around the disk were defined as being susceptible to penicillin. The minimal inhibitory concentration (MIC) of the oxacillin-resistant strains was determined by apolipoprotein-E testing for penicillin. MIC interpretive standards were defined according to the 1997 NCCLS breakpoints.9
Isolates were defined as being susceptible, as having intermediate resistance, or as being highly resistant to the agent tested, according to the NCCLS definitions.10
A immunochromatographic assay (Binax-Now; Binax; Portland, ME)11
12
and an enzyme immunoassay (Biotest AG; Dreieich, Germany)12
13
14
were the urine antigen tests used for detecting S pneumoniae and L pneumophila, respectively. The paired sera, which had been stored at -30°C, were tested for L pneumophila antibodies (immunofluorescence), for Chlamydia pneumoniae (microimmunofluorescence for IgM, IgG, and IgA), and for Mycoplasma pneumoniae (particle agglutination; Serodia-Myco II; Fujirebio; Tokyo, Japan).15
16
17
Criteria for Etiologic Diagnosis
The agents causing CAP were considered as being definite, presumptive, and unknown. A definite etiology was defined if one of the following conditions was present: (1) blood or pleural fluid culture yielding the presence of a bacterial or fungal pathogen; (2) urinary antigen test results positive for L pneumophila or S pneumoniae11
12
13
14
; (3) a fourfold increase in the antibody titer for L pneumophila (to
1:128), M pneumoniae (to
1:160), or C pneumoniae (IgM, IgG, or IgA)15
16
17
; and (4) a single increase in the IgM titer for C pneumoniae
1:32. A presumptive etiology was considered if any of the following conditions was present: (1) growth of a predominant bacterial pathogen on sputum culture in combination with the same finding on sputum Gram stain; (2) an antibody titer of
1:1,024 for L pneumophila in either the acute-phase or convalescent-phase serum specimen15
; (3) an antibody titer of
1:160 for M pneumoniae in either the acute-phase or convalescent-phase serum specimen16
; or (4) an IgG antibody titer of
1:512 or an IgA antibody titer of
1:512 for C pneumoniae.17
Patients in whom more than one microorganism was detected, according to the above-mentioned criteria, were defined as having mixed infection. When no causative agent was found according to the criteria, the pneumonia was classified as having unknown etiology.
Statistical Analysis
Data are presented as mean ± SD. Continuous variables were compared by the Student t test, and categoric variables were compared by the
2 test or Fisher exact test where appropriate. A p value of
0.05 was considered to be statistically significant.
| Results |
|---|
|
|
|---|
|
|
|
|
2 µg/mL). Only one isolate had an MIC of > 4 µg/mL. No significant differences were found between patients with penicillin-sensitive or penicillin-resistant pneumococcal pneumonia regarding age, gender, prior antibiotic treatment, and underlying diseases including diabetes mellitus, cardiopulmonary diseases, liver diseases, and neurologic diseases. The initial antimicrobial treatment for 13 patients with penicillin-resistant strains of S pneumoniae consisted of a macrolide (1 patient), fluoroquinolones (2 patients), third-generation cephalosporins alone (5 patients), and a third-generation cephalosporin plus a macrolide (2 patients), metronidazole (2 patients), or aminoglycoside (1 patient). The initial therapy was considered to be effective in 11 patients with penicillin-resistant pneumococci. In the remaining two patients, antibiotic therapy was changed from a third-generation cephalosporin to vancomycin. Seven of 46 patients (15.2%) with pneumococcal pneumonia died in the hospital, and the mortality rates were not significantly different between the patients with penicillin-resistant S pneumoniae and penicillin-sensitive S pneumoniae (23.1% vs 13.6%, respectively; p = 0.65).
Other Bacterial Pathogens
Only one presumptive case of Haemophilus influenzae infection was found in an outpatient (Table 3)
. Among the hospitalized patients, K pneumoniae was the most common Gram-negative pathogen isolated, representing 9.5% of the patients (14 of 147 patients; definite diagnosis, 9 patients; presumptive diagnosis, 5 patients). The other Gram-negative bacilli were H influenzae (2.7%), Escherichia coli (1.4%), Acinetobacter baumannii (1.4%), Burkholderia pseudomallei (1.4%), Proteus mirabilis (0.7%), and Pseudomonas spp (0.7%). The Gram-negative infections were significantly higher in hospitalized patients than in outpatients (17.7% vs 1%, respectively; p < 0.001). These infections were more common in the patients who had underlying comorbid illnesses (19% vs 3.9%, respectively; odds ratio, 5.8; 95% confidence interval, 2.1 to 15.9; p < 0.001). The mortality rate of hospitalized patients with Gram-negative infections was significantly higher than that of patients with Gram-positive or atypical pathogen infections (53.8% vs 24.4% vs 12.1%, respectively; p = 0.001).
Atypical Pathogens
C pneumoniae infection was significantly higher among outpatients (36.7%; 36 of 98 patients; definite diagnosis, 32 patients; presumptive diagnosis, 4 patients) than among inpatients (16.3%; 24 of 147 patients; definite diagnosis, 21 patients; presumptive diagnosis, 3 patients; p < 0.001). The rate of infection caused by M pneumoniae was also higher among outpatients (29.6%; 29 of 98 patients; definite diagnosis, 26; presumptive diagnosis, 3 patients) than among hospitalized patients (6.8%; 10 of 147 patients; definite diagnosis, 6; presumptive diagnosis, 4 patients; p < 0.001). The incidence of L pneumophila infection was not different between the groups (8 of 98 outpatients [8.2%]; 8 of 147 inpatients [5.4%]).
Mixed Infections
Mixed infection was demonstrated in 13.3% of the outpatients (13 of 98 patients) and in 6.1% of the hospitalized patients (9 of 147 patients) [Table 4
], without a significant difference between the groups. C pneumoniae was found to be the most common pathogen coinfecting patients with another organism.
|
| Discussion |
|---|
|
|
|---|
The high infection rates caused by C pneumoniae (36.7%) and M pneumoniae (29.6%) in our ambulatory patients could be explained by many factors. First, paired sera were usually collected from the patients for the diagnosis of atypical pathogens in our study, and this has been shown to improve the diagnostic yield.5 Second, C pneumoniae and M pneumoniae often cause a mild clinical disease severity, therefore patients are more likely to be seen as outpatients.18 19 20 21 Moreover, the infection by the atypical microorganisms is more common among persons in a younger age group, as was seen in our outpatients.12 22 23 Finally, the differences in geographic distribution and in patient characterization may contribute to a higher infection rate. However, the incidence of infection caused by L pneumophila, which is another atypical pathogen, in our study was 8.2%, and this rate was not different from the previous reported incidence of 1 to 13%.24 In the literature, the pathogens responsible for CAP were discovered in < 50 to 70% of patients,2 3 4 18 22 depending on the types of diagnostic tests. In an ambulatory setting, S pneumoniae, M pneumoniae, Chlamydia spp, viruses and Legionella spp were described as the most common organisms found, accounting for 9 to 23%, 12 to 33%, 4 to 28%, 7 to 15%, and 1 to 13% of patients, respectively.18 19 20 21 22 25 Other less common pathogens included H influenzae (0.2 to 10%) and Coxiella burnetii (2 to 3%).18 20 21 22
With regard to S pneumoniae infection in the outpatient setting, our incidence was rather low (13.3%) despite the use of a urine immunochromatographic pneumococcal antigen test. The test has been shown to be sensitive (80%) and specific (97%) for the detection of bacteremic and nonbacteremic pneumococcal pneumonia.11 Although the role of the test in patients who had received prior antipneumococcal therapy has not been clearly defined, Dominguez and colleagues11 demonstrated positive tests in three of five patients after they had received antibiotic therapy. Pneumococcal pneumonia has been reported to be more common in elderly and hospitalized patients who had a variety of medical conditions including alcoholism, chronic cardiovascular disease, COPD, and hematologic malignancy, while most of our outpatients were young and had no comorbidity.26 In one study,27 33% of case of CAP of unknown etiology that had been diagnosed by routine methods were found to be due to S pneumoniae based on findings from transthoracic needle lung aspiration. Younger age, lower incidence of cardiopulmonary diseases, and the limitation of the tests may have contributed to the lower incidence of S pneumoniae in the ambulatory patients.
While the most commonly identified pathogen causing CAP among hospitalized patients was S pneumoniae (22.4%), the rate of infection caused by atypical pathogens was also high (28.5%). This result is similar to those of several studies3 5 27 28 in which S pneumoniae and atypical pathogens were reported to be the common organisms in 20 to 55% and 18 to 55% of all episodes, respectively. C pneumoniae infection was found in 16.3% of our patients, and this rate was similar to the previous incidences reported (ie, 7 to 18%).5 12 23 27 28 As many patients with Chlamydia infection were also seen as outpatients, which implies that the infection caused by C pneumoniae can present with a wide range of clinical severity. Although a high incidence of C pneumoniae infection may suggest an epidemic,3 there was no evidence of a higher prevalence of respiratory tract infection in Bangkok during the period of the study.
The infection caused by Gram-negative bacteria was more common in the hospitalized patients. This result is in agreement with those of previous studies1 6 that Gram-negative pathogens, usually H influenzae and Enterobacteriaceae, were common in up to 20% of patients. Interestingly, we found two cases of B pseudomallei infection, and both of them came from the northeastern part of Thailand, which is an endemic area for this organism. B pseudomallei, a habitat organism living in the soil, was considered to be a common cause of CAP in hospitalized patients in the northeastern part of Thailand and also was considered to be the most frequent pathogen causing severe CAP in Singapore.29 30 This emphasizes the importance of epidemiologic considerations while considering empirical antibiotic therapy.
The observation that more than one causative pathogen can be identified in a patient with CAP has been demonstrated in several studies.31 The exact rate of polymicrobial infection, which depends on the number of the pathogens tested for and the laboratory techniques used, has been reported to vary from 3 to 40%, and C pneumoniae seems to be the most common organism of coinfection.31 C pneumoniae can cause ciliostasis in the human bronchial epithelial cell,32 and therefore primes the airways for another coinfection. M pneumoniae also exerts a toxic effect on the ciliated human epithelium.33 It is not known whether the first pathogen, in addition to inducing ciliary injury, may cause an additional manifestation of the infectious disease in conjunction with another copathogen that penetrates in its tracks.
In our study, the etiologic pathogen could not be identified in 27% of patients, which is the same rate as reported in previous studies. This could be due to viral infection as a cause of CAP or to the limitation of the tests used for identifying S pneumoniae.27
In conclusion, our epidemiologic data indicated that the microbial agents causing CAP in Thailand, in general, are not different from those in Western countries. The data support the current guidelines that the treatment of CAP in both outpatients and hospitalized patients should cover atypical pathogens and S pneumoniae. To achieve the best outcome of therapy, however, physicians must be aware of a possible local pathogen, which may not be covered by standard empiric therapy.
| Acknowledgements |
|---|
| Footnotes |
|---|
Received for publication March 8, 2002. Accepted for publication September 20, 2002.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
N. Miyashita, H. Fukano, K. Mouri, M. Fukuda, K. Yoshida, Y. Kobashi, Y. Niki, and M. Oka Community-acquired pneumonia in Japan: a prospective ambulatory and hospitalized patient study J. Med. Microbiol., April 1, 2005; 54(4): 395 - 400. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |