TEXT

Top Abstract Text References

Chlamydia pneumoniae is one of the new, emerging infectious agents, with a spectrum of clinical manifestations, including upper and lower respiratory tract infections, and it has recently been tentatively linked to atherosclerosis (2, 4, 15, 21). C. pneumoniae is spread through person-to-person transmission by droplet, and outbreaks of infection have been reported in families, schools, military barracks, and nursing homes (1, 8, 13). Infection with C. pneumoniae is usually mild or asymptomatic, but it can be severe, especially in the elderly, probably as a result of underlying illness (11).

C. pneumoniae is notoriously difficult to cultivate in a cell culture system. PCR assay, antigen enzyme immunoassay, and direct immunofluorescence tests have been described for C. pneumoniae: however, their efficiencies have not been properly validated (6, 11). The microimmunofluorescence (MIF) test is specific for C. pneumoniae and is the standard method for Chlamydia serology today. Little is known about the prevalence of C. pneumoniae antibodies in healthy individuals and the association between C. pneumoniae and respiratory infection in Korea. This study aimed to evaluate the prevalence of C. pneumoniae antibodies in healthy individuals and patients with acute respiratory disease in Seoul, Korea.

Specimens. We collected a total of 564 serum samples for C. pneumoniae antibody testing. Three hundred forty-nine serum samples were obtained from healthy individuals who had no acute respiratory diseases. Cord blood samples were collected from 70 healthy babies whose mothers did not have pelvic inflammatory disease or acute respiratory disease. One hundred forty-five patients with acute respiratory disease (98 with acute pneumonia, 24 with acute bronchitis, and 23 with acute pharyngitis) who visited Hanyang University Hospital in Seoul, Korea, from January 1996 to November 1997 were enrolled in this study. A diagnosis of acute pneumonia was made if there was a compatible clinical illness and if a pulmonary opacity was present on radiographs. All patients with acute bronchitis and acute pharyngitis were examined clinically.

Antigen preparation. C. pneumoniae (TW-183) was provided by the Centers for Disease Control and Prevention and propagated in HeLa cells. The elementary bodies (EBs) of C. pneumoniae were partially purified by differential centrifugation followed by gradient centrifugation in Percoll (17). The EBs were resuspended in a solution of 2% yolk sac in phosphate-buffered saline containing 0.02% formalin. The EBs and sonicated HeLa cells (negative control) were dotted on clean slides. The slides were dried at room temperature for 2 h and fixed in acetone for 15 min (3).

Serologic testing. An MIF test was used to measure chlamydial antibodies (3). The presence of chlamydial antibodies in the immunoglobulin M (IgM) or IgG serum fractions was detected with fluorescein isothiocyanate conjugates of anti-human IgM or IgG (Dako, Copenhagen, Denmark). Dots were evaluated for homogeneity and intensity of the fluorescence with a fluorescence microscope. The endpoint was the highest serum dilution with positive fluorescence. Fluorescence in the negative control (HeLa cells) negated other reactions at that dilution. Serological diagnosis of a previous infection was made when IgG antibody titers were 1:32 or higher. A single titer of anti-C. pneumoniae antibody of 1:512 for IgG or 1:16 for IgM was considered to indicate a recent infection. When high titers of antibody (IgG  1:512) or positive IgM results were observed in healthy individuals, their records were examined to determine whether respiratory infection had occurred within 3 months of serum collection. The significance of the data was determined by the chi-square test. A probability value (P) of <0.05 was considered significant.

Rheumatoid factor assay. A false-positive MIF IgM antibody test may occur if the patient has circulating rheumatoid factor, the prevalence of which increases with age (24). Rheumatoid factor was detected with the Array 360 system (Beckman, Fullerton, Calif.). The sera positive for rheumatoid factor were absorbed with a goat anti-human IgG antibody reagent (Gaullosorb; Gull Laboratories, Inc., Salt Lake City, Utah) as prescribed by the manufacturer and retested for the presence of C. pneumoniae-specific IgM antibody in the MIF test.

Results. The prevalence of antibodies to C. pneumoniae in the sera of healthy individuals was evaluated by the MIF test (Table 1). For cord blood, the antibody was detected in 50% of the samples. The antibody detection rate in healthy individuals was 52%, broken down by age as follows: 1 year old, 39%; 2 to 5 years old, 11%; 6 to 10 years old, 22%; 11 to 20 years old, 44%; 21 to 40 years old, 53%; 41 to 50 years old, 64%; 51 to 60 years old, 71%; and 61 years old, 80%. Overall, C. pneumoniae antibody was present in 53% of males and 51% of females. However, in subjects over 21 years of age, C. pneumoniae antibody was present in 66 of 94 males (70%) and 82 of 146 females (56%). There were 11 cases (3%) of recent infection with C. pneumoniae in healthy individuals, as determined by IgG, and 2 cases (0.6%) for IgM. Antibody titers of 1:256 or higher were observed in 18% of the healthy individuals who tested positive for the IgG antibody. On reviewing the past history of these individuals with a high titer of C. pneumoniae antibody (IgG  1:512) or a positive IgM (1:16), 2 of 13 individuals had developed acute pneumonia, 3 had suffered from acute pharyngitis, another 3 had asthma, and the remaining 5 had no illness.

Discussion. Grayston et al. (9) suggested that antibody titers of IgG (1:16) indicate past infection, and this antibody (IgG) is used in population antibody prevalence surveys. However, IgG titers of 1:32 or higher have been reported in several populations around the world (12, 19, 22), prompting us to consider a serum titer of 1:32 or higher as positive in order to compare our results with other studies. In our study, the antibody was detectable in 50% of cord blood samples, and this rate was similar to that of the maternal age group (21 to 40 years old). In subjects less than 1 year old, the antibody prevalence was about 40%. This is probably attributable to transfer of maternal antibodies. The prevalence of the antibody decreased to 11% in those between 2 and 5 years old, which may indicate that the IgG antibody transferred from the mother had slowly disappeared and that primary infection with C. pneumoniae in this age group is rare. This decline was statistically significant (P <0.05). The detection rate of C. pneumoniae antibody increased rapidly in subjects between the ages of 6 and 10 years, and 44% of subjects aged 11 to 20 years showed the antibody. These results show that primary infection with C. pneumoniae begins in young children and in the early teens and that the serum antibodies are not protective. The prevalence of C. pneumoniae antibody also began to increase in subjects between the ages of 4 and 7 years in other countries as well (5, 12, 18, 25). However, the prevalence continued to increase even in adults, reaching up to 80% in subjects over the age of 61 years. Antibody persistence in old age might be a result of infection and reinfection during life (20). Saikku et al. (23) suggested that the prevalence of C. pneumoniae is associated with population density, which may explain the higher rate of positivity in Seoul, a city with a high population density. Antibody titers of 1:256 or higher were observed in 18% of healthy individuals who tested positive for the antibody. This implies that infection is frequent and that the level of antibody titers probably depends on exposure and time elapsed between the last encounter and monitoring of the organism (14). We found no relationship between gender and incidence of C. pneumoniae antibody until the age of 20 years. However, the incidence of C. pneumoniae antibody in males was higher than that in females after 21 years of age. The prevalence of C. pneumoniae antibody in elderly males (61 years old) was 91%, compared to 70% in females of the same age. This difference between the sexes was statistically significant (P < 0.05). To date, no explanation for the higher frequency among males has been found (14).