Asthma and Chlamydial Infection: A Case Series David L. Hahn, MD, and Rjurik Golubjatnikov, PhD, MPH Madison, Wisconsin

Background. Adult-onset asthma is frequently associ­ asthma for the first time during the study period. ated with antecedent respiratory symptoms that could Thirty' (22.9%) patients were classified with acute asth­ represent either previously undiagnosed asthma or pre­ matic bronchitis, and 89 (67.9%) had nonwheezing ill­ vious lung infections that result in subsequent asthma. ness. Two of the newly diagnosed asthmatics met sero­ To further investigate a reported association of Chla­ logic criteria for acute C pneumoniae infection, and one mydia pneumonias infection and adult reactive airway had serologic evidence for acute M pneumoniae infec­ disease, we looked for evidence o f atypical infections in tion. Compared with patients with nonwheezing respi­ patients with acute wheezing and nonwheezing respira­ ratory' illnesses, C pneumoniae seroreactivity was signifi­ tory illnesses. cantly (P < .001) associated with both chronic asthma and with acute asthmatic bronchitis. Methods. Pharyngeal cultures and acute and convales­ cent serology for C pneumoniae and Mycoplasma pneu­ Conclusions. Acute wheezing illness was encountered moniae were obtained from 131 primary care outpa­ frequendy in this primary care setting. Although most tients (mean age, 36 years) with acute wheezing or acute wheezing respiratory' illness resolved without ob­ nonwheezing respiratory illnesses. Peak flow measure­ vious chronic sequelae, some patients had persistent ments were obtained in patients with cough or wheeze. symptoms and were diagnosed with chronic asthma. C Spirometry before and after bronchodilator use was pneumoniae seroreactivity was associated with both obtained to substantiate the diagnosis of chronic acute and chronic wheezing, suggesting that pulmo­ asthma in patients who had persistent wheezing and nary' infection with this intracellular pathogen plays a dyspnea after enrollment. role in the natural history of reactive airway disease. Results. Twelve (9.2%) of 131 patients were classified Key words. Asthma; bronchitis; Chlamydia pneumoniae; as having chronic asthma, 5/12 developed chronic Mycoplasma pneumoniae. (/ Fam Pract 1994; 38:589-595)

The cause of asthma, a common chronic inflammatory bacteria (Mycoplasma pneumoniae and Chlamydia pneumo­ airway condition, is not well understood. It was once niae) may play a role in these reported associations.5-9 thought that infection played an important causative role Given these preliminary' results and our limited under­ in asthma,1"3 but there are no convincing scientific stud­ standing of asthma etiology, it is worthwhile to investi­ ies documenting “bacterial allergy” as a cause.4 More gate further the possibility that atypical infections arc- recent epidemiologic studies and clinical reports have involved in the initiation and exacerbation o f asthma, or documented that respiratory diagnoses including bron­ even the promotion of asthma in persistently infected, chiolitis,5 acute bronchitis,6 chronic bronchitis7 and susceptible individuals. pneumonia8 are associated with the development of sub­ Mycoplasma pneumoniae infection is a recognized sequent asthma in both adults and children. Some re­ cause of acute respiratory illnesses, including atypical ports have presented preliminary evidence that atypical pneumonia and tracheobronchitis, and may also cause asymptomatic infection in about 20% of cases.10 M pneu­ moniae infection is more common in children than in Submitted, revised, December 30, 1993. adults,11 and has been associated with varying propor­ From the Arcand Park Clinic, Division o f Dean Medical Center (D.L.H.), and tions of acute infectious exacerbations of asthma, usually Wisconsin State Laboratory o f Hygiene (R.G.), Madison, Wisconsin. Requests for reprints should be addressed to David L. Hahn, MD, Arcand Park Clinic, 3434 East in children12-14 but also sometimes in adults.15 A pro­ Washington Ave, Madison, W I 53704. longed carrier state beyond several months postinfection

® 1994 Appleton & Lange ISSN 0094-3509

The Journal of Family Practice, Vol. 38, No. 6(Jun), 1994 5 8 9 Asthma and Chlamydial Infection H ahn and Golubjatnikov

in immunocompetent patients has not been document­ study are greater, therefore, than comparable figures ed,16 making it unlikely that persistent M pneumoniae (13% and 3%, respectively) from our previous study infection frequently promotes chronic asthma.5 which enrolled all patients systematically.9 Another atypical respiratory pathogen, Chlamydia Our definition of “asthmatic bronchitis” differs from pneumoniae, has emerged as an important cause of acute that used in some standard references of pulmonary med­ respiratory illness, including bronchitis and pneumo­ icine. Thus, asthmatic bronchitis has been used to de­ nia.17 Like M pneumoniae, most primary infections occur scribe patients in whom the diagnoses of chronic bron­ in younger age groups. Unlike M pneumoniae, however, chitis and asthma coexist, or are difficult to distinguish.^ the majority of patients with Cpneumoniae infections are The term “wheezy bronchitis,” as used in another refer­ asymptomatic or only mildly symptomatic,18 and repeat ence,27 denotes episodes of wheezing during acute bron­ infections are common and occur throughout adult­ chitis, occuring mainly in children. Acute asthmatic hood.19 Chronic intracellular infection is typical for chla- bronchitis, as used in this paper, refers to the latter type mydiae.20 Persistent infection has been documented for of wheezing during acute bronchitis but is not limited to C pneumoniae21 and has been implicated in some chronic children, since there is mounting evidence that a signif­ cardiopulmonary diseases including sarcoidosis,22 coro­ icant proportion (15% to 30%) of episodes of acute nary artery disease,23-25 and adult-onset asthma.9 bronchitis in adult outpatients from primary care settings In a previous prospective study designed to assess may involve bronchospasm or wheezing.6'9-28'29 the role o f acute C pneumoniae infection in bronchitis and atypical pneumonia, we found post hoc associations of C Pulmonary Function Measurement and pneumoniae scroreactivity with wheezing, asthmatic Diagnosis of Asthma bronchitis, and adult-onset asthma.9 We now report re­ sults of culture and serology for M pneumoniae and C Patients with acute asthmatic bronchitis had peak expi­ pneumoniae in a series of patients with asthma confirmed ratory flow measurements documented (best of three by pulmonary function testing, studied during a M pneu­ values) using a Wright peak flow meter (Armstrong moniae epidemic. Asthma patients arc compared with Industries, Inc, Northbrook, 111). Standardized normal patients who had acute asthmatic bronchitis or acute values for age, sex, and height were also recorded. Pa­ nonwheezing respiratory illnesses. tients with persistent wheezing and dyspnea in whom a diagnosis of asthma was suspected had spirometric test­ ing performed using a spirometer (System 21, Gould Methods Medical Products Inc, Dayton, OH). This study em­ ployed American Thoracic Society (ATS) guidelines for Study Setting the diagnosis of asthma and chronic obstructive pulmo­ One hundred thirty-one patients who visited a primary nary disease (COPD).30-31 care office with complaints o f respiratory illness between Clinical diagnoses of acute asthmatic bronchitis and October 1989 and January 1991 (16 months) were en­ asthma and serologic results o f patients (n = 50) enrolled rolled in the study. The study site was a community- between October and December 1989 were included in based, four-physician family practice office affiliated with our first report associating C pneumoniae infection with a large multispecialty, multisite group practice located in reactive airway disease.9 The resuits of pulmonary func­ a metropolitan area of south central Wisconsin (Dane tion testing for this group have not been published County), with a predominantly white, middle-class pop­ previously. When these original patients were compared ulation. with the 81 patients enrolled after December 1989, there were no significant differences in mean age, sex, history Enrollment Criteria of smoking, Cpneumoniae seroreactivity, titer category' ot acute antibody, or diagnoses of acute asthmatic bronchi­ Enrollment was offered to all patients who were encoun­ tis, asthma, or COPD. tered by the clinician-investigator during the course of usual practice and who had laryngitis, biphasic illness (pharyngitis and/or laryngitis, followed by bronchitis or Bacteriologic and Serologic Data pneumonia), wheezing, or atypical pneumonia. Patient Acute sera for microimmunofluorescence (MIF) testing refusal was not documented but is believed to be mini­ for C pneumoniae and complement fixation (CF) testing mal. Enrollment of patients with other respiratory ill­ for At pneumoniae and oropharyngeal culture for C pneu­ nesses was less systematic. The proportions of acute moniae and M pneumoniae were obtained at the time of asthmatic bronchitis (23%) and asthma (9%) in this study enrollment for all patients. Because 17 patients

5 9 0 The Journal of Family Practice, Vol. 38, No. 6(Jun), 1994 \sthma and Chlamydial Infection Hahn and Golubjamikov

failed to return for follow-up testing, convalescent sera Respirator)^ diagnoses at enrollment and for the were available from only 114 patients (87%). Serologic 6 -month period before and after enrollment were tabu­ criteria for acute C pneumoniae infection were (1) a lated by a medical record reviewer who was unaware of four-fold titer rise in either IgM seroreactivity or poly­ the culture and serologic results. Serologic results were valent (mixture of IgM, IgG, and IgA) seroactivity in the unavailable at the time of study enrollment, and therefore VIIF test (using the TW -183 strain as antigen), or (2) could not have influenced enrollment or preenrollment any IgM titer > 1:16, or (3) any polyvalent titer >1:512. diagnoses. Because the clinician-investigator became Currently, there are no clearcut serologic criteria for aware o f serologic results sometime during the 6 -month chronic C pneumoniae infection. C pneumoniae titer cat­ postenrollment period, diagnostic bias after that time egory of patients not meeting serologic criteria for acute cannot be excluded. However, other clinicians in the infection was defined as the greater of the acute or practice setting who may have made respiratory diag­ convalescent titer, or the acute titer if the convalescent noses in study patients were not aware of serologic re­ was not available (< 1 :1 6 [seronegative], 1:16, 1:32, sults. Also, other than those enrolled, no other adult 1:64 and 3:1:128). Serologic criteria for acute M pneu­ patient encountered by the principal investigator in this moniae infection were (1) a four-fold or greater difference primary care practice reported asthma symptoms in as­ between acute and convalescent CF titers, or (2) a single sociation with acute respirator)' illness during the study acute CF titer of 1:128 or greater. Oropharyngeal cul­ period. Therefore, bias related to exclusion of seronega­ tures were transported from the study clinic to the Wis­ tive asthma patients with respiratory illness is possible consin State Laboratory of Hygiene in Madison for cul­ but seems unlikely. ture on McCoy cells as described previously.9 Statistical Methods Clinical Categorization of Patients The chi-square test was used to analyze tabular data Respiratory diagnoses made at enrollment and during containing more than five items per category; otherwise, the 6 -month periods preceding and following enroll­ Fisher’s exact test was used to analyze 2 x 2 tables, and ment were recorded for all subjects. A 6 -month period a Monte-Carlo procedure (Bill Engels, Genetics Depart­ was chosen to allow identification of patients who devel­ ment, University of Wisconsin, Madison) was used to oped persistent asthma following the enrollment illness. estimate the probabilities for NxK tables (minimum of Medical records were also completely reviewed for evi­ 10,000 iterations per table). Analysis of variance was dence of previous asthma and COPD. Based on this used to test differences in mean ages, and the Kruskal- information, patients were classified into one o f three Wallis test and the Mann-Whitney U test were used to diagnostic groups: compare geometric mean titers of subgroups. In the analysis of geometric mean titers, titers reported as Group 1, Asthma (n = 12): This group included < 1:16 were coded as 1:8. Associations between diagnos­ patients who had persistent wheezing and dyspnea and tic groups and seroreactivity were controlled for age, sex, who met ATS criteria for reversible airway obstruction and current smoking status using logistic regression.32 after inhaling albuterol or ingesting oral steroids. Five Two-sided P values <.05 are reported as significant. (41.7%) group 1 asthma patients were given a diagnosis This study was approved by the human subjects of asthma for the first time during the postcnrollment committee of St Marys Hospital Medical Center, Madi­ period, and the other 7 (58.3%) had exacerbations of son, Wisconsin, and all patients gave informed consent. previously diagnosed asthma. Further details of the study population, data collection methods, and microbiologic and serologic techniques Group 2, Acute asthmatic bronchitis (n = 30): This have been published elsewhere.9 group consisted of patients who had cough and wheez­ ing, a diagnosis o f acute bronchitis with wheezing, or a diagnosis of asthmatic bronchitis at enrollment, and who Results did not have persistent asthma symptoms during the 6 months following enrollment. Patient Characteristics Group 3, Nonwheezinp respiratory illness (n = 89): Mean age of the 131 patients was 36.4 years (range 11 to This group included other patients with acute respiratory 78, standard deviation [SD] 13.5), 43.5% were male and illness and no clinical evidence for cither acute asthmatic 29.1% were smokers. One hundred twenty-seven (97%) bronchitis or asthma. subjects were 15 years of age or older. Eighty-three

The Journal of Family Practice, Vol. 38, No. 6(Jun), 1994 591 Asthma and Chlamydial Infection Hahn and Golubjatnikov

fable 1. Characteristics of Asthma Patients and Description of Respiratory' Illnesses That Preceded the Diagnosis of Asthma

Chlamydia Case Age Titer No. (y) Sex Category* COPD Category' Antecedent Respiratory Illnesses I t 31 F 32 None Bronchitis 2 t 36 F 32 None Pneumonia 3 t 46 M 256 None Bronchitis 4 t t 52 M 256 Moderately severe Episodes of asthmatic bronchitis 5 t§ 57 M 32 Moderate Pneumonia 6 58 F 64 Very severe Wheezing exacerbations of COPD 7 59 F 32 Moderate Bronchitis 8 t 60 M 512 Moderately severe Sore throat, gradual onset of wheezing, and dyspnea 9 63 M 32 Severe Wheezing with a “cold” 10 73 F 128 Mild Episodes of asthmatic bronchitis n 73 F 256 Mild Chronic cough and sinusitis 12 75 F 64 Severe Chronic bronchitis with episodes o f pneumonitis; possible interstitial lung disease * Greater o f the C pneumoniae acute or convalescent polyvalent titer. fNervly diagnosed asthma. pAcute C pneumoniae infection (serologic criteria). §Acute M pneumoniae infection (serologic criteria). COPD denotes chronic obstructive pulmonary disease.

(63.4%) o f 131 patients had C pneumoniae seroreactivity and dyspnea. He required oral theophylline and inhaled o f 1:16 or greater in the polyvalent MIF test (72% o f 49 bronchodilator therapy because chronic asthma devel­ men and 57% of 82 women, P = .10). Serologic evi­ oped and persisted until his death from other causes. dence for acute C pneumoniae infection was present in 7 Clinical data for asthma patients arc individually (5.3%) patients, and acute At pneumoniae infection was summarized in Table 1. The asthma diagnosis was pre­ found in an additional 10 (7.6%) patients. There were no ceded by a variety o f acute or chronic respiratory illnesses positive oropharyngeal cultures for C pneumoniae or M in each case. COPD tended to be age related, as younger pneumoniae. patients did not have it. Cases 2, 3, 7, and 10 had strong family histories of asthma. None of the asthma patients Clinical Findings in Patients with Atypical described in Table 1 had a history of childhood asthma, Infections and Asthma nor did any of them have histories implicating aeroaller- gens as triggers of asthma symptoms. In case 1, skin Nine of 10 patients with acute M pneumoniae infection testing with a battery o f common aeroallcrgens produced had atypical pneumonia confirmed by an infiltrate on negative results. IgE and blood eosinophils were not chest x-ray film. The remaining patient with acute At measured in this study. Therefore, the atopic status of pneumoniae infection had a diagnosis o f bronchitis, but these patients, as indicated by skin testing, IgE levels, and since he did not have a chest radiograph, it is possible eosinophilia, was not determined. All asthma patients that pneumonia might have been missed. Two patients improved clinically with standard antiasthma medication with acute AA pneumoniae infection had coexisting (inhaled albuterol, oral theophylline, and/or steroids). COPD, and one of these patients had prolonged dyspnea with spirometric evidence for reversible airway obstruc­ Comparison of Wheezing and Nonwheezing tion, which persisted during the postenrollment period but subsequently resolved (Table 1). Respiratory Illness Groups The clinical presentations of the seven patients with Table 2 compares clinical and serologic data for patients acute C pneumoniae infection were heterogeneous. Three with asthma, acute asthmatic bronchitis, and nonwheez­ patients had bronchitis (two also had pharyngitis or ing respiratory illness. Asthma patients were significantly laryngitis), one patient had atypical pneumonia, and two older (P < .001) and had COPD more often (P < .001) patients had acute asthmatic bronchitis (one of these had than did patients with asthmatic bronchitis or nonwheez­ a biphasic illness presentation). One of the patients with ing respiratory illness. Four asthma patients had mild to acute asthmatic bronchitis at enrollment later presented moderate COPD, and five had moderately severe to very with another wheezing episode and rapidly developed severe COPD accompanying their asthma diagnosis. severe, disabling, steroid-dependent asthma. The seventh There were no significant differences between groups in patient with acute C pneumoniae antibody presented with the prevalence of previous asthma, which did not appear mild pharyngitis and a history o f gradual onset o f wheeze to be reactivated by acute respiratory illness in group 3,

5 9 2 The lournal of Family Practice, Vol. 38, No. 6(Iun), 1994 \sthma and Chlamydial Infection Hahn and Golubjatnikov

Table 2. Study Group Characteristics and Associations with Atypical Infection by Diagnostic Category

Group 1 Group 2 Group 3 Asthmatic Bronchitis in Nonwheezing Patients Without Respiratory Asthma Asthma Illnesses* P Characteristic (n = 12) (n = 30) (n = 89) Value

Age, (y), mean (range) 56.9 (31-75) 32.2 (15-50) 34.9 (11-78) < 0 0 1 Sex (% male) 41.7 40.0 44.9 NS Current smoker (%) 41.7 43.3 23.3 NS Previous diagnoses (%) Asthmaf 58.3 16.7 7.9 NS COPD 75.0 6 .7 4.5 < .0 0 1 C pneumoniae Serologic acute infection (%) 16.7 3.3 4.5 NS Seroreactivity (%) 100 80.0 52.8 < .0 0 1 Geometric mean titer 76.1 29.2 18.9 .0001 M pneumoniae Serologic acute infection (%) 8.3 0.0 6 .9 NS 'Pharyngitis (28.1%), laryngitis (22.5%), sinusitis (27.0%), influenza-like illness (6.7%), biphasic illness (5.6%), bronchitis (43.8%), atypical pneumonia (25.8%). Excluding bronchitis, distribution o f diagnoses was not significantly different fo r other groups. Percentage adds to greater than 100 because some patients had multiple diagnoses. ‘Group 1: 7 (58.3%) o f 12 had an exacerbation o f previous asthma, and 5 (41.7%) had asthma diagnosed for the first time during this study; Group 2: 5 (16.7%) o f 30 patients with a previous diagnosis o f asthma did not have persistent symptoms during this study; Group 3: 7 (7.9%) o f 89 had a previous diagnosis o f inactive asthma. NS denotes not significant.

C pneumoniae seroreactivity, whether measured as Discussion seroprevalence or as geometric mean titer, was signifi­ cantly (P < .001 and P = .0001, respectively) more The case series reported here is consistent w ith previous common in asthma patients and in patients with asth­ observations that bronchitis often precedes the diagnosis matic bronchitis than in patients with nonwheezing ill­ of adult-onset asthma6’33 and that asthma presenting nesses. Using nonwheezing patients as controls, logistic after age 40 is often difficult to distinguish from, or regression analysis showed that these significant differ­ coexists with, obstructive airway disease.34 The associa­ ences persisted after adjusting for age, sex, and smoking tion of asthma with previous respiratory disease has been status. Significant dose-response associations of C pneu­ interpreted as evidence that preceding symptoms were moniae titer with both acute asthmatic bronchitis (P < actually due to undiagnosed asthma.7 Currently, there is .05) and asthma (P < .01) were also noted. After exclud­ little evidence that bacterial respiratory illnesses arc re­ ing the four asthmatic patients with documented im­ lated to the development of asthma or that any one provement in FE V j ranging from 12% to 14%, a signif­ specific pathogen is responsible for a significant propor­ icant (P = .02) difference in the prevalence of C tion of the respiratory illnesses preceding the develop­ pneumoniae seroreactivity' between group 1 and group 3 ment of asthma in adults.4 patients persisted. Our study coincided with an epidemic o f M pneu­ Comparable logistic regression models were devel­ moniae. Nevertheless, M pneumoniae infection was asso­ oped for indicators of M pneumoniae infection. In these ciated with only one case o f asthma in our study, and no analyses, no significant associations with asthma or acute antibody associations could be demonstrated. In studies asthmatic bronchitis were found for (1) CF seroreactivity' by other investigators, M pneumoniae infection has been (1:8 or greater), or (2) CF titer for M pneumoniae. associated with infectious exacerbations o f asthma in less “Preexisting” M pneumoniae infection could not be mea­ than 1% of older adults (mean age, 48 years),35 in 21% sured because o f the nature of the test, and therefore any of younger adults (mean age, 34 years),15 and in 25% of possible relationship of “old” antibody with reactive air­ children and adults less than 30 years of age.14 Our way disease could not be assessed. results provide further evidence that M pneumoniae in­ Dividing the study group, into patients enrolled be­ fection plays a minor role in asthma in older adults. tween October and December 1989, and those enrolled Observations on our case series of patients with after December 1989, acute asthmatic bronchitis re­ asthma and acute asthmatic bronchitis confirm and ex­ mained significantly associated with C pneumoniae sero­ tend previous findings,9 suggesting that C pneumoniae reactivity, titer category, and geometric mean titer within infection, as measured by seroreactivity, is associated each subgroup analyzed separately. with reactive airway disease. Our initial study, which did

The Journal of Family Practice, Vol. 38, No. 6(Jun), 1994 5 93 Asthma and Chlamydial Infection H ahn and Golubjatnikov

not employ pulmonary function testing, showed an as­ infection without having nasopharyngeal infection. If so sociation of C pneumoniae seroreactivity with wheezing bronchoscopic sampling o f pulmonary macrophages42 or and with the clinical diagnosis of acute asthmatic bron­ of tissue samples from the lower respiratory tract is chitis.9 Serologic evidence for acute secondary C pneu­ necessary to exclude chronic C pneumoniae infection in moniae infection was also found in some patients with asthma patients. newly diagnosed asthma as well as in patients with an Since patients with chronic lung disease may be exacerbation of previously diagnosed asthma. The case susceptible to infection by a variety of microorganisms, it series reported here included a significant association of is possible that C pneumoniae seroreactivity is a result of C pneumoniae seroreactivity with asthma confirmed by asthma rather than an antecedent risk factor for asthma. pulmonary function tests. This association of C pneumo­ However, seroreactivity was also associated with acute niae seroreactivity with asthma persisted after adjusting wheezing illness in many patients without evidence of for age, sex, and smoking status, or when a more strin­ chronic lung disease in our study. Either persistent in­ gent criterion for reversible airway obstruction was used fection with C pneumoniae or sensitization of T cells by (15% or greater increase in FEV, following bronchodi- previous chlamydial infection could plausibly contribute lator therapy, rather than the 12% or greater value cur­ to asthmatic inflammation. Ongoing chlamydial infec­ rently suggested by the ATS).30 tions cause inflammatory reactions at the site of infec­ It is unclear whether these serologic associations tion.43 Inflammatory damage might result from reactiva­ reflect previous exposure or persistent infection. The tion o f previously sensitized lymphocytes by cross usefulness o f serologic diagnosis without culture confir­ reaction with other infectious agents44 or with human mation o f C pneumoniae has been questioned recently, as antigens to cause autoimmune inflammation.45 As exam­ serologic criteria have been found to be problematic in ples, chlamydial infection causes inflammatory damage some studies in children.36 However, most investigators responsible for blindness in trachoma and infertility and performing microimmunofluorescence testing for C ectopic pregnancy in pelvic inflammatory disease.20 De­ pneumoniae feel that serologic criteria for acute primary layed hypersensitivity to a highly immunogenic chlamyd­ infection in adults are sensitive.37 Serologic diagnosis of ial protein (called “heat shock protein”) has been strongh secondary or chronic C pneumoniae infection, on the implicated in the pathogenesis o f trachoma,45-46 probable other hand, is difficult when four-fold titer changes or a is involved in tubal damage in chlamydial pelvic inflam­ convalescent titer is absent because of lack of an IgM matory disease,44-47-48 and may possibly be involved in response. There are also inherent problems in interpret­ the pathogenesis of sexually acquired reactive arthritis.49 ing an arbitrary serologic cutpoint.37 Although the serologic associations reported here Our studies employed a polyvalent (mixture of IgM, do not distinguish current infection from previous expo­ IgG and IgA) antibody test. Since most of the patients sure, they do suggest that C pneumoniax infection plays a reported here had IgM antibody titers of less than 1:16, role in the natural history o f adult-onset asthma. Further the findings in this study reflect associations of reversible studies of this important association are warranted. airway disease with IgG and/or IgA class antibody. IgA antibody has recently been suggested as a marker for chronic C pneumoniae infection in coronary heart dis­ Acknowledgments

ease.23 Further study is needed to determine whether the This study was supported by the Dean Foundation for Health, Re­ polyvalent antibody associations with asthma and asth­ search and Education, Madison, Wisconsin, and by the Wisconsin matic bronchitis reported here reflect IgA or IgG anti­ Academy o f Family Physicians, under the auspices o f the Wisconsin body activity. Research Network (WReN). The authors thank Bridget Pribbenow for technical assistance with data We did not succeed in culturing C pneumoniae from collection and data entry. the oropharynx of any of our patients. It is possible that nasopharyngeal swabbing rather than oropharyngeal swabbing, as well as improved transport of specimens to References

the laboratory, would have more likely yielded positive 1. Bivings L. Asthmatic bronchitis following chronic upper respira­ cultures. Optimization of sampling, specimen handling38 tory infection. JAMA 1940; 115:1434-6. and culture techniques,39~41 and use of polymerase chain 2. Chobot R, Uvitsky IH, Dundy H. The relationship of the etiologic reaction testing18 are necessary before concluding that C factors in asthma in infants and children. J Allergy 1951; 22:106- 10. pneumoniae is absent from the upper respiratory tract of 3. Fox JL. Infectious asthma treated with triacetyloleandomvein. symptomatic asthma patients. Chronic chlamydial infec­ Penn Med J 1961; 64:634-5. 4. Stenius-Aarniala B. The role of infection in asthma. Chest 1987; tions may involve deep tissues only.25 It is possible that 9 1 :1 3 0 -6 S . some categories of asthma patients have deep-seated lung 5. Korppi M, Rcijonen T, Poysa L, Juntunen-Backman K. A 2- to

5 9 4 The Journal of Family Practice, Vol. 38, No. 6(Jun), 1994 \sthma and Chlamydial Infection Hahn and Golubjatnikov

3-year outcome after bronchiolitis. Am J Dis Child 1993; 147: 27. Baum GL, Wolinsky E, eds. Textbook o f pulmonary diseases. Vol 628—31. 1. Boston: Little, Brown, 1989:358. 6. Williamson HA, Schultz P. An association between acute bronchi­ 28. Williamson HA. Pulmonary' function tests in acute bronchitis: tis and asthma. J Fam Pract 1987; 24:35-8. evidence for reversible airway obstruction. J Fam Pract 1987; 7. Dodge RR, Burrows B, Lebowitz MD, Cline MG. Antecedent 2 5 :2 5 1 -6 . features of children in whom asthma develops during the second 29. Hueston WJ. A comparison of albuterol and erythromycin for the decade of life. J Allergy Clin Immunol 1993; 92:744-9. treatment of acute bronchitis. J Fam Pract 1991; 33:476-80. 8. Weiss SG, Newcomb R W , Beem MO. Pulmonary assessment of 30. American Thoracic Society. Lung function testing: selection of children after chlamydial pneumonia o f infancy, j Pediatr 1986; reference values and interpretive strategies. Am J Resp Dis 1991; 108:659-64. 144:1202-18. 9. Hahn DL, Dodge R, Golubjatnikov R. Association of Chlamydia 31. American Thoracic Society'. Recommended standardized proce­ pneumoniae (strain TW A R) infection with wheezing, asthmatic dures for pulmonary function testing. Am J Resp Dis 1978- bronchitis and adult-onset asthma. JAMA 1991; 266:225-30. 1 1 8 :5 5 -7 7 . 10. Clyde WJ. Clinical overview o f typical Mycoplasma pneumoniae 32. The GLIM Working Party'. The GLIM (Generalised Linear Inter­ infections. Clin Infect Dis 1993; 17:S32-6. active Modelling) system. Version 3.77. London: Royal Statistical 11. Foy HM, Cooney MK, McMahan R, Grayston JT. Viral and Society 1986. mycoplasmal pneumonia in a prepaid medical care group during an 33. Hallett JS, Jacobs RL. Recurrent acute bronchitis: the association eight-year period. Am J Epidemiol 1973; 9 7 :9 3 -1 0 2 . with undiagnosed bronchial asthma. Ann Allergy' 1985; 55:568- 12. Leaver R, Weinberg EG. Is Mycoplasma pneumoniae a precipitat­ 70. ing factor in acute severe asthma in children? S Afr Med J 1985; 34. Dodge RR, Burrows B. The prevalence and incidence of asthma 68:78 -9 . and asthma-like symptoms in a general population sample. Am J Resp Dis 1980; 122:567-75. 1 3 . Teo J, Vellayappan K, Yip W C L, Doraisingham S. Mycoplasma 35. Huhti E , Mokka T , Nikoskelainen J. Association o f viral and pneumoniae and viral infections in childhood asthma. J Trop Pedi­ my'coplasma infections with exacerbations o f asthma. Ann Allergy atr 1986; 32:87-9. 1974; 33:145-9. .4. Gil JC, Cedillo RL, Mayagoitia BG, Paz MD. Isolation of Myco­ 36. Hammerschlag MR. Chlamydia pneumoniae infections. Pediatr In­ plasma pneumoniae from asthmatic patients. Ann Allergy' 1993; fect Dis J 1993; 1 2 :2 6 0 - L 70:23-5. 37. Grayston JT, Golubjatnikov R, Hagiwara T, Hahn DL, Leinonen 5. Seggev JS, Lis I, Siman-Tov R, Gutman R, Abu-Samara H, Schey M, Persson K, et al. Serologic tests for Chlamydia pneumoniae. G, Naot Y. Mycoplasma pneumoniae is a frequent cause o f exacer­ Pediatr Infect Dis J 1993; 12:790-1. bation of bronchial asthma in adults. Ann Allergy' 1986; 57: 38. Kuo C-C, Grayston JT. Factors affecting viability' and growth in 263-5. HeLa 229 cells o f Chlamydia sp. strain TW AR. J Clin Microbiol 6. Foy HM. Infections caused by Mycoplasma pneumoniae and possi­ 1988; 26:812-5. ble carrier state in different populations o f patients. Clin Infect Dis 39. Roblin PM, Dumornay W, Hammerschlag MR. Use of HEp-2 1993; 17:S37-46. cells for improved isolation and passage o f Chlamydia pneumoniae. 7. Grayston JT, Campbell LA, Kuo C-C, Mordhorst CH, Saikku P, J Clin Microbiol 1992; 30:1968-71. Thom DH, Wang S-P. A new respiratory tract pathogen: Chla­ 40. Weber PA, Buck M L, Hooper DG. Comparison o f six cell lines for mydia pneumoniae strain TWAR. J Infect Dis 1990; 161:618-25. the culture of C. pneumoniae. In: Abstracts of the 92nd American 8. Grayston JT. Infections caused by Chlamydia pneumoniae strain Society of Microbiology. Washington, DC: American Society of Mi­ TWAR. Clinical Infectious Diseases 1992; 1 5 :7 5 7 -6 3 . crobiology, 1992; C-423:491. 9. Aldous MB, Grayston JT, Wang S-P, Foy HM. Seroepidemiology 41. Wong KH, Skelton SK, Chan YK. Efficient culture of Chlamydia of Chlamydia pneumoniae TW A R infection in Seattle families, pnemnoniae with cell lines derived from the human respiratory 1966-1979. J Infect Dis 1992; 166:646-9. tract. J Clin Microbiol 1992; 30:1625-30. 5. Schachter J. Chlamy'dia as pathogens. Overview o f human diseases. 42. Black CM, Perez R. Chlamydia pneumoniae multiplies within hu­ In: Barron AL, ed. Microbiology of Chlamydia. Boca Raton, Fla, man pulmonary macrophages. In: Abstracts of the 90th Annual CRC Press, 1 9 8 8 :1 5 3 -6 5 . Meeting of the American Society for Microbiology. Washington, DC: f Hammerschlag M R, Chirgwin K, Roblin PM, Gelling M, Dumor- American Society of Microbiology', 1990; D -l:80. nay W, Mandel L, Schachter J. Persistent infection with Chlamydia 43. Kuo C-C. Chlamy'dia as pathogens. Host response. In: Barron AL, pneumoniae following acute respiratory illness. Clin Infect Dis ed. Microbiology o f Chlamydia. Boca Raton, Fla, C R C Press, 1992; 14:178-82. 1988:193-208. ' !. Gronhagen-Riska C, Saikku P, Riska H , Froseth B, Gray'ston JT. 44. Witkin SS, Jeremias J, Toth M, Ledger WJ. Cell-mediated immune Antibodies to TW A R, a novel type o f chlamydia, in sarcoidosis. response to the recombinant 57-kDa heat-shock protein of Chla­ In: Grassi C, Rizzato G, Pozzi E , eds. Sarcoidosis and other mydia trachomatis in women with salpingitis. J Infect Dis 1993; granulomatous disorders. Amsterdam: Elsevier Science Publishers 167:1379-83. BV, 1 9 8 8 :2 9 7 -3 0 1 . 45. Morrison RP, Belland RJ, Lyng K, Caldwell H D . Chlamydial 1. Saikku P, Leinonen M, Tenkanen L, Linnanmaki E, Ekman M-R, disease pathogenesis: the 57-kD chlamydial hypersensitivity anti­ Manninen V, et al. Chronic Chlamydia pneunwniae infection as a gen is a stress response protein. J Exp Med 1989; 1 7 0 :1 2 7 1 -8 3 . risk factor for coronary heart disease in the Helsinki Heart Study. 46. Morrison RP, Lyng K, Caldwell HD. Chlamydial disease patho­ Ann Intern Med 1992; 116:273-8. genesis: ocular hypersensitivity elicited by a genus-specific 57-kD ■■ Thom DH, Wang SP, Grayston TJ, Siscovick DS, Stewart DK, protein. J Exp Med 1989; 169:663-75. Kronmal RA, Weiss NS, et al. Chlamydia pneumoniae strain 47. Rice PA, Schacter J. Pathogenesis o f pelvic inflammatory' disease: TWAR antibody and angiographically demonstrated coronary ar­ what are the questions? JAMA 1991; 2 6 6 :2 5 8 7 -9 3 . tery disease. Arterioscler Thomb 1991; 1 1 :5 4 7 -5 1 . 48. Toye B, Laferriere C, Claman P, Jessamine P, Peeling R. Associ­ ■ Kuo C-C, Shor A, Campbell LA, Fukushi H, Patton DL, Grayston ation between antibody to the chlamydial heat-shock protein and JT. Demonstration of Chlamydia pneumoniae in atherosclerotic tubal infertility. J Infect Dis 1993; 1 6 8 :1 2 3 6 -4 0 . lesions of coronary arteries. J Infect Dis 1993; 167:841-9. 49. Inman RD, Johnston MEA, Chiu B, Falk J, Petrie M. Immuno­ • Fraser RG, Pare JAP, Pare PD, Fraser RS, Genereux GP, eds. chemical analysis o f immune response to Chlamydia trachomatis in Diagnosis o f diseases o f the chest. Vol. 3. Philadelphia: WB Reiter’s syndrome and nonspecific urethritis. Clin Exp Immunol Saunders, 1990:2089. 1987; 69:246-54.

IC Journal of Family Practice, Vol. 38, No. 6(Jun), 1994 595