15 February 1992
Volume 116
Number 4
Annals of Internal Medicine Chronic Chlamydia pneumoniae Infection as a Risk Factor for Coronary Heart Disease in the Helsinki Heart Study Pekka Saikku, MD, PhD; Maija Leinonen, PhD; Leena Tenkanen, MSc; Eila Linnanmaki, MSc; Marja-Riitta Ekman, MD; Vesa Manninen, MD, PhD; Matti Manttari, MD, PhD; M. Heikki Frick, MD, PhD; and Jussi K. Huttunen, MD, PhD
• Objective: To investigate in the prospective Helsinki Heart Study, whether chronic Chlamydia pneumoniae infection, indicated by elevated antibody titers against the pathogen, chlamydial lipopolysaccharide-containing immune complexes, or both, is a risk factor for coronary heart disease. • Design and Setting: The Helsinki Heart Study was a randomized, double-blind, 5-year clinical trial to test the efficacy of gemfibrozil in reducing the risk for coronary heart disease. Participants were randomized to receive either gemfibrozil (2046 patients) or placebo (2035 patients). Fatal and nonfatal myocardial infarction and sudden cardiac death were the main study end points. Serum samples were collected at 3-month intervals from all patients. • Patients: One hundred forty cardiac events occurred during the follow-up period. Serum samples from 103 case patients obtained 3 to 6 months before a cardiac end point were matched with those from controls for time point, locality, and treatment. Samples were tested for markers of chronic chlamydial infection. • Measurements: Immunoglobulin A (IgA) and G (IgG) antibodies to C. pneumoniae were measured using the microimmunofluorescence method. Lipopolysaccharide-containing immune complexes were measured using two antigen-specific enzyme immunoassays, the lipopolysaccharide-capture and immunoglobulin M (IgM)-capture methods. • Main Results: Using a conditional logistic regression model, odds ratios for the development of coronary heart disease were 2.7 (95% CI, 1.1 to 6.5) for elevated IgA titers, 2.1 (CI, 1.1 to 3.9) for the presence of immune complexes, and 2.9 (CI, 1.5 to 5.4) for the presence of both factors. If we adjusted for other coronary heart disease risk factors such as age, hypertension, and smoking, the corresponding values would be 2.3 (CI, 0.9 to 6.2), 1.8 (CI, 0.9 to 3.6), and 2.6 (CI, 1.3 to 5.2), respectively. • Conclusion: The results suggest that chronic C. pneumoniae infection may be a significant risk factor for the development of coronary heart disease. Annals of Internal Medicine. 1992;116:273-278. From the University of Helsinki, the National Public Health Institute, Helsinki; and the University of Tampere, Tampere, Finland. For current author addresses, see end of text.
{chlamydia pneumoniae, previously known as TWAR, is a recently recognized respiratory pathogen and a common causative agent of pneumonia, bronchitis, and mild respiratory infection (1, 2). The prevalence of antibodies to C. pneumoniae increases with age and is about 50% in middle-aged adults throughout the world (2). In Finland and other Scandinavian countries, C. pneumoniae causes widespread epidemics every 4 to 5 years (2-5), and estimates from antibody prevalence data indicate that every person will experience at least one C. pneumoniae infection during their lifetime. Chlamydia are obligatory intracellular, gram-negative bacteria known for their ability to cause chronic infections with severe sequelae; trachoma and infertility caused by C. trachomatis (6) are major public health concerns throughout the world. Approximately 500 million persons in developing countries have blinding trachoma, and 20 000 women per year in the United States are estimated to become infertile because of chlamydiainduced tubal scarring. We have previously observed an association between chronic C. pneumoniae infection and coronary heart disease as shown by the presence of elevated antibody levels or chlamydial lipopolysaccharide-containing immune complexes in the sera of 50% to 60% of patients with acute myocardial infarction or chronic coronary heart disease, compared with 7% to 12% among controls (7, 8). Recent studies in Seattle have confirmed our preliminary findings by showing an association between elevated C. pneumoniae titers and angiographically verified coronary heart disease (9). The concept of an association between microbial infections and coronary heart disease is not new; immunologic and microbiologic factors have been suggested to induce, directly or indirectly, the pathologic events leading to atherosclerosis (10). Chlamydia pneumoniae, a recently discovered common respiratory pathogen, is now suspected to play a role in the pathogenesis of atherosclerosis. Our previous studies could not exclude the possible activation of a latent C. pneumoniae infection by acute myocardial infarction. This prospective study describes C. pneumoniae antibody and lipopolysaccharide-containing immune complexes as risk factors for the development of coronary heart disease using patients from the Helsinki Heart Study (11). © 1992 American College of Physicians
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19744/ by a Universite Laval Biblioteque User on 06/11/2017
273
Methods
Chlamydial Studies
General Design
Antibody and immune complex determinations were done in a blinded fashion, with sera from both case patients and controls included in all test series. Immunoglobulin (Ig) classspecific antibodies to C. pneumoniae were measured using the microimmunofluorescence (micro-IF) method (15) with elementary bodies of TW 183-strain (University of Washington Research Foundation, Seattle, Washington) as antigen, using fluorescein-conjugated antihuman IgG, IgM (Dakopatts, Glostrup, Denmark), and IgA (Orion Diagnostica, Espoo, Finland). Serum samples obtained at 6 months were tested twice in a blinded fashion by two of the investigators using different antigen lots. A strong correlation was found for both IgG (r = 0.86) and IgA (r = 0.92) titers. When the titers differed, sera were titrated a third time, and the result obtained twice was considered to be correct. Baseline sera were tested only once in the same test series with the 6-month sera. Lipopolysaccharide-containing immune complexes were measured in serum samples at baseline and at six months using two antigen-specific enzyme immunoassays, the lipopolysaccharide-capture and IgM-capture methods. The details of these methods have been described previously (8). Briefly, the lipopolysaccharide-capture method uses a monoclonal antibody specific to chlamydial lipopolysaccharide to bind immune complexes to the solid phase, and the presence of immune complexes is shown by alkaline phosphatase-conjugated antihuman IgM. The IgM-capture method binds immune complexes to the solid phase using rabbit antibodies to the /i-chain of human IgM, and the presence of immune complexes is shown by the alkaline phosphatase-conjugated F(ab)2 fragment of the monoclonal antibody to chlamydial lipopolysaccharide.
The design and main results of the Helsinki Heart Study have been described in detail elsewhere (11-14). Briefly, the Helsinki Heart Study was a randomized, double-blind, 5-year clinical trial to test the efficacy of gemfibrozil, a fibric acid derivative, in reducing coronary heart disease risk by altering the levels of serum triglycerides and high-density lipoprotein (HDL) and low-density lipoprotein (LDL) cholesterol. The study began in 1980 and was completed in 1987. Study participants were selected from 23 531 men between 40 and 55 years of age who were employed by two state agencies and five industrial companies in Finland. Volunteers were enrolled in the trial if their non-HDL cholesterol level exceeded 5.2 mmol/L and if they had no evidence of coronary heart disease or other significant disease. The participants were randomized to receive either gemfibrozil (2046 patients) or placebo (2035 patients). Fatal and nonfatal confirmed myocardial infarction and sudden cardiac death were the principal end points of the study. Follow-up examinations, including laboratory studies of serum samples, were done at 3-month intervals and included a complete annual check-up. Data on smoking habits and physical activity (at baseline), symptoms suggesting acute myocardial infarction, and body weight and blood pressure were recorded. Total and HDL-cholesterol and triglyceride levels were determined from baseline and follow-up serum samples as described elsewhere (11-14). Serum Specimens We observed 140 cardiac end points during the 5 years of follow-up. Serum specimens from 103 of these men, obtained during follow-up visits 3 to 6 months (6-month serum) before the cardiac end point, were available for chlamydial studies. Serum samples were obtained from 103 controls who had no cardiac end point and who were individually matched with case patients for treatment, locality, and time point at the end of the trial. In addition, baseline serum samples were available from 96 of 103 cases and from 96 controls. All the sera were not available for all determinations.
Statistical Analysis The association between diagnostic criteria for chronic C. pneumoniae infection and coronary heart disease was analyzed using conditional logistic regression analysis of the pairwise matched data, with the classic risk factors as covariates in the model (16). For the analyses we used the statistical package EGRET (Statistics and Epidemiology Research Corporation, Seattle, Washington). We used age, systolic blood pressure, and the ratio of baseline levels of total cholesterol to HDL cholesterol as continuous variables. Smoking status was assessed by classifying participants as current smokers or nonsmokers.
Table 1. Prevalence of IgA and IgG Antibodies against Chlamydia pneumoniae and of Chlamydial Lipopolysaccharide Immune Complexes at Baseline and Six Months Results before Cardiac Event* The prevalences of positive IgG (> 16) and IgA (> 8) Diagnostic Baseline Six Months before antibody titers to C. pneumoniae and of lipopolysacchaCriteria Cardiac End Point ride immune complexes were high in serum samples at Case Controls Case Controls baseline and at 6 months: IgG antibodies were found in Patients (n = 87) Patients {n = 102) 76% to 78%, IgA antibodies in 41% to 53%, and lipo(n = 87) (n = 102) polysaccharide immune complexes in 30% to 50% of < % > serum samples (Table 1). No substantial changes in prevalences of elevated IgG (> 128) and IgA (> 64) IgA titer titers occurred during the follow-up period: elevated 64 23 14 22 10 a coronary event in 33% and 31% of case patients and Total 100 100 100 100 in 22% and 21% of controls, respectively. The correIgG titer sponding values for elevated IgA titers were 23% and < 16 22 24 25 24 22% in case patients and 14% and 10% in controls. The 32 to 64 45 54 44 55 > 128 33 22 31 21 presence of lipopolysaccharide immune complexes was Total 100 100 100 100 more variable. In the beginning of the study, 33% of ICt case patients and 30% of controls had immune comAbsent 67 70 48 67 plexes, whereas 6 months before the coronary event the Present 33 30 52 33 corresponding values were 52% and 33%, respectively. Total 100 100 100 100 We also analyzed the persistence of elevated anti* IC = Chlamydial lipopolysaccharide-containing immune complexes. body titers and the presence of lipopolysaccharide imt Samples were from 84 study participants each at baseline and 95 participants at 6 months before cardiac end point. mune complexes in individual case patients and controls 274
15 February 1992 • Annals of Internal Medicine • Volume 116 • Number 4
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19744/ by a Universite Laval Biblioteque User on 06/11/2017
Table 2. "Degree of Chronicity" of Chlamydia pneumoniae Infection Measured as the Persistence of High IgA Antibody Titers and Risk for Coronary Heart Disease IgA Titers
< 64 both at baseline and at 3 to 6 months before cardiac end point* > 64 either at baseline or at 3 to 6 months before cardiac end point > 64 both at baseline and at 3 to 6 months before cardiac end point
Odds Ratio (95% CI) Unadjusted Adjusted for Smoking
1
1
1.8t(0.7to4.6)
1.7$ (0.6 t© 4.4)
2.5t (0.9 to 6.5)
1-8* (0.7 to 5.1)
* Reference group. t Test for linear trend, P - 0.04. t Test for linear trend, P - 0.17.
during the follow-up period. Persistently elevated IgG titers were evident in 27.9% of case patients and 19.9% of controls, and persistently elevated IgA titers were found in 17.4% of case patients and 8.1% of controls. A persistent finding of immune complexes was seen in 25.3% and 16.4% of case patients and controls, respectively. The persistence of antibody and immune complex findings was considered to be a measure of the chronic nature of the infection. Table 2 shows the doseresponse analysis for the "degree of chronicity" for elevated IgA antibody levels. If elevated IgA antibody levels were present in only one of the two sera samples tested, the odds ratio for the development of a coronary event was 1.8; however, this ratio increased to 2.5 if elevations were present in both baseline and 6-month sera (P = 0.04 for linear trend). We then studied the relation of IgG and IgA C. pneumoniae antibody titers and lipopolysaccharide immune complexes to some classic risk factors for coronary heart disease (Table 3). As expected, the overall prevalence rates of the classic risk factors of smoking, age, hypertension, and dyslipidemia were greater among case patients than among controls. In addition, the
prevalences of smoking and age were positively associated with increasing levels of antibodies to C. pneumoniae among both case patients and controls. For example among participants with IgA titers of greater than 64, smokers comprised 82% of the patient group and 60% of the control group, whereas among participants with IgA titers of less than 8, the corresponding values were 48% and 34%, respectively. The prevalence of dyslipidemia, defined as a ratio of total cholesterol to HDL-cholesterol greater than 6, was negatively associated with increased levels of antibodies to C. pneumoniae. These findings indicate that age, dyslipidemia, and especially smoking must be considered as possible confounders of the association between chronic chlamydial infection and the risk for coronary heart disease. The odds ratios for the development of coronary heart disease according to conditional logistic regression analysis using different diagnostic criteria for chronic C. pneumoniae infection, with and without adjustment for classic risk factors, are shown in Table 4. At baseline, none of our diagnostic criteria was a significant risk factor for the development of coronary heart disease. Six months before infarction, however, the odds ratios for the presence of an IgA titer of > 64 were 2.7 {P = 0.02) and 2.3 (P = 0.09), respectively, without and with adjustment for other risk factors. The corresponding odds ratios for the presence of lipopolysaccharide immune complexes were 2.1 (P = 0.03) and 1.8 (P = 0.11). The combined presence of lipopolysaccharide immune complexes and an IgA titer of > 64 emerged as the most important risk factor, with an odds ratio of 2.9 (P = 0.001) without adjustment and 2.6 (P = 0.007) with adjustment for other risk factors. Because the adjustment did not essentially change the estimated odds ratio, the effect of C. pneumoniae infection on the risk for coronary heart disease-related events appears to be independent of the classic risk factors for coronary heart disease. Table 5 shows the dose-response pattern between IgA and IgG titers to C. pneumoniae and the risk for coronary heart disease. The association between increasing
Table 3. Prevalence of Classic Risk Factors for Coronary Heart Disease Compared with IgA and IgG Titers to Chlamydia pneumoniae and Presence of Chlamydial Lipopolysaccharide-containing Immune Complexes* Risk Factors for Coronary Heart Disease
< 8
IgAt 16-32
> 64
< 16
IC* > 128
Absent
Present
>
-%
< Current smoker Cases Controls Systolic blood pressure > 130 mm Hg and diastolic > 90 mm Hg Cases Controls Age > 47 years! Cases Controls Total cholesterol/HDL-cholesterol > 6 Cases Controls
IgGt 32-64
48 34
59 42
82 60
52 44
53 34
72 48
59 39
59 48
46 26
38 16
18 40
36 24
44 27
28 19
39 23
35 19
58 43
59 55
68 60
60 40
56 52
69 48
61 44
61 55
67 46
66 48
36 30
60 64
69 38
47 43
61 47
57 39
* IC = Chlamydial lipopolysaccharide-containing immune complexes; HDL = high-density lipoprotein. t Measured 3 to 6 months before cardiac end point. t Median age in the whole trial population (n = 4081).
15 February 1992 • Annals of Internal Medicine
• Volume 116 • Number 4
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19744/ by a Universite Laval Biblioteque User on 06/11/2017
275
Table 4. Odds Ratio for Developing Coronary Heart Disease during the 5-Year Follow-up by Categories of IgA and IgG Titers to Chlamydia pneumoniae and by Presence of Chlamydial Lipopolysaccharide Immune Complexes* Diagnostic Criteria
Not Adjusted by the Model
Baseline IgA > 64 IgG > 128 Six months before end point IgA > 64 IgG > 128 IC Combined IgA > 32 and/or IgG > 128 IgG > 128 and/or IC IgG > 128 and/or IgA > 32 and/or IC IgA > 64 and/or IC
Adjusted for Classical Risk Factorst P Odds Ratio Value (95% CI)
Odds Ratio (95% CI)
P Value
1.8 (0.8 to 3.9) 1.9 (0.9 to 4.0)
0.14 0.08
1.1 (0.5 to 2.7) 2.1 (0.9 to 5.2)
> 0.2 >0.2
2.7(1.1 to 6.5) 1.8 (0.9 to 3.4) 2.1 (1.1 to 3.9)
0.02 0.08 0.03
2.3 (0.9 to 6.2) 1.5 (0.7 to 3.2) 1.8 (0.9 to 3.6)
0.09 > 0.2 0.11
1.8 (1.0 2.6(1.3 2.3 (1.2 2.9 (1.5
0.04 0.005 0.01 0.001
1.6 (0.9 2.2(1.1 1.9 (0.9 2.6(1.3
to to to to
3.0) 5.0) 4.4) 5.4)
to to to to
3.0) 4.5) 3.9) 5.2)
0.12 0.04 0.08 0.007
* The reference group in each model consists of those with a titer lower than the cut-off point or without immune complexes. IC = Chlamydial lipopolysaccharide-containing immune complexes. t Age, smoking, systolic blood pressure, and baseline value of total cholesterol/high-density lipoprotein-cholesterol.
IgG titer and coronary heart disease risk was not significant. However, a dose response was observed for IgA antibodies; odds ratios were 1.1 and 1.3 for moderately elevated IgA titer and 1.9 and 2.9 (P = 0.02) for high titer (> 64) at baseline and 6 months before infarction, respectively. Discussion The results of this prospective study showed that chronic C. pneumoniae infection, as indicated by the presence of elevated IgA titer and lipopolysaccharide immune complexes 3 to 6 months before myocardial infarction, is an independent risk factor for the development of coronary heart disease. The presence of elevated IgG antibody titers as a risk factor did not, Table 5. Dose-Response Pattern between Risk for Coronary Heart Disease during a 5-Year Follow-up and Levels of IgG and IgA Titers to Chlamydia pneumoniae Titer
Odds Ratio (95% CI) Unadjusted Adjusted*
Baseline IgG ^ 16 1 32-64 1.0(0.5 to 2.0) > 128 1.9 (0.7 to 4.7) IgA 64 1.8 (0.8 to 4.0) Six months before end point IgG < 16 1 32-64 0.8 (0.4 to 1.7) > 128 1.6 (0.7 to 3.6) IgA < 8 1 16-32 1.3(0.7to2.4)t > 64 2.9(1.2to7.0)t
1 0.9 (0.4 to 2.2) 1.4 (0.5 to 4.1) 1 1.2 (0.5 to 2.7) 1.2 (0.5 to 2.9) 1 1.0(0.5 to 2.1) 1.5 (0.6 to 3.8) 1 1.1 (0.6 to 2.3) 2.4 (0.9 to 6.6)
* Age, smoking, blood pressure, and the ratio of total cholesterol to high-density lipoprotein-cholesterol were adjusted for by the model. t P = 0.024 for linear trend. 276
however, reach statistical significance. Our earlier discovery of an association of elevated C. pneumoniae titers (7) and presence of circulating immune complexes containing chlamydial lipopolysaccharide (8) with acute myocardial infarction is confirmed by our current results and is in agreement with an angiographic study published recently by Thorn and colleagues (9). Further, our results argue against the possibility that our previous findings could have been caused by the activation of a latent chlamydial infection by an acute myocardial infarction. The diagnosis of chronic infections is problematic. After an acute infection, IgG antibody titers rise and usually decrease slowly, whereas IgA antibodies tend to disappear more rapidly. Levels of antibody titers in patients with chronic infection are less well known. The persistence of elevated antibody titers is generally considered to be a sign of a chronic infection process. Recent evidence has suggested an association between elevated IgA titers and chronic bacterial infections, such as yersinial reactive arthritis (17), cystic fibrosis caused by Pseudomonas aeruginosa (18, 19), and chronic pelvic inflammatory disease caused by C. trachomatis (20, 21). Further, circulating immune complexes containing microbial components have also been observed in several chronic infections, such as syphilis, tuberculosis, leprosy, reactive arthritis, and cystic fibrosis (18, 22, 23). Similarly, we have shown the presence of immune complexes with chlamydial lipopolysaccharide in patients with pelvic inflammatory disease (24). In our present study we used elevated antibody titers and chlamydial lipopolysaccharide-containing immune complexes as diagnostic criteria for chronic C. pneumoniae infection. The presence of lipopolysaccharide immune complexes and a high IgA titer were shown to be significantly associated with the development of a coronary heart disease event. The risk for such coronary events associated with chronic chlamydial infection was shown to be independent of the classic risk factors of age, smoking, hypertension, and the ratio of total cholesterol to HDL-cholesterol levels.
15 February 1992 • Annals of Internal Medicine • Volume 116 • Number 4
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19744/ by a Universite Laval Biblioteque User on 06/11/2017
At the start of the study, signs of chlamydial infection were present in about 60% of case patients and in 40% of controls. In our earlier studies, 50% to 60% of case patients and only 10% to 20% of controls were shown to have either elevated antibody titers or lipopolysaccharide immune complexes in their sera after acute myocardial infarction (7, 8). The high percentage of signs of chlamydial infection observed among control patients in the Helsinki Heart Study can be explained to some extent by epidemiologic factors. An extensive C. pneumoniae epidemic occurred in Finland between 1977 and 1980 (3, 4, 25); therefore, when the Helsinki Heart Study began in 1980, the antibody prevalence in the Finnish population was at a high level (25). It is possible that both case patients and controls had both lipopolysaccharide immune complexes and elevated antibody titers caused by a recent infection and that positive findings persisted during follow-up only in those with chronic infection. Persistent elevations in IgA and IgG titers and the presence of immune complexes were clearly more common among case patients than among controls, thus supporting an association between chronic C. pneumoniae infection and the risk for coronary heart disease. Several mechanisms could allow a chronic chlamydial infection to affect the development of coronary heart disease and acute myocardial infarction. Chlamydia are gram-negative bacteria with a notable tendency to cause chronic infections (6). Chlamydia pneumoniae is capable of invading the lungs, causing about 5% to 10% of pneumoniae in adults (2). Chlamydia pneumoniae has also been shown to multiply in mononuclear phagocytes (26). A persistent infection of alveolar macrophages may be of special importance because of their close association with the circulatory system, which would allow easy access to the bloodstream for bacteria and bacterial components such as lipopolysaccharide. Lipopolysaccharide has powerful effects on vascular permeability and the clotting mechanism. Bacterial components can form immune complexes with antibodies currently present, a process that can lead to tissue injury. Lipopolysaccharide binds in human serum to both HDL and LDL, depending on their relative concentrations (27). Chlamydial lipopolysaccharide consists of a short, hydrophilic, strongly acidic 2-keto-3-deoxyoctulosonic acid (KDO) trisaccharide linked to lipid A with exceptionally long fatty acid chains (28, 29). Binding of chlamydial lipopolysaccharide to LDL could modify the lipoprotein, making it immunogenic or toxic to endothelial cells. Modified (30) or antibody-associated (31, 32) LDL is known to cause foam-cell formation in vitro, a process considered to be the first phase in the development of atherosclerosis. Chlamydia and their lipopolysaccharide component are also potent inducers of tumor necrosis factor (TNF) and IL-2 (33-35). Tumor necrosis factor inhibits the activity of lipoprotein lipase (36), leading to altered lipid metabolism and accumulation of triglycerides in the bloodstream. Continuous production of TNF in chronic chlamydial infection could contribute to the serum lipid pattern typical of coronary heart disease. Final verification of a pathogenetic role of chronic
chlamydial infection in the development of coronary heart disease is still lacking. Clinical intervention trials are needed as well as data directly showing the presence of C. pneumoniae in coronary lesions. A causal association between chronic C. pneumoniae infection and coronary heart disease could have widespread implications for public health, because the infection is common and amenable to treatment with appropriate antibiotics. Acknowledgments: The authors thank Anja Ratilainen, RN, and Teija Jaakkola, RN, for technical assistance. Grant Support: By grants from the Medical Council, Academy of Finland, Sigrid Juselius Foundation (P.S.), and the Meilahti Foundation. Requests for Reprints: Pekka Saikku, MD, PhD, Department of Virology, University of Helsinki, Haartmaninkatu 3, SF-00290 Helsinki, Finland. Current Author Addresses: Drs. Saikku and Ekman and Ms. Linnanmaki: Department of Virology, University of Helsinki, Haartmaninkatu 3, SF-00290 Helsinki, Finland. Drs. Leinonen and Huttunen: National Public Health Institute, Mannerheimintie 166, SF 00300 Helsinki, Finland. Ms. Tenkanen: Department of Public Health, University of Tampere, P.O. Box 607, SF 33101 Tampere, Finland. Drs. Manttari, Manninen, and Frick: First Department of Medicine, Helsinki University Central Hospital, Haartmaninkatu 4, SF-00290 Helsinki, Finland. References 1. Grayston JT, Kuo CC, Campbell LA, Wang SP. Chlamydia pneumoniae sp. nov. for Chlamydia sp. strain TWAR. Int J Syst Bacteriol. 1989;39:88-90. 2. Grayston JT, Campbell LA, Kuo CC, Mordhorst CH, Saikku P, Thorn DH, et al. A new respiratory tract pathogen: Chlamydia pneumoniae strain TWAR. J Infect Dis. 1990;161:618-25. 3. Saikku P, Wang SP, Kleemola M, Brander E, Rusanen E, Grayston JT. An epidemic of mild pneumonia due to an unusual strain of Chlamydia psittaci. J Infect Dis. 1985;151:832-9. 4. Kleemola M, Saikku P, Visakorpi R, Wang SP, Grayston JT. Epidemics of pneumonia caused by TWAR, a new Chlamydia organism, in military trainees in Finland. J Infect Dis. 1988;157:230-6. 5. Mordhorst CH, Wang SP, Myhra W, Grayston JT. Chlamydia pneumoniae, strain TWAR, infections in Denmark 1975-1987. In: Bowie WR, Caldwell HD, Jones RP, Mardh PA, Ridgway GL, Schachter J, et al.; eds. Chlamydial Infections. Cambridge:Cambridge University Press; 1990:418-21. 6. Mardh PA, Paavonen J, Puolakkainen M. Chlamydia. New York: Plenum Medical Book Company; 1989. 7. Saikku P, Leinonen M, Mattila K, Ekman MR, Nieminen MS, Makela PH, et al. Serological evidence of an association of a novel Chlamydia, TWAR, with chronic coronary heart disease and acute myocardial infarction. Lancet. 1988;2:983-6. 8. Leinonen M, Linnanmaki E, Mattila K, Nieminen MS, Valtonen V, Leirisalo-Repo M, et al. Circulating immune complexes containing chlamydial lipopolysaccharide in acute myocardial infarction. Microb Pathog. 1990;9:67-73. 9. Thorn DH, Wang SP, Grayston JT, Siscovick DS, Stewart DK, Kronmal RA, et al. Chlamydia pneumoniae strain TWAR antibody and angiographically demonstrated coronary artery disease. Arterioscler Thromb. 1991;11:547-51. 10. Lopes-Virella M, Virella G. Immunological and microbial factors in the pathogenesis of atherosclerosis. Clin Immunol Immunopathol. 1985;37:377-86. 11. Manninen V, Wood C; eds. Helsinki Heart Study: a controlled coronary prevention trial. Design and baseline findings. Eur Heart J. 1987;8:(Suppl 1)1-43. 12. Frick MH, Elo O, Haapa K, Heinonen OP, Heinsalmi P, Helo P, et al. Helsinki Heart Study: primary-prevention trial with gemfibrozil in middle-aged men with dyslipidemia. Safety of treatment, changes in risk factors, and incidence of coronary heart disease. N Engl J Med. 1987;317:1237-45. 13. Huttunen JK, Manninen V, Tenkanen L, Heinonen OP, Koskinen P, Frick MH. Drug-induced changes in HDL cholesterol and coronary heart disease. Experiences from the Helsinki Heart Study. In: Miller NE; ed. High Density Lipoproteins and Atherosclerosis. Amsterdam: Elsevier Science Publishers; 1989:191-8. 14. Manninen V, Huttunen JK, Tenkanen L, Heinonen OP, Manttarim, Frick MH. High density lipoprotein cholesterol as a risk factor for coronary heart disease in the Helsinki Heart Study. In: Miller NE; ed. High Density Lipoproteins and Atherosclerosis. Amsterdam: Elsevier Science Publishers; 1989:35-42.
15 February 1992 • Annals of Internal Medicine • Volume 116 • Number 4
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19744/ by a Universite Laval Biblioteque User on 06/11/2017
277
15. Wang SP, Grayston JT. Microimmunofluorescence serological studies with the TWAR organism. In: Oriel JD, Ridgway G, Schachter J, Taylor-Robinson D, Ward M; eds. Chlamydial Infections. Cambridge: Cambridge University Press; 1986;329-32. 16. Breslow NE, Day NE. Statistical Methods in Cancer Research. Vol. 1. The analysis of case-control studies. Lyon, France: International Agency for Research on Cancer, 1980:192-246. 17. Granfors K. Measurement of immunoglobulin M (IgM), IgG and IgA antibodies against Yersinia enterocolitica by enzyme-linked immunosorbent assay: persistence of serum antibodies during disease. J Clin Microbiol. 1979;9:336-41. 18. Hoibv N, Doring G. Immunoglobulin responses and immune complex formation in chronic lung diseases. Clin Microbiol Newsletter. 1988;9:133-6. 19. Brett MM, Ghoneim AT, Littlewood JM. Serum IgA antibodies against Pseudomonas aeruginosa in cystic fibrosis. Arch Dis Child. 1990;65:259-63. 20. Sarov I, Sarov B, Lunefeld E, Hagai Z, Chaim W, Figura B. The significance of chlamydia-specific serum IgA antibodies in Chlamydia trachomatis infections. 1st Proc Eur Soc Chlamydia Res. Stockholm: Almqvist & Wiksell International; 1988:234-8. 21. Chaim W, Sarov B, Sarov I, Piura B, Cohen A, Insler V. Serum IgG and IgA antibodies to chlamydia in ectopic pregnancies. Contraception. 1989;40:59-71. 22. Hojby N, Doring G, Schiotz PO. The role of immune complexes in the pathogenesis of bacterial infections. Ann Rev Microbiol. 1986; 40:29-53. 23. Lahesmaa-Rantala R, Granfors K, Kekomaki R, Toivanen A. Circulating yersinia specific immune complexes after acute yersiniosis; a follow-up study of patients with and without arthritis. Ann Rheum Dis. 1987;46:121-6. 24. Puolakkainen M, Linnanmaki E, Paavonen J, Leinonen M, Saikku P. Chlamydial lipopolysaccharide containing circulating immune complexes in pelvic inflammatory disease. In: Bowie WR, Caldwell HD, Jones RP, Mardh PA, Ridgway GL, Schachter J, et al.; eds. Chlamydial Infections. Cambridge: Cambridge University Press; 1990: 319-22. 25. Puolakkainen M, Ukkonen P, Saikku P. Seroepidemiology of chlamydia in Finland 1971-1987 by complement fixation test. Epidemiol Infect. 1989;102:287-95. 26. Black CM, Perez R. Chlamydia pneumoniae multiplies within hu-
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
man pulmonary macrophages. In: Abstracts of the 90th Annual Meeting of the American Society for Microbiology. Washington, DC: American Society for Microbiology, 1990:80. Van Lenten BJ, Fogelman AM, Haberland ME, Edwards PA. The role of lipoproteins and receptor-mediated endocytosis in the transport of bacterial lipopolysaccharide. Proc Natl Acad Sci USA. 1986;83:2704-8. Brade H, Brade L, Nano FE. Chemical and serological investigations on the genus-specific lipopolysaccharide epitope of Chlamydia. Proc Natl Acad Sci USA. 1987;84:2508-12. Nurminen ME, Rietschel ET, Brade H. Chemical characterization of Chlamydia trachomatis lipopolysaccharide. Infect Immun. 1985;48: 573-5. Steinberg D, Parthasarathy S, Carew TE, Khoo JC, Witzum JL. Beyond cholesterol. Modifications of low-density lipoprotein that increase its atherogenicity. N Eng J Med. 1989;320:915-24. Klimov AV, Denisenko AD, Popov AV, Nagorne VA, Pleskov VM, Vinogradov AG, et al. Lipoprotein-antibody immune complexes. Their catabolism and role in foam cell formation. Atherosclerosis. 1985;58:1-15. Griffith RL, Virella GT, Stevenson HC, Lopes-Virella MF. Low density lipoprotein metabolism by human macrophages activated with low density lipoprotein immune complexes. A possible mechanism of foam cell formation. J Exp Med. 1988;168:1041-59. Williams DM, Bonewald LF, Roodman GD, Byrne GI, Magee DM, Schachter J. Tumor necrosis factor alpha is a cytotoxin induced by murine Chlamydia trachomatis infection. Infect Immun. 1989;57: 1351-5. Williams DM, Magee DM, Bonewald LF, Smith JG, Bleicker CA, Byrne GI, et al. A role in vivo of tumor necrosis factor alpha in-host defense against Chlamvdia trachomatis. Infect Immun. 1990;58: 1572-6. Shemer-Avni Y, Holtmann H, Wallach D, Sarov I. Cytokine-cell interactions resulting in Chlamydia trachomatis inhibition—mechanisms involved. In: Bowie WR, Caldwell HD, Jones RP, Mardh PA, Ridgway GL, Schachter J, et al.; eds. Chlamydial Infections. Cambridge: Cambridge University Press; 1990:173-6. Kawakami M, Pekala PH, Lance MD, Cerami A. Lipoprotein lipase suppression in 3T3 LI cells by an endotoxin-induced mediator from exudate cells. Proc Natl Acad Sci USA. 1983;79:912-6.
People hardly ever make use of the freedom they have, for example freedom of thought; instead they demand freedom of speech as a compensation. Soren Kierkegaard A Kierkegaard Anthology Princeton Publishers
278
15 February 1992 • Annals
of Internal
Medicine
• V o l u m e 116 • N u m b e r 4
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19744/ by a Universite Laval Biblioteque User on 06/11/2017
Volume 116
Number 4
Annals of Internal Medicine Chronic Chlamydia pneumoniae Infection as a Risk Factor for Coronary Heart Disease in the Helsinki Heart Study Pekka Saikku, MD, PhD; Maija Leinonen, PhD; Leena Tenkanen, MSc; Eila Linnanmaki, MSc; Marja-Riitta Ekman, MD; Vesa Manninen, MD, PhD; Matti Manttari, MD, PhD; M. Heikki Frick, MD, PhD; and Jussi K. Huttunen, MD, PhD
• Objective: To investigate in the prospective Helsinki Heart Study, whether chronic Chlamydia pneumoniae infection, indicated by elevated antibody titers against the pathogen, chlamydial lipopolysaccharide-containing immune complexes, or both, is a risk factor for coronary heart disease. • Design and Setting: The Helsinki Heart Study was a randomized, double-blind, 5-year clinical trial to test the efficacy of gemfibrozil in reducing the risk for coronary heart disease. Participants were randomized to receive either gemfibrozil (2046 patients) or placebo (2035 patients). Fatal and nonfatal myocardial infarction and sudden cardiac death were the main study end points. Serum samples were collected at 3-month intervals from all patients. • Patients: One hundred forty cardiac events occurred during the follow-up period. Serum samples from 103 case patients obtained 3 to 6 months before a cardiac end point were matched with those from controls for time point, locality, and treatment. Samples were tested for markers of chronic chlamydial infection. • Measurements: Immunoglobulin A (IgA) and G (IgG) antibodies to C. pneumoniae were measured using the microimmunofluorescence method. Lipopolysaccharide-containing immune complexes were measured using two antigen-specific enzyme immunoassays, the lipopolysaccharide-capture and immunoglobulin M (IgM)-capture methods. • Main Results: Using a conditional logistic regression model, odds ratios for the development of coronary heart disease were 2.7 (95% CI, 1.1 to 6.5) for elevated IgA titers, 2.1 (CI, 1.1 to 3.9) for the presence of immune complexes, and 2.9 (CI, 1.5 to 5.4) for the presence of both factors. If we adjusted for other coronary heart disease risk factors such as age, hypertension, and smoking, the corresponding values would be 2.3 (CI, 0.9 to 6.2), 1.8 (CI, 0.9 to 3.6), and 2.6 (CI, 1.3 to 5.2), respectively. • Conclusion: The results suggest that chronic C. pneumoniae infection may be a significant risk factor for the development of coronary heart disease. Annals of Internal Medicine. 1992;116:273-278. From the University of Helsinki, the National Public Health Institute, Helsinki; and the University of Tampere, Tampere, Finland. For current author addresses, see end of text.
{chlamydia pneumoniae, previously known as TWAR, is a recently recognized respiratory pathogen and a common causative agent of pneumonia, bronchitis, and mild respiratory infection (1, 2). The prevalence of antibodies to C. pneumoniae increases with age and is about 50% in middle-aged adults throughout the world (2). In Finland and other Scandinavian countries, C. pneumoniae causes widespread epidemics every 4 to 5 years (2-5), and estimates from antibody prevalence data indicate that every person will experience at least one C. pneumoniae infection during their lifetime. Chlamydia are obligatory intracellular, gram-negative bacteria known for their ability to cause chronic infections with severe sequelae; trachoma and infertility caused by C. trachomatis (6) are major public health concerns throughout the world. Approximately 500 million persons in developing countries have blinding trachoma, and 20 000 women per year in the United States are estimated to become infertile because of chlamydiainduced tubal scarring. We have previously observed an association between chronic C. pneumoniae infection and coronary heart disease as shown by the presence of elevated antibody levels or chlamydial lipopolysaccharide-containing immune complexes in the sera of 50% to 60% of patients with acute myocardial infarction or chronic coronary heart disease, compared with 7% to 12% among controls (7, 8). Recent studies in Seattle have confirmed our preliminary findings by showing an association between elevated C. pneumoniae titers and angiographically verified coronary heart disease (9). The concept of an association between microbial infections and coronary heart disease is not new; immunologic and microbiologic factors have been suggested to induce, directly or indirectly, the pathologic events leading to atherosclerosis (10). Chlamydia pneumoniae, a recently discovered common respiratory pathogen, is now suspected to play a role in the pathogenesis of atherosclerosis. Our previous studies could not exclude the possible activation of a latent C. pneumoniae infection by acute myocardial infarction. This prospective study describes C. pneumoniae antibody and lipopolysaccharide-containing immune complexes as risk factors for the development of coronary heart disease using patients from the Helsinki Heart Study (11). © 1992 American College of Physicians
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19744/ by a Universite Laval Biblioteque User on 06/11/2017
273
Methods
Chlamydial Studies
General Design
Antibody and immune complex determinations were done in a blinded fashion, with sera from both case patients and controls included in all test series. Immunoglobulin (Ig) classspecific antibodies to C. pneumoniae were measured using the microimmunofluorescence (micro-IF) method (15) with elementary bodies of TW 183-strain (University of Washington Research Foundation, Seattle, Washington) as antigen, using fluorescein-conjugated antihuman IgG, IgM (Dakopatts, Glostrup, Denmark), and IgA (Orion Diagnostica, Espoo, Finland). Serum samples obtained at 6 months were tested twice in a blinded fashion by two of the investigators using different antigen lots. A strong correlation was found for both IgG (r = 0.86) and IgA (r = 0.92) titers. When the titers differed, sera were titrated a third time, and the result obtained twice was considered to be correct. Baseline sera were tested only once in the same test series with the 6-month sera. Lipopolysaccharide-containing immune complexes were measured in serum samples at baseline and at six months using two antigen-specific enzyme immunoassays, the lipopolysaccharide-capture and IgM-capture methods. The details of these methods have been described previously (8). Briefly, the lipopolysaccharide-capture method uses a monoclonal antibody specific to chlamydial lipopolysaccharide to bind immune complexes to the solid phase, and the presence of immune complexes is shown by alkaline phosphatase-conjugated antihuman IgM. The IgM-capture method binds immune complexes to the solid phase using rabbit antibodies to the /i-chain of human IgM, and the presence of immune complexes is shown by the alkaline phosphatase-conjugated F(ab)2 fragment of the monoclonal antibody to chlamydial lipopolysaccharide.
The design and main results of the Helsinki Heart Study have been described in detail elsewhere (11-14). Briefly, the Helsinki Heart Study was a randomized, double-blind, 5-year clinical trial to test the efficacy of gemfibrozil, a fibric acid derivative, in reducing coronary heart disease risk by altering the levels of serum triglycerides and high-density lipoprotein (HDL) and low-density lipoprotein (LDL) cholesterol. The study began in 1980 and was completed in 1987. Study participants were selected from 23 531 men between 40 and 55 years of age who were employed by two state agencies and five industrial companies in Finland. Volunteers were enrolled in the trial if their non-HDL cholesterol level exceeded 5.2 mmol/L and if they had no evidence of coronary heart disease or other significant disease. The participants were randomized to receive either gemfibrozil (2046 patients) or placebo (2035 patients). Fatal and nonfatal confirmed myocardial infarction and sudden cardiac death were the principal end points of the study. Follow-up examinations, including laboratory studies of serum samples, were done at 3-month intervals and included a complete annual check-up. Data on smoking habits and physical activity (at baseline), symptoms suggesting acute myocardial infarction, and body weight and blood pressure were recorded. Total and HDL-cholesterol and triglyceride levels were determined from baseline and follow-up serum samples as described elsewhere (11-14). Serum Specimens We observed 140 cardiac end points during the 5 years of follow-up. Serum specimens from 103 of these men, obtained during follow-up visits 3 to 6 months (6-month serum) before the cardiac end point, were available for chlamydial studies. Serum samples were obtained from 103 controls who had no cardiac end point and who were individually matched with case patients for treatment, locality, and time point at the end of the trial. In addition, baseline serum samples were available from 96 of 103 cases and from 96 controls. All the sera were not available for all determinations.
Statistical Analysis The association between diagnostic criteria for chronic C. pneumoniae infection and coronary heart disease was analyzed using conditional logistic regression analysis of the pairwise matched data, with the classic risk factors as covariates in the model (16). For the analyses we used the statistical package EGRET (Statistics and Epidemiology Research Corporation, Seattle, Washington). We used age, systolic blood pressure, and the ratio of baseline levels of total cholesterol to HDL cholesterol as continuous variables. Smoking status was assessed by classifying participants as current smokers or nonsmokers.
Table 1. Prevalence of IgA and IgG Antibodies against Chlamydia pneumoniae and of Chlamydial Lipopolysaccharide Immune Complexes at Baseline and Six Months Results before Cardiac Event* The prevalences of positive IgG (> 16) and IgA (> 8) Diagnostic Baseline Six Months before antibody titers to C. pneumoniae and of lipopolysacchaCriteria Cardiac End Point ride immune complexes were high in serum samples at Case Controls Case Controls baseline and at 6 months: IgG antibodies were found in Patients (n = 87) Patients {n = 102) 76% to 78%, IgA antibodies in 41% to 53%, and lipo(n = 87) (n = 102) polysaccharide immune complexes in 30% to 50% of < % > serum samples (Table 1). No substantial changes in prevalences of elevated IgG (> 128) and IgA (> 64) IgA titer titers occurred during the follow-up period: elevated 64 23 14 22 10 a coronary event in 33% and 31% of case patients and Total 100 100 100 100 in 22% and 21% of controls, respectively. The correIgG titer sponding values for elevated IgA titers were 23% and < 16 22 24 25 24 22% in case patients and 14% and 10% in controls. The 32 to 64 45 54 44 55 > 128 33 22 31 21 presence of lipopolysaccharide immune complexes was Total 100 100 100 100 more variable. In the beginning of the study, 33% of ICt case patients and 30% of controls had immune comAbsent 67 70 48 67 plexes, whereas 6 months before the coronary event the Present 33 30 52 33 corresponding values were 52% and 33%, respectively. Total 100 100 100 100 We also analyzed the persistence of elevated anti* IC = Chlamydial lipopolysaccharide-containing immune complexes. body titers and the presence of lipopolysaccharide imt Samples were from 84 study participants each at baseline and 95 participants at 6 months before cardiac end point. mune complexes in individual case patients and controls 274
15 February 1992 • Annals of Internal Medicine • Volume 116 • Number 4
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19744/ by a Universite Laval Biblioteque User on 06/11/2017
Table 2. "Degree of Chronicity" of Chlamydia pneumoniae Infection Measured as the Persistence of High IgA Antibody Titers and Risk for Coronary Heart Disease IgA Titers
< 64 both at baseline and at 3 to 6 months before cardiac end point* > 64 either at baseline or at 3 to 6 months before cardiac end point > 64 both at baseline and at 3 to 6 months before cardiac end point
Odds Ratio (95% CI) Unadjusted Adjusted for Smoking
1
1
1.8t(0.7to4.6)
1.7$ (0.6 t© 4.4)
2.5t (0.9 to 6.5)
1-8* (0.7 to 5.1)
* Reference group. t Test for linear trend, P - 0.04. t Test for linear trend, P - 0.17.
during the follow-up period. Persistently elevated IgG titers were evident in 27.9% of case patients and 19.9% of controls, and persistently elevated IgA titers were found in 17.4% of case patients and 8.1% of controls. A persistent finding of immune complexes was seen in 25.3% and 16.4% of case patients and controls, respectively. The persistence of antibody and immune complex findings was considered to be a measure of the chronic nature of the infection. Table 2 shows the doseresponse analysis for the "degree of chronicity" for elevated IgA antibody levels. If elevated IgA antibody levels were present in only one of the two sera samples tested, the odds ratio for the development of a coronary event was 1.8; however, this ratio increased to 2.5 if elevations were present in both baseline and 6-month sera (P = 0.04 for linear trend). We then studied the relation of IgG and IgA C. pneumoniae antibody titers and lipopolysaccharide immune complexes to some classic risk factors for coronary heart disease (Table 3). As expected, the overall prevalence rates of the classic risk factors of smoking, age, hypertension, and dyslipidemia were greater among case patients than among controls. In addition, the
prevalences of smoking and age were positively associated with increasing levels of antibodies to C. pneumoniae among both case patients and controls. For example among participants with IgA titers of greater than 64, smokers comprised 82% of the patient group and 60% of the control group, whereas among participants with IgA titers of less than 8, the corresponding values were 48% and 34%, respectively. The prevalence of dyslipidemia, defined as a ratio of total cholesterol to HDL-cholesterol greater than 6, was negatively associated with increased levels of antibodies to C. pneumoniae. These findings indicate that age, dyslipidemia, and especially smoking must be considered as possible confounders of the association between chronic chlamydial infection and the risk for coronary heart disease. The odds ratios for the development of coronary heart disease according to conditional logistic regression analysis using different diagnostic criteria for chronic C. pneumoniae infection, with and without adjustment for classic risk factors, are shown in Table 4. At baseline, none of our diagnostic criteria was a significant risk factor for the development of coronary heart disease. Six months before infarction, however, the odds ratios for the presence of an IgA titer of > 64 were 2.7 {P = 0.02) and 2.3 (P = 0.09), respectively, without and with adjustment for other risk factors. The corresponding odds ratios for the presence of lipopolysaccharide immune complexes were 2.1 (P = 0.03) and 1.8 (P = 0.11). The combined presence of lipopolysaccharide immune complexes and an IgA titer of > 64 emerged as the most important risk factor, with an odds ratio of 2.9 (P = 0.001) without adjustment and 2.6 (P = 0.007) with adjustment for other risk factors. Because the adjustment did not essentially change the estimated odds ratio, the effect of C. pneumoniae infection on the risk for coronary heart disease-related events appears to be independent of the classic risk factors for coronary heart disease. Table 5 shows the dose-response pattern between IgA and IgG titers to C. pneumoniae and the risk for coronary heart disease. The association between increasing
Table 3. Prevalence of Classic Risk Factors for Coronary Heart Disease Compared with IgA and IgG Titers to Chlamydia pneumoniae and Presence of Chlamydial Lipopolysaccharide-containing Immune Complexes* Risk Factors for Coronary Heart Disease
< 8
IgAt 16-32
> 64
< 16
IC* > 128
Absent
Present
>
-%
< Current smoker Cases Controls Systolic blood pressure > 130 mm Hg and diastolic > 90 mm Hg Cases Controls Age > 47 years! Cases Controls Total cholesterol/HDL-cholesterol > 6 Cases Controls
IgGt 32-64
48 34
59 42
82 60
52 44
53 34
72 48
59 39
59 48
46 26
38 16
18 40
36 24
44 27
28 19
39 23
35 19
58 43
59 55
68 60
60 40
56 52
69 48
61 44
61 55
67 46
66 48
36 30
60 64
69 38
47 43
61 47
57 39
* IC = Chlamydial lipopolysaccharide-containing immune complexes; HDL = high-density lipoprotein. t Measured 3 to 6 months before cardiac end point. t Median age in the whole trial population (n = 4081).
15 February 1992 • Annals of Internal Medicine
• Volume 116 • Number 4
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19744/ by a Universite Laval Biblioteque User on 06/11/2017
275
Table 4. Odds Ratio for Developing Coronary Heart Disease during the 5-Year Follow-up by Categories of IgA and IgG Titers to Chlamydia pneumoniae and by Presence of Chlamydial Lipopolysaccharide Immune Complexes* Diagnostic Criteria
Not Adjusted by the Model
Baseline IgA > 64 IgG > 128 Six months before end point IgA > 64 IgG > 128 IC Combined IgA > 32 and/or IgG > 128 IgG > 128 and/or IC IgG > 128 and/or IgA > 32 and/or IC IgA > 64 and/or IC
Adjusted for Classical Risk Factorst P Odds Ratio Value (95% CI)
Odds Ratio (95% CI)
P Value
1.8 (0.8 to 3.9) 1.9 (0.9 to 4.0)
0.14 0.08
1.1 (0.5 to 2.7) 2.1 (0.9 to 5.2)
> 0.2 >0.2
2.7(1.1 to 6.5) 1.8 (0.9 to 3.4) 2.1 (1.1 to 3.9)
0.02 0.08 0.03
2.3 (0.9 to 6.2) 1.5 (0.7 to 3.2) 1.8 (0.9 to 3.6)
0.09 > 0.2 0.11
1.8 (1.0 2.6(1.3 2.3 (1.2 2.9 (1.5
0.04 0.005 0.01 0.001
1.6 (0.9 2.2(1.1 1.9 (0.9 2.6(1.3
to to to to
3.0) 5.0) 4.4) 5.4)
to to to to
3.0) 4.5) 3.9) 5.2)
0.12 0.04 0.08 0.007
* The reference group in each model consists of those with a titer lower than the cut-off point or without immune complexes. IC = Chlamydial lipopolysaccharide-containing immune complexes. t Age, smoking, systolic blood pressure, and baseline value of total cholesterol/high-density lipoprotein-cholesterol.
IgG titer and coronary heart disease risk was not significant. However, a dose response was observed for IgA antibodies; odds ratios were 1.1 and 1.3 for moderately elevated IgA titer and 1.9 and 2.9 (P = 0.02) for high titer (> 64) at baseline and 6 months before infarction, respectively. Discussion The results of this prospective study showed that chronic C. pneumoniae infection, as indicated by the presence of elevated IgA titer and lipopolysaccharide immune complexes 3 to 6 months before myocardial infarction, is an independent risk factor for the development of coronary heart disease. The presence of elevated IgG antibody titers as a risk factor did not, Table 5. Dose-Response Pattern between Risk for Coronary Heart Disease during a 5-Year Follow-up and Levels of IgG and IgA Titers to Chlamydia pneumoniae Titer
Odds Ratio (95% CI) Unadjusted Adjusted*
Baseline IgG ^ 16 1 32-64 1.0(0.5 to 2.0) > 128 1.9 (0.7 to 4.7) IgA 64 1.8 (0.8 to 4.0) Six months before end point IgG < 16 1 32-64 0.8 (0.4 to 1.7) > 128 1.6 (0.7 to 3.6) IgA < 8 1 16-32 1.3(0.7to2.4)t > 64 2.9(1.2to7.0)t
1 0.9 (0.4 to 2.2) 1.4 (0.5 to 4.1) 1 1.2 (0.5 to 2.7) 1.2 (0.5 to 2.9) 1 1.0(0.5 to 2.1) 1.5 (0.6 to 3.8) 1 1.1 (0.6 to 2.3) 2.4 (0.9 to 6.6)
* Age, smoking, blood pressure, and the ratio of total cholesterol to high-density lipoprotein-cholesterol were adjusted for by the model. t P = 0.024 for linear trend. 276
however, reach statistical significance. Our earlier discovery of an association of elevated C. pneumoniae titers (7) and presence of circulating immune complexes containing chlamydial lipopolysaccharide (8) with acute myocardial infarction is confirmed by our current results and is in agreement with an angiographic study published recently by Thorn and colleagues (9). Further, our results argue against the possibility that our previous findings could have been caused by the activation of a latent chlamydial infection by an acute myocardial infarction. The diagnosis of chronic infections is problematic. After an acute infection, IgG antibody titers rise and usually decrease slowly, whereas IgA antibodies tend to disappear more rapidly. Levels of antibody titers in patients with chronic infection are less well known. The persistence of elevated antibody titers is generally considered to be a sign of a chronic infection process. Recent evidence has suggested an association between elevated IgA titers and chronic bacterial infections, such as yersinial reactive arthritis (17), cystic fibrosis caused by Pseudomonas aeruginosa (18, 19), and chronic pelvic inflammatory disease caused by C. trachomatis (20, 21). Further, circulating immune complexes containing microbial components have also been observed in several chronic infections, such as syphilis, tuberculosis, leprosy, reactive arthritis, and cystic fibrosis (18, 22, 23). Similarly, we have shown the presence of immune complexes with chlamydial lipopolysaccharide in patients with pelvic inflammatory disease (24). In our present study we used elevated antibody titers and chlamydial lipopolysaccharide-containing immune complexes as diagnostic criteria for chronic C. pneumoniae infection. The presence of lipopolysaccharide immune complexes and a high IgA titer were shown to be significantly associated with the development of a coronary heart disease event. The risk for such coronary events associated with chronic chlamydial infection was shown to be independent of the classic risk factors of age, smoking, hypertension, and the ratio of total cholesterol to HDL-cholesterol levels.
15 February 1992 • Annals of Internal Medicine • Volume 116 • Number 4
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19744/ by a Universite Laval Biblioteque User on 06/11/2017
At the start of the study, signs of chlamydial infection were present in about 60% of case patients and in 40% of controls. In our earlier studies, 50% to 60% of case patients and only 10% to 20% of controls were shown to have either elevated antibody titers or lipopolysaccharide immune complexes in their sera after acute myocardial infarction (7, 8). The high percentage of signs of chlamydial infection observed among control patients in the Helsinki Heart Study can be explained to some extent by epidemiologic factors. An extensive C. pneumoniae epidemic occurred in Finland between 1977 and 1980 (3, 4, 25); therefore, when the Helsinki Heart Study began in 1980, the antibody prevalence in the Finnish population was at a high level (25). It is possible that both case patients and controls had both lipopolysaccharide immune complexes and elevated antibody titers caused by a recent infection and that positive findings persisted during follow-up only in those with chronic infection. Persistent elevations in IgA and IgG titers and the presence of immune complexes were clearly more common among case patients than among controls, thus supporting an association between chronic C. pneumoniae infection and the risk for coronary heart disease. Several mechanisms could allow a chronic chlamydial infection to affect the development of coronary heart disease and acute myocardial infarction. Chlamydia are gram-negative bacteria with a notable tendency to cause chronic infections (6). Chlamydia pneumoniae is capable of invading the lungs, causing about 5% to 10% of pneumoniae in adults (2). Chlamydia pneumoniae has also been shown to multiply in mononuclear phagocytes (26). A persistent infection of alveolar macrophages may be of special importance because of their close association with the circulatory system, which would allow easy access to the bloodstream for bacteria and bacterial components such as lipopolysaccharide. Lipopolysaccharide has powerful effects on vascular permeability and the clotting mechanism. Bacterial components can form immune complexes with antibodies currently present, a process that can lead to tissue injury. Lipopolysaccharide binds in human serum to both HDL and LDL, depending on their relative concentrations (27). Chlamydial lipopolysaccharide consists of a short, hydrophilic, strongly acidic 2-keto-3-deoxyoctulosonic acid (KDO) trisaccharide linked to lipid A with exceptionally long fatty acid chains (28, 29). Binding of chlamydial lipopolysaccharide to LDL could modify the lipoprotein, making it immunogenic or toxic to endothelial cells. Modified (30) or antibody-associated (31, 32) LDL is known to cause foam-cell formation in vitro, a process considered to be the first phase in the development of atherosclerosis. Chlamydia and their lipopolysaccharide component are also potent inducers of tumor necrosis factor (TNF) and IL-2 (33-35). Tumor necrosis factor inhibits the activity of lipoprotein lipase (36), leading to altered lipid metabolism and accumulation of triglycerides in the bloodstream. Continuous production of TNF in chronic chlamydial infection could contribute to the serum lipid pattern typical of coronary heart disease. Final verification of a pathogenetic role of chronic
chlamydial infection in the development of coronary heart disease is still lacking. Clinical intervention trials are needed as well as data directly showing the presence of C. pneumoniae in coronary lesions. A causal association between chronic C. pneumoniae infection and coronary heart disease could have widespread implications for public health, because the infection is common and amenable to treatment with appropriate antibiotics. Acknowledgments: The authors thank Anja Ratilainen, RN, and Teija Jaakkola, RN, for technical assistance. Grant Support: By grants from the Medical Council, Academy of Finland, Sigrid Juselius Foundation (P.S.), and the Meilahti Foundation. Requests for Reprints: Pekka Saikku, MD, PhD, Department of Virology, University of Helsinki, Haartmaninkatu 3, SF-00290 Helsinki, Finland. Current Author Addresses: Drs. Saikku and Ekman and Ms. Linnanmaki: Department of Virology, University of Helsinki, Haartmaninkatu 3, SF-00290 Helsinki, Finland. Drs. Leinonen and Huttunen: National Public Health Institute, Mannerheimintie 166, SF 00300 Helsinki, Finland. Ms. Tenkanen: Department of Public Health, University of Tampere, P.O. Box 607, SF 33101 Tampere, Finland. Drs. Manttari, Manninen, and Frick: First Department of Medicine, Helsinki University Central Hospital, Haartmaninkatu 4, SF-00290 Helsinki, Finland. References 1. Grayston JT, Kuo CC, Campbell LA, Wang SP. Chlamydia pneumoniae sp. nov. for Chlamydia sp. strain TWAR. Int J Syst Bacteriol. 1989;39:88-90. 2. Grayston JT, Campbell LA, Kuo CC, Mordhorst CH, Saikku P, Thorn DH, et al. A new respiratory tract pathogen: Chlamydia pneumoniae strain TWAR. J Infect Dis. 1990;161:618-25. 3. Saikku P, Wang SP, Kleemola M, Brander E, Rusanen E, Grayston JT. An epidemic of mild pneumonia due to an unusual strain of Chlamydia psittaci. J Infect Dis. 1985;151:832-9. 4. Kleemola M, Saikku P, Visakorpi R, Wang SP, Grayston JT. Epidemics of pneumonia caused by TWAR, a new Chlamydia organism, in military trainees in Finland. J Infect Dis. 1988;157:230-6. 5. Mordhorst CH, Wang SP, Myhra W, Grayston JT. Chlamydia pneumoniae, strain TWAR, infections in Denmark 1975-1987. In: Bowie WR, Caldwell HD, Jones RP, Mardh PA, Ridgway GL, Schachter J, et al.; eds. Chlamydial Infections. Cambridge:Cambridge University Press; 1990:418-21. 6. Mardh PA, Paavonen J, Puolakkainen M. Chlamydia. New York: Plenum Medical Book Company; 1989. 7. Saikku P, Leinonen M, Mattila K, Ekman MR, Nieminen MS, Makela PH, et al. Serological evidence of an association of a novel Chlamydia, TWAR, with chronic coronary heart disease and acute myocardial infarction. Lancet. 1988;2:983-6. 8. Leinonen M, Linnanmaki E, Mattila K, Nieminen MS, Valtonen V, Leirisalo-Repo M, et al. Circulating immune complexes containing chlamydial lipopolysaccharide in acute myocardial infarction. Microb Pathog. 1990;9:67-73. 9. Thorn DH, Wang SP, Grayston JT, Siscovick DS, Stewart DK, Kronmal RA, et al. Chlamydia pneumoniae strain TWAR antibody and angiographically demonstrated coronary artery disease. Arterioscler Thromb. 1991;11:547-51. 10. Lopes-Virella M, Virella G. Immunological and microbial factors in the pathogenesis of atherosclerosis. Clin Immunol Immunopathol. 1985;37:377-86. 11. Manninen V, Wood C; eds. Helsinki Heart Study: a controlled coronary prevention trial. Design and baseline findings. Eur Heart J. 1987;8:(Suppl 1)1-43. 12. Frick MH, Elo O, Haapa K, Heinonen OP, Heinsalmi P, Helo P, et al. Helsinki Heart Study: primary-prevention trial with gemfibrozil in middle-aged men with dyslipidemia. Safety of treatment, changes in risk factors, and incidence of coronary heart disease. N Engl J Med. 1987;317:1237-45. 13. Huttunen JK, Manninen V, Tenkanen L, Heinonen OP, Koskinen P, Frick MH. Drug-induced changes in HDL cholesterol and coronary heart disease. Experiences from the Helsinki Heart Study. In: Miller NE; ed. High Density Lipoproteins and Atherosclerosis. Amsterdam: Elsevier Science Publishers; 1989:191-8. 14. Manninen V, Huttunen JK, Tenkanen L, Heinonen OP, Manttarim, Frick MH. High density lipoprotein cholesterol as a risk factor for coronary heart disease in the Helsinki Heart Study. In: Miller NE; ed. High Density Lipoproteins and Atherosclerosis. Amsterdam: Elsevier Science Publishers; 1989:35-42.
15 February 1992 • Annals of Internal Medicine • Volume 116 • Number 4
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19744/ by a Universite Laval Biblioteque User on 06/11/2017
277
15. Wang SP, Grayston JT. Microimmunofluorescence serological studies with the TWAR organism. In: Oriel JD, Ridgway G, Schachter J, Taylor-Robinson D, Ward M; eds. Chlamydial Infections. Cambridge: Cambridge University Press; 1986;329-32. 16. Breslow NE, Day NE. Statistical Methods in Cancer Research. Vol. 1. The analysis of case-control studies. Lyon, France: International Agency for Research on Cancer, 1980:192-246. 17. Granfors K. Measurement of immunoglobulin M (IgM), IgG and IgA antibodies against Yersinia enterocolitica by enzyme-linked immunosorbent assay: persistence of serum antibodies during disease. J Clin Microbiol. 1979;9:336-41. 18. Hoibv N, Doring G. Immunoglobulin responses and immune complex formation in chronic lung diseases. Clin Microbiol Newsletter. 1988;9:133-6. 19. Brett MM, Ghoneim AT, Littlewood JM. Serum IgA antibodies against Pseudomonas aeruginosa in cystic fibrosis. Arch Dis Child. 1990;65:259-63. 20. Sarov I, Sarov B, Lunefeld E, Hagai Z, Chaim W, Figura B. The significance of chlamydia-specific serum IgA antibodies in Chlamydia trachomatis infections. 1st Proc Eur Soc Chlamydia Res. Stockholm: Almqvist & Wiksell International; 1988:234-8. 21. Chaim W, Sarov B, Sarov I, Piura B, Cohen A, Insler V. Serum IgG and IgA antibodies to chlamydia in ectopic pregnancies. Contraception. 1989;40:59-71. 22. Hojby N, Doring G, Schiotz PO. The role of immune complexes in the pathogenesis of bacterial infections. Ann Rev Microbiol. 1986; 40:29-53. 23. Lahesmaa-Rantala R, Granfors K, Kekomaki R, Toivanen A. Circulating yersinia specific immune complexes after acute yersiniosis; a follow-up study of patients with and without arthritis. Ann Rheum Dis. 1987;46:121-6. 24. Puolakkainen M, Linnanmaki E, Paavonen J, Leinonen M, Saikku P. Chlamydial lipopolysaccharide containing circulating immune complexes in pelvic inflammatory disease. In: Bowie WR, Caldwell HD, Jones RP, Mardh PA, Ridgway GL, Schachter J, et al.; eds. Chlamydial Infections. Cambridge: Cambridge University Press; 1990: 319-22. 25. Puolakkainen M, Ukkonen P, Saikku P. Seroepidemiology of chlamydia in Finland 1971-1987 by complement fixation test. Epidemiol Infect. 1989;102:287-95. 26. Black CM, Perez R. Chlamydia pneumoniae multiplies within hu-
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
man pulmonary macrophages. In: Abstracts of the 90th Annual Meeting of the American Society for Microbiology. Washington, DC: American Society for Microbiology, 1990:80. Van Lenten BJ, Fogelman AM, Haberland ME, Edwards PA. The role of lipoproteins and receptor-mediated endocytosis in the transport of bacterial lipopolysaccharide. Proc Natl Acad Sci USA. 1986;83:2704-8. Brade H, Brade L, Nano FE. Chemical and serological investigations on the genus-specific lipopolysaccharide epitope of Chlamydia. Proc Natl Acad Sci USA. 1987;84:2508-12. Nurminen ME, Rietschel ET, Brade H. Chemical characterization of Chlamydia trachomatis lipopolysaccharide. Infect Immun. 1985;48: 573-5. Steinberg D, Parthasarathy S, Carew TE, Khoo JC, Witzum JL. Beyond cholesterol. Modifications of low-density lipoprotein that increase its atherogenicity. N Eng J Med. 1989;320:915-24. Klimov AV, Denisenko AD, Popov AV, Nagorne VA, Pleskov VM, Vinogradov AG, et al. Lipoprotein-antibody immune complexes. Their catabolism and role in foam cell formation. Atherosclerosis. 1985;58:1-15. Griffith RL, Virella GT, Stevenson HC, Lopes-Virella MF. Low density lipoprotein metabolism by human macrophages activated with low density lipoprotein immune complexes. A possible mechanism of foam cell formation. J Exp Med. 1988;168:1041-59. Williams DM, Bonewald LF, Roodman GD, Byrne GI, Magee DM, Schachter J. Tumor necrosis factor alpha is a cytotoxin induced by murine Chlamydia trachomatis infection. Infect Immun. 1989;57: 1351-5. Williams DM, Magee DM, Bonewald LF, Smith JG, Bleicker CA, Byrne GI, et al. A role in vivo of tumor necrosis factor alpha in-host defense against Chlamvdia trachomatis. Infect Immun. 1990;58: 1572-6. Shemer-Avni Y, Holtmann H, Wallach D, Sarov I. Cytokine-cell interactions resulting in Chlamydia trachomatis inhibition—mechanisms involved. In: Bowie WR, Caldwell HD, Jones RP, Mardh PA, Ridgway GL, Schachter J, et al.; eds. Chlamydial Infections. Cambridge: Cambridge University Press; 1990:173-6. Kawakami M, Pekala PH, Lance MD, Cerami A. Lipoprotein lipase suppression in 3T3 LI cells by an endotoxin-induced mediator from exudate cells. Proc Natl Acad Sci USA. 1983;79:912-6.
People hardly ever make use of the freedom they have, for example freedom of thought; instead they demand freedom of speech as a compensation. Soren Kierkegaard A Kierkegaard Anthology Princeton Publishers
278
15 February 1992 • Annals
of Internal
Medicine
• V o l u m e 116 • N u m b e r 4
Downloaded From: http://annals.org/pdfaccess.ashx?url=/data/journals/aim/19744/ by a Universite Laval Biblioteque User on 06/11/2017
