2026 年 33 巻 4 号 p. 428-440
Aims: Persistent Chlamydia pneumoniae (C. pneumoniae) infection has been suggested to be a risk factor for cardiovascular events; however, only findings from studies on small populations are available so far. This study investigated this hypothesis in a large general population through a longitudinal analysis.
Methods: We included 9,064 community residents who participated in the Nagahama study (mean age: 52.8 years). C. pneumoniae infection (seropositivity) was determined by serum levels of immunoglobulin A and immunoglobulin G assessed by enzyme-linked immunoassay. The incidence rates of cardiovascular diseases (CVDs), including stroke and coronary artery diseases, were determined by reviewing participants’ hospital records and death certificates. Basic clinical parameters were obtained using the baseline survey of the Nagahama study.
Results: During a mean follow-up duration of 4,390 days, we observed 323 cases of CVDs. The incidence rates of CVDs were 45.0 and 24.5 per 10,000 person-years in the seropositive and seronegative groups, respectively (log-rank test: p＜0.001). The results of the Cox proportional hazard model analysis indicated that C. pneumoniae seropositivity was remarkably associated with CVDs (1.30, 95% confidence interval: 1.04−1.64) after adjusting for established risk factors, including arterial stiffness (p = 0.023). The hazard ratio was higher in the subpopulation aged ≤ 55 years (2.62, 95% confidence interval: 1.45−4.75, p = 0.001) and reached 3.66 (95% confidence interval: 1.39–9.65, p = 0.009) in the subpopulation aged ≤ 45 years.
Conclusion: C. pneumoniae seropositivity was significantly associated with CVDs incidence, especially in adolescents and middle-aged individuals.
Chlamydia pneumoniae (C. pneumoniae) is the most common nonviral human respiratory pathogen, accounting for approximately 10% of community-acquired pneumonia cases1-4). Persistent C. pneumoniae infection may also be associated with pathological changes in large arteries5, 6), including carotid hypertrophy7), atherosclerotic lesions as identified by angiography8), and coronary artery calcification progression9), although contradictory findings have also been reported10, 11). A possible mechanism by which C. pneumoniae promotes atherosclerosis is the enhancement of adhesion molecule expression via migration of circulating monocytes infected with C. pneumoniae through the endothelium6). We previously reported that plasma concentrations of soluble forms of intercellular adhesion molecule were significantly higher in community residents infected with C. pneumoniae12).
In contrast, in a meta-analysis of longitudinal studies on coronary heart disease, C. pneumoniae immunoglobulin (Ig) G titers did not show a statistically significant association13), whereas IgA titers showed a weak association14); however, the studies included in the meta-analysis had major limitations. Several retrospective studies that adopted the nested case–control study design did not adjust for smoking13, 14), an important risk factor for coronary artery diseases and C. pneumoniae respiratory tract infection, or conventional risk factors for CVDs, including blood pressure and plasma lipid levels15). Limited statistical power in either a nested case–control study13-17) or a longitudinal study18, 19) due to a smaller sample size may also hinder the elucidation of the relationship between C. pneumoniae infection and CVD. Furthermore, except for a well-designed nested case–control study in Japan that reported a remarkable association between C. pneumoniae infection and coronary heart disease17), most studies were based on populations in the 1980s; thus, their clinical features and environmental conditions may differ somewhat from those of the current population.
This study aimed to investigate the possible association between persistent infection with C. pneumoniae, as determined by IgA and IgG seropositivity, and CVD development, through a longitudinal study of a large general population with a baseline survey conducted between 2008 and 2010. Because a previous epidemiological study20) and an animal study21) suggested that the harmful effects of C. pneumoniae infection may vary with age, this study also investigated the interaction of age in the association between C. pneumoniae seropositivity and CVDs. We also investigated the association between infection with Mycoplasma pneumoniae (M. pneumoniae), another atypical pathogen that causes community-acquired pneumonia, and CVDs because M. pneumoniae predisposes patients to risk of CVDs22).
The study participants were consenting community residents who voluntarily participated in the Nagahama study23, 24), a longitudinal cohort study aimed at investigating factors associated with noncommunicable diseases. Participants were recruited between 2008 and 2010 from physically independent Nagahama City residents aged between 30 years and 75 years. Nagahama City, located in central Japan, had an approximate population of 85,000 inhabitants in 2008. Among 9,764 individuals who participated in the baseline survey, 9,064 were eventually included in this study after applying the following exclusion criteria: pacemaker implantation (n = 12), hemodialysis therapy (n = 5), histories of cardiovascular diseases including stroke and myocardial infarction (n = 262), pregnancy (n = 42), large bilateral differences (Mahalanobis distance ＞5) in brachial-ankle pulse wave velocity (PWV) (n = 327), and incomplete or widely deviating clinical values required for this study (n = 52).
All study procedures were approved by the ethics committee of Kyoto University Graduate School of Medicine and by the Nagahama Municipal Review Board. Written informed consent was obtained from all participants.
Outcome DefinitionThe incidence rates of coronary artery events (including myocardial infarction, coronary artery bypass grafting, or angioplasty25, 26)) and stroke (including cerebral infarction and intracerebral and subarachnoid hemorrhage) were obtained from personal medical records stored at three emergency care hospitals (Nagahama City Hospital, Nagahama Red Cross Hospital, and Nagahama City Kohoku Hospital). Stroke was defined as the sudden onset of neurological symptoms lasting for at least 24 h, including death associated with neurological deficits, as defined by the World Health Organization Monitoring of Trends and Determinants in Cardiovascular Disease (WHO-MONICA) project27). Myocardial infarction was defined according to the WHO-MONICA project criteria27) by considering the symptoms at onset, electrocardiographic findings, relevant blood markers (including creatine kinase), and coronary angiographic findings. Participants who underwent coronary angioplasty or coronary artery bypass grafting were also identified from the medical record and considered incident cases of coronary artery diseases. Death records were also reviewed at the local healthcare center to identify individuals who suddenly died of acute myocardial infarction or stroke.
Follow-Up PeriodThe follow-up period was calculated as the interval (in days) from the date of participation in the baseline survey to the earliest date of incidence of CVDs or the end of the current follow-up period (December 15, 2022). Individuals who either died or relocated to other municipalities were classified as censored cases.
C. pneumoniae SeropositivityA commercially available enzyme-linked immunoassay kit (HITAZYME C. pneumoniae, Hitachi Chemical, Tokyo, Japan) that detects antibodies to the chlamydial outer membrane complex was used to determine serum levels of IgA and IgG for C. pneumoniae. IgA and IgG levels in each sample were expressed as the IgA or IgG index, respectively. Seropositivity to C. pneumoniae was diagnosed as an IgA and IgG index of ＞1.1 for both. Details of the enzyme-linked immunoassay assay and the validity of the cut-off point have been described previously12). The mean intra-assay coefficients of variation ranged from 3.3% to 7.7% for the IgA index and from 6.3% to 9.0% for the IgG index, and the mean inter-assay coefficients of variation ranged from 3.8% to 10.7% and from 4.0% to 10.7%, respectively.
M. pneumoniae SeropositivityM. pneumoniae seropositivity was assessed using a semiquantitative particle agglutination test kit (Serodia-Myco II, Fujirebio, Tokyo, Japan) consisting of gelatin particles coated with cell membrane components of M. pneumoniae (Mac strain)28). Serum samples were serially diluted to yield final dilutions of 1:40 to 1:20,480. According to the manufacturer’s instructions, antibody titers of ≥ 1:40 were considered seropositive29, 30).
Clinical ParametersClinical parameters used in this study were obtained at the baseline survey. The brachial blood pressure was measured twice using a cuff-oscillometric device (HEM-9000AI, Omron Healthcare, Kyoto, Japan) after a few minutes of rest in a sitting position, and the average of the two measurements was used for the analysis. PWV measured between the brachium and the ankle was used as an index of arterial stiffness. For PWV measurement, cuffs were attached to the brachia and ankles, and pulse volume waveforms were simultaneously recorded while the participants were in a supine position using a plethysmographic sensor connected to the cuffs (Vasera-1500, Fukuda Denshi Co., Ltd., Tokyo, Japan). The PWV was calculated from the time interval between the wavefronts of the brachial and ankle waveforms and the path length from the brachia to the ankles calculated from the participant’s body height31). The PWV measured between the brachium and the ankle has been reported to be closely correlated with carotid–femoral PWV, a standard measure of arterial stiffness32, 33). Basic plasma markers were measured at commercial laboratories (SRL Inc., Tokyo, Japan, or Medic Inc., Shiga, Japan) using blood specimens drawn at the baseline survey. Data on smoking habits, histories of CVD and any cancer, and medication use were obtained using a structured questionnaire.
Statistical AnalysisValues are shown as mean±standard deviation or frequency. Group differences in continuous variables were assessed using the analysis of variance, and frequency differences were assessed using the chi-square test. CVD incidence was calculated per 10,000 person-years. The survival curve was produced using the Kaplan–Meier method, and survival curve differences between the C. pneumoniae-seropositive and -seronegative groups were assessed using a log-rank test. The Cox proportional hazard model analysis was used to clarify whether the association between C. pneumoniae seropositivity and CVD incidence was independent of age (years), sex, and established conventional risk factors for CVDs, including current smoking, body mass index, systolic blood pressure, plasma glycemic (hemoglobin A1c), lipids (high-density lipoprotein cholesterol and low-density lipoprotein cholesterol), inflammatory markers (C-reactive protein), and renal function (estimated glomerular filtration rate). We also adjusted for arterial stiffness (brachial-to-ankle PWV), which underlies the relationship between conventional risk factors and CVDs34). The Cox proportional hazard model analysis was performed with stratification at age 55, the study population’s median age. Several ages were employed as stratification points to determine the threshold age of stratification that affected the findings. We also performed Fine-Gray’s competing risk model analysis for CVDs with non-cardiovascular mortality as a competing risk factor. In this model, the same covariates included in the Cox hazard model analysis were adjusted for.
All statistical analyses were performed using statistical software (JMP® Pro 17.2.0, SAS Institute, Cary, NC, or STATA 19.5, StataCorp, College Station, TX). A p-value of less than 0.05 was considered statistically significant.
Table 1 summarizes the clinical characteristics of the study participants. During a mean follow-up duration of 4,390 days, 323 cases of CVDs were observed. The incidence of CVDs, as well as that of stroke and coronary artery diseases, was approximately twice as high in the seropositive group as in the seronegative group (Table 2). The Kaplan–Meier curves for the incidence of CVDs (Fig.1), stroke (Supplementary Fig.1), and coronary artery diseases (Supplementary Fig.2) exhibited a higher event rate in the seropositive group than in the seronegative group. Table 3 summarizes the results of the Cox proportional hazard model analysis for CVDs incidence. Clinical characteristics at the baseline survey differed considerably between the seropositive and seronegative groups (Supplementary Table 1); the participants in the seropositive group were older, more likely to be male and smokers, and had slightly poorer clinical characteristics. However, C. pneumoniae seropositivity was significantly associated with CVDs after adjusting for conventional risk factors. These associations remained statistically significant when PWV was excluded from the model (total population: hazard ratio (HR) = 1.29, p = 0.029; younger subpopulation: HR = 2.63, p = 0.001), indicating that the association between C. pneumoniae seropositivity and CVDs was not biased by potential collinearity between PWV and conventional risk factors. When participants with a history of cancer were excluded from the analysis, C. pneumoniae seropositivity did not attain statistical significance in the total population (HR = 1.23, p = 0.089), whereas it remained statistically significant in the younger subpopulation (HR = 2.77, p = 0.001), suggesting that the association between C. pneumoniae and CVDs may not have cancer as a confounding factor. Similar trends were observed in the analysis for stroke and coronary artery diseases (Table 3), though the p-values were below the threshold for statistical significance. Supplementary Tables 2, 3, and 4 present the complete Cox proportional hazard model analysis results for CVDs, stroke, and coronary artery diseases, respectively.
| Age, years | 52.8±13.2 |
| Sex, men % | 31.6 |
| Body mass index, kg/m2 | 22.2±3.3 |
| Smoking, current/past/never, n | 1,323/1,767/5,974 |
| History of any cancer, % | 3.1 |
| Systolic BP, mmHg | 123±17 |
| Diastolic BP, mmHg | 76±11 |
| Antihypertensive medication, % | 14.9 |
| Glucose, mg/dL | 90±12 |
| Hemoglobin A1c, % | 5.4±0.5 |
| Antihyperglycemic medication, % | 2.4 |
| HDL cholesterol, mg/dL | 66±17 |
| LDL cholesterol, mg/dL | 123±31 |
| Lipid-lowering medication, % | 10.9 |
| Creatinine, mg/dL | 0.7±0.2 |
| eGFR, mL/min/1.73 m2 | 79.8±15.5 |
| C-reactive protein, μg/mL | 0.83±2.92 |
| Pulse wave velocity, cm/sec | 1,240±191 |
| Chlamydia pneumoniae IgG index | 1.17±0.74 |
| Chlamydia pneumoniae IgA index | 0.98±0.64 |
| Mycoplasma pneumoniae§, % | |
| ＜1:40 | 79.0 |
| 1:40 | 11.8 |
| 1:80 | 6.1 |
| ≥ 1:160 | 3.1 |
Values are presented as the mean±standard deviation or frequencies. BP: blood pressure, HDL: high-density lipoprotein, LDL: low-density lipoprotein, eGFR: estimated glomerular filtration rate. §Values are available for 9,063 participants.
| Total population | ≤ 55 years of age | ＞55 years of age | ||||
|---|---|---|---|---|---|---|
| Seropositive (2,298) | Seronegative (6,766) | Seropositive (983) | Seronegative (3,456) | Seropositive (1,315) | Seronegative (3,310) | |
| Cardiovascular disease | ||||||
| Number of events, n | 123 | 200 | 21 | 24 | 102 | 176 |
| Follow-up duration, person-years | 27,353 | 81,670 | 11,802 | 41,637 | 15,551 | 40,033 |
| Incidence rate | 45.0 | 24.5 | 17.8 | 5.8 | 65.6 | 44.0 |
| Stroke | ||||||
| Number of events, n | 56 | 101 | 15 | 16 | 41 | 85 |
| Follow-up duration, person-years | 27,674 | 82,206 | 11,830 | 41,676 | 15,844 | 40,530 |
| Incidence rate | 20.2 | 12.3 | 12.7 | 3.8 | 25.9 | 21.0 |
| Coronary artery diseases | ||||||
| Number of events, n | 69 | 105 | 6 | 8 | 63 | 97 |
| Follow-up duration, person-years | 27,603 | 82,153 | 11,867 | 4,1721 | 15,736 | 40,432 |
| Incidence rate | 25.0 | 12.8 | 5.1 | 1.9 | 40.0 | 24.0 |
The incidence rate was calculated as per 10,000 person-years. Cardiovascular diseases include stroke and coronary artery disease.
Cardiovascular diseases include stroke and coronary artery disease. Chlamydia pneumoniae seropositivity was defined as IgA and IgG index values of ＞1.1. Group differences were assessed by the log-rank test.
Seropositivity of Chlamydia pneumoniae was defined as IgA and IgG index values of both more than 1.1. Group differences were assessed by log-rank test.
Seropositivity of Chlamydia pneumoniae was defined as IgA and IgG index values of both more than 1.1. Group differences were assessed by log-rank test.
| Total population | ≤ 55 years of age | ＞55 years of age | ||||
|---|---|---|---|---|---|---|
| HR (95% CI) | P | HR (95% CI) | p | HR (95% CI) | p | |
| Cardiovascular diseases | 1.30 (1.04−1.64) | 0.023 | 2.63 (1.45−4.75) | 0.001 | 1.15 (0.90−1.47) | 0.269 |
| Stroke | 1.29 (0.92−1.79) | 0.137 | 2.70 (1.32−5.52) | 0.006 | 1.03 (0.71−1.51) | 0.862 |
| Coronary artery diseases | 1.26 (0.92−1.71) | 0.146 | 2.31 (0.77−6.93) | 0.136 | 1.21 (0.88−1.67) | 0.248 |
Values are hazard ratio (HR) and 95% confidence interval (CI). Adjusted factors include age, sex, current smoking, body mass index, systolic blood pressure, hemoglobin A1c, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, estimated glomerular filtration rate, C-reactive protein, and brachial-to-ankle pulse wave velocity.
| Chlamydia pneumoniae | p | ||
|---|---|---|---|
| Seropositive 2,298 | Seronegative 6,766 | ||
| Age, years | 55.0±13.0 | 52.1±13.2 | ＜0.001 |
| Sex, men % | 36.8 | 29.9 | ＜0.001 |
| Body mass index, kg/m2 | 22.4±3.3 | 22.2±3.2 | 0.002 |
| Smoking, current/past/never % | 17.2/23.2/59.5 | 13.6/18.3/68.1 | ＜0.001 |
| Systolic BP, mmHg | 125±18 | 123±17 | ＜0.001 |
| Diastolic BP, mmHg | 77±12 | 75±11 | ＜0.001 |
| Antihypertensive medication, % | 17.1 | 14.1 | 0.001 |
| Glucose, mg/dL | 91±13 | 89±12 | ＜0.001 |
| Hemoglobin A1c, % | 5.5±0.5 | 5.4±0.4 | ＜0.001 |
| Antihyperglycemic medication, % | 3.0 | 2.1 | 0.017 |
| HDL cholesterol, mg/dL | 64±17 | 66±17 | ＜0.001 |
| LDL cholesterol, mg/dL | 123±31 | 123±31 | 0.795 |
| Lipid-lowering medication, % | 10.4 | 11.1 | 0.343 |
| Creatinine, mg/dL | 0.7±0.2 | 0.7±0.1 | ＜0.001 |
| eGFR, ml/min/1.73 m2 | 78.8±15.3 | 80.2±15.6 | ＜0.001 |
| C-reactive protein, μg/mL | 0.93±3.18 | 0.80±2.83 | ＜0.001 |
| Pulse wave velocity, cm/sec | 1,266±195 | 1,231±189 | ＜0.001 |
Values are mean±standard deviation or frequency.
Seropositivity of Chlamydia pneumoniae was defined as IgA and IgG index values of both more than 1.1. Statistical significance was assessed by analysis of variance or chi-squared test. Pulse wave velocity was assessed between brachia and ankle.
BP: blood pressure, HDL: high-density lipoprotein, LDL: low-density lipoprotein, eGFR: estimated glomerular filtration rate.
| Total population | ≤ 55 years of age | >55 years of age | ||||
|---|---|---|---|---|---|---|
| HR (95% CI) | p | HR (95% CI) | p | HR (95% CI) | p | |
| Age, 10 years | 1.77 (1.52−2.07) | ＜0.001 | 1.87 (1.14−3.06) | 0.013 | 2.03 (1.57−2.64) | ＜0.001 |
| Sex, men | 1.85 (1.44−2.38) | ＜0.001 | 1.40 (0.69−2.84) | 0.351 | 1.95 (1.49−2.56) | ＜0.001 |
| Current smoking | 1.55 (1.15−2.08) | 0.004 | 1.54 (0.75−3.15) | 0.236 | 1.51 (1.09−2.09) | 0.014 |
| Body mass index, kg/m2 | 1.00 (0.96−1.04) | 0.967 | 1.01 (0.92−1.10) | 0.906 | 1.00 (0.96−1.04) | 0.956 |
| Systolic blood pressure, 10mmHg | 1.13 (1.05−1.22) | 0.001 | 1.21 (0.99−1.49) | 0.068 | 1.12 (1.04−1.22) | 0.004 |
| Hemoglobin A1c, % | 1.24 (1.03−1.48) | 0.020 | 0.48 (0.20−1.15) | 0.099 | 1.34 (1.12−1.62) | 0.002 |
| HDL cholesterol, 5 mg/dL | 0.95 (0.91−0.98) | 0.005 | 0.90 (0.81−1.00) | 0.045 | 0.95 (0.92−0.99) | 0.026 |
| LDL cholesterol, 10 mg/dL | 1.00 (0.97−1.04) | 0.920 | 0.96 (0.87−1.06) | 0.462 | 1.01 (0.97−1.05) | 0.584 |
| eGFR, 5 ml/min/1.73 m2 | 0.98 (0.94−1.03) | 0.482 | 1.02 (0.92−1.14) | 0.663 | 0.98 (0.93−1.02) | 0.340 |
| C-reactive protein, 0.1 mg/dL | 1.00 (1.00−1.00) | 0.535 | 0.98 (0.94−1.02) | 0.351 | 1.00 (1.00−1.00) | 0.373 |
| Pulse wave velocity, m/sec | 1.16 (1.07−1.26) | ＜0.001 | 1.48 (1.15−1.89) | 0.002 | 1.13 (1.03−1.22) | 0.006 |
| Chlamydia pneumoniae seropositive | 1.30 (1.04−1.64) | 0.023 | 2.63 (1.45−4.75) | 0.001 | 1.15 (0.90−1.47) | 0.269 |
Values are hazard ratio (HR) and 95% confidence interval (CI). Pulse wave velocity was assessed between brachia and ankle. Seropositivity of Chlamydia pneumoniae was defined as IgA and IgG index values of both more than 1.1.
HDL: high-density lipoprotein, LDL: low-density lipoprotein, eGFR: estimated glomerular filtration rate.
| Total population | ≤ 55 years of age | >55 years of age | ||||
|---|---|---|---|---|---|---|
| HR (95% CI) | p | HR (95% CI) | p | HR (95% CI) | p | |
| Age, 10 years | 1.55 (1.26−1.91) | ＜0.001 | 1.91 (1.06−3.46) | 0.032 | 1.85 (1.26−2.72) | 0.002 |
| Sex, men | 1.21 (0.84−1.74) | 0.298 | 0.80 (0.33−1.95) | 0.630 | 1.39 (0.93−2.07) | 0.110 |
| Current smoking | 1.66 (1.07−2.57) | 0.023 | 1.93 (0.80−4.65) | 0.143 | 1.53 (0.92−2.56) | 0.104 |
| Body mass index, kg/m2 | 1.01 (0.96−1.07) | 0.615 | 1.04 (0.93−1.15) | 0.491 | 1.01 (0.95−1.08) | 0.701 |
| Systolic blood pressure, 10mmHg | 1.12 (1.00−1.24) | 0.042 | 1.10 (0.84−1.44) | 0.479 | 1.12 (0.99−1.26) | 0.068 |
| Hemoglobin A1c, % | 0.80 (0.57−1.12) | 0.198 | 0.36 (0.12−1.11) | 0.075 | 0.92 (0.65−1.31) | 0.644 |
| HDL cholesterol, 5 mg/dL | 0.98 (0.93−1.03) | 0.389 | 0.90 (0.80−1.02) | 0.096 | 0.99 (0.93−1.05) | 0.758 |
| LDL cholesterol, 10 mg/dL | 0.97 (0.92−1.03) | 0.305 | 0.92 (0.81−1.04) | 0.174 | 0.99 (0.93−1.06) | 0.831 |
| eGFR, 5 ml/min/1.73 m2 | 0.98 (0.92−1.04) | 0.513 | 1.04 (0.92−1.17) | 0.537 | 0.96 (0.90−1.03) | 0.279 |
| C-reactive protein, 0.1 mg/dL | 1.00 (1.00−1.00) | 0.132 | 0.91 (0.81−1.01) | 0.086 | 1.00 (1.00−1.01) | 0.050 |
| Pulse wave velocity, m/sec | 1.18 (1.05−1.33) | 0.004 | 1.62 (1.19−2.19) | 0.002 | 1.12 (0.99−1.27) | 0.068 |
| Chlamydia pneumoniae seropositive | 1.29 (0.92−1.79) | 0.137 | 2.70 (1.32−5.52) | 0.006 | 1.03 (0.71−1.51) | 0.862 |
Values are hazard ratio (HR) and 95% confidence interval (CI). Pulse wave velocity was assessed between brachia and ankle. Seropositivity of Chlamydia pneumoniae was defined as IgA and IgG index values of both more than 1.1.
HDL: high-density lipoprotein, LDL: low-density lipoprotein, eGFR: estimated glomerular filtration rate.
| Total population | ≤ 55 years of age | >55 years of age | ||||
|---|---|---|---|---|---|---|
| HR (95% CI) | p | HR (95% CI) | p | HR (95% CI) | p | |
| Age, 10 years | 2.23 (1.76−2.83) | ＜0.001 | 1.93 (0.77−4.83) | 0.162 | 2.35 (1.66−3.33) | ＜0.001 |
| Sex, men | 2.63 (1.85−3.75) | ＜0.001 | 5.47 (1.30−22.97) | 0.020 | 2.47 (1.71−3.56) | ＜0.001 |
| Current smoking | 1.51 (1.03−2.22) | 0.037 | 0.82 (0.23−2.87) | 0.756 | 1.54 (1.02−2.32) | 0.041 |
| Body mass index, kg/m2 | 1.00 (0.95−1.05) | 0.971 | 0.98 (0.82−1.17) | 0.808 | 1.00 (0.94−1.06) | 0.960 |
| Systolic blood pressure, 10mmHg | 1.14 (1.03−1.26) | 0.009 | 1.34 (0.96−1.86) | 0.085 | 1.11 (1.00−1.24) | 0.042 |
| Hemoglobin A1c, % | 1.55 (1.26−1.90) | ＜0.001 | 1.03 (0.36−2.92) | 0.959 | 1.60 (1.30−1.97) | ＜.0001 |
| HDL cholesterol, 5 mg/dL | 0.93 (0.88−0.98) | 0.008 | 0.85 (0.68−1.07) | 0.166 | 0.93 (0.88−0.99) | 0.019 |
| LDL cholesterol, 10 mg/dL | 1.03 (0.98−1.08) | 0.237 | 1.10 (0.93−1.31) | 0.258 | 1.02 (0.97−1.08) | 0.417 |
| eGFR, 5 ml/min/1.73 m2 | 0.99 (0.93−1.05) | 0.751 | 1.00 (0.80−1.25) | 0.992 | 0.99 (0.93−1.05) | 0.818 |
| C-reactive protein, 0.1 mg/dL | 1.00 (1.00−1.00) | 0.909 | 1.00 (0.98−1.03) | 0.832 | 1.00 (1.00−1.00) | 0.917 |
| Pulse wave velocity, m/sec | 1.15 (1.03−1.28) | 0.014 | 1.27 (0.80−2.01) | 0.313 | 1.13 (1.01−1.27) | 0.029 |
| Chlamydia pneumoniae seropositive | 1.26 (0.92−1.71) | 0.146 | 2.31 (0.77−6.93) | 0.136 | 1.21 (0.88−1.67) | 0.248 |
Values are hazard ratio (HR) and 95% confidence interval (CI). Pulse wave velocity was assessed between brachia and ankle. Seropositivity of Chlamydia pneumoniae was defined as IgA and IgG index values of both more than 1.1.
HDL: high-density lipoprotein, LDL: low-density lipoprotein, eGFR: estimated glomerular filtration rate.
When the study population was subdivided by the median age (55 years) (Supplementary Table 5), the HR of C. pneumoniae seropositivity for CVDs was approximately twice as high in the younger subgroup (Table 3) than in the older subgroup, irrespective of the lower incidence rate (Table 2). Fig.2 shows the HR for CVD stratified by the various ages, indicating that the HR of C. pneumoniae infection was inversely correlated with age. A similar trend was observed in the analysis in which non-cardiovascular death was considered a competing risk factor (total population: HR = 1.30, p = 0.026, aged ≤ 45 years: HR = 3.61, p = 0.011, aged ≤ 50 years: HR = 2.93, p = 0.007, aged ≤ 55 years: HR = 2.60, p = 0.003, aged ≤ 65 years: HR = 1.45, p = 0.040). When participants were stratified by 10-year age categories, the adjusted HR of C. pneumoniae seropositivity was significant only in the population aged below 45 years (Supplementary Fig.3).
| Chlamydia pneumoniae | p | ||
|---|---|---|---|
| Seropositive 983 | Seronegative 3,456 | ||
| Age, years | 41.8±7.4 | 40.8±7.5 | ＜0.001 |
| Sex, men % | 25.8 | 28.0 | 0.185 |
| Body mass index, kg/m2 | 21.9±3.6 | 21.7±3.4 | 0.291 |
| Smoking, current/past/never % | 20.4/18.2/61.4 | 17.8/18.2/64.0 | 0.182 |
| Systolic BP, mmHg | 117±16 | 115±15 | 0.003 |
| Diastolic BP, mmHg | 74±12 | 73±11 | 0.001 |
| Antihypertensive medication, % | 3.3 | 3.0 | 0.728 |
| Glucose, mg/dL | 87±11 | 86±9 | 0.274 |
| Hemoglobin A1c, % | 5.3±0.4 | 5.3±0.4 | 0.873 |
| Antihyperglycemic medication, % | 0.5 | 0.3 | 0.380 |
| HDL cholesterol, mg/dL | 67±17 | 67±17 | 0.931 |
| LDL cholesterol, mg/dL | 117±31 | 117±31 | 0.664 |
| Lipid-lowering medication, % | 3.0 | 3.1 | 0.780 |
| Creatinine, mg/dL | 0.7±0.1 | 0.7±0.1 | 0.474 |
| eGFR, ml/min/1.73 m2 | 86.0±14.9 | 86.7±14.9 | 0.209 |
| C-reactive protein, μg/mL | 0.68±2.08 | 0.68±2.32 | 0.930 |
| Pulse wave velocity, cm/sec | 1,128±127 | 1,116±123 | 0.007 |
Values are mean±standard deviation or frequency.
Seropositivity of Chlamydia pneumoniae was defined as IgA and IgG index values of both more than 1.1. Statistical significance was assessed by analysis of variance or chi-squared test. Pulse wave velocity was assessed between brachia and ankle.
BP: blood pressure, HDL: high-density lipoprotein, LDL: low-density lipoprotein, eGFR: estimated glomerular filtration rate.
The hazard ratio was calculated in the total study population and in the subpopulations excluding older adults in a stepwise manner. The factors adjusted for were age, sex, current smoking, body mass index, systolic blood pressure, hemoglobin A1c, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, estimated glomerular filtration rate, C-reactive protein, and brachial-to-ankle pulse wave velocity. The table summarizes the number of participants (N), follow-up duration (person-years), number of incident cases of cardiovascular diseases, and incidence rate (per 10,000 person-years) by seropositivity (P: seropositive, N: seronegative) for each group. Cardiovascular diseases include stroke and coronary artery disease. Chlamydia pneumoniae seropositivity was defined as IgA and IgG index values of ＞1.1.
Adjusted factors were age, sex, current smoking, body mass index, systolic blood pressure, hemoglobin A1c, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, estimated glomerular filtration rate, C-reactive protein, and brachial to ankle pulse wave velocity. The table summarizes the number of participants (N), follow-up duration (person-years), number of incident cases of cardiovascular diseases, and incidence rate (per 10,000 person-years) by seropositivity (P: seropositive, N: seronegative) for each group. Cardiovascular diseases include stroke and coronary artery diseases. Seropositivity of Chlamydia pneumoniae was defined as IgA and IgG index values of more than 1.1.
In contrast, M. pneumoniae seropositivity was not significantly associated with CVD incidence in the Cox hazard model analysis adjusted for the same covariates (HR = 1.13, p = 0.453), even in the analysis of the younger subpopulation (HR = 0.81, p = 0.582).
This longitudinal study of a large general population showed that C. pneumoniae seropositivity was significantly associated with CVD incidence independently of established risk factors. To the best of our knowledge, this study was the first to investigate the possible association in a prospective study setting, including more than 9,000 individuals. The HR for CVDs was inversely correlated with age, identifying C. pneumoniae infection as a possible risk factor for CVDs, especially in adolescents and middle-aged individuals.
A previous prospective longitudinal study, which included 1,187 participants in the Framingham Heart Study, reported that the seropositivity of C. pneumoniae IgG and IgA was not associated with CVD incidence during the 10-year follow-up period18). However, the mean age of this study population (69 years) exceeded that of our study population. Given the age-related changes in the HR, the CVD risk of C. pneumoniae infection remained unidentified in this elderly population with a small number of participants. Indeed, another prospective study of 1,773 men aged 45–59 years, a population younger than that of the Framingham Heart Study, observed statistically significant associations between C. pneumoniae IgA seropositivity and incident ischemic heart disease and mortality17). Unfortunately, other studies that investigated C. pneumoniae infection prognosis were conducted in a case–control setting35); therefore, further investigations are needed to conclude the epidemiological significance of the age−C. pneumoniae infection interaction.
The reason for the high HR of C. pneumoniae infection in the younger group remains uncertain. In an animal study in which mice were intranasally infected with C. pneumoniae, the spread of this pathogen to the heart and ascending aorta—as assessed by immunohistochemistry and C. pneumoniae titer—was more pronounced in older mice than in younger ones. Thus, C. pneumoniae infection may be more severe in older animals21), implying that C. pneumoniae infection increases the risk of CVDs in elderly individuals, which was contrary to our findings. The full results of the Cox proportional hazard model analysis for CVDs indicate that, in addition to C. pneumoniae seropositivity, conventional risk factors (including old age, high blood pressure, high hemoglobin A1c levels, and low high-density lipoprotein cholesterol levels) were significant determinants in the entire population. In contrast, only age, PWV, and C. pneumoniae seropositivity were identified in the younger subpopulation. Given that the HR of PWV in the younger subgroup was higher than that in the older subgroup, similar to the C. pneumoniae seropositivity, the risk of C. pneumoniae infection may have been higher relative to other factors in a younger population in which traditional risk factors had not worsened enough to increase the risk of CVDs.
Pathophysiological changes in vasculature—including increases in macrophage adhesion to the endothelium, smooth muscle cell proliferation, and foam cell formation—and consequent atherosclerotic vascular change are postulated mechanisms for the association between C. pneumoniae infection and CVD events6). This study used PWV as an index of arterial stiffness, finding that the prognosis of C. pneumoniae infection was independent of PWV. The adverse effects of C. pneumoniae infection were not solely mediated by changes in arterial characteristics. PWV mainly represents decreased arterial wall distensibility caused by decreased vascular wall elastin fiber content36). Hypertrophy of intima-media thickness or the plaque score at the carotid artery, representing atherosclerotic vascular changes, would be more appropriate as adjusting factors in this analysis; however, we did not perform carotid echography at the baseline survey of the Nagahama study. Nonetheless, significant correlations have been documented between carotid intima-media thickness and PWV measured between the carotid−femoral artery37) or the brachial−ankle artery38). C. pneumoniae infection may also increase CVD risk through pathways independent of atherosclerosis.
The microimmunofluorescence test is currently the objective standard for C. pneumoniae serodiagnosis39). An enzyme-linked immunoassay is another convenient method of serodiagnosis; however, discrepancies in the detection ratio between microimmunofluorescence test and enzyme-linked immunoassay remain to be resolved40). The detection rates of IgG and IgA antibodies to C. pneumoniae by the enzyme-linked immunoassay technique employed in this study were higher than those achieved by the microimmunofluorescence method (sensitivity: 90.4% for IgG and 84.6% for IgA, specificity: 89.9% for IgG and 84.7% for IgA)41). Random errors in the measurement of risk factors will introduce regression dilution; thus, measurement error in C. pneumoniae seropositivity (if any) would not introduce considerable bias in the present findings.
This study has several limitations that warrant mention. First, we could not accurately distinguish between persistent infection and past infection by measuring seropositivity. We diagnosed seropositivity as an IgA and IgG index value of more than the threshold for both. Because IgA levels increase during the acute phase of an infection, our method seems to have excluded cases of past infection. Instead, cases with persistent infection may be underestimated due to the time-dependent decline in IgG levels, indicating that the seropositivity observed in this study may be biased toward cases with evident infection. A guideline from a workshop on the standardization of C. pneumoniae diagnostic methods39) suggested that no valid serological marker specific for chronic C. pneumoniae infection is available; hence, the infection status based on single-titer readings should be carefully interpreted. Although assessment of seropositivity using paired specimens obtained in the follow-up survey of the Nagahama study may provide clues to elucidate this issue. However, the present findings are remarkable as they demonstrate that even a single assessment of C. pneumoniae seropositivity may be a risk factor for future CVD events. Second, we did not consider socioeconomic status because this information was not available for the baseline population of the Nagahama study. Since socioeconomic status may be associated with both C. pneumoniae infection and the development of CVD, studies considering the socioeconomic status would further clarify the importance of C. pneumoniae seropositivity. Third, our study participants were Japanese community residents. Considering possible regional differences in infection rates of atypical pathogens, including C. pneumoniae42), further studies in different populations are warranted.
In summary, the seropositivity of C. pneumoniae increased the risk for incidence of CVDs in the general Japanese population. This study’s findings reaffirm the CVD risk of C. pneumoniae infection and will encourage further studies.
We are extremely grateful to the Nagahama City Office and the nonprofit organization Zeroji Club for their help in performing the Nagahama study. We would also thank the editors of Crimson Interactive Pvt. Ltd. for their help in the preparation of this manuscript.
Author Takeo Nakayama has received honoraria from Ohtsuka Pharmaceutical Co., ONO PHARMACEUTICAL CO., and AbbVie Inc. All other authors declare no conflicts of interest related to this study.
This study was supported by a university grant, the Center of Innovation Program, the Global University Project, and the Grant-in-Aid for Scientific Research (25293141, 26670313, 26293198, 17H04182, 17H04126, and 21H04850) from the Ministry of Education, Culture, Sports, Science and Technology of Japan; the Practical Research Project for Rare/Intractable Diseases (ek0109070, ek0109283, ek0109196, and ek0109348), the Comprehensive Research on Aging and Health Science Research Grants for Dementia R&D (dk0207006 and dk0207027), the Program for an Integrated Database of Clinical and Genomic Information (kk0205008), the Practical Research Project for Lifestyle-related Diseases including Cardiovascular Diseases and Diabetes Mellitus (ek0210066, ek0210096, and ek0210116), the Research Program for Health Behavior Modification by Utilizing IoT (le0110005), and the Research and Development Grants for Longevity Science (dk0110040) from the Japan Agency for Medical Research and Development (AMED); Welfare Sciences Research Grants, Research on Region Medical from the Ministry of Health, Labour and Welfare of Japan; and the Takeda Medical Research Foundation, Daiwa Securities Health Foundation, Sumitomo Foundation, and the Mitsubishi Foundation.
