ince Fabricant et al first reported avian herpes virus in the arterial walls of chickens,1 there has been the concept that pathogens could contribute to athero- sclerosis. Multiple studies, both epidemiologic and clinical research, have demonstrated associations of infection with coronary artery disease (CAD),2–4and the suggested mecha- nisms are endothelial dysfunction secondary to endotoxins released from the pathogen5and disturbance in blood lipid levels.6,7Much of the recent interest has focused specifically on the role of pathogen infection in hypertension (HT).8–13
HT is considered to be an inflammatory process and is associated with increased expression of systemic markers of inflammation, such as C-reactive protein (CRP) and fibrinogen.14–17Pathogens that have been implicated include Chlamydia pneumoniae (C. pneumoniae),11 Helicobacter pylori (H. pylori)8–10 and herpes simplex virus (HSV).13 However, a recent study did not support the hypothesis linking specific pathogens to HT,18and more recently, the hypothesis has broadened to include the infectious pathogen burden; that is, the notion that is not a single pathogen but the aggregate pathogen load that increases the risk of HT.
Pathogen burden has been shown to be associated with
Circulation Journal Vol.71, November 2007
atherosclerosis and long-term prognosis in patients with documented CAD,19–21but to our knowledge, whether the pathogen burden is associated with HT has not yet been reported. Therefore, in the present study, we examined the hypotheses that the risk of HT is associated with: (1) each spe- cific pathogen, defined as seroprevalence of IgG antibodies for easily acquired bacterial and viral infections, such as C. pneumoniae, Mycoplasma pneumoniae(M. pneumoniae), H. pyloriand Coxsackie virus, or (2) the aggregate number of pathogens (pathogen burden).
Methods
Study Subjects
We designed a cross-sectional survey on essential HT to be performed through July to September in 2003 in a rural population of 14 administrative villages in Naiman Banner, Inner Mongolia Autonomous Region, P.R. China, which is the common area in Inner Mongolia Autonomous Region.
All participants were Chinese Mongolians living as farmers and herders, which differs from other Chinese people’s life- styles. The demographic data of participants were collected by a self-designed questionnaire. People aged 30 and older were eligible, and only those who signed informed consent were recruited. Informed consent was in accordance with the Soochow University ethics committee guidelines. HT was defined as systolic blood pressure ≥140 and/or dias- tolic blood pressure ≥90 mmHg according to the WHO/ISH criteria of 1999. Exclusion criteria were known or suspected liver or kidney disease, or connective tissue or autoimmune disease. All study subjects gave a 5-ml sample of fasting venous blood, from which the serum was isolated and frozen at –80°C until analysis.
Circ J2007; 71:1761 – 1764
(Received April 26, 2007; revised manuscript received June 20, 2007;
accepted July 10, 2007)
*State Key Laboratory of Pathogen and Biosecurity, Beijing Institute of Microbiology and Epidemiology, Beijing, **Department of Epidem- iology, Public Health College, Harbin Medical University, Harbin,
†Epidemiology Department, School of Radiation Medicine and Public Health, Soochow University, Suzhou and ††Epidemiology Department, Center for Disease Prevention and Control of Naiman Banner, Inner Mongolia, People’s Republic of China
Mailing address: Yonghong Zhang, MD, PhD, Epidemiology De- partment, School of Radiation Medicine and Public Health, Soochow University, Suzhou 215021, PROC. E-mail: zhangyh388@sohu.com
Pathogen Burden in Essential Hypertension
Lijuan Liu, MD, PhD*,**; Yanbin Liu, MSc†; Weijun Tong, MD†; Hong Ye, MD, PhD**; Xianyu Zhang, MBc††;
Wuchun Cao, MD, PhD*; Yonghong Zhang, MD, PhD†
Background Associations between pathogens and hypertension (HT) has been reported, but few studies have focused on the relationship between aggregate pathogens and HT. The present study explored whether the risk of HT is associated with each pathogen (defined as Chlamydia pneumoniae (C. pneumoniaee), Mycoplasma pneumoniae(M. pneumoniae), Helicobacter pylori(H. pylori) and Coxsackie virus) or with aggregate pathogens in Chinese Mongolians.
Methods and Results One thousand and thirty Chinese Mongolians aged 30 years or more were recruited, including 488 hypertensive and 942 normotensive subjects. Enzyme-linked immunosorbent assay was used to detect IgG antibodies for C. pneumoniaee, M. pneumoniae, H. pyloriand Coxsackie virus. The seroprevalence of Coxsackie virus was significantly associated with HT (odds ratio (OR) 3.7 after adjustment for risk factors), but no significant association was found for C. pneumoniae, M. pneumoniaeand H. pylori(OR 1.32, 0.75 and 1.19, respectively). The results also showed that the risk of HT was associated with the aggregate pathogens: it increased with the increasing number of pathogens, and the ORs were 1.629, 2.653, 2.129, and 5.146 for 1, 2, 3 and 4 pathogens, respectively, after controlling for risk factors.
Conclusion The risk of HT is associated with Coxsackie virus and aggregate pathogen load. The mecha- nism(s) underlying the associations remain to be elucidated further. (Circ J2007; 71:1761 – 1764)
Key Words: Aggregate pathogens; Coxsackie virus; Hypertension; Risk factor
S
1762 LIU L et al.
Circulation Journal Vol.71, November 2007 Risk Factors for HT
The following potential risk factors for HT were ana- lyzed: age, male gender, alcohol intake, cigarette smoking, body mass index (calculated as weight in kilograms divided by square of height in meters), waist – hip ratio (calculated as waist circumference divided by hip circumference), and blood lipid and CRP levels. Drinking status was classified as drunk usually or seldom, smoking status was classified as current smoker or never smoker. Dyslipidemia was defined as total cholesterol value ≥220 mg/dl (5.72 mmol/L) and/or triglycerides ≥200 mg/dl (2.26 mmol/L) and/or high-density lipoprotein-cholesterol <40 mg/dl (1.04 mmol/L) and/or low- density lipoprotein-cholesterol ≥140 mg/dl (3.64 mmol/L).
Serum CRP concentrations were measured by fluorescence polarization immunoassay.
Measurement of Specific IgG Antibodies for Pathogens Each blood sample was labeled only with a serial num- ber, so that the examiners were unaware of the subject’s data at the time of testing, and remained so until statistical analysis of the results. Each serum sample was tested for
specific anti-C. pneumoniae, M. pneumoniae, H. pylori and Coxsackie virus IgG antibodies by enzyme-linked immu- nosorbent assay with a commercially available kit (Nero- Hemin biotechnology Co, Hang Zhou). The presence or absence of antibodies was determined by comparing the absorbency value of the sample to a cutoff value, which was calculated from the ratio of the sample/negative con- trol absorbency value according to the manufacturer’s pro- tocols. Samples with absorbance values/negative control greater than or equal to 2.1 were considered positive for each pathogen, whereas less than 2.1 were negative. IgG seropositivity indicated prior infection of C. pneumoniae, M. pneumoniae, H. pylori and Coxsackie virus.
Statistical Analysis
For skewed distribution, CRP level was transformed logarithmically for the normal distribution. In the univariate analysis, an independent 2-sample t-test was used for the normally distributed continuous variables, andχ2test was used to compare categorical variables, and each pathogen, with HT. In the multivariate analysis, non-conditional logistic regression analysis was used to control the poten- tial confounding factors and the adjusted odds ratio (OR) and 95% confidence interval (CI) expressing the associa- tion of the pathogen or the aggregate number of pathogens with the presence or absence HT were calculated. Values of p<0.05 were considered to indicate statistical significance.
Data were analyzed using SPSS13.0 software (Chicago, IL, USA).
Results
General Characteristics of the Study Subjects
In total, 1,832 inhabitants were eligible for inclusion in the study and of them 1,430 (78%) signed informed con- sent. Among the 1,430 subjects, there were 488 (34.2%) with HT and 942 without. Table 1 summarizes the main risk factors for HT in both groups. Systolic and diastolic BP levels in the hypertensive and normotensive subjects were significant different, reflecting the correct group allocation.
Table 1 General Characteristics of the Hypertensive and Normotensive Subjects
Hypertensive Normotensive p value
N 488 942
Age (years) 51.65±11.73 43.93±11.29 <0.001
Male gender (%) 52.5% 35.9% <0.001
SBP (mmHg) 149.69±23.89 114.70±10.57 <0.001 DBP (mmHg) 96.70±10.35 76.56±6.87 <0.001 BMI (kg/m2) 22.92±3.62 21.61±3.18 <0.001 Waist-hip ratio 0.98±0.06 0.86±0.06 <0.001
Current smoker (%) 48.4% 44.7% 0.103
Drinker (%) 52.7% 30.3% <0.001
Dyslipidemia (%) 30.5% 13.6% <0.001
Log-CRP (mmol/L) 0.81±0.27 0.72±0.23 <0.001 Data are mean±SD or percent.
SBP, systolic blood pressure; DBP, diastolic blood pressure; BMI, body mass index; CRP, C-reactive protein.
Table 3 Seroprevalence for C. Pneumoniae, M. Pneumoniae, H. Pylori and Coxsackie Virus in Hypertensive and Normotensive Subjects After Controlling Risk Factors
Hypertensive Normotensive OR (95%CI) p value adjusted adjusted Seroprevalence of IgG antibodies for C. pneumoniae 16.7% 12.7% 1.32 (0.96, 1.82) 0.088 Seroprevalence of IgG antibodies for M. pneumoniae 4.3% 5.2% 0.75 (0.44, 1.30) 0.312 Seroprevalence of IgG antibodies for H. pylori 38.7% 33.7% 1.19 (0.94, 1.50) 0.157 Seroprevalence of IgG antibodies for Coxsackie virus 76.2% 50.3% 3.17 (2.31, 4.35) 0.001 Risk factors: age, male gender, drinking status, overweight, dyslipidemia and log-CRP level.
Abbreviations see in Tables 1,2.
Table 2 Seroprevalence of C. Pneumoniae, M. Pneumoniae, H. Pylori and Coxsackie Virus in the Hypertensive and Normotensive Subjects
Hypertensive Normotensive OR (95%CI) p value Seroprevalence of IgG antibodies for C. pneumoniae 16.7% 12.7% 1.38 (1.02, 1.88) 0.038 Seroprevalence of IgG antibodies for M. pneumoniae 4.3% 5.2% 0.82 (0.49, 1.31) 0.461 Seroprevalence of IgG antibodies for H. pylori 38.7% 33.7% 1.25 (0.99, 1.57) 0.054 Seroprevalence of IgG antibodies for Coxsackie virus 76.2% 50.3% 3.17 (2.48, 4.0) 0.001 Data are percent. *p<0.05.
C. pneumoniae, Chlamydia pneumoniae; M. pneumoniae, Mycoplasma pneumoniae; H. pylori, Helicobacter pylori; OR, odds ratio;
CI, confidence interval.
Aggregate Pathogen Load in HT 1763
Circulation Journal Vol.71, November 2007
As compared with the normotensive group, participants with HT as a whole had greater age, body mass index, waist – hip ratio and higher CRP level (p<0.05). There were also significant differences in the prevalence of male gender, drinking and dyslipidemia between the groups (p<0.05), but no significant difference in smoking was found between them (p>0.05).
Relationship Between Each Pathogen and HT
Table 2 shows the seroprevalence in the hypertensive and normotensive groups for C. pneumoniae, M. pneumoniae, H. pylori, and Coxsackie virus. There were significant differences between the 2 groups with regard to the sero- prevalences of H. pylori (38.7% vs 33.7%), C. pneumoniae (16.7% vs 12.7%), and Coxsackie virus (76.2% vs 50.3%), but no significant difference was found between the groups forM. pneumoniae (4.3% vs 5.2%).
Table 3 outlines the ORs for HT by seroprevalence IgG to C. pneumoniae, M. pneumoniae, H. pyloriand Coxsackie virus, adjusted for risk factors (age, male gender, drinking status, overweight, dyslipidemia and log-CRP level). The significant association of C. pneumoniae with HT was atten- uated after controlling the risk factors, and there was still no evidence of increased risk of HT with M. pneumoniae or H. pylori. However, the association of Coxsackie virus with HT was retained, even after adjustment by risk factors.
Relationship Between Aggregate Pathogens and HT Table 4 depicts the unadjusted and adjusted ORs for HT by increasing number of pathogens (as defined by seropreva- lence IgG antibodies to C. pneumoniae, M. pneumoniae, H. pylori and Coxsackie virus) compared with none. The ORs of pathogen burden with HT were 1.63, 2.65, 2.13, and 5.15, increased by 1, 2, 3 and 4 pathogens infection, respectively, after adjustment for risk factors (age, male gender, drinking status, overweight, dyslipidemia, and log-CRP level). There was an increased risk of HT with increasing number of pathogens. The p-value for trend was less than 0.001.
Discussion
In the present study, we found that Coxsackie virus in- fection was consistent with HT, as in previous studies.8–11,13 The OR of Coxsackie virus infection with HT was 3.17 after adjustment for age, male gender, drinking status, over- weight, dyslipidemia, and log-CRP level. Coxsackie virus, a member of the enterovirus genus, is the most common cause of human myocarditis. Recently, a significant elevated level of the vasoconstrictor endothelin-1, which is secreted from the endothelium, has been demonstrated in the heart in some models of murine myocarditis.22,23 That increase occurs early after infection and persists until at least 14 days post infection.27Therefore, it has been proposed that Coxsackie virus infection may be related to cardiovascular disease24and HT. With respect to the latter, we also found that the CRP level, a systemic marker of inflammation, was higher in hypertensives than in normotensives, which sug- gested that inflammation is related to HT. Indeed, pathogen infection could lead to a chronic inflammatory response.
Many studies have focused on the relationship between HT and specific pathogen infection. Kristensen et al re- ported that the seroprevalence of HSV was significant higher in patients with WHO stage III HT (39.5%) and those with stage I-II HT (25.9%), than that in normotensives
(9.3%).25Another study also reported serum IgG antibodies for Chlamydia, HSV, cytomegalovirus, Epstein-Barr virus, and showed that each specific pathogen was associated with HT.26Cook et al also revealed the association between C. pneumoniae and severe HT.11 However, most of them just detected the influence of a single pathogen and did not explore the potential aggregation of multiple pathogens.
Their different results provoked our interest in the possi- bility that aggregate pathogen infection contributed to HT.
We thought that if infection did play a part in HT, it was highly unlikely to be derived from a single pathogen’s inflammatory response, and we also believed that aggregate pathogens carried by the individual would contribute to the risk potential of HT. Accordingly, we examined the hypo- thesis that the risk of HT was incrementally related to the pathogens. The present results show that whether adjusted by risk factors or not, the risk of HT is still increased with an increasing number of pathogens (p for trend <0.001).
Study Limitations
Although our results were compatible with the causal role of pathogen burden infection in the development of HT, there are still some limitations.
First, in this cross-sectional study, it is possible that infection played no causal role and these pathogens were associated with HT only because hypertensive persons are predisposed to infection. Second, approximately 20% of the subjects who were eligible for the inclusion in the study did not participate and this would underestimate or overestimate the relationship between pathogen burden and HT (selec- tion bias); however, the study site was a fairly remote district from the city in Inner Mongolia in China, all of the Chinese Mongolian inhabitants knew little about their blood pressure value, and those did not participate in the study would be unlikely to know their blood pressure values well, and this, perhaps, could retrieve the above-mentioned deficiency. Finally, IgG seropositivity/seronegativity may be more appropriate for distinguishing between subjects with frequent, recent, severe, or chronic infection and those with less frequent, less recent, or milder infection.
In summary, our results suggest that the risk of HT is associated with Coxsackie virus infection and aggregated pathogen load, but the hypothesis needs to be further dem- onstrated in a prospective study.
Acknowledgments
We are deeply appreciative of the subjects who participated in the study and of the many field-workers who organized them and collected the data.
This study was supported by the National Nature and Science Founda- tion of the People’s Republic of China (Grant No. 30471484).
Table 4 ORs of Pathogen Burden With Hypertension
No. of OR (95%CI) OR (95%CI) p value
pathogens adjusted adjusted
0 1.00 1.00
1 1.73 (1.23, 2.42) 1.629 (1.159, 2.290) 0.005 2 2.78 (1.91, 4.05) 2.653 (1.819, 3.869) 0.001 3 2.27 (1.27, 4.04) 2.129 (1.189, 3.812) 0.011 4 6.11 (1.03, 36.338) 5.146 (0.857, 30.883) 0.073
Trend test 0.001 0.001
Adjusted for age, male gender, drinking status, overweight, dyslipidemia and log-CRP level.
Abbreviations see in Tables 1,2.
1764 LIU L et al.
Circulation Journal Vol.71, November 2007 Dr Liu is an Associate Professor of Epidemiology at the State Key Labo-
ratory of Pathogen and Biosecurity, Beijing Institute of Microbiology and Epidemiology. Her research interests focus on chronic and infectious dis- eases.
References
1. Fabricant CG, Fabricant J, Litrenta MM, Minick CR. Virus-induced atherosclerosis. J Exp Med1978; 148:335 – 340.
2. Hilgers KF. Monocytes/macrophages in hypertension. J Hypertens 2002; 20:593 – 596.
3. Ross R. Atherosclerosis-a inflammatory disease. N Engl J Med1999;
340:115 – 126.
4. Nakanishi N, Sato M, Shiral K, Suzuki K, Tatara K. White blood cell count as risk factor for hypertension: A study of Japanese male office workers. J Hypertens2002; 20:851 – 857.
5. Patel P, Mendall MA, Carrington D, Strachan DP, Leatham E, Molineaux N, et al. Association of Helicobacter pylori and Chlamydia pneumoniae infections with coronary heart disease and cardiovascu- lar risk factors. BMJ1995; 311:711 – 714.
6. Niemala S, Karttunen T, Korhonen T, Laara E, Karttunen R, Ikaheimo M, et al. Could Helicobacter pylori infection increase the risk of coronary heart disease by modifying serum lipid concentra- tions? Heart1996; 75:573 – 575.
7. Scragg RK, Fraser A, Metcalf PA. Helicobacter pylori seropositivity and cardiovascular risk factors in a multicultural workforce. J Epidemiol Community Health1996; 50:578 – 579.
8. Barnes RJ, Uff JS, Dent JC, Gear MW, Wilkinson SP. Long term follows up of patients with gastritis associated with helicobacter pylori infection. Br J Gen Pract1991; 41:286 – 288.
9. Whincup PH, Mendall MA, Perry IJ, Strachan DP, Walker M. Pro- spective relations between helicobacter pylori infection, coronary heart disease and stroke in middle aged men. Heart1996; 75:568 – 572.
10. Lip GYH, Wise R, Becvers G. Association between H. pylori infec- tion and hypertension. BMJ1996; 312:250 – 251.
11. Cook PJ, Lip GYP, Davies P, Beevers DG, Wise R, Honeybourne D.
Chlamydia pneumoniae antibodies in severe essential hypertension.
Hypertension1998; 31:589 – 594.
12. Liu DX, Peng DX, Chen R, Lei HT, Che DY, Zhao SY. Chronic pulmonary infection caused by Mycoplasma pneumoniae leading to pulmonary arteriole remodeling and pulmonary hypertension in rats.
J Tongji Med Univ1995; 15:223 – 226.
13. Sun Y, Pei W, Wu Y, Jing Z, Zhang J, Wang G. Herpes simplex virus type 2 infection is a risk factor for hypertension. Hypertens Res2004, 27:541 – 544.
14. Conen D, Dieterle T, Utech K, Rimner M, Martina B. C-reactive protein and B-type natriuretic peptides in never-treated white coat hypertensive. Hypertens Res 2006; 29:411 – 415.
15. Sesso HD, Buring JE, Rifai N, Blake GJ, Gaziano JM, Ridker PM.
C-reactive protein and the risk of developing hypertenison. JAMA 2003; 29:2945 – 2951.
16. Schillaci G, Pirro M, Gemelli F, Pasqualini L, Vaudo G, Marchesi S.
Increased C-reactive protein concentrations in never-treated hyper- tension: The role of systolic and pulse pressures. Hypertension2003, 21:1841 – 1846.
17. Bautista LE, Atwood JE, O’Malley PG, Taylor AJ. Association be- tween C-reactive protein and hypertension in healthy middle-aged men and women. Coronary Heart Dis2004, 15:331 – 336.
18. Harvey R, Lane A, Murray L, Harvey I, Nair P, Donovan J. Effect of Helicobacter pylori infection on blood pressure: A community based cross sectional study. Br J Med2001; 323:7307.
19. Espinola-Klein C, Rupprecht HJ, Blankenberg S, Bickel C, Kopp H, Rippin G. Impact of infectious burden on extent and long-term prog- nosis of atherosclerosis. Circulation2002; 105:15 – 21.
20. Zhu J, Nieto FJ, Horne BD, Anderson JL, Muhlestein JB, Epstein SE.
Prospective study of pathogen burden and risk of myocardial infarc- tion or death. Circulation2001; 103:45 – 51.
21. Rupprecht HJ, Blankenberg S, Bickel C, Rippin G, Hafner G, Prellwitz W, et al. Impact of viral and bacterial infectious burden on long-term prognosis in patients with coronary artery disease. Circu- lation2001; 104:25 – 31.
22. Clozel M. Endothelin receptor antagonists: Current status and per- spectives. J Cardiovasc Pharmacol2000; 35:S65 – S68.
23. De Vriese AS, Verbeuren TJ, Van de Voorde J, Lameire NH, Vanhoutte PM. Endothelial dysfunction in diabetes. Br J Pharmacol 2000; 130:963 – 974.
24. Choy JC, Lui AH, Moien-Afshari F, Wei K, Yanagawa B, McManus BM, et al. Coxsackievirus B3 infection compromises endothelial- dependent vasodilation of coronary resistance arteries. J Cardiovasc Pharmacol2004; 43:39 – 47.
25. Kristensen BO, Andersen PL, Vestergaard BF. Herpes virus anti- bodies and vascular complications in essential hypertension. Acta Med Scand1982; 212:375 – 377.
26. Omura Y. Microbial infection or trauma at cardiovascular represen- tation area of medulla oblongata as some of the possible causes of hypertension or hypotension. Acupunct Electrother Res1988; 13:
131 – 145.
27. Ono K, Matsumori A, Shioi T, Furukawa Y, Sasayama S. Contribu- tion of endothelin-1 to myocardial injury in a murine model of myo- carditis: Acute effects of bosentan, an endothelin receptor antagonist.
Circulation1999; 100:1823 – 1829.
