中國醫藥大學機構典藏 China Medical University Repository, Taiwan:Item 310903500/50312

全文

(1)中國醫藥大學臨臨床醫學研究所碩士學位論論文. 肺炎鏈球菌感染增加急性冠心症的風險. Pneumococcal Pneumonia and the Risk of Acute Coronary Syndrome. 指導教授：汪貴珍. 教. 授. 共同指導教授：高嘉鴻. 教. 授. 研究生：王駿丞. 中華民國一 ○ 二年年. 七. 月.

(2) 1.

(3) Abstract Introduction A link between infection and the incidence of acute coronary syndrome (ACS) has been suggested. The reason of infection leading to ACS maybe explained by increased inflammation, hemoconcentration, imbalance between oxygen demand and supply. Of all bacterial infections, pneumonia is mostly associated with the development of ACS.Streptococcus pneumonia is the most common causative pathogen of community acquired pneumonia, and Streptococcus pneumonia infection is asscociated with clinically severe form of pneumonia. However, relatively few studies have specifically discussed the association between pneumococcal pneumonia and ACS. This study focuses on investigating the association between pneumococcal pneumonia and the development of ACS. This is an important research topic, because it may provide a rationale for implementing pneumococcal vaccination in preventing ACS, especially in Asian populations. Materials and methods We conducted a longitudinal cohort study from the Taiwan National Health Institute Research Database (NHIRD). The study sample consisted of 20111 patients who received the first diagnoses of pneumococcal pneumonia between 1997 and 2010. We age and sex-matched these participants with 80444 control patients without a previous diagnosis of either pneumococcal pneumonia or ACS. We first compared the differences of baseline demographics between the two groups with chi-square test.We consider hypertension, 2.

(4) diabetes, hyperlipidemia, and chronic obstructive pulmonary disease as potential covariates. We used the follow-up person-years to assess the incidence density rates until ACS was either identified or censored.We used Poisson regression models to evaluate the ratios of the pneumococcal pneumonia cohort to the controls (relative risk) and 95% confidence intervals (CI). We used the Cox proportional-hazards model to investigate whether pneumococcal pneumonia is independently associated with ACS after adjusting for all potential risk factors. We then used Kaplan-Meier analysis and the log-rank test to compare the cumulative risk of developing ACS between the 2 groups. Results The incidence of ACS was 43.1 per 10000 person-years in the pneumococcal pneumonia and 22.4 per 10000 person-years in the control group.(Incidence rate ratio: 1.92; 95% confidence interval: 1.70-2.17). Higher proportions of hypertension, diabetes, hyperlipidemia and chronic obstructive pulmonary disease are noted in the pneumococcal pneumonia group. After adjusting for age, sex, and comorbidities, we found that the risk of ACS was 47% higher in the pneumococcal pneumonia group than in the control group (95% confidence interval: 1.24-1.73). We divided the time lag into 3 periods, (≤ 3 mo, 3 mo to 1 y and >1 y), and found that the highest relative risk of incidence of ACS between the 2 groups was within the first 3 months after infection with pneumococcal pneumonia (Incidence rate ratio: 3.90; 95% confidence interval: 2.46-6.18). The Kaplan-Meier survival curve showed that the risk of ACS was higher in the pneumococcal pneumonia group than in the control group (Log-rank test, P<0.0001). 3.

(5) Conclusion Pneumococcal pneumonia is associated with an increased incidence of ACS, and the relative risk of incidence of ACS between the two groups is highest within the first three months. Our study implied that pneumococcal vaccination may be considered as an option for prevention of incidence of ACS.. 4.

(6) 中文摘要: 簡介: 臨臨床上感染與急性冠心症之間的關聯聯過去的研究已被提及。感染會造成急性冠心症的病理理機轉包含發炎反應，血液容積濃縮，與氧氣供需的失衡。在所有感染中，肺炎感染與急性冠心症之間最為相關肺炎鏈球菌是社區性肺炎最好發的致病菌，且往往造成臨臨床上表現較嚴重的社區性肺炎。然而,肺炎鏈球菌的感染與急性冠心症的關聯聯則較少被提及。我們的研究主要目的是探討肺炎鏈球菌的感染與急性冠心症的關聯聯。我們的研究可以提供一個想法，就是肺炎鏈球菌疫苗可能減少未來來得到急性冠心症的風險。研究方法: 我們利利用台灣健保資料料庫作一個回溯性世代研究。我們搜索索從民國 86 年年至民國 99 年年共 20111 位病人初次診斷為肺炎鏈球菌感染，且之前未曾患有急性冠心症的病人。對照組則是 80444 位過去未曾診斷有任何形式肺炎或是急性冠心症的病人。肺炎鏈球菌組與對照組的樣本是依 1:4 的比例例且分別依年年齡與性別去配對。兩兩組的基本資料料我們用卡方分析作比較，我們考慮高血壓、糖糖尿尿病、高血脂及慢性呼吸道阻塞塞疾病當作共變項。我們使用 Cox-Proportional Hazard model 來來評估經過多變項分析，肺炎鏈球菌感染是否是得到急性冠心症的一個危險因子。我們用 Kaplan-Meier curve 與 log-rank test 來來探討肺炎鏈球菌組與對照組長期得到急性冠心症的風險是否有所不不同。研究結果: 5.

(7) 在肺炎鏈球菌組中得到急性冠心症的發生率率率是 43.1/10000 人年年，在對照組中得到急性冠心症的發生率率率是 22.4/10000 人年年。(發生率率率比例例為 1.92,95%信賴區間為 1.70-2.17) 經過年年紀、性別與其它共變項的校正後，肺炎鏈球菌組長期得到急性冠心症的風險仍比對照組增加 47%，且其風險的增加呈統計上有意義相關。(95%信賴區間為 1.24-1.73) 我們把臨臨床上從感染肺炎鏈球菌到得到急性冠心症的時間分為三個時段(小於 3 個月, 3 個月至 1 年年,及大於一年年)，我們發現肺炎鏈球菌組在感染的前三個月得到急性冠心症的相對風險與對照組比較起來來增加最多。(發生率率率比例例為 3.9, 95%信賴區間為 2.46-6.18) 我們用 Kaplan-Meier curve 比較兩兩組發生急性冠心症,肺炎鏈球菌組得到急性冠心症的風險仍顯著的比對照組高。（Log-rank test, P<0.0001）結論論: 肺炎鏈球菌感染會增加得到急性冠心症的風險。病人在得到肺炎鏈球菌感染的前三個月得到急性冠心症的相對風險最高。我們的研究也建議未來來發展合適的疫苗可能可以減少得到急性冠心症的風險。. 6.

(8) 誌. 謝. 時光飛逝, 兩兩年年的碩士班學業已結束。回首這兩兩年年,每星期往來來於學校與醫院,實驗室間,常忙得不不可開交。然而, 從師長的手中拿下畢業證書,內心著實有一種苦盡甘來來之感觸。今日能如願畢業, 首先需感謝中國醫藥大學臨臨床醫學研究所汪貴珍教授與醫學系高嘉鴻教授,感謝他們對我論論文的指導。感謝中國醫藥大學公共衛生學系,感謝他們對統計方面的幫忙與分析。另外,感謝台中慈濟醫院人體試驗委員會與研究部,感謝他們對我碩士研究的支持。感謝台中慈濟醫院心臟內科林林茂仁主任對我唸研究所的大力力支持,以及科內同仁幫忙分擔科內臨臨床與行行政事物。也感謝中國醫藥大學藥學系吳介信主任,台中榮民總醫院核醫科林林萬鈺主任,彰賓秀傳核醫科洪光威主任,感謝他們百忙中撥空擔任我的碩士口試委員,並給予寶貴的建議。最後,感謝我的家人能體量量我兼顧學業與工作的辛苦,幫我許多生活上的事,讓我無後顧之憂。王駿丞中華民國. 謹誌. 102 年年 7 月 24 日. 7.

(9) 目. 錄錄. 1. 封面…………………………………………………………………………… 2. 學位考試委員審定書影本……………………………………………………. P1. 3. 英文摘要………………………………………………………………………. P2. 4. 中文摘要………………………………………………………………………. P5. 5. 誌謝……………………………………………………………………………. P7. 6. 論論文正文………………………………………………………………………. P10. 第一章. 前言…………………………………………………………………. P10. 第一節. 研究背景…………………………………………………………………………. P10. 第二節研究目的…………………………………………………………………………. P10. 第二章. 研究方法…………………………………………………………………………… P12. 第一節. 研究材料料…………………………………………………………………………… P12. 第二節. 研究設計…………………………………………………………………………… P12. 第三節. 統計方法…………………………………………………………………………… P14. 第三章. 研究結果…………………………………………………………………………… P15. 第一節. 描述性統計分析 Table 1 文字描述…………………………………………………………………. 第二節. P15. 推論論性統計分析 Table 2 文字描述…………………………………………………………………. P15. Table 3 文字描述…………………………………………………………………. P16. Table 4 文字描述…………………………………………………………………. P16 8.

(10) Figure 1 文字描述…………………………………………………………………. P16. 第四章討論論…………………………………………………………………………………. P17. 第一節結果討論論……………………………………………………………………………. P17. 第二節其他相關性討論論……………………………………………………………………. P19. 第三節研究限制……………………………………………………………………………. P20. 第五章結論論與建議…………………………………………………………………………… P22 第一節結論論…………………………………………………………………………………… P22 第二節建議…………………………………………………………………………………… P22. 7. 參參考文獻………………………………………………………………………… P23 8. 圖表附錄錄………………………………………………………………………… P27 8-1 Table 1……………………………………………………………………………………… P27 8-2 Table 2……………………………………………………………………………………… P28 8-3 Table 3……………………………………………………………………………………… P29 8-4 Table 4……………………………………………………………………………………… P30 8-5 Figure 1……………………………………………………………………………………… P31. 9. 論論文電子檔案上網授權書. 9.

(11) Introduction: Study Background Previous research has proposed an association between acute bacterial infection and an increased incidence of acute coronary syndrome (ACS).1 The reason of infection leading to ACS may be explained by increased inflammation, hemoconcentration, and imbalance betweeb oxygen demand and supply. Of all acute bacterial infections, acute bacterial pneumonia is most widely discussed. Pneumonia is still the leading cause of death in Western countries and the mortality rate of community acquired pneumonia(CAP) remains high at approximately 5%-15%.2 Pneumonia may cause a ventilation-perfusion mismatch, hypoxemia, or hemodynamic compromise, aggravating the imbalance between the oxygen supply and demand of the myocardium. Several studies have proposed an association between CAP and an increased incidence of ACS. The most common pathogen of CAP is Streptococcus pneumonia. Streptococcus pneumonia may cause severe form of CAP and often required hospitalization. Though the association of CAP and ACS has been discussed before, few studies have focused on the specific bacterial agents that cause pneumonia. This is an important research topic because vaccination against specific infectious agents may improve cardiovascular outcomes if the association between acute cardiovascular events and the specific bacterial infection can be established. Study Purpose Limited publications have suggested an association between pneumococcal pneumonia and the incidence of ACS.6,7In addition, the results of pneumococcal vaccination for prevention 10.

(12) of incidence of cardiovascular outcomes published before are controversial. Therefore, we attempted to use large database from the Taiwan National Health Institute Research Database (NHIRD) to investigate if pneumococcal pneumonia is associated with risk of ACS, especially in the Taiwan population.. 11.

(13) Materials and Methods Data source We established a longitudinal cohort study based on data from the Taiwan National Health Institute Research Database (NHIRD), which is the largest and most comprehensive population-based medical benefit claims. Taiwan launched a single-payer compulsory national health insurance (NHI) program in the beginning of 1995. This program covered over 99% of Taiwan’s entire 23 million residents.8,9 All of the NHI data sets can be interlinked throughde-identifications of people, making the NHI reimbursement data suitable for public academicresearch. The information in the NHIRD, such aspatient identification number,sex, birthday, and discharge date,. have been thoroughly described. in various studies.10,11The accuracy and validity of diagnosis identified in the NHIRD has been strictly implemented and certificated.12,13This study was exempted by the Institutional Review Board of China Medical University (CMU-REC-101-012). Study design Patients with pneumococcal pneumonia. The International Classification of Disease, Ninth Revision (ICD-9) was used for the diagnosis. We selected all adult pneumococcal pneumonia (ICD-9 code 481) patients (≥20 years old), identifying 20111patients with a first-time diagnosis of pneumococcal pneumonia and with at least one health care admission from 1997 to 2010. The index date was the date of pneumococcal pneumonia registration. We excluded patients with a history of ACS before the index date, and those with incomplete age or sex information. We then 12.

(14) randomly selected patients without a history of pneumonia (ICD-9 codes 480-487) or ACS in the NHIRD. To increase the statistical power, we created four comparison controls(80444) for each case, and frequency- matched the patients based on age 5 years each, sex, and index year of enrollment. We followed all study patients were until the date at which they were diagnosed with ACS. Identification of acute coronary syndrome cases. We linked study patients to the admission claims datato identify the first episode of ACS (ICD-9-CM codes410, 411.1) after pneumococcal pneumonia. Thepatients newly diagnosed with ACS were confirmed by the cardiology specialistand the index date was the date of ACS registration. After providing comprehensive supporting information anda rigorous cardiology assessments, we enrolled patients who met the criteria for diagnoses. Variables of exposure. We identified potentially confounding factors based on established risk factors, and performed an analysis to establish whether these variables were substantially associated with ACS. Baseline comorbidities–including hypertension (ICD-9 codes 401-405), diabetes mellitus (DM, ICD-9 code 250), hyperlipidemia (ICD-9-CM 272), and chronic obstructive pulmonary disease (COPD)–are important factors affecting ACS episodes. Therefore, we assessed patients for the factors at the startof pneumococcal pneumonia. Hypertension, diabetes, and hyperlipidemia are conventional risk factors of ACS.12,13,14,15 In addition, COPD, which is characterized by a chronic airway inflammatory process, has been linked to ACS.16 13.

(15) Statistical Analysis We used the chi-square test to examine and compare the distributions of the categorical characteristics between the pneumococcal pneumonia group and the control group. We calculated the follow-up person-years to assess the incidence density rates until ACS was either identified or censored. We used Poisson regression models to evaluate the ratios of the pneumococcal pneumonia cohort to the controls(relative risk) and 95% confidence intervals (CI). To estimate the effect of age on the absolute and relative risk of ACS, we divided the study patients in categories based on age (20-40 years old, 40-54 years old, 55-64 years old, and ≥65 years old) at the index date of pneumococcal pneumonia. We also used Cox proportional-hazards analysis to investigate the association between pneumococcal pneumonia and the risk of developing ACS over time, and adjusted for any cofactors significantly related to pneumococcal pneumonia. A further analysis was performed to assess whether the association of ACS varied according to the length of the follow-up period after pneumococcal pneumonia was diagnosed. We divided the time lag into the following 3 periods: ≤ 3 months, 3 months to 1 year and >1 years. We performed all statistical analyses using the SAS package (Version 9.1 for Windows; SAS institute, Inc., Cary, NC, USA). We adopted a two-tailed P value lower than .05 as the statistical significance level.. 14.

(16) Study Results Descriptive statistical analysis Table 1 shows the demographic characteritics of the study sample.More male patienta were present in our study, and almost 75% of the patients were more than 55 years old (mean age 65.0±17.8 years and 64.9±17.6 years in controls, respectively). The prevalence of comorbidities was greater in the pneumococcal pneumonia group. Statistical analysis inference Table 2 shows the incidence rate ratio (IRR, or relative risk) and adjusted hazard ratio (aHR, or absolute risk) between the pneumococcal pneumonia group and the controls.The overall incidence rate of ACS was 92% higher in the pneumococcal pneumonia group than in the controls (43.1vs22.4 per 10000 person-years) with an aHR of 1.47 (95% CI = 1.24-1.73) in the following 14 years.For women, the incidence densities of the 2 groups are 39.3 and 19.0 per 10000 person-years, with a 2.07-fold relative risk of developing ACS (95% CI = 1.68-2.54). Men have a significantly higher absolute risk (18%) of developing ACS compared to women (95% CI = 1.02-1.37). When stratified by age, the incidence density rates of ACS increase with age, and are the highest in the oldest patients of both groups (70.2 and 35.0, respectively, per 10000 person-years). The 40-54 year-old group had a 3.52-fold relative risk of developing ACS (95% CI = 2.32-5.35) in the pneumococcal pneumonia group than in the controls. However, in the 20-40 year-old age group, the relative risk of developing ACS in the pneumococcal pneumonia group than in the controls is not significantly increased. After adjusting for cofactors, the risk of developing ACS 15.

(17) increased with age (patients 20-40 years of age were the reference group) with an aHR of 24.7 (95% CI = 12.5-48.5). We calculated the aHR according to the length of the follow-up period after pneumococcal pneumonia diagnosis. The relative risk of developing ACS decreased over the follow-up period. We found a 3.90-fold greater relative risk of developing ACS (the highest value) within the first 3-months follow-up period (95% CI = 2.46-6.18). Table 3 shows the specific analysis of the comorbidities of the IRRs and aHRs between the pneumococcal pneumonia group and the controls. In patients without any conventional cardiovascular risk factors, the incidence rate of ACS was 29%higher in the pneumococcal pneumonia group than in the controls (95% CI = 1.03–1.60).When stratified by comorbidites, patients with hypertension, DM, and hyperlipidemia had statistically significantly higher absolute risks of developing ACS compared to those without (aHR = 2.07,95% CI = 1.77-2.41 in hypertension; aHR = 1.77,95% CI = 1.48-2.11 in DM; aHR = 2.14,95% CI = 1.70-2.68 in hyperlipidemia). Table 4 shows the comorbidity effects (joint effects) on ACS. For example, pneumococcal pneumonia patients who had hyperlipidemia in their medical history have a 4.8-fold increased risk of developing ACS than patients without any comorbidities. The Kaplan-Meier survival analysis showed that patients with pneumococcal pneumonia had significantly higher ACS rates than the controls (Log-rank test, P<0.0001) (Fig 1).. 16.

(18) Discussion: Discussion of the study results The result of this study show that patients infected with pneucoccal pneumonia have increased risk of ACS in the long-term follow up. We noted that patients infected with pneumococcal pneumonia had the highest hazard ratio of cumulative incidence of ACS than patients without pneumococcal pneumonia in the first 3 months. The hazard ratio of cumulative incidence of ACS gradually decreased after 3 months but the risk was persistently significantly higher in pneumococcal pneumonia patients. Relatively few studies have discussed the long-term association between patients with pneumococcal pneumonia and ACS. Smeeth et al. presented a within-person comparison by using a case-series method and proposed that patients had an increased risk of acute myocardial infarction (AMI) and stroke within 90 days of the exposure period after an acute respiratory tract infection compared to the baseline period.17 Using the same method, Corrales-Medina et al. proposed that the risk of ACS increased significantly within 15 days of the exposure period after infection with CAP compared to the baseline period. 6 However, we showed that patients infected with pneumococcal pneumonia have a persistently higher risk of a cumulative incidence of ACS than the control group even longer than 1 year after exposure. Using a meta-analysis method, Corrales-Medina VF et al. proposed that patients had an increased risk of incidence of ACS within 30 days of CAP diagnosis.18 Our explaination is that after the antibiotics treatment of the acute infection, the streptococcus pneumonia may still reside within the nasopharynx and lead to asymptomatic 17.

(19) persistence of infection.19Johnstone et al. conducted a retrospective cohort study to investigate long-term outcomes of patients hospitalized for CAP and concluded that up to 5.4 years of follow-up, approximately 31% of patients died of cardiovascular disease.20 Although this study did not compare the cohort of CAP patients with a control group, the result implied that CAP patients may have an increased risk of cardiovascular events even after an acute infectious stage. Our data showed that the cumulative risk of ACS in patients with pneumococcal pneumonia is persistently significantly higher in the long-term follow up compared to the control group. Several studies have discussed the association between CAP and ACS.6,17,19,21,22,23However, few of these studies have focused on the association between pneumococcal pneumonia and ACS,6,7,24 and most of them are limited by a relatively small sample size. This prevents any definite conclusion regarding the association between pneumococcal pneumonia and ACS. Therefore, we used a subset of thedatabase from the reimbursement claims authorized by the National Health Research Institute (NHRI) in Taiwan, which encompasses a larger sample size, to investigate this association. The results of this study are justifiable because of several reasons. First, Streptococcus is the most common pathogen in CAP.3-5 Second, several studies had pointed out that patients with severe form of CAP might have increased risk of ACS. 23 Sahuquillo-Arce et al. proposed that Streptococcus pneumonia infection produces higher serum procalcitonin levels and had a higher inflammatory expression than other atypical pathogen.25 Therefore, streptococcus pneumonia is likely associated with the severe form of 18.

(20) CAP and with ACS. In our study, we do not find any significant difference between the two groups regarding the risk of developing ACS in the 20-40 year-old subgroup. This may be explained by the fact that patients of young age are more immunocompetent and have less comorbidities. Therefore, the severity of the pneumococcal pneumonia in the 20-40 year-old age group may be less severe and less likely to complicate with developing ACS. Discussion of other associated aspects This study has important clinical implications because it primarily focuses on a specific pathogen instead of the broader spectrum of CAP. If we cannot validate that streptococcus pneumonia is associated with an increased risk of ACS, then pneumococcal vaccination might not be effective in preventing ACS. Previous studies have produced controversial results regarding the effectiveness of pneumococcal vaccination in reducing the cumulative incidence of ACS.26-29 The different results of these studies might be due to various study designs. Thus, in order to identify whether pneumococcal vaccination actually reduces the incidence of ACS, we need to consider two aspects. First, we need to establish the association between pneumococcal pneumonia and ACS. In our study, we used a large retrospective cohort database to validate this hypothesis. Second, we need to discuss the effectiveness of the current pneumococcal vaccination in patients with high cardiovascular risk factors. Animal models show that pneumococcal vaccination can reduce atherosclerosis through molecular mimicry between streptococcus pneumonia and oxidized low-density lipoprotein. However, the current 23-valent pneumococcal vaccine offers incomplete 19.

(21) protection and has poor immunogenicity in immunocompromised patients or patients with multiple comorbidities. Therefore, a new generation of a pneumococcal vaccine that could provide good immunogenicity in patients with high cardiovascular risk awaits. 18 Study limitations This study has limitations. First, the NHI database did not disclose patients’ personal histories such as smoking, which is associated with both an increased risk of pneumococcal pneumonia and ACS. However, we attempted to control for this potential confounder by including the COPD covariate because the cause-effect relationship between smoking and COPD had been well validated. Pneumococcal pneumonia is still associated with an increased risk of ACS after adjusting the covariate COPD. In addition, the smoking confounder cannot explain the temporal relationship of the higher hazard ratio of cumulative risk of ACS within 3 months of exposure to pneumococcal pneumonia. Second, previous studies have proposed that in approximately 10% CAP cases, the causative pathogens are mixed. We could not exclude the possibility that some patients in the pneumococcal pneumonia group had a coexistent influenza virus infection. Influenza pneumonia has been linked to an increased risk of ACS. However, viral infection is not routinely checked in daily practice, and we do not know the numbers of cases in our study cohort with both streptococcus pneumonia and influenza virus infection. Third, the evidence derived from a retrospective cohort study is generally lower in statistical quality than that from randomized trials because of potential biases related to adjustments for confounding variables. Fourth, all data in the NHIRD are anonymous. Thus, relevant 20.

(22) clinical variables, such as blood pressure, imaging results, pathology findings, serum laboratory data, and inflammatory markers (especially C-reactive protein), were unavailable for our study cohort.. 21.

(23) Conclusion and Commentary Conclusion In conclusion, our study demonstrates that pneumococcal pneumonia is associated with an increased risk of a cumulative incidence of ACS in the long-term follow up, and the hazard ratio is the highest within the first 3 months after exposure to pneumococcal pneumonia. Our study result implied that pneumococcal vaccination might be used as an option in preventing the occurrence of ACS. Commentary 1. For future studies, we planned to use the same large database to compare if patients with pneumococcal vaccination could reduce the risk of ACS compared with patients without pneumococcal vaccination. 2. We plan to check some inflammatory marker that is associated with the occurrence of ACS in patients with active pneumococcal pneumonia infection and compare with patients without active infection. In addition, we plan to check inflammatory markers in patients with active infection and compare with the convalescent period.. 22.

(24) References: 1. Sims JB, de Lemos JA, Maewal P, et al. Urinary tract infection in patients with acute coronary syndrome: A potential systemic inflammation connection. Am Heart J 2005;149:1062-5. 2. Singanayagam A, Elder DHJ, Chalmers JD, et al. Is community-acquired pneumonia an independent risk factor for cardiovascular disease? EurRespir J 2012;39:187-96. 3. Vila-Corcoles A, Ochoa-Gondar O, Rodriguez-Blanco T, et al. Epidemiology of community-acquired pneumonia in older adults: A population-based study. Respiratory Medicine 2009;103:309-16. 4. Álvarez RF, Toste IS, Cuadrado GR, et al. Community-acquired pneumonia: aetiologic changes in a limited geographic area. An 11-year prospective study. Eur J ClinMicrobiol Infect Dis 2007;26:495-9. 5. Cillóniz C, Ewig S, Polverino E, et al. Community-acquired pneumonia in outpatients: aetiology and outcomes. EurRepir J 2012;40:931-8. 6. Corrales-Medina VF, Serpa J, Rueda AM, et al. Acute Bacterial Pneumonia is associated with the occurrence of acute coronary syndrome. Medicine 2009;88:154-9. 7. Musher DM, Rueda AM, Kaka AS, et al. The association between pneumococcal pneumonia and acute cardiac events. Clin Infect Dis 2007;45:158-65. 8. Yu YH, Liao CC, Hsu WH, et al. Increased lung cancer risk among patients with pulmonary tuberculosis: a population cohort study. J ThoracOncol 2011;6:32-7. 9.Liang JA, Sun LM, Yeh JJ, et al. Malignancies associated with systemic lupus 23.

(25) erythematosus in Taiwan: a nationwide population-based cohortstudy. Rheumatol Int 2012;32:773-8. 10. Cheng CL, Kao YH, Lin SJ, Lee CH, et al. Validation of the National HealthInsurance Research Database with ischemic stroke cases in Taiwan Pharmacoepidemiol Drug Saf 2011;20:236-42. 11. Kang JH, Chen YH, Lin HC. Comorbidity profiles among patients withankylosing spondylitis: A nationwide population-based study. Ann Rheum Dis 2010;69:1165–8. 12. Walston J, McBurnie MA, Newman A, et al. Frailty and activation of the inflammation and coagulation systems with and without clinical comorbidities: results from the Cardiovascular Health Study. Arch Intern Med 2002;162:2333-41. 13. Alexander KP, Newby LK, Armstrong PW, et al. Acute coronary care in the elderly, part II: ST-segment-elevation myocardial infarction: a scientific statement for healthcare professionals from the American Heart Association Council on Clinical Cardiology: in collaboration with the Society of Geriatric Cardiology.. Circulation 2007;115:2570-89.. 14. Sanon S, Patel R, Eshelbrenner C, et al. Acute coronary syndrome in patients with diabetes mellitus: perspectives of an interventional cardiologist. Am J Cardiol 2012;110(9 Suppl):13B-23B. 15. Davidson MH.Cardiovascular risk factors in a patient with diabetes mellitus and coronary artery disease: therapeutic approaches to improve outcomes: perspectives of a preventive cardiologist. Am J Cardiol 2012:110(9 Suppl):43B-49B. 16. Hadi HA, Zubaid M, Al Mahmeed W, et al. Prevalence and prognosis of chronic 24.

(26) obstructive pulmonary disease among 8167 Middle Eastern patients with acute coronary syndrome. ClinCardiol 2010;33:228-35. 17. Smeeth L, Thomas SL, Hall AJ, et al. Risk of myocardial infarction and stroke after acute infection or vaccination. N Engl J Med 2004;351:2611-8. 18. Corrales-Medina VF, Suh KN, Rose G, et al. Cardiac complications in patients with community-acquired pneumonia: A systemic review and meta-analysis of observational studies. PLoS Med 2011;8:e1001048. 19. Villa-Córcoles A. Vaccinate your child and save its grandparents from a heart attack? Current perspectives in antipneumococcal vaccination. J Intern Med 2009;266:432-44. 20. Johnstone J, Eurich DT, Majumdar SR, et al. Long-term morbidity and mortality after hospitalization with community-acquired pneumonia. A population-based cohort study. Medicine 2008;87:329-34. 21. Viasus D, Garcia-Vidal C, Manresa F, et al. Risk stratification and prognosis of acute cardiac events in hospitalized adults with community-acquired pneumonia.. J Infect. 2013;66:27-33. 22. Mandal P, Chalmers JD, Choudhury G, et al. Vascular complications are associated with poor outcome in community-acquired pneumonia. QJM 2011;104:489-95. 23. Corrales-Medina VF, Musher DM, Wells GA, et al. Cardiac complications in patients with community-acquired pneumonia: Incidence, timing, risk factors, and association with short-term mortality. Circulation. 2012;125:773-81. 24. Chen LF, Chen HP, Huang YS, et al. Pneumococcal pneumonia and the risk of stroke: A 25.

(27) population-based follow-up study. PLoS ONE. 2012;7:e51452. 25. Sahuquillo-Arce JM, Reyes S, Martínez R, et al. Cytokine activation patterns and biomarkers are influenced by mocroorganisms in community-acquired pneumonia. Chest. 2012;141:1537-45. 26. Lamontagne F, Garant MP, Carvalho JC, et al. Pneumococcal vaccination and risk of myocardial infarction. CMAJ. 2008;179:773-7. 27. Siriwardena AN, Gwini SM, Coupland CAC. Influenza vaccination, pneumococcal vaccination and risk of acute myocardial infarction: matched case-control study. CMAJ. 2010;182:1617-23. 28. Eurich DT, Johnstone JJ, Minhas-Sandhu JK, et al. Pneumococcal vaccination and risk of acute coronary syndrome in patients with pneumonia: population-based cohort study. Heart. 2012;98:1072-77. 29. Tseng HF, Slezak JM, Quinn VP, et al. Pneumococcal vaccination and risk of acute myocardial infarction and stroke in men. JAMA. 2010;303:1699-706. 30. Binder CJ, Hörkkö S, Dewan A, et al. Pneumococcal vaccination decreases atherosclerotic lesion formation: molecular mimicry between streptococcus pneumonia and oxidized LDL. Nat. Med. 2003;9:736-42. 31. Madjid M, Miller CC, Zarubaev VV, et al. Influenza epidemics and acute respiratory tract disease activity are associated with a surge in autopsy confirmed coronary heart disease death: results from 8-years of autopsies in 34892 subjects. Eur Heart J. 2007;28:1205-10. 26.

(28) Table1. Demographic characteristics and co-morbidity in patients with and without pneumococcal pneumonia Pneumococcal pneumonia Variables. No (N=80444) n. Sex Male Female. %. Yes (N=20111) n. % 1.00. 52640 27804. 65.4 34.6. 13160 6951. 65.4 34.6. Age, years. 1.00. 20-40 40-54. 9856 10668. 12.3 13.3. 2646 2667. 12.3 13.3. 55-64. 11408. 14.2. 2852. 14.2. ≥65 Hypertension Yes. 48512. 60.3. 121218. 60.3. 12501. 15.5. 7858. 39.1. No Diabetes mellitus. 47943. 84.5. 12253. 60.9. Yes No Hyperlipidemia Yes. 6213 74231. No COPD Yes No. p-value†. <0.0001. <0.0001 7.7 92.3. 5090 15021. 25.3 74.7 <0.0001. 2327. 2.9. 1438. 7.1. 78117. 97.1. 18673. 92.9 <0.0001. 2819 77625. 3.5 96.5. 3773 16338. 18.8 81.2. †Chi-square test; occupation missing n=29481. 27.

(29) Table 2. Incidence rate ratio and HR of acutecoronary syndrome and pneumococcal pneumoniacohort to non-pneumococcal pneumonia cohort. 7 54. 328. 716. 1044. event. 76538. 68255 71888. 172381. 292892. 465272. Person years. 35.0. 14.9. 1.0 7.5. 19.0. 24.5. 22.4. rate†. 34. 237. 53. 5 37. 123. 209. 332. event. 14264 61643. 4441. 33780. 12974. 16316 13987. 31267. 45789. 77056. Person years. 36.5 40.0. 76.6. 70.2. 40.9. 3.1 26.5. 39.3. 45.6. 43.1. rate†. 2.43 (1.75-3.38)*** 1.74 (1.51-2.00)***. 3.90 (2.46-6.18)***. 2.01 (1.74-2.32)***. 2.74 (1.98-3.80)***. 2.99 (0.95-9.41) 3.52 (2.32-5.35)***. 2.07 (1.68-2.54)***. 1.86 (1.60-2.18)***. 1.92 (1.70-2.17)***. IRRa (95% CI). -. -. 24.7 (12.5-48.5)***. 10.6 (5.28-21.3)***. reference 8.07 (4.03-16.2)***. reference. 1.18 (1.02-1.37)*. 1.47 (1.24-1.73)***. aHRb (95% CI). Pneumococcal pneumonia. Female. 114 248591. 19.6. 52 246. Yes (N=20111). Age, years 20-40 40-54 869 19861. 15.0 23.0. No (N=80444). 55-64. 39. 73965 389278. Variables. ≥65 Follow up period Within 3 months 111 894. Total Sex Male. Within 1 year > 1 year. Rate† per 10000 person-year; IRRa represented incidence rate ratio; aHRbrepresented adjusted hazard ratio: mutually adjusted for age, gender, comorbidities in Cox proportional hazard regression;CI, confidence interval;*P<0.05, **P<0.01, ***P<0.001. 28.

(30) Table 3. Incidence rate ratio and HR of acutecoronary syndrome and pneumococcal pneumoniacohort to non-pneumococcal pneumonia cohort stratify by comorbidity Pneumococcal pneumonia. 632. event. 418368 46904. 400098. Person years. 19.5. 17.2 68.9. 15.8. rate†. 115. 217. 143 189. 92. event. 72730 4326. 14261. 62796. 55203 21853. 45226. Person years. 38.6 73.6. 37.8 131.8. 80.6. 34.6. 25.9 86.5. 20.3. rate†. 1.78 (1.55-2.04)** 1.22 (0.86-1.72). 1.80 (1.57-2.06)** 1.40 (1.00-1.95). 0.99 (0.79-1.25). 1.77 (1.53-2.06)**. 1.50 (1.26-1.80)** 1.26 (1.05-1.50)*. 1.29 (1.03-1.60)*. reference 1.09 (0.87-1.38). reference 2.14 (1.70-2.68)**. 1.77 (1.48-2.11)**. reference. reference 2.07 (1.77-2.41)**. -. aHRb (95% CI). 721 323 442984. 81.2. 275 57. 67144 9913. IRRa (95% CI). 863 22288. 21.0 94.4. 259 73. Yes (N=57,958). None§. 181. 456264 9008. 21.7 60.5. No (N=231,832). Hypertension No Yes Diabetes mellitus. 959 85. 456012 9260. Variables. Yes Hyperlipidemia No Yes 988 56. No. COPD No Yes. Rate†per 10000 person-year; §Adjusted for age and gender. IRRa represented incidence rate ratio; aHRbrepresented adjusted hazard ratio: mutually. adjusted for age, gender, and comorbidities in Cox proportional hazardregression; CI, confidence interval; *P<0.05, **P<0.01. 29.

(31) 632. Event. 2.99 (2.59-3.44)*** 3.78 (3.18-4.48)***. 1.00 (reference). aHR (95% CI). Non-pneumococcal pneumonia. 1.00 (reference). 323 181. 4.76 (3.79-5.99)*** 2.24 (1.69-2.97)***. aHR (95% CI) 2.67 (2.05-3.48)*** 2.81 (2.12-3.73)***. 85 56. Event 189 115. 4.80 (3.42-6.75)*** 1.91 (1.33-2.73)**. 92. 57 73. Pneumococcal pneumonia. Table 4. Joint effects of associated comorbidities on pneumococcal pneumonia and non- pneumococcal pneumonia for acute coronary syndrome Variables None withhypertension withdiabetes mellitus withdyslipidaemia with COPD. aHRrepresented adjusted hazard ratio: mutually adjusted for age, gender CI, confidence interval; *P<0.05,**P<0.01, ***P<0.001. 30.

(32) Figure 1. Cumulative incidence of acutecoronary syndrome for pneumococcal pneumonia and comparison cohorts in Taiwan.. 31.

(33) 32.

(34)