Chronic rhinosinusitis confers an increased risk of acute myocardial infarction Pi-Chieh Wang, M.D.,1 Herng-Ching Lin, Ph.D.,2 and Jiunn-Horng Kang, M.D., Ph.D.3.4
Y P
ABSTRACT
Background: The link between chronic inflammatory disease and cardiovascular disease (CVD) is recognized. Chronic rhinosinusitis (CRS) is one of the most common chronic inflammatory diseases. However, whether CRS increases the risk for CVD is still unknown. This epidemiology study investigated the risk for acute myocardial infarction (AMI) in patients with CRS using a large-scale population-based cohort study. Methods: Data on all study cohorts were retrieved from the Longitudinal Health Insurance Database in Taiwan. In total, data on 7975 CRS subjects from 2001 to 2003 were extracted for the study cohort. We selected 39,875 comparison subjects whose demographic variables matched those of the study cohort. We individually tracked each subject for a 6-year period (from 2001 to 2009) to identify which subjects subsequently received a diagnosis of AMI. A stratified Cox proportional hazards regression was used to compare the 6-year risk of a subsequent AMI after a diagnosis of CRS. Results: Among the 47,850 sampled subjects, the incidence rate of AMI during the 6-year follow-up period was 5.66/1000 person-years; rates were 8.49 and 5.09/1000 person-years for the study and comparison cohort, respectively. The hazard ratio (HR) for AMI during the 6-year follow-up period for subjects with CRS was 1.70 (95% confidence interval [CI], 1.52⬃1.91). In addition, after adjusting for cardiovascular risk factors, the HR of AMI for subjects with CRS was 1.48 (95% CI, 1.32⬃1.67) compared with subjects without CRS. Conclusion: Patients with CRS were at higher risk for AMI occurrence in the 6-year follow-up. (Am J Rhinol Allergy 27, e178 –e182, 2013; doi: 10.2500/ajra.2013.27.3952)
G
rowing evidence has indicated that chronic inflammation plays an important role in the development of atherosclerosis and cardiovascular diseases (CVDs). CVDs remain a major cause associated with morbidity and mortality worldwide. Several systemic rheumatic diseases, such as rheumatoid arthritis and ankylosing spondylitis, are recognized as being associated with an elevated risk for CVDs.1,2 In addition, certain infections associated with chronic inflammation were also reported to increase the risk for CVDs.3 Chronic rhinosinusitis (CRS) is one of most prevalent chronic diseases in the world.4,5 CRS is a chronic regional inflammatory disease that involves the paranasal sinuses.6 Although the pathogenesis of CRS is still not well understood, CRS is associated with impaired mucociliary transport and obstruction and stasis of passages of the paranasal sinus. The mucosa of an involved paranasal sinus becomes chronically inflamed and may have a superimposed infection. A recent epidemiological study showed that CRS patients are also at higher risk to develop subsequent stroke.7 It may be caused by the proximity between the inflamed paranasal sinus and the vessels in the intracranial space of rhinosinusitis patients. Nevertheless, involvement of distant vessels such as coronary arteries associated with arthrosclerosis and thrombosis in CRS patients is still unknown. Because CRS appeared as a chronic inflammatory disease, we hypothesize that CRS may exhibit not only focal but also systemic effects to aggravate atherosclerosis. In the present study, we investigated whether patients with CRS are at higher risk of having a subsequent acute myocardial infarction (AMI), one of the most serious consequences of CVDs, using a largescale cohort study. Because CRS is a common chronic disease in the general population, our study can provide an important epidemiological survey of the association between CRS and CVDs.
O D
T
O N
From the 1Department of Family Medicine, PoJen General Hospital, Taipei, Taiwan, and 2School of Health Care Administration, 3Department of Physical Medicine and Rehabilitation, 4School of Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan The authors have no conflicts of interest to declare pertaining to this article Address correspondence and reprint requests to Jiunn-Horng Kang, MD, Ph.D., School of Medicine, College of Medicine, Taipei Medical University, 250 Wu-Hsing Street, Taipei 110, Taiwan E-mail address: [email protected] Copyright © 2013, OceanSide Publications, Inc., U.S.A.
e178
METHODS Database
O C
The sampled subjects analyzed in this cohort study were retrieved from the Longitudinal Health Insurance Database (LHID2000). The LHID2000 is derived from medical claims records of the Taiwan National Health Insurance (NHI) program, which was inaugurated in 1995. This data set consists of all of the original medical claims and registration files for 106 enrollees under the Taiwan NHI program randomly sampled from all the enrollees listed in the 2000 Registry of Beneficiaries (n ⫽ 23.72 million). Each year, the Taiwan National Health Research Institute collects data from the NHI program and sorts it into the LHID2000. The Taiwan National Health Research Institute and some researchers have shown the high validity of the data form the NHI program,8,9 and hundreds of papers using the LHID2000 have been published in internationally peer-reviewed journals.10 This study was exempt from full review by the Institutional Review Board of Taipei Medical University because the LHID2000 consists of de-identified secondary data released to the public for research purposes.
Study Sample This retrospective matched-cohort study featured a study cohort and a comparison cohort. For selection of the study cohort, we first included 25,673 subjects who had received a first-time principal diagnosis of CRS (ICD-9-CM codes 473, 473.0, 473.1, 473.2, 473.3, 473.8, and 473.9) during ambulatory care visits (including outpatient departments of hospitals and clinics) between January 1, 2001 and December 31, 2003. We excluded patients ⬍40 years old (n ⫽ 16,549) because of the very low incidence of AMI in this age group. We then assigned the date of their first ambulatory care visit for receiving a CRS diagnosis as the index date. In addition, we excluded those subjects who had a history of coronary heart disease (CHD; ICD9-CM codes 410–414 or 429.2) before their index date (n ⫽ 1149). However, we were unable to exclude those subjects who had received a diagnosis of CHD before 1995 because the NHI program was initiated in 1995. Ultimately, 7975 CRS subjects were included in the study cohort. For the comparison cohort, we also retrieved subjects from the LHID2000. Similarly, we first excluded all subjects aged ⬍40 years old
November–December 2013, Vol. 27, No. 6
Delivered by Ingenta to: Economics Dept IP: 185.46.86.242 On: Wed, 29 Jun 2016 04:31:00 Copyright (c) Oceanside Publications, Inc. All rights reserved. For permission to copy go to https://www.oceansidepubl.com/permission.htm
Table 1 Distributions of demographic characteristics and comorbid medical disorders for patients in Taiwan with CRS and comparison subjects, 2001⬃2003 (n ⴝ 47,850) Variable
Subjects with CRS n ⴝ 7975
Gender Male Female Age (yr) 40⬃44 45⬃49 50⬃54 55⬃59 60⬃64 65⬃69 70⬃74 ⬎74 Urbanization level 1 (most urbanized) 2 3 4 5 (least urbanized) Stroke Obesity Renal disease Hypertension COPD Hyperlipidemia Diabetes Alcohol abuse/alcohol dependence syndrome Asthma Chronic bronchitis Tobacco use disorder Oral cancer Lung cancer
Comparison Subjects n ⴝ 39,875
Total No.
%
Total No.
%
3712 4263
46.6 53.4
18,560 21,315
46.6 53.4
2216 1634 1216 846 685 559 448 371
27.8 20.5 15.2 10.6 8.6 7.0 5.6 4.7
11,080 8170 6080 4230 3425 2795 2240 1855
27.8 20.5 15.2 10.6 8.6 7.0 5.6 4.7
2720 2308 1283 1010 654 598 64 461 2495 3216 1784 1315 48 1,406 1,845 335 39 37
34.1 28.9 16.1 12.7 8.2 7.5 0.8 5.8 31.3 40.3 22.4 16.5 0.6 17.6 23.1 4.2 0.5 0.5
13,600 11,540 6415 5050 3270 2193 168 1539 9900 7369 6175 5127 60 2,704 3,543 877 171 152
34.1 28.9 16.1 12.7 8.2 5.5 0.4 3.9 24.8 18.5 15.5 12.9 0.4 6.8 8.9 2.2 0.4 0.4
p Value
1.000
Y P 1.000
T
O N
O C
1.000
⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 0.016 ⬍0.001 ⬍0.001 ⬍0.001 0.239 0.126
CRS ⫽ chronic rhinosinusitis; COPD ⫽ chronic obstructive pulmonary disease. Table 2 Incidence rate and HRs for AMI among sample subjects during the 6-yr follow-up period starting from the index date
O D Presence of AMI
Six-year follow-up period Incidence rate per 1000 person-years (95% CI) Crude HR# (95% CI) Adjusted HR§ (95% CI)
Total Sample
No.
Subjects with CRS %
5.66 (5.39⬃5.94) — —
No.
Comparison Subjects %
8.49 (7.69⬃9.34) 1.70* (1.52⬃1.91) 1.48* (1.32⬃1.67)
No.
%
5.09 (4.81⬃5.38) 1.00 1.00
*p ⬍ 0.001. #The HR was calculated by the stratified Cox proportional hazards regression (stratified by age group, gender, urbanization level, and index year). §The HR was calculated by the stratified Cox proportional hazards regression (stratified by age group, gender, urbanization level, and index year) after adjusting for hypertension, diabetes, stroke, hyperlipidemia, renal disease, COPD, alcohol abuse/alcohol dependence syndrome, obesity, asthma, chronic bronchitis, and tobacco use disorder. AMI ⫽ acute myocardial infarction; HRs ⫽ hazard ratios; CI ⫽ confidence interval; COPD ⫽ chronic obstructive pulmonary disease.
and those who had been diagnosed with CRS since the beginning of the NHI program in 1995. Thereafter, 39,875 comparison subjects (5 comparison subject per study subject) were randomly selected to match the study subjects in terms of gender, age group (40⬃44. 45⬃49, 50⬃54, 55⬃59, 60⬃64, 65⬃69, 70⬃74, and ⬎74 years old), urbanization level (5 levels, with 1 referring to the most urbanized and 5 referring to the least), and the year of the index date using the
SAS program proc SurveySelect (SAS System for Windows, Version 8.2; SAS Institute, Inc., Cary, NC). We matched the distribution of urbanization levels between study subjects and comparison subjects to help assure that the study cohort and comparison cohort were reasonably similar in terms of unmeasured neighborhood socioeconomic characteristics. For the comparison cohort, we assigned the date of their first use of medical services occurring during that
American Journal of Rhinology & Allergy
Delivered by Ingenta to: Economics Dept IP: 185.46.86.242 On: Wed, 29 Jun 2016 04:31:00 Copyright (c) Oceanside Publications, Inc. All rights reserved. For permission to copy go to https://www.oceansidepubl.com/permission.htm
e179
matched year as the index date. For study subjects, the year of the index date was the year in which they received their first CRS diagnosis; for comparison subjects, the year of index date was simply a matched year in which controls had medical use. We further assured that none of the selected comparison subjects had received a diagnosis of CHD before their index date. Finally, we individually tracked each subject in this study (n ⫽ 47,850) for a 6-year period (from 2001 to 2009) following their index date to identify which subjects had subsequently received a diagnosis of AMI (ICD-9-CM codes 410 and 410.0⬃410.9) during the follow-up period. Additionally, we censored subjects who had died from nonAMI causes during the 6-year follow-up period in the regression model. Of the total 47,850 sampled subjects, 4777 died from non-AMI causes, including 710 in the study cohort (8.9% of study cohort) and 4067 in the comparison cohort (8.5% of the comparison cohort).
Y P
Statistical Analysis We used the SAS statistical package (SAS System for Windows, Version 8.2) to perform all statistical analyses in this study. We performed chi-squared tests to explore differences in the distribution of medical comorbidities between the study and comparison cohorts. The selected medical comorbidities included hypertension, diabetes, stroke, hyperlipidemia, renal disease, chronic obstructive pulmonary disease (COPD), alcohol abuse/alcohol dependence syndrome, obesity, asthma, chronic bronchitis, and tobacco use disorder. We only counted these comorbidities if they occurred in an inpatient setting or in two or more ambulatory care claims coded within 6 months before and after the index date. We used the Kaplan-Meier method and log-rank test to examine differences in 6-year AMI-free survival rates between the study and comparison cohorts. In addition, we used Cox proportional hazards regressions (stratified by gender, age group, urbanization level, and index year) to calculate the 6-year hazard ratio (HR) of subsequent AMI between the study and comparison cohorts. In this study, we found that the proportional hazards assumption was satisfied because the survival curves for both strata (patients in the study and comparison cohorts) had hazard functions that were proportional over time. Furthermore, we calculated the adjusted HR by using Cox proportional hazards regression after adjusting for hypertension, diabetes, stroke, hyperlipidemia, renal disease, COPD, alcohol abuse/ alcohol dependence syndrome, obesity, asthma, chronic bronchitis, and tobacco use disorder and censoring cases whom had died during the 6-year follow-up period. These selected medical comorbidities were regarded as potential risk factors for CHD. We considered the ␣-level at p ⬍ 0.05 as significant.
O D
Of the 47,850 sampled subjects, the mean age was 53.4 ⫾ 11.1 years. The mean age for men and women was 54.6 ⫾ 11.9 and 53.1 ⫾ 10.3 years, respectively. After matching for gender, age group, urbanization level, and year of the index date, Table 1 shows that subjects with CRS were more likely to have hypertension, diabetes, stroke, hyperlipidemia, renal disease, COPD, alcohol abuse/alcohol dependence syndrome, obesity, asthma, chronic bronchitis, and tobacco use disorder than subjects without CRS. Table 2 shows the incidences and HRs of AMI during the 6-year follow-up period between subjects with and without CRS. Of all 47,850 sampled subjects, the incidence rate of AMI during the 6-year follow-up period was 5.66 (95% confidence interval [CI], 5.39⬃5.94) per 1000 person-years; respective rates were 8.49 (95% CI, 7.69⬃9.34) and 5.09 (95% CI: 4.81⬃5.38) per 1000 person-years for the study and comparison cohorts. Of the sampled subjects, the mean age of AMI for men and women was 60.3 ⫾ 11.4 years and 59.1 ⫾ 10.1 years, respectively. However, in the year of 2008 in Taiwan, the mean age of AMI for men and women was 65.3 ⫾ 13.0 years and 62.4 ⫾ 14.3 years, respectively. The log-rank test revealed that subjects with CRS had a
e180
T
O N
RESULTS
O C
Figure 1. Acute myocardial infarction (AMI)–free interval rates for subjects with chronic rhinosinusitis (CRS) and comparison subjects.
significantly lower 6-year AMI-free survival rate than subjects without CRS (p ⬍ 0.001). The 6-year AMI-free interval curves for the study and comparison groups by the Kaplan-Meier method are displayed in Fig. 1. The stratified Cox proportional hazards regression (stratified by gender, age group, urbanization level, and year of the index date) shows that compared with subjects without CRS, the HR for AMI during the 6-year follow-up period for subjects with CRS was 1.70 (95% CI, 1.52⬃1.91) after censoring cases that had died during the 6-year follow-up period. In addition, after adjusting for hypertension, diabetes, stroke, hyperlipidemia, renal disease, COPD, alcohol abuse/ alcohol dependence syndrome, obesity, asthma, chronic bronchitis, and tobacco use disorder and censoring cases that had died during the 6-year follow-up period, the HR of AMI for subjects with CRS was 1.48 (95% CI, 1.32⬃1.67) compared with subjects without CRS. Table 3 further shows the HR for AMI according to gender. Among men, the HR of AMI for subjects with CRS was only 1.31 (95% CI, 1.06⬃1.63) compared with subjects without CRS. However, among women, the adjusted HR for AMI for subjects with CRS was as high as 1.55 (95% CI, 1.34⬃1.79) over comparison subjects.
DISCUSSION We found that patients with CRS were at higher risk of having a subsequent AMI in the 6-year follow-up period. The Cox proportional hazards regression indicated that HRs for AMI in subjects with CRS were 1.48 (95% CI, 1.32⬃1.67) after adjusting for cardiovascular risks. To the best of our knowledge, this is the first study to report an epidemiological association between CRS and AMI. AMI is a serious condition resulting in significant medical burdens and even mortality.11 Therefore, the association between CRS and AMI should be further evaluated. Our results should not simply be interpreted that CRS can cause AMI. We propose several mechanisms for the observed association between CRS and AMI. First, from a statistical viewpoint, the association between CRS and AMI we observed may suffer from ascertainment bias, in that a subject who has a known medical disease may garner more medical attention and has a higher chance that other unrelated medical conditions may be discovered. However, because
November–December 2013, Vol. 27, No. 6
Delivered by Ingenta to: Economics Dept IP: 185.46.86.242 On: Wed, 29 Jun 2016 04:31:00 Copyright (c) Oceanside Publications, Inc. All rights reserved. For permission to copy go to https://www.oceansidepubl.com/permission.htm
Table 3 HRs for AMI among sample subjects during the 6-yr follow-up period by gender Variable
Gender Group Male Subjects with CRS n%
Presence of AMI Yes Incidence rate per 1000 person-years (95% CI) Crude HR (95% CI) Adjusted HR (95% CI)#§
119 (3.2) 5.34 (4.45⬃6.37)
Female
Comparison subjects n%
Subjects with CRS n%
382 (2.1) 3.43 (3.10⬃3.79)
287 (6.7) 11.22 (9.98⬃12.58)
1.58** (1.28⬃1.94) 1.31* (1.06⬃1.63)
1.00 1.00
Comparison subjects n%
Y P
1.77** (1.54⬃2.03) 1.55** (1.34⬃1.79)
836 (3.9) 6.57 (6.11⬃6.99) 1.00 1.00
*p ⬍ 0.05; ** p ⬍ 0.001. #The HR was calculated by the stratified Cox regression model (stratified by age group, gender, urbanization level, and index year). §Adjustments were made for hypertension, diabetes, stroke, hyperlipidemia, renal disease, COPD, alcohol abuse/alcohol dependence syndrome, obesity, asthma, chronic bronchitis, and tobacco use disorder. AMI ⫽ acute myocardial infarction; CRS ⫽ chronic rhinosinusitis; HRs ⫽ hazard ratios; CI ⫽ confidence interval; COPD ⫽ chronic obstructive pulmonary disease.
of the natural course and clinical presentation of AMI, a diagnosis of AMI would generally not be easily neglected. Thus, the magnitude of the ascertainment bias might be minor in the present study. Second, the association between CRS and AMI may reflect these two diseases possibly sharing some common risks. Asthmas are highly associated with CRS. Previous studies showed that patients with asthma may carry a higher risk for CVDs and stroke.12 Thus, excessive cardiovascular risks in CRS patients may be mediated by underlying comorbidities such as asthma. Thus, excessive cardiovascular risks in CRS patients may be mediated by underlying comorbidities such as asthma. Third, the focal accumulation of fluids containing proinflammatory cytokines and altered activation of the coagulation system in the paranasal sinus was reported in patients with CRS.13,14 Some authors hypothesized that exposure to these proinflammatory cytokines may trigger juxtavascular inflammation and promote intravascular thromboses.15 Elevation of the systemic proinflammatory marker, C-reactive protein, was reported to be an independent risk for CVDs.16 Nevertheless, whether systemic proinflammatory markers are increased in CRS patients is still unclear. One recent study did not find elevated C-reactive protein in patients with CRS.17 Fourth, some consequences of CRS may have adverse effects in aggravating cardiovascular events. For example, some CRS patients may suffer from superimposed infection. Sleep impairment and snoring are also common and notable problems in CRS patients. In addition, sleep-disordered breathing is associated with rhinosinusitis involving complex interactions including focal anatomic changes and immunologic-mediated responses.18,19 Growing evidence showed sleep-disordered breathing has direct proatherosclerotic effects and exacerbate other cardiovascular risks.20 These factors may all act as mediating factors in the development of atherosclerosis and CVDs. Fifth, it has been reported that patients with CRS who suffered from decreased quality of life involved physical, emotional, and functional domains.21 Therefore, these patients may suffer from decreased physical activity, which may be linked to obesity and additional cardiovascular risks. However, the mechanism regarding this association may be complex and requires further exploration. Finally, treatment for CRS may be also linked with excessive risk for subsequent CVDs and AMI. Corticosteroids are often used to treat CRS. Elevated cardiovascular risks may be associated with long-term steroid usage.22 Other decongestion medications, such as sympathomimetic medications, may be associated with elevated blood pressure and have adverse effects in CHDs. Interestingly, we found that patients with CRS had a higher frequency of several coexisting cardiovascular risk factors, such as hy-
O D
T
O N
O C
pertension, hyperlipidemia, obesity, and diabetes compared with the population without CRS. Although some studies noted comparative results,7 the mechanism regarding CRS patients with higher comorbid cardiovascular risks is still unclear. Although we found that the risk for a subsequent AMI remained significant after adjusting for cardiovascular risks in the present study, we suggest that intensive monitoring and management of underlying cardiovascular risk factors is still necessary in CRS patients. Several limitations in our study should be noted. First, some patients with CRS may have been uncoded in our database, particularly patients with minor symptoms and mild disease who did not seek medical service. Thus, the estimated risk for a subsequent AMI in the study cohort may have been diluted and underestimated compared with the comparison cohort. Second, CRS is a heterogeneous disease. Different subgroups, such as those with CRS with and without nasal polyps, may exhibit specific clinical and local immunologic features.23,24 The large sample size of our study cohort may have strengthened our ability to provide a representative sample despite the heterogeneity of CRS and prevent selection bias. Nevertheless, further evaluation of the risk for AMI in different subgroups of CRS patients should be considered. Third, the main ethnic group in Taiwan is Han Chinese. Exploration of our findings in other ethnic groups is recommended. Fourth, although we adjusted for the presence of cardiovascular risks as binary variables in our study, disease severity and disease control for these cardiovascular risks could not be determined. Fifth, some variables associated with CRS, such as environmental exposure to certain air pollutants, could not be determined in the present study. These variables may have confounded the occurrence of CVDs.25 These confounding variables should be further investigated. Sixth, because of the limitation of the database, we were unable to identify the patients with diagnosed CVDs before 1995, which may result in the failure to exclude some patients who had preexisting CVDs and, therefore, misestimate the risk. Seventh, although we performed the adjustment for associated cardiovascular risk factors, the severity and control status of underlying cardiovascular risk factors can not be determined, which could result in overor underadjustment for these risk factors. Finally, like many epidemiologic studies, this investigation may have been victim to a surveillance bias in which subjects with one condition were more likely to be diagnosed with separate and possibly unrelated comorbidities purely based on their increased exposure to the medical community. However, it is unlikely under the NHI system that increased exposure of a patient to a clinician who is diagnosing or treating CRS would result in a greater likelihood for the detection of AMI, a condition that needs urgent attention and that is unlikely to go undiagnosed. In
American Journal of Rhinology & Allergy
Delivered by Ingenta to: Economics Dept IP: 185.46.86.242 On: Wed, 29 Jun 2016 04:31:00 Copyright (c) Oceanside Publications, Inc. All rights reserved. For permission to copy go to https://www.oceansidepubl.com/permission.htm
e181
10.
addition, we further analyzed the prevalence of oral and lung cancer, conditions that are generally not expected to be associated with CRS, between subjects with and without CRS. We found that there was no difference in the prevalence of oral cancer (0.49% versus 0.43%; p ⫽ 0.239) and lung cancer (0.46% versus 0.38%; p ⫽ 0.126) between subjects without CRS (Table 1). Therefore, it is reasonable to assume that there were no differences in health care seeking behavior between the two cohorts.
11. 12.
CONCLUSIONS
13.
14.
We found that patients with CRS were at a higher risk of having a subsequent AMI compared with patients without CRS in a 6-year follow-up period. Additional study is advised to confirm our findings and clarify the pathomechanism.
15. 16.
REFERENCES 1.
2.
3.
4. 5. 6.
7.
8.
9.
Zinger H, Sherer Y, and Shoenfeld Y. Atherosclerosis in autoimmune rheumatic diseases-mechanisms and clinical findings. Clin Rev Allergy Immunol 371:20–28, 2009. Sarzi-Puttini P, Atzeni F, Gerli R, et al. Cardiac involvement in systemic rheumatic diseases: An update. Autoimmun Rev 9:849–852, 2010. Rosenfeld ME, and Campbell LA. Pathogens and atherosclerosis: Update on the potential contribution of multiple infectious organisms to the pathogenesis of atherosclerosis. Thromb Haemost 106:858–867, 2011. Van Cauwenberge P, and Watelet JB. Epidemiology of chronic rhinosinusitis. Thorax 55:S20–S21, 2000. Wang X, and Cutting GR. Chronic rhinosinusitis. Adv Otorhinolaryngol 70:114–121, 2011. Van Crombruggen K, Zhang N, Gevaert P, et al. Pathogenesis of chronic rhinosinusitis: inflammation. J Allergy Clin Immunol 128: 728–732, 2011. Wu CW, Chao PZ, Hao WR, et al. Risk of stroke among patients with rhinosinusitis: A population-based study in Taiwan. Am J Rhinol Allergy 26:278–282, 2012. Lin HC, Xirasagar S, Chen CH, and Hwang YT. Physician’s case volume of intensive care unit pneumonia admissions and in-hospital mortality. Am J Respir Crit Care Med 177:989–994, 2008. Cheng CL, Kao YH, Lin SJ, et al. Validation of the national health insurance research database with ischemic stroke cases in Taiwan. Pharmacoepidemiol Drug Saf 20:236–242, 2011.
O D e182
17.
18. 19. 20.
21.
Chen YC, Wu JC, Chen TJ, and Wetter T. Publicly available databases accelerate academic production. BMJ 342:297–298, 2011. White HD, and Chew DP. Acute myocardial infarction. Lancet 372: 570–584, 2008. Onufrak SJ, Abramson JL, Austin HD, et al. Relation of adult-onset asthma to coronary heart disease and stroke. Am J Cardiol 101:1247– 1252, 2008. Shimizu S, Gabazza EC, Ogawa T, et al. Role of thrombin in chronic rhinosinusitis-associated tissue remodeling. Am J Rhinol Allergy 25: 7–11, 2011. Sejima T, Holtappels G, and Bachert C. The expression of fibrinolytic components in chronic paranasal sinus disease. Am J Rhinol Allergy 25:1–6, 2011. Perez Barreto M, Sahai S, Ameriso S, et al. Sinusitis and carotid artery stroke. Ann Otol Rhinol Laryngol 109:227–230, 2000. Hamirani YS, Pandey S, Rivera JJ, et al. Markers of inflammation and coronary artery calcification: A systematic review. Atherosclerosis 201:1–7, 2008. YıldırımYS, Apuhan T, Koc¸og˘luE, et al. High sensitivity C-reactive protein levels in chronic rhinosinusitis and allergic rhinitis. Kulak Burun Bogaz Ihtis Derg 21:266–269, 2011. Leo G, Triulzi F, and Incorvaia C. Diagnosis of chronic rhinosinusitis. Pediatr Allergy Immunol 23:20–26, 2012. Arter JL, Chi DS, M G, et al. Obstructive sleep apnea, inflammation, and cardiopulmonary disease. Front Biosci 9:2892–2900, 2004. Drager LF, Polotsky VY, and Lorenzi-Filho G. Obstructive sleep apnea: An emerging risk factor for atherosclerosis. Chest 140:534–542, 2011. Birch DS, Saleh HA, Wodehouse T, et al. Assessing the quality of life for patients with chronic rhinosinusitis using the “Rhinosinusitis Disability Index.” Rhinology 39:191–196, 2001. Fardet L, Kassar A, Cabane J, and Flahault A. Corticosteroid-induced adverse events in adults: Frequency, screening and prevention. Drug Saf 30:861–881, 2007. Van Zele T, Claeys S, Gevaert P, et al. Differentiation of chronic sinus diseases by measurement of inflammatory mediators. Allergy 61: 1280–1289, 2006. Huvenne W, van Bruaene N, Zhang N, et al. Chronic rhinosinusitis with and without nasal polyps: What is the difference? Curr Allergy Asthma Rep 9:213–220, 2009. Kunzli N, and Tager IB. Air pollution: From lung to heart. Swiss Med Wkly 135:697–702, 2005. e
O N
T 22.
23.
24.
25.
Y P
O C
November–December 2013, Vol. 27, No. 6
Delivered by Ingenta to: Economics Dept IP: 185.46.86.242 On: Wed, 29 Jun 2016 04:31:00 Copyright (c) Oceanside Publications, Inc. All rights reserved. For permission to copy go to https://www.oceansidepubl.com/permission.htm
Y P
ABSTRACT
Background: The link between chronic inflammatory disease and cardiovascular disease (CVD) is recognized. Chronic rhinosinusitis (CRS) is one of the most common chronic inflammatory diseases. However, whether CRS increases the risk for CVD is still unknown. This epidemiology study investigated the risk for acute myocardial infarction (AMI) in patients with CRS using a large-scale population-based cohort study. Methods: Data on all study cohorts were retrieved from the Longitudinal Health Insurance Database in Taiwan. In total, data on 7975 CRS subjects from 2001 to 2003 were extracted for the study cohort. We selected 39,875 comparison subjects whose demographic variables matched those of the study cohort. We individually tracked each subject for a 6-year period (from 2001 to 2009) to identify which subjects subsequently received a diagnosis of AMI. A stratified Cox proportional hazards regression was used to compare the 6-year risk of a subsequent AMI after a diagnosis of CRS. Results: Among the 47,850 sampled subjects, the incidence rate of AMI during the 6-year follow-up period was 5.66/1000 person-years; rates were 8.49 and 5.09/1000 person-years for the study and comparison cohort, respectively. The hazard ratio (HR) for AMI during the 6-year follow-up period for subjects with CRS was 1.70 (95% confidence interval [CI], 1.52⬃1.91). In addition, after adjusting for cardiovascular risk factors, the HR of AMI for subjects with CRS was 1.48 (95% CI, 1.32⬃1.67) compared with subjects without CRS. Conclusion: Patients with CRS were at higher risk for AMI occurrence in the 6-year follow-up. (Am J Rhinol Allergy 27, e178 –e182, 2013; doi: 10.2500/ajra.2013.27.3952)
G
rowing evidence has indicated that chronic inflammation plays an important role in the development of atherosclerosis and cardiovascular diseases (CVDs). CVDs remain a major cause associated with morbidity and mortality worldwide. Several systemic rheumatic diseases, such as rheumatoid arthritis and ankylosing spondylitis, are recognized as being associated with an elevated risk for CVDs.1,2 In addition, certain infections associated with chronic inflammation were also reported to increase the risk for CVDs.3 Chronic rhinosinusitis (CRS) is one of most prevalent chronic diseases in the world.4,5 CRS is a chronic regional inflammatory disease that involves the paranasal sinuses.6 Although the pathogenesis of CRS is still not well understood, CRS is associated with impaired mucociliary transport and obstruction and stasis of passages of the paranasal sinus. The mucosa of an involved paranasal sinus becomes chronically inflamed and may have a superimposed infection. A recent epidemiological study showed that CRS patients are also at higher risk to develop subsequent stroke.7 It may be caused by the proximity between the inflamed paranasal sinus and the vessels in the intracranial space of rhinosinusitis patients. Nevertheless, involvement of distant vessels such as coronary arteries associated with arthrosclerosis and thrombosis in CRS patients is still unknown. Because CRS appeared as a chronic inflammatory disease, we hypothesize that CRS may exhibit not only focal but also systemic effects to aggravate atherosclerosis. In the present study, we investigated whether patients with CRS are at higher risk of having a subsequent acute myocardial infarction (AMI), one of the most serious consequences of CVDs, using a largescale cohort study. Because CRS is a common chronic disease in the general population, our study can provide an important epidemiological survey of the association between CRS and CVDs.
O D
T
O N
From the 1Department of Family Medicine, PoJen General Hospital, Taipei, Taiwan, and 2School of Health Care Administration, 3Department of Physical Medicine and Rehabilitation, 4School of Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan The authors have no conflicts of interest to declare pertaining to this article Address correspondence and reprint requests to Jiunn-Horng Kang, MD, Ph.D., School of Medicine, College of Medicine, Taipei Medical University, 250 Wu-Hsing Street, Taipei 110, Taiwan E-mail address: [email protected] Copyright © 2013, OceanSide Publications, Inc., U.S.A.
e178
METHODS Database
O C
The sampled subjects analyzed in this cohort study were retrieved from the Longitudinal Health Insurance Database (LHID2000). The LHID2000 is derived from medical claims records of the Taiwan National Health Insurance (NHI) program, which was inaugurated in 1995. This data set consists of all of the original medical claims and registration files for 106 enrollees under the Taiwan NHI program randomly sampled from all the enrollees listed in the 2000 Registry of Beneficiaries (n ⫽ 23.72 million). Each year, the Taiwan National Health Research Institute collects data from the NHI program and sorts it into the LHID2000. The Taiwan National Health Research Institute and some researchers have shown the high validity of the data form the NHI program,8,9 and hundreds of papers using the LHID2000 have been published in internationally peer-reviewed journals.10 This study was exempt from full review by the Institutional Review Board of Taipei Medical University because the LHID2000 consists of de-identified secondary data released to the public for research purposes.
Study Sample This retrospective matched-cohort study featured a study cohort and a comparison cohort. For selection of the study cohort, we first included 25,673 subjects who had received a first-time principal diagnosis of CRS (ICD-9-CM codes 473, 473.0, 473.1, 473.2, 473.3, 473.8, and 473.9) during ambulatory care visits (including outpatient departments of hospitals and clinics) between January 1, 2001 and December 31, 2003. We excluded patients ⬍40 years old (n ⫽ 16,549) because of the very low incidence of AMI in this age group. We then assigned the date of their first ambulatory care visit for receiving a CRS diagnosis as the index date. In addition, we excluded those subjects who had a history of coronary heart disease (CHD; ICD9-CM codes 410–414 or 429.2) before their index date (n ⫽ 1149). However, we were unable to exclude those subjects who had received a diagnosis of CHD before 1995 because the NHI program was initiated in 1995. Ultimately, 7975 CRS subjects were included in the study cohort. For the comparison cohort, we also retrieved subjects from the LHID2000. Similarly, we first excluded all subjects aged ⬍40 years old
November–December 2013, Vol. 27, No. 6
Delivered by Ingenta to: Economics Dept IP: 185.46.86.242 On: Wed, 29 Jun 2016 04:31:00 Copyright (c) Oceanside Publications, Inc. All rights reserved. For permission to copy go to https://www.oceansidepubl.com/permission.htm
Table 1 Distributions of demographic characteristics and comorbid medical disorders for patients in Taiwan with CRS and comparison subjects, 2001⬃2003 (n ⴝ 47,850) Variable
Subjects with CRS n ⴝ 7975
Gender Male Female Age (yr) 40⬃44 45⬃49 50⬃54 55⬃59 60⬃64 65⬃69 70⬃74 ⬎74 Urbanization level 1 (most urbanized) 2 3 4 5 (least urbanized) Stroke Obesity Renal disease Hypertension COPD Hyperlipidemia Diabetes Alcohol abuse/alcohol dependence syndrome Asthma Chronic bronchitis Tobacco use disorder Oral cancer Lung cancer
Comparison Subjects n ⴝ 39,875
Total No.
%
Total No.
%
3712 4263
46.6 53.4
18,560 21,315
46.6 53.4
2216 1634 1216 846 685 559 448 371
27.8 20.5 15.2 10.6 8.6 7.0 5.6 4.7
11,080 8170 6080 4230 3425 2795 2240 1855
27.8 20.5 15.2 10.6 8.6 7.0 5.6 4.7
2720 2308 1283 1010 654 598 64 461 2495 3216 1784 1315 48 1,406 1,845 335 39 37
34.1 28.9 16.1 12.7 8.2 7.5 0.8 5.8 31.3 40.3 22.4 16.5 0.6 17.6 23.1 4.2 0.5 0.5
13,600 11,540 6415 5050 3270 2193 168 1539 9900 7369 6175 5127 60 2,704 3,543 877 171 152
34.1 28.9 16.1 12.7 8.2 5.5 0.4 3.9 24.8 18.5 15.5 12.9 0.4 6.8 8.9 2.2 0.4 0.4
p Value
1.000
Y P 1.000
T
O N
O C
1.000
⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 0.016 ⬍0.001 ⬍0.001 ⬍0.001 0.239 0.126
CRS ⫽ chronic rhinosinusitis; COPD ⫽ chronic obstructive pulmonary disease. Table 2 Incidence rate and HRs for AMI among sample subjects during the 6-yr follow-up period starting from the index date
O D Presence of AMI
Six-year follow-up period Incidence rate per 1000 person-years (95% CI) Crude HR# (95% CI) Adjusted HR§ (95% CI)
Total Sample
No.
Subjects with CRS %
5.66 (5.39⬃5.94) — —
No.
Comparison Subjects %
8.49 (7.69⬃9.34) 1.70* (1.52⬃1.91) 1.48* (1.32⬃1.67)
No.
%
5.09 (4.81⬃5.38) 1.00 1.00
*p ⬍ 0.001. #The HR was calculated by the stratified Cox proportional hazards regression (stratified by age group, gender, urbanization level, and index year). §The HR was calculated by the stratified Cox proportional hazards regression (stratified by age group, gender, urbanization level, and index year) after adjusting for hypertension, diabetes, stroke, hyperlipidemia, renal disease, COPD, alcohol abuse/alcohol dependence syndrome, obesity, asthma, chronic bronchitis, and tobacco use disorder. AMI ⫽ acute myocardial infarction; HRs ⫽ hazard ratios; CI ⫽ confidence interval; COPD ⫽ chronic obstructive pulmonary disease.
and those who had been diagnosed with CRS since the beginning of the NHI program in 1995. Thereafter, 39,875 comparison subjects (5 comparison subject per study subject) were randomly selected to match the study subjects in terms of gender, age group (40⬃44. 45⬃49, 50⬃54, 55⬃59, 60⬃64, 65⬃69, 70⬃74, and ⬎74 years old), urbanization level (5 levels, with 1 referring to the most urbanized and 5 referring to the least), and the year of the index date using the
SAS program proc SurveySelect (SAS System for Windows, Version 8.2; SAS Institute, Inc., Cary, NC). We matched the distribution of urbanization levels between study subjects and comparison subjects to help assure that the study cohort and comparison cohort were reasonably similar in terms of unmeasured neighborhood socioeconomic characteristics. For the comparison cohort, we assigned the date of their first use of medical services occurring during that
American Journal of Rhinology & Allergy
Delivered by Ingenta to: Economics Dept IP: 185.46.86.242 On: Wed, 29 Jun 2016 04:31:00 Copyright (c) Oceanside Publications, Inc. All rights reserved. For permission to copy go to https://www.oceansidepubl.com/permission.htm
e179
matched year as the index date. For study subjects, the year of the index date was the year in which they received their first CRS diagnosis; for comparison subjects, the year of index date was simply a matched year in which controls had medical use. We further assured that none of the selected comparison subjects had received a diagnosis of CHD before their index date. Finally, we individually tracked each subject in this study (n ⫽ 47,850) for a 6-year period (from 2001 to 2009) following their index date to identify which subjects had subsequently received a diagnosis of AMI (ICD-9-CM codes 410 and 410.0⬃410.9) during the follow-up period. Additionally, we censored subjects who had died from nonAMI causes during the 6-year follow-up period in the regression model. Of the total 47,850 sampled subjects, 4777 died from non-AMI causes, including 710 in the study cohort (8.9% of study cohort) and 4067 in the comparison cohort (8.5% of the comparison cohort).
Y P
Statistical Analysis We used the SAS statistical package (SAS System for Windows, Version 8.2) to perform all statistical analyses in this study. We performed chi-squared tests to explore differences in the distribution of medical comorbidities between the study and comparison cohorts. The selected medical comorbidities included hypertension, diabetes, stroke, hyperlipidemia, renal disease, chronic obstructive pulmonary disease (COPD), alcohol abuse/alcohol dependence syndrome, obesity, asthma, chronic bronchitis, and tobacco use disorder. We only counted these comorbidities if they occurred in an inpatient setting or in two or more ambulatory care claims coded within 6 months before and after the index date. We used the Kaplan-Meier method and log-rank test to examine differences in 6-year AMI-free survival rates between the study and comparison cohorts. In addition, we used Cox proportional hazards regressions (stratified by gender, age group, urbanization level, and index year) to calculate the 6-year hazard ratio (HR) of subsequent AMI between the study and comparison cohorts. In this study, we found that the proportional hazards assumption was satisfied because the survival curves for both strata (patients in the study and comparison cohorts) had hazard functions that were proportional over time. Furthermore, we calculated the adjusted HR by using Cox proportional hazards regression after adjusting for hypertension, diabetes, stroke, hyperlipidemia, renal disease, COPD, alcohol abuse/ alcohol dependence syndrome, obesity, asthma, chronic bronchitis, and tobacco use disorder and censoring cases whom had died during the 6-year follow-up period. These selected medical comorbidities were regarded as potential risk factors for CHD. We considered the ␣-level at p ⬍ 0.05 as significant.
O D
Of the 47,850 sampled subjects, the mean age was 53.4 ⫾ 11.1 years. The mean age for men and women was 54.6 ⫾ 11.9 and 53.1 ⫾ 10.3 years, respectively. After matching for gender, age group, urbanization level, and year of the index date, Table 1 shows that subjects with CRS were more likely to have hypertension, diabetes, stroke, hyperlipidemia, renal disease, COPD, alcohol abuse/alcohol dependence syndrome, obesity, asthma, chronic bronchitis, and tobacco use disorder than subjects without CRS. Table 2 shows the incidences and HRs of AMI during the 6-year follow-up period between subjects with and without CRS. Of all 47,850 sampled subjects, the incidence rate of AMI during the 6-year follow-up period was 5.66 (95% confidence interval [CI], 5.39⬃5.94) per 1000 person-years; respective rates were 8.49 (95% CI, 7.69⬃9.34) and 5.09 (95% CI: 4.81⬃5.38) per 1000 person-years for the study and comparison cohorts. Of the sampled subjects, the mean age of AMI for men and women was 60.3 ⫾ 11.4 years and 59.1 ⫾ 10.1 years, respectively. However, in the year of 2008 in Taiwan, the mean age of AMI for men and women was 65.3 ⫾ 13.0 years and 62.4 ⫾ 14.3 years, respectively. The log-rank test revealed that subjects with CRS had a
e180
T
O N
RESULTS
O C
Figure 1. Acute myocardial infarction (AMI)–free interval rates for subjects with chronic rhinosinusitis (CRS) and comparison subjects.
significantly lower 6-year AMI-free survival rate than subjects without CRS (p ⬍ 0.001). The 6-year AMI-free interval curves for the study and comparison groups by the Kaplan-Meier method are displayed in Fig. 1. The stratified Cox proportional hazards regression (stratified by gender, age group, urbanization level, and year of the index date) shows that compared with subjects without CRS, the HR for AMI during the 6-year follow-up period for subjects with CRS was 1.70 (95% CI, 1.52⬃1.91) after censoring cases that had died during the 6-year follow-up period. In addition, after adjusting for hypertension, diabetes, stroke, hyperlipidemia, renal disease, COPD, alcohol abuse/ alcohol dependence syndrome, obesity, asthma, chronic bronchitis, and tobacco use disorder and censoring cases that had died during the 6-year follow-up period, the HR of AMI for subjects with CRS was 1.48 (95% CI, 1.32⬃1.67) compared with subjects without CRS. Table 3 further shows the HR for AMI according to gender. Among men, the HR of AMI for subjects with CRS was only 1.31 (95% CI, 1.06⬃1.63) compared with subjects without CRS. However, among women, the adjusted HR for AMI for subjects with CRS was as high as 1.55 (95% CI, 1.34⬃1.79) over comparison subjects.
DISCUSSION We found that patients with CRS were at higher risk of having a subsequent AMI in the 6-year follow-up period. The Cox proportional hazards regression indicated that HRs for AMI in subjects with CRS were 1.48 (95% CI, 1.32⬃1.67) after adjusting for cardiovascular risks. To the best of our knowledge, this is the first study to report an epidemiological association between CRS and AMI. AMI is a serious condition resulting in significant medical burdens and even mortality.11 Therefore, the association between CRS and AMI should be further evaluated. Our results should not simply be interpreted that CRS can cause AMI. We propose several mechanisms for the observed association between CRS and AMI. First, from a statistical viewpoint, the association between CRS and AMI we observed may suffer from ascertainment bias, in that a subject who has a known medical disease may garner more medical attention and has a higher chance that other unrelated medical conditions may be discovered. However, because
November–December 2013, Vol. 27, No. 6
Delivered by Ingenta to: Economics Dept IP: 185.46.86.242 On: Wed, 29 Jun 2016 04:31:00 Copyright (c) Oceanside Publications, Inc. All rights reserved. For permission to copy go to https://www.oceansidepubl.com/permission.htm
Table 3 HRs for AMI among sample subjects during the 6-yr follow-up period by gender Variable
Gender Group Male Subjects with CRS n%
Presence of AMI Yes Incidence rate per 1000 person-years (95% CI) Crude HR (95% CI) Adjusted HR (95% CI)#§
119 (3.2) 5.34 (4.45⬃6.37)
Female
Comparison subjects n%
Subjects with CRS n%
382 (2.1) 3.43 (3.10⬃3.79)
287 (6.7) 11.22 (9.98⬃12.58)
1.58** (1.28⬃1.94) 1.31* (1.06⬃1.63)
1.00 1.00
Comparison subjects n%
Y P
1.77** (1.54⬃2.03) 1.55** (1.34⬃1.79)
836 (3.9) 6.57 (6.11⬃6.99) 1.00 1.00
*p ⬍ 0.05; ** p ⬍ 0.001. #The HR was calculated by the stratified Cox regression model (stratified by age group, gender, urbanization level, and index year). §Adjustments were made for hypertension, diabetes, stroke, hyperlipidemia, renal disease, COPD, alcohol abuse/alcohol dependence syndrome, obesity, asthma, chronic bronchitis, and tobacco use disorder. AMI ⫽ acute myocardial infarction; CRS ⫽ chronic rhinosinusitis; HRs ⫽ hazard ratios; CI ⫽ confidence interval; COPD ⫽ chronic obstructive pulmonary disease.
of the natural course and clinical presentation of AMI, a diagnosis of AMI would generally not be easily neglected. Thus, the magnitude of the ascertainment bias might be minor in the present study. Second, the association between CRS and AMI may reflect these two diseases possibly sharing some common risks. Asthmas are highly associated with CRS. Previous studies showed that patients with asthma may carry a higher risk for CVDs and stroke.12 Thus, excessive cardiovascular risks in CRS patients may be mediated by underlying comorbidities such as asthma. Thus, excessive cardiovascular risks in CRS patients may be mediated by underlying comorbidities such as asthma. Third, the focal accumulation of fluids containing proinflammatory cytokines and altered activation of the coagulation system in the paranasal sinus was reported in patients with CRS.13,14 Some authors hypothesized that exposure to these proinflammatory cytokines may trigger juxtavascular inflammation and promote intravascular thromboses.15 Elevation of the systemic proinflammatory marker, C-reactive protein, was reported to be an independent risk for CVDs.16 Nevertheless, whether systemic proinflammatory markers are increased in CRS patients is still unclear. One recent study did not find elevated C-reactive protein in patients with CRS.17 Fourth, some consequences of CRS may have adverse effects in aggravating cardiovascular events. For example, some CRS patients may suffer from superimposed infection. Sleep impairment and snoring are also common and notable problems in CRS patients. In addition, sleep-disordered breathing is associated with rhinosinusitis involving complex interactions including focal anatomic changes and immunologic-mediated responses.18,19 Growing evidence showed sleep-disordered breathing has direct proatherosclerotic effects and exacerbate other cardiovascular risks.20 These factors may all act as mediating factors in the development of atherosclerosis and CVDs. Fifth, it has been reported that patients with CRS who suffered from decreased quality of life involved physical, emotional, and functional domains.21 Therefore, these patients may suffer from decreased physical activity, which may be linked to obesity and additional cardiovascular risks. However, the mechanism regarding this association may be complex and requires further exploration. Finally, treatment for CRS may be also linked with excessive risk for subsequent CVDs and AMI. Corticosteroids are often used to treat CRS. Elevated cardiovascular risks may be associated with long-term steroid usage.22 Other decongestion medications, such as sympathomimetic medications, may be associated with elevated blood pressure and have adverse effects in CHDs. Interestingly, we found that patients with CRS had a higher frequency of several coexisting cardiovascular risk factors, such as hy-
O D
T
O N
O C
pertension, hyperlipidemia, obesity, and diabetes compared with the population without CRS. Although some studies noted comparative results,7 the mechanism regarding CRS patients with higher comorbid cardiovascular risks is still unclear. Although we found that the risk for a subsequent AMI remained significant after adjusting for cardiovascular risks in the present study, we suggest that intensive monitoring and management of underlying cardiovascular risk factors is still necessary in CRS patients. Several limitations in our study should be noted. First, some patients with CRS may have been uncoded in our database, particularly patients with minor symptoms and mild disease who did not seek medical service. Thus, the estimated risk for a subsequent AMI in the study cohort may have been diluted and underestimated compared with the comparison cohort. Second, CRS is a heterogeneous disease. Different subgroups, such as those with CRS with and without nasal polyps, may exhibit specific clinical and local immunologic features.23,24 The large sample size of our study cohort may have strengthened our ability to provide a representative sample despite the heterogeneity of CRS and prevent selection bias. Nevertheless, further evaluation of the risk for AMI in different subgroups of CRS patients should be considered. Third, the main ethnic group in Taiwan is Han Chinese. Exploration of our findings in other ethnic groups is recommended. Fourth, although we adjusted for the presence of cardiovascular risks as binary variables in our study, disease severity and disease control for these cardiovascular risks could not be determined. Fifth, some variables associated with CRS, such as environmental exposure to certain air pollutants, could not be determined in the present study. These variables may have confounded the occurrence of CVDs.25 These confounding variables should be further investigated. Sixth, because of the limitation of the database, we were unable to identify the patients with diagnosed CVDs before 1995, which may result in the failure to exclude some patients who had preexisting CVDs and, therefore, misestimate the risk. Seventh, although we performed the adjustment for associated cardiovascular risk factors, the severity and control status of underlying cardiovascular risk factors can not be determined, which could result in overor underadjustment for these risk factors. Finally, like many epidemiologic studies, this investigation may have been victim to a surveillance bias in which subjects with one condition were more likely to be diagnosed with separate and possibly unrelated comorbidities purely based on their increased exposure to the medical community. However, it is unlikely under the NHI system that increased exposure of a patient to a clinician who is diagnosing or treating CRS would result in a greater likelihood for the detection of AMI, a condition that needs urgent attention and that is unlikely to go undiagnosed. In
American Journal of Rhinology & Allergy
Delivered by Ingenta to: Economics Dept IP: 185.46.86.242 On: Wed, 29 Jun 2016 04:31:00 Copyright (c) Oceanside Publications, Inc. All rights reserved. For permission to copy go to https://www.oceansidepubl.com/permission.htm
e181
10.
addition, we further analyzed the prevalence of oral and lung cancer, conditions that are generally not expected to be associated with CRS, between subjects with and without CRS. We found that there was no difference in the prevalence of oral cancer (0.49% versus 0.43%; p ⫽ 0.239) and lung cancer (0.46% versus 0.38%; p ⫽ 0.126) between subjects without CRS (Table 1). Therefore, it is reasonable to assume that there were no differences in health care seeking behavior between the two cohorts.
11. 12.
CONCLUSIONS
13.
14.
We found that patients with CRS were at a higher risk of having a subsequent AMI compared with patients without CRS in a 6-year follow-up period. Additional study is advised to confirm our findings and clarify the pathomechanism.
15. 16.
REFERENCES 1.
2.
3.
4. 5. 6.
7.
8.
9.
Zinger H, Sherer Y, and Shoenfeld Y. Atherosclerosis in autoimmune rheumatic diseases-mechanisms and clinical findings. Clin Rev Allergy Immunol 371:20–28, 2009. Sarzi-Puttini P, Atzeni F, Gerli R, et al. Cardiac involvement in systemic rheumatic diseases: An update. Autoimmun Rev 9:849–852, 2010. Rosenfeld ME, and Campbell LA. Pathogens and atherosclerosis: Update on the potential contribution of multiple infectious organisms to the pathogenesis of atherosclerosis. Thromb Haemost 106:858–867, 2011. Van Cauwenberge P, and Watelet JB. Epidemiology of chronic rhinosinusitis. Thorax 55:S20–S21, 2000. Wang X, and Cutting GR. Chronic rhinosinusitis. Adv Otorhinolaryngol 70:114–121, 2011. Van Crombruggen K, Zhang N, Gevaert P, et al. Pathogenesis of chronic rhinosinusitis: inflammation. J Allergy Clin Immunol 128: 728–732, 2011. Wu CW, Chao PZ, Hao WR, et al. Risk of stroke among patients with rhinosinusitis: A population-based study in Taiwan. Am J Rhinol Allergy 26:278–282, 2012. Lin HC, Xirasagar S, Chen CH, and Hwang YT. Physician’s case volume of intensive care unit pneumonia admissions and in-hospital mortality. Am J Respir Crit Care Med 177:989–994, 2008. Cheng CL, Kao YH, Lin SJ, et al. Validation of the national health insurance research database with ischemic stroke cases in Taiwan. Pharmacoepidemiol Drug Saf 20:236–242, 2011.
O D e182
17.
18. 19. 20.
21.
Chen YC, Wu JC, Chen TJ, and Wetter T. Publicly available databases accelerate academic production. BMJ 342:297–298, 2011. White HD, and Chew DP. Acute myocardial infarction. Lancet 372: 570–584, 2008. Onufrak SJ, Abramson JL, Austin HD, et al. Relation of adult-onset asthma to coronary heart disease and stroke. Am J Cardiol 101:1247– 1252, 2008. Shimizu S, Gabazza EC, Ogawa T, et al. Role of thrombin in chronic rhinosinusitis-associated tissue remodeling. Am J Rhinol Allergy 25: 7–11, 2011. Sejima T, Holtappels G, and Bachert C. The expression of fibrinolytic components in chronic paranasal sinus disease. Am J Rhinol Allergy 25:1–6, 2011. Perez Barreto M, Sahai S, Ameriso S, et al. Sinusitis and carotid artery stroke. Ann Otol Rhinol Laryngol 109:227–230, 2000. Hamirani YS, Pandey S, Rivera JJ, et al. Markers of inflammation and coronary artery calcification: A systematic review. Atherosclerosis 201:1–7, 2008. YıldırımYS, Apuhan T, Koc¸og˘luE, et al. High sensitivity C-reactive protein levels in chronic rhinosinusitis and allergic rhinitis. Kulak Burun Bogaz Ihtis Derg 21:266–269, 2011. Leo G, Triulzi F, and Incorvaia C. Diagnosis of chronic rhinosinusitis. Pediatr Allergy Immunol 23:20–26, 2012. Arter JL, Chi DS, M G, et al. Obstructive sleep apnea, inflammation, and cardiopulmonary disease. Front Biosci 9:2892–2900, 2004. Drager LF, Polotsky VY, and Lorenzi-Filho G. Obstructive sleep apnea: An emerging risk factor for atherosclerosis. Chest 140:534–542, 2011. Birch DS, Saleh HA, Wodehouse T, et al. Assessing the quality of life for patients with chronic rhinosinusitis using the “Rhinosinusitis Disability Index.” Rhinology 39:191–196, 2001. Fardet L, Kassar A, Cabane J, and Flahault A. Corticosteroid-induced adverse events in adults: Frequency, screening and prevention. Drug Saf 30:861–881, 2007. Van Zele T, Claeys S, Gevaert P, et al. Differentiation of chronic sinus diseases by measurement of inflammatory mediators. Allergy 61: 1280–1289, 2006. Huvenne W, van Bruaene N, Zhang N, et al. Chronic rhinosinusitis with and without nasal polyps: What is the difference? Curr Allergy Asthma Rep 9:213–220, 2009. Kunzli N, and Tager IB. Air pollution: From lung to heart. Swiss Med Wkly 135:697–702, 2005. e
O N
T 22.
23.
24.
25.
Y P
O C
November–December 2013, Vol. 27, No. 6
Delivered by Ingenta to: Economics Dept IP: 185.46.86.242 On: Wed, 29 Jun 2016 04:31:00 Copyright (c) Oceanside Publications, Inc. All rights reserved. For permission to copy go to https://www.oceansidepubl.com/permission.htm
