| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the School of Health Care Administration, Taipei Medical University, Taipei, Taiwan (H.-Ching. L.); Department of Psychiatry, Taipei Medical University Hospital, Taipei, Taiwan (S.-Y.T., H.-Chien. L.); Department of Psychiatry, School of Medicine, Taipei Medical University, Taipei, Taiwan (S.-Y.T., H.-Chien. L.).
Address correspondence and reprint requests to Hsin-Chien Lee, Department of Psychiatry, Taipei Medical University Hospital, 252 Wu-Hsing St., Taipei 110, Taiwan. E-mail: ellalee{at}tmu.edu.tw
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Methods: We used administrative claims data from the Taiwan National Health Insurance Research Database covering the years 1997–2002, with the two study cohorts comprising patients hospitalized for bipolar disorder (n = 1429) or appendectomies (n = 4993) in 1997. Multiple logistic regression analyses were performed to compare the crude odds ratio of patients in these cohorts developing AMI following the index discharge by gender.
Results: A total of 2.24% of the bipolar disorder patients developed AMI during the 6-year follow-up period, when compared with 1.72% of the appendectomy patients. The multiple logistic regression analyses revealed that there were no significant relationships between the patients in the two cohorts developing AMI, regardless of gender.
Conclusions: There were no significant differences in the risk of developing AMI between patients with bipolar disorder and patients undergoing appendectomy operations, when compared either by gender or as whole groups.
Key Words: acute myocardial infarction • cardiovascular disease • bipolar disorder • acute mood episode • population-based study
Abbreviations: CVD = cardiovascular disease; AMI = acute myocardial infarction; NHIRD = National Health Insurance Research Dataset.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Nevertheless, despite extensive evidence indicating a much higher prevalence of risk factors for CVD among bipolar disorder patients, including metabolic syndrome, obesity, hypertension, dyslipidemia, diabetes, and smoking (6–9), surprisingly few studies have directly explored the risk of CVD for this particular group of patients. Beyer et al. (10) reported that 10.7% of a clinical population of bipolar outpatients was affected by cardiac disease or hypertension. However, their study did not specify whether the patients were affected by CVD, per se, or did they carry out any direct comparison with a control population. Baune et al. (11) subsequently evaluated the relationship between bipolar disorder and CVD. Using a survey of 4181 participants in Germany, they found that there was no significant increase in the risk of CVD among patients suffering from bipolar disorder. In addition to the major limitation of its cross-sectional design, their study findings were weakened by the rather small number of participants who actually suffered from bipolar disorder (n = 42), a fact clearly undermining the statistical power of the results for detecting differences, and the generalizability of the findings to the population as a whole.
The purpose of this study is to estimate the risk of bipolar disorder patients developing acute myocardial infarction (AMI—the major cause of cardiovascular mortality) during a 6-year follow-up of acute mood episodes, comparing them with another cohort of patients undergoing appendectomies during the same period. Patients were identified from a nationwide population-based dataset in Taiwan, with follow-up carried out from 1997 until the end of 2002. The risks of developing AMI for these two cohorts were calculated and compared.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Study Sample
Our two cohorts were the study cohort, comprised all patients aged 45 and above hospitalized for bipolar disorder in Taiwan between January and December 1997 (with ICD-9-CM codes 296.0X, 296.1X, 296.4X, 296.5X, 296.6X, 296.7X, 296.80, and 296.89 as the primary diagnoses) and the control cohort, comprised all patients aged 45 and above who underwent appendectomies over the same period (with ICD-OP code 47.0 as the primary operative procedure). The reason for excluding patients below the age of 45 years was the low incidence of AMI among this age group. Patients with a previous diagnosis of AMI (ICD-9-CM code 412.XX) were also excluded from this study, as were patients undergoing an appendectomy who had had any diagnosis of a major psychiatric disorder (ICD-9-CM codes 290, 294, 295, 296, and 297). Ultimately, a total of 6422 patients were eligible for inclusion in this study, with 1429 suffering from bipolar disorder and 4993 undergoing appendectomies.
The reason patients who underwent appendectomy were selected as the control group was that they were relatively indistinguishable from the general population; indeed, there were no statistically significant differences with regard to either gender (p = .17) or age (p = .21) between the control cohort and the general population, according to data for the year 1997 released by the Population Affairs Administration of the Ministry of the Interior in Taiwan (12). Furthermore, to the best of our knowledge, appendectomy procedures have no known long-term impact on brain or cardiovascular functioning, and there has never been any study reporting increased risk of bipolar disorder or cardiovascular disease among patients undergoing appendectomies.
These two cohorts were followed-up until the end of 2002 to determine whether they had ever received medical services for treatment of AMI (ICD-9-CM codes 410.XX). Because a substantial number of AMI victims do not survive their visit to the emergency department, the use of emergency department services and hospitalization are included to better reflect the actual incidence of AMI.
Statistical Analysis
The SAS statistical package (SAS System for Windows, Version 8.2; SAS Institute, Cary, NC) was used to perform the analyses. Descriptive statistical analyses, including frequency and percentage, were performed on all the identified variables, with 
2 tests used to examine differences between the two cohorts in terms of demographic characteristics and comorbid medical disorders.
Demographic characteristics included gender and age, with patient age categorized into three groups, 45 to 64, 65 to 74, and >74 years. Comorbid medical disorders including hypertension, diabetes, hyperlipidemia, chronic obstructive pulmonary disease, and renal disease were determined from claims data at the time of the index discharge, because these suggest an increased risk of developing AMI.
The crude odds ratios between these two cohorts of developing AMI from time of inclusion to the end of 2002 were calculated by gender. Then odds ratios obtained by multiple logistic regression after adjusting for patient's age and comorbid medical disorders were found to be significant in univariate analysis. A significance level of p < .05 has been adopted for this study.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
|
Details of the likelihood of developing AMI during the 6-year follow-up period for the two cohorts by gender are provided in Table 2, with AMI occurring among 2.24% of patients suffering from bipolar disorder between 1997 and 2002, and 1.72% of patients undergoing appendectomies during the same period. The 2 test illustrates that there were no significant relationships between the cohorts and the incidence of AMI (p = .20).
|
Similarly, no significant relationship was observed between the two cohorts developing AMI during the follow-up period for both genders, although the study cohort had consistently higher AMI incidences than the control cohort (p = .71 for men and p = .12 for women). Patients with bipolar disorder had greater odds of developing AMI during the follow-up period than those undergoing appendectomies, regardless of gender; however, these relationships did not achieve statistical significance. The results remained after adjusting for age and selected comorbid medical disorders.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
The association between somatic risk factors and bipolar disorder is considered to predispose bipolar patients to an increased risk of vascular disease (8,9). Indeed, we earlier reported an approximate two-fold increase in the risk of stroke for patients suffering from bipolar disorder (13). As for CVD, to the best of our knowledge, no study has yet reported incidences of AMI among bipolar disorder patients.
Although depression has consistently been indicated as an important cardiovascular risk factor in a number of prior studies (14,15), the models linking CVD to bipolar disorder are mainly extrapolated from findings on depression (16). However, the excess cardiovascular comorbidity previously reported for bipolar patients receives very little support from our direct observations in this study.
Despite the higher rate of cardiovascular mortality among patients suffering from bipolar disorder, there was no increased AMI comorbidity found in this study. One possible explanation is that although the odds for bipolar disorder patients of developing AMI did not differ from those of the general population, the cause-specific mortality from CVD could be aggravated by barriers to appropriate care or nonadherence to treatment protocols among this disadvantaged group of patients (17). Indeed, focusing solely on chronic medical conditions, patients with bipolar disorder were significantly more likely to have cardiovascular comorbidity (18).
Although much lower than many Western countries, CVD incidence and prevalence in Taiwan continue to rise with continuing economic development and more Westernized lifestyle (19). It is not clear whether the difference in AMI incidence between bipolar patients and controls will eventually become significant as overall AMI incidence increases.
Increased rates of smoking have been proposed as contributing to the elevated odds for CVD among bipolar disorder patients (18). However, the smoking rate was estimated at no more than 30% among bipolar patients in Taiwan, much less than the smoking rate of at least 45% among Western patients (3,20). Thus, the relationship between bipolar disorder and comorbid medical disorders may be less confounded by smoking and the use of other substances among our selected population.
Our study reveals slightly higher incidences of AMI among female bipolar disorder patients than males. This finding might concur with Swedish observations that women with bipolar disorder had higher rates of cardiovascular mortality than men did (1). The higher risk of AMI in women with bipolar disorder departs from the common beliefs regarding gender and CVD. Nonetheless, information was scant with regard to this gender-specific difference. Obesity, which contributes significantly to the risk of AMI, was found to be associated with female gender among patients with severe mental illness, including bipolar disorder (21). Though male bipolar patients were more frequently overweight than females, female bipolar patients had significantly higher rates of extreme obesity than males (22). Women may exhibit disproportionate weight gain related to the use of psychotropic agents for bipolar disorder as suggested by one prospective study (23). Female bipolar patients would consequently be at an increased risk for developing AMI.
This study also finds that patients with bipolar disorder were more likely to have comorbid hypertension, hyperlipidemia, and renal disease than those undergoing appendectomies, a finding similar to prior studies reporting a higher prevalence of cardiovascular risk factors among patients with bipolar disorder than among the general population. However, at 11.3%, the prevalence of hypertension found in our study is much lower than the 39% reported by Fagiolini et al. (24) and the 35.0% to 39.0% reported by Newcomer (8). One possible reason for this discrepancy is that our study used service utilization data as opposed to the clinical screening used by prior studies. Hypertension could therefore remain undiagnosed because of negligence by healthcare providers or patients themselves, and indeed, could be much more prevalent among bipolar patients, as is diabetes among patients with schizophrenia (25).
A particular strength of this study is the use of a nationwide population-based dataset that enables us to trace all AMI incidents. Nevertheless, this study does suffer from three limitations that need to be addressed. First, the psychiatric and medical diagnoses come from administrative data reported by physicians or hospitals; thus, questions regarding validity of the diagnoses could compromise our findings. However, independently developing a population-based dataset containing this sort of information would be extremely costly and difficult to achieve. Second, the sample of patients with bipolar disorder in this study was based on acute mood episodes that occurred during the first year of the study period; it may not be a truly representative cohort of all bipolar patients, because symptom-free patients would have been excluded from the study sample.
Furthermore, the 6-year follow-up period used in our study could be insufficient to capture the real relationship between bipolar disorder and the development of AMI. The small number of AMI victims in the study cohorts may not provide adequate statistical power to detect significant difference. It may well be that given a longer follow-up period, bipolar disorder patients are in fact more likely to develop AMI than the general population.
Despite these limitations, we found that during a 6-year follow-up period, there were no significant differences in the risk of AMI between patients suffering from bipolar disorder and patients undergoing appendectomies, either as whole groups or by gender. However, because this is a pioneering study, we urge caution regarding attempts to generalize the findings of this study to other populations. A further follow-up study over a longer period is recommended to confirm the findings of this study.
This study is based in part on data from the National Health Insurance Research Database provided by the Bureau of National Health Insurance, Department of Health and managed by the National Health Research Institutes, Taiwan. The interpretations and conclusions contained herein do not represent those of the Bureau of National Health Insurance, Department of Health, or the National Health Research Institutes.
|
NOTES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
DOI:10.1097/PSY.0b013e31815c1e93
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. G. Fiedorowicz, D. A. Solomon, J. Endicott, A. C. Leon, C. Li, J. P. Rice, and W. H. Coryell Manic/Hypomanic Symptom Burden and Cardiovascular Mortality in Bipolar Disorder Psychosom Med, July 1, 2009; 71(6): 598 - 606. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.-C. Lin, P.-Z. Chao, and H.-C. Lee Sudden Sensorineural Hearing Loss Increases the Risk of Stroke: A 5-Year Follow-Up Study Stroke, October 1, 2008; 39(10): 2744 - 2748. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
