CLINICAL RESEARCH STUDY
Depression and Disease Severity in Patients with Premature Acute Coronary Syndrome Roxanne Pelletier, PhD,a Kim L. Lavoie, PhD,b Simon L. Bacon, PhD,c George Thanassoulis, MD, MSc,d Nadia A. Khan, MD, MSc,e Louise Pilote, MD, MPH, PhDa,f; for the GENESIS-PRAXY Investigators* a Division of Clinical Epidemiology, The Research Institute of the McGill University Health Centre, Montréal, Québec, Canada; bUniversity of Quebec in Montreal and Research Centre, Hôpital du Sacré-Coeur de Montréal, Québec, Canada; cConcordia University and Research Centre, Hôpital du Sacré-Coeur de Montréal, Montréal, Québec, Canada; dDepartment of Medicine and Research Institute, McGill University Health Center, Montréal, Québec, Canada; eDepartment of Medicine, Center for Health Evaluation and Outcomes Science, University of British Columbia, Vancouver, British Columbia, Canada; fDivision of Internal Medicine, McGill University, Montréal, Québec, Canada.
ABSTRACT OBJECTIVES: The association between depression and cardiovascular disease severity in younger patients has not been assessed, and sex differences are unknown. We assessed whether major depression and depressive symptoms were associated with worse cardiovascular disease severity in patients with premature acute coronary syndrome, and we assessed sex differences in these relationships. METHODS: We enrolled 1023 patients (aged 55 years) hospitalized with acute coronary syndrome from 26 centers in Canada, the United States, and Switzerland, through the GENdEr and Sex determInantS of cardiovascular disease: From bench to beyond-Premature Acute Coronary Syndrome study. Left ventricular ejection fraction, Killip class, cardiac troponin I, and Global Registry of Acute Coronary Events score data were collected through chart review. RESULTS: The sample comprised 248 patients with major depression and 302 women. In univariate analyses, major depression was associated with a lower likelihood of having an abnormal left ventricular ejection fraction (odds ratio, 0.70; 95% confidence interval, 0.51-0.97; P ¼ .03) and lower troponin I levels (estimate, 4.04; 95% confidence interval, 8.01 to 0.06; P ¼ .05). After adjustment for sociodemographic and clinical characteristics, neither major depression nor depressive symptoms were associated with disease severity indices, and there were no sex differences. CONCLUSION: The increased risk of adverse events in depressed patients with premature acute coronary syndrome is not explained by disease severity. Ó 2014 Elsevier Inc. All rights reserved. The American Journal of Medicine (2014) 127, 87-93 KEYWORDS: Acute coronary syndrome; Depression; Disease severity; Epidemiology; Women
The relationship between depression and cardiovascular disease has been demonstrated.1,2 Depression is a risk factor for mortality and cardiovascular disease events both in the general population3 and in patients with cardiovascular
disease.4,5 Patients who are depressed after myocardial infarction are at high risk for subsequent events, because depression leads to a 2-fold increased risk of fatal and nonfatal events compared with nondepressed patients.6
Funding: This study was funded by the Canadian Institutes of Health Research and the Heart and Stroke Foundations of Quebec, Nova Scotia, Alberta, Ontario, Yukon and British Columbia, Canada. RP is supported by a Canadian Institute of Health Research Award. KLL is funded by a Canadian Institutes of Health Research New Investigator award and a Fonds de recherche en santé Québec chercheur-boursier award. SLB is supported by a Canadian Institutes of Health Research New Investigator award and a Fonds de recherche en santé Québec chercheur-boursier award. GT is funded by a Fonds de recherche en santé Québec chercheur-boursier clinicien junior I award. NK is funded by a Michael Smith Foundation for Health Research
Career Scientist award. LP is supported by a Fonds de recherche en santé Québec award and holds a James McGill Chair in medicine. Conflict of Interest: None. Authorship: All authors had access to the data and played a role in writing this manuscript. * See Appendix A for a list of Co-Investigators for GENESIS-PRAXY. Requests for reprints should be addressed to Louise Pilote, MD, MPH, PhD, Division of Clinical Epidemiology, Royal Victoria Hospital, 687 Pine Ave West, V Building, Room 2.17, Montreal, Quebec, H3A 1A1. E-mail address: [email protected]
0002-9343/$ -see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjmed.2013.09.026
88
The American Journal of Medicine, Vol 127, No 1, January 2014
However, the relationship between depression and cardioStudy Population and Data Sources vascular prognosis may be explained partly by cardiac disPatients who presented with acute coronary syndrome (see ease severity. Results of a meta-analysis suggested that the “Acute Coronary Syndrome Case Definition”) to one of the relative risk for future coronary heart disease associated participating hospitals and were aged 18 to 55 years, fluent with depression was reduced by 48% when models were in English or French, and able to provide informed consent adjusted for left ventricular ejection fraction.4 Researchers were eligible for the study. Eligible patients with acute also have observed lower left vencoronary syndrome were tricular ejection fraction7-9 and approached by the research nurse CLINICAL SIGNIFICANCE higher Killip classes2 in patients in the Coronary Care Unit or carwith high levels of depressive Women with premature acute coronary diology wards at the earliest symptoms. However, others have possible time after admission. A syndrome have reduced access to care. observed no relationship between total of 1156 patients were Depression seems to delay access to care major depression or depressive recruited, and for the present subin women. symptoms and various markers of study, 1023 patients with comdisease severity, including cardiac plete baseline data were included. Given that mortality is highest in young troponin T,10 left ventricular ejecwomen with acute coronary syndrome, tion fraction,11,12 and Killip depressed women who present with 13 Procedure class. It has been suggested that acute coronary syndrome should be tarAfter the completion of the consent disparities between studies may be geted for prompt management. form, participants were asked to explained by inadequate statistical complete a self-administered quesadjustments, different measures of tionnaire. Anthropometric measudepression, and incomplete mearements and a blood sample were then taken by a research sures of cardiovascular disease severity.11 nurse. Medical chart reviews were carried out by a research The incidence of acute coronary syndrome in younger nurse to collect participants’ medical history and current acute adults, particularly women, is increasing.14 In the INTERcoronary syndrome characteristics, including disease severity HEART study, depression also was more prevalent among data. younger than older participants.15 Furthermore, previous evidence indicates that depression may contribute to worse cardiovascular disease outcomes in women relative to Acute Coronary Syndrome Case Definition men.1,16 Therefore, the relationship between depression and Eligible patients had symptoms consistent with acute corodisease severity in younger male and female patients warrants nary syndrome on hospital presentation and at least 1 of further investigation. Accordingly, in a cohort of more than the following conditions: (1) significant electrocardiogram 1000 patients with premature acute coronary syndrome, we changes: transient ST-segment elevations of 1 mm, STsought to examine whether major depression or depressive segment depressions of 1 mm, new T-wave inversions symptoms are associated with disease severity and to ascerof 1 mm, pseudo-normalization of previously inverted T tain whether the association differs between men and women. waves, new Q waves (one third the height of the R wave or
MATERIALS AND METHODS Study Design The GENdEr and Sex determInantS of cardiovascular disease: From bench to beyond-Premature Acute Coronary Syndrome (GENESIS-PRAXY) study is a multicenter project, for which 24 centers across Canada, 1 in the United States, and 1 in Switzerland participated in the recruitment of patients (See Appendix B, online, for the list of recruiting centers and sites’ principal investigator). Detailed methods of the GENESIS-PRAXY study have been described.17
Ethics In Quebec, a multicenter ethics review allowed for the McGill University Heath Centre to act as the central review board and coordinate ethics approval for all centers. All other centers received ethics approval from their respective hospital ethics review boards.
0.04 seconds), new R wave > S wave in lead V1 (posterior myocardial infarction), new left bundle branch block (changes had to be seen in 2 contiguous leads); or (2) increase in cardiac enzymes: creatine kinase-MB >2 upper limit of the hospital’s normal range or if no creatine kinase-MB available, then total creatine phosphokinase >2 upper limit of the hospital’s normal range, positive troponin I, positive troponin T (Global Registry of Acute Coronary Events [GRACE] Variable Definitions, Version of March 2006).
Demographic and Medical Characteristics Age, sex, education level, marital status, income, antidepressant use, type of acute coronary syndrome (ST-elevation myocardial infarction, noneST-elevation myocardial infarction, unstable angina), hypertension, diabetes, dyslipidemia, smoking status, and previous myocardial infarction were determined using a combination of self-report and chart review data. Body mass index was calculated from height and weight (kilograms/meters squared).
Pelletier et al
Depression and Disease Severity in Acute Coronary Syndrome
Disease Severity Measures Measures of acute coronary syndrome severity included type of acute coronary syndrome, left ventricular ejection fraction, Killip class, cardiac troponin I level, and GRACE score. Left ventricular ejection fraction and Killip class were determined using chart review data. Left ventricular ejection fraction >50% was considered “normal,” and values 50% were considered as “abnormal” left ventricular function. Cardiac troponin I was determined using a blood sample drawn by the research nurse within 24 hours of hospital admission. Collected blood samples were immediately centrifuged and stored locally at 80 C until being transported in dry ice to the McGill University Health Centre. Cardiac troponin I levels were analyzed centrally at the core laboratory of the Royal Victoria Hospital in Montreal, using a highly sensitive enzyme-linked immunosorbent assay.18 The GRACE scores were calculated using chart review data and the Risk Calculator for 6-Month Postdischarge Mortality After Hospitalization for Acute Coronary Syndrome.19
Assessment of Depression Major depression before the index acute coronary syndrome was assessed using the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) criteria.20 As part of the self-report questionnaire, participants had to refer to how they were feeling before their acute coronary syndrome and answer 9 questions representing DSM-IV criteria for major depression. Major depression was identified when patients answered “yes” to 5 or more of the 9 questions and where 1 of the reported symptoms was a loss of interest in daily activities or feeling blue, sad, or depressed most days of the week for 2 weeks or more in a row.20 Recall of depressive symptoms before the index acute coronary syndrome was assessed using the depression subscale of the Hospital Anxiety and Depression Scale (HADSD).21 The HADS is a self-report questionnaire developed to detect adverse anxiety and depressive states. Participants were asked to choose 1 response from the 4 given on a Likert-type scale for each of the 7 questions on the depression subscale. Validity studies have demonstrated sensitivity and specificity values around 80% for the depression subscale of the HADS.22
Statistical Analyses Baseline demographic and medical characteristics of patients with and without major depression were compared using t tests and chi-square tests. In the regression analyses, major depression was a binary independent variable, whereas depressive symptoms were considered as a continuous variable. Because of the bimodal distribution of left ventricular ejection fraction and Killip class, the latter 2 variables were considered as dichotomous outcomes. Left ventricular ejection fraction values >50% were considered “normal,” and values 50% were considered as “abnormal” left ventricular function. Killip class II, III, and IV were
89
grouped and compared against Killip class I. Troponin I levels and GRACE score were used as continuous outcomes. To assess the main effect of major depression, as well as its interaction with sex, on left ventricular ejection fraction, Killip class, troponin I levels, and GRACE score, a series of logistic and linear regressions were performed. For each outcome, univariate and then multivariate analyses were conducted with no interaction term first, and then with the addition of an interaction term between major depression/depressive symptoms and sex. All multivariable models were adjusted for age, sex, marital status, education level, family income, previous myocardial infarction, type of current acute coronary syndrome, diabetes, dyslipidemia, hypertension, body mass index, antidepressant use, and cigarette smoking. All covariates were selected a priori on the basis of previous literature.8,10,11,23,24 All statistical analyses were performed using SAS version 9.2 (SAS Institute Inc, Cary, NC). Statistical tests were 2-sided; a P value .05 was considered statistically significant.
RESULTS Baseline Characteristics Among our sample, 302 (30%) were women and 248 participants (25%) reported symptoms consistent with major depression. As detailed in Table 1, participants with major depression were more likely to be female; to have hypertension, diabetes, dyslipidemia, and a history of myocardial infarction; to be current smokers; and to be taking antidepressants. They were less often married, had lower education levels, and had lower family incomes. They also were less often diagnosed with an ST-elevation myocardial infarction and more often with a none ST-elevation myocardial infarction.
Major Depression and Disease Severity Table 2 shows mean levels of markers of disease severity according to major depression status. In univariate analyses, major depression was associated with less severe cardiac disease, as assessed by left ventricular ejection fraction and troponin I levels. There was no univariate effect of major depression on Killip class and GRACE score (Table 3). However, in the adjusted models, major depression was no longer associated with left ventricular ejection fraction and troponin I levels, and there was no association with Killip class and GRACE score. Analyses including the interaction term between major depression and sex revealed no interaction effect on any of the outcomes, both in the univariate and adjusted models.
Depressive Symptoms (Hospital Anxiety and Depression Scale D) and Disease Severity When the HADS-D score was entered as a continuous variable, depressive symptoms were not associated with left
90
The American Journal of Medicine, Vol 127, No 1, January 2014
Table 1
Patients’ Characteristics According to Major Depression Status
% (n) Sociodemographic Female Age, y, mean (SD) Married Post-secondary education Family income, mean, $ (SD) Risk factors Hypertension Dyslipidemia Diabetes BMI, mean (SD) (weight [kg]/height [m2]) Smoking Previous myocardial infarction Antidepressant use Type of index ACS STEMI NSTEMI Unstable angina
Major Depressionn ¼ 248
No Major Depressionn ¼ 753
P Value
40 48 62 52 58,000
27 48 70 63 76,000
(202) (5.9) (530) (475) (36,000)
Depression and Disease Severity in Patients with Premature Acute Coronary Syndrome Roxanne Pelletier, PhD,a Kim L. Lavoie, PhD,b Simon L. Bacon, PhD,c George Thanassoulis, MD, MSc,d Nadia A. Khan, MD, MSc,e Louise Pilote, MD, MPH, PhDa,f; for the GENESIS-PRAXY Investigators* a Division of Clinical Epidemiology, The Research Institute of the McGill University Health Centre, Montréal, Québec, Canada; bUniversity of Quebec in Montreal and Research Centre, Hôpital du Sacré-Coeur de Montréal, Québec, Canada; cConcordia University and Research Centre, Hôpital du Sacré-Coeur de Montréal, Montréal, Québec, Canada; dDepartment of Medicine and Research Institute, McGill University Health Center, Montréal, Québec, Canada; eDepartment of Medicine, Center for Health Evaluation and Outcomes Science, University of British Columbia, Vancouver, British Columbia, Canada; fDivision of Internal Medicine, McGill University, Montréal, Québec, Canada.
ABSTRACT OBJECTIVES: The association between depression and cardiovascular disease severity in younger patients has not been assessed, and sex differences are unknown. We assessed whether major depression and depressive symptoms were associated with worse cardiovascular disease severity in patients with premature acute coronary syndrome, and we assessed sex differences in these relationships. METHODS: We enrolled 1023 patients (aged 55 years) hospitalized with acute coronary syndrome from 26 centers in Canada, the United States, and Switzerland, through the GENdEr and Sex determInantS of cardiovascular disease: From bench to beyond-Premature Acute Coronary Syndrome study. Left ventricular ejection fraction, Killip class, cardiac troponin I, and Global Registry of Acute Coronary Events score data were collected through chart review. RESULTS: The sample comprised 248 patients with major depression and 302 women. In univariate analyses, major depression was associated with a lower likelihood of having an abnormal left ventricular ejection fraction (odds ratio, 0.70; 95% confidence interval, 0.51-0.97; P ¼ .03) and lower troponin I levels (estimate, 4.04; 95% confidence interval, 8.01 to 0.06; P ¼ .05). After adjustment for sociodemographic and clinical characteristics, neither major depression nor depressive symptoms were associated with disease severity indices, and there were no sex differences. CONCLUSION: The increased risk of adverse events in depressed patients with premature acute coronary syndrome is not explained by disease severity. Ó 2014 Elsevier Inc. All rights reserved. The American Journal of Medicine (2014) 127, 87-93 KEYWORDS: Acute coronary syndrome; Depression; Disease severity; Epidemiology; Women
The relationship between depression and cardiovascular disease has been demonstrated.1,2 Depression is a risk factor for mortality and cardiovascular disease events both in the general population3 and in patients with cardiovascular
disease.4,5 Patients who are depressed after myocardial infarction are at high risk for subsequent events, because depression leads to a 2-fold increased risk of fatal and nonfatal events compared with nondepressed patients.6
Funding: This study was funded by the Canadian Institutes of Health Research and the Heart and Stroke Foundations of Quebec, Nova Scotia, Alberta, Ontario, Yukon and British Columbia, Canada. RP is supported by a Canadian Institute of Health Research Award. KLL is funded by a Canadian Institutes of Health Research New Investigator award and a Fonds de recherche en santé Québec chercheur-boursier award. SLB is supported by a Canadian Institutes of Health Research New Investigator award and a Fonds de recherche en santé Québec chercheur-boursier award. GT is funded by a Fonds de recherche en santé Québec chercheur-boursier clinicien junior I award. NK is funded by a Michael Smith Foundation for Health Research
Career Scientist award. LP is supported by a Fonds de recherche en santé Québec award and holds a James McGill Chair in medicine. Conflict of Interest: None. Authorship: All authors had access to the data and played a role in writing this manuscript. * See Appendix A for a list of Co-Investigators for GENESIS-PRAXY. Requests for reprints should be addressed to Louise Pilote, MD, MPH, PhD, Division of Clinical Epidemiology, Royal Victoria Hospital, 687 Pine Ave West, V Building, Room 2.17, Montreal, Quebec, H3A 1A1. E-mail address: [email protected]
0002-9343/$ -see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjmed.2013.09.026
88
The American Journal of Medicine, Vol 127, No 1, January 2014
However, the relationship between depression and cardioStudy Population and Data Sources vascular prognosis may be explained partly by cardiac disPatients who presented with acute coronary syndrome (see ease severity. Results of a meta-analysis suggested that the “Acute Coronary Syndrome Case Definition”) to one of the relative risk for future coronary heart disease associated participating hospitals and were aged 18 to 55 years, fluent with depression was reduced by 48% when models were in English or French, and able to provide informed consent adjusted for left ventricular ejection fraction.4 Researchers were eligible for the study. Eligible patients with acute also have observed lower left vencoronary syndrome were tricular ejection fraction7-9 and approached by the research nurse CLINICAL SIGNIFICANCE higher Killip classes2 in patients in the Coronary Care Unit or carwith high levels of depressive Women with premature acute coronary diology wards at the earliest symptoms. However, others have possible time after admission. A syndrome have reduced access to care. observed no relationship between total of 1156 patients were Depression seems to delay access to care major depression or depressive recruited, and for the present subin women. symptoms and various markers of study, 1023 patients with comdisease severity, including cardiac plete baseline data were included. Given that mortality is highest in young troponin T,10 left ventricular ejecwomen with acute coronary syndrome, tion fraction,11,12 and Killip depressed women who present with 13 Procedure class. It has been suggested that acute coronary syndrome should be tarAfter the completion of the consent disparities between studies may be geted for prompt management. form, participants were asked to explained by inadequate statistical complete a self-administered quesadjustments, different measures of tionnaire. Anthropometric measudepression, and incomplete mearements and a blood sample were then taken by a research sures of cardiovascular disease severity.11 nurse. Medical chart reviews were carried out by a research The incidence of acute coronary syndrome in younger nurse to collect participants’ medical history and current acute adults, particularly women, is increasing.14 In the INTERcoronary syndrome characteristics, including disease severity HEART study, depression also was more prevalent among data. younger than older participants.15 Furthermore, previous evidence indicates that depression may contribute to worse cardiovascular disease outcomes in women relative to Acute Coronary Syndrome Case Definition men.1,16 Therefore, the relationship between depression and Eligible patients had symptoms consistent with acute corodisease severity in younger male and female patients warrants nary syndrome on hospital presentation and at least 1 of further investigation. Accordingly, in a cohort of more than the following conditions: (1) significant electrocardiogram 1000 patients with premature acute coronary syndrome, we changes: transient ST-segment elevations of 1 mm, STsought to examine whether major depression or depressive segment depressions of 1 mm, new T-wave inversions symptoms are associated with disease severity and to ascerof 1 mm, pseudo-normalization of previously inverted T tain whether the association differs between men and women. waves, new Q waves (one third the height of the R wave or
MATERIALS AND METHODS Study Design The GENdEr and Sex determInantS of cardiovascular disease: From bench to beyond-Premature Acute Coronary Syndrome (GENESIS-PRAXY) study is a multicenter project, for which 24 centers across Canada, 1 in the United States, and 1 in Switzerland participated in the recruitment of patients (See Appendix B, online, for the list of recruiting centers and sites’ principal investigator). Detailed methods of the GENESIS-PRAXY study have been described.17
Ethics In Quebec, a multicenter ethics review allowed for the McGill University Heath Centre to act as the central review board and coordinate ethics approval for all centers. All other centers received ethics approval from their respective hospital ethics review boards.
0.04 seconds), new R wave > S wave in lead V1 (posterior myocardial infarction), new left bundle branch block (changes had to be seen in 2 contiguous leads); or (2) increase in cardiac enzymes: creatine kinase-MB >2 upper limit of the hospital’s normal range or if no creatine kinase-MB available, then total creatine phosphokinase >2 upper limit of the hospital’s normal range, positive troponin I, positive troponin T (Global Registry of Acute Coronary Events [GRACE] Variable Definitions, Version of March 2006).
Demographic and Medical Characteristics Age, sex, education level, marital status, income, antidepressant use, type of acute coronary syndrome (ST-elevation myocardial infarction, noneST-elevation myocardial infarction, unstable angina), hypertension, diabetes, dyslipidemia, smoking status, and previous myocardial infarction were determined using a combination of self-report and chart review data. Body mass index was calculated from height and weight (kilograms/meters squared).
Pelletier et al
Depression and Disease Severity in Acute Coronary Syndrome
Disease Severity Measures Measures of acute coronary syndrome severity included type of acute coronary syndrome, left ventricular ejection fraction, Killip class, cardiac troponin I level, and GRACE score. Left ventricular ejection fraction and Killip class were determined using chart review data. Left ventricular ejection fraction >50% was considered “normal,” and values 50% were considered as “abnormal” left ventricular function. Cardiac troponin I was determined using a blood sample drawn by the research nurse within 24 hours of hospital admission. Collected blood samples were immediately centrifuged and stored locally at 80 C until being transported in dry ice to the McGill University Health Centre. Cardiac troponin I levels were analyzed centrally at the core laboratory of the Royal Victoria Hospital in Montreal, using a highly sensitive enzyme-linked immunosorbent assay.18 The GRACE scores were calculated using chart review data and the Risk Calculator for 6-Month Postdischarge Mortality After Hospitalization for Acute Coronary Syndrome.19
Assessment of Depression Major depression before the index acute coronary syndrome was assessed using the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) criteria.20 As part of the self-report questionnaire, participants had to refer to how they were feeling before their acute coronary syndrome and answer 9 questions representing DSM-IV criteria for major depression. Major depression was identified when patients answered “yes” to 5 or more of the 9 questions and where 1 of the reported symptoms was a loss of interest in daily activities or feeling blue, sad, or depressed most days of the week for 2 weeks or more in a row.20 Recall of depressive symptoms before the index acute coronary syndrome was assessed using the depression subscale of the Hospital Anxiety and Depression Scale (HADSD).21 The HADS is a self-report questionnaire developed to detect adverse anxiety and depressive states. Participants were asked to choose 1 response from the 4 given on a Likert-type scale for each of the 7 questions on the depression subscale. Validity studies have demonstrated sensitivity and specificity values around 80% for the depression subscale of the HADS.22
Statistical Analyses Baseline demographic and medical characteristics of patients with and without major depression were compared using t tests and chi-square tests. In the regression analyses, major depression was a binary independent variable, whereas depressive symptoms were considered as a continuous variable. Because of the bimodal distribution of left ventricular ejection fraction and Killip class, the latter 2 variables were considered as dichotomous outcomes. Left ventricular ejection fraction values >50% were considered “normal,” and values 50% were considered as “abnormal” left ventricular function. Killip class II, III, and IV were
89
grouped and compared against Killip class I. Troponin I levels and GRACE score were used as continuous outcomes. To assess the main effect of major depression, as well as its interaction with sex, on left ventricular ejection fraction, Killip class, troponin I levels, and GRACE score, a series of logistic and linear regressions were performed. For each outcome, univariate and then multivariate analyses were conducted with no interaction term first, and then with the addition of an interaction term between major depression/depressive symptoms and sex. All multivariable models were adjusted for age, sex, marital status, education level, family income, previous myocardial infarction, type of current acute coronary syndrome, diabetes, dyslipidemia, hypertension, body mass index, antidepressant use, and cigarette smoking. All covariates were selected a priori on the basis of previous literature.8,10,11,23,24 All statistical analyses were performed using SAS version 9.2 (SAS Institute Inc, Cary, NC). Statistical tests were 2-sided; a P value .05 was considered statistically significant.
RESULTS Baseline Characteristics Among our sample, 302 (30%) were women and 248 participants (25%) reported symptoms consistent with major depression. As detailed in Table 1, participants with major depression were more likely to be female; to have hypertension, diabetes, dyslipidemia, and a history of myocardial infarction; to be current smokers; and to be taking antidepressants. They were less often married, had lower education levels, and had lower family incomes. They also were less often diagnosed with an ST-elevation myocardial infarction and more often with a none ST-elevation myocardial infarction.
Major Depression and Disease Severity Table 2 shows mean levels of markers of disease severity according to major depression status. In univariate analyses, major depression was associated with less severe cardiac disease, as assessed by left ventricular ejection fraction and troponin I levels. There was no univariate effect of major depression on Killip class and GRACE score (Table 3). However, in the adjusted models, major depression was no longer associated with left ventricular ejection fraction and troponin I levels, and there was no association with Killip class and GRACE score. Analyses including the interaction term between major depression and sex revealed no interaction effect on any of the outcomes, both in the univariate and adjusted models.
Depressive Symptoms (Hospital Anxiety and Depression Scale D) and Disease Severity When the HADS-D score was entered as a continuous variable, depressive symptoms were not associated with left
90
The American Journal of Medicine, Vol 127, No 1, January 2014
Table 1
Patients’ Characteristics According to Major Depression Status
% (n) Sociodemographic Female Age, y, mean (SD) Married Post-secondary education Family income, mean, $ (SD) Risk factors Hypertension Dyslipidemia Diabetes BMI, mean (SD) (weight [kg]/height [m2]) Smoking Previous myocardial infarction Antidepressant use Type of index ACS STEMI NSTEMI Unstable angina
Major Depressionn ¼ 248
No Major Depressionn ¼ 753
P Value
40 48 62 52 58,000
27 48 70 63 76,000
(202) (5.9) (530) (475) (36,000)
