Association Between Anger and Mental Stress-induced MI
Between July 2009 and April 2012, the Myocardial Infarction and Mental Stress (MIMS) study enrolled 98 patients between the age of 38 and 60 years with a documented history of MI within the previous 6 months (range 1.3–6 months). Men and women were matched for age (±2 years), type of MI (ST-elevation MI or non–ST-elevation MI), and time since the MI (±2 months). Other inclusion and exclusion criteria and details of sample construction have been described elsewhere.
Subjects underwent 3 single-photon emission computed tomography (SPECT) imaging studies; 1 at rest, 1 with mental stress, and 1 with exercise or pharmacologic stress. The 2 stress scans were obtained in separate days within 1 week of each other (the order was balanced), and the rest scan was obtained during the first session. All testing was done after an overnight fast, and antiischemic medications were held for 24 hours before testing. Sociodemographic and psychosocial data were collected at the first visit before stress testing. At the end of the study protocol, medical records were abstracted for clinical information. The study protocol was approved by the Emory University Institutional Review Board, and informed consent was obtained from all participants.
Mental stress was induced by a standardized social stressor using a public speaking task as previously described. Briefly, subjects were asked to imagine a real-life stressful situation and to make up a realistic story around this scenario. They were given 2 minutes to plan the story and 3 minutes to present it in front of a video camera and a small audience wearing white coats. Subjects were told that their speech would be evaluated by the laboratory staff for content, quality, and duration. For physical stress, subjects underwent a Bruce protocol by walking on a treadmill, with exercise target set at 85% of maximum predicted heart rate based on the patient's sex and age. For 16 subjects who were unable to reach the heart rate target, we performed a pharmacologic stress test with regadenoson (Astellas, Northbrook, IL), an adenosine receptor agonist. Blood pressure and heart rate were monitored during each stress test. Subjective ratings of distress were obtained at baseline and after mental stress with the Subjective Units of Distress Scale on a linear scale of 0 to 100 (100 = highest level of distress). We also obtained visual analog ratings of nervousness, anxiety, fear, and anger with a scale of 0 to 4, with 4 being extreme.
Subjects underwent [Tc]-sestamibi SPECT myocardial perfusion imaging at rest, during mental stress, and during physical stress on a dedicated ultrafast solid-state camera (Discovery NM 530c; GE, Milwaukee, WI) without attenuation correction. We used [Tc]-sestamibi dosages of 10 to 15 mCi for the rest scan and of 30 to 45 mCi for the stress scans, according to body weight and with a dose ratio of rest to stress of 1:3. For mental stress, the radioisotope was injected 1 minute after the onset of the speech; whereas for exercise stress, it was injected at peak exertion after the Bruce protocol. Following standard procedures, stress images were acquired 45 to 60 minutes after [Tc]-sestamibi injection.
Myocardial perfusion was quantified by means of the Emory Cardiac Toolbox software, which provides objective (operator-independent) quantitative assessment of perfusion with established validity and reproducibility. Briefly, the 3-dimensional tracer uptake distribution in the left ventricle was oriented along the short axis and sampled onto a 2-dimensional polar map. A summed score, quantifying the extent and severity of perfusion defects across 17 myocardial segments, was computed. In each region, defect severity was quantified using a 4-point scale from normal (score 0) to absent perfusion (score 4). The regional severity scoring was then summed up across the 17 myocardial segments. Separate scores were obtained for the rest images (summed-rest score [SRS]) and the stress images (summed-stress score [SSS]). For each stress, a summed-difference score (SDS), quantifying the number and severity of reversible (ischemic) myocardial perfusion defects, was obtained by subtracting the rest score from the stress score; a positive SDS would indicate presence of ischemia. We also calculated the percentage of myocardial involvement by dividing the number of myocardial segments with perfusion defects (score >0) by the total number of segments (17). The use of automated image analysis has specific advantages for our study. Quantitative SPECT image analysis is equivalent to visual analysis from expert readers but is more reproducible because it eliminates interpreter variability. Thus, it is better suited for protocols with serial SPECT scans such as in our study.
Anger was assessed using the Spielberger's State-Trait Anger Expression Inventory, a 57-item questionnaire, which measures the following anger dimensions: (1) state anger (intensity of anger at a particular time); (2) trait anger (disposition to experience angry feelings as a trait); and (3) anger expression, including anger out (anger expressed toward others or the environment), anger in (suppression of anger), and anger control; the latter consisting of 2 subscales: anger control (out), the ability to limit expression of anger and anger control (in), the ability to calm down. Larger scores for each dimension indicate more severity of anger, except for the anger-control subscales (higher score indicating better anger control). All scales have good internal consistency (α ranging from 0.70 to 0.87) and validity.
Sociodemographic factors and medical history were assessed using standardized questionnaires. Angiographic data were obtained from the coronary angiogram performed in conjunction with the index MI. Coronary artery disease severity was quantified using the Gensini semiquantitative angiographic scoring system, which takes into account the degree of luminal narrowing along with a multiplier for specific coronary tree locations. If a patient underwent revascularization, the percentage of coronary obstruction used in the scoring reflected the postrevascularization results. Depressive symptoms were assessed with the Beck Depression Inventory II (BDI-II). We also administered the State-Trait Anxiety Inventory to measure state and trait anxiety and the Seattle Angina Questionnaire to assess angina symptoms.
Multiple linear regression models were used to assess the association between summed scores of myocardial perfusion with mental/physical stress and the 6 anger subscales, adjusting for possible confounding factors. The SDS for ischemia quantification was our main outcome variable of interest. Because the SDS for both mental and physical stress was skewed, whereas the SSS for both conditions was normally distributed, we used the SSS scores as dependent variables while adjusting for the rest score (SRS). Because of the mathematical relationship between these scores, the coefficient from a model with SSS as dependent variable, adjusted for SRS, is identical to that from a model, where the dependent variable is the SDS. This strategy allowed us to obtain nonbiased standard errors and P values.
In cumulative hierarchical models, we adjusted for a set of factors that were considered a priori either possible confounding factors or mediators of the relationships under study. Because of the relatively small sample size, we were careful to develop parsimonious models. Adjustment factors included sociodemographic and lifestyle characteristics (age, sex, race, and current cigarette smoking), CAD severity (Gensini score), depressive symptoms (BDI-II score), and trait anxiety. To allow comparison of effects across different anger subscales with unequal score range, we used the interquartile range (IQR) as scaling factor, that is, the distance between the 25th and 75th percentiles. We also assessed the interaction of sex and age (≤50 and >50 years) for each anger subscale in the final models. We performed thorough regression diagnostics to rule out collinearity and outliers[ these analyses showed that our models were appropriate, and no influential data points were present. In additionally, we repeated the analysis using nonparametric generalized additive modeling, which yielded similar results.
This work was supported by the National Institutes of Health (R21-HL093665, R21-HL093665-01A1S1, R01-HL109413, 2R01-HL068630, 2 K24-HL077506, K24-MH076955, R01-MH056120, R01-HL088726, and P01-HL 101398). The authors are solely responsible for the design and conduct of this study, including all study analyses, the drafting-editing of the manuscript, and its final content.
Methods
Subjects
Between July 2009 and April 2012, the Myocardial Infarction and Mental Stress (MIMS) study enrolled 98 patients between the age of 38 and 60 years with a documented history of MI within the previous 6 months (range 1.3–6 months). Men and women were matched for age (±2 years), type of MI (ST-elevation MI or non–ST-elevation MI), and time since the MI (±2 months). Other inclusion and exclusion criteria and details of sample construction have been described elsewhere.
Study Design
Subjects underwent 3 single-photon emission computed tomography (SPECT) imaging studies; 1 at rest, 1 with mental stress, and 1 with exercise or pharmacologic stress. The 2 stress scans were obtained in separate days within 1 week of each other (the order was balanced), and the rest scan was obtained during the first session. All testing was done after an overnight fast, and antiischemic medications were held for 24 hours before testing. Sociodemographic and psychosocial data were collected at the first visit before stress testing. At the end of the study protocol, medical records were abstracted for clinical information. The study protocol was approved by the Emory University Institutional Review Board, and informed consent was obtained from all participants.
Mental and Physical Stress Procedures
Mental stress was induced by a standardized social stressor using a public speaking task as previously described. Briefly, subjects were asked to imagine a real-life stressful situation and to make up a realistic story around this scenario. They were given 2 minutes to plan the story and 3 minutes to present it in front of a video camera and a small audience wearing white coats. Subjects were told that their speech would be evaluated by the laboratory staff for content, quality, and duration. For physical stress, subjects underwent a Bruce protocol by walking on a treadmill, with exercise target set at 85% of maximum predicted heart rate based on the patient's sex and age. For 16 subjects who were unable to reach the heart rate target, we performed a pharmacologic stress test with regadenoson (Astellas, Northbrook, IL), an adenosine receptor agonist. Blood pressure and heart rate were monitored during each stress test. Subjective ratings of distress were obtained at baseline and after mental stress with the Subjective Units of Distress Scale on a linear scale of 0 to 100 (100 = highest level of distress). We also obtained visual analog ratings of nervousness, anxiety, fear, and anger with a scale of 0 to 4, with 4 being extreme.
Myocardial Perfusion Imaging
Subjects underwent [Tc]-sestamibi SPECT myocardial perfusion imaging at rest, during mental stress, and during physical stress on a dedicated ultrafast solid-state camera (Discovery NM 530c; GE, Milwaukee, WI) without attenuation correction. We used [Tc]-sestamibi dosages of 10 to 15 mCi for the rest scan and of 30 to 45 mCi for the stress scans, according to body weight and with a dose ratio of rest to stress of 1:3. For mental stress, the radioisotope was injected 1 minute after the onset of the speech; whereas for exercise stress, it was injected at peak exertion after the Bruce protocol. Following standard procedures, stress images were acquired 45 to 60 minutes after [Tc]-sestamibi injection.
Myocardial perfusion was quantified by means of the Emory Cardiac Toolbox software, which provides objective (operator-independent) quantitative assessment of perfusion with established validity and reproducibility. Briefly, the 3-dimensional tracer uptake distribution in the left ventricle was oriented along the short axis and sampled onto a 2-dimensional polar map. A summed score, quantifying the extent and severity of perfusion defects across 17 myocardial segments, was computed. In each region, defect severity was quantified using a 4-point scale from normal (score 0) to absent perfusion (score 4). The regional severity scoring was then summed up across the 17 myocardial segments. Separate scores were obtained for the rest images (summed-rest score [SRS]) and the stress images (summed-stress score [SSS]). For each stress, a summed-difference score (SDS), quantifying the number and severity of reversible (ischemic) myocardial perfusion defects, was obtained by subtracting the rest score from the stress score; a positive SDS would indicate presence of ischemia. We also calculated the percentage of myocardial involvement by dividing the number of myocardial segments with perfusion defects (score >0) by the total number of segments (17). The use of automated image analysis has specific advantages for our study. Quantitative SPECT image analysis is equivalent to visual analysis from expert readers but is more reproducible because it eliminates interpreter variability. Thus, it is better suited for protocols with serial SPECT scans such as in our study.
Measurement of Anger and Other Covariates
Anger was assessed using the Spielberger's State-Trait Anger Expression Inventory, a 57-item questionnaire, which measures the following anger dimensions: (1) state anger (intensity of anger at a particular time); (2) trait anger (disposition to experience angry feelings as a trait); and (3) anger expression, including anger out (anger expressed toward others or the environment), anger in (suppression of anger), and anger control; the latter consisting of 2 subscales: anger control (out), the ability to limit expression of anger and anger control (in), the ability to calm down. Larger scores for each dimension indicate more severity of anger, except for the anger-control subscales (higher score indicating better anger control). All scales have good internal consistency (α ranging from 0.70 to 0.87) and validity.
Sociodemographic factors and medical history were assessed using standardized questionnaires. Angiographic data were obtained from the coronary angiogram performed in conjunction with the index MI. Coronary artery disease severity was quantified using the Gensini semiquantitative angiographic scoring system, which takes into account the degree of luminal narrowing along with a multiplier for specific coronary tree locations. If a patient underwent revascularization, the percentage of coronary obstruction used in the scoring reflected the postrevascularization results. Depressive symptoms were assessed with the Beck Depression Inventory II (BDI-II). We also administered the State-Trait Anxiety Inventory to measure state and trait anxiety and the Seattle Angina Questionnaire to assess angina symptoms.
Statistical Analysis
Multiple linear regression models were used to assess the association between summed scores of myocardial perfusion with mental/physical stress and the 6 anger subscales, adjusting for possible confounding factors. The SDS for ischemia quantification was our main outcome variable of interest. Because the SDS for both mental and physical stress was skewed, whereas the SSS for both conditions was normally distributed, we used the SSS scores as dependent variables while adjusting for the rest score (SRS). Because of the mathematical relationship between these scores, the coefficient from a model with SSS as dependent variable, adjusted for SRS, is identical to that from a model, where the dependent variable is the SDS. This strategy allowed us to obtain nonbiased standard errors and P values.
In cumulative hierarchical models, we adjusted for a set of factors that were considered a priori either possible confounding factors or mediators of the relationships under study. Because of the relatively small sample size, we were careful to develop parsimonious models. Adjustment factors included sociodemographic and lifestyle characteristics (age, sex, race, and current cigarette smoking), CAD severity (Gensini score), depressive symptoms (BDI-II score), and trait anxiety. To allow comparison of effects across different anger subscales with unequal score range, we used the interquartile range (IQR) as scaling factor, that is, the distance between the 25th and 75th percentiles. We also assessed the interaction of sex and age (≤50 and >50 years) for each anger subscale in the final models. We performed thorough regression diagnostics to rule out collinearity and outliers[ these analyses showed that our models were appropriate, and no influential data points were present. In additionally, we repeated the analysis using nonparametric generalized additive modeling, which yielded similar results.
This work was supported by the National Institutes of Health (R21-HL093665, R21-HL093665-01A1S1, R01-HL109413, 2R01-HL068630, 2 K24-HL077506, K24-MH076955, R01-MH056120, R01-HL088726, and P01-HL 101398). The authors are solely responsible for the design and conduct of this study, including all study analyses, the drafting-editing of the manuscript, and its final content.
Source...