Anger Expression and Stroke Subtypes
To the Editor:
I have read with interest the article of Everson et al1 and would like to comment on a few points. First, the authors found that the subjects with an “anger-out” score in top one third had a 2-fold increased risk for subsequent stroke compared with those having a low score. The results remained significant after risk factors such as hypertension, diabetes mellitus, cigarette smoking, serum lipids, and fibrinogen were adjusted. However, the presence of coronary heart disease (CHD) was not included in their multivariate analysis. They then divided the subjects according to the presence of CHD and found that the anger-out score was a strong risk factor for stroke in the subjects having CHD but not others. I am curious whether the high anger-out score would have been a significant factor if the presence of CHD had been included in their multivariate analysis. If the result proves to be negative, the anger-out tendency cannot be regarded as a risk factor for stroke in general.
Second, I think that stroke subtype differences may be the reason that the anger-out score was a significant risk factor only in the subjects with CHD. My colleagues and I previously reported that the Tenseness dimension of Eysenk and Fulker’s type A score2 was significantly higher in stroke patients than in control subjects.3 The subjects with a high Tenseness dimension are those who easily become irritated, nervous, and angry. The increased score remained significant even after risk factors such as hypertension, diabetes mellitus, cigarettes smoking, and habitual alcohol drinking were controlled. Thus, our results appear to agree with those of Everson et al.1 In our study, we further divided the strokes into large-vessel infarction (LVI), small-vessel infarction (SVI), and intracerebral hemorrhage (ICH) and found that the increased score was evident only in the patients with LVI, not in those with SVI or ICH. We therefore proposed that these behavioral characteristics are related to the pathogenesis of atherosclerotic large-vessel disease but not small-artery diseases. In the study of Everson et al, stroke subtypes were not studied in detail. However, considering that extracranial atherosclerotic carotid/vertebral diseases are closely associated with CHD,4 5 we may speculate that a majority of strokes occurring in the patients with CHD were LVI while the subjects without CHD more often developed SVI or ICH. This assumption, in concert with our own data, may explain why a high anger-out score was a risk factor for stroke only in the subjects with CHD.
- Copyright © 1999 by American Heart Association
Everson SA, Kaplan GA, Goldberg DE, Lakka TA, Sivenius J, Salonen JT. Anger expression and incident stroke: prospective evidence from the Kuopio Ischemic Heart Disease study. Stroke.. 1999;30:523–528.
Fisher CM, Gore I, Okabe N, White PD. Atherosclerosis of the carotid and vertebral arteries: extracranial and intracranial. J Neuropathol Exp Neurol.. 1965;24:455–476.
Nishino M, Sueyoshi K, Yasuno M, Yamada Y, Abe H, Hori M, Kamada T. Risk factors for carotid atherosclerosis and silent cerebral infarction in patients with coronary heart disease. Angiology.. 1993;44:432–440.
We appreciate Dr Kim’s comments regarding our recent report on anger expression and incident stroke.R1 In response, we recalculated the multivariate analyses of anger-out and incident stroke, including a dummy-coded variable for prevalent ischemic heart disease. As seen in the Table⇑, results were little changed from our original models, with men in the top tertile of anger-out experiencing a nearly 2-fold and significant increased risk of any stroke and a 2.47-fold increased risk of ischemic stroke. In our original paper article,R1 we noted that high levels of anger-out predicted excess risk only in the subgroup of men with a history of prevalent ischemic heart disease. Thus, we have demonstrated that prevalent ischemic heart disease modifies the effect of outward anger expression on stroke risk but does not confound this association.
With respect to stroke subtypes, our original report and the analyses noted above show that the effect of anger expression on stroke risk was stronger for ischemic strokes than for all strokes combined. Too few hemorrhagic strokes occurred in our population to allow reliable assessment of the association between anger-out and hemorrhagic stroke. Unfortunately, we do not currently have the information to examine large-vessel versus small-vessel infarction, as Kim and colleagues did.R2 However, in our sample it is true that nearly all of the strokes that occurred in men with a history of ischemic heart disease were ischemic strokes (20 of 21 events), whereas 30% of the strokes in the men without ischemic heart disease were not ischemic (13 of 43 events). Kim’s hypothesis regarding the significance of behavioral characteristics in the pathogenesis of atherosclerotic large-vessel disease certainly is plausible and remains to be adequately tested.
Everson SA, Kaplan GA, Goldberg DE, Lakka TA, Sivenius J, Salonen JT. Anger expression and incident stroke. Prospective evidence from the Kuopio Ischemic Heart Disease Study. Stroke. 1999;30:523–528.
Kim JS, Yoon SS, Lee Si, Yoo HJ, Choi-Kwon S, Lee BC. Type A behavior and stroke: high tenseness dimension may be a risk factor for cerebral infarction. Eur Neurol. 1998;39:168–173.