Subgroups of patients with atypical circadian patterns of symptom onset in acute myocardial infarction

Abstract

Circadian variation of onset of acute myocardial infarction has been noted in many studies and may carry important implications. However, only a few previous studies have attempted subgroup analysis. In 4,796 patients with documented acute myocardial infarction, the time of symptom onset was recorded. As in other studies, a peak occurred in the morning hours from 6 A.M. to noon with 28% of the population (1.16 times the average percent for the other time periods) experiencing symptom onset in that period (p < 0.001). There was a second lower peak (25%) in the evening hours between 6:01 P.M. and midnight, also observed in some previous studies. Whether the presence of subgroups with specific clinical characteristics would exhibit different patterns and thereby contribute to these peaks in the overall population was then determined. In patients with a history of congestive heart failure (n = 606) or with non-Q-wave infarction (n = 832), a pronounced peak (29%) occurred only in the evening hours. In patients >70 years of age (n = 1,422), smokers (n = 2,057), diabetics (n = 767), women (n = 1,213) and patients taking β blocking drugs (n = 847), 2 nearly equal peaks were observed. Finally, in patients with previous myocardial infarction (n = 1,104), no peaks were observed. In a subgroup of patients (n = 1,084) free from the most important modifying factors, there was a single very pronounced late morning peak (32%), 1.39 times the average percent for the other time periods (p < 0.001), without evidence of a second evening peak. Marked differences in diurnal patterns of myocardial infarction onset occur in subgroups of patients with modifying factors, particularly non-Q-wave infarction, smoking, β blocker use, diabetes, prior congestive heart failure and prior myocardial infarction. The circadian pattern observed in a given total population reflects the contributions of these subgroups.