Inﬂuence of Circadian Rhythm on Mortality After Myocardial Infarction: Data From a Prospective Cohort of Emergency Calls ROBERTO MANFREDINI, MD,* BENEDETTA BOARI, MD,* SABRINA BRESSAN, MD,‡ MASSIMO GALLERANI, MD,* RAFFAELLA SALMI, MD,§ FRANCESCO PORTALUPPI, MD,† AND RAJENDRA H. MEHTA, MD, MS¶ Myocardial infarction (MI) occurs more frequently in the morning as a result of the concomitant unfavorable timing of several physiological parameters and/or biochemical conditions. However, little is known about the possible inﬂuence of this circadian pattern on prognosis. To evaluate whether the time of symptom onset could potentially inﬂuence mortality from acute MI, this prospective study considered all consecu- tive MIs admitted to the ED of Ferrara, Italy, after a call to the Emergency Coordinating Unit from January 1, 1998, to December 31, 2001. The total sample consisted of 442 MIs (mean age, 68.7 years; males, 72%). Eighty patients (males, 82.5%) died in the ED; the remaining 362 were admitted to the hospital. Of these, 50 (males, 60%) died during their hospital stay. Based on the timing of their symptom onset, cases were categorized both into 24 1-hour intervals and four 6-hour intervals (midnight to 5:59 AM, 6:00 AM to 11:59 AM, noon to 5:59 PM, and 6:00 PM to 11:59 PM), and the circadian distributions of fatal versus nonfatal MIs were compared. The circadian variation of MI peaked between 6:00 AM and noon (P < .001), and in this period, there was a trend toward a higher frequency of fatal cases (41.5% vs. 35.2%;  2  1.911, P  .167). To verify whether this higher frequency of fatal events in the morning hours could be related to possible higher severity of cases observed in that hours, a further sep- arate analysis considering age, infarct site, and peak levels of MB was made. Again, no signiﬁcant temporal differences among the four 6-hour intervals were found between fatal and nonfatal Mis, although a trend toward older age was observed in morning MIs. Not only the frequency, but also the mortality, of acute MI could be increased in the morning hours. This could be of practical interest for emergency doctors and could have signiﬁcant implications for acute treatment, because several studies have reported a lowered efﬁcacy of thrombolytic drugs in the morning hours. (Am J Emerg Med 2004;22:555-559. © 2004 Elsevier Inc. All rights reserved.) Many acute cardiovascular events, for example, myocar- dial infarction (MI), sudden cardiac death, cardiac arrest, pulmonary thromboembolism, rupture of aortic aneurysms, ischemic and hemorrhagic stroke, 1-11 show a circadian pat- tern of occurrence with a peak in the morning hours. It has been estimated that approximately one of every 11 MIs and one of every 15 sudden cardiac deaths are attributable to this morning excess in onset. 12 A synergistic effect in triggering of the morning occurrence is played by the concomitant increases in blood pressure, 13 heart rate, 14 sympathetic ac- tivity, 15 and basal vascular tone, 16 together with an imbal- ance between coagulation and ﬁbrinolysis characterized by an increased platelet aggregability 17,18 and a reduced endog- enous ﬁbrinolysis. 19 However, little is known about the possible inﬂuence of this circadian pattern on prognosis. The purpose of this study was to evaluate whether the time of onset of symptoms could inﬂuence mortality from acute MI. METHODS Study Design and Sample Ferrara is a town in northeastern Italy, with a population of approximately 150,000 and demographic and socioeco- nomic characteristics similar to those of Italy as a whole. One single hospital (St. Anna General Hospital) serves the entire community. The population is almost exclusively white. The organization of both the local emergency service and the Emergency Coordinating Unit (ECU) is reported elsewhere. 20 All the emergency calls are addressed to the phone number 118 (the equivalent of 911 in the United States). Ambulances are strategically distributed throughout the territory with an average time of arrival at the ED of 15 minutes or less from the time the call is received by the ECU where a highly trained nurse identiﬁes and classiﬁes the level of severity, then decides the type and the starting point of the intervention needed. In collaboration with the hospital departments participating in the Myocardial Infarc- tion Study on Timing and Outcome (MISTO) (see the “Acknowledgments” section), we prospectively analyzed all acute cardiologic cases referred to the ED of St. Anna General Hospital, Ferrara, Italy, after an emergency call from January 1, 1998, to December 31, 2001. All cases were (1) discharged after exclusion of acute coronary event or other acute disease suitable of hospitalization, or (2) admit- ted to hospital departments. Moreover, (3) some patients died in the ED after an unsuccessful attempt of resuscita- tion. For this study, we considered only cases belonging to the (2) and (3) groups, in which the ﬁnal discharge diagnosis was an acute MI. The diagnosis was always made by an internist and/or a cardiologist on the basis of history and clinical signs, speciﬁc myocardial enzymes, electrocardio- From the Sections of *Internal Medicine and ‡Cardiology, and †Hypertension Center, Department of Clinical and Experimental Medicine, University of Ferrara, Ferrara, Italy; the §Hospital Depart- ment of Internal Medicine, St. Anna Hospital, Ferrara, Italy; and the ¶Division of Cardiology, University of Michigan, Ann Arbor, Michi- gan. Manuscript received July 16, 2003, accepted October 6, 2003. Supported by a research grant (ex-60%) of the University of Ferrara. Address correspondence to Roberto Manfredini, MD, Section of Internal Medicine, Department of Clinical and Experimental Medi- cine, University of Ferrara, via Savonarola 9, I-44100 Ferrara, Italy. Email: mfr@unife.it Key Words: Myocardial infarction, circadian rhythm, prognosis, mortality. © 2004 Elsevier Inc. All rights reserved. 0735-6757/04/2207-0010$30.00/0 doi:10.1016/j.ajem.2004.08.014 555