Circadian variation of acute st segment elevation myocardial infarction by anatomic location in a Turkish cohort

Background

Circadian rhythms have long been known to occur in many biologic phenomena, including secretion of hormones and activities of the nervous system. Circadian rhythms have also been noted to occur in the incidences of certain cardiac and cerebrovascular events, including transient myocardial ischemia, myocardial infarction (MI), sudden cardiac death, and stroke [1].

Acute myocardial infarction (AMI) is one of the leading cause of mortality and mortality worldwide. Heart attacks are more common in the morning hours than at any other time of day. A circadian rhythm with a peak in the early morning has been reported for the frequency of myocardial infarction [2], giving rise to the idea that circadian variation plays a prominent role in the genesis of AMI. This role is commonly attributed to increased sympathetic activation as a result of physiological changes and biochemical parameters [3]. The changes in blood pressure, blood fibrinolytic activity, platelet aggregability, endothelial function and coronary tone during the daily cycle may increase the risk of heart attack, especially in the early morning hours [4–9]. As a result of increased sympathetic activation and changes in hemostatis, atherosclerotic plaque might be disrupted and thrombosis may occur. Autonomic nervous system activity may also play a role [3].

The occurrence of variations in the spectrum of cardiovascular diseases between different regions of the world and ethnic groups is well known [10–12]. Differences in the circadian variation of AMI have also been reported in different regions of the world and in different ethnic groups [10,11,13]. Although several studies have shown the association between acute coronary syndromes and circadian rhythm, studies investigating the association between circadian rhythm and coronary artery involvement in the genesis of AMI are limited [3]. Thus, we sought to evaluate the relationship between site of infarction (anterior vs. inferior) and beginning time of chest pain in patients with ST segment elevation myocardial infarction (STEMI) in a Turkish cohort.

Material and Methods

PATIENTS:

This restrospective study enrolled 465 patients (407 male, mean age 65±7 years) with acute ST elevation MI (STEMI) between April 2002 and January 2007. Of 535 patients, timing of the onset of AMI could not be definitely identified in 70 patients, and therefore only the data obtained from 465 patients were analyzed. Of 465 patients, 72 (15%) underwent thrombolytic treatment and 430 (85%) underwent primary percutaneous coronary ntervention (pPCI). The diagnosis and the treatment of AMI was made according to the ACC/AHA/ESC Guidelines for the Management of Patients With ST-Elevation Myocardial Infarction [14]. Characteristics of patients by site of infarction (anterior vs. inferior) were compared. The study was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice (GCP)/International Conference on Harmonization (ICH) guidelines. The time of AMI onset was determined by the attending physician on the basis of patients’ self-reports. The time in which chest pain had started was considered as the beginning of AMI, within ±30 min. Patients experiencing MI without chest pain, those who could not exactly remember the time of onset of chest pain, patients with non-ST elevation MI and unstable angina pectoris, patients experiencing MI after coronary bypass grafting or any invasive cardiac procedures, and patients with occupations requiring work during various shifts (changing hours of work) were not included into the current study.

EVALUATION OF ST SEGMENT RESOLUTION:

Twelve-lead surface ECG (Hewlett Packard M 1700A, Houston, USA) was recorded from all patients before, and an average of 55 (± 12 min) after, primary PCI. The sum of ST elevation was assessed in 3 contiguous leads in the infarct zone, 60 miliseconds (ms) from the J point (anterior: leads I, aVL, V1–V6; nonanterior: leads II, III, aVF, V5–V6). The extent of ST segment resolution was assessed an average of 55 min (±12 min) after primary PCI and is expressed as a percentage of the ST elevation measured on the ECG recorded before primary PCI.

ECHOCARDIOGRAPHY:

Transthoracic echocardiography was performed by using EASOTE 2.5 Mhz probe (ESAOTE, Genova, Italy) at the left lateral decubitis position before primary PCI.

BLOOD CHEMISTRY:

Venous blood samplings were withdrawn into the tubes containing K₃ EDTA and the tubes containing no anticoagulant agent. After all tubes were spun at 5000 rpm for 15 minutes, plasma and serum samplings were stored at –80°C until analyses were made. For the measurement of lipid profile, fasting blood samples were withdrawn from the patients with AMI within 24 hours of admission. Total plasma cholesterol, triglyceride and high-density lipoprotein [HDL] cholesterol concentrations were measured with a spectrophotometric technique by the Olympus AU-2700 autoanalyzer using commercial kits (Olympus, Hamburg, Germany). Low-density lipoprotein [LDL] cholesterol levels were calculated by the Friedwald formula.

CIRCADIAN ANALYSIS:

The standard hourly profile of the onset of AMI was obtained over a 24-h period. Patients were then categorised into 4 6-h increments according to the time that the symptoms began (12:00 AM–06:00 AM, 06:00 AM–12:00 PM; 12:00 PM–06:00 PM and 06:00 PM–12:00 AM hours). The time of onset and the electrocardiographic site of acute myocardial infarction, previous history of coronary artery disease (CAD), age, coronary risk factors (smoking, diabetes mellitus, hypertension, hypercholesterolemia, family history of CAD), coronary culprit lesion, prevalence of male sex, alcohol consumption, marital status, aspirin usage, beta-blocker usage and employment status were recorded.

STATISTICAL ANALYSIS:

Statistical analysis was performed using the SPSS 15.0 Statistical Package Program for Windows (SPSS Inc., Chicago, Illinois, USA). We used One-Sample Kolmogorov-Smirnov and Levene tests to determine the distribution characteristics of variables and variance homogeneity. Results are expressed as the mean ±SD, median and percentages. The means of groups were compared with each other. The differences between groups were tested by chi-square, independent samples t test and Mann-Whitney U tests. Differences were considered significant at p<0.05.

Results

Time of onset of symptoms of myocardial ınfarction

ANTERIOR MYOCARDIAL INFARCTION: The frequency of onset of acute anterior MI as determined by onset of pain demonstrated significant circadian variation among the 4 time periods, demonstrating bimodal peaks (afternoon and morning) and a trough between 06:00 PM to 06:00 AM (Table 4, Figure 1). The onset of MI was between 12:00 PM and 06:00 PM hours in 119 patients (54.1%); between 06:00 AM and 12:00 PM hours in 79 patients (36%); between 12:00 AM and 06:00 AM hours in 13 patients (5.9%) and between 06:00 PM and 12:00 AM hours in 9 patients (4.1%). The incidences of MI between 06:00 AM and 06:00 PM hours were significantly higher when compared with other 2 6-h periods (p<0.001). The incidence of occurrence of MI between 06:00 AM to 06:00 PM hours was 4.50 times that of the average frequency of the remaining 12 hours of the day.

INFERIOR MYOCARDIAL INFARCTION: The frequency of onset of acute inferior MI as determined by onset of pain exhibited significant circadian variation among the 4 time periods, demonstrating bimodal peaks (midnight to 06:00 AM and 06:00 AM to noon) and a trough between noon to midnight (Table 4; Figure 2). The onset of MI was between 12:00 AM and 06:00 AM hours in 85 patients (38.3%); 06:00 AM and 12:00 PM hours in 103 patients (46.5%); between 12:00 PM and 06:00 PM hours in 24 patients (10.7%) and between 06:00 PM and 12:00 AM hours in 11 patients (4.5%). The incidences of MI between 12:00 AM and 12:00 PM hours were significantly higher when compared with the 2 other 6-h periods (p<0.001). The incidence of occurrence of MI between midnight to noon was 4.25 times that of the average frequency of the remaining 12 hours of the day.

Discussion

The current study showed different circadian periodicity in patients with acute anterior MI compared to that of inferior MI in a Turkish cohort. Compared with other times of the day, anterior MIs were found to be more common from noon to 06:00 PM, whereas inferior MIs were found to be more common from midnight to noon in our patient population.

Studies on the circadian rhythm of AMI onset have suggested that the morning peak of incidence is related to other known daily rhythms [2,5,15]. One obvious explanation is the increase in physical and mental stress that occurs after waking. An increase in sympathetic activity occurs after waking and a rise occurs in plasma levels of catecholamine and of cortisol, heart rate, blood pressure, coronary vascular tone, and platelet aggregability [7,16,17]. The reasons for a second, late evening peak are less obvious. The evening peak may reflect altered circadian rhythms due to late working hours or reflect that some individuals are “night people” rather than “morning people” [2]. Some investigators have postulated that the morning and evening peaks are related to sleep-mediated factors, with changes in autonomic regulation, as a bimodal rhythm for the tendency to fall asleep has been described [18].

Circadian variation of the onset of AMI has been noted in many studies and may carry important pathophysiologic implications. A circadian variation in the onset of cardiac events and stroke is well recognised, with prominent clustering of MI events between 06:00 AM and 12:00 PM hours in the studies came from Western populations [2,19]. However, Lopez et al demonstrated a different circadian rhythm in MI onset between ethnic groups, with a significantly higher number of acute MI onsets occurring between midnight and noon in British Caucasians and Indo-Asians [20]. In contrast, Mediterranean Caucasians showed the opposite circadian pattern, with most acute MI events happening between noon and midnight. Although a morning peak of MI onset is very common, some populations studied have demostrated different results. For example, a maximal peak of onset of MI between 04:00 PM and 12:00 AM hours, with a second smaller peak between 06:00 AM and 08:00 AM hours, was found in a Bulgarian cohort [21], and a peak between 01:00 AM and 07:00 AM hours and a trough between 01:00 PM and 07:00 PM hours was demonstrated in a Chinese cohort [22]. On the other hand, a double-peak distribution of MI was reported in a South American patient population, with a lower MI incidence between 03:00 AM and 07:00 AM hours, a first maximum between 08:00 AM and 12:00 PM hours, and a second maximum between 03:00 PM and 10:00 PM hours [23]. The data on circadian variation of acute myocardial infarction in the Turkish population came from the very well-designed study by Sari et al, who examined circadian variation of AMI in a Turkish cohort [24]. Although the authors showed afternoon predominance in circadian variation of AMI in the Turkish cohort, they did not examine circadian distribution of AMI by anatomic location and coronary artery involvement.

Environmental-genetic background, socio-economic factors and customs could possibly underlie ethnic disparities in cardiovascular risk factor profiles [25,26]. However, ethnicity is not only one’s genetic background, but also includes social and cultural habits that might alter the cardiovascular disease spectrum in different populations. Ethnic disparities may be partly responsible for cardiovascular risk prevalence.

The precise mechanisms for these circadian differences remains uncertain. Ethnicity and lifestyle may also be responsible in the discrepancy in circadian variation of MI among different populations. Differences in eating habits, daytime rhythm, sunshine hours, sleep patterns and disparities in cardiovascular risk prevalence might also contribute to this discrepancy. Although controversial, eating a large meal is considered by some to be a trigger of MI [27–29]. There are variations in which the main meal of the day is breakfast in some populations, but lunch or supper is the largest meal in others. Although we did not clearly document that lunch was the main meal of our study population, afternoon peak in patients with anterior MI can be partly explained by eating a large lunch. However, we were not able to explain the morning peak in patients with inferior MI in the the current study population.

The current study had several limitations. First, this study is a single-centre study with a relatively small sample size, making it difficult to generalise the results to the entire Turkish population, and needs to be clarified with large-scale multicentre studies. Secondly, data were gathered retrospectively, and we could not determine whether patients were awake or asleep at the time of symptom onset. We also could not distinguish REM and non-REM sleep. Thirdly, we did not fully evaluate the socioeconomic status, physical activity level, eating and sleep habits that might affect the circadian distribution. Fourthly, we did not perform subgroup analysis. Presence of DM, beta blockers and aspirin usage have been reported to blunt the circadian variation of MI, although the current study sample showed circadian variation in both study groups regarding those clinical charecteristics. More importantly, DM frequency, aspirin and beta blocker usage were not statistically different between the patients with anterior and inferior MI. Lastly, a potential weakness of our study is its subjective reporting of the exact time of onset of symptoms, although we believe that our data are as accurate as such retrospectively gathered information can be.

Conclusions

In conclusion, different circadian periodicity in the time of onset of ST elevation MI was found regarding infarction site in a Turkish cohort. The present study suggests that circadian variation of AMI with morning peak might not be the pattern throughout the world; instead, one can refer to ‘population rhythm’. In fact, the factors known to affect circadian rhythm of MI may not be equally effective in all populations. Further large-scale clinical studies are needed to analyze the underlying pathophysiological mechanisms causing these differences in the chronobiology of MI.