This case-control prospective follow-up study aimed to evaluate sleep and circadian rhythm in patients with acute MI and their correlations with prognosis. Patients who met the inclusion criteria without the exclusion criteria were recruited from the intensive care unit (ICU) following admission. They received standard MI treatments and light therapy during their ICU stay, alongside assessments by a team of cardiologists and psychiatrists experienced in sleep medicine during hospitalization.

The inclusion criteria were as follows: (1) age between 35 and 85 years, (2) diagnosis of acute MI (Killip I-III), and (3) agreement to participate in this study and provide informed consent. The exclusion criteria were as follows: (1) unclear consciousness, (2) use of sedative or hypnotic drugs, (3) blindness or severe cataracts, (4) severe neurological diseases (eg, seizure, brain injury, or severe organic brain diseases), (5) severe psychiatric disorders (eg, schizophrenia, bipolar disorder, mental retardation, or substance use disorder), (6) unwillingness to participate, and (7) high mortality risk.

An age- and gender-matched healthy control group was also recruited for comparison.

Upon admission to the ICU, patients were screened and invited if they met the inclusion criteria and were not excluded based on the criteria. Physical evaluation and the diagnosis and treatments of acute MI were conducted by experienced cardiologists, while mental status evaluation was performed by psychiatrists to rule out severe psychiatric disorders. One-hour light therapy was administered each morning during ICU stay to prevent delirium. Baseline evaluation included age, gender, BMI, and blood pressure. Participants also underwent blood sampling for tests, including blood routine tests (white blood cell count, red blood cell count, hemoglobin, hematocrit, mean corpuscular volume, mean corpuscular hemoglobin, mean corpuscular hemoglobin concentration, platelet, platelet distribution width, and mean platelet volume), renal and liver function tests (blood urea nitrogen, creatinine, alanine aminotransferase, and aspartate aminotransferase), electrolyte tests (sodium and potassium), thyroid function tests (thyroid-stimulating hormone and free T4), and vitamin D tests. Actigraphy was used to evaluate sleep and circadian rhythm, while HRV was assessed within 24 hours of ICU admission to evaluate autonomic nerve function (Figure 1). Participants’ conditions were monitored throughout their ICU stay and subsequent general ward stay. Two delirium screening scales (the Confusion Assessment Method for the ICU and the Intensive Care Delirium Screening Checklist) were administered by psychiatrists daily to screen delirium during ICU stay and every 2 days in the general ward [24,25], and only 1 participant had delirium during the study period. Prognosis was evaluated with follow-ups at 3 and 6 months after discharge, and data on prognosis, including the duration of current admission, the number and duration of readmission and ICU admission, and catheterization during the 6-month follow-up period, were collected and confirmed via phone contact and by medical records (Figure 1).

Actigraphy, a nonintrusive approach involving a wristwatch-like device, can monitor participants’ motions and light exposures over extended periods [26]. Data analysis by the widely used parametric method provided sleep parameters, including sleep onset latency, total sleep time, sleep efficiency, wake after sleep onset, and times of awakening, and all 34 participants were included in the analysis. The estimation of these parameters is based on the observation that there is less motion during sleep than during wakefulness [27]. Participants wore the device from enrollment until discharge or withdrawal from the study, which could occur if they met exclusion criteria such as hypnotic use or delirium. Healthy controls wore the device for 2 weeks. To analyze circadian rhythm, participants with actigraphy recording of less than 3 days were excluded. Consequently, 28 of the 34 patients with acute MI and 17 healthy controls were included in the circadian rhythm analysis (Figure 2). Representative actograms for the MI and control groups are shown in Figures 3A and 3B, respectively.

Since the circadian rhythms of patients with MI can be disrupted and are not easily modeled by a 24-hour sinusoidal wave, we calculated several nonparametric indicators based on the hourly averaged activity from minute-by-minute actigraphy epochs. These indicators are designed to quantify the active-rest patterns and include interdaily stability, intradaily variability, the most active 10 consecutive hours (M10), the least active 5 consecutive hours (L5), the relative amplitude, and the actigraphic dichotomy index (I<O). Detailed mathematical formulas can be found elsewhere [20,28-30]. The interdaily stability and intradaily variability values quantify the stability and fragmentation of the rest-active patterns, respectively, and lower interdaily stability and higher intradaily variability indicate a disrupted circadian rhythm. While M10 and L5 identify the peak and trough of daily activity intensities, relative amplitude quantifies the daily circadian activity amplitude by calculating the difference between M10 and L5, normalized to their means. The midpoints of M10 and L5 represent the central times of these periods. The average daily values of M10, L5, and relative amplitude, and the midpoints of M10 and L5 were used to represent the circadian activity profiles of each participant (Multimedia Appendix 1). The actigraphic dichotomy index (I<O) measures the percentage of in-bed activity counts that fall below the median of out-of-bed activity counts. A lower index suggests a weaker circadian rest-activity rhythm [31].

HRV, reflecting autonomic nervous system tone, was monitored for 24 hours in patients with acute MI after enrollment. Healthy controls underwent HRV assessment for 15 minutes. It has a well-established role as a marker of cardiovascular risk [32]. Previous studies have found that HRV can predict the mortality rate of patients with acute MI [33-35]. Six patients with a history of arrhythmia were excluded, and 22 patients with acute MI and 16 healthy controls were included in the HRV analysis (Figure 4).

Data were analyzed using SPSS version 20.0 (IBM Corp). Variables were presented by mean and percentage. Independent sample t tests and analysis of covariance (ANCOVA) were used to compare differences between the MI group and healthy control group, and age, gender, and BMI were controlled. Pearson correlation tests measured correlations between the parameters of actigraphy and HRV and the prognosis of patients with MI, and partial correlation coefficient (r) analysis was used to adjust age and gender. A P value of <.05 was considered statistically significant, while a P value of <.01 was considered highly statistically significant.

The study was conducted in accordance with the Declaration of Helsinki and was approved by the Institutional Review Board of Chang Gung Memorial Hospital (201902108A3). Before enrollment, participants were fully informed about the purpose, voluntary participation, and confidentiality of the study, and all signed an informed consent form. Data were anonymized and deidentified. Participants were not offered any form of compensation for their involvement.

Previous studies have used actigraphy to evaluate activities and sleep in patients with MI [23,36,37]. However, to our knowledge, no study has examined the circadian rhythm of patients after the onset of acute MI using actigraphy and analyzed its correlations with prognosis. Therefore, our findings are significant as they not only reveal sleep and circadian rhythm disturbances after acute MI but also support the application of actigraphy and HRV in evaluating the prognosis.

Unsurprisingly, baseline laboratory work-up assessments showed significant differences between patients with MI and healthy controls. Patients with MI exhibited more obesity and more abnormalities, including a lower vitamin D level, which is known to be negatively associated with the incidence of cardiovascular diseases [38,39]. This study further revealed significant differences in sleep patterns between the 2 groups. After acute MI, patients had significantly worse sleep, characterized by more wake after sleep onset and number of awakenings, and lower sleep efficiency. These findings align with a previous study, which has reported sleep disturbances in patients after acute MI, including more wakefulness and awakening, more stage shifts, less rapid eye movement, decreased sleep efficiency, and increased daytime sleep [40]. These changes can be attributed to MI itself rather than differences in a hospital setting [40] and can be normalized with an improved physical condition. Sleep has been proposed as a vital sign and a marker for disease and recovery [41], and thus, routine monitoring and evaluation of sleep are recommended in patients with MI.

The circadian rhythm analysis using actigraphy data supports that both sleep quality and circadian rhythm can be disturbed in patients with MI. Compared with controls, the activities of patients with MI were significantly less during the daytime, while nighttime activities remained unchanged. Moreover, there was less variability (SD) in both daytime and nighttime activities, indicating less change in activity levels throughout the day. Although these findings may be linked to the bed rest and activity restrictions commonly advised for patients with acute MI, the higher intradaily variability contradicts this hypothesis and instead suggests greater fluctuations in hourly activities during both the daytime and nighttime. Furthermore, a lower relative amplitude and actigraphic dichotomy index suggested less distinct activity changes between sleep and wakefulness and a weak circadian rest-activity rhythm. Patients with MI also exhibited a significant delay in their circadian rhythm, as indicated by a higher time of M10.

Most studies on circadian rhythm and MI have focused on the circadian variances of MI [42-44], which can be explained by the fact that circadian rhythm–related coronary vasoconstriction and increased sympathetic tone may disrupt atherosclerotic plaque [45]. In this study, 5 out of 34 participants (18%) experienced initial symptoms within 1 hour after awakening, but we did not find any correlation between their onset time and prognosis. The nonsignificant result can be influenced by several factors. We only enrolled patients with Killip I to III and excluded those with severe comorbidities and a high mortality risk. Some participants could have recall bias due to their atypical clinical presentations. Circadian variances of MI and their correlations with prognosis necessitate further investigation.

Delayed and disturbed circadian rhythm has been found to be associated with cardiovascular diseases [46,47]. Consistent with our findings, a cross-sectional study also found a lower 24-hour rest-activity rhythm amplitude and overall rhythmicity in patients with coronary artery disease, using actigraphy [48]. Although little is known about circadian rhythm after the onset of MI, our findings indicate that several parameters derived from the circadian rhythm analysis have significant correlations with prognosis, including the total days of current admission, ICU readmission, and the number and duration of readmissions during the 6-month follow-up period. Patients with a more delayed circadian rhythm (higher times of L5 and M10) were more likely to have a longer duration of current admission, more frequent readmissions, and a longer duration of readmission. Since gender and age can have impacts on both sleep and the prognosis of MI [49-51], we adjusted age and gender in the correlation analysis. Most of the significant correlations could still be found (Multimedia Appendix 4). Although a causal relationship cannot be confirmed, these findings are consistent with those of previous studies that suggest abnormal circadian rhythm as a core vulnerability factor for MI [15,52]. Interventions targeting the circadian rhythm of patients with MI, especially those with a disturbed and delayed sleep phase, may be beneficial. Exercise programs and rehabilitations have already been demonstrated to improve prognosis and reduce mortality in patients with MI [53], and they can also help improve sleep. During the acute phase, light therapy and environmental modifications may aid in the regulation of circadian rhythm and are worthy of further investigation.

We found that patients with MI had lower parasympathetic activities compared to healthy controls, and there were correlations between several HRV parameters and prognosis. HRV is an indirect measure of autonomic nervous activity [29] and has been investigated for its ability to predict prognosis in patients with acute MI. Previous studies have reported associations between HRV and mortality or cardiovascular events [54-57]. In this study, better parasympathetic nerve activity after acute MI could be correlated with a better prognosis. However, we also found positive correlations between the ratio of SD1 to SD2 (SD ratio) and prognosis, including recatheterization and the duration of readmission, contradicting other findings. A previous study found a correlation between a high SD ratio and arrhythmia such as ventricular tachycardia [58]. Thus, although a higher parasympathetic nerve activity can correlate with a better prognosis, the SD ratio may not always be “the higher, the better.”

Comparing correlations with the prognosis of MI, the circadian rhythm analysis using actigraphy showed more significant findings than the HRV analysis, especially in nighttime activity and the time of M10. Moreover, fewer significant correlations between HRV parameters and prognosis were found after adjusting for age and gender (Multimedia Appendix 5). These findings suggest that the evaluation of sleep and circadian rhythm after the onset of MI is a valuable prognosis indicator. Previous studies have demonstrated the potential of actigraphy to predict the outcomes of various diseases [19,20]. The application of actigraphy in the evaluation and treatment of patients with MI is promising, as it has already been suggested to play a role in the prognostication of heart failure [59]. Another study showed that the evaluation of subjective sleep complaints through questionnaires may serve as a prognostic marker in patients after acute MI [60], emphasizing the importance of sleep in the risk stratification of MI. Although our study found fewer HRV correlations with prognostic factors, it differed from past research as there was no morality after excluding patients with Killip IV. Furthermore, instead of analyzing mortality, our study focused on correlations with prognostic factors such as admission duration, readmission, and recatheterization. Given that only 22 cases were included in the HRV analysis, factors, including age, gender, and medications such as beta-blockers [61], might have affected the findings. Further research is required to compare the 2 evaluation tools in terms of their correlations with the prognosis of MI.

The study has some limitations. First, the study had a small sample size, which limits detailed subgroup analysis by age or gender. Second, we only evaluated participants after the onset of MI during their admission, and therefore, the observed delayed and disturbed circadian rhythm may not represent their premorbid sleep and circadian rhythm. They did not undergo actigraphy monitoring during follow-up, and thus, subsequent changes in sleep and circadian rhythm were not known. Further studies following MI patients with actigraphy after discharge can increase the understanding of the long-term changes in sleep and circadian rhythm after MI and their impacts on prognosis. Third, patients in our study received light therapy during their stay in the ICU to prevent delirium. However, the duration of light therapy was short (1 hour per day for only 1-2 days), limiting its potential impacts. Fourth, some patients with MI were excluded, such as those classified as Killip IV. Therefore, caution is required when generalizing our findings, especially to severe cases with high mortality rates. Lastly, comorbidities that can contribute to sleep disturbance were not controlled. Sleep disorders, such as obstructive sleep apnea, cannot be ruled out, as polysomnography is challenging to perform in patients with acute MI in the ICU setting. Nevertheless, our findings not only support but also extend the use of actigraphy beyond its conventional role in sleep medicine to include applications within cardiology.

Patients with acute MI experienced significantly worse sleep and disturbed circadian rhythm compared with healthy controls. Actigraphy-based analysis revealed a disturbed circadian rhythm, including reduced daytime activities, a greater fluctuation in hourly activities, and a weak rest-activity rhythm, which were correlated with prognosis. The evaluation of sleep and circadian rhythm in patients with acute MI can serve as a valuable indicator for prognosis and warrants further investigation.