Obstructive Sleep Apnea Increases the Risk of Perioperative Myocardial Infarction Following Off-Pump Coronary Artery Bypass Grafting

Background and Aims: The impact of obstructive sleep apnea (OSA) on perioperative myocardial infarction (PMI) following coronary artery bypass grafting (CABG) remains unclear. Off-pump CABG (OPCABG) has become a common practice for CABG in China. The present study investigated mainly the correlation between OSA and PMI following OPCABG. Methods: In this prospective observational single-center study, consecutive eligible patients listed for elective OPCABG underwent cardiorespiratory polygraphy before surgery between January 2019 and June 2020. OSA was defined as an apnea-hypopnea index (AHI) ≥15 events/h. The primary end point was perioperative myocardial infarction (PMI) following OPCABG (type 5 MI). Results: Patients with OSA accounted for 42.2% (62/147) of the cohort. Twenty-four patients (16.3%) met the protocol criteria for PMI: 17 (27.4%) in the OSA group and 7 (8.2%) in the non-OSA group (P = 0.002). Multivariate logistic regression analysis revealed that AHI (OR = 1.115, 95% CI 1.066 to 1.166, P < 0.001), high-sensitivity c-reactive protein (hs-CRP) (OR = 1.080, 95% CI 1.025 to 1.138, P = 0.004), and SYNTAX score (OR = 1.098, 95% CI 1.056 to 1.141, P < 0.001) were associated with PMI incidence. Furthermore, ROC analysis revealed that the AHI (AUC = 0.766, 95% CI 0.689 to 0.832, P < 0.001) and SYNTAX score (AUC = 0.789, 95% CI 0.715 to 0.852, P < 0.001) had predictive value for PMI. In addition, multiple linear regression analysis showed that the AHI was an independent influencing factor of hs-CRP (B = 0.176, 95% CI 0.090 to 0.263, P < 0.001) and the SYNTAX score (B = 0.553, 95% CI 0.397 to 0.709, P < 0.001). Conclusions: OSA was independently associated with a higher incidence of PMI following OPCABG, and the formation of severe coronary atherosclerotic lesions aggravated by an enhanced inflammatory response might be the potential mechanism. The efficacy of CPAP treatment for improving prognosis after CABG remains to be further investigated.


INTRODUCTION
Coronary artery bypass grafting (CABG) is the standard of care for patients with extensive coronary artery disease (CAD). Techniques to perform CABG without cardiopulmonary bypass (off-pump CABG, OPCABG) have emerged to reduce complications associated with cardiopulmonary bypass and manipulation of the aorta, especially for higher-risk patients who could benefit from short-term prognosis improvement (1). Despite decades of refinement, perioperative myocardial infarction (PMI) following CABG, with an incidence of 1.9-30%, remains a major predictor of increased cardiovascular mortality and healthcare resource utilization (2)(3)(4). Therefore, it is necessary to identify high-risk patients with PMI and initiate targeted treatment in advance.
Obstructive sleep apnea (OSA), characterized by recurrent episodes of partial or complete collapse of the upper airway during sleep, is highly prevalent in patients with CAD (5). OSA has been proven to be closely related to the degree of coronary atherosclerosis and myocardial infarction through activation of proinflammatory responses, oxidative stress, sympathetic excitation, platelets, or metabolic dysregulation (6). Compared with the general population, OSA, diagnosed by an apneahypopnea index (AHI) ≥15, has a significantly higher prevalence of ∼50% in CABG patients (7). Considerable evidence has confirmed that OSA is strongly correlated with increased risks of new revascularization, typical angina, atrial fibrillation and prolonged length of hospital stay after CABG (7,8); however, the impact of OSA on PMI following CABG is still unclear. Therefore, we investigated mainly the correlation between OSA and PMI following OPCABG (type 5 MI), which was defined by the fourth universal definition of MI (9).

Study Design
This prospective observational single-center study was conducted in Beijing Anzhen Hospital, Capital Medical University. We enrolled consecutive patients with coronary artery disease who were scheduled to undergo elective OPCABG and had not been diagnosed with OSA from January 2019 to June 2020. The study protocol, which was approved by the institutional review board (Approval No.: 2013025), was explained to all patients, all of whom signed informed consent.

Patients
All participating patients, who were between 40 and 75 years old, underwent an overnight sleep study before OPCABG. The exclusion criteria were as follows: preoperative treatment with an intra-aortic balloon pump (IABP), pulmonary insufficiency (e.g., chronic obstructive pulmonary disease), central sleep apnea (≥50% central events or central AHI ≥10/h), body temperature >37.5 • C, combination with other surgeries, onpump CABG, and sleep study failure (patients without adequate signal recording).

Overnight Sleep Study
Overnight sleep studies were performed after clinical stabilization (within 1 week after admission) by portable cardiorespiratory polygraphy (ApneaLink, Resmed, Australia), which might provide high diagnostic accuracy and offer an accessible alternative to full polysomnography. The reported sensitivity and specificity were 0.88 and 0.84, respectively (10). Nasal airflow, arterial oxygen saturation, thoracoabdominal movements, and snoring episodes were recorded. Sleep studies were scored according to the American Academy of Sleep Medicine 2007 guidelines (11). Apnea was defined as an absence of airflow lasting ≥10 s (obstructive if thoracoabdominal movement was present and central if thoracoabdominal movement was absent). Hypopnea was defined as a reduction in airflow of >30% for ≥10 s and associated with a decrease in arterial oxygen saturation >4%. AHI was defined as the number of apneas and hypopneas per hour of total recording time. The oxygen desaturation index (ODI) was the number of times that oxygen saturation decreased by >3% per hour. Recruited patients were categorized into OSA (AHI ≥ 15 events/h) and non-OSA (AHI < 15 events/h) groups. A satisfactory polygraphy signal recorded for a minimum of 3 h was considered a valid test. All studies were double-scored manually by independent sleep technologists. Further, scoring was performed in cases of discrepancy by a senior consultant in sleep medicine. All OSA patients were recommended for further evaluation and consideration of continuous positive airway pressure (CPAP) therapy after discharge.

OPCABG
Surgical indications were in accordance with the guidelines (12). The same experienced cardiac surgeon (Y.Y.) performed all surgeries. The quality of graft anastomosis met the criteria recommended by the Operation Quality Committee of Beijing Anzhen Hospital. Measurement of graft flow and pulsatility index (PI) was performed by placing a handheld flow probe (MediStim, Oslo, Norway) around the graft. After surgery, the patients were monitored in the intensive care unit (ICU) until ventilator removal, and vital signs were stable. Perioperative management was conducted according to the guidelines (13). Within 48 h after OPCABG, myocardial markers were routinely tested every 6 h, and electrocardiogram (ECG) and echocardiography were performed every 12 h.

Outcome
The primary end point was CABG-related MI ≦48 h after the operation (type 5 MI), also known as perioperative myocardial infarction (PMI), defined by the fourth universal definition of MI [cardiac troponin (cTn) >10 times the 99th percentile upper reference limit or a rise of cTn values >20% if the baseline values were elevated or falling, in addition to ECG changes, angiographic findings or new regional wall motion abnormalities] (9). All relevant data were independently judged by two experienced doctors (L.H. L, D.X.H.), who were blinded to sleep study results.

Statistical Analyses
For continuous variables, a normal distribution was assessed using the Kolmogorov-Smirnov test. Continuous variables are shown as the mean ± SD or median (first and third quartiles) and were compared by using Student's t-test or the Mann-Whitney U-test. Categorical variables are presented as percentages and were compared using the chi-square test. To analyze the association between PMI and other variables, multivariate logistic regression analysis was implemented (demographic characteristics, cardiovascular risk factors and variables with p < 0.1 in the univariate logistic regression were included in the model). Additionally, we assessed the predictive value of AHI, SYNTAX score and high-sensitivity c-reactive protein (hs-CRP) for PMI through receiver operating characteristic (ROC) curve analysis. The predictive values were evaluated with the area under the curve (AUC), and the difference in the AUC with statistical significance was assessed with the DeLong test. Furthermore, the correlation between AHI and hs-CRP or SYNTAX score was determined by multivariate linear regression analysis (the model included demographic characteristics, cardiovascular risk factors and variables with p < 0.1 in the univariate linear regression). All analyses were conducted with SPSS 24.0 (SPSS., Chicago, IL, USA) and MedCalc V.11.4 (MedCalc, Inc., Ostend, Belgium) assuming a 2-sided test with a 5% level of significance.

Basic Characteristics and Overnight Sleep Study Results
A total of 215 consecutive eligible patients scheduled for OPCABG were prospectively enrolled. After exclusion of 68 patients, 147 patients were included (Figure 1). The mean age of the patients recruited was 62.0 ± 8.9 years, and 74.8% were male. Patients with OSA accounted for 42.2% of the cohort 3), and they had longer mean and maximum apnea times, higher oxygen desaturation indices, and lower lowest and mean arterial oxygen saturations than those without OSA ( Table 2).

Angiographic and Surgical Characteristics
Angiographic and OPCABG surgical characteristics are shown in Table 3. The mean SYNTAX score of all patients was 52.6 ± 15.6, and in the OSA group, it was significantly higher than that in the non-OSA group (P < 0.001). The number of target vessels, distal anastomoses/patient, graft type, and assessment of grafts did not differ between the OSA and non-OSA groups.

DISCUSSION
To the best of our knowledge, this is the first study to evaluate the impact of OSA on PMI following OPCABG. In this prospective cohort of 147 patients who underwent OPCABG, the prevalence of OSA was 42.2%, and compared with in non-OSA patients, PMI was more likely to occur in patients with OSA. AHI, hs-CRP, and SYNTAX score were independent risk factors for PMI. In regard to identifying patients at high risk of having PMI, the AHI, and SYNTAX score could provide a superior discriminatory ability to that of hs-CRP. In addition, AHI was associated with changes in hs-CRP and SYNTAX score. Along with the in-depth understanding of the pathogenesis and prognostic significance of PMI following CABG, different definitions have emerged, and the incidence of PMI ranges from 1.9 to 30% depending on different diagnostic criteria and patient populations (2, 3). The Fourth Universal Definition of Myocardial Infarction (2018) (9), applied in this study, remains the accepted standard in contemporary clinical practice. In addition, the diagnosis of PMI in several large randomized controlled trials was based on the creatine kinase-myocardial band (CK-MB) (14), but recently, the ESC Study Group on Cardiac Biomarkers of the Association for Acute Cardiovascular Care suggested that CK-MB be eliminated from the menu of biomarkers available since it was less sensitive and clinically instructive than cTn (15). We diagnosed patients based on cTn combined with other presentations and found that PMI following OPCABG occurred in 16.3% of patients with severe coronary lesions (mean SYNTAX score was 52. 6).

An analysis from the EXCEL (Evaluation of XIENCE vs. Coronary Artery Bypass Surgery for Effectiveness of Left Main
Revascularization) trial showed that PMI was associated with cardiovascular death and all-cause death at 3 years (16). In a Korean registry of 3,183 patients who underwent CABG, the presence of PMI was associated with increased risks of major cardiovascular events at the 5-year follow-up regardless of the PMI definition (2). PMI can be ascribed to multiple causes. In the EXCEL trial, age, chronic obstructive pulmonary disease, and aortic cross-clamp duration were independent predictors of PMI following CABG (16). Other well-established risk factors for PMI include left main coronary artery disease, recent MI, unstable angina, emergent operations, poor target-artery quality, inadequate revascularization, and prolonged bypass duration (17). However, few studies have focused on OPCABG. Our findings extend those of previous studies by proposing that OSA increases the risk of PMI in patients who underwent  OPCABG. Of note, we found no significant correlation between the incidence of PMI and the measurement index of graft (flow and PI), the reason for this result might be that all indices met the quality control standard (flow>15/min and PI < 5.0) (18). OSA has been advocated as a new emerging contributor to the development of CAD by guidelines and expert consensus (19,20), and intravascular ultrasound has indicated that patients with OSA present a larger total atheroma volume in the coronary artery (21). The SYNTAX (SYNergy between PCI with TAXUS and Cardiac Surgery) score was created, more than a decade ago, to provide an angiographic tool grading the complexity of coronary artery disease in patient with left main or multivessel disease, and it relates to CAD outcomes (22). Evidence has revealed that OSA significantly correlates with the SYNTAX score (23). Our study supported this finding and further demonstrated that both the AHI and SYNTAX score could identify patients at high risk for developing PMI following OPCABG; however, the accuracy is still moderate, and their performance should be further explored.
Intermittent hypoxia (IH), sleep fragmentation, and intrathoracic pressure swings in OSA appear to promote the development of cardiovascular disease, and inflammation is proposed as an intermediate mechanism linking OSA with the severity of coronary lesions (6). IH appears to facilitate atheromatous lesion formation by promoting systemic inflammation, partially via activation of the transcription factor NF-κB (24). The most studied biomarker in this context is hs-CRP, which has been demonstrated to participate actively in atherosclerosis via the induction and enhanced expression of adhesion molecules (25). Our findings were consistent with previous literature: hs-CRP levels were strongly associated with the severity of OSA. We added further evidence that the hs-CRP level was an independent risk factor for PMI following OPCABG; however, its limited predictive value for PMI might be due to its susceptibility to multiple interference factors.
In addition, evidence has confirmed that patients with OSA show reduced fibrinolytic capacity (26) and increased platelet activation and aggregation (27), all of which may contribute to PMI via thrombotic events of grafts or native vessels after CABG.
CPAP remains the standard of care for patients with OSA, and evidence has revealed that it is correlated with a reduced inflammatory response (28) and coagulability (29), as well as improvement in early signs of atherosclerosis (30), suggesting that CPAP may improve the prognosis of coronary revascularization. Wu et al. (31) reported that untreated moderate-to-severe OSA was independently correlated with a significantly increased risk of repeat revascularization after percutaneous coronary intervention (PCI). Cassar et al. (32) found that screening for and treating OSA (CPAP) in patients who underwent PCI might result in decreased cardiac death. However, we know little about the role of CPAP in reducing the incidence of complications after CABG. In consideration of our previous clinical practice, we speculate that patient compliance and timing of CPAP treatment after CABG may be the key to curative effects. Additional trials evaluating the effects of CPAP in patients with OSA who underwent CABG are warranted.
This study has several limitations. First, this was a singlecenter study with a limited sample size, and the results need to be validated through multicenter, large-sample studies. Follow-up of the cohort is needed. Second, compared with polysomnography, portable cardiorespiratory polygraphy may underestimate the severity of OSA, but it is safe and simple to identify OSA in a high-risk patient population (11). Finally, mechanical ventilation treatment after surgery might weaken the short-term effect of OSA to a certain extent.

CONCLUSIONS
This prospective cohort study demonstrated for the first time that OSA was independently associated with a higher incidence of PMI following OPCABG, and the formation of severe coronary atherosclerotic lesions aggravated by an enhanced inflammatory response might be the potential mechanism. The efficacy of CPAP treatment for improving prognosis after CABG remains to be further explored.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

ETHICS STATEMENT
This study was approved by the Institutional Review Board of Beijing Anzhen Hospital of Capital Medical University (Approval No: 2013025). The patients/participants provided their written informed consent to participate in this study.

AUTHOR CONTRIBUTIONS
KF, MG, and YY: study concept and design. KF, MG, WY, HL, LC, and XD: acquisition, analysis, or interpretation of data. KF and MG: drafting of the manuscript. WY and YY: obtained funding. All authors: critical revision of the manuscript for important intellectual content, read and approved the final manuscript.

FUNDING
This study was supported by the National Natural Science Foundation of China (81770343). The funds were used to acquire portable sleep monitors for the study. The sponsor had no role in the design or conduct of this research.