Severe Obstructive Sleep Apnea Is Associated With Left Ventricular Diastolic Dysfunction

Original scientific paper Severe obstructive sleep apnea is associated with coronary microvascular dysfunction and obstruction in patients with ST-elevation myocardial infarction European Heart Journal: Acute Cardiovascular Care 0(0) 1–9 ! The European Society of Cardiology 2020 Article reuse guidelines: sagepub.com/journals-permissions DOI: 10.1177/2048872620919946 journals.sagepub.com/home/acc Yoshitaka Ichikawa1, Yasuhiro Izumiya1, Koichi Tamita2, Hiroya Hayashi1, Hirotoshi Ishikawa1, Atsushi Shibata1, Atsushi Yamamuro2 and Minoru Yoshiyama1 Abstract Background: Coronary microvascular dysfunction and obstruction (CMVO) is a strong predictor of a poor prognosis in patients with ST-segment elevation myocardial infarction (STEMI). Although research has suggested that obstructive sleep apnea (OSA) exacerbates CMVO after primary percutaneous coronary intervention, data supporting a correlation between OSA and CMVO are limited. This study was performed to investigate whether OSA is associated with CMVO, detected as microvascular obstruction on cardiovascular magnetic resonance images, in patients with STEMI. Methods: Patients (N ¼ 249) with a first STEMI underwent primary percutaneous coronary intervention. CMVO was evaluated on cardiovascular magnetic resonance images based on the presence of microvascular obstruction. OSA was classified into four levels of severity based on the respiratory event index (REI): absent (REI of <5), mild (REI of 5 to <15), moderate (REI of 15 to <30) and severe (REI of 30). Results: The REI was significantly higher in the presence of microvascular obstruction (n ¼ 139) than in its absence (n ¼ 110) (REI of 12.8 vs. 10.7, respectively; p ¼ 0.023). Microvascular obstruction was observed in 42%, 58%, 57% and 70% of patients in the absent, mild, moderate and severe OSA groups, respectively. Multiple logistic regression analysis showed that severe OSA was associated with increased odds of microvascular obstruction (odds ratio (OR), 5.10; 95% confidence interval (CI),1.61–16.2; p ¼ 0.006). Mild and moderate OSA were also associated with increased odds of microvascular obstruction (mild OSA: OR, 2.88; 95% CI, 1.19–7.00; p ¼ 0.019 and moderate OSA: OR, 3.79; 95% CI, 1.43–10.1; p ¼ 0.008). Conclusion: Severe OSA was associated with CMVO after primary percutaneous coronary intervention in patients with STEMI. Keywords Cardiovascular magnetic resonance imaging, coronary microvascular dysfunction and obstruction, microvascular obstruction, obstructive sleep apnea, ST-segment elevation myocardial infarction Date received: 21 January 2020; accepted: 29 March 2020 1 Department of Cardiovascular Medicine, Osaka City University Graduate School of Medicine, Japan 2 Department of Cardiovascular Medicine, Nishinomiya Watanabe Cardiovascular Center, Japan Corresponding author: Yasuhiro Izumiya, Department of Cardiovascular Medicine, Osaka City University Graduate School of Medicine, 1-4-3 Asahimachi, Abeno-ku, Osaka 545-8585, Japan. Email: izumiya.yasuhiro@med.osaka-cu.ac.jp 2 Introduction For patients with ST-elevation myocardial infarction (STEMI), immediate reperfusion via primary percutaneous coronary intervention (PCI) can salvage the myocardium and reduce mortality.1,2 However, the area of myocardial hypoperfusion persists in some patients despite prompt epicardial recanalization of the infarctrelated artery.3 This phenomenon, known as no reflow, is caused by coronary microvascular dysfunction and obstruction (CMVO), which has been shown to be associated with adverse ventricular remodeling and a poor prognosis after STEMI.2,4–7 CMVO can be detected as microvascular obstruction using cardiovascular magnetic resonance imaging (CMR).2,4–7 Microvascular obstruction is revealed by the lack of gadolinium enhancement within the hyper-enhanced infarcted area. It appears in two patterns, referred to as early and late microvascular obstruction. Although late microvascular obstruction is less sensitive than early microvascular obstruction and may lead to underestimation of the presence of CMVO, it is a strong predictor of clinical endpoints after the STEMI has been reperfused during primary PCI.6–9 The discovery of a crucial target for preventive therapy for CMVO may lead to an improved prognosis after STEMI; however, the mechanisms and causes of CMVO are complex and are not yet fully understood.4 Increasing evidence indicates that obstructive sleep apnea (OSA) is a significant risk factor for cardiovascular disease, including acute myocardial infarction, and is associated with increased morbidity and mortality.10,11 OSA is characterized by repetitive episodes of apnea or reduced inspiratory airflow due to upper airway obstruction during sleep. These events provoke intermittent hypoxemia and hypercapnia and are associated with hemodynamic alterations, oxidative stress, sympathetic hyperactivity, inflammatory response, endothelial dysfunction and hypercoagulability.12 These alterations are major contributors to cardiovascular diseases and may exacerbate CMVO. Although a previous study showed that OSA impairs CMVO in patients with STEMI after primary PCI,13 the association between OSA and CMVO remains controversial, partially because of the difficulty of evaluating CMVO. Because CMR enables precise detection of CMVO, we investigated the relation between OSA and CMVO determined by CMR in patients with acute STEMI. European Heart Journal: Acute Cardiovascular Care 0(0) with STEMI and who underwent primary PCI within 24 h after symptom onset. STEMI was diagnosed based on the presence of symptoms consistent with myocardial ischemia and signs of ST-segment elevation (measured at the J-point) of 1 mm in at least two contiguous leads on a 12-lead electrocardiogram. The exclusion criteria were death, previous myocardial infarction, unsuccessful PCI, heart failure and/or cardiogenic shock treated with intravenous inotropes and/or mechanical support, complications requiring surgery, other severe disease, lack of CMR, OSA that was already being treated with continuous positive airway pressure (CPAP) therapy, lack of sleep studies, and central sleep apnea. Lack of CMR was due to patients who had refused the test, had claustrophobia, or had a contraindication for contrast medium (hypersensitivity or severe kidney disease defined as an estimated glomerular filtration rate of <30 mL/min per 1.73 m2). Lack of sleep studies was due to patients who had refused the study or who had had an incomplete sleep study. Study design This prospective observational study was conducted in accordance with the amended Declaration of Helsinki and was approved by the Ethics Committee of Nishinomiya Watanabe Cardiovascular Center (approval number 2009-006). All patients provided written informed consent. Patients were treated and medicated following contemporary clinical practice and guidelines by each cardiologist in our hospital. Unless contraindicated, b-blockers, angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, and statins as lipid-lowering therapy were given as early as possible after PCI. Definitions Hypertension was defined as use of antihypertensive medication and/or a systolic blood pressure of 140 mmHg and/or diastolic blood pressure of 90 mmHg on admission. Dyslipidemia was defined as a low-density lipoprotein cholesterol level of 140 mg/dL and/or use of lipid-lowering agents on admission. Diabetes mellitus was defined as a glycated hemoglobin A1c concentration of 6.5% and/or current use of insulin or oral hypoglycemic agents on admission. Smokers were defined as patients who were previous and current smokers. Methods Patients We enrolled patients admitted to Nishinomiya Watanabe Cardiovascular Center in Japan from January 2010 to December 2016 who were diagnosed Primary PCI All patients received a bolus injection of unfractionated heparin (5000 U) in the emergency room as soon as diagnosed. Following oral administration of aspirin (200 mg) and an adenosine diphosphate receptor 3 Ichikawa et al. inhibitor (clopidogrel 300 mg or prasugrel 20 mg), primary PCI was performed according to standard methods. Thrombectomy was performed at the operator’s discretion. Glycoprotein IIb and IIIa were not used because these drugs were not approved for use in Japan. Whether to use bare-metal stents, drug-eluting stents or neither in the infarct-related artery was determined by each operator. Spontaneous recanalization of the infarct-related artery was defined as a Thrombolysis In Myocardial Infarction (TIMI) flow grade of 2 to 3 at the pre-PCI time point. Unsuccessful PCI was defined as a TIMI flow grade of 0 to 1 after PCI. CMR Within four days after primary PCI, eligible patients underwent contrast-enhanced CMR to establish the presence or absence of late microvascular obstruction. CMR studies were performed on a whole-body 1.5-T MR scanner (Intera Achieva; Philips Medical Systems, Best, The Netherlands) equipped with a six-element cardiac phased-array coil for signal reception. Patients were examined at rest in the supine position. Images were gated to the electrocardiogram and obtained during repeated breath-holds. Localizers and left ventricular (LV) function assessment included determining the LV ejection fraction, LV end-diastolic volume and LV endsystolic volume using steady-static, free-procession images. In the short-axis orientation, the left ventricle was completely encompassed by contiguous slices. Late gadolinium enhancement images were obtained from contiguous short-axis slices and representative longaxis slices of the left ventricle 10 to 15 min after intravenous injection of gadolinium-diethylenetriamine pentaacetic acid at 0.1 mmol/kg (Magnevist; Schering AG, Berlin, Germany). A breath-hold, three-dimensional inversion recovery gradient-echo pulse sequence (recovery time, 4.0 ms; echo time, 1.93 ms; flip angle, 20 ; typical spatial resolution, 1.56 mm 1.56 mm 10 mm) was used for image acquisition. We optimized the inversion time (250–300 ms) to null the normal myocardium. All analyses were interpreted by consensus of two blinded observers at an offline workstation (ViewForum; Philips Medical Systems). Sleep study During the first week, eligible patients underwent an overnight sleep study to measure the severity of their OSA. The sleep study was performed using a portable sleep apnea type 3 test (Somte; Compumedics, Melbourne, Australia) (SAS-3200; Nihon Kohden, Tokyo, Japan). These devices measured cardiopulmonary parameters, naso-oral airflow and thoracoabdominal movements to determine the type of apnea. Arterial oxyhemoglobin saturation was recorded using a pulse oximeter, and electrocardiographic recordings were obtained from a single lead. Apnea was defined as the cessation of inspiratory air flow lasting 10 s. Hypopnea was defined as a 50% reduction in airflow lasting 10 s associated with a 4% decrease in oxygen saturation (i.e. the oxygen saturation index) and a state of arousal. The respiratory event index (REI) was defined as the number of apnea and hypopnea events per hour. OSA was defined as the absence of airflow despite respiratory movement or exertion and an REI of 5 events/h, with >50% of them obstructive. OSA was thus classified into four categories based on the REI: absent (REI of <5), mild (REI of 5 to <15), moderate (REI of 15 to <30) and severe (REI of 30). Central sleep apnea was defined as the absence of both airflow and respiratory movement. Statistical analysis Continuous variables are expressed as mean  standard deviation or median (first and third quartiles). Categorical variables are expressed as number and percentage. Patients were divided into two groups according to the absence or presence of microvascular obstruction. Comparison analyses between groups were performed using Student’s t test or the Mann–Whitney U test for continuous variables and by v2 statistics or Fisher’s exact test for categorical variables, as appropriate. In the next analysis, patients were divided into four groups based on the severity of their OSA. Comparison analyses among groups were performed by trend testing according to the Jonckheere–Terpstra test for continuous variables and the Cochran–Armitage test for categorical variables. Univariate logistic regression analysis was performed to assess the relation between the presence of microvascular obstruction and multiple factors. A multivariate logistic regression analysis was performed to investigate the association between the severity of OSA and the presence of microvascular obstruction, adjusting for age, male sex, body mass index, hypertension, diabetes mellitus, dyslipidemia, smoking, spontaneous recanalization, anterior infarct, multivessel disease and a final TIMI flow grade of 3. A two-sided p-value of <0.05 was considered statistically significant. All statistical analyses were performed with EZR version 1.37 software (Saitama Medical Center, Jichi Medical University, Saitama, Japan), which is a graphic user interface for R version 3.4.1 (The R Foundation for Statistical Computing, Vienna, Austria). Results During our study period, 370 patients with STEMI underwent primary PCI. The study flowchart is shown in Figure 1. Among the 370 patients, 121 were 4 European Heart Journal: Acute Cardiovascular Care 0(0) STEMI patients with a primary PCI within 24 h from onsets: n = 370 Death: n = 3 Unsuccessful PCI: n = 4 Previous MI: n = 11 Heart failure and/or cardiogenic shock treated with intravenous injection inotropes and/or mechanical supports: n = 24 Complication required surgical indication: n = 3 Other severe disease: n = 10 Lack of CMR: n = 22 (contraindication: n = 8, claustrophobia: n = 2, refuse: n = 5, others: n =7) Patients with a successful CMR: n =293 Obstructive sleep apnea already receiving continuous positive airway pressure therapy: n =2 No sleep study: n =25 Patients with a successful sleep study: n =266 Central sleep apnea: n =17 Analyzed patients: n = 249 Absent: n = 55 Mild: n = 92 Moderate: n = 65 Severe: n = 37 Figure 1. Study flowchart. STEMI: ST-segment elevation myocardial infarction; PCI: percutaneous coronary intervention; MI: myocardial infarction; CMR: cardiovascular magnetic resonance imaging excluded from this study. The final study population comprised 249 patients. To evaluate the relation between the presence of microvascular obstruction and the extent of myocardial injury after STEMI, patients were divided into two groups according to the presence (n ¼ 139) or absence (n ¼ 110) of microvascular obstruction (Table 1). Patients with microvascular obstruction were significantly younger and comprised significantly more males than patients without microvascular obstruction. The peak creatine kinase level was higher and the incidence of anterior infarcts was significantly higher among patients with than without microvascular obstruction. The number of patients with spontaneous recanalization, the number of patients with a final TIMI flow grade of 3 and the LV ejection fraction measured on CMR were all significantly lower in patients with than without microvascular obstruction. The REI was significantly higher in patients with than without microvascular obstruction (p ¼ 0.023). These results suggest that the presence of microvascular obstruction reflects exacerbated myocardial damage after STEMI. To investigate the relation between OSA and microvascular obstruction, the patients were divided into four groups according to the severity of OSA based on their REI. The patients’ baseline characteristics are shown in Table 2. Age, body mass index and the incidences of diabetes mellitus and dyslipidemia were significantly higher in patients with more severe OSA. There were no significant differences in angiographic findings or in the LV function parameters (ejection fraction, enddiastolic volume or end-systolic volume) measured by CMR among the four groups. Microvascular obstruction was present in 42%, 58%, 57% and 70% of the absent, mild, moderate and severe OSA groups, respectively (for trend, p < 0.001) (Figure 2). These results indicate a significantly greater presence of microvascular obstruction in patients with more severe OSA. The univariate logistic analysis showed that age, spontaneous recanalization and a final TIMI flow grade of 3 were significantly associated with decreased odds of microvascular obstruction (Table 3). In contrast, male sex and an anterior infarct were significantly associated with increased odds of microvascular obstruction. Severe OSA was significantly associated with increased odds of microvascular obstruction (odds ratio (OR), 3.29; 95% confidence interval (CI), 1.36–7.97; p ¼ 0.008). The multivariate logistic regression analysis adjusted for age, sex, body mass index, dystension, diabetes mellitus, hyperlipidemia, smoking, spontaneous recanalization, anterior infarct, multivessel disease and final TIMI flow grade of 3 revealed that OSA was independently associated with the presence of microvascular obstruction in a severity-dependent 5 Ichikawa et al. Table 1. Patient characteristics according to the absence or presence of microvascular obstruction. Basal characteristics Age, years Male BMI, kg/m2 Peak CK, IU/L Cardiovascular risk factors Hypertension Diabetes mellitus Dyslipidemia Smoking Angiographic findings Onset to reperfusion time, h Infarct-related artery: LAD/LCX/RCA Anterior infarct Stent Multivessel disease Spontaneous recanalization Final TIMI flow grade 3 CMR results LV EF, % LV EDV, mL LV ESV, mL Sleep study REI, events/h 4% ODI, events/h Minimum SaO2, % MO absence n¼110 MO presence n ¼ 139 p-value 68  11 81 (74%) 24.1  3.6 929 (605–1409) 64  12 123 (89%) 24.0  3.5 3279 (2285–4787) 0.008 0.002 0.820 <0.001 66 20 46 66 76 28 69 87 (60%) (18%) (42%) (60%) (55%) (20%) (50%) (63%) 0.399 0.697 0.219 0.677 2.9 (2.0–4.5) 45 (41%)/8 (7%)/57 (52%) 3.3 (2.1–5.4) 80 (58%)/13 (9%)/46 (33%) 0.866 0.013 45 (41%) 103 (94%) 37 (34%) 45 (41%) 107 (97%) 80 (58%) 131 (94%) 49 (35%) 15 (11%) 124 (89%) 0.011 1.000 0.895 <0.001 0.015 54.7  10.9 108.2  31.9 50.7  25.1 46.2  9.3 120.6  34.1 66.6  27.8 <0.001 0.004 <0.001 10.7 (4.2–21.0) 6.7 (2.5–13.6) 86 (83–89) 12.8 (7.3–24.3) 9.8 (3.9–18.3) 87 (83–89) 0.023 0.063 0.811 Data are presented as mean  SD, median (first quartile–third quartile), or n (%). MO: microvascular obstruction; BMI: body mass index; CK: creatine kinase; LAD: left anterior descending artery; LCX: left circumflex artery; RCA: right coronary artery; TIMI: Thrombolysis In Myocardial Infarction; CMR: cardiovascular magnetic resonance imaging; LV: left ventricular; EF: ejection fraction; EDV: end-diastolic volume; ESV: end-systolic volume; REI: respiratory event index; ODI: decrease in oxygen saturation. manner (mild: OR, 5.10; 95% CI, 1.61–16.2; p ¼ 0.006, moderate: OR, 3.79; 95% CI, 1.43–10.10; p ¼ 0.008; and severe: OR, 2.88; 95% CI, 1.19–7.00; p ¼ 0.019, respectively) (Table 4). Discussion This prospective observational study showed that OSA is associated with the presence of microvascular obstruction in a severity-dependent manner in patients with acute STEMI. We measured microvascular obstruction by CMR, which enables precise detection of localized CMVO. It has been reported that CMVO is a strong independent prognosticator in patients with acute STEMI.2,4,5 Thus, the correlation between OSA and the presence of microvascular obstruction indicates that OSA is a predictor of adverse events in patients with acute STEMI after primary PCI. Previous studies have shown that OSA is associated with poor outcomes after primary PCI in patients with STEMI during follow-up.10,14–16 Additionally, the mechanisms of OSA-related cardiovascular disease are considered to be associated with microvascular dysfunction.17 However, little clinical evidence is available to support these speculations in the setting of STEMI. Our study supports previous findings and may explain the causes of a poor prognosis from the viewpoint of the coronary microcirculation. Previous studies have shown that coronary microvascular dysfunction can result from functional and/or structural alterations, the relative importance of which can vary depending on the clinical setting.18 We speculate that the mechanisms underlying the CMVO detected by CMR in patients with STEMI and OSA in the present study involved both functional and structural abnormalities. Morra and Roubille17 reported that OSA causes vascular remodeling (structural) and dysfunction of both endothelial cells and smooth muscle cells (functional). OSA causes intermittent and chronic transmural pressure variations that increase circumferential wall stress in the left ventricle. These changes might trigger structural vascular remodeling 6 European Heart Journal: Acute Cardiovascular Care 0(0) Table 2. Patient characteristics according to the severity of obstructive sleep apnea. Basal characteristics Age, years Male BMI, kg/m2 Peak CK, IU/L Cardiovascular risk factors Hypertension Diabetes mellitus Dyslipidemia Smoking Angiographic findings Onset to reperfusion time, h Infarct-related artery: LAD/LCX/RCA Anterior infarct Stent Multivessel disease Spontaneous recanalization Final TIMI flow grade 3 CMR results LV EF, % LV EDV, mL LV ESV, mL MO presence Sleep study REI, events/h 4% ODI, events/h Minimum SaO2, % Absent n¼ 55 Mild n¼ 92 Moderate n¼ 65 Severe n¼ 37 p-value 60  12 42 (76%) 23.3  3.3 1627 (828–3326) 65  13 79 (86%) 23.8  3.3 2062 (829–3569) 69  11 51 (79%) 24.7  4.0 2114 (1333–3678) 68  9 32 (87%) 24.3  3.7 2135 (1247–3917) <0.001 0.483 0.017 0.126 47 18 34 35 19 9 21 24 24 (44%) 5 (9%) 18 (33%) 34 (62%) 52 16 42 60 (57%) (17%) (46%) (65%) (72%) (28%) (52%) (54%) (51%) (24%) (57%) (65%) 0.088 0.015 0.013 0.729 3.1 (2.1–5.5) 32, 1, 22 32 (58%) 52 (95%) 12 (22%) 16 (29%) 52 (95%) 3.1 (2.0–5.1) 39, 10, 43 39 (42%) 86 (94%) 44 (48%) 20 (22%) 86 (94%) 3.2 (2.1–5.5) 31, 7, 27 31 (48%) 62 (95%) 21 (32%) 17 (26%) 58 (89%) 3.3 (2–4.6) 23, 3, 11 23 (62%) 34 (92%) 9 (24%) 7 (19%) 35 (95%) 0.866 0.381 0.715 0.799 0.785 0.429 0.631 49.7  9.6 124.1  35.0 64.2  29.3 23 (42%) 51.7  11.3 112.8  35.9 57.3  30.5 53 (58%) 49.1  11.3 112.3  30.9 58.5  24.6 37 (57%) 47.7  10.7 112.5  29.4 60.5  23.2 26 (70%) 0.288 0.242 0.923 0.013 2.3 (0.6–3.3) 1.1 (0.5–2.1) 91 (88–92) 9.7 (7.3–11.6) 5.5 (3.5–7.9) 87 (84–90) 22.0 (18.9-25.8) 14.7 (12–19.7) 85 (82–88) 36.8 (32.8–39.4) 28.5 (22.2–34.9) 83 (77–85) <0.001 <0.001 <0.001 Data are presented as mean  SD, median (first quartile–third quartile), or n (%). BMI: body mass index; CK: creatine kinase; LAD: left anterior descending artery; LCX: left circumflex artery; RCA: right coronary artery; TIMI: Thrombolysis In Myocardial Infarction; CMR: cardiovascular magnetic resonance imaging; LV: left ventricular; EF: ejection fraction; EDV: end-diastolic volume; ESV: end-systolic volume; MO: microvascular obstruction; REI: respiratory event index; ODI: decrease in oxygen saturation P for trend < 0.001 70% 58% 57% Mild Moderate 42% Absent Severe Figure 2. Incidence of microvascular obstruction at various levels of obstructive sleep apnea (OSA) severity. The incidence of microvascular obstruction was 42%, 58%, 57% and 70% in the absence of OSA and with mild, moderate, and severe OSA, respectively (for trend, p<0.001). processes such as vascular smooth muscle cell hypertrophy and wall thickening.17 Additionally, OSA-induced intermittent hypoxia and generation of reactive oxygen species cause oxidative stress and stimulate proinflammatory processes, which result in endothelial dysfunction.19 Moreover, hypoxia itself can cause the deterioration of vascular endothelial function because oxygen is essential for nitric oxide biosynthesis from L-arginine, and hypoxia may reduce nitric oxide production.13 One study showed that endothelium-dependent vasodilation was significantly impaired in patients with OSA,20 and endothelial dysfunction is a well-known prognostic factor for various cardiovascular events.21,22 In our study population, structural and functional abnormalities in the microcirculation were already present prior to the onset of STEMI in patients with OSA. These preexisting coronary microvascular dysfunctions might contribute to the development and poor prognosis of STEMI. Ischemia–reperfusion injury and distal embolism are other mechanisms of CMVO after primary PCI in patients with STEMI.4 These alterations result from various intracellular and extracellular responses, such as neutrophil infiltration, swelling of vascular cells and cardiomyocytes, and micro-obstruction of small arteries and arterioles.18 Oxidative stress, inflammation and 7 Ichikawa et al. Table 3. Univariate logistic analysis according to microvascular obstruction presence. Univariate factors OR 95% CI p-value Age Male BMI Hypertension Diabetes mellitus Dyslipidemia Smoking Onset to reperfusion time Spontaneous recanalization Anterior infarct Multivessel disease Final TIMI flow grade 3 REI Severity of OSA Absent Mild Moderate Severe 0.972 2.75 0.992 0.804 1.14 1.37 1.12 1.05 0.175 1.96 1.07 0.232 1.020 0.951–0.993 1.41–5.39 0.924–1.06 0.484–1.34 0.600–2.15 0.828–2.27 0.668–1.86 0.980–1.13 0.091–0.337 1.18–3.25 0.634–1.82 0.065–0.822 1.00–1.04 0.009 0.003 0.819 0.400 0.697 0.219 0.677 0.162 <0.001 0.009 0.790 0.023 0.038 1 (Reference) 1.89 1.840 3.29 – 0.961–3.72 0.889–3.80 1.36–7.97 – 0.065 0.101 0.008 OR: odds ratio; CI: confidence interval; BMI: body mass index; TIMI: Thrombolysis In Myocardial Infarction; REI: respiratory event index; OSA: obstructive sleep apnea Table 4. Multivariate logistic analysis according to microvascular obstruction presence. Multivariate factors OR 95% CI p-value Severity of OSA Absent Mild Moderate Severe 1 (Reference) 2.88 3.79 5.10 – 1.19–7.00 1.43–10.1 1.61–16.2 – 0.019 0.008 0.006 Adjusting for age, sex, body mass index, hypertension, diabetes mellitus, dyslipidemia, smoking, spontaneous recanalization, anterior infarct, multivessel disease, final Thrombolysis In Myocardial Infarction flow grade 3. OR: odds ratio; CI: confidence interval; OSA: obstructive sleep apnea vasoconstriction, which are triggered by OSA-mediated intermittent hypoxia, are functionally involved in these mechanisms;12 therefore, in our study, CMVO could be exacerbated in STEMI patients with OSA because of the presence of both functional and structural alterations. Diabetes and dyslipidemia also reportedly cause endothelial dysfunction.23 As shown in our study, patients with severe OSA were highly susceptible to complications associated with these diseases. These data suggest that the adverse interaction between these comorbidities and OSA synergistically impairs endothelial dysfunction, leading to microvascular obstruction in a severity-dependent manner. Interventions to address the impaired microcirculation caused and exacerbated by OSA may be a new therapeutic approach to preventing CMVO, which would improve the prognosis of patients with STEMI. Administration of CPAP is the standard treatment for OSA. Although several studies have shown that treatment of OSA with CPAP reduces the incidence of acute coronary syndrome,24 the effect of acute coronary syndrome on patients’ prognosis has not been established. As shown in this study, severe OSA exacerbates CMVO and may lead to a poor prognosis in patients with STEMI. Therefore, research that focuses on alleviating the OSA-related CMVO is needed. In addition, none of the patients in our study had been diagnosed with OSA at the time of their inclusion in the study. Hence, we cardiologists should be more attuned to diagnosing OSA at an earlier time and start an intervention before the onset of cardiovascular events. Limitations This study has several limitations. First, myocardial infarction and heart failure may exacerbate breathing disorders during sleep.25 In this study, whether OSA was the cause or the result of the cardiac problems was unclear because the sleep studies were performed in patients with acute STEMI only after primary PCI. To minimize the effects of the acute phase as much as possible, we excluded patients with heart failure and/or cardiogenic shock treated with intravenous injection of inotropes and/or mechanical support, patients with central sleep apnea, and performed sleep studies 1 week after onset. Second, although we clearly showed 8 that the presence of microvascular obstruction was associated with severe OSA, microvascular obstruction was not evaluated quantitatively. Therefore, the association between the extent of microvascular obstruction and OSA could not be investigated. Because quantitative evaluation of microvascular obstruction may explain the strong correlation between microvascular obstruction and OSA, additional studies are needed. Third, after the onset of STEMI, CMVO is caused by structural and functional cardiovascular abnormalities. Which factor is more strongly involved in these processes remains unclear. CMR might provide additional clues to this question by evaluating the presence of intramyocardial hemorrhage (IMH), which appears to be a consequence of microvascular injury.4 IMH is caused by the destruction of endothelial walls due to the sudden appearance of intravascular positive pressure after reperfusion, resulting in microvascular dysfunction.5 Because the presence of IMH reflects irreversible myocardial damage and may be affected by structural rather than functional cardiovascular abnormalities, assessing the presence of IMH may help to identify whether structural or functional abnormalities are having a stronger effect. However, we could not evaluate IMH in the present study, and further research is required. Fourth, the serum levels of inflammatory markers such as C-reactive protein, tumor necrosis factor-alpha, interleukin 6 and interleukin 8 may provide additional information to clarify the mechanism of CMVO in the setting of STEMI. Moreover, recent studies have shown that the skin and retinal microcirculation, which is closely related to OSA, is structurally and functionally similar to the microcirculation of other organs, such as the coronary artery.26,27 Thus, non-invasive evaluation of the microcirculation in the skin and retina can reflect coronary microvascular dysfunction in patients with OSA. Overall, using these modalities to assess functional and/or structural alterations may clarify the mechanisms of CMVO; further research is needed. Finally, although we showed the association between OSA and CMVO, evaluation of clinical follow-up data can reveal the patients’ actual prognosis. However, followup data were available for only some patients; therefore, we were unable to analyze the patients’ prognosis in this study. We plan to continue accumulating more data and provide new findings in the future. Conclusion The severity of OSA is independently associated with CMVO as evaluated by CMR in patients with acute STEMI after primary PCI. Because severe OSA causes CMVO and may lead to a poor prognosis in patients with STEMI, therapeutic interventions to European Heart Journal: Acute Cardiovascular Care 0(0) alleviate the CMVO caused by OSA should be investigated. Acknowledgment We thank Nancy Schatken, BS, MT(ASCP) and Angela Morben, DVM, ELS, from Edanz Group (https://en-auth or-services.edanzgroup.com/), for editing a draft of this manuscript. Conflict of interest The authors have no conflicts of interest to declare. Funding The authors received no financial support for the research, authorship, and/or publication of this article. References 1. Keely EC, Boura JA and Grines CL. Primary angioplasty versus intravenous thrombolytic therapy for acute myocardial infarction: A quantitative review of 23 randomised trials. Lancet 2003; 361: 13–20. 2. Bekkers SC, Yazdani SK, Virmani R, et al. Microvascular obstruction: Underlying pathophysiology and clinical diagnosis. J Am Coll Cardiol 2010; 55: 1649–1660. 3. Ito H, Tomooka T, Sakai N, et al. Lack of myocardial perfusion immediately after successful thrombolysis. A predictor of poor recovery of left ventricular function in anterior myocardial infarction. Circulation 1992; 85: 1699–1705. 4. Niccoli G, Scalone G, Lerman A, et al. Coronary microvascular obstruction in acute myocardial infarction. Eur Heart J. 2016; 37: 1024–1033. 5. Ibanez B, Aletras AH, Arai AE, et al. Cardiac MRI endpoints in myocardial infarction experimental and clinical trials: JACC Scientific Expert Panel. J Am Coll Cardiol 2019; 74: 238–256. 6. Nijveldt R, Hofman MB, Hirsch A, et al. Assessment of microvascular obstruction and prediction of short-term remodeling after acute myocardial infarction: Cardiac MR imaging study. Radiology 2009; 250: 363–370. 7. Wu KC. CMR of microvascular obstruction and hemorrhage in myocardial infarction. J Cardiovasc Magn Reson 2012; 14: 68. 8. Hombach V, Grebe O, Merkle N, et al. Sequelae of acute myocardial infarction regarding cardiac structure and function and their prognostic significance as assessed by magnetic resonance imaging. Eur Heart J 2005; 26: 549–557. 9. De Waha S, Desch S, Eitel I, et al. Impact of early vs. late microvascular obstruction assessed by magnetic resonance imaging on long-term outcome after ST-elevation myocardial infarction: A comparison with traditional prognostic markers. Eur Heart J 2010; 31: 2660–2668. 10. Fan J, Wang X, Ma X, et al. Association of obstructive sleep apnea with cardiovascular outcomes in patients 9 Ichikawa et al. 11. 12. 13. 14. 15. 16. 17. 18. 19. with acute coronary syndrome. J Am Heart Assoc 2019; 8: e010826. Mohananey D, Villablanca PA, Gupta T, et al. Recognized obstructive sleep apnea is associated with improved in-hospital outcomes after ST elevation myocardial infarction. J Am Heart Assoc 2017; 6: e006133. Bradley TD and Floras JS. Obstructive sleep apnoea and its cardiovascular consequences. Lancet 2009; 373: 82–93. Nakashima H, Muto S, Amenomori K, et al. Impact of obstructive sleep apnea on myocardial tissue perfusion in patients with ST-segment elevation myocardial infarction. Circ J 2011; 75: 890–896. Yumino D, Tsurumi Y, Takagi A, et al. Impact of obstructive sleep apnea on clinical and angiographic outcomes following percutaneous coronary intervention in patients with acute coronary syndrome. Am J Cardiol 2007; 99: 26–30. Buchner S, Satzl A, Debl K, et al. Impact of sleepdisordered breathing on myocardial salvage and infarct size in patients with acute myocardial infarction. Eur Heart J 2014; 35: 192–199. Mazaki T, Kasai T, Yokoi H, et al. Impact of sleepdisordered breathing on long-term outcomes in patients with acute coronary syndrome who have undergone primary percutaneous coronary intervention. J Am Heart Assoc 2016; 5: e003270. Morra S and Roubille F. Obstructive sleep apnoea: From respiratory events to coronary microvascular dysfunction. Acta Cardiol 2018; 73: 319–324. Camici PG, d’Amati G and Rimoldi O. Coronary microvascular dysfunction: Mechanisms and functional assessment. Nat Rev Cardiol 2015; 12: 48–62. Shpilsky D, Erqou S, Patel SR, et al. Association of obstructive sleep apnea with microvascular endothelial 20. 21. 22. 23. 24. 25. 26. 27. dysfunction and subclinical coronary artery disease in a community-based population. Vasc Med 2018; 23: 331–339. Kato M, Roberts-Thomson P, Phillips BG, et al. Impairment of endothelium-dependent vasodilation of resistance vessels in patients with obstructive sleep apnea. Circulation 2000; 102: 2607–2610. Suwaidi JA, Hamasaki S, Higano ST, et al. Long-term follow-up of patients with mild coronary artery disease and endothelial dysfunction. Circulation 2000; 101: 948–954. Marks DS, Gudapati S, Prisant LM, et al. Mortality in patients with microvascular disease. J Clin Hypertens (Greenwich) 2004; 6: 304–309. Bakker W, Eringa EC, Sipkema P, et al. Endothelial dysfunction and diabetes: Roles of hyperglycemia, impaired insulin signaling and obesity. Cell Tissue Res 2009; 335: 165–189. Marin JM, Carrizo SJ, Vicente E, et al. Long-term cardiovascular outcomes in men with obstructive sleep apnoea-hypopnoea with or without treatment with continuous positive airway pressure: An observational study. Lancet 2005; 365: 1046–1053. Chami HA, Resnick HE, Quan SF, et al. Association of incident cardiovascular disease with progression of sleep-disordered breathing. Circulation 2011; 123: 1280–1286. Shankar A, Peppard PE, Young T, et al. Sleep-disordered breathing and retinal microvascular diameter. Atherosclerosis 2013; 226: 124–128. Lin K-H, Lin H-Z, Lin Y-P, et al. Obstructive sleep apnea and retinal microvascular characteristics: A brief review. Neuropsychiatry 2017; 7: 12–21.