Incidence of thrombosis in perioperative and non-operative myocardial infarction

• Published in: British journal of anaesthesia : BJA Vol. 120; no. 4; pp. 725 - 733
• Main Authors: Sheth, T., Natarajan, M.K., Hsieh, V., Valettas, N., Rokoss, M., Mehta, S., Jolly, S., Tandon, V., Bezerra, H., Devereaux, P.J.
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 01.04.2018
• Subjects: Aged; Cohort Studies; Comorbidity; coronary artery disease; Female; Humans; Incidence; Male; myocardial infarction; Myocardial Infarction - epidemiology; optical coherence tomography; perioperative; Perioperative Period; Prospective Studies; thrombosis; Tomography, Optical Coherence; Venous Thrombosis - diagnostic imaging; Venous Thrombosis - epidemiology; coronary artery disease; perioperative; thrombosis; optical coherence tomography; myocardial infarction
• ISSN: 0007-0912; 1471-6771
• DOI: 10.1016/j.bja.2017.11.063

The contribution of thrombosis to the aetiology of perioperative myocardial infarction (MI) is uncertain. We used optical coherence tomography (OCT) to determine the presence of thrombus and plaque morphology in patients experiencing a perioperative MI and matched patients experiencing a non-operative MI using OCT. We conducted a single-centre, prospective, cohort study. Thirty patients experiencing a perioperative MI and 30 matched patients experiencing a non-operative MI, without ST elevation, underwent OCT to determine the presence of thrombus and culprit lesion plaque morphology. Angiography and OCT were performed a mean of 1.93(1.09) days and 1.53(0.68) days after the onset of perioperative and non-operative MI, respectively. OCT images were evaluated by an independent core laboratory without knowledge of whether the patient had suffered a perioperative or non-operative MI. We identified thrombus at the culprit lesion in four of 30 patients (13.3%) who experienced a perioperative MI and in 20 of 30 patients (66.7%) who experienced a non-operative MI, P<0.01. The only non-culprit lesion with thrombus was in a perioperative MI patient who also had a culprit lesion thrombus. Perioperative and non-operative MI culprit lesions demonstrated fibroatheroma in 18 patients (60.0%) us 20 patients (66.7%), respectively (P=0.52) and thin cap fibroatheroma in one patient (3.3%) us five patients (16.7%), respectively (P=0.11). One perioperative MI patient (3.3%) suffered a cardiac death and no non-operative MI patient died during the 30-day follow-up. Thrombosis was less common in perioperative than non-operative MI, despite similar underlying plaque morphology.