Case report

Neonatal myocardial infarction from a coronary artery thrombus is a rare entity with a high mortality rate and often unknown aetiology. Association with prothrombotic conditions, myocarditis, prematurity, and umbilical venous line placement has been reported.Reference Poonai, Kornecki, Buffo and Pepelassis ^{1 ,} Reference Farooqi, Sutton, Weinstein, Menegus, Spindola-franco and Pass ² Treatment strategies are sparse, including extracorporeal membrane oxygenation and supportive care. Surgical thrombectomy with left ventricular assist device support has also been reported.Reference Ramlogan, McKee, Lofland and Carlson ³ The use of thrombolytic therapy in the management of neonatal myocardial infarction is controversial and not well established, but it has been used successfully in rare reported cases.Reference Cesna, Eicken, Juenger and Hess ^{4 ,} Reference Deutsch, Cleuziou and Noebauer ⁵ We used local tissue plasminogen activator at a dose several folds lower than had been previously reported, with subsequent thrombus resolution, normalisation of ventricular function by 4 weeks of age, no neurological sequelae, and no utilisation of extracorporeal membrane oxygenation support.

The patient was born at term following an uncomplicated pregnancy with unremarkable prenatal ultrasounds. At 10 minutes of life, he developed cyanosis and poor perfusion and was found to be in wide complex tachycardia. An electrocardiogram in sinus rhythm shortly thereafter showed a right bundle branch block with non-specific T-wave abnormalities, but no significant ST segment changes and no Q waves. Despite return of sinus rhythm and medical management of metabolic acidosis, he remained in cardiogenic shock. Echocardiogram showed a structurally normal heart with severe left ventricular dysfunction, regional wall motion abnormality of the left ventricular free wall, and moderate mitral regurgitation. The coronary arteries were not well visualised. He was transferred to our tertiary care children’s hospital owing to concerns for a coronary artery anomaly.

At ~24 hours of life, cardiac catheterisation with selective left coronary artery angiography revealed normal coronary anatomy but with the presence of a large filling defect at the bifurcation of the left anterior descending artery and left circumflex artery. This filling defect was consistent with a large thrombus, likely embolic in nature, as suggested by its “saddle embolus”-like location (Supplementary Movie 1). He developed transient ST segment changes, bradycardia, and hypotension with each selective angiogram. The left ventricular end-diastolic pressure was elevated (16 mmHg). Angiographically, there was severe left ventricular dysfunction, minimal free wall contractility, and severe mitral regurgitation. In addition to systemic heparinisation, tissue plasminogen activator (Genentech Inc., San Francisco, California, United States of America) was delivered locally via a 3.3 Fr Judkins Left-1 catheter (Pedia Vascular Inc., Cleveland, Ohio, United States of America) engaged in the proximal left coronary artery. In all, two tissue plasminogen activator injections of 0.2 mg (0.05 mg/kg) were administered with no significant acute angiographic changes. Owing to a perceived increased risk of intracranial bleeding in a newborn treated with tissue plasminogen activator, and as the patient’s haemodynamics were relatively stable, our plan was to repeat local tissue plasminogen activator delivery in a few days if the thrombus did not improve, as determined by planned repeat angiography. He was maintained on an unfractionated heparin infusion at therapeutic level (anti Xa levels 0.5–0.7) in the cardiac ICU.

For the next 3 days, he was kept sedated, intubated, and kept on medical afterload reduction to minimise myocardial oxygen demand. During this time, his left ventricular function improved from severely to moderately depressed and his troponin-I levels decreased. At 4 days of life, repeat coronary angiography showed complete thrombus resolution (Supplementary Movie 2). This repeat catheterisation was undertaken as the thrombus had not acutely resolved at the time of tissue plasminogen activator administration and ventricular function remained moderately depressed, despite lowered myocardial oxygen demands – sedated, intubated, and medical afterload reduction. The initial tissue plasminogen activator was lower than previously reported to minimise the risk of intracranial haemorrhage in the setting of relative clinical stability. The plan was to repeat intracoronary tissue plasminogen activator dosing if indicated at the time of repeat catheterisation. Instead, the patient was transitioned to subcutaneous lovenox and aspirin, and oral medical management of left ventricular dysfunction. At 2 weeks of age, his left ventricular ejection fraction had improved to 50%. By 4 weeks, it had normalised to 65% with no regional wall motion abnormalities, and there was no mitral regurgitation. Troponin-I levels had decreased from initial 10.5 to 0.15 ng/ml – <0.3 not consistent with acute myocardial infarction – and brain natriuretic peptide levels had decreased from 10,582 to 2533 pg/ml (normal <100). Head ultrasounds remained normal. He was discharged home at 3 weeks of life.

The aetiology of the coronary thrombus remains unclear, and limited hypercoagulable testing has been unrevealing. Factor V leiden and prothrombin gene mutations were negative, and anti-thrombin III levels were appropriate for age. Maternal testing for anti-phospholipid antibodies was normal. The patient will remain on lovenox until at least 6 months of age, given the severity of the clinical presentation and the assumption that maternal antibodies may have influenced the initial thrombosis. Testing at 6 months of age will reflect the patient’s antibody status and will influence the decision regarding continued anticoagulation, potentially lifetime therapy.

Prompt coronary angiography is crucial for the potentially life-saving diagnosis of a neonatal myocardial infarction. In this case, the clinical presentation and regional left ventricular wall motion abnormalities on echocardiogram lead us to early catheterisation, despite the absence of expected electrocardiogram changes, possibly because of ischaemic right bundle branch block, and before laboratory markers of myocardial ischaemia were available. In this patient, local “low-dose” tissue plasminogen activator (0.1 mg/kg versus 0.5–1 mg/kg in prior reportsReference Cesna, Eicken, Juenger and Hess ^{4 ,} Reference Deutsch, Cleuziou and Noebauer ⁵ ) and systemic therapeutic heparinisation were successful at resolving the large left coronary artery thrombus. This dose of thrombolytics presumably lowered the risk of intracranial haemorrhage in this 24-hour-old patient. Furthermore, he did not require extracorporeal membrane oxygenation support, experienced no adverse events, and his left ventricular systolic function normalised. In summary, in this case of a neonatal myocardial infarction owing to a large left coronary artery thrombus, the combination of intracoronary low-dose tissue plasminogen activator and systemic anticoagulation was effective in resolving the thrombus with subsequent normalisation of ventricular function.