Case report
Congenital diaphragmatic hernia in the fetus of a 29-year-old pregnant woman at 21 weeks’ gestation was detected on routine fetal ultrasonography. At 41 + 1 weeks’ gestation, the patient was delivered by cesarean section. Birthweight was 3260 g, and Apgar score was 9 at 1 minute and 9 at 5 minute. A chest radiograph demonstrated collapsed left lung due to intestine gas in the left thoracic cavity. Echocardiogram revealed a structurally normal heart, a patent ductus arteriosus (0.27 cm, bi-directional), and a patent foramen ovale and elevated pulmonary artery pressure.
Surgery for the left diaphragmatic hernia was carried out by a thoracoscope at 50 hours after birth. Operation time was 1 hour, and the size of the diaphragmatic hernia was 4 × 3 cm2, almost half size of the intact diaphragm. Three hours after surgery, severe pulmonary hypertension and haemodynamic instability developed. Severe hypoxaemia lasted for 6 hours and resulted in a pronounced metabolic acidosis (pH 7.16, lactate 7.3 mmol/L). A head ultrasound showed no evidence of intracranial haemorrhage. The decision was made to initiate extracorporeal membrane oxygenation support. The neonate was placed on veno-arterial extracorporeal membrane oxygenation via common carotid artery and internal jugular vein in a cut-down method.
Pump flows were stable during the first 30 hours, and metabolic acidosis was improved. However, in the next 5 hours, extracorporeal membrane oxygenation flow notably decreased and accompanied by deteriorating haemoglobinuria as a sign of haemolysis. Intermittent extracorporeal membrane oxygenation pump interruptions were dealt with fluid administration. However, an hour later, extracorporeal membrane oxygenation flow was completely interrupted. The emergent echocardiography demonstrated a thrombus developed in aortic arch near the arterial cannula with a size of 0.6 × 2.8 cm2 (Fig 1a). Clots in the circuit and arterial cannulas were confirmed. Therefore, the whole set of the circuit including cannulas was immediately exchanged with a loading dose of unfractionated heparin (100 U/kg). Extracorporeal membrane oxygenation flow was restored, but the thrombus posed a great threat to cerebral embolism. And if the thrombus continued to enlarge, it might interrupt the extracorporeal membrane oxygenation flow again. As soon as activated clotting time optimised to 160–200 seconds, the unfractionated heparin infusion was continued and increased to a therapeutic dose of 20 U/kg/hour. However, 4 hours passed by, there was no evidence of dissolving the clot. Collaboration with a multi-disciplinary team considered administration of alteplase as a possible non-invasive option for treatment of the thrombus.
Figure 1. Aortic arch ultrasound photographs: ( a ) echocardiogram pre-alteplase administration demonstrated a large clot; ( b ) complete resolution of aortic arch clot after thrombolytic therapy combining alteplase and unfractionated heparin.
There is no protocol to follow for alteplase therapy, and this was the first administration of alteplase in our hospital. Before initiation of alteplase, the fibrinogen level was 176 mg/dl and platelet counts were 165,000/µl. Fibrinogen was administered when it dropped below 100 mg/dl. Platelet counts were aimed to maintain above 100,000/µl. Intravenous alteplase infusion commenced at 0.1 mg/kg/hour systematically without a bolus. Meanwhile, the unfractionated heparin infusion was decreased to 5 U/kg/hour. Four hours later, echocardiography showed no significant lessening of the clot. Alteplase and unfractionated heparin were increased to 0.15 mg/kg/hour and 10 U/kg/hour, respectively. About 6 hours later, the clot reduced to half size. After another 6-hour infusion, owing to a perceived increased risk of intracranial bleeding in a newborn treated with alteplase, it was reduced to 0.025 mg/kg/hour. Four hours later, alteplase was discontinued and unfractionated heparin alone was prescribed as antithrombin treatment. Therefore, the dose of unfractionated heparin was adjusted to 20 U/kg/hour. Fifteen hours later, excessive bleeding in thoracic cavity occurred (thoracic drainage 240 ml, 74 ml/kg) and required blood product transfusion including packed red blood cell, fresh frozen plasma, and platelet. Twenty-two hours after completion of alteplase infusion, repeat echocardiography demonstrated complete resolution of the aortic arch thrombus (Fig 1b). Extracorporeal membrane oxygenation was weaned off after 6 days of support when the patient’s pulmonary function improved and pulmonary hypertension was controlled. Subsequent ultrasound demonstrated slight thrombus in common iliac artery, and the head ultrasound showed no evidence of intracranial haemorrhage. Afterwards, low-molecular-weight heparin was subcutaneously injected (100 u/kg, q12h) to treat the thrombus in common iliac artery. Three days later, physical examination revealed that the pupil size of both eyes was not equal. Cranial CT demonstrated slight subdural haemorrhage. Low-molecular-weight heparin was reduced to 100 u/kg, qd, for 11 days. Finally, the general condition of the patient was improved gradually. He was discharged with better pulmonary function and without complications.
Discussion
In this case, we present a neonatal case of the use of alteplase for the lysis of a large aortic arch thrombus formed during extracorporeal membrane oxygenation support after congenital diaphragmatic hernia surgery, in which condition, many clinicians would be reluctant to attempt. Anticoagulation and thrombolysis therapy included alteplase low dose 0.1–0.15 mg/kg/hour combined with unfractionated heparin 5–10 u/kg/hour. Despite the great challenge of bleeding, alteplase was successfully administered and the patient survived without apparent catastrophic long-term complications. In this case, the benefits of clot resolution ended up exceeding the risks of significant bleeding. Therefore, it is reasonable to consider alteplase for thrombolysis in this setting.
Alteplase has been used with success but is contraindicated in patients with recent surgical procedures and may be associated with catastrophic bleeding complications such as intracranial haemorrhage, especially in neonates. Reference Garcia, Gander and Gross1 Alteplase’s pharmacokinetics have not been established in infants and children. Its pharmacokinetics in adults shows that more than 50% present in plasma cleared approximately 5 minutes after infusion terminated, approximately 80% cleared within 10 minutes; fibrinolytic activity persists for up to 1 hour after infusion terminated. Compared to adult, the decreased level of plasminogen in neonates (≈50% of that in adults) slows the generation of plasmin and reduces the effect of thrombolytics in an in vitro system. Reference Andrew, Vegh, Johnston, Bowker, Ofosu and Mitchell2 Thrombolytics have a low margin of safety and variable efficacy in children and therefore should be used with caution. Pediatric Cardiac Intensive Care Society 2014 Consensus (United States of America) suggests that high-dose alteplase may be considered for patients with arterial or more critical thrombotic events. It suggests an initial dose of 0.1–0.6 mg/kg/hour. Reference Giglia, Witmer, Procaccini and Byrnes3 With regard to the ongoing extracorporeal membrane oxygenation support, the use of low-dose unfractionated heparin may help in preventing thrombus formation but will increase the risk of bleeding. In addition, fibrinogen drops dramatically during alteplase administration. Hence, in this condition, the neonate should be monitored with frequent measurements of activated clotting time, prothrombin time, partial thromboplastin time, fibrinogen, platelet counts, and D-dimer levels. Fibrinogen should be transfused when it drops below 100 mg/dl. Platelet counts should be maintained above 100,000/µl using an appropriate transfusion.
Although no intracranial haemorrhage occurred during alteplase delivery, excessive bleeding from the surgical site happened when high-dose heparin was infused afterwards. It gives us a meditation that whether an aggressive antithrombin regimen is suitable in neonates during extracorporeal membrane oxygenation support. Usually, in our hospital, targeting activated clotting time 160–200 seconds for neonatal extracorporeal membrane oxygenation patients after surgery, the unfractionated heparin dose was below 20 u/kg/hour. Instead, a moderate dose of unfractionated heparin at 10 U/kg/hour is safer. After weaning off from extracorporeal membrane oxygenation, slight subdural haemorrhage happened due to low-molecular-weight heparin injection for a slight thrombus in common iliac artery. Although the patient recovered without tragic complications, it warns us that low-molecular-weight heparin use in neonates still has a high risk of haemorrhage. Serious consideration of the risk and benefits in each situation prior to initiation of therapy is necessary.
There are no studies that describe the risk of bleeding while administering alteplase during extracorporeal membrane oxygenation. Even fewer cases for alteplase systemic administration in neonates during extracorporeal membrane oxygenation support were reported. Reference Olarte, Glover and Totapally4 This case presents detailed thrombolytic therapy in a neonate supported by extracorporeal membrane oxygenation after congenital diaphragmatic hernia surgery. Successful medical thrombolysis despite the small size and very high severity of illness and comorbidities was also the unique aspect of this case. Our experience learned from this case for thrombolysis during extracorporeal membrane oxygenation in neonates is treating with alteplase at 0.1–0.15 mg/kg/hour combined with unfractionated heparin at 5–10 U/kg/hour for a certain period such as 6 hours. And, the whole process should be under frequent monitoring of coagulation spectrum and platelet counts.
Acknowledgements
My deepest gratitude goes first and foremost to Professor Ru Lin, for her constant encouragement and guidance. Second, I would like to express my thanks to my colleagues who have contributed to this article.
Financial Support
This research received no specific grant from any funding agency, commercial, or not-for-profit sectors.
Conflicts of Interest
None.
Ethical Standards
Zhejiang University School of Medicine Children's Hospital Committee on Clinical Investigation approved the review of patient medical records.
