Discussion

In the absence of additional pulmonary blood flow, acute shunt occlusion is catastrophic with low survivability. Although the aetiology of acute shunt thrombosis cannot be identified in all cases, known risk factors include anatomical stenosis in the shunt, arrhythmia, infection, dehydration, perioperative shunt clamping, thrombophilia and small shunt size.^{Reference Bonnet, Petit and Lambert2}

In our patient, pulmonary vein angioplasty was likely the trigger for shunt thrombosis. The transient complete left pulmonary vein occlusion during balloon angioplasty likely resulted in acute increase in pulmonary artery pressure. Lack of concurrent change in systemic arterial pressure decreased the driving pressure gradient across the shunt and led to stagnated blood flow, which in turn induced thrombosis. Another possible mechanism, although unlikely, is the embolisation of a thrombus in the shunt that initially formed somewhere else.

Shunt thrombosis can be treated surgically or percutaneously.^{Reference Sasikumar, Hermuzi and Fan1,Reference Bonnet, Petit and Lambert2,Reference Kogon, Villari and Shah5} Transcatheter methods include balloon angioplasty, stent placement, thrombolysis and/or rheolytic thrombectomy.^{Reference Bonnet, Petit and Lambert2,Reference Kogon, Villari and Shah5,Reference Gillespie and Rome6} Historically, tissue plasminogen activator has been used systemically to treat thrombosis at different locations including shunt thrombosis. Most recent reports have described successful transcatheter tissue plasminogen activator use in treating thrombosis in the shunt as well in the coronary artery. Nevertheless, in these reports, relatively high doses were utilised and reflect institutional practice rather than universal guidelines.^{Reference Patange, Blake and Gowda3,Reference Moszura, Zubrzycka and Michalak4,Reference Mohammad Nijres, Huntington, Baliulis and Vettukattil7} Our patient responded well to local low-dose tissue plasminogen activator (0.1 mg/kg) in line with recent evidence of low-dose efficacy in small children with thrombosis.^{Reference Patange, Blake and Gowda3,Reference Mohammad Nijres, Huntington, Baliulis and Vettukattil7} In hindsight, high-dose heparin (330 IU/kg) administered after confirming shunt thrombosis was probably not necessary and 100 IU/kg would have been sufficient. However, total procedure heparin dose of 405 IU/kg was within previously reported range (300–500 IU/kg) in patients who underwent transcatheter recanalisation of shunt thrombosis.^{Reference Bonnet, Petit and Lambert2} Although angioplasty alone has the potential to dislodge and fragment the thrombus, it cannot dissolve it. Therefore, the authors agree with Moszura and colleagues’ approach where local tissue plasminogen activator was administered in all cases with shunt thrombosis unless a contraindication was present.^{Reference Moszura, Zubrzycka and Michalak4}

Use of tissue plasminogen activator is not risk free. Major complications are estimated to affect 11.5–40% of patients. Bleeding is the most commonly encountered complication typically at the access site, although it can occur at other locations in the body including intracranial haemorrhage. Incident of major complications correlates with younger age, prolonged treatment and increased decline in fibrinogen level. It is worth mentioning that this high complication rate was extrapolated from studies that used relatively high systemic doses (0.25–0.5 mg/kg/h) for a prolonged period of time.^{Reference Gupta, Leaker and Andrew8,Reference Newall, Browne, Savoia, Campbell, Barnes and Monagle9}