Discussion

The phrenic nerve is a bilateral nerve that originates in the neck and descends through the thorax to the diaphragm. ^{Reference El-Boghdadly, Chin and Chan1} The pericardiophrenic artery is a branch of the internal mammary artery that runs adjacent to the phrenic nerve, near the pericardium. In the current case, arterial bleeding was identified to be from the pericardiophrenic artery. ^{Reference Henriquez-Pino and Prates2}

A nerve injury can result from either compression to the haematoma or direct trauma from the needle or catheter. Compression injuries secondary to bleeding usually improve with resolution of the associated haematoma, while a direct injury may take much longer to improve. ^{Reference Depierraz, Essinger, Morin, Goy and Buchser3} In our patient, haemorrhage caused by the rupture of the pericardiophrenic artery compressed the phrenic nerve, which resulted in transient phrenic nerve palsy.

Transient phrenic nerve palsy induced by cardiac catheterisation is rare. However, there are some reports about catheterisation-induced phrenic nerve palsy. Catheterisation-related phrenic nerve palsy was well-known complications for pulmonary vein isolation of radiofrequency catheter ablation. ^{Reference Abbadessa, Lavorgna and Cirillo4} Phrenic nerve paralysis is also a rare complication associated with central venous catheterisation or cardiac electronic device implantation due to the anatomical position between the central vein and the phrenic nerve. ^{Reference Depierraz, Essinger, Morin, Goy and Buchser3,Reference Godcharles, Safarzadeh, Oliver, Roman and Al-Kouatly5,Reference López-Gil, Fontenla, Juliá, Parra and Arribas6}

It may be difficult to detect phrenic nerve palsy under general anaesthesia because the symptoms will be masked by mechanical respiration. A chest X-ray or fluoroscope could also be useful for detecting phrenic nerve palsy. In unilateral phrenic nerve palsy as in our case, patients are usually asymptomatic, have a good prognosis, and do not always need treatment. ^{Reference Abbadessa, Lavorgna and Cirillo4} However, some patients experience severe heart failure or severe respiratory failure, are sometimes symptomatic, and need oxygen therapy or respiratory support.

Generally, coil embolisation is a better choice to stop bleeding from a ruptured artery. In our case, bleeding from the perforated site was minimal, and the patient’s vital signs were stable during careful observation. Thus, we continued the diagnostic catheterisation without performing coil embolisation.

In this case, it was difficult to access the left-modified Blalock–Taussig shunt because the entrance of the shunt was very close to that of the left internal mammary artery. We were not able to immediately recognise the wrong pathway, and the guidewire was inserted into a branch of the left internal mammary artery, although we had checked the lateral view. To avoid these complications, we suggest that operators perform angiography without hesitation if they lose their way.

In conclusion, phrenic nerve palsy is a rare complication associated with cardiac catheterisation. Cardiologists should keep in mind that the phrenic nerve runs adjacent to the pericardiophrenic artery when they access this artery.