2 The procedure
2.1.1 Atrial fibrillation is the irregular and rapid activation of the atria. It is caused by uncoordinated electrical stimulation of the atrial muscle. Atrial fibrillation may be classified as paroxysmal, persistent or permanent. Patients may be asymptomatic or have symptoms such as palpitations, fatigue and chest pain.
2.1.2 Patients considered to be at high risk of embolic stroke from thrombus in the fibrillating left atrium are usually treated with anticoagulation therapy. Medical treatment for atrial fibrillation includes drugs to control heart rate or to help maintain a normal cardiac rhythm after cardioversion. Ablation procedures, designed to disrupt abnormal conduction pathways, may be used when drug therapy is either not tolerated or is ineffective. Several methods are available to deliver cardiac ablation including cryoablation with or without a balloon.
2.2.1 Percutaneous balloon cryoablation for pulmonary vein isolation in atrial fibrillation helps maintain normal heart rhythm by isolating the electrical impulses originating in the pulmonary veins that are thought to be responsible for 'triggering' atrial fibrillation.
2.2.2 With the patient under general anaesthesia, or using local anaesthesia and sedation, catheters are introduced percutaneously through one or both femoral veins. One or more electrode catheters are placed in the heart to allow pacing. An additional electrode catheter is placed in a vein or the heart to allow stimulation of the phrenic nerve. One or two sheaths are advanced into the left atrium transseptally. A multipolar circular mapping catheter (to record electrical signals from the pulmonary vein ostia) and the balloon cryoablation catheter are passed through the two sheaths. The balloon cryoablation catheter is placed at one of the pulmonary vein ostia and the balloon is inflated to allow continuous contact between the balloon and the atrial myocardium. Good contact is confirmed fluoroscopically by injecting contrast into the pulmonary vein through the lumen of the balloon catheter.
2.2.3 When the balloon catheter has been positioned satisfactorily, it is cooled in bursts of approximately 4 minutes, to achieve circumferential isolation of the cells responsible for the arrhythmia. This is assessed using the mapping catheter. Each of the pulmonary veins is treated in the same way, until all are electrically isolated.
Sections 2.3 and 2.4 describe efficacy and safety outcomes from the published literature that the Committee considered as part of the evidence about this procedure. For more detailed information on the evidence, see the overview.
2.3.1 A systematic review with 1308 patients (23 studies) reported acute procedural success (defined as complete isolation of all targeted pulmonary veins) in 99% (95% confidence interval [CI] 98 to 99) of patients (19 studies; n = 924; I2 = 0% [no heterogeneity]) (timing of assessment unclear). A comparative case series of 94 patients treated by balloon cryoablation (n = 30), radiofrequency ablation (n = 29) or robotically assisted radiofrequency ablation (n = 35) reported procedure success (defined as 'no atrial tachyarrhythmias [symptomatic or asymptomatic] lasting ≥ 30 seconds, identified on surface electrocardiogram [ECG], Holter monitoring or 7-day ECG recording') without anti-arrhythmic drugs in 66% (23), 66% (19) and 67% (20) of patients respectively at a mean follow-up of 13 months (p = 0.625) (denominator not reported).
2.3.2 The systematic review reported 1-year freedom from atrial fibrillation in 73% of patients (95% CI 69 to 77) with paroxysmal atrial fibrillation after a 3-month blanking period (time period during which transient episodes of arrhythmia were not considered recurrences) (5 studies; n = 519; I2 = 0% [no heterogeneity]); in 60% of patients (95% CI 55 to 66) with paroxysmal atrial fibrillation without a 3-month blanking period (3 studies; n = 316; I2 = 0% [no heterogeneity]); and in 45% of patients (95% CI 32 to 58) with persistent atrial fibrillation after a 3-month blanking period (2 studies; n = 62; I2 = 0% [no heterogeneity]).
2.3.3 In a case series of 346 patients treated by balloon cryoablation (treatment completed by cryoablation catheter if isolation was not achieved), sinus rhythm was maintained without anti-arrhythmic drugs in 74% (159, denominator not reported) of patients with paroxysmal atrial fibrillation and 42% (13/31) of patients with persistent atrial fibrillation (follow-up not stated).
2.3.4 A case series of 141 patients reported recurrence of atrial fibrillation in 51% (71/139) of patients after 1 procedure (follow-up 457 days).
2.3.5 A non-randomised comparative case series of 177 patients treated by balloon cryoablation or radiofrequency ablation reported that 14% (17) and 23% (12) of patients respectively needed retreatment for recurrent atrial fibrillation (denominators and timing of events not reported; mean follow-up 13 months).
2.3.6 The Specialist Advisers listed key efficacy outcomes as electrical isolation of 'all pulmonary veins' or 'all 4 pulmonary veins', procedure duration, freedom from atrial fibrillation, avoidance of repeat procedures and reduced use of anti-arrhythmic drugs.
2.4.1 The case series of 346 patients reported periprocedural pericardial tamponade in 2 patients, both successfully treated by pericardial drainage and without the need for surgery. A comparative case series of 133 patients reported pericardial effusion within 24 hours in 11% (5/46) of patients treated by balloon cryoablation and in 16% (14/87) of patients treated by radiofrequency ablation (drainage was needed in 1 patient in each group; all the other effusions resolved spontaneously).
2.4.2 The case series of 346 patients reported right phrenic nerve injury in 8% (26/346) of patients during balloon cryoablation of the right superior pulmonary vein. Of these, 2 resolved during the procedure and full function recovered in all patients during follow-up of less than 1 year.
2.4.3 A comparative case series of 74 patients treated by balloon cryoablation (n = 67) or cryocatheter (n = 7) reported oesophageal ulceration identified by endoscopy in 17% (6/35) of patients and 0/7 patients respectively within 1 week of the procedure. All were asymptomatic and resolved within 3 months.
2.4.4 Stroke or transient ischaemic attack was reported in less than 1% of patients (4/1241) in the systematic review. Three of the 4 cerebrovascular events were observed in the same study and resolved within 24 hours.
2.4.5 In the case series of 141 patients, 2 patients had haemoptysis during the first month after pulmonary vein isolation. This resolved after temporary cessation of oral anticoagulation therapy.
2.4.6 The Specialist Advisers listed adverse events reported in the literature as pulmonary vein stenosis, groin haematoma at venous entry site and pseudoaneurysm. They reported anecdotal adverse events of air embolus and phrenic nerve injury, and they considered theoretical adverse events to include death, atrio-oesophageal fistula, permanent phrenic nerve palsy, damage to structures anatomically close to pulmonary veins and deep vein thrombosis.
2.5.1 The Committee noted the advances in the understanding of the causes of atrial fibrillation and acknowledged that this procedure is likely to be more effective in paroxysmal than persistent atrial fibrillation.
2.5.2 The Committee noted that asymptomatic cerebral lesions have been shown after percutaneous balloon cryoablation for pulmonary vein isolation in atrial fibrillation but that it is unknown whether these have clinical consequences.