Percutaneous closure of patent foramen ovale to prevent recurrent cerebral embolic events

5 Safety

This section describes 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.

5.1 The occurrence of any serious adverse event was reported in 23% (numbers not stated), 17% (68/402), and 21% (43/204) of patients treated by percutaneous closure and 22% (numbers not stated), 17% (76/458), and 18% (37/210) of patients treated by medical therapy in the randomised controlled trials (RCTs) of 980, 909, and 414 patients respectively (p=0.7, 0.9 and 0.4 respectively).

5.2 Death related to the procedure (not further defined) was reported in 0.1% (95% confidence interval [CI] 0 to 0.3) of patients in the meta‑analysis of 58 studies (8185 patients treated by percutaneous patent foramen ovale closure and 2142 patients treated by medical therapy).

5.3 Pericardial effusion or tamponade was reported in 0.3% (95% CI 0 to 0.6) of patients in the meta-analysis of 58 studies. Details of treatment and outcome were not provided.

5.4 Perforation of the left atrium and cardiac perforation (not further described) were each reported in 1 patient treated by percutaneous closure in the RCTs of 909 and 980 patients respectively. Vascular surgical repair (not further defined) was reported in 1 patient in the RCT of 909 patients.

5.5 Device embolisation or malposition (not further described) was reported in 0.4% (95% CI 0.2 to 0.7) of patients in the meta-analysis of 58 studies.

5.6 Bleeding (described as serious or major) was reported in 3% (10/378) and 0.5% (1/204) of patients treated by percutaneous closure and 1% (4/374 and 3/210) of patients treated by medical therapy in the RCTs of 909 and 414patients respectively. Major bleeding that was considered to be device- or procedure-related was reported in 0.4% (2/499) of patients treated by percutaneous closure in the RCT of 980 patients.

5.7 Air embolism was reported in 0.6% (95% CI 0.2 to 1.0) of patients in the meta-analysis of 58 observation studies (not further described).

5.8 Infective or bacterial endocarditis that was considered to be device- or procedure-related was reported in 1 patient in the RCT of 980 patients (no further information was given).

5.9 Cardiac thrombus, together with deep vein thrombosis and pulmonary embolism, was detected 4 months after the closure procedure in 1 patient in the RCT of 980 patients.

5.10 A fistula between the aortic root and right atrium was described in 1 patient in a case report. There was incomplete patent foramen ovale obliteration with residual shunting in both directions 6 months after the closure procedure. The device was removed through a mini-thoracotomy and the fistula was closed.

5.11 Atrial fibrillation was reported in 6% (23/402) of patients treated by percutaneous patent foramen ovale closure and 0.7% (3/458) of patients treated by medical therapy in the RCT of 909 patients (p<0.001). In the closure group, 61% (14/23) of the patients with atrial fibrillation developed it within 30 days of the procedure; atrial fibrillation was transient in 17 patients and persistent in 6 patients. Serious atrial fibrillation (not further defined) was reported in 1% (2/204 and 2/210) of patients in each group in the RCT of 414 patients.

5.12 The specialist advisers listed additional adverse events reported in the literature as embolism of clots attached to the device, device erosion, and transient worsening of migraine.