Mechanical clot retrieval for treating acute ischaemic stroke

Death within 7 days was reported in 12% (27 of 233) of patients with acute ischaemic stroke treated by intra‑arterial treatment (intra‑arterial thrombolysis, mechanical clot retrieval, or both) with usual care and in 12% (33 of 267) of patients treated by usual care alone in a randomised controlled trial (RCT) of 500 patients. Death within 30 days was reported in 19% (44 of 233) and 18% (49 of 267) of patients respectively, in the same study. Death within 7 days was reported in 10% (10 of 103) of patients treated by mechanical clot retrieval with medical therapy and in 5% (5 of 103) of patients treated by medical therapy alone in a RCT of 206 patients (risk ratio 2.0, 95% confidence interval [CI] 0.7 to 5.6). Death within 1 week was reported in 6% (6 of 106) of patients in a case series of 106 patients.

A systematic review of 8 RCTs (2,423 patients) reported no significant difference in the rate of symptomatic intracerebral haemorrhage between the treatment groups (odds ratio 1.03, 95% CI 0.71 to 1.49, p=0.88). Symptomatic intracerebral haemorrhage was reported in 8% (18 of 233) of patients with acute ischaemic stroke treated by intra‑arterial treatment (intra‑arterial thrombolysis, mechanical clot retrieval, or both) with usual care and in 6% (17 of 267) of patients treated by usual care alone in the RCT of 500 patients in the systematic review. Hemicraniectomy was done in 6% (14 of 233) and 5% (13 of 267) of patients respectively.

Symptomatic intracerebral haemorrhage was reported in 4% (6 of 165) of patients treated by mechanical clot retrieval with standard care and in 3% (4 of 150) of patients treated by standard care alone in an RCT of 315 patients in the systematic review (rate ratio 1.2, 95% CI 0.3 to 4.6); 1 patient in the intervention group was treated by hemicraniectomy.

Symptomatic intracranial haemorrhage at 27 hours was reported in no patients treated by intravenous tissue plasminogen activator (tPA) with mechanical clot retrieval and in 3% (3 of 97) of patients treated by intravenous tPA alone (p=0.12) in a RCT of 196 patients in the systematic review. Subarachnoid haemorrhage was reported in 4% (4 of 98) and 1% (1 of 97) of patients respectively (p=0.37) and parenchymal haematoma in 5% (5 of 98) and 7% (7 of 97) of patients respectively, in the same study.

Symptomatic intracranial haemorrhage diagnosed using the SITS‑MOST criteria was reported in 2% (2 of 103) of patients both in the mechanical clot retrieval with medical therapy group and in the patients treated by medical therapy alone in the RCT of 206 patients in the systematic review. Subarachnoid haemorrhage was reported in 5% (5 of 103) and 2% (2 of 103) of patients, respectively, and parenchymal haematoma was reported in 6% (6 of 103) of patients in both groups, in the same study.

Symptomatic intracranial haemorrhage was reported in 1% (1 of 89) of patients treated by mechanical clot retrieval using a stent retriever and in 11% (6 of 55) of patients treated by mechanical clot retrieval using a coil retriever in a RCT of 113 patients (p=0.013). In the same study, symptomatic subarachnoid haemorrhage was reported in 1% (1 of 89) and 7% (4 of 55) of patients respectively (p=0.07).

New ischaemic stroke in a different vascular territory was reported in 6% (13 of 233) of patients with acute ischaemic stroke treated by intra‑arterial treatment (intra‑arterial thrombolysis, mechanical clot retrieval, or both) with usual care and in less than 1% (1 of 267) of patients treated by usual care alone in the RCT of 500 patients (p<0.001).

Large or malignant middle cerebral artery stroke was reported in 5% (8 of 165) of patients treated by mechanical clot retrieval with standard care and in 11% (16 of 150) of patients treated by standard care alone in the RCT of 315 patients (rate ratio 0.3, 95% CI 0.1 to 0.7); 1 patient in the control group was treated by hemicraniectomy.

Symptomatic ischaemic stroke was reported in 3% (3 of 89) of patients treated by mechanical clot retrieval using a stent retriever and in 13% (7 of 55) of patients treated by mechanical clot retrieval using a coil retriever in the RCT of 113 patients (p=0.04).

Embolisation into new vascular territories was reported in 9% (20 of 233) of patients with acute ischaemic stroke treated by intra‑arterial treatment (intra‑arterial thrombolysis, mechanical clot retrieval, or both) with usual care in the RCT of 500 patients.

Asymptomatic embolisation into a different vascular territory occurred in 6% (2 of 35) of patients treated by mechanical clot retrieval in an RCT of 70 patients. Distal embolisation in a different territory was reported in 5% (5 of 103) of patients treated by mechanical clot retrieval with medical therapy in the RCT of 206 patients.

Vessel dissection and vessel perforation were reported in 2% (4 of 233) and 1% (2 of 233) of patients, respectively, with acute ischaemic stroke treated by intra‑arterial treatment (intra‑arterial thrombolysis, mechanical clot retrieval, or both) with usual care in the RCT of 500 patients. Perforation of the middle cerebral artery was reported in 1 patient treated by mechanical clot retrieval in the RCT of 315 patients.

Arterial dissection and arterial perforation were reported in 4% (4 of 103) and 5% (5 of 103) of patients treated by mechanical clot retrieval in the RCT of 206 patients.

Dissection of the access vessel was reported in 4% (4 of 106) of patients in a case series of 106 patients and dissection of the target vessel was reported in 5% (5 of 106): all 5 patients also had subarachnoid haemorrhage.

Vessel dissection was reported in 4% (4 of 89) of patients treated by mechanical clot retrieval using a stent retriever and in 1 patient treated by mechanical clot retrieval using a coil retriever in the RCT of 113 patients (p=0.65): 3 patients were managed conservatively, 1 patient was treated by balloon angioplasty and 1 patient was treated by stent placement.

Malignant cerebral oedema was reported in 11% (11 of 103) of patients with acute ischaemic stroke treated by mechanical clot retrieval with medical therapy and in 10% (10 of 103) of patients treated by medical therapy alone in the RCT of 206 patients (risk ratio 1.1, 95% CI 0.5 to 2.5): 3 patients in the intervention group and 6 patients in the control group were treated by decompressive hemicraniectomy.

Unintended stent detachment was reported in 3% (3 of 106) of patients in the case series of 106 patients. Technical difficulty with the device was reported in 10% (9 of 89) of patients treated by mechanical clot retrieval using a stent retriever and in 7% (4 of 55) of patients treated by mechanical clot retrieval using a coil retriever in the RCT of 113 patients (p=0.77). A trapped cerebral thrombectomy device was reported in a case report: the patient was treated by carotid endarterectomy to remove a carotid stent with the trapped clot retrieval device inside. The patient showed no clinical deterioration and was discharged from hospital 6 days later.

In addition to safety outcomes reported in the literature, specialist advisers are asked about anecdotal adverse events (events which they have heard about) and about theoretical adverse events (events which they think might possibly occur, even if they have never done so). For this procedure, specialist advisers listed the following anecdotal adverse events: vasospasm, distal embolisation beyond site of initial thrombus, worsened stroke, groin haemorrhage and retroperitoneal haemorrhage. They considered that the following were theoretical adverse events: device malfunction (degeneration of internal structure), reperfusion injury and contrast allergy.