Article Text

Download PDFPDF
P-019 Role of collateral circulation in branch vessel occlusion from flow diversion
  1. S Raymond1,
  2. M Koch2,
  3. C Stapleton2,
  4. C Torok3,
  5. A Patel2
  1. 1Radiology, Massachusetts General Hospital, Boston, MA
  2. 2Neurosurgery, Massachusetts General Hospital, Boston, MA
  3. 3St. Paul Radiology, St. Paul, MN


Introduction/Purpose Flow diversion with the Pipeline Embolization Device (PED) often necessitates covering branch vessels. A number of studies suggest a low rate of branch vessel occlusion with only rare clinical complications from these occlusions. We and others hypothesize that branch vessel occlusion is generally clinically silent due to collateral circulation from ECA to ICA anastomoses (e.g. in the case of the ophthalmic artery) or via the circle of Willis.

Materials and Methods We reviewed a consecutive retrospective cohort of 64 patients from 2011–2016, who had branch vessel coverage associated with aneurysm flow diversion. Immediate post-treatment angiography and interval follow-up angiography was evaluated for branch vessel opacification. Branch vessels demonstrated either normal, slow, or absent contrast opacification. Collateral circulation was assessed for all branch vessels with slow or absent flow when selective angiography was available.

Results In our cohort, we identified 106 branch vessels covered by the PED construct in 64 patients. These were primarily anterior circulation branches (99 of 106) and of those, most were ophthalmic arteries (56 of 99). Slow flow was seen in 11 of 106 branches (10%), only 1 of which progressed on follow up to full occlusion. Angiographic occlusion was seen in 11 vessels (10%). Seven patients had new or worsening neurologic deficits, two of which were associated with stent thrombosis. Three patients had new visual deficits although the covered ophthalmic artery remained patent; two patients had new or worsening 6th nerve palsy related to mass effect from coils and/or the thrombosed aneurysm.

The majority of branch vessels with altered flow (slow or absent) had angiographic evidence of collateral circulation (15 of 22, 68%). Altered branch vessel flow was not associated with new or worsening neurologic deficit. In the subset of branch vessels with altered flow, lack of collateral circulation was associated with new or worsening neurologic deficit (p<0.03, Fisher exact test), which in all cases occurred in the setting of PED construct thrombosis. Altered branch vessel flow was not associated with the use of adjunctive coils or the use of more than one PED construct.

Conclusion Branch vessel occlusion is a well-known ramification from flow diversion, but rarely results in clinical deficits. Most patients with altered branch vessel flow (either slow or absent) have distal supply via collateral circulation. In our cohort with altered branch vessel flow, new or worsening neurologic symptoms were associated with absent collateral circulation.

Disclosures S. Raymond: None. M. Koch: None. C. Stapleton: None. C. Torok: None. A. Patel: 2; C; Medtronic, Penumbra.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.