Article Text
PDF
Abstract
Purpose We present a series of cases featuring re-rupture of posterior communicating artery aneurysms in the immediate post-coiling period, where clots obscured the neck of the aneurysm during the procedure. We report on the incidence of re-rupture and discuss possible mechanisms behind this phenomenon.
Methods This study included 462 aneurysms over a 10-year period which were treated with endovascular coiling. 137 of these were ruptured aneurysms. Radiographic and clinical information pertaining to the patients was collected. The immediate post-coiling period was defined as ≤3 days following the procedure.
Results The incidence of re-rupture in the immediate post-coiling period was 1.4% (2/137). Of these two, one received recombinant tissue plasminogen activator and both received acetylsalicylic acid. One patient presented with adjacent hematoma on re-rupture. In both cases, clot obscured the aneurysm neck during the procedure.
Conclusions Re-rupture of intracranial aneurysms in the immediate post-coiling period is an exceedingly rare complication, and the events leading to it have not been discussed at length. Repeat treatment, follow-up imaging and close observation are merited in the setting of re-rupture of a posterior communicating artery aneurysm.
- Aneurysm
- Coil
- Complication
- Hemorrhage
- Thrombolysis
Statistics from Altmetric.com
Introduction
Rupture in the immediate post-coiling period is a rare but extremely serious complication of endovascular treatment for intracranial aneurysms.1 Following subarachnoid hemorrhage (SAH), treatment is considered urgent as re-rupture carries a mortality of approximately 70%. There is an estimated 3–4% chance of re-rupture within the first 24 h, which increases by 2% each subsequent day for the first month.2 Although endovascular coiling represents the standard of care in many institutions for aneurysms suitable for treatment by either this method or microsurgical clipping, there has been little discussion of re-rupture in the immediate post-coiling period. Early re-treatment may prevent further hemorrhage with resultant morbidity or mortality, but is complicated by the presence of coils.
Methods
In a review of 462 aneurysm coilings performed between January 2002 and July 2012 including 138 ruptured aneurysms, we found two cases of aneurysm rupture in the immediate post-coiling period related possibly to clot obscuring the aneurysmal neck during the intervention. Seventy-three of the 462 were posterior communicating artery aneurysms.
Results
Case 1
A woman aged >65 years presented to the emergency room with sudden severe headache, vomiting and a Glasgow Coma Scale score of 14 with no focal neurological deficit. CT angiography (CTA) revealed SAH secondary to rupture of a left posterior communicating artery aneurysm (World Federation of Neurological Societies (WFNS) grade 2, Fisher grade 3). The aneurysm measured 5.3×3 mm with the neck measuring 2.5 mm. The dome-to-neck ratio was 2.1 (figure 1A). She required an external ventricular drain for management of hydrocephalus and was then transferred to our center for coil embolization. The aneurysm was treated using a double microcatheter technique because the first coil would not remain in the aneurysm, possibly due to its relatively wide neck. Multiple bare platinum coils were used to achieve a packing density of 24%. There was stretching of the last coil, which was eventually withdrawn. The coiling was complicated further by small non-occlusive clot formation at the neck of the aneurysm (figure 1B,C) and she was placed on acetylsalicylic acid (ASA). A post-interventional CT showed early resolution of hemorrhage (figure 1D). She remained neurologically stable.
(A) Initial working angle pre-embolization angiogram. (B) Oblique angiogram of left internal carotid artery showing no residual filling of the aneurysm with a small clot near the neck of the aneurysm. (C) Subtracted view of the coil mass following the procedure. (D) CT scan of the head immediately after coiling. (E) CT scan of the head 2 days later, when the patient started vomiting and experienced loss of consciousness, showing an increase in subarachnoid hemorrhage in a similar location.
Three days later she experienced new headache, vomiting and right-sided weakness, and a CT scan of the head showed more SAH (figure 1E). She was taken to the operating room and underwent clipping of the coiled aneurysm (Yasargil, B Braun, Melsungen, Germany). Intraoperative observations suggested the point of rebleeding was probably at the neck of the aneurysm. Her weakness resolved, but she developed focal motor seizures of the right face and upper limb at the time of discharge which were treated with phenytoin.
Case 2
A middle-aged woman presented with slurred speech, right-sided headache and gradually resolving left hemiparesis over 1 h. CT/CTA revealed a right temporal hematoma secondary to saccular right posterior communicating artery aneurysm rupture (WFNS grade 3, Fisher grade 3). The aneurysm measured 7.5×4.7 mm with the neck measuring 2.5 mm. The dome-to-neck ratio was 3 (figure 2A). She was transferred to our center and underwent endovascular coiling of the aneurysm. The aneurysm was treated using multiple bare platinum coils with a packing density of 21%. The coiling was complicated by thrombus formation along the distal and medial aspect of the coil mass, projecting into the lumen of the internal carotid artery (figure 2B,C). This was treated using 4 mg recombinant tissue plasminogen activator (rtPA) through the microcatheter with resolution of the clot. No further coils were placed. She was treated with 325 mg ASA daily.
(A) Initial working angle pre-embolization angiogram. (B) Angiogram right internal carotid artery showing satisfactory occlusion of the aneurysmal done, with clot formed along the lumenally-protruding coils. (C) Subtracted view of the coil mass following the procedure. (D) CT scan of the head 6 h later, when the patient experienced recurrent hemiparesis and reduced level of consciousness, showing new temporal hematoma.
Several hours later her level of consciousness began to deteriorate and her left hemiparesis recurred along with a new finding of pinpoint pupils. CT showed new blood in the temporal hematoma (figure 2D). She underwent craniotomy for evacuation of the hematoma and clipping of the aneurysm (Yasargil clip). Intraoperatively, the re-rupture point appeared to be over the dome rather than the neck of the aneurysm.
She subsequently developed severe vasospasm in the right middle cerebral artery (MCA). She was managed with hypertensive therapy, balloon angioplasty and intra-arterial verapamil, but suffered a new MCA territory infarct and was left with corresponding persistent mild weakness.
Discussion
Rebleeding in the immediate post-coiling period occurs at a rate of 1.4% or less, and is usually due to high-risk aneurysm location, incomplete occlusion or poor clinical condition at the time of treatment.3 These risk factors also apply to mortality, in addition to the presence of intracerebral hematoma and small aneurysm size.
Instances of rebleeding from anterior communicating artery aneurysms after coiling have been reported, with a mortality rate of 100%.3 As our cases show, rebleeding of posterior communicating artery aneurysms may not be fatal but may result in increased morbidity. At our center, these were the only cases of bleeding in the post-coiling period in a series of 138 ruptured aneurysms, giving an incidence of 1.4% for early rebleeding of coiled ruptured aneurysms. Both of these patients presented with SAH from posterior communicating artery aneurysms and had thrombus formation during coiling, for which rtPA was used in one case and ASA in both. Incomplete treatment of the aneurysm has been described as an important risk factor for rebleeding.4 ,5 In the setting of thromboembolic complications, visualization of residual filling is difficult and packing density may help in the guidance of complete treatment in this setting. The calculation of packing density becomes controversial in the presence of pseudoaneurysm, particularly in cases of ruptured aneurysms with associated adjacent hematoma. Some suggest not occluding the pseudoaneurysm.3 Good packing density may not reliably prevent rebleeding, but Sluzewski et al found incomplete occlusion to be less of a risk factor than aneurysm size and position.
Thrombus formation during coiling may prevent complete visualization of residual aneurysm filling, pseudoaneurysms or neck remnants, and may indirectly correlate with early re-hemorrhage in previously ruptured aneurysms. It may also conceal stretching or injury to the aneurysm; one possible explanation for early rebleeding in our first patient could be an undetected tear at the neck from use of the double microcatheter technique which re-hemorrhaged after the clot resolved a few days later. Use of antiplatelet or antithrombotic agents may further predispose to rebleeding, particularly in cases of incomplete aneurysm occlusion.4 Caution is advised in assessing the result of coiling in patients who develop thrombus intraprocedurally as they may re-bleed in the post-intervention period. Risk factors for unexpected clot formation are not known with certainty. This could be multifactorial, with some studies suggesting that iodinated contrast medium used during angiography may react with vascular endothelium to promote clot formation.6 However if this were the case, we would expect to see this complication more often. Thrombus may also be induced by the coil mass at the base of a very wide-necked aneurysm or on coil loops that protrude into the parent vessel.
The incidence of rebleeding is so low that it is difficult to propose any guidelines. However, patients whose treatment is characterized both by low packing density and thromboembolic phenomena should be managed with caution in the postoperative period. However, in our case both of the patients’ aneurysms had a good packing density. It might also be appropriate to modify the triple-H treatment should vasospasm develop in these patients, or to do very early imaging follow-up for timely detection of any reopening of the aneurysm lumen. A non-invasive imaging method such as contrast-enhanced magnetic resonance angiography (CEMRA) should be used for this, as CEMRA has been shown to demonstrate residual filling of a coiled aneurysm better than conventional angiograms.7 This may not prevent extremely early cases of rebleeding such as occurred in our second patient, but could potentially have detected early aneurysm reopening in our first patient.
It may be beneficial to refrain from postprocedural anticoagulation after coiling of a ruptured aneurysm with an adjacent hematoma. Although this is controversial, it has been suggested that the presence of thrombosis during coiling may disguise the presence of pseudoaneurysm on angiography. These pseudoaneurysms can remain undetected following apparent resolution of the thrombosis, and this fragile area can subsequently be ruptured by migrating coils.4 ,8 ,9 Our experience is equivocal with regard to adjacent hematoma as one patient exhibited adjacent temporal hematoma while the other did not have any. However, posterior communicating artery aneurysms do not usually present with adjacent hematoma and thus it is difficult to compare this against related studies. We believe that periprocedural anticoagulation was possibly related to the rebleeding in our patients.
Lastly, the clinical condition prior to intervention has been reported as a dependent risk factor. Not only might particularly unwell patients be inherently prone to bleeding, but they may also not achieve candidacy for surgery for aneurysms that would be favorable for clipping.10
Rebleeding requires immediate surgical intervention. Repeat endovascular coiling may be a viable option if there is significant residual filling of the aneurysm. However, in the setting of rebleeding in the post-coiling period, assessment of residual filling is not straightforward. Improving on the original coiling result may require a new strategy such as balloon remodeling. In the presence of other findings such as intracerebral hematoma, clipping represents the preferred treatment. However, clip placement may be rendered more difficult due to the presence of substantial coil mass at the base of a very recently treated aneurysm.
Conclusions
Re-rupture of recently coiled aneurysms is rare. Incomplete treatment of the aneurysm may be an important risk factor for re-rupture. Visualization of residual filling becomes difficult in the setting of intraoperative thromboembolic complication. Shorter and more frequent follow-up imaging may help in preventing these re-ruptures.
References
Footnotes
-
Contributors Each author contributed equally and meaningfully to the preparation of this manuscript. This reflects both time and effort invested. AAD: patient data acquisition, interpretation and analysis; primary manuscript preparation and editing, critical review and intellectual contribution; approval of the final version. GEP: expert clinical, neurosurgical and neurointerventional care and analysis; manuscript preparation and editing, critical review and intellectual contribution; approval of the final version. JJSS: expert neurointeventional care and analysis; expert radiologlogical interpretation and analysis; manuscript preparation and editing, critical review and intellectual contribution; approval of the final version.
-
Competing interests None.
-
Ethics approval Capital District Health Authority.
-
Provenance and peer review Not commissioned; externally peer reviewed.
-
Data sharing statement We possess a record of all intracranial aneurysms coiled from 2002 to 2008. Data contained therein that are not relevant to the study are available only to the Department of Radiology at the QEII Health Sciences Centre.