Intracranial aneurysms can present with cranial nerve

Transcription

1 J Neurosurg 121: , 2014 AANS, 2014 Resolution of cranial neuropathies following treatment of intracranial aneurysms with the Pipeline Embolization Device Clinical article Karam Moon, M.D., Felipe C. Albuquerque, M.D., Andrew F. Ducruet, M.D., R. Webster Crowley, M.D., and Cameron G. McDougall, M.D. Division of Neurological Surgery, Barrow Neurological Institute, St. Joseph s Hospital and Medical Center, Phoenix, Arizona Object. Intracranial aneurysms, especially those of the cavernous segment of the internal carotid artery (ICA), can present with cranial nerve (CN) palsies. The Pipeline Embolization Device (PED) has demonstrated safety and efficacy in the treatment of cerebral aneurysms by flow diversion, but little data exist reporting the outcomes of cranial neuropathies following treatment with the device. Methods. The prospectively maintained Barrow Neurological Institute s endovascular database was reviewed for all patients treated with the PED after presenting with one or more CN palsies secondary to a cerebral aneurysm since May Patient charts and digital subtraction angiograms were reviewed to report clinical and angiographic outcomes. Only patients with clinical follow-up were included in the analysis. Results. A total of 127 patients were treated with the PED at the authors institution after FDA approval. Twentytwo patients presented with cranial neuropathies, for initial inclusion in this study. Of these, 20 had sufficient followup for analysis. Cranial neuropathies included those of CN II, III, V, and VI, with presenting symptoms of diplopia, decreased visual acuity, and facial numbness and/or pain. Thirteen lesions were cavernous segment ICA aneurysms, whereas the remainder included supraclinoid and petrous segment ICA, posterior communicating artery, and basilar trunk aneurysms. At an average clinical follow-up of 9.55 months, 15 patients (75%) had resolution or significant improvement of their cranial neuropathies, and the remaining 5 had stable symptoms. Of the 18 patients with angiographic follow-up, 12 (66.7%) demonstrated complete obliteration or small neck residual, whereas 6 (33.3%) had residual lesion. Patients with complete or near-complete obliteration of their lesion were significantly more likely to demonstrate symptomatic improvement at follow-up (p = 0.009). Two patients with persistent symptoms were eventually treated with microsurgical bypass. Transient complications in this series included 6 (30%) extracranial hemorrhagic complications related to dual-antiplatelet therapy, all of which were managed medically. There was 1 delayed right ICA occlusion following retreatment that led to microsurgical bypass. Conclusions. Intracranial aneurysms presenting with one or more CN palsies show a high rate of clinical improvement after treatment with the PED. Clinical outcomes must be weighed against the risks and challenges faced with flow diverters. Further research is warranted for patients whose symptoms do not respond optimally to device placement. ( Key Words Pipeline Embolization Device cranial nerve palsy cerebral aneurysm vascular disorders Abbreviations used in this paper: CN = cranial nerve; DSA = digital subtraction angiography; ICA = internal carotid artery; MCA = middle cerebral artery; MRA = MR angiography; PCoA = posterior communicating artery; PED = Pipeline Embolization Device; STA = superficial temporal artery. J Neurosurg / Volume 121 / November 2014 Intracranial aneurysms can present with cranial nerve (CN) palsies, or cranial neuropathies, secondary to either mass effect or flow-related dynamic effects. Commonly reported deficits include those of CN II, III, IV, V, VI, and VII, although frequency varies by location of the lesion. 3,5,6,8,15,22 Cavernous segment ICA aneurysms, in particular, often present with intolerable symptoms, such as diplopia and severe headaches invoking their treatment, despite an otherwise relatively benign natural history. 11,19 Flow diverters have gained significant traction in the treatment of large or complex intracranial aneurysms not amenable to other endovascular or microsurgical modalities, with a recent meta-analysis quoting an occlusion rate of 76% and a total procedure-related complication rate of 9%. 2 Whereas microsurgical treatment of large or giant cavernous segment ICA aneurysms has typically involved trapping and bypass, flow diversion has made the overall treatment of these lesions less invasive and more safe. 1085

2 K. Moon et al. The Pipeline Embolization Device (PED) (ev3) has been the most studied and used flow diverter stent on the market, but little information and data exist regarding its effects on cranial neuropathies. We report follow-up on this subset of patients, from our larger institutional experience with the PED, with an emphasis on the correlation between clinical outcomes and angiographic findings. Methods Data Collection Since use of the PED was approved by the FDA in May 2011, all patients who undergo treatment of ruptured or unruptured aneurysms with the PED at our institution are entered into a prospectively maintained database containing basic demographic data, clinical presentation, aneurysm characteristics, treatment details, complications, and clinical and/or radiographic follow-up. Following institutional review board approval, a retrospective review of this database was performed to identify patients with an intracranial aneurysm treated with the PED after presenting with one or more CN deficits. Of 22 patients who met these criteria, 20 had medical records with clinical follow-up sufficient for analysis and inclusion in this study. Cranial nerve palsies were included only if attributable to the aneurysm of interest. All preoperative cranial neuropathies were assessed by performing a detailed history and physical examination. Neuroophthalmological evaluation was not routinely used, although 5 patients underwent assessment for confirmation of their cranial neuropathies, 2 of whom had established care with an ophthalmologist prior to treatment. Patients who did not respond symptomatically to PED placement underwent a formal workup; this was recommended in 3 patients postprocedure for further follow-up of their deficits. Treatment With the PED and Coil Embolization All aneurysms were treated with the PED construct, with or without adjunctive coil embolization. All PEDs were delivered through a in Marksman (ev3 Endovascular) or XT27 (Stryker) microcatheter. When coil embolization was performed simultaneously, a microcatheter was jailed in the aneurysm prior to PED insertion, and coil embolization commenced following PED delivery. The decision to perform adjunctive coil embolization was made at the time of treatment by the principal neurosurgeon. Patients underwent loading with and were started on 325 mg aspirin and 75 mg clopidogrel daily up to 1 week prior to the procedure; those who did not undergo preloading received loading during the procedure and were treated prophylactically with abciximab immediately following device placement. All patients were maintained on a dualantiplatelet regimen until 6- to 12-month follow-up unless they had a specific contraindication to one agent or the other, at which point they were transitioned to aspirin alone if no significant in-stent stenosis was noted. Response to aspirin and clopidogrel was assessed using platelet inhibition assays (Accumetrics). Adequate clopidogrel response was defined as P2Y12 Reaction Units (PRU) < 187, or > 40% inhibition value. One patient was noted to be a nonresponder to clopidogrel, and this individual was maintained on prasugrel as an alternative. Clinical and Angiographic Follow-Up Both clinical symptoms and physical examination findings were assessed preoperatively and at follow-up. Both clinical and angiographic follow-up were initially carried out at 6 months postoperatively, the first time point at which treatment effects on cranial neuropathies were noted and assessed. Clinical response was classified as resolved, improved, stable, or worse, based on both symptomatic profile and physical examination findings. Patients were classified as resolved only if they were asymptomatic, with no physical examination findings of cranial neuropathy, whereas they were classified as improved if they claimed improvement of symptoms but still displayed residual signs of mild cranial neuropathy on examination. Aneurysm obliteration at follow-up was categorized as complete or, in cases with a small neck residual, near complete (> 95% obliteration). Patients were followed clinically, and further radiographic follow-up was carried out with MR angiography (MRA). Patients underwent further angiographic workup if they experienced recurrent or worsening symptoms, or if their initial angiographic follow-up demonstrated residual aneurysm or construct stenosis. If any patients required retreatment, similar data were collected at the time of retreatment. Statistical Analysis Descriptive analyses were performed for all variables of interest. Chi-square analyses were used to understand the association of variables with categorical outcomes, and ANOVAs were used for continuous variables. Statistical analyses were performed using SPSS version 21. The p values were 2-tailed and considered significant at < Results Patient and Aneurysm Characteristics There were 20 of 22 patients with sufficient clinical follow-up for analysis; of the remaining 2 patients, one was lost to follow-up due to the patient s wishes to transfer care for geographical reasons, and the other was awaiting follow-up at the time of this report (Tables 1 and 2). The mean age for patients in this cohort was 66.3 years (range years), and there were 15 women and 5 men. Several patients were treated with the PED for recurrence of a previously treated lesion; 5 had previous clipping or coil embolization, whereas 1 patient had undergone a craniotomy for exploration and 1 had been treated with the Gamma Knife for suspected trigeminal neuralgia. Cranial nerve deficits on presentation included those of CN II, III, V, and VI. The most common symptom exhibited was diplopia; others included decreased visual acuity and facial pain or numbness. Headache and chemosis were also seen as common presenting symptoms, but were not classified as cranial neuropathies. The average symptomatic time interval before treatment was days (range 1 day 5 years). The average aneurysm size was 17.9 mm (range J Neurosurg / Volume 121 / November 2014

3 Pipeline device for aneurysms with cranial neuropathies TABLE 1: Characteristics of 20 patients treated with the PED after presenting to the BNI with an intracranial aneurysm and associated cranial neuropathy* Characteristic No. (%) sex male 5 (25) female 15 (75) previous Tx clip 1 (5) coil 5 (25) exploration 1 (5) Gamma Knife 1 (5) symptoms diplopia 16 (80) visual field/acuity 3 (15) facial numbness/pain 4 (20) other (HA) 7 (35) aneurysm location cavernous segment 13 (65) paraclinoid segment 4 (20) petrous segment 1 (5) PCoA 1 (5) basilar trunk 1 (5) * BNI = Barrow Neurological Institute; HA = headache; Tx = treatment. mm), treated with a mean of 2.35 devices (range 1 11). The majority of lesions were cavernous segment ICA aneurysms (n = 13). The remainder included paraclinoid or ophthalmic segment ICA (n = 4), petrous segment ICA (n = 1), PCoA (n = 1), and basilar trunk (n = 1). Device sizes used ranged from 3.5 to 5 mm in diameter and 16 to 35 mm in length. Six patients (30%) underwent adjunctive coil embolization at the time of initial treatment, although 2 more had coils placed at time of retreatment. When cohort characteristics were analyzed by clinical outcome, no significant differences were seen with regard to mean age, aneurysm size, duration of symptoms, or number of CNs involved (Table 3). In a distribution of clinical outcomes by specific CN involved, no significant associations were noted (Table 4). Clinical and Angiographic Follow-Up The average clinical follow-up time was 9.55 months (range 6 28 months). All but 2 patients (90%) had undergone follow-up angiography by the time of this subset analysis. Nine (45%) had complete obliteration, 3 (15%) had near-complete obliteration or small neck remnant, and 6 (30%) had residual aneurysm dome filling. Of all patients with angiographic follow-up, 12 of 18 (66.7%) had complete or near-complete obliteration. Fifteen patients (75%) reported improvement or resolution of their cranial neuropathies at their last clinical follow-up, whereas 5 (25%) had stable or persistent symptoms (Table 5). Five patients had transient worsening of their symptoms, including 3 cases of retroorbital pain, 1 J Neurosurg / Volume 121 / November 2014 case of proptosis, and 1 of worsened diplopia, all of which resolved over a few weeks with a tapered regimen of hydrocortisone. All patients with symptomatic improvement were noted as such at their 6-month follow-up. Of the patients with improvement in their symptoms, 10 of 15 (66.7%) had complete or near-complete obliteration of their aneurysm, whereas 3 (20%) had residual dome filling, and 2 had not yet undergone follow-up angiography. Of patients with stable symptoms, 3 of 5 (60%) had residual aneurysm, whereas 2 (40%) had complete or near-complete obliteration. One patient was noted to have recurrence of his symptoms at 7 months postprocedure, associated with residual lesion (see Retreatment for Residual Aneurysm). When angiographic outcomes were dichotomized, the proportion of patients classified as having complete or near-complete obliteration were noted to have a significantly higher rate of symptomatic improvement or resolution (84.6%) compared with those with residual aneurysm (15.4%) (p = 0.009). The odds of symptomatic improvement at follow-up were 22 times higher for the group with complete or near-complete obliteration (OR 22; 95% CI ). Interestingly, of the 6 patients who underwent adjunctive coil embolization, 5 had improvement of their symptoms. Angiographic outcomes for this subset included 4 patients with complete or near-complete obliteration and 2 with residual aneurysm. The single patient with stable symptoms following PED and coil placement was noted to have an angiographic cure at follow-up. Retreatment for Residual Aneurysm Four patients (20%) underwent retreatment for residual aneurysm. Two of these patients underwent placement of additional devices with adjunctive coil embolization due to persistent symptoms; one of them was lost to follow-up and the other continued to have symptoms and residual lesion on last follow-up. The remaining 2 patients ultimately underwent microsurgical bypass. One had experienced improvement of her cranial neuropathies after initial treatment but still had severe daily headaches despite multiple retreatments (see Case 12 under Illustrative Cases). The other also had initial improvement but presented with an occluded parent ICA and recurrence of his symptoms approximately 1 month after retreatment for residual aneurysm. The latter patient demonstrated a symptomatic perfusion deficit and underwent an occipital artery middle cerebral artery (MCA) bypass, resulting in immediate improvement of his hemiparesis, as well as delayed resolution of his cranial neuropathies. Illustrative Cases Case 4 A 47-year-old woman presented with a 3-day history of acute-onset headache and double vision. She was noted on examination to have left-sided CN III and VI palsies, and imaging demonstrated a giant 30-mm cavernous segment ICA aneurysm (Fig. 1A and B). She was considered for clip occlusion with bypass grafting, but parent vessel sacrifice was not initially desirable given her young age. 1087

4 K. Moon et al. TABLE 2: Detailed characteristics, complications, and outcomes in 20 patients treated with the PED after presenting to the BNI with an intracranial aneurysm and associated cranial neuropathy between May 2011 and June 2013* Case No. Age (yrs), Sex Prior Tx Location Lesion Size (mm) Cranial Neuropathy Coil Embo Complications FU (mos) Clinical Outcome Obliteration Re-Tx Bypass 1 44, M SAC embo cavernous 19 CN VI intraop OphA occl, recanalized w/ abciximab 6 improved, mild ROP complete 2 72, F SAC embo cavernous 16 CN VI 6 improved, mild ROP complete 3 74, F SAC embo ophthalmic 10 partial CN III incomplete opening of PED; retro hemorrhage, easy bruising 6 improved, mild ROP complete 4 47, F cavernous 30 partial CN III & VI 28 resolved CN III & VI palsy, persistent HA 5 28, M cavernous 40 CN VI, V2/V3 (numbness) alopecia after 1st Tx, asymp rt ICA occl at 9-mo FU 9 improved symptoms initially, recurrent at 7 mos residual at 18 & 24 mos yes residual at 6 mos yes 6 83, F cavernous 20 CN VI groin hematoma, self-limited 7 stable OP residual at 6 mos 7 56, F cavernous 13 CN VI multi episodes of epistaxis, 6 improved, mild ROP complete self-limited 8 68, F cavernous 10 partial CN III 13 stable OP residual at 6 mos residual 9 81, F paraclinoid 15 CN II (field cut & decreased acuity) yes 16 stable field cut & decreased acuity 10 79, F cavernous 20 CN VI yes proximal endoleak; retro hematoma after FU angio 11 66, F exploratory craniotomy 19 resolved CN VI palsy near-complete ophthalmic 14 CN II 8 improved visual acuity near-complete 12 64, M petrous 16 CN II (field cut), V (facial numbness), & VI yes 18 improved, mild residual facial numbness & OP complete 13 75, F cavernous 19 CN VI 6 improved, mild ROP complete 14 85, F cavernous 25 partial CN III 6 improved, mild ROP no DSA FU 15 74, F cavernous 18 CN VI GI bleed 1 mo postop 4 improved, mild ROP no DSA FU 16 64, M cavernous 22 partial CN III GI bleed postop 7 resolved CN III palsy complete 17 59, M cavernous & CCF 18 72, F microsurgical clipping 9 partial CN III, VI, V1/V2 (numbness) PCoA 9 partial CN III (ptosis only) yes 7 improved, mild residual facial numbness & OP complete 6 improved ptosis near-complete 19 75, F Gamma Knife basilar trunk 20 CN V2/V3 (TN) yes 7 improved facial pain residual 20 60, F coil embo paraclinoid 12 partial CN III yes mild proximal migration of 6 stable OP complete PED at FU * angio = angiography; asymp = asymptomatic; CCF = carotid-cavernous fistula; embo = embolization; FU = follow-up; GI = gastrointestinal; multi = multiple; occl = occlusions; OP = ophthalmoparesis; OphA = ophthalmic artery; retro = retroperitoneal; ROP = residual ophthalmoparesis; SAC = stent-assisted coil; TN = trigeminal neuralgia. Managed conservatively J Neurosurg / Volume 121 / November 2014

5 Pipeline device for aneurysms with cranial neuropathies TABLE 3: Comparison of cohort characteristics by clinical outcome Characteristic Mean ± SD p Value age in yrs 0.83 resolved 64.0 ± 13.1 improved 68.2 ± 11.4 stable 64.0 ± 22.2 aneurysm size in mm 0.40 resolved 21.5 ± 6.6 improved 15.8 ± 5.1 stable 19.4 ± 12.1 symptom duration in days 0.36 resolved 13.7 ± 10.3 improved ± stable ± no. of CNs 0.90 resolved 1.25 ± 0.5 improved 1.36 ± 0.8 stable 1.2 ± 0.4 TABLE 5: Clinical and angiographic outcomes in 20 patients treated with the PED after presenting to BNI with an intracranial aneurysm and associated cranial neuropathy Outcome No. (%) clinical resolved 4 (20) improved 11 (55) stable 5 (25) worse 0 (0) angiographic resolved/improved symptoms complete/near-complete 10 (50) residual 3 (15) no angiographic FU 2 (10) total 15 (75) stable symptoms complete/near-complete 2 (10) residual 3 (15) total 5 (25) She underwent treatment with 2 devices measuring mm. Shortly after her initial treatment, she developed significant left-sided proptosis, which resolved over the course of a few weeks. As ascertained by phone follow-up at 3 months she had near resolution of her visual complaints, with some mild persistent double vision on extreme right-sided lateral gaze. However, she still suffered from constant daily headaches. An MRA study obtained at 16 months demonstrated some residual filling, and she underwent follow-up angiography at 18 months, which demonstrated separation of the overlapping constructs, with prolapse into the aneurysm dome (Fig. 1C and D). She underwent placement of 2 additional devices at this time. Despite being instructed to discontinue her clopidogrel at 3 months following retreatment, she continued to demonstrate residual aneurysm on angiography, requiring additional retreatment with 2 devices. At 28 months following her initial treatment, she underwent her final follow-up angiogram, which demonstrated continued filling of her lesion. Her clinical status remained stable, with only mild residual diplopia but burdensome chronic daily headaches. She underwent microsurgical clip occlusion of her left ICA with a superficial temporal artery (STA) MCA bypass, complicated by transient right-sided weakness and speech difficulty following surgery (Fig. 2). Her headaches improved dramatically for several months fol- TABLE 4: Comparison of clinical outcome by CN CN No. Clinical Outcome Resolved (%) Improved (%) Stable (%) J Neurosurg / Volume 121 / November 2014 p Value II 1 (25) 1 (9) 1 (20) 0.70 III 2 (50) 4 (36) 2 (40) 0.89 V 0 (0) 3 (27) 1 (20) 0.51 VI 2 (50) 7 (64) 2 (40) 0.66 Fig. 1. Case 4. A: Preoperative DSA. Lateral projection of left ICA injection, demonstrating a giant cavernous segment ICA aneurysm. B: Preoperative CT angiography. Axial plane image with contrast, demonstrating a giant left cavernous segment ICA aneurysm. C: Postoperative MRA study. Axial plane image with contrast, demonstrating residual contrast filling of the aneurysm dome (arrow) at 16-month follow-up. D: Follow-up DSA. Anteroposterior projection of left ICA injection, demonstrating residual angiographic filling of a giant cavernous segment ICA aneurysm, secondary to a prolapsed PED construct. 1089

6 K. Moon et al. lowing surgery, but have returned to a lesser extent at last follow-up 9 months postoperatively. Case 10 A 79-year-old woman presented with a 3-week history of acute-onset headache, associated with nausea and horizontal diplopia. Imaging demonstrated a 20-mm cavernous segment ICA aneurysm (Fig. 3 left). Due to a history of bleeding ulcers from aspirin administration, the patient underwent preloading with clopidogrel 7 days prior to her procedure. She underwent treatment with 2 devices measuring mm and mm, along with adjunctive coil embolization. The 6-month followup evaluation revealed near-complete occlusion of her lesion and resolution of her symptoms (Fig. 3 right). She remains neurologically intact at 19 months. Case 12 A 64-year-old man with a yearlong history of double vision presented with recent development of left face and tongue numbness as well as a left eye visual field cut, which was found on formal ophthalmological workup. He was initially evaluated for microsurgical clip occlusion of a 16-mm petrous segment ICA aneurysm and referred to the endovascular neurosurgery service for definitive management (Fig. 4 left). Treatment was performed with 2 devices measuring mm and mm and coil embolization. The 6-month follow-up evaluation revealed cure of his lesion and interval improvement of his visual acuity, facial numbness, and diplopia (Fig. 4 right). At his last follow-up at 18 months, he remains clinically stable. Discussion Treatment for cavernous segment ICA aneurysms is typically reserved for symptomatic lesions or those with sequelae of rupture, such as subarachnoid hemorrhage or carotid-cavernous fistula. The majority of these lesions remain asymptomatic, and the risk of subarachnoid hemorrhage is often quoted as less than 0.5% per year, reserved for lesions large enough to extend into the subarachnoid space or the junction of the cavernous and Fig. 3. Case 10. Left: Preoperative DSA. Lateral projection of left ICA injection, demonstrating a large cavernous segment ICA aneurysm. Right: Follow-up DSA. Lateral projection of left ICA injection, demonstrating near-complete occlusion of a previously treated cavernous segment ICA aneurysm at 6 months. intradural segments of the ICA. 10,13,19 Nevertheless, these and other intracranial aneurysms can present with cranial neuropathies that impose a significant burden on a patient s quality of life, including those of diplopia, loss of visual acuity, and pain. Given the recent development of the PED and the relatively uncommon incidence of cranial neuropathies secondary to an intracranial aneurysm, data for follow-up of these patients are sparse but promising. Small series of symptomatic lesions treated without adjunctive coil embolization found improvement in 71% 94% of patients. 20,24 Puffer et al. found a 90% rate of symptomatic improvement for cavernous segment lesions, but did not define symptoms as attributable to cranial neuropathies or other nonspecific symptoms, nor were cases with coil embolization included. 18 Most recently, the multicenter Canadian experience found improvement or resolution in 72% of patients with cavernous segment aneurysms, although only oculomotor palsies were evaluated in this subset. 17 The current report represents the largest institutional series of intracranial aneurysms presenting with cranial neuropathies and treated with the PED with or without adjunctive coil embolization. Fig. 2. Case 4. Left: Follow-up CT angiography study obtained with contrast. Axial plane image demonstrating clip occlusion of the left ICA for a previously seen cavernous segment ICA aneurysm. Right: Postoperative CT angiography with contrast. Coronal plane image demonstrating patent left STA-MCA bypass following clip occlusion of the left ICA for residual giant cavernous segment aneurysm. Fig. 4. Case 12. Left: Preoperative DSA. Lateral projection of left ICA injection, demonstrating a large petrous segment ICA aneurysm. Right: Follow-up DSA. Lateral projection of left ICA injection, demonstrating complete occlusion of a previously treated petrous segment ICA aneurysm at 6 months J Neurosurg / Volume 121 / November 2014

7 Pipeline device for aneurysms with cranial neuropathies J Neurosurg / Volume 121 / November 2014 The mechanism by which CN deficits occur in the setting of intracranial aneurysms is debated. Mass effect is a likely explanation for de novo neuropathies in the setting of large or giant aneurysms with proximity to CNs, such as cavernous or ophthalmic segment ICA and PCoA aneurysms. This phenomenon is supported by data suggesting a higher risk of mass effect or transient worsening of symptoms in patients undergoing coil embolization of large lesions. 7,14,25 However, this risk seems to be mitigated in the setting of concurrent PED placement, which may be attributable to our practice of decreased coil packing when used in conjunction with the PED. Notably, only 1 patient in this series had persistent symptoms after undergoing adjunctive coil embolization, despite an angiographic cure at follow-up for a 12-mm supraclinoid lesion. The remaining 5 patients experienced improvement of their symptoms, and no patients in whom coils were placed reported transient worsening of their symptoms. Furthermore, PED-induced thrombosis has been shown to significantly decrease the size of large aneurysms on follow-up imaging, which is likely to reduce gross mass effect, regardless of the presence or absence of coils. 4,20 At the same time, it is clear that mass effect in and of itself is not solely responsible. Flow-related or dynamic effects are probably responsible for persistent or recurrent symptoms in many lesions, especially those with residual angiographic filling following treatment or recurrence of the aneurysm. 23 This point is underscored by the often diminutive size of the aneurysmal remnant or recurrence without significant mass effect, yet with objective evidence of CN palsy. Furthermore, 4 patients in this series with residual aneurysmal dome filling on follow-up still achieved improvement in their symptomatology. These data probably highlight the novel mechanism by which flow-diverter stents bring about aneurysm thrombosis. By promoting immediate stagnation and cessation of arterial inflow, the PED not only induces thrombosis but probably also mitigates the effects of arterial pulsatility on adjacent CNs. Given the high rate of incomplete occlusion or recurrence in the treatment of large or giant aneurysms with primary coil embolization, these findings provide support for the PED as a valuable tool in achieving symptomatic resolution. What remains difficult to explain or predict is the phenomenon of persistent or recurrent symptoms, especially in cavernous segment ICA aneurysms. The anatomy of these lesions is obviously complex, with the intimate relationship and proximity to the parent artery of CNs and other neural elements playing a large role in the symptomatology of these patients, especially in the setting of lesional recurrence. 23 It is likely that these patients remain far more sensitive to dynamic or flow-related effects as well as the smallest changes in arterial wall architecture the latter of which is probably affected by placement of a flow-diverter stent. It is also likely, as has been shown in series of coil embolization for symptomatic lesions, that longer duration of symptoms results in a more persistent degree of neural injury. Although this was not demonstrated in this subset analysis, a larger series powered to study this phenomenon may reveal a difference. Although the results of this series demonstrate comparable rates of efficacy for these lesions, conventional endovascular and surgical techniques are well established and remain among the mainstays of therapy. Rates of symptomatic improvement after coil embolization are high, ranging from 64% 76% in several series. 6,8,9,14,15,22 Clinical improvement after parent carotid artery occlusion after balloon test occlusion is probably even higher, ranging from 73% 90%. 16,21,22 Treatment with an open-cell stent has been shown to ameliorate symptoms as well. 5 Surgery for large or giant cavernous segment ICA aneurysms is most commonly performed with Hunterian ligation or aneurysmal trapping with or without concurrent bypass, although some lesions may be treated with clipping or even wrapping. Nonetheless, endovascular or surgical occlusion of a parent ICA has several potential disadvantages. Patients must pass a balloon test occlusion on clinical and radiographic grounds beforehand and, even when deemed appropriate, face a risk of ischemic complications. Bypass and revascularization carries additional risk of morbidity in low-volume centers. In addition, therapeutic options for mirror lesions on the contralateral ICA become limited in the face of occlusion on one side. Finally, flow-related de novo aneurysm formation has been estimated to be 10% in patients undergoing Hunterian ligation. 12 As long-term results and follow-up emerge from ongoing series of PED placement for large symptomatic lesions, further questions regarding its durability for cranial neuropathies will be answered. There are known challenges in using the PED when compared with outcomes for other endovascular modalities. Complications in this series included 6 (30.0%) transient hemorrhagic complications related to dual-antiplatelet therapy 2 gastrointestinal bleeds, 2 retroperitoneal hematomas, 1 groin hematoma, and 1 case of epistaxis all of which were self-limited and/or managed medically, as well as 1 case of radiation-induced alopecia. There was 1 permanent complication in the form of a delayed right ICA occlusion following retreatment that was tolerated symptomatically but that ultimately led to microsurgical bypass due to persistent perfusion deficit noted radiographically. Improvement of the patient s cranial neuropathies was noted following microsurgical bypass. There were no deaths or cases of intraprocedural aneurysmal rupture. Early results of our entire institutional experience with the PED demonstrate a complication rate of approximately 20%, with 3% being permanent. 1 Accordingly, the PED should remain a viable option for treatment of lesions that are otherwise not amenable to coil embolization or microsurgery. Counseling for patients with these lesions should weigh the high probability of clinical improvement after PED placement against the device s emerging complication profile. Conclusions The risks and challenges faced with the PED must be weighed against not only anatomical outcomes but also variable rates of clinical improvement. Treatment of symptomatic intracranial aneurysms with the PED results in a high rate of improvement of cranial neuropathies at follow-up, with a small subset of patients achieving complete resolution. There remains a subset of patients who do 1091

8 K. Moon et al. not respond optimally to PED placement, for whom further research and/or device modification may be indicated. Although standard practices for adjunctive coil embolization are not yet established, placement of coils does not appear to increase the risk of mass effect and symptomatic worsening in the setting of concurrent PED placement. Disclosure Dr. McDougall is a consultant for Microvention, Covidien, and Codman. Author contributions to the study and manuscript preparation include the following. Conception and design: Albuquerque. Acquisition of data: Moon, Ducruet, Crowley. Analysis and interpretation of data: Albuquerque, Moon, Ducruet, McDougall. Drafting the article: Moon. Critically revising the article: all authors. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Albuquerque. Statistical analysis: Moon, Ducruet. Administrative/ technical/material support: Albuquerque, McDougall. Study supervision: Albuquerque. References 1. Albuquerque F, Ducruet A, Crowley R, McDougall C: The Barrow Neurological Institute experience with pipeline embolisation device: results in the first 100 patients treated after FDA approval. J Neurointervent Surg 5:A6 A7, 2013 (Abstract) 2. Brinjikji W, Murad MH, Lanzino G, Cloft HJ, Kallmes DF: Endovascular treatment of intracranial aneurysms with flow diverters: a meta-analysis. Stroke 44: , Ferns SP, van Rooij WJ, Sluzewski M, van den Berg R, Majoie CB: Partially thrombosed intracranial aneurysms presenting with mass effect: long-term clinical and imaging follow-up after endovascular treatment. AJNR Am J Neuroradiol 31: , Fiorella D, Lylyk P, Szikora I, Kelly ME, Albuquerque FC, McDougall CG, et al: Curative cerebrovascular reconstruction with the Pipeline embolization device: the emergence of definitive endovascular therapy for intracranial aneurysms. J Neurointerv Surg 1:56 65, Fischer S, Weber W, Henkes H, Kühne D: Regression of cranial nerve palsy due to a cavernous carotid artery aneurysm after deployment of a covered stent. Clin Neuroradiol 18: , Güresir E, Schuss P, Setzer M, Platz J, Seifert V, Vatter H: Posterior communicating artery aneurysm-related oculomotor nerve palsy: influence of surgical and endovascular treatment on recovery: single-center series and systematic review. Neurosurgery 68: , Halbach VV, Higashida RT, Dowd CF, Barnwell SL, Fraser KW, Smith TP, et al: The efficacy of endosaccular aneurysm occlusion in alleviating neurological deficits produced by mass effect. J Neurosurg 80: , Hanse MC, Gerrits MC, van Rooij WJ, Houben MP, Nijssen PC, Sluzewski M: Recovery of posterior communicating artery aneurysm-induced oculomotor palsy after coiling. AJNR Am J Neuroradiol 29: , Kim DJ, Kim DI, Lee SK, Kim SY: Unruptured aneurysms with cranial nerve symptoms: efficacy of endosaccular Guglielmi detachable coil treatment. Korean J Radiol 4: , Kupersmith MJ, Hurst R, Berenstein A, Choi IS, Jafar J, Ransohoff J: The benign course of cavernous carotid artery aneurysms. J Neurosurg 77: , Kupersmith MJ, Stiebel-Kalish H, Huna-Baron R, Setton A, Niimi Y, Langer D, et al: Cavernous carotid aneurysms rarely cause subarachnoid hemorrhage or major neurologic morbidity. J Stroke Cerebrovasc Dis 11:9 14, Lawton MT, Hamilton MG, Morcos JJ, Spetzler RF: Revascularization and aneurysm surgery: current techniques, indications, and outcome. Neurosurgery 38:83 94, Linskey ME, Sekhar LN, Hirsch WL Jr, Yonas H, Horton JA: Aneurysms of the intracavernous carotid artery: natural history and indications for treatment. Neurosurgery 26: , Malisch TW, Guglielmi G, Viñuela F, Duckwiler G, Gobin YP, Martin NA, et al: Unruptured aneurysms presenting with mass effect symptoms: response to endosaccular treatment with Guglielmi detachable coils. Part I. Symptoms of cranial nerve dysfunction. J Neurosurg 89: , Mansour N, Kamel MH, Kelleher M, Aquilina K, Thornton J, Brennan P, et al: Resolution of cranial nerve paresis after endovascular management of cerebral aneurysms. Surg Neurol 68: , Matouk CC, Kaderali Z, terbrugge KG, Willinsky RA: Longterm clinical and imaging follow-up of complex intracranial aneurysms treated by endovascular parent vessel occlusion. AJNR Am J Neuroradiol 33: , O Kelly CJ, Spears J, Chow M, Wong J, Boulton M, Weill A, et al: Canadian experience with the pipeline embolization device for repair of unruptured intracranial aneurysms. AJNR Am J Neuroradiol 34: , Puffer RC, Piano M, Lanzino G, Valvassori L, Kallmes DF, Quilici L, et al: Treatment of cavernous sinus aneurysms with flow diversion: results in 44 patients. AJNR Am J Neuroradiol 35: , Stiebel-Kalish H, Kalish Y, Bar-On RH, Setton A, Niimi Y, Berenstein A, et al: Presentation, natural history, and management of carotid cavernous aneurysms. Neurosurgery 57: , Szikora I, Marosfoi M, Salomváry B, Berentei Z, Gubucz I: Resolution of mass effect and compression symptoms following endoluminal flow diversion for the treatment of intracranial aneurysms. AJNR Am J Neuroradiol 34: , van der Schaaf IC, Brilstra EH, Buskens E, Rinkel GJ: Endovascular treatment of aneurysms in the cavernous sinus: a systematic review on balloon occlusion of the parent vessel and embolization with coils. Stroke 33: , van Rooij WJ, Sluzewski M: Unruptured large and giant carotid artery aneurysms presenting with cranial nerve palsy: comparison of clinical recovery after selective aneurysm coiling and therapeutic carotid artery occlusion. AJNR Am J Neuroradiol 29: , Xu DS, Hurley MC, Batjer HH, Bendok BR: Delayed cranial nerve palsy after coiling of carotid cavernous sinus aneurysms: case report. Neurosurgery 66:E1215 E1216, Yu SC, Kwok CK, Cheng PW, Chan KY, Lau SS, Lui WM, et al: Intracranial aneurysms: midterm outcome of pipeline embolization device a prospective study in 143 patients with 178 aneurysms. Radiology 265: , Yu SC, Wong WC, Chung AC, Lee KT, Wong GK, Poon WS: Does endoluminal coil embolization cause distension of intracranial aneurysms? Neuroradiology 48: , 2006 Manuscript submitted December 4, Accepted July 22, Please include this information when citing this paper: published online September 5, 2014; DOI: / JNS Address correspondence to: Felipe C. Albuquerque, M.D., c/o Neuroscience Publications, Barrow Neurological Institute, St. Joseph s Hospital and Medical Center, 350 W. Thomas Rd., Phoenix, AZ neuropub@dignityhealth.org J Neurosurg / Volume 121 / November 2014