Portal de Periódicos da CAPES

Microsurgical Management of Residual and Recurrent Aneurysms After Coiling and Clipping

2015; Lippincott Williams & Wilkins; Volume: 62; Issue: Supplement 1 Linguagem: Inglês

10.1227/neu.0000000000000791

1524-4040

Christopher Owen, Nicola Montemurro, Michael T. Lawton,

Vascular Malformations Diagnosis and Treatment

Residual and recurrent aneurysms are difficult quandaries for patients and clinicians because curative occlusion is not achieved and a decision about retreatment must be made, weighing the risks of rupture from a partially treated aneurysm against the risks of further intervention, which can be exponentially more challenging than an untreated aneurysm. In a review of incompletely coiled and recurrent aneurysms managed microsurgically during a 10-year period from 1997 to 2007, we observed a spike in the annual number of treated patients from an average of 3 patients per year to 11 patients in the last year of the study.1 We were concerned that this sharp increase in the incidence of residual and recurrent aneurysms requiring surgery might portend a dangerous problem in the durability of endovascular repair, particularly when more than half of recurrent aneurysms had coils extruded from the aneurysm into the subarachnoid space. Since this review, endovascular case volume has grown steadily as results from the International Subarachnoid Aneurysm Trial (ISAT)2 were replicated in similar randomized trials like the Barrow Ruptured Aneurysm Trial (BRAT).3 In addition, market forces and patient preferences for minimally invasive therapies have reinforced this shift in clinical practice away from open craniotomy.4,5 Increasing endovascular case volume will magnify the problem of aneurysm recurrence and retreatment after initial coiling unless advances in endovascular technology and technique such as stent-assisted or balloon-assisted coiling and flow diverters address this issue. Therefore, we conducted an update of our experience with incompletely coiled and recurrent aneurysms managed microsurgically to measure the magnitude of this problem. In addition, we examined a concurrent experience with incompletely clipped and recurrent aneurysms after clipping also managed microsurgically. This review was intended to better define the indications for surgery, to help select the best first treatment modality, and to elaborate on microsurgical techniques for these challenging aneurysms. METHODS The study was approved by the Institutional Review Board and conducted in compliance with Health Insurance Portability and Accountability Act regulations. The prospectively maintained database for the Vascular Neurosurgery Service at the University of California-San Francisco was searched for patients with aneurysm who had microsurgical treatment after previous endovascular intervention during the 17-year period from December 1997 to December 2014. Medical records, radiographic studies, operative reports, intraoperative photographs, and clinical follow-up evaluations were reviewed retrospectively. Patients were included if their surgery was preceded by incomplete aneurysm coiling or their aneurysm recurred after prior coiling (Gurian group B). Patients were excluded if endovascular therapy was attempted but aborted (Gurian group A) or if they required surgical intervention as a result of a complication arising from an endovascular procedure (Gurian group C). Additionally, the database was searched for patients with aneurysm who had microsurgical treatment for residual or recurrent aneurysm after previous microsurgery either by the senior author or by other referring neurosurgeons. Residual or remnant aneurysms were defined as treated aneurysms, whether endovascularly or microsurgically, with incomplete obliteration or persistent aneurysm filling angiographically immediately after treatment. Recurrent aneurysms were defined as treated aneurysms, whether endovascularly or microsurgically, with incomplete obliteration or persistent filling angiographically at later follow-up (>3 months). Recurrent aneurysms may have arisen from aneurysms that were completely occluded initially and subsequently underwent coil compaction or regrowth or from residual aneurysms that were incompletely occluded initially and subsequently compacted or regrew. Recurrent aneurysms presented as rehemorrhages, with new compressive symptoms, or as findings on surveillance angiography. Aneurysm outcomes were evaluated with postoperative catheter angiography that was interpreted by independent neuroradiologists. Patient outcomes were evaluated by independent research clinicians preoperatively, postoperatively, and at last follow-up with the Glasgow Outcome Scale score. RESULTS Residual and Recurrent Aneurysms After Coiling During the 17-year study period, 3600 aneurysms were treated in a single-surgeon series of 2748 consecutive patients by the senior author (M.T.L.). Of those, 73 patients (2.7%) underwent microsurgical treatment of 73 aneurysms (2.0%) that were either residual or recurrent after endovascular coiling. A female predominance was observed, with 58 women and 15 men at a mean age of 49 years (range, 15-77 years; Table 1). Overall, 64 patients (88%) presented with subarachnoid hemorrhage (SAH), and the remainder presented with mass effect symptoms from giant aneurysms, with headache, or with incidental aneurysms.TABLE 1: Residual and Recurrent Aneurysms After CoilingAnterior cerebral artery (ACA) aneurysms were the most common (22, 30%), followed by posterior cerebral artery (PCA; 12, 16%) and middle cerebral artery (MCA; 10, 14%) aneurysms. Seventeen aneurysms (23%) were in the posterior circulation. Four aneurysms (5%) were large, and 7 (10%) were giant. Thirty-two patients (44%) had known residual aneurysms after endovascular coiling and were referred immediately for microsurgical treatment. Four of these patients (12%) rehemorrhaged after the coiling procedural before their referral. The remaining 41 patients (56%) had recurrent aneurysms after endovascular coiling. The median interval from initial endovascular treatment to microsurgery for a recurrent aneurysm was 2 years. Of the recurrent aneurysms, 35 (85%) were referred after the recurrence was demonstrated on surveillance angiography; 3 (7%) were referred after developing new or progressive symptoms from mass effect; and 3 patients (7%) presented with recurrent SAH. Most residual and recurrent aneurysms (59, 81%) were retreated with microsurgical clipping alone. Of those, 72% were clipped below coils that compacted enough to reform a soft, clippable neck (Figures 1 and 2). Another 6% had coils in the neck, and the clips were placed across loops of coil. Another 9% also had coils in the neck, and the clips were placed against the coil mass. Aneurysm transection with coil mobilization and/or extraction was required in 12% to facilitate clip placement (Figures 3 and 4). Bypass was required in 8 patients (11%) (Figure 5), and 1 giant posterior cerebral artery aneurysm was trapped without bypass (Table 1). Wrapping was used early in the series (the last wrapped aneurysm was in 2005) as a last resort when other methods of aneurysm exclusion failed.FIGURE 1: This case demonstrates simple clipping of a recurrent aneurysm. A, this 47-year-old woman had a ruptured left ophthalmic artery that was coiled successfully (left internal carotid artery injection, 3-dimensional reconstruction, lateral view). B, surveillance angiography demonstrated coil compaction at 6 months (left internal carotid artery injection, digital subtraction angiogram, lateral view). C, the anterior clinoid process covered the proximal neck. D, clinoidectomy exposed the neck and optic nerve. E, the aneurysm was clipped with 2 clips below the compacted coil mass, and (F) the compressive coil mass was removed. G, postoperative angiography (left internal carotid artery injection, 3-dimensional rotational angiogram, anteroposterior view) confirmed complete aneurysm occlusion.FIGURE 2: This case demonstrates tandem clipping of a recurrent left posterior communicating artery aneurysm after stent-assisted coiling. A, surveillance angiography in this 49-year-old woman demonstrated aneurysm recurrence and regrowth (left internal carotid artery injection, digital subtraction angiogram, lateral view and [B] anteroposterior view). C, intraoperatively, the Neuroform stent was seen through the wall of the internal carotid artery, and coils were extruded into the subarachnoid space. D, the aneurysm was clipped with a fenestrated clip on the proximal neck and a straight clip closing the fenestration distally. E, extruded coils extended into the temporal lobe, which was stained with hemosiderin from the initial subarachnoid hemorrhage. F, indocyanine green videoangiography showed the tines of the stent, complete aneurysm occlusion, and preservation of anterior choroidal artery.FIGURE 3: This case demonstrates clipping of a residual anterior communicating artery (ACoA) aneurysm after coil extraction. A, this 34-year-old man's coiling procedure after subarachnoid hemorrhage was complicated by aneurysm perforation and rehemorrhage. The procedure was stopped, and residual aneurysm remained (left carotid injection, digital subtraction angiography, lateral view and [B] 3-dimensional reconstruction, lateral view). C, intraoperatively, coils perforated through the dome of the aneurysm superiorly, and intrasaccular coils filled the neck. D, after trapping of the ACoA complex with temporary clips, the aneurysm was transected, and the coil mass was pulled upward out of the aneurysm. E and F, the softened neck was clipped with a straight clip, and the untreated anterior lobe was clipped with 4 stacked clips. G, postoperative angiography (right internal carotid artery injection, digital subtraction angiography, right anterior oblique view, and [H] 3-dimensional reconstruction, left anterior oblique view) confirmed complete aneurysm occlusion.FIGURE 4: This case demonstrates clipping of a recurrent anterior communicating artery aneurysm after coil extraction. A, this 44-year-old woman presented with subarachnoid hemorrhage, was initially coiled, and had 2 recurrences that were retreated with additional coiling (right internal carotid artery [ICA] injection, digital subtraction angiography, lateral view). B, an attempted retreatment of her third recurrence with stent-assisted coiling resulted in intraprocedural perforation before stent placement (right ICA injection, digital subtraction angiography, anteroposterior view; note guidewires in both ICAs). C, intraoperatively, the aneurysm was densely packed with some coil extrusion inferior in the interoptic triangle. D, the aneurysm was transected, and coils were densely adherent to the posterior neck. E, most but not all of these adherent coil fragments were removed, and (F) the transected neck was closed with clips across these strands of coil. Note the bilateral A2 anterior cerebral arteries (ACAs) and a midline accessory ACA. G, postoperative angiography (right internal carotid artery injection, digital subtraction angiography, right anterior oblique view, and [H] 3-dimensional reconstruction, right anterior oblique view) confirmed complete aneurysm occlusion.FIGURE 5: This case demonstrates treatment of a recurrent posterior inferior cerebellar artery aneurysm with trapping and bypass. A, this 69-year-old woman had coil compaction identified on surveillance angiography (left vertebral artery injection, digital subtraction angiography, lateral view and [B] anteroposterior view). C, the aneurysm neck was atherosclerotic, and the inflow and outflow arteries were at right angles, making clipping difficult. D, these arteries were transected and reanastomosed end to end. E, the coiled aneurysm was left in situ to avoid manipulation of the lower cranial nerves, and (F) the anastomosis filled well, (G) as confirmed on postoperative angiography (left vertebral artery injection, digital subtraction angiography, lateral view).Bypasses included 2 extracranial-to-intracranial bypasses: superficial temporal artery-to-PCA (1 patient with a giant basilar artery apex aneurysm) and a cervical internal carotid artery (ICA)-to-MCA bypass with a saphenous vein graft (1 patient with a giant ophthalmic artery aneurysm). Three in situ bypasses were performed: reimplantation of the posterior inferior cerebellar artery (PICA) onto the vertebral artery (2 patients with PICA aneurysms, 1 of which was giant) and an anterior temporal artery-to-MCA bypass (1 patient with a fusiform MCA aneurysm). One PICA aneurysm recurrence was excised and reanastomosed. Two intracranial-to-PICA bypasses were performed. In the first case, a radial artery graft was connected proximally to the A1 anterior cerebral artery (ACA), and the MCA efferent trunks coming from the aneurysm were reimplanted onto the graft (double-reimplantation technique). In the second case, an allograft saphenous vein graft was connected proximally to the M2 MCA and distally to the A2 ACA branches of an anterior communicating artery (ACoA) aneurysm. Once the bypasses were completed, 4 aneurysms were trapped, 3 were proximally clip occluded, and 1 was treated with additional coils. Complete angiographic aneurysm occlusion was achieved in 65 patients (89%). Incomplete aneurysm occlusion was observed in 2 residual and 1 recurrent aneurysm (4%) treated with direct clipping. Incomplete aneurysm occlusion was observed in the 5 wrapped aneurysms, of which 4 had angiographic follow-up and these 4 aneurysms remained radiographically stable. One patient with a large, broad-necked recurrent basilar bifurcation aneurysm was treated with a superficial temporal artery-PCA bypass only to aggressively coil the recurrent aneurysm and to protect the PCA territory if the involved P1 origin was occluded. This aneurysm was incompletely obliterated after coiling. Seven of the 8 bypasses were patent. Three patients died in the perioperative period (surgical mortality, 4%). One patient had an MCA aneurysm that was incompletely coiled at an outside institution and reruptured, and she was transferred while in coma. She died despite uncomplicated clipping of her aneurysm. One patient with a giant basilar bifurcation aneurysm recurred after coiling with aneurysm growth, coil compaction, and progressive obtundation. Superficial temporal artery-PCA bypass and additional coiling of the aneurysm failed to reverse his deterioration, and support was withdrawn. One patient died of ischemic complications related to vasospasm. Patient outcomes were favorable, with good outcomes (Glasgow Outcome Scale score, 5 or 4) observed in 65 patients (89%). At the last follow-up evaluation (mean duration, 6 years), 67 patients (92%) were improved or unchanged. Three patients (4%) were worse neurologically after surgery. One patient had a new visual field deficit after clipping of a recurrent ophthalmic artery aneurysm. One patient developed unilateral leg weakness caused by her MCA-to-ACA bypass occluding and causing an ischemic infarct. One patient developed short-term memory difficulty after clipping of an enlarging residual, previously ruptured ACoA aneurysm. Residual and Recurrent Aneurysms After Microsurgical Clipping During the 17-year study period in which 3600 aneurysms were treated in a single-surgeon series of 2748 consecutive patients, 6 patients (0.2%) returned to the operating room for re-exploration and clipping of a residual aneurysm seen on postoperative angiography. The mean patient age was 53 years (range, 38-72 years), and patients were evenly divided between men and women. All 6 patients presented with SAH. Aneurysms were located as follows: ACoA, 4 patients; superior hypophyseal artery, 1 patient; and midbasilar trunk, 1 patient. Five residual aneurysms were completely clipped during the second surgery, and the aneurysm in 1 patient reruptured during re-exploration of a midbasilar artery aneurysm, and the patient later died. An additional 14 patients were initially treated microsurgically by an outside neurosurgeon and were referred with recurrent aneurysms for microsurgical treatment. Their mean age was 54 years (range, 29-69 years); all but 1 were female; and all of them presented with SAH. Twelve of these patients' aneurysms were previously clipped, and 2 were wrapped. The median time between initial surgery and reoperation for recurrence was 13 years (range, 0.4-32 years). Six recurrences were discovered with surveillance imaging; 2 were discovered on imaging for headaches; and 6 patients (40%) presented with new SAH. Aneurysm recurrences occurred at the posterior communicating artery (PCoA) in 4 patients, supraclinoid ICA in 3, basilar bifurcation in 2, ACoA in 2, MCA in 2, and SCA in 1 patient. Overall, 12 recurrent aneurysms were completely clipped. One fusiform ICA aneurysm required a cervical ICA-MCA bypass and trapping, and 1 recurrent giant basilar artery aneurysm could not be clipped and later underwent successful stent-assisted coil embolization. Twelve patients with recurrent aneurysms (86%) were improved or unchanged after surgery. One patient developed significant vasospasm and mild aphasia after clipping of a recurrent PCoA aneurysm. Another patient's recurrent SCA aneurysm presented with a Hunt-Hess grade IV SAH, and although the aneurysm was completely clipped, she remained in coma, and care was withdrawn. Four additional patients were initially treated microsurgically by the senior author, had recurrence of their clipped aneurysms, and were subsequently treated endovascularly. One 29-year-old patient had a giant and dysplastic basilar bifurcation aneurysm that ruptured and was clip reconstructed with a good angiographic result, but the aneurysm recurred 4 years later on the efferent P1 PCA segment with the formation of a serpentine aneurysm. This was treated endovascularly with coiling of the channel and has been stable for 3 years. A second 66-year-old patient with a dolichoectatic, giant basilar trunk aneurysm had a large PCoA that was proximally clip occluded. Her aneurysm failed to thrombose completely, and she was treated with 2 separate stages of coil embolization of the enlarging aneurysm sac. A third 34-year-old woman presented with a ruptured P1-P2 PCA aneurysm that was proximally clip occluded. Although the aneurysm thrombosed completely, she returned with a new oculomotor nerve palsy and recanalization of the aneurysm through the PCoA. She tolerated coiling of the aneurysm despite occlusion of the contribution of the PCoA to the PCA territory. Finally, a fourth 53-year-old patient presented with a supraclinoid ICA aneurysm that recurred 23 years after PCoA aneurysm clipping. The aneurysm was clipped completely but recurred a second time 7 years later and was treated with stent-assisted coiling. All 4 of these patients remained stable clinically after their endovascular therapies. Rehemorrhage of Recurrent Aneurysms A total of 14 patients (14%) in this experience rehemorrhaged from residual or recurrent aneurysms after coiling and clipping (Table 2), including 4 patients with residual aneurysms after coiling, 3 patients with recurrent aneurysms after coiling, and 6 patients with recurrent aneurysms after surgical clipping. The median time to rehemorrhage was 1.1 years for previously coiled recurrent aneurysms and 15.5 years for previously clipped recurrent aneurysms. Aneurysm recurrence and rehemorrhage adversely affected patient outcomes, with a median final Glasgow Outcome Scale score of 4 for patients experiencing SAH from recurrent aneurysms compared with a median of 5 for those with recurrent aneurysms that were diagnosed with surveillance imaging.TABLE 2: Hemorrhage Rates Associated With Residual and Recurrent AneurysmsDISCUSSION Clinical Trends This microsurgical experience with 97 residual and recurrent aneurysms in 97 patients, 73 of whom had residual/recurrent aneurysms after coiling, is one of the largest reported in the literature.6-13 However, this review spanned 17 years with an average of 4 to 5 coil residual/recurrence cases per year and accounted for only 2% of all aneurysms treated microsurgically. Therefore, aneurysm recurrence and retreatment after coiling have not become the expansive problem that we once feared, even with the frequent extrusion of coils observed intraoperatively. In an earlier publication, we projected that residual/recurrent aneurysms would increase in incidence as a result of the following: a growing endovascular case volume; neurologists and neurosurgeons joining interventional neuroradiologists in acquiring endovascular competencies; patients preferring less invasive aneurysm therapy; community hospitals building comprehensive stroke and aneurysm centers with low case volumes and diminished technical expertise; and cost savings derived from shortened hospitalizations that appeal to administrators and policy makers.1 Surprisingly, the incidence of residual/recurrent aneurysms has remained stable from our surgical perspective. Although we did not analyze retreatment rates of residual/recurrent aneurysms from an endovascular perspective, our findings suggest that advances in coil technology, balloon- and stent-assisted coiling, and flow diverters, together with increasing endovascular expertise generally, have kept recurrences and microsurgical retreatment rates in check. Comparative Efficacy and Durability of Aneurysm Repair Coil remnants and recurrences were treated almost 4 times more than clip remnants and recurrences in our experience, which is consistent with the superior efficacy and durability of microsurgical clipping3,14-24 (Table 3). A clip across an aneurysm neck mechanically closes the sac, reinforces the bifurcation structurally, and apposes healthy arterial tissues that can repair and re-endothelialize the wall defect. In contrast, luminal coils obliterate the aneurysm sac but hold the neck open and separate arterial tissues, exposing the coils to pathological hemodynamics, compacting or shifting the coils over time, and preventing the parent artery from repairing and re-endothelializing the wall defect in some cases.25-28 Incomplete aneurysm occlusion is a significant problem after coiling that leads to both remnants and recurrences.21-23,29 In ISAT, complete occlusion was achieved in 584 of 881 endovascular patients (66%) compared with 370 of 450 microsurgical patients (82%). The large number of neck remnants and incomplete occlusions in 297 endovascular patients contributed to 28 rehemorrhages and 15 deaths during the first year after treatment and an additional 7 rehemorrhages and 2 deaths after the first year. The relative rehemorrhage risk was 2.5 times greater in endovascular patients than in microsurgical patients, in whom there were 13 rehemorrhages in total.2,19TABLE 3: Rates of Recurrent and Residual Aneurysms, Retreatment, and Recurrent Subarachnoid Hemorrhage Across Prospective Studies and Large Case SeriesaThe Cerebral Aneurysm Rerupture After Treatment (CARAT) study found a similar difference in rehemorrhage rates: 3.4% in 295 endovascular patients and 1.3% in 706 microsurgical patients. The complete aneurysm occlusion rate was 92% in microsurgical patients and only 39% in endovascular patients, and the annual rehemorrhage rate varied inversely with degree of coil occlusion: 1.1% with completely occluded aneurysms, 5.9% with 70% to 90% occlusion, and 17.6% with <70% occlusion. Rehemorrhage rates in coiled patients declined from 4.9% in the first year to 0.1% afterward, but recurrences and retreatments were significant: 13.3% retreatment during the first year, 4.5% retreatment during the second year, and 1.1% later. Retreatment rates were substantially less in microsurgical patients: 2.6% in the first year and none thereafter.20 Late results from BRAT also identified significant differences in aneurysm occlusion rates (52% and 87% for coiling and clipping at 3 years, respectively; P < .001) and retreatment rates (10.6% and 4.5% for coiling and clipping at 1 year, respectively; P = .03; and 13% and 5% for coiling and clipping at 3 years, respectively; P = .01).30 Even more important, late outcomes in BRAT demonstrated that the early advantages of coiling disappeared over time, with no difference in outcomes with anterior circulation aneurysms. Compared with 34.0% of patients randomized to clipping, 28.7% of patients randomized to coiling had modified Rankin Scale scores >2 (P = .27). Therefore, randomized comparative trials attract attention for their early primary patient outcomes, but incomplete aneurysm occlusion, recurrences, rehemorrhages, and retreatments profoundly affect final outcomes and appear to erase the early advantages associated with endovascular therapy. The problems of efficacy and durability with endovascular aneurysm repair may not translate to a large or growing number of microsurgical cases of residual/recurrent aneurysms, but these problems remain important considerations in the selection of initial aneurysm therapy. Treatment Indications for Residual and Recurrent Aneurysms The data summarized above and our own (Table 2) demonstrate real rehemorrhage risks from residual/recurrent aneurysms after coiling and clipping and argue for intervention when detected. Vigilant long-term surveillance of aneurysm patients is essential. Patients with coiled aneurysms should be studied angiographically at 6 months and 2 years after treatment, with additional follow-up recommended when even minimal residual/recurrent aneurysm is observed. Patients with clipped aneurysms should be studied angiographically immediately postoperatively and 5 years after treatment, looking not only for recurrences but also for new aneurysm formation. Patients with multiple aneurysms, suspicious angiographic findings, and long life expectancy are studied further at 10 years and beyond. The rate of de novo aneurysm formation is low, estimated by Tsutsumi et al17 to be 1%/y in patients with treated aneurysm. Rehemorrhages from clipped aneurysm recurrences occur late, with median times of 13 years in our experience and 9 years in the experience of Tsutsumi et al, which emphasizes the need for prolonged surveillance. Coil recurrences present new anatomy that differs from a virgin aneurysm. The uniformly soft, collapsible sac of an untreated aneurysm is replaced by a hard dome, limited collapsibility, and a variable amount of compressibility at the neck. These features can make both endovascular and microsurgical retreatments more difficult. The risk of intraprocedural or intraoperative rupture is probably less with coiled aneurysms (none in our experience), but the coil mass is problematic. Aneurysm recurrences typically lack the spherical morphology that facilitates coiling. Insufficient coil compaction may not create enough space for clip application. A large coil mass will resist closure of clip blades or slide the clip down the neck and occlude the parent or branch arteries. The morphology of aneurysm recurrence influences the selection of endovascular or microsurgical retreatment. Coiling is preferred when recurrent aneurysm can be packed safely and densely, whereas clipping is preferred when the recurrence is sizable and there is sufficient soft neck for clipping. Clipping is also preferred when the recurrence is not simply compaction but regrowth, because regrowing aneurysms have pathological walls that are at increased risk of further recurrences with coiling. Clipping excludes this dysplastic tissue and is likely to be more durable. Although sometimes difficult to determine angiographically, coil extrusion may also indicate pathological aneurysm wall and indicate surgery over additional coiling.1,24 A predictor of clipability would help with surgical planning. We found a useful relationship between compaction height and coil width (Figure 6). Compaction height is the height of the aneurysm lumen beneath the coils, not the distance to the true dome. Coil width is the widest diameter of the coil mass, usually at the fundus of the aneurysm at the plane of compaction, parallel to the neck but perpendicular to the direction of anticipated clip application. Coils resist or oppose clip closure; therefore, coil width is measured in the plane perpendicular to the clip blades or, in other words, along the line of closure between the tips of opened clip blades. A ratio of coil width to compaction height (width/height [compaction ratio]) 90° (103° for compaction ratio of 2.5), which impedes clip blade closure, slides the clip down the aneurysm neck, and can occlude branch or parent arteries. In a previous report, 9 of 13 aneurysms (70%) with a compaction ratio >2.5 were not amenable to simple clipping and instead required complex clipping (clipping after thrombectomy or coil mobilization) or alternative techniques (bypass or wrapping). In contrast, 6 of 8 aneurysms (75%) with a compaction ratio ≤2.5 were amenable to simple clipping.1FIGURE 6: Compaction ratio may predict the clipability of recurrent aneurysms after coiling. Compaction height (H) is the height of the aneurysm lumen beneath the coils, not the distance to the true dome. Coil width (C) is the widest diameter of the coil mass, usually at the aneurysm fundus at the plane of compaction, parallel to the neck but perpendicular to the direction of anticipated clip application. A ratio of coil width to compaction height of <2.5 correlated with aneurysm clipability.Other morphological factors also affect clipability. Aneurysms with fusiform morphology or aberrant branches at the base might be unclippable. Aneurysms with minimal compaction (compaction height <2 mm) might not have room for clip application, and a delay in surgery might allow more compaction and easier clipping later. Microsurgical Techniques The preferred microsurgical treatment of coiled aneurysms is direct clipping, with the clip below coils, across a strand or two in the neck, or against a larger coil mass in the neck. A previously coiled aneurysm does not seem to require the perfect and complete crossing of the neck with clips like a virgin aneurysm does. Clipping a residual/recurrent aneurysm is generally simple and straightforward when it is clippable, but unclippable aneurysms require contingency plans. First, an unclippable aneurysm can be opened to mobilize coils away from the neck or extract coils, converting it into a clippable one. This maneuver is best accomplished by transecting the sac at the equator to maximize access to the coil mass, to visualize adhesions to the walls, and to leave plenty of aneurysmal tissue at the neck. Pulling coils through a small opening in the dome is unpredictable, and the extraction is poorly visualized. Complete extraction of coils is usually not necessary; coils are safest when left within the aneurysm walls, and coils need to be mobilized only enough to soften the neck and apply the clip. Coils can be removed easily from residual aneurysms that are operated acutely, but late recurrences have intraluminal fibrosis that makes coils adherent. Aneurysm transection requires complete trapping with temporary clips and sometimes debulking of intraluminal thrombus. This technique can be complicated by incomplete proximal and distal control, back-bleeding during the repair, adherent branch arteries or perforators, difficulty in reconstructing the neck after transecting the aneurysm, and dense scarring of the coils to the neck. Ischemia time can be prolonged; branch arteries are susceptible to thromboembolic occlusion; unexpected problems are inevitable; and the decision to pursue this option is irrevocable. Second, an unclippable aneurysm can be bypassed in combination with aneurysm trapping or proximal occlusion, thereby eliminating or reversing intra-aneurysmal blood flow. Proximal or distal occlusion does not exclude the aneurysm completely but causes aneurysm thrombosis through altered hemodynamics. With the current variety of extracranial-to-intracranial bypasses, in situ bypasses, and high-flow bypasses with saphenous veins or radial artery grafts, the territory of the parent artery of any unclippable aneurysm can be revascularized effectively.31-33 In contrast to aneurysm transection, a bypass strategy can be executed methodically, ischemia time is predictable, and direct aneurysm attack can be avoided. We were more likely to opt for the bypass contingency over the extraction contingency with unclippable aneurysms. Third, wrapping with muslin or cotton induces inflammation and scarring of the aneurysm wall and is a treatment of last resort. Although none of the wrapped aneurysms in our experience has enlarged or hemorrhaged since surgery, we have moved away from this technique and have not wrapped any residual/recurrent aneurysms during the last 10 years. Initial Selection of Aneurysm Treatment Perhaps the best way to deal with residual/recurrent aneurysms is to prevent them by initially selecting the better aneurysm treatment. Factors associated with residual/recurrent aneurysms after coiling include wide necks, low aspect ratios, and large and giant size.28,29,34 These factors influence packing density and completeness of obliteration. Other factors like intraluminal thrombosis make coiled aneurysms recanalize and recur, whereas clipping more definitively closes the aneurysm. Aneurysm location affects recurrence through morphological peculiarities such as at the MCA bifurcation where necks are broader than at other locations and through hemodynamic peculiarities such as at the basilar bifurcation where shear forces are more likely to cause compaction. ACoA aneurysms were more susceptible to remnants and recurrences in our experience. Curvature of the parent artery or aberrant origins of branch arteries can influence packing density and obliteration rates. The initial choice of aneurysm treatment is best made by a collaborative team of aneurysm specialists with expertise in both endovascular and microsurgical therapies. Consideration must be given not just to treatment risks but also to efficacy and durability. It is reasonable to select endovascular therapy preferentially for those aneurysms with favorable anatomy that optimizes treatment risks, efficacy, and durability. It is also reasonable to select microsurgery for those aneurysms with unfavorable endovascular anatomy that could be coiled safely but may not be obliterated completely and may have an increased recurrence risk. Careful treatment selection will avoid the complexities of retreating a coil recurrence later. Established algorithms or protocols for selecting aneurysm therapy are still lacking, and in the meantime, these challenging decisions demand individualized evaluations, multidisciplinary input, and good judgment that weighs the risk of aneurysm recurrence with other critical factors. CONCLUSION In the current endovascular era, microsurgical treatment of residual and recurrent aneurysms is necessary in a limited number of cases. Endovascular therapy is associated with significant rates of residual/recurrent aneurysms with a defined risk of rehemorrhage and neurological morbidity. Microsurgical treatment of residual/recurrent aneurysms is simple, safe, and effective in the majority of cases, with a small minority requiring contingency maneuvers, including dome transection with coil extraction and trapping with bypass. Regardless of the initial treatment modality, patients with treated aneurysms remain at risk for aneurysm recurrence, de novo aneurysm formation, and rehemorrhage, and aggressive management is recommended when long-term angiographic surveillance is positive. Disclosure The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article.