Association of endovascular therapy of very small ruptured aneurysms with higher rates of procedure-related rupture

Transcription

1 See the Editorial and the Response in this issue, p J Neurosurg 108: , 2008 Association of endovascular therapy of very small ruptured aneurysms with higher rates of procedure-related rupture THANH N. NGUYEN, M.D., 1 JEAN RAYMOND, M.D., 2 FRANÇOIS GUILBERT, M.D., 2 DANIEL ROY, M.D., 2 MAXIME D. BÉRUBÉ, M.D., 3 MOSTAFA MAHMOUD, M.D., 4 AND ALAIN WEILL, M.D. 2 1 Department of Neurology, Neurosurgery, and Radiology, Boston University Medical Center, Boston, Massachusetts; 2 Department of Radiology and 3 Division of Neurology, Centre Hospitalier de l Université de Montréal, Notre-Dame Hospital, Montreal, Quebec, Canada; and 4 Department of Radiology, Ains Shams University, Cairo, Egypt Object. Procedure-related rupture during endovascular therapy of intracranial aneurysms is associated with a mortality rate of more than one third. Previously ruptured aneurysms are a known risk factor for procedure-related rupture. The objective of this study was to evaluate whether very small, ruptured aneurysms are associated with more frequent intraprocedural ruptures. Methods. This was a retrospective cohort study in which the investigators examined consecutive ruptured aneurysms treated with coil embolization at a single institution. The study was approved by the institutional review board. Very small aneurysms were defined as 3 mm. Procedure-related rupture was defined as contrast extravasation during treatment. Univariate analysis with the Fisher exact test and the Mann Whitney U test was performed. Results. Between August 1992 and January 2007, 682 aneurysms were selectively treated with coils in 668 patients. Procedure-related rupture occurred in 7 (11.7%) of 60 aneurysms 3 mm, compared with 14 (2.3%) of 622 aneurysms 3 mm (relative risk 5.2, 95% confidence interval ; p 0.001). Among cases with procedure-related rupture, inflation of a compliant balloon was associated with better outcome (Glasgow Outcome Scale Score 4) compared with patients treated without balloon assistance (5 of 5 compared with 7 of 16; p = 0.05). Death resulting from procedure-related rupture occurred in 8 (38%) of 21 patients, and a vegetative state occurred in 1 patient. Clinical outcome was good in the other 12 patients (57%). Conclusions. Endovascular coil embolization of very small ( 3 mm) ruptured cerebral aneurysms is 5 times more likely to result in procedure-related rupture compared with larger aneurysms. Balloon inflation for hemostasis may be associated with better outcome in the event of intraprocedural rupture and merits further study. (DOI: /JNS/2008/108/6/1088) KEY WORDS cerebral aneurysm complication outcome procedure-related rupture ruptured aneurysm subarachnoid hemorrhage P ROCEDURE-RELATED rupture, one of the most feared complications of endovascular therapy for ruptured aneurysms, is associated with high rates of neurological disability (5 63%) 2,7 and mortality (20 63%). 4,12,13,16 In a large meta-analysis, ruptured aneurysm was confirmed as a risk factor for procedure-related rupture. 2 Small size has also been reported as a risk factor, 14,18 but this was recently challenged. 15 Confirmation of size as a risk factor for procedure-related rupture is important for better estimates of patient risk. Identification of endovascular strategies that could improve clinical outcome in the event of intraprocedural rupture can help to reduce the high morbidity and mortality rates associated with this complication. In this Abbreviations used in this paper: GOS = Glasgow Outcome Scale; ICP = intracranial pressure; MCA = middle cerebral artery; PICA = posterior inferior cerebellar artery; SCA = superior cerebellar artery. single-center study, we report on endovascular coil embolization of 682 ruptured aneurysms, in which there were 21 procedure-related ruptures. The purpose of the study was to evaluate primarily whether very small ruptured aneurysms are associated with higher procedure-related rupture rates, and secondarily, the impact of the use of a balloon-assisted coil insertion strategy on clinical consequences of procedure-related rupture. Clinical Materials and Methods Patients and Aneurysms This was a retrospective cohort study in which we examined consecutive patients with ruptured aneurysms treated with coil embolization, as extracted from a prospectively collected database containing cases treated between August 1992 and January The study was approved by our institutional review committee with waiver of informed con J. Neurosurg. / Volume 108 / June 2008

2 Higher procedure-related rupture rate in very small aneurysms sent. Patients were referred from neurosurgical centers for lesions that were judged to be too difficult for surgical treatment, after failed surgical attempts, or in most cases, after consensus was reached between the attending neurosurgeon and neuroradiologist in favor of endovascular treatment. In general, patients with advanced age, high Hunt and Hess grade, or multiple medical comorbidities were triaged to endovascular treatment. Patients with MCA aneurysms or the presence of a large hematoma tended to go to surgery. When possible, large or giant ruptured aneurysms, dissecting, or distal aneurysms (SCA, PICA) were treated with parent-vessel occlusion and were excluded from the analysis. Aneurysm size was estimated by visual estimation of the longest axis or by 3D reconstructive measurement and corrected according to a comparison with the diameter of the first coil used. The largest aneurysm size was recorded, even if the first coil was sized to a smaller transverse diameter. All aneurysm sizes were evaluated by neuroradiologists who had 10 years (A.W., J.R., D.R.) or 5 years (F.G.) of experience. Very small aneurysms were arbitrarily predefined in our institution as 3 mm. Endovascular Treatment The procedure for endovascular coil embolization has been previously described. 12 All patients were treated after induction of general anesthesia with systemic heparinization, most often after femoral artery puncture, with the goal activated clotting time 250 seconds (Medtronic). Protamine sulfate was always available for rapid anticoagulation reversal in anticipation of intraprocedural rupture. Procedures were performed on a monoplane C-arm angiographic system without 3D reconstruction between 1992 and 2004, and on biplane with 3D reconstruction after Multiple projections were used to identify the best angle of incidence to display the aneurysm neck. Coil insertion was performed using platinum coils (Guglielmi detachable coils, Boston Scientific; or Micrus coils, Micrus Therapeutics). Modified coils (Cerecyte, Micrus Therapeutics) were used in 11 cases (once, in a 3-mm aneurysm) in the context of a randomized study. In selected cases (aneurysms with large necks), balloon-assisted coil embolization (Hyperglide or Hyperform, Microtherapeutics) was performed according to the previously described remodeling technique. 10 Procedure-related rupture was defined as contrast extravasation during treatment. For this paper, the terms procedure-related rupture and perforation are used interchangeably. For statistical purposes, we grouped all causes of intraprocedural rupture together. Management of Procedure-Related Rupture On recognition of intraprocedural rupture, protamine was administered immediately in all cases. If there was no balloon on site prior to coil insertion, then coils were rapidly placed in the aneurysm while an assistant prepared for balloon catheterization. If a balloon was already in place prior to coil delivery, then it would be temporarily inflated and the aneurysm subsequently embolized with coils. In some cases, the rupture was so severe that a rapid rise in ICP ensued, with compromised cerebral blood flow and hemodynamic changes (hypertension, bradycardia). All efforts to control the ruptured site were rapidly undertaken J. Neurosurg. / Volume 108 / June 2008 via the endovascular route. At the same time, the rise in ICP was treated by mannitol, hyperventilation, and emergency ventricular drain insertion. The concomitant spike in blood pressure was not treated aggressively to avoid hampering cerebral perfusion pressure. Clinical Follow-Up Clinical follow-up for patients with procedure-related rupture was assessed using the GOS. Good outcome was defined as GOS 4 (moderate disability but independent) or 5 (good recovery). Statistical Analysis Univariate analysis with the Fisher exact test and the Mann Whitney U test was performed when appropriate, and 95% confidence intervals were calculated using statistical software (Version 2.5.6, StatsDirect). A probability value 0.05 was considered significant. Patients and Aneurysms Results We attempted treatment of 707 ruptured aneurysms in 693 patients. In 13 patients, multiple lesions were treated acutely (2 aneurysms in 12 patients, 3 aneurysms in 1 patient) because it was unclear which aneurysm had ruptured by the distribution of hemorrhage on the computed tomography scan. Treatment failure occurred in 14 aneurysms that were included in the analysis. Causes of treatment failure included the following: large aneurysm neck (9 lesions), vessel tortuosity (3), vasospasm and thrombosis (1), or unstable microcatheter positioning (1). Deliberate occlusion of the parent artery containing the aneurysm sac was performed in 25 patients. These patients were excluded from analysis, leaving 682 aneurysms in 668 patients. There were 477 women (71%) and 191 men (29%) with a mean age of 55.8 years (median 54 years, range years). The Hunt and Hess classification at presentation was Grade I in 179 patients, Grade II in 199, Grade III in 169, Grade IV in 98, and Grade V in 23. There were 60 very small aneurysms (46 aneurysms were 3 mm and 14 were 2 mm). There were 448 (66%) anterior circulation aneurysms and 234 (34%) were in the posterior circulation. The location for ruptured aneurysms treated with selective coil embolization is summarized in Table 1. Aneurysm Size and Procedure-Related Rupture The cumulative frequency of intraprocedural rupture was 21 (3.1%) of 682. The median size of aneurysms with procedure-related rupture was 5 mm compared with 7.5 mm of nonprocedure-related ruptured aneurysms. Intraprocedural rupture occurred in 7 (11.7%) of 60 aneurysms that were 3 mm compared with 14 (2.3%) of 622 aneurysms that were 3 mm, resulting in a relative risk of 5.2 (95% confidence interval ). Smaller size was significantly associated with procedure-related rupture according to the Mann Whitney U test (p 0.001). A summary of 21 patients with procedure-related rupture, including demographic characteristics, the lesion size and location, concomitant use of a balloon, and GOS score is given in Table

3 T. N. Nguyen et al. TABLE 1 Location of 682 ruptured aneurysms treated with coil embolization Very Small Overall Aneurysm Aneurysm Lesion Location No. (%) ( 3 mm) 3 mm ACoA 200 (29) A 1 segment 8 (1) 0 8 PCoA/AChA 103 (15) 5 98 carotido-ophthalmic artery 62 (9) 4 58 CA tip 15 (2) 1 14 MCA 35 (5) 5 30 PerA 25 (4) 4 21 total anterior circulation 448 (66) BA tip 164 (24) PICA 30 (4) 5 25 PCA 7 (1) 2 5 SCA 19 (3) 2 17 VBJ 13 (2) 2 11 AICA 1 (0.1) 1 0 total posterior circulation 234 (34) * AChA = anterior choroidal artery; ACoA = anterior communicating artery; AICA = anterior inferior cerebellar artery; BA = basilar artery; CA = carotid artery; PCA = posterior cerebral artery; PCoA = posterior communicating artery; PerA = pericallosal artery; VBJ = vertebrobasilar junction. Balloon Assistance and Procedure-Related Rupture Reasons for using balloon assistance were noted and included remodeling (102 cases), hemostasis (predefined in the reports) prior to coil insertion (4 procedures), hemostasis after procedure-related rupture (2), angioplasty for vasospasm (4), and balloon test occlusion (1). There were 3 procedure-related ruptures of the 102 cases in which a balloon was used for remodeling. Impact of Balloon Assistance in Management of Procedure-Related Rupture In 4 of the 7 patients with very small aneurysms that perforated, a balloon had been positioned adjacent to the aneurysm prior to procedure-related rupture. The reason for using a balloon was the fear of perforation in 2 cases, and in 2 cases it was balloon-assisted coil embolization of a wide-necked aneurysm. Following recognition of procedure-related rupture, immediate inflation of the balloon for hemostasis was performed in 3 patients, resulting in arrest of bleeding in all 3. The balloon permitted bleeding to be arrested for as long as 45 minutes in one of these 3 patients, allowing enough time to insert a ventricular drain, puncture the opposite groin, and insert another balloon via the opposite vertebral artery to trap and insert coils in the PICA-V 4 arterial segment that was ruptured. In the fourth patient, bleeding stopped rapidly while coils were placed. The GOS score was good in all 4 patients. Of the remaining 3 patients with very small aneurysms that ruptured during coil embolization performed without balloon assistance, two had undergone coil treatment in 1993 and Another patient, who was treated in 2005, developed increased ICP so rapidly that access for balloon navigation could not be visualized. All 3 patients died of the procedure-related rupture. In the other 14 patients with procedure-related rupture in aneurysms 3 mm, the balloon was placed prior to aneurysm coil embolization in 2 patients, and it was used with the intention of remodeling in one patient and hemostasis in the other. When aneurysm perforation was identified, the balloon was used for temporary hemostasis in both patients. Altogether, among patients with intraprocedural rupture, inflation of a compliant balloon as a hemostatic adjunct at TABLE 2 Summary of ruptured aneurysms perforated by coil embolization in 21 patients* Age EVD Case (yrs), Yr of Lesion Size H & H Balloon Precoil GOS No. Sex Procedure Location (mm) Grade Hemostasis Mechanism of Rupture Tx Score 1 58, M 1992 PCoA 30 III no coil yes , M 1992 BA 6 III no coil yes , F 1993 BA 3 III no coil yes , F 1993 MCA 4 I no coil no , F 1994 ACoA 3 II no coil no , F 1997 BA 6 II no coil no , F 1999 PICA 6 III no microguidewire yes , F 2002 BA 5 IV no coil yes , M 2002 PerA 5 IV no coil no , F 2003 PICA 2 III yes coil no , M 2003 BA 2 III yes microcatheter & coil yes , F 2003 ACoA 3 III no microcatheter withdrawal no , F 2004 ACoA 6 II yes coil no , F 2004 PerA 8 IV no coil yes , M 2004 MCA 5 I no microcatheter & coil no , M 2005 ACoA 5 I no coil no , M 2005 ACoA 2 I no coil no , F 2006 PCoA 10 III yes coil yes , F 2006 SCA 2 I yes coil no , F 2006 PICA 6 II no coil no , F 2006 ACoA 5 I no microcatheter no 1 * EVD = external ventricular drain; H & H = Hunt and Hess; Tx = treatment J. Neurosurg. / Volume 108 / June 2008

4 Higher procedure-related rupture rate in very small aneurysms the aneurysm was associated with better outcome (GOS Score 4 or 5) compared with patients in whom embolization was performed without balloon assistance (5 of 5 compared with 7 of 16, p = 0.05). Death caused by intraprocedural rupture occurred in 8 of 21 patients, and a vegetative state occurred in 1 patient. Of these 9 patients with bad outcomes, 5 had presented in good Hunt and Hess grade (I III). Clinical outcome was good (GOS Score 4 or 5) in the other 12 patients. Discussion J. Neurosurg. / Volume 108 / June 2008 Procedure-related rupture occurred in 3% of patients, and was associated with a high mortality rate (38%). These rates are comparable to those reported in the literature. 2 As noted by Sluzewski et al., 14 the clinical outcome of intraprocedural rupture reflects 2 extremes: patients either do very well or they die, which is probably related to the rapidity with which hemostasis can be achieved (or not) and ICP controlled (or not). The faster the hemostasis, the more rapid the control of ICP, and the better the neurological outcome (and vice versa). Very small aneurysms ( 3 mm) were found to be 5 times as likely to be associated with procedure-related rupture as larger aneurysms. In earlier published series, there was limited experience with endovascular coil embolization of very small aneurysms. In one of the first studies involving Guglielmi detachable coils, aneurysms 4 mm were not included in a series of 403 patients with subarachnoid hemorrhage. 18 In this report, small aneurysms were classified as 4 10 mm, and 9 of the 11 aneurysm perforations were seen within this subgroup. Also, the International Subarachnoid Aneurysm Trial 9 did not include aneurysms 3 mm (A Molyneux, personal communication, 2006). With the development of smaller, softer coils, aneurysms as small as 2 mm have been treated with coils, with varying success. 3,11 15,17 Suzuki et al. 15 recently published a series describing endovascular treatment of 21 patients with very small ruptured intracranial aneurysms ( 3 mm diameter) without procedure-related complications. All patients had to fulfill the criterion of a fundus/ neck ratio 1.5 to be treated. There are several potential explanations for the higher risk of intraprocedural rupture with smaller aneurysms. A smaller size of aneurysm sac restricts the freedom of movement for microcatheter and coil positioning. 14 Random displacement of a device by a few millimeters could lead to rupture in the more confined lumen of a small lesion, whereas this would be trivial in a larger aneurysm. 2 Another hypothesis is an increased probability that the coil or microcatheter will be inadvertently positioned in the vicinity of the initial rupture site in smaller compared with larger aneurysms. In surgical series of ruptured aneurysms, small size does not appear to affect the risk of intraoperative rupture, although very small aneurysms ( 3 mm) have not been isolated as a study group. 1,5,6 The high mortality and morbidity rates associated with procedure-related aneurysm perforation call for optimum preparation for cases considered high risk (that is, very small, ruptured aneurysms). In our series, in the event of procedure-related rupture, placement of a balloon adjacent to the aneurysm prior to coil embolization, with the intention of producing hemostasis or remodeling was associated with better outcome compared with results in patients treated without balloon assistance. Although the number of events for procedure-related rupture was small, multiple confounding factors can contribute to patient outcome, and further study with a larger series of patients is required to confirm this finding. It is likely that immediate balloon hemostasis helped to attenuate the instant rush of blood into the subarachnoid space along with the dangerous and rapid rise in ICP, which, in 1 patient, obscured our pathway for intracranial navigation. Having a balloon ready for hemostasis allows the practitioner time to place the coil rapidly in the aneurysm and its perforated site or plan for other strategies, such as insertion of a ventricular drain or trapping of an arterial segment. The regular use of balloon assistance in the treatment of small, ruptured aneurysms has been advocated by several teams (J. Moret and S. Cekirge, personal communications, both 2005). Initially conceived by Moret et al., 10 the use of balloon assistance was not associated with increased risk of procedure-related rupture in their early series (3 [5%] of 58 compared with 6 [2.5%] of 249, p = 0.38). In the series published by Moret et al., of the 3 cases of procedure-related rupture treated with balloon assistance, the balloon was inflated in 1 patient for hemostasis over 15 minutes, during which time a second coil was placed. All 3 patients were asymptomatic from the procedure-related rupture. Other endovascular strategies that can be considered in the management of procedure-related rupture include placement of additional coils, 4,8 a second microcatheter, 19 and a second balloon to trap an artery segment. There are several limitations to our study. This was a single-center, retrospective study. The longest diameter of the aneurysm was recorded to define its size, but this crude estimation probably does not capture the real risks of intraprocedural rupture related to size. Thus, some 5 2 mm aneurysms may be at higher risk of perforation than certain round 4-mm aneurysms. The small number of cases of intraprocedural rupture did not permit us to evaluate other possible predictors of such ruptures: aneurysm morphological features, location, and time between initial rupture and treatment. For example, certain morphological characteristics are believed to herald a higher risk of procedure-related rupture; these include small, bilobed (the so-called Mickey Mouse type) aneurysms that look like 2 daughter sacs on a common neck with no aneurysm body. Conclusions Procedure-related ruptures were 5 times as frequent during coil embolization of very small ( 3 mm) cerebral aneurysms compared with larger ones. Balloon-assisted hemostasis may be associated with better outcome and merits further study in a larger series. If the balloon can be navigated safely, balloon-assisted coil embolization may be considered in cases thought to be at high risk for perforation. Disclaimer We received no financial support for this study. References 1. Batjer H, Samson D: Intraoperative aneurysmal rupture: inci- 1091

5 T. N. Nguyen et al. dence, outcome, and suggestions for surgical management. Neurosurgery 18: , Cloft HJ, Kallmes DF: Cerebral aneurysm perforations complicating therapy with Guglielmi detachable coils: a meta-analysis. AJNR Am J Neuroradiol 23: , Cognard C, Weill A, Castaings L, Rey A, Moret J: Intracranial berry aneurysms: angiographic and clinical results after endovascular treatment. Radiology 206: , Doerfler A, Wanke I, Egelhof T, Dietrich U, Asgari S, Stolke D, et al: Aneurysmal rupture during embolization with Guglielmi Detachable Coils: causes, management, and outcome. AJNR Am J Neuroradiol 22: , Fridriksson S, Saveland H, Jakobsson KE, Edner G, Zygmunt S, Brandt L, et al: Intraoperative complications in aneurysm surgery: a prospective national study. J Neurosurg 96: , Houkin K, Kuroda S, Takahashi A, Takikawa S, Ishikawa T, Yoshimoto T, et al: Intra-operative premature rupture of the cerebral aneurysms. Analysis of the causes and management. Acta Neurochir (Wien) 141: , Johnston SC, Elijovich L: Predictors and outcome of intraprocedural rupture in patients treated for ruptured intracranial aneurysms: the CARAT study. Stroke 38:P489, 2007 (Abstract) 8. McDougall CG, Halbach VV, Dowd CF, Higashida RT, Larsen DW, Hieshima GB: Causes and management of aneurysmal hemorrhage occurring during embolization with Guglielmi detachable coils. J Neurosurg 89:87 92, Molyneux A, Kerr R, Stratton I, Sandercock P, Clarke M, Shrimpton J, et al: International Subarachnoid Aneurysm Trial (ISAT) of neurosurgical clipping versus endovascular coiling in 2143 patients with ruptured intracranial aneurysms: a randomised trial. Lancet 26: , Moret J, Cognard C, Weill A, Castaings L, Rey A: Reconstruction technique in the treatment of wide-neck intracranial aneurysms. Long-term angiographic and clinical results. Apropos of 56 cases. J Neuroradiol 24:30 44, Piotin M, Mounayer C, Spelle L, Williams MT, Moret J: Endovascular treatment of anterior choroidal artery aneurysms. AJNR Am J Neuroradiol 25: , Raymond J, Roy D: Safety and efficacy of endovascular treatment of acutely ruptured aneurysms. Neurosurgery 41: , Ricolfi F, Le Guerinel C, Blustajn J, Combes C, Brugieres P, Melon E, et al: Rupture during treatment of recently ruptured aneurysms with Guglielmi Electrodetachable Coils. AJNR Am J Neuroradiol 19: , Sluzewski M, Bosch JA, van Rooij WJ, Nijssen PCG, Wijnalda D: Rupture of intracranial aneurysms during treatment with Guglielmi detachable coils: incidence, outcome, and risk factors. J Neurosurg 94: , Suzuki S, Kurata A, Ohmomo T, Sagiuchi T, Niki J, Yamada M, et al: Endovascular surgery for very small ruptured aneurysms. Technical note. J Neurosurg 105: , Van Rooij WJ, Sluzewski M, Beute GN, Nijssen PC: Procedural complications of coiling of ruptured intracranial aneurysms: incidence and risk factors in a consecutive series of 681 patients. AJNR Am J Neuroradiol 27: , Vanninen R, Koivisto T, Saari T, Hernesniemi J, Vapalahti M: Ruptured intracranial aneurysms: acute endovascular treatment with electrolytically detachable coils a prospective randomized study. Radiology 211: , Viñuela F, Duckwiler G, Mawad M: Guglielmi detachable coil embolization of acute intracranial aneurysm: perioperative anatomical and clinical outcome in 403 patients. J Neurosurg 86: , Willinsky R, terbrugge K: Use of a second microcatheter in the management of a perforation during endovascular treatment of a cerebral aneurysm. AJNR Am J Neuroradiol 21: , 2000 Manuscript submitted August 11, Accepted October 5, Address correspondence to: Alain Weill, M.D., Department of Radiology, Notre Dame Hospital, 1560 Sherbrooke Street East, Montreal, Quebec H2L 4M1, Canada. alain.weill.chum@ssss. gouv.qc.ca J. Neurosurg. / Volume 108 / June 2008