British Journal of Neurosurgery, 2015; Early Online: 1–6 © 2015 The Neurosurgical Foundation ISSN: 0268-8697 print / ISSN 1360-046X online DOI: 10.3109/02688697.2015.1023774
ORIGINAL ARTICLE
Clipping of anterior communicating artery aneurysms in the early post-rupture stage via transorbital keyhole approach—Chinese neurosurgical experience Hui Wang†, Lun Luo†, Zhuopeng Ye†, Wensheng Li, Chuan Chen, Yueyang Ba, Xinjie Ning & Ying Guo
Br J Neurosurg Downloaded from informahealthcare.com by University New South Wales on 08/10/15 For personal use only.
Department of Neurosurgery, the Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, Guangdong province, China
all the aneurysms found in the anterior circulatory structure, AComA aneurysms are regarded as the most complex due to the angioarchitecture and flow dynamics of the AComA region, the frequent presence of anatomical variations, its deep interhemispheric location, and the danger of triggering neurological deficits resulting from the severance of perforators.2 Usually, conservative management of AComA aneurysms has a high mortality rate of over 44%.3 Recently developed methods of treatment focus on microsurgical and endovascular procedures.4 Apart from complications related to the magnitude of conventional craniotomy and the extensive brain retraction done without protection, less invasive methods such as supraorbital and transorbital keyhole approaches5 have begun to receive increasingly more attention. With rapid developments being made in neuroanesthesia and microneurosurgery, the transorbital keyhole approach can provide increased ventral access, thereby combining the advantages of minimal invasiveness with those of cranial base techniques6,7 and has begun to be used in the management of intracranial aneurysms.7,8 However, treatment of AComA aneurysms using the transorbital approach has only been reported three times thus far.6,9,10 In addition, due to the difficulty of managing possible intraoperative rupture, many surgeons are greatly concerned about implementing the keyhole approach, especially when treating those cases with ruptured aneurysms. In the present study, we report our success in managing 52 AComA aneurysms using transorbital approaches. Our experience proves that with careful preoperative evaluation, AComA aneurysms can indeed be sufficiently treated using the keyhole approach.
Abstract The anterior communicating artery (AComA) complex is the site at which intracranial aneurysms occur most frequently. At present, effective treatments for AComA aneurysms are yet to be developed. Here, we present our experience in successfully managing AComA aneurysms via the transorbital keyhole approach. A total of 52 patients having a history of aneurysm rupture received surgery. All patients were assigned a Hunt–Hess grade prior to surgery. The cistern was opened to expose the AComA complex using a keyhole approach, and aneurysms were then surgically clipped with the assistance of neuroendoscopy or indocyanine green angiography. Surgery outcomes were confirmed using computed tomography angiography (CTA). Each of the 52 AComA aneurysms was successfully clipped with a single operation. Three of these patients experienced intraoperative aneurysm rupture. Five had postoperative hydrocephalus which was successfully treated with ventriculoperitoneal shunt. All patients survived the surgical procedure. Using the Glasgow Outcome Scale scores for evaluation, 39 patients (75.0%) had good recovery, 9 (17.3%) had moderate disability, 2 (3.8%) had severe disability, and 2 patients who had been in preoperative comas (3.8%) remained in a vegetative state. During the follow-up period, CTA showed no recurrence of rupture or bleeding in all cases. Results of logistic analysis indicated that the transorbital keyhole approach was feasible based on the patients’ preoperative Hunt–Hess grades, which should be considered a priority in using this approach in the treatment of ruptured AComA aneurysms. Keywords: anterior communicating artery aneurysms; hunt–hess grade; microsurgery; transorbital approach
Introduction
Patients and methods
The anterior communicating artery (AComA) complex is the most common single location for the occurrence of intracranial aneurysms in surgical and autopsy series.1 Of
Patient population and data collection
†Hui
Between October 2009 and January 2014, a total of 52 patients having ruptured AComA aneurysms underwent transorbital
Wang, Lun Luo, and Zhuopeng Ye contributed equally to the study.
Correspondence: Ying Guo, Department of Neurosurgery, the Third Affiliated Hospital of Sun Yat-sen University, No. 600 Tianhe Rd, Tianhe District, Guangzhou, Guangdong Province, 510630 China. Tel: ⫹ 86-20-85253333. Fax: ⫹ 86-20-85253336. E-mail: [email protected] Received for publication 11 June 2014; accepted 21 February 2015
1
Br J Neurosurg Downloaded from informahealthcare.com by University New South Wales on 08/10/15 For personal use only.
2
H. Wang et al.
craniotomy. The demographic and clinical characteristics of these patients are presented in Table I. All patients had a history of subarachnoid hemorrhage (SAH), of which two cases had experienced SAH twice. The average age of the patients was 44.2 (range: 24–76 ) years. The presence of AComA aneurysm was confirmed in all the patients by a 320-slice computed tomography angiography (CTA) (Aquilion ONETM, Toshiba, Tokyo, Japan). The total number of aneurysms was 52 with diameters ranging from 2.4 to 12 mm. The dominant side of blood supply for aneurysms was found on the right in 23 cases (44.2%) and left in 29 cases (55.8%). Contralateral absence of the anterior cerebral artery was observed in eight cases (15.4%). Projections of aneurysms in all patients are summarized in Table I. Saccular or fusiform aneurysms were found in 37 patients (71.2) and irregular lobulated aneurysms were found in 15 patients (28.8%). All patients were operated on within 3 days of experiencing their last hemorrhage. Prior to surgery, the Hunt–Hess grading scale was applied for each patient as a means to evaluate their neurological conditions. The three patients having the complication of hydrocephalus received external ventricular drainage.
Transorbital “eyebrow” craniotomy The patient was in a supine position. The head was elevated 15° and turned 20°–30° to the side contralateral to the location of the planned craniotomy. The head was then slightly retroflexed 20° and fixed in place using a head Table I. Demographic and clinical characteristics of patients included in this study. Characteristics No. of patients (%) Mean age (range), year Gender Male Female Symptoms Sudden headache Nausea Vomit Disturbance of consciousness Meningeal irritation sign positive Hydrocephalus Intracranial hematomas Intraventricular hemorrhage Hunt–Hess grade I II III IV Fisher grade 1 2 3 4 Median aneurysm diameter (range), mm Projection Anterior Superior Posterior Inferior Complex Dominant blood supply On the right On the left Contralateral absence of anterior cerebral artery Shape Saccular or fusiform Irregular lobulated
44.2 (24–76) 21 (40.4) 31 (59.6) 52 (100) 48 (92.3) 39 (75.0) 9 (17.3) 52 (100) 5 (9.6) 6 (11.5) 7 (13.5) 12 (23.1) 28 (53.8) 9 (17.3) 3 (5.8) 0 (0.0) 24 (46.2) 23 (44.2) 5 (9.6) 7.4 (2.4–12) 9 (17.3) 17 (32.7) 12 (23.1) 9 (17.3) 5 (9.6) 23 (44.2) 29 (55.8) 8 (15.4)
holder (Fig. 1A). A skin incision was made starting at the outside edge of the supraorbital foramen on the eyebrow, continued along the supraorbital margin, and extended slightly outward from the lateral border of the eyebrow, going five millimeters beyond the zygomatic process of the frontal bone (Fig. 1B). The upper frontalis muscle was pulled up while the temporalis muscle was pulled down, thus exposing the zygomatic process of the frontal bone. The lower frontalis muscle and the orbicularis oculi muscle were pulled down to the orbital rim and fixed in place with temporal sutures (Fig. 1C). A single burr hole was placed closely posterior to the zygomatic process. Then, the frontal bone was cut from outside to inside along the supraorbital margin using a craniotome; the cutting process was stopped at the outside edge of the supraorbital foramen. A half-moon-shaped bone flap, approximately 25 mm by 20 mm, was then elevated to allow for adequate visualization of the surgical site (Fig. 1D). The dura was opened in a C-shape with its base toward the base of skull. After the cerebrospinal fluid (CSF) was drained, the carotid cistern, chiasmatic cistern, interpeduncular cistern, and the cistern of the lamina terminalis were each opened in turn. In cases of high intracranial pressure, a ventricular puncture was conducted. Once the aneurysm and parent artery were sufficiently exposed, an appropriate aneurysm clip was chosen to clip the neck of the aneurysm (Fig. 2A and 2B). Neuroendoscopy or indocyanine green (ICG) angiography was employed to monitor the state of the aneurysm, parent artery, and perforating artery. Following surgical clipping of the aneurysm, the dura was closed in standard watertight fashion. The bone flap was repositioned and fixed with titanium plates. The incision was sutured layer by layer, and the scalp was closed using a routine intradermal pattern with 4–0 gut suture.
Postoperative care and follow-up Postoperative care was carried out using intravenous administration of nimodipine for 10–14 days. Starting on the second day after surgery, daily lumbar punctures were carried out to facilitate drainage of bloody CSF. Drainage should be performed for at least 1 week and until CSF turns clear and colorless. Patients’ Glasgow Outcome Scale (GOS) scores were assessed at the time of hospital discharge. After hospital discharge, periodic follow-up was performed on all patients to monitor surgery-related complications and assess the recovery of neurological functions. Briefly, patients received follow-up at 1, 2, and 4 weeks in the first month after discharge. After which, the follow-up schedule was once every month for 3 months and then once every 3 months thereafter. All patients were instructed to report all and any clinical abnormalities that occurred during the entire follow-up period. CTA was conducted at least once in 3 months postoperatively to evaluate the outcome of the treatment. The follow-up period ranged between 3 and 50 (median: 28) months.
Statistical analysis 37 (71.2) 15 (28.8)
Statistical analysis was performed using SPSS software version 16 (SPSS, Chicago, IL, United States). Categorical variables
Br J Neurosurg Downloaded from informahealthcare.com by University New South Wales on 08/10/15 For personal use only.
Key-hole approach for AComA aneurysms 3
Fig. 1. Intraoperative photos showing head positioning (A), marked skin incision (B), surgical flap (C), and bone flap (D) during the craniotomy procedures in a representative patient. Seven days after surgery, good postoperative cosmetic results were achieved in this patient (E).
are expressed as counts and percentages. Kendall’s tau-b test was used to determine the correlation between ordinal categorical variables. Logistic regression analysis was carried out to explore factors that can influence and predict the outcome of surgery. The dependent variable used in logistic regression was the GOS score given at hospital discharge. All patients were divided into two groups according to this dependent variable: the good recovery group (GOS score ⫽ 5) and the poor recovery group (GOS score from 1 to 4). The independent variables used in logistic regression included age, sex,
Hunt–Hess grade, aneurysm diameter, aneurysm projection, and side of dominant blood supply. Age and sex were used as covariates in the analysis. Results are expressed as an odds ratio (OR) with a 95% confidence interval (CI).
Results Each of the 52 AComA aneurysms was successfully clipped in a single operation. Postoperative CTA examination confirmed complete occlusion of these aneurysms, as well as good patency of the parent and perforating arteries (Fig. 2C and 2D). The surgical outcomes are summarized in Table II. Table II. Clinical outcomes and postoperative complications of patients receiving surgery via the transorbital keyhole approach. No. of patients (%)
Fig. 2. Intraoperative photos showing the surgical view of an aneurysm (indicated by black arrow) in a representative case located between the A2 segments of both sides (A), and the clipping of this aneurysm (indicated by black arrow) (B). Preoperative CTA identified the location of this aneurysm (C). Postoperative CTA examination confirmed the complete occlusion of this aneurysm, as well as good patency of the parent and perforating arteries (D).
GOS scorea 5 (good recovery) 4 (moderate disability) 3 (sever disability) 2 (vegetable state) 1 (death) Postoperative complicationsc Permanent paralysis Temporary limb paralysis Mental disorder Cerebral infarction Subcutaneous fluid collection Infection CSF leakage Intracranial hematoma Acute hydrocephalus Delayed hydrocephalus
39 (75) 9 (17.3) 2 (3.8) 2 (3.8)b 0 (0) 0 (0) 1 (1.9) 6 (11.5) 0 (0) 1 (1.9) 0 (0) 0 (0) 0 (0) 5 (9.6) 1 (1.9)
GOS, Glasgow Outcome Scale; CSF, cerebrospinal fluid aGOS score was assessed at hospital discharge. bThese two patients had preoperative comas. cPostoperative complications refer to those complications that occurred during the hospital stay and follow-up period.
Br J Neurosurg Downloaded from informahealthcare.com by University New South Wales on 08/10/15 For personal use only.
4
H. Wang et al.
During surgery, 3 patients experienced aneurysmal rupture. The aneurysms in these patients had inferior/ anterior projections and irregular shapes. However, we did not identify any other special features when compared with the unruptured aneurysms. In these cases, the parent artery was temporarily occluded for 9 min, 5 min, and 6 min, respectively, before neck clipping was successfully applied. Additionally, 4 patients experienced gyrus rectus hemorrhage during surgery. Since the gyrus rectus is located in the midline of the brain, surgical treatment for these cases was conducted through a transorbital approach on the side where bleeding took place and hemorrhage was evacuated during surgery. Furthermore, neuroendoscopy was applied in 11 patients (21.2%), among which two cases experienced a change in clip position due to incorrect clipping of the dominant perforating artery. ICG angiography was used in 23 patients (44.2%), among which one case experienced clip position change due to stenosis at the A2 segment of the ipsilateral anterior cerebral artery. After surgery, all of the 52 patients experienced various degrees of swelling at the incision site, but swelling subsided within 1 week. One patient experienced subcutaneous fluid collection and was then treated by puncture and compression bandaging. Five patients suffered acute hydrocephalus during their hospital stay and were treated using a ventriculoperitoneal shunt. Of these cases, two developed mild mental disorders and the other three recovered completely. In addition, we observed eight patients in this study with angiographic vasospasm, five of whom had clinical vasospasm. In the five cases with clinical vasospasm, four developed mental disorders after surgery and recovered without sequelae; the remaining one experienced temporary limb paralysis on one side and, thus far, has not recovered completely. By the end of follow-up, no infection, CSF leakage, permanent paralysis, intracranial hematoma, or cerebral infarction had occurred. As shown in Table II, 39 patients (75.0%) had GOS scores of 5 (good recovery) at discharge with good cosmetic outcomes (Fig. 1E). Nine patients (17.3%) had GOS scores of 4 (moderate disability) and two (3.8%) had GOS scores of 3 (severe disability). Additionally, the two patients who had been in preoperative comas (3.8%) remained in a vegetative state (GOS score 2) after surgery. One patient, who had a Hunt–Hess grade of III, developed hydrocephalus 6 months after surgery and was then treatment with a ventriculoperitoneal shunt. Moreover, we explored the correlation between Fisher grades and clinical outcomes in patients overall using Kendall’s tau-b analysis. Briefly, all Fisher grade 2 patients (24 cases) had GOS scores of 5 after surgery. Among the 23 patients with a Fisher grade of 3, 14 showed good outcomes (GOS scores of 5 ), 8 showed GOS scores of 4, and only one case had a GOS score of 3. Of the patients with Fisher grades of 4 (5 cases), one had a GOS score of 5 and one had a score of 4; the other three showed poor outcomes (one with a GOS score of 3 and two with GOS scores of 2). The results of Kendall’s tau-b analysis showed that there was a significant correlation between both parameters (Kendall’s tau-b coefficient ⫽ ⫺ 0.567, P ⬍ 0.0001). Results of multiple logistic regression analysis are presented in Table III. The Hunt–Hess grade was the only
Table III. Logistic regression analysis for risk factors that influence surgical outcomes. Variable B S.E. OR (95% CI)* P value Age Sex Hunt–Hess score Aneurysm diameter Aneurysm projection Dominant blood supply
⫺ 0.901 ⫺ 1.474 1.867 ⫺ 0.473 ⫺ 0.256 ⫺ 0.085
0.825 0.899 0.626 0.763 0.271 0.938
0.406 (0.081–2.048) 0.229 (0.039–1.333) 6.466 (1.896–22.055) 0.623 (0.140–2.777) 0.774 (0.455–1.317) 0.918 (0.146–5.774)
0.275 0.101 0.003 0.535 0.345 0.928
B, regression coefficient; S.E. standard error; OR, odds ratio; CI, confidence interval. GOS score at hospital discharge as the dependent variable; age, sex, Hunt–Hess score, aneurysm diameter, aneurysm projection, and dominant blood supply as independent variables. *Adjusted by age and sex.
independent risk factor that was found to significantly influence the outcome of transorbital surgery (OR ⫽ 6.466, 95% CI ⫽ 1.896–22.055, P ⫽ 0.003), thereby implying that the Hunt–Hess grade should be assessed preoperatively and could serve as a predictor for surgical outcomes.
Discussion The transorbital keyhole approach provided minimal exposure and disruption of normal anatomy. In the current study, we presented our experience in successfully managing 52 ruptured AComA aneurysms by performing keyhole transorbital craniotomies. The key step to our success was assessing patients’ preoperative Hunt–Hess grades, which is demonstrated to be the independent risk factor that significantly influences surgical outcomes. According to our experience, we found that patients with Hunt–Hess grades of I–III are suitable for application of the transorbital approach because these patients are usually in a stable conscious condition without the presence of high intracranial pressure or severe brain swelling. In these patients, the keyhole approach was able to provide a path by which the lesion could be accessed with limited exposure, which allows most of the brain to remain under the protection of the skull. For patients with Hunt–Hess grades IV and V, however, additional exposure of the surrounding cisterns is usually required in order to obtain sufficient space for operation, due to the frequent presence of significant brain swelling in these patients. Therefore, we recommend that the inclusion criteria for use of supraorbital craniotomy versus a standard pterional or orbitozygomatic craniotomy should be based on the patient’s respective Hunt–Hess grade. The transorbital approach is generally suitable for patients with a Hunt–Hess grade of III or lower, and having no significant cerebral hemispheric swelling or shift of the midline structures due to an intracranial hematoma. For patients with Hunt–Hess grades IV and clinically significant cerebral edema, the pterional approach should be considered, especially for those who have shifted midline structures caused by a unilateral hematoma or require decompressive craniectomy. Special attention should be paid to acquiring a surgical view that offers the best exposure of the parent artery and aneurysm neck. The transorbital approach may not be suitable for AComA aneurysms that are 15 mm or more away from
Br J Neurosurg Downloaded from informahealthcare.com by University New South Wales on 08/10/15 For personal use only.
Key-hole approach for AComA aneurysms 5 the anterior clinoid process, especially for those with inferior projection, due to the poor surgical view of the neck portion. In such cases, part of the orbital roof needs to be removed in order to gain more exposure over the anterior cranial base, or other approaches should be considered. In addition, we found that use of neuroendoscopy or ICG angiography can help greatly to distinguish the aneurysm neck, perforating arteries surrounding the aneurysm, as well as clip position during surgery, thereby reducing the risk of accidental injury and improving the success rate of the surgery. However, in this study, two patients who had undergone neuroendoscopy experienced a change in clip position due to incorrect clipping of the dominant perforating artery, thereby indicating that further study is still needed to optimize the process of manipulation in exposing the perforating arteries in the anterior circle of Willis via a transorbital approach aided by neuroendoscopy or ICG angiography. Aneurysmal SAH is associated with high early mortality rates.11 There are concerns about the possible deleterious effects of early clipping surgery.12 However, Lan et al. claim that surgery can be performed any time after SAH provided that there is no occurrence of vasospasm.13 In line with this suggestion, all 52 cases reported here received surgery within 3 days of the latest hemorrhage and had good prognoses, as confirmed by the GOS grade given at the time of hospital discharge. Our results confirmed the benefits of early surgery in reducing the risk of rebleeding and facilitating the prevention and management of cerebral vasospasm and hydrocephalus following aneurysmal SAH. Additionally, it should be noted that vasospasm is not a contraindication to transorbital surgery, as in our study, five patients with clinical vasospasm had satisfactory clinical outcomes (three had GOS scores of 5, two had GOS scores of 4, and one had a GOS score of 3) and there was no significant correlation between the incidence of vasospasm and GOS scores. However, vasospasm may be associated with the occurrence of postoperative complications, because all the mentioned five cases developed postoperative complications. However, due to the small sample size of subjects in this study, this finding requires further validation. The most dangerous situation during the course of transorbital surgery is intraoperative aneurysmal rupture. In our study, in order to avoid such intraoperative ruptures, aneurysms were exposed in the subarachnoid cavity rather than inside the brain parenchyma. Sharp dissection is usually helpful. Proximal and distal ends of the parent artery should be isolated prior to isolation of the aneurysm. When necessary, the parent artery should be temporally occluded to prevent bleeding caused by intraoperative aneurysmal rupture. For small aneurysms or those with a broad and expansive neck, the A1 segment can first be occluded to reduce intracranial pressure prior to isolation. An aspirator should be used with great caution in order to avoid rupture of the aneurysm. Instead, bipolar electrocoagulation can be used to reshape the neck of the aneurysm to facilitate clipping. In addition, the clamp should match the neck of aneurysm and the patient’s systolic pressure should be maintained at around 80 mmHg during anesthesia. In case aneurysm rupture occurs, a brain retractor should still be mounted and the
surgical field should be immediately cleared by suction. The A1 segments of both sides are then temporally occluded and the neck of the aneurysm is isolated and quickly clipped. In our experience, intraoperative rupture of aneurysm occurred in three cases: after 9 min, 5 min, and 6 min, respectively; emergency treatments were performed according to the aforementioned procedures, and all ruptured aneurysms were clipped successfully, which was confirmed by CTA. Good recovery was achieved in all the patients. The following issues should also be taken into consideration for the prevention and management of intraoperative aneurysmal rupture: (1) based on preoperative analysis of image data, the projection of the aneurysm provides a reasonable means to predict the risk of intraoperative rupture; (2) the side of dominant blood flow is selected as the surgical side to ensure effective control of proximal blood supply prior to exposure of the aneurysm; and (3) for large aneurysms or those with wide necks, thrombi, calcified walls, or connections with parent arteries and perforating vessels, sufficient exposure and a larger operative space are necessary.
Conclusion For AComA aneurysms in the early postrupture stage, surgery via the transorbital keyhole approach is a valuable treatment option by virtue of its limited skin incision, favorable clinical and cosmetic outcomes, and low incidence rates of postoperative complications. However, its feasibility should be assessed based on the patient’s preoperative Hunt–Hess grade, which should be considered a top priority in the treatment of ruptured AComA aneurysms using this approach.
Acknowledgments This study was supported by the National Natural Science Foundation of China (No. 30901542), the Science and Technology Planning Project of Guangdong Province (No. 20120314), and the Science and Technology Planning Project of Guangzhou (No. 132000116).
Declaration of interest: The authors report no declarations of interest. The authors alone are responsible for the content and writing of the paper.
References 1. .Agrawal A , Kato Y, Chen L, et al. Anterior communicating artery aneurysms: an overview. Minim Invasive Neurosurg 2008;51: 131–5. 2. .Hernesniemi J, Dashti R, Lehecka M, et al. Microneurosurgical management of anterior communicating artery aneurysms. Surg Neurol 2008;70:8–28. 3. .Lee KC, Lee KS, Chung SS, Kim YS, Choi JU. Surgical treatment of anterior communicating artery aneurysms. Yonsei Med J 1982;23:131–45. 4. Proust F, Debono B, Hannequin D, et al. Treatment of anterior communicating artery aneurysms: complementary aspects of microsurgical and endovascular procedures. J Neurosurg 2003;99:3–14.
6
H. Wang et al.
Br J Neurosurg Downloaded from informahealthcare.com by University New South Wales on 08/10/15 For personal use only.
5. Fischer G, Stadie A , Reisch R, et al. The keyhole concept in aneurysm surgery: results of the past 20 years. Neurosurgery 2011;68:45–51. 6. Steiger HJ, Schmid-Elsaesser R, Stummer W, Uhl E. Transorbital keyhole approach to anterior communicating artery aneurysms. Neurosurgery 2001;48:347–51. 7. Beretta F, Andaluz N, Chalaala C, Bernucci C, Salud L, Zuccarello M. Image-guided anatomical and morphometric study of supraorbital and transorbital minicraniotomies to the sellar and perisellar regions: comparison with standard techniques. J Neurosurg 2010;113:975–81. 8. Ramos-Zúñiga R, Velázquez H, Barajas MA , López R, Sánchez E, Trejo S. Trans-supraorbital approach to supratentorial aneurysms. Neurosurgery 2002;51:125–30.
9. Zhou Y, Ao XS, Huang X, et al. Application of keyhole approach in operation of intracranial aneurysms. Zhonghua Yi Xue Za Zhi 2005;85:2250–3. 10. Grand W, Landi MK, Daré AO. Transorbital keyhole approach to anterior communicating artery aneurysms. Neurosurgery 2001;49:483–4. 11. Alaraj A , Charbel FT, Amin-Hanjani S. Peri-operative measures for treatment and prevention of cerebral vasospasm following subarachnoid hemorrhage. Neurol Res 2009;31:651–9. 12. Whitfield PC1, Kirkpatrick PJ. Timing of surgery for aneurysmal subarachnoid haemorrhage. Cochrane Database Syst Rev 2001; 2: CD001697. 13. Lan Q, Gong Z, Kang D, et al. Microsurgical experience with keyhole operations on intracranial aneurysms. Surg Neurol 2006; 66 Suppl 1: S2–9.
ORIGINAL ARTICLE
Clipping of anterior communicating artery aneurysms in the early post-rupture stage via transorbital keyhole approach—Chinese neurosurgical experience Hui Wang†, Lun Luo†, Zhuopeng Ye†, Wensheng Li, Chuan Chen, Yueyang Ba, Xinjie Ning & Ying Guo
Br J Neurosurg Downloaded from informahealthcare.com by University New South Wales on 08/10/15 For personal use only.
Department of Neurosurgery, the Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, Guangdong province, China
all the aneurysms found in the anterior circulatory structure, AComA aneurysms are regarded as the most complex due to the angioarchitecture and flow dynamics of the AComA region, the frequent presence of anatomical variations, its deep interhemispheric location, and the danger of triggering neurological deficits resulting from the severance of perforators.2 Usually, conservative management of AComA aneurysms has a high mortality rate of over 44%.3 Recently developed methods of treatment focus on microsurgical and endovascular procedures.4 Apart from complications related to the magnitude of conventional craniotomy and the extensive brain retraction done without protection, less invasive methods such as supraorbital and transorbital keyhole approaches5 have begun to receive increasingly more attention. With rapid developments being made in neuroanesthesia and microneurosurgery, the transorbital keyhole approach can provide increased ventral access, thereby combining the advantages of minimal invasiveness with those of cranial base techniques6,7 and has begun to be used in the management of intracranial aneurysms.7,8 However, treatment of AComA aneurysms using the transorbital approach has only been reported three times thus far.6,9,10 In addition, due to the difficulty of managing possible intraoperative rupture, many surgeons are greatly concerned about implementing the keyhole approach, especially when treating those cases with ruptured aneurysms. In the present study, we report our success in managing 52 AComA aneurysms using transorbital approaches. Our experience proves that with careful preoperative evaluation, AComA aneurysms can indeed be sufficiently treated using the keyhole approach.
Abstract The anterior communicating artery (AComA) complex is the site at which intracranial aneurysms occur most frequently. At present, effective treatments for AComA aneurysms are yet to be developed. Here, we present our experience in successfully managing AComA aneurysms via the transorbital keyhole approach. A total of 52 patients having a history of aneurysm rupture received surgery. All patients were assigned a Hunt–Hess grade prior to surgery. The cistern was opened to expose the AComA complex using a keyhole approach, and aneurysms were then surgically clipped with the assistance of neuroendoscopy or indocyanine green angiography. Surgery outcomes were confirmed using computed tomography angiography (CTA). Each of the 52 AComA aneurysms was successfully clipped with a single operation. Three of these patients experienced intraoperative aneurysm rupture. Five had postoperative hydrocephalus which was successfully treated with ventriculoperitoneal shunt. All patients survived the surgical procedure. Using the Glasgow Outcome Scale scores for evaluation, 39 patients (75.0%) had good recovery, 9 (17.3%) had moderate disability, 2 (3.8%) had severe disability, and 2 patients who had been in preoperative comas (3.8%) remained in a vegetative state. During the follow-up period, CTA showed no recurrence of rupture or bleeding in all cases. Results of logistic analysis indicated that the transorbital keyhole approach was feasible based on the patients’ preoperative Hunt–Hess grades, which should be considered a priority in using this approach in the treatment of ruptured AComA aneurysms. Keywords: anterior communicating artery aneurysms; hunt–hess grade; microsurgery; transorbital approach
Introduction
Patients and methods
The anterior communicating artery (AComA) complex is the most common single location for the occurrence of intracranial aneurysms in surgical and autopsy series.1 Of
Patient population and data collection
†Hui
Between October 2009 and January 2014, a total of 52 patients having ruptured AComA aneurysms underwent transorbital
Wang, Lun Luo, and Zhuopeng Ye contributed equally to the study.
Correspondence: Ying Guo, Department of Neurosurgery, the Third Affiliated Hospital of Sun Yat-sen University, No. 600 Tianhe Rd, Tianhe District, Guangzhou, Guangdong Province, 510630 China. Tel: ⫹ 86-20-85253333. Fax: ⫹ 86-20-85253336. E-mail: [email protected] Received for publication 11 June 2014; accepted 21 February 2015
1
Br J Neurosurg Downloaded from informahealthcare.com by University New South Wales on 08/10/15 For personal use only.
2
H. Wang et al.
craniotomy. The demographic and clinical characteristics of these patients are presented in Table I. All patients had a history of subarachnoid hemorrhage (SAH), of which two cases had experienced SAH twice. The average age of the patients was 44.2 (range: 24–76 ) years. The presence of AComA aneurysm was confirmed in all the patients by a 320-slice computed tomography angiography (CTA) (Aquilion ONETM, Toshiba, Tokyo, Japan). The total number of aneurysms was 52 with diameters ranging from 2.4 to 12 mm. The dominant side of blood supply for aneurysms was found on the right in 23 cases (44.2%) and left in 29 cases (55.8%). Contralateral absence of the anterior cerebral artery was observed in eight cases (15.4%). Projections of aneurysms in all patients are summarized in Table I. Saccular or fusiform aneurysms were found in 37 patients (71.2) and irregular lobulated aneurysms were found in 15 patients (28.8%). All patients were operated on within 3 days of experiencing their last hemorrhage. Prior to surgery, the Hunt–Hess grading scale was applied for each patient as a means to evaluate their neurological conditions. The three patients having the complication of hydrocephalus received external ventricular drainage.
Transorbital “eyebrow” craniotomy The patient was in a supine position. The head was elevated 15° and turned 20°–30° to the side contralateral to the location of the planned craniotomy. The head was then slightly retroflexed 20° and fixed in place using a head Table I. Demographic and clinical characteristics of patients included in this study. Characteristics No. of patients (%) Mean age (range), year Gender Male Female Symptoms Sudden headache Nausea Vomit Disturbance of consciousness Meningeal irritation sign positive Hydrocephalus Intracranial hematomas Intraventricular hemorrhage Hunt–Hess grade I II III IV Fisher grade 1 2 3 4 Median aneurysm diameter (range), mm Projection Anterior Superior Posterior Inferior Complex Dominant blood supply On the right On the left Contralateral absence of anterior cerebral artery Shape Saccular or fusiform Irregular lobulated
44.2 (24–76) 21 (40.4) 31 (59.6) 52 (100) 48 (92.3) 39 (75.0) 9 (17.3) 52 (100) 5 (9.6) 6 (11.5) 7 (13.5) 12 (23.1) 28 (53.8) 9 (17.3) 3 (5.8) 0 (0.0) 24 (46.2) 23 (44.2) 5 (9.6) 7.4 (2.4–12) 9 (17.3) 17 (32.7) 12 (23.1) 9 (17.3) 5 (9.6) 23 (44.2) 29 (55.8) 8 (15.4)
holder (Fig. 1A). A skin incision was made starting at the outside edge of the supraorbital foramen on the eyebrow, continued along the supraorbital margin, and extended slightly outward from the lateral border of the eyebrow, going five millimeters beyond the zygomatic process of the frontal bone (Fig. 1B). The upper frontalis muscle was pulled up while the temporalis muscle was pulled down, thus exposing the zygomatic process of the frontal bone. The lower frontalis muscle and the orbicularis oculi muscle were pulled down to the orbital rim and fixed in place with temporal sutures (Fig. 1C). A single burr hole was placed closely posterior to the zygomatic process. Then, the frontal bone was cut from outside to inside along the supraorbital margin using a craniotome; the cutting process was stopped at the outside edge of the supraorbital foramen. A half-moon-shaped bone flap, approximately 25 mm by 20 mm, was then elevated to allow for adequate visualization of the surgical site (Fig. 1D). The dura was opened in a C-shape with its base toward the base of skull. After the cerebrospinal fluid (CSF) was drained, the carotid cistern, chiasmatic cistern, interpeduncular cistern, and the cistern of the lamina terminalis were each opened in turn. In cases of high intracranial pressure, a ventricular puncture was conducted. Once the aneurysm and parent artery were sufficiently exposed, an appropriate aneurysm clip was chosen to clip the neck of the aneurysm (Fig. 2A and 2B). Neuroendoscopy or indocyanine green (ICG) angiography was employed to monitor the state of the aneurysm, parent artery, and perforating artery. Following surgical clipping of the aneurysm, the dura was closed in standard watertight fashion. The bone flap was repositioned and fixed with titanium plates. The incision was sutured layer by layer, and the scalp was closed using a routine intradermal pattern with 4–0 gut suture.
Postoperative care and follow-up Postoperative care was carried out using intravenous administration of nimodipine for 10–14 days. Starting on the second day after surgery, daily lumbar punctures were carried out to facilitate drainage of bloody CSF. Drainage should be performed for at least 1 week and until CSF turns clear and colorless. Patients’ Glasgow Outcome Scale (GOS) scores were assessed at the time of hospital discharge. After hospital discharge, periodic follow-up was performed on all patients to monitor surgery-related complications and assess the recovery of neurological functions. Briefly, patients received follow-up at 1, 2, and 4 weeks in the first month after discharge. After which, the follow-up schedule was once every month for 3 months and then once every 3 months thereafter. All patients were instructed to report all and any clinical abnormalities that occurred during the entire follow-up period. CTA was conducted at least once in 3 months postoperatively to evaluate the outcome of the treatment. The follow-up period ranged between 3 and 50 (median: 28) months.
Statistical analysis 37 (71.2) 15 (28.8)
Statistical analysis was performed using SPSS software version 16 (SPSS, Chicago, IL, United States). Categorical variables
Br J Neurosurg Downloaded from informahealthcare.com by University New South Wales on 08/10/15 For personal use only.
Key-hole approach for AComA aneurysms 3
Fig. 1. Intraoperative photos showing head positioning (A), marked skin incision (B), surgical flap (C), and bone flap (D) during the craniotomy procedures in a representative patient. Seven days after surgery, good postoperative cosmetic results were achieved in this patient (E).
are expressed as counts and percentages. Kendall’s tau-b test was used to determine the correlation between ordinal categorical variables. Logistic regression analysis was carried out to explore factors that can influence and predict the outcome of surgery. The dependent variable used in logistic regression was the GOS score given at hospital discharge. All patients were divided into two groups according to this dependent variable: the good recovery group (GOS score ⫽ 5) and the poor recovery group (GOS score from 1 to 4). The independent variables used in logistic regression included age, sex,
Hunt–Hess grade, aneurysm diameter, aneurysm projection, and side of dominant blood supply. Age and sex were used as covariates in the analysis. Results are expressed as an odds ratio (OR) with a 95% confidence interval (CI).
Results Each of the 52 AComA aneurysms was successfully clipped in a single operation. Postoperative CTA examination confirmed complete occlusion of these aneurysms, as well as good patency of the parent and perforating arteries (Fig. 2C and 2D). The surgical outcomes are summarized in Table II. Table II. Clinical outcomes and postoperative complications of patients receiving surgery via the transorbital keyhole approach. No. of patients (%)
Fig. 2. Intraoperative photos showing the surgical view of an aneurysm (indicated by black arrow) in a representative case located between the A2 segments of both sides (A), and the clipping of this aneurysm (indicated by black arrow) (B). Preoperative CTA identified the location of this aneurysm (C). Postoperative CTA examination confirmed the complete occlusion of this aneurysm, as well as good patency of the parent and perforating arteries (D).
GOS scorea 5 (good recovery) 4 (moderate disability) 3 (sever disability) 2 (vegetable state) 1 (death) Postoperative complicationsc Permanent paralysis Temporary limb paralysis Mental disorder Cerebral infarction Subcutaneous fluid collection Infection CSF leakage Intracranial hematoma Acute hydrocephalus Delayed hydrocephalus
39 (75) 9 (17.3) 2 (3.8) 2 (3.8)b 0 (0) 0 (0) 1 (1.9) 6 (11.5) 0 (0) 1 (1.9) 0 (0) 0 (0) 0 (0) 5 (9.6) 1 (1.9)
GOS, Glasgow Outcome Scale; CSF, cerebrospinal fluid aGOS score was assessed at hospital discharge. bThese two patients had preoperative comas. cPostoperative complications refer to those complications that occurred during the hospital stay and follow-up period.
Br J Neurosurg Downloaded from informahealthcare.com by University New South Wales on 08/10/15 For personal use only.
4
H. Wang et al.
During surgery, 3 patients experienced aneurysmal rupture. The aneurysms in these patients had inferior/ anterior projections and irregular shapes. However, we did not identify any other special features when compared with the unruptured aneurysms. In these cases, the parent artery was temporarily occluded for 9 min, 5 min, and 6 min, respectively, before neck clipping was successfully applied. Additionally, 4 patients experienced gyrus rectus hemorrhage during surgery. Since the gyrus rectus is located in the midline of the brain, surgical treatment for these cases was conducted through a transorbital approach on the side where bleeding took place and hemorrhage was evacuated during surgery. Furthermore, neuroendoscopy was applied in 11 patients (21.2%), among which two cases experienced a change in clip position due to incorrect clipping of the dominant perforating artery. ICG angiography was used in 23 patients (44.2%), among which one case experienced clip position change due to stenosis at the A2 segment of the ipsilateral anterior cerebral artery. After surgery, all of the 52 patients experienced various degrees of swelling at the incision site, but swelling subsided within 1 week. One patient experienced subcutaneous fluid collection and was then treated by puncture and compression bandaging. Five patients suffered acute hydrocephalus during their hospital stay and were treated using a ventriculoperitoneal shunt. Of these cases, two developed mild mental disorders and the other three recovered completely. In addition, we observed eight patients in this study with angiographic vasospasm, five of whom had clinical vasospasm. In the five cases with clinical vasospasm, four developed mental disorders after surgery and recovered without sequelae; the remaining one experienced temporary limb paralysis on one side and, thus far, has not recovered completely. By the end of follow-up, no infection, CSF leakage, permanent paralysis, intracranial hematoma, or cerebral infarction had occurred. As shown in Table II, 39 patients (75.0%) had GOS scores of 5 (good recovery) at discharge with good cosmetic outcomes (Fig. 1E). Nine patients (17.3%) had GOS scores of 4 (moderate disability) and two (3.8%) had GOS scores of 3 (severe disability). Additionally, the two patients who had been in preoperative comas (3.8%) remained in a vegetative state (GOS score 2) after surgery. One patient, who had a Hunt–Hess grade of III, developed hydrocephalus 6 months after surgery and was then treatment with a ventriculoperitoneal shunt. Moreover, we explored the correlation between Fisher grades and clinical outcomes in patients overall using Kendall’s tau-b analysis. Briefly, all Fisher grade 2 patients (24 cases) had GOS scores of 5 after surgery. Among the 23 patients with a Fisher grade of 3, 14 showed good outcomes (GOS scores of 5 ), 8 showed GOS scores of 4, and only one case had a GOS score of 3. Of the patients with Fisher grades of 4 (5 cases), one had a GOS score of 5 and one had a score of 4; the other three showed poor outcomes (one with a GOS score of 3 and two with GOS scores of 2). The results of Kendall’s tau-b analysis showed that there was a significant correlation between both parameters (Kendall’s tau-b coefficient ⫽ ⫺ 0.567, P ⬍ 0.0001). Results of multiple logistic regression analysis are presented in Table III. The Hunt–Hess grade was the only
Table III. Logistic regression analysis for risk factors that influence surgical outcomes. Variable B S.E. OR (95% CI)* P value Age Sex Hunt–Hess score Aneurysm diameter Aneurysm projection Dominant blood supply
⫺ 0.901 ⫺ 1.474 1.867 ⫺ 0.473 ⫺ 0.256 ⫺ 0.085
0.825 0.899 0.626 0.763 0.271 0.938
0.406 (0.081–2.048) 0.229 (0.039–1.333) 6.466 (1.896–22.055) 0.623 (0.140–2.777) 0.774 (0.455–1.317) 0.918 (0.146–5.774)
0.275 0.101 0.003 0.535 0.345 0.928
B, regression coefficient; S.E. standard error; OR, odds ratio; CI, confidence interval. GOS score at hospital discharge as the dependent variable; age, sex, Hunt–Hess score, aneurysm diameter, aneurysm projection, and dominant blood supply as independent variables. *Adjusted by age and sex.
independent risk factor that was found to significantly influence the outcome of transorbital surgery (OR ⫽ 6.466, 95% CI ⫽ 1.896–22.055, P ⫽ 0.003), thereby implying that the Hunt–Hess grade should be assessed preoperatively and could serve as a predictor for surgical outcomes.
Discussion The transorbital keyhole approach provided minimal exposure and disruption of normal anatomy. In the current study, we presented our experience in successfully managing 52 ruptured AComA aneurysms by performing keyhole transorbital craniotomies. The key step to our success was assessing patients’ preoperative Hunt–Hess grades, which is demonstrated to be the independent risk factor that significantly influences surgical outcomes. According to our experience, we found that patients with Hunt–Hess grades of I–III are suitable for application of the transorbital approach because these patients are usually in a stable conscious condition without the presence of high intracranial pressure or severe brain swelling. In these patients, the keyhole approach was able to provide a path by which the lesion could be accessed with limited exposure, which allows most of the brain to remain under the protection of the skull. For patients with Hunt–Hess grades IV and V, however, additional exposure of the surrounding cisterns is usually required in order to obtain sufficient space for operation, due to the frequent presence of significant brain swelling in these patients. Therefore, we recommend that the inclusion criteria for use of supraorbital craniotomy versus a standard pterional or orbitozygomatic craniotomy should be based on the patient’s respective Hunt–Hess grade. The transorbital approach is generally suitable for patients with a Hunt–Hess grade of III or lower, and having no significant cerebral hemispheric swelling or shift of the midline structures due to an intracranial hematoma. For patients with Hunt–Hess grades IV and clinically significant cerebral edema, the pterional approach should be considered, especially for those who have shifted midline structures caused by a unilateral hematoma or require decompressive craniectomy. Special attention should be paid to acquiring a surgical view that offers the best exposure of the parent artery and aneurysm neck. The transorbital approach may not be suitable for AComA aneurysms that are 15 mm or more away from
Br J Neurosurg Downloaded from informahealthcare.com by University New South Wales on 08/10/15 For personal use only.
Key-hole approach for AComA aneurysms 5 the anterior clinoid process, especially for those with inferior projection, due to the poor surgical view of the neck portion. In such cases, part of the orbital roof needs to be removed in order to gain more exposure over the anterior cranial base, or other approaches should be considered. In addition, we found that use of neuroendoscopy or ICG angiography can help greatly to distinguish the aneurysm neck, perforating arteries surrounding the aneurysm, as well as clip position during surgery, thereby reducing the risk of accidental injury and improving the success rate of the surgery. However, in this study, two patients who had undergone neuroendoscopy experienced a change in clip position due to incorrect clipping of the dominant perforating artery, thereby indicating that further study is still needed to optimize the process of manipulation in exposing the perforating arteries in the anterior circle of Willis via a transorbital approach aided by neuroendoscopy or ICG angiography. Aneurysmal SAH is associated with high early mortality rates.11 There are concerns about the possible deleterious effects of early clipping surgery.12 However, Lan et al. claim that surgery can be performed any time after SAH provided that there is no occurrence of vasospasm.13 In line with this suggestion, all 52 cases reported here received surgery within 3 days of the latest hemorrhage and had good prognoses, as confirmed by the GOS grade given at the time of hospital discharge. Our results confirmed the benefits of early surgery in reducing the risk of rebleeding and facilitating the prevention and management of cerebral vasospasm and hydrocephalus following aneurysmal SAH. Additionally, it should be noted that vasospasm is not a contraindication to transorbital surgery, as in our study, five patients with clinical vasospasm had satisfactory clinical outcomes (three had GOS scores of 5, two had GOS scores of 4, and one had a GOS score of 3) and there was no significant correlation between the incidence of vasospasm and GOS scores. However, vasospasm may be associated with the occurrence of postoperative complications, because all the mentioned five cases developed postoperative complications. However, due to the small sample size of subjects in this study, this finding requires further validation. The most dangerous situation during the course of transorbital surgery is intraoperative aneurysmal rupture. In our study, in order to avoid such intraoperative ruptures, aneurysms were exposed in the subarachnoid cavity rather than inside the brain parenchyma. Sharp dissection is usually helpful. Proximal and distal ends of the parent artery should be isolated prior to isolation of the aneurysm. When necessary, the parent artery should be temporally occluded to prevent bleeding caused by intraoperative aneurysmal rupture. For small aneurysms or those with a broad and expansive neck, the A1 segment can first be occluded to reduce intracranial pressure prior to isolation. An aspirator should be used with great caution in order to avoid rupture of the aneurysm. Instead, bipolar electrocoagulation can be used to reshape the neck of the aneurysm to facilitate clipping. In addition, the clamp should match the neck of aneurysm and the patient’s systolic pressure should be maintained at around 80 mmHg during anesthesia. In case aneurysm rupture occurs, a brain retractor should still be mounted and the
surgical field should be immediately cleared by suction. The A1 segments of both sides are then temporally occluded and the neck of the aneurysm is isolated and quickly clipped. In our experience, intraoperative rupture of aneurysm occurred in three cases: after 9 min, 5 min, and 6 min, respectively; emergency treatments were performed according to the aforementioned procedures, and all ruptured aneurysms were clipped successfully, which was confirmed by CTA. Good recovery was achieved in all the patients. The following issues should also be taken into consideration for the prevention and management of intraoperative aneurysmal rupture: (1) based on preoperative analysis of image data, the projection of the aneurysm provides a reasonable means to predict the risk of intraoperative rupture; (2) the side of dominant blood flow is selected as the surgical side to ensure effective control of proximal blood supply prior to exposure of the aneurysm; and (3) for large aneurysms or those with wide necks, thrombi, calcified walls, or connections with parent arteries and perforating vessels, sufficient exposure and a larger operative space are necessary.
Conclusion For AComA aneurysms in the early postrupture stage, surgery via the transorbital keyhole approach is a valuable treatment option by virtue of its limited skin incision, favorable clinical and cosmetic outcomes, and low incidence rates of postoperative complications. However, its feasibility should be assessed based on the patient’s preoperative Hunt–Hess grade, which should be considered a top priority in the treatment of ruptured AComA aneurysms using this approach.
Acknowledgments This study was supported by the National Natural Science Foundation of China (No. 30901542), the Science and Technology Planning Project of Guangdong Province (No. 20120314), and the Science and Technology Planning Project of Guangzhou (No. 132000116).
Declaration of interest: The authors report no declarations of interest. The authors alone are responsible for the content and writing of the paper.
References 1. .Agrawal A , Kato Y, Chen L, et al. Anterior communicating artery aneurysms: an overview. Minim Invasive Neurosurg 2008;51: 131–5. 2. .Hernesniemi J, Dashti R, Lehecka M, et al. Microneurosurgical management of anterior communicating artery aneurysms. Surg Neurol 2008;70:8–28. 3. .Lee KC, Lee KS, Chung SS, Kim YS, Choi JU. Surgical treatment of anterior communicating artery aneurysms. Yonsei Med J 1982;23:131–45. 4. Proust F, Debono B, Hannequin D, et al. Treatment of anterior communicating artery aneurysms: complementary aspects of microsurgical and endovascular procedures. J Neurosurg 2003;99:3–14.
6
H. Wang et al.
Br J Neurosurg Downloaded from informahealthcare.com by University New South Wales on 08/10/15 For personal use only.
5. Fischer G, Stadie A , Reisch R, et al. The keyhole concept in aneurysm surgery: results of the past 20 years. Neurosurgery 2011;68:45–51. 6. Steiger HJ, Schmid-Elsaesser R, Stummer W, Uhl E. Transorbital keyhole approach to anterior communicating artery aneurysms. Neurosurgery 2001;48:347–51. 7. Beretta F, Andaluz N, Chalaala C, Bernucci C, Salud L, Zuccarello M. Image-guided anatomical and morphometric study of supraorbital and transorbital minicraniotomies to the sellar and perisellar regions: comparison with standard techniques. J Neurosurg 2010;113:975–81. 8. Ramos-Zúñiga R, Velázquez H, Barajas MA , López R, Sánchez E, Trejo S. Trans-supraorbital approach to supratentorial aneurysms. Neurosurgery 2002;51:125–30.
9. Zhou Y, Ao XS, Huang X, et al. Application of keyhole approach in operation of intracranial aneurysms. Zhonghua Yi Xue Za Zhi 2005;85:2250–3. 10. Grand W, Landi MK, Daré AO. Transorbital keyhole approach to anterior communicating artery aneurysms. Neurosurgery 2001;49:483–4. 11. Alaraj A , Charbel FT, Amin-Hanjani S. Peri-operative measures for treatment and prevention of cerebral vasospasm following subarachnoid hemorrhage. Neurol Res 2009;31:651–9. 12. Whitfield PC1, Kirkpatrick PJ. Timing of surgery for aneurysmal subarachnoid haemorrhage. Cochrane Database Syst Rev 2001; 2: CD001697. 13. Lan Q, Gong Z, Kang D, et al. Microsurgical experience with keyhole operations on intracranial aneurysms. Surg Neurol 2006; 66 Suppl 1: S2–9.
