- Email: [email protected]
Risk Factor Analysis for Poor Outcomes in Supraorbital Keyhole Aneurysm Clipping for Ruptured Anterior Circulation Aneurysms
Accepted Manuscript Risk factor analysis for poor outcomes in supraorbital keyhole aneurysm clipping for ruptured anterior circulation aneurysm Jun Sang Park, MD, Hoon Kim, MD, Min Woo Baik, MD, Ik Seong Park, MD, PhD, MBA PII:
S1878-8750(17)32187-3
DOI:
10.1016/j.wneu.2017.12.071
Reference:
WNEU 7084
To appear in:
World Neurosurgery
Received Date: 10 September 2017 Revised Date:
10 December 2017
Accepted Date: 11 December 2017
Please cite this article as: Park JS, Kim H, Baik MW, Park IS, Risk factor analysis for poor outcomes in supraorbital keyhole aneurysm clipping for ruptured anterior circulation aneurysm, World Neurosurgery (2018), doi: 10.1016/j.wneu.2017.12.071. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1 Park
ACCEPTED MANUSCRIPT
Risk factor analysis for poor outcomes in supraorbital keyhole aneurysm clipping for ruptured anterior circulation aneurysm
RI PT
Jun Sang Park, MD, Hoon Kim, MD, Min Woo Baik, MD, Ik Seong Park, MD, PhD,MBA
SC
Department of Neurosurgery, Bucheon St. Mary’s Hospital, College of Medicine, The
M AN U
Catholic University of Korea, Bucheon, Republic of Korea.
Correspondence author: Ik Seong Park, MD
Department of Neurosurgery, Bucheon St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, 327, Sosa-ro, Bucheon-si, Gyeonggi-do, 14647, Republic of
Tel: +82-32-340-7212 Fax: +82-32-340-7391
TE D
Korea
AC C
EP
E-mail: [email protected]
2 Park
ACCEPTED MANUSCRIPT Key words: Supraorbital keyhole surgery, Aneurysm clipping, Subarachnoid hemorrhage
RI PT
Abbreviations and Acronmyms
AC C
EP
TE D
M AN U
SC
anterior cerebral artery (ACA), cerebrospinal fluid (CSF), digital subtraction angiography (DSA), intracranial internal carotid artery (ICA), magnetic resonance imaging (MRI), middle cerebral artery (MCA). modified Rankin Scale (mRS), ruptured intracranial aneurysms (RIA), unruptured intracranial aneurysms (UIA), three-dimensional computed tomography angiography (3D CTA), World Federation of Neurological Surgery(WFNS),
3 Park
ACCEPTED MANUSCRIPT ABSTRACT
OBJECTIVE: The transciliary keyhole approach has been actively employed for unruptured intracranial aneurysms in many institutions, although applying this technique to ruptured
RI PT
aneurysms remains controversial. We investigated the risk factors related to poor surgical outcomes in ruptured aneurysm and attempted to clarify the differences between conventional craniotomy and keyhole surgery.
SC
METHODS: A total of 188 patients who underwent keyhole surgery due to ruptured anterior circulation aneurysms were enrolled at a single institution. A retrospective review of medical
M AN U
records and images from surgeries completed between July 2007 and February 2015 was performed.
RESULTS: This study included 116 (62%) female and 72 (38%) male patients, and their age ranged from 23 to 86 years. Preoperative clinical grades were good in almost all patients
TE D
except for a few patients with poor clinical grades. The mean aneurysm size was 5.5 mm and the most common aneurysm location was the anterior communicating artery (n=82). Most patients (n=158, 91.5%) showed good clinical outcomes. Univariate analysis of risk factors
EP
associated with poor grade outcomes after 3 months showed the following results: Hunt-Hess grade (odds ratio (95% confidence interval)=13.50, P<.0001), WFNS scale (OR=7.69,
AC C
P<.0001), aneurysm size (OR=1.21, P=0.019) and vasospasm (OR=6.43, P=0.0003) were statistically significant, while Fisher grade, skin to skin time (operation time), re-bleeding and ventricle puncture were not statistically significant. CONCLUSIONS: Because the poor surgical outcome of keyhole surgery is not different from known conventional craniotomy, we suggest that this approach is an acceptable treatment option in a good grade ruptured anterior circulation aneurysm.
4 Park
ACCEPTED MANUSCRIPT INTRODUCTION
As minimally invasive medical procedures grow in popularity, recent trends in intracranial aneurysm treatment have also changed. During the past two decades, there has been a
RI PT
significant evolution in minimally invasive keyhole aneurysm surgery.1-7 The trans-eyebrow approach is different from a standard craniotomy in terms of the skin incision site and craniotomy size. Trans-eyebrow keyhole surgery has certain potential advantages such as
SC
minimal brain exposure, shorter operation time, faster recovery, lower occurrence of wound problems, shorter hospital stay and excellent cosmetic outcome. This approach has been
M AN U
actively employed for clipping of unruptured intracranial aneurysms (UIA) in many institutions, although application of this approach for ruptured intracranial aneurysms (RIA) remains controversial. Surgeons worry about the narrow surgical view and difficulty in manipulation due to brain swelling in ruptured cases. Another concern is the management of
TE D
an unexpected rupture before proximal control. Due to these issues, almost all neurosurgeons consider a standard craniotomy rather than keyhole surgery as the first choice in ruptured aneurysmal conditions.8
EP
However, the trans-eyebrow keyhole approach has benefits due to its high cost-
AC C
effectiveness and patient satisfaction.7-9
CLINICAL MATERIALS AND METHODS After a retrospective review of medical records and images obtained between July 2007 and February 2015, a total of 188 patients who underwent keyhole surgery due to ruptured anterior circulation aneurysms were enrolled at a single institution. Preoperatively, we evaluated the clinical status of the patients and conducted a surgical view and clipping simulation with three-dimensional computed tomography angiography (3D CTA) at the CT
5 Park
ACCEPTED MANUSCRIPT
workstation, and then selected candidates for the transciliary supraorbital keyhole approach with the following selection criteria: 1) Good clinical grade (Hunt-Hess grade I, II, III), 2) Poor grade without brain swelling, 3) Poor grade due to hydrocephalus.
RI PT
After surgery, the patient was asked to make an outpatient visit on the scheduled date, and neurosurgeons who work at outpatient clinic recorded clinical outcomes. Postoperative clinical outcomes were evaluated on the day of discharge, after 3 months and after 6 months
SC
by the modified Rankin Scale (mRS). Brain CTA, magnetic resonance imaging (MRI) and digital subtraction angiography (DSA) were performed postoperatively for identification of
M AN U
complete clipping and brain injury. We investigated the related risk factors of poor outcome (mRS 3-6) after 3 months, such as age, gender, Hunt-Hess grade, Fisher grade, World Federation of Neurological Surgery (WFNS) grade, unexpected rupture, aneurysm size, operation time, location of aneurysm, ventricular puncture during operation, intraoperative
Statistical analyses
TE D
re-bleeding and vasospasm.
EP
The data were analyzed using SAS 9.4 for Windows (SAS Institute, Cary, NC, USA). Quantitative variables were presented as the median (interquartile range, IQR) in cases of
AC C
non-normal distributions or as the mean ± standard deviation for normal distributions. Categorical variables were presented as absolute numbers and percentages. Group comparisons were analyzed with the Student’s t-test or Wilcoxon rank-sum test for numeric variables and the χ2 or Fisher’s exact test for categorical variables. The variables found to be statistically significant in the univariate logistic regression analysis were used to identify the risk factors for poor clinical grade with odds ratios (ORs) and 95% confidence intervals. A P value less than 0.05 indicated statistical significance.
6 Park
ACCEPTED MANUSCRIPT Surgical technique Positioning
All patients underwent operations under general anesthesia. The head was placed on a horse
RI PT
shoe head rest with a slightly vertex down posture and could be easily rotated during the operation if necessary to provide an adequate surgical view. A skull fixing pin head-holder was not used. The degree of head rotation varied depending upon aneurysm location or
SC
craniotomy area but generally was between 20 and 60 degrees from the center.
M AN U
Skin incision
Eyebrow shaving was not performed for the purpose of infection control since it could cause skin injury. Prior to the operation, only alcohol was used to disinfect the skin. Povidone was not used because of the risk of pigmentation. Before covering with the drape, a mark was
TE D
placed in the middle of the eyebrows for a natural incision line. The skin incision varied according to the length of the eyebrows. For patients with thick and long eyebrows, a linear skin incision was made to the outside from the point two thirds outside of the eyebrow or
EP
supraorbital notch in order to prevent supraorbital nerve damage. For patients with short eyebrows, an incision was made in the whole eyebrow with resection of the orbicularis oculi
AC C
muscle below the skin incision line. The supraorbital nerve was then located, exfoliation was performed around it and the nerve was retracted towards the medial side. This process can prevent nerve damage while securing a sufficient cranial window for the operational view. In order to prevent damage to the frontal branch of the facial nerve, the skin incision should not pass the lateral margin of the eyebrow. Sub-periosteal dissection along the superior temporal line was conducted for making a burr hole under the temporalis muscle to protect the frontal branch of the facial nerve.
7 Park
ACCEPTED MANUSCRIPT Craniotomy and bone work
A small burr hole was made on the exposed superior temporal line. A 2x3 cm size free bone flap was made as close to the anterior skull base as possible while preserving the superior
RI PT
orbital rim (Figure 1). To minimize skull defects due to sawing, the smallest diameter electrical saw for pediatrics was used during the craniotomy. If a patient has a large expanded frontal sinus up to the lateral side, his or her posterior wall may be open. In such a case, the
SC
mucous membrane of the frontal sinus should be preserved. The small bony defect of the ○ (Stryker, frontal sinus may be repaired with a polyethylene reconstruction implant, MedporR
M AN U
○ (B. Braun, Kalamazoo, MI, USA) and secured by the topical skin adhesive, HistoacrylR Aesculap, Germany) to prevent infection or CSF leakage. After craniotomy and prior to opening the dura mater, the inner edge of the skull and protruded part of the orbital roof should be drilled out to provide a wider view under the microscope. The outer edge of the
TE D
skull should be preserved for minimizing skull defects. This process is indispensable to minimize brain retraction (Figure 2).
EP
Microsurgical work
The dura mater was incised in a semicircular shape and reflected to the outside. With gentle
AC C
retraction of the orbitofrontal gyrus, a gradual cerebrospinal fluid (CSF) drain with a suction device provided a direct view over the arachnoid membrane of the chiasmatic and carotid cisterns. Opening these cisterns can then accelerate rapid evacuation of cerebrospinal fluid by suction to expose the intracranial internal carotid artery (ICA) for proximal control. Sometimes it is difficult to drain sufficient CSF in cases of a ruptured aneurysm because the subarachnoid space is filled with a blood clot instead of CSF. In this situation, a ventricle puncture and CSF drain are needed for brain relaxation. A ventriculostomy catheter was introduced to the frontal horn of the lateral ventricle in the same operation field from the
8 Park
ACCEPTED MANUSCRIPT
frontal cortex. To avoid a rapid drop in intracranial pressure, CSF drainage should be performed slowly and gently. Aspiration of 40 to 60 ml of CSF provided an adequate view around the carotid cistern without frontal lobe retraction. With gradual dissection between the
RI PT
frontal lobe and optic nerve, the laminar terminalis behind the optic chiasm can be exposed. The opening of the laminar terminalis was performed to increase the CSF drain and further enhance CSF circulation. The next step was the same as conventional aneurysm surgery,
SC
which was basically the subfrontal approach requiring frontal lobe retraction. We applied this keyhole surgery for anterior circulation aneurysms of the anterior communicating artery,
M AN U
proximal anterior cerebral artery (ACA), supraclinoid segments of the ICA and M1 segments and bifurcation of the middle cerebral artery (MCA). In other types of lesions, we used other treatment options such as standard craniotomy or coil embolization.
TE D
Closure
Dura closure was conducted with an interrupted silk tie. Dura tear defects which were made during the craniotomy due to periosteal adhesion were repaired with a Fibrin Sealant Patch,
EP
TachoSil® (Baxter Health Care Corporation, Westlake Village, California, USA) . A small free bone flap was fitted and fixed with a miniplate-screw. Indwelling of an epidural drainage
AC C
catheter was not needed. Orbicularis oculi and levator palpebrae muscles were repaired by primary closure. Meticulous skin closure was performed with absorbable subcutaneous sutures and Steri-Strip™ adhesive skin closures (3M, Two Harbors, Minnesota, USA).
RESULTS
A total of 188 patients with ruptured anterior circulation aneurysms who underwent the supraorbital keyhole approach were enrolled in this study. There were 116 (62%) females and
9 Park
ACCEPTED MANUSCRIPT
72 (38%) males, and patient age ranged from 23 to 86 years. According to the preoperative Fisher grades for brain CT, 8 patients (4.3%) were grade 1, 44 (23.4%) were grade 2, 115 (61.2%) were grade 3, and 21 (11.1%) were grade 4. The preoperative clinical grade was
RI PT
good in almost all patients except for some patients with poor grades. Thirteen patients (6.9%) were Hunt-Hess grade I, 130 patients (69.1%) were grade II, 29 patients (15.4%) were grade III, 13 patients (6.9%) were grade IV, and 3 patients (1.6%) were grade V. The mean
SC
aneurysm size was 5.5 mm, with a largest diameter of 20 mm and smallest of 1.4 mm. The most common aneurysm location was the anterior communicating artery (82 patients), while
M AN U
the second most common location was the MCA (51 patients), and the chief site among them was a MCA bifurcation (46 patients). There were 46 patients with an ICA aneurysm, 44 with a posterior communicating artery aneurysm, 3 with ICA bifurcation, 2 with an anterior choroidal artery aneurysm, and only 1 patient with an ophthalmic artery aneurysm (Table 1).
TE D
In seven patients, opening of the frontal sinus was required due to its prominent size. However, no patient suffered from CSF leakage postoperatively despite having a frontal sinus violation. Three patients suffered from meningitis, but completely recovered with antibiotic
EP
therapy. Overall infection rate was 1.6% (3 in 188, meningitis) but there was no revision due to post-operative infection. And we had no experience of frontalis palsy. Intraoperative
AC C
ventriculostomy and CSF drainage were performed in 62 (33%) patients and no patient had intracerebral hemorrhage related to ventricle tapping. Premature aneurysm rupture occurred in 17 patients during arachnoid dissection or parent vessel dissection, but uneventful clippings were done in all patients with the preparation of proximal control. Bone flap malposition which needed revision surgery occurred in 3 patients. After the operation, 23 patients began complaining of hyposmia and anosmia and the symptoms improved gradually, although 11 patients had permanent anosmia. Twenty-five cases of vasospasm were verified in post-operative transcranial Doppler (TCD), and 3 of these received chemical angioplasty.
10 Park
ACCEPTED MANUSCRIPT
Only 1 patient had an acute cerebral infarction in the territory of the anterior cerebral artery (Table 2). At the day of discharge, 53 of 188 patients (28.2%) were mRS grade 0, while 93 (49.5%)
RI PT
were grade 1. Patients who had disabilities were as follows: 14 were grade 2 (7.4%), 11 were grade 3 (5.9%), 9 were grade 4 (4.8%), and 3 were grade 5 (1.6%). Five patients expired, but 3 of them expired due to aggravation of combined systemic disease such as systemic lupus
SC
erythematosis, acute renal failure, and pneumonia. At 3 and 6 months after the operation, the number of mRS grade 0 patients increased to 105 patients (59.3%) and 121 patients (70.0%),
M AN U
respectively. At the 6 month follow-up, 158 patients were MRS 0-2 (91.5%). Most patients showed good clinical outcomes with transciliary supraorbital keyhole surgery (Table 3, Figure3). Apart from 5 patients who died, there was additional data loss for 6 patients who were not available for an outcome checkup at 6 months follow-up because of their location or
TE D
as a personal choice; these 6 patients who live far from our institute had bled near our institute and visited emergency room. They moved to another hospital near their home with good condition after surgery but we couldn’t follow them .For investigation of related risk
EP
factors for poor clinical outcomes (mRS 3-6), 177 patients were enrolled in univariate logistic regression analysis. As a result of our investigation of the risk factors influencing the
AC C
differences between poor outcomes and good outcomes, the characteristics of patients with poor grade outcomes (mRS 3-6) included the following: the effects of age, Hunt-Hess grade, WFNS scales, aneurysmal size and vasospasm. These factors were identified using the Chisquare test, Fisher's exact test or Student's t-test (P = 0.0001, P < .0001, P = 0.001, P = 0.0001 and P = 0.090, respectively). Other risk factors such as gender, Fisher grade, skin to skin time (operation time), premature aneurysm rupture during operation, location of aneurysm, and ventricle puncture during the operation did not show significant differences (P = 0.432, P = 0.083, P = 0.744, P = 0.421, P = 0.467 and P = 0.924) (Table 4). Univariate analysis of risk
11 Park
ACCEPTED MANUSCRIPT
factors associated with poor grade outcomes (mRS 3-6) after 3 months showed the following: Hunt-Hess grade (odds ratio (95% confidence interval) = 13.50, P < .0001), WFNS scale (OR = 7.69, P < .0001), aneurysmal size (OR = 1.21, P = 0.019) and vasospasm (OR = 6.43, P =
RI PT
0.0003) were statistically significant but Fisher grade, skin to skin time (operation time), rebleeding and ventricle puncture were not statistically significant (Table 5).
SC
DISCUSSION
M AN U
The main purpose of the keyhole approach is to minimize the size of a craniotomy for optimal intracranial exposure without compromising safety and effectiveness in surgical treatment. Many neurosurgeons have developed methods and techniques for the keyhole approach based on this principle for the last several decades. The first description of the
TE D
supraorbital approach was a pituitary tumor case by Frazier in 1913.10 In 1971, Wilson described using limited exposure to operate on various intracranial lesions with a 2-inch trephine to open the skull.11 A small fronto-lateral approach for aneurysms of the anterior
EP
circulation was described by Brock in 1978.12 This approach was in a similar location as the supraorbital approach, but it had a significantly larger diameter for the craniotomy window.
AC C
To minimize brain retraction and skin incisions, Jane JA described techniques for the supraorbital approach in 1982.13 Since the late 1980s, several groups began developing minimally invasive supraorbital approaches.1-7 In the report by Lindert and Perneczky et al., there were no complications in 139 supraorbital keyhole approaches performed for aneurysms of the anterior cerebral circulation.7 Development of surgical microscopes and endoscopes facilitated these attempts. The supraorbital approach is basically a subfrontal approach of the anterior skull base giving access to both supratentorial and basilar aneurysms.3,6,11 It is performed either through
12 Park
ACCEPTED MANUSCRIPT
an incision on the scalp behind the hairline superficial to the temporalis fascia or an eyebrow or transciliary incision, preserving the supraorbital nerve. After the report from Lindert and Perneczky et al.,7 a minicraniotomy with trans-eyebrow incision was called the “supraorbital
RI PT
approach”. This supraorbital approach is different from the lateral supraorbital approach with a scalp incision on the temporal area which was reported by Hernesniemi et al. in terms of its eyebrow incision, relatively small bone flap and more anterior location of approaching
SC
route.14 To avoid confusion of terminology in this paper, we defined trans-eyebrow minicraniotomies including the supraorbital sub-frontal approach as “transciliary keyhole surgery.”
M AN U
Based on the experience of many neurosurgeons, transciliary keyhole surgery has the following advantages: 1) a small size skin incision on the eyebrow for excellent cosmetic outcome, 2) less brain exposure, 3) shorter operation time and reduction in general anesthetic agents, 4) faster recovery, 5) preservation of the superficial temporal artery and frontal
TE D
branches of the facial nerve which have a chance of injury in the standard pterional approach, 6) lower occurrence of wound problems, and 7) the economic benefits of a shorter hospital stay.15,16
EP
In 2005, Mitchell reported 56 anterior circulation aneurysms that were clipped via a supraorbital eyebrow mini-craniotomy. Six patients had ruptured aneurysms and 40 patients
AC C
who were admitted for elective clipping had unruptured aneurysms. This study had a morbidity of 4% at 1 year.17 A comparative study between keyhole and standard craniotomy was performed by J. Paladino in 2005.8 The keyhole group consisted of 482 patients with 565 aneurysms, while 146 patients with 167 aneurysms underwent a standard craniotomy. The mortality rate was 0.83% (4 patients) in the keyhole group and 2.05% (3 patients) in the standard craniotomy group. The majority of researchers selected a target of operation that included both unruptured and ruptured aneurysms, or they simply selected elective clipping with unruptured cases. The above studies were mainly reports with only UIAs or reports with
13 Park
ACCEPTED MANUSCRIPT no discrimination between UIAs and RIAs.
In 2014, Gupta et al. reported long-term outcome in surviving patients after clipping of ruptured intracranial aneurysms without distinguish of surgical approach. 74(15%) of 494
RI PT
clip-assigned patients had a poor outcome (mRS 3-6) at 1 year. 18 In the Barrow ruptured aneurysm trial, 62(36.5%) of 170 patients after clipping had a poor outcome (mRS 3-6) at 6 month.19 In our study, only 15 (8.7%) of 173 had a poor outcome (mRS 3-6) with keyhole
SC
surgery at 6 month. But actually, our result is good compared with overall other outcomes because we selected good grade patient with this keyhole approach. Because of this selection
M AN U
bias we cannot say that keyhole approach is better than conventional approach. A preconception about transciliary keyhole surgery is that it is difficult to apply this approach for operation on ruptured aneurysms because of brain swelling, while another preconception is that if unexpected re-bleeding occurs before proximal control of the parent
TE D
vessel, the operator would not be able to properly manage the bleeding. As a result, keyhole surgery has been attempted mostly in patients with UIAs. Thereby, it is hard to find any reports on the surgical outcomes of transciliary keyhole surgery for ruptured aneurysms, so
EP
the feasibility of this type of keyhole surgery on RIAs has not been established. Here, we report good surgical results of ruptured anterior circulation aneurysms with transciliary
AC C
keyhole surgery. Our study showed that this approach was feasible with good grade patients even with ruptured aneurysms. The degree of brain swelling in good grade patients which can be estimated by the extent of cortical sulci effacement on preoperative brain CT was not higher compared with that of poor grade patients, so adequate surgical room could be obtained with arachnoid dissection and CSF drainage with keyhole surgery as with unruptured aneurysms. Some poor grade patients with anterior communicating artery aneurysm ruptures who do not show cortical sulci effacement can be candidates for keyhole surgery. Poor grade patients who gradually deteriorate due to hydrocephalus but with an
14 Park
ACCEPTED MANUSCRIPT
initial good grade can be managed with keyhole surgery by first attempting intraoperative ventriculostomy. Unexpected rupture of an aneurysm is a major risk in all aneurysmal surgeries. Lindert
RI PT
reported 4 (3%) of 139 patients had an unexpected rupture of an aneurysm during supraorbital keyhole surgery.7 The incidence of re-bleeding is related to large aneurysm size and amount of subarachnoid hemorrhage. The cohort in our keyhole surgery group had
SC
relatively small aneurysms (5.5 mm in size) and had a lower SAH grade compared with conventional surgery reports. We experienced only two cases of unexpected bleeding during
M AN U
craniotomy before proximal control of the parent artery. One case that deteriorated during the anesthetic procedure underwent a craniectomy instead of the planned keyhole surgery. Another case developed sudden brain bulging after opening the dura suggesting re-bleeding. This patient was managed with gentle compression of the craniotomy site with gauze to control bleeding by increased intracranial pressure. With the CSF drain via ventriculostomy,
TE D
partial frontal lobe resection provided adequate space for clipping.20,21 A pre-operative imaging study and accurate clinical assessment are important when
EP
deciding the appropriate surgical approach. Three-dimensional CT-angiography imaging is very helpful in determining the approaching route. The surgical corridor simulating an
AC C
operative craniotomy can be made by the operator in the CT work station with surface rendering images from the CT angiography (Figure 2). With this computational reconstruction technique, we can determine which surgical approach is more appropriate for an individual patient`s situation. Over the past few decades, there have been many reports on the clinical outcomes of RIAs and risk factors of poor outcomes. According to the majority of previous reports, admission clinical status (Hunt-Hess grade, WFNS), patient age and aneurysm size were associated with poor prognoses. These findings are similar to our results. The only major
15 Park
ACCEPTED MANUSCRIPT
difference from some reports is the relation between Fisher grade and poor outcome. Ogilvy et al. reported Fisher grade as a meaningful poor outcome predictor, but it did not show any meaningful results in our study.22 We conducted conventional pterional craniotomy or
RI PT
decompressive craniectomy instead of keyhole surgery in patients with intracerebral hemorrhage, which may have led to this difference. For the location of the aneurysm, there was no significant difference in outcomes in anterior circulation between previous reports and
SC
our keyhole cohorts. Overall, our surgery results showed good prognoses because we mainly selected operable patients for minimally invasive surgery.
M AN U
There have been many reports concerning intracranial aneurysmal size over the past few decades. The mean (median) maximum diameter of aneurysms varied in these previous studies: 8.2±4 mm in RIAs in the Cooperative Aneurysm Study from 1970 to 1977,23 11.4±1.2 mm in RIAs in the Johns Hopkins series,24 and 8.0 mm in RIAs in a recent study by
TE D
Weir.25 The mean maximum diameter of 5.5 mm in RIAs from our study was relatively smaller than the size in these other studies.
In all patients operated using the transciliary supraorbital keyhole approach, there was
EP
sufficient room to identify the rupture site, as well as for clip application. The area of exposure of the parasellar region through the smaller supraorbital keyhole approach was as
AC C
adequate as the larger pterional or orbitozygomatic skull base approach and was sufficient for arachnoid dissection and CSF drainage.26,27 Even in ruptured aneurysmal keyhole surgery, the surgeon obtained enough surgical space with the aid of a head up position, hyperventilation, hyperosmolar agents as well as arachnoid dissection if there was no severe brain swelling. An in situ CSF drain by ventricular tapping is helpful when the above procedures cannot provide sufficient brain relaxation. However, rapid CSF drainage may cause unexpected re-bleeding of an aneurysm, so drainage should be done slowly and gently. One third of our series underwent intraoperative ventriculostomy and drainage of CSF amounts of 40 – 60 ml was
16 Park
ACCEPTED MANUSCRIPT enough to obtain access to the carotid cistern.
Several patients had prominent frontal sinuses. The posterior wall of the frontal sinus can be injured during craniotomy which is conducted on the frontal bone in the case of a wide
RI PT
frontal sinus. This defect can be a source of post-operative CSF leakage and wound infection. To avoid frontal sinus wall injury during craniotomy, the lateral extent of the frontal sinus should be evaluated by a preoperative skull X-ray. If a patient had a prominent frontal sinus,
SC
we measured the lateral margin of the frontal sinus and marked it on the skin before incision to avoid opening the frontal sinus. However, 7 patients required opening of the frontal sinus
M AN U
due to its prominent size. We tried to keep the mucous membrane intact and seal the defect ○ and HistoacrylR ○ to prevent infection completely with an artificial sealant such as MedporR or cerebrospinal fluid leakage. Therefore, a wide frontal sinus is not an absolute contraindication of the transciliary keyhole approach. Frontal sinus wall defects can be
surgery.
TE D
secured easily by a surgical sealant because the defect was not wide in transciliary keyhole
Anosmia due to olfactory nerve injury is a possible postoperative complication. It usually
EP
occurs because of frontal lobe retraction. The anterior communicating artery is a well-known vulnerable lesion for postoperative olfactory nerve damage. Park et al. reported on the
AC C
incidence of olfactory dysfunction after pterional approaches for ruptured cerebral aneurysms. Four patients (4%) 18 of 104 showed olfactory dysfunction with an ipsilateral pterional approach for ruptured aneurysms of anterior circulation, except for the anterior communicating artery. The ipsilateral pterional approach for a ruptured anterior communicating artery had a higher incidence of olfactory dysfunction, which developed in 10 of 70 patients (14%).28 Eriksen et al. reported a high incidence (89%) of anosmia after pterional approaches for clipping ruptured aneurysms.29 A 15% incidence of olfactory dysfunction, including hyposmia and anosmia, was reported by Aydin et al.30 A 39%
17 Park
ACCEPTED MANUSCRIPT
incidence of anosmia after aneurismal SAH was reported by Wermer et al.31 In our case, 23 patients of 188 (12%) presented with postoperative olfactory dysfunction. Since there were no previous studies on olfactory nerve injury after supraorbital keyhole surgery, our report
RI PT
has great value. Because of massive frontal lobe retraction, the transciliary keyhole approach has a considerable risk of olfactory nerve injury. In our study, there were a large number of patients with olfactory dysfunction similar to the incidence in the previous above-mentioned
SC
studies. However, most of the patients with olfactory dysfunction were in the early series of the study. Once we started to use the sub-olfactory approach with olfactory nerve dissection
cases with olfactory dysfunction.
M AN U
from the frontal lobe in transciliary keyhole surgery, it considerably decreased the number of
The frontal or frontotemporal branch of the facial nerve is especially vulnerable to injury during neurosurgical procedures such as cranioorbitozygomatic, pterional, and supraorbital
TE D
keyhole approaches that involve an incision and soft-tissue dissection in the frontal and temporal regions. Reisch and Perneczky previously reported a 5.5% incidence of permanent palsy of the frontalis muscle.15 But there was no frontalis palsy during our study. Performing
EP
sub-periosteal dissection without the temporalis muscle incision, and the lack of a lateral extension of 1cm or more from the superior temporal line made it possible to prevent damage
AC C
in the frontal branches of the facial nerves in the superficial fascial layer. Bony malposition happens during the rigid fixation of free bone flap not because of screw loosening. The skin incision is so small that the bone flap margin is difficult to accurately identify and there is a risk of incorrect fixation So surgeons should bear in mind that free bone flap is replaced in correction position.
CONCLUSION
18 Park
ACCEPTED MANUSCRIPT
Thus far, a standard craniotomy is typically selected as the surgical treatment of choice for ruptured intracranial aneurysm surgery. However, we demonstrated a favorable outcome with
RI PT
transciliary keyhole surgery in ruptured anterior circulation aneurysms. Our results showed that initial Hunt-Hess grade, WFNS scale, aneurysmal size and post-operative vasospasm were related risk factors of poor outcomes. Considering these related risk factors, surgical
SC
candidates with good clinical grades and operable patients from a preoperative imaging study can be selected. Because the minimally invasive transciliary supraorbital approach has many
M AN U
advantages, we suggest that this approach is an acceptable treatment option for selected case of ruptured anterior circulation aneurysm.
Limitation
TE D
The validity of the logistic model becomes problematic when the ratio of the numbers of events per variable analyzed becomes small. The parameter estimates may be biased and the usual tests of significance may not be valid. Because the number of poor outcome events was
EP
small (poor outcome n=23/total n=177), we could not use multivariate logistic regression analysis and only used univariate logistic regression analysis, which is the limitation of this
AC C
study. This is a retrospective cohort study so there should be selection bias. We selected good grade patients for surgery and it may affect to good surgical outcome.
19 Park
ACCEPTED MANUSCRIPT
Acknowledgement Statistical consultation was supported by a grant of the Korea Health Technology R&D
RI PT
Project through the Korea Health Industry Development Institute (KHIDI), funded by the
Author Disclosures
SC
Ministry of Health & Welfare, Republic of Korea (grant number: HI14C1062).
The authors report no conflict of interest concerning the materials or methods used in this
M AN U
study or the findings specified in this paper.
Conflicts of Interest and Ethical Adherence
AC C
EP
TE D
The authors have no conflicts of interest and adhered to ethical standards in this study.
20 Park
ACCEPTED MANUSCRIPT REFERENCES 1.
Delashaw JB, Jr., Jane JA, Kassell NF, Luce C. Supraorbital craniotomy by fracture of the anterior orbital roof. Technical note. J Neurosurg. 1993;79:615-618. Goel A. Supraorbital craniotomy. J Neurosurg. 1994;81:642-643.
3.
Jho HD. Orbital roof craniotomy via an eyebrow incision: a simplified anterior skull
RI PT
2.
base approach. Minim Invasive Neurosurg. 1997;40:91-97.
Ko Y, Yi HJ, Kim YS, Oh SH, Kim KM, Oh SJ. Eyebrow incision using tattoo for
SC
4.
20. 5.
Ramos-Zuniga R. The trans-supraorbital approach. Minim Invasive Neurosurg. 1999;42:133-136.
6.
M AN U
anterior fossa lesions: technical case reports. Minim Invasive Neurosurg. 2001;44:17-
Steiger HJ, Schmid-Elsaesser R, Stummer W, Uhl E. Transorbital keyhole approach to communicating
artery
aneurysms.
Neurosurgery.
2001;48:347-351;
TE D
anterior
discussion 351-342. 7.
van Lindert E, Perneczky A, Fries G, Pierangeli E. The supraorbital keyhole approach
EP
to supratentorial aneurysms: concept and technique. Surg Neurol. 1998;49:481-489; discussion 489-490.
Paladino J, Mrak G, Miklic P, Jednacak H, Mihaljevic D. The keyhole concept in
AC C
8.
aneurysm surgery--a comparative study: keyhole versus standard craniotomy. Minim
Invasive Neurosurg. 2005;48:251-258.
9.
Czirjak S, Nyary I, Futo J, Szeifert GT. Bilateral supraorbital keyhole approach for multiple aneurysms via superciliary skin incisions. Surg Neurol. 2002;57:314-323; discussion 323-314.
10.
Frazier CH. I. An Approach to the Hypophysis through the Anterior Cranial Fossa.
21 Park
ACCEPTED MANUSCRIPT Ann Surg. 1913;57:145-150. 11.
Wilson DH. Limited exposure in cerebral surgery. Technical note. J Neurosurg. 1971;34:102-106. Brock M, Dietz H. The small frontolateral approach for the microsurgical treatment of
RI PT
12.
intracranial aneurysms. Neurochirurgia (Stuttg). 1978;21:185-191. 13.
Jane JA, Park TS, Pobereskin LH, Winn HR, Butler AB. The supraorbital approach:
14.
SC
technical note. Neurosurgery. 1982;11:537-542.
Hernesniemi J, Ishii K, Niemela M, Smrcka M, Kivipelto L, Fujiki M, et al. Lateral
M AN U
supraorbital approach as an alternative to the classical pterional approach. Acta Neurochir Suppl. 2005;94:17-21. 15.
Reisch R, Perneczky A, Filippi R. Surgical technique of the supraorbital key-hole craniotomy. Surg Neurol. 2003;59:223-227.
Yasargil MG, Reichman MV, Kubik S. Preservation of the frontotemporal branch of
TE D
16.
the facial nerve using the interfascial temporalis flap for pterional craniotomy. Technical article. J Neurosurg. 1987;67:463-466. Mitchell P, Vindlacheruvu RR, Mahmood K, Ashpole RD, Grivas A, Mendelow AD.
EP
17.
Supraorbital eyebrow minicraniotomy for anterior circulation aneurysms. Surg Neurol.
18.
AC C
2005;63:47-51; discussion 51. Gupta SK, Chhabra R, Mohindra S, Sharma A, Mathuriya SN, Pathak A, et al. Long-
Term Outcome in Surviving Patients After Clipping of Intracranial Aneurysms. World Neurosurg. 2014;81:316-321.
19.
Spetzler RF, McDougall CG, Zabramski JM, Albuquerque FC, Hills NK, Russin JJ et al. The Barrow Ruptured Aneurysm Trial: 6-year results. J Neurosurg. 2015;123:609617.
20.
Solanki C, Pandey P, Rao KV. Predictors of aneurysmal rebleed before definitive
22 Park
ACCEPTED MANUSCRIPT
surgical or endovascular management. Acta Neurochir (Wien). 2016;158:1037-1044. 21.
van Donkelaar CE, Bakker NA, Veeger NJ, Uyttenboogaart M, Metzemaekers JD, Luijckx GJ, et al. Predictive Factors for Rebleeding After Aneurysmal Subarachnoid
RI PT
Hemorrhage: Rebleeding Aneurysmal Subarachnoid Hemorrhage Study. Stroke. 2015;46:2100-2106. 22.
Ogilvy CS, Carter BS. A proposed comprehensive grading system to predict outcome
SC
for surgical management of intracranial aneurysms. Neurosurgery. 1998;42:959-968; discussion 968-970.
Weir B. Unruptured intracranial aneurysms: a review. J Neurosurg. 2002;96:3-42.
24.
Sjoberg RJ. Risk factors for the development and rupture of intracranial berry
M AN U
23.
aneurysms. Am J Med. 1986;80:A65. 25.
Weir B, Amidei C, Kongable G, Findlay JM, Kassell NF, Kelly J, et al. The aspect
TE D
ratio (dome/neck) of ruptured and unruptured aneurysms. J Neurosurg. 2003;99:447451. 26.
Cheng CM, Noguchi A, Dogan A, Anderson GJ, Hsu FP, McMenomey SO, et al.
EP
Quantitative verification of the keyhole concept: a comparison of area of exposure in the parasellar region via supraorbital keyhole, frontotemporal pterional, and
27.
AC C
supraorbital approaches. J Neurosurg. 2013;118:264-269. Yagmurlu K, Safavi-Abbasi S, Belykh E, Kalani MY, Nakaji P, Rhoton AL, Jr., et al.
Quantitative anatomical analysis and clinical experience with mini-pterional and mini-orbitozygomatic approaches for intracranial aneurysm surgery. J Neurosurg.
2016:1-14. 28.
Park J, Lee SH, Kang DH, Kim JS. Olfactory dysfunction after ipsilateral and contralateral
pterional
approaches
2009;65:727-732; discussion 732.
for
cerebral
aneurysms.
Neurosurgery.
23 Park
ACCEPTED MANUSCRIPT 29.
Eriksen KD, Boge-Rasmussen T, Kruse-Larsen C. Anosmia following operation for cerebral aneurysms in the anterior circulation. J Neurosurg. 1990;72:864-865.
30.
Aydin IH, Kadioglu HH, Tuzun Y, Kayaoglu CR, Takci E, Ozturk M. Postoperative
approach. Minim Invasive Neurosurg. 1996;39:71-73.
Wermer MJ, Donswijk M, Greebe P, Verweij BH, Rinkel GJ. Anosmia after
SC
aneurysmal subarachnoid hemorrhage. Neurosurgery. 2007;61:918-922; discussion
EP
TE D
M AN U
922-913.
AC C
31.
RI PT
anosmia after anterior communicating artery aneurysms surgery by the pterional
24 Park
ACCEPTED MANUSCRIPT
Figure 1. Post-operative skull 3D CT image: A 2x3 cm size free bone flap was made close to the anterior skull base with preservation of the superior orbital rim.
RI PT
Figure 2. Pre-operative brain 3D CT angiography is a helpful tool for surgical stimulation. The surgical corridor simulating operative craniotomy can be made by the operator in the CT
SC
work station with surface rendering images from the CT angiography.
AC C
EP
TE D
M AN U
Figure 3. Modified Rankin Scale of post-operative patients.
ACCEPTED MANUSCRIPT TABLE 1. Summary of cases with transciliary aneurysm neck clipping in ruptured anterior circulation aneurysm 188
Male/Female
72/116
Age
23-86 years; mean 58.5
Preoperative Fisher grade on CT
1: 8 (4.3%) 2: 44 (23.4%)
4: 20 (10.6%)
Preoperative Hunt-Hess grade
M AN U
5: 1 (0.5%)
SC
3: 115 (61.2%)
RI PT
Total cases
I: 13 (6.9%)
II: 130 (69.1%) III: 29 (15.4%) IV: 13 (6.9%)
Mean aneurysm size
5.6 mm Middle cerebral artery
AC C
EP
Location
TE D
V: 3 (1.6%)
51
MCA bifurcation
Anterior communicating artery
46 82
Anterior cerebral artery
5
Internal cervical artery
46
ICA bifurcation Posterior communicating artery Ophthalmic artery Anterior choroidal artery
3 44 1 2
ACCEPTED MANUSCRIPT TABLE 2. Adverse events during the follow-up period after transciliary keyhole surgery Complications
Acute/Delayed Acute
7
Hyposmia and anosmia
Acute
9
Unexpected rupture of aneurysm during surgery
Acute
2
Delayed
25
Vasospasm Vasospasm with chemical angioplasty
RI PT
Frontal sinus opening
19
Delayed
3
SC
Wound and bone flap complications bone flap extrusion
Meningitis Hygroma Visual disturbance
AC C
EP
TE D
Cerebral infarction (ACA infarction)
M AN U
wound swelling
1 2
Delayed
3
Delayed
3
Acute
3
Acute
1
ACCEPTED MANUSCRIPT TABLE 3. Modified Rankin Scale of post-operative patients At discharge
3 months
6 months
0 - No symptoms
53 (28.2%)
105 (59.3%)
121 (70.0%)
1 - No significant disability
93 (49.5%)
44 (24.9%)
31 (18.0%)
2 - Slight disability
14 (7.4%)
10 (5.6%)
6 (3.5%)
3 - Moderate disability
11 (5.9%)
8 (4.5%)
9 (5.2%)
4 - Moderately severe disability
9 (4.8%)
7 (4.0%)
4 (2.3%)
5 - Severe disability
3 (1.6%)
2 (1.1%)
2 (2.7%)
6 – Death
5 (2.7%)
SC
(SLE 1, pneumonia 1, ARF 1)
1 (0.6%)
188/0
AC C
EP
TE D
Total/follow-up loss
0 (0%)
(ARF 1)
M AN U
(combined systemic disease)
RI PT
mRS
177/6
173/2
ACCEPTED MANUSCRIPT TABLE 4. Characteristics of patients with poor grade outcomes (mRS 3-6) mRS Scale Total good grade
poor grade
(n=154)
(n=23)
P value
Age mean±sd
51.7±12.3
49.7±10.5
median (min-max)
50 (23-86)
49 (22-77)
64.7±15.2
0.0001
67 (37-86)
0.432
67 (37.8)
60 (39.0)
Female
110 (62.2)
94 (61.0)
Hunt-Hess grade grade Ⅰ, Ⅱ, Ⅲ
161 (91.0)
grade Ⅳ, Ⅴ
16 (9.0)
grade Ⅰ, Ⅱ, Ⅲ grade Ⅳ, Ⅴ
16 (69.6) <.0001
147 (95.5)
14 (60.9)
7 (4.5)
9 (39.1) 0.083
50 (28.2)
47 (30.5)
3 (13.0)
127 (71.8)
107 (69.5)
20 (87.0)
EP
WFNS scale
TE D
Fisher Grade
7 (30.4)
M AN U
Male
SC
Gender
RI PT
(n=177)
0.0001
153 (86.4)
140 (90.9)
13 (56.5)
grade Ⅳ, Ⅴ
24 (13.6)
14 (9.1)
10 (43.5)
6.0±2.4
5.8±2.1
7.1±3.6
5.5 (1.0-20.0)
5.0 (1.0-15.7)
6.0 (2.0-20.0)
151.5±51.0
151.0±51.6
154.8±47.7
140 (60-400)
139 (60-400)
145 (80-240)
AC C
grade Ⅰ, Ⅱ, Ⅲ
Aneurysm size mean±sd
median (min-max)
0.090
Skin to skin time mean±sd median (min-max) Location of aneurysm
0.744
0.421
ACCEPTED MANUSCRIPT ICA
46 (26.0)
41 (26.6)
5 (21.8)
ACA
82 (46.3)
73 (47.4)
9 (39.1)
MCA
49 (27.7)
40 (26.0)
9 (39.1)
Re-bleeding
0.467 160 (90.4)
140 (90.9)
Yes
17 (9.6)
14 (9.1)
20 (87.0)
RI PT
No
3 (13.0)
Ventricle puncture
0.924
117 (66.1)
102 (66.2)
Yes
60 (33.9)
52 (33.8)
No
154 (87.0)
Yes
23 (13.0)
M AN U
Vasospasm
15 (65.2)
SC
No
8 (34.8)
140 (90.9)
14 (60.9)
14 (9.1)
9 (39.1)
Values are numbers (percentages) or mean±sd.
AC C
EP
TE D
P values were calculated using the Chi-square test, Fisher's exact test or Student's t-test.
0.001
ACCEPTED MANUSCRIPT TABLE 5. Univariate analysis of risk factors associated with poor grade outcomes (mRS 3-6). OR (95% CI)
P value
Hunt-Hess grade ,
grade
,
,
Reference
13.50 (4.36-41.78)
Fisher Grade
grade
,
,
Reference
SC
,
2.93 (0.83-10.33)
M AN U
grade
WFNS scale ,
grade
,
,
Aneurysm size Skin to skin time Re-bleeding
EP
No Yes
<.0001
0.10
Reference
TE D
grade
RI PT
grade
7.69 (2.86-20.72)
<.0001
1.21 (1.03-1.42)
0.019
1.00 (0.99-1.01)
0.743
Reference 1.50 (0.40-5.68)
0.551
No Yes
AC C
Ventricle puncture
Reference 1.05 (0.42-2.63)
0.924
Vasospasm No
Reference
Yes
6.43 (2.36-17.50)
OR: Odds Ratio, CI: Confidence Interval
0.0003
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Trans-eyebrow keyhole aneurysmal neck clipping is one of minimal invasive cranial surgery.
This procedure has the advantage of less damage to normal tissues and a shorter recovery period. The application of this technique to ruptured aneurysms is still controversial.
The authors experienced more than 200 cases of keyhole surgery on ruptured aneurysms and
RI PT
obtained good surgical outcome.
We analyzed for poor outcome factors and wanted to provide selection indication for keyhole
AC C
EP
TE D
M AN U
SC
surgery on ruptured aneurysms.
ACCEPTED MANUSCRIPT Disclosure-Conflicts documentation
Title : Risk factor analysis for poor outcomes in supraorbital keyhole aneurysm clipping
RI PT
for ruptured anterior circulation aneurysm
SC
Disclosure of Funding : The authors have no financial support or interest.
M AN U
Conflicts of Interest and Ethical Adherence : The authors have no conflict of interest or ethical adherence
AC C
EP
TE D
Statement of Authorship : Ik Seong Park has a corresponding authorship
S1878-8750(17)32187-3
DOI:
10.1016/j.wneu.2017.12.071
Reference:
WNEU 7084
To appear in:
World Neurosurgery
Received Date: 10 September 2017 Revised Date:
10 December 2017
Accepted Date: 11 December 2017
Please cite this article as: Park JS, Kim H, Baik MW, Park IS, Risk factor analysis for poor outcomes in supraorbital keyhole aneurysm clipping for ruptured anterior circulation aneurysm, World Neurosurgery (2018), doi: 10.1016/j.wneu.2017.12.071. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1 Park
ACCEPTED MANUSCRIPT
Risk factor analysis for poor outcomes in supraorbital keyhole aneurysm clipping for ruptured anterior circulation aneurysm
RI PT
Jun Sang Park, MD, Hoon Kim, MD, Min Woo Baik, MD, Ik Seong Park, MD, PhD,MBA
SC
Department of Neurosurgery, Bucheon St. Mary’s Hospital, College of Medicine, The
M AN U
Catholic University of Korea, Bucheon, Republic of Korea.
Correspondence author: Ik Seong Park, MD
Department of Neurosurgery, Bucheon St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, 327, Sosa-ro, Bucheon-si, Gyeonggi-do, 14647, Republic of
Tel: +82-32-340-7212 Fax: +82-32-340-7391
TE D
Korea
AC C
EP
E-mail: [email protected]
2 Park
ACCEPTED MANUSCRIPT Key words: Supraorbital keyhole surgery, Aneurysm clipping, Subarachnoid hemorrhage
RI PT
Abbreviations and Acronmyms
AC C
EP
TE D
M AN U
SC
anterior cerebral artery (ACA), cerebrospinal fluid (CSF), digital subtraction angiography (DSA), intracranial internal carotid artery (ICA), magnetic resonance imaging (MRI), middle cerebral artery (MCA). modified Rankin Scale (mRS), ruptured intracranial aneurysms (RIA), unruptured intracranial aneurysms (UIA), three-dimensional computed tomography angiography (3D CTA), World Federation of Neurological Surgery(WFNS),
3 Park
ACCEPTED MANUSCRIPT ABSTRACT
OBJECTIVE: The transciliary keyhole approach has been actively employed for unruptured intracranial aneurysms in many institutions, although applying this technique to ruptured
RI PT
aneurysms remains controversial. We investigated the risk factors related to poor surgical outcomes in ruptured aneurysm and attempted to clarify the differences between conventional craniotomy and keyhole surgery.
SC
METHODS: A total of 188 patients who underwent keyhole surgery due to ruptured anterior circulation aneurysms were enrolled at a single institution. A retrospective review of medical
M AN U
records and images from surgeries completed between July 2007 and February 2015 was performed.
RESULTS: This study included 116 (62%) female and 72 (38%) male patients, and their age ranged from 23 to 86 years. Preoperative clinical grades were good in almost all patients
TE D
except for a few patients with poor clinical grades. The mean aneurysm size was 5.5 mm and the most common aneurysm location was the anterior communicating artery (n=82). Most patients (n=158, 91.5%) showed good clinical outcomes. Univariate analysis of risk factors
EP
associated with poor grade outcomes after 3 months showed the following results: Hunt-Hess grade (odds ratio (95% confidence interval)=13.50, P<.0001), WFNS scale (OR=7.69,
AC C
P<.0001), aneurysm size (OR=1.21, P=0.019) and vasospasm (OR=6.43, P=0.0003) were statistically significant, while Fisher grade, skin to skin time (operation time), re-bleeding and ventricle puncture were not statistically significant. CONCLUSIONS: Because the poor surgical outcome of keyhole surgery is not different from known conventional craniotomy, we suggest that this approach is an acceptable treatment option in a good grade ruptured anterior circulation aneurysm.
4 Park
ACCEPTED MANUSCRIPT INTRODUCTION
As minimally invasive medical procedures grow in popularity, recent trends in intracranial aneurysm treatment have also changed. During the past two decades, there has been a
RI PT
significant evolution in minimally invasive keyhole aneurysm surgery.1-7 The trans-eyebrow approach is different from a standard craniotomy in terms of the skin incision site and craniotomy size. Trans-eyebrow keyhole surgery has certain potential advantages such as
SC
minimal brain exposure, shorter operation time, faster recovery, lower occurrence of wound problems, shorter hospital stay and excellent cosmetic outcome. This approach has been
M AN U
actively employed for clipping of unruptured intracranial aneurysms (UIA) in many institutions, although application of this approach for ruptured intracranial aneurysms (RIA) remains controversial. Surgeons worry about the narrow surgical view and difficulty in manipulation due to brain swelling in ruptured cases. Another concern is the management of
TE D
an unexpected rupture before proximal control. Due to these issues, almost all neurosurgeons consider a standard craniotomy rather than keyhole surgery as the first choice in ruptured aneurysmal conditions.8
EP
However, the trans-eyebrow keyhole approach has benefits due to its high cost-
AC C
effectiveness and patient satisfaction.7-9
CLINICAL MATERIALS AND METHODS After a retrospective review of medical records and images obtained between July 2007 and February 2015, a total of 188 patients who underwent keyhole surgery due to ruptured anterior circulation aneurysms were enrolled at a single institution. Preoperatively, we evaluated the clinical status of the patients and conducted a surgical view and clipping simulation with three-dimensional computed tomography angiography (3D CTA) at the CT
5 Park
ACCEPTED MANUSCRIPT
workstation, and then selected candidates for the transciliary supraorbital keyhole approach with the following selection criteria: 1) Good clinical grade (Hunt-Hess grade I, II, III), 2) Poor grade without brain swelling, 3) Poor grade due to hydrocephalus.
RI PT
After surgery, the patient was asked to make an outpatient visit on the scheduled date, and neurosurgeons who work at outpatient clinic recorded clinical outcomes. Postoperative clinical outcomes were evaluated on the day of discharge, after 3 months and after 6 months
SC
by the modified Rankin Scale (mRS). Brain CTA, magnetic resonance imaging (MRI) and digital subtraction angiography (DSA) were performed postoperatively for identification of
M AN U
complete clipping and brain injury. We investigated the related risk factors of poor outcome (mRS 3-6) after 3 months, such as age, gender, Hunt-Hess grade, Fisher grade, World Federation of Neurological Surgery (WFNS) grade, unexpected rupture, aneurysm size, operation time, location of aneurysm, ventricular puncture during operation, intraoperative
Statistical analyses
TE D
re-bleeding and vasospasm.
EP
The data were analyzed using SAS 9.4 for Windows (SAS Institute, Cary, NC, USA). Quantitative variables were presented as the median (interquartile range, IQR) in cases of
AC C
non-normal distributions or as the mean ± standard deviation for normal distributions. Categorical variables were presented as absolute numbers and percentages. Group comparisons were analyzed with the Student’s t-test or Wilcoxon rank-sum test for numeric variables and the χ2 or Fisher’s exact test for categorical variables. The variables found to be statistically significant in the univariate logistic regression analysis were used to identify the risk factors for poor clinical grade with odds ratios (ORs) and 95% confidence intervals. A P value less than 0.05 indicated statistical significance.
6 Park
ACCEPTED MANUSCRIPT Surgical technique Positioning
All patients underwent operations under general anesthesia. The head was placed on a horse
RI PT
shoe head rest with a slightly vertex down posture and could be easily rotated during the operation if necessary to provide an adequate surgical view. A skull fixing pin head-holder was not used. The degree of head rotation varied depending upon aneurysm location or
SC
craniotomy area but generally was between 20 and 60 degrees from the center.
M AN U
Skin incision
Eyebrow shaving was not performed for the purpose of infection control since it could cause skin injury. Prior to the operation, only alcohol was used to disinfect the skin. Povidone was not used because of the risk of pigmentation. Before covering with the drape, a mark was
TE D
placed in the middle of the eyebrows for a natural incision line. The skin incision varied according to the length of the eyebrows. For patients with thick and long eyebrows, a linear skin incision was made to the outside from the point two thirds outside of the eyebrow or
EP
supraorbital notch in order to prevent supraorbital nerve damage. For patients with short eyebrows, an incision was made in the whole eyebrow with resection of the orbicularis oculi
AC C
muscle below the skin incision line. The supraorbital nerve was then located, exfoliation was performed around it and the nerve was retracted towards the medial side. This process can prevent nerve damage while securing a sufficient cranial window for the operational view. In order to prevent damage to the frontal branch of the facial nerve, the skin incision should not pass the lateral margin of the eyebrow. Sub-periosteal dissection along the superior temporal line was conducted for making a burr hole under the temporalis muscle to protect the frontal branch of the facial nerve.
7 Park
ACCEPTED MANUSCRIPT Craniotomy and bone work
A small burr hole was made on the exposed superior temporal line. A 2x3 cm size free bone flap was made as close to the anterior skull base as possible while preserving the superior
RI PT
orbital rim (Figure 1). To minimize skull defects due to sawing, the smallest diameter electrical saw for pediatrics was used during the craniotomy. If a patient has a large expanded frontal sinus up to the lateral side, his or her posterior wall may be open. In such a case, the
SC
mucous membrane of the frontal sinus should be preserved. The small bony defect of the ○ (Stryker, frontal sinus may be repaired with a polyethylene reconstruction implant, MedporR
M AN U
○ (B. Braun, Kalamazoo, MI, USA) and secured by the topical skin adhesive, HistoacrylR Aesculap, Germany) to prevent infection or CSF leakage. After craniotomy and prior to opening the dura mater, the inner edge of the skull and protruded part of the orbital roof should be drilled out to provide a wider view under the microscope. The outer edge of the
TE D
skull should be preserved for minimizing skull defects. This process is indispensable to minimize brain retraction (Figure 2).
EP
Microsurgical work
The dura mater was incised in a semicircular shape and reflected to the outside. With gentle
AC C
retraction of the orbitofrontal gyrus, a gradual cerebrospinal fluid (CSF) drain with a suction device provided a direct view over the arachnoid membrane of the chiasmatic and carotid cisterns. Opening these cisterns can then accelerate rapid evacuation of cerebrospinal fluid by suction to expose the intracranial internal carotid artery (ICA) for proximal control. Sometimes it is difficult to drain sufficient CSF in cases of a ruptured aneurysm because the subarachnoid space is filled with a blood clot instead of CSF. In this situation, a ventricle puncture and CSF drain are needed for brain relaxation. A ventriculostomy catheter was introduced to the frontal horn of the lateral ventricle in the same operation field from the
8 Park
ACCEPTED MANUSCRIPT
frontal cortex. To avoid a rapid drop in intracranial pressure, CSF drainage should be performed slowly and gently. Aspiration of 40 to 60 ml of CSF provided an adequate view around the carotid cistern without frontal lobe retraction. With gradual dissection between the
RI PT
frontal lobe and optic nerve, the laminar terminalis behind the optic chiasm can be exposed. The opening of the laminar terminalis was performed to increase the CSF drain and further enhance CSF circulation. The next step was the same as conventional aneurysm surgery,
SC
which was basically the subfrontal approach requiring frontal lobe retraction. We applied this keyhole surgery for anterior circulation aneurysms of the anterior communicating artery,
M AN U
proximal anterior cerebral artery (ACA), supraclinoid segments of the ICA and M1 segments and bifurcation of the middle cerebral artery (MCA). In other types of lesions, we used other treatment options such as standard craniotomy or coil embolization.
TE D
Closure
Dura closure was conducted with an interrupted silk tie. Dura tear defects which were made during the craniotomy due to periosteal adhesion were repaired with a Fibrin Sealant Patch,
EP
TachoSil® (Baxter Health Care Corporation, Westlake Village, California, USA) . A small free bone flap was fitted and fixed with a miniplate-screw. Indwelling of an epidural drainage
AC C
catheter was not needed. Orbicularis oculi and levator palpebrae muscles were repaired by primary closure. Meticulous skin closure was performed with absorbable subcutaneous sutures and Steri-Strip™ adhesive skin closures (3M, Two Harbors, Minnesota, USA).
RESULTS
A total of 188 patients with ruptured anterior circulation aneurysms who underwent the supraorbital keyhole approach were enrolled in this study. There were 116 (62%) females and
9 Park
ACCEPTED MANUSCRIPT
72 (38%) males, and patient age ranged from 23 to 86 years. According to the preoperative Fisher grades for brain CT, 8 patients (4.3%) were grade 1, 44 (23.4%) were grade 2, 115 (61.2%) were grade 3, and 21 (11.1%) were grade 4. The preoperative clinical grade was
RI PT
good in almost all patients except for some patients with poor grades. Thirteen patients (6.9%) were Hunt-Hess grade I, 130 patients (69.1%) were grade II, 29 patients (15.4%) were grade III, 13 patients (6.9%) were grade IV, and 3 patients (1.6%) were grade V. The mean
SC
aneurysm size was 5.5 mm, with a largest diameter of 20 mm and smallest of 1.4 mm. The most common aneurysm location was the anterior communicating artery (82 patients), while
M AN U
the second most common location was the MCA (51 patients), and the chief site among them was a MCA bifurcation (46 patients). There were 46 patients with an ICA aneurysm, 44 with a posterior communicating artery aneurysm, 3 with ICA bifurcation, 2 with an anterior choroidal artery aneurysm, and only 1 patient with an ophthalmic artery aneurysm (Table 1).
TE D
In seven patients, opening of the frontal sinus was required due to its prominent size. However, no patient suffered from CSF leakage postoperatively despite having a frontal sinus violation. Three patients suffered from meningitis, but completely recovered with antibiotic
EP
therapy. Overall infection rate was 1.6% (3 in 188, meningitis) but there was no revision due to post-operative infection. And we had no experience of frontalis palsy. Intraoperative
AC C
ventriculostomy and CSF drainage were performed in 62 (33%) patients and no patient had intracerebral hemorrhage related to ventricle tapping. Premature aneurysm rupture occurred in 17 patients during arachnoid dissection or parent vessel dissection, but uneventful clippings were done in all patients with the preparation of proximal control. Bone flap malposition which needed revision surgery occurred in 3 patients. After the operation, 23 patients began complaining of hyposmia and anosmia and the symptoms improved gradually, although 11 patients had permanent anosmia. Twenty-five cases of vasospasm were verified in post-operative transcranial Doppler (TCD), and 3 of these received chemical angioplasty.
10 Park
ACCEPTED MANUSCRIPT
Only 1 patient had an acute cerebral infarction in the territory of the anterior cerebral artery (Table 2). At the day of discharge, 53 of 188 patients (28.2%) were mRS grade 0, while 93 (49.5%)
RI PT
were grade 1. Patients who had disabilities were as follows: 14 were grade 2 (7.4%), 11 were grade 3 (5.9%), 9 were grade 4 (4.8%), and 3 were grade 5 (1.6%). Five patients expired, but 3 of them expired due to aggravation of combined systemic disease such as systemic lupus
SC
erythematosis, acute renal failure, and pneumonia. At 3 and 6 months after the operation, the number of mRS grade 0 patients increased to 105 patients (59.3%) and 121 patients (70.0%),
M AN U
respectively. At the 6 month follow-up, 158 patients were MRS 0-2 (91.5%). Most patients showed good clinical outcomes with transciliary supraorbital keyhole surgery (Table 3, Figure3). Apart from 5 patients who died, there was additional data loss for 6 patients who were not available for an outcome checkup at 6 months follow-up because of their location or
TE D
as a personal choice; these 6 patients who live far from our institute had bled near our institute and visited emergency room. They moved to another hospital near their home with good condition after surgery but we couldn’t follow them .For investigation of related risk
EP
factors for poor clinical outcomes (mRS 3-6), 177 patients were enrolled in univariate logistic regression analysis. As a result of our investigation of the risk factors influencing the
AC C
differences between poor outcomes and good outcomes, the characteristics of patients with poor grade outcomes (mRS 3-6) included the following: the effects of age, Hunt-Hess grade, WFNS scales, aneurysmal size and vasospasm. These factors were identified using the Chisquare test, Fisher's exact test or Student's t-test (P = 0.0001, P < .0001, P = 0.001, P = 0.0001 and P = 0.090, respectively). Other risk factors such as gender, Fisher grade, skin to skin time (operation time), premature aneurysm rupture during operation, location of aneurysm, and ventricle puncture during the operation did not show significant differences (P = 0.432, P = 0.083, P = 0.744, P = 0.421, P = 0.467 and P = 0.924) (Table 4). Univariate analysis of risk
11 Park
ACCEPTED MANUSCRIPT
factors associated with poor grade outcomes (mRS 3-6) after 3 months showed the following: Hunt-Hess grade (odds ratio (95% confidence interval) = 13.50, P < .0001), WFNS scale (OR = 7.69, P < .0001), aneurysmal size (OR = 1.21, P = 0.019) and vasospasm (OR = 6.43, P =
RI PT
0.0003) were statistically significant but Fisher grade, skin to skin time (operation time), rebleeding and ventricle puncture were not statistically significant (Table 5).
SC
DISCUSSION
M AN U
The main purpose of the keyhole approach is to minimize the size of a craniotomy for optimal intracranial exposure without compromising safety and effectiveness in surgical treatment. Many neurosurgeons have developed methods and techniques for the keyhole approach based on this principle for the last several decades. The first description of the
TE D
supraorbital approach was a pituitary tumor case by Frazier in 1913.10 In 1971, Wilson described using limited exposure to operate on various intracranial lesions with a 2-inch trephine to open the skull.11 A small fronto-lateral approach for aneurysms of the anterior
EP
circulation was described by Brock in 1978.12 This approach was in a similar location as the supraorbital approach, but it had a significantly larger diameter for the craniotomy window.
AC C
To minimize brain retraction and skin incisions, Jane JA described techniques for the supraorbital approach in 1982.13 Since the late 1980s, several groups began developing minimally invasive supraorbital approaches.1-7 In the report by Lindert and Perneczky et al., there were no complications in 139 supraorbital keyhole approaches performed for aneurysms of the anterior cerebral circulation.7 Development of surgical microscopes and endoscopes facilitated these attempts. The supraorbital approach is basically a subfrontal approach of the anterior skull base giving access to both supratentorial and basilar aneurysms.3,6,11 It is performed either through
12 Park
ACCEPTED MANUSCRIPT
an incision on the scalp behind the hairline superficial to the temporalis fascia or an eyebrow or transciliary incision, preserving the supraorbital nerve. After the report from Lindert and Perneczky et al.,7 a minicraniotomy with trans-eyebrow incision was called the “supraorbital
RI PT
approach”. This supraorbital approach is different from the lateral supraorbital approach with a scalp incision on the temporal area which was reported by Hernesniemi et al. in terms of its eyebrow incision, relatively small bone flap and more anterior location of approaching
SC
route.14 To avoid confusion of terminology in this paper, we defined trans-eyebrow minicraniotomies including the supraorbital sub-frontal approach as “transciliary keyhole surgery.”
M AN U
Based on the experience of many neurosurgeons, transciliary keyhole surgery has the following advantages: 1) a small size skin incision on the eyebrow for excellent cosmetic outcome, 2) less brain exposure, 3) shorter operation time and reduction in general anesthetic agents, 4) faster recovery, 5) preservation of the superficial temporal artery and frontal
TE D
branches of the facial nerve which have a chance of injury in the standard pterional approach, 6) lower occurrence of wound problems, and 7) the economic benefits of a shorter hospital stay.15,16
EP
In 2005, Mitchell reported 56 anterior circulation aneurysms that were clipped via a supraorbital eyebrow mini-craniotomy. Six patients had ruptured aneurysms and 40 patients
AC C
who were admitted for elective clipping had unruptured aneurysms. This study had a morbidity of 4% at 1 year.17 A comparative study between keyhole and standard craniotomy was performed by J. Paladino in 2005.8 The keyhole group consisted of 482 patients with 565 aneurysms, while 146 patients with 167 aneurysms underwent a standard craniotomy. The mortality rate was 0.83% (4 patients) in the keyhole group and 2.05% (3 patients) in the standard craniotomy group. The majority of researchers selected a target of operation that included both unruptured and ruptured aneurysms, or they simply selected elective clipping with unruptured cases. The above studies were mainly reports with only UIAs or reports with
13 Park
ACCEPTED MANUSCRIPT no discrimination between UIAs and RIAs.
In 2014, Gupta et al. reported long-term outcome in surviving patients after clipping of ruptured intracranial aneurysms without distinguish of surgical approach. 74(15%) of 494
RI PT
clip-assigned patients had a poor outcome (mRS 3-6) at 1 year. 18 In the Barrow ruptured aneurysm trial, 62(36.5%) of 170 patients after clipping had a poor outcome (mRS 3-6) at 6 month.19 In our study, only 15 (8.7%) of 173 had a poor outcome (mRS 3-6) with keyhole
SC
surgery at 6 month. But actually, our result is good compared with overall other outcomes because we selected good grade patient with this keyhole approach. Because of this selection
M AN U
bias we cannot say that keyhole approach is better than conventional approach. A preconception about transciliary keyhole surgery is that it is difficult to apply this approach for operation on ruptured aneurysms because of brain swelling, while another preconception is that if unexpected re-bleeding occurs before proximal control of the parent
TE D
vessel, the operator would not be able to properly manage the bleeding. As a result, keyhole surgery has been attempted mostly in patients with UIAs. Thereby, it is hard to find any reports on the surgical outcomes of transciliary keyhole surgery for ruptured aneurysms, so
EP
the feasibility of this type of keyhole surgery on RIAs has not been established. Here, we report good surgical results of ruptured anterior circulation aneurysms with transciliary
AC C
keyhole surgery. Our study showed that this approach was feasible with good grade patients even with ruptured aneurysms. The degree of brain swelling in good grade patients which can be estimated by the extent of cortical sulci effacement on preoperative brain CT was not higher compared with that of poor grade patients, so adequate surgical room could be obtained with arachnoid dissection and CSF drainage with keyhole surgery as with unruptured aneurysms. Some poor grade patients with anterior communicating artery aneurysm ruptures who do not show cortical sulci effacement can be candidates for keyhole surgery. Poor grade patients who gradually deteriorate due to hydrocephalus but with an
14 Park
ACCEPTED MANUSCRIPT
initial good grade can be managed with keyhole surgery by first attempting intraoperative ventriculostomy. Unexpected rupture of an aneurysm is a major risk in all aneurysmal surgeries. Lindert
RI PT
reported 4 (3%) of 139 patients had an unexpected rupture of an aneurysm during supraorbital keyhole surgery.7 The incidence of re-bleeding is related to large aneurysm size and amount of subarachnoid hemorrhage. The cohort in our keyhole surgery group had
SC
relatively small aneurysms (5.5 mm in size) and had a lower SAH grade compared with conventional surgery reports. We experienced only two cases of unexpected bleeding during
M AN U
craniotomy before proximal control of the parent artery. One case that deteriorated during the anesthetic procedure underwent a craniectomy instead of the planned keyhole surgery. Another case developed sudden brain bulging after opening the dura suggesting re-bleeding. This patient was managed with gentle compression of the craniotomy site with gauze to control bleeding by increased intracranial pressure. With the CSF drain via ventriculostomy,
TE D
partial frontal lobe resection provided adequate space for clipping.20,21 A pre-operative imaging study and accurate clinical assessment are important when
EP
deciding the appropriate surgical approach. Three-dimensional CT-angiography imaging is very helpful in determining the approaching route. The surgical corridor simulating an
AC C
operative craniotomy can be made by the operator in the CT work station with surface rendering images from the CT angiography (Figure 2). With this computational reconstruction technique, we can determine which surgical approach is more appropriate for an individual patient`s situation. Over the past few decades, there have been many reports on the clinical outcomes of RIAs and risk factors of poor outcomes. According to the majority of previous reports, admission clinical status (Hunt-Hess grade, WFNS), patient age and aneurysm size were associated with poor prognoses. These findings are similar to our results. The only major
15 Park
ACCEPTED MANUSCRIPT
difference from some reports is the relation between Fisher grade and poor outcome. Ogilvy et al. reported Fisher grade as a meaningful poor outcome predictor, but it did not show any meaningful results in our study.22 We conducted conventional pterional craniotomy or
RI PT
decompressive craniectomy instead of keyhole surgery in patients with intracerebral hemorrhage, which may have led to this difference. For the location of the aneurysm, there was no significant difference in outcomes in anterior circulation between previous reports and
SC
our keyhole cohorts. Overall, our surgery results showed good prognoses because we mainly selected operable patients for minimally invasive surgery.
M AN U
There have been many reports concerning intracranial aneurysmal size over the past few decades. The mean (median) maximum diameter of aneurysms varied in these previous studies: 8.2±4 mm in RIAs in the Cooperative Aneurysm Study from 1970 to 1977,23 11.4±1.2 mm in RIAs in the Johns Hopkins series,24 and 8.0 mm in RIAs in a recent study by
TE D
Weir.25 The mean maximum diameter of 5.5 mm in RIAs from our study was relatively smaller than the size in these other studies.
In all patients operated using the transciliary supraorbital keyhole approach, there was
EP
sufficient room to identify the rupture site, as well as for clip application. The area of exposure of the parasellar region through the smaller supraorbital keyhole approach was as
AC C
adequate as the larger pterional or orbitozygomatic skull base approach and was sufficient for arachnoid dissection and CSF drainage.26,27 Even in ruptured aneurysmal keyhole surgery, the surgeon obtained enough surgical space with the aid of a head up position, hyperventilation, hyperosmolar agents as well as arachnoid dissection if there was no severe brain swelling. An in situ CSF drain by ventricular tapping is helpful when the above procedures cannot provide sufficient brain relaxation. However, rapid CSF drainage may cause unexpected re-bleeding of an aneurysm, so drainage should be done slowly and gently. One third of our series underwent intraoperative ventriculostomy and drainage of CSF amounts of 40 – 60 ml was
16 Park
ACCEPTED MANUSCRIPT enough to obtain access to the carotid cistern.
Several patients had prominent frontal sinuses. The posterior wall of the frontal sinus can be injured during craniotomy which is conducted on the frontal bone in the case of a wide
RI PT
frontal sinus. This defect can be a source of post-operative CSF leakage and wound infection. To avoid frontal sinus wall injury during craniotomy, the lateral extent of the frontal sinus should be evaluated by a preoperative skull X-ray. If a patient had a prominent frontal sinus,
SC
we measured the lateral margin of the frontal sinus and marked it on the skin before incision to avoid opening the frontal sinus. However, 7 patients required opening of the frontal sinus
M AN U
due to its prominent size. We tried to keep the mucous membrane intact and seal the defect ○ and HistoacrylR ○ to prevent infection completely with an artificial sealant such as MedporR or cerebrospinal fluid leakage. Therefore, a wide frontal sinus is not an absolute contraindication of the transciliary keyhole approach. Frontal sinus wall defects can be
surgery.
TE D
secured easily by a surgical sealant because the defect was not wide in transciliary keyhole
Anosmia due to olfactory nerve injury is a possible postoperative complication. It usually
EP
occurs because of frontal lobe retraction. The anterior communicating artery is a well-known vulnerable lesion for postoperative olfactory nerve damage. Park et al. reported on the
AC C
incidence of olfactory dysfunction after pterional approaches for ruptured cerebral aneurysms. Four patients (4%) 18 of 104 showed olfactory dysfunction with an ipsilateral pterional approach for ruptured aneurysms of anterior circulation, except for the anterior communicating artery. The ipsilateral pterional approach for a ruptured anterior communicating artery had a higher incidence of olfactory dysfunction, which developed in 10 of 70 patients (14%).28 Eriksen et al. reported a high incidence (89%) of anosmia after pterional approaches for clipping ruptured aneurysms.29 A 15% incidence of olfactory dysfunction, including hyposmia and anosmia, was reported by Aydin et al.30 A 39%
17 Park
ACCEPTED MANUSCRIPT
incidence of anosmia after aneurismal SAH was reported by Wermer et al.31 In our case, 23 patients of 188 (12%) presented with postoperative olfactory dysfunction. Since there were no previous studies on olfactory nerve injury after supraorbital keyhole surgery, our report
RI PT
has great value. Because of massive frontal lobe retraction, the transciliary keyhole approach has a considerable risk of olfactory nerve injury. In our study, there were a large number of patients with olfactory dysfunction similar to the incidence in the previous above-mentioned
SC
studies. However, most of the patients with olfactory dysfunction were in the early series of the study. Once we started to use the sub-olfactory approach with olfactory nerve dissection
cases with olfactory dysfunction.
M AN U
from the frontal lobe in transciliary keyhole surgery, it considerably decreased the number of
The frontal or frontotemporal branch of the facial nerve is especially vulnerable to injury during neurosurgical procedures such as cranioorbitozygomatic, pterional, and supraorbital
TE D
keyhole approaches that involve an incision and soft-tissue dissection in the frontal and temporal regions. Reisch and Perneczky previously reported a 5.5% incidence of permanent palsy of the frontalis muscle.15 But there was no frontalis palsy during our study. Performing
EP
sub-periosteal dissection without the temporalis muscle incision, and the lack of a lateral extension of 1cm or more from the superior temporal line made it possible to prevent damage
AC C
in the frontal branches of the facial nerves in the superficial fascial layer. Bony malposition happens during the rigid fixation of free bone flap not because of screw loosening. The skin incision is so small that the bone flap margin is difficult to accurately identify and there is a risk of incorrect fixation So surgeons should bear in mind that free bone flap is replaced in correction position.
CONCLUSION
18 Park
ACCEPTED MANUSCRIPT
Thus far, a standard craniotomy is typically selected as the surgical treatment of choice for ruptured intracranial aneurysm surgery. However, we demonstrated a favorable outcome with
RI PT
transciliary keyhole surgery in ruptured anterior circulation aneurysms. Our results showed that initial Hunt-Hess grade, WFNS scale, aneurysmal size and post-operative vasospasm were related risk factors of poor outcomes. Considering these related risk factors, surgical
SC
candidates with good clinical grades and operable patients from a preoperative imaging study can be selected. Because the minimally invasive transciliary supraorbital approach has many
M AN U
advantages, we suggest that this approach is an acceptable treatment option for selected case of ruptured anterior circulation aneurysm.
Limitation
TE D
The validity of the logistic model becomes problematic when the ratio of the numbers of events per variable analyzed becomes small. The parameter estimates may be biased and the usual tests of significance may not be valid. Because the number of poor outcome events was
EP
small (poor outcome n=23/total n=177), we could not use multivariate logistic regression analysis and only used univariate logistic regression analysis, which is the limitation of this
AC C
study. This is a retrospective cohort study so there should be selection bias. We selected good grade patients for surgery and it may affect to good surgical outcome.
19 Park
ACCEPTED MANUSCRIPT
Acknowledgement Statistical consultation was supported by a grant of the Korea Health Technology R&D
RI PT
Project through the Korea Health Industry Development Institute (KHIDI), funded by the
Author Disclosures
SC
Ministry of Health & Welfare, Republic of Korea (grant number: HI14C1062).
The authors report no conflict of interest concerning the materials or methods used in this
M AN U
study or the findings specified in this paper.
Conflicts of Interest and Ethical Adherence
AC C
EP
TE D
The authors have no conflicts of interest and adhered to ethical standards in this study.
20 Park
ACCEPTED MANUSCRIPT REFERENCES 1.
Delashaw JB, Jr., Jane JA, Kassell NF, Luce C. Supraorbital craniotomy by fracture of the anterior orbital roof. Technical note. J Neurosurg. 1993;79:615-618. Goel A. Supraorbital craniotomy. J Neurosurg. 1994;81:642-643.
3.
Jho HD. Orbital roof craniotomy via an eyebrow incision: a simplified anterior skull
RI PT
2.
base approach. Minim Invasive Neurosurg. 1997;40:91-97.
Ko Y, Yi HJ, Kim YS, Oh SH, Kim KM, Oh SJ. Eyebrow incision using tattoo for
SC
4.
20. 5.
Ramos-Zuniga R. The trans-supraorbital approach. Minim Invasive Neurosurg. 1999;42:133-136.
6.
M AN U
anterior fossa lesions: technical case reports. Minim Invasive Neurosurg. 2001;44:17-
Steiger HJ, Schmid-Elsaesser R, Stummer W, Uhl E. Transorbital keyhole approach to communicating
artery
aneurysms.
Neurosurgery.
2001;48:347-351;
TE D
anterior
discussion 351-342. 7.
van Lindert E, Perneczky A, Fries G, Pierangeli E. The supraorbital keyhole approach
EP
to supratentorial aneurysms: concept and technique. Surg Neurol. 1998;49:481-489; discussion 489-490.
Paladino J, Mrak G, Miklic P, Jednacak H, Mihaljevic D. The keyhole concept in
AC C
8.
aneurysm surgery--a comparative study: keyhole versus standard craniotomy. Minim
Invasive Neurosurg. 2005;48:251-258.
9.
Czirjak S, Nyary I, Futo J, Szeifert GT. Bilateral supraorbital keyhole approach for multiple aneurysms via superciliary skin incisions. Surg Neurol. 2002;57:314-323; discussion 323-314.
10.
Frazier CH. I. An Approach to the Hypophysis through the Anterior Cranial Fossa.
21 Park
ACCEPTED MANUSCRIPT Ann Surg. 1913;57:145-150. 11.
Wilson DH. Limited exposure in cerebral surgery. Technical note. J Neurosurg. 1971;34:102-106. Brock M, Dietz H. The small frontolateral approach for the microsurgical treatment of
RI PT
12.
intracranial aneurysms. Neurochirurgia (Stuttg). 1978;21:185-191. 13.
Jane JA, Park TS, Pobereskin LH, Winn HR, Butler AB. The supraorbital approach:
14.
SC
technical note. Neurosurgery. 1982;11:537-542.
Hernesniemi J, Ishii K, Niemela M, Smrcka M, Kivipelto L, Fujiki M, et al. Lateral
M AN U
supraorbital approach as an alternative to the classical pterional approach. Acta Neurochir Suppl. 2005;94:17-21. 15.
Reisch R, Perneczky A, Filippi R. Surgical technique of the supraorbital key-hole craniotomy. Surg Neurol. 2003;59:223-227.
Yasargil MG, Reichman MV, Kubik S. Preservation of the frontotemporal branch of
TE D
16.
the facial nerve using the interfascial temporalis flap for pterional craniotomy. Technical article. J Neurosurg. 1987;67:463-466. Mitchell P, Vindlacheruvu RR, Mahmood K, Ashpole RD, Grivas A, Mendelow AD.
EP
17.
Supraorbital eyebrow minicraniotomy for anterior circulation aneurysms. Surg Neurol.
18.
AC C
2005;63:47-51; discussion 51. Gupta SK, Chhabra R, Mohindra S, Sharma A, Mathuriya SN, Pathak A, et al. Long-
Term Outcome in Surviving Patients After Clipping of Intracranial Aneurysms. World Neurosurg. 2014;81:316-321.
19.
Spetzler RF, McDougall CG, Zabramski JM, Albuquerque FC, Hills NK, Russin JJ et al. The Barrow Ruptured Aneurysm Trial: 6-year results. J Neurosurg. 2015;123:609617.
20.
Solanki C, Pandey P, Rao KV. Predictors of aneurysmal rebleed before definitive
22 Park
ACCEPTED MANUSCRIPT
surgical or endovascular management. Acta Neurochir (Wien). 2016;158:1037-1044. 21.
van Donkelaar CE, Bakker NA, Veeger NJ, Uyttenboogaart M, Metzemaekers JD, Luijckx GJ, et al. Predictive Factors for Rebleeding After Aneurysmal Subarachnoid
RI PT
Hemorrhage: Rebleeding Aneurysmal Subarachnoid Hemorrhage Study. Stroke. 2015;46:2100-2106. 22.
Ogilvy CS, Carter BS. A proposed comprehensive grading system to predict outcome
SC
for surgical management of intracranial aneurysms. Neurosurgery. 1998;42:959-968; discussion 968-970.
Weir B. Unruptured intracranial aneurysms: a review. J Neurosurg. 2002;96:3-42.
24.
Sjoberg RJ. Risk factors for the development and rupture of intracranial berry
M AN U
23.
aneurysms. Am J Med. 1986;80:A65. 25.
Weir B, Amidei C, Kongable G, Findlay JM, Kassell NF, Kelly J, et al. The aspect
TE D
ratio (dome/neck) of ruptured and unruptured aneurysms. J Neurosurg. 2003;99:447451. 26.
Cheng CM, Noguchi A, Dogan A, Anderson GJ, Hsu FP, McMenomey SO, et al.
EP
Quantitative verification of the keyhole concept: a comparison of area of exposure in the parasellar region via supraorbital keyhole, frontotemporal pterional, and
27.
AC C
supraorbital approaches. J Neurosurg. 2013;118:264-269. Yagmurlu K, Safavi-Abbasi S, Belykh E, Kalani MY, Nakaji P, Rhoton AL, Jr., et al.
Quantitative anatomical analysis and clinical experience with mini-pterional and mini-orbitozygomatic approaches for intracranial aneurysm surgery. J Neurosurg.
2016:1-14. 28.
Park J, Lee SH, Kang DH, Kim JS. Olfactory dysfunction after ipsilateral and contralateral
pterional
approaches
2009;65:727-732; discussion 732.
for
cerebral
aneurysms.
Neurosurgery.
23 Park
ACCEPTED MANUSCRIPT 29.
Eriksen KD, Boge-Rasmussen T, Kruse-Larsen C. Anosmia following operation for cerebral aneurysms in the anterior circulation. J Neurosurg. 1990;72:864-865.
30.
Aydin IH, Kadioglu HH, Tuzun Y, Kayaoglu CR, Takci E, Ozturk M. Postoperative
approach. Minim Invasive Neurosurg. 1996;39:71-73.
Wermer MJ, Donswijk M, Greebe P, Verweij BH, Rinkel GJ. Anosmia after
SC
aneurysmal subarachnoid hemorrhage. Neurosurgery. 2007;61:918-922; discussion
EP
TE D
M AN U
922-913.
AC C
31.
RI PT
anosmia after anterior communicating artery aneurysms surgery by the pterional
24 Park
ACCEPTED MANUSCRIPT
Figure 1. Post-operative skull 3D CT image: A 2x3 cm size free bone flap was made close to the anterior skull base with preservation of the superior orbital rim.
RI PT
Figure 2. Pre-operative brain 3D CT angiography is a helpful tool for surgical stimulation. The surgical corridor simulating operative craniotomy can be made by the operator in the CT
SC
work station with surface rendering images from the CT angiography.
AC C
EP
TE D
M AN U
Figure 3. Modified Rankin Scale of post-operative patients.
ACCEPTED MANUSCRIPT TABLE 1. Summary of cases with transciliary aneurysm neck clipping in ruptured anterior circulation aneurysm 188
Male/Female
72/116
Age
23-86 years; mean 58.5
Preoperative Fisher grade on CT
1: 8 (4.3%) 2: 44 (23.4%)
4: 20 (10.6%)
Preoperative Hunt-Hess grade
M AN U
5: 1 (0.5%)
SC
3: 115 (61.2%)
RI PT
Total cases
I: 13 (6.9%)
II: 130 (69.1%) III: 29 (15.4%) IV: 13 (6.9%)
Mean aneurysm size
5.6 mm Middle cerebral artery
AC C
EP
Location
TE D
V: 3 (1.6%)
51
MCA bifurcation
Anterior communicating artery
46 82
Anterior cerebral artery
5
Internal cervical artery
46
ICA bifurcation Posterior communicating artery Ophthalmic artery Anterior choroidal artery
3 44 1 2
ACCEPTED MANUSCRIPT TABLE 2. Adverse events during the follow-up period after transciliary keyhole surgery Complications
Acute/Delayed Acute
7
Hyposmia and anosmia
Acute
9
Unexpected rupture of aneurysm during surgery
Acute
2
Delayed
25
Vasospasm Vasospasm with chemical angioplasty
RI PT
Frontal sinus opening
19
Delayed
3
SC
Wound and bone flap complications bone flap extrusion
Meningitis Hygroma Visual disturbance
AC C
EP
TE D
Cerebral infarction (ACA infarction)
M AN U
wound swelling
1 2
Delayed
3
Delayed
3
Acute
3
Acute
1
ACCEPTED MANUSCRIPT TABLE 3. Modified Rankin Scale of post-operative patients At discharge
3 months
6 months
0 - No symptoms
53 (28.2%)
105 (59.3%)
121 (70.0%)
1 - No significant disability
93 (49.5%)
44 (24.9%)
31 (18.0%)
2 - Slight disability
14 (7.4%)
10 (5.6%)
6 (3.5%)
3 - Moderate disability
11 (5.9%)
8 (4.5%)
9 (5.2%)
4 - Moderately severe disability
9 (4.8%)
7 (4.0%)
4 (2.3%)
5 - Severe disability
3 (1.6%)
2 (1.1%)
2 (2.7%)
6 – Death
5 (2.7%)
SC
(SLE 1, pneumonia 1, ARF 1)
1 (0.6%)
188/0
AC C
EP
TE D
Total/follow-up loss
0 (0%)
(ARF 1)
M AN U
(combined systemic disease)
RI PT
mRS
177/6
173/2
ACCEPTED MANUSCRIPT TABLE 4. Characteristics of patients with poor grade outcomes (mRS 3-6) mRS Scale Total good grade
poor grade
(n=154)
(n=23)
P value
Age mean±sd
51.7±12.3
49.7±10.5
median (min-max)
50 (23-86)
49 (22-77)
64.7±15.2
0.0001
67 (37-86)
0.432
67 (37.8)
60 (39.0)
Female
110 (62.2)
94 (61.0)
Hunt-Hess grade grade Ⅰ, Ⅱ, Ⅲ
161 (91.0)
grade Ⅳ, Ⅴ
16 (9.0)
grade Ⅰ, Ⅱ, Ⅲ grade Ⅳ, Ⅴ
16 (69.6) <.0001
147 (95.5)
14 (60.9)
7 (4.5)
9 (39.1) 0.083
50 (28.2)
47 (30.5)
3 (13.0)
127 (71.8)
107 (69.5)
20 (87.0)
EP
WFNS scale
TE D
Fisher Grade
7 (30.4)
M AN U
Male
SC
Gender
RI PT
(n=177)
0.0001
153 (86.4)
140 (90.9)
13 (56.5)
grade Ⅳ, Ⅴ
24 (13.6)
14 (9.1)
10 (43.5)
6.0±2.4
5.8±2.1
7.1±3.6
5.5 (1.0-20.0)
5.0 (1.0-15.7)
6.0 (2.0-20.0)
151.5±51.0
151.0±51.6
154.8±47.7
140 (60-400)
139 (60-400)
145 (80-240)
AC C
grade Ⅰ, Ⅱ, Ⅲ
Aneurysm size mean±sd
median (min-max)
0.090
Skin to skin time mean±sd median (min-max) Location of aneurysm
0.744
0.421
ACCEPTED MANUSCRIPT ICA
46 (26.0)
41 (26.6)
5 (21.8)
ACA
82 (46.3)
73 (47.4)
9 (39.1)
MCA
49 (27.7)
40 (26.0)
9 (39.1)
Re-bleeding
0.467 160 (90.4)
140 (90.9)
Yes
17 (9.6)
14 (9.1)
20 (87.0)
RI PT
No
3 (13.0)
Ventricle puncture
0.924
117 (66.1)
102 (66.2)
Yes
60 (33.9)
52 (33.8)
No
154 (87.0)
Yes
23 (13.0)
M AN U
Vasospasm
15 (65.2)
SC
No
8 (34.8)
140 (90.9)
14 (60.9)
14 (9.1)
9 (39.1)
Values are numbers (percentages) or mean±sd.
AC C
EP
TE D
P values were calculated using the Chi-square test, Fisher's exact test or Student's t-test.
0.001
ACCEPTED MANUSCRIPT TABLE 5. Univariate analysis of risk factors associated with poor grade outcomes (mRS 3-6). OR (95% CI)
P value
Hunt-Hess grade ,
grade
,
,
Reference
13.50 (4.36-41.78)
Fisher Grade
grade
,
,
Reference
SC
,
2.93 (0.83-10.33)
M AN U
grade
WFNS scale ,
grade
,
,
Aneurysm size Skin to skin time Re-bleeding
EP
No Yes
<.0001
0.10
Reference
TE D
grade
RI PT
grade
7.69 (2.86-20.72)
<.0001
1.21 (1.03-1.42)
0.019
1.00 (0.99-1.01)
0.743
Reference 1.50 (0.40-5.68)
0.551
No Yes
AC C
Ventricle puncture
Reference 1.05 (0.42-2.63)
0.924
Vasospasm No
Reference
Yes
6.43 (2.36-17.50)
OR: Odds Ratio, CI: Confidence Interval
0.0003
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Trans-eyebrow keyhole aneurysmal neck clipping is one of minimal invasive cranial surgery.
This procedure has the advantage of less damage to normal tissues and a shorter recovery period. The application of this technique to ruptured aneurysms is still controversial.
The authors experienced more than 200 cases of keyhole surgery on ruptured aneurysms and
RI PT
obtained good surgical outcome.
We analyzed for poor outcome factors and wanted to provide selection indication for keyhole
AC C
EP
TE D
M AN U
SC
surgery on ruptured aneurysms.
ACCEPTED MANUSCRIPT Disclosure-Conflicts documentation
Title : Risk factor analysis for poor outcomes in supraorbital keyhole aneurysm clipping
RI PT
for ruptured anterior circulation aneurysm
SC
Disclosure of Funding : The authors have no financial support or interest.
M AN U
Conflicts of Interest and Ethical Adherence : The authors have no conflict of interest or ethical adherence
AC C
EP
TE D
Statement of Authorship : Ik Seong Park has a corresponding authorship