Clinical Neurology and Neurosurgery 125 (2014) 182–188

Contents lists available at ScienceDirect

Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro

Surgical outcomes after posterior fossa decompression with and without duraplasty in Chiari malformation-I Silky Chotai, Azedine Medhkour ∗ Division of Neurosurgery, Department of Surgery, University of Toledo Medical Center, Toledo, USA

a r t i c l e

i n f o

Article history: Received 9 June 2014 Received in revised form 8 July 2014 Accepted 20 July 2014 Available online 12 August 2014 Keywords: Chiari malformation Syringomyelia Dura-splitting Extra-dural Intradural

a b s t r a c t Introduction: Chiari malformation-I (CM) is one of the most controversial entities in the contemporary neurosurgical literature. Posterior fossa decompression (PFD) is the preferred treatment for CM with and without syringomyelia. A variety of surgical techniques for PFD have been advocated in the literature. The aim of this study was to evaluate our results of surgically treated patients for CM-I with and without syringomyelia; using extradural dura-splitting and intradural intraarachnoid techniques. Methods: A retrospective review of the medical records of all the patients undergoing PFD was conducted. Symptomatic patients with tonsillar herniation ≥ 3-mm below the foramen magnum on neuroimaging, and CSF flow void study demonstrating restricted or no CSF flow at the craniocervical junction, were offered surgical treatment. In patients without syringomyelia, extradural decompression with thinning of the sclerotic tissue at the cervicomedullary junction and splitting of outer dural layer was performed. In patients with syringomyelia, the dura was opened and an expansile duraplasty was performed. Results: The mean age of 8 males and 34 females was 33.8 years (range, 16–58 years). Headache (39/41; 95%), and/or tingling and numbness (17/41; 42%) were the most common presenting symptoms. The syrinx was associated with CM-I in 5/41 (12%) patients. PFD without durotomy was performed in 29/41 (73%) patients. The mean duration of preoperative symptoms was significantly longer in duraplasty group (4.6 versus 1.7 years, P = 0.005, OR = 0.48, CI = 0.29–0.8). The use of duraplasty was significantly associated with presence of complications (P = 0.004, OR = 0.5, CI = 0.3–0.8) and longer duration of hospital stay (P = 0.03, OR = 2.7, CI = 1.1–6.8). The overall complication rate was 6/41(15%) patients. The overall improvement rate was evident in 84% (36/41); 12% (5/41) were stable; and 5% (2/41) had worsening of symptoms. The history of prior CM decompression was associated with unfavorable outcomes (P = 0.04, OR = 14, CI = 1.06–184). One patient experienced recurrence one year after the PFD with duraplasty. Conclusion: The present study reports favorable surgical outcomes with extra-dural decompression of the posterior fossa in patients CM-I without syringomyelia. For patients with syringomyelia and history of prior PFD, intradural intra-arachnoid decompression is required. The prior history of decompression was associated with unfavorable outcomes. The use of duraplasty was associated with longer duration of hospital stay and higher complication rate. Further large cohort prospective study is needed to provide any recommendation on the indication of intra or extradural decompression for a given CM-I patient. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Chiari malformation-I (CM-I) is herniation of cerebellar tonsil, 3-mm or greater, below the foramen magnum with associated developmental abnormality of posterior fossa [1–6]. The herniated tonsil causes obstruction at the cervico-vertebral junction,

∗ Corresponding author at: Division of Neurosurgery, Department of Surgery, University of Toledo Medical Center, Toledo, OH 43614, USA. Tel.: +1 419 383 3547; fax: +1 419 383 6570. E-mail address: [email protected] (A. Medhkour). http://dx.doi.org/10.1016/j.clineuro.2014.07.027 0303-8467/© 2014 Elsevier B.V. All rights reserved.

which may lead to formation of syringomyelia [4,7–9]. Various theories have been contemplated for pathogenesis, natural history, and diagnostic as well as treatment modality for CM-I with and without syringomyelia [2,4,8–15]. It is generally accepted that the patients with symptomatic CM-I with and without syrinx are benefited by posterior fossa decompression (PFD). The technique for decompression, however, varies based on the varied understanding of pathophysiology, surgeons’ preference and patient characteristics [10,16–23]. Several techniques that have been advocated for PFD includes bone decompression only; extradural lysis of sclerotic tissue and removal of the outer dural layer [17,24–26]; intradural extra-arachnoid durotomy with and without duraplasty;

S. Chotai, A. Medhkour / Clinical Neurology and Neurosurgery 125 (2014) 182–188

183

Fig. 1. (A–F) Intraoperative images showing: (A) thick sclerotic band is thinned and removed at the cervicomedullary junction; (B) extradural removal of the sclerotic band; (C) opening of outer dural layer; (D) a small nick was created inadvertently on the inner dural layer; (E) the dural nick was sutured and (F) surgicel and surgifoam are placed over the dura and hemostasis perfected.

intra-arachnoid lysis of the scarring and adhesions around the herniated tonsil [10,12,27–31]; reduction of the herniated tonsil by coagulation or partial tonsillectomy [11,21,31]; as well as opening the foramen of magendie and obex plugging [9,32,33]. The intradural techniques are associated with increased risk of complications including wound infection, CSF leak, pseudomeningocele, meningitis and complication associated with dural graft [5,13,16,29,34–39]. Due to a greater risk of the associated complications, more conservative extradural decompression techniques; dura-splitting, dura scoring [37], or dural delamination, have been advocated. PFD with extradural lysis of the sclerotic band and opening of the outer dural layer offers a minimally invasive decompression technique with lower risk of complications. Several neurosurgeons have adopted this technique with the caveat that there might be an increased risk of reoperation, as published in earlier studies [8,20,25,26,35,39–43]. Recent studies have demonstrated that extradural decompression is effective as PFD with duraplasty in a select group of patients [25,26]. The selection of surgical technique based on the clinical presentation of CM-I with and without syringomyelia might be a reasonable approach. The aim of this study was to evaluate and report our results of surgically treated patients for CM-I with and without syringomyelia; using extradural dura-splitting and intradural intraarachnoid techniques. 2. Methods 2.1. Patients We conducted a retrospective review of medical record data of all the patients, between January 2000 and June 2012, treated with surgical decompression for CM-I at our institute. The institutional review board approved the study. Preoperative parameters that were reviewed included age, sex, symptoms, duration of symptoms, neurological deficits, primary or recurrent CM-I, preoperative imaging showing the degree of tonsillar descent, presence or absence of syrinx, CSF flow on cine-flow MRI. Intraoperative notes and inpatient charts were reviewed for the details of surgical strategy employed including the durotomy, duraplasty, type of graft used, reduction of cerebellar tonsils, placement of shunt

for syrinx, postoperative complications including any neurological, medical and surgical complications. 2.2. Surgical procedure The criteria for offering surgical treatment included: (1) symptomatic patients; (2) neuroimaging demonstrating tonsillar herniation ≥ 5-mm below the foramen magnum; and (3) cine flow MRI study demonstrating restricted or no flow around the foramen magnum. All the patients underwent standard preoperative evaluation and perioperative antibiotics were administered to all patients. Under general anesthesia, patients were placed in a prone position with slight flexion of the neck to allow for visualization of the occipital bone. A midline incision extending from the inion to C2 was carried out. The subperiosteal dissection of the soft tissue, occipital and paraspinal muscle was carried out. The dissection was continued to expose the hemi-lamina of C1, the tip of spinous process of C2, and the suboccipital bone. A midline suboccipital craniectomy was performed with two paramedian burr holes at about 1.5 cm below the inion. The suboccipital craniectomy was completed by connecting the burr holes via a craniotome along with opening of the foramen magnum. The size of craniectomy was 2.5 cm × 3 cm in the circular fashion all the way down to the foramen magnum. A C1 laminectomy was carried out and up to 2 cm of C1 lamina was removed using a power drill and Kerrison. In cases where decompression was carried out without durotomy, the dura was inspected under the microscope; a thick sclerotic tissue compressing the dura was identified (Fig. 1A). The epidural microdissection was executed by sharp transaction of the adhesive sclerotic band (Fig 1B). The outer layer of the dura was divided up to the inferior and lateral extent of the bony decompression and the intact arachnoid was visualized through the intact inner dural layer (Fig. 1C). Care was taken to avoid any inadvertent opening of the dura. In cases where a rent was created in the dura during the epidural lysis, the opening was sealed with either a low voltage bipolar, one 4.0 suture or a running suture (Fig. 1D and E). Thinning of the dural layer led to a release of pressure at the cervicomedullary junction giving more room for expansion of herniated cerebellar tonsil. Once the dura was thinned, the cervicomedullary junction elements could

184

S. Chotai, A. Medhkour / Clinical Neurology and Neurosurgery 125 (2014) 182–188

Fig. 2. (A) Pre-operative MRI T2-image showing cerebellar descent and a cervical C1–C4 syrinx and (B) postoperative 1-year follow-up MRI T2-imaging after suboccipital decompression, C1 laminectomy and duraplasty, demonstrating an artificial cisterna magna, rounded shape of tonsil and reduction in syrinx size.

be visualized pulsating under the microscope. Dural integrity was confirmed and the Surgicel was placed over the dura (Fig. 1F). The dura was opened in all the patients with syrinx, a few patients without syrinx operated on during the early years in the series, as well as those patients with a prior history of CM-I decompression. The dura was incised in a Y-shape under the microscope and held open by sutures. The arachnoid was inspected for any scarring and adhesions; if present, a sharp dissection was carried out for lysis of adhesions and scarring. The cerebellar tonsils were coagulated at their tips with a low-voltage bipolar. Duraplasty using fascia lata and occipital muscle was performed in cases operated on in earlier years; the artificial dura (DuraGen) and pericranial graft was used for the duraplasty performed during later years in the series. A central tacking suture was placed in the dural graft through the fascia or muscle to avoid adherence of the dural graft to the underlying arachnoid. The dural graft was covered with TissealTM and the closure was performed in standard layered fashion. All the patients were admitted to neurosurgical ICU postoperatively.

patient. The patients with symptom improvement were considered to have favorable outcome whereas patients with no change or worse symptoms were classified as unfavorable outcome. The modified Rankin Score (mRS) scale was used to grade the functional outcomes of the patients. 2.4. Statistical analysis All the continuous variables were represented with a mean ± SD. Univariate analysis was employed to analyze the association of the clinical parameters to the outcome. The correlation between complications and various clinical parameters was performed using Pearson’s and Spearman correlation test. The association between duraplasty and clinical parameters was analyzed using Chi-square test. Statistical analysis was performed with a statistical software package (SPSS Version 20.0, SPSS Inc., IBM). A P-value of less than or equal to 0.05 was considered statistically significant. 3. Results

2.3. Follow-up and outcomes 3.1. Preoperative parameters All the patients had a postoperative follow-up visit at 4 weeks, with subsequent follow-ups scheduled at 6 months and 1 year after surgery. The outcomes were evaluated based on the last available follow-up. The parameters evaluated at the follow-up included symptomatic improvement and functional status of the

The study included 8 males and 33 females with the male to female ratio of 1:4.2. The mean age of the study population was 33.8 years (range 16–58 years). Fig. 4 shows the comparison of age and overall postoperative improvement. The mean duration

Fig. 3. (A) Pre-operative MRI T1-image showing cerebellar descent and (B) postoperative MRI demonstrating T1-imaging after suboccipital decompression, C1 laminectomy without dural opening, demonstrating a cisterna magna and rounded shape of tonsil.

S. Chotai, A. Medhkour / Clinical Neurology and Neurosurgery 125 (2014) 182–188

185

Fig. 4. Graph showing correlation between age and improvement.

Table 1 Presenting symptoms in CM-I patients with and without syringomyelia. Total Headache Tremors Gait disturbance Sensory loss Dysphagia Hoarseness Dizziness Tingling numbness Blurred vision Memory loss Hearing difficulty Muscle weakness Gagging Neck Pain Photophobia Back pain Facial numbness Facial drooling Speech difficulty Extremity Paresthesia Tinnitus

41 (100%) 40 (98%) 1 (2%) 7 (17%) 3 (7%) 1(2%) 1(2%) 6 (15%) 18 (44%) 6 (15%) 3 (7%) 4 (10%) 5 (12%) 1 (2%) 16 (39%) 1 (2%) 5 (12%) 3 (7%) 1 (2%) 1 (2%) 2 (5%) 1 (2%)

of symptoms was 2.7 years (range, 0.25–12 years). All the patients demonstrated restricted or no flow on cine-MRI study. The mean descent of the cerebellar tonsil on preoperative MRI was 9.13 mm (range, 5–30 mm). Twelve percent of the patients (5/41) had syrinx with associated CM-I. The characteristic of five patients with syrinx is demonstrated in Table 2. Headache (39/41; 95%) as well as tingling and numbness (17/41; 42%) were the most common presenting symptoms (Table 1). Neurological exam demonstrated sensory, motor or gait abnormalities in 29/41 (71%) patients. Two patients had communicating hydrocephalus that was shunted before CM decompression. A past history of PFD was present in 3/41 (7%) patients. The mean duration of PFD after prior decompression was 16.3 months. The prior PFD was significantly associated with unfavorable outcomes (P = 0.02, OR = 0.2, CI = 0.01–0.7). There was no correlation of prior PFD and occurrence of complications during the present decompression. 3.2. Surgical details PFD without durotomy was performed in 29/41 (73%) patients. There were no unintentional durotomy in any case of extradural decompression. In 4/29 (14%) cases, small punctate holes were inadvertently created in the dura, either a low voltage bipolar or a running suture, Surgical and Tseal were used to secure the opening. A total of 12/41(33%) patients underwent opening of the dura; of this 5 (42%) had syrinx; three (25%) had a history of prior CM decompression that failed; and four patients (33%) were operated on early in the series. In all the cases with durotomy, shrinking of cerebellar tonsils was performed. Fascia lata was used for duraplasty in the

earliest patients in the study (3/12). For patients operated later on in the study, occipital pericranial graft was used in 3/12 patients and artificial dura (DuraGen, Gore-Tex) was used in 6/12 patients. 3.3. Duraplasty Twelve patients (29%) underwent duraplasty. The mean duration of preoperative symptoms was significantly longer, 4.6 years, in duraplasty group compared to1.7 years in non-duraplasty group (P = 0.005, OR = 0.48, CI = 0.29–0.8). The use of duraplasty was associated with occurrence of complications (P = 0.004, OR = 0.5, CI = 0.3–0.8) and longer duration of hospital stay (P = 0.03, OR = 2.7, CI = 1.1–6.8). The mean duration of hospital stay was 3.9 days (range, 3–6 days); mean 4.13 days in PFD with duraplasty patients and mean 3.1 days for patients undergoing PFD without duraplasty. 3.4. Complications The overall complication rate was 12% (5/41); four of the 12 patients in duraplasty group had surgical or transient neurological complications. Overall, surgical complications developed in 3/41 (7%) patients. Two patients developed wound infection; one was in the duraplasty group and the other in the nonduraplasty group. Both patients were treated by wound irrigation and debridement. Another patient in duraplasty group (1/12) had CSF leak. Two patients had transient neurological complications; one patient had speech disturbance and another had blurred vision. Both patients improved prior to discharge. The use of duraplasty (P = 0.01, OR = 18.3, CI = 1.9–18.7) and longer duration of preoperative symptoms were associated with the occurrence of complications (P = 0.007). 3.5. Outcomes The mean follow-up period was 14.7 months (range, 1–60 months). The outcomes were evaluated based on the last available follow-up. All patients with syringomyelia demonstrated postoperative reduction in syrinx volume (Fig. 2). The presence or absence of syrinx was not associated with outcomes or complications. Fig. 3 demonstrates a typical decompression in CM-1 without syrinx. The mRS at the last follow-up was 0, 1 and 2 in 39% (16/41), 46% (19/41), 14% (6/41) respectively. The overall symptomatic improvement was evident in 85% (35/41); 10% (4/41) were stable; and 5% (2/41) had worsening of symptoms. The symptoms that worsened at follow-up were neck stiffness in one patient and headache in the other patient. Of the two patients with symptom worsening, one had PFD with duraplasty and the other had PFD without duraplasty. At the last follow-up, 4/41 patients had headache; of these one patient had headache of similar pattern as preoperative headache and was worse post-operatively. Other patients had

Syrinx improved Tingling numbness 7-mm 10 years headache and back pain, 1 week numbness in upper extremity

5-mm 1 year

Lysis of tonsilar adhesions 4 years

11-mm

None Pericranial graft Cauterization of tonsils

16 months

2

Syrinx improved Intermittant neck pain CSF leak Fascia lata Lysis of adhesions and cauterization of tonsils

24 months

0

Syrinx improved Pain in the neck persist None Pericranial graft

16 months

1

Syrinx improved Intermittent headaches Lysis of adhesions and cauterization of tonsils 7 years

12-mm

Pericranial graft

None

23 months

2

Complete resolution of syrinx 0 54 months Lysis of adhesions Duragen and mobilization of tonsils

5 42/F

4 58/F

3 45/F

2 27/F

1 26/F

Decreased sensation of right side of the body, low back pain Headache, low-back pain, limb dysthesia, gagging Neck pain, left sided tingling and numbness Neck and Shoulder pain, swallowing difficulty, decrease in gag reflex Headache, low-back pain, numbness in upper extremity

1 year

8-mm

None

mRS Symptom outcome Follow-up Complications Dural graft Surgical technique CTD Symptom duration Symptoms Age/sex

Table 2 Characteristics of five patients with syringomyelia.

Resolution of symptoms

S. Chotai, A. Medhkour / Clinical Neurology and Neurosurgery 125 (2014) 182–188

MRI at last follow-up

186

headache that was attributed to migraine (2/6) or refractory error (1/6). Fifteen percent (6/41) of patients reported cervicocephalgia which was attributed to the greater occipital nerve irritation. One patient experienced recurrence of symptoms with hydrocephalus one year after PFD with duraplasty. This patient underwent external lumbar drain placement and later a redo-PFD with duraplasty was performed. The patient demonstrated complete symptomatic and functional improvement at 60 months follow-up. The only clinical parameter associated with unfavorable outcomes was a history of prior CM decompression (P = 0.04, OR = 14, CI = 1.06–184). 4. Discussion 4.1. Outcomes The symptomatic improvement following extradural durasplitting decompression in CM with and without syrinx have been reported in the range of 61.5–93% [1,17,25,26,30,35,43,44]. In present study, 85% patients demonstrated overall improvement and 10% were stable at mean 20 months after surgery. The history of prior PFD was associated with unfavorable outcomes. There was no significant difference in outcomes between extradural durasplitting and intradural duraplasty techniques. The intra-dural decompression with duraplasty and longer duration of preoperative symptoms was associated with occurrence of complication. This finding suggests that early decompression in patients with and without syrinx might be associated with fewer complications. The rate of overall complications in patients undergoing PFD with duraplasty is reported up to 42%; whereas it is 10% in patients where dura is left intact [17,20,21,23,25–27,35,39,41,45]. In present study, 4/41 (9.7%) patients in duraplasty group had complications, only one of these four patients developed CSF leak, one had wound infection, two had transient neurological complications. One patient in non-duraplasty group developed wound infection, which was treated with wound debridement and antibiotics. Most complications are related to CSF exposure to blood and muscle cellular debris and the type of dural graft used. The complications includes the need for reoperation, post-operative CSF leak, pseudomeningocele formation, meningitis, irritation and scarring of the arachnoid leading to obstruction of CSF dynamics, wound breakdown, graft dissolution, encapsulation, foreign-body reaction, excessive scarring, and adhesion formation [16]. Thus, extradural decompression without duraplasty might be associated with lesser complication rates. 4.2. Extradural decompression Extradural decompression is a minimally invasive technique for PFD in CM-I patients. Bone-only decompression has demonstrated improved conduction on brainstem auditory evoked potential without any additional gain with dural opening [46]. In most cases after the osseous decompression, an abnormally thickened sclerotic atalantooccipital membrane or the dural band is found compressing the craniocervical junction and contributing to the abnormal CSF flow. Removal of this sclerotic band would then decompress the posterior fossa and allow for the expansion of the posterior fossa volume [30]. The dura mater is predominantly composed of collagen fibers with a few elastin fibers which are oriented cranio-caudally. The dura mater at the craniocervical junction can be separated into two separate layers; the outer and the inner layer, each have distinct biochemical properties. The outer dura has an elastic fragile behavior while, after splitting the dura, the inner layer has an elasto-plastic property which has the capability to enlarge in suitable stress conditions [24]. Thus, the splitting of the outer dural layer leads to expansion of the posterior fossa

S. Chotai, A. Medhkour / Clinical Neurology and Neurosurgery 125 (2014) 182–188

and formation of the cisterna magna. The young children have more distensible duramater that would expand further, allowing optimal decompression [30]. Most authors have reported their experience with the dura-splitting decompression in the pediatric age group patients [26,35,43,44,47]. However, only few authors have reported favorable outcomes in adult CM patients with durasplitting decompression technique [8,17,25]. In the present study, the outcomes did not differ between the pediatric (

Surgical outcomes after posterior fossa decompression with and without duraplasty in Chiari malformation-I.

Chiari malformation-I (CM) is one of the most controversial entities in the contemporary neurosurgical literature. Posterior fossa decompression (PFD)...
2MB Sizes 5 Downloads 7 Views

Recommend Documents