- Email: [email protected]
Time to Resolution of Symptoms After Suboccipital Decompression with Duraplasty in Children with Chiari Malformation Type I
Original Article
Time to Resolution of Symptoms After Suboccipital Decompression with Duraplasty in Children with Chiari Malformation Type I Eveline Teresa Hidalgo, Yosef Dastagirzada, Cordelia Orillac, Svetlana Kvint, Emily North, Ramona Bledea, Michelle W. McQuinn, Gabriel Redel-Traub, Crystalann Rodriguez, Jeffrey H. Wisoff
BACKGROUND: Duraplasty is a technique successfully used to treat Chiari malformation type I (CM-I). This study describes the timely manner of clinical outcomes and the postoperative course after craniectomy and duraplasty for the treatment of symptomatic CM-I in children.
-
METHODS: A retrospective chart review was performed in 105 consecutive children who underwent surgical decompression of symptomatic CM-I with dural opening by a single surgeon between 1999 and 2015.
months, headaches were resolved in all cases, and syrinx was resolved or decreased in 79% of cases.
-
RESULTS: In 16 of 28 children (57%) with typical Valsalva-related/tussive and mixed headaches, the symptoms resolved before discharge; by 6 months, all children were headache-free. Two of 28 children (7%) had recurrent headaches 9 months after surgery. Among the 78 children with syrinx, syrinx resolved or decreased in 68 (87%), recurred in 8 (10%), and was stable in 2 children (3%). Syrinx was resolved or decreased by 3 months in 51 children (65%) and by 6 months in 62 children (79%). Complications included aseptic meningitis requiring reoperation in 3 children (3%) and infection in one child (1%). Twelve children underwent reoperation, none within the first 30 days. No child had a major morbidity or mortality.
-
CONCLUSIONS: In carefully selected children with CM-I, a high success rate can be achieved with suboccipital decompression and duraplasty. Valsalva-related/tussive headaches resolved by the time of discharge from the hospital in the majority of children, and syrinx resolved or decreased in two-thirds of the children by 3 months. By 6
-
INTRODUCTION
A
variety of techniques for posterior fossa decompression have been described, all associated with favorable outcomes. Although some reports on the resolution of syrinx have been published,1-3 the factors influencing the timely resolution of symptomatic Chiari malformation type I (CM-I) and resolution of syrinx have not been consistently identified. The resolution of headaches over time has not been addressed in a large series of children undergoing decompression with opening of the dura for CM-I. Bony decompression with a suboccipital craniectomy and C1 laminectomy is a standard approach in nearly all series, but the need for durotomy and duraplasty for adequate decompression remains highly controversial. Surveys suggest that 75% of pediatric neurosurgeons routinely open the dura, whereas others perform bony decompression alone or perform durotomy and duraplasty only in the setting of specific clinical, radiographic, or intraoperative factors.4-7 Opponents of dural opening cite significantly increased risks compared with bony decompression alone, including cerebrospinal fluid (CSF) leakage, bacterial meningitis, aseptic meningitis, pseudomeningocele, and hydrocephalus.8-11 Proponents of dural opening contend that duraplasty increases the likelihood of resolving presenting symptoms and improving syringomyelia while concurrently decreasing reoperation rates.12
Key words Chiari malformation type I - Duraplasty - Headache - Outcome - Suboccipital decompression - Syringomyelia
Division of Pediatric Neurosurgery, Department of Neurosurgery, NYU Langone Health, New York, New York, USA
Abbreviations and Acronyms CM-I: Chiari malformation type I CSF: Cerebrospinal fluid MRI: Magnetic resonance imaging
Available online: www.sciencedirect.com
-
WORLD NEUROSURGERY -: e1-e8, - 2018
To whom correspondence should be addressed: Eveline Teresa Hidalgo, M.D. [E-mail: [email protected]] Citation: World Neurosurg. (2018). https://doi.org/10.1016/j.wneu.2018.06.073 Journal homepage: www.WORLDNEUROSURGERY.org 1878-8750/$ - see front matter ª 2018 Elsevier Inc. All rights reserved.
www.WORLDNEUROSURGERY.org
e1
ORIGINAL ARTICLE EVELINE TERESA HIDALGO ET AL.
SUBOCCIPITAL DECOMPRESSION WITH DURAPLASTY IN CHIARI MALFORMATION TYPE I
Here we report the surgical outcomes, with a low incidence of complications, and the consistent resolution of symptoms over time in a consecutive series of children treated for CM-I with dural opening and duraplasty. METHODS Upon approval by New York University Langone Health’s Institutional Review Board (s15-0091), a retrospective analysis was conducted of all pediatric patients who underwent surgical management of CM-I. The children were identified using the Division of Pediatric Neurosurgery’s database; all children treated by the senior author (J.H.W.) between 1999 and 2015 were included. The data were collected through electronic medical records, office charts, operative reports, and magnetic resonance imaging (MRI) results. All children underwent preoperative and postoperative MRI. The following variables were analyzed: clinical history/diagnosis, presenting symptoms, neurologic deficits, presence of syrinx, neuroimaging, number of operations, surgical technique, perioperative complications, resolution of symptoms, and follow-up information. Presenting Symptoms Clinical presentation included headache disorder, neurologic deficits, scoliosis, and serendipitous discovery of Chiari/syrinx on MRI performed for unrelated disease or symptomatology. Headaches were divided into 3 categories: 1) Valsalva-related/tussive headaches, associated with situations that increase intracranial pressure (e.g., coughing, exercise, defecating); 2) mixed headaches, containing both a Valsalva-related component and another type of primary headache (e.g., cluster, migraine, tension), and 3) atypical headaches, those without a Valsalva-related component. Children presenting with neurologic deficits were classified as having gait, balance, motor, sensory, or cranial nerve abnormalities. Serendipitous discovery of CM-I was defined as cases in which brain imaging was obtained for reasons unrelated to the CM-I. These were distinguished from children who were diagnosed as part of a scoliosis evaluation. Surgical Indications Among the children with headaches as the initial presentation, the indication for surgical decompression was Valsalva-related/tussive headaches and mixed headaches. In the children with syringomyelia, the indication for surgery was involvement of at least 3 spinal levels and expansion of the spinal cord >4 mm, irrespective of presenting symptoms. Scoliosis and serendipitous discovery of a Chiari malformation without a syrinx were not an indication for surgery. Surgical Technique All surgeries were performed with the patient in the prone position with the head secured in a Sugita headrest. Somatosensory evoked potentials were monitored continuously. A midline incision was made from the inion to C2, and dissection was performed to expose the occiput and C1. Care was taken to avoid dissecting the muscle and fascia from the superior nuchal line, to permit a watertight closure at the end of the procedure. A midline suboccipital craniectomy including the foramen magnum was
e2
www.SCIENCEDIRECT.com
performed, followed by a C1 laminectomy. In most children, the craniectomy was 4 cm in the horizontal plane and 2e3 cm in longitudinal diameter. In all cases, ultrasound was used to determine the location of the cerebellar tonsils. The dura was opened in a Y fashion cranially, and the caudal margin was opened in a T fashion similar to the technique described by Pirouzmand et al.13 This technique provides greater expansion of the dural opening at the caudal end, preventing an area of constriction when suturing the duraplasty. The cisterna magna was opened, and arachnoidal adhesions were lysed until the foramen of Magendie and obex were identified. Coagulation of the cerebellar tonsils was not performed routinely. At this point, a duraplasty was fashioned out of Durepair (Medtronic, Minneapolis, Minnesota, USA) and sewn into place using running locking sutures of 5-0 PDS. In some cases, the dural suture line was reinforced with a small amount of Bioglue (CryoLife, Kennesaw, Georgia, USA) or fibrin glue. Valsalva maneuver was performed to assess for CSF leakage, and the operative site was irrigated copiously with bacitracin solution. The wound was then closed in a layered fashion with absorbable sutures, including a watertight approximation of the fascia. Children were transferred directly to the pediatric intensive care unit for overnight observation and then transferred to the regular ward. Follow-Up The children had scheduled postoperative follow-up visits at 2 weeks, 3 months, and 6 months postoperatively, and then at subsequent 12-month intervals. In addition, in children with a syrinx at presentation, MRI was obtained at 3-, 6-, 12-, 18-, and 24-month intervals. Children with headaches had follow-up visits until 2 years after surgery, and those with syringomyelia were followed continuously. Surgical outcomes were assessed, including parameters according to the presenting symptoms, change in syrinx size, and postoperative complications. Syrinx “resolution” was defined as a decrease in initial size of >75%; “improvement,” as a decrease of 50%e75%, and “stable,” as a decrease of <50%. Complications Complications were analyzed, including hydrocephalus, aseptic meningitis, CSF leak, pseudomeningocele, wound dehiscence, wound infection, hematoma requiring reoperation, bone defects, neurologic deficits, and any other medical complications related to surgery. Statistical analyses were conducted with SPSS, Vassar Stats, GraphPad, and Social Science Statistics, and P < 0.05 was considered to indicate statistical significance. Chi-squared (2 2) contingency analysis and binomial logistic regression analysis without correction were used to identify the correlation of resolution of headaches and syrinx with clinical factors. RESULTS This series comprised a total of 105 consecutive children who underwent suboccipital decompression with duraplasty for the treatment of symptomatic CM-1. Characteristics The characteristics of the study population are detailed in Table 1. Seven children had undergone previous decompression surgery at
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.06.073
ORIGINAL ARTICLE EVELINE TERESA HIDALGO ET AL.
SUBOCCIPITAL DECOMPRESSION WITH DURAPLASTY IN CHIARI MALFORMATION TYPE I
Table 1. Patient Characteristics and Clinical Presentation (n ¼ 105)
Table 2. Indications for Imaging Leading to Diagnosis of Chiari Malformation Type I
Characteristic
Indication
Age at surgery (years), median
Value 10
Sex, number (%) Male Female
Number of Patients (%)
Primary (imaging obtained for symptomatic patients) Headaches
45 (43)
48 (46)
Valsalva-related/tussive
16 (15)
57 (54)
Mixed headache
11 (10)
Presence of syrinx, number (%)
78 (74)
Atypical
18 (17)
Presence of scoliosis, number (%)
31 (30)
21 (20)
Motor
7 (7)
Mixed
11 (10)
Cranial nerve involvement None Associated genetic syndromes, number (%)
10 (10)
Secondary (imaging obtained for asymptomatic patients)
Neurologic deficit on initial physical examination, number (%) Sensory
Neurologic deficit
3 (3) 63 (60) 8 (8)
Crouzon syndrome
3
Neurofibromatosis-1
2
EhlerseDanlos syndrome
1
RubinsteineTaybi syndrome
1
Achondroplasia
1
outside institutions. Eight children had hydrocephalus before consultation; all underwent shunting procedures before the decompression surgery. The median duration of postoperative follow-up was 33 months (range, 3e272 months), and the mean duration of follow-up was 46 months. Forty-nine children (47%) were followed for >3 years. Of the 78 children with syrinx, 28 (36%) had scoliosis on examination. Of the 45 children with headache at presentation, 4 (13%) had scoliosis on examination; thus, the incidence of concomitant scoliosis was significantly higher in the children with syrinx than in those with headache (P ¼ 0.01). Similarly, in the 31 children with scoliosis, the incidence of syrinx on imaging was significantly higher than the incidence of headaches on presentation (90% vs. 13%; P ¼ 0.0003).
Serendipitous
28 (27)
Scoliosis workup
22 (21)
children (40%), the headaches resolved by 3 months. In 6 children (14%), headaches resolved by 6 months, and in 1 child (2%), headaches resolved by 9 months (Figure 1). The mean and median times to headache resolution were 2.3 and 3 months, respectively. Comparing the children with tussive headaches and those with atypical headaches, the former cohort exhibited significantly faster resolution of headaches in the immediate postoperative period (P ¼ 0.039); 57% of children with isolated tussive headaches experienced resolution of pain before the time of discharge, compared with 17% of children with atypical headaches (e.g., migraine, chronic headache disorder, tension). Seven of the 42 children (17%) had partial resolution of headaches, all of whom had mixed headaches. In these children, the Valsalva-related/tussive headaches resolved, but all continued to experience forms of their preexisting primary headache. Fourteen children exhibited postoperative clinical signs of
Presentation The indications for the imaging leading to the diagnosis of CM-1 are presented in Table 2. The indications for surgical intervention included pure Valsalva-related headaches, mixed headache disorders with a Valsalva-related headache component, and syringomyelia. No child underwent surgery for atypical headaches alone; all 18 of these children also had syringomyelia. Clinical Outcomes Headaches. Of the 45 children who experienced preoperative headaches, 42 children (93%) experienced partial or complete long-term resolution of symptoms. In 18 children (43%), headaches resolved by the time of discharge from the hospital and in 17
WORLD NEUROSURGERY -: e1-e8, - 2018
Figure 1. Proportion of patients with 50%e75% radiographic improvement of syrinx (green) or resolution of syrinx (black dotted) according to time after surgery.
www.WORLDNEUROSURGERY.org
e3
ORIGINAL ARTICLE EVELINE TERESA HIDALGO ET AL.
SUBOCCIPITAL DECOMPRESSION WITH DURAPLASTY IN CHIARI MALFORMATION TYPE I
mild meningeal irritation with neck pain and low-grade fever. These children were treated with oral steroids and nonsteroidal anti-inflammatory drugs; their hospital stay was not significantly different than that of children without postoperative meningeal irritation. Three children experienced recurrence of their headaches after initial postoperative headache free periods. One was successfully treated with antimigraine medication, and 1 underwent successful reexploration of the suboccipital decompression. The third child experienced return of tussive headaches at 1 year after decompression. This 17-year-old child with preoperative suboccipital headaches exacerbated with Valsalva maneuvers experienced complete relief of his headaches after decompression. After 2e3 years, the tussive headaches returned, limiting his athletic activities. Reexploration with C1 laminectomy, partial C2 laminectomy, and microsurgical intradural exploration with coagulation of the right cerebellar tonsil and fenestration of the foramen of Magendie was performed. He again experienced complete relief immediately after the operation but returned 6 months later with tussive headaches and magnetic resonance imaging showing progressive basilar invagination. He underwent a successful transnasal odontoidectomy and occipitocervical fusion. At the most recent follow-up (12 months after fusion), he reported no headaches and had returned to his baseline activity. This is the only child in our series who required fusion and/or an anterior decompression. Syrinx. Of the 78 children with syrinx, 68 (87%) experienced resolution or significant decrease over time after a single operation. The median time to syrinx improvement was 3 months (range, 3e72 months), and the mean time was 4.8 months. Syrinx size improved in 65% of children in the first 3 months and in 79% in the first 6 months (Figure 2). There was a statistically significant negative correlation between the presence of an associated genetic syndrome (Crouzon syndrome, neurofibromatosis type I, EhlerseDanlos syndrome, RubinsteineTaybi syndrome, or achondroplasia) and syrinx resolution in the first 3 months after surgery (18% vs. 55%; P ¼ 0.03).
Figure 2. Proportion of patients with clinical resolution of Valsalva-related/tussive headaches (red) and all types of headaches (blue dotted) according to time after surgery.
e4
www.SCIENCEDIRECT.com
Similarly, a statistically significant negative correlation was found in children with previous surgeries and syrinx resolution at 1- and 2-year follow-ups (P ¼ 0.01). No statistically significant correlations were found between age at the time of surgery, sex, presence of scoliosis, focal neurologic deficit at diagnosis, presence of a shunt before surgery, and chemical meningitis postoperatively. Eight children (10%) had recurrent syrinx, 7 of whom underwent reoperation. Among the children who underwent reoperation for recurrent syrinx, the average time to reoperation was 37.4 months. After reoperation, 5 of these children experienced complete resolution of syrinx, 1 child showed a significant decrease on follow-up imaging, and 1 child with Crouzon syndrome had an initial decrease in syrinx size but subsequent reexpansion 7 months later (Table 3). Two children (3%) showed evidence of a stable syrinx, similar in size to that seen on preoperative imaging and remained asymptomatic throughout follow-up. At last follow-up, 95% of the children with syrinx at presentation had decreased or resolved syrinx, 2 children had small, stable syrinxes and 1 child had stable disease. Complications Among the 105 children, there was no permanent morbidity or mortality. The complication rate was 3.8%, including 3 cases of clinically significant aseptic meningitis requiring removal of the allogenic duraplasty, and one Serratia wound infection with subsequent meningitis. The 3 cases of aseptic meningitis requiring reoperation occurred at 1 month, 2 months, and 3 months after the first surgery (Table 3). The child with a Serratia wound infection and meningitis developed the infection 2 weeks postoperatively after swimming in the ocean 1 week after surgery against medical advice. He was successfully treated with antibiotic therapy and did not require surgical intervention. At the time of this report, he is nearly 12 years out of surgery, has completed college in the interim, and is working full-time with no symptoms. Reoperations Twelve of the 105 children (11%) required reoperation, with a median interval between operations of 8.9 months (range, 1.3e77.9 months) (Table 3). Indications for reoperation included recurrent syrinx (symptomatic and/or 3 level disease), aseptic meningitis, and recurrent Valsalva-related headache. In the 7 children with recurrent syrinx, reoperation included repeat craniotomy, dissection of the fibrous band, and dissection of arachnoid adhesions, depending on the presentation. If there were no visible arachnoid adhesions or if there was a lack of spontaneous pulsations after arachnoid dissection, a syringo-subarachnoid or fourth ventricle to subarachnoid shunt was placed. In 6 out of 7 children, a >75% improvement in syrinx size was achieved with this approach. Children with prolonged aseptic meningitis were treated in a uniform manner, with removal of the Durepair graft and placement of new autologous pericranial duraplasty. In these children, Bioglue was used in all cases during the initial operation. Fisher exact test showed no significant correlation between the use of Bioglue and aseptic meningitis (P ¼ 1.64). Partial C2 laminectomy and coagulation of cerebellar tonsils was used to treat children with recurrent tussive headache, and both children experienced complete symptom resolution after surgery.
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.06.073
Age at Surgery Patient (Years)
Initial Presentation
Presence of Syrinx
Interval Between Operations (Months)
Intraoperative Findings
Surgery
Evolution of Syrinx
Outcome
Follow-Up (Months)
Collapse
Asymptomatic
87
Recurrent syrinx 6
Serendipitous
Yes
78
Posterior arch of C1 completely reossified and fibrocalcific dural scarring at the foramen magnum causing compression
2
6
Serendipitous
Yes
7
Fibrous band compressing the dural graft at Dissection of fibrous band, placement of the foramen magnum; no spontaneous fourth ventricle to subarachnoid shunt pulsation of the dura; no significant arachnoidal scarring; foramen of Magendie open
Collapse
Asymptomatic
37
3
4
Scoliosis workup
Yes
16
No reossification; capacious subarachnoid space visible after opening the duraplasty; foramen of Magendie obstructed by arachnoidal scarring
Collapse
Scoliosis improved
36
4
4
Valsalva- related/ tussive headaches
Yes
21
Reexploration suboccipital craniectomy; Epidural pseudomeningocele with fibrotic microsurgical dissection of arachnoidal walls causing compression; foramen of Magendie obstructed by arachnoidal scarring adhesions
Decrease Headache-free
48
5
11
Scoliosis workup
Yes
9
Foramen of Magendie obstructed by arachnoidal scarring
Collapse
Asymptomatic
96
6
6
Serendipitous
Yes
4
Partial reossification of craniectomy; dural Decrease Expansion of suboccipital craniectomy, graft markedly fibrotic and compressing the partial C2 laminectomy, Durepair duraplasty; cervicomedullary junction. T6 laminectomy and placement of syringosubarachnoidal shunt
Unknown
7
7
13
Gait instability and focal neurologic deficit
Yes
127
Collapse
Asymptomatic
272
Markedly fibrotic dural graft and reduced subarachnoid space; arachnoidal scarring
Laminectomy C1 and expansion of reossified craniectomy; microsurgical dissection of arachnoidal adhesions until exposure of foramen of Magendie
Microsurgical dissection of arachnoidal adhesions until exposure of foramen of Magendie, T5 laminectomy, and insertion of syringo-subarachnoidal shunt
Microsurgical dissection of arachnoidal adhesions until exposure of foramen of Magendie; placement of fourth ventriclesubarachnoidal shunt
Microsurgical dissection of arachnoidal adhesions until exposure of foramen of Magendie, coagulation of tonsils
Prolonged aseptic meningitis 9
Scoliosis workup
Yes
2
3-mm perforation in the dural graft, epidural Removal of Durepair graft and placement of pseudomeningocele new autologous pericranial duraplasty
Collapse
Headache-free
43
9
8
Mixed headache
Yes
3
Small dural opening, epidural pseudomeningocele
Removal of Durepair graft and placement of new autologous pericranial duraplasty
Collapse
Headache-free
51
10
16
Mixed headache
No
1
Small epidural pseudomeningocele
Removal of Durepair graft and placement of new autologous pericranial duraplasty
N/A
Headache-free
47
Recurrent Valsalva-related/tussive headache N/A, not applicable.
e5
Continues
ORIGINAL ARTICLE
8
SUBOCCIPITAL DECOMPRESSION WITH DURAPLASTY IN CHIARI MALFORMATION TYPE I
www.WORLDNEUROSURGERY.org
1
EVELINE TERESA HIDALGO ET AL.
WORLD NEUROSURGERY -: e1-e8, - 2018
Table 3. Characteristics and Outcomes of 12 Patients Requiring Reoperation
ORIGINAL ARTICLE
e6
SUBOCCIPITAL DECOMPRESSION WITH DURAPLASTY IN CHIARI MALFORMATION TYPE I
Thick band of fibrous scar tissue at foramen Partial C2 laminectomy, coagulation of both magnum and C1 compressing the underlying cerebellar tonsils duraplasty
Decrease Headache-free
12
In our study, we report the outcomes of 105 consecutive children treated with a standardized surgical technique including duraplasty, which proved to be efficient in resolving headaches and syrinx with no permanent morbidity or mortality. The majority of headaches resolved by 3 months, and syrinx improvement was noted by 6 months. We believe that our narrow window of surgical indications—specifically, Valsalva-related headaches and syringomyelia with involvement of at least 3 spinal levels and expansion of the spinal cord by >4 mm—contributed to the favorable outcomes. The surgical indication for asymptomatic large syringomyelias in young children is based on the unknown natural history and is partially predicated on low rates of new neurologic injury, infection, and CSF leakage. Resolution of Symptoms Over Time Headaches. Valsalva-related headaches completely resolved in 93% of the children within the first year after surgery. Fifty-seven percent of the children experienced complete relief before discharge from the hospital. In our subanalysis, children with classic Valsalva-related/tussive headaches had a significantly faster resolution of symptoms than children with atypical headaches. In a study of 500 children with dural opening reported by Tubbs et al.,10 the overall resolution rate of preoperative signs and symptoms was 83%, similar to that our series. In their series of 156 children without dural opening, Kennedy et al. found that 62% of children who presented with occipital and/or tussive headache had complete resolution of headache; when including asymptomatic and improved but minimally symptomatic children, 91% of children were considered therapeutic successes.11 In a large retrospective review of 256 children treated with suboccipital decompression, McGirt et al.5 found a recurrence rate of 22%, and noted that headaches were 70% more likely to recur than cranial nerve and brainstem symptoms. We did not find such a high recurrence rate; in our series, only 3 of 45 children (7%) with headaches had a recurrence, with a mean follow-up time of 46 months. In contrast to that study, our children were treated in a uniform manner with dural opening and stringent criteria for the headaches that we believed were Chiari-related and appropriate for surgical intervention. Children with atypical headaches underwent surgery because of their coexisting syrinx detected on imaging, not because of their headache etiology. Although we did not expect them to have headache improvement, we did discover some improvement that might possibly be attributed to medical management of a coincidental non-Chiari headache disorder. The results of this study may reflect a narrower definition of headaches that are appropriate for surgical management and achieve rapid and long-lasting resolution of symptoms.
www.SCIENCEDIRECT.com
N/A, not applicable.
9 Yes Valsalva- related/ tussive headache 20 12
No 17
Valsalva- related/ tussive headache
37
Thick band of scar tissue at the foramen Partial C2 laminectomy, coagulation of right magnum extending to C2; cerebellar tonsil on cerebellar tonsil the right extended to the upper border of C2
N/A
Headache-free
24
DISCUSSION
11
Age at Surgery Patient (Years)
Table 3. Continued
Initial Presentation
Presence of Syrinx
Interval Between Operations (Months)
Intraoperative Findings
Surgery
Evolution of Syrinx
Outcome
Follow-Up (Months)
EVELINE TERESA HIDALGO ET AL.
Syrinx. In this study, we found that children with an associated genetic syndrome or previous surgery had a statistically significant delay in resolution of syrinx compared with children without these conditions. This may be helpful in clinical practice when evaluating the radiologic outcomes of these children. No other clinical factor was correlated with timing of syrinx resolution. We observed an improvement or resolution of syrinx in 95% of our children, with a mean time to resolution of 8.1 months. These findings are similar to those reported in previous surgical series of
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.06.073
ORIGINAL ARTICLE EVELINE TERESA HIDALGO ET AL.
SUBOCCIPITAL DECOMPRESSION WITH DURAPLASTY IN CHIARI MALFORMATION TYPE I
decompression and duraplasty. Wu et al.14 reported syrinx improvement in 82% of 44 children with CM-I undergoing dural opening by 6 months postoperatively and in 98% of children over the course of follow-up. As in our study, Yeh et al.15 reported a notable improvement in 17 of 20 (85%) children after posterior fossa decompression with dural opening and tonsillar shrinkage. We found a median time to syrinx improvement of 3 months, with >79% of cases improving in the first 6 months after surgery. This is comparable to the results of a prospective study reported by Wetjen et al.,1 with a median time to syrinx narrowing of 3.6 months following dural opening. We believe that if there is no evidence of improvement in syrinx size within 12 months, reoperation should be considered, because the majority of children improve in this time period (Figure 2). Our present series and the previously mentioned reports in the literature suggest a more robust improvement in syringomyelia over a shorter period when duraplasty is performed. In a series of children with CM-I without dural opening, Kennedy et al.11 reported that 70% of children experienced radiologic improvement of syrinx, 23% had a stable syrinx, and 7% had new a syrinx or growth of an old syrinx.11 For children with demonstrated radiographic improvement, the median and mean times to improvement were 21 and 31 months, respectively.2 The other technical aspect that has been advocated at the initial surgery is extensive coagulation of the herniated cerebellar tonsils. Stanko et al.3 compared to syrinx improvement in children with tonsillar cautery and those without cautery.3 They found that 81% of the children with tonsillar coagulation had improvement of syrinx after a mean of 8.0 months, compared with 13.1 months in children without tonsillar coagulation. In their series, children who underwent tonsillar cautery had 6-fold greater odds of improvement in syrinx. In our series of duraplasty without significant tonsillar cauterization, the rate of improvement in syrinx was similar if not better. This raises the question of whether coagulation of the tonsils with the low attendant risk of parenchymal or vascular injury should be avoided at the initial operation. Complications The overall complication rate in our series was 3.8%. We believe this low complication rate is due to the absence of wound healing problems, which is a result of our modified surgical technique. In a meta-analysis conducted by Durham et al.,11 CSF-related complications were described in 18.5% of cases with dural opening, compared with 1.8% of cases without dural opening.12 The absence of CSF leak may be due to the technique of dural closure with a Durepair patch, which was secured with a
REFERENCES 1. Wetjen NM, Heiss JD, Oldfield EH. Time course of syringomyelia resolution following decompression of Chiari malformation type I. J Neurosurg Pediatr. 2008;1:118-123. 2. Kennedy BC, Nelp TB, Kelly KM, Phan MQ, Bruce SS, McDowell MM, et al. Delayed resolution of syrinx after posterior fossa decompression without dural opening in children with Chiari malformation type I. J Neurosurg Pediatr. 2015;16: 599-606.
running, locking absorbable suture, in many cases along with a dural sealant (Bioglue) or fibrin glue. Parker et al.16 investigated the associations between complications and the use of duraplasty and dural sealant. Although their sample was quite small (n ¼ 15), they noted an overall complication rate of 42%, with the most common complication, aseptic meningitis, seen in 21% of their children.16 They also reported CSF leak in 8.4% of their children, pseudomeningocele in 3.9%, and infection in 3%. Although 3 children in our series developed aseptic meningitis that required reoperation, we did not find any statistically significant difference between the type of sealant and the rate of complications. The same type of allogenic Duragraft was used in all cases. Our finding that all 3 children with aseptic meningitis had subclinical CSF leakage and dehiscence of the dural graft at reoperation, along with the lack of statistical correlation between sealant use and aseptic meningitis, suggests that the sealant alone might not be the cause of the inflammatory response. Reoperation Rate Lower rates of reoperation have been reported in children who undergo posterior fossa decompression with duraplasty, ranging from 2.1% to 3.1%, compared with the 9%e12.6% in children without duraplasty.11,12,17 In our series with duraplasty, reoperation was necessary in 11% of children. This higher reoperation rate may be due to a longer duration of follow-up; 5 of the 12 children requiring reoperation had a late recurrence of symptoms or syrinx (range, 16e127 months). Of note, 2 of those children underwent previous surgery at outside institutions. Limitations of this study are its retrospective nature and potentially missing data on late recurrence, as in this historical cohort some children may be lost to follow-up. CONCLUSIONS Chiari decompression with duraplasty is an efficacious procedure in carefully selected children. This study confirms that this technique rapidly resolves both classic tussive headaches and syringomyelia, at average intervals of 3 and 6 months, respectively, and provides durable remission for these children. Further prospective studies are needed to identify the subgroup of children at risk for recurrence or delayed resolution of symptoms. ACKNOWLEDGMENTS The authors thank Christopher Hernandez for his collaboration in data collection.
3. Stanko KM, Lee YM, Rios J, Wu A, Sobrinho GW, Weingart JD, et al. Improvement of syrinx resolution after tonsillar cautery in pediatric children with Chiari type I malformation. J Neurosurg Pediatr. 2016;17:174-181.
5. McGirt MJ, Attenello FJ, Datoo G, Gathinji M, Atiba A, Weingart JD, et al. Intraoperative ultrasonography as a guide to patient selection for duraplasty after suboccipital decompression in children with Chiari malformation type I. J Neurosurg Pediatr. 2008;2:52-57.
4. Attenello FJ, McGirt MJ, Gathinji M, Datoo G, Atiba A, Weingart J, et al. Outcome of Chiariassociated syringomyelia after hindbrain decompression in children: analysis of 49 consecutive cases. Neurosurgery. 2008;62:1307-1313 [discussion: 1313].
6. Haroun RI, Guarnieri M, Meadow JJ, Kraut M, Carson BS. Current opinions for the treatment of syringomyelia and chiari malformations: survey of the Pediatric Section of the American Association of Neurological Surgeons. Pediatr Neurosurg. 2000; 33:311-317.
WORLD NEUROSURGERY -: e1-e8, - 2018
www.WORLDNEUROSURGERY.org
e7
ORIGINAL ARTICLE EVELINE TERESA HIDALGO ET AL.
SUBOCCIPITAL DECOMPRESSION WITH DURAPLASTY IN CHIARI MALFORMATION TYPE I
7. Rocque BG, George TM, Kestle J, Iskandar BJ. Treatment practices for Chiari malformation type I with syringomyelia: results of a survey of the American Society of Pediatric Neurosurgeons. J Neurosurg Pediatr. 2011;8:430-437.
12. Durham SR, Fjeld-Olenec K. Comparison of posterior fossa decompression with and without duraplasty for the surgical treatment of Chiari malformation type I in pediatric children: a metaanalysis. J Neurosurg Pediatr. 2008;2:42-49.
8. Hankinson T, Tubbs RS, Wellons JC. Duraplasty or not? An evidence-based review of the pediatric Chiari I malformation. Childs Nerv Syst. 2011;27:35-40.
13. Pirouzmand F, Tucker WS. A modification of the classic technique for expansion duroplasty of the posterior fossa. Neurosurgery. 2007;60(2 Suppl 1): ONS60-ONS62 [discussion: ONS62].
9. Galarza M, Sood S, Ham S. Relevance of surgical strategies for the management of pediatric Chiari type I malformation. Childs Nerv Syst. 2007;23:691-696. 10. Tubbs RS, Beckman J, Naftel RP, Chern JJ, Wellons JC 3rd, Rozzelle CJ, et al. Institutional experience with 500 cases of surgically treated pediatric Chiari malformation type I. J Neurosurg Pediatr. 2011;7:248-256. 11. Kennedy BC, Kelly KM, Phan MQ, Bruce SS, McDowell MM, Anderson RC, et al. Outcomes after suboccipital decompression without dural opening in children with Chiari malformation type I. J Neurosurg Pediatr. 2015;16:150-158.
e8
www.SCIENCEDIRECT.com
14. Wu T, Zhu Z, Jiang J, Zheng X, Sun X, Qian B, et al. Syrinx resolution after posterior fossa decompression in patients with scoliosis secondary to Chiari malformation type I. Eur Spine J. 2012; 21:1143-1150. 15. Yeh DD, Koch B, Crone KR. Intraoperative ultrasonography used to determine the extent of surgery necessary during posterior fossa decompression in children with Chiari malformation type I. J Neurosurg. 2006;105(1 Suppl):26-32.
children: dural graft or sealant? J Neurosurg Pediatr. 2011;8:177-183. 17. Mutchnick IS, Janjua RM, Moeller K, Moriarty TM. Decompression of Chiari malformation with and without duraplasty: morbidity versus recurrence. J Neurosurg Pediatr. 2010;5:474-478.
Conflict of interest statement: This publication is supported in part by the NYU CTSA grant UL1TR001445 from the National Center for Advancing Translational Sciences, National Institutes of Health. Received 31 March 2018; accepted 11 June 2018 Citation: World Neurosurg. (2018). https://doi.org/10.1016/j.wneu.2018.06.073 Journal homepage: www.WORLDNEUROSURGERY.org Available online: www.sciencedirect.com
16. Parker SR, Harris P, Cummings TJ, George T, Fuchs H, Grant G. Complications following decompression of Chiari malformation type I in
1878-8750/$ - see front matter ª 2018 Elsevier Inc. All rights reserved.
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.06.073
Time to Resolution of Symptoms After Suboccipital Decompression with Duraplasty in Children with Chiari Malformation Type I Eveline Teresa Hidalgo, Yosef Dastagirzada, Cordelia Orillac, Svetlana Kvint, Emily North, Ramona Bledea, Michelle W. McQuinn, Gabriel Redel-Traub, Crystalann Rodriguez, Jeffrey H. Wisoff
BACKGROUND: Duraplasty is a technique successfully used to treat Chiari malformation type I (CM-I). This study describes the timely manner of clinical outcomes and the postoperative course after craniectomy and duraplasty for the treatment of symptomatic CM-I in children.
-
METHODS: A retrospective chart review was performed in 105 consecutive children who underwent surgical decompression of symptomatic CM-I with dural opening by a single surgeon between 1999 and 2015.
months, headaches were resolved in all cases, and syrinx was resolved or decreased in 79% of cases.
-
RESULTS: In 16 of 28 children (57%) with typical Valsalva-related/tussive and mixed headaches, the symptoms resolved before discharge; by 6 months, all children were headache-free. Two of 28 children (7%) had recurrent headaches 9 months after surgery. Among the 78 children with syrinx, syrinx resolved or decreased in 68 (87%), recurred in 8 (10%), and was stable in 2 children (3%). Syrinx was resolved or decreased by 3 months in 51 children (65%) and by 6 months in 62 children (79%). Complications included aseptic meningitis requiring reoperation in 3 children (3%) and infection in one child (1%). Twelve children underwent reoperation, none within the first 30 days. No child had a major morbidity or mortality.
-
CONCLUSIONS: In carefully selected children with CM-I, a high success rate can be achieved with suboccipital decompression and duraplasty. Valsalva-related/tussive headaches resolved by the time of discharge from the hospital in the majority of children, and syrinx resolved or decreased in two-thirds of the children by 3 months. By 6
-
INTRODUCTION
A
variety of techniques for posterior fossa decompression have been described, all associated with favorable outcomes. Although some reports on the resolution of syrinx have been published,1-3 the factors influencing the timely resolution of symptomatic Chiari malformation type I (CM-I) and resolution of syrinx have not been consistently identified. The resolution of headaches over time has not been addressed in a large series of children undergoing decompression with opening of the dura for CM-I. Bony decompression with a suboccipital craniectomy and C1 laminectomy is a standard approach in nearly all series, but the need for durotomy and duraplasty for adequate decompression remains highly controversial. Surveys suggest that 75% of pediatric neurosurgeons routinely open the dura, whereas others perform bony decompression alone or perform durotomy and duraplasty only in the setting of specific clinical, radiographic, or intraoperative factors.4-7 Opponents of dural opening cite significantly increased risks compared with bony decompression alone, including cerebrospinal fluid (CSF) leakage, bacterial meningitis, aseptic meningitis, pseudomeningocele, and hydrocephalus.8-11 Proponents of dural opening contend that duraplasty increases the likelihood of resolving presenting symptoms and improving syringomyelia while concurrently decreasing reoperation rates.12
Key words Chiari malformation type I - Duraplasty - Headache - Outcome - Suboccipital decompression - Syringomyelia
Division of Pediatric Neurosurgery, Department of Neurosurgery, NYU Langone Health, New York, New York, USA
Abbreviations and Acronyms CM-I: Chiari malformation type I CSF: Cerebrospinal fluid MRI: Magnetic resonance imaging
Available online: www.sciencedirect.com
-
WORLD NEUROSURGERY -: e1-e8, - 2018
To whom correspondence should be addressed: Eveline Teresa Hidalgo, M.D. [E-mail: [email protected]] Citation: World Neurosurg. (2018). https://doi.org/10.1016/j.wneu.2018.06.073 Journal homepage: www.WORLDNEUROSURGERY.org 1878-8750/$ - see front matter ª 2018 Elsevier Inc. All rights reserved.
www.WORLDNEUROSURGERY.org
e1
ORIGINAL ARTICLE EVELINE TERESA HIDALGO ET AL.
SUBOCCIPITAL DECOMPRESSION WITH DURAPLASTY IN CHIARI MALFORMATION TYPE I
Here we report the surgical outcomes, with a low incidence of complications, and the consistent resolution of symptoms over time in a consecutive series of children treated for CM-I with dural opening and duraplasty. METHODS Upon approval by New York University Langone Health’s Institutional Review Board (s15-0091), a retrospective analysis was conducted of all pediatric patients who underwent surgical management of CM-I. The children were identified using the Division of Pediatric Neurosurgery’s database; all children treated by the senior author (J.H.W.) between 1999 and 2015 were included. The data were collected through electronic medical records, office charts, operative reports, and magnetic resonance imaging (MRI) results. All children underwent preoperative and postoperative MRI. The following variables were analyzed: clinical history/diagnosis, presenting symptoms, neurologic deficits, presence of syrinx, neuroimaging, number of operations, surgical technique, perioperative complications, resolution of symptoms, and follow-up information. Presenting Symptoms Clinical presentation included headache disorder, neurologic deficits, scoliosis, and serendipitous discovery of Chiari/syrinx on MRI performed for unrelated disease or symptomatology. Headaches were divided into 3 categories: 1) Valsalva-related/tussive headaches, associated with situations that increase intracranial pressure (e.g., coughing, exercise, defecating); 2) mixed headaches, containing both a Valsalva-related component and another type of primary headache (e.g., cluster, migraine, tension), and 3) atypical headaches, those without a Valsalva-related component. Children presenting with neurologic deficits were classified as having gait, balance, motor, sensory, or cranial nerve abnormalities. Serendipitous discovery of CM-I was defined as cases in which brain imaging was obtained for reasons unrelated to the CM-I. These were distinguished from children who were diagnosed as part of a scoliosis evaluation. Surgical Indications Among the children with headaches as the initial presentation, the indication for surgical decompression was Valsalva-related/tussive headaches and mixed headaches. In the children with syringomyelia, the indication for surgery was involvement of at least 3 spinal levels and expansion of the spinal cord >4 mm, irrespective of presenting symptoms. Scoliosis and serendipitous discovery of a Chiari malformation without a syrinx were not an indication for surgery. Surgical Technique All surgeries were performed with the patient in the prone position with the head secured in a Sugita headrest. Somatosensory evoked potentials were monitored continuously. A midline incision was made from the inion to C2, and dissection was performed to expose the occiput and C1. Care was taken to avoid dissecting the muscle and fascia from the superior nuchal line, to permit a watertight closure at the end of the procedure. A midline suboccipital craniectomy including the foramen magnum was
e2
www.SCIENCEDIRECT.com
performed, followed by a C1 laminectomy. In most children, the craniectomy was 4 cm in the horizontal plane and 2e3 cm in longitudinal diameter. In all cases, ultrasound was used to determine the location of the cerebellar tonsils. The dura was opened in a Y fashion cranially, and the caudal margin was opened in a T fashion similar to the technique described by Pirouzmand et al.13 This technique provides greater expansion of the dural opening at the caudal end, preventing an area of constriction when suturing the duraplasty. The cisterna magna was opened, and arachnoidal adhesions were lysed until the foramen of Magendie and obex were identified. Coagulation of the cerebellar tonsils was not performed routinely. At this point, a duraplasty was fashioned out of Durepair (Medtronic, Minneapolis, Minnesota, USA) and sewn into place using running locking sutures of 5-0 PDS. In some cases, the dural suture line was reinforced with a small amount of Bioglue (CryoLife, Kennesaw, Georgia, USA) or fibrin glue. Valsalva maneuver was performed to assess for CSF leakage, and the operative site was irrigated copiously with bacitracin solution. The wound was then closed in a layered fashion with absorbable sutures, including a watertight approximation of the fascia. Children were transferred directly to the pediatric intensive care unit for overnight observation and then transferred to the regular ward. Follow-Up The children had scheduled postoperative follow-up visits at 2 weeks, 3 months, and 6 months postoperatively, and then at subsequent 12-month intervals. In addition, in children with a syrinx at presentation, MRI was obtained at 3-, 6-, 12-, 18-, and 24-month intervals. Children with headaches had follow-up visits until 2 years after surgery, and those with syringomyelia were followed continuously. Surgical outcomes were assessed, including parameters according to the presenting symptoms, change in syrinx size, and postoperative complications. Syrinx “resolution” was defined as a decrease in initial size of >75%; “improvement,” as a decrease of 50%e75%, and “stable,” as a decrease of <50%. Complications Complications were analyzed, including hydrocephalus, aseptic meningitis, CSF leak, pseudomeningocele, wound dehiscence, wound infection, hematoma requiring reoperation, bone defects, neurologic deficits, and any other medical complications related to surgery. Statistical analyses were conducted with SPSS, Vassar Stats, GraphPad, and Social Science Statistics, and P < 0.05 was considered to indicate statistical significance. Chi-squared (2 2) contingency analysis and binomial logistic regression analysis without correction were used to identify the correlation of resolution of headaches and syrinx with clinical factors. RESULTS This series comprised a total of 105 consecutive children who underwent suboccipital decompression with duraplasty for the treatment of symptomatic CM-1. Characteristics The characteristics of the study population are detailed in Table 1. Seven children had undergone previous decompression surgery at
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.06.073
ORIGINAL ARTICLE EVELINE TERESA HIDALGO ET AL.
SUBOCCIPITAL DECOMPRESSION WITH DURAPLASTY IN CHIARI MALFORMATION TYPE I
Table 1. Patient Characteristics and Clinical Presentation (n ¼ 105)
Table 2. Indications for Imaging Leading to Diagnosis of Chiari Malformation Type I
Characteristic
Indication
Age at surgery (years), median
Value 10
Sex, number (%) Male Female
Number of Patients (%)
Primary (imaging obtained for symptomatic patients) Headaches
45 (43)
48 (46)
Valsalva-related/tussive
16 (15)
57 (54)
Mixed headache
11 (10)
Presence of syrinx, number (%)
78 (74)
Atypical
18 (17)
Presence of scoliosis, number (%)
31 (30)
21 (20)
Motor
7 (7)
Mixed
11 (10)
Cranial nerve involvement None Associated genetic syndromes, number (%)
10 (10)
Secondary (imaging obtained for asymptomatic patients)
Neurologic deficit on initial physical examination, number (%) Sensory
Neurologic deficit
3 (3) 63 (60) 8 (8)
Crouzon syndrome
3
Neurofibromatosis-1
2
EhlerseDanlos syndrome
1
RubinsteineTaybi syndrome
1
Achondroplasia
1
outside institutions. Eight children had hydrocephalus before consultation; all underwent shunting procedures before the decompression surgery. The median duration of postoperative follow-up was 33 months (range, 3e272 months), and the mean duration of follow-up was 46 months. Forty-nine children (47%) were followed for >3 years. Of the 78 children with syrinx, 28 (36%) had scoliosis on examination. Of the 45 children with headache at presentation, 4 (13%) had scoliosis on examination; thus, the incidence of concomitant scoliosis was significantly higher in the children with syrinx than in those with headache (P ¼ 0.01). Similarly, in the 31 children with scoliosis, the incidence of syrinx on imaging was significantly higher than the incidence of headaches on presentation (90% vs. 13%; P ¼ 0.0003).
Serendipitous
28 (27)
Scoliosis workup
22 (21)
children (40%), the headaches resolved by 3 months. In 6 children (14%), headaches resolved by 6 months, and in 1 child (2%), headaches resolved by 9 months (Figure 1). The mean and median times to headache resolution were 2.3 and 3 months, respectively. Comparing the children with tussive headaches and those with atypical headaches, the former cohort exhibited significantly faster resolution of headaches in the immediate postoperative period (P ¼ 0.039); 57% of children with isolated tussive headaches experienced resolution of pain before the time of discharge, compared with 17% of children with atypical headaches (e.g., migraine, chronic headache disorder, tension). Seven of the 42 children (17%) had partial resolution of headaches, all of whom had mixed headaches. In these children, the Valsalva-related/tussive headaches resolved, but all continued to experience forms of their preexisting primary headache. Fourteen children exhibited postoperative clinical signs of
Presentation The indications for the imaging leading to the diagnosis of CM-1 are presented in Table 2. The indications for surgical intervention included pure Valsalva-related headaches, mixed headache disorders with a Valsalva-related headache component, and syringomyelia. No child underwent surgery for atypical headaches alone; all 18 of these children also had syringomyelia. Clinical Outcomes Headaches. Of the 45 children who experienced preoperative headaches, 42 children (93%) experienced partial or complete long-term resolution of symptoms. In 18 children (43%), headaches resolved by the time of discharge from the hospital and in 17
WORLD NEUROSURGERY -: e1-e8, - 2018
Figure 1. Proportion of patients with 50%e75% radiographic improvement of syrinx (green) or resolution of syrinx (black dotted) according to time after surgery.
www.WORLDNEUROSURGERY.org
e3
ORIGINAL ARTICLE EVELINE TERESA HIDALGO ET AL.
SUBOCCIPITAL DECOMPRESSION WITH DURAPLASTY IN CHIARI MALFORMATION TYPE I
mild meningeal irritation with neck pain and low-grade fever. These children were treated with oral steroids and nonsteroidal anti-inflammatory drugs; their hospital stay was not significantly different than that of children without postoperative meningeal irritation. Three children experienced recurrence of their headaches after initial postoperative headache free periods. One was successfully treated with antimigraine medication, and 1 underwent successful reexploration of the suboccipital decompression. The third child experienced return of tussive headaches at 1 year after decompression. This 17-year-old child with preoperative suboccipital headaches exacerbated with Valsalva maneuvers experienced complete relief of his headaches after decompression. After 2e3 years, the tussive headaches returned, limiting his athletic activities. Reexploration with C1 laminectomy, partial C2 laminectomy, and microsurgical intradural exploration with coagulation of the right cerebellar tonsil and fenestration of the foramen of Magendie was performed. He again experienced complete relief immediately after the operation but returned 6 months later with tussive headaches and magnetic resonance imaging showing progressive basilar invagination. He underwent a successful transnasal odontoidectomy and occipitocervical fusion. At the most recent follow-up (12 months after fusion), he reported no headaches and had returned to his baseline activity. This is the only child in our series who required fusion and/or an anterior decompression. Syrinx. Of the 78 children with syrinx, 68 (87%) experienced resolution or significant decrease over time after a single operation. The median time to syrinx improvement was 3 months (range, 3e72 months), and the mean time was 4.8 months. Syrinx size improved in 65% of children in the first 3 months and in 79% in the first 6 months (Figure 2). There was a statistically significant negative correlation between the presence of an associated genetic syndrome (Crouzon syndrome, neurofibromatosis type I, EhlerseDanlos syndrome, RubinsteineTaybi syndrome, or achondroplasia) and syrinx resolution in the first 3 months after surgery (18% vs. 55%; P ¼ 0.03).
Figure 2. Proportion of patients with clinical resolution of Valsalva-related/tussive headaches (red) and all types of headaches (blue dotted) according to time after surgery.
e4
www.SCIENCEDIRECT.com
Similarly, a statistically significant negative correlation was found in children with previous surgeries and syrinx resolution at 1- and 2-year follow-ups (P ¼ 0.01). No statistically significant correlations were found between age at the time of surgery, sex, presence of scoliosis, focal neurologic deficit at diagnosis, presence of a shunt before surgery, and chemical meningitis postoperatively. Eight children (10%) had recurrent syrinx, 7 of whom underwent reoperation. Among the children who underwent reoperation for recurrent syrinx, the average time to reoperation was 37.4 months. After reoperation, 5 of these children experienced complete resolution of syrinx, 1 child showed a significant decrease on follow-up imaging, and 1 child with Crouzon syndrome had an initial decrease in syrinx size but subsequent reexpansion 7 months later (Table 3). Two children (3%) showed evidence of a stable syrinx, similar in size to that seen on preoperative imaging and remained asymptomatic throughout follow-up. At last follow-up, 95% of the children with syrinx at presentation had decreased or resolved syrinx, 2 children had small, stable syrinxes and 1 child had stable disease. Complications Among the 105 children, there was no permanent morbidity or mortality. The complication rate was 3.8%, including 3 cases of clinically significant aseptic meningitis requiring removal of the allogenic duraplasty, and one Serratia wound infection with subsequent meningitis. The 3 cases of aseptic meningitis requiring reoperation occurred at 1 month, 2 months, and 3 months after the first surgery (Table 3). The child with a Serratia wound infection and meningitis developed the infection 2 weeks postoperatively after swimming in the ocean 1 week after surgery against medical advice. He was successfully treated with antibiotic therapy and did not require surgical intervention. At the time of this report, he is nearly 12 years out of surgery, has completed college in the interim, and is working full-time with no symptoms. Reoperations Twelve of the 105 children (11%) required reoperation, with a median interval between operations of 8.9 months (range, 1.3e77.9 months) (Table 3). Indications for reoperation included recurrent syrinx (symptomatic and/or 3 level disease), aseptic meningitis, and recurrent Valsalva-related headache. In the 7 children with recurrent syrinx, reoperation included repeat craniotomy, dissection of the fibrous band, and dissection of arachnoid adhesions, depending on the presentation. If there were no visible arachnoid adhesions or if there was a lack of spontaneous pulsations after arachnoid dissection, a syringo-subarachnoid or fourth ventricle to subarachnoid shunt was placed. In 6 out of 7 children, a >75% improvement in syrinx size was achieved with this approach. Children with prolonged aseptic meningitis were treated in a uniform manner, with removal of the Durepair graft and placement of new autologous pericranial duraplasty. In these children, Bioglue was used in all cases during the initial operation. Fisher exact test showed no significant correlation between the use of Bioglue and aseptic meningitis (P ¼ 1.64). Partial C2 laminectomy and coagulation of cerebellar tonsils was used to treat children with recurrent tussive headache, and both children experienced complete symptom resolution after surgery.
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.06.073
Age at Surgery Patient (Years)
Initial Presentation
Presence of Syrinx
Interval Between Operations (Months)
Intraoperative Findings
Surgery
Evolution of Syrinx
Outcome
Follow-Up (Months)
Collapse
Asymptomatic
87
Recurrent syrinx 6
Serendipitous
Yes
78
Posterior arch of C1 completely reossified and fibrocalcific dural scarring at the foramen magnum causing compression
2
6
Serendipitous
Yes
7
Fibrous band compressing the dural graft at Dissection of fibrous band, placement of the foramen magnum; no spontaneous fourth ventricle to subarachnoid shunt pulsation of the dura; no significant arachnoidal scarring; foramen of Magendie open
Collapse
Asymptomatic
37
3
4
Scoliosis workup
Yes
16
No reossification; capacious subarachnoid space visible after opening the duraplasty; foramen of Magendie obstructed by arachnoidal scarring
Collapse
Scoliosis improved
36
4
4
Valsalva- related/ tussive headaches
Yes
21
Reexploration suboccipital craniectomy; Epidural pseudomeningocele with fibrotic microsurgical dissection of arachnoidal walls causing compression; foramen of Magendie obstructed by arachnoidal scarring adhesions
Decrease Headache-free
48
5
11
Scoliosis workup
Yes
9
Foramen of Magendie obstructed by arachnoidal scarring
Collapse
Asymptomatic
96
6
6
Serendipitous
Yes
4
Partial reossification of craniectomy; dural Decrease Expansion of suboccipital craniectomy, graft markedly fibrotic and compressing the partial C2 laminectomy, Durepair duraplasty; cervicomedullary junction. T6 laminectomy and placement of syringosubarachnoidal shunt
Unknown
7
7
13
Gait instability and focal neurologic deficit
Yes
127
Collapse
Asymptomatic
272
Markedly fibrotic dural graft and reduced subarachnoid space; arachnoidal scarring
Laminectomy C1 and expansion of reossified craniectomy; microsurgical dissection of arachnoidal adhesions until exposure of foramen of Magendie
Microsurgical dissection of arachnoidal adhesions until exposure of foramen of Magendie, T5 laminectomy, and insertion of syringo-subarachnoidal shunt
Microsurgical dissection of arachnoidal adhesions until exposure of foramen of Magendie; placement of fourth ventriclesubarachnoidal shunt
Microsurgical dissection of arachnoidal adhesions until exposure of foramen of Magendie, coagulation of tonsils
Prolonged aseptic meningitis 9
Scoliosis workup
Yes
2
3-mm perforation in the dural graft, epidural Removal of Durepair graft and placement of pseudomeningocele new autologous pericranial duraplasty
Collapse
Headache-free
43
9
8
Mixed headache
Yes
3
Small dural opening, epidural pseudomeningocele
Removal of Durepair graft and placement of new autologous pericranial duraplasty
Collapse
Headache-free
51
10
16
Mixed headache
No
1
Small epidural pseudomeningocele
Removal of Durepair graft and placement of new autologous pericranial duraplasty
N/A
Headache-free
47
Recurrent Valsalva-related/tussive headache N/A, not applicable.
e5
Continues
ORIGINAL ARTICLE
8
SUBOCCIPITAL DECOMPRESSION WITH DURAPLASTY IN CHIARI MALFORMATION TYPE I
www.WORLDNEUROSURGERY.org
1
EVELINE TERESA HIDALGO ET AL.
WORLD NEUROSURGERY -: e1-e8, - 2018
Table 3. Characteristics and Outcomes of 12 Patients Requiring Reoperation
ORIGINAL ARTICLE
e6
SUBOCCIPITAL DECOMPRESSION WITH DURAPLASTY IN CHIARI MALFORMATION TYPE I
Thick band of fibrous scar tissue at foramen Partial C2 laminectomy, coagulation of both magnum and C1 compressing the underlying cerebellar tonsils duraplasty
Decrease Headache-free
12
In our study, we report the outcomes of 105 consecutive children treated with a standardized surgical technique including duraplasty, which proved to be efficient in resolving headaches and syrinx with no permanent morbidity or mortality. The majority of headaches resolved by 3 months, and syrinx improvement was noted by 6 months. We believe that our narrow window of surgical indications—specifically, Valsalva-related headaches and syringomyelia with involvement of at least 3 spinal levels and expansion of the spinal cord by >4 mm—contributed to the favorable outcomes. The surgical indication for asymptomatic large syringomyelias in young children is based on the unknown natural history and is partially predicated on low rates of new neurologic injury, infection, and CSF leakage. Resolution of Symptoms Over Time Headaches. Valsalva-related headaches completely resolved in 93% of the children within the first year after surgery. Fifty-seven percent of the children experienced complete relief before discharge from the hospital. In our subanalysis, children with classic Valsalva-related/tussive headaches had a significantly faster resolution of symptoms than children with atypical headaches. In a study of 500 children with dural opening reported by Tubbs et al.,10 the overall resolution rate of preoperative signs and symptoms was 83%, similar to that our series. In their series of 156 children without dural opening, Kennedy et al. found that 62% of children who presented with occipital and/or tussive headache had complete resolution of headache; when including asymptomatic and improved but minimally symptomatic children, 91% of children were considered therapeutic successes.11 In a large retrospective review of 256 children treated with suboccipital decompression, McGirt et al.5 found a recurrence rate of 22%, and noted that headaches were 70% more likely to recur than cranial nerve and brainstem symptoms. We did not find such a high recurrence rate; in our series, only 3 of 45 children (7%) with headaches had a recurrence, with a mean follow-up time of 46 months. In contrast to that study, our children were treated in a uniform manner with dural opening and stringent criteria for the headaches that we believed were Chiari-related and appropriate for surgical intervention. Children with atypical headaches underwent surgery because of their coexisting syrinx detected on imaging, not because of their headache etiology. Although we did not expect them to have headache improvement, we did discover some improvement that might possibly be attributed to medical management of a coincidental non-Chiari headache disorder. The results of this study may reflect a narrower definition of headaches that are appropriate for surgical management and achieve rapid and long-lasting resolution of symptoms.
www.SCIENCEDIRECT.com
N/A, not applicable.
9 Yes Valsalva- related/ tussive headache 20 12
No 17
Valsalva- related/ tussive headache
37
Thick band of scar tissue at the foramen Partial C2 laminectomy, coagulation of right magnum extending to C2; cerebellar tonsil on cerebellar tonsil the right extended to the upper border of C2
N/A
Headache-free
24
DISCUSSION
11
Age at Surgery Patient (Years)
Table 3. Continued
Initial Presentation
Presence of Syrinx
Interval Between Operations (Months)
Intraoperative Findings
Surgery
Evolution of Syrinx
Outcome
Follow-Up (Months)
EVELINE TERESA HIDALGO ET AL.
Syrinx. In this study, we found that children with an associated genetic syndrome or previous surgery had a statistically significant delay in resolution of syrinx compared with children without these conditions. This may be helpful in clinical practice when evaluating the radiologic outcomes of these children. No other clinical factor was correlated with timing of syrinx resolution. We observed an improvement or resolution of syrinx in 95% of our children, with a mean time to resolution of 8.1 months. These findings are similar to those reported in previous surgical series of
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.06.073
ORIGINAL ARTICLE EVELINE TERESA HIDALGO ET AL.
SUBOCCIPITAL DECOMPRESSION WITH DURAPLASTY IN CHIARI MALFORMATION TYPE I
decompression and duraplasty. Wu et al.14 reported syrinx improvement in 82% of 44 children with CM-I undergoing dural opening by 6 months postoperatively and in 98% of children over the course of follow-up. As in our study, Yeh et al.15 reported a notable improvement in 17 of 20 (85%) children after posterior fossa decompression with dural opening and tonsillar shrinkage. We found a median time to syrinx improvement of 3 months, with >79% of cases improving in the first 6 months after surgery. This is comparable to the results of a prospective study reported by Wetjen et al.,1 with a median time to syrinx narrowing of 3.6 months following dural opening. We believe that if there is no evidence of improvement in syrinx size within 12 months, reoperation should be considered, because the majority of children improve in this time period (Figure 2). Our present series and the previously mentioned reports in the literature suggest a more robust improvement in syringomyelia over a shorter period when duraplasty is performed. In a series of children with CM-I without dural opening, Kennedy et al.11 reported that 70% of children experienced radiologic improvement of syrinx, 23% had a stable syrinx, and 7% had new a syrinx or growth of an old syrinx.11 For children with demonstrated radiographic improvement, the median and mean times to improvement were 21 and 31 months, respectively.2 The other technical aspect that has been advocated at the initial surgery is extensive coagulation of the herniated cerebellar tonsils. Stanko et al.3 compared to syrinx improvement in children with tonsillar cautery and those without cautery.3 They found that 81% of the children with tonsillar coagulation had improvement of syrinx after a mean of 8.0 months, compared with 13.1 months in children without tonsillar coagulation. In their series, children who underwent tonsillar cautery had 6-fold greater odds of improvement in syrinx. In our series of duraplasty without significant tonsillar cauterization, the rate of improvement in syrinx was similar if not better. This raises the question of whether coagulation of the tonsils with the low attendant risk of parenchymal or vascular injury should be avoided at the initial operation. Complications The overall complication rate in our series was 3.8%. We believe this low complication rate is due to the absence of wound healing problems, which is a result of our modified surgical technique. In a meta-analysis conducted by Durham et al.,11 CSF-related complications were described in 18.5% of cases with dural opening, compared with 1.8% of cases without dural opening.12 The absence of CSF leak may be due to the technique of dural closure with a Durepair patch, which was secured with a
REFERENCES 1. Wetjen NM, Heiss JD, Oldfield EH. Time course of syringomyelia resolution following decompression of Chiari malformation type I. J Neurosurg Pediatr. 2008;1:118-123. 2. Kennedy BC, Nelp TB, Kelly KM, Phan MQ, Bruce SS, McDowell MM, et al. Delayed resolution of syrinx after posterior fossa decompression without dural opening in children with Chiari malformation type I. J Neurosurg Pediatr. 2015;16: 599-606.
running, locking absorbable suture, in many cases along with a dural sealant (Bioglue) or fibrin glue. Parker et al.16 investigated the associations between complications and the use of duraplasty and dural sealant. Although their sample was quite small (n ¼ 15), they noted an overall complication rate of 42%, with the most common complication, aseptic meningitis, seen in 21% of their children.16 They also reported CSF leak in 8.4% of their children, pseudomeningocele in 3.9%, and infection in 3%. Although 3 children in our series developed aseptic meningitis that required reoperation, we did not find any statistically significant difference between the type of sealant and the rate of complications. The same type of allogenic Duragraft was used in all cases. Our finding that all 3 children with aseptic meningitis had subclinical CSF leakage and dehiscence of the dural graft at reoperation, along with the lack of statistical correlation between sealant use and aseptic meningitis, suggests that the sealant alone might not be the cause of the inflammatory response. Reoperation Rate Lower rates of reoperation have been reported in children who undergo posterior fossa decompression with duraplasty, ranging from 2.1% to 3.1%, compared with the 9%e12.6% in children without duraplasty.11,12,17 In our series with duraplasty, reoperation was necessary in 11% of children. This higher reoperation rate may be due to a longer duration of follow-up; 5 of the 12 children requiring reoperation had a late recurrence of symptoms or syrinx (range, 16e127 months). Of note, 2 of those children underwent previous surgery at outside institutions. Limitations of this study are its retrospective nature and potentially missing data on late recurrence, as in this historical cohort some children may be lost to follow-up. CONCLUSIONS Chiari decompression with duraplasty is an efficacious procedure in carefully selected children. This study confirms that this technique rapidly resolves both classic tussive headaches and syringomyelia, at average intervals of 3 and 6 months, respectively, and provides durable remission for these children. Further prospective studies are needed to identify the subgroup of children at risk for recurrence or delayed resolution of symptoms. ACKNOWLEDGMENTS The authors thank Christopher Hernandez for his collaboration in data collection.
3. Stanko KM, Lee YM, Rios J, Wu A, Sobrinho GW, Weingart JD, et al. Improvement of syrinx resolution after tonsillar cautery in pediatric children with Chiari type I malformation. J Neurosurg Pediatr. 2016;17:174-181.
5. McGirt MJ, Attenello FJ, Datoo G, Gathinji M, Atiba A, Weingart JD, et al. Intraoperative ultrasonography as a guide to patient selection for duraplasty after suboccipital decompression in children with Chiari malformation type I. J Neurosurg Pediatr. 2008;2:52-57.
4. Attenello FJ, McGirt MJ, Gathinji M, Datoo G, Atiba A, Weingart J, et al. Outcome of Chiariassociated syringomyelia after hindbrain decompression in children: analysis of 49 consecutive cases. Neurosurgery. 2008;62:1307-1313 [discussion: 1313].
6. Haroun RI, Guarnieri M, Meadow JJ, Kraut M, Carson BS. Current opinions for the treatment of syringomyelia and chiari malformations: survey of the Pediatric Section of the American Association of Neurological Surgeons. Pediatr Neurosurg. 2000; 33:311-317.
WORLD NEUROSURGERY -: e1-e8, - 2018
www.WORLDNEUROSURGERY.org
e7
ORIGINAL ARTICLE EVELINE TERESA HIDALGO ET AL.
SUBOCCIPITAL DECOMPRESSION WITH DURAPLASTY IN CHIARI MALFORMATION TYPE I
7. Rocque BG, George TM, Kestle J, Iskandar BJ. Treatment practices for Chiari malformation type I with syringomyelia: results of a survey of the American Society of Pediatric Neurosurgeons. J Neurosurg Pediatr. 2011;8:430-437.
12. Durham SR, Fjeld-Olenec K. Comparison of posterior fossa decompression with and without duraplasty for the surgical treatment of Chiari malformation type I in pediatric children: a metaanalysis. J Neurosurg Pediatr. 2008;2:42-49.
8. Hankinson T, Tubbs RS, Wellons JC. Duraplasty or not? An evidence-based review of the pediatric Chiari I malformation. Childs Nerv Syst. 2011;27:35-40.
13. Pirouzmand F, Tucker WS. A modification of the classic technique for expansion duroplasty of the posterior fossa. Neurosurgery. 2007;60(2 Suppl 1): ONS60-ONS62 [discussion: ONS62].
9. Galarza M, Sood S, Ham S. Relevance of surgical strategies for the management of pediatric Chiari type I malformation. Childs Nerv Syst. 2007;23:691-696. 10. Tubbs RS, Beckman J, Naftel RP, Chern JJ, Wellons JC 3rd, Rozzelle CJ, et al. Institutional experience with 500 cases of surgically treated pediatric Chiari malformation type I. J Neurosurg Pediatr. 2011;7:248-256. 11. Kennedy BC, Kelly KM, Phan MQ, Bruce SS, McDowell MM, Anderson RC, et al. Outcomes after suboccipital decompression without dural opening in children with Chiari malformation type I. J Neurosurg Pediatr. 2015;16:150-158.
e8
www.SCIENCEDIRECT.com
14. Wu T, Zhu Z, Jiang J, Zheng X, Sun X, Qian B, et al. Syrinx resolution after posterior fossa decompression in patients with scoliosis secondary to Chiari malformation type I. Eur Spine J. 2012; 21:1143-1150. 15. Yeh DD, Koch B, Crone KR. Intraoperative ultrasonography used to determine the extent of surgery necessary during posterior fossa decompression in children with Chiari malformation type I. J Neurosurg. 2006;105(1 Suppl):26-32.
children: dural graft or sealant? J Neurosurg Pediatr. 2011;8:177-183. 17. Mutchnick IS, Janjua RM, Moeller K, Moriarty TM. Decompression of Chiari malformation with and without duraplasty: morbidity versus recurrence. J Neurosurg Pediatr. 2010;5:474-478.
Conflict of interest statement: This publication is supported in part by the NYU CTSA grant UL1TR001445 from the National Center for Advancing Translational Sciences, National Institutes of Health. Received 31 March 2018; accepted 11 June 2018 Citation: World Neurosurg. (2018). https://doi.org/10.1016/j.wneu.2018.06.073 Journal homepage: www.WORLDNEUROSURGERY.org Available online: www.sciencedirect.com
16. Parker SR, Harris P, Cummings TJ, George T, Fuchs H, Grant G. Complications following decompression of Chiari malformation type I in
1878-8750/$ - see front matter ª 2018 Elsevier Inc. All rights reserved.
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.06.073