The natural history of the Chiari Type I anomaly

Transcription

1 See the corresponding editorial in this issue, pp J Neurosurg Pediatrics 2: , 2008 The natural history of the Chiari Type I anomaly FEDERICA NOVEGNO, M.D., MASSIMO CALDARELLI, M.D., ANTONIO MASSA, M.D., DANIELA CHIEFFO, PH.D., LUCA MASSIMI, M.D., BENEDETTA PETTORINI, M.D., GIANPIERO TAMBURRINI, M.D., AND CONCEZIO DI ROCCO, M.D. Department of Pediatric Neurosurgery, Catholic University Medical School, Rome, Italy Object. Since the advent of MR imaging, an increasing number of asymptomatic or oligosymptomatic patients have been diagnosed with Chiari malformation Type I (CM-I). The decision of whether or not to operate is more difficult in these patients than in those with clear symptoms because of the lack of information about the natural course of this disease. Methods. The authors report on their experience in a series of 22 patients with CM-I who were evaluated at the authors institution, and for whom a conservative approach to treatment was adopted. The patients ranged in age from 1 to 16 years (mean 6.3 years) at diagnosis. Neuroradiological and complete clinical examinations were performed in all patients 6 months after the first observation and every year thereafter. The follow-up period ranged from 3 to 19 years (mean 5.9 years). Results. Chiari malformation Type I was incidentally detected on MR images in 11 of 22 patients. The remaining 11 patients had minimal clinical signs at presentation that were not regarded as necessitating immediate surgical treatment. Seventeen patients (77.3%) showed progressive improvement in their symptoms or remained asymptomatic at the last follow-up whereas 5 patients (22.7%) experienced worsening, which was mild in 2 cases and required surgical correction in the remaining 3 cases. On MR imaging a mild reduction in tonsillar herniation was appreciated in 4 patients (18.18%), with complete spontaneous resolution in 1 of these. In 16 patients, tonsillar herniation remained stable during follow-up. Conclusions. The authors data confirm the common impression that in both asymptomatic and slightly symptomatic patients with CM-I, a conservative approach to treatment should be adopted with periodic clinical and radiological examinations. (DOI: /PED/2008/2/9/179) KEY WORDS Chiari malformation Type I clinical observation syringomyelia T HE Chiari anomaly constitutes a heterogeneous and multifactorial entity, comprising variants on a malformative base as well as acquired forms. It is a rare pathological condition with many controversial aspects, the unifying findings being the abnormal position and to some extent the abnormal shape of the cerebellar tonsils which herniate outside the cranial cavity and into the upper cervical canal, possibly leading to an obliteration of the subarachnoid spaces at the level of foramen magnum. 5,6 In most cases, especially in the acquired variant that arises secondary to a reduction in intracranial cavity volume, the caudal herniation of the cerebellar tonsils is associated with a cranial herniation of the upper cerebellar vermis into the cistern of the great vein of Galen. 8 This phenomenon may play a role in the cerebrospinal fluid dynamics of the affected patients. However, the impaction of the herniated cerebellar tonsils at the level of the foramen magnum is believed to cause occlusion of the subarachnoid spaces and consequently functional separation between the cranial and spinal Abbreviations used in this paper: CM = Chiari malformation; CM-I = CM Type I. compartments. The genesis of the often-associated spinal cord cavitation is attributed to this kind of dynamic anomaly. 25 The signs and symptoms in such patients are generally related to the compression of the neural structures by the herniated tonsils (long tract deficits, cranial nerve dysfunction, and cerebellar involvement) and/or by the presence of an associated syrinx. 31 Once the condition has been recognized and confirmed on a neuroradiological investigation, the choice of the most appropriate management remains debatable. Clinical observation may be the best option in asymptomatic patients, whereas surgical correction is generally suggested for symptomatic children. In asymptomatic patients, CM-I is a frequent incidental finding on MR imaging in children investigated because of behavioral disorders, developmental delay, seizures, or head injuries. The management of incidentally detected CM-I is particularly questionable; some authors 23 believe that CM-I is a dynamic disease, based on the observation that the number of symptomatic patients may increase with age, and they recommend prophylactic surgery. However, most surgeons still prefer to conduct clinical and radiological surveillance. 28 Even more controversial is defining the man- 179

2 F. Novegno et al. agement of children with CM-I who, although neurologically intact, present with syringomyelia on MR imaging. Similarly, it is difficult to decide whether mild and relatively aspecific clinical signs and symptoms (such as headache, vertigo, and dizziness) may be regarded as reliable indications for surgical treatment. For these reasons CM-I remains poorly understood; in particular, the natural history of this disease has not yet been established. In fact, despite the contributions of many authors, 6,8,9,14,20 23,29,36,37 a basic understanding of the pathogenesis and progression of this condition remains elusive. Our aim in the present study is to contribute to the understanding of the evolution of CM-I in children. We therefore present a series of 22 pediatric patients with CM-I (both asymptomatic and oligosymptomatic) for whom a surgical procedure was not advised and who underwent clinical observation for at least 3 years of follow-up. Methods Ninety-four children with CM-I were observed at the Department of Pediatric Neurosurgery at the Catholic University Medical School in Rome between January 1988 and November Among these, 22 consecutive patients with mild or absent clinical manifestations were not considered suitable for immediate surgical treatment and were enrolled in a follow-up program. Their ages ranged from 1 to 16 years (mean 6.3 years) at diagnosis. Cases of acquired Chiari Type I anomaly (such as those with faciocraniostenosis, osteopetrosis, or long-term extrathecal cerebrospinal fluid shunting) were excluded from the study. As to the remaining 72 children, 39 patients underwent posterior fossa decompression, which consisted in suboccipital craniectomy in 32 cases (30 of these have been previously reported 6 ) and bone removal associated with intradural procedures in the remaining 7 patients. Three patients affected by CM-I and tethered cord underwent terminal filum resection. Eighteen patients were evaluated for CM-I after shunt placement, and 7 of these underwent an enlarging bilateral cranioplasty. A further 12 children with CM-I associated with syndromic craniosynostosis involving the lambdoid sutures were initially treated by means of cranial vault remodeling. Neuroradiological investigation was performed in all patients and included CT scanning and MR imaging; in addition to the cranial compartment, thorough craniospinal axis examination was performed in each patient for evidence of syringomyelia or spinal cord tethering. Cranial and spinal MR imaging was performed initially 6 months after the first observation and every year thereafter. We decided to perform serial MR imaging examinations to evaluate changes (for example, ascent of cerebellar tonsils or further descent into the spinal canal, or the novel appearance of syringomyelia) despite clinical stability. Complete clinical examinations were conducted at least once a year. The follow-up varied from 3 to 19 years (mean 5.9 years); follow-up data are summarized in Table 1. To obtain information about psychomotor development, the last 10 children of the series underwent routine neuropsychological assessment. Cognitive performance, receptive and expressive language skills, verbal fluency, selective and sustained visual attention, visual motor integration and praxia, and gross and fine motor function were investigated using a series of specific tests. The assessment of cognitive development was performed using different scales as appropriate for the child s chronological age and level of functioning: the Griffith Mental Development Scales or Uzgiris Hunt Scales in children younger than 4 years old, Wechsler Preschool and Primary Scale of Intelligence in children 4 6 years of age, and the Wechsler Intelligence Scale for Children Revised for those older than 6 years. Behavioral disorders were classified according to the Diagnostic and Statistical Manual of Mental Disorders IV Revised criteria. The Movement Assessment Battery for Children measures 30 were adopted to evaluate fine motor coordination and dexterity, and static and dynamic balance. The language was explored by receptive and naming lexical and syntactic tasks. 4,27 The Developmental Test of Visual Motor and Perceptual Integration 16 and Rey Osterrieth Complex Figure Test were utilized to assess fine-motor coordination, visual perceptual organization, and spatial planning. Phrase recall of the Wechsler Intelligence Scale for Children Revised and Wechsler Preschool and Primary Scale of Intelligence were applied to measure short-term auditory memory. Finally, the Gauthier Test was used to evaluate sustained and selective visual attention in patients 4 14 years of age. Patient data were transferred into a spreadsheet (Excel, Microsoft) for tracking, comparison, and statistical examination. Results Clinical Findings at Diagnosis Among our 22 patients, CM-I was detected in 11 on MR images obtained for investigation of completely unrelated symptoms or signs as follows: paranasal sinusitis, developmental delay, precocious thelarche, neck injury at birth, and visual disturbances in 1 patient each; and seizure disorder, macrocrania, and head injury in 2 patients each. The average age in this group was 5 years (range 1 14 years). The remaining 11 patients (50%) had clinical manifestations for which they underwent neuroradiological investigation. The average age in this group was 7.5 years (range 3 16 years). The most frequent complaint was occipital and/or nuchal pain reported by 8 patients, followed by vertigo which was noted in 2 patients. Clinical signs included upper extremity pain, gait instability, and papilledema in 1 patient each. Two of these patients presented with epileptic disorders and cortical electric anomalies revealed on electroencephalography and required antiepileptic drugs that provided good control of seizures. Neuroradiological Investigation at Diagnosis All patients had a frank tonsillar herniation below the basion opisthion line that was associated with narrowing or obliteration of the cisterna magna. The degree of tonsillar herniation is displayed in Table 2. Syringomyelia was demonstrated in only 1 patient (4.5%), and this was limited to the cervical level. Hydrocephalus was detected in 5 patients (22.7%); it was moderate in 4 patients and mild and unilateral in 1 patient. Other associated anomalies were platybasia, facial dysmorphism with microcephaly, and cavum septum pellucidum detected in 1 patient each. Clinical Follow-Up Among the 11 asymptomatic patients, 7 (63.6%) did not 180

3 Natural history of Chiari malformation Type I TABLE 1 Summary of patient data* Incidental Degree of FU Degree of Case Age (yrs), Discovery; Tonsillar Period Evolution of Tonsillar Syrinx Ventricular No. Sex Symptoms Reason for MRI Associated Anomaly Herniation (mm) (yrs) Symptoms Herniation at FU Dilation 1 2, M gait instability NA none asymptomatic complete res- no no olution 2 3, F headache & NA epileptic disorder 10 5 worsened stable no no vertigo 3 4, M neck pain NA none 10 5 asymptomatic improved no no 4 5, F headache NA none asymptomatic improved no no 5 6, F headache NA none 10 4 asymptomatic stable no no 6 7, F vertigo NA none 5 3 asymptomatic stable no no 7 7, M neck pain NA none 10 3 improved stable no no 8 9, F headache NA none 5 5 improved stable no no 9 10, F nausea & NA none 5 7 asymptomatic stable no no vomiting 10 10, F headache & NA mild unilat ventriculo asymptomatic stable no stable papilledema megaly 11 16, F headache NA Sturge Weber, elephanti- 5 6 mild worsened stable no no asis & cavum septum 12 1, F NA neck injury at none 10 7 asymptomatic stable no no birth 13 1, M NA macrocrania none asymptomatic stable no no 14 2, M NA macrocrania mild ventriculomegaly asymptomatic worsened no stable 15 2, M NA epileptic disor- none asymptomatic improved no no der 16 5, M NA developmental platibasia & mild ven asymptomatic stable stable stable delay triculomegaly 17 5, F NA paranasal si- none new-onset stable no no nusitis neck pain 18 5, F NA trauma mild ventriculomegaly 5 4 new-onset stable new onset worsened headache (cervical) 19 6, F NA visual defect craniofacial dysmorph new-onset worsened new onset worsened ism & mild ventriculo- neck pain (cervical) megaly 20 6, F NA precocious none 10 3 asymptomatic stable new onset no thelarche (cervical) 21 9, M NA trauma none 5 5 asymptomatic stable no no 22 14, F NA epileptic disor- none 10 3 asymptomatic stable no no der * FU = follow-up; NA = not applicable. This patient requried a suboccipital craniectomy and C-1 laminectomy. Postoperatively the patient improved but headaches persisted. A syrinx was initially present only in Case 16. These patients required endoscopic third ventriculostomies. The patient in Case 18 was subsequently asymptomatic and the patient in Case 19 was dramatically improved. show any changes in follow-up clinical examinations. The remaining 4 patients (36.4%) experienced the onset of clinical symptoms between 3 and 5 years after the first diagnosis; these symptoms included headache and neck pain in 1 patient each, and headache/neck pain and vertigo in 2 patients. At the last follow-up examination, 8 (72.7%) of the 11 asymptomatic patients did not show clinical complaints. In the symptomatic group, 7 (63.6%) of 11 patients had no clinical manifestations of disease at last follow-up. In 2 patients (18.2%) a mild improvement was observed, and worsening symptoms were noted in the remaining 2 patients (18.2%). One of these 2 patients (Case 2) was judged to need an operation to relieve her symptoms; in fact, she showed the progressive increasing of the symptoms (headache and vertigo) without any relevant change of the neuroradiological findings. The second patient (Case 11) presented a mild worsening of the headache; the MR imaging findings were stable. She underwent intracranial pressure monitoring that revealed values within the normal range and, on these grounds, a further observational policy was adopted. Of all 22 patients, both asymptomatic and oligosymptomatic groups, 17 (77.3%) had progressive symptom improvement or remained asymptomatic at last follow-up and 5 patients (22.7%) experienced a worsening of their clinical conditions. This worsening was mild in 2 cases (9%) and required a surgical procedure in the remaining 3 (13.6%). The evolution of clinical manifestations in all 22 patients is summarized in Table 3. Findings on MR Imaging at Late Follow-Up All patients underwent periodic radiological evaluation. No radiological changes were demonstrated on MR images obtained 6 months after diagnosis (short-term follow-up). In the long term, however, a mild reduction in tonsillar herniation was appreciable in 4 (18.18%) patients (3 were symptomatic and 1 was asymptomatic) and complete spontaneous resolution occurred in 1 of these patients (Fig. 1). In 2 pa- 181

4 F. Novegno et al. TABLE 2 Evolution of tonsillar herniation degree in the asymptomatic and oligosymptomatic groups No. of Extent of Condition Patients Tonsillar Group (%) Herniation (mm) Stable Improved Worsened asymptomatic 2 (18.1) 5 2 (11 patients) 6 (54.5) (27.2) 10 3 oligosymptomatic 4 (36.4) 5 4 (11 patients) 1 (9) (54.5) tients (9%) a progressive increase in the degree of tonsillar herniation was observed. The patient in Case 14 (asymptomatic at diagnosis) did not show any progression of clinical findings, and intracranial pressure monitoring revealed values within the normal range. The patient in Case 19, who had initially presented with a visual defect, presented with new-onset progressive neck pain and apnea. Magnetic resonance images obtained in this patient revealed progression of the preexisting hydrocephalus associated with an increase in the degree of tonsillar herniation and the new appearance of syringomyelia (Fig. 2). Sixteen patients showed stable tonsillar herniation during the entire follow-up period; among these 8 were symptomatic and 8 were asymptomatic. Syringomyelia, present in 1 child, did not change over 19 years of follow-up. In contrast, the de novo appearance of a cervical syrinx was observed in 3 patients. Two of these patients also experienced worsening of their hydrocephalus and had had mild ventriculomegaly at diagnosis. The other 3 patients who had presented with mild ventriculomegaly appeared stable at last follow-up (at 4, 5, and 19 years after diagnosis). Need for Surgical Correction After a period of 3 6 years after the first diagnosis (mean 4.6 years), 3 of the 22 patients required surgery (Cases 2, 18, and 19); 2 patients underwent endoscopic third ventriculocisternostomy because of progressive worsening of the supratentorial hydrocephalus associated with new-onset cervical syringomyelia in both patients, and an increasing degree of the tonsillar herniation in 1. The third patient underwent a suboccipital craniectomy associated with C-1 TABLE 3 Summary of the evolution of clinical manifestations over long-term follow-up* Evolution of Symptoms Age at Case Diagnosis Last Group No. (yrs), Sex Initial Complaint 1st Yr 2nd Yr 3rd Yr 4th Yr 5th Yr 6th Yr 7th Yr 8th Yr 9th Yr FU (yr) A 12 1, F none asymp , M none asymp asymp, , M none asymp , M none asymp asymp, , F none asymp , M none asymp , F none asymp 3 B 17 5, F none neck pain , F none headache , F none neck pain, 3 vertigo, apnea C 3 4, M neck pain improved asymp 5 4 5, F headache improved asymp 6 5 6, F headache improved asymp 4 6 7, F vertigo improved asymp 3 7 7, M neck pain improved improved 3 8 9, F headache improved improved , F nausea & vomiting improved asymp asymp , F headache & papille- improved asymp 4 dema D 2 3, F headache & vertigo worsened 5 apnea 11 16, F headache worsened stable 6 E 1 2, M gait instability asymp neck pain improved asymp, , M none headache & improved asymp 9 vertigo * In this table the series has been divided into 5 subgroups according to clinical evolution: A) 7 incidentally diagnosed patients with no changes at the last clinical examination; B) 3 asymptomatic patients with new onset of clinical symptoms 3 5 years after the first diagnosis (note that the worsening of the clinical symptoms in 2 of these patients was related to the progression of supratentorial hydrocephalus); C) 8 oligosymptomatic patients who experienced improvement from the 2nd and 3rd years after diagnosis; D) 2 oligosymptomatic patients with worsening of symptoms; and E) 2 patients (1 asymptomatic and 1 oligosymptomatic) with new onset of headache/neck pain at 5 and 6 years after first diagnosis which disappeared after 2 years. Abbreviation: asymp = asymptomatic. 182

5 Natural history of Chiari malformation Type I FIG. 1. Case 1. Representative T1-weighted MR images. Tonsillar herniation is demonstrated in the image in panel A obtained in 1997 and performed for gait instability. Additional images were obtained in 2000 (B), 2003 (C), 2005 (D), and 2007 (E). In 2007 the neuroradiological findings changed totally, showing complete resolution of the malformation (E). FIG. 2. Case 19. Representative MR images. Sagittal T1-weighted (A) and axial T2-weighted (B) images obtained at diagnosis, showing hindbrain herniation associated with mild ventriculomegaly. After 3 years, the patient presented with progression of hydrocephalus associated with an increase in tonsillar herniation shown on sagittal T1-weighted (C) and axial T2-weighted (D) images. laminectomy. The postoperative course was uneventful in all cases and the patients experienced a drastic improvement in their symptoms in the first 2 cases, and a mild improvement in the third. No significant changes were observed on postoperative MR images except for ventricular dimensions, which were reduced in the first 2 patients. Neuropsychological Findings Seven of the 10 children who underwent neuropsychological testing were right-handed, 1 was left-handed, and 2 had incomplete dominance. Eight children showed a normal level of intelligence (group mean IQ 106.2; mean of the corresponding population [standard deviation]). Two children presented with a borderline level of global intelligence (mean IQ 81.5). Naming speech delay was found in 5 cases, visual attention disturbances in 4 cases, dyspraxia in 2 cases, and visual memory deficits in 2 cases. Four of the 5 patients with naming speech delays had tonsillar herniations 10 mm. The visual attention disorders were associated with neurovisual deficits in only 1 case (Case 19). Language comprehension, short-term verbal memory, and fine movement were normal in all patients. Two patients presented with behavioral disorders such as hyperkinesia. Discussion In recent years, the wide availability of cerebral and spinal MR imaging technology has led nonneurosurgeon physicians, such as pediatricians, neurologists, and other clinicians, to use MR imaging to evaluate patients with varied complaints such as learning disabilities, behavioral problems, and epileptic disorders, with the incidental discovery of CM-I in some of them. For this reason, neurosurgeons are increasingly faced with the decision of how to manage the care of asymptomatic patients and patients who present with symptoms that may or may not be referable to an underlying hindbrain herniation, in addition to the patients who present with the typical presentation of syringomyelia and severe neurological manifestations. Moreover, in the pre MR imaging era the diagnosis of CM-I was usually not made in children younger than 12 years old because of the invasiveness of the traditional radiological investigations 183

6 F. Novegno et al. which were reserved only for patients with signs and symptoms clearly referable to spinal cord or craniocervical junction lesions. 31 The expanding use of MR imaging has led, therefore, to a progressive reduction in patient age at diagnosis, as CM-I is increasingly identified in younger children with minimal neurological disorders. 12 The proper management of these types of cases of Chiari anomaly Type I is still debatable because of the lack of precise knowledge concerning the natural history of the anomaly. In their survey of the members of the Pediatric Section of the American Association of Neurological Surgeons in 1998, Haroun et al. 13 reported that most respondents (63%) believed that asymptomatic patients rarely develop symptoms, and 83% of neurosurgeons would not operate in asymptomatic patients unless symptoms developed. A similar conclusion was reached by Schijman and Steinbok 28 in their 2004 international survey on the management of CM and syringomyelia. This study demonstrated a general agreement among practitioners for a nonoperative approach to the treatment of asymptomatic patients. Only 8% of those surveyed would recommend prophylactic surgery in the absence of an associated syrinx; the presence of a spinal cord cavity would lead the surgeon to operate even without correlating symptoms in 28% of cases, if the size of the syrinx was 2-mm wide, and in 75% of cases, if the syrinx were 8- mm wide. It is worth noting, however, that despite this consensus, no objective data were described in these studies to support clinical decision making on scientific grounds. On the other hand, some authors have speculated that CM-I is a dynamic disease, a hypothesis based on the observation that the number of symptomatic patients may increase with age. 10,23 The question raised by these considerations then is not whether or not to operate but when and which procedure should be performed. The lack of reliable information about the natural evolution of the CM-I does not allow confirmation or denial of the validity of these opposing views, and few articles in the literature are dedicated to these issues. In 2000 Meadows et al. 18 reported on a retrospective analysis of MR images obtained in 21 asymptomatic patients with CM-I, 6 of whom were children. Meadows and colleagues observed that the degree of the tonsillar herniation and the overall MR imaging characteristics are similar to those observed in the symptomatic patients reported on in the literature. We observed that in the symptomatic group, more than half of the patients (54.6%) presented with herniation 10 mm, while in the asymptomatic group 54.6% had tonsils descend between 5 and 10 mm. Among the 21 asymptomatic patients reported on by Meadows et al., only 1 (4.8%) presented with syringomyelia. In the present study only 1 child (Case 16) had a cervical syrinx without any neurological dysfunction. Magnetic resonance images obtained in this patient demonstrated the presence of platybasia, 5 10 mm tonsillar ectopia, and mild hydrocephalus. Meadows et al. did not conduct follow-up in their patients, as this was not the purpose of their analysis. Genitori et al. 10 reported on a series of 53 children with CM-I, 27 of whom were asymptomatic. These 27 patients did not undergo surgery and instead received clinical surveillance for 1 12 months. None of these patients developed symptoms or signs related to brainstem compression or the presence of a syrinx. However, the authors noted that the average age at diagnosis was significantly lower in the asymptomatic patients than in those with symptoms, a finding observed also in our series. Nevertheless, even if we accept that symptomatic cases are more common in the older age group, the clinical significance of this phenomenon in terms of prognosis remains to be clarified. Nishizawa et al. 24 published a unique report providing new information on the long-term outcomes in purely asymptomatic patients with CM-I and syringomyelia in adults. These authors described the cases of 9 adult patients who underwent evaluation for incidentally identified syringomyelia associated with tonsillar herniation and regularly underwent neurological and MR imaging examinations for at least 10 years. At the last follow-up only 1 patient had symptoms and required surgery, with prompt recovery. The analysis of these data may give the impression that an even wider syrinx associated with CM-I would follow a benign course, thus justifying conservative management until a neurological change is detected. In our series the patient in Case 16 represents the only case of asymptomatic syringomyelia associated with hindbrain herniation. This patient underwent follow-up for 19 years without any changes in his clinical condition. In this case the syrinx extends from C- 5 to C-7, with a width of 4 mm. Tokunaga et al. 35 performed a retrospective review of 27 patients with scoliosis who had syringomyelia associated with cerebellar tonsil herniation detected on MR imaging. The mean age of these patients at the first MR imaging session was 10.9 years (range 5 18 years). In ~ 50% of these cases a decrease in syrinx size was observed; in this group there was a significant elevation of the cerebellar tonsils (complete resolution in 3 cases) and improvement in scoliosis and neurological impairments. Spontaneous resolution of tonsillar herniation is rare (Table 4). Although sporadic cases have been reported in adult patients, 17 most authors describe this phenomenon as typically a childhood condition. 2,3,7,15,26,32 35 Among the patients in our study, we observed spontaneous resolution in only 1 case (Case 1). At 2 years of age, this child underwent cerebral MR imaging to investigate progressive gait instability, and a tonsillar herniation of 10 cm below the foramen magnum was detected. After a follow-up period of 10 years, the radiological appearance in this child changed totally, showing complete resolution of the malformation. During follow-up the gait instability progressively disappeared, and after 6 years from the initial diagnosis the patient presented with sporadic episodes of neck pain that eventually resolved. Many theories have been proposed to explain the spontaneous resolution of CM-I. It has been shown that in children tonsil position depends on age, and that the tonsils gradually ascend as children grow. 19 These observations raise the hypothesis that as the skull growth accelerates, it expands to accommodate the cerebellum, and the cerebellar tonsils assume their normal position. 15,19,33 Sun et al. 33 reported on an interesting case of spontaneous resolution and recurrence of CM-I with associated syringomyelia in a child. The theory of the differential growth of the osseous compartment in relation to the brain and spinal cord responsible for initial improvement cannot be applied in this case because the CM recurred despite the patient s continued growth. On this basis, the authors suggested that the venous sinus occlusion had improved and then worsened, or that the resolution of the CM and later advent of venous occlusion could have caused a recurrence of hindbrain herniation. Tonsillar her- 184

7 Natural history of Chiari malformation Type I TABLE 4 Spontaneous improvement in pediatric patients reported in the literature Authors & Year Age (yrs), Sex Associated Syrinx FU (yrs) CM-I Syrinx Sudo et al., , M cervicothoracic 2.3 complete resolution complete resolution Castillo & Wilson, , F absent 4 complete resolution absent Avellino et al., , M cervical & thoracic 5.5 improved improved Avellino et al., , F cervicothoracic 11 improved improved Sun et al., 2000* 11, M cervicothoracic 2 complete resolution improved Sun et al., , M cervical 7 complete resolution complete resolution Tokunaga et al., , F present 2 improved improved 6, M present 9 improved complete resolution 7, M present 5 complete resolution improved 7, M present 6 improved improved 8, M present 8 improved complete resolution 10, M present 3 improved complete resolution 10, M present 9 improved complete resolution 10, F present 2 improved complete resolution 10, M present 8 complete resolution complete resolution 11, M present 4 improved complete resolution 13, F present 2 improved improved 14, F present 8 improved complete resolution 15, F present 4 improved improved 16, F present 2 complete resolution complete resolution Rafia & Pascual-Castroviejo, , M cervical & dorsolumbar 4 improved complete resolution Jatavallabhula et al., , M absent 4.6 complete resolution absent present study 6, M absent 10 complete resolution absent * Patient s condition worsened after 1 year. niation due to obstruction of the posterior fossa venous outflow, in the presence of vein of Galen malformations, has already been reported. 11 Moreover, embolization in those cases led to complete resolution or significant improvement in the extent of tonsillar herniation. Among our patients with an incidental diagnosis, 4 (36.4%) presented with new-onset clinical symptoms 3 5 years after the initial diagnosis; these symptoms were headache in 1 case, neck pain in a second case, and headache, neck pain, and vertigo in the third and fourth cases. In the first patient (Case 18), there was progressive worsening of hydrocephalus without any change in the degree of tonsillar herniation, but this was associated with the onset of cervical syringomyelia and surgery was advised. The second patient (Case 17) reported neck pain at the last follow-up examination (5 years after presentation), with no evidence of progression on neuroradiological studies. The third patient (Case 15) showed progressive spontaneous improvement over the last 4 years of a 9-year follow-up period. The last patient (Case 19) experienced progression of preexisting hydrocephalus that was associated with an increase in the degree of tonsillar herniation and the novel appearance of syringomyelia; this patient underwent a surgical procedure. One of the symptomatic patients (Case 1) who initially presented with gait instability experienced rapid improvement in coordination, but complained of sporadic episodes of neck pain 6 years after the original diagnosis. During the last 3 years of follow-up in this patient, the neck pain disappeared. All these 5 patients, after a period of no less than 3 years, showed the appearance of symptoms probably related to CM-I, which resolved spontaneously in 2 of these patients. It is worth noting that the worsening of the clinical symptoms was related to the progression of supratentorial hydrocephalus in 2 of the remaining 3 patients. In our patients, the most frequent complaint was occipital and/or nuchal pain, present in 8 patients, and less frequently vertigo, in 2 patients. These are nonspecific symptoms, and although they are frequently encountered in patients with CM-I, it is still a matter of debate as to whether they are sufficient to justify surgery. 28,38 According to a survey of surgeons conducted by Schijman and Steinbok, 28 46% of respondents would operate in patients with CM-I in the absence of a syringomyelia if headaches were the only complaint. In the present series, of the 8 patients with headaches at presentation, 6 significantly improved; at the last followup 4 were asymptomatic and 2 had improved. One patient with vertigo at diagnosis was asymptomatic at the last follow-up, while the other experienced general worsening of symptoms and required surgery. From these data, it seems reasonable to consider conservative treatment if pain symptoms are the only clinical manifestations of CM-I. Three of our patients required surgery (Cases 2, 18, and 19); 2 of these underwent third ventriculocisternostomies because of progressively worsening supratentorial hydrocephalus. Independent of clinical manifestations, evidence of a slight ventricular dilation should alert the surgeon to the possibility of CM progression. However, 3 other patients (Cases 10, 14, and 16) with evidence of mild ventricular dilation at diagnosis, who were asymptomatic at the last follow-up examination, did not require surgery. One patient (Case 2) underwent suboccipital craniectomy; her initial symptoms included headache and vertigo. Once the CM was identified she underwent polysomnography testing, which revealed cortical electrical anomalies and she was started on an antiepileptic drug regimen; the vertigo subsequently improved. The new onset of sporadic episodes of apnea, which progressively increased in frequency, led the surgeons to decompress the posterior fossa. One year postoperatively, her vertigo and apnea had disappeared but the headaches persisted. 185

8 F. Novegno et al. Of our 22 patients, 5 experienced a worsening of clinical conditions. However, in only 3 of these was the worsening caused by the evolution of the CM itself. In the remaining 2 cases, the worsening was due to hydrocephalus and associated ventriculomegaly, which caused increasing hindbrain herniation. In Cases 11 and 17, headache was the main symptom that presented with a new onset or that had worsened at the last follow-up (at 6 and 5 years, respectively). In these 2 cases we chose a conservative approach because we have observed that these symptoms may eventually disappear within 2 3 years unless associated with hydrocephalus. The real question is why the patient in Case 2 required surgical decompression of the foramen magnum. Unfortunately, at this time we still do not know why this patient s condition progressed to this point, as there had been no radiological evidence of worsening during follow-up. In conclusion, 17 (77.3%) of our patients experienced progressive improvement of symptoms or remained asymptomatic at the last follow-up without evidence of significant corresponding changes on neuroimaging studies. If we exclude the 2 patients with worsening hydrocephalus, the percentage rises to 85% (17 of 20 patients). Our last comment concerns the neuropsychological assessment. Normal general development was observed in all patients. Despite having a normal mean IQ, neuropsychological processing deficits in language skills were evident in 50% of the patients who underwent neuropsychological assessment (5 of 10 patients). Qualitative analysis demonstrated slowed speech, lack of fluency, and word-finding difficulties. In these cases, we observed the patients verbal IQ scores to be within the average range, and language comprehension and production were adequate. These results are supported by similarities in the pattern of neuropsychological executive functioning deficits with dorsolateral prefrontal cortex lesions and patients with cerebellar lesions. They include deficits in planning, problem solving, verbal fluency, and other word generation abilities. 1 Conclusions In the attempt to clarify the natural history of CM-I, we present one of the first studies to analyze the evolution (for (3 years) of incidentally discovered cases of CM-I in which the patient has no or minimal clinical signs regarded as not requiring immediate surgical treatment. Our data seem to confirm the general impression that in asymptomatic patients a conservative approach may be adopted. Clinical follow-up may also be advisable for patients whose symptoms are extremely nonspecific or only doubtfully referable to the malformation. In both of these types of cases, we recommend periodic clinicoradiological examinations. However, we are as yet unable to establish how long these examinations should continue. Our study has 2 main faults: the limited number of patients and the lack of a homogeneous longterm follow-up. A multicenter cooperative study should be undertaken to provide valuable information on the natural course of these malformations. Disclaimer The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper. Acknowledgment We thank Daniele Petricca for help with the illustrations, and the association "Ali di Scorta onlus" ( References 1. Appollonio IM, Grafman J, Schwartz V, Massaquoi S, Hallett M: Memory in patients with cerebellar degeneration. Neurology 43: , Avellino AM, Britz GW, McDowell JR, Shaw DW, Ellenbogen RG, Roberts TS: Spontaneous resolution of a cervicothoracic syrinx in a child. Case report and review of the literature. Pediatr Neurosurg 30:43 46, Avellino AM, Kim DK, Weinberger E, Roberts TS: Resolution of spinal syringes and Chiari I malformation in a child. Case illustration. J Neurosurg 84:708, Bottari P, Cipriani P, Chilosi AM, Pfanner L: The italian determiner system in normal acquisition, specific language impairment, and childhood aphasia. Brain Lang 77: , Caldarelli M, Di Rocco C: Diagnosis of Chiari I malformation and related syringomyelia: radiological and neurophysiological studies. Childs Nerv Syst 20: , Caldarelli M, Novegno F, Massimi L, Romani R, Tamburrini G, Di Rocco C: The role of limited posterior fossa craniectomy in the surgical treatment of Chiari malformation Type I: experience with a pediatric series. J Neurosurg 106 (3 Suppl): , Castillo M, Wilson JD: Spontaneous resolution of a Chiari I malformation: MR demonstration. AJNR Am J Neuroradiol 16: , Di Rocco C, Velardi F: Acquired Chiari type I malformation managed by supratentorial cranial enlargement. Childs Nerv Syst 19: , Galarza M, Sood S, Ham S: Relevance of surgical strategies for the management of pediatric Chiari type I malformation. Childs Nerv Syst 23: , Genitori L, Peretta P, Nurisso C, Macinante L, Mussa F: Chiari type I anomalies in children and adolescents: minimally invasive management in a series of 53 cases. Childs Nerv Syst 16: , Girard N, Lasjaunias P, Taylor W: Reversible tonsillar prolapse in vein of Galen aneurysmal malformations: report of eight cases and pathophysiological hypothesis. Childs Nerv Syst 10: , Greenlee JD, Donovan KA, Hasan DM, Menezes AH: Chiari I malformation in the very young child: the spectrum of presentations and experience in 31 children under age 6 years. Pediatrics 110: , Haroun RI, Guarnieri M, Meadow JJ, Kraut M, Carson B: Current opinions for the treatment of syringomyelia and Chiari malformations: survey of the Pediatric section of the American Association of Neurological Surgeons. Pediatr Neurosurg 33: , James HE, Brant A: Treatment of the Chiari malformation with bone decompression without durotomy in children and young adults. Childs Nerv Syst 18: , Jatavallabhula NS, Armstrong J, Sgouros S, Whitehouse W: Spontaneous resolution of isolated Chiari I malformation. Childs Nerv Syst 22: , Kulp MT, Sortor JM: Clinical value of the Beery visual-motor integration supplemental tests of visual perception and motor coordination. Optom Vis Sci 80: , Kyoshima K, Bogdanov EI: Spontaneous resolution of syringomyelia: report of two cases and review of the literature. Neurosurgery 53: , Meadows J, Kraut M, Guarnieri M, Haroun RI, Carson BS: Asymptomatic Chiari Type I malformations identified on magnetic resonance imaging. J Neurosurg 92: , Mikulis DJ, Diaz O, Egglin TK, Sanchez R: Variance of the position of the cerebellar tonsils with age: preliminary report. Radiology 183: ,

9 Natural history of Chiari malformation Type I 20. Milhorat TH, Bolognese PA: Tailored operative technique for Chiari type I malformation using intraoperative color Doppler ultrasonography. Neurosurgery 53: , Milhorat TH, Chou MW, Trinidad EM, Kula RW, Mandell M, Wolpert C, et al: Chiari I malformation redefined: clinical and radiographic findings for 364 symptomatic patients. Neurosurgery 44: , Munshi I, Frim D, Stine-Reyes R, Weir BK, Hekmatpanah J, Brown F: Effects of posterior fossa decompression with and without duraplasty on Chiari malformation-associated hydromyelia. Neurosurgery 46: , Navarro R, Olavarria G, Seshadri R, Gonzales-Portillo G, Mc- Lone DG, Tomita T: Surgical results of posterior fossa decompression for patients with Chiari I malformation. Childs Nerv Syst 20: , Nishizawa S, Yokoyama T, Yokota N, Tokuyama T, Ohta S: Incidentally identified syringomyelia associated with Chiari I malformations: is early interventional surgery necessary? Neurosurgery 49: , Oldfield EH, Muraszko K, Shawker TH, Patronas NJ: Pathophysiology of syringomyelia associated with Chiari 1 malformation of the cerebellar tonsils. Implications for diagnosis and treatment. J Neurosurg 80:3 15, Rafia S, Pascual-Castroviejo I: Syringohydromyelia. Report of a case which resolved spontaneously. Rev Neurol 32: , Riva D, Nichelli F, Devoti M: Developmental aspects of verbal fluency and confrontation naming in children. Brain Lang 71: , Schijman E, Steinbok P: International survey on the management of Chiari I malformation and syringomyelia. Childs Nerv Syst 20: , Sivaramakrishnan A, Alperin N, Surapaneni S, Lichtor T: Evaluating the effect of decompression surgery on cerebrospinal fluid flow and intracranial compliance in patients with Chiari malformation with magnetic resonance imaging flow studies. Neurosurgery 55: , Spanò M, Mercuri E, Randò T, Pantò T, Gagliano A, Henderson S, et al: Motor and perceptual-motor competence in children with Down syndrome: variation in performance with age. Eur J Paediatr Neurol 3:7 13, Steinbok P: Clinical features of Chiari I malformations. Childs Nerv Syst 20: , Sudo K, Doi S, Maruo Y, Tashiro K, Terae S, Miyasaka K, et al: Syringomyelia with spontaneous resolution. J Neurol Neurosurg Psychiatry 53: , Sun JC, Steinbok P, Cochrane DD: Spontaneous resolution and recurrence of a Chiari I malformation and associated syringomyelia. Case report. J Neurosurg 92 (2 Suppl): , Sun PP, Harrop J, Sutton LN, Younkin D: Complete spontaneous resolution of childhood Chiari I malformation and associated syringomyelia. Pediatrics 107: , Tokunaga M, Minami S, Isobe K, Moriya H, Kitahara H, Nakata Y: Natural history of scoliosis in children with syringomyelia. J Bone Joint Surg Br 83: , Tubbs RS, Lyerly MJ, Loukas M, Shoja MM, Oakes WJ: The pediatric Chiari I malformation: a review. Childs Nerv Syst 23: , Tubbs RS, McGirt MJ, Oakes WJ: Surgical experience in 130 pediatric patients with Chiari I malformations. J Neurosurg 99: , Weinberg JS, Freed DL, Sadock J, Handler M, Wisoff JH, Epstein FJ: Headache and Chiari I malformation in the pediatric population. Pediatr Neurosurg 29:14 18, 1998 Manuscript submitted March 4, Accepted June 18, Address correspondence to: Federica Novegno, M.D., Department of Pediatric Neurosurgery, Catholic University Medical School, Largo A. Gemelli 1, Rome, Italy. federica novegno@hotmail.it. 187