Chiari Malformation Associated With Craniosynostosis

Jennifer Strahle, M.D.; Karin M. Muraszko, M.D.; Steven R. Buchman, M.D.; Joseph Kapurch, B.S.; Hugh J. L. Garton, M.D., M.H.Sc.; Cormac O. Maher, M.D.


Neurosurg Focus. 2011;31(3):e2 

In This Article


An association between CM and craniosynostosis has been recognized for many years.[36] Chiari malformation is especially likely to be found in those with multisuture or syndromic craniosynostosis. Cinalli et al.[3,4] examined 95 patients with syndromic craniosynostosis and found CM in 70% of those with Crouzon syndrome, 75% of those with oxycephaly, 50% of those with Pfeiffer syndrome, and 100% of those with the Kleeblattschädel deformity.[26,43] Chiari malformation was found in only 1.9% of patients with Apert syndrome in that series.[4] Importantly, that group used cerebellar tonsillar descent of less than 2 mm below the basion-opisthion line as the diagnostic criterion, perhaps leading to a larger number of diagnosed cases of CM. Other reports, however, have confirmed the frequent association of CM with craniofacial syndromes.[7,9,33] Francis et al.[9] found an associated CM in 5 of 10 patients with Crouzon syndrome. Fearon and Rhodes[7] found that 84% of the 28 patients they treated for Pfeiffer syndrome had associated CM. Half of the children with Pfeiffer syndrome underwent placement of a VP shunt for hydrocephalus, and CM was diagnosed in every patient in whom a VP shunt was placed. Because CM has been found so frequently in children with Pfeiffer syndrome, Fearon and Rhodes have advocated routine screening of these children with MR imaging.

Many cases of single-suture nonsyndromic craniosynostosis associated with CM have also been reported.[2,17,25,41] Most of these cases involve the lambdoid suture, but CM has been reported even in conjunction with sagittal, metopic, and unilateral coronal synostosis.[2,17,25,41] Leikola et al.[17] reported on a series of 124 patients with single-suture craniosynostosis and, on imaging, found that 7 (5.6%) had CM. Sgouros et al.[38] showed that cranial base growth is altered in both multi- and single-suture craniosynostosis, even in cases that do not exhibit premature fusion of a cranial base synchondrosis. Tubbs et al.[41] found that 30% of those with metopic ridges had an associated CM, and they postulated that this may be the result of reduced anterior fossa volume. In our series, 15 (52%) of 29 patients with CM and craniosynostosis had single-suture disease, including syndromic and nonsyndromic cases. Five of these cases involved the lambdoid suture alone. We found CM in over half of all cases of isolated lambdoid synostosis. Chiari malformation was also found in isolated sagittal and coronal synostosis in 5 patients each when syndromic cases were included. Three patients with single-suture, nonsyndromic sagittal craniosynostosis and no patients with single-suture, nonsyndromic coronal craniosynostosis had associated CM. Because these sutures are involved much more often than the lambdoid suture, CM is significantly less associated with single-suture disease in these locations.

On the basis of the previously reported series and our own experience, we believe it is clear that syndromic and multisuture synostosis is frequently associated with CM. Furthermore, the frequent association of lambdoid synostosis with CM is well established.[4] Lambdoid suture involvement is predictive of CM formation even in patients with multisuture or syndromic forms of craniosynostosis.[4,43] Cinalli et al.[4] have postulated that CM is much more likely in patients with Crouzon syndrome than those with Apert syndrome, because the former is associated with significantly earlier closure of the sagittal and lambdoid sutures.

The frequency of CM in various types of craniosynostosis can guide the practitioner in the decision to obtain an MR imaging study to screen for CM. Cinalli et al.[5] found that approximately one-third of their patients who had CM associated with craniofacial disorders were either symptomatic or had a spinal cord syrinx. They advocated screening for CM in all patients with complex or syndromic craniosynostosis. We generally screen patients with syndromic craniosynostosis and patients with lambdoid synostosis with brain MR imaging prior to surgical correction of the craniosynostosis. We do not routinely screen asymptomatic individuals with single-suture craniosynostosis at locations other than the lambdoid suture.

Although some brain malformations associated with craniofacial disorders may be the primary result of a common genetic abnormality, most now believe that CM associated with craniosynostosis is acquired postnatally as a result of abnormal skull development.[27] Chiari malformation has been associated with underdevelopment of the occipital bone and a small posterior fossa,[21,22,24] as well as with primary bone disorders that affect the posterior fossa skull such as osteopetrosis and fibrous dysplasia.[30] Craniofacial disorders that lead to a small or deformed posterior fossa may result in crowding of posterior fossa contents and tonsillar descent in affected patients.[3] Most cases of CM associated with craniosynostosis are found after the prematurely fused suture has resulted in a significant skull deformity.[5] This assertion is supported by several reported cases of de novo CM formation following diagnosis of a craniofacial syndrome. Hopkins and Haines[14] reported on a case of Seckel syndrome in a patient in whom rapid CM development was demonstrated on serial imaging. Ranger et al.[26] reported on a patient with Pfeiffer syndrome whose initial brain MR imaging findings were normal; 2 months after craniofacial reconstruction and VP shunt placement, however, MR imaging showed the development of CM, providing more evidence for an acquired pathogenesis. In patients with Crouzon syndrome, premature closure of the lambdoid suture is associated with an increased risk of CM.[4,11,20] Apert syndrome, however, which typically presents with a normal or larger than normal posterior fossa, is not associated with CM.[32]

Hydrocephalus is frequently associated with craniofacial anomalies and may occasionally play a role in CM pathogenesis in these patients.[4,9,10,43] Children with craniofacial disorders as well as hydrocephalus are more likely to have a CM than children with a craniofacial disorder in the absence of hydrocephalus.[4,9,34] Hydrocephalus has also been reported to develop after repair of craniosynostosis, with subsequent development of tonsillar herniation.[9,43] In our own series, 4 of the 5 patients with de novo CM on serial imaging developed hydrocephalus after craniosynostosis repair and were treated with VP shunt placement prior to the CM diagnosis. Our own results, combined with those of prior reports, suggest that hydrocephalus plays a role in the pathogenesis of CM in some cases.[4,9,10,34,43] We recommend routine screening for CM in patients with both craniosynostosis and hydrocephalus.

Craniofacial anomalies may be associated with venous outflow impairment that may lead to venous hypertension and increased ICP.[1,9,12,13,28,31,33,39,40] Some surgeons suggest that cerebellar tonsillar herniation may result from increased venous turgor.[9,28] Venous abnormalities are more often associated with complex and syndromic forms of craniosynostosis.[33,39] Abnormal venous anatomy must be accounted for when considering surgical treatment of CM associated with complex craniofacial syndromes. For this reason, we now screen all craniosynostosis and CM patients with preoperative CT venography to identify any transosseous venous channels. Furthermore, opening the dura for CM decompression in these cases may increase the risk of hemorrhage due to abnormal venous sinuses. In our experience, opening the dura has not been necessary for the treatment of CM in these patients. In some cases, preoperative detection of major venous collaterals at the site of a proposed CM decompression may suggest that even bone removal at the foramen magnum is not safe and should not be performed.[1,37]

In general, most surgeons agree that CM should not be treated unless it is symptomatic or, in some cases, associated with a spinal syrinx. It is possible that the indications for surgical treatment of CM may be different for individuals with associated craniosynostosis. In some cases, CM decompression can be done simply at the time of a planned craniosynostosis repair. In this retrospective analysis, it is impossible to accurately describe all of the factors that went into each surgical decision. In general, we approach patients with both lambdoid synostosis and CM with an eye to expanding the posterior vault with over-correction. When a significant CM is seen on preoperative imaging, we often will include a posterior fossa decompression in the operation to repair the craniosynostosis. To the limits of this retrospective analysis, we do not believe that CSF flow was ever used as an important criterion for performing a posterior fossa decompression. Several groups recommend posterior fossa expansion surgery as the treatment of choice for all cases of CM identified prior to craniosynostosis correction, even in the absence of symptoms.[5,38,44] It is possible that, in many instances, the CM will resolve or improve following craniosynostosis repair (Table 3). Di Rocco and Velardi[6] reported on a single case in which a supratentorial cranial expansion resulted in resolution of an acquired CM. Given the very young age of patients undergoing craniosynostosis repair, it is important to consider that any bony decompression may be less durable compared with CM decompression performed later.[35] For this reason, Fearon and Rhodes[7] have advocated delaying posterior remodeling procedures in patients with Pfeiffer syndrome until at least 13 months of age. For patients with CM diagnosed before craniosynostosis repair, we reserve CM decompression for patients who exhibit symptoms of CM or a spinal cord syrinx after an extended interval following surgical correction of the skull deformity. Exceptions may be made if the CM is symptomatic or associated with a spinal syrinx. The presence of posterior fossa transosseous venous collaterals, however, would lead us to reconsider any posterior fossa decompression.

In cases associated with hydrocephalus or intracranial hypertension, it may be necessary to treat the hydrocephalus or relieve the ICP to treat the CM. Pouratian et al.[25] have reported on 2 cases of acquired CM associated with craniosynostosis and hydrocephalus that were initially treated with CSF diversion. Others have also reported on CM development after treatment of hydrocephalus.[9,28] The role of endoscopic third ventriculostomy is not clear. Fearon et al.[8] reported on a small series of patients with craniofacial dysostosis, CM, and hydrocephalus; patients who did not require surgical management of the CM underwent endoscopic third ventriculostomy, in contrast to those treated with a VP shunt for their hydrocephalus. In most cases, we prefer to treat the hydrocephalus with a shunt procedure prior to treating CM.

There are several limitations to our study. In this retrospective analysis, the treating physicians may have recommended a different treatment for patients depending on the perceived severity of the craniosynostosis or the CM. Patients with symptoms, a spinal syrinx, or a greater degree of tonsillar cerebellar descent were more likely to be surgically treated. Therefore, any attempt to compare outcomes of patients by treatment pattern will reflect this selection bias. Furthermore, MR imaging was not performed in all patients presenting with craniosynostosis. Our selective use of MR imaging in patients with craniosynostosis has resulted in a sampling bias that must be considered in any analysis of these results. Sampling bias also needs to be considered when pondering the role of hydrocephalus in these patients, because individuals with hydrocephalus were more likely to undergo repeated imaging. Although we reported on a large number of patients with craniosynostosis, CM presentations and treatments were diverse in this group of patients. This diversity resulted in relatively small sample sizes for the analysis of individual treatment strategies. Finally, none of the patients in this series were treated using minimally invasive strip craniectomy techniques that are increasingly used for treatment of craniosynostosis.[15,16,19] It is possible that these techniques will result in different rates of occurrence and different treatment outcomes for CM.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: