Arnold-Chiari Malformation

Current understanding of cerebellar ectopia is primarily based on the detailed study and classification of this rare rhombencephalon deformity by Hans Chiari, MD, more than a century ago. Recent advances in neuroradiology, especially since the introduction of magnetic resonance imaging, have enabled the incidental diagnosis of cerebellar ectopia. This radiographic finding, increasingly observed in asymptomatic adults, presents a different clinical scenario than that characterized by Chiari, making application of Chiari's original classification scheme often difficult. This article highlights Chiari's observations of cerebellar ectopia and those of other seminal figures, relating their findings to modern recognition of this entity.

Hans Chiari (1851-1916), born in Vienna, Austria, was the son of J. B. V. L. Chiari, an Austrian gynecologist, and the brother of Ottokar Chiari, a rhinolaryngologist. After graduating in 1875, Chiari became an assistant to Karl Rokitansky, the famous pathologist at the Institute of Pathology in Vienna. Chiari later served as professor of pathology in Prague, Czechoslovakia, in 1882 and in Strasbourg, France, in 1906. In addition to observations on cerebellar deformities, to my knowledge, he was the first to relate cerebral embolism to arteriosclerosis of the carotid bifurcation.¹

Although the spectrum of craniocervical junction abnormalities is designated Arnold-Chiari malformations, to my knowledge, Chiari first characterized and classified these alterations. His earliest article² on the subject, published in 1891, was presented as a preliminary report introducing 3 types of cerebellar changes as a result of chronic hydrocephalus. An English translation of the German text was presented in 1971,³ a more extensive translation was provided in 1987,⁴ and translation corrections were published in 1991.⁵ Chiari acknowledged that, although there was vast pathological literature concerning hydrocephalus, only one author (John Cleland, MD) had previously mentioned changes of the cerebellum and brainstem as a result of hydrocephalus. With repeated recognition of abnormal anatomical patterns, Chiari began to give more attention to the relationships of the cerebellum, pons, and medulla. In describing the occurrence of these structural aberrations, he wrote: "I have found it in a relatively large percentage of cases of chronic congenital hydrocephalus, but never without hydrocephalus or in cases of acute or later developing hydrocephalus. Apparently it is related to a quite gradual growth of an anomalous form of the cerebellum."^{4 (p4)}

The malformation he labeled type 1 was characterized by elongation of the cerebellar tonsils and medial inferior lobules into cone-shaped projections that accompanied the medulla oblongata into the spinal canal. Chiari surmised that bulbar symptoms could result from the cerebellar changes; however, in the case of a 17-year-old adolescent girl used to illustrate this malformation, specific symptoms related to the cerebellum or brainstem were lacking. Although evidence of congenital hydrocephalus occurred in this case, Chiari questioned the causal relationship of hydrocephalus to the changes in the cerebellum. "In this case I do not intend to attribute the abnormal configuration of the cerebellum, and distal displacement and flattening of the medulla and cervical cord directly to hydrocephalus."^{4 (p4)}

After reviewing 14 additional cases characterized as type 1 alterations, Chiari offered an alternative explanation in a subsequent study⁶ published in 1896. He concluded that a mechanism of deficient bone growth and insufficient enlargement of the skull caused increased local pressure on intracranial contents, resulting in elongation of the cerebellar tonsils.⁶

The Chiari type 2 anomaly was depicted by the case of a 6-month-old girl with a history of bladder and rectum paralysis and complete paraplegia. In this example, there was marked cerebral hydrocephalus.

In summary, gross inspection showed downward displacement of the pons medulla and cervical cord as well as parts of the cerebellum inside the fourth ventricle into the spinal canal; a step-like formation at the border between medulla and cervical cord; a shortened cervical cord; a lengthened inferior segment of the thoracic cord; a lumbar myelomeningocele with diastematomyelia, and a hydromyelia in the thoracic cord.^{4 (p5)}

Chiari considered 2 sequences of anomalies in the central nervous system to underlie lumbar diastematomyelia, rhombencephalon changes, cellular dysgenesis, and cerebral hydrocephalus. He stated that the most striking changes in this case were the crowding of the cervical cord and the deposition of the cerebellum within the fourth ventricle. Having witnessed these changes before in patients with congenital hydrocephalus, he believed that cerebral hydrocephalus contained within an inadequately sized skull resulted in downward descent of the basal brain.

The skull cavity becomes too small for the hydrocephalic brain despite the expansion of the skull. The basal parts of the brain are displaced downward into the widened spinal canal. Gradually in the course of their further growth, the described changes of architecture and structure are produced. That this type of consecutive change in hydrocephalus could become clinically significant need not be further emphasized.^{4 (p6)}

The third type of alteration was represented by the case of a 5-month-old girl who presented with marked cerebral hydrocephalus, converging strabismus, and a cerebellar-cervical encephalocele, without clinical paralysis. The sylvian aqueduct and fourth ventricle were dilated, the latter communicating eccentrically with markedly hydromyelic cavities of the cervical and thoracic cord, extending to the second lumbar vertebra.

Between 1889 and 1892, Chiari collected pathological material from 4276 cadavers, including 63 cases of congenital hydrocephalus.⁶ From these data, presented in an article published in 1896,⁶ Chiari amended the description of the type 2 anomaly based on observations in 5 of 7 cases in neonates, and he offered a fourth classification entity to the scheme. For the type 2 class, he described the inferior vermis, pons, and medulla descending dorsal to (not necessarily within, as previously described) an elongated fourth ventricle. The type 4 anomaly was characterized by cerebellar hypoplasia based on 2 cases. In both, Chiari considered hydrocephalus as the cause of the pathological changes.^{5 ,7} In the concluding remarks of this 1896 article,⁶ Chiari acknowledged Arnold and Cleland, who previously published on this specific subject.

John Cleland of Edinburgh, Scotland, presented an account in 1883 of an infant with spina bifida, a displaced nodular cerebellum into an elongated fourth ventricle, and hydrocephalus.⁸ His writing focused on the dysraphia and visceral abnormalities. This description corresponded to the Chiari type 2 malformation.

Julius Arnold (1835-1915) was a student of Nikolaus Friedreich, MD, and Rudolf Virchow, MD. After graduation in 1859 in Heidelberg, Germany, he rapidly rose in the academic hierarchy and became a professor of pathological anatomy in 1866. He published an article in 1894⁹ that portrayed an infant with spina bifida. Arnold described an elongated inferior portion of the cerebellum that covered the fourth ventricle and extended halfway into the spinal canal, and an unchanged medulla. He did not detect widening of ventricles or other signs of hydrocephalus. Arnold defined parts of the myelomeningocele and characterized the spina bifida in detail. His discussion and concluding remarks focused primarily on spina bifida and theories on the cause of this malformation. Chiari referred to this case in his 1896 publication⁶ and believed that the abnormal formation of the rhombencephalon in the presence of spina bifida was similar to his type 2 classification. Chiari also thought that prior hydrocephalus caused the described alterations of the rhombencephalon.

In 1907, 2 students from Arnold's laboratory (Schwalbe and Gredig) first applied the Arnold-Chiari eponym to patients previously characterized as having the Chiari type 2 malformation.^{3 ,5 ,10} They differentiated the Chiari type 2 classification from Arnold's description by attributing initial detection of cerebellar elongation to Arnold and the medullary steplike deformity (kink) to Chiari. This improper modification ignored the Chiari original type 2 description, which included cerebellar and medullary abnormalities. This eponym continued to modern times, however, creating ambiguity when attempting to define rhombencephalon deformities with a uniform definition. Incontestably, Chiari conducted the more thorough study of rhombencephalon deformities and provided the more detailed anatomical analysis of the lesions, classifying malformations based on patterns of alterations. Based on these seminal contributions, one could argue that the nosologic descriptive scheme should be attributed to Chiari alone.

Modern neuropathologists largely continue to accept Chiari's classification design. Today's routine clinical use of neuroradiological imaging facilitates determination of malformation type without the need for invasive techniques. With this technological advance, however, type 1 malformations with varying degrees of tonsillar ectopia are increasingly recognized on midsagittal magnetic resonance images in asymptomatic subjects and in some patients with a wide variety of clinical symptoms.¹¹ Chiari presented his classification scheme as a continuum, with emergence of each type based on downward displacement of the cerebellum, its relationship to the medulla and pons, and the presence of spina bifida and hydromyelia. Chiari did not attribute clinical symptoms to the cerebellar changes in subjects he classified as having type 1 disease. In addition, he offered a mechanism other than hydrocephalus to account for these cerebellar alterations, indicating a different and perhaps less severe entity. There is continued debate regarding the significance of cerebellar ectopic changes incidentally observed on magnetic resonance images. Neurologists face a challenge that Chiari did not consider, and must determine the significance of incidental Chiari malformations in the context of often vague and difficult to localize symptoms.

Corresponding author and reprints: Bridgette Arnett, MD, PhD, South Suburban Neurology, Ltd, 3235 Vollmer Rd, Suite 110, Flossmoor, IL 60422 (e-mail: bridgette@hotmail.com).

Accepted for publication October 25, 2002.