To the Editor:
This article adds to the list of impressive publications of Milhorat et al. on craniovertebral junction and related anomalies. The present work is based on a massive, and probably the largest, series of 364 symptomatic patients having Chiari I malformation (7). The article provides inroads to embryological, radiological, and clinical understanding of Chiari I malformation. On the basis of our understanding of the embryological pathology of this region and our experience with approximately 2000 cases of congenital craniovertebral anomalies evaluated and surgically treated from 1950 to 1999, we wish to make some comments.
The major thrust of the article by Milhorat et al. (7) is the currently well-accepted premise that Chiari I malformation and related pathological events could be primarily attributable to maldevelopment of the occipital bone and overcrowding of the normally developed cerebellum within a smaller posterior cranial fossa (PCF). On the basis of a literature review and our own study (3), we have observed that the maldevelopment results in a reduced length of the clivus (i.e., the sphenoid part of the clivus is formed relatively normally, whereas the occipital part is formed incompletely) and platybasia, occipital condylar and adjoining occipital bone hypoplasia, nonformation or inadequate formation of the occipitoaxial joint, and, frequently, occipitalization of the atlas. Fusion of the atlantoaxial joint and C2–C3 spinal elements and a range of Klippel Fiel spinal abnormalities are also frequently associated. The entire complex of the odontoid process, the atlas, and the clivus is rostrally located, and effectively the volume of the PCF is reduced. Although the authors mention that the clivus was reduced in length in 180 patients (50%), there is no mention of its rostral positioning or of platybasia, which is almost constantly associated with reduced PCF volume.
There is also no mention of partial or complete assimilation of the atlas, which is an important and frequent component of mesodermal maldevelopment. We have observed partial or complete assimilation of the atlas in approximately 50% of our cases. Basal mesodermal maldevelopment will result in rostral positioning of the plane of the foramen magnum and significantly severe basilar invagination if the measurements are taken on the basis of parameters laid down by Chamberlain (1), McGregor (5), and Fischgold et al. (2). However, the tip of the odontoid process will remain below both Wackenheim’s clival line (8) and McRae’s line (6) of the foramen magnum in a large percentage of cases.
The authors do not mention what criteria were used to measure basilar invagination. In the present series, basilar invagination was observed only in 12% of cases, which in our view is an unusually low incidence. We are convinced that more than one classically described criterion should have been used by the authors to evaluate basilar invagination. The use of sagittal images on computed tomographic scans or even plain radiography could have assisted the authors, as the bony architecture is not clearly delineated on magnetic resonance imaging (MRI). The MRI criteria described by us (3) and based on measurement of the distance of the tip of the odontoid process and the pontomedullary junction also could have been useful. Although the authors mention that the height of the “supraocciput” was reduced, there is no mention as to what criteria were used in its measurement. The term “supraocciput” used by the authors is interesting. The authors refer to the area of occipital bone adjoining the foramen magnum as supraocciput. According to our information, this area is commonly referred to as “subocciput.”
The retroflexion of the odontoid process was measured as an inclination of the tip of the odontoid process in relationship to the line drawn from the midbody of C2 to the basion in the midsagittal plane. In our view, in Chiari I malformation the rostral position of the odontoid is more predominant than retroflexion. Considering that the relationship of the tip of the odontoid process and the inferior lip of the clivus remains essentially unchanged in the maldevelopment, the criterion of the omega angle as described by Klaus in 1957 (4) and a modified omega angle described by us (3) seem to be better choices when one is measuring the retroflexion of the odontoid process.
The authors could have differentiated between retroflexion of the odontoid process and presence of a “fixed” atlantoaxial dislocation, which is also a frequently observed anomaly. The authors mention that the brainstem was “kinked” in 140 cases (39%). The exact meaning and the site of the kink are not elaborated. It also is not clear whether the anteroposterior dimension of the brainstem was compromised, either at the level of the tip of the odontoid process or at the level of the foramen magnum. We have observed that platybasia was as important as invagination of the odontoid process in causing the anterior “concavity” of the brainstem and in reducing the volume of the PCF. In our series, the anterior concavity of the brainstem was usually in the form of a smooth curvature and not a “kink,” and the anteroposterior length of the brainstem was only marginally reduced or unaffected in these cases, thereby indicating that there was no direct brainstem compression by the odontoid process (3).
The authors did not mention the volume of the foramen magnum. Some authors consider it an important parameter that would dictate surgical treatment. We observed that although the dimensions of the foramen magnum were large, and in the majority of cases larger than normally found, the volume of its contents and probably the longstanding “pulsatile” compression of the structures at the foramen magnum resulted in the neurological symptoms.
The report of 12% association of positive family history is also interesting and rather unusual. To date, no genetic abnormality has been detected in these cases. Only isolated case reports are available in the literature on familial relationship. In our series, there has been no associated positive family history in any case, but we accept that Milhorat et al. have evaluated this issue much more extensively.
The unusually high incidence of visual and otological problems (more than 75% of cases) is also unusual. The authors have placed extensive stress on even “minor” clinical features, but there is no mention of short neck or low hairline, which are the features of small PCF. Finally, no relationship with the radiological and clinical observations has been identified with the type of surgical therapy that would be indicated in these cases.
Atul Goel
1. Chamberlain WE: Basilar impression (platybasia): A bizarre developmental anomaly of the occipital bone and upper cervical spine with striking and misleading neurologic manifestations. Yale J Biol Med 11: 487–496, 1939.
2. Fischgold H, Lievre JA, Simon J: Indice radiog- raphique de profil de l’impression basilaire. Rev Rhum Ed Fr 26: 72, 1959.
3. Goel A, Bhatjiwale M, Desai K: Basilar invagination: A study based on 190 surgically treated patients. J Neurosurg 88: 962–968, 1998.
4. Klaus E: Rontgendiagnostik der platybasic und basilaren Impression. Fortschr Rontgenstr 86: 469, 1957.
5. McGregor M: The significance of certain measurements of the skull in the diagnosis of basilar impression. Br J Radiol 21: 171–181, 1948.
6. McRae DL: Bony abnormalities in the region of the foramen magnum: Correlation of anatomic and neurologic findings. Acta Radiol 40: 335–354, 1953.
7. Milhorat TH, Chou MW, Trinidad EM, Kula RW, Mandell M, Wolpert C, Speer MC: Chiari I malformation redefined: Clinical and radiographic findings for 364 symptomatic patients. Neurosurgery 44: 1005–1017, 1999.
8. Wackenheim A: Radiological diagnosis of congenital forms, intermittent forms and progressive forms of stenosis of the spinal canal at the level of the atlas [in French]. Acta Radiol Diagn (Stockh) 9: 481–486, 1969.
