2025 年 12 巻 p. 283-288
Foramen magnum decompression is an established surgical method for the treatment of Chiari type 1 malformation with syringomyelia. However, in some cases, neurologic symptoms that improve only after foramen magnum decompression may deteriorate again, and it has been suspected that increased instability of the craniocervical junction may be a factor. We report a case of Chiari type 1 malformation accompanied by basilar invagination and syringomyelia in which atlantoaxial facet joint distraction and fixation was performed as a revision surgery for deteriorating neurologic symptoms after foramen magnum decompression. The patient was a 16-year-old boy with chief complaints of gait disturbance and repeated falls. He was diagnosed with Chiari type 1 malformation accompanied by basilar invagination and syringomyelia. The clivo-axial angle was narrow at 105.8°. Initially, only foramen magnum decompression was performed, and the symptoms were relieved after the procedure but re-exacerbated within 2 weeks. The condition was speculated to have worsened instability at the atlantoaxial segment and ventral medullary compression; then, atlantoaxial facet joint distraction and fixation was performed secondarily, and symptoms improved. There is no clear surgical standard for performing foramen magnum decompression, atlantoaxial facet joint distraction and fixation, or a combination of both for Chiari type 1 malformation. Foramen magnum decompression provides horizontal decompression at the craniocervical junction, and atlantoaxial facet joint distraction and fixation achieves vertical indirect decompression of the ventral medulla and stabilization of the atlantoaxial segment in case with basilar invagination. Our experience suggests that combined foramen magnum decompression and atlantoaxial facet joint distraction and fixation may be particularly beneficial in Chiari type 1 malformation cases with basilar invagination and narrow clivo-axial angle.
Foramen magnum decompression (FMD) is an established surgical method for typical Chiari type 1 malformation (CM-1).1-3) This procedure enlarges the subarachnoid space and improves syringomyelia by resolving the obstruction of cerebrospinal fluid (CSF) flow at the craniocervical junction (CCJ). However, neurologic symptoms that improve after FMD may deteriorate again,3) and it has been suspected that the increased instability of the CCJ after posterior decompression may be a factor.4)
Some CM-1 cases are accompanied by basilar invagination (BI), where the tip of the odontoid process compresses the ventral medulla.2,4) Odontoidectomy as an anterior bony decompression has been used to effectively decompress the ventral medulla, but its higher possibility of complications such as surgical site infection, pneumonia, and respiratory troubles requiring reintubation is a major issue.5)
Goel reported the efficacy of atlantoaxial facet joint distraction and fixation (AFDF) for BI and CM-1 and stated that odontoidectomy was not required for BI.4,6-8) In this method, an artificial cage or allograft bone is implanted into the lateral atlantoaxial joint, and the tip of the odontoid process is withdrawn from the skull base (Fig. 1). This procedure achieves 3 objectives: decompression of the ventral medulla without odontoidectomy, improved CSF flow at the CCJ, and stabilization of the atlantoaxial segment.
In atlantoaxial facet joint distraction and fixation (AFDF), an artificial cage or allograft bone (asterisk) is implanted into the lateral atlantoaxial joint, and the tip of the odontoid process is withdrawn from the skull base (arrow) (A, B). Combined foramen magnum decompression and AFDF provides craniocervical junction decompression horizontally and vertically, respectively (C).
The efficacy of AFDF and its modifications for BI or atlantoaxial dislocation has been reported from various institutions.9-15) However, there are only a few reports that describe AFDF as a revision surgery for BI after FMD.12,16)
Herein, we present a case of CM-1 with BI and syringomyelia, where neurologic symptoms initially improved after FMD but subsequently deteriorated. AFDF was performed as a revision surgery, resulting in sustained symptom improvement.
The patient was a 16-year-old boy with psychomotor retardation, with a height of 1.48 m and body weight of 55 kg. Although he was able to complete a 3-km marathon at the age of 14 years, he developed an unsteady gait, with repeated falls at the age of 15 years. Clinical examination revealed clear consciousness, nystagmus, and mild dysphagia.
Manual muscle testing showed grade 4 weakness in the extremities, necessitating walker use. Deep tendon reflexes were normal with no pathologic reflexes. The plain radiograph demonstrated atlantoaxial intervals measuring 1.5 mm in the neutral position, 6.8 mm on flexion, and 1.4 mm on extension. Computed tomography showed that the tip of the odontoid process did not project above the foramen magnum; however, the clivo-axial angle (CXA) was narrow at 105.8° (Fig. 2). Magnetic resonance imaging showed ventral medullary compression by the odontoid process, descent of the cerebellar tonsils, and syringomyelia (Fig. 2).
Preoperative computed tomography image showing the narrow clivo-axial angle (A). Preoperative magnetic resonance T2-weighted image showing a syringomyelia in the cervical and thoracic spinal cord (B).
We speculated that CSF flow and ventral medullary compression would be improved by posterior bony decompression of the CCJ, and we only performed FMD initially. In the FMD procedure, the bony margin of the foramen magnum was removed in an oval shape, measuring approximately 20 mm rostrocaudal and 40 mm transverse dimension. The C1 laminectomy was also executed. The dura was incised in a Y-shaped fashion. Due to significant herniation of the cerebellar tonsils, partial tonsillar resection was performed. Duraplasty was accomplished using an artificial dural substitute.
Within 1 week after the operation, the patient was able to walk 60 m continuously using a walker, and the syrinx had shrunk (Fig. 3). However, 2 weeks after FMD, the patient developed acute obstructive hydrocephalus, resulting in loss of consciousness, respiratory dysfunction, and quadriplegia (Fig. 3). After artificial respirator management with intubation and ventricular drainage, consciousness improved but quadriplegia persisted. A ventriculoperitoneal shunt was performed approximately 2 weeks after ventricular drainage.
Magnetic resonance T2-weighted image showing shrinkage of the syrinx after foramen magnum decompression (A). Head computed tomography (CT) image obtained before foramen magnum decompression (B) and head CT image taken before ventricular drainage surgery showing ventricular enlargement (C).
This clinical course suggested that atlantoaxial instability was aggravated after FMD, leading to exacerbated ventral medullary compression and subsequent obstructive hydrocephalus and quadriplegia. Therefore, we performed AFDF to address ventral medullary compression by the odontoid process and atlantoaxial instability.
AFDFBecause titanium cages specialized for the procedure are not commercially offered in Japan, approval for the creation and use of custom-made cages was obtained from the institutional individual case review committee. The sizes of the titanium cages and dedicated inserters used in this procedure were determined based on preoperative images, and they were manufactured by a medical device manufacturer (Meistec Co., Ltd., Tokyo, Japan).
Surgery was performed under general anesthesia, with the patient placed in the prone position. The previous skin incision of the initial FMD was reopened, and the bilateral lateral atlantoaxial joints were exposed. Bilateral C2 nerves were transected during this process. The cartilage of the atlantoaxial joints was removed with a curette, and a titanium cage (width, 8.0 mm; length, 10.0 mm; height, 5.0 mm) was inserted into both joints using intraoperative fluoroscopy. The lateral mass and lamina screws for C1 and C2, respectively, were inserted and fixed (Fig. 4).
Custom-made titanium cages and a dedicated inserter (A, B). Plain radiograph and computed tomography (CT) images after atlantoaxial facet joint distraction and fixation (AFDF) showing titanium cages placed into the lateral atlantoaxial facet joint, C1 lateral mass screw, and C2 lamina screw (C, D, E). CT image after AFDF showing increased CXA (F).
After AFDF, quadriplegia improved and CXA increased by 126.0° (Fig. 4). However, dropped head posture gradually developed because of muscle weakness in the occipital and cervical regions. Therefore, posterior fixation was performed from the occiput to the upper thoracic spine. The patient was able to walk >100 m with a walker 1 year after the final surgery and 300 m without a cane and 1,200 m with a walker 2 years after. The 10-m walk test, Japanese Orthopedic Association score, and functional independence measure 2 years after surgery improved from preoperative measurements (18 to 8 s, 13.5 to 14 points, and 92 to 107 points, respectively).
Informed consent was obtained from the patient and his families for the establishment of this paper.
This case demonstrated the successful application of AFDF as revision surgery for CM-1 with BI and syringomyelia, where initial post-FMD improvement was followed by symptom re-exacerbation. Although FMD achieved syrinx shrinkage, it potentially exacerbated atlantoaxial instability and ventral medullary compression. AFDF, followed by supplementary posterior fixation for dropped head posture, led to sustained neurologic improvement over a 30-month follow-up period.
FMD's effectiveness in typical CM-1 cases is well-established, particularly, in enlarging the subarachnoid space at the CCJ, improving CSF flow, and shrinking syringomyelia.1,3) However, post-FMD symptom recurrence is a recognized phenomenon.3,7) Goel suggests that underlying atlantoaxial instability may be a fundamental factor in CM-1, noting that intraoperative findings often reveal instability not evident on preoperative imaging.4) This observation suggests that bony decompression alone, whether through FMD or odontoidectomy, may be insufficient or potentially detrimental in cases with atlantoaxial instability.3,4,15) However, current literature offers limited guidance on revision strategies post-FMD, with few reports specifically addressing AFDF as a revision option for BI or atlantoaxial dislocation cases with post-FMD deterioration.12,16) Klekamp2) recommended FMD combined with posterior fixation for CM-1 with instability, ventral medullary compression, or skeletal anomalies at the CCJ, whereas Goel4) stated that FMD was not necessary for CM-1, regardless of the presence or absence of BI or syringomyelia and that the symptoms could be improved by AFDF alone.
Although there is underlying instability, fixation surgery such as AFDF alone may not be sufficient to treat all types of cases, and some may require a bony decompression procedure. The combination of FMD and AFDF potentially offers complementary benefits: FMD provides horizontal decompression at the CCJ, whereas AFDF achieves vertical decompression (Fig. 1). Feng et al.15) reported the efficacy of atlantoaxial distraction and fixation for 30 cases of BI, stating that "stabilization without decompression is not perfect, and decompression without stabilization is not persistent."
CXA, the angle between the clivus and axis, may also be a factor in determining the surgical strategy. CM-1 cases with BI and a CXA <130°, as presented here, show that FMD alone may result in residual obstruction of CSF flow and delayed shrinkage of the syrinx.1) This possibly reflects that BI with a CXA <130° is associated with a narrower CSF space in the CCJ and stronger medullary compression. Therefore, combined FMD and AFDF may be particularly beneficial in such cases.
Technical considerations in AFDF include cage height optimization. If only a low-height cage/bone graft can be inserted into the lateral atlantoaxial joint, odontoid tip extraction may be insufficient, and combined FMD may play a complementary role in decompressing the CCJ. However, excessive cage/bone graft height risks neurologic complications from cord stretching. Limiting cage/bone graft height to 10 mm under intraoperative neurologic monitoring may be safe.17)
Potential AFDF complications warrant attention. C2 nerve resection during lateral atlantoaxial joint exposure typically causes minimal postoperative sensory disruption,9,18) similar to the present case. More serious risks include dural and vertebral artery injury during soft tissue dissection and cage deviation to the outside of the lateral atlantoaxial joint, resulting in obstruction of vertebral artery flow. Patients should understand that fixation procedures restrict cervical range of motion. There has been a case report in which additional odontoidectomy was required for neurologic deterioration due to cage subsidence after AFDF.17) Considering that a cylindrical cage was used in this case, a cage/bone graft with a larger footprint may be preferred to prevent postoperative cage subsidence. The dropped head syndrome observed in our case, although previously unreported, may relate to the patient's small stature and a small amount of muscle mass. This experience suggests that 1-stage FMD and AFDF may be preferable to minimize cumulative muscle damage.
In conclusion, this case demonstrates successful AFDF revision surgery after post-FMD symptom re-exacerbation in a patient with CM-1 with BI and syringomyelia. Our experience suggests that combined FMD and AFDF may be particularly beneficial for CM-1 cases with BI and narrow CXA. Further research should focus on defining precise surgical indications and verification of long-term outcomes of this combined approach.
All authors have no conflict of interest.
