Dysplasia and anomalies of atlas result in pediatric torticollis: A case report and literature review

INTRODUCTION

Torticollis may result from a congenital craniovertebral junction (CVJ) anomaly. Here, in a 6-year-old child with progressive torticollis, we recognized a dysplastic C1 vertebra showing assimilation to the right occiput and C2, accompanied by a nonseparated left odontoid, and discontinuity in both anterior and posterior arches of the atlas. This was successfully managed with a C1-C2 Goel and Harm’s fusion.

CASE DESCRIPTION

A 6-year-old female presented with a 3-year history of a progressive torticollis toward the right side. It was associated with facial asymmetry including a smaller right eye [Figure 1]. There was no history of any prior untoward events during the pregnancy. The patient had no neurological deficits. The radiographs revealed evidence of fusion between the C2 and C3 vertebral bodies with a dysplastic C1, while the CT scan confirmed an absent C1 spinous process and adjacent lamina (posterior arch deficiency) [Figure 2]. The CT scan showed additional bony anomalies as explained in [Table 1] and [Figures 3 and 4]. MRI showed no evidence of cord compression [Figure 5].

Close

Figure 1:: Clinical photograph showing right-sided torticollis with features of facial asymmetry including smaller eye size.

Export to PPT

Close

Figure 2:: 3D CT scan showing absent C1 spinous process with adjacent lamina (posterior arch deficiency) and complete fusion of posterior elements of C2 and C3.

Export to PPT

Close

Figure 3:: 3D CT scan showing occipitalization of the right side of lateral mass of C1, discontinuity in anterior arch of C1, left side lateral mass of C1 with residual anterior arch of C1 showed fusion with body of C2 anteriorly with minimal space between left lateral mass of C1 and C2 in contrast to the right side.

Export to PPT

Close

Figure 4:: CT scan showing subtraction view showing odontoid nonseparation and its assimilation with anterior arch of atlas.

Export to PPT

Close

Figure 5:: T2 Sagittal image sequence of MRI showing no evidence of cord compression.

Export to PPT

Table 1:: Anomalies as evident on CT scan.

Surgical management

The progressive torticollis was treated with a C1 and C2 Goel and Harm fusion. During the surgery, on the right side, the interfacet space between C1 and C2 was large in contrast to the left side. Articular cartilage was debrided and cancellous bone grafting (e.g., harvested from the posterior iliac crest) was performed.

Similar steps were followed on the left side. Next, C1 lateral mass and C2 pedicle screw were inserted with increased attention to the dysplastic C1 and C2 anatomy. The screw trajectory on the right side was kept high in sagittal plane to adjust for occipitalization of the right lateral mass of the atlas. As there was fusion between the C1 and C2 bodies, we passed larger bilateral pedicle screws into the C2 vertebral body [Figure 6].

Close

Figure 6:: Immediate postoperative lateral and anteroposterior X-ray showing C1 C2 fixation by Goel and Harm technique.

Export to PPT

Intraoperative reduction/compression was performed on the right side between the C1 and C2 screws to correct the torticollis [Figure 7]. Two years later, the patient remains asymptomatic [Figure 8].

Close

Figure 7:: Schematic representation of intraoperative reduction maneuver.

Export to PPT

Close

Figure 8:: Clinical photograph on postoperative day 2 showing correction of torticollis and maintained in cervicothoracic orthosis.

Export to PPT

DISCUSSION

It is extremely rare to find dysplastic C1 vertebrae causing torticollis without rotatory subluxation of C1 and C2. A report by Geipel[3] stated that incidence of defects in posterior arch of atlas was about 4%, and defects of the anterior arch of the atlas were 0.1%.

Cave had classified atlantoaxial congenital fusion into three types: (1) fusion of a separated odontoid process with the anterior atlantal arch; (2) complete (bilateral) fusion of atlas and axis; and (3) incomplete (unilateral) fusion, with or without some degree of assimilation.[1] The present case was fitting in to subclass of 3 of Cave classification with regard to atlantoaxial congenital fusion in addition to other congenital anomalies. It is extremely rare to find fusion of a nonseparated odontoid process with the anterior arch of the atlas.

Previously published literature has reports about bipartite atlas. However, in our case, atlas cannot be called bipartite, as defect on anterior side was not in midline and there was unilateral assimilation of atlas with axis and occiput. As C2-C3 had fused from both anterior and posterior side symmetrically, it did not have any effect on torticollis caused at CVJ.

Occipitalization of the atlas occurs as a result of the failure of segmentation between the fourth occipital sclerotome and the first cervical sclerotome.[4] Hemiaplasia or hypoplasia of the atlas can cause a marked torticollis, which if untreated, constitutes an unacceptable deformity. Dubousset described three forms of the anomaly.[2] Type I is characterized by an isolated hemiatlas; Type II, by partial or complete aplasia of one hemiatlas with congenital fusions in the cervical spine; and Type III, by partial or complete occipitoatlantal fusion and symmetrical or asymmetrical hemiatlas aplasia with or without anomalies of the odontoid or other cervical vertebrae.

The distinguishing feature in our case was the torticollis caused by differential growth between C1 and C2. (minimal growth due to fusion of C1 and C2 on the left half and continuous growth on the right side between occipitalized C1 lateral mass and C2). Our patient underwent a C1-C2 fusion using the Goel and Harm technique to reduce the torticollis and achieve stability [Figure 6].

CONCLUSION

Torticollis caused by anomalous/differential growth of C1-C2 vertebrae resulting in a nonrotatory subluxation is exceedingly rare. Here, a 6-year-old female with a dysplastic C1 and multiple other congenital anomalies required a C1-C2 Goel and Harm fusion to reduce the torticollis.