in Craniovertebral Junction Instrumentation: Hardware Removal Can Be Associated with Long-Lasting Stability. Personal Experience

Fig. 1

Sagittal T2-weighted magnetic resonance (MR) image (a) and computed tomography (CT) scan sagittal reconstruction; (b) demonstrates platybasia, cerebellar tonsil herniation into the foramen magnum, and upper displacement of the odontoid process, along with compression of the medulla oblongata. Postoperative CT scan sagittal reconstruction (c) shows craniovertebral junction (CVJ) decompression by means of odontoidectomy performed with the transoral approach. CT scan sagittal reconstruction (d) shows complete bone fusion and secondary stabilization of C0-C1-C2-C3. Dynamic sagittal T2-weighted MR images (e–f) confirm the CVJ stability

Case Presentation 2

A 52-year-old man, with a few months’ history of cervical pain, underwent CT and MR scans of the CVJ with contrast medium, showing an osteolytic process involving the axis, with complete infiltration and ballooning of C2 extending towards the anterior arch of C1 and the inferior edge of the clivus (Fig. 2a–b). Possible multiple localizations of the neoplastic disease were ruled out by a total-body CT scan and Technetium-99 m sestamibi bone scintigraphy. Neither monoclonal gammopathy nor Bence-Jones proteinuria was found. Radiological findings were suggestive of a high risk of axis fracture with potential CVJ instability. Biopsy of the lesion was performed and extemporary histopathological examination was consistent with plasmacytoma. A CVJ instrumentation and fusion procedure was performed via C0-C3-C4-C5 with lateral mass screws (SUMMIT™ SI OCT System; Codman Johnson & Johnson, Leeds, England) and rods; also in this patient a synthetic bone graft substitute (ß-TCP; VITOSS® Synthetic Cancellous Bone Void Filler) was locally applied to achieve fusion. A post-operative CT scan documented the correct placement of the device, in the absence of CVJ dislocation (Fig. 2c). After 1 month, the patient underwent radiotherapy (RT) consisting of the administration of 200 centigray (cGy) for 20 consecutive days (6 MV-photons) to C1–C2–C3 vertebrae (total dose on the tumor core 4,000 cGy). Prednisone administration was started (25 mg daily for the first week; 12.5 mg daily for the second week; 12.5 mg on alternate days for 3 weeks). Six months after the RT, the patient complained of left cervical pain and local paresthesia. Cervical CT scan showed partial bone resorption with the dislocation of screws and rods, along with atlantoaxial rotatory subluxation (Fig. 2d). The dislocated screws and rods were removed and Songer’s titanium sublaminar wires (SUMMIT™ SI OCT System; Codman Johnson & Johnson USA) were placed at the C3–C4–C5 levels and connected with an occipital plate with rods. Two cross-link bars were also placed, at the C0 and C5 levels, in order to reduce the risk of rotatory subluxation. Autologous bone was harvested from the posterior left iliac crest, cut in a double-wing shape, and fixed to the CVJ using a silk suture, synthetic bone dust, and human fibrin glue (Tissucol; Baxter, West Lake Village, CA, USA). A Halo-Vest System was then applied. No signs of local infection were found. The patient regained walking ability quickly. Postoperative CT scan confirmed the correct placement of the device and the restoration of vertebral alignment. Oral bisphosphonate (BP) therapy with zoledronic acid was instituted (Zometa®; Novartis Pharmaceuticals) at a dose of 4 mg IV every 4 weeks. The Halo-Vest was maintained for 3 months. A 30-month follow-up CT scan and dynamic X-ray study documented correct vertebral alignment and fusion due to successful instrumented surgery, with no signs of CVJ instability. A cervical MR scan showed tumor remission, and a new bone marrow needle biopsy excluded metastatic diffusion.

Fig. 2

Sagittal CT scan (a) and MR studies (b) show the advanced osteolytic process exclusively involving the axis at the level of the soma and the dens, especially the trabecular spongiosa region, evidencing the distorted profile of the axis without spinal cord compression. Specific axial CT scan study of the C5 vertebra after the first intervention demonstrates the right position of the screws in the lateral masses; absence of osteolytic areas around the intraosseous screw course is also evident (c). Differently, axial CT scan after radiotherapy (RT) shows signs of bone resorption and evidence of left screw dislocation on the C5 vertebra (white arrow) (d)

Case Presentation 3

A 54-year-old patient presented with a medical history of neck pain with lower limb weakness, ataxia, and paresthesia in four limbs, of several months’ duration. A CVJ MRI examination demonstrated os odontoideum instability with compression of the medulla oblongata. In May 2000, CVJ instrumentation with C0 suboccipital and C2-C3 sublaminar wires (Songer cables) and U-shaped rods was performed, along with synthetic bone graft substitute fusion (ß-TCP; VITOSS® Synthetic Cancellous Bone Void Filler). The patient was advised to wear a hard collar. Some days after the surgery patient experienced acute neck pain associated with the fixed neck position in hyperextension. Neck X-ray examination documented a displacement of the occipital wires, sliding caudally on the U-shaped rods, which had resulted in deformation—hyperlordosis (Fig. 3a). The patient underwent a second surgical procedure to correct the dislocation; this was done by performing a new operation with a different U-shaped construct (Fig. 3b–d).

Fig. 3

Early postoperative lateral X-ray demonstrates the sliding of the titanium construct associated with the pulling out of the hardware (a). Postoperative X-rays (CT scout view) (b–d) show control after reoperation

Discussion

The causes of CVJ instability include CVJ trauma, rheumatological diseases, tumors, infections, congenital malformations, and degenerative disease processes. Considering dysmorphic pathologies, basilar invagination is the most common bony abnormality (38–74 %) of the CVJ, followed by platybasia, often associated with other findings such as Chiari malformation, syringomyelia, and os odontoideum [23, 30]. Instability of the CVJ may lead to significant pathological problems, including cervical pain at first, followed by all the consequences of cervical cord compression: respiratory distress, cranial nerve dysfunction, paresis, and plegia, or even sudden death. Fusion procedures at the CVJ must be capable of withstanding the forces of compression, axial loading, flexion, extension, lateral rotation, and lateral bending. A variety of techniques are used for CVJ instrumentation and fusion: rigid rod-screw fixation, rod-wire systems, occipital hooks, and cervical claws. These techniques have shown comparable effectiveness (fusion rates from 89 to 100 %). Despite possible alternative less invasive surgical procedures sparing occipital bone fusion, CVJ instrumentation and fusion remains the most appropriate treatment when dealing with instability or in cases of widespread bone destruction, fractures, or progressive inflammatory or metabolic diseases [6, 8, 17, 20, 21]. The choice of surgical procedure is mainly based on radiological parameters, such as occiput bone density or the firmness of posterior cervical elements, but also considering the surgeon’s experience. Biomechanical in-vitro experiments have demonstrated for many years that screws provide better immediate stability than wires, but it is still debated whether this results in a true higher rate of fusion. Reports of a fusion rate of 80 % after wiring and of 94 % after screwing do not seem to influence markedly the final clinical results [9, 18, 19, 24]. Wiring constructs remain an excellent method for stabilizing the CVJ and upper cervical spine. When wired to the spine and skull, bone struts or metal implants provide reasonably good mechanical stabilization properties with a low incidence of complications [2]. Ductility, resistance to stress, and the possibility of using MR postoperatively convey advantages in using titanium, compared with other metals, in osteosynthesis. The most frequent complications related to CVJ instrumentation and fusion include dislocation and rupture of the fixation system, screw loosening, dural fistula, neural or vascular damage, wound infection, and the persistence of neurological pain [1, 13]. In more detail, the most commonly encountered perioperative complications have been related to instrumentation failure after nonunion, with rates as high as 7 % during CVJ instrumentation and fusion and 6.7 % during atlantoaxial fusion (quite the same!). Other commonly encountered complications have included injury to the vertebral artery (1.3–4.1 % during the placement of C1-C2 transarticular screws, most commonly in the case of high-riding vertebral artery [13]. In our experience, we can affirm that CVJ instrumentation procedures can be considered efficient and substantially safe; in the 48 patients who underwent CVJ posterior instrumented procedures at our Institution, the three cases of instrumentation failure, are discussed and compared with related results in the current literature as outlined below.

Case Considerations

Case 1

Postoperative infection of a stabilization system is considered a serious complication, with an incidence rate of 0.1–3 %, and it mostly occurs 3–7 days after surgery, with a higher incidence via the transoral approach [10]. Pediatric patients with postoperative wound infections requiring surgical debridement have higher surgical failure rates after CVJ instrumentation and fusion. Those with skeletal dysplasia and congenital spinal anomalies are more likely to require reoperation for hardware failure, as confirmed in our case [15]. Moreover, in a series of pediatric patients, the efficacy of surgical wash-out associated with antibiotics was demonstrated; this treatment was successful in curing wound infections in many patients without hardware removal [11, 13]. Bathia et al., in a total of 100 patients, reported three cases of infections treated with wound washout, and in one case only, hardware removal was performed (3 %), without any subsequent radiological confirmation of long-lasting stability. Choi et al. also reported one case of hardware removal with referred CVJ bone fusion, but no radiological evidence of the fusion was shown in their report [5, 7]. Similar conclusions were reached by Ahmed et al., who presented the case of a 20-year-old patient who developed postoperative infection after CVJ instrumentation and fusion in Chiari I malformation. After removal of the instrumentation, solid bony fusion was evidenced, but it was not confirmed by later dynamic cervical radiographs [1]. As far as we know, our case is the first documented condition of hardware removal followed by stability due to bone fusion confirmed by postoperative dynamic neuroradiological investigations. These data support our finding: we can hypothesize that the infection, “long-lasting” as in our patient, had a role in the ossification process involved in the CVJ fusion, since the fusion occurred 33 months after onset of the infection. Finally, and very surprisingly, in our case, the post-infective bone fusion not only produced a good fixation but also resulted in a sort of odontoid regeneration, never reported before. Although we recently reported a “true” odontoid process regeneration (along with clival regeneration and Chiari malformation recurrence) after transoral decompression, in the present case we observed the union of the remaining C2 with C3 bodies, strongly mimicking a concomitant quite complete axis and clival regeneration [26].

Only gold members can continue reading. Log In or Register to continue