Posterior decompression and occipitocervical fixation followed by intraoperative vertebroplasty for metastatic involvement of the axis

The spine ranks first in the list of cancer metastasis in the skeletal system. Approximately 8-20% of metastases occur in the cervical spine [1‐3]. However, metastases to the upper cervical spine were rarely reported in the literature [4, 5]. Due to the importance of this area, metastatic involvement of the axis may cause spinal instability and cord compression, which in turn can result in respiratory failure and neurological dysfunction. For advanced patients, the aim of surgical intervention was not to cure the disease but to relieve the pain and spinal cord compression, reconstruct spinal stability, and improve the quality of remaining life. The treatments included external immobilization, conventional surgical resection and percutaneous cement augmentation. However, conventional surgical treatment itself had risks of mixed infection, speech change, tongue oedema, cerebrospinal fluid leakage, etc., due to poor clinical conditions, comorbidities, and its transoral approach, while percutaneous vertebroplasty in this area was challenging [6, 7].

Therefore, combining intraoperative vertebroplasty and posterior decompression and occipitocervical fixation may be advisable, which is less invasive and could provide strong stability. To our knowledge, palliative surgical treatment of such disease has rarely been reported. This palliative method was used to treat 10 patients with metastatic involvement of the axis in our study and produced desirable outcomes.

Metastasis to the spine are commonly spread through the Batson plexus, a valveless vertebral venous complex. Additionally, the vertebral body is the most common site to be affected compared with the posterior elements [10]. In our study, all cases had an affected vertebral body of C2. The clinical picture of metastatic involvement of the upper cervical region ranges from localized pain to a variable extent of the neurological deficits and even death induced by spinal cord compression [11, 12]. Occipitocervical pain occurred in approximately 89% to 100% of patients, was not linked to daily activities and worsened in the night [13‐15]. All patients in our study presented with occipitocervical pain. Furthermore, given the relatively larger sagittal diameter of the C1 and C2 level compared to the counterpart of the subaxial cervical spine, the neurological deficit was overt only in the later stages. Hence, the preoperative JOA score was 10.1 ± 2.2 in our study.

With the development of internal fixators and surgical techniques, more invasive approaches were recommended for patients with upper cervical metastases who met strict selective criteria. The aim of the surgery for upper cervical metastases was to decompress the spinal cord, reconstruct spinal stability and obtain pain relief. The main goal was improving the quality of the remaining life rather than cure the disease. Previous studies suggested that metastases to the cervical spine have a worse prognosis than metastases to other sites [16, 17]. However, a recent matched pair analysis showed that metastatic spinal cord compression had similar clinical outcomes to those of thoracic and lumbar spine metastasis in terms of postoperative survival, surgery-related complication, and pain outcome [18].

Total en bloc spondylectomy (TES) was widely used to avoid local recurrence of spinal metastases. However, TES also had adverse aspects, such as a longer operation time, massive blood loss and difficulty to perform in the upper cervical spine because of the distinct anatomy of the vertebral arteries and the complex bony architecture. In addition, the transoral approach to the operative field was deep and narrow and many complications may occur, such as mixed infection, speech change, tongue oedema, retropharyngeal haematoma or cerebrospinal fluid leakage [6]. Sakaura et al. [19] described that the recurrence rate was similar for palliative surgery compared with TES. In our study, we only observed one patient with recurrence.

Vertebroplasty or kyphoplasty was widely used on spinal metastases and had gained effective clinical results in patients with osteolytic spinal metastases. Recently, kyphoplasty was even applied in osteoblastic spinal metastases and caused a significant reduction of pain [20]. The mechanism of pain relief was thought to be that bone cement stopped vertebral micro- or macromovement and the destruction of pain receptors through exothermic reactions and chemical and toxic effects [21]. Percutaneous vertebroplasty (PVP) has a shorter operation period, minimal blood loss, rare intra-operative injuries and low cost [22]. The relief of pain was commensurate with an increase in activity, which in turn decreased the risk of concomitant infections and the need for further hospitalization. In our study, all patients experienced a certain extent improvement in occipitocervical pain and a decrease in VAS from 8.2 ± 0.4 to 2.3 ± 0.2. However, this method could not address the problem of spinal cord compression. Interestingly, several studies demonstrated that PMMA had an antitumoural effect, which may be due to the cytotoxicity, thermal effects and ischaemic effects [23, 24]. However, others proposed that the low recurrence rate was attributed to the short survival time rather than the antitumoural effect. Hence, the antitumoural effect needs further study. In addition, transoral PVP also had the demerits itself.

Therefore, posterior surgery combined with intraoperative vertebroplasty may be a feasible and less invasive method. Intraoperative vertebroplasty was accurate and visual to inject bone cement. In our study, cement extravasation was only detected in 1 case, but with no clinical significance (10%).This was overtly lower than PVP (51%) [25]. According to our experience, it was important that the cement injection was under high viscosity, with intraoperative fluoroscopy and preoperative evaluation of the integrity of the vertebral endplates and posterior wall of the vertebrae. Furthermore, laminectomy and removing the pedicles could provide a better way to perform vertebroplasty and the possibility of an immediate clearing of the leaked bone cement.

In our study, our patients had either impending instability or an unstable spine according to SINS, of which sensitivity and specificity was 96% and 80%, respectively [26]. If patients were in good clinical and stable oncological condition, extended posterior fixation (O-C5) was highly recommended [14]. Patients in our study were long-term bedridden and malnourished, with a mean age of 68.9 years. In addition, the upper cervical spine lacked the support of ribs, which was different from the thoracic spine. Thus, all these factors contributed to the decision to perform an extended posterior fixation. This could achieve the maximum and most long-lasting stability. The shorter posterior fixation might later become insufficient and increases the risk of reoperation. Due to the nature of palliative surgery, we do not ordinarily perform a bone graft. This was consistent with a previous study [27].

Lung, breast and prostate cancer were the 3 main tumours that metastasized to the spine [28, 29], which accounted for 80% in our study. The average survival time of our study was 12.8 months, which was within the range of previous studies, 3-18 months [2, 30]. Patients with primary tumours from lung and prostate had the worst prognosis and a relatively shorter survival time. This was consistent with previous studies [31]. It was worth noting that postoperative radiotherapy and chemotherapy were also important in terms of prolonging life and improving the quality of life. Treatment of spinal metastases is challenging and needs multidisciplinary collaboration.

There were several limitations. First, there was the nature of the retrospective study. Second, the number of patients included was relatively limited due to the rarity of this disease. Third, there was the absence of the control group. A larger controlled study is required to identify the efficacy of this method.

Posterior decompression and occipitocervical fixation followed by intraoperative vertebroplasty was a safe and valuable palliative method with relatively less invasion to treat metastatic involvement of the axis.