Subaxial cervical metastases are known to have a poorer prognosis than metastases to the thoracic or lumbar regions.[15] Despite the potential complications, operative treatment for subaxial cervical metastasis has been shown to be more effective than nonoperative treatment for relieving pain, and may also reverse neurologic deficits and improve ambulatory function.[16, 17] Operative treatment options in this region include the anterior approach, the posterior approach, or the combined anterior and posterior approach. Fehlings, et al. conducted a systematic review and recommended the anterior approach for most subaxial cervical spine metastases, and the posterior approaches for most craniovertebral metastases.[7] Combined anterior and posterior approach is recommended in circumferential tumor involvement and poor bone quality.[7] However, data on the perioperative complication profiles of these three approaches is lacking. Therefore, the purpose of the current study was to compare the clinical results among these three standard approaches. This information may help to guide decision-making for surgical treatment of subaxial cervical spine metastases.
Previous studies reported low complication rates for the anterior surgical approach.[2-4, 8, 9] Jonsson, et al. reported complications, such as dysphagia, reversible vocal cord paralysis, and a postoperative mortality rate of about 2%.[3] Heidecke, at al. reported complications in 62 patients including reversible vocal cord paralysis (8%), early instrumentation failure (4.8%), thrombosis and embolism (4.8%), wound infection (3.2%), and neurological deterioration (6.5%).[2] Oda, et al. reported results from 32 patients (25 for posterior fixation alone, and 7 for combined approach) and found an overall complications rate of 19% [Radiculopathy from screw malposition (3%), deep wound infection (3%), postoperative hematoma (3%), cerebrospinal leakage (3%), and one patient in combined group (14%) had displacement of an anterior strut and required revision surgery]. In the current study, the overall perioperative complication rate was slightly higher because more types of complications were included in our analysis. For the subgroup analysis, even though there was no difference in preoperative patient characteristics among groups, the perioperative complication rate was highest in the combined approach group (75%). This may be due to the fact that the combined approach is the most extensive surgical approach, and it has the longest operative time and the most blood loss.
There are three main types of pain that are experienced by patients with spinal metastasis. An expanding tumor can cause periosteal ‘stretching’, which leads to constant localized pain and compression of nerve roots, which in turn leads to radicular pain. Axial pain is associated with pathological vertebral body fractures that can cause spinal instability.[18] Surgical treatment can reduce the tumor size and stabilize the vertebral column, which results in pain reduction. Previous studies reported decreased pain as measured by pain scale after surgical treatment. Cho, et al. analyzed the results of 46 patients treated with surgery for cervical spine metastasis and reported reduced pain in 37 of 44 patients (84.1%). They found that the mean visual analog scale (VAS) for pain decreased from a preoperative 7.86±1.05 points to a postoperative 4.48±2.09 points, which represents a decrease of 3.39±2.14 points (p=0.001). The only factor found to influence pain improvement was neural foramina invasion by preoperative MRI.[19] Rao, et al. reported on a series of 11 patients who underwent surgical treatment for symptomatic cervical spinal metastasis (5 anterior approach, 4 posterior approach and 2 combined approach). All patients experienced reduced axial neck pain according to patient postoperative VAS scores taken at the 1-month follow-up after surgery.[20] The current study found greater than 75% improvement in the VNS in all approaches, with the greatest improvement in pain observed in the combined approach group (87.5%). Due to extensive soft tissue dissection, the combined approach provides more decompression and stability compared to the two other approaches, so this may explain the greater decrease in pain in this group. Gallazzi, et al. reported a case series that treated thirty cervical spinal metastasis patients using posterior-only laminectomy and posterior stabilization. Their results showed a strong significant improvement in pain score (5.5±1.8 to 2.1±1.0, p<0.00001). After average follow up of 13.7±14.8 months, fifteen (50%) patients died, 2 (6.9%) had surgical-site infection that required reoperation, and there were no mechanical failures reported.[21]
Cho, et al. reported neurological improvement in 75.7% (28/37) preoperative neurological deficit patients.[19] In the current study, neurological recovery was highest in the combined approach group (33.3%), which may be explained by more adequate decompression. However, the high rate of neurological recovery in the combined group could have been confounded by preoperative neurological status. Most patients in the combined group were ASIA class D (75%) compared to 55.6% in the anterior group, and 42.9% in the posterior group. The overall neurological recovery rate in our study was quite low when compared to the rate reported by Cho, et al.; however, this difference between studies may be due to the fact that our follow-up period was shorter.
Regarding ambulatory status improvement, Denaro, et al. reported improvement in 5 of 18 (27.8%) nonambulatory patients (Frankel B/C) who later became ambulatory (Frankel D) after surgery.[22] In the current study, 12 of 70 nonambulatory patients (17.1%) became ambulatory with the highest rate of improvement in the combined group (33.3%). This result may be related to the better neurological recovery rate observed in the combined group. Heidecke, at al. reported survival after surgery of 58% and 21% at 1 year and 2 years, respectively, after surgery.[2] Cho, et al. found primary tumor growth rate, preoperative Tomita score, radiotherapy (RT), timing of RT, and postoperative adjuvant treatment to be related to longer survival.[19] In the current study, despite the statistically similar preoperative Tomita score among the three approaches, the median survival time was longest in the combined group.
Limitations
The primary limitation of this study is it retrospective design, which rendered it vulnerable to missing or incomplete data in some cases. Another notable limitation is that our relatively small study population was recruited from only two centers. This could limit the generalizability of our findings to other care setting, and the small sample size may have impeded our ability to statistically reveal all existing differences and associations among the 3 evaluated surgical approaches. Third and last, the relatively short follow-up time could have prevented us from uncovering advantages or disadvantages of one approach over the other that may emerge over time. A prospective multi-center study with a longer follow-up period is warranted to confirm the findings of this study, and to identify additional information that may enhance surgical approach-related decision-making in this patient population.