Methods
Approval of the institutional review board of our institution was obtained prior to the current study. Patients with spinal metastases or proposed surgery involving C7-T2 were included. There were 36 patients with CTJ metastases who underwent surgery in our department between May 2012 and December 2015. However, two patients lost their follow-up at 3 and 7 months after surgery respectively, and they were not included in this research. At last, a total of 34 patients were included and retrospectively analyzed in our study. The diagnosis of spinal metastasis was confirmed histologically, and diagnostic imaging including X-ray, computed tomography (CT), or magnetic resonance imaging (MRI), as well as PET-CT. The selection criteria for surgical intervention required the patients to have at least one of the following conditions: (1) Spine Instability Neoplastic Score (SINS) > 6, indicating spinal instability requiring surgical reconstruction; (2) significant or progressive neurological deficits/or no neurological recovery under 2-weeks’ non-surgical treatment; and (3) intractable pain under 2-weeks’ treatment of pain relief medication. Patients’ choice is also an important consideration, e.g., patients who rejected surgery were not included in our study. Patients excluded from surgery were those with an estimated survival less than 3 months or poor health situation to undergo surgery. All patients who underwent surgery have met the above indications.
The medical characteristics were retrospectively analyzed for demographic, clinical, radiographic, and histological data. The location of the spinal lesions was assessed using magnetic resonance (MR) imaging and plain X-rays. The modified Tokuhashi score and SINS were applied to evaluate the patients’ prognosis and spinal stability [
8,
9]. The Frankel grading system and visual analog scale (VAS) were used to assess the neurological signs and the severity of pain [
10,
11]. The patients’ life quality was assessed by Karnofsky Performance Status (KPS). The demographic and clinical characteristics of these patients are illustrated in Table
1. The operative data included surgical approach/procedure, blood loss, and perioperative complications. Blood loss included direct blood loss during surgery and blood loss until removing the tube.
Table 1
Demographics and clinical characteristics of study patients
Age (years) |
Median | 54 |
Range | 34–72 |
Gender |
Male | 17 (50) |
Female | 17 (50) |
Primary tumors |
Lung | 13 (38.2) |
Breast | 7 (21.9) |
Prostate | 6 (17.6) |
Gastrointestinal | 2 (5.9) |
Other | 6 (17.6) |
Number of involved vertebrae |
1–2 | 26 (76.5) |
> 2 | 8 (23.5) |
Visceral metastases |
No | 23 (67.6) |
Yes | 11 (32.4) |
Frankel score at entry |
A, B, C | 18 (52.9) |
D, E | 16 (47.1) |
Modified Tokuhashi score |
0–8 | 22 (64.7) |
9–12 | 12 (35.3) |
SINS score |
7–12 | 23 (67.6) |
13–18 | 11 (32.4) |
Surgical procedures
The aim of surgery was to provide immediate direct circumferential decompression of the spinal cord and reconstruction of the spinal stability. There were four surgeons involved in the surgery. Prof. HL performed all the cases as primary surgeon; Dr. ZL, YH, and JX were assistants during the surgeries. The surgical implants were chosen from different companies based on the operative strategies and surgical approaches (companies including Stryker Spine, MI, USA and DePuy Synthes Spine, MA, USA). The protocol did not specify operative techniques or fixation devices. The surgical strategy was determined for each patient depending on the involved level, tumor location, and the patient’s condition. In general, three approaches were applied in our study: (1) anterior approach: for anteriorly located tumors, e.g., tumors involving the vertebral body and/or encroached the spinal canal anteriorly, the approach was anterior; (2) posterior approach: mainly posteriorly located tumors, e.g., tumors mainly involving lamina, pedicle, spinous process, or other posterior elements, and/or encroached the spinal canal posteriorly or laterally, while the anterior structures were minor or not involved, a laminectomy and decompression were done and any other posterior elements involved were removed; and (3) combined posterior-anterior approach: tumors involving both the vertebral body and the lamina, or preexisting spinal deformity, a combined posterior-anterior approach was used. Fixation devices including screws, plate, titanium rods, mesh, or artificial vertebral body were used.
Intra-operative neuromonitoring (IOM) were applied during all the surgical procedures routinely using Xltek® Protektor32 IOM System (Natus Medical Incorporated, Ontario, Canada). Changes in transcranial motor-evoked potentials (tcMEPs) and somatosensory-evoked potentials (SSEPs) were monitored.
Postoperative data
All patients received systemic tumor treatment after surgery. Patients were evaluated at the time of discharge and at postoperative time points of 1 month, 3 months, 6 months, and 1 year after surgery. Comparisons were made between preoperative and postoperative Frankel grade, VAS pain score, and KPS score. Postoperative spine MR images and X-rays were evaluated. Survival time after surgical treatment was also recorded.
Statistical analysis
Postoperative survival as a function of time was expressed using Kaplan-Meier estimates with death as the failure event. Survival curves were compared using the Mantel-Cox test, and hazard ratios (HR) and 95% confidence intervals (CI) were computed. The Wilcoxon signed-rank test was used to compare non-parametric paired variables. Pearson correlation analysis was used to further evaluate the correlation between VAS and KPS scores. Statistical analyses were performed using GraphPad Prism 5.01 (GraphPad Software Inc., La Jolla, CA) and IBM SPSS 22.0 (SPSS Inc., Chicago, IL). For all analyses, probability values < 0.05 were considered statistically significant. Median values were reported with range.
Discussion
In recent years, multiple studies have highlighted the effectiveness of surgical treatment for patients with spinal metastases [
12‐
14]. Patchell et al. [
13] reported that decompressive surgical resection treatment is superior in regaining ambulatory function to treatment with radiotherapy alone (median 122 vs. 13 days), as well as prolonging survival time (median 126 vs. 100 days) for patients with spinal cord compression caused by metastatic cancer. The better survival time for patients with surgery was probably because a greater proportion of patients were remained or regained ambulatory than those with radiation therapy. Rades et al. [
14] reported that motor function and overall survival at the end points after decompressive surgery + radiotherapy displayed better results than those after radiotherapy alone, but there was no significant difference. Surgery allows most patients to remain ambulatory for the remainder of their lives, whereas patients treated with radiation spend a substantial proportion of their remaining time paraplegic. Surgical treatment also reduces the need for corticosteroids and opioid pain relief [
13]. However, radiation therapy remains an option for tumor control in patients who choose not to have surgery or are deemed not to be suitable surgical candidate or supplement. There may also be a role for radiation therapy following decompressive surgery [
15]. The results in our study shows a median postoperative survival time of 12.4 months and 44% neurological improvement, which highlights the beneficial effects of surgery for patients with spinal metastases.
During the decision-making process of spinal metastases, primary tumor is an important predictive factor. Primary tumor has been included in most of the prognostic scoring systems, e.g., most commonly used Tokuhashi and Tomita scores. These scores have included the effects of primary tumors in predicting survival and making surgical plan. Recently, Bollen et al. [
16] has assessed the predictive value of six prognostic scoring systems for spinal metastases based on a retrospective study of 1379 patients. In this study, the percentage and median survival of most prevalent primary tumors were also presented: breast cancer (28%, 18.6 months), lung cancer (23%, 2.0 months), prostate cancer (19%, 7.4 months), kidney cancer (7%, 7.8 months), and colon cancer (5%, 3.1 months). The results of this study indicated that survival time could be varied in different primary tumors. Our results did not show any significant difference in survival time in different primary tumors, which may due to the relative small data size. However, in our clinical experience, patients of spinal metastases with breast cancer usually do better than other primary tumors (including lung or gastrointestinal cancer).
The CTJ site accounts for approximately 10% of all spinal metastases [
6]. The most common primary cancers include lung, breast, and prostate cancer [
4]. However, spinal metastases involving the cervicothoracic junction and surgical treatment to this region have been sparsely described in the literature [
17,
18]. Mazel et al. [
18] retrospectively reviewed 32 patients who underwent posterior fixation for cervicothoracic junctional tumors (30 metastasis, 1 chondrosarcoma, and 1 myeloma); the average survive time for patients with vertebrectomy and palliative decompression was 16 (range 3–54) and 11 (range 5–19) months respectively. In the current study, we retrospectively analyzed the surgical outcomes of 34 patients with cervicothoracic junction metastases. Compared with previous studies, the results in this study indicate an acceptable median postoperative survival time (12.4 months) and neurological improvement (44% improved, 56% remained stable). Moreover, the KPS and VAS scores were also significantly improved after surgical treatment. The follow-up data in this report is encouraging and likely reflects the comprehensive effects of prompt and aggressive surgery, radiation, and improved systemic therapies.
In accordance with previous study [
13], differences in postoperative ambulatory function showed predictive value for postoperative survival in this study. Neurological impairment in CTJ metastases is favored by narrowing of the spinal canal or vascular damage of the spinal cord, which would deteriorate rapidly in a short period [
19]. If the compression is of short duration, the effects are reversible [
20]. However, secondary vascular injury occurs with prolonged compression; recovery would be impossible [
20]. Thus, prompt surgical involvement is of great importance for CTJ metastases especially with spinal cord compression. The groups were similar in preoperative SINS, Tokuhashi, and KPS scores, and it is possible that the survival difference was not significant due to the small sample size.
The cervicothoracic spine is a junctional area of the vertebral column with its own unique anatomic and biomechanical characteristics [
21‐
23]. Progression from cervical lordosis to thoracic kyphosis at C7-T1 results in transfer of weight from the posterior aspect to the anterior aspect of the spinal column [
17,
18]. Furthermore, metastases involving this junctional area are prone to segmental instability, which can lead to excessive kyphosis with subsequent narrowing of the spinal canal and injury to the spinal cord [
18,
23]. Therefore, surgical treatment for patients with the CTJ metastases requires proper opportunity and represents unique challenges.
Anterior, posterior, or combined reconstruction can provide stabilization, either as part of a palliative procedure or curative procedure [
17]. In the current study, surgical strategy was determined depending on the involved level, tumor location, and the patient’s condition. Anterior approach, specially to the cranial region of the thoracic spine (T1-T4), is technically demanding and not risk-free [
24]. Further, a recent study reported that patients with T1 lesions showed biomechanical failure of the anterior construct and subsequently underwent posterior fixation [
25]. Thus, anteriorly located cervical lesions were chosen for the anterior approach alone. In this study, 23.52% (8/34) of patients underwent anterior surgery; one patient (12.5%) had cardio-respiratory worsening which was successfully treated.
Posterior fixation techniques have been demonstrated as ideal methods of stabilization for CTJ instability associated with spine tumors [
18]. It is especially important to avoid anterior approach in cases in which there is potential posterior ring disruption or pedicle lesion. Therefore, in this series, posterior approach was used for posteriorly located tumors, circumferential lesions, or in regions that were anatomically difficult to access anteriorly. The C7 screw placement is somewhat unique and is considered separately [
21‐
23]. Transpedicular C7 screw insertion is associated with an increased risk of nerve injury due to the unique anatomical morphology and difficulty in achieving clear intraoperative images [
21]. In this study, we recommended a pedicle entry point for the C7 pedicle 1 mm inferior to the mid portion of the facet joint with 25° to 30° medial angulation and perpendicular to the posterior arch as previously described [
22]. In our series, 64.7% (22/34) of patients underwent posterior surgery. One patient had deep wound infection which was successfully treated by surgical debridement. Another patient had acute epidural hematoma 6 h after surgery, which was manifested as rapid muscle strength worsening confirmed by MRI. Immediate surgical debridement and decompression were performed, and the patient was recovered with no neurological deterioration. This case may indicate that aggressive surgical treatment could save the neurological function for patients with acute epidural hematoma.
Combined posterior-anterior approach was indicated in lesions involving the vertebral body or preexisting spinal deformity [
23]. However, surgical risk, patients’ general status, and life expectancy must be evaluated. 11.8% (4/34) of patients underwent posterior-anterior approach surgery. One of them (2.9%) had post-operative cerebrospinal fluid leakage/effusion confirmed by MRI, whereas 1 year follow-up did not show any special symptom. There were no instrumentation failures in all patients.
Indeed, tumor resection and surgical reconstruction in the cervicothoracic junction (CTJ) area pose challenges; the exact surgical indications for different approach are controversial. Kulkarni et al. indicated that posterior fixation is a gold standard for the treatment of cervicothoracic instability in spine tumors, considering anatomic and biomechanical goals [
7]. However, other researchers suggested that the surgical approach should be individualized depending on tumor location and involved site [
13,
25]. Similarly, in our study, surgical indications are determined for each patient, based on the concept to provide immediate direct decompression and spinal stability reconstruction in the CTJ area. As the follow-up outcomes are acceptable and relatively satisfied, we hope the surgical indications and strategies proposed in this study could provide helpful information.
We have also analyzed the Pearson correlation between VAS and KPS scores. However, no significant correlation was detected, which may partly due to the relatively small data size. The clinical evaluation indexes such as VAS, ODI, and SF-36 are very important and effective to assess the post-operative outcomes for the spinal metastasis patients. They have different features and priorities: VAS is a visual and liner parameter to evaluate neck or back pain, the ODI is a sensitive measure for spinal disorders because it includes domains that are specific to patients with back pain-related disability [
26], and SF-36 is a generic multidimensional scale [
27], which is reliability and validity for use in the general population and in patients with symptoms of back pain and sciatica with and without surgical intervention. To further clarify the relationship between different indexes, properly designed and larger cohort studies will be needed in the future. Besides, it would also be of great importance to explore a more scientific, effective, and feasible evaluation system for the patients with spinal disorders.
There are some limitations to this study. First, this study is a retrospective research of a selected cohort. Future studies should aim to enroll prospective control clinical trials, in order to determine best practices and guide clinical decision-making. Indexes such as ODI were not included, which should be a limiting factor of this study. Involvement of different companies’ implants and different surgeons in each surgery would have confounding impacts to our study. Further, the overall patient number is small and may be underpowered for statistical significance in some variables due to the small patient cohort. The power of this study is also limited by the use of patients at a single institution and within a restricted time period.
Remarkably, the socio-economic effect is an important dilemma for the surgeon and the patient. The socio-economic condition of each patient must be taken into consideration before surgery. Firstly, strict surgical indications (based on spinal metastasis related evaluating systems, especially significant neurological deficits or intractable pain) must be evaluated for each patient before surgery. Patients with an estimated short survival or poor health situation will be excluded. In this series, most of the patients need long-term use of analgesic drugs or even additional care from their family. It will lead to heavy social and economic burden on the patients, families, and communities. Thus, the aims of the surgery are to improve the function, prevent drug use, maintain the dignity of life, enhance the self-care ability, and provide opportunity for further comprehensive treatment of primary tumor. After surgery, necessary rehabilitation including physiotherapy, electrotherapy, and functional exercise will be given to the patients to promote their recovery and improve their functional status. From our point of view, the decision of a such surgery requires careful consideration based on the strict individualized evaluation and the socio-economic effect for each spinal metastasis patient.