Comparison of Outcomes between Single-Level and Double-Level Corpectomy in Thoracolumbar Reconstruction: A Retrospective Study of 16 Patients Using Expandable Cages

Abstract

Background

Reconstruction surgery using titanium vertebral body replacements aids in spinal stability after damage. Functional evaluation includes Nurick and ASIA grading systems. This study compares outcomes of single and double-level corpectomy in patients treated with thoracolumbar reconstruction.

Material/Methods

Records of 16 patients who underwent vertebral reconstruction with expandable cages after single (n=9) and double (n=8) corpectomy were analyzed retrospectively. Thoracal and lumbar cases were approached posteriorly (n=14). Clinical evaluation was performed by neurological examination, Nurick Scale, American Spinal Injury Association (ASIA) Neurological Scores, and Visual Analog Scale (VAS). Preoperative and postoperative 3-month scores were recorded. Radiological evaluation was performed by calculation of regional angulation and postoperative recovery of vertebral body height.

Results

Preoperatively, 2 patients had no motor or functional sensory deficits (ASIA grade E) and 6 patients were ASIA grade D. Immediately after surgery, 4 of the ASIA grade D patients improved to ASIA grade E, while 2 patients remained ASIA grade D during follow-up. Four patients were ASIA grade A; their conditions showed no improvement postoperatively. Preoperative and postoperative 3-month Nurick grade was statistically significant (P=0.022). Postoperative Cobb angle improved by a mean of 5.4° (P=0.001). Improvement by at least 1 grade in neurological recovery was observed in 6 (38%) of the cases. Pain scores analyzed using the VAS changed from a mean of 7 to a mean of 2.63 (P<0.001).

Conclusions

In our experience, expandable cages are useful in the restoration of different pathologies of the thoracolumbar trunk with low complication and revision rates. With expandable cages it is possible to achieve restoration of the Cobb angle, improvement in Nurick Grade and effective pain palliation.

Introduction

The anterior spinal column can be damaged by primary spinal tumors, metastases, infections, deformities, and trauma [1]. To prevent vertebral body collapse and spinal instability after spinal column damage, reconstruction surgery with stabilization, decompression, single-level and double-level corpectomy, and titanium vertebral body replacement can be used [2]. These titanium grafts consist of 2 telescoping, internally threaded cylinders with wide windows to help pack bone graft material together. Following the corpectomy, the implant’s length is precisely adjusted with optimal in situ distraction using fluoroscopic guidance to accommodate the vertebral defect. Spikes on the cage’s endplates help it dock securely into the nearby vertebral bodies’ subchondral bones. By having a wide axial profile, the cage maximizes the surface area that comes into touch with the vertebral bodies both above and below the operating level. Rather than using an additional apparatus, the cage provides distraction to the vertebral endplates.

There are several measures available to evaluate patients’ functional impairment. One of the most used ones that primarily evaluates ambulatory status is Nurick, which was presented in 1972 for cervical spondylosis [3,4]. The other measure commonly used is American Spinal Injury Association (ASIA), which is a grading and classification system for spinal cord injuries, released in 1982 by the American Spinal Injury Association [5]. In our study, to evaluate the functional disability of patients, we used both the Nurick and ASIA grading.

Single-level corpectomy entails the excision of 1 vertebral body and disc, often used to decompress the spinal cord from large disc herniations or localized tumors. This procedure restores spinal stability and function by replacing the excised vertebral body with bone grafts or implants. Double-level corpectomy involves the removal of 2 adjacent vertebral bodies and discs. It is used for more extensive spinal pathologies such as multi-level degenerative disc disease or expansive spinal tumors. Reconstruction following double-level corpectomy requires larger bone grafts, additional implants, or fusion techniques. This retrospective study from a single center included 16 patients treated with thoracolumbar reconstruction with expandable cages, and aimed to compare outcomes from single-level and double-level corpectomy.

Material and Methods

Between 2018 and 2021, 16 patients who underwent vertebral reconstruction with expandable cages after single- or double-corpectomy were included. A titanium-made expandable corpectomy cage (Alton Sapimed Onspine^®) was used. Our study was approved by the SBU Istanbul Research Hospital Clinical Research Ethics Committee (approval number: 2023/128). All participants gave informed consent.

Patient Inclusion Criteria

Any patient receiving anterior column reconstruction with an expandable cage met the inclusion criteria. Severe mechanical back pain, pathological fractures with >50% loss of height, and >50% vertebral body resection were among the indications for anterior column reconstruction. Patients who did not require anterior column reconstruction were not included.

Clinical Evaluation

Neurological functional assessment was performed with the Nurick and ASIA impairment scales, preoperatively and postoperatively after 3 months [6]. In Nurick grading, patients with signs or symptoms of root involvement, yet lacking evidence of spinal cord disease, fall under grade 0. Grade 1 indicates the presence of spinal cord disease without hindrance in walking. Grade 2 represents patients experiencing mild difficulty in walking but still able to maintain full-time employment. Grade 3 denotes significant walking difficulty, inhibiting full-time employment or the ability to complete household tasks independently, yet not severe enough to necessitate assistance for walking. Grade 4 marks patients who rely on assistance or walking aids, such as frames, to move. Grade 5 characterizes those who are confined to a chair or bedridden due to the severity of their condition; unable to walk independently [3].

In the ASIA Impairment Scale, the loss of all motor and sensory abilities distal to the site of the lesion, including sacral roots, is referred to as a total spinal cord injury. These injuries are classified as grade A. A partial loss of motor or sensory function below the site of injury is referred to as an incomplete injury. These are ranked B through E. Individuals with lesions classified as grade B have no motor function and partial sensory function. Less than 3 motor grades below the neurologic level of injury characterize grade C injuries, whereas at least 3 motor grades below the neurologic level of injury characterize grade D injuries. Even when the motor and sensory exams of patients with Grade E injuries are normal, the patients may nonetheless exhibit aberrant reflexes or other neurologic abnormalities [5]. Visual Analog Scale (VAS) was used to assess pain [7]. Neurologic status was assessed using the Frankel grading scale.

Radiologic Evaluation

All patients underwent evaluations preoperatively, right after surgery (before being discharged), and 3, 6, and 12 months after surgery. To evaluate the stability and fusion status of the operated section, serial radiographs of the pertinent segment were acquired. After surgery, postoperative CT imaging scans of the affected area were carried out to record the cage and screw locations and to assess the degree of spinal decompression and fusion. Using the Cobb angle, kyphotic deformity was assessed on thoracolumbar spine lateral radiographs. The Cobb angle was measured from the top endplate of the vertebra at the upper level (where the corpectomy was done) to the bottom endplate of the vertebra at the lower level. The kyphotic angle was assessed preoperatively, immediately following surgery, 1 month after surgery, and during the final follow-up appointment (8–28 months; mean follow-up time was 18 months). The lateral radiograph was used to measure the height of the structure as the separation between the lower endplate of the upper and lower vertebrae. Three months following surgery, all patients had dynamic X-ray and CT scans performed to evaluate stability, fusion, subsidence, and potential hardware displacement. It can be difficult to determine fusion when using anterior thoracolumbar instrumentation. If there was no motion on flexion-extension films, zero radiolucency at the junction of the interbody graft and vertebral body, and no indication of an angulation shift in spacing after a follow-up of 1 year, the construct was deemed stable. The mean radiographic and clinical follow-up time was 18 months (range, 8–28 months).

Surgery

In 10 cases, thoracoabdominal surgery was conducted, and in 6 cases, retroperitoneal surgery was conducted. To prevent the liver or inferior vena cava from retracting, a left-sided technique was recommended. To obtain an adequate working area for decompression and instrument placement, exposure above and below the diaphragm was necessary in the thoracolumbar approach. Each patient had a corpectomy, the spinal canal was decompressed, and the anterior column was rebuilt utilizing an Alton Sapimed Onspine^® cylinder-shaped titanium mesh cage. When preparing the endplates following corpectomy, great care was taken to maintain the bone endplates as much as feasible. The superior and inferior ends of the cage were cut to fit the vertebral endplates’ sagittal alignment. Bone chips containing both demineralized bone matrix as well as synthetic bone as an allograft, particularly in the case of a metastatic spine, were placed within and around each cage, both anteriorly and laterally (Figure 1). In such situations, autograft was taken from the shattered vertebral body and the iliac bone. Iliac bone grafts were utilized to fill the cage in cases of tuberculosis and malignant diseases. To prevent damage to the contralateral segmental vessels, in traumatic instances, the corpectomy was subtotal, leaving a little rim of the vertebral cortex on the other side of the approach. After positioning the plate device, distractive forces were used to lessen the kyphotic deformity.

Figure 1.

(A) Pre-operative T2-weighted MRI showing a tumor located at thoracal eighth vertebra. (B) Pre-operative CT image showing kyphotic angulation caused by tumor. (C) Postoperative CT image after reconstruction of kyphosis with an expandible cage insertion.

Follow-Up

Before the patient was discharged, the clinical examination was carried out in the preoperative and early postoperative phases, and at least 1 year after the surgery. The mean radiographic and clinical follow-up time was 18 months (range, 8–28 months).

Statistical Analysis

Demographic and clinical variables were analyzed with descriptive statistics. When data conformed to a normal distribution, measurements were represented by the mean and range. Preoperative and postoperative Nurick, VAS, and ASIA scores, as well as angulations, were analyzed with 2-tailed paired t-tests. The sign test was used to analyze the change in Cobb angle from preoperative to postoperative. SPSS version 21 for Windows (IBM Corp. Armonk, NY) was used for all statistical analyses, and P<0.05 was considered the level for statistical significance. The ethics review committee approved the research protocol. All participants gave informed consent.

Results

Baseline and Operative Parameters

Clinical and radiological data for 16 patients were analyzed, retrospectively. Their demographic and clinical characteristics are summarized in Table 1. The median age was 51.7 years (range 24–76). The most common symptom was back pain (87%). A neurological deficit was detected in 62% (10/16) of patients. Four patients had deformities due to Pott disease (25%), and 12 patients had tumors 75%). Three patients had lung cancer, which is the most common tumor etiology.

Table 1.

Patient demographic and medical characteristics.

Characteristic	Value or mean
Age (mean±standard deviation)	51.6±15.1
Gender
Male	7
Female	9
Surgery duration (h)	5.6
Blood loss (mL)	1200
Complications
Cerebrospinal fluid leakage	1
Deep venous thrombosis	1
Epidural hemorrhage	1

Thoracal and lumbar cases were approached posteriorly (n=14). The mean follow-up time was 28 months. Two patients had revision surgery due to early complications (1 epidural hematoma and 1 screw malposition), and 1 patient had revision surgery for an L2 burst fracture. Two patients were re-operated 5 and 6 months later, respectively, because of tumor recurrence.

The mean operation time was 5.6 hours (range 5–8). In the thoracolumbar approach, the mean blood loss was 1200 mL (range 800–1700). Surgical complications were observed in 18.75% of the cases. The complications are listed in Table 2. Transient neurological deterioration was seen in 1 case, which occurred after epidural hemorrhage. After detection of neurological damage, the patient underwent revision surgery in 48 hours. In 2 weeks, the patient recovered without permanent deficit.

Table 2.

Number of patients with clinical characteristics at preoperative and 3-month postoperative assessments: Nurick grade, ASIA scale, and VAS score.

Variable	Preoperative (N)	3-month postoperative (N)
Nurick grade
0	1	1
1	3	7
2	5	4
3	1	1
4	2	0
5	4	3
ASIA Scale
A	4	4
B	1	0
C	3	0
D	6	5
E	2	7
VAS (mean)	7	2.63

ASIA – American Spinal Injury Association Impairment Scale; VAS – visual analog scale.

Clinical Efficacy Assessment

Preoperative and 3-month postoperative Nurick and ASIA scores are listed in Table 2. The improvement in Nurick grade was statistically significant (P=0.022). Preoperatively, 2 patients had no motor or functional sensory deficits (ASIA grade E), while 6 patients were ASIA grade D. Immediately after surgery, 4 ASIA grade D patients improved to ASIA grade E, while 2 patients remained at ASIA grade D during follow-up. Four patients were ASIA grade A; their conditions showed no improvement postoperatively. Pain scores were analyzed with VAS, which is listed in Table 2. All patients had pain reduction, which changed from a mean of 7 to a mean of 2.63 (P<0.001).

Comparing Radiography of Patients with and without Expandable Cages

The difference between mean preoperative and postoperative Cobb angle showed a clear improvement, improving from 16.5 to 10.13, indicating that thoracal kyphosis was restored (P=0.001).

Discussion

In our study, a retrospective analysis of 16 patients undergoing spinal surgery revealed diverse demographic and clinical profiles, predominantly presenting with back pain (87%) and neurological deficits (62%). The cohort consisted of cases primarily attributed to tumors (75%) and Pott disease deformities (25%). Surgical approaches were primarily posterior (n=14), with a mean follow-up of 28 months. Complications were observed in 18.75% of cases, including transient neurological deterioration necessitating revision surgery in 1 instance. Postoperative assessments demonstrated significant improvements in Nurick grade (P=0.022) and ASIA scores, with notable pain reduction (P<0.001). Additionally, utilizing expandable cages resulted in significant improvements in thoracal kyphosis (P=0.001), as evidenced by radiographic comparisons. These findings underscore the efficacy of surgical interventions in ameliorating symptoms and restoring spinal alignment in patients with various spinal pathologies.

Up to 80% of the axial force is transmitted through the spinal column, which is essential for preserving the biomechanical balance of the spine. Pathological processes such as trauma, malignancy, and infection can affect the stability of the vertebral body. These processes can affect the stability of the vertebral body and the spinal column such that the patient may require vertebrectomy and anterior reconstruction [8]. Also, besides they cause instability, these pathological processes cause reduction in disc and foraminal height. Particularly when the disc height is reduced to 4 mm with loss of height at the bottom of the vertebral space, this may result in compression of the nerve roots within the foraminal area, causing pain and neurological motor deficits [9]. Yet, the disc and foraminal height can be efficiently restored with decompression and application of an expandable cage [10]. In our study, adequate restoration was achieved by correcting the Cobb angle from a mean of 16.5 to 10.5 degrees with expandable cages.

Adequate decompression and prompt stabilization of the spine can be achieved by anterior techniques. Direct anterior decompression is more efficient than indirect decompression of neural components via ligamentotaxis. For the anterior column, the best reconstruction tool should preserve mechanical stability, spinal alignment, and allow for stable fusion. The anterolateral approach allows for more direct and complete decompression of the spinal canal. Bone fragments can be removed from the canal with direct visualization, potentially providing a better neurological outcome. This technique maintains the integrity of the posterior column and allows for continuous correction of angular deformities [11]. However, in our study, 14 thoracal and lumbar cases were approached posteriorly, because the anterior approach has many potentially morbid complications like chylothorax, pulmonary problems, and chest pain. Some cases may require joint surgical intervention with other clinics such as general surgery and thoracic surgery, which causes organizational difficulties. Also, our clinic has more experience in the posterior approach, so that is the preference among most of our surgeons.

In our study, back pain (87%) was the most common symptom. Most of our patients had spine tumors (75%) which required stabilization, corpectomy, and anterior support to correct instability and kyphotic deformities. Prior research has demonstrated that improving sagittal alignment leads to better clinical outcomes [12]. An expandable cage was shown to be an excellent option to fill this gap after tumor resection and corpectomy [6]. In our clinic, to provide adequate stabilization and fusion with a cage, first, transpedicular screws were placed in the vertebra 3 levels above and 3 levels below the corpectomy in thoracic cases. Transpedicular screws were placed 2 levels above and 2 levels below in patients who underwent lumbar corpectomy. Our fusion rate was 81% in our early postoperative and 3-month CT scans. At a year’s follow-up following cage placement, our study’s radiographic results were largely positive.

In the recent literature, for correction of tumoral deformities, spine radiosurgery (SRS) has gained acceptance [6]. In our cases, SRS was not applied because vertebrectomy was required because corpus erosion was present in our tumoral cases; therefore, SRS was not preferred.

Deep vein thrombosis, CSF leakage, and epidural hematoma were among our complications. Similar complications were observed in the literature. Our complication rate was 19%.

In thoracolumbar corpectomy, the posterior approach is preferred because anterior approaches from the thoracic cavity have high morbidity due to the risk of vascular complications, pneumothorax, pneumonia, respiratory collapse, and cerebrospinal fluid leakage into the thoracic cavity [13,14]. Advantages of the posterior approach include reduced operation time, with lower rates of complications related to avoiding position changes required for combined approaches [15].

After corpectomy, autologous iliac crest and fibular bone grafts have been widely used for reconstruction. Yet, bone grafts have risks of pseudarthrosis and pain in the donor site as complications [16]. In recent years, expandable cages have been safely used to reconstruct the spinal column after corpectomy instead of bone grafts [17]. Potential advantages are ease of implanting during surgery and providing adequate correction of kyphosis and height restoration [18].

Limitations

One of our study’s limitations was the limited sample size. A small sample size is insufficient to assess the various results and complications that can arise early or late in cases of surgical treatment such as these. It will take longer-term research (with long-term follow-up) to evaluate the impact of expandable cages over time. Secondly, selection bias may be present since patient pathological examination and surgeon preference play major roles in decision making. Depending on the unique patient’s features, including overall stability, bone quality, and underlying disease, we can decide whether to proceed with further plating or dorsal stabilization. Also, potential biases in information or selection were introduced, and postoperative problems were assessed based on potentially incomplete clinical records. The use of lateral X-ray data to analyze subsidence, and increased CT study of subsidence during and after operations, will be beneficial in future research.

Conclusions

Expandable cages are useful in the restoration of different pathologies of the thoracolumbar spine, with low complication and revision rates. With expandable cages, it is possible to achieve restoration of the Cobb angle, improvement in Nurick grade, and effective pain palliation. Thus, expandable cages are deserving of more study and implementation because they have a promising clinical application future.