Abstract
Background
Percutaneous vertebroplasty (PVP) is an effective measure for painful metastases or impending pathological fractures of the spine with cement leakages being the most frequent complication. Posterior extrusion of cement into the spinal canal may result in neurological symptoms and deficits.
Purpose
To compare the occurrence of intraspinal canal cement leakage between vertebrae with posterior wall disruption and vertebrae without posterior wall disruption.
Material and Methods
A single-center retrospective study was conducted of all PVP for spine metastases between June 2020 and November 2021. All leaks were analyzed by a postprocedural computed tomography scan or cone-beam computed tomography.
Results
A total of 77 patients with 143 vertebrae treated by PVP were included. Posterior wall disruption was observed in 64 (44.8%) vertebrae while 79 (55.2%) had a complete posterior wall. Spinal canal cement leakage occurred in 36 (25.2%) vertebrae and was comparable in both groups, occurring in 16 (25.0%) vertebrae with posterior wall disruption and 20 (25.3%) vertebrae without posterior wall disruption (P = 1). No risk factors for spinal canal leakage were found in the univariate and multivariate analyses. One spinal leak was symptomatic with intercostal neuralgia.
Conclusion
Our results suggest that an incomplete vertebral posterior wall does not increase the rate of spinal canal cement leakage during PVP.
Introduction
Percutaneous vertebroplasty (PVP) is an effective palliative or preventive measure for painful metastases or impending pathological fractures of the spine (1,2). PVP was first described by Galibert et al. in 1987 (3) and has greatly expanded throughout the years following the growing prevalence of bone metastases (4). During PVP, polymethylmethacrylate (PMMA) cement is injected percutaneously through a needle to fortify a pathological fracture or consolidate a lytic tumor. Although PVP is a safe technique, some complications may occur with cement leakages being the most frequent complication (5,6). Most cement leakages are not associated with serious complications such as paravertebral soft tissue leaks and discal leaks. However, posterior extrusion of cement into the spinal canal may result in neurological symptoms and deficits (5,7–9). According to some authors, cortical disruption increases the risk of cement leakage (10–13) while epidural tumor extension is considered a relative contraindication for PVP (1). Breast cancer, injected volume of cement, and vertebral collapse were also reported risk factors of cement leakage during PVP for spine metastases (10,11,13). The treatment of cement leakage is usually conservative with anti-inflammatory agents while surgical removal may be pursued in patients with persistent symptoms (6).
The aim of the present study was to compare the occurrence of intraspinal canal cement leakage between vertebrae with posterior tumoral cortical disruption and vertebrae without posterior tumoral cortical disruption.
Material and Methods
Data collection
The study was approved by our institutional ethics committee (No. IRB00012437). We retrospectively retrieved all patients treated by PVP in our single institution center between June 2020 and November 2021. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Inclusion criteria were as follows: (i) histologically confirmed diagnosis of metastatic cancer and (ii) treatment by PVP for either symptomatic pathological fracture or impeding pathological fracture of a vertebra, symptomatic or not.
Exclusion criteria were as follows: (i) lack of spine preprocedural computed tomography (CT) scan and (ii) lack of spine postprocedural CT scan or cone-beam CT (CBCT).
Patient records were reviewed for patient demographics, primary tumor site, indication for PVP, and procedural complications. Preprocedural CT scans were reviewed for number, location, size, and existence of posterior wall disruption of spine metastases treated by PVP. Postprocedural CT scans and/or CBCT were reviewed for intraspinal canal cement leakage, foraminal cement leakage, and volume of injected cement.
Posterior wall disruption was quantified with measurement of the axial long axis (in mm) and with the percentage of disruption of the posterior wall on axial reformations (Fig. 1).
Assessment of the posterior wall disruption by a spinal metastasis showed on (a) round dotted line. (b) Measurement of the axial long axis in millimeters of the posterior wall disruption (dotted double arrow) and the whole posterior wall (full double arrow).
The volume of intraspinal canal cement leakage was evaluated with measurement of the long axis (in mm) of the leak, and with the angle (in degrees) of the epidural space occupied by the cement leak with the vertex being the center of the spinal canal (Fig. 2).
(a, d) Assessment of the volume of intraspinal canal cement leaks evaluated with measurement of (b, e) the long axis in millimeters of the leak, and (c, f) with the angle in degrees of the epidural space occupied by the cement leak with the vertex being the center of the spinal canal.
The volume of injected cement was estimated according to the Saliou classification (14).
Procedures
All procedures were performed with image guidance using a digital subtraction angiography unit with a C-arm (Allura Clarity, Philips, Amsterdam, Netherlands) by a senior interventional radiologist. Procedures were performed either under conscious sedation with local anesthesia or under general anesthesia depending on the patient's clinical condition, the site of the vertebra, and the number of vertebrae to treat. Patients with more than three vertebrae to be treated or cervical PVPs were usually performed under general anesthesia. Beveled 10–13-G vertebroplasty needles (T’CD II, Thiebaud, Margencel, France or Special vertebroplasty needle, Optimed, Ettlingen, Germany) were inserted into the vertebral body with fluoroscopic guidance in all cases associated to CBCT guidance on a case-by-case basis. Pedicular approaches, trans-costo-vertebral approaches, and anterolateral approaches were performed respectively for lumbar, thoracic, or cervical vertebrae. A bilateral approach was performed on a case-by-case basis when the cement filling of the vertebra was considered insufficient. Vertaplex bone cement (Stryker, Kalamazoo, Michigan, USA) was used in all procedures. Injection of PMMA was performed under continuous fluoroscopic screening using the PCD mixer and delivery system (Stryker, Kalamazoo, Michigan, USA).
Endpoints
The primary endpoint of our study was to compare the occurrence of spinal canal leakage during PVP for spine metastases between vertebrae with posterior wall disruption and vertebrae without posterior wall disruption.
The secondary endpoints were to report risk factors for spinal canal leakage during PVP and the rate of symptomatic spinal cement leaks.
Statistical analysis
Statistical analysis was performed using EZR software (15). Continuous variables were studied using the Welch t-test. For nominal variables, the chi-square test or Fisher test was used. We carried out a multivariate analysis using logistic regression. The covariates were predefined based on data from the literature. A P value <0.05 was considered significant.
Results
Patients
Between June 2020 and November 2021, 190 consecutive patients had PVP in our institution and were screened for eligibility (Fig. 3). A total of 77 patients (33 men, 44 women; age range = 27–91 years) with 143 vertebrae treated met the selection criteria. The primary tumor was the breast in 33 (42.3%) patients, lung in 15 (19.5%) patients, myeloma in 6 (7.8%) patients, colon in 5 (3.5%) patients, kidney in 2 (2.6%) patients and other miscellaneous origins in 18 (23.4%) patients. Patients had a mean body mass index (BMI) of 25.1 ± 6.1 kg/m2. Lesions were all lytic with a mean size of 22.2 ± 8.2 mm. Posterior wall disruption was observed in 64 (44.8%) vertebrae and involved <25% of the posterior wall in 24 (16.8%) vertebrae, 25%–50% in 15 (10.5%) vertebrae, 50%–75% in 25 (17.5%) vertebrae, and >75% in 6 (4.2%) vertebrae. A vertebral collapse was seen in 93 (65.0%) vertebrae. The site of metastases was cervical in 6 (4.2%) of cases, thoracic in 78 (54.5%) of cases, and lumbar in 59 (41.3%) of cases. Patients’ and metastases’ characteristics are presented in Tables 1 and 2, respectively.
Flow chart.
Characteristics of patients.
Values are given as n (%) or median (range).
BMI, body mass index.
Characteristics of metastases.
Values are given as n (%) or median (range).
Spinal canal cement leakage
After PVP, a total of 61 treated vertebrae were evaluated by CBCT and CT scans, 50 by CBCT alone, and 32 by CT scan alone. Posterior cement leakage occurred in 36 (25.2%) vertebrae. The cement leak measured 9 ± 4.9 mm and the angle of the epidural space occupied by the cement leak was <45° in 17 (11.9%) vertebrae, 45°–90° in 13 (9.1%) vertebrae, and >90° in 6 (4.2%) vertebrae. The angle of the epidural space occupied by the cement leak never reached ≥180°. Foramen cement leakage occurred in 6 (4.2%) vertebrae and was associated with posterior spinal canal leak in all cases.
Posterior wall disruption
Spinal canal cement leakage occurred in 16 (25.0%) vertebrae with posterior wall disruption and was comparable with the spinal canal cement leakage rate in vertebrae without posterior wall disruption (20 vertebrae, 25.3%) (P = 1) (Figs. 4 and 5). Both groups were comparable in characteristics except for breast cancer and lung cancer which were overrepresented in the group with posterior wall disruption and the group without posterior wall disruption respectively (P < 0.001). Comparisons between PVP for spine metastases with and without posterior wall disruption are presented in Table 3.
Axial computed tomography scan reformations before and after percutaneous vertebroplasty in a 55-year-old woman with a T3 metastasis of breast cancer. (a) Disruption of 25%–50% of the posterior wall is present (white arrow). (b, c) Spinal canal cement leakage (white arrows) occurred through the posterior wall lysis (black arrowhead).
Axial and sagittal computed tomography scan reformations before and after percutaneous vertebroplasty in a 43-year-old woman with L3 and L4 metastases of breast cancer. (a, b) Disruption of 50%–75% of the posterior wall is present (white arrows). (c, d) Proper cement filling of the lysis (stars) is obtained without spinal canal cement leakage.
Comparison between percutaneous vertebroplasty for spine metastases with and without posterior wall disruption.
Values are given as n (%) or mean ± SD.
BMI, body mass index.
Risk factors for spinal canal cement leakage and symptomatic spinal cement leaks
No risk factor for cement leakage during PVP was found in the univariate and multivariate analyses (Tables 4 and 5). Of the 36 spinal canal leaks, 1 (3%) was symptomatic with intercostal neuralgia caused by a foraminal leak. All other 35 cement leaks were asymptomatic.
Univariate analysis for risk factors for spinal canal leakage during percutaneous vertebroplasty.
Values are given as n (%) or mean ± SD.
BMI, body mass index.
Multivariate analysis for risk factors for spinal canal leakage during percutaneous vertebroplasty.
Values in parentheses are 95% confidence intervals.
Discussion
Cement leakage, defined as the presence of extra vertebral cement, is a common adverse event after PVP, with an incidence rate of approximately 60% according to a recent systematic review (16). Although perivertebral vascular leaks can cause pulmonary embolism, the most severe complications may occur when cement leaks into the spinal canal and/or the foramen with the potential of causing neurological compression. Spinal canal leakage represents cement leakage through the anterior internal venous plexus and basivertebral vein or directly through the incomplete posterior wall. Posterior cement leakage responsible for neuropathy requiring decompression surgery rarely occurs, taking place in approximately 3% of cases according to the systematic review by Chew et al. (17). In our study, a symptomatic spinal cement leakage occurred after only one PVP (0.7%) due to the extension of the cement to a foramen resulting in intercostal neuralgia. The majority of the cement leaks (52.7%) occupied >45° of the epidural circumference, meaning more than half of the anterior epidural space, while the mean diameter of leaks was 9 mm, confirming that spinal canal leakage is at low risk of complications even with relatively large leaks.
In our study, the spinal canal leakage rate was 25.2% (36 vertebrae), higher than in some previous studies (11–13), in the range of 7%–15%. However, in the studies by Cui et al. and Shi et al., CT scans were not routinely used for postoperative examination resulting in a possible underestimation of spinal canal leakage incidence. However, our results were consistent with Saliou et al.’s epidural leakage rate of 30% (14). In our series, all patients had a post-PVP CT scan or CBCT allowing a precise evaluation of the existence of a spinal canal cement leak.
Risk factors of cement leakage are very diverse in the literature. A meta-analysis by Zhan et al. (16) reported intravertebral cleft, cortical disruption, low cement viscosity, and high volume of injected cement as being risk factors after PVP for osteoporotic vertebral compression fractures or spinal metastases. When considering PVP for vertebral metastases exclusively, posterior wall disruption, breast cancer, injected volume of cement, and vertebral collapse were reported risk factors of cement leakage (10,11,13).
Our results show that posterior wall disruption was not at a higher risk of spinal canal leakage when compared to vertebrae with a complete posterior wall. When comparing both groups, the patients’ characteristics were mostly comparable except for the metastases’ largest diameter, primary tumor histology, and the number of levels treated during the same session. Metastases being larger in the posterior wall disruption group is explained by the fact that larger vertebral lesions are more likely to reach the posterior wall. Breast cancer was not equally distributed in both groups. However, even though breast cancer is believed by some to increase the risk of cement leakage (11) and the over-representation was in the posterior wall disruption group, cement leakage rates were identical in both groups. In addition, breast cancer was not found to be a risk factor for spinal leakage in the univariate or multivariate analyses.
The amount of cement leakage (in mm), and with the angle of the epidural space occupied by the cement, was comparable in both groups. However, knowledge of the existence of an incomplete posterior wall, all the more caused by a breast cancer metastasis, may have led to great care by the interventional radiologist, therefore helping avoid spinal canal leakage.
The volume of cement injected was evaluated in two different ways in our study: first, with the Saliou score (14) introduced to assess the quality of cement filling by dividing the vertebra into nine sections of the same size in the front and lateral views, giving a score out of 18; and second, with the level treated knowing that the vertebral volume increases from the cervical spine to the lumbar spine. Vertebral cement filling was comparable in both groups and our results show that the same rate of cement filling may be obtained whether a posterior wall disruption exists or not. Furthermore, we did not find an increased risk of spinal cement leakage according to the volume of cement injected in the univariate or multivariate analyses.
Low cement viscosity may increase the rate of cortical and vascular cement leakage according to some authors (18–20). However, in our study all PVPs were performed using the same PMMA cement, eliminating a potential bias when comparing both groups.
When performing PVP for spinal metastases with posterior wall disruption, some interventional radiologists may choose to inject cement exclusively in the adjacent normal bone to reduce the risk of cement leakage through the incomplete posterior wall. However, cement filling of the lesion was not found to be a risk factor for spinal cement leakage in the univariate analysis, and approximately 80% of lesions were filled with cement, comparable in both groups.
Several techniques have been described to minimize cement leakage during PVP, such as the use of balloon catheters to control cement flow or the use of polyetheretherketone (PEEK) coiled metallic implants (6). Some authors use combined techniques with ablation in association with cement injection for the treatment of bone fracture and impending fracture (21). Such combined techniques might have a positive impact on the cement leakage rate and should be evaluated.
The present study has some limitations. First, the analysis was retrospective, exposing the study to selection bias. However, it included a relatively large number of spine metastases with groups being comparable in number and most characteristics. Second, the metastatic nature of the spine lesions was not based on histologic analyses in all cases. However, all patients had histologically confirmed diagnoses of metastatic cancer, and lesions were assumed to be malignant metastases related to imaging appearance and progression.
In conclusion, our results suggest that an incomplete vertebral posterior wall does not increase the rate of spinal cement leakage during PVP and that PVP can be performed with a high degree of confidence when considering spinal canal leakage in patients with posterior wall disruption.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
