Percutaneous vertebroplasty (PVP) was first used in 1985 by French radiologists Deramond and Galibert in the treatment of a case of cervical vertebra hemangioma.1 Subsequently, this technique has been widely used in osteoporosis vertebral compression fractures, vertebral metastasis of malignant tumors, multiple myeloma, and other spinal diseases.2,3 The goal is to strengthen the vertebral body by injecting bone cement, to restore spinal stability, and to alleviate and eliminate pain. Several studies have proved that PVP has many advantages, such as good safety, effective pain relief, and a short course of treatment.4,5
The efficacy of PVP is related to many factors, such as the treatment time after fracture, fracture severity, and magnetic resonance imaging (MRI) of bone marrow edema. Recently, some scholars have suggested that the diffusion of bone cement is the key to the success of PVP.6,7 The use of balloons in percutaneous kyphoplasty (PKP) limits the diffusion of bone cement in all directions, confining it to the center of the vertebral body. In some studies, patients using PKP on the surface were more likely to have postoperative vertebral height loss than patients using PVP, which was due to the different distribution of bone cement and the different stress in the vertebral bodies.8,9
Compared with PKP, when PVP is injected with bone cement, the bone cement can spread freely around. In clinical practice, it is often found that some patients inject more bone cement, but the diffusion volume (DV) in the vertebral body is small. However, some patients injected a small amount of bone cement but showed a large DV on imaging, indicating that the diffusion of bone cement was affected by some factors.10 Patients with osteoporotic vertebral compressive fractures (OVCF) have different degrees of osteoporosis; the number and quality of the trabecular bone in the vertebral body are different, which may be the reason for the difference in bone cement diffusion. This study attempted to analyze the effect of bone mineral density (BMD) on the diffusion coefficient (DC) during PVP and to explore the correlation between bone DC and the efficacy after PVP.
Materials and Methods
This research received institutional review board approval. This retrospective study involved patients with OVCF who were hospitalized before December 2018. Inclusion criteria were confirmed thoracolumbar OVCF (T11-L2), fresh fracture within 1 month, T2-weighted imaging high signal on MRI, 1 and 2 Genant grading, and BMD examination results included in clinical data (QDR 4500; Hologic). The postoperative 3-dimensional computed tomography (CT) scan data of the operative vertebra, single-segment OVCF, were included. Postoperative follow-up was more than 1 year, and radiograph examination and visual analog scale (VAS) and Oswestry Disability Index (ODI) scores were recorded on follow-up. Exclusion criteria were multi-segment OVCF, MRI showing air sign, severe compression (Genant 3), bone cement leakage, incomplete imaging data and scores, and nerve damage.
Patients with a T value of −2.5 or more were assigned to the BMD decrease group. Patients with a T value between −2.5 and −3.5 were assigned to the osteoporosis group. Patients with a T value of −3.5 or less were assigned to the severe osteoporosis group.
All patients were immediately treated with anti-osteoporosis therapy after hospitalization, including 50 IU of salmon calcitonin injected intramuscularly once a day, 70 mg of alendronate sodium D3 orally once a week, 1.25 g of calcium carbonate orally twice a day, and 60 mg of celecoxib orally twice a day. They also underwent BMD test, MRI, and blood test for bone markers. The assessment system of thoracolumbar osteoporotic fracture (ASTLOF) is used to assess the treatment choice after the OVCF diagnosis is clear.11 This system is divided into 2 parts: symptoms and examination. Symptoms include pain evaluation. Examination includes 3 evaluation parts—vertebral morphology, MRI, and BMD—and each evaluation part has different scores (Table 1). The total score obtained by adding the scores of all parts is the ASTLOF score (T value, from 0 to 8). Conservative treatment was performed when T was 3 points or less. If T was 4 points, treatment was according to the patient's wishes. If T was 5 points or greater, PVP surgery was recommended.
Assessment System of Thoracolumbar Osteoporotic Fracture
The PVP procedure involved preoperative establishment of venous channels; electrocardiogram monitoring; monitoring of blood oxygen saturation, blood pressure, and respiration; and intramuscular injection of 0.1 g of phenobarbital and 0.1 g of tramadol. The patient was in the prone position on the operating bed, and the overextension position was performed. A C-arm radiograph machine was used to locate the fractured vertebral body, and gentian violet was used to mark the position of the bilateral vertebral pedicles. With a routine disinfection towel, 1% lidocaine infiltration anesthesia, and the assistance of positive and lateral fluoroscopy, the puncture needle was inserted into the anterior one-third of the vertebral body through the pedicle. After mixing bone cement (polymethyl methacrylate) with powder (0.5 mL/g) as recommended by the manufacturer, at a temperature of 24 °C and mixed with 80 seconds, polymethyl methacrylate was injected into the vertebra. All of the operations were monitored through fluoroscopy to observe whether the bone cement leaked into the vertebral body. When the bone cement reached the anterior two-thirds of the vertebral body, the injection was stopped, and the injection amount of bone cement was recorded. After the bone cement solidified, the puncture needle was pulled out.
After the operation, 64-slice CT was used for vertebral body examination. Tube voltage was 120 kV, 300 mAs, layer thickness was 5 mm, layer spacing was 5 mm, and volume scanning was performed. After scanning, the reconstructed layer thickness is 1 mm and the interval between layers is 1 mm. The scope of the scan included injured vertebra and 2 adjacent upper and lower vertebra. The original data were imported to the Vitrea (Toshiba Corporation) workstation to measure and calculate the DV. The CT value of the bone cement in the vertebral body was 1000 to 3000 HU, and the bone cement in the vertebral body was outlined in the area at each level. The DV was calculated by volume calc volumetric measurement function, and the DC was obtained based on the actual clinical injection volume (IV): (DC=DV/IV). To reduce error, the boundary drawing of DV was performed by 2 experienced radiologists and averaged (Figure 1).
Diagram of diffusion volume calculation software.
When the patient returned as an outpatient 12 months after the operation, reexamination of the vertebral radiograph was conducted to measure whether there was height loss in the injured vertebra. The criteria for determining whether vertebral height loss occurred after PVP were as follows: compared with preoperatively, postoperative vertebral height loss was 15% or less or the local kyphosis angle had increased 10° or greater.
The VAS pain score was used to evaluate the degree of pain in patients, and the ODI score was used to evaluate the degree of daily activity dysfunction in patients. Both were recorded before the operation and at each follow-up visit.
SPSS, version 18.0, statistical software (IBM Corporation) was used. Values were presented as mean±SD. Chi-square test was used for the comparison of rates between groups. The least significant difference method was used for pairwise comparison of counting data between groups. Linear correlation was used to analyze the correlation between BMD and DC, and P<.05 was considered statistically significant.
In the review of the cases from January 2017 to December 2018, 132 cases were eligible for inclusion, and all patients were diagnosed as fresh OVCF by MRI and vertebral compression height less than 40%. There were 34 males and 98 females with a mean age 76.5±9.6 years and fracture segments from T11 to L2.
In this study, after PVP, patients were assigned to the BMD decrease group, osteoporosis group, or severe osteoporosis group according to different BMDs. All of the cases were successfully operated on without bone cement leakage or serious postoperative complications, and there was no significant difference in VAS and ODI scores between groups preoperatively (P>.05). The DV in the 3 groups was larger than the IV (P<.05). There was no statistically significant difference in the IV among the 3 groups (P>.05). There were significant differences in the DC among the 3 groups (P<.05). There was no significant difference in VAS and ODI scores in 12 months after surgery (P>.05), and there was a significant difference in the ratio of vertebral height loss between the 3 groups (P<.05). There were 12 cases of vertebral height loss in the BMD decrease group (33.3%), which was the lowest among the 3 groups, followed by 32 cases of vertebral height loss in the osteoporosis group (59.2%) and 35 cases of vertebral height loss in the severe osteoporosis group (83.3%) (Table 2, Figure 2).
Comparison of Data Between Groups
Bar chart comparing data between groups. *Differences among all 3 groups. Abbreviation: BMD, bone mineral density.
Correlation analysis results showed that there were significant correlations between BMD and IV (−0.716), BMD and DC (0.754), IV and DV (0.502), and IV and DC (−0.666) (P<.01). There was no significant correlation between BMD and DV (P>.05). The IV and BMD correlation was r=−0.716, R2=0.513. The DC and BMD correlation was r=0.754, R2=0.568. Both correlation coefficients were close to 0.8 (strong correlation), as shown in Table 3 and Figure 3.
Correlation Analysis of Bone Mineral Density, Injection Volume, Diffusion Volume, and Diffusion Coefficient
Scatter plots of injection volume and diffusion coefficient with bone mineral density (BMD).
Volume calc software was used to measure the DV, and the differences in IV, DV, and DC between the groups were analyzed. The results showed that there was a significant difference among the 3 groups in IV and DV (P<.01), and the IV was inversely proportional to BMD (R=−0.716). In other words, the IV was the smallest in the BMD decrease group, larger in the osteoporosis group, and largest in the severe osteoporosis group. The DV was positively proportional to BMD (0.754), In other words, the IV was the largest in the BMD decrease group, smaller in the osteoporosis group, and smallest in the severe osteoporosis group. This is because BMD directly influences the quality and quantity of the vertebral body. The higher the BMD, the more trabeculae there are per unit area. Meanwhile, the larger the trabeculae, the more bone mineral content there is. The smaller the bone trabecular clearance, the less bone cement can be filled per unit volume. For the same IV of bone cement, in the same fluidity state, a larger space is required for diffusion. On the other hand, the lower the bone density, the smaller the trabecular bone, the less bone mineral content. The larger the trabecular space, the more bone cement can be contained per unit volume, and the smaller the diffusion range of bone cement.
In the early stage of PVP to treat vertebral compression fracture, as much bone cement as possible is injected into the vertebral body, resulting in a high incidence of complications such as bone cement leakage and allergic reactions.1 In recent years, some clinical and experimental data have shown that a small amount of bone cement injection can achieve satisfactory analgesic effect. Kaufmann et al12 and Molloy et al13 both found that there is no correlation between IV of bone cement and efficacy. Tohmeh et al14 used the diffusion area of bone cement in the vertebral body as an indicator to evaluate the therapeutic effect, and the results showed that the diffusion area was not significantly correlated with the improvement of clinical symptoms. Molloy et al13 found that the filling rate of bone cement in the vertebral body was weakly correlated with the strength and stiffness of the vertebral body. However, the filling rate of the vertebral body in this study was the ratio of the IV to the volume of the vertebral body, which could not fully reflect the distribution of bone cement in the vertebral body. Ye et al8 suggest that insufficient distribution of bone cement may be the cause of unsatisfactory pain relief after the treatment of OVCF by PVP and that the distribution of bone cement is closely related to the efficacy of PVP. Zhang et al15 suggested that controlling the distribution of bone cement during surgery could provide satisfactory postoperative efficacy and reduce the risk of postoperative vertebral recompression. Therefore, scholars gradually began to pay attention to the diffusion of bone cement. Lin et al16 studied the distribution pattern of bone cement and found that it is a potential influencing factor for predicting the postoperative reconstruction effect in unilateral PKP, and the distribution of bone cement is related to the recovery rate of anterior vertebral height and the risk of bone cement extravasation. The higher the distribution of bone cement, the better the vertebral repair effect but the higher the risk of bone cement extravasation. Wang et al17 found that preoperative imaging data were measured to determine the puncture point and insertion point of PKP. According to the measured data of intraoperative puncture, unilateral puncture can achieve the effect of bilateral puncture in the treatment of thoracolumbar compression fracture. According to the existing research, neither IV, diffusion area, nor bone cement vertebra filling rate is suitable for evaluating efficacy.
Due to the irregular diffusion of bone cement in the vertebral body and its 3-dimensional distribution, this spatial diffusion may affect the physical performance of the vertebral body. Poor bone cement diffusion may cause many complications. Many factors influence the diffusion of bone cement in vertebral bodies. For example, the injection system of bone cement, injection timing, injection pressure, physical and chemical properties of bone cement, fracture type, and injection site all affect the diffusion of bone cement.8 To study the effect of bone density on the diffusion volume of bone cement, these influencing factors need to be unified to reduce the interference factors, but this is difficult to achieve. Therefore, the DC was selected to evaluate the efficacy in this study, because it can eliminate the differences in influencing factors among different groups and only focus on the distribution pattern of bone cement at the same volume. Therefore, the DC is more suitable to reflect the reconstruction of vertebral structure after PVP.
Bone mineral density is an absolute value and an important indicator of bone strength (grams per cubic centimeter). The higher the BMD, the greater the vertebral body strength, and the larger and richer the trabecular bone in the vertebral body. Therefore, the smaller the inter-space volume in the vertebral body, the larger the diffusion range of bone cement, and the larger the diffusion, which is not easy to form stress concentration. On the contrary, the smaller the BMD, the weaker the vertebral body strength, and the sparser and more deficient the trabeculae in the vertebral body. Therefore, the larger the space volume in the vertebral body, the smaller the diffusion range of bone cement and the smaller the diffusion, and the easier to form stress concentration. In this study, the BMD was proportional to the DC, and the incidence of vertebral height loss was lower in patients with high DC during follow-up, which could predict the vertebral height loss of patients after PVP in clinic, and advance intervention could be carried out, such as wearing lumbar back braces, or strengthening anti-osteoporosis treatment.
Generally speaking, the higher the degree of fracture compression, the more serious the damage to the internal structure of the vertebral body.18 Too much compression is not suitable for the study of the diffusion effect of bone cement in the vertebral body.15 Therefore, in this study, the compression height of the vertebral body was not more than 40%, which reduced the volume change of the vertebral body and led to the variation of diffusion volume. Therefore, the results of this study are not applicable to the cases with severe compression. In addition, cases with vacuum or effusion were also excluded, because bone cement could not diffuse and gather in the middle of the gap, it will affect the results.
Percutaneous vertebroplasty can rapidly relieve pain in patients with OVCF, but the biomechanics of bone cement materials and vertebral bodies are significantly different.19 Therefore, complications such as vertebral height loss, proximal segment fracture, and bone cement loosening occur during long-term follow-up. This situation has led to the current controversy on PVP treatment. In this study, the DC of bone cement could predict the efficacy of patients after short-term PVP surgery, but the basic biomechanical research was not enough to support this view, with a larger sample being needed.
The DV is larger than the IV in PVP, and the DC is proportional to BMD. This short-term follow-up of 12 months showed that the higher the DC, the lower the vertebral height loss ratio.
- Martin JB, Jean B, Sugiu K, et al. Vertebroplasty: clinical experience and follow-up results. Bone. 1999;25(2)(suppl):11S–15S. doi:10.1016/S8756-3282(99)00126-X [CrossRef] PMID:10458267
- Mansour A, Abdel-Razeq N, Abuali H, et al. Cement pulmonary embolism as a complication of percutaneous vertebroplasty in cancer patients. Cancer Imaging. 2018;18(1):5. doi:10.1186/s40644-018-0138-8 [CrossRef] PMID:29422089
- Hatzantonis C, Czyz M, Pyzik R, Boszczyk BM. Intracardiac bone cement embolism as a complication of vertebroplasty: management strategy. Eur Spine J. 2017;26(12):3199–3205. doi:10.1007/s00586-016-4695-x [CrossRef] PMID:27535287
- Wang D, Zheng S, Liu A, et al. The role of minimally invasive vertebral body stent on reduction of the deflation effect after kyphoplasty: a biomechanical study. Spine. 2018;43(6):E341–E347. doi:10.1097/BRS.0000000000002317 [CrossRef] PMID:28678108
- Lin D, Hao J, Li L, et al. Effect of bone cement volume fraction on adjacent vertebral fractures after unilateral percutaneous kyphoplasty. Clin Spine Surg. 2017;30(3):E270–E275. doi:10.1097/BSD.0000000000000204 [CrossRef] PMID:28323711
- Xu W, Wang S, Chen C, et al. Correlation analysis between the magnetic resonance imaging characteristics of osteoporotic vertebral compression fractures and the efficacy of percutaneous vertebroplasty: a prospective cohort study. BMC Musculoskelet Disord. 2018;19(1):114. doi:10.1186/s12891-018-2040-8 [CrossRef] PMID:29650014
- Yu W, Xiao X, Zhang J, et al. Cement distribution patterns in osteoporotic vertebral compression fractures with intravertebral cleft: effect on therapeutic efficacy. World Neurosurg. 2019;123:e408–e415. doi:10.1016/j.wneu.2018.11.181 [CrossRef] PMID:30503288
- Ye LQ, LiangJiang XB, et al. Risk factors for the occurrence of insufficient cement distribution in the fractured area after percutaneous vertebroplasty in osteoporotic vertebral compression fractures. Pain Physician. 2018;21(1):E33–E42. PMID:29357338
- Bae JS, Park JH, Kim KJ, Kim HS, Jang IT. Analysis of risk factors for secondary new vertebral compression fracture following percutaneous vertebroplasty in patients with osteoporosis. World Neurosurg. 2017;99:387–394. doi:10.1016/j.wneu.2016.12.038 [CrossRef] PMID:28012889
- Henderson R. Expert's comment concerning Grand Rounds case entitled “Intracardiac bone cement embolism as a complication of vertebroplasty: management strategy” by Hatzantonis C, Czyz M, Pyzik R, Boszczyk BM. (Eur Spine J; 2016. doi:10.1007/s00586-016-4695-x). Eur Spine J. 2017;26(12):3206–3208. doi:10.1007/s00586-017-5089-4 [CrossRef] PMID:28405848
- Xu Z, Hao D, He L, et al. An assessment system for evaluating the severity of thoracolumbar osteoporotic fracture and its clinical application: a retrospective study of 381 cases. Clin Neurol Neurosurg. 2015;139:70–75. doi:10.1016/j.clineuro.2015.09.006 [CrossRef] PMID:26383865
- Kaufmann TJ, Trout AT, Kallmes DF. The effects of cement volume on clinical outcomes of percutaneous vertebroplasty. AJNR Am J Neuroradiol. 2006;27(9):1933–1937. PMID:17032870
- Molloy S, Mathis JM, Belkoff SM. The effect of vertebral body percentage fill on mechanical behavior during percutaneous vertebroplasty. Spine. 2003;28(14):1549–1554. doi:10.1097/01.BRS.0000076831.38265.8D [CrossRef] PMID:12865843
- Tohmeh AG, Mathis JM, Fenton DC, Levine AM, Belkoff SM. Biomechanical efficacy of unipedicular versus bipedicular vertebroplasty for the management of osteoporotic compression fractures. Spine. 1999;24(17):1772–1776. doi:10.1097/00007632-199909010-00004 [CrossRef] PMID:10488505
- Zhang L, Wang Q, Wang L, Shen J, Zhang Q, Sun C. Bone cement distribution in the vertebral body affects chances of recompression after percutaneous vertebroplasty treatment in elderly patients with osteoporotic vertebral compression fractures. Clin Interv Aging. 2017;12: 431–436. doi:10.2147/CIA.S113240 [CrossRef] PMID:28260871
- Lin J, Qian L, Jiang C, Chen X, Feng F, Lao L. Bone cement distribution is a potential predictor to the reconstructive effects of unilateral percutaneous kyphoplasty in OVCFs: a retrospective study. J Orthop Surg Res. 2018;13(1):140. doi:10.1186/s13018-018-0839-5 [CrossRef] PMID:29880007
- Wang XF, Fan YF, Shi RF, Deng Q, Li ZF. [The relationship between angle of puncture and distribution of bone cement of unilateral percutaneous kyphoplasty for the treatment of thoracolumbar compression fractures]. Zhongguo Gu Shang. 2015;28(8):704–707. PMID:26502519
- Widmer Soyka RP, Helgason B, Hazrati Marangalou J, van den Bergh JP, van Rietbergen B, Ferguson SJ. The effectiveness of percutaneous vertebroplasty is determined by the patient-specific bone condition and the treatment strategy. PLoS One. 2016;11(4):e0151680. doi:10.1371/journal.pone.0151680 [CrossRef] PMID:27100630
- Zhang J, Fan Y, He X, Du J, Hao D. Bracing after percutaneous vertebroplasty for thoracolumbar osteoporotic vertebral compression fractures was not effective. Clin Interv Aging. 2019;14:265–270. doi:10.2147/CIA.S192821 [CrossRef] PMID:30787602
Assessment System of Thoracolumbar Osteoporotic Fracture
| Compression fracture||1|
| Burst fracture||2|
|Magnetic resonance imaging|
| Long T1 and T2 signal||1|
| Vacuum or effusion in vertebral body||2|
|Bone mineral density|
| T between −2.5 and −3.5||1|
| No obvious pain||0|
| Pain does not need pain-killers||1|
| Pain needs painkillers||2|
Comparison of Data Between Groups
|Group||No. of cases||Mean±SD||No. vertebral height loss|
|BMD||IV, mL||DV, mL||DC||VAS-12M||ODI-12M|
|A: BMD decrease group||36||−1.36±0.71a||4.84±0.55a||11.26±1.92d||2.57±0.31a||2.5±1.2a||31.7±11.8a||12 (33.3%)a|
|B: Osteoporosis group||54||−2.7±0.20b||5.53±0.63b||12.29±1.74e||2.22±0.22b||2.3±1.3b||30.4±10.6b||32 (59.2%)b|
|C: Severe osteoporosis group||42||−3.55±0.33c||6.61±0.73c||11.91±1.57f||1.80±0.17c||2.4±1.5c||33.4±9.6c||35 (83.3%)c|
Correlation Analysis of Bone Mineral Density, Injection Volume, Diffusion Volume, and Diffusion Coefficient