Experimental analysis of vertebroplasty: biomechanic and techinical safety analysis

Abstracts

Analisou-se a segurança e a resistência mecânica de 30 corpos vertebrais, de cadáveres humanos, submetidos à vertebroplastia. Após a densitometria óssea, foram distribuídos em dois grupos A e B. Os do grupo A foram comprimidos até o segundo pico de força (Gráfico Força por Deformação), enquanto os do grupo B, até o primeiro. Injetou-se cimento acrílico no corpo vertebral por via transpedicular, observando-se seu extravasamento e a temperatura do corpo vertebral. Testou-se a resistência e comparou-se com a da própria vértebra quando íntegra. Esta técnica traz riscos às estruturas próximas, devido ao extravasamento do cimento.Quanto à temperatura, ela é segura desde que não ocorra o extravasamento. A resistência da vértebra, após o procedimento, depende do grau do achatamento inicial. Aquelas que tiveram achatamento menor apresentaram resistência maior.

vertebroplastia; cimento ósseo

ARTIGO ORIGINAL

Experimental analysis of vertebroplasty: biomechanic and techinical safety analysis

Alexandre Felipe França^I; Tarcísio Eloy Pessoa de Barros Filho^II; César Augusto Martins Pereira^III

^IOrthopaedics Doctor. Teacher in charge of Orthopaedics in FAFICA from Catanduva

^IIAssociate Professor and PhD at Spine Surgery Service

^IIITecnologist at LIM-41

^{Correspondence}

SUMMARY

It was analyzed the safety and mechanical resistance of 30 vertebral bodies obtained from human cadavers submitted to "vertebroplasty" were analyzed. After bone "densitometry", they were distributed into two groups: A and B. In group A, they were compressed to the second apex of force (graph by Force of Deformation), while in the group B, to the first. Acrylic cement was injected in the vertebral body by transpedicular access. It was observed the cement leakage and the temperature of the vertebral body. The resistance was tested and compared to the one of vertebrae not compressed before. The surrounding structures can be damaged due to the extravasation of the cement. Considering the temperature, this technique is safe, since extravasation does not take place. The resistance of the vertebra after the procedure is dependent on the degree of initial flattening, since bodies which had smaller flattening, presented the largest resistances.

Key words: vertebroplasty, bony cement.

INTRODUCTION

Osteoporosis is related to bone fractures, leading to pain, impairment and, indirectly even death. Thus, it can be stated that it produces a high social and financial cost to the society. The USA had in 1995 a total expense of 13 billion USD with osteoporosis, being 6 billion USD of direct expenditure with patients, 4 billion USD with nursery assistance and two with other services⁽³²⁾.

This disease⁽¹⁶⁾ is said to cause a new compression fracture of a vertebral body every 45 seconds, with a total of 700.000 cases each year, 260,000 of them with no symptoms, resulting in a hospital cost of 800 million USD in 150.000 in hospital treatments. In average, patients with a compressive fracture of the spine remain in hospital for eight days, and 14 days in bed rest due to pain.

Painful symptoms of patients with osteoporotic spine fracture ranges very much. Authors report since totally asymptotic patients, up to pain not possible to manage by means of a conservative treatment. It is in general considered that the pain usually remits with analgesics, rest and physiotherapy, and eventually any kind of orthesis immobilizing the spine⁽³¹⁾. A stabilizing surgery of a spine fracture due to osteoporosis is very rare, since in most of the cases there is no medullar or radicular compression that is able to produce a neurologic alteration. Surgery indication is restricted to the neurologically compromised patients, or those with pain that is untreatable through the above-mentioned methods⁽¹⁹⁾.

Cases with surgical indication are restricted to 0.2% of all osteoporosis cases related to spine fracture. However, when it is necessary to perform medullar decompression and fixation of osteoporotic spine, we face an important problem, since these patients generally have other associated clinical problems limiting the surgery extent. Additionally, the quality of the bone in osteoporotic patients is not favorable to allow a rigid fixation of a spine fracture, and consequently an adequate stabilization is hardly achieved⁽¹⁹⁾.

Some authors have lately recommended the use of a technique called vertebroplasty, which consists in injecting bone cement (polymethylmetacrilate) inside the vertebral body, using a percutaneous way for cervical spine and transpedicular for dorso-lumbar. This kind of approach allows to rapidly restoring mechanical stability of the spine, as well as fast relief of patient's pain^{(11,15,21,22)}.

In our work we aim to demonstrate if the vertebroplasty is a safe procedure for anatomical structures next to vertebral body, as well as to compare the biomechanical resistance of the fractured vertebra after submitted to vertebroplasty with the resistance of an healthy vertebra.

MATERIALS AND METHODS

Eight segments of lumbar spine, LI-LV were obtained from human cadavers, seven males and one female, ages Rangoon from 35 to 72 years, average 51.87 years. According to race, five were white, two non-whites, one black, all of them from the Serviço de Verificação de Óbitos da Capital (SVOC - São Paulo).

The parts were got from fresh bodies, positioned in ventral decubitus, accessing the spine through a posterior median approach, over spinous processes, dissecting skin, subcutaneous and paravertebral muscles from tenth thoracic vertebra to sacrum. Lumbar segment was carefully removed as a block, and soft tissues removed, remaining only bones and intervertebral discs. The pieces were numbered according to the SVOC number and packed in plastic bags, and then frozen at -20° C.

Identified and frozen spine segments were defrosted immersed in saline solution at 0.9% and took for bone densitometry in order to detect whether or not osteoporosis was present, according to World Health Organization standards.

Bone densitometry was performed by immersion of the spine segment in a plastic recipient with five liters of water. The segment was immobilized using plastic bricks with higher density in relation to water. The equipment used was a Lunar® model DPX, and the analysis result took into consideration weight, height, age and gender of the donator.

The two samples withdrawn due to densitometry results with more than 2.5 standard error bellow reference population, i.e. up to forty years, were used in preliminary assays however not included in statistical analysis.

Preliminary assays were performed to study the biomechanical behavior of lumbar vertebrae submitted to compressive loads. These assays were recorded in graphs of force versus deformation. It was observed that when a vertebra is compressed the graph presents an ascending curve up to reach a first force peak, and following a descending curve, called trough and finally a retake of the applied force up to a second force peak. After this last force peak, there is a load plateau related to the maximal compactation of the vertebra (Figure 1).

The six spine segments left were divided into two groups: A and B, with three spine segments each. Vertebrae from group A were submitted to the following: preparation of proof body, intact vertebra compression assay, vertebroplasty, observation of safety parameters and post vertebroplasty compression assay.

In group B, vertebrae were submitted to the same procedures of those from group A, however the compression over the intact vertebra was smaller. Intact vertebral bodies from group A were compressed up to the start of the force plateau in the graph and in group A compression ceased at the first force peak.

Each spine segment was composed by five vertebrae called LI, LII, LIII, LIV and LV, which were separated at intervertebral disc level resulting in units. Each vertebral unit was identified by a letter from its group, A or B, and for an arabian number one to three corresponding to the number of the lumbar segment, plus its anatomical designation, for instance, A2LI and B3LIV.

PREPARATION OF THE PROOF BODY

The proof body for the assay was prepared initially measuring the height of the vertebral body at its central portion, using a Mitutoy^® digital pachimeter with a centi-mm precision. Next step was to cement the vertebral body to two metal cylinders, above and below. For this it was used 150 mg of powder of acrylic polymer with 50 ml of methylmetacrylate liquid monomer brand Jet^® for odontologic use. Cementing the vertebral bodies, besides fixing it to the metal cylinders allows fulfilling eventual irregularities of the superior and inferior surfaces of the vertebral bodies, allowing a better distribution of the tensions. The alignment and adjust of height of vertebral bodies during hardening of methylmetacrylate were granted by a device called cementation device (Figure 2). This device has two metallic rings, with an internal diameter allowing fitting the metallic cylinders. The metallic cylinders of the proof body were connected to each other by screws, fixing them by three points, thus fixating the cylinders to the rings. The uniform height between the sets metallic cylinders/rings was kept by three threaded rods placed perpendicularly to the sets. The rods crossed the rings through holes and fixed to them by nuts.

COMPRESSION ASSAY

Biomechanical assays were performed in a Kratos 5002 Universal Mechanical Assay Machhine, with a load cell of five tf and connected to a PC with a software developed by the LIM-41 Biomechanics Laboratory from Instituto de Ortopedia e Traumatologia do Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo, for saving data from the assay machine, such as force displacement and measured force by load cell (Figure 3).

Compression assays were performed by using a mechanical device allowing to apply the load application and correct positioning of proof body together in relation to the assay machine.

As the mobile bar of the assay machine moves down the load cell (Figure 4-A) applies a compressive load over the proof body threaded rod (Figure 4-B), connected to the actuator (Figure 4-C) through a ball and socket joint. The proof body (Figure 4-F) was involved by a superior cylinder (Figure 4-D) as well as by an inferior (Figure 4-E) and the latest was connected to a cylinder (Figure 4-G) and to a tube (Figure 4-H) which served as support.

Compression assay of intact vertebrae for groups A and B differentiated in regard to the deformation they were submitted, since in group A compression was performed up to a deformation related to the start of the force plateau, and in group B compression was performed up to the first force peak (Figure 1).

After the vertebroplasty, vertebrae of each group were again submitted to compression assay. Criterion for stopping the assay was related to the recorded force, that is, when this load was equal or superior to the maximal force found in the previous intact vertebrae assay.

Compression assay of intact vertebra and submitted to vertebroplasty for groups A and B were related to the objective of comparing the rigidity and the deformations.

Rigidity is defined as the relation between force variation and deformation variation between two points belonging to the linear region of the curve (Figure 5) and expressed in N/mm.

For determining the deformations, during the compression assay it was used as reference the peak force in intact assay (FPico) and its deformation (DPint). By projecting FPico over vertebroplasty assay curve, it was possible to find the deformation (DPVer) related to the same applied force (Figures 6, 7, 8, 9).

VERTEBROPLASTY

The vertebral pedicle was perforated with a 3.5 mm drill, and cannulated by a plastic tube 5cm long up to reach inside vertebral body, thus granting the acrylic cement to be injected inside the vertebral body. Perforation of the pedicle was made at the intersection point between two imaginary lines, the first horizontal through the middle of the transversal processes of the vertebra and the second, vertical, crossing the middle of the articular facet.

A mixture of 40 g of a powder (composed by 30 g of co-polymer of methylmetacrylate, 6g of polymethylmetacrylate and 4g of barium sulfate) with 20 ml of liquid monomer (19.5 ml of methyl metacrylate, 0.5 ml N,N-dimethyl paratoluidine and 1.5 mg of hydroquinone). The use acrylic cement is the Simplex^® from Howmedica^®. After the mixture a soft paste was obtained which was injected inside the vertebral body through the vertebral pedicle cannulated with the plastic tube. The injected volume of acrylic cement was of 9 ml and for injecting it was used a plastic syringe of 20 ml connected to the plastic tube. Just after the vertebroplasty the safety parameters were checked for both groups, A and B, consisting in observing the places from where the cement leaked and measuring the temperature next to posterior wall of vertebral canal.

SAFETY PARAMETERS

The safety parameters were divided regarding temperature and cement leakage to the canal during the process of vertebroplasty.

Temperature measurement was performed just after transpedicular injection of acrylic cement by the placement of two temperature sensors (thermo-resistors model PT-100) in contact with the posterior wall of the vertebral body (Figure 10), one at the right side, the other one at the left side. Recording of the temperature was started by a RoberShaw® pyrometer model LDT 900. As measure it was considered the obtained just before injecting the cement, called T0. The remaining measurements were performed just after the vertebroplasty with one-minute interval between them, for a period of 15 minutes.

Temperature sensors were involved by a paste, when placed in vertebral canal. This paste improves thermic conductibility between the wall and the sensors.

The other parameter used in observing safety of vertebroplasty was checking the places where the cement leaked from after its injection. We observed thus if the cement leaked outside of vertebral canal, into the canal or if there was no leakage.

ANALYSIS STATISTICS

Frequency distribution both absolute and relative of the qualitative parameters, and descriptive of the quantitative dada was performed, and classified as mean (M), standard deviation (DP), standard deviation of the mean (EPM), maximum value (Max), minimum value (Min), interval (Inter) and total number of samples (N).

Exact Fischer's test was use for comparing qualitative parameters, and Wilcoxon, Mann Withney's U, and t Student test were used for quantitative parameter comparison.

The quantitative parameters analyzed were the measured temperatures at posterior lateral right and left parts of vertebral bodies; regarding the compression assays of lumbar vertebral bodies (groups A and B) it was also analyzed the rigidity, the percentual range of initial vertebral height and its deformation up to the 1^st peak, percentual range between initial vertebra height and its final deformation and the percentual range between rigidity of groups A and B.

In all comparisons, a significance level of 5% (=0.05) was used, and significant results noted with asterisk.

RESULTS

Results regarding safety parameters, local of leakage and posterior cortical temperature after vertebroplasty, are listed in (Tables 1 to 5), stressing that N = 16 in (Tables 3 and 4) refer to the number of measurements of temperature performed in each tested vertebral body.

Biomechanical results are listed in (Tables 6 to 12). Tables 6 to 9 compare biomechanical findings within each group, while (Tables 10 to 12) compare results between groups A and B.

DISCUSSION

The spine can be affected by a number of diseases resulting in bone mass reduction, thus increasing the fragility of the vertebral body. Among these, we highlight hemangiomas, lythic tumors and osteoporosis. We understand that bone mass reduction leads to a bone that is unable to resist mechanical and physiological requirements, resulting in vertebral body fractures and their consequences, such as nerve compressions, kyphosis, scoliosis and frequently an important picture of pain⁽¹⁾.

Among the above mentioned diseases, osteoporosis is deserving special attention from several specialties, due to an increasing number of these conditions among world population, mostly in the countries experiencing improvement in food and health conditions. As a consequence, an increase in life expectation has been noticed by some authors^(25,32).

They defined⁽¹⁷⁾ osteoporosis as an skeletal condition characterized by reduction of bone mineral density (mass/volume) of a initially well mineralized bone, enough to produce fractures at minimal traumas.

Densitometer is the equipment of choice for bone mineral densitometry, for having a standard error of 5 to 10% for sensitivity and 2% for specificity⁽¹³⁾.

World Health Organization classified bone density got from densitometry in three categories: normal, as a patient up to one standard deviation from median of young population, up to forty years old, used as reference. The other category is osteopenia, comprising individuals with a densitometric result 1 to 2.5 SD below the reference population, and the last category, as osteoporosis, those with more than 2.5 SD below the reference parameters.

Spine fracture associated to osteoporosis is found in 16% of the white American women after menopause⁽²⁵⁾. These patients should be treated, in their vast majority, in a conservative manner, with painkillers, NSAIDs, orthesis, rest and physiotherapy.

Surgical treatment of a spine fracture related to osteoporosis is very rare, being indicated only in those patients with untreatable pain with the conservative methods, and associated nerve compression^(19,31).

However, when it is necessary to stabilize a spine with reduced mineral density, we face a great problem and stabilization should include multiple points of fixation in the spine, using for this transpedicular screws⁽¹⁹⁾. However, this is a too aggressive surgical practice for patients who generally have frequent association of other pathologies. Furthermore, when these transpedicular screws are inserted in less mineralized bones, the fixation is not as good as when they are placed in a normally mineralized bone⁽³⁾.

A new technique was described⁽¹⁵⁾ for spine stabilization, able also to improve pain in patients with hemangioma and lythic metastatic tumors involving the region. This technique is called vertebroplasty, and consists in injection of polymethylmetacrylate inside the vertebral body under a fluoroscopic view.

Other authors started to indicate vertebroplasty for treatment of patients with vertebral fracture associated to osteoporosis with pain not improving with conventional treatments, or with neurologic compression associated to these fractures^(6,11,21).

Most of the publications we found in literature regarding vertebroplasty were reviews of clinical cases^(2,24,33). For this reason we decided to perform a biomechanic study to test the safety and resistance of vertebras to compression forces, before and after injection of the acrylic cement inside the vertebral body, through a transpedicular way.

We choose to use models obtained from human cadavers, even though knowing the impossibility to have a group with homogeneous characteristics of race, gender, age, vertebra geometry and bone mass.

We agree with⁽²⁸⁾ that it is thanks to this variability, normally found between individuals of the same species in nature that is the importance of works using models got from human cadavers, thus demonstrating the behavior of these structures when submitted to a biomechanic assay. For example, the response of human cadaver's lumbar vertebrae to a compressive force after injection of acrylic cement inside the vertebral body.

We disagree from⁽²⁷⁾ who proposed the use of an experimental study model in computer, called finite elements model, as a way to reduce the variability found between individuals in a same species. We highlight that it is not found in nature any individual, from any species two identical individuals. For this, we believe that an experimental work must reflect the findings of a group of the same species to the same problem. However we also believe that it is important limit as possible this variability. For this we did not choose human cadavers with vertebral tumor or previous spine fracture; a bone densitometry was performed and spines with results under 2.5 SD were discharged, and additionally we used the vertebra itself as its own control.

To perform the densitometry, we faced the problem of devices used to measure bone densitometry are drawn taking into account existing soft tissues. This was solved with immersion of the lumbar segment in a plastic recipient with water, since, according to the supplier of Lunar® DPX densitometer, soft tissues surrounding the bones have a similar density to water.

The tests we performed in the two spine segments not included in the study were useful for determining vertebra deformation when a compression force was applied. A similar behavior as described by⁽²⁶⁾ was found, characterized by a graph with a peak of force, followed by a drop in force and a second peak in force increase up to reach a plateau.

The first peak means the resistance limit of the vertebral body cortical. The drop, that we called trough, is correspondent to the resistance of the cancellous bone of the vertebral body. The second peak happens due to the complete compactation of both corticals of the vertebral body and cancellous bone, producing a graphic image of a plateau, a smaller deformation of the vertebral body when submitted to growing compressive loads.

It is necessary to explain that the vertebral body has a physical behavior of deformation similar to coil connections, where the superior and inferior corticals have a different behavior from the cancellous layer. Vertebroplasty introduces a third element inside the vertebral body, the polymethylmetacrylate, which can or not influence the ability of resisting a growing compression force.

Our word is divided into two groups of lumbar vertebrae, and what differentiates both groups is the moment of the graph force versus deformation, above described, we chose to inject bone cement inside the vertebral body. In group A, the chosen moment was after the second peak of force versus deformation curve and in group B, just after the first peak of force.

In our results we demonstrate that there is a statistical difference between the degree of compression of the intact vertebra between groups A and B (Table 8). This means that the moment of the graph force versus deformation chosen for both groups in truly different. Introduction of polymethylmetacrylate inside the vertebral body in different degrees of vertebral compression can lead to different responses, when submitted to a compression force.

Preparation of the bodies for the assay was a very important step, as stressed by⁽¹³⁾, since we need to reduce the irregularities of the vertebral body so load application is uniform over them. We used odontologic use polymethylmetacrylate as a cushion between the vertebral body and the metallic cylindrical recipients receiving compressive load. For granting uniform distribution of the cement, we developed a device to keep a homogeneous cementation during the drying process of the cement.

Another important phase in our work that deserves to be discussed it preparation of the polymethylmetacrylate to be injected inside the vertebral body. Several authors^(7,8,18,20) described bone cement as a substance possessing among its qualities to be easy to handle, ductility, elasticity and resistance. These features made bone cement very much used in orthopedic procedures, mainly arthroplasties.

We believe that the use of acrylic cement is not free of risks, both for the patient and the team. Heat release⁽⁴⁾ during the polymerization process can cause damage to tissues in direct contact with it, and can lead even to necrosis. Besides this, gases released during preparation can cause hepatic injuries both to the patient and the surgical team.

The use of polymethylmetacrylate in spine started as a substance able to replace a vertebral body affected by a vertebral tumor, and to give immediate stability to posterior arthrodesis in patients with instability and no conditions for obtaining bone graft. The use of bone cement in this fashion⁽²³⁾ does not give a durable stability and risk of future problems is high and difficult to solve. It is stated⁽³⁸⁾, also after experimental assays that bone cement should not be used as bone graft in places with high mechanical requirements.

The way to use bone cement, proposed by⁽¹⁵⁾ has a huge difference to those who replace the vertebral body for polymethylmetacrylate cylinders, since the vertebral body is not removed and the acrylic cement is used to reinforce bony structures of the vertebral body, theoretically increasing its resistance.

Polymethylmetacrylate we used in our study was the proposed by the brand Howmedica^®, and is the same used in arthroplastic surgeries. During preparation process, we preferred to have the cement in a paste consistency, according to recommendation by several authors^{(10,21,24,33)}.

The volume of injected polymethylmetacrylate in our work was of 9 ml. Regarding this aspect, there is no consensus in literature, since the authors above mentioned used volumes ranging from 2 to 9 ml. They defined⁽¹¹⁾ the technique as percutaneous injection of bone cement in a vertebral body which was involved in a process that reduced its resistance and recommended a volume of bone cement ranging from 2 to 9 ml.

The transpedicular way was chosen for bone cement injection due to, unlike cervical region, where the vertebral body is closer to the skin, in lumbar region the anterior approach of a percutaneous access can lead to injuries of structures surrounding the spine. We perforated only one of the vertebral pedicles, and cannulated it with a plastic tube to inside the vertebral body, thus allowing the bone cement to reach it. This choice is supported by authors like^(10,12).

Place of leakage of bone cement is an important parameter for safety evaluation of vertebroplasty, since according to the local of the leakage, serious damage can be caused to neighbor structures. This injury can happen during polymerization of the cement, which releases heat, or after its ending, when a root or medulla compression can take place⁽³⁵⁾.

In out results presented in (Tables 1 and 2), there was no statistically significant difference for distribution frequency of place of leakage of the bone cement. However it is worthy to highlight that only 6.67% of bone cement injected vertebrae didn't have any leakage and 26.66% of injected vertebrae had cement leakage into vertebral canal.

Analysis of these data leads us to conclude that the consistency of injected bone cement is an important factor regarding leakage, since a given volume of cement at a more solid consistency reduces the risks of its leakage through the vertebral body, even making difficult its injection through transpedicular way, when compared to a more liquid consistency. The used consistency of bone cement still allowed leakage through clefts presents after the compression assay of intact vertebral body. We agree with⁽¹⁰⁾ who says that it is due to these lesions of vertebral body that polymethylmetacrylate leaks during vertebroplasty.

These findings are helpful to understand clinical complications, such as neuritis and radicular or medullar compressions found in review papers of vertebroplasty^{(2,10,11,34,35)}. So, we believe that diseases producing destruction of the posterior cortical of vertebral body which are detectable via radiograph or CT constitute an absolute contra-indication of vertebroplasty.

The use of a balloon via transpedicular to vertebral body, and injection of bone cement inside it (warranting cement not to leak) has been recommended in the last two years⁽³⁶⁾. However this balloon is not available in Latin America yet.

Finally, the use of continuous fluoroscopy during the injection of bone cement inside the vertebral body, together with use of polymethylmetacrylate mixed to radio-opaque substances is as well a way to prevent leakage and their complications⁽¹⁰⁾.

Polymerization of polymethylmetacrylate is an exothermic reaction resulting in temperature increase ranging from 40 to 90° C in tissues surrounding the cement^(4,20).

In our study we measured vertebral body temperature, before and after cement injection by the use of two temperature sensors at posterior wall of vertebral body, inside vertebral canal, close to right and left pedicle. After vertebroplasty excess of cement was removed when necessary, from inside the canal and measured the temperature at the right pedicle (where the cement was injected). In the other sensor, placed at the other side, we measured the pedicle temperature (where cement was not injected). It is known that polymerization reaction takes 15 minutes. For this, in our study we monitored the temperature during this period⁽¹⁸⁾.

Results in (Table 3) demonstrate that in all tested vertebrae there was a statistically significant difference in the sensor located close to the pedicle where the cement was injected and the other one, on the opposite side. In the first, temperature was higher than in the opposite side. (Table 4) also demonstrates that there was a temperature difference between the sensors, which ranged from 1.35 to 2.66°C.

We stress that observation of this data can lead us to think that polymethylmetacrylate injected at cake icing consistency doesn't have the ability to distribute itself inside all the vertebral body, thus there is a higher concentration of bone cement close to the pedicle were it was injected, thus contributing to temperature increase in this area. Therefore we disagree from the authors who in revision of clinical studies believed that the acrylic cement could be distributed inside the vertebral body^(11,15). We recommend that, for an homogeneous distribution, vertebroplasty should be performed with injection through both pedicles.

We analyzed as well the time vertebral body takes in average to reach the highest temperature after cement injection (Table 3). We observed that the average highest temperature observed was 31.92° C, in an average time of 7.97 minutes. This mean temperature observed reflects how much the vertebral body increased its temperature in relation to the initial temperature. Besides this, we should during this time be particularly careful regarding the release of toxic substances during cement polymerization. This temperature increase of the vertebral body in not able to harm anatomical structures close to vertebral body. However, it does not reflect the real temperature of bone tissue closest to polymethylmetacrylate, nor of the leaked cement, however we do know that this temperature can reach 90° C⁽⁴⁾.

We believe that a high temperature during bone cement polymerization is one of the factors responsible not only for pain relief not only due to mechanic stabilization of vertebral body, as reported in several clinical studies^(9,14,15,22). According to the work of⁽⁴⁾, the high temperature produces necrosis of tissues closest to the cement, including nervous terminations responsible for pain, thus, supporting our position.

When the degree of deformation of intact vertebrae from groups A and B in relation to the initial assay, that is, without vertebroplasty, we find in our results (Table 10) that there was a statistically significant difference between the groups. Group A vertebrae had an average deformation of 23.34% in relation to intact vertebra, while in Group B the deformation was of 9.37%. This way we made sure that vertebroplasty was performed in different degrees of compression.

The authors^(1,28,29) stressed in their works the importance of bone mass in vertebral body resistance to bear its physiological loads. However, there are diseases like lythic tumors, hemangiomas and marked osteoporosis that reduce the natural resistance of the vertebral body. The use of bone cement to increase bone resistance as suggested^(5,11), doesn't seem to be a completely true statement, due to the results found in our work. We observed that vertebrae in Group A that received polymethylmetacrylate injection, after total compression presented as less resistant than intact vertebra. We understand that, from a vertebroplasty point of view, resistance of this vertebral body to compressive load is dependent of the degree of flattening the vertebra is.

The results of rigidity measurement that we found in biomechanical assays in Group A (Table 4) demonstrate that there is an statistical difference between intact vertebra rigidity and after vertebroplasty. This means that introduction of bone cement does not contributes to restore the resistance of the vertebral body when submitted to compression load, since vertebral body resistance after vertebroplasty was approximately 66.10% lesser than intact vertebra.

In group B we observed a different biomechanical behavior. There was no significant statistic difference between average rigidity of intact vertebra and after vertebroplasty. We found that polymethylmetacrylate helped keeping vertebral body resistance to a compressive load which was close to intact vertebra.

Comparison of variation of the rigidity between groups A and B demonstrates that there was an statistically significant difference between them. Thus, group B presented less variation of the rigidity, whit a more uniform behavior than observed in group A.

Another important data we found is in regard to percentual variation of initial height of the vertebra to the first peak of compression assay of groups A and B (Table 12). We noticed that the percentage of initial height lost at group A (10.57%) was statistically significant in relation to group B (5.29%). This means that vertebrae of this group suffered more deformation when underwent the same compressive load.

All these findings demonstrate that vertebroplasty, when indicated before vertebrae are completely flattened,group B can contribute to keeping the resistance of the vertebral body, without however being able to keep the same resistance of an intact vertebra.

Vertebroplasty is a valid technique for treatment of diseases reducing vertebral body resistance, when indicated in vertebrae not very much flattened, respecting the clinical criteria for method indication. However, we need to improve the safety of the method to reduce risks of injury to nervous structures. For this we mentioned the use of a balloon to avoid leakage^(16,36). And the use of acrylic cements releasing lower amounts of heat and ability to integrate to the bone without loosing the resistance conventional polymethylmetacrylate has^(5,6) are measures more recently introduced which, in a short term will bring benefits helping to perform more safe vertebroplasties.

CONCLUSIONS

1) Vertebroplasty is a technique that can harm anatomical structures that are close to vertebral body, due to the possibility of polymethylmetacrylate leakage.

2) Once there is no leakage, temperature observed at posterior cortical of vertebral body, after vertebroplasty, does not reach values considered as harmful to anatomical structures inside the vertebral canal.

3) Biomechanical resistance after vertebroplasty is dependent on how much the vertebral body was flattened. Vertebrae with less initial flattening, when submitted to vertebroplasty presented a higher biomechanical resistance than those with a more important deformation.

REFERÊNCIAS BIBLIOGRÁFICAS

1. Alho, A: Mineral and mechanics of bone fractures. A review of fixation methods. Acta Orthop Scand 64: 227-32 , 1993.

2. Barr, J.D., Barr, M.S., Lemley, T.J. & MacCann, R.M.: Percutaneous vertebroplasty for pain relief and spinal stabilization. Spine 25: 923-28, 2000.

3. Bennett, G.J., Serhan, H.A., Sorini, P.M. & Willis, B.H.: An experimental study of lumbar destabilization. Restabilization and bone density. Spine 22: 1448-53, 1997.

4. Berman, A.T, Reid, J.S., Yanicko, D.R., Sih, G.C. & Zimmerman, M.R.: Thermally induced bonenecrosis in the rabbits. Clin. Orthop. 186: 284-92, 1984.

5. Bo, B., Laith, M.J., Kummer, F.J. & Spivak, J.M.: The use of na injectable, biodegradable calcium phosphate bone substitute for prophylactic augmentation of osteoporotic vertebrae and the management of vertebral compression fractures. Spine 24: 1521-1526, 1999.

6. Bostrom, M.P.G. & Lane, J.M.: Future directions augmentation of osteoporotic vertebral bodies. Spine 22: 38-42, 1997. Supplement 24.

7. Cloft, H.J., Easton, D.N., Jensen, M.E., Kallmes: D.F. & Dion, J.E.: Exposure of medical personnel to methylmethacrylate vapor during percutaneous vertebroplasty. J. Neuroradiol Am 20: 352-53, 1999.

8. Convery, F.R., Gunn, D., Hugues, D. & Martin, W.E.: The relative safety of polymethylmethcrylate. J. Bone and Joint Surg.A 57:57-64, 1975.

9. Cortet, B., Cotten, A., Deprez, X., Deramond, H., Lejeune, J.P., Leclerc, X., Chastanet, P., Duquesnoy, B. & Delcambre, B.: Intérêt de la vertébroplastie couplée á une décompression chirurgicale dans le traitement des angiomes vertébraux agressifs. A propos de trois cãs. Rev. Rhum.Fr 61:16-22, 1994.

10. Cotten, A., Dewatre, F., Cortet, B., Assaker, R., Leblond, D., Duquesnoy, B., Chastanet, P. & Clarisse, J.: Percutaneuous vertebropasty for osteolytic metastases and myeloma: effcts of the percentage of lesion filling and leakage of methylmetacrilato at clinical folow-up. Radiology 200: 525-30, 1996.

11. Cotten, A. & Duquesnoy, B.: Vertebroplasty: current data and future potencial. Rev. Rhum Engl 64: 645-9, 1997.

12. Dufresne, A.C., Brunet, E., Sola-Martinez, M.T., Rose, M. & Chiras, J.: Vertébroplastie percutanée de la charniére cervico dorsale par voie antérieure. Technique et résultats. À propos de neufcas. J. Neuroradiol 25: 123-28, 1998.

13. Edmondston, S.J., Singer, K.P., Day, R.E., Price R.I. & Breidahl, P.D.: Ex vivo estimation of thoracolumbar vertebral body compressive strength: the relative contributions of bone densitometry and vertebral morphometry. Osteoporosis. Int 7:142-48, 1997.

14. Feydy, A., Cognard, C., Miaux, Y., Sola Martinez, M.T., Weill, A., Rose M. & Chiras, J.: Acrylic vertebroplasty in symptomatic cervical vertebral haemangiomas: report of 2 cases. Neuroradiology 38:389-391, 1996.

15. Galibert, P., Deramond, H., Rosat, P. & Le Gars, D.: Note préliminaire sur le traitement desangiomes vertébraux par vertébroplastie acrylique percutanée. Neurochirurgie 33:166-68, 1987.

16. Garfim, R.S., Lane, J.M., Reiley, M.A. & Yuan, H.A.: Symposium on Advances in treating osteoporotic vertebral bod compression fractures. Chicago, 1999.

17. Glaser, D.L. & Kaplan, F.S.: Definition and clinical presentation. Spine 22:12-16, 1997. Supplement 24.

18. Hass, S.S., Braver, G.M. & Dickson, G.: Characterization of polymethy-methacrylate bone cement. J. Bone Joint Surg.Am 57:380-391, 1975.

19. Hu, S.S.: Internal fixation in the osteoporotic spine. Spine 22:43 48, 1997. Supplement 24.

20. Jefferriss, C.D., Lee, A.J.C. & Ling, R.S.M.: Thermal aspects of self-curing polymethylmethacrylate. J. Bone Joint. Surg.Br 57: 511-518, 1975.

21. Jensen, M.E., Evans, A. J., Mathis, J.M., Kallmes, D.F., Cloft, H.J. & Dion, J.E.: Percutaneous polymethylmethacrylate vertebroplasty in the treatment of osteoporotic vertebral compression fractures: techinical aspects. Am. J. Neuroradiol 18:1987-904, 1997.

22. Lapras, C., Motolesse, C., Deruty, R., Lapras, C., Jr., Remond, J. & Dusquesnel, J.: Injectin percutaneé de méthil métaccrylate dans traitement de l'ostéoprose et ostéolyse vertébralegrave. Ann. Chir. Fr 43:371-76, 1989.

23. Macfee, P.C., Bohlman, H.H. & Ducker, T.: Failure of stabilization of the spine with methylmethacrylate. J. Bone Joint Surg.Am 68: 1145-1157, 1986.

24. Mathis, J.M., Petri, M. & Naff, N.: Percutaneous vertebroplasty treatment of steroid-induced osteoporotic compression fractures. Arthritis Rheum 41: 171-5, 1998.

25. Melton, L. J., III.: Epidemiology of spinal osteoporosis. Spine 22: 2-11, 1997. Supplement 24. 33. Menezes, L.A.F.: Compressão axial de corpos vertebrais humanos ensaios com cargas lentas e progressivas. São Paulo, 1987. 72p. Dissertação (Mestrado) - Faculdade de Medicina, Universidade de São Paulo.

26. Menezes, L.A.F.: Compressão axial de corpos vertebrais humanos ensaios com cargas lentas e progressivas. Sâo Paulo, 1987. 72 p. Dissertação (Mestrado) - Faculdade de Medicina, Universidade de São Paulo.

27. Mizrahi, J., Silva, M., Keaveny, T. & Edwards, W.: Finite element stress anaysis of the normal and osteoprotic lumbar vertebral body. Spine 18: 2088-2096, 1993.

28. Mosekilde, L.: Vertebral structure and strength in vivo and in vitro. Calcif Tissue Ind 53: 121-26, 1993. Supplement 1.

29. Myers, E.R. & Wilson, S.E.: Biomechanics of osteoporosis and vertebral fracture. Spine 22: 25-31, 1997. Supplement 24.

30. Rossi, J.D.M.B.A., Costa, R.C. & Leivas, T.P.: Avaliação mecânica de compósitos de hidroxi-apatita polimetilmetacrilato. Rev. Bras Ortop 28: 167-171, 1993.

31. Tamayo-Orozco, J., Arzac-Palumbo, P., Héctor, P.V. H., Mota Bolfeta, R. & Fuentes, F.: VertebraL fractures associated with osteoporosis: patient management. Am J Med. 103: 44-50, 1997.

32. Tosteson, A.N.A.: Quality of life in the economic evaluation of osteoporosis prevention and

33. Wälchli, B., Berlemann, U.& Heini, P.F.: Vertebroplasty techique and first clinical results on the treatment of osteoporotic. Spine 25: 26-27, 2000. Supplement 26.

34. Weill, A., Chiras, J., Simon, L.M., Rose, M., Sola-Martinez, T.& Enkaoua, E.: Spinal metastases: indications for and results of percutaneous injection of acrylic surgical cement. Radiology 199: 241-247, 1996.

35. Wilkes, R.A., Mackinnom, J.G. & Thomas, W.G.: Neurological deterioration after cement injection into a vertebral body. J bone Joint Surg. {Br} 76: 155, 1994.

36. Wilson, D.R., Myers, E.R., Mathis, J.M., Scribner, R.M., Conta, J.A., Reiley, M.A., Talmadge, K.D. & Hayes, W.C.: Effect of augmentation on the mechanics of vertebral wedge fractures. Spine 25: 158-165. 2000.

*Work performed at Biomechanic Lab (LIdoM-41) from IOT - HC - FMUSP

Publication Dates

• Publication in this collection
  21 Sept 2005
• Date of issue
  Mar 2002

History

• Received
  04 July 2001
• Accepted
  27 Nov 2001