The effect of mechanical aspiration of the vertebral body on pulmonary arterial pressure before cement injection in the vertebroplasty procedure

Introduction

Percutaneous vertebroplasty (VP) is the injection of polymethyl methacrylate into a weakened or osteoporotic compression fractured vertebra under fluoroscopy in order to give pain relief and mechanical strengthening of the vertebral body. It is also used in traumatic pure vertebral compression fractures (VCFs), multiple myeloma metastatic tumors, and symptomatic hemangioma. ¹ The main purpose of VP is pain relief and the strengthening of the vertebrae bones weakened by disease. It is increasingly accepted as one of the options in the management of intractable back pain due to VCFs. ^2,3 Most studies have shown substantial pain relief and increased mobility during the 72 h after VP in up to 90% of patients. ^{1,4 –6} It is an efficient treatment but not free of complications. The main complication of VP is due to cement leaks, fat or thromboemboli, inaccurate needle placement, pain exacerbation, infection, hematoma, and blooding. Other less seen complications are advert cement reaction, anaphylaxis, pedicle fracture, adjacent organs injury, and epidural hematoma. ^{7 –10} Cement leaks into the adjacent structures may be seen in 41% of routine cases. These leaks are generally small and are usually asymptomatic. The symptoms are regardless of the location and amount of the leak. ¹¹ Approximately 2% of the observed leaks are pulmonary emboli. These reported incidences of pulmonary cement embolism are not reliable because most of them are asymptomatic and more leaks are detected by computed tomography. ^{12 –14} In addition to unwanted cement leakage, pulmonary fat embolism also occurs as a result of the displacement of intraosseous bone marrow into the bone vessels. Bone marrow particles cause microembolization of the arterioles and capillaries of the lung resulting in an increase in pulmonary arterial pressure (PAP). In healthy lungs, small embolies can be tolerated. Multiple embolies or large cement emboli can cause a pulmonary infarct and may lead to pulmonary compromise or even death. The acute consequences of fat embolism may be hypotension, cardiac arrest, and in serious cases, sudden death. ^{15 –17} In this prospective and comparative study, we describe a new technique to decrease these types of complications.

Methods

Forty patients having multilevel VP (more than three levels) were included in this study. All of the VP were for osteoporotic compression fractures. Traumatic compression fractures were excluded from the study (Figure 1). We also excluded patients with pulmonary and cardiovascular diseases. The main complaint was pain in the fracture area with or without motion. All the procedures were done under local plus sedation anesthesia. All of the procedures were performed using anterior-posterior and lateral fluoroscopy. The procedure was done by one surgeon and the cement injection was performed manually. PAPs were measured by standard echocardiography preoperatively, after 24 h and 3 days after the procedure and blood d-dimer values were also recorded preoperatively, after 24 h and 3 days after the procedure. In 20 patients (group A), mechanical aspiration of the cavity was done with a flexible aspiration tube just prior to the cement application and in the remaining 20 patients (group B) aspiration of vertebral body did not perform. The stan-dard VP procedure was done with the same size cannula, injection force, injection speed, and cement viscosity. The amount of injected cement was 3 ml for thor-acic levels and 4 ml for lumbar levels. The effect of injecting bone cement was statistically examined using a paired Student’s t-test for the arterial pressure and d-dimer data, comparing the minimum value against the baseline value.

OPEN IN VIEWER

Figure 1.

Aspiration of the vertebral body with a flexible aspirator tube. At the upper segment of the aspirated vertebra, the cement has been injected into the vertebral body.

VP technique

All of the patients were positioned in a prone position on a fluoroscopy table and were under intravenous sedation and analgesics. The skin overlying the area was prepared and made sterile. The skin and periosteum of the pedicle were locally anesthetized with fluoroscopic guidance. After the skin incision, a Jamshidi-type bone trocar was inserted into the pedicle under biplanar fluoroscopy. We removed the needle of the trocar and put a guide wire onto the trocar to the corpus of the vertebra and then the Jamshidi-type trocar was removed. A VP cannula was put into the vertebra on a guide wire. We removed the Kirschner and aspirated the vertebra body with a flexible aspirator of tube size 8fr over the VP cannula in 20 patients and did not aspirate in 20 patients (Figure 1). Cement injection performed with the same size cannula, same injection force, same injection speed and same cement viscosity into the vertebral body.

Results

The mean age was 71 (range: 62–87) in group A and 70 (range: 64–88) in group B. The number of augmented levels was 6.7 in group A and 6.9 in group B. Cement leakage was present in four patients in group A and six patients in group B. The cement leakage toward the neural foramina or spinal canal was not seen in both groups. All of the cement leaks were asymptomatic, acute hypotension was observed immediately after the cement injection in one patient in group A and four patients in group B. The preoperative mean PAP in group A was 35 mm/Hg and elevated to 48 mm/Hg on the first postoperative day and decreased to 42 mm/Hg on the third postoperative day. On the other hand, the mean PAP in group B was 36 mm/Hg preoperatively, 71 mm/Hg on the first postoperative day, and 58 mm/Hg on the third postoperative day (p < 0.05) (Figure 2). The average blood d-dimer values in group A increased from 1.94 to 2.31 and in group B increased from 2.14 to 5.72 (p < 0.05; Figure 3).

OPEN IN VIEWER

Figure 2.

This chart shows changing of PAP before procedure, at first day and at the third day after vertebroplasty. PAP: pulmonary arterial pressure.

OPEN IN VIEWER

Figure 3.

This chart shows changing of d-dimer before and after vertebroplasty.

Discussion

VP is increasingly used for pain relief in patients with osteoporotic compression fractures. VP showed pain relief in 80–90% of patients treated. ^4,18 It was first described by Gilbert and Deramond in 1984 for symptomatic vertebral hemangiomas. ¹⁸ Most of the complications are related to the cement leakage area, which may result in radiculopathy, spinal cord compression, and pulmonary embolism (PE). ^1,3 An acute PE is a cardiovascular emergency with high morbidity and mortality. ¹⁹ If the occlusion affects more than 30–50% of the pulmonary arterial bed, PE becomes hemodynamically relevant, with increased systolic PAP. ¹⁹ Symptoms can include dyspnea, tachypnea, cyanosis, chest pain, cough, hemoptysis, dizziness, or sweating. ²⁰ Consequently cement PE may cause a sustained increase in the PAP and thus will lead to a cardiopulmonary failure. ²¹

In our study, we measure PAP and plasma d-dimer to determine PE. The d-dimer level in plasma is a marker of fibrinolysis and elevated in the presence of acute venous thromboembolism as well as in other clinical conditions. d-dimer blood tests have high sensitivity and have been proven a safe test to rule out PE in outcome studies. ²²

PE can occur due to injected cement leakage or the fat of intramedullary marrow. ²³ During cement applications to the bone, the cement increases intramedullary pressure. Increased intramedullary pressure compresses the intramedullary contents, especially bone marrow fats. They may enter the medullar vessels and embolize the pulmonary vessels. ¹⁶ In patients undergoing total hip arthroplasty, most emboli occur during femoral reaming and the insertion of the femoral stem. Femoral reaming and femoral stem insertion increase intramedullary pressure and result in PE. ²⁴ Hemodynamic changes in total hip arthroplasty were seen during intramedullary femoral stem insertion, conversely during the manual packing of the acetabulum with low pressure not been found in changes in hemodynamics. It also shows that most of the emboli are related to intramedullary pressure. ²⁵ These studies on hip arthroplasty show that fat embolism occurs due to high intramedullary pressure. Also, a study using sheep as models showed that fat embolism occurs after VP regardless of the augmentation of material used. Aebli et al. showed that the reasons for the cardiovascular changes in sheep are associated with the increase in intraosseous pressure during the augmentation, causing a release of bone marrow contents into the circulation. ²⁶ Benneker et al. investigated the effect of bone marrow removal using pulsed jet lavage. They were able to prove that the application of pulsed jet lavage in combination with a vacuum applied contralateral pedicle did not cause any cardiovascular reaction during and after the removal of intravertebral fat. ¹⁵ All these studies show that decreasing intramedullary pressure decreases fat embolism, thromboembolism, and cement leakage.

Most patients are able to tolerate the fat and bone marrow embolization that occurs and are thus asymptomatic. But high-risk patients, those with compromised cardiopulmonary reserve, specifically patients with chronic obstructive pulmonary disease, pulmonary arterial hypertension/cor pulmonale, and any patient with a history of deep venous thrombosis or PE may not tolerate even it is a small amount. ²⁷

For this purpose, we mechanically aspirated the vertebra body over VP cannula with a flexible aspiration tube just prior to the cement application. This decreased the PAP and d-dimer value that compared with the not aspirated VP procedure.

Conclusion

Aspiration of the vertebral body can be easily used to decrease the risk of cement leakage and the migration of fatty particles to the pulmonary circulation. It is very easy to perform, does not need extra time, and does not need fluoroscopy.