Vertebroplasty is largely being used for metastatic lesions from several neoplasms and from multiple myeloma (MM); however, literature data regarding patients with painful sacral metastases are limited. Furthermore, there is no consensus on the indications or degree of sacral involvement for percutaneous treatment. Sacroplasty was initially used to stabilize sacral-insufficiency fractures [1–3].

Even if the exact mechanism by which sacroplasty causes pain relief is unknown, two theories have been proposed: (1) stabilization obtained with cement injection decreases movement at the fracture site and thus pain by augmenting the strength and stiffness of the sacrum; and (2) the analgesic effect of osteoplasty in vertebral metastases results from reinforcing the trabecular bone, which leads to stabilization of the vertebral body [4], thus obviating deformation of the periostium [5]. This mechanism has been suggested for other skeletal bone, including the sacrum.

Materials and Methods

We retrospectively reviewed the records of eight patients (four women, four men; age range 47–68 years; mean age 57.8) who underwent osteoplasty for painful osteolytic localization of MM between April 2007 and May 2009. Patient characteristics are listed in Table 1. Five patients had monolateral lesions (Fig. 1), in particular three of these had small painful lesions (Figs. 2, 3, 4); the other patients had massive involvement of the sacrum (Figs. 5, 6). After evaluating the various treatment options and possible contraindications, the indication for treatment was established by a team of hematologists, radiotherapists, surgeons, and radiologists on the basis of a detailed correlation between clinical history, particular current symptoms (PS), and CT or MRI findings. All patients had difficulty walking because of increasing pain, and patient no. 8 (Fig. 3), also affected by an iliac bone lesion, could not even get out of bed. Six patients had persistent pain despite having had other osteoplasties for vertebral and femur involvement, whereas two patients referred at the time of diagnosis had a sacral lesion only (Table 1).

Table 1 Population
Full size table
Fig. 1
figure 1

Patient no. 4. Coronal reformatted CT images showing a focal area of osteolysis with involvement of ipsilateral foramina that was treated by (a) radiofrequency ablation needle (white arrow) coaxially inserted through a (b) vertebroplasty needle (black arrow). c and d Postprocedural coronal and sagittal reformatted CT images

Full size image
Fig. 2
figure 2

Patient no. 3. Coronal reformatted CT image showing (a) a focal area of osteolysis referred to localization of MM (arrow) that was treated by (b) insertion of vertebroplasty needle under fluoroscopic guidance. c and d Postprocedural fluoroscopic and CT images

Full size image
Fig. 3
figure 3

Patient no. 8. a and b Sagittal reformatted CT images showing a focal area of osteolysis in the left sacrum (arrow) with another osteolitic lesion in controlateral iliac bone (asterisk) that was treated by (c and d) insertion of two vertebroplasty needles (arrows). Postprocedural fluoroscopic (e) and CT images (f–h)

Full size image
Fig. 4
figure 4

Patient no. 7. T1 and T2 MRI scans showing (a and b) a right sacral lesion (arrows). c Sagittal plane fluoroscopic image showing needle inserted in the right sacrum with posttreatment fluoroscopic (d) and coronal and sagittal reformatted CT images (e and f)

Full size image
Fig. 5
figure 5

Patient no. 2. a CT image showing massive sacral osteolysis. Postprocedural fluoroscopic (b), coronal (c), and sagittal (d) CT images

Full size image
Fig. 6
figure 6

Patient no. 1. Coronal reformatted CT images showing (a) massive sacral osteolysis treated by bilateral insertion of 11-gauge needles showed in coronal (b) and sagittal (c) planes. d Postprocedural CT and fluoroscopic images showing the results with small cement leak (arrow)

Full size image

The clinical indication for treatment was a pain intensity score ≥5 or higher on VAS and pain totally or partially refractory to analgesic treatment in patients with a life expectancy >3 months. Technical planning was based on computed tomography (CT) and/or magnetic resonance imaging (MRI). Seven patients had previously undergone radiotherapy or chemotherapy and were receiving varying doses of analgesics (Table 2), whereas for patient nos. 4 and 5, osteoplasty, performed after radiofrequency ablation (RFA) in the former case, represented the first treatment. In that case we used an electrode with 1-cm active tip to decrease the risk of neurologic damage due to the cortical invasion operated by the neoplasm. Written informed consent to treatment in this study was obtained from all patients and their families. All procedures were performed with the patient under local anaesthesia or moderate sedation. Osteoplasty was performed with 11- or 13-gauge bevel-tip needles 15 cm in length plus cement (polymethylmethacrylate; Cardinal Health, France). All procedures were carried out under fluoroscopic guidance in the angiographic suite. CT (Brilliance 16; Philips, The Netherlands) was used in all cases for immediate postprocedural evaluation. A subjective assessment of perceived pain intensity was obtained through analysis of VAS scores (on which 0 indicates total absence of pain and 10 indicates the worst pain ever experienced), which were recorded before and immediately after treatment and during the follow-up period (range 3–27 months; mean 12.5) by ambulatory visits or telephone interviews (Table 3). Imaging follow-up was carried out according to the standard follow-up schedule for the tumour and was anticipated in the event of worsening of pain. Technical success was defined as the ability to access the lesion using the approach established at previous CT and MR imaging. Clinical success was defined as a quantitative decrease in analgesic medication and improved VAS score and PS compared with baseline. Complications were classified, based on the standards of the Society of Interventional Radiology, into major and minor according to length of hospitalization after the procedure, where the lowest grade of major complication involved therapy and hospitalization for <48 h.

Table 2 Therapy
Full size table
Table 3 Results
Full size table

Results

Technical success was achieved in all cases. A total amount of cement, ranging between 3 and 4 cc for focal monolateral procedures and 8–10 cc for bilateral interventions, was injected. We had only a small and asymptomatic foraminal leak in patient no. 1 (Fig. 6). All patients experienced improvement in symptoms after the procedure as demonstrated by improved VAS scores and PS and decreased analgesic doses (Tables 2, 3). All patients were followed-up for a period ranging from 3 to 27 months (mean 12.5). VAS scores in all patients were significantly lower, between 3 and 8 points (mean 5) within 24 h of the procedure, and they remained substantially unchanged throughout the follow-up.

Discussion

Osteoplasty currently represents one of the standard treatments of vertebral neoplasms, and in recent years its use has been proposed for extraspinal lesions on the basis of the same pathologic mechanism (i.e, activation of pain nerves of the periosteum). Even if vertebroplasty is widely used for localization of MM, reported literature on the use of percutaneous stabilization in extraspinal lesions in such patients is lacking. If we consider three more recent literature series on percutaneous treatment of extraspinal bone neoplasms published in 2008 by Anselmetti et al. [6] (50 patients), by Hoffman et al. [7] (22 patients), and by Basile et al. [8] (13 patients), the investigators reported, respectively, 1, 4, and 5 cases of MM. The localization of MM to the sacrum is rare, as is local lesion or massive involvement. To our knowledge, only 17 cases of sacroplasty for bone metastases have been published up until now in the English literature [6–14] (Table 4). Of these, only two cases of sacroplasty for MM have been published. The first case was published by Butler et al. [12] in a patient with sacral fracture secondary to bilateral MM involvement. The investigators performed a double procedure on the right sacrum because cement distribution during the initial therapy was insufficient; however, they did not specify whether or not the patient had clinical improvement after the first procedure. The second case was published by Wee et al. [14] in a patient with extensive infiltration of the S1 segment, who had already been treated with surgical intervention for a large destructive mass in the S1 and S2 segments, extending to involve the L5 vertebral body and obliterating the sacral thecal sac.

Table 4 Literature data
Full size table

In the largest reported series of extraspinal bone lesions treated percutaneously by cementoplasty alone or after RF ablation, published by Anselmetti et al. [6], the investigators performed eight sacroplasties for seven monolateral and one bilateral metastases without using previous RF ablation, with good results in terms of pain relief and decreased analgesic therapy.

In our patients, sacral lesions were discovered at the exordius of the disease or as a late involvement, and they were treated despite any sign of fracture. All patients experienced almost immediate and dramatic pain relief after osteoplasty for the duration of follow-up, and they were able to leave the hospital few days after the procedure and returned to home care.

In particular, patient no. 4 was treated at the exordius first with percutaneous techniques instead of standard treatment (chemotherapy and radiotherapy), and this represents a new indication. We performed radiofrequency ablation before osteoplasty because there was a large neoplasm involving the bone; we use a 1-cm active-tip electrode to decrease the risk of neurologic damage. In our institution, according to the protocol implemented in collaboration with haematologists, we prefer to perform immediate percutaneous treatment in highly symptomatic patients with MM or at risk of fracture who need immediate therapy and who cannot wait for radiotherapy for those reasons. Frequently we treat patients with proximal long-axis bone lesions, such as the femoral neck, in this way. These patients would otherwise undergo surgical treatment or radiotherapy and have to wait weeks before starting chemotherapy. The important concept of this approach is that such patients undergo a “bridge” percutaneous stabilization, which allows immediate pain relief, decreased risk of fracture, and immediate start up of chemotherapy.

In conclusion, based on our knowledge of English literature, this is the largest series published this far of patients with MM of the sacrum treated by sacroplasty. On the basis of our experience, percutaneous stabilization can be used effectively and safely in patients with focal or extensive involvement of sacrum; however, more patients and longer follow-up are necessary to confirm our experience.