Osteoblastoma (OB) and osteoid osteoma (OO) are benign bone-forming tumors, with nearly identical basic microscopic features. The main difference is dimension (OO has usually a nidus measuring <2 cm in diameter). In addition, OB is biologically more active than OO, with a tendency to grow in size.
1. Magnetic resonance-guided focused ultrasound surgery (MRgFUS) Technique
MRgFUS is a novel non-invasive procedure, which allows for ablating bone lesions without even touching the skin of the patient, and without antibiotic prophylaxis or radiation exposure. This revolutionary system uses the thermal ablation power of ultrasound (US) combined with the radiological guidance of MR
[1].
The US beam is focused on the lesion and generates sufficient heat (57 °C for at least one second) to induce coagulative necrosis and cell death with minimal heat generation around the target. Treatment consists in a series of “sonications”, where this term indicates the act of delivering an amount of high-intensity sound waves to agitate matter and it is used to describe each energy deposition within a focal point into the region of interest. The procedure involves some preliminary low-energy sonications that are used as a preliminary test to verify the correct position of the target area and the correct focalization of the US beam within the target area before starting with the full-energy ablations. Each sonication usually lasts for a few seconds, ablating few cubic millimeters of lesion
[2].
The procedure is performed under MRI guidance through which the operator can accurately evaluate the anatomy of the region. This is particularly helpful, allowing a meticulous evaluation of both target lesion and surrounding structures throughout the treatment. In addition, the system allows a real-time control of the temperature thanks to a thermometry using specific sequences, proton resonance frequency (PRF), and by detecting all temperature-dependent MR phase variables, showing areas where the temperature is high enough to generate coagulation necrosis. The temperature increase is measured on the adjacent peri-skeletal tissue. In fact, due to the absence of moving protons in the cortical zone, PRF sequence is not able to detect temperature on bone surface, whereas it reveals a temperature increase which propagates by conductive processes from the bone to the adjacent tissues. Prior to treatment, the patients were submitted to MR imaging to determine their eligibility and plan the treatment. In fact, it is fundamental to evaluate accessibility to the lesion
[3].
Acoustic attenuation in bone is 30–60 times higher than in soft tissue; therefore, the lesions that benefit from MRgFUS are the most superficial ones
[4]. Lesions deep in the bone more than few millimeters are not suitable to be treated directly
[5].
Another important factor is the distance of the lesion from the skin surface, because due to the interposition of soft tissues, it is possible that the US beam loses its effectiveness. If the distance between the skin surface and the target lesion exceeds 10 cm, the ablation can be compromised
[4].
Osteoid osteomas (OO) are mostly located superficially (at cortical or subendosteal level) and in the appendicular skeleton (50% at femoral and tibial level); this is an important advantage as it can be easily reached and is far from a sensitive structure
[6].
The presence of neurovascular structures along the pathway of the US beam is not a limitation when using MRgFUS, since the US beams pass through these structures without damaging them. Differing from the needle-based techniques (like RFA), when using MRgFUS it is not possible to move away critical structures using hydrodissection because there is the risk of injecting microbubbles of air in the area of treatment, and so the effectiveness of the procedure could be affected. In case critical structures are strictly closed to the target lesion (touching them or at the distance of a few millimeters), an accurate control of the temperature is needed to avoid thermal injuries. This may also be applied when tendons are close to the lesion.
The proper acoustic field is a key point for the clinical success of the treatment and is defined by these findings:
-
Cortical or subperiosteal location of the lesion which must not be deeper than 2/3 mm in the cortex;
-
Absence of interfaces between skin surface and lesion;
-
Absence of any obstacles between skin and lesions
Another crucial aspect in planning MRgFUS is patient positioning, as it allows the US beam to optimally reach the target area. It is important that the skin above the lesion is in close contact with the transducer
[4].
Different anesthesiologic approaches are used, depending on the location of the lesion, whose ablation is painful. Patients with lesions located in the upper arms are administered US-guided peripheral nerve block, whereas for lesions located in the hip joint, spinal anesthesia is used. After treatment, the patient recovery time is minimal (average time of 24 h) and an additional cortisone therapy is administered during the following week
[7].
2. Literature and State of the Art
When applicable, MRgFUS is considered the first-choice treatment of benign bone tumors, due to absence of skin incision and ionizing radiation exposure
[7].
Patients with non-spinal OOs are those who benefit more from this needleless approach.
Nevertheless, there is scarce literature describing features and treatment of benign bone lesions other than osteoid osteoma.
After an initial experience in 2013 based on six non-spinal OOs, all successfully treated in terms of alleviation of pain, and without complications
[8], a larger and multicentric prospective study was carried out: 29 cases of non-spinal OOs
[5] were treated and at 12 months of follow-up, complete clinical success (defined by absence of pain and interruption of analgesics) was obtained in 90%. Partial success was recorded in 10% of patients. In cases of complete clinical success, at 12-month imaging follow-up, a marked reduction in perilesional bone marrow edema and nidus vascularization was registered. No complications related to treatment were observed, demonstrating good accuracy and high safety of MRgFUS
[5]. From these experiences it emerges how an accurate selection of patients is important as well as a comfortable patient position to achieve immobility.
Bazzocchi et al.
[4] described 7 cases of superficial OOs of the lower limb treated with MRgFUS. During 12 months of follow up, no severe events or recurrences were registered. Primary outcome was pain relief: the Visual Analogue Scale (VAS) dropped to 0 after 1 month. In six cases, VAS remained at 0 during the study period, while in 1 patient VAS dropped from 9 to 0 after 1 month, rising to 2 after 3 months. When the clinical response in MRgFUS was compared to that of a cohort treated with CT-RFA, the clinical results were almost the same.
In line with literature, which describes good clinical response in terms of lack of recurrence in the long-term follow-up, currrent experience confirms a positive outcome of MRgFUS. researchers compared two groups of 15 patients, each affected by non-spinal OO. One group was submitted to MRgFUS and the other to CT-RFA. No significant differences were found in the primary and secondary outcome measures in terms of pain relief and recovery rate of compromised motor function. Complete response was obtained in 94% (MRgFUS) and 100% (RFA) of patients. No major complications were observed in both groups. The retrospective analysis on the 3 cases of 15 treated with MRgFUS resulted non-respondent. It was found that they presented a thick cortex and a thick layer of subcutaneous soft tissues. These results underline how the therapeutic success of MRgFUS is strictly dependent on the correct selection of patients
[9].
More recently a bicentric retrospective study was conducted, analyzing 116 patients who underwent either CT-RFA or MRgFUS, over a mean of 2 years of follow-up
[10]. This bicentric study confirmed that CT-RFA and MRgFUS were equally safe and effective in the treatment of OO. It was stressed that MRgFUS requires a strict anatomical selection to guarantee good results. In particular, a lesion is suitable for MRgFUS when it is well exposed to the US beam penetration with and adequate acoustic window. In this way, patients with scars or metallic devices which can interfere with the transmission of US beam are excluded. Similarly, OOs located deeply in the bone (intramedullary or with a thick layer of reactive bone) are not suitable for MRgFUS. Furthermore, this treatment has the benefit of treating larger areas with less invasiveness in presence of lesions on the bone surfaces.
Another important aspect is the treatment of bone lesions in pediatric population, where an ionizing radiation-free technique is particularly attractive. A recent multicentric experience in children (age <18 years) reported results on 33 OOs, performed in three centers, with 24 months of follow-up. The clinical outcome showed a primary success of 97%. One patient repeated the treatment obtaining success. No major complications were recorded. During the follow-up period no clinical relapses or further complications were observed. It was demonstrated that MRgFUS for OO has a clinical response comparable to CT-RFA, with the benefit of being a less invasive, ionizing radiation free technique
[11].
To the best of currrent knowledge, when considering the application of MRgFUS for OBs, literature is very scarce. The larger of these published studies is a single-center retrospective analysis on 6 cases of intra-articular OBs, based on 2-year follow-up. The minimally invasive nature of MRgFUS was confirmed, and the absence of chondral or subchondral damage was reported. The authors also reported good results in terms of pain relief and imaging analysis, showing reduction of bone edema and synovitis with complete resolution. Conversely, they found a lack of significant correlation between VAS-determined pain change and VAS-determined functional impairment, perhaps because pain resolution does not mean complete recovery. This suggests the need of physical rehabilitation particularly in those cases where a long time elapsed between the onset of symptoms and the treatment
[12].