Introduction

Osteoid Osteomas (OO) and Osteoblastomas (OB) are relatively rare benign bone lesions, accounting for approximately 12% and 3%, respectively, of all benign skeletal tumors [1,2,3]. While these lesions are infrequent, they are often encountered in the first three decades of life. Despite their benign nature, they necessitate treatment due to the significant associated pain symptoms, which can lead to considerable functional limitations and adversely affect the quality of life in young patients.

Historically, surgical resection or curettage were the primary therapeutic options [4, 5]. However, while effective, these techniques often required surgery with resection margins larger than the actual size of the lesion, prolonged hospital stays, rehabilitation, and increased risk of pathological fractures. Consequently, minimally invasive techniques have gained prominence in recent years as the preferred treatments for these lesions [6,7,8]. Among these, Radiofrequency Thermal Ablation (RFA) and Magnetic Resonance-guided Focused Ultrasound (MRgFUS) have emerged as the most utilized [9, 10], with RFA now considered the gold standard [11,12,13,14]. Both techniques have demonstrated high efficacy, reduced complication rates, and minimal invasiveness [15, 16].

Using both RFA and MRgFUS in our clinical practice, we undertook a retrospective evaluation to assess the selection criteria guiding the choice between these treatments and determine whether statistically significant differences exist between them.

We examined a series of variables, including the intra/extra-articular localization of the lesion, maximum nidus size, presence of intranidal calcifications, type of sclerosis, proximity to critical vascular-nervous structures, maximum extension of intraosseous edema, and presence of an adequate acoustic window. It is important to note that the evaluations of the proposed selection criteria are influenced by the experience and preferences of the operator performing the procedure. With the exception of the presence of an acoustic window, the other parameters assessed may not absolutely preclude the use of radiofrequency thermal ablation (RFA) over magnetic resonance-guided focused ultrasound (MRgFUS), but rather reflect individual clinical decisions based on subjective interpretation of the imaging data.

Methods

The objective of our study was to evaluate the selection criteria for determining the optimal treatment modality, either RFA or MRgFUS, for Osteoid Osteomas and/or Osteoblastomas, considering the cost/benefit ratio and the statistical significance of these criteria.

We retrospectively analyzed 102 cases of OO/OB treated in our Radiology Department between January 2013 and July 2020 using RFA under CT guidance or MRgFUS. Of the 102 patients evaluated, 82 had Osteoid Osteomas and 20 had Osteoblastomas.

We performed 51 MRgFUS treatments and 51 RFA treatments. However, ten patients opted for RFA despite the recommended alternative, and thus, they were excluded from the statistical analysis (Table 1).

Table 1 Sample characteristics
Full size table

Eight potential factors influencing treatment selection for these lesions were evaluated (Table 2):

  1. 1.

    Maximum nidus diameter (mm)

  2. 2.

    Nidal depth from bone surface (mm)

  3. 3.

    Presence of calcifications within the nidus (yes/no)

  4. 4.

    Perilesional sclerosis (absence, non-compact, compact)

  5. 5.

    Intra- or extra-articular lesion localization (yes/no)

  6. 6.

    Maximum extension of bone edema surrounding the lesion (mm)

  7. 7.

    Proximity of critical vascular-nervous structures to the lesion (≤ 1 cm cutoff: yes/no)

  8. 8.

    Acoustic Window, defined as the absence of structures obstructing the progression of the Focused Ultrasound beam (yes/no).

Table 2 Analyzed criteria
Full size table

All variables were assessed via CT and/or MRI prior to treatment; For RF ablation, a 320-Slices Spiral CT Scanner (TAC Toshiba Aquilion One) was utilized, while a 3 Tesla MRI system (GE) was used for HIFU treatment. The parameters were evaluated by experienced operators trained in the respective techniques.

The inclusion criteria for patient selection included a confirmed diagnosis of OO/OB and treatment with either RFA or MRgFUS during the specified period.

Statistical analysis

Statistical analysis involved calculating frequency and dispersion indices and performing statistical regression using the Wilcoxon test. The patients treated were divided into three groups for statistical comparison: Group A (patients treated with MRgFUS), Group B (patients treated with RFA), and Group C, which evaluated the statistical significance of parameters arbitrarily defined by the operator to justify RFA treatment even in the presence of an acoustic window: compact sclerosis and non-compact sclerosis with depth > 7 mm.

Results

Of the 102 patients evaluated, 82 were diagnosed with Osteoid Osteomas and 20 with Osteoblastomas. Diagnosis was obtained through a combination of clinical-radiological assessment and biopsy in cases where necessary for confirmation.

The study included both Osteoid Osteomas and Osteoblastomas to encompass a broader spectrum of benign bone lesions. While the type of tumor could indeed influence the choice of treatment modality, our aim was to evaluate the selection criteria for both types of lesions and determine whether statistically significant differences exist between MRgFUS and RFA treatments.

Our study did not delve into the analysis of complications associated with RFA and MRgFUS treatments; however, the complication rate was poor, typically ranging from 1 to 6% [17, 18].

We performed 51 MRgFUS treatments and 51 RFA treatments, with 10 patients opting for RFA despite the recommended alternative and thus being excluded from statistical analysis.

We subsequently analyzed the data of 92 patients, 51 treated with MRgFUS and 41 with RFA. The most common lesion locations were the femur (41), tibia (16), humerus (13), vertebral bodies (8), and talus (7). The femur was the most frequently affected skeletal segment, with 25 and 16 patients treated with RFA and MRgFUS, respectively. Lesions were found to be cortical in 50 patients, subperiosteal in 28, and intramedullary in 14.

In Group A (MRgFUS), 24 patients had intra-articular lesions, 27 had extra-articular lesions, and 34 exhibited calcifications within the nidus. Additionally, 45 lesions were located more than 1 cm from critical vascular-nervous structures, and all patients had an adequate acoustic window. In Group B (RFA), 24 lesions were intra-articular, 30 had calcifications, and 24 were more than 1 cm from critical structures, but only 18 had an adequate acoustic window. Group C (RFA with an adequate acoustic window) comprised 18 patients, with similar distributions of intra/extra-articular lesions and calcifications, and all patients had an adequate acoustic window.

Statistical comparisons between Groups A and B revealed no significant differences in intra/extra-articular localization, presence of intranidal calcifications, sclerosis, or bone edema extension (p > 0.05). However, significant differences were observed in lesion proximity to vascular-nervous structures, lesion depth, nidus size, and presence of an acoustic window (p < 0.001).

Significant differences were also noted between Groups A and C in lesion depth (p < 0.001), sclerosis (p = 0.03), and nidus size (p = 0.02). Conversely, no significant differences were observed in intra/extra-articular localization, intranidal calcifications, proximity to vascular-nervous structures, or bone edema extension (p > 0.05).

Discussion and conclusions

Our study provides insights into the selection criteria guiding the choice between RFA and MRgFUS for the treatment of OO and OB [19]. While both techniques have demonstrated high efficacy and minimal complication rates [20,21,22], our analysis aimed to identify key factors influencing treatment selection and any statistically significant differences between them.

RFA has established itself as highly effective, with primary success rates ranging from 76 to 100% and secondary success rates, following repeated ablations, between 87 and 100% [23, 24] (Fig. 1).

Fig. 1
figure 1

OO treated by RFA. a OO localized to the femoral head: the absence of an adequate acoustic window and the intramedullary localization do not allow treatment using MRgFUS. b The electrode tip is located perfectly in the center of the nidus: in this way, it is possible to obtain an adequate necrotic area to cover the entire lesion

Full size image

Similarly, MRgFUS has shown efficacy rates between 89 and 100%, with absence of pain after 1 year of follow-up and minimal complications [25, 26]. Notably, MrRgFUS offers advantages such as the absence of needles and exposure to ionizing radiation. However, it does have limitations, including the requirement for an adequate acoustic window, the inability to perform biopsies, and the inability to treat deep lesions (Fig. 2).

Fig. 2
figure 2

Tibial OO treated with MRgFUS. The presence of an acoustic window and the superficial localization of the lesion allow the treatment to be carried out easily. The MRI (PRF) sequences acquired during the treatment provide a real-time thermometric map of the region of interest, allowing the temperature reached at the focus and at the level of the surrounding structures to be assessed

Full size image

Our retrospective evaluation focused on several variables, including intra/extra-articular lesion localization, nidus size, presence of intranidal calcifications, type of sclerosis, proximity to critical vascular-nervous structures, maximum extension of intraosseous edema, and presence of an adequate acoustic window.

Our analysis identified the presence of an adequate acoustic window as the primary selection criterion, crucial for enabling treatment with Focused Ultrasound. We found that the absence of an acoustic window significantly impacted treatment feasibility, highlighting the importance of considering alternative options in such cases.

Additionally, sclerosis emerged as a significant factor, with the presence of compact sclerosis favoring RFA due to the potential difficulty of achieving adequate concentrations of Focused Ultrasound energy at the nidus (Fig. 3).

Fig. 3
figure 3

Wilcoxon Score distribution in relation to the sclerosis variable between groups A and B and groups A and C, with relative p values

Full size image

Nidus size also played a crucial role, with larger lesions often favoring RFA due to the need for broader Focused Ultrasound action, which carries the risk of necrosis in healthy bone tissue (Fig. 4).

Fig. 4
figure 4

Wilcoxon Score distribution in relation to the nidus size variable between groups A and B and groups A and C, with relative p values

Full size image

While proximity to critical vascular-nervous structures was considered, it was found to be less influential compared to other parameters and should be evaluated in conjunction with them.

However, our study has several limitations that should be considered when interpreting the results. First, it is important to note that measurements of the variables analyzed, such as nidus size, lesion depth, and presence of sclerosis, may be influenced by the operator performing them. Despite careful efforts to standardize measurement procedures, there is always the risk of intra-observer and inter-observer variability, which could impact the consistency and reliability of the collected data.

Additionally, our study did not delve into the analysis of complications associated with RFA and MRgFUS treatments. Although both techniques have minimal complication rates, it is important to recognize that adverse events, such as skin burns, local infections, or post-treatment bone fractures, can occur. The absence of a detailed analysis of complications may limit a comprehensive understanding of the risks associated with treatments and safety considerations.

In conclusion, our findings underscore the importance of considering multiple factors when selecting a treatment modality for OO and OB. While both RFA and MRgFUS offer comparable efficacy and complication rates, careful assessment of patient-specific features and lesion characteristics, particularly the presence or absence of an acoustic window and sclerosis, should guide treatment decisions. These findings provide valuable insights for clinicians in optimizing treatment outcomes for patients with OO and OB.