European Journal of Radiology 83 (2014) 696–702

Contents lists available at ScienceDirect

European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad

Postinterventional MRI findings following MRI-guided laser ablation of osteoid osteoma S. Fuchs a,∗ , B. Gebauer a , L. Stelter a , M.L. Schäfer a , D.M. Renz a , I. Melcher b , K. Schaser b , B. Hamm a , F. Streitparth a a b

Department of Radiology, Charité, Humboldt University, Berlin, Germany Center for Musculoskeletal Surgery, Charité, Humboldt University, Berlin, Germany

a r t i c l e

i n f o

Article history: Received 12 September 2013 Received in revised form 9 December 2013 Accepted 12 December 2013 Keywords: Muskuloskeletal Osteoid osteoma Thermal ablation MRI guidance MRI characteristics

a b s t r a c t Objective: To evaluate postinterventional magnetic resonance imaging (MRI) characteristics following MRI-guided laser ablation of osteoid osteoma (OO). Materials and methods: 35 patients treated with MRI-guided laser ablation underwent follow-up MRI immediately after the procedure, after 3, 6, 12, 24, 36, and up to 48 months. The imaging protocol included multiplanar fat-saturated T2w TSE, unenhanced and contrast-enhanced T1w SE, and subtraction images. MR images were reviewed regarding the appearance and size of treated areas, and presence of periablation bone and soft tissue changes. Imaging was correlated with clinical status. Results: Mean follow-up time was 13.6 months. 28/35 patients (80%) showed a postinterventional “targetsign” appearance consisting of a fibrovascular rim zone and a necrotic core area. After an initial increase in total lesion diameter after 3 months, a subsequent progressive inward remodeling process of the zonal compartments was observed for up to 24 months. Periablation bone and soft tissue changes showed a constant decrease over time. MR findings correlated well with the clinical status. Clinical success was achieved in 32/35 (91%). Conclusions: Evaluation of long-term follow-up MRI after laser ablation of OO identified typical postinterventional changes and thus may contribute to the interpretation of therapeutic success and residual or recurrent OO in suspected cases. © 2013 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Osteoid osteomas are circumscribed bone tumors and account for up to 12% of benign primary bone lesions. In more than half of the cases, they are centered in the cortex of the diaphysis of lower extremity bones, although virtually any bone can be affected. There is a male predominance, and more than 75% of osteoid osteomas occur in people younger than 25 [1]. Characteristic morphological feature is a nidus, a small area of osteolysis surrounded by a halo of reactive osteosclerosis. The dominant clinical symptom is nocturnal pain that responds well to nonsteroidal anti-inflammatory drugs (NSAID). Pain is thought to be due to nerve fibers marginating the nidus as well as local inflammation with increased expression of prostaglandins by the tumor, whereas tumor prostaglandin levels correlate with the extent of peritumoral edema [2,3].

∗ Corresponding author at: Department of Radiology, Charité, Humboldt University Medical School, Augustenburger Platz 1, 13353 Berlin, Germany. Tel.: +49 30 450657056; fax: +49 30 4505 57900. E-mail address: [email protected] (S. Fuchs). 0720-048X/$ – see front matter © 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ejrad.2013.12.018

The diagnosis is usually made based on plain radiographs, computed tomography (CT), and/or magnetic resonance imaging (MRI) in conjunction with the typical clinical findings. In ambiguous cases, bone scintigraphy may be helpful to confirm the diagnosis. Total surgical removal of the lesion was considered to be the definitive treatment of choice for many years. In the last 2 decades though, image-guided thermal ablation procedures with radiofrequency ablation (RFA) and laser ablation (LA) have turned out to be reasonable treatment alternatives with high success rates and few complications, which is why most authors consider thermal ablation as method of choice today [4–7]. Among the available imaging modalities, CT is the most common technique for guidance of minimally invasive treatment [8]. Recently, open MRI has been shown to allow efficient, safe, and cost-effective guidance of osteoid osteoma treatment [5,9–11]. Immediate postinterventional contrast-enhanced MRI including subtraction images may be of additional benefit for confirming the success of the intervention and preventing recurrence [5]. With its excellent soft tissue contrast, MR imaging appears to be superior to CT for monitoring the effects of thermal ablation because it provides precise information on the ablated area and its surroundings. In addition to this, MRI lacks radiation exposure.

S. Fuchs et al. / European Journal of Radiology 83 (2014) 696–702

Although clinical status is considered the gold standard for assessing outcome, the understanding of normal and abnormal MRI findings in the follow up period is of the utmost importance particularly when there is concern for recurrence or postprocedural complication [12–14]. Therefore, the purpose of this prospective study was to evaluate postinterventional MR imaging characteristics after MRI-guided laser ablation of osteoid osteoma lesions. 2. Materials and methods 2.1. Patients During a 4-year period, 35 patients with a mean age of 26.7 ± 15.5 years (range, 5–64 years) with a clinical and imagingbased diagnosis of osteoid osteoma were treated with MRI-guided percutaneous laser ablation. Inclusion criteria for treatment were pathognomonic clinical history with located, sharp pain that could not be clearly related to preceding trauma or physical exposure, with nocturnal predominance, typically responding to salicylates or other NSAID. Radiologic features included typical findings such as a radiolucent nidus with a mean nidal diameter smaller than 15 mm, surrounded by a sclerotic rim representing reactive osteosclerosis to tumor growth. Cortical lesions of the diaphysis of lower extremity bones were considered to be typical, all other locations including spinal lesions as well as joint-associated and primarily spongious lesions were considered to be atypical osteoid osteomas. Imaging modalities used in the patients included plain film radiographs (n = 35, 100%), CT (n = 18, 51%), MRI (n = 35, 100%), and bone scintigraphy (n = 3, 9%). One patient had clinical and radiologic recurrence of osteoid osteoma after preceding drilling procedure ex domo; all other patients had a new diagnosis of osteoid osteoma before entering this study. One patient showed multicentric (bifocal) osteoid osteoma. 2.2. Interventional procedure All patients included in our study were treated with MRIguided percutaneous laser ablation using an open 1.0 T MRI system (Panorama HFO, Philips, Best, Netherlands). Patients were operated on while under general as well as under local subcutaneous and periosteal anesthesia (Xylonest 1%, AstraZeneca, Wedel, Germany) to reduce postinterventional pain. Additionally, the patients received preinterventional prophylactic intravenous antibiosis (1.5 g cefuroxim) to avoid periinterventional infection. Interactive lesion localization, instrument guidance, drilling, and positioning of the laser fiber within the lesion were guided using near real-time imaging with a fast T1-weighted turbo spin echo (TSE) sequence (TE/TR 5.7/200 ms, TF 7, FA 90◦ , scan duration 3 s). Multiplanar imaging provided (para-) axial, (para-) sagittal or (para-) coronal planes for full anatomical orientation. Depending on the extent of perifocal ossification and localization of the target lesion within the bone, an MRI-compatible bone biopsy drill (Invivo, Schwerin, Germany) was used in some of the cases, and bone specimens were optionally obtained (18/35 patients). After that, a 16–18 G needle (Somatex, Teltow, Germany) was introduced. The needle position was usually verified with a proton densityweighted (PDw) TSE sequence (TE/TR 30/383 ms, TF 11, FA 90◦ , scan duration 41 s) in two planes. A 600 ␮m bare laser fiber (Frank Optic Products® , Berlin, Germany) was then introduced through the needle. Subsequent laser treatment was performed using a Nd:YAG laser (1064 nm, Fibertom Medilas, Dornier MedTech, Wessling, Germany) with continuous energy flow and an effective output ranging from 2 to 3 W. Total energy deposition was 360–4300 J, depending on size and localization of the target lesion. A T1w

697

gradient echo (GRE) sequence (TE/TR 2/4.3 ms; FA 27◦ ), enabling image update every 4 s, was used for online monitoring of the T1 temperature tissue effects. Target temperature at the center of the lesion was at least 60–90 ◦ C. For efficacy control, immediate postinterventional subtraction images of unenhanced and contrast-enhanced (Gadovist; BayerSchering, Berlin, Germany) T1w SE sequences (TE/TR 12/503 ms, FA 90◦ , TA 92 s) were obtained to ensure nidal signal loss. Contrast agent was administered using a weight-based injection protocol (0.1 ml/kg body weight). After the procedure, all patients were observed carefully to identify possible postinterventional neurovascular damage, bleeding, or burns. All patients were discharged from hospital within 48 h after the procedure. 2.3. Follow-up analysis In this prospective study, 35 patients underwent follow-up with clinical evaluation and MR imaging directly postablational as well as at 3, 6, 12, 24, 36 and up to 48 months after the procedure. At each visit, patients were asked to grade residual pain using the numeric rating scale (NRS 1–10). Alternatively, if patients were unable to visit our department for logistic reasons, external MRI was accepted after instructing the respective radiologist about the sequences required. Patients were then contacted by phone to ask about residual pain and clinical status. The standard follow-up imaging protocol included fat-saturated T2-weighted TSE sequences (SPIR, TE/TR 60/1600 ms, FA 90◦ ) and T1-weighted SE sequences (TE/TR 12/503 ms, FA 90◦ , TA 92 s) before and after weight-based intravenous contrast administration (Gadovist, Bayer Schering, Germany). Subtraction images were computed to rule out residual or recurring nidal enhancement and to relate the diameter of the ablation area to the preinterventional extent of the target lesion. Additional parameters for image analysis were signal changes of the ablation zone and surrounding bone marrow as well as of adjacent soft tissues. Complications such as joint effusion, signs of infection, hematoma were also recorded. For follow-up evaluation, the most adequate slice orientation for evaluation of the appropriate signal changes was assessed and used for comparison within the follow-up series. In order to evaluate the dynamics of the ablated area, lesion size was measured and periablation changes such as bone edema, soft tissue enhancement, or joint effusion in joint-related lesions were graded qualitatively (0 = none, 1 = mild, 2 = moderate and 3 = severe). Two radiologists reviewed the MR images. Discrepant findings were re-evaluated and rated in consensus. Statistical analyses were performed using Microsoft Excel 2007. Pearson correlation tests were performed to assess the correlation between total energy deposit and postablational lesion size. Statistical significance was tested by using a two-tailed Student t test. 3. Results Mean nidal diameter was 6.8 ± 2.9 mm. Most osteoid osteomas were found in long lower extremity bones, mainly in femur (n = 9) and tibia (n = 9). Eighteen patients showed typical locations and 17 showed atypical locations (Table 1). Four patients showed primarily spongious osteoid osteoma, whereas 31 patients had cortical lesions. Eight lesions were joint-related. From 18 patients, bone specimens were obtained for histopathological analysis. Diagnosis was confirmed in 10 patients. In 7 patients, biopsy results were inconclusive. In the MRI analysis, reviewers reached a consensus on dividing the ablation area into three different zones (Z0–Z2) according to

698

S. Fuchs et al. / European Journal of Radiology 83 (2014) 696–702

Table 1 Data of all patients with osteoid osteoma treated by MRI-guided laser ablation. M = male, f = female, W = Watts, J = Joule, EEO = effective energy output, M = months, NRS = numeric rating scale. Mean follow-up was 13.6 ± 11.6 months. Patient no.

Sex/age (years)

Location

Typical vs. atypical

EEO (W)

Total energy deposition (J)

Ablation time (min)

Target sign

Follow up (M)

NRS at last visit

Histol. specimen

Patient data 1 2 3 4 5 6 7 8 9 10 11a

m/20 f/64 m/25 m/46 f/61 f/22 m/26 m/22 f/21 m/18 m/19 m/20 m/20 m/29 m/26 m/46 m/19 m/30 m/7 f/19 m/21 m/26 m/5 m/11 f/15 f/34 f/12 m/25 f/20 m/59 m/12 f/11

33 34 35

m/62 m/25 m/23

Typ. Atyp. Atyp. Atyp. Typ. Typ. Atyp. Typ. Typ. Atyp. Typ. Typ. Atyp. Typ. Atyp. Atyp. Atyp. Atyp. Atyp. Typ. Atyp. Typ. Atyp. Typ. Typ. Typ. Typ. Atyp. Typ. Atyp. Atyp. Typ. Typ. Typ. Atyp. Atyp. Typ.

2.3 2.8 2.8 2.8 2.8 2.4 2.5 2.5 2.2 2.3 2.0 2.0 2.0 2.3 2.0 1.5 2.0 2.0 2.5 2.3 2.3 2.3 2.6 2.4 2.2 2.3 2.8 2.2 2.4 2.9 2.9 2.3 2.5 2.5 2.5 2.4 2.5

1498 2352 2016 4320 1404 2268 1350 1350 1320 1258 1440 1220 1260 1380 360 380 1200 360 800 1380 1380 1380 1092 1440 1320 1680 1680 1320 1300 1392 1218 1380 1200 750 1125 1000 1300

11 14 12 24 9 12.5 9 9 10 9 12 10 10 10 3 4 10 3 5 10 10 10 7 10 10 12 10 10 9 8 7 10 8 5 7.5 7 8

Pos. Pos. Pos. Pos. Pos. Pos. Pos. Pos. Pos. Pos. Neg. Neg. Pos. Pos. Neg. Pos. Pos. Neg. Pos. Neg. Neg. Pos. Pos. Pos. Pos. Neg. Pos. Pos. Neg. Pos. Pos. Pos. Pos. Pos. Pos. Pos. Pos.

48 36 36 12 24 24 24 24 3 24

12 13 14 15 16 17a 18 19 20a 21 22 23 24 25 26 27 28 29 30 31 32

Fibular diaphysis Talus MT4 basis Tibial plateau Tibial diaphysis Femoral diaphysis Femoral neck Femoral diaphysis Fibular diaphysis Calcaneus Tibial diaphysis Tibial diaphysis Tibial head Tibial diaphysis Distal phalanx Vertebra Os ilium Distal phalanx Calcaneus Tibial diaphysis Femoral neck Femoral diaphysis Os cuneiforme Femoral diaphysis Femoral diaphysis Tibial diaphysis Femoral diaphysis Acetabulum Humeral diaphysis Femoral neck Talus Tibial diaphysis Femoral diaphysis Femoral diaphysis Patella Tibial head Tibial metaphysis

0 0 0 0 0 0 0 0 1 0 3 1 0 2 0 0 0 6 0 0 3 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0

Yes Yes No No Yes Yes Yes No Yes Yes No Yes No No No No Yes No Yes No Yes No No Yes Yes Yes Yes Yes No No No No No No Yes No Yes

a

24 3 12 12 24 3 6 12 6 6 6 12 12 6 3 6 6 12 6 6 6 3 3

Residuum/recurrence.

a specific, recurrent pattern of signal changes. This “target sign” appearance was detected in 28/35 patients (80%) (Fig. 1): Z0 refers to a zone in the center of the ablation area, which is usually hypointense on T1w images and hyperintense on T2w SPIR sequences (Fig. 1). The Z0 area occurred in 10/35 patients (29%), and disappeared within the first 12 months of the intervention in all cases. Z1 refers to the actual ablation zone. On initial postablation images, Z1 appeared as an ovoid area of hypointensity relative to the surrounding bone marrow in T1w TSE sequences. In follow-up images obtained after >3 months, Z1 has a more hyperintense or isointense appearance relative to bone on T1w and a hypointense appearance on T2w TSE SPIR images. Z2 is seen as a distinct band surrounding the Z0/Z1 complex, which is best visible after >3 months with a hypointense signal in T1w and a markedly hyperintense signal in T2w TSE SPIR as well as in contrast-enhanced (CE) T1w TSE sequences (Figs. 1 and 2). The ablation zone (Z1) expanded during the initial follow-up period from a mean diameter of 7.3 ± 3.4 mm directly postinterventional up to 9.5 ± 3.8 mm at 3 months, whereas a constant shrinking of the different zonal compartments over time was detected subsequently, with a mean Z1 diameter of 5.7 ± 2.8 mm after 12 months and 2.6 ± 2.9 mm after 24 months. A total loss of Z1 has been detected in 5/35 (14%) of the patients so far. Mean diameter of the entire thermal lesion (Z0–Z2) increased from 10.1 ± 5.3 mm initially to 16.5 ± 5.2 mm at 3 months. Further follow-up showed a steady decrease to 10.7 ± 4.1 mm after 12 months and 7.0 ± 3.3 mm

after 24 months (Fig. 3). Mean total lesion size tended to be larger in the “atypical” subset, though not significantly different: mean total lesion size at 12 months was 11.8 ± 5.3 mm for atypical lesions and 9.2 ± 3.2 mm for typical lesions (p = 0.12). A significant positive correlation of total energy deposit and maximum total lesion diameter on follow up was detected (r = 0.37, p < 0.0005). Eighty-nine percent of the patients showed at least mild preinterventional bone edema, 60% of the patients showed at least mild preinterventional soft tissue enhancement. After ablation, a strong decrease in periablation bone and soft tissue changes was observed: mean intensity of perifocal bone edema changed from 2.2 ± 1.0 postinterventionally to 0.6 ± 0.7 after 12 months; none of the patients showed considerable soft tissue enhancement after 24 months. Mean intensity of periablation changes in follow-up MRI is shown in Fig. 4. Joint effusion was found in 7 patients preinterventionally, of whom 4 showed complete resolution within the first 12 months, 3 patients had residual joint effusion at the time of editing. Thirty-four/thirty-six nidi in 35 patients (one multicentric osteoid osteoma, patient No. 32) were located within Z1 on immediate postablational images and showed no enhancement in subtraction imaging in the last follow-up. Thirty-three of thirty-five patients showed primary clinical success, defined as an NRS score of

Postinterventional MRI findings following MRI-guided laser ablation of osteoid osteoma.

To evaluate postinterventional magnetic resonance imaging (MRI) characteristics following MRI-guided laser ablation of osteoid osteoma (OO)...
1MB Sizes 0 Downloads 0 Views