RPS 1410b - Tumours and bone density

Purpose:

To identify MRI findings, pathological findings, and tumour-related factors that predict the response to treatment and its association with survival of soft tissue sarcomas.

Methods and materials:

41 patients with 23 complete cases were retrospectively analysed. All patients had a multiparametric MRI 1.5T, including conventional (T1-weighted, T2-weighted, and contrast-enhanced T1-weighted imaging) and functional (DCE-MR imaging and DW imaging with ADC mapping) sequences at the diagnoses, from 2011-2018. The analysis was done from the diagnosis MRI until the last MRI, chest-CT, or death. ADC values were reported using global regions of interests (gROIs) including all the tumuor in one slide and specific ROIs (sROIs) in the areas of lower ADC values. Other data was also reported (T1-T2 heterogeneity, presence of blood or fat, and contrast enhancement type). Pathological data after the percutaneous biopsy and the resected specimen was reported.

Results:

Higher values of gROIs were related with less mortality (p=0.03) and less fibrosis post-treatment (p<0.05). Higher sROIs values were related with more fibrosis post-treatment (p=0.02) and less mortality (p=0.01). Higher T1-heterogeneity was related to less fibrosis post-treatment (p<0.05). We also found that higher values of fibrosis post-treatment were related to less mortality.

Conclusion:

Higher values of diffusion are related to less mortality and the absence of areas with low ADC values are related to a good response to therapy. Higher values of T1-heterogeneity are related to a bad response to treatment (less fibrosis post-treatment). There is a positive association with fibrosis post-treatment and survival.

Limitations:

The subjective reporting of the non-quantitative data and that not all patients had all data.

Ethics committee approval

n/a

Funding:

No funding was received for this work.

Purpose:

To analyse the feasibility of diffusion-weighted MRI (DWI) for predicting treatment response in multiple myeloma (MM) lesions compared to FDG PET-CT.

Methods and materials:

20 patients with MM undergoing same-day whole-body MRI and FDG-PET/CT for the evaluation of treatment response were retrospectively recruited. Overall, 63 whole-body MRIs were evaluated. 3 significant bone lesions were analysed for each patient. Quantitative analysis of ADC maps and SUVmax of each lesion were evaluated in the subsequent studies. Variation of ADC values was compared with the variation of SUVmax for each lesion, which was considered the reference standard for this study. Differences between the mean ADC value in lesions from patients with clinical progression and patients with treatment response were also analysed, considering the criteria of the international myeloma work group (IMWG). Data analysis was performed using SPSS 21.0 software (Spearman’s rank correlation coefficient and Student’s t-test). Two-tailed p values were used for all statistical assessment and p<0.05 was considered statistically significant.

Results:

An inverse, high, and significant correlation between the variations of the ADC value and SUVmax (r=-0.686; p<0.01) was found. There were also significant differences in mean ADC values between lesions from patients with clinical progression and patients with treatment response (757.69 vs 1377.95, p<0.01).

Conclusion:

Variations of ADC values in DWI can be used to assess treatment response in patients with MM.

Limitations:

A retrospective study limited to a single centre and MRI protocol. The sample consisted of 20 patients in advanced stages.

Ethics committee approval

n/a

Funding:

No funding was received for this work.

Purpose:

To evaluate the diagnostic value for osteoporosis of dual-energy CT (DECT) based on the virtual non-calcium (VNCa) technique.

Methods and materials:

Prospective dual-energy CT scanning was performed on the lumbar region of 55 patients with chronic low back pain, with a standard QCT phantom placed at the waist during scanning. Using the QCTPro analysis system, the BMD of each vertebral body was measured. Based on the VNCa technique, the quantitative parameters were measured, including the CT value of calcium (contrast media, CM), CT value of mixed energy images (regular CT value, rCT), calcium density(CaD), and fat fraction (FF), by using Siemens dual-energy analysis software with a specially modified configuration file. Pearson's test was used to analyse the correlation between BMD and these parameters, and to establish a regression equation. The diagnostic efficiency for osteoporosis of the equation was evaluated by ROC curve.

Results:

The CM, rCT, CaD, and FF were significantly correlated with BMD (r-values were 0.885, 0.947, 0.877, and 0.492, respectively, and all P<0.01), the regressive BMD (rBMD) could be expressed with an equation as rBMD=54.82-0.19*CM+20.03*CaD-1.24*FF. Taking rBMD<81.94mg/cm³ as the threshold, the sensitivity and specificity were 90.0% and 92.0%, respectively, and the AUC was 0.966±0.009 (P<0.01).

Conclusion:

The calcium and fat quantitative related parameters of DECT have a good correlation with the BMD measured by QCT, and it may serve as an alternative method for the quantitative evaluation of osteoporosis in terms of mineral content and bone marrow fat composition.

Limitations:

We try our best to ensure ROI "point-to-point" correspondence to reduce the generation of system errors.

Ethics committee approval

This research was approved by the Ethics Committee of Xi’an Jiaotong University Affiliated Honghui Hospital.

Funding:

The Natural Science Foundation of Shaanxi Province (2017JM8152).

Purpose:

We assessed 3-dimensional changes to bone-mineral density during decalcification and fracture healing processes. Characterising these changes has direct relevance to a wide range of clinical situations including fracture risk, bone healing, and other diseases affecting bone.

Methods and materials:

3D calcium maps at 90 µm voxel size were obtained using a MARS spectral photon-counting CT. Measured calcium densities ranged from 0-1200 mg/ml.

In a decalcification study, 4 excised ovine femur bones were demineralised in 2.5% nitric acid at different time intervals (0, 3, 6, and 16 days). Atomic absorption spectrometry was performed on 251 times diluted nitric acid solution (collected at each interval) and compared with MARS 3D calcium maps of the biological sample.

In a fracture healing study, 4 sheep were administered strontium daily until sacrifice at week 2, 4, 6, and 8 after an induced fracture. Excised tibias were scanned and mineral content values from MARS were compared with DXA.

Results:

For decalcification, measured cortical BMD and total BMD (on 100 slices) decreased across the time 880±170 to 440±200 mg/cm³ and 410±230 to 213±180 mg/cm³, respectively. The trend was consistent with the atomic absorption spectrometry measurements of calcium loss into the nitric acid solution.

In assessing bone healing, MARS calcium maps show an increasing trend in cortical bone-mineral content as the fracture heals in all 4 sheep (1.04, 1.17, 1.31, and 1.72g). Total bone mineral content measurements with MARS and DXA correlated within 5 and 8% of each other.

Conclusion:

Spectral photon-counting CT with a MARS scanner can quantify and visualise in 3D the changes in bone mineralisation. This will impact research, diagnosis, and management of a wide range of bone conditions.

Limitations:

n/a

Ethics committee approval

Animal ethics committee: AEC2018-31A.

Funding:

No funding was received for this work.

Purpose:

To explore the typical magnetic resonance imaging (MRI) pattern of osteoblastoma (OB) after radiofrequency ablation treatment (RFA) and to derive signs that indicate treatment success or relapse.

Methods and materials:

A total of 43 follow-up MRI examinations on 15 OB patients that underwent a total of 19 RFAs were analysed retrospectively. An early follow-up group (1-4 months after RFA) and late follow-up group (8-131 months after RFA) were created. Groups were differentiated depending on clinical treatment success. Images were analysed for the presence of central nidus enhancement (CNE), peripheral nidus enhancement (PNE), perifocal bone marrow oedema (PBME), and fatty nidus conversion (FNC).

Results:

Every patient (n=14/14) in the treatment success group showed a target-like appearance with negative CNE and positive PNE or PBME at early follow-up. PNE was present in 93% (n=13/14) and PBME in 71% (n=10/14) of early follow-up images. At late follow-up, the target-like appearance was observed in 25% (n=5/20) of exams and 20% (n=4/20) for both PNE and PBME. FNC was not observed at early follow-up but occurred in 55% (n=11/20) of late follow-up exams. All three included MRI exams at clinical recurrence showed strong CNE, PNE, and extensive PBME in contrast to exams of the treatment success group.

Conclusion:

The target-like appearance of OB indicates treatment success in early follow-up MRI exams. PNE and PBME typically reduce over time and can lead to FNC in successfully treated patients. CNE recurrence with PNE and extensive PBME are signs of relapse. PNE and PBME do not indicate treatment failure at early follow-up.

Limitations:

The retrospective study design and limited patient number due to the rareness of osteoblastoma.

Ethics committee approval

The study was performed with a waiver of informed consent by the local review board.

Funding:

No funding was received for this work.

Purpose:

To investigate the value of MR texture analysis (TA) to differentiate malignant from benign soft tissue tumours at 3T MRI including IVIM DWI.

Methods and materials:

69 patients with pathologically confirmed soft tissue tumours (29 benign and 40 malignant) who underwent 3T MRI including IVIM DWI were retrospectively enrolled. TA was performed on axial T1WI, T2WI, CET1, high-b value DWI, and ADC map using software (TexRAD). Mean, SD, entropy, MPP, skewness, and kurtosis were calculated according to different Laplacian of Gaussian spatial filters: SSF 0, 2, 4, and 6. The ROC curve was obtained. Student’s t-test and multivariate logistic regression analysis were performed. The ROC curve with AUC was obtained.

Results:

SSF 4 showed overall higher AUCs than other SSFs, although there were no significant differences in different SSFs. AUCs of the mean of DWI, ADC, T2WI, and CET1 were 0.709 (95% CI, 0.586-0.831), 0.851 (0.751-0.951), and 0.772 (0.656-0.887), which were significantly different between benign and malignant soft tissue tumours (p<0.05). AUCs of kurtosis of DWI, T1WI, and CET1 were 0.748 (0.632-0.864), 0.732 (0.610-0.855), and 0.806 (0.700 -0.913), which were significantly higher in malignant tumors (p<0.05). AUCs of entropy of DWI, ADC, and T2WI were 0.763 (0.653-0.874), 0.763 (0.653-0.874), and 0.769 (0.657-0.880), which were significantly higher in malignant tumours (p<0.05). AUCs of MPP of ADC and CET1 were 0.734 (0.605-0.864) and 0.744 (0.624-0.864), which was significantly lower in malignant tumours (p<0.05). AUC of combined ADCmean and CEkurtosis was 0.883 (0.783-0.943) with a sensitivity and specificity of 83% and 86%.

Conclusion:

MR TA may be reliable and accurate in differentiating malignant from benign soft tissue tumours at 3T including IVIM-DWI.

Limitations:

A retrospective study with selection bias and relatively small study population.

Ethics committee approval

Approved by IRB and informed consent was waived.

Funding:

No funding was received for this work.

Purpose:

To evaluate treatment effects on osteoporotic vertebral compression fractures (VCFs) by using percutaneous vertebroplasty (PFP) and percutaneous kyphoplasty (PKP) and to document whether these effects are still detectable several years after the end of therapy.

Methods and materials:

From January 2002-February 2015, 49 patients who received PVP or PKP were evaluated. The surgical protocols and radiologically assisted CTs were retrospectively evaluated related to the occurrence of cement leakage. A special questionnaire regarding pain development before and after the procedure was analysed prospectively. The measurement of pain quality was based on the visual analogue scale (VAS).

Results:

VAS score was on average 7.0 before and 3.7 after therapy with a significant difference (p≤0.001) in the change of VAS scores in all patients. Considering the reduction of VAS over time of observation, there was no significant difference in the occurrence of symptoms (p=0.694) and no difference in the decrease in VAS over time. In 10 patients (22.73%), at least one cement leakage was recorded by CT scan. There was no significant difference in the occurrence of cement leakage and the change in VAS (p=0.146). After therapy, 79.6% of all patients reported an increase in mobility and 83.7% reported a rise in quality of life.

Conclusion:

Both PVP and PKP significantly reduce the pain in VCFs. These results are still being observed more than 10 years after the end of treatment.

Limitations:

Other interventions overlapped with the therapeutic approach documented here. Evaluation of pain before and after therapy was based on the subjective feelings of the individual patients. Comparison with non-surgical fracture treatment or a placebo group is missing.

Ethics committee approval

This study was approved by our local ethical committee. Written informed consent was obtained.

Funding:

No funding was received for this work.

Purpose:

To evaluate the diagnostic performance of dual-energy computed tomography (DECT) for sacroiliac bone marrow oedema (BME) detection in patients suspected for spondyloarthritis.

Methods and materials:

Patients aged 18-55 years with suspicion for spondyloarthritis underwent DECT and 3.0 Tesla MRI of the sacroiliac joints on the same day. Virtual non-calcium (VNCa) images were calculated from DECT images for BME detection, using a three-material decomposition algorithm for bone mineral, red bone marrow, and yellow bone marrow. Each sacroiliac joint was divided into four quadrants for analysis. VNCa images were scored by two readers independently: 0=normal bone marrow and 1=BME. A quadrant was recorded as BME present if either of the readers scored ‘1’ for the quadrant. Diagnostic performance was assessed with reference to MRI. CT numbers were measured on VNCa images. ROC analyses were performed to determine the cut-off values for BME detection.

Results:

40 patients (16 men, 24 women) were included. The sensitivity and specificity of BME detection by DECT were 65.4% and 94.2%. AUCs for CT numbers in the ilium and sacrum were 0.90 and 0.87, respectively. Cut-off values of -44.4 HU (iliac quadrants) and -40.8 HU (sacral quadrants) yielded sensitivities of 76.9% and 76.7% and specificities of 91.5% and 87.5%, respectively.

Conclusion:

VNCa images are useful for BME detection in patient suspected for sacroiliitis.

Limitations:

The sample size of the present study is small and the prevalence of BME is relatively low. The final clinical diagnosis was not taken into consideration for analysis.

Ethics committee approval

Institutional review board approved this study. Written informed consent was obtained.

Funding:

The study is partially funded by the Young Researchers Grant awarded by the European Society of Musculoskeletal Radiology.